NUREG-1549, Requests Placement of Matl Into Nudocs Sys & Into NRC Pdr. Matl Includes Working Paper on Dose Modeling & Draft NUREG-1549 Using Decision Methods for Dose Assessment to Comply W/Radiological Criteria for License Termination
| ML20199H584 | |
| Person / Time | |
|---|---|
| Issue date: | 02/03/1998 |
| From: | Cardile F NRC OFFICE OF NUCLEAR REGULATORY RESEARCH (RES) |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| RTR-NUREG-1549, RTR-REGGD-GENERA NUDOCS 9802050081 | |
| Download: ML20199H584 (490) | |
Text
{{#Wiki_filter:From Frank Cardile MS/MA/I To TWD1.TWPS.JDB3 h/M Date: 2/3/98 2:58pm
Subject:
MATERIAL FOR PDR Please place the attached material into the NUDOCS system and into NRC's Public Document Room. The material attached includes the " Working Paper on Dose Modeling" thar. is part of a future Regulatory Guide en " Demonstrating Compliance with the Radiological Criteria for License Termination." The attached Working Paper consists of the following pieces:
- 1) Section C.1 of the Guide "Doce Modeling"
- 2) Draft NUREG-1549 "Using Decision Methods for Dose Assessment to Comply with Radiological Criteria for License Termination" This Working Paper will be the subject of a public workshop to be held in the NRC Auditorium on February 19, 1998.
The workahop was announced in a Federal Register Notice dated January 28, 1998 (63 FR 4307). That FRN also announced that the Working Paper would be placed in NRC's Public Document Room prior to the workshop. Please also provide an advance copy of the korking Paper (containing Section C.1 and NUREG-1549 as noted above) to the NPC's Public Document Room Thanks. / CC: CXD, CAT 1 d s
) h' ' l s /,.! s+t f' - f f S chts
[pe, N i \ 9802050001 9802d3 PDR ORO NREB PDR lk!,kkf}ll! i
C. REGULATORY POSITION 1.0 Dose modelina Dose modeling is used to estimate the total effective dose equivalent to an average member of the critical group from concentrations of residual radioactivity. The critical group means the group of individuals reasonably
- expected to~ receive the greatest exposure to residual radioactivity for any applicable set of circumstances. The concentration of residual radioactivity distinguishable from background which. if distributed uniformly throughout a-survey _ unit, would result in a total effective dose equivalent of 25 millirems
() per year to an average member of the critical group is called the derived concentration guideline (Wilcoxon) (DCGly). NUREG-1549 (Reference 1) describes ucceptable methods to calculate DCGL, i values, including default DCGlu values for buildings and soil.' methods.to use modles and parameters to calculate site-specific DCGly values for buildings
- and soil methods.to calculate area factors for use with the elevated measurement concentration.-methods for handling special circumstances in buildings (embedded residual radioactivity, s' ers, waste plumbing systems, floor- drains. ventilation ducts, and embedded piping), and methods for handling special circumstances in soil (subsurface residual radioactivity.
DRAFT: February 2. 1998 6 G-ar 1
l l .h LJ rocks and debris, paved surfaces, non soil materials like asphalt or fly ash, groundwater, surface water, and sediments). Some facilities may have residual radioactivity comprised of more than one radionuclide. When there are multiple radionuclides rather than a single radionucl.ide, the dose contribution from each radionuclide needs to be considered. An acceptable method to do this is described in Chapter 11 of NUREG-1505 (Reference 2). In some cases the dose from one or more of the radionuclides in a mixture will dominate the total projected potential dose to s the extent that the uncertainty in the potential dose from other radionuclides in the mixture may be negligible. When there is a fixed relationship between f3 V the concentrations of the nuclides, such as for radionuclides that are in secular equilibrium. 611 the radionuclides should be considered in determining DCGly. The measurement to demonstrate compliance with the radiological criteria for license termination may be based on the single radionuclide that is casiest to measure. However, when there is no fixed relationship, it may be impractical to consider all constituents. For example, at a nuclear power plant, many different radionuclides could be present, and there may be no fixed relationship between their concentrations. However, it may be that almost all the dose would come from just e few of the radionuclides. In this situation, the presence of radioquclides that would contribution less than 10% O] \' DRAFT: February 2, 1998 7
f) O of the total effective dose equivalent may be ignored as long as the total contribution from all radionuclides not considered does not exceed 25%. In general, it is to the advantage nf the licensee to obtain NRC approval of its DCGL, prior to remediating the site and conducting the final radiation status survey because the remediation plan and the final radiation survey design will depend on the DCGly. Thus, if the remediation and the final radiation survey were conducted using a DCGly that was subsequently not accepted by the NRC, the remediation effort and final status survey might not t . be acceptable. Therefore. if the licensee will submit a decommissioning plan or license termination plan, the licensee should include the proposed DCGLy as O
!,j a part of the plan. If neither of these documents need to be submitted, the licensee may submit a proposed site-specific DCGly for NRC review anj approval prior to remediation.
References
- 1. NUREG-1549, " Guidance on Using Decision Methods for Dose Assessment to Comply With Radiological _ Criteria for License Termination." (To be published in 1988. Draft available on the web at http : / / techcon f. ll nl . gov /cgi - bi n/topi cs )
/m I i
V' DRAFT: February 2. 1998 8
4
- 2. C. V. Gogolak. E. E. Powers, and A. M. Huffert. "A Nonparametric Statistical Methodology for the design and Analysis of Final Status Decomissioning Surveys." NUREG-15^5 (publication exoected in summer 1998).
o f 1 i l l l 1 l DRAFT: February 2. 1998 9 4 4
.- .._c. , .,
l J l
- l l .
i i i i e i i 1 i i i i t-l l a l a DRAFT NUREG-1549: Using Decision Methods for Dose Assessment j to Comply With Radiological Criteria for License Termination e l i i l 4 i
~
t ( i l i, i i i i i i 1 4 A k m._. _., ..m,
Table of Contents t 1.0 I ntrod u ctio n . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 2.0 Overview of the decision framework . . . . . . . . . . . . . . . . . . . . . . , , . . . . . . . . . . . 4 3.0 Use of the Framework for Licensees That Use Generic Screening , . . . . . . . . . . . . . . 9 4.0 Use of the framework for licensees that use Site Specific Information to modify site parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 5.0 Use of the framework for licensees that use Site Specific Information and Consider a range of Options for using that information . . . . . . . , , , . . . . . . . , , 17 RE F E RE N C E S . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 Appendix A Default Concentr%+f Values Equivalent To 25 mrem /y Appendix B Annual Total Effec:<r Dose Equiva'ent Factors Appendix C - Scenarios. Pathways, and Critical Groups Appendix D - Dose Models Appendix E: Parameter Descriptions and Information for Changing Parameters Appendix F - Area Factors Appendix G - Examples I c i i i l
p) ( v 1.0 Introduction 1.1 Use of Decision-Making Framswork for Complying with NRC Regulations on Radiological Criteria for License Termination This NUREG contains an overall framework for dose assessment and decision making for site characterization, dose assessments, and remedial actions at sites licensed by the U.S. Nuclear Regulatory Commission (NRC). The framework can be used throughout the decommissioning and license termination process for sites ranging from the more simple sites to the most complex or contaminated sites. This document represents information for using the framework to step through the decommissior5g and license termination process. This framework is designed for coordinated and mutual use by the licensee, the NRC, and other stakeholders and is intended to streamline the process of coming to closure on decisions while providing a sound technical basis for decisionmaking. By diing so, the process allows the licensee to coordinate its planning efforts with the NRC's input, to conduct dose assessments and site characterization activities that are directly related to rQulatory decisions, to optimize cost decisions related to site characterization, remediation, and land-use restrictions, to integrate analyses for As Low As Reasonably Achievable (AL/< RA) requirements; and to elicit other stakeholders' input at crucial points. The framework also provides a comprehensive approach for treating the uncertainty associated with contaminated sites, including quantification, propagation, and reduction of uncertainty. 1.2 Content of the NRC regulations on Radiological Criteria for License Termination t ; u) On July 21,1997, the NRC published in the Federal Regis'er (62 FR 39058) a final rule incorporating a new Subpart E into 10 CFR Part 20 that inc'udes radiological criteria for license termination. Subpart E provides the regulatory basis for determining the extent to which lands and structures must be 9 mediated before decommissioning of a site can be considered complete and the license terminated. Subpart E of Part 20 includes requirements for ur; restricted and restricted use of facilities efter license termination in Sections 20.1402 and 20.1403, respectively. Subpart E also addresses public participation in the license termination process, the finality of Ibense termination decisions, time periods for dose calculations, alternate dose criteria, and minimization of contamination. The criteria for releasing a site for unrestricted or restricted use are listed here (and summarized in Table 1.1); 6 20.1402 - Criteria for unrestricted use - a site is considered acceptable for unrestncted use if the residual radioactivity that is distinguishable from background tr.diation results in a Total Effective Dose Equivalent (TEDE) to an average member of the critical group that does not exceed 20 mrem /yr, including that from groundwater sources of drinking water, and the residual radioactivity has been reduced to levels that are as low as is reasonably achievable (ALARA). () d Draft NUREG-1549 1 February 2,1998
A 6 20.1403 - Criteria for license termination under restricted conditions - a site is considered acceptable for license termination under restricted conditions if: (a) A licensee can demonstrate that further reductions in residual radioactivity necessary to comply with the provisions of 9 20.1402 would result in net public or environmental harm or were not made because the residual levels associated with restricted conditions are ALARA; (b) A licensee has m 'e provisions for legally enforceable institutional controls that provide reasonable auurance that the TEDE from residual radioactivity distinguishable from background to the average member of the critical group will not exceed 25 mremlyr; (c) A licensee has provided sufficient financial assurance to enable an independent third party to assume and carry out responsibilities for any necessary cuntrol and maintenance of the site. (d) A licensee has submitted a decommissioning plan or license termination plan specifying that the licensee intends to decommission by restricting use of the site and documenting how the advice of individuals and institutions in the community who may be affected by the decommissioning has been sought and incorporated into the plan. (e) Residual radioactivity at the site has been reduced so that if the institutional [) controls were no longer in effect, there is reasonable assurance that the TEDE Cl from residual radioactivity distinguishable from background to the average member of the critical group is as low as is reasonably achievable and would noi exceed either: (1) 100 mrernlyr; or (2) 500 mrem /yr provided the licensee: (a) demonstrates that further reduction'. in residual radioactivity necessary to comply with 100 mremly are not technically achievable, would be prohibitively expensive, or would result in net public or environmental harm; (b) makes provisions for durable institutional controls; and (c) provides sufficient financial assurance to enable an independent third party both to carry out periodic rechecks of the site every 5 years to assure that the institutional controls remain in place and to assume and carry out responsibilities for any necessary control and maintenance of those controls. This NUREG provides information on demonstrating compliance with the dose criteria and ALARA provisions of the unrestricted and restricted use requirements in Sections 20.1402 and 20.1403. Other requirements described above (e.g., public participation) are covered in Regulatory Guide 1.xxx. V Draft NUREG-1549 2 February 2,1998
4 lO Table 1.1 - Summary of 10 CFR Part 20 Subpart E 4 Unrestricted Release Restricted Release Dose Criterion 25 mrem TEDE per 25 mrem TEDE per 100 mrem or 500 4 year peak annual dose year peak annual mrem TEDE per year to the average member dose to the average peak annual dose to - of the critical group member of the critical the average member group while controls of the critical group are in place upon failure of controls
- Time Frame 1000 years 1003 years 1000 years Other ALARA ALARA, financial ALARA, financial Requirements assurance, public assurance, public participation participation d
i i 4 1 d Draft NUREG-1549 3 February 2,1998 (
-e- - -,
2.0 - Overview of the decision framework (n) V 2.1 Rationale for the decision framework The NRC is responsible for evaluating requests from licensees for the termination of the NRC license for their facilities. As part of the regulatory process, these facilities must demonstrate
. that they meet the radiological criteria for license termination in Subpart E of 10 CFR 20 (summarized in Table 1.1).
A logical, consistent decision process is viewed as a useful tool that will support licensee planning of decommissioning activities and NRC review of license termination requests. To support this process, this NUREG describes a decision methodology, or " framework," to support implementation of the dose assessment requirements in Subpart E. Dose assessments of facilities are typically used to demor' strate compliance with the criteria of Subpart E and generally rely on the use of site characterization and modeling and analytical tools. The principal components of the dose assessments are: (a) models for transport of radionuclides through tiie environment to a receptor, and (b) the parameters used in those models in these dose assessments, a reasonable treatment of uncertainty is needod to provide the regulator with the confidence that the actions taken and the decisions made to terminate the facility license are consistent with the regulations. The steps and decision points of the decision framework support cssessment of the entire , range of dose modeling options from which a licensee may choose, whether it involves using generic screening, parameters, changing parameters, or modifying pathways or models. The l cd, decision framewcrk, including its steps and decision points, is illustrated in Figure 1. P 2 Phased approach in using the decision frarnework 2.2.1 Contents of the phased approach in using the decision framework l To facilitate the preparation and evaluation of the dose assessments, this NUREG describes a phased approach to decision making for license termination. A phased approach is necessary because of the very wide range of levels of contamination and complexity of analysis and potential remeciation necessary at NRC-licensed sites. The phased approach consists of generic screening and of making use of site specific information as appropriate. These phases are described in broad terms below;
- 1) Generic screenino: In this phase, licensees would demonstrate compliance with the dose criteria of the rule by using: (a) pre-defined models, and (b) generic screening parameters.
Pre-defined models which use generic exposure scenurios and pathways are based on the NUREG/CR-5512 methodology and can be used with minimal justification by licensees who are applying generic screening scenarios and parameters using the DandD software. The minimum justification for the use of the default scenarios and parameters consists of a statement by the licensee that no conditions exist at the site, O b) Draft NUREG-1549 4 February 2,1998 . t
outside those incorporated in the default scenarios and modeling assumptions, that (pv) would cause the calculated dose to increar 3 Examples of site specific features that may require modeling beyond the defaults inuv,is (but are not limited to) known groundwater contamination, large quantities of contaminated material (such as slag piles), or buried wastes. The generic scenarios and pathwap of the pre-defined models provide the licensee with a simple method to demonstrate compliance using little or no site-specific information other than the source term. The pre-defined models and generic screening parameters are intended to approximate the upper range of the dose that the average member of the critical group could receive. The default screening parameters were selected probabilistically to control the regulatory risk associated with releasing a site based on source term data alone. Uncertainty in site conditions across sites is treated by the NRC through their use of a systematic quantification of the uncertainty that assures the regulator will not make an incorrect decision. In generic screening, the licensee need only provide site specific residual contamination information which is combined with NRC's pre-defined mocels and generic screening parameter values. If comphance is demonstrated using NRC's screoning models and parameter values, then progression to more site specific analysis is unnecessary. It is anticipated that the majority of NRC's licensees will be able to use generic screening to demonstrate that their site is acceptable for license termination, Use of the framework for facilities using generic screening will be relatively simple and [)
%.)
straightforward and penerally results in use of only Steps 1 - 7 of Figure 1.
- 2) Use of site soecific information as aoorooriate: If compliance cannot be demonstrated with generic screening , then there can be progression to a later phase in which defensible site specific values are obtained and applied. Depending on the complexity of the site contamination, the licensee can use site specific information by:
(a) using the NRC's pre-defined models but replacing generic screening parameters with site-specific parameter values to allow site specific factors to be taken into account. Thus, the dose estimates would be more realistic, but will still be conservative for a particular site based on the use of the pre-defined models. or (b) using both site-specific parameter values and site-specific model assumptions; (c) using some combination of a and b and also remediating the site; (d) using some combination of a, b, and c, and also restricting use of the site in any of the cases a - d, site specific data are used to support modifying or eliminating a particular scenario or pathway, or to demonstrate that a parameter or group of parameters can be better represented by site speciSc values. Altemative exposure scenarios may be appropriate based on site-specific factors that affect the likelihood and extent of potential future exposure to residual radioactivity. In cases a - d, the licensee O Draft NUREG-1549 5 February 2,1998
would have to provide justification for the site specific parameter values or for the (v f alternative model, as appropriate. Thus use of the framework for these situations can range from faiily simple site assessments to fairly complex analyses. In either case, use of all 12 steps of the framework in Figure 1 are likely used in these cases, although the range of options analyzed in Step 8 can be fairly simple (e.g. modification of parameters) to fairly complicated (e.g., use of restrictions on site use). 2.2.2 General concepts regarding the phased approach The following general concepts apply to using the phased approach with the decision framework: a) The approach provides a logical, iterative process for screening sites and for directing additional data collection efforts where necessary. It provides the licensee with a variety of options for performing dose assessments from simple screening to more detailed site specific analyses. The framework is designed such that the level of complexity and rigor of analysis conducted for a given site shou;d be commensurate with the level of risk that the site poses. Although all sites are expected to step through Steps 1 through 5, and steps 6 and 7, the amount of work that goes into each of these steps should be based on the s expected levels of contamination and the health risks they pose. Note that in this ,[V) framework, all sites may start at the same level of very simple analyses (not a requirement for successful implementation), but it is expected that only certain sites l would progress to very complex dose assessment and options analyses. Some sites may not need to conduct any options analyses (Step 8) and some sites may need to evalt. ate a limited set of relatively simple and inexpensive options. For example, a site with a contained source of contamination that is obviously simple to remove would not spend time analyzing large suites of alternative dats collection and remediation options. On the other hand, a site with high levels of contamination that are widely distributed may use this process to analyze a variety of simple and complex options to define the best decontamination and decommissioning strategy. Thus, the approach ensures that the NRC's, the licensee's, and other affected partiec' efforts and expenses are commensurate with the level of risk posed by the site; b) The licensee need not start the process with generic screening but can move directly to use of site specific information, as appropriate c) Consideration of risk is implicit in the methodology in that the likelihood of receiving less than the simulated dose is accomplished through the use of probabilistic treatment of parameter uncertainty (implicitly and/or explicitly). The NRC will not require licensees to conduct probabilistic analyses in their evaluation of compliance; however, a robust treatment of uncertainty will be needed to lend credibility to the results and confidence to fi O Draft NUREG-1549 6 February 2,1998
p re0ulatory decision making when site specific information is used. Data collection V) ( activities can be tied directly to the regulatory dose-based performance objectives A key point in implementing this framework is that, as new data is collected or the site is remediated, the simulated dose should decrease with each subsequent dose assessment. d) For the process to work correctly and efficiently, the bcensee is encouraged to involve the NRC from the very first step through the end of the decision making process. Chapters 3,4 and 5 of this NUREG describe how to apply the decision framework of Figure 1 to the wide range of NRC licensees, from those witn relativcly simple decommissioning situations, to those with intermediate levels of contamination, and to those with more complex contamination patterns, respectively, Chapters 3,4 and 5 provide descriptions of each of the framework steps (see Figure 1) in some detail, and how they integrate to define a process for moving through the framework to treat uncertainty and define a license termination strategy. It is important to note that these chapters and the process of considering them by any particular licensee should be fluid, that is a licensee may, in considering options for dose assessment and license termination, use any one of the chapters or all of them. Detail on each step is provided in Chapter 3 for sites that use the generic screening approach developed by NRC, in Chapter 4 for sites that use an approach of incorporating site specific information because they have intermediate situations, and in Chapter 5 for more complex situations. Licensees using codes and modeling approaches other than generic screening a . O. should use Chapters 4 and 5. Chapter 4 is presented separately from Chapter 5 because of Q the large differences in the level of analysis and evaluation necessary for the wide range of NRC licensees that use site specific information. This may cause some repetition but it is l expected to be most useful to licensees to be presented in this manner. CN b Draft NUREG-1549 7 February 2,1998
_ . _ . _ . . _ _ _ _ _ _ _ . _ . . . . . . . _ . . _ . . _ _ . _ . _ . . _ _ _ . _ . . _ . . . _ _ . _ _ _ _ . . . ~ . _ _ _ _ _ 4 , AssimiladonofExistingData andinformation
=
0 _ Scenarlo Definitioni PathwayIdenefication j I ReviseMode! Assumptions, i System p ParameterValues,&Pathwsys ! Conceptualization . an4EvaluateResults Y I Dose Assessmert l If O implement I Profondoption j Define "
, Can
' ! Site Charactertzstion, I k Site be RoleeseJ?
"O
? Romedation, and RestrictedUse0ptions U
yes $ O Select ! Analyze Optforisinterms of Cost Preferred O g j g ALARA Requir. meres andLikelihoodofSuccess Option i e LicenseT. Instionand S.eRe,e..e ucenseWai ned l Figure 1 Decommissioning and License Termination Framework i Draft NUDIG-1549 8 February 2,1998 l
3.0 Use of the Framework for Licenrees That Use Generic Screening U As notec above, this chapter describes the use of the framework for licensees that use the generic screening approach described in Section 2.2.1 (as noted in Section 2.2.2, licensees may use other models or codes but should use the steps in Chapters 4 or 5 of this NUREG with those approaches), in general, a licensee with a relatively simple decommissioning situation would follow Steps 1 through 7 of the framework of Figure 1. Such licensees would i likely include those NRC licen.ees with contained or short-lived radionuclide scWes that have ' small amounts of buiding or soil contaminatio.t An example of the use of the framework for such a situation is given in Appendix G as Case 1. Licensees of this type would step through the framework as foliows:
- 11. Sten 1:
This step involves gathering and evaluating existing data and information. Licensees should check their records to determine the types and amounts of radioac:ive material they possesseo on their site. They should also gather information about any surveys and leak tests thct had been performed, as well as any recorr's that would support their ability to terminate the license uader 10 CFR Parts 30,40,50,70, or 72, as appropriate. For example, Part 30 licensees should gather information sufficient to " Certify the disposition of all licensed matcrial, including accumulated wastes, by submitting a completed NRC Form 314 or equivalent information' [10 CFR 30.36(j)(1)). 7_. Licensees using generic screening would be making use of the NRC developed DandD models / \ and generic screening concentration values, and therefore only site-specific source term data d quantifying the amount of contamination present would need to be gathered.
- 12. Sten 2:
This step involves defining the scenarios and pathways that are important for the site dose assessment. For a licensee using the generic screening pararneters, this step has already been completed by the NRC, based on the generic scenarios and pathvvays for screening that have been defined and described in NUREG/CR 5512, Volume. Information on generic scenarios and pathways is presented in Appendix C.1. 1;L Steo 3: This step involves system conceptualization, which includes conceptual and mathematical model development and assessment of parameter uncertainty, For a licensee using generic screening, this step has already been completed by NRC, using the models described in NUREG/CR-5512, Volume 1, and imp lemen%d in the DandD softwaro, information on generic models for system conceptualization is presented in Appendix D.1. Thus, a licensee using generic screening could use the DandD software containing pre-defined models and default parameters, or as noted in Step 4, could instead use the radionuclide-specific default concentrations contained in Tables A-1 or A-2 of Appendix A of this NUREG. The minimum justification for the use of the default models, scenarios, and parameters consists U Draft NUREG-1549 9 February 2,1998
i of a statement by the licensee that no conditions exist at the site, outside those incorporated in i (n} C the default scenarios and modeling assumptions, that would cause the calculated cose to increase. Examples of site specific features that may require modeling beyond the defaults i include (but are not limited to) known groundwater contamination, large quantities of l contaminated material (such as slag piles), or buried wastes. i
- 14. SitP__41 This step i. wolves the dose assessment for the site, For a licensee using generic sci vening, the licensee can either: (a) use the generic screening concentrations that correspond to the 25 mremlyr dose criterion which have already been calculated by the NRC (see Tables A-1 or A-2 of Appendix A of this NUREG) to compare against the site contamination levels obtained in Step 1, or (b) run DandD with the appropriate site specific source term.
Defense and justification orovided by the licensee for the selection and use of the DandD code would not be necessary. The licensee would provide to the NRC a copy of the DandD generated report to verify the version of DandD that was used in the analysis. Defense is needed of the source characterization and to show that DandD is applicable for the site conditions (discussed under Step 3). M_ Sina 5; TNs is the first major decision point in the framework and involves answerin w me question of whether the dose assessment results of Step 4 are less than the dose criterion of 25 mremlyr in (s) C/ 10 CFR 20, Subpart E. For a licensee using DandD or equivalent code with default values and site source term, or using the generic screening concentrations of Tables A-1 or A-2, the licensee would find either that: a) The result in Step 5 is that the calculated dose is less than 25 mrem /yr or site contaminatioa is less than or equal to the values in Tables A-1 or A 2. If this is the case, proceed to Step 6 b) The result in Step 5 is that the calculated dose is greater than 25 mrem /yr er the site contamination is greater than the values in Table A-1 or A-2. If this is the case, it means that the contamination at the site is such thht the licensee cannot use the generic screening approach to terminate the licensee. Rather,in order to terminate the license, the licensee would need to evaluate other options such as incorporating site specific information into the dose assessment. Thus, if this result is found, the licensee should proceed to Chapters 4 or 5 and use the framework steps applicable to use of site specific information.
- 16. Step 6:
If the result in Step 5 is that the calculated dose is less than 25 mrem /yr or the contamination at the site is less than the values in Tables A-1 or A-7 .ne licensee can proceed to satisfy ALARA requirements, if not already addressed (see Section 4.0 of Regulatory Guide x.xxx) (3
-V) Draft NUREG-1549 10 February 2,1998
LZ. - Step 7; in this step the final paperwork requirements are completed, including documenting any survey results used to calculate the source term and the results of the dose calculations, and the licensee would request that their license be terminated by the NRC. (- l
\
l D Draft NUREG-1549 11 February 2,1998
4.0 Use of the framework for licensees that use Site Specific Information to modify site (v j parameters This chapter describes the use of the framework for the wide range of licensees that may use site specific information in their dose assessment. As described in Section 2.2, there are a wide range of options for using site specific data ranging from modifying parameters, to modifying models, to remediating the site, to res'.ricting site use. This chapter describes use of the framework specifically for those licensees that choose the option of modifying parameters without further considerttion of other options. This chapter is prepared separately from Chapter 5 (which includes a more in-depth evaluation of options) because it is thought that a number of licensees will have relatively low levels of contamiriation and will seek to perform a dose assessment by changing default parameters to more adequately represent the:r site. This could include those licensees who attempt to use the generic screening approach of Chapter 3 but who do not meet the unrestricted release criter'a in step 5. This chapter only describes the option of changing modeling parameters but is not intended to limit the options a licensee may pursue. For example, it is possible that a licensee could combine obtaining additional site data to revise parameters with remediaong a site, or could even proceed directly to remediate a site. A licensee who is uncertain of what option is most appropriate should proceed to Chapter 5. This chapter provides information for licensees who,
" assessing relatively simple contamination patterns, have used a correspondingly sin:ple onsideration of their options to conclude that modifying parameters from the screening values
\
,I will provide a simple, cost effective means to comply with the dose criteria of Subpart E. It l () should be further noted that licensees who proceed through the framework as cutiined in Chapter 4 can still proceed to Chapter 5 if necessary. An example of the use of the framework for the situation discussed in this chapter is given in Appendix G as Case 2. M Steps 1- 5 Licensees using this approach are assumed to have little information about their site initially and are assumed to go though the process of generic screening to determine if their site can be released ueng the generic screening concentrations. Thus, Steps 1 - 5 would be the same as described in Chapter 3. It is further assumed that theso licensees on their initial pass would end up in Step Sb in which the contamination at the site is such that the licensee cannot use the generic screening approach to meet the dose criteria of Subpart E. Thus, rather than proceeding to Step 6 and 7, these li:ensees would proceed to Step B. 4.2 Step 8 - Define Site Characterization, Remediation, And Restricted Use Options The purpose of Step 8 is to define options for proceeding with the license termination process. These options are presented here as information for licensees in planning their dose assessments and their submittals to the NRC. As described in Chapter 1 above, it is thought that a well thought out consideration of options for compliance with Subpart E and for submittals V Draft NUREG-1549 12 February 2,1998
& to NRC will enhance the process of decision-making on both on the licensee's and the NRC's h)
V part by allowing the licensee to make decisions in a timely manner that are both cost-effective and have a sound technical basis. There are basically three options that the licensee could apply either alone or in combination: a) Option 1 - Activities that reduce uncertainty (information/ data collection) in the calculated dose through use of source terms, pathways, models, and/or parameters that better represent the site based on some additional site information gathering or characterization , b) Option 2 - Activities that reduce contamination (remediation), and c) Option 3 - Activities that reduce exposure (land use restrictions). Chapter 4 assumes that the licensee will proceed to use Option 1. Most sites would perform an analysis of the options that is relatively simple and arrive at Option 1 because the nature of the contamination or the site conditions appear likely to support a lower estimated dose. Licensees might elect to use Option 1 before proceeding to other more complex adivities such as excavating, transporting and disposing of soil from the site that would be involved in Option 2 or establishing institutional controls for restricted use that would be involved in Option 3. An example of a process of considering options that a licensee might ase before arriving at a decision to use Option 1 is shown in Table 5.2. For Option 1, licensees should do the following: [j L a) Review the rarameters in the NUREG/CR-5512 model and what they recraigp2 The parameter distributions and their rationa'e are presented in Attachment 1. The rationale for parameter selection for the generic screening approach is presented in Section 2.2.1 of this NUREG where it is noted that generic screening analysis is intended by NRC to provide a specified level of confidence in the dose estimate and to control the amount by which the dose could exceed the criterion. b) Consider how to modify the carameters to incoroorate site soecific 'nformation and determine the data needs to modify the carameters. Attachment 1 provides information regarding the valid ranges for site specific parameter changes that a license could propose without an additional uncertainty analysis. As a consequence, the licensee needs little supporting information to defend changes to the parameter values that are within the limits specified in the parameter analysis. This is important in evaluating the relative worth of collecting additional data on these parameters under Step 9 of the decisic.n framework. Appendix E provides information on how to modify the parameters used in the dose assessment . gy ! ) V Draft NUREG-1549 13 February 2,1998
l 4,3 Step 9 - Analyze Options in terms of Cost and Likelihood of Success d This step involves the analysis of options in terms of cost and the likelihood of success. As noted above in Step 8, the purpose of this step is to provide information for the licensee so that the evaluation of options considers both the probability that a desired result will be achieved, (i.e., meeting the criteria of Subpart E), and that acileving the desired result is done in a cost-effective manner. For the iicensee choosing Option 1, Step 9 should consist of the following: a) an evaluation of the level of detail and information sources to use to better estimate values for the model parameters that will be updated with site-specific information. . Such an evaluation is important because there are many options for modifying parameters which range in cost and complexity depending on whether low or no cost information is easily accessible or will require expensive or specialized laboratory analysis , This evaluation can be done by reviewing the parameters in Appendix E. b) The cost and time needed to review each parameter should tm estimated, along with the likelihood that the approach will be successful in meeting the desired endpoint (i.e., meeting the criteria of Subpart E). The analysis should also address the uncertainty associated with each potential outcome. If the activity is successful, then the revised calculation of dose will meet the Subpart E criteria,, no follow on activities are necessary, and no other significant costs would be ( incurred. On the other hand, if the activity is unsuccessful, the esentual totai cost ends ( up being the cost to conduct the activity plus the cost to conduct any necessary follow-on activities to get the dose to an acceptable level. c) A decision should be made regarding the method for gathering information to revise parameters based on a and b, above. Note that actualsuccess or failure of this effort will not be realized until the second iteration of Steps 4 and 5. 4,4 Step 10 - Select Preferred Option
- The activities in Step 9 provide information for the licensee using Option 1. In this case, at this step, the licensee would choose the method for revising the parameters given the cost, timeliness and likelihood of success.
4.5 Step 11 -Implement Preferred Option Under Step 11, the preferred option is implemented. The licer see obtains the information necessary to support revisions to the parameters that will be modi %d. 4,6 Step 12 - Revise Model Assumptions, Parameter Values, and Pathways I C Draft NUREG-1549 14 February 2,1998
Under Option 1, the parameter values for the pre defined models are revised as appropriate, O i To support a future request for license termination, any site survey results, parameter data, y laboratory tests should be carefully documented. The process that the licensee shouM go through to justify ne.v parameter values or refine parameter distributions is presented in Appendix E. M Reiteration of sten 4: The revised parameter values are used in iteration 2 of the dose assessment. For the licensee only changing parameters, the original default model assumptions and pathways would remain - unchanged. M Reiteration of Sten 5: The revised dose assessment is evaluated to determine if the calculated dose meets the requirements in 10 CFR 20, Subpart E. For a licensee using site specific information to modify the parameter values, the licensee would find either that: a) The result in Step 5 is that the calculated dose is less than or equal to the 25 mrem /yr dose criterion of 10 CFR 20.1402. If this is the case, proceed to Step 6 - i b) The result in Step 5 is that the calculated dose is greater than the 25 mrem /yr dose criterion of 10 CFR 20.1402.. If this is the case, it means that the contamination at l the site is such that the licensee cannot simply revise the parameters of the dose assessment to comply with Subpart E. Rather, in order to terminate the license, the licenses would need to incorporate additional site spec.ific information into the dose asvassment, or possibly consider remediation or restricting site use. Since the initial simple approach of revising parameters has not proven acceptable, licensees should proceed to Chapter 5 and use the framework steps applicable to further analysis of options . The licensee is encouraged to actively work with the NRC during this step to evaluate the appropriateness and adequacy of the analyses before moving on and expending resources on follow on steps. 4.9 Step 6 - ALARA Requirements If the result in Step 5 is that the 25 mrem /y criterion has been met, the licensee can proceed to satisfy ALARA requirements, if not already addressed. ALARA actions at this step can be based on Section 3 of Reg Guide x.xx. The licensee is encourage to actively work with the NRC to discuss, define, and concur on alternative ALARA actions under this step prior to
- implementing any actions.
1 4.10 Step 7 - License Termination and Site Release Draft NUREG-1549 15 February 2,1998
in this step the licensee would complete final paperwork requirements, including documenting ( any survey results used to calculate the source term and the results of the dose calculations, and would request that their license be terminated by the NRC. b O Draft NUREG-1549 16 February 2,1998 l
p 5.0 Use of the framework for licensees that use Site Specific Information and Consider a range of Options for using that information This chapter describes the use of the framework for the wide range of licensees that may use site specific information in their dose assessment. As described Section 2.2, there are a wide range of options for using site specific data. Chapter 4 discussed the framework for those licensees who take the option of merely modifying model parameters. However, there may be sites with complex or substantial contamination for which it may be necessary to: a) change the models, scenarios, pathways, and/or parameters used for assessing nuclide j behavior to support release of the site for unrestricted use, b) remediation of the site by removal of soil or concrete, c) restrict future use of the site under the requirements of 10 CFR 20.1403, d) perform some combinction of a, b, or c. Licensees with fairly complex situations may already have considerable information about their site. Such licensees may choose not to use generic screening, preferring instead to immediately utilize as much existing site-specific information as possible. Therefore the discussion of the use of the framework for these sites begins with the licensees using site specific mformation in Steps 1 - 4 rather than using the generic screening approach of Chapter 7 3 (alternatively, even a licensee with significant site-specific information may find it useful to
/
\ start with the generic screening in the initial iteration (see Section 4.1)). Licensees using this V approarn would step through the framework as follows:
L1 Sten 1 - Assimilate Existina Data and Information: This step involves gathering and evaluating existing data and information. Licensees should check their records to determine the typer and amounts of radioactive material they possessed on their site. They should also gather information about any surveys and leak tests : hat had been performed, as well as any records that would support their ability to terminate the license under 10 CFR Parts 30,40,50,70, or 72, as appropriate. Data gathered in this snp are used to support Step 3 which is development of a conceptual model, and model assumptions and model parameter values. As described above, the licensee has 3 options in this anelysis: (1) use the pre-defined DandD models and the specified set of site-specific parameter values (2) use other existing models and codes and site-specific parameter values (3) develop site-specific models and codes and accompanying parameter values Additional information is needed to support and defend the conceptual model of Step 3 if models other than DandD are used or if site specific parameter values are used. Methods for
/% / \
V Draft NUREG-1549 17 February 2,1998
(p
%,)
obtaining the necessary additional information to support the site specific parameters and models used are dessibed in Appendix E.2. 54 SteD 2 - Scenarlo Definition / Pathway Identification: This step involves defining the scenarios and pathways that are important for the site dose assessment. In this step, the licensee defines potential human activities and identifies migration and exposure pathways that need to be considered. Each of the site release conditions defined in Subpart E (cnrestricted use or restricted use) involve potentially different considerations with respect to humar. activities on or near the site (the critical group) and radionuci,Je migration pathways. These should be considered as follows: a) Scenarios are defineo as plausible sets of human activities and future uses of the site. As such, scenarios provide a description of the reasonable future land uses and human activities over the period of interest. With an understanding of the physical system and potent.at human activities, one can then develop conceptual models of the sita (Step 3). Those conceptual models are translated into mathematical models and implemented in (and solved by) corresponding analytical or numerical models and computer codes. The objective is to calculate a dose (Step 4) which is then compared with dose criteria (Step 5) to assess whether the site complies with requirements of Subpart E. b) The definition of scenarios and identification of pathways can be generic or site specific; [,,\ however, these definitions should be such that they adhere to the ccnstraints of the iterative approach as defined in this document. That is, the simulated dose should decrearn with each iteration if the scenarios and pathways are changed based on site-specific information. The generic scenarios, critical groups, and pathways acceptable to the NRC are described in Appendix C.1 Licensees choosing to modify the generic scenarios, critical groups, and/or pathways should use the information in Appendix C.2 which describes the method the licensee would use to consider appropriate critical groups for their site and to develop site specific scenarios and pathways.
- 52. Stoo 3 System Conceptualipitom System Conceptualization, as defined here, includes conceptual and mathematical model development and assessment of parameter uncertainty. This assessment of uncertainty includes a process of systematically evaluating the level of uncertainty associated with a specific site and the qu3ntification of that uncertainty. In order to manage the treatment of uncertainty associate:i with dose assessment at a given site, the four steps of scenario definition, pathway iuentification, model devclopment, and assessment of parameter uncertainty are treated as a hierarchy, moving from the former of these to the latter.
As with the pathways, conceptual and mathematical models have been defined for the NUREG/CR-5512 methodology and these models (codified in the DandD code) are acceptable p ( Draft NUREG-1549 18 February 2,1998
for making cose assessments Information on the generic models is contained in Appendix (q - D.1. In add. tion, Appendix D.1 provides information that the licensee should uso in evaluating b) whether N nat the generic models are appropriate for their site given the assumptions made in NUREG-5512 end the nature of their site, if the licensee chooses to develop site specific models (either through changes to the default parameter values, model assumptions or development of new models), then the licensee would need to defend the model and associated parameters. Information on methods for defend site specific models is contained in Appendix D.2, including information on develooment of the model (D.2.1), use of a deterministic or probabalistic approach (D.2.2), and selection of codes j (D.2.3). 5.4 Step 4 - Dose assessment This step involves the dose assessment for the site, which means running the DandD or equivalent software with the appropriate site specific source term. In this step, the licensee will calculate potential doses using mathematical representations of the conceptual models. This step involves the execution of the numerical model(s) that implement the mathematical equations and vrill provide the basis for (1) assessing compliance with the individual dose criteria and (2) an analysis of the impact of uncertainty in models and input parameters on the model output. In drMg so, this step includes the propagation of uncertainty in parameters through eposure models and should provide a quantitative representation of the uncertainty in the dose given those models and parameters. NRC has implemented the default scenarios, critical groups, pathways, model assumptions and parameter values from Steps 2 and 3 in the DandD code. The licensee has the option of using DandD for the dose assessment, under the conditions discussed in Appendix D 1 for justifying assumptions and parameter values for their site. The rationale and approach for using the generic models and code selection are contained .n Appendix D.1.1 and D.1.2. Information on methods to perform site specific dose assessments is contained in Appendix D.2.1 through D.2.3. 5.5 Step 5 - Can the site be released The dose assessment using the site specific information generated in Steps 1 - 4 is evaluated to determine if the calc . lated dose meets the requirements in 10 CFR 20, Subpart E. For a licensee using the site specific information, the licensee would find either that: a) The result in Step 5 is that the calculated dose is less than or equal to the 25 mrem /yr dose criteria of 10 CFR 20.1402. If this is the case. oroceed to Steo 6 b) The result in Step 5 is that the calculated dose is greater than the 25 mrem /yr dose criteria of 10 CFR 20.1402.. If this is the case. It means that the contamination at the site is such that the licensee would need to consider additional options to terminate the licensee. In order to terminate the license. the licensee would need Draft NUREG-1549 19 February 2,1998
jnraroorate additional amounts of site specific information into the dose ( Rs191 ament. or nossibly consider remediation or restrictina site use. Thus. If this. reeult is found. the licensee should would oroceed to Steo 8. The licensee is encouraged to actively work with the NRC during this step to eva'Jate the appropriateness and adequacy of the analyses before moving on and expending resources on follow on steps. 5.6 Step 8 - Define Site Characterization, Remedletion, And Restricted Use Options The purpose of this step is to for the licensee to better define its options for proceeding with the license termination process. These options are presented here as information for licensees in planning their dose assessments and their submittals to the NRC. As described in Chapter 1 above, it is thought that a well thought nut consideration of options for compliance with Subpart E and for submittals to NRC will enhance the process of decision-making both on the licensee's and the NRC's part by allowing the licensee to make decisions in a timely manner that are both cost-effective and have a sound technical basis. It w.ll also allow the licensee to define the ' most effective and cost efficient decontamination and decomm ssioning strategy. Section 5.6.1 presents the principal options and Section 5.6.2 present the actions that a licensee should take in considering these options. Table 5.1 presents a summary of a licensee's possible process for considering the options. 5.6.1 Defining options There are basically three options that the licensee could use. Generically, the options are: (
\ 1) Option 1 - Activities that lower dose by reducing uncertainty (information/ data collection)
- 2) Option 2 - Activities that lower dose by reducing contamination (remediation)
- 3) Option 3 - Activities that lower dose by reducir.y exposure (land-use restrictions)
The options can be implemented singly or in combination, and the combinations can be l performed either in parallel or in series to provide the optimal solution. In addition, there could be a large number of combinations of site characterization data collection options. Examoles of combined alternatives include: site characterization for revision of source term (Option 1) combined with remediation (Option 2) followed by unrestricted release,
. site characterization for revision of source term (Option 1) combined with literature / database review to support default parameter replacement (Option 1) followed by unrestricted release, site characterization for revision of source term (Option 1) combined with literature / database review to support default parameter replacement (Option 1) combined with remediation (Option 2) followed by unrestricted release, L Draft NUREG-1549 20 February 2,1998
p) ( V remediation (Option 2) combined with land-use restrictions (Option 3) followed by restricted release. Another example is the application of land-ure restrictions to some portions of the site and remediation and unrestricted release to other portions of the site to reduce long term maintenance, monitoring and assurance costs. 5.6.2 Specific Licensee Actions Under the Options A licensee in Step 8 should go through a consideration of the options in a manner similar to the following (Table 5,1 presents a summary of a licensee's possible process for considering the options):
- 1) Option 1 - Activities that lower dose by reducina uncertainty finformation/ data collection and revision of source term. narameters. and/or models)
This cption wculd be pursued if existing information does not result in t.1e dose criterion being met, but further reduction of uncertainties and relaxation of conservatism through use of site-specific data are likely to result in a calculation of dose that meets the enteria of Subpart E. Specifically, these activities are data and information collection activities that would result in a reduction of uncertainty in the calculated doses through use of source terms, pathways, models, and/or parameters that better represent the site, in Option 1 licensees should do the following: a) Review the default carameter values in the NUREG/CR 3512 model and what [.
\
thev reoresent: The default parameter distrib.;tions and their rationale are presented in Appendix C. The rationale t< the default parameter selection is presented in Section 2.2.1 where it 5 noted that the screening parameters are selected by NRC to provide a sr.,cified level of confidence in the dose estimate and control the amount by W1lch the dose could exceed the criterion. b) Consider how to modify the default carameters to consider site soecific information and determine the data needs to modify the carameters: The parameter analysis indicated in Appendix C provides information regarding the valid ranges for site specific parameter changes that a license could propose without an additional uncertainty analysis. As a consequence, the licensee needs little supporting infomiation to defend changes to the carameter values that are within the limits specified in the parameter analysis. This is important in evaluating the relative worth of collecting additional data on these parameters under Step 9 of the decision framework. c) Consider whether modification of the critical arouo is accrooriate: Various site uses ar.d scenarios can be postulated within the f: nits of reasonable future uses for the site and surrounding properties. Licensees need to specifically define the critical group. Initialiterations of the decision process defined by this document may simply involve use of the screening scenarios and screening groups listed in the previous section. Subsequent iterations may involve site specific scenarios and critical groups (referred to as " site specific critical groups"). Background D Draft NUREG-1549 21 February 2,1998
information on critical groups is contahed in Appendix C.3 and ir. formation on developing eite specific scenarios and critical groups is in Appendix C.2 (2) Qation 2 Activities that lower dose by teducina contamination (rmediation) This option involves remediation activities that remove actual contamination from the site or reconfigure contamination (physically or chemically) such that transport to a receptor is decreased. Option 2 results in actual reduction of the gimntity of residual radioactivity remaining on the site. (3) Option 3 - Activities that lower dose by reducina exnosure fland use restrictions {_ This option would be pursued if the licensee is considering restricting use of the site as a means of terminating the license. If Option 3 is pursued, the liconsee is required by 10 CFR 20.1403 to conduct the following additional analyses and activities: (1) demonstrate that achieving unrestricted release is not ALARA, (2) provide legally enforceable institutional controls that would limit the exposure to individuals to 25 mrem /yr, (3) provide financial assurance for the controls, and (4) seek advice from those in the community who may be affected by the decommissioning. In order to comply with these requirements, a licensee should do the following as part of considering Option 3: a) Because use of Option 3 requires a demonstration to the NRC that further s reduct;on in dose levels to unrestricted use is not ALARA (i.e. NRC would prefer unrestricted use), licensees should fully evaluate unrestricted release for the first iteration through the decision process by fully considering Option 1 and/or Option
- 2. Any site-specific information gathered to support Options 1 or 2 can be used in a later iteration analyzing restricted release.
b) The dose modeling for Option 3 should include as much site specific information (gathered as part of Option 1) as necessary to provide a reasonable evaluatbn of future impacts, both with and without inst!iutional controls in effect, to show compliance with restricted release criteria, i.e., the screening parameters are not sufficient to su sport a decision to select restricted use c) The dose assessment under Option 3 should evaluate site specific critical groups as foi!ows: (1) the site specific critical group present as defined by the institutional controls used to restrict use. For example, if a site is restricted to industrial use, the site specific critical group would be a group of workers occupying the building. (2) the site specific critical group possibly affected by transport to exposed indeluals outside the control boundary (and possibly physically off site somewhat removed froin the actuallocation of the contamination, r, groundwater movement of radionuclides offsite). C Draft NUREG-154 22 February 2,1998
(3) the site specific critical group which would be exposed in the event of i failure of institutional controls ana which thus effectively has access to the site as if the site were unrestricted. The site specific critical group in this case would likely be the resident farmer scenario used in generic screening, unless the licensee is able to define and defend an alternate site specific critical group and scena 'J. d) Conduct the regulato~ activities that will need to be completed prior to NRC granting a license termination under 20,1403 Guidance on these aspects are contained in Regulatory Guide 1.xxx. For a set of hypothetical options, Table 5.1 provides an example of how a licensee might
' identify and summarize their options. In making such a tab:3, column 1 would be the expected outcome following the activities in each of the options even though the this would be uncertain at this point in the process. Column 2 of the table would be a translation of the expected outcome, at least qualitatively, into a beneficialimpact on the calculated dose. Column 3 would define the action that would be needed to demonstrate the new state of knowledge or to realize the new state of the site.
In preparing a table like Table 5.1, if there are limits on the time or funds available, then only the options that are below such constraints (if successful), should be considered. If such constraints do not exist then all viable and realistic options should be considered and presented to the NRC. O d Dd Draft NUREG 1549 23 February 2,1998
O
'Q Table 5.1. Example Options Definition Table Expectation Effect on Dose Action Source is believed to be of Simulated dose expected to Collect field data to better lower concentration than decrease as concentrations characterize source currently modeled decrease distribution Experimentally measured kd Simula:ed dose expected to Collect laboratory data to for this site expected to be decrease as availability to reduce uncertainty in the kd higher than default value the receptor is decreased Soil type is expected to '1 ave
. Simulated dose expected to Collect literature and soil higher kds than default value decrease as availability to map data to defend the receptor is decreased alternative soil type / texture Soil permanently removed to Available mass of Remediation by soil removal decrease source contaminare. decreases, concentrations hence simulated dose would decrease Controls are expected to restrictions willlimit access Dispose of waste and remain in place for the to disposal areas on site stabilize on current site and p duration of the compliance while controls are in place; apply for restricted release i l period (if controls fall, simulated dose will decrease simulated doses are between 25 mrem and 100 mrem)
Controls are expected to restrictions willlimit uses for Set land use restrictions and remain in place for the site while controls are in apply for restricted release duration of the compliance place to limit exposure time period (if controls fail, and pathways to individual; simulated doses are between simulated dose will decrease 25 mrem and 100 mrem) , 5.7 Step 9 - Analyze Options in terms of Cost and Likelihood of Success This step involves the analysis of options in terms of cost and the likelihood of success. As noted above in Stop 8, the purpose of this step is to provide information for the licensee so that the evaluation of options considers both the probability that a desired result will be achieved. (i.e., meeting the criteria of Subpart E), and that achieving that result is done in a cost effective manner. Step 9 should be performec in the following manner: a) An a:'alysis of the potential outcome should be performed for each of the options
- identified in Step 8.
Draft NUREG-1549 24 February 2,1998
b) The analysis of outcomes should be no more complex than necessary to support a U}
,- reasonable and cost effective evaluation of the options. The analysis of outcomes could be very simple (e g., the option is remediation and the result is meeting the criteria of Subpart E) to as complicated as further refining and expanding the analysis of Step 4.
c) The cost and time necessary to complete each option should be estimated, along with the likelihood that the option will be successfulin meeting the desired endpoint (meeting the criteria of Subpart E). The analysis should also address the uncertainty associated with each potential outcome, and potential for success and failure at achieving the desired endpoint. i For example, if the licensee chose to spend money to collect some additional informstion on some specific soil properties at their site and spend some money on remediating a small portion of the site, and after this were able to defensibly demonstrate that the dose was below 25 mrem, then their activities would have been successful and the site could be released as unrestricted. Analysis of options would include explicit evaluation of the associated regulatory requirements. This may mean that options need to be executed in a specified order, or that certain options are not allowed until specific conditions are met (e.g., as noted in Section 5.6.2, Option 3 is not permitted unless the cost or risk of Options 1 or 2 is too high (10 CFR 20.1403(a))).
/9 With regard to costs, the licensee should consider that if the option (s) selected are C successful, the license will be released and further costs will be minimized. However, if the selected option (s) are unsuccessful, it may be necessary to perform additional characterization or remediation, or there may need to be an evaluation of restricted use (with its associated costs).
This step should also include ALARA considerations based on the guidance in Regulatory Guide xxx, in terms of cost / benefit calculations as well as qualitative considerations. d) A list should be prepared of the options with their corresponding cost, probability of success (i.e., meeting Subpart E criteria), and other important considerations. An example of such a list is shown in Table 5.2. e) Make a decision regarding the preferred option (Step 10). In some cases, the decision regarding the preferred option will be obvious; for example, a low cost, high probability of success option will generally be selected over a high cost, low probability of success option, assuming the regulatory requirements are equal. However, the preferred option will not a lways be obvious, and additional analysis may be needed for sites attempting to balan:e complex issues. At this point in the decision process, the idea is not to perman sntly eliminate options from further consideration, but rather to select the optimurn approach for the current state of knowledge. O b Draft NURM31549 25 February 2,1998
t'kKw" J m(?5thk S q Note that actual success or failure would not be realized until the sece Steps 4 and 5. s t AMn of The licensee in making a decision regarding the options should consider the following: a) for Option 1 the likelihood of being successfulin collecting the data that is needed to reduce the uncertainty enough to change from an unacceptable dose to an acceptable dose (within specified constraints of time and cost); b) for Option 2, the likelihood that contamination will be reduced to a level thM will result in acceptable dose (within specified constraints of time and cost); or c) for Option 3, the likelihood that a specified restriction will be durable and effective in reducing exposure for the necessary time period (within a specified cost). An example of how the options could be organized is provided in Table 5.2 (for a set of hypothetical alternative actions). The decision as to which option to select may be the joint responsibility of a number of parties, including the licensee, the NRC, and perhaps other stakeholders. The decision process could include other factors in addition to probability of success and cost (e.g., time to complete the activity, environmental justice, etc ). These other influencing factors can be articulated and presented as part of the results of each of the options defined in the options analysis table. Consequently, the result of Step 9 is a logically represented list of options, and the corresponding cost, likelihood of site release, and V other important considerations given that the option is pursued. This analysia will provide information necessary in Step 10. d Draft NUREG-1549 26 February 2,1998
-. __- _ .= . . _ _ - - - - . . _ - - - - - _ - - - - _ _ -
Table 5.2. Example Options Analysis Table (Hypothetical) l l' Alternative Cost Cost Probability of Required Action (if successful) (if unsuccessful) Success Outcome ! Collect field data $$ high dose less than to better 25 mrem characterize source distribution Collect $$ high dose less than laboratory data 25 mrem to reduce uncertainty in Thorium Kd Colhet literature $ medium dose less than data to defend 25 mrem alternative soil type / texture Remediation by
$$$ high dose less than soll removal 25 mrem CN Stabilize or $$
t'j medium dose w/ dispose of controls less waste on site than 25 mrem; and apply for dose w/o restricted controls less release than 100 mrem Set land use $$ low cose w/ restrictions and controls less apply for than 25 mrem; restricted dose w/o release controls less than 100 mrem 5.8 Step 10 Select Preferred Option in Step 10, the decision makers choose the option that will be pursued given the cost, timeliness and likelihood of success, and regulatory requirements of the options identified in Steps 8 and 9, in addition to factors outside the scope of this process. , Draft NUREG 1549 27 February 2,1998
p 6.9 Step 11 Implement Preferred Option Under Step 11 the preferred option is implemented. A licensee should conduct the following activities: a) If a decision is made to use Option 1, then Step 11 is where the data collection would occur. The concentration limits to which the site is cleaned up are based on tha scenarios and consequence analysis simulations conducted in the nrevious steps. To support a future request for license termination, any site survey results, parameter data, or laboratory tests should be carefully documented, b) If a decision is made to use Option 2, then the remedial action is performed and additional data are collected to verify that the remediation reduced the extent and amount of residual contamination to the targeted levels (through a Confirmatory Survey). If the Confirmatory Survey demonstrates that contamination and potential exposure have been reduced to acceptable levels, then the site proceeds to the stage of either restricted or unrestricted release. c) If a decision is made to conduct both Options 1 and 2, remediation would be performed in combination with data collection for the purposes of uncertainty reduction. 5,10 Step 12 Revise Model Assumptions, Parameter Values, and Pathways c
/ Once the preferred option has been implemented, the model assurnptions, parameter values, V) and pathways (as appropriate) are revised. Depending on the results of data collection, the new data can be used to eliminate potential pathways, refute certain model assumptions, to justify new parameter values or refine parameter distributions, or to reduce the estimated extent and amount of residual contamination.
If remediation if performed to portions of the sita or to levels that are less than complete, then new parameter values, refined parameter distributions, and/or new model assumptions should be defined to reduce the estimated extent and amount of residual contamination. 5.12 Reiteration of Sten 4: As appropriate, revised scenarios, pathways, parameters, and source terms are used in a second iteration of the dose assessment. Depending on the application, the licensee can leave the original default model assumptions and pathways unchanged, or in other more complicated situations modify assumptions and pathways or apply different models. l 5413 E93RDtion of Sten 5: l The revised dose assessment is evaluated to determine if the calculated dose rneets the i requirements in 10 CFR 20, Subpart E. The licensee would find either that: (3 Draft NUREG-1549 28 February 2,1998
a) The result in Step 5 is that the calculated dose is less than or equal to the 25 mremlyr g dose criterion of 10 CFR 20.1402. If this is the case, proceed to Step 6 b) The result in Step 5 is that the calculated dose is greater than the 25 mremlyr dose criterion of 10 CFR 20.1402., If this is the case,it means that the contamination at the site is such that the licensee would need to consider additional options to terminate the licensee, in order to terminate the license, the licensee would need incorporate additional amounts of site specific Information into the dose assessment, or possibly consider further remediation or restricting site use. Thus, if this result is found, the licensee should proceed to Step 8 again. The licensee is encouraged to actively work with the NRC during this step to evaluate the appropriateness and adequacy of the analyses before moving on and expending resources on follow on steps. 5.14 Step 6 - ALARA Requirements If the result in Step 5 is that the 25 mremly criterion has been met, the licensee can proceed to
- satisfy ALARA requirements, if not already addressed. ALARA actions at this step can be based on Section 4 of Reg Guide x.xx. The licensee is encourage to actively work with the NRC to discuss, define, and concur on alternative ALARA actions under this step prior to implementing any actions.
5.15 Step 7 - License Termination and Site Release O in this step the licensee would complete final paperwork requirements, including documenting any survey results used to calculate the source term and the results of the dose calculations, and would request that their license be terminated by the NRC. Draft NUREG 1549 29 February 2,1998
p REFERENCES \ k Berger, J.D. (1992), Manual for Conducting Radiological Surveys in Support of License Termination, NUREG/CR 5849, U.S. Nuclear Regulatory Commission, Washington, DC. Daily, M.C., Huffert, A., Cardile, F., and Malaro, J.C. (1994), Working Draft Regulatory Guide on Release Criteria for Decommissioning: NRC Staff's Draft for Comment, NUREG 1500, U.S. Nuclear Regulatory Commission, Washington, DC. Davis, P.A., Bonano, E.J., Wahl, K.K., Price, L.L. (1990), Uncertainties Associated wi'h Performance Assessment of High Level Radioactive Waste Repositories, NUREG/CR-5211, U.S. Nuclear Regulatory Commission Washington, DC. EPA 97 Draft Federal Radiation Protection Guidance for Exposure of the General Public Gogolak, C.V., Huffert, A.M., and Powers, G.E. (1995), A Nonparametric Statistical Methodology for the Design and Analysis of Final Status Decommissioning Surveys, NUREG-1505, U.S. Nuclear Regulatory Commission, Washington, DC. ICRP 1991 *1990 Recommendations of the International Commission of Radiological Protection *, ICRP-60,1991 7.s ICRP 1985 *Radiatinn Protection Principles for the Disposal of Solid Radioactive Waste *, ICRP-h 46, July,1985. (O ICRP 1984,
- Principles of Monitoring for the Radiation Protection of the Populatiori', ICRP-43, May,1984.
lAEA 1995 International Atomic Energy Agency Safety Series No. 57," Technical Bases for Yucca Mountain Standards *, National Research Council,1995 Iman, R.L. and Shortencarier, M.J. (1984), A FORTRAN 77 Program and User's Guide for the Generation of Latin Hypercube and Random Samples for Use in computer Models, NUREG/CR 3624, U.S. Nuclear Regulatory Commission, Washington, DC. Kennedy. W.E., Jr., and Strenge, D.L. (1992), Residual Radioactive Contamination From Decommissioning, Technical Basis for Translating Oontamination Levels to Annual Total Effective Dose Equivalent, NUREG/CR 5512, PNL-7994, U.S. Nuclear Regulatory Commission, Washington, DC. NRC (U.S. Nuclear Regulatory Commission) (1993), Site Decommissioning Management Plan, U.S. Nuclear Regulatory Commission, Washington, DC. NRC (U.S. Nuclear Regulatory Commission)(1995), Site Decommissioning Management Plan, Supplemant 1, U.S. Nuclear Regulatory Commission, Washington, DC. b Draft NUREG 1549 30 February 2,1998
NRC (U.S. Nuclear Regulatory Commission) (1996), Draft Branch Technical Position, Screening Methodology for Assessing Prior Land Burials of Radioactive Waste D authorized Under Former 10 CFR 20.304 and 20.302, Low Level Waste and Decommissioning Projects Branch, Division of Waste Management, NMSS, U.S. Nuclear Regulatory Commission, Washington, DC. NRC (U.S. Nuclear Regulatory Commission) (1997), ' Final Rule on Radiological Criteria for License Termination', need the rest of this reference NRC (U.S. Nuclear Regulatory Commission) (1997), NUREG 1496, ' Generic Environmental impact Statement in Support of Rulemaking on Radiological Criteria for License Termination of NRC Licensed Nuclear Facilities,' Final Report, U.S. Nuclear Regulatory Commission, Washington, DC. NRC (US Nuclear Regulatory Commission) (undated), Chapter XXXX, Draft Decommissioning Procedures for Fuel Cycle and Materials Licensees, and *NMSS Handbook for Decommissioning Fuel Cycle and Materials Licensees *, Silling, S.A. (1983), Final Technical Position on Documentation of Computer Codes for High. l Level Waste Management," NUREG 0856, U.S. Nuclear Regulatory Commission, Washington, DC. Davis, P.A., Price, L.L. Wahl, K.K., Goodrich, M.T., Gallegos, D.P., Bonano, E.J., and Guzowski, R.V., (1990), Components of an Overall Performance Assessment O Methodology, NUREG/CR 5256, U.S. Nuclear Regulatory Commission, Washington, V DC. O e,a NeRee.,e49 s, ,ee,ua,,2.,99. ,
APPENDICES D aft NUREG 1549 32 February 2,1998
Appendix A Default Concentration Values Equivalent To 25 mremly 5 1 l l l . Draft NUREG 1549 33 Febraary 2,1998
Default Concentration Values To Achieve 25 mremly For the Residential Scenarlo
\\
Table A.1 Concentration (pC6/g) equivalent to 26 mrom'y for the specified percentile ci the dose distributlen [Not intended for use et cleanup goals) Source 75th, _ 90th 95th 8ource 75th 90th 95th 3H 177E+r2 108E+02 806E+01 166mHo 557E+00 556E+00 $$6E+00 10Be 169E+01 151E + 03 133E+03 181W 152E+03 151E+03 141E+03 _ 14C 410E +01 116E+01 6 50E+00 185W 1 34E+04 103E+04 4 54E+03 22Na 455E+00 4 25E+00 3 65E+00 187Re 612E + 04 4 20E+04 3 03E+04 35S 3 87E+02 2 70E+02 2 0BE+02 1850s 386E+01 3 85E+01 3 85E+01 36CI $ 61E 01 3 62E 01 2 93E 01 192ir 414 E +01 413E +01 413E +01 40K 913E +00 3 60E+00 169E+00 210Pb 9 $0E 01 8 46E-01 7 90E 01 41Cs 1.10E +02 6 63E+01 515E+01 210Po 946E+00 887E+00 841E+00 - 45Ca 929E+01 5 67E+01 4 2BE+01 226Ra 7 77E-01 6 94E-0. 6 48E-01 46Sc 147E+01 147E+01 147E+01 226Ra+C 603E+00 545E+00 516E+00 54Mn 157E+01 148E+01 139ti+ 01 228Ra 3 85E+00 3 65E+00 3 $4E+00 55Fe 113E+04 103E+04 9 35E+03 227Ac $ 92E 01 5 31E 01 4 85E-01 57Co 151E+02 148E +02 144E+02 227Ac+C 474E+00 4 25E+00 3 89E+00 58Co 3 49E+01 3 47E+01 3 4P+01 228Th 489E+00 4 73E+00 4 61E+00 60Co 3 85E+00 3 79E+00 3 68E+00 228Th+C 3 39E+01 328E+01 3 20E+01 l SDNi 121E+04 554E+03 185E+03 229Th 2 04E+00 185E+00 1.71 E +00 63Ni 443E+03 211E+03 717E+02 229Th+C 163E+01 148E+01 136E+01 65Zn 1.36E +01 108E +01 8 93E+00 230Th 210E +00 183E +00 165E+00 75Se 589E+01 583E+01 5 78E+01 230Th+C 6 44E+00 57BE+00 536E+00 79Se 2 39E+02 2 07E+02 185E+02 232Th 122E +00 113E+00 1.0BE +00 90Sr 2 84E+00 172E+00 122E +00 232Th+C 1.18 E + 01 1.10E+01 104E +01 ( 93Zr 1.3BE +03 108E + 03 6 48E+02 231Pa 3 66E 01 3 27E-01 2 77E 01 93Zr+C 254E+03 188E+03 125E+03 231Pa+C 3 03E+00 2 67E+00 2 36E+00 93mNb 2 02'i+03 181 E +03 149E+03 232U 2 47E+00 196E+00 5 8BE 01 94Nb 581E+00 5 79E+00 $76E+00 2320+C 174E +01 1.46E+01 4 80E+00 93Mo 421E+02 213E +02 149E+02 233U 147E+01 911 E+00 3 70E+00 99Tc 2 92E+01 187E +01 149E+01 233D+C 163E +01 140E+01 9 81E+00 106Ru 5 28E+01 5 06E+01 4 83E+01 234U 2 23E+01 132E +01 3 78E+00 107Pd 907E+03 6 43E+03 4 09E+03 235U 113E+01 8 04E+00 3 35E+00 110mA0 507E+00 4 92E+00 4 78E+00 235U+C 3 58E+00 316E+00 2 75E+00 109Cd 154E+02 106E +02 7 23E+01 236U 2 36E+01 140E+01 3 99E+00 113mCd 8 80F+00 4 95E+00 2 76E+00 238U 2 26E+01 139E+01 3 95E+00 119mSn 3 60E+03 3 09E+03 2 26E+03 238U+C 821E+00 713E+00 544E+00 121mSn 1.37E+03 570E+02 129E+02 237Np 1.77E-01 918E-02 5 81E 02 123Sn 8 74E+02 7.71 E +02 616E +02 237Np+C 184E+00 9 81E 01 5 75E 01 126Sn 472E+00 4 70E+00 4 66E+00 236Pu 9.11 E + 00 817E+00 7 45E+00 126Sn+C 101 E + 01 100E+01 089E+00 - 238Pu 2 81E+00 2 54E+00 2 39E+00 125Sb 2 57E+01 2 56E+01 255E+01 239Pu 2.53E +00 2 28E+00 215E +00 123mTo 186E+02 185E+02 184E +02 240Pu 2 53E+00 2 28E+00 215E+00 127mTe 1.52E +03 143E+03 133E+03 241Pu 8 2BE+01 7.16E+01 4 30E+01 1291 1.70E+00 5 38E-01 2 47E 01 242Pu 2 66E+00 2 41E+00 2 26E+00 134Cs 598E+00 56BE+00 5 36E+00 244Pu 2 42E+00 2 22E+00 2 07E+00 135Cs 2 80E+02 183E+02 1.15E+02 241Am 2 39E+00 2 08E+00 152E+00 137Cs 122E +01 1.10E+01 9 83E+00 243Am 2 30E+00 2.01E+00 149E +00 144Ce 193E+02 184E+02 1.74 E + 02 242C,n 181E+02 164E+02 156E+02 147Pm 9 08E+03 8 20E+03 7 71E+03 243Cm 3 50E+00 3 20E+00 3 03E+00 0 t 6
\.J Draft NUREG 1549 34 February 2,1998
gx Table A 1 Concentration (pC6/g) equivalent to 26 mrom/y for the specified percentile of the dose distribution [Not intended for uso as cleanup goals) Source 76th 90th 96th Source 76th 90th 96th 147Sm 412E + 01 3 62E+01 2 89E*01 244Cm 4 68E+00 417E+00 3 94E+00 151Sm 2 01E+04 176E +04 1 $0E+04 24bCm 163E+00 138E +00 1 18E +00 152Eu 8 68E+00 867E+00 866E+00 246Cm 2 42E+00 2 20E+00 2 09E+00 154Eu 8 02E +00 8 01E+00 800E+00 247Cm 2 33E+00 212E + 00 2 02E+00 155Eu 2 86E+02 2 84E+02 2 82E+02 248Cm 6 $7E-01 6 9BE 01 $ 67E 01 153Gd 3 27E+02 315E+02 2 83E+02 9'.?rf Annr+nn nAAr+nn sacr +nn innis s nyr +ni 3nar+n, 3n2r+ni l C' ( O) e V Draft NUREG-1549 35 February 2,1998
Default Concentration Values To Achieve 26 mromly For Building occupancy Scenario Table A.2 Concentration (dpm/100 cm2) Equivalent to 25 mromly for the Specified Quantile of the Dose Distribution Source I 0.76 0.9 0.96 0.99 3H 1.72e+08 - 1.40e+ 0B 1.26e+08 1.14e+08 10Be 4.30e+04 3 22e+04 2.82e+04 2ABe+04 14C 5.13e+06 4.15e+06 3.75e+06 3.39e+06 22Na 9.58e+03 9.54e+03 9.54e+03 0.54e+03 35S 1.68e+ 07 1.30e+ 07 1.15e+07 1.02e+07 36Cl 6 44e+05 4.91e+05 4.34e+05 3.84e+05 40K 1.05e+05 1.02e+05 1.01e+05 9 92e+04 41Ca 8.12e+06 6.54e+06 5.90e+06 5.33e+0S ASCa 3.78e+06 2.95e+ 06 2.63e+06 2.36e+06 46Sc 3.09e+04 2.87e+04 2.86e+04 2.85e+04 54Mn 3.16e+04 3.16e+04 3.15e+ 04 3.14e+04 55Fe 5.91e+06 4.51e+ 06 3.99e+06 3.54e+06 57Co 2.17e+05 2.10e+05 2.08e+05 2.05e+05 58Co 6.79e+04 6.78e+04 6.76e+04 6.74e+04 60Co 7.27e+03 7.04e+03 6 91e+03 6.78e+03 59Ni 5 49e+06 4.13e+06 3.63e+06 3.21e+06 i 63Ni 2.36e+06 1.77e+06 1.56e+06 1.38e+06 65Zn 4.90e+04 4.04e+04 4.80e+04 4.76e+04 75Se 1.09e+05 1.08e+05 1.08e+05 1.07e+05 79Se 1.13e+06 9.09e+05 8.17e+05 7.37e+05 90Sr 1.13e+04 8.53e+03 7.51e+03 6.65e+03 93Zr 4.76e+04 3.56e+04 3.12e+04 2.75e+04 93Zr+C 4.36e+04 3.26e+ 04 2.86e+04 2.52e+04 93mNb 5.21e+05 3.92e+05 3.44e+05 3.04e+05 94Nb 8.87e+03 6.20e+03 7.86e+03 7.53e+03 93Mo 4 47e+05 147e+05 3.09e+05 2.75e+05 99Tc 1.70e+06 1.30e+06 1.15e+06 1.01e+06 106Ru 3.16e+04 2.55e+04 2.29e+04 2.08e+04 107Pd 1.20e+06 8.96e+05 7.84e+0S 6.91e+05 110 mag 1.03e+04 1.02e+04 1.01e+04 1.00e+04 109Cd 1.43e+05 1.12e+ 05 9.96e+04 8.93e+04 113mCd 9.84e+03 7.44e+03 6.54e+03 5.79e+03 119mSn 1.45e+06 1.28e+06 1.20e+06 1.12e+06 121mSn 9.03e+05 7.29e+05 6.60e+05 5.97e+05 123Sn 8.09e+05 6 44e+05 5.80e+05 5.22e+05 126Sn 8.62e+03 8.45e+03 8.36e+03 8.28e+03 12GSn+C 8 53e+03 8 36e+03 8.28e+03 8.20e+03 Draft NUREG 1549 36 February 2,1998
Og Table A.2 Concentration (dpm/100 cm2) Equivalent to 25 mromly for the Specified Quantile of the Dose Distribution Source 0.78 0.9 0.96 0.99 125Sb 4.49e+ 04 4.43e+ 04 4 41e+04 4.38e+04 123mTe 2.72e+05 2.65e+05 2.61e+05 2.57e+05 127mTe 9.96e+05 8.39e+05 7.72e+05 7.10e+05 1291 4.90e+04 4.13e+04 3.79e+04 3.48e+04 134Cs 1.31e+04 1.29e+04 1.28e+04 1.28e+ 04 135Cs 2.02e+06 1.68e+06 1.53e+06 1.40e+06 137Cs 2.92e+04 2.87e+04 2.83e+04 2.80e+04 144Ce 5.35e+04 4.15e+04 3 68e+04 3.28e+ 04 147Pm 4.40e+05 3 29e+05 2.89e+05 2.55e+05 147Sm 2.05e+02 1.53e+02 1.34e+02 1.18e+02 151Sm 5.11e+05 3.82e+05 3.35e+05 2.95e+05 152Eu 1.34e+ 04 1.26e+04 1.22e+04 4.18e+04 154Eu 1.22e+04 1.14e+04 1.10e+04 1.05e+04 155Eu 1.77e+05 1.53e+05 1.43e+05 1.33e+05 153Gd 2.16e+05 2.00e+05 1.94e+05 1.87e+05 160Tb 5.79e+04 5.73e+04 5.71e+ 04 5.68e+04 166mHo 6.85e+03 0.13e+03 5.79e+03 5.46e+03 181W 1.07e+06 1.07e+06 1.07e+06 1.07e+06 185W 3.05e+07 2.66e+07 2.48e+07 2.31e+07 Q 187Re 1850s 2.63e+08 7.18e+04 2.00e+08 1.76e+08 1.56e+08 7.14e+04 7.12e+04 7.10e+04 192tr 7.53e+04 7.42e+04 7.37e+04 7.31e+04 210Pb 7.27e+02 5.61e+02 4.97e+02 4.42e+02 210Po 3.29e+03 2.50e+03 2.21e+03 1.95e+03 220Ra 1.41e+03 1.11e+03 9.88e+02 8.83e+02 226Ra+C 4.13e+02 3.19e+02 2.83e+02 2.52e+02 228Ra 2.57e+02 1.94e+02 1.70e+02 1.50e+02 227Ac 2.31e+00 1.74e+00 1.52e+00 1.34e+00 227Ac+C 2.31e+00 1.74e+00 1.52e+00 1.34e+00 228Th 5.29e+01 3.95e+01 3.46e+01 3.05e+01 228Th+C 5.29e+01 3.95e+01 3 46e+01 3.05e+01 229Th 7.10e+00 5.30e+00 4.64e+00 4.08e+00 229Th+C 7.08e+00 5.30e+00 4.64e+00 4.08e+00 230Th 4.71e+01 3.52e+01 3.08e+01 2.71e+01 230Th+C 4.22e+01 3.16e+01 2.78e+01 2.45e+01 232Th 9.36e+00 6.98e+00 6.11e+00 5.39e+00 232Th+C 7.69e+00 5.75e+00 5,04e+00 4.44e+00 231Pa 1.10e+01 8.22e+00 7.20e+00 6.35e+00 231Pa+C 1.91e+00 1.43e+00 1.26e+00 1.10e+00 O U Draft NUREG-1549 37 February 2,1998
1 1 Table A 2 Concentration (dpm/100 cm2) Equivalent to 26 mremly l for the Specified Quantile of the Dose Distribution Source 0.75 0.9 0.96 l 0.99 232U 2.16e+01 1.61e+01 1.41e+01 1.24e+ 01 , 232V+C 1.52e+01 1.14e+01 9.96e+00 8.77e+00 g 233U 1.13e+02 8,45e+01 7.42e+ 01 0.53e+01 233U+C 6.38e+00 4.76e+00 4.17e+00 3.68e+00 l 234U 1.16e+02 8 65e+01 7.58e+01 6.67e+01 , 235U 1.25e+02 9.33e+01 8.17e+01 7.18e+01 235U+C 1.88e+00 1.41e+00 124e+00 1.09e+00 ' 236U 1.23e+02 9.12e+01 8 01e+01 - 7.04e+01 , 238U 1.30e+02 9.65e+ 01 8.47e+01 7.46e+01 238U+C 2.50e+01 1.87e+ 01 1.64e+01 1.45e+01 237Np 2.83a+01 2.12e+01 1.85e+01 1.63e+01 237Np+C 5.06e+00 3.78e+00 3 31e+00 2.91e+00 236Pu 1.16e+02 8.68e+01 7.62e+01 6.72e+01 ~ 238Pu 3 92e+01 2.93e+01 2.57e+01 2.25e+01 239Pu 3.57e+01 2.66e+01 2.34e+ 01 2.05e+01 240Pu 3.57e+01 2.66e+01 2.34e+01 2.05e+01 241Pu 1.82e+03 1.36e+03 1.19e+03 1.05e+03 242Pu 3.73e+01 2.78e+01 2.45e+01 2.16e+01 244Pu 3.79e+01 2.83e+01 2.48e+01 2.19e+01 ( 241Am 3 45e+01 2.58e+01 2.25e+01 1.98e+01 242mAm 3.54e+01 2.65e+01 2.31e+01 2.05e+01 [ 243Am 3 47e+01 2.60e+01 2.27e+01 2.00e+01 242Cm 1.57e+03 1.17e+03 1.03e+03 9.06e+02 243Cm 5.04e+01 3.77e+01 - 3.30e+01 2.91e+01 244Cm 6.30e+01 4.70e+01 4.12e+01 3.63e+01 ! 245Cm 3.36e+01 2.51e+01 2.19e+01 1.94e+01 246Cm 3 39e+01 2.53e+01 2.21e+01 1.95e+01 247Cm 3.69e+01 2.76e+01 2.43e+01 2.14e+01 248Cm 9.26e+ 00 6 91e+00 6.07e+00 5 34e+00 252Cf 1.11e+02 8.28e+01 7.27e+01 6.39e+0 ! Draft NUREG-1549 38 February 2,1998
Appendix B Annual Total Effective Dose Equivalent Factors l l O O- Draft NUREG 1549 39 February 2,1998
l I i (7) Table 81 - Selected percentiles for the TEDE distnbutions for I the residential scenario (mremly per pCl/g) Source 76th 90th 96th l Dose @ 69thi Dose @ 60th 3H 1.41 E-01 2.32E-01 3.10E 01 4.15 10Be 1.48E 02 1.65E-02 1.88E-02 13.17 14C 6.10E-01 2.15E+ 0G 3.85E+00 33.61 22Na 5.49E+00 5.88E+00 6.85E+00 2.98 35S 6.46E 02 9.26E 02 1.20E-01 4.56 36Cl 4.46E+01 6.91 E+01 8.54E+01 4.87 40K 2.74E+00 6.94E+00 1.48E+01 19.24 41Ca 2.28E-01 3.77E 01 4.86E-01 6.51 45Ca 2.69E-01 4.41 E-01 5.84E-01 6.88 46Sc 1.70E4 00 1.70E+00 1.70E+00 1.01 54Mn 1.60E+00 1.69E+00 1.79E+00 1.37 55Fe 2.21 E-03 2.43E 03 2.67E-03 10.10 57Co 1.66E 01 1.69E-01 1.73E-01 1.25 58Co 7.17E 01 7.20E-01 7.24E-01 1.05 60Co 6.49E+00 6.60E+00 6.79E+00 1.24 59Ni 2.07E 03 4.51 E-03 1.35E-02 39.83 63Ni 5.65E-03 1.19E 02 3.49E-02 39.30 65Zn 1.84E+00 2.32E+00 2.80E+00 3.38 O 75Se 4.24E 01 4.29E 01 4.32E 01 1.05 79So 1.05E-01 1.21E 01 1.35E-01 1.92 90Sr 8.80E+00 1.46E+01 2.05E+01 8.42 93Zr 1.82E-02 2.32E 02 3.86E-02 ~ 13.38 93Zr+C 9.84E-03 1.33E-02 2.01E-02 12.60 93mNb 1.24E 02 1.38E 02 1.67E 02 7.68 94Nb 4.30E+00 4.32E+00 4.34E+00 1.03 93Mo 5.94E 02 1.17E-01 1.67E-01 11.03 99Tc 8.57E-01 1.34E+00 1.68E+00 5.63 106Ru 4.73E-01 4.94E-01 5.18E-01 1.29 107Pd 2.76E-03 3.89E-03 6.11E-03 12.35 110 mag 4.93E+00 5.08E+00 5.23E*00 1.20 109Cd 1.63E-01 2.35E-01 3.46E-01 10.77 113mCd 2.84E+00 5.05E+00 9.07E+00 19.52 119mSn 6.95E-03 8.10E-03 1.10E-02 14.69 121mSn 1.83E-02 4.39E-02 1.94E-01 61.15 123Sn 2.86E-02 3.24E-02 4.06E-02 5.76 126Sn 5.30E+00 5.32E+00 5.36E+00 2.13 p., Draft NUREG-1549 40 February 2,1998
.~ - - . -. - . - - - .. - . . /]/
Table B1 Selected percentiles for the TEDE distnbutions for the residential scenario (mremly per pCl/g) source 75t.: Doth 95th Dose G 99thi
~ Dose e 50th d
126Sn+C 2.48E+00 2.49E+ 00 2.53E+00 2.13 125Sb 9.71 E-01 9.76E 01 9.82E 01 1.16
] 123mTe 1.34E 01 1.35E 01 1.36E 01 1.22 127mTe 1.64E 02 1.75E-02 1.88E 02 3.54 i 1291 1.47E+01 4.65E+01 1.01 E+02 49.83 ' j 134Cs 4.18E+00 4.40E+00 4.66E+00 1.92 135Cs 8.94E 02 1.36E 01 2.18E-01 29.74 137Cs . 06E400 2.27E+00 2.54E+00 5.67 144Ce, 1.29E-01 1.36E-01 1.44E 01 1.36 l 147Pm 2.75E 03 3.05E 03 3.24E-03 1.92 i 147Sm 6.07E-01 6.91E 01 8.66E 01 6.61 151Sm 1.24E 03 1.42E 03 1.67E-03 6.26 152Eu 2.88E+00 2.88E+00 2.89E+00 1.01 154Eu 3.12E+ 00 3.12E+00 3.12E+00 1.01 155Eu 8.75E-02 8.80E-02 5.86E 02 1.07 ;
153Gd 7.66E 02 7.93E 02 8.83E 02 1.66 160Tb 8.29E 01 8.29E 01 8.29E 01 1.02 166mHo 4.49E+00 4.49E+00 4.50E+00 1.01 181W 1.64E 02 1.66E 02 1.77E 02 1.95 185W 1.87E-03 2.43E 03 5.51 E-03 27.86 187Re 4.09E 04 5.95E 04 8.25E-04 7.15 1850s 6.48E 01 6.49E-01 6.50E-01 1.03 192lr 6.04E 01 6.05E-01 6.05E 01 1.01 210Pb 2.63E+01 2.95E+01 3.17E+01 5.37
' 210Po 2.64E+00 2.82E+00 2.97E+00 1.69 226Ra 3.22E+01 3.60E+01 3.86E+01 5.60 226Ra+C 4.15E+00 4.58E+00 4.85E+00 5.24 228Ra 6.49E+00 6.84E+00 7.05E+00 1.24 227Ac 4.22E+01 4.70E+01 5.16E+01 ,
8.85 227Ac+C 5.28E+00 5.88E+00 6.43E+00 8.83 228Th 5.12E+00 5.29E+00 5.43E+00 1.15 228Th+C 7.37E-01 7.62E 01 7.81E-01 1.15 229Th 1.22E+01 1.35E+01 1.46E+01 5.99 229Th+C 1.53E+00 1.69E+00 1.83E+00 5.98 230Th 1.19E+01 1.36E+01 1.51E+01 9.18 230Th+C 3.88E+00 4.33E+00 4.67E+00 6.32 O Draft NUREG-1549 41 February 2,1998
O Table B1 Selected percentiles for the TEDE distnbutions for the residential scenario (mrem /y per pCl/g) source 75th 90th 95th Do.e 9 99thi oo.. e soth 232Th 2.05E+01 2.21 E+01 2.32E+01 3.0'i 232Th+C 2.12E+00 2.27E+00 2.40E+00 3.15 231Pa 6.82E+01 I 7.66E+01 9.01 E+01 7.49 231Pa+ C 8.24E+00 9.38E+00 1.06E $01 7.36 232U 1.01 E+01 1.28E+01 4.25E+01 17.79 232U+C 1.43E+00 1.72E+00 5.21E+00 15.34 233U 1.71 E+00 2.74E+00 6.76E+00 21.20 233U+C 1.53E+00 1.79E+00 2.55E400 6.90 234U 1.12E+00 1.89E+00 6.62E+00 22.71 235U 2.22E+00 3.11 E+00 7.47E+00 20.57 235U+C 6.99E+00 7.91 E+00 9.09E+00 7.19 l 236U 1.06E+00 1.79E+00 6.27E+00 22.81 238U 1.11E+00 1.80E+00 6.33E+00 21.80 238U+C 3.04E+00 3.51 E+00 4.59E+00 7.03 237Np 1.41 E+02 2.72E+02 4.30E+02 11.45 237Np+C 1.36E+01 2.55E+01 4.35E+01 10.12 236Pu 2.74E+00 3.06E+00 3.35E+00 2.87 238Pu 8.88E+00 9.83E+00 1.05E+01 1.93
\ 239Pu 9.88E+00 1.09E+01 1.17E+01 2.47 240Pu 9.88E+00 1.09E+01 1.17E+01 2.46 241Pu 3.02E 01 3.49E-01 5.81E 01 11.58 242Pu 9.38E+00 1.04E+01 1.11 E+01 2.47 244Pu 1.03E+01 1.13E+01 1.21E+01 2.23 241Am 1.05E+01 1.20E+01 1.65E+01 10.28 243Am 1.09E+01 1.24E+01 1.68E+01 9.96 242Cm 1.38E-01 1.53E-01 1.61 E-01 1.43 243Cm 7.15E+00 7.82E+00 8.26E+00 1.41 244Cm 5.46E+00 6.00E+00 6.34E+00 1.42 245Cm 1.53E+01 1.81 E+01 2.12E+01 3.93 246Cm 1.03E+01 1.14E+01 1.20E+01 1.42 247Cm 1.07E+01 1.18E+01 1.24E+01 1.35 248Cm 3.80E+01 4.18E+01 4.41 E+01 1.42 252Cf 312E+00 3_ B4E +00 4 42E+00 12 34 O_ Draft NUREG 1549 42 February 2,1998
Table 52 . Selected percentiles for the TEDE distributions for the Building Occupancy scenario (dpm/100 cm9 Source 0.78 0.9 0.96 0.99 3H 1.45e-07 1.79e-07 1.98e-07 2.19e-07 10Be 5 81e-04 7.77e-04 8.86e 04 1.01e 03 14C 4.87e 06 6.02e 06 6.66e-06 7.37e-06 22Na 2.61e-03 2.62e-03 2.62e-03 2.62e 03 35S 1.49e-06 1.93e-06 2.18e-06 2 46e-06 36Cl 3 88e-05 5.09e-05 5.76e-05 6.51e-05 40K 2.38e-04 2.45e-04 2.48e-04 2.52e-04 41Ca 3.08e-06 3.82e 06 4 24e-06 4.69e-06 45Ca 6.62e 03 8.48e-06 9 51e-06 1.06e-05 46 Ic 8.66e-04 8.71e 04 8.74e-04 8.77e-04 54Mn 7.90e-04 7.92e 04 7.93e-04 7.95e-04 55Fe 4.23e-06 5 54e-06 6.27e-06 7.07e 06 ) 57Co 1.15e-04 1.19e 04 1.20e 04 1.22e 04 58Co 3 68e-04 3.69e-04 3.70e-04 3.71e 04 60Co 3.44e. 03 3.55e-03 3.62e-03 3,69e-03 59Ni 4.55e-06 6.05e-06 6.88e-06 7.79e-06 63Ni 1.06e-05 1.41e 05 1.60e-05 1.81e-05 65Zn 5.10e-04 5.17e-04 5.21e-04 5.25e-04 75Se 2.29e-04 2.31e 04 2.32e 04 2.33e 04 79Se 2.21e-05 2.75e-05 3.06e 05 3.39e-05 90Sr 2.22e-03 2.93e-03 3 33e-03 3.76e-03 93Zr 5.25e 04 7.02e-04 8.01e 04 9.10e-04 932r+C 5.73e-04 7.66e-04 8.74e-04 9.92e-04 93mNb 4.80e-05 6.38e-05 7.26e-05 8.22e-05 94Nb 2.82e 03 3.05e-03 3.18e 03 3.32e-03 93Mo 5.59e-05 7.20e 05 8.10e-05 9.08e 05 99Tc 1.47e-05 1.93e-05 2.18e-05 2.47e-05 106Ru 7.91e-04 9.81e-04 1.09e-03 1.20e-03 107Pd 2.09e-05 2.79e 05 3.19e-05 3.62e-05 110 mag 2.43e-03 2.45e 03 2.47e-03 2.49e-03 109Cd 1.75e-04 2.24e-04 2.51e-04 2.80e-04 113mCd 2.54e-03 3.36e-03 3.82e-03 4.32e-03 119msn 1.73e-05 1.96e-05 2.09e 05 2.23e-05 121mSn 2.77r 05 3.43e-05 3.79e-05 4.19e 05 123Sn 3.09>05 3.88e-05 4.31e-05 4.79e-05 126Sn 2.90e-03 2.96e-03 2.99e-03 3.02e-03 126Sn+C- 2.93e 03 2.99e-03 3.02e-03 3.05e-03 125Sb 5 57e-04 5 64e-04 5 67e-04 5 71e-04 Draft NUREG 1549 43 February 2,1998
Table 82 . Selected percontiles for the TEDE distributions for the Building Occupancy scenario (dpm/100 cm') Source 0.75 0.9 0.96 0.99 123mTe 9.20e-05 9 45e-05 9.58e-05 9.73e-05 127mTe 2 51e-05 2.98e 05 3.24e 05 3.52e 05 1291 5.10e-04 6.06e-04 6.60e-04 7.18e-04 134Cs 1.91e-03 1.94e-03 1.95e-03 1.96e-03 135Cs 1.24e-05 1.49e-05 1.63e-05 1.78e-05 137Cs 8.55e-04 8.72e 04 8.82e 04 8.93e-04 144Ce 4.67e-04 6.03e-04 6.79e-04 7.63e-04 l 147Pm 5.68e-05 7.59e-05 8.64e 05 9.81e-05 147Sm 1.22e-01 1.63e-01 1.86e-01 2.11e-01 151Sm 4.89e-05 6.54e-05 7.46e-05 8.47e-05 152Eu 1.86e-03 1.98e-03 2.05e-03 2.12e-03 r 154Eu 2.05e-03 2.20e-03 2.28e-03 2.38e-03 155Eu 141e-04 1.63e-04 1.75e-04 1.88e-04 153Gd 1.16e-04 1.25e-04 1.29e-04 1.34e-04 160Tb 4 32e 04 4.36e-04 4.38e-04 4.40e-04 166mHo 3 65e-03 4.08e-03 4.32e-03 4.58e-03 181W 2.33e-05 2.33e 05 2.33e-05 2.33e-05 185W 8.20e-07 9.39e 07 1.01e-06 1.08e-06 187Re 9.52e-08 1.25e-07 1.42e 07 1.60e-07 1850s 3.48e-04 3.50e-04 3.51e-04 3 52e-04 192lr 3.32e-04 3.37e-04 3.39e-04 3 42e-04 210Pb 3 44e-02 4.46e-02 5.03e-02 5 65e-02 310Po 7.61e-03 9.99e-03 1.13e-02 1.28e-02 226Ra 1.77e-02 2.26e-02 2.53e-02 2.83e-02 226Ra+C 6 06e-02 7.84e-02 8.82e-02 9 91e-02 228Ra 9.71e 02 1.29e-01 1.47e 01 1.67e-01 227Ac 1.08e+01 1.44e+01 1.65e+01 1.87e+01 227Ac+C 1.08e+01 1.44e+01 1.65e+01 1.87e+01 228Th 4.73e-01 6.33e-01 7.22e-01 8.20e-01 228Th+C 4.73e-01 6.33e-01 7.22e-01 8.20e-01 229Th 3.52e+00 4.72e+00 5.39e+00 6.12e+00 229Th+C 3.53e+00 4.72e+00 5.39e+00 6.12e+00 230Th 5 31e-01 7.11e-01 8.11e-01 9.21e-01 230Th+C 5.92e-01 7.90e-01 9.00e-01 1.02e+00 232Th 2.67e+00 3.58e+00 4.09e+00 4.64e+00 232Th+C 3.25e+00 4.35e+00 4.96e+00 5.63e+00 231Pa 2.27e+00 3.04e+00 3.47e+00 3 94e+00 231Pa+C 1.31e+01 1.75e+01 1.99e+01 2 27e+01 Draft NUREG 1549 44 February 2,1998 m !
)
Table 52 Selected percontiles for the TEDE distributions for the i O.. Building Occuparcy scenario (dpm/100 cm9 Source 0.75 0.9 0.96 0.99 232U 1.16e+00 1.55e+00 1.77e+00 2.01e+ 00 232V+C 1.64e+00 2.20e+ 00 2.51e+00 2.85e+00 l 233U 2.21e-01 2.96e-01 3.37e-01 3.f.3e-01 233U+C 3 92e+00 5.25e+00 599e+00 6.80e+00
. 234U 2.16e 01 2.89e-01 3.30e 01 3.75e-01 I
235U 2.00e-01 2.68e-01 3.06e 01 3 48e 01 235U+C 1.33e+01 1.77e+01 2.02e+01 2.30e+ 01 236U 2.04e-01 2.74e-01 3.12e 01 3.55e-01 4 238U 1.93e 01 2.59e-01 2.95e 01 3.35e-01 238U+C 1.00e+00 1.34e+00 1.52e+ 00 1.73e+00 237Np 8.83e 01 1.18e+00 1.35e+00 1.53e+00 237Np+C 4.94e+00 6 62e+00 7.55e+00 8.58e+00 j 236Pu 2.15e 01 2.88e-01 3.28e 01 3.72e-01 i 238Pu 6.38e-01 8.54e 01 9.74e 01 1.11e+ 00 239Pu 7.01e 01 9.39e-01 1.07e+00 1.22e+00 240Pu 7.01e-01 9.39e-01 1.07e+00 1.22e4 00 3 241Pu 1.37e-02 1.84e 02 2.10e 02 2.38e-02 O 242Pu 6.71e 01 ' 98e 01 1.02e+00 1.16e+00
, 244Pu 6.59e-01 u.82e-01 1.01e+00 1.14e+00 241Am 7.25e-01 9.70e-01 1.11e+00 1.26e+00 . 242mAm 7.06e 01 9 45e-01 1.08e+00 1.22e+00
. 243Am 7.20e-01 9 63e-01 1.10e+00 1.25e+00 d 242Cm 1.59e-02 2.13e-02 2.43e-02 2.76e-02 243Cm 4.96e-01 6.64e-01 7.57e-01 8.59e-01 ! 244Cm 3 97e-01 5.32e 01 6.07e-01 6 89e-01 245Cm 7.44e-01 9.96e-01 1.ide+00 1.29e+00 246Cm 7.37e-01 9.87e-01 1.13e+00 1.28e+00 j 247Cm 6.78e-01 9.07e-01 1.03e+00 1.17e+00 248Cm 2.70e+00 3.62e+00 4.12e+00 4.68e+00 l~ 252Cf 2.25e-01 3.02e 01 3.44e-01 3 91e-01 L 4 Draft NUREG 1549 45 February 2,1998 i t
, ,m. . . , ,- , - - , , - , , .---e. , - - - -- - - , ,,.,rw ... __ ,,, ,.vr y ..w, -o -m . - ~ . - _- -
(~N Appendix C Scenarlos, Pathways, and Critical Groups
\
This appendix provides information br defining the scenarios, pathways, and critical groups that are important for the site dose assessment. This allows for identification of a) potential human activities on or near the site which can result in exposure (scenarios) b) migration and exposure pathways of the radionuclides (pathways) c) critical receptors (the critical group), Scenarios are defined as plausible sets of human activities and of future uses of the site. As such, scenarios provide a description of the plausible future land uses, human activities and behavior of the natural system. With an understanding of the potential human activities and the physical system, one can then develop conceptual models of the site (See main text, Figure 1, Step 3, and Appendix D). Those conceptual models are translated into mathematical models and implemented in (and solved by) corresponding analytical or numerical models and computer codes. The objective is to calculate a dose (main text, Figure 1, Step 4) which is then compared with dose criteria (main text, Figure 1 Step 5) to assess whether the site complies with requirements. The definition of scenarios and identification of pathways and the dose assessment based on p that definition, can be generic or site specific, A licensee should use a critical group that is j( appropriate for its site. Licensees might: (a) if they have very simple situations, use screening scenarios, critical groups, and pathway parameters developed in NUREG 5512 and described in this NUREG, (b) if they have relatively simple situations, use the default screening scenarios but develop more site specific parameters and/or pathway analyses, or I (c) if they have more complex situations, define and use site specific scenarios and site specific critical groups for use with site specific pathway analysis and parameters. If the generic approach of 'a' is used, a licensee would not need to provide justification for use of the anoroach. For either situations *b" or "c" t.t,0ve, the licensee would need to provide site specific data to defend the use of alternative scenarios, critical group definitions, and more complex pathway analysis, models, and parameter revision, Section C.1 describes the rationale for and licensee actions in using the generic approach. Section C.2 describes the method the licensee would use in developing site specific scenarios, critical groups, end pathways. Section C.3 provides background information regarding the critical group, including its regulatory basis and the concept of use of the term critical group by standards setting bodies like ICRP Description of methods for changing parameters is contained in Appenoix E. Draft NUREG-1549 46 February 2,1998 1 m I
C,1 Generic Scenarios, Critical Groups, and Pathways Scenario descriptions acceptable to NRC for use in generic screeninq are developed and contained in NUREG/CR 5512 [ Kennedy and Strenge,1992). NUREG/CR 5512 provides the rationale for applicability of the generic scenarios, critical groups, and pathways at a site, :.ie rationale and assumptions for teenarios and pathways included (and excluded), the conceptual modeling approaches, the default parameter values and bases for revising parameters and pathways based on site specific information. There are two critical groups used for screening (referred to here as
- screening groups" based on the default scenarios of NUREGICR 5512:
- 1) Buildina occuoant for reuse of structures. This scenario accounts for exposure to fixed and removable thin layer or surface contamination sources. The building occupant is dafined as a person who workc in a commercial building following license terminatiori.
The pathways that apply to the building occupant include: a) extemal exposure to penetrating radiation from surface sources, b) inhalation of resuspended curface contamination, c) inadvertent ingestion of surface contamination. The models tha' en be used to mathematically represent these pathways are described in Appendix D. .. a parameters that are used to describe these pathways are presented in Appendix E. It is possible to modify the parameters for the building occupant based ( on Appendices E.
- 2) Resident farmer for contaminated soil sitet This scenario accounts for potential exposure to residual radioactive contamination in soll. For this scenario, the soil contamination is assumed to be contained in a surface-layer. The resident farmer is defined as a person who lives on the site following license termination, grows some portion of their diet on the site, and drinks water from an on site well. The pathways that apply to the resident farmer include:
a) external exposure to penetrating radiation from volume soil sources while outdoors b) external exposure to penetrating radiation from volume soil sources while indoors c) inhalation exposure to resuspended soll while outdoors d) inhalation exposure to resuspended soil while indoors e) inhalation exposure to resuspended surface sources of soil tracked indoors f) direct ingestion of soil
/3 p) v Draft NUREG-1549 47 February 2,1998
g) inadvertent ingestion of soil tracked indoors s' h) ingestion of drinking water from a groundwater source
- 1) ingestion of plant products grown in contaminated soll j) ingestion of plant product; irrigated with contaminated groundwater k) Ingestion of animal products grown onsite (i.e., after animals ingest contaminated drinking water, plant products, and soll) l} ingestion of fish from a contaminated surface water source The models that can be used to mathematicany represent these pathways are described in Appendix D. The parameters that are used to describe these pathways are presented in Appendix E. It is possible to modify the parameters for the resident based on Appendix E.
Thus, the licensee can use the screening groups listed above which NRC has developed in NUREG/CR 5512. When using the NUREG/CR 5512 scenarios, screening groups, or pathways, licensees generally do not need to provide justification or documenwtion for selecting and using the scenarios and pathways included. A licensee has the option of using the screening group with default parameters or modifying the parameters used for the screening group based an site specific information and providing justification for the parameter O
, revision (as described in Appendix E),
C.2 Site Specific Scenarios, Critical Groups, and Pathways A licensee may develop site specific scenarios, critical groups, and pathways based on site-specific information. This information could descr;5e a critical group, referred to here as a
' site-specific critical group," which is different from the screening group. Use of a site specific critical group would occur in cases where, for example:
a) the licensee could justify that the site was such that major pathways (e.g., the agricultural pathway) of the screening group could be eliminated, either because of physical reasons or site use ieasons, b) there was a specific sensitive group on the site, c) restricted use was proposed for a site
- Modifying scenarios and developing site specific critical groups requires information regarding plausible uses of the oite and demographic information. Such information might include considerations of the prevailing (and future) uses c,f the land and site specific issues such as historical and planned future land use, and physical characteristics that constrain site use. It will often be necessary to evaluate several potential critical groups, based on different combinations of site-specific sce.narios developed from expected pathways ano demographics, (U
Draft NUREG 1549 48 February 2,1998
+
to determine the group receiving the highest exposure. It is especially important to evaluate the C' / homogeneity of specific groups to determine if what appears to be one group is actually multiple groups. The following guidelines for defining the site specific critical group and the homogeneity of the critical group should be followed. a) If the distribution of dose equivalents for the workers ranges over more than a factor of ten, then they should be treateu as two or more different groups. b) If the dose evaluation is being done to determine the site-specific cleanup level, then the distribution of dose equivalents should not range over 'nore than a factor of three, since the fraction of the source upper bound in that case has been defined to be one. For restricted release, similar considerations apply. However, now the nature of the critical group changes due to site restrictions and institutional controls which can restrict certain kinds of activities or land or water uses. The detailed definition of the scenarios coriAidered for restricted release need to include the impact of the control provisions on the location and behavior of the average member of the appropriate critical group. In developing site specific scenarios, critical groups, and pathways, a licensee's analysis should encompass the following: a) An evaluation oi whether the generic scenarios of NUREG/CR 5312 applicable to its site and, if not, for each scenario, whether major exposure pathways can be modified or eliminated from further consideration based on site-specific conditions (NUREG/CR-5512 notes that pathways can be added or eliminated, as appropriate, using site-s specific data and that possibly different, scenarios and associated pathways may be necessary for complex site specific analyses beyond those in NUREG/CR 5512). This evaluation should include adequate justification, based on site specific data, for eliminating scenarios and/or pathways from the analysis. As examples, for a site in a predominantly urban or industrial location or for a site in a particularly rocky environment, a licensee may want to defend not using the screening group in favor of a scenarios more representative of prevailing (and future) t.ses of the land. The licensee in this cases might indicate that the historical and planned future land use or the physical characteristics of the site were such as to preclude the generic resident farmer scenario of Appendix C.1. Such a demonstration would be enhanced in cases where the radionuclides at the site were relatively short lived and the time period over which such u situation might need to last were therefore also relatively shcrt. This approach could be appropriate for the situations noted here based on their characteristics (and therefore be an unrestricted use of the site), and would not require the licensee establish institutional controls to restrict site use under 10 CFR 20.1403. Similarly, a licensee should consider other aspects of the site and critical groups that might be exposed including factors related to plumbing systems, floor drains, and embedded piping, ventilation ducts, building external surfaces, and embedded contamination in surfaces. 4
.O b Draft NUREG 1549 49 February 2,1998
h Tabic C.1 provides a possible set of scenarios that licensees may consider for use in
; site specific dose assessments.
- Table C.1 Potential Scenarios for use in Dose Assessments These scenarios are applicable for unrestricted releasc of the site and for analyzing restricted release sites assuming institutional controls fail. The NUREG/CR 5512 scenarios may be based on the screening group, but the scenario definition and pathways may be changed due to site specific considerations (e.g. no drinking water, no pond, etc.). Seme of these scenarios are also appropriate for restricted release of the site. In addition, they may be considered for unrestricted sites for which geography or realistic future uses of the site would preclude certain usas (such as agriculture).
Building occupancy (Generic screening - NUREG/CR-5512 based). Residential farmer (Generic screening - NUREG/CR 5512 based). Urban construction (contaminated soil, no suburban or agricultural uses).This scenario is meant for small urban sites cleared of all original buildings; only contaminated land and/or buried waste remains. Residential (a more restricted subset of the residential farmer scenario, for those urban or suburban sites where farming is not a realistic projected future use of the
- site).
Recreational (where the site is preserved for recreational uses only). Hybrid Building occupancy (adds contaminatM soil, building may or may not be contaminated). Drinking water (no on-site use of groundwater; off-site impacts of contaminated
\ plume).
b) An analysis of exposure pathways. For this analysis, the licensee should begin with at least the pathways prescribed by NUREG/CR-5512 (and as listed for the building occupant and resident farmer scenarios in Aonendix C.1 of this NUREG). After-considering those pathways, the licensee may then wish to conduct a more thorough pathway analysis. The objective of this approach (i.e., proceeding from generic to more site specific pathways) is to focus resources on the pathways, and models associated - with those pathways, that have the highest likelihood of significant exposures to the critical group. Applying this pathway analysis process results in a set of the dominant pathways for the site-specific scenarios (see Table C.1) that could be further pared down using site-specific conditions and screening criteria. Licensees will need to document their pathway analyses and provide justification for the elimination of pathways from dose assessments. A brief summary of the NRC-recommended pathway analysis process is as follows:
- 1) Compile a list of exposure pathways applicable to any type of contaminated site (this list is developed in NUREG/CR-5512 and summarized in Appendix C.1 of this NUREG) b' Draft NUREG-1549 50 February 2,1998
p 2) Categonze the genera l types of contamination at the site (e.g. sediment or soil, I deposits in buildings and equ;pment, ',ar. ace contamination, surface waters, groundwater, industrial products such as slag).
- 3) Screen out insignificant pathways for each contaminant type.
- 4) Identify the physical processes pertinent to the pathways for the site.
- 5) Separate the list of exposure pathways into unique pairs of exposure media (e.g.
source to groundwater, groundwater to surface water, etc.). Determine the physical processes that are relevant for each exposure media pair and combine the processes with the pathway links.
- 6) Reassemble exposure pathways for each source type, using the exposure media pairs as building blocks, thus associating all the physical processes identified
- with the individual pairs with the complete pathway.
C.3 Background information Related to " Critical Group" This section provides background information on the critical group which a licensee can use in understanding the terms ' critical group,"
- screening group', and ' site specific critical group."
C.3.1 The requirements in Subpart E for Critical Groups ( The dose calculated from residual radioactivity at a decommissioned site is dependent upon d how the receptor and the physical characteristics of the site are defined. With regard to the receptor, Subpart E contains the following specific requirements: i
- 1) 20.1402 states that the crite..on for unrestricted release is 25 mrem /y to the average member of the critical group:
- 2) 20.1403, in setting criteria for restricted release, addresses two separate critical groups and hence a licensee would have to evaluate two separate critical groups for restricted use as follows:
a) 20.1403(b) states that the criterion for restricted release is 25 mrem /y to the average member of the critical group with institutional controls in place (per 20.1403(b), because site restrictions limiting or eliminating certain kinds of i activities are highly site specific, the nature of the critical group is also highly site-specific (see Section C.2) b) 20.1403(e) states that , if the institutional controls are no longer in effect, the criterion is that the dose to the average member of the critical group is less tnan either 100 mrem (1 mSv) per year or 500 mrem (5 mSv) per year : A second critical group would have to evaluated based on consideration of the restrictions failing and essentially unrestricted use occurr;ng. The considerations as to the ! critical group for this situation would be similar as those noted above for 20.1402. I fh Draft NUREG-1549 51 February 2,1998
A The terms " critical group" and " average member" are defined and discussed in the regulations in the following way; a) The critical group for decommissioning is defined in 10 CFR 20.1003 as "the group of individuals reasonably expected to receive the greatest exposure to residual radioactivity for any applicable set of circumtaances." NUREG/CR 5512, Volume 1, similarly describes the Critical Group as an individual or relati,ely homogeneous group of individuals expected to receive the highest exposure within the assumptions of the particular scenario, b) The average member of the Critical Group is an individual who in turn is assumed to represent the most likely exposure situation based on prudently conservative exposure assumptions and parameter values within the model calculations. C.3.2 Backaround information on Critical Groaos The definition and evaluation of a critical group is a site-specific and often complex process that has been discussed in NRC documents as well as documents produced by national and international organizations. The practice of defining and using a Critical Group when assessing individual public dose from low levels of radioactivity similar to those expected from a decommissioned site is proposed in Section 5.5.1 of the 1990 recommendations of the International Commission on Radiological Protection (ICRP 1991) and has been adopted in the current draft of the Environmental Protection Agency " Draft Federal Radiation Protection Guidance for Exposure of the General Public"(EPA 97). l [s\ U ICRP 46 (ICRP 1985)contains a detailed and useful definition of the critical group that could be applied to decommissioning sites:
"46. The critical group should be representative of those individuals in the population expected to receive the highest dose equivalent, and should be relatively homogeneous with respect to the location, habits and metabolic characteristics that affect the doses received. It may comprise existing persons, or a future group of persons who will be exposed at a higher level than the general population. When an actual group cannot be defined, a hypothetical group or representative individual should be considered who, due to location and time, would receive the greatest dose. The habits and characteristics of the group should be based upon present knowledge using cautious, but reasonable, assumptions."
Similar definitions can be found in IAEA Safety Series No. 57 (IAEA 1995) and several NRC documents related to low and high level waste. ICRP 43 (ICRP 1984) contains an approach for defining homogeneity for the critical group:
"69. ..it is suggested that, in general, to satisfy the homogeneity requirement the ratio of maximum to minimum values should not exceed an order of mcgnitude. For many distributions, therefore, the mean will be a factor of two to three lower than the maximum postulated. The necessary degree of homogeneity in the critical group I \
V Draft NUREG-1549 52 February 2,1998
' depends on the magnitude of the mean dose equivalent in the group as a fraction of the relevant source upper bound, if that fraction is less than about one tenth, a critical group should be regarded as homogeneous if the distribution of individual dose equivalents
- lies substantially within a total range of a factor of 10, i.e., a factor of about 3 on either side of the mean. At higher fractions, the total range should be less, preferably no more than a factor of 3."
's
] Draft NUREG-1549 53 February 2,1998
p Appendix D - Dose Models System Conceptualization (see main text, Figure 1, Step 3) includes conceptual and mathematical model development and assessment of parameter uncertainty. The system conceptualization represents the process of systematically evaluating the level of uncertainty associated with a specific site and the quantification of that uncertainty, In order to manage the treatment of uncertainty associated with dose assessment at a given site, the four steps of scenario 6finition, pathway identification, model development, and assessment of parameter uncertainty are treated as a hierarchy, moving from the former of these to the latter. This appendix discusses development of models for calculating dose. The licensee uses the dose models to perform dose assessments (see main text, Figure 1, Step 4) using the mathematical representations of the conceptual models (codified in DandD or equivalent software). The dose assessment involves the execution of the numerical model(s) that implement the mathematical equations and will provide the basis for (1) assessing compliance with the individual dose criteria and (2) an analysis of the impact of uncertainty in models and input parameters on the model output. As is the case for the scenarios and pathy: . /s (see Appendix C), a licensee can use models in dose assessment that are either generic or site-specific. The following sections describe the process which a licensee should use in selecting models for dose assessment at its site. D.1 Generic models D.1.1 Mathematical models V) ( As with scenarios and pathways 'see Appendix C), conceptual and mathematical models have s been defined for the NUREG/CR-5512 methodology and these models (codified in the DandD code) are acceptable for making generic dose assessments. A licensee can use these models (and DandD) for its dose assessment based on an evaluation of whether or not the NUREG/CR- 5512 models are appropriate for their site given the followm.g assumptions made in developing the 5512 models and any change in the model assumptions or scenarios for site-specific analyses (NUREG/CR-5512, Section 4.1.2): a) If the NUREG/CR-5512 models and default parameters are used, the licensee would only need to provide information that demonstrates the models are appropriate and to describe and defend the source term. b) Initial radioactivity is contained in the top layer (building surface or soil) c) The remainder of the unsaturated zone and groundwater are initially free of contamination ri) The activity in the aquifer is diluted by the volume of water in the aquifer O r i V Draft NUREG-1549 54 February 2,1998
D.1.2 Selection of Codes O As noted above the mathematical models in NUREG/CR 5512 are codified in the DandD code. As noted in NUREG-0856 [Silling,1983), it is important that codes and databases used in the analysis be properly verified and documented according to a rigorous quality assurance (QA)/ quality control (OC) program. If the NUREG 5512 screening scenarios and groups are used and the DandD code is used to conduct the analysis, the licensee does not need to perform QA/QC, as the NRC has already completed this for this code, if the DandD is used, the licensee does not need to provide defense and justification for the selection and use of this code or for the use of default pathways, assumptions, and parameter values. NRC has already defined these default assumptions and parameter values such that their implementation provides NRC the necessary confidence that if the site meets the unrestricted dose criteria, the likelihood of an incorrect regulatory decision is very low. The licensea would only need provide to the NRC a copy of the DandD generated report to verify the version of DandD that was used in the analysis. Defense would be needed for the source characterization and for the alternative parameter values used (see Appendix E). D.2 Site Specific Models Site specific models might be developed by a licensee because they either find that the generic models do not describe their site, or because they choose to use a model different from the generic model developed in NUREG-5512 (and codified in DandD). O If site-specific models are developed (either through changes to the default parameter values, model assumptions or development of new models), then the licensee needs to defend the model and associated parameters. D.2.1 Site specific Model development D.2.1.1 Conceptual models if site-specific models and parameters are used, the licensee needs to develop and defend conceptual models of the physical system that describe the specific physical processes and exposure mechanisms for each pathway. The conceptual modelincludes the set of assumptions of how the described system can be simplified for representation with a mathematical model. The simplification of the physical system into a mathematical model requires the analyst to make consistent, defensible assumptions. The licensee needs to present and provide adequate defense for each assumption. In general, a defensible simplifying assumption is one for which the simulated outcome (dose) would not be increased by a more complex (realistic) representation of the system. It is likely that there is uncertainty in the conceptual model, that more than one possible interpretation of the system can be justified based on the existing information. This uncertainty should be addressed by developing multiple alternative models of the system and proceeding forward through the framework with all the conceptual models that are consistent with available data and result in doses that exceed the dose criteria. The conceptual models that result in Draft NUREG-1549 55 February 2.1998
doses that exceed the dose criteria will be determined under Step 4. If the conceptual model r uncertainty is incorporated in the dose assessment, then the value of data that would reduce or eliminate the uncertainty in the conceptual model can be estic.iated. D.2.1.2 Mathematical models i The conceptual model describes how the contaminants move from the source to the receptor. The mathematical models, and the numerical links between those models, are the equations that implement the conceptual model. Each transport and exposure pathway may require a separate conceptual and mathematical model. The source model generally describes a boundary condition for a contaminant transport model or the concentration for a model of direct exposure to the source. The pathway models provide an estimate of the amount and distribution (concentration) of the contaminant. The exposure model translates the concentration into an amount of energy (or mass) absorbed or ingested as a function of human behaviors. Finally, the exposure is translated into a dose based on the ICRP 26, 30 and 48 models (a regulatory based requirement for TEDE). D.2.1.3 SoLrce models Source models are developed based on the following: a) Possible mathematical representations of the source include constant concentration, specified mass flux and time variant concentration or flux boundary conditions. If the (p) NUREG/CR-5512 models are used, then the source is represented with an initial activity V density or concentration (the total amount of activity for each isotope per unit area on a building surface or per unit volume in the upper soil layer) which changes over time due to radioactive decay (depletion due to decay, production from decay of the parent) and transport away from the source area (by leaching from soil or resuspension from the building surface). The teaching and resuspension processes are modeled as fractional releases of the total source mass. b) in the analysis of the dose due to contamination of building surfaces, the DandD models estimate the dose due to inhalation as a function of the concentration in air. A resuspension factor is used to estimate the concentration in air as a function of the concentration on the surface. The licensee may choose to propose a site or contamination specific resuspension factor. [ Insert text on the type of support that they would need to provide) c) In the DandD models, soil contaminatica is divided into two components: sorbed mass and leached mass. All the mass that is not retarded by sorption is leached from the source and transported to the groundwater system during the first simulated year. In reality, the amount of mass that is transported to the groundwater system in the first year will be a function of the infiltration rate and the contaminant solubility which is a function of the geochemical conditions and the physical and chemical nature of the source of contamination. The licensee may choose to perform laboratory experiments or conduct geochemical modeling to support a more realistic representation of the
'd Draft NUREG-1549 56 February 2,1998 I
l
O source. It is recommended that the identification and selection of options for site (' specific analyses be weighed in terms of the potential benefit, time frame and costs (Steps 8-10). D.2.1.4 Transport models l l The potential transport mechanisms for moving the contaminant from the source to the receptor include mechanical disturbances by the receptor (direct exposure to the source) and diffusive and advective transport via air (wind), surface water and groundwater (unsaturated and saturated). The models for these processes can be very complex (e.g. three-dimensional, transient, advection-dispersion equations for flow through heterogeneous media with source and sink terms) or simplified empirical models (e.g., transfer functions like resuspension factor). The level of complexity of the model that can be justified depends on the nature of the simplifying assumptions (conservative, reasonably conservative) and the information available to support the model (a complex model may be more realistic, but the data necessary to support the development of parameter values may not be available or obtainable). Multiple, simple alternative models may be necessary to evaluate the system when the relative conservatism cannot be determined a pn'ori. D.2.1.5 Exposure models The conceptual model describes the human behaviors (scenario and pathways) that lead to, and control the amount of, exposure, it includes the consumption rates (e.g. rates of respiration times the volume of intake per inhalation) for each media and the time and duration of exposure. (,mV) D.2.1.6 Dose models The dose criterion in 10 CFR 20.1402 is based on the TEDE concept. The TEDE is to be calculated based on the definition of TEDE in Subpart E and the models referred to in NUREG/CR-5512. Once the numerical models are developed, the licensee presents all the mathematical models and how each modelis linked. The model parameters are defined in this process. D.2.2 Use of deterministic or probabilistic approach for site specific models in preparing site specific models, the licensee can either conduct the analyses deterministically or probabilistically. A deterministic estimate of dose clearly and demonstrably bounds the potential doses, whereas a probabilistic approach quantitatively depicts system performance as a distribution of potential outcomes based on uncertainty and variation in models and parameters. Regardless of the type of analyses the licensee chooses to use, the same level of defense would be needed to demonstrate that the analyses provide sufficient information for the license termination decision. In addition, the licensee would need to demonstrate that their analyses are consistent with the framework such that if additions' data were collected, the simulated dose would decrease. These two approaches are:
/ T Draft NUREG-1549 57 February 2,1998
a): Option 1 - Deterministic analysis Deterministic analysis involves the calculation of a single value of the dose using single values for input parameter values. Single estimates of dose often can be conducted - easily, but the selection of appropriate models and parameter values may be difficult. When performance is measured against a single estimate, uncertainty is addressed by providing reasonable assurance that this estimate conservatively bounds actual performance. Given the uncertainties inherent in these dose assessments, it is i expected that bounding analyses will use simple modeling approaches, assumptions, and parameter values that readily can be demonstrated as being conservative (i.e., produce simulated doses that are consistently greater than actual doses). 4 b) Option 2 - Probabilistic analysis i i Probabilistic approaches encompass a wide range of analysis techniques and methods. For this report, the probabilistic approach refers to the use of a formal, systematic uncertainty analysis to quantify the uncertainty in performance estimates because of uncertainty and variability in models and parameters. Probabilistic analyses under this framework would involve the analysis of individual scenarios, each with multiple possible pathways, and possibly with attemative models for certain pathways. Parameter uncertainty would likely be quantified and propagated through the dose assessment models. Parameter uncertainty is often evaluated using a Monte Carlo analysis where the input variables representing parameter uncertainty and the output of model(s) are in O the form of distribution functions (see Davis, et al.,1990). An output distribution is produced by evaluating the performance many times, using sets of input values based on random and Latin Hypercube Sampling (LHS)[lman and Shortencarier,1984)._ The
- specification of the parameter distribution should reflect the level of knowledge about the parameter or "degrea of belief" rather than concentrate on rigorous statistical efforts to determine distributions. As a result, this approach does not require extreme amounts of site specific data to specify the parameter distributions, and in fact can be conducted with small amounts of data /information. : Assigning probabilities to scenarios, which is !
characteristic of some probabilistic approaches, is not recommended for dose assessments under this framework. That is, compliance will be assessed and demonstrated for each scenario independent of other scenarios. Similarly, assigning probabilities to attemative models is not recommended under this framework. Probabilistic analyses may be used to support compliance determination based on a deterministic value taken from the resulting distribution of output or compliance - determination based on a comparison of the entire output distribution to the performance objective. D.2. Selection of site specific codes
' The licensee will need to defend and justify their selection and use of a given computer code in order for their analyses to be acceptable to the NRC. In principle, the selection of a given computer code should be based on the scenarios, critical groups, and pathways defined in Draft NUREG-1549 58 February 2,1998
Appendix C and the mathematical model defined in Section D.1 of this appendix. To do this.
.\ the licensee will need to demonstrate that the mathematical representation of a given fate or transport process as implemented within the selected code is not inconsistent with the set of assumptions defined in Appendix C and will have to verify that the mathematical representation as implemented in the code is correct.
If enough uncertainty exists such that alternative conceptual models exist (i.e., alternative sets of assumptions are proposed), then it will probably be necessary to select alternative codes or alternative configurations of the same code and conduct the analyses with each of these. The licensee will need to provide results from the conceptual models with doses that exceed the performance objective or from all the conceptual models. Often times, it will not be possible to deduce, until after the quantitative dose assessment, which model yields the highest doses.
-The options for code selection for a site specific analysis and the defense needed under each option are:
a) Use DandD with alternative carameter values and modified / eliminated oathwavs if the licensee elects to use DandD but modify or eliminate the generic pathways listed developed in NUREG/CR-5512 (and listed in Appendix C.1), the licensee will need to describe the modifications to the DandD pathways for their site and justify that the l modified site representation in DandD is appropriate to use for their proposed
- conceptual model. The licensee will need to provide to the NRC a copy of the DandD generated report to verify the version of DandD that was used in the analysis and to l
O describe the DandD code's representation of the licensee's conceptual model, t Q b) Licensee-selected code
)
If appropriate, the licensee can elect to use a code other than DandD. As described above this may occur if the site is such that the DandD (i.e., NUREG-5512) models are not appropriate for the site or if the licensee chooses or prefers another code. Other codes that might be used include other standard codes (e.g., RESRAD) or codes which are developed by the licensee. In either case, the licensee will have to: (1) demonstrate that the set of implicit assumptions associated with the code that they have chosen are consistent with the site specific scenario and pathways (see Appendix C) and the site conceptual model(s) (see Section D.1 above). (2) if the licensee uses a code that has default parameter values built in, defend the appropriateness of those parameter values for their site within the context of the framework. That is, the licensee would have to demonstrate that if additional data were collected, the simulated doses would be very unl:kely to increase. (3) defend the model assumptions implied by the use of the code (i.e., the model assumptions should be consistent with the framework). V Draft NUREG-1549 59 February 2,1998
(4) provide to the NRC, as necessary, a copy of the code executable, user's manual i-for the code, an electronic copy of the input file, and an electronic copy of the I output file. As noted in NUREG-0956 (Silling,1983), it is important that codes and databases used in the analysis be properly verified and documented according to a rigorous quality assurance (QA)/ quality control (QC) program. Thus for either case a or b above, the license would need to perform QA/QC for the code used. j O b Draft NUREG-1549 60 February 2,1998 1 __J
A Appendix E: Parameter Descriptions and information for Changing Parameters E.1 Parameter Descriptions and Information for Changing Parameters Tables E 1 through E-6 list parameters to be evaluated if model parameters are changed from the defaults. Each of the tables indicates a definition of the parameter and also considerations involved in modifying the parameter. More details about the parameter distributions are contained in Attachment 1 to this NUREG. The evaluation and potential modification of the parameter will be different depending upon whether the parameter is physical, behavioral, or metabolic, and upon whether a deterministic or probabilistic analysis is performed. Note that, for deterministic calculations,~ parameters that are not modified using regional or site-specific information will be set to the value of the 95th or 5th percentile of their original distribution, as noted in the parameter descriptions below. Physical parameters are presented in Tables E-1 through E-3 as follows: Table E-1 Physical Parameters That Need to be Evaluated if Water Pathway Parameters are changed Table E-2: Physical parameters Which Should Be Evaluated if Diet or ingestion Parameters Are Changed Table E-3: Physical Parameters Which do not need to Be Changed if Other Parameters Are Changed '(\ These parameters were originally defined to encompass the variability expected across all licensees in all regions of the country. These parameters usually depend on physical features of the site that may vary based on local geological and meteorological characteristics.
. Modifications to these parameters can be based on the development of a narrower distribution that better represents site-specific features or location, or selection of a more realistic but still bounding deterministic value from within the distribution developed for the default analysis.-
Some physical parameters are surrogates for multiple processes within the model and are not correlated to specific physical processes that will be significantly different from site to site, or development of site-specific information may require complex or expensive specialized analyses that would not normally be justified for a decommissioning action.- These parameters are in a separate table to clarify which parameters need to be changed and which parameters may be changed whenever parameter modification is chosen as the preferred option. Behavioral parameters represent the average member of the screening group iad are contained in Tables E-4 and E-5 as follows. : Table E-4 Behavioral parameters that need to be evaluated for site specific critical groups Table E-5 Behavioral parameters that may be changed to account for modifications to screening group assumptions These parameters are based on the variability between individuals in the screening group. The metabolic parameters are contained in Table E-6, which also includes discussion of dependent C Draft NUREG-1549 61 February 2,1998
parameters, represent the physiological variability between individuals in the screening group. O These parameters were defined by development of distributions representing the screening group, then selecting the mean of the distribution to represent the average member of the group for the deterministic value to be used in the default modeling. These mean values and underlying distributions are not expected to change based on site-specific information unless the licensee proposes a site-specific critical group which is different from the screening group. Therefore, a licensee who chooses the option of modifying parameters will generally not need to modify the behavioral and metabolic parameters. However, a critical group may be defined for restricted use scenarios, or to account for physical features or legal requirements which cause the screening group to not be representative of the current and future use of the site. If the screening group definition is modified or replaced with a site-specific critical group, licensees should evaluate, and modify as appropriate, all behavioral and metabolic parameters related to the critical group. Table E-1: Parameters That Need to be Evaluated if Water Pathway Parameters are changed - Physical Parameter Description Discussion g H, Thickness of Definition: the The thickness of the unsaturated zone is used in determining unsaturated radionuclide leach rates from the unsaturated zone to the zone saturated zone. The default distribution was developed from 9 area-weighted data from observation wells across the U.S. Information on H2 (also called water table depth) is readily available from state or city governments and the USGS. Site Soecific carameters: Because oata are easily available and because it is not possible, a priori, to determine whether a thick or thin unsaturated zone is more conservative, licensees using deterministic modeling should use the best estimate of the minimum value for their site. Draft NUREG-1549 62 February 2,1998
)
D Table E 1: Parameters That Need to be Evaluated if Water Pathway Parameters are changed - Physical Parameter Description Discussion I, fi , f, infiltration rate Definition:
& saturation Infiltration rate is measured as the volume of water per unit ratios area per unit time that percolates deeply beneath the root zone -
and becomes infiltration.' The saturation ratio is the volume of water relative to the volume of the pore space, and also the ratio of the moisture content to the porosity. Both these parameters will vary based on regional climate characteristics and site soil texture. A full discussion of these parameters and their derivation, as well as possible information sources for site-specific values, is contained in Attachment 1. Site soecific carameters: Because data are easily available, and because it is not possible, a priori, to determine whether high or low values are more conservative, licensees using deterministic modeling should use the best estimate of the median value for their site. IR Irrigation water Definition: application This parameter represents the annual average quantity of rate groundwater used to irrigate on site agricultural products.- It is used, along with the area of land cultivated (A,) to calculate the volume of water removed from the aquifer per year for irrigation. Site soecific carameters: Licensees may propose changes to this parameter based on regional precipitation and regional soil moisture levels and other soil properties, and data that support alternative irrigation rates for certain forage crups or edible foods that may be supported due to prevailing dietary pattems or land use patterns. Because it is not possible, o priori, to determine whether high or low values are more c:nservative, licensees using deterministic modeling should use the best estimate of the median value for their site, based on a multi-year state-specific annual average irrigation rate (attached parameter description report contains such data for twenty-seven states). Draft NUP.EG-1549 63 February 2,1998
p g Table E-1: Parameters That Need to be Evaluated if Water Pathway Parameters are changed - Physical Parameter Description Discussion l i, n,, p,, Porosities, soil Definition: r>,,P, bulk densities, Porosity is a measure of the relative pore volume in the soil and and soil areal is the ratio of the volume of the voids to the total volume. Soil density of the bulk density relates the mass of dried soil to its total volume surface plow (solids and pores together). Soil areal density of the surface layer plow layer is a measure of the mass of soil per square meter in the surface layer, with an assumed depth of 15 cm for the DandD model. Porosity varies with soil texture, and distributions based on the 12 Soil Conservation Service textural classifications are listed in the attached parameter descriptions. Bulk density can be defined as functionally related to porosity: Bulk density = (1 - porosity)*2.65. Soil areal density is calculated as a conversion of units from bulk density plus the 15 cm depth assumption: Areal density = 150* bulk density or Areal density = 397.5*(1 - porosity). Site soecific carameters: Because it is not possible, a priori, to determine whether high or low values are more conservative, licensees using deterministic modeling should use the best estimate of the median value for O their site, based on the site specific soil texture. x I
'b Draft NUREG-1549 64 February 2,1998
i g Table E 2: Parameters Which Should Be Evaluated N If Diet or Ingestion Parameters Are Changed - Physical Parameter Description Discussion Animal 6ed Definition: intake m,ec These parameters represent the average daily quantities of on-for site produced foods and on-site well water consur..ed by livestock. Default values were developed based on the Q, forage assumption that the total annual diet for the animals is derived from on site contaminated feed and water ' rom the on-site well. Q, grain Site Soecific carameters Licensees may propose parameter modifications based on On hay lim' .ns on the types or quantities of feed that can be raised on the w 3 and the existence and quality of the on-site well, intake Q,, water rates can be used to directly account for the contaminated fraction of feed and water in the animal diet. (Deterministic j calculations should be based on the 95th percentile value of the default or revised distribution] Y, Crop yields Definition: (grain) This parameter represents the average yield of all grain crops consumed by each of the four food-producing animals evaluated in the model, per unit area of cultivated land at the site. The distribution was based on the production of three main grain b) crops (corn, sorghum, and oats) in direct proportion to the production across the United States. Site soecific carameters Licensees may modify this parameter by limiting the distribution to crop types likely to be grown in the area of their site, as well as incorporating climatic conditions and soil features that may affect production. [ Deterministic calculations should be based on the 95th percentile value of the default or revised distribution] Yn Crop yields Definition (stored hay) This parameter represents the average yield of all hay crops consumed by each of the four food-producing animals evaluated in the model, per unit area of cultivated land at the site. Site soecific carameters Licensees may modify this parameter by limiting the distribution to crop types likely to be grown in the area of their site, as well as incorporating climatic conditions and soil features that may affect production. [ Deterministic calculations should be based on the 95th percentile value of the defau!t or revised distribution) Draft NUREG-1549 65 February 2,1998 i
Table E 2: Parameters Which Should Be Evaluated g If Diet or Ingestion Parameters Are Changed Physical Parameter Description - Discussion Y, Crop yields Definition (stored This parameter represents the amounts of garden produce grown i vegetables, per unit area of cultivated land at the site and is based on the fruits, & production of all crops in direct proportion to the production grains) across the United States, Site soecific carameters Licensees may mody this paiameter by limiting the distribution to crop types !; haly to be grown in the area of their site. [ Deterministic calculations should be based on the 95th percentile value of the & fault or revised distribution) L 1 L o f
Table E 3: Parameters Which do not need to Be Changed
/m) If Other Parameters Are Changed * - Physical Parameter Description Discussion B, Vegetation Definition concentration This parameter is affected by multiple factors that vary factors for uptake non linearly in t;me and across locations.
Site soecific carameters Licensees are not expected to modify the default without specialized site-specific analysis. Licensees may propose different values based on published, peer reviewed data not evaluated in the parameter analysis. However, no further analysis is required by the licensee, and this parameter does not have to be modified if other l parameters are changed. [ Deterministic calculations l should be based on the 95th percentile value of the default or revised distribution) f.c Fraction of carbon Site soecific carameters , in animal products Licensees are not expected to modify the default without specialized site-specific analysis. Licensees may propose different values based on published, peer reviewed data not evaluated in the parameter analysis. However, no further analysis is required by the licensee, and this f) (j parameter does not have to be modified if other parameters are changed. [ Deterministic calculations should be based on the 95tn percentile value of the default or revised distribution! CDO,CDG Air dust-loading Definition outdoors & These parameters represent the long-term averages for gardening respirable particulate materialin outdoor air. Site soecific carameterji Licensees may propose alternate values based on site-specific, local climatic conditions which impact Just loading such as wind speed, soil moisture, soil type, i topography, and vegetation cover. Table 3.2.2 in the attached parameter description provides additional informatten. [ Deterministic calculations should be based on the 95th percentile value of the default or revised distribution) (C) m Draft NUREG-1549 67 February 2,1998 j
Table E 3: Parameters Which do not need to Be Changed if Other Parameters Are Changed' Physical Parameter Description Discussion fen, te,, fe, Fraction of carbon Site _ specific carameters in forage, stored Licensees are not expected to modify the default without grain, and stored specialized site-specific analysis. Licensees may propose hay different values based on published, peer reviewed data not evaluated in the parameter analysis. However, no further analysis is cequired by the licensee, and this parameter does not have to be modified if other parameters are changed. The one exception is fee because of the different forage crops that grow in different regions throughout the U.S. Regional data may support a different value based on specific forage crop growth. [ Deterministic calculations should be based on the 95th percentile value'of the default or revised distribution] O 4 Draft NUREG-1549 68 February 2,1998
\1
n Table F-3 Parameters Which do not need to Be Chang 3d
! If OtP4r Parameters Are Changed * - Physral Parameter Descriotion Discussion KD, Partition coefficients Definition Partition coefficients define the ratio between radionuclide solid concentrations (radionuclide quantity adsorbed on the soil / rock particles) and radionuclide liquid concentrations (redionuclide quantity dissolved in the soil / rock pore water) under equilibrium conditions. These coefficients are used to calculate radionuclide retardation and define the transport velocities in the soit layer and unsaturated zone. Transport velocities determine the radionuclide leaching rates. Partition coefficients noticeably affect doses because they significantly influence the mass transfer rates between soil, unsaturated zone, and aquifer and the subsequent concentrations in soil, drinking water, and water used for agricultural purposes. Radionuclides most sensitive to this parameter tend to be those whose leaching rates are l comparable to or greater than the radionuclide radioactive l decay constant. Partition coefficients are not correlated to l soil type or texture, or other easily measurable site l
O characteristics. t
\ 'j Site soecific oarameters Licensees using deterministic analyses may only replace the default values with values determined from site-specific testing or propose different values based on published, peer reviewed data not evaluated in the parameter analysis. However, no further analysis is required by the licensee, and this parameter does not have to be modified if other parameters are changed.
[ Deterministic calculations should be based on the 95th or 5th percentile value of the default or revised distribution. depending on the specific radionuclide) p ; j Draft NUREG-1549 69 February 2,1998
r g Table E 3: Parameters Which do not need to Be Changed If Other Parameters Are Changed * - Physical Parameter Description Discussion RF, Resuspension Definition: factor This parameter represents the ratio of the long term average resoirable contaminant concentration in air to the long term average floor surface contaminant concentration due to contaminated soil tracked indoors. Site soecific car 3 meters Licensees are nut expected to modify the default without specialized site specific analysis. Licensees may propose different values based on published, peer reviewed data not evaluated in the parameter analysis. However, no further analysis is required by the licensee, and this parameter does not have to be modified if other parameters are changed. [ Deterministic calculations should be based on the 95th percentile value of the default or revised diMribution) l r, Interception fraction Definition for vegetation This parameter represents the average fraction of all ! deposited contaminates retained on all plants grown for food and animal feed after above-ground irrigation with t contaminated groundwater, ( Sjfe soecific carameters Licensees may modify this parameter based on the chemical form of their source term, since different distributions can be supported based on contaminants which are negatively charged versus positively-charged or insoluble (see attached parameter discussion for details), (Deterministic calculations should be based on the 95th percentile % e of the default or revised distribution) Draft NUREG-1549 70 February 2,1998 i
,q , Table E 3: Parameters Which do not need to be Changed ! ) If Other Parameters Are Changed * - Physical V
parameter Description Discussion 6 Volume of water Definition removed from the This parameter represents the annual volume of aquifer per year for groundwater removed from the aquifer for domestic uses, domestic uses including such things as showers, washing, and water used for drinking and cooking. V, includes the volume of water used for drinking, defined by U , and along with the volume of water used for irrigation, establishes the total volume of water in the aquifer. Site soecific oarameters Since this parameter is influenced by site specific considerations such as climate, rainfall, and societal : restrictions on water use, licensees may propose alternative values for this parameter based on the State-specific values in the attached parameter description document, USGS county data, or other equivalent information. [ Deterministic calculations should be based on the 95th percentile value of the default or revised distribution) wet to-dry Definition conversion factors Wet to-dry conversion factors correspond to the fraction )[mV} of dry matter in the particular crop, and varies with the (forage) type of crop and the growing conditions. The va!ue for N, grain, both as used for animal feed and as consumed by (grain) humans, is proposed as a constant because there is so j N, little variability between different grain crops. (hay) Site soecific carameters Nn Conversion factors for fruits, vegetables, and hay / forage (vegetables, fruits, crops do vary t,ased vn the crop type, and licensees may N, & grains) propose different distributions from the defaults based on site-specific information about the specific crops that could be grown in that area. [ Deterministic calculations should be based on the 95th percentile value of the default or revised distribution) n. Draft NUREG-1549 71 February 2,1998 l
Table E-3: Parameters Which do not need to Be Changed
,, ) If Other Parameters Are Changed * - Physical w/
Parameter Description Discussion 4 Crop yields (forage) Definition , This parameter represents the average yiek of all forage crops consumed by each of the four food-producing animals evaluated in the model, per unit area of cultivated land at the site. The default distribution is based on the production of hay, as that was determined w be most representative. Site soecific carameters Licensees may modify this parameter by limiting the distribution to crop types likely to b grown in the area of their site, as well as incorporating climatic conditions and soil features that may affect production. [ Deterministic calculations should be based on the 95th percentile value of the default or revised distribution] P. Floor dust-loading Definition This parameter represents the long term average mass of contaminated soil per unit area of floor inside the residence, it is used with the resuspension factor to calculate the airborne particulate concentration due to . b resuspension of soil tracked indoors. V Site soecific parameters Licensees are not expected to modify the default without specialized site-specific analysis. Licensees may propose different values based on published, peer reviewed data not evaluated in the parameter analysis. However, no r further analysis is required by the licensee, and this parameter does not have to be modified if other parameters are changed. [ Deterministic calculations should be based on the 95th percentile value of the default or revised distribution)
- Licensees performing probabilistic analyses may use the original distributions developed for the default analyses in their calculations. Licensees using deterministic calculations should use the value of the 95th or 5th percentile of the original distr;bution or the value recommended in the parameter discussion, as stated in this table.
l () Draft NUREG-1549 72 February 2,1998 { I
l l 1 l Table E 4: Parameters That Need to be Evaluated for Site Specific Critical Groups - Behavioral (O) v Parameter Description Discussion t,, t,, t, Exposure periods DefinitiOD During the one year scenarie period, the average member of the screening group is assumed to divide their on site time between indoor, outdoor, and gardening activities. Site soecific carameters if the screening group definition is modified or replaced with a site-specific critical group, licensees should re-evaluate this parameter and modify it as appropriate. For example, if the critical group does not engage in agricultural activities, gardening time, alone with ingestion rates of domestic produce, cultivated area, and irrigation rate would be 0. [ Deterministic calculations ehould be based on the mean value of the default distribution) U,,U ,U, Ingestion rates of Definition home produced food These parameters represent ingestion rates of home produced leafy vegetables, other vegetables, fruits, grains (U,); beef, poultry, milk, eggs (U.); and fish (U). i The default ingestion rates represent the diet of the average p member of the screening group. These parameters are
, also important for defining the area of land cultivated k parameter A,.
Site soecific carameters While the defaults represent values developed from information in national surveys, site-specific values may be different based on regional and meteorologi:al conditions that impact agricultural practices and local dietary habits. U, can be set to zero if the site does not contain a pond or surface water that could support fish, or if any existing pond or surface water will not be contaminated with residual radioactivity during the 1000 year period following license termination. [ Deterministic calculations should be based on the mean value of the default distribution) V Draft NUREG-1549 73 February 2,1998 l __________a
p) i U Table E-4: Parameters That Need to be Evaluated for Site-Specific Critical Groups - Behavioral Parameter Description Discussion U. Drinking water Definition ingestion rate This parameter represents the long-term average daily ingestion of drinking water from an on-sae well. Site soecific 02rameters Licensees may modify (reduce or set to zero) this parameter based on site specific physical factors that affect the existence or quality of the well, or based on information supporting a finding that an on-site well would not become contaminated by residual radioactivity during the 1000 year analysis period. [ Deterministic calculations should be based on the mean value of the default distribution unless this pathway is completely eliminated) SFI Indoor shielding Definition factor This parameter represents the attenuation of gamma radiation by structural materials such as walls, floors, and foundations in residential buildings. The model uses a single, constant value for all radionuclides and all structural materials. Site soecific carameters a Licensees may substitute alternative values for this I \ parameter from Table X.XX based on a shielding factor for the specific energy range for the radionuclides in their , source term. It will usually not be acceptable to limit the l structural requirements for future structures that may be built on the site unless the licensee proposes restricted release, and such restrictions would not hold for the analysis of dose when controls fail. GR Soilingestion Definition transfer rate This parameter represents the quantity of soilingested per day, averaged over the one year duration of the scenario, by inadvertent transfer from hands or other objects that have been in contact with a contaminated surface, such as food, cigarettes, etc. into the mouth. Site soecific carameters if the screening group definition is modified or replaced with a site-specific critical group, licensees should re-evaluate this parameter and modify it as appropriate. [ Deterministic cakulations should be based on the mean value of the dem t distribution) Draft NUREG-1549 74 February 2,1998
Table E-5: Parameters That May be Changed to Account for Modifications to Screening Group Assumptions - Behavioral Parameter Description Discussion U,, U , U, Ingestion rates Definition of home These parameters represent ingestion rates of home produced food produced leafy vegetables, other vegetables, fruits, grains (U,); beef, poultry, milk, eggs (U ); and fish (U,). The default 3 ingestion rates represent the diet of the average member of the screening group. These parameters are also important for defining the area of land cultivated parameter A,. Site soecific oarameters While the defaults represent values developed from information in national surveys, site-specific values may be different based on regional and meteorological conditions that impact agricultural practices and local dietary habits. U, can be set to zero if the site does not contain a pond or sJrface water that could support fish, or if any existing pond or surface water will not be contaminated with residual radioactivity during the 1000 year period following license termination. [ Deterministic calculations should be based on l the mean value of the default distribution] U, Drinking water Definition C, ingestion rate This parameter represents the long-term average daily ingestion of drinking water kom an on site well. Site soecific carameters Licensees may modify (reduce or set to zero) this parameter based on site-specific physical factors that affect the existence or quality of the well, or based on ino 'ation supporting a finding that an on-site well would not become contaminated by residual radioactivity during the 1000 year analysis period. [ Deter-inistic calculations shcfd be based on the mean value of the default distribution unless this pathway is completely eliminated] r O Draft NUREG-1549 75 February 2,1998 I
\
p) t
\s Parameter Type Table E-6: Parameters That May Need to be Evaluated - Other Description Discussion V,, physical Volume of Definition (dependant) watei removed Tnis parameter represents the volume of water from the removed from the aquifer for irrigation of all crops aquifer per grown on site.
year for Site soecific oarameters irrigation use It is calculated as a function of the irrigation rate (IR) and the land area under cultivation (A,) and must be changed if either 'R or A,, or both, are changed. A, physical Area of land Definition (dependant) cultivated This parameter represents the area of land that is used for the production of agricultural products for both human and animal consumption. A,is calculated as a function of the nurnber of food and animal products considered in the diet, the ingestion rates for those products by the individual, and the yields for the food and animal products. Site soecific carameters pi Licensees may propose changes to the food and i t animal products that compose the on-site L) resident's diet based on the types of products that can be raised on the site, or physicallimits on the site area that can be cultivated. A, should be recalculated if the types of foods, ingestion rates, or yields are changed. In addition, if the screening group definition is modified or replaced with a site-specific critical group, licensees should re-evaluate this parameter and modify it as appropriate. CDI physical Air dust- Definition (dependent) loading This parameter represents the process of indoorn infiltration of contaminated airborne particles into the house (mass-loading) as the mass of infiltrating particles per unit volume of air. Site soecific carameters It is calculated as a function of CDO (air dust-loading outdoors) and PF (penetration factor) and must be changed if either CDO or PF, or both, are changed.
- \.y Draft NUREG-1549 76 February 2,1998 o w _ _ _ _ - _ _ _
Table E-6: Parameters That May Need to be Evaluated - Other
'd Par meter Type Description Discussion DIET behavioral Fraction of Definition (constant) annual diet This parameter wa9 originally intended to derived from represent the frsction of the average member of home-grown the screening group's diet that was derived from foods food grown on site in the contaminated area.
However, it was determined during the parameter analysis that a single diet fraction value for all food types was not representative of the screening group. Therefore. this parameter was set to 1, and the behavior of the screening group, which is expected to produce different fractions of each food product, is represented by the consumption rates U,, U,, and U,. The consumption rates have been redefined to represent the consumption of food derived from on-site production rather than the rate of consumption in generci. Site soecific carameters Therefore, this parameter should normally not be changed.
/3 SFO physical Outdoor (d)
(constant) shielding Definition This parameter represents attenuation of the factor external dose rate during periods outdoors based j on shielding by clean cover or other materials. Under normal circumstances associated with unrestricted release, and for evaluation of restricted release following failure of controls, this parameter should not be changed from 1. Site soecific carameters This parameter can be changed to account for physical controls under restricted release conditions. V,, V,, V, metabolic Volumetric Definition breathing These paramcters represent the annual average rates while breathing rate of the average member of the indoors, screening group while indoors, outdoors, and outdoors, and gardening. gardening Site soecific carameters if the screening group definition is modified or replaced with a site-specific critical group, licensees should re-evaluate this parameter and modify it as appropriate.
,0 h Draft NUREG-1549 77 February 2,1998 l
The following table lists shkrlding factors based on the maximum energy of the source term.
, Licensees may modify the SFl parameter in the model(E[SFl]) based on the maximum energy ,
-\ for their site specific source term. For example. if the source term maximum energy is less i than 0.4 MeV, the default value for SFl can be r@ laced with 0.574.
Table E.7 Shielding Factors For Various Materials vs, Energy; SFl Replacement Values Based on Maximum Energy Energy Energy (MeV) 3.5" 5.25" 7.0" 1.0" (MeV) E[SFl] , 0.015 1.36e 12 2.55e 24 2.55e 24 2.05e 06 0.015 5.13e 07 0.03 8.10e 03 8.10e 03 8.10e 03 9.67e 02 0.03 3.03e 02 0.06 2.41e 01 2.41e 01 2.41e 01 6.08e 01 0.06 3.33e-01 O.08 3.80e-01 3.77e-01 3.77e-01 7.22e 01 0.08 4.64e 01 0.1 4.38e-01 4.32e-01 4.31e 01 7.67e 01 0.1 5.17e 01 0.2 5.07e-01 4.86e 01 4.79e 01 8.07e 01 0.2 5.70e 01 ,
~
0.4 5.17e 01 4.78e-01 4.62e 01 8.14e-01 0.4 5.74e-01 0.8 4.89e 01 4.25e 01 3.94e 01 8.24e 01 0.8 5.77e-01 1.5 4.91e 01 4.05e 01 3.59e 01 8.45e-01 1.5 5.82e-01_ 2.25 5.14e 01 4.22e 01 3.69e 01 8.57e 01 2.25 5.85e-01 swsmewhensweerear Aearamennvma tanwwwwosme Ensw pg-._.. ..7,.,..,,, _,__ . _ i ; og _ 4- _. . . ._._ .
.j l
3, __ _ . _ _ __ _ i g g3 _ . . -. _ . _ . , . _ . - - . , - _ ..
, ll l l .
o, ..p 4 i. jp . p _ _ 3
- I l l I
- ot 0
om 01 1 # w e e-. m
- . - ~ . , - . , . - - - . . . - - - - . . - - - . - .
1 Draft NUREG 1549 78 February 2,1998
. - - , . . . . . - - . v --r..., --
3-.,,-,,,yg3 m -4_v-y , . _ , .--,--f. ,,.,--,.,-.-.m..r., 7--, -.-. i. _. ,-y
E.2 Assimilating Existing Data and Information l',~\ V This step involves gathering and evaluating existing data and information. Licensees should check their records to determine the types and amounts of radioactive material they possessed on their site. They should also gather information about any surveys and leak tests that had been performed, as well as any records that would support their ability to
- Certify the disposition of alllicensed material, including accumulated wastes, by submitting a completed NRC Form 314 or equivalent ini~ormation'[10 CFR 30.36(J)(1)).
Data are used to support Step 3 which is development of a conceptual model, and model asFumptions and model parameter values. As desenbed above, the licensee has 2 options in this analysis: (1) use the pre defined DandD models and one or more site specific parameter values (2) develop site specific models and accompanying parameter values Additional information is needed to support and defend the conceptual model of Step 3 if models other than DandD are used or if site specific parameter values are used. Types of potentially useful information include: processes that utilized the potential contaminants, releases and mitigative actions, hydroloCi c conditions (soil moisture content, conductivities, depth to groundwater, tydraulic gradients, hydraulic conductivities), soll type and texture, clay content, geochemical conditions (Kd, pH) , atmospheric conditions (annual averages or time and date specific conditions), geology (unconsolidated sediments, fractured rock). Methods for obtaining the necessary r.dditionalinformation to support the site specific parameters and /O models used are described in Sections 4.1.1 through 4.1.4. The 12/6/96 Draft of MARSSIM chapter 3 discusses the Historical Site Assessment (HSA) process. Only one of the five objectives of the HSA apply to Step 1 of the D&D decision framework. The common objective of Step 1 and the HSA is to identify the potential, likely or known sources of radioactive material and radioactive contamination based on existing information. Section 3.4 and Appendix G of the MARSSIM draft provide useful guidance on sources of information and Section 3.6 discusses how to identify potentially contaminated media. (Note: it may be more appropriate to reference Sec. tion 3.6 in the discussinn on the applicability of the 5512 pathway assumptions (Step 2) and models (Step 3) to the site.) (The other objectives of the HSA (identifying sites that pose a threat to human balth and those that do not, assessing the likelihood of contaminant migration, providing usefulinformation for developing and analyzing surveys, and providing an initial classification of the site or specific areas of the site as impacted or non-impacted) are similar to objectives of later steps in the framework. These inconsistencies arise because the MARSSIM methodology assumes that the DCGLs are established prior to beginning the MARSSIM process. A dose assessment is needed in order to establish the DCGLs. In order to perform that dose assessment, knowledge about the potential contamination and exposure mechanisms is needed. In the D&D framework, DCGLs would be proposed in Step 6 as part of the ALARA analysis, or in Step 8 when remediation options are defined.) E.2.1 Source data O V Draft NUREG 1549 79 February 2,1998
The licensee gathers and interprets all pertinent and legitimate existing site data and other g} ( C relevant information that can be used to define characteristics of the residual radioactive (and non radioactive) contamination at the site. In defining the residual contamination, all existing information on the amount, location and distribution of all possible contaminants should be evaluated. Where data are unavailable, the licensee may estimate the amount and distribution of the potential contaminant based on initialinventories (mass balance approach) and the processes involved in generating the original materials (e.g., ore processing, contained source, laboratory analyses). The uncertainty in the extent and amount of residual cor tamination for each substance will depend on the amount and variability of the data. The uncertainty in the magnitude and distribution of the source should be represented or bounded in the later dose assessment in order to evaluate the worth of collecting additional data about the residual contamination. The uncertainty in the extent and amount of residual contamination can be accounted for in the dose assessment by employing conservative assumptions about the source magnitude and distribution. As noted in Chapter 3 of the 12/6/90 draft of MARSSIM useful sources of information about the potential amount, form and distribution of radioactive contaminants include licenses, site permits, authorization documents, operating records, financial records, site plots, bluepnnts, photographs, aerial photos and maps. Licenses, permits and authorizations may indicate the quantities of radioactive material. chemical and physical form and types of operations. Operation records may include c ecounts of intentional and accidental releases of radioactivity (leaks, spills, disposal, storage, routine emicsions). These accounts may include estimates of the amount, distribution, and the chemical and physical form of the potential contaminants. Financial records may provide evidence of the amount of material entering and leaving the site. O Maps, figures and photos provide information for evaluating the location of potential V contamination based on operations. E.2.2 Hydrogeologic data Existing data on the geology, surface water and groundwater systems at the site are used to support the conceptual model, defend the use of the DandD models and support the dose assessment model parameter values. The data used to develop the default parameter values for the 5512 models provides a data base for estimating the uncertainty in the model parameter values (Step 3) and for evaluating how that uncertainty might be reduced given site specific information (Step 8). The hydrologic data in the 5512 parameter analysis include the range in observed unsaturated zone thickness (depth to groundwater), unsaturated zone and soil porosity, saturation ratios (volumetric moisture contents), infiltration rates and volume of surface water pond. As noted in the 12/6/96 draft of MARSSIM, potentially useful site specific information includes rainfall; the location of nearby wetlands, intermittent streams, drainages and surface water bodies (rivers, lakes, oceans, coastal tidal waters) relative to the potential sources of contamination; flooding potential; runoff rates; runoff barriers; infiltration rates; soil / subsurface permeabilities; depth to groundwater; and type of groundwater system (karst, fractured rock, porous media; confined; unconfined), Table G.1 in the 12/6/96 draft of MARSSIM, indicates useful sources of hydrogeologic data (in this document it includes geologic, hydrologic and O b Draft NUREG 1549 80 February 2,1998
groundwater characteristics). In addition to data collected during operations, other agencies
, i that may have useful hydrogeologic information and experts include: the USGS, state 'v' geological surveys, state environmental agencies, state departmonts of transportation, local colleges and universities, local well drillers, local water authonties, local health departments, EPA regional offices, U.S. Army Corps of Engineers, FEMA, US Fish and Wildlife, and national databases (WATSTORE, STORET, GRIDS, National Wetland Inventory Maps).
E.2.3 Chemical data Existing data on the chemical properties of the potentially contaminated material are used to support the conceptual model and dose assessment model parameter values. The data used to develop the defau;t parameter values for the 5512 models provides a data base for estimating the uncertainty in the distribution coefficients (Step 3) and for evaluating how that uncertainty might be reduced given site specific information (Step 8). This data set can be evaluated in terms of the soil type and site specific information on the soil type can be used to justify reducing the uncertainty in this parameter value. The Soil Conservation Service is the agency that may have useful information and experts to contact regarding soil type and potentially useful data bases include the National Soll Geographic Database, State Soll Geographic Database, and the Soil Survey Geographic Database. All of the databases are available through EPA's website. Other sources of site specific information include local experts at universities or colleges, state geological surveys and environmental agencies. E.2.4 Land Use data v if a site specific critical group is proposed, land use data will be used to defend the characteristics of the critical group and model parameter values. I As noted in Appendix G of the 12/6/06 draft of MARSSIM, local planning and zoning officials, tax assessor, and local university or college geography departments are potential sources of land-use information. The USGS is a source of land use and land cover information and the U.S. Bureau of the Census TIGER Map Seivice is a source of demographics information. A key point of this framework is that new site data collection does not take place until Step 12. New data collection is deferred until the data that would make a difference in decision making and are cost effective to collect can be defined through cost / benefit and data worth activities (Steps 8,9 and 10). Otherwise, money may be spent on collection of superfluous data. To start this decision process using the modeling approach den.cribed in Volume 1 of. NUREG/CR 5512, only information on the nature and extent of the residual contamination is needed. For new sites of any type the same approach is recommenJed. However, other sites may have evolved further in the process prior to using this approach. In this case, all relevant site data should be included and evaluated. This information would provide the bulk of the primary background information for any NEPA analysis and documentation for these sites. This information may be augmented under later steps where additional data collection activities occur. O V Draft NUREG 1549 81 February 2,1998 9
l 1 l 1 A I i l -- i Draft NUREG-1549 82 February 2,1998 l I i
p Appendix F Area Factors i l V F.5.0 Area Factors / Elevated Measurement Criteria:Inte0 ration of Modeled Risk with Areal Extent of Contamination Area Factors are used to calculate the maximum concentration, distributed over a specific area, that can remain following decommissioning without requiring additional clean up. They are used to determine the elevated measurement comparison value, as described in NUREG 1505, Chapter 5. Area factors are calculated as a ratio of the dose conversion factor (DCF) based on the default contaminated area to the DCF based on the contaminated area of interest. Since area factors can be applied to any site and are calculated generically, they are based on default parameters developed at a Pm level of 0.05 (Beyeler, et. al,1996). This level of conservatism is reasonable in the context of developing allowable multiples of the guideline levels for use at sit.is that will be released from license. All calculations for area factors in this report were done using the Residential and Building Occupancy Scenarios in DandD version 1.0. The source term in all cases is based on a unit concentration, equivalent to 1 pCl/g in the Residential Scenario and 1 dpm/100 cm'in the Building Occupancy Scenario. DandD parameters with links to area are shown on the spreadsheets included in Appendix A, along with proposed modifications based on area of contamination. If the parameter is not listed, no change was made to the Level 1 (L1) default. For the purposes of these calculations, the L1 parameters are set at the P = 0.05 level. F.5.1 General Assumptions ( F.5.1.1 Areal Distribution A) Residential Scenario For contamination under a house (the house scenario), it is assumed that the house has an area of about 2,000 square feet (-186 m2). The contamination is assumed to be completely covered by the house until it exceeds a size of 186 m2, at which time the contaminated area exceeding 186 m2 is assumed to be in the cultivated area (garden). For contamination in a garden (the garden scenario), the contaminated area is assumed to be comp!etely in the garden until the size exceeds the default garden size or a garden size associated with the area needed to support 50% of the individual's diet. Once the contaminated area exceeds the garden size, the excess is assumed to be under the house. B) Building Occupancy Scenario For contaminated areas inside buildings, the baseline room is assumed to have a floor area of 4 meters x 4 meters and a ceiling height of 3 meters. External dose is based on the assumption of an infinite flat plane with uniform contamination. 7.1.2 Diet (Residential Scenario Only) (3 tb Draft NUREG 1549 83 February 2,1998
. - - - . - -_ _ ~ _- -- -- --- - - - -.-_ _- -
1 l l i The assumption is made that no more than 50% of a person's diet would be from the contaminated area. Beyond 50%, site specific adjustments should be made to the parameters because the scenario has been extended beyond the original assumptions made in the i construction of the resident farmer scenario. The fraction of the diet is related to area using the L1 baseline area and diet fraction. For contaminated areas other than the default area, the fraction ls calculated as the ratio of the default diet fraction to the default area, multiplied by the contaminated area. As explained above, the maximum fraction is limited to 0.5. Fraction of Diel from Contaminated Area (OIET) s L1 Fraction of Olet u Contaminated Area (A), {1) L1 Contaminated Area DIET s 0.5 L1.3 Time A) Residential Scenario This model is structured in such a way that it is not simple to modify the time of exposure to an external source without also affecting the inhalation and ingestion pathways. The time variables used to control time spent indoors and outdoors affect both the time of exposure to external sources, as well as time inhaling resuspended dust and secondary ingestion. Time of exposure is important because it is used as a surrogate for modification of the source geometry. This model currently only supports ( an infinite flat plane geometry. The time of exposure to an external source is important for evaluating the effect of contaminated areas smaller or larger than the default area. For example, if it is assumed that a person has an equal probability of being at any location on the site at any time during the analysis period, then the time of exposure to the source can be related to the size of the contaminated area versus the entire site area, if the entire site is contaminated, the person is exposed to the source the entire time they are on the site. If one quarter of the site is contaminated, the person can be assumed to be exposed to the source for one quarter of their time on site. It is important to note that these simplifying assumptions are only valid within the context of this model, which was designed to evaluate distributed, relatively homogeneous low activity sources. It would not be valid, for example, to apply these assumptions or this model to an exposure assessment for a high energy gamma sealed source. While it is easiest to adjust the external exposure pathway by changing the duration of exposure, other pathways are best adjusted by applying a correction based on the ratio of the contaminated area to the site area while using the default exposure time. In addition, the external exposure pathway is complicated by the fact that it is divided into three components and uses two shielding factors. The components are gardening, outside activities other than gardening, and indoors. Separate shielding factors are applied to indoor and outdoor activities. Since both contamination in the garden and ( Draft NUREG 1549 84 February 2,1998
contamination under the house are being evaluated, it is important to be able to change both shielding factors and time spent in each of the three locations. In addition, this allows the time indoors, for example, to be used as a surrogate for time of exposure , without impacting the time exposed to resuspended dust from soil tracked indoors. Given these complications, adjustments to the time of exposure for the external dose pathway are made after the model has first been run with adjustments to all other parameters'. The external dose in the residential scenario is calculated by summing the time spent indoors, outdoors on site, and gardening. Additional details regarding the external dose pathway and how it is integrated into the residential scenario can be found in NUREG/CR 5512 Volume 1, page 5.52 to 5.54. The equation used to calculate external dose'is as follows: J, , DEXR,a 24 (t,/t,,) SFO C,,[ S {A,y, t ,}DFER, i l'1 . J, 24 (l atti,) SFO C,i [ S {A,y, t ,}DFER i (2) ji J,
+
24 (t,lti ,) SFl C., [ S {A,y, t ,}i DFER; jet where DEXR, = external dose from 1 year of residential scenario exposure to radionuclide I in soils (mrem for a year of residential scenario) DFER, = external dose rate factor for radionuclide j for exposure to contamination uniformly distributed in the top 15 cm of residential soil (mrem /h per pCi/g) A,, = concentration factor for radionuclide j in soil at the beginning of the current annual exposure period per initial unit concentration of parent radionuclide I in soil at time of site release (pCilg per pCilg) l l 'A Quattro workbook containing adjusted parameter sets and all calculations is attached. Names of workbook pages containing calculations associated with adjustments to the external exposure pathway have a standard format consisting of the radionuclide name followed by " ext fix". For example, the page associated with Cobalt 60 is named 'Co ext fix".
' Equation 5.69, NUREG/CR 5512, Volume 1 Draft NUREG 1549 85 February 2,1998
C., = concentration of parent radionuclide I in soll at time of site release (pCi/g dry weight soil) SFl = shielding factor by which external dose rate is reduosd during periods of indoor residence (dimension less) SFO = shielding factor by which external dose rate is reduced during periods of outdoor residence and gardening (dimension less) 4 = number of explicit members of the decay chain for parent radionuclide l S(A,,,t,) = time-integral operator used to develop the concentration time integral of radionuclide j for exposure over a 1-year period per unit initial concentration of parent radionuclide I in soil (pCl d/g per pCi/g dry weight soil) S{A,y,t,,} = time integral operator used to develop the concentration time integral of radionuclide j for exposure outdoors over one gardening season during 1 year period per unit initial concentration of parent radionuclide i in soil (pCi*d/g per pCilg dry weight soll) O t, = time during the gardening period that the individual spends outdoors gardening (d for a year of residential scenario) t, = time in the 1 year exposure period that the individual spends indoors (d for a year of residential scenario) t, = time in the 1-year exposure period that the individual spends outdoors, other than gardening (d for a year of residential s& nano) t,, = total time in the gardening period (d) t, = total time in the residential exposure period (d) 24 = unit conversion factor (h/d). The concentration time-integral factors, S{}, are evaluated for all radionuclides in a decay chain. The factors represent the time integral of concentration during the exposure period of interest. The concentration factor, A,,, defines the concentration of each radionuclide in soilin a decay chain at the beginning of the current year of the dose evaluation. The concentration includes materialinitially present in d Draft NUREG 1549 86 February 2,1998
, the soil, plus material that has migrated to ground water and been
( redeposited onto the farmland soil by irrigation with the contaminated water (- during the previous year. Equation 2 can be reorganized and simplified to isolate the times and shielding factors of interest: DEXR, = K = [(t,= SF/) + (t,x SFO) + (t,= SFO)) (3) Where K= combined L1 parameters and other variables are as defined above. Assuming that the receptor has an equal probability of being at any point on the site, the time of exposure to the contaminated area can be calculated by multiplying the default exposure time by the ratio of the size of the contaminated area to the Level 1 (L1) default area size. The external dose due to exposure to a contaminated area of any size is calculated by applying the times and shielding factors associated with the area of interest. The shielding factors are not adjusted in the same way as time of exposure for area of O("/ contamination They are only used to turn the indoor or outdoor external exposure pathway completely on or off. When the pathway needs to be turned off, the shielding factor is set to zero. If the pathway is on, the shielding factor is set to the L1 level. Therefore, the ilma of exposure is the primary way that the external exposure is varied to account for the size of the contaminated area. The revised external dose is calculated by multiplying K, which is composed of known Li values, by the modified exposure times and shielding factors: DEXR,(A) = K x [(t,(A) x SFI) + (t,(A) = SFO) + (t,(A) = SFO) (4) Where A = contaminated area (m'), DEXR,(A) = external dose based on area A from 1 year of residential scenario exposure to radionuclide I in soils (mrem for a year of residential scenario) and other variables are as described above. C'\ Ij x Draft NUREG-1549 87 February 2,1998
p Once the revised external dose has been calculated. the area corrected DCF, DCFg,, C) ! can be calculated. DCFmis calculated by first running DandD with parameters (except time) adjusted for the contaminated area of interest. The resulting DCFg ,(DCF without time of exposure modification)is then adjusted by first subtracting the external dose contribution calculated without accounting for the time factor, then adding the corrected external dose: DCF,,=lOCFm - DEXRi)
- DEXRi(A) (5)
The area factor can then be calculated by dividing the baseline DCFw yb the area corrected DCFg, for the specific contaminated area of interest.
- 8) Building Occupancy Scenario Calculation of external exposure for the building occupancy scenario is simpler than the residential scenario because all exposure occurs inside the building and no shielding factors are used, However, the same need exists to separate the time of exposure to external sources from inhalation and ingestion. Therefore, the external dose is modified after the model is run, in the same general way as described above, and the area-corrected DCF is calculated as shown in equation 5, 7.2 Parameter Specific Assumptions y/ Most parameters in both the residential and building occupancy scenarios are modified by being multiplied by the ratio of the contaminated area of interest to the L1 default contaminated area. This provides a reasonable and repeatable method for adjusting the impact of various pathways, based on the assumption that such a ratio can act as a reasonable surrogate for variations in the contaminated fraction based on area.
An example of the application of the ratio of contaminated area to L1 default area is demonstrated with the air dust loading factors. These factors are described in NUREG/CR-5512, Volume 1, pages 6.10 through 6.12. The use of dust loading rather than resuspension was originally selected because it was assumed to be the most straight forward approach for prospective screening, and would require the least number of assumptions regarding input parameters. The base assumption is that the dust loading parameter represents contaminated, respirable dust. Unfortunately, dust loading does not allow direct incorporation of the impact of contaminated area size on the contaminated fraction of resuspended dust. However, a crude approximation of the impact of area can be incorporated by assuming that as the contaminated area decreases in size, the amount of contaminated material versus clean material available for resuspension also decreases. Therefore, while the total amount of dust in the breathing zone would remain the same, the fraction contributed by contaminated soil could be assumed to decrease in direct proportion to the contaminated area. This is approximated by modifying the dust loading parameters by the ratio of contaminated area to the L1 default area. V Draft NUREG 1549 88 February 2,1998
The resuspension fcetor used in the building occupancy scenario, is difficult to adjust because it is insensitive to the distribution of contamination and the size of the contaminated area As a first approximation, and within the constraints of this study, it is assumed that the resuspension factor can vary between the minimum value assumed in the parameter analysis (1E 6 m"), and a maximum of the Li default for areas equal to or greater than the assumed default room size. Analogous to the discussion of dust loading, this approach is based on the assumption that while the resuspansion factor may remain constant, the contaminated fraction of material that is resuspended decreases with a reduction in the size of the contaminated area. For the house scenario, the ratio is modified by the area (186 m') that is assumed to be under the house, and which therefore does not contribute to any pathway except external exposure. The fish ingestion parameter is only used to turn aquatic food ingestion on or off, as is the contaminated water ingestion pathway. Shielding factors are set to either the L1 default value or zero, since they are only used to turn the external exposure pathway on or off. For example, when no contamination is located under the house, the indoor shielding factor is set to zero, and when all contamination is located under the house, the outdoor shielding factor is set to zero. In most cases, the L1 default parameter value is assumed to be the maximum reasonable value, and areas larger than the default do not cause the parameter value to increase Since the default is set at a known conservative value, it is not necessary and would likely be unduly unrealistic to assume higher values. Exceptions are the fraction of the diet from the on site garden, which can increase to a maximum of 0.5, and the time spent gardening, which is tied to garden size, \ Draft NUREG 1549 89 February 2,1998 ____ . - - - - J
n Appendix G Examples i
- 1. Example applications A logical, consistent decision process is viewed as a useful tool that will support licensee planning of decommissioning activities and NRC review of license termination requests. To support this process, Chapter 2 of this NUREG describes a decision framework to support
, implementation of the dose criteria of Subpart E of 10 CFR 20. Three example applications are described in this Appendix which illustrate the cases described in Chapters 3,4, and 5 of this NUREG. 2.1 Case 1 Use of the Framework for licensees who use Generic screening j Slen 1 Assimilating existing data _and InformA110n. In checking records to determine the types and amounts of radioactive material they possessed on their site, and gathering information about any surveys and leak tests that had been performed, the licensee in this example determines that: a) all wiste has been properly disposed, b) sources have been properly transferred to another licensee, c) minor amounts of contamination have been detected inside a laboratory building during [3 routine surveys. Sten 2 - Scenarlo Definitiontpathway identification The licensee would note that: a) The building occupancy scenario applies, with the associated inhalation, secondary ingestion, and external exposure pathways (building occupancy applies to situations where contamination exists on interior builciing surfaces (but not in the soil) and where the building will be re-used for commercial (not residential) purposes following license termination. b) for the simple case considered here, Step 2 has already been completed by the NRC, based on the generic scenarios and pathways for screening that have been defined and described in NUREG/CR 5512, Volume 1. Steo 3 - Syttem Conceptualization For the simple case considered here, Step 3 (conceptual and mathematical model development and assessment of parameter uncertainty) has already been completed by NRC, using the models described in NUREG/CR 5512, Volume 1, by its preparation of the DandD software and the generic screening tables of Appendix A and B. ( \ V' Draft NUREG 1549 90 February 2,1998
Ette_4.: D. Die.Atatsamatit In this example, the licensee could either: a) run DandD and plug in the maximum surface contamination concentrations from the, existing building surveys b) compare the maximum surface contamination concentrations from the existing build;ng surveys to the generic screening concentrations in Tables A 1 or A 2 of Appendix A. The maximum curvey results should be used because, if the dose assessment using these values indicates that the dose is below the 25 mremlyr criterion, there will be a high assurance that the site meets the dose requirements and aciditional refinement of the source term will be unnecessary. Slep 5 Datorminina if Site can be reltaand Based on Step 4, the licensee can then simply answer the question of whether the dose assessment results from the model are less than the dose criterion of 25 mrem /yr in 10 CFR 20, Subpart E. 1 in this example, the model results are much less than the 25 mrem criterion. Eten 8 - ALARA reauirementa In Step 6 the licensee would satisfy any remaining ALARA requirements (see Reg Guide xxx, Section 3). Eten 7 License Termination and SiteRelesem The licensee would: a) complete paperwork requirements, including documenting the survey results used to calculate the source term and the model output, b) submit necessary forms and request to have their license terminated by the NRC. Draft NUREG 1549 91 February 2,1990 a
l 12 Sate 2 - Llgensnes _who ustsite_spasific information but only_modifv_ site g paramettra This example illustrates use of the framework for a licensee that uses site specific information in their dose ascessment. As described Section 2.2, there are a wide range of options for using site specific data ranping from modifying parameters, to modifying models, to remediating the site, to restricting site use. , I This example describes use of the framework specifically for those licensees that conclude that
- the option of modifying parameters will provide a simple, cost effective means to comply with the dose criteria of Subpart E with only limited consideration of other options, This example is prepared sepa*ately from Case 3 (which includes a more in-depth eva'uation of options) because it is thought that a number of licensees will have relatively low levels and patterns of
, contamination and will seek to perform a dose assessment by changing certain parameters to more adequately represeni their site. This example is not intended to limit the options a t
licensee may pursue. j In this example, the licensee is interested in terminating the license for an outdoor location that is believed to have areas of soll contamination from leaks in a waste tank. Although this licensee has a more complex situation than that described in Case 1, they would still follow the same processes in Steps 1 5 described for Case 1, at least for the first iteration. J Bien 1. Analmilate Exlating.Date and.Existina Data and Information O 3 V The licensee would gather as much information as possible about their site. This might include: a) radionuclides and processes used, b) quantities and forms of material that might still remain on site, c) other information (e.g., ) useful for performing a site dose assessment.
- 3. ten 2 Scenario definition and nathway identiflgatiga In this example:
a) because some small amount of soll contamination exists, the residential farmer scenario applies, with the associated inhalation, ingestion, and extemal exposure pathways (the residential farming scenario applies to situations where contamination exists on soil surfaces to a depth of less than 15 cm with potential for use of the land for residential purposes following license termination). b) The licensee decides to begin the decision process by using the pro-defined scenarios and pathways in the residentia! scenario (soil contamination) described in NUREG/CR-5512, Volume 1. As for Case 1, for the simple case considered here, Step 2 has
\ Draft NUREG-1549 92 February 2,1998
,..m_. -
. . . , . --. __,. ._ . -_rm. , _ . _ ,_y. - y x
(g) already been completed by the NRC, based on the generic scenarios and pathways for screening that have been defined and described in NUREG/CR 5512, Volume 1.
'O S.teo 3 System Conteg1LIAllzadon The licensee continues the process of using the pre-defined methods by using the default parameters and the DandD software. For ths :,mp!e case considered here, Step 3 has already been completed by NRC, using the models described in NUREG/CR 5512. Volume 1, by its preparation of the DandD software and the generic screening tables of Appendix A and B.
Sten 1done_Astaannittit. The licensee runs DandD using a source term developed from the information gathered in step one, and which is the maximum reasonable value they believe they can defend. Slep 5 Cautado be released Based on the results of the dose assessment in Step 4, it is clear that the site does not meet the Subpart E dose criterion of 25 mrem /yr, The licensee would therefore proceed to Step 8. 1 Slep 8 Define Options for Stig (9 There are three options that the licensee could apply either alone or in combination: O a) Option 1 - Activities that reduce uncertainty (irfo.
- ntion/ data collection),
7 b) Option 2 - Activities that reduce contamination (remediation), and c) Option 3 Activities that reduce e>posure (land use restrictions). Table 2.2.1 lists some of the options that a license could consider, the first two related to Option 1, and the next two related to Opt;0ns 2 and 3, respectively, in this example, the nature of the soil contamination is relatively simple, and the options are relatively straightforward, in this case the licensee conducts the following fairly simple thought process regarding the options ibn Table 2.2.1: a) The 1st item in the table would reduce uncertainty in the source term (Option 1) and would require additional site characterization; b) The 2nd item would replace the default kd with a more site specific value based on the site soil type (Option 1) and would require collection of some additional data; j c) The 3rd option in the table would result in an actual reduction of the quantity of residual radioactivity remaining on the site by use of soil removal activities such as excavating, transporting, and disposing of the soil at a licensed burial site (Option 2).
/~N b Draft NUREG-154g 93 February 2,1998 i
l p....... .. .
n d) The 4th item in the tsble, reduction of exposure by restricting use, would require the ( licensee (per 10 CFR 20.1403) to demonstrate that unrestricted release was not ALARA L and to convene an SSAB. This would require additional site specific modeling to ensure that the decision has a sufficient basis (Option 3). Based on the review, the licensee the licensee chooses Option 1 (and specifically b above), and considers the following in determining what type of information to collect: a) Reviews the paramet,er distributions and their rationale as presented in Appendix A.1.2; b) Considers how to medify the parameters to consider site specific information and determine the data needs to modify the parameters. This would involve review of Appendix A.1.2 which provides information regarding the valid ranges for site specific parameter changes that a license could propose without an additional uncertainty analysis and for which the licensee would need little supporting information to defend changes. This is important in evaluating tha relative worth of collecting additional data on these parameters under Step 9 of the decision i framework. Table 2.2.1 Example Options Definition Table Expectation Effect on Dose Action Source is believed to be Simulated dose expected to Collect field data to better lower concentration than decrease as concentrations characterize source currently modeled decrease distribution Soil type is expected to be Simulated dose expected to Collect literature and soil predominantly clay and decrease as availability of map data to defend consequently have higher radionuclides to the receptor alternative soil type / texture Kds is decreased Enough soilis expected to Actual available mass of Remediation by soil removal + be permanently removed to contaminant decreases, decrease source - hence simulated dose would concentrations so dose level decrease is acceptable Controls are expected to Restrictions willlimit uses for Set land use restrictions and remain in place for the site while controls are in apply for restricted release duration of the compliance place to limit exposure time period (if contcols fail, and pathways to individual; simulated doses are between simulated dose will decrease 25 mrem and 100 mrem) A b Draft NUREG 1549 94 February 2,1998
p Sho 9 AnajysionLQpilong. U To evaluate the likelihood of success, an analysis of the potential outcome (consequence analysis) will need to be performed for each of the options. Depending on the option, this consequence analysis could be very simple (e.g., the option is complete remediation and the consequence is effectively restoring the syttem to an acceptable condition) to as complicated as refining and expanding the dose assessment. The cost and time necessary to complete each option would also need to be estimated. The consequence analysis should also address the uncertainty associated with each potential outcome. The desired endpoint is a determination of the likelihood or probability that employing a given option will result in meeting the criteria of 10 CFR 20, Subpart E. The result of the activ. ties performed under Step 9 lo a logically organized list of options, and the corresponding cost, likelihood of site release (probability of success), and other important considerations given that the option is pursued. Table 2.2.2 contains examples of how the options could be organized. In some cases, the decision regarding the preferred option will be obvious; for example, a low cost of success and failure, high probability of success option will always be selected over a high cost, low probability of success option. However, the preferred option will not always be obvious, and additional analysis may be needed for sites attempting to balance complex issues. Table 2.2.2. Example Options Analysis Table Alternative Action Cost (if Cost (if Prebability Required Outcome
- successful) unsuccessful) of Success Collect field data to botter $$ $$ medium dose less than 25 characterize source mrem distribution Collect literature dsta to S $ medium dose less than 25 defend alternative soll mrem type / texture Remediation by soil $$$ $$$ high dose less than 25 removal mrem Set land use restrictions dose w/ controls less and apply for restricted than 25 mrem; dose release w/o controls less than 100 mrem
'These assume each option is performed in isolation. If performed in combination with other options, each option on its own would not need to achieve a dose less than 25 mrem To analyze the potential outcome of the selected options, the licensee can use the DandD software to perform some low cost "what if" calculations. For example, they can review the existing information about their source term and try to estimate how it would change based on additional characterization. Based on the quality of the existing information, they may be able
~ V Draft NUREG 1549 95 February 2,1998 l
to modify the source term and obtain a less bounding value This modified source term would l [m then ou input into the model and a revised dose estimate calculated. ' in the same way, the licensee could review site specific or regional data to determine the I predominant soil type at their site. If the soil type is not well characterized by a clean sand, as was used to define the default soil parameters, the licensee could investigate the impact of changing parameters associated with soil type, such as kd. This process can be continued for other model parameters inat the licensee believes could be changed based on site specific ) informat:on. This is similar to performing an informal sensitivity analysis, and will help focus ; atten'. ion to those parameters likely to have the most impact on the calculation of dose. The licensee can then direct resources to reducing the uncertainty in those parameters, or can determine that a different approach is necessary before any higher cost activities, such as soil removal or site surveys, are begun. For this example case, it is assumed that a preliminery evaluation of the remediation option indicates that it is not cost effective to remove the contaminated soil and transport it off site. However, the preliminary analysis is based on the default dose screening and initial bounding estimate of the source term, both of which impact the estimated soil volume requinng remediation, and the cost of remediation. These estimates will change as more site specific data is obtained, which may make remediation a more reasonable option at another point in the decision process At this point in the decision process, the idea is not to permanently eliminate optioris from further consideration, but rather to select the optimum approach for the current state of knowledge. ( y/ This step in the decision framework should support an evaluation of the cost and time impacts of both success and failure. Generally, low cost / high likelihood of success options, or combinations of options, are preferred. This step should also include ALARA considerations, in terms of cost / benefit calculations as well as qualitative considerations. With regard to costs, the licensee should consider that if the option (s) selected are successful, the license will be released and further costs will be minimized. However,if the selected option (s) are unsuccessful, it may be necessary to perform additional characterization or remediation, or there may need to be an evaluation of restricted use (with its associated costs). Sito 10 - Selent Preferred Ootion Based on the DandD analysis and cost estimates for this example, the licensee decides choose Option 1 and specifically to: a) perform additional characterization of the source term, with the expectation that this will result in the source term estimate being reduced. b) use the additional characterization that will also involve obtaining data on the site soil type to support revision of the default kd. The combination of these two actions should have a medium cost and a high likelihood of success.
/3 Draft NUREG 1549 96 February 2,1998
SitD 11
- Impltment prefRRid opilon The licensee; a) develops a characterization plan that will support both radiological and soil data requirements, b) obtains regional soil maps c) performs a radiological site survey. If the licensee has a very high expectation that the additionalinformation will be sufficient to support a revised dose assessment that is less than or equal to 25 mrem, it may be wodhwhile to design the site survey so that it can be used an a final site survey. However, it is important to note that the final site survey has more extensive requirements than may be needed if the site requires remediation.
The extra cost of a final site survey should be weighed against the need to repeat the survey at a later time. Step 12 Revise Model Assumotions; In this example, the licensee revises the parameter values associated with soil type (kd) and I source term are modified based on the site data. To support the future request for license termination, the site survey results, soil maps, and methods used to revise Kd are carefully l documented, l Reiteration of Sleo 4 Iteration 2 Dose Asstalment The revised source term and parameter values are used in iteration 2 of the dose assessment in step 4. In this example, the licensee decides to leave the original default model assumptions and pathways unchanged, and continues to use the DandD software. In this example, when the revised parameter values are input into the model, the result is a dose less equal to 25 mrem /y. Relioration of steo s Since the dose assessment result is equal to 25 mremly, and the site survey met the minimum requirements for a final release survey, the site can be released. Sleo 6 - ALARA the licensee can move on to consider any remaining Al. ARA requirements. The licensee can document that best practice procedures were applied as part of its operational program. In addition, ALARA was incorporated and documented in the options definition (step 8), analysis of options (step 9), and selection of the preferred option (step 10). Draft NUREG 1549 97 February 2,1998
7_-.._-. l Slep 7_ .RtJaate_gLSlie Based on the above, the license can be terminated and the site released. The licensee submits O- all required forms, including NRC Form 314, and documentation of the decision process, and the site is releast,d for unrestricted use. i i I I f t I ()
% Draft NUREG 1549 98 February 2,1998
Case 3: Uranium Contaminated Soil This example will demonstrate the use of the decision methodology to evaluate compliance with the 25 mremly dose criterion for a site with residual soil contamination consisting of depleted uranium. The fictitious site, for the purposes of this example, has been placed in south central Pennsylvania, in an area that is used for both industrial and agricultural purposes, to support a demonstration of how regional and site specific data can be used to support parameter changes within the dose model. StPJL1 The licensee in this example had processed uranium metals for many years, and several outdoor locations are contaminated from that processing. Although this licensee faces a more complex situation than that described in Cases 1 and 2, they would still follow the same steps described above, at least for the first iteration. As before in step one, they would gather as much existing information as possible about their site, including radionuclides and processes used, quantities and forms of material that might still remain on site, and anything else that would be useful for performing a site dose assessment. Based on the information gathered in step one, the licensee determines that although uranium
- of various isotopic ratios had been used over several years, operational and special purpose surveys have generally indicated that the contaminant in soil is depleted uranium, and is well characterized by the following activity percentages
- 90% Ur",9% U'", and 1% U225 For this example, the licensee is evaluating two separate soil contamination areas. One area, (area A),
is directly adjacent to an existing storage area, and the other, (area B), is a large open area that had contained a large structure. The structure was demolished and removed several years ago, Area A is approximately 10 m', and contains a localized area of highly elevated residual radioactivity; area B is 10,000 m' with contamination expected to be relatively uniform and primarily in the top few inches. Step 2 For the scenario definition and pathway identification in step two, the licensee in this example decides to begin the decision process by using the pre-defined scenarios and pathways in the residential scenario (soil contamination) described in NUREG/CR 5512, Volume 1. Step 3 in step three, the licensee decides to use the existing default parameters for the NUREG/CR-5512 models and to perform the analysis using the DandD software. S129A For step four, the dose assessment, the licensee runs DandD using the source term developed from the information gathered in step one. This source term is the maximum reasonable value that is defensible given the existing data sources. The result of the initial dose assessment is as follows: Area A,20 pCi/g DU,127 mrei,,!y; Area B,9.5 pCilg DU,60.5 mremly. Draft NUREG 1549 99 February 2,1998 4
Meo 5 / Steo 8 V Based on the results of step four, in step five it is c! car that the site does not meet the Subpart E dose criterion of 25 mrem /yr for either area A or B. The licensee therefore proceeds to step eight and begins defining options for meeting the 10 CFR Part 20 requirements for license termination. Note that there are basically three options that the licensee can apply either alone or in combination: Option 1 Activities that reduce uncertainty (information/ data collection). Option 2 - Activities that reduce contamination (remediation), and Option 3 - Activities that reduce exposure (land use restrictions). Table 3.1 i.
- some of the options that a license could consider, including three related to reduction of unewtainty, one related to reducing contamination, and one related to reducing exposure.
As mentioned in the Case 2 discussion, when evaluating activities that reduce uncertainty under Option 1, it is useful to begin by looking at the default parameter values and dose conversion factor datasets used in the NUREG/CR 5512 model and what they represent. The default parameter values for the NUREG/CR 5512 modeling (that have been implemented in DandD) were developed based on probability distributions representing the expected variability across the country. A probabilistic parameter analysis was performed to develop default radionuclide-specific concentrations and which also provided information regarding the valid ranges for site specific parameter changes that a license could propose without an additional uncertainty analysis. Therefore, the licensee needs minimal supporting information to defend changes to the parameter values that are within the limits specified in the parameter analysis. This is important in evaluating the relative worth of collecting additional data on these O parameters under Step 9 of the decision framework, b For example, in evaluating the default parameter values the licensee could look at parameters which impact the watur pathway, and which can easily be modified based on site specific information. For this example, the water pathway parameters listed below were changed since easily obtainable site specific information was availaole. [ Note that, as discussed in Appendix E, these parameters should be modified as a group to avoid introducing inconsistencies into the modelj The associated cost for this activity could, for example, be the cost of accessing USGS and state sponsored sites on the Internet, or the cost of obtaining copies directly from those agencies or the library. This approach of moving away from the reasonably conservative values used in the NUREG/CR 5512 modeling based on site specific information could be used by all sites until the point that further reduction in simulated dose would require model changes. At that point, probability distributions for the new model parameters would have to be developed and defended by the licensee. For example, in evaluating the default dose conversion factor datasets the licensee could investigate the values for uranium and associated chain radionuclides that are used in the model. The dose conversion data set in the modelis taken directly from Federal Guidance Report 11, and is based on International Commission on Radiological Protection (ICRP) Report
- 30. In 1994, the ICRP published report 68, which incorporates updated dosimetric information and modeling that resulted in significant changes to the dose factors for uranium and its associated chain radionuclides. While most licensees should use the ICRP 30 values during operations to avoid conflicts with current reporting requirements under 10 CFR Part 20,
( Draft NUREG-1549 100 February 2,1998
licensees engaged in decommissioning activities may wish to propose the use of mort recent dosimetne information and models to support the best technically defensible approach for estimating the dose from residual radioactivity. Such propossis would not conflict with current reporting requirements for operational facilities. The model output can be adjusted using the updated ingestion and inhalation (1 pm AMAD) CEDE factors in ICRP 68, based on the Table B.1 values to match as closely as possible the assumptions used in 10 CFR Part 20 (i.e. adult male workers). Model Parameters That Will be Modified Using Site Specific laformation Hg Thickness of the unsatu.tated zone The thickness of the unsaturated zone is used in determining radionuclide leach rates from the unsaturated zone to the saturated zone. The default distribution was developed from area-weighted data from observation wells across the U.S. Infermation on H,(also called water table depth)is readily available from state or city governments and the USGS. Data for this parameter are easily available, and licensees using deterministic modeling should use the minimum value (thinest unsaturated zone) applicable to their site. U/ Incestion rate for fish from an on-site cond if the site does not currently support a pond or surface water source (that is or could be impacted by residual contamination from the site during the 1000 year analysis period) that contains edible fish, this parameter should be set to zero. This is equivalent to rietting the pond volume to zero. (Note that, in this case, setting this parameter to zero directly eliminates the aquatic pathway.) If a pond does exist at the site, this parameter should be left at the default o value. LIJg infiltration rate a saturation ratio 3 Infiltration rate is defined as the volume of water per unit area per unit time that percolates deeply beneath the root zone and becomes infiltration. The saturation ratio is the volume of water relative to the volume of the pore space, and also the ratio of the moisture content to the purosity. Both these parameters will vary based on regional climate characteristics and site soil texture. A full discussion of these parameters and their derivation, as well as possible information sources fcr site specific values, is contained in the attached parameter definitions. Because data are easily available, and because it is not possible, a priori, to determine whether high or low values are more conservative, licensees using deterministic modeling should use the best estimate of the median value for their site. IR. Irrigation water aoolication rate This parameter represents the annual average quantity of gwindwater used to irrigate on site agricultural products. It is used, along with the area of land cultivated (A,) to calculate the volume of water removed from the aquifer per year for irrigation. Licensees may propose changes to this parameter based on regional precipitation and regional soil moisture levels and other soil properties, and data that support alternative irrigation rates for certain forage crops or edible foods that may be supported due to prevailing dietary pattems or land use pattems. Because it is not possible, a priori, to determine whether high or low values are more conservative, licensees using deterministic modeling should use the best estimate of the median value for their site, based on a multi-year state specific annual average irrigation rate Draft NUREG-1549 101 February 2,1998
D4&Jd Porosities coil bulk densities. and soil areal density of the surface olow layer Porosity is a measure of the relative pore volume in the soil and is the ratio of the volume of the volds to the total volume. Soil bulk density relates the mcts of dried soil to its total volume (solids and pores together). Soil areal density of the surface plow layer is a measure of the rnass of soit per square meter in the surface layer, with an assumed depth of 15 cm for the DandD model. Porosity varies with soil texture, and distnbutions based on the 12 Soil Conservation Service textural classifications are listed in the attached parameter desenptions, Bulk density can be defined as functionally related to porosity: Bulk density = (1 - porosity)*2.65 Soil areal density is calculated as a conversion of units from bulk density plus the 15 cm depth assumption: Areal density = 150* bulk density or Areal density = 397.5'(1 - porosity). Because it is not possible, a priori, to determine whether high or low values are more conservative, hcensees using deterministic modeling should use the best estimate of the median value for their site, based on the site specific soil texture. As stated above, the options that have been identified in this iteration include three related to i reduction of uncertainty. One option is related to reduction of the estimated source term, one is related to reduction of the me Aled exposure through use of site specific parameter values, and one would update the dose conversion factors. The . arth option listed in Table 3.4 would result in an actual reduction of the quantity of residual radioactivity remaining on the site, if the l final option, reduction of exposure through restricted release, were pursued, the licensee would be required by 10 CFR 20, Subpart E, io demonstrate that unrestricted release was not ALARA. This would require additional site specific modeling to ensure that the decision had a sufficient basis. Table 3,3 Options Definition Table v Expectation dffect on Dose Action Source is believed to be a Simulated dose expected to Collect field data to better lower concentration than decrease ac concentrations characterize source currently modeled decrease distribution Getter estimaka of Simulated dose expected to Collect literature and soit parameter values based on decrease as availability of map data to defend site-specific information will radionuclides to the receptor alternative soil parameter be less restrictive is decret. sed values Updated dosimetry is Simulated dose is expected Collect literature values and expected to reduce the to decrease based on better adjust model output estimated dose per unit characterization of uranium intake dosimetry Enough soilis expected to Actual available mass of Remediation by soil removal be permanently removed to contaminant decreases, decrease source hence simulated dose wculd concentrations so dose level decrease is acceptable Draft NUREG 1549 102 February 2,1998
Table 3,3 Options Definition Table
]
Expectation Effect on Dose Action Controls are expected to Restrictions willlimit uses for Set land use restrictions and remain in place for the site while controls are in apply for restricted release duration of the compliance place to limit exposure time period (if controls fail, and pathways to individual, simulated doses are between simulated dose will decrease 25 inrem and 100 mrem) Sten _0 The licensee now moves to step 9, analysis of options in terms of cost and the likelihood of success. To evaluate the likelihood of success, an analysis of the potential outcome (consequence analysis) will need to b: performed for each of the options. Depending on the option, this censquence analysis could be very simple (e.g., we option is complete remediation and the consequence is a demonstration of complinnee with the 10 CFR 20, Subpart E requirements) to as complicated as refining and expaMing the dose assessment. The cost and time required to complete each option should be estimated. The consequence l analysis should also address the uncertainty associated with each potential outcome, The desired endpoint is a determination of the likelihood or probability that employing a given option will result in meeting the criteria of 10 CFR 20, Subpart E. O
) The result of the activities performed under Step 9 is a logically organized list of options, and I the corresponding cost, likelihood of site release (probability of success), and other important considerations given that the option is pursued. Table 3.4 contains examples of how the options could be organized, in some cases, the decision regarding the preferred option will be obvious, however, this may not be true for certain situations and additional analysis may be required for sites attempting to balance complex issues.
Table 3.4 Options Analysis Table Attemative Action Cost (if cost (if Probability of Required Outcome' successful) unsuccessful) Success Collect field data to better $$ $$ low (A') dose less than 25 characterize source medium (B') mrem distribution Collect literature data to $ $ low (A) dose loss than 25 defend alternative soil medium (B) mrem type / texture Collect literature values S $ medium (A) dose less than 25 and adjust model output medium (B) mrem Draft NUREG-1549 103 February 2,1998
1
- Table 3.4. Options Analysis Table
, Alternauve Action Cost (if Cos*, (if Probability of I Required Outcome' successful) unsuccessful) SuccesJ s Remediation by soil $$$ $$$ high (A) dose less than 25 removal $$$$ $$$$ high (B) mrem Set land use restrictions dose w/ controls less and apply for restricted than 25 mrem; dose released w/o controls less than 1 100 mrem
'These Lssume each option is performed in isolation. If performed in combination with other options, each option on its own would not need to achieve a dose less than 2$ mrem 8 Area A s Area B
- See discussion under Case 2 for an explanation of this option
,o analyze the potential outcome, of the selected options, tne liceosee can use the DandD software to perform some low cost "what if" calculations. For example, they can review the existing information about their source term and try to estimate how it would change based on
] additional characterization. Based on the quality of the existing information, they may be able to modify the source term and obtain a less bounding value. This modified source term would then be input into the model and a revised dose estimate calculated. A in the same way, the licensee could review site specific or regional data to deterrnine the Q predominant soil type at their site, the depth to groundwater, and average precipitation rates, Using this information, the licensee could investigate the impact of changing parameters affecting water pathways. This process can be enntinued for other model parameters that the licensee believes could be changed based on site specific information. This is similar to
; performing an informal sensitivity analysis, and w,il help focus attention to those parameters likely to have the most impact on the calculation of dose. The licensee can then direct resources to reducing the uncertainty in those parameters, or can determine that a different approach is necessary before any higher cost activitics, such as soil removal or site surveys.
- are begun.
For this example case, a preliminary evaluation of the remediation option indicates that it is not cost effective to remove the contaminated soil and transport it off site for area 8, but is cost effective for area A. This preliminary analysis is based on the initial dose screening and initial bounding estimate of the source term, both of which impact the estimated soit volume requiring remediation, and the cost of remediation. These estimates will change as more site specific data is obtained, which may make remediation a more reasonable option for area B at another point in the decommissioning process. At this point in the decision process, the idea is not to permanently eliminate options from further consideration, but rather to select the optiT,um approach for the current state of knowledge. Step 9 in the decision framework should support an evaluation of the cost and time impacts of both success and failure. Assuming all options meet the regulatory requirements, in general, (O) Draft NUREG 1549 104 February 2,1998
low cost / high likelihood of success options, or combinations of optionrr are preferred This [^N step should also include ALARA considerations, in terms vf cost / benefit calculations as well as
! ) qualitative considerations With re?ard to costs, the licensee should consider that if the option (s) selected are succest'% : license will be released and further costs will be minimized. However, if the selecwd option (s) are unsuchsful,it may be necessarf to perform additional characterization or remediation, or there may need to be an evaluation of r6stricted use (with its associated costs).
Sten 10 Once the various options have been evaluated, the preferred option can be selected in step 10. Based on the DandD analysis, quality of the survey data available for area A, and cost estimates, the licensee decides to remediate area A. This involves removal of a relatively Onall volume of soil that has been well characterized, and is expected to result in the area easily meeting the unrestricted release criterion. The decision to remediate in this case is based primari'y on information specific to the licensee's %siness practices and plans related to the future use of area A. For area B, the licensee dectues to perfolm additional charactert' tion to obtain dma on the site soil type to support revision of the parameters asswiated with soils and groundwater. The dose model results will also be modified by the dose factors obtained from ICRP 38. The combination of these options should have a medium cost and a high likelihood of success. At this stage in the analysis, unrestricted release is preferred, and therefore restricted release not considered further at this time. Step 11 p ( Under step 11, the preferred option is implemented. The contaminated soilin area A is
\
removed and $ posed of off site. Following the remediation, a final survey is performed and documented, and a revised source term for area A is developed from the survey data. The licensee also develops a characterization plan for area B that supports the soil oata requirements, then obtains regional soil maps and other data associated with the site geology and bdrology. Sten 12 Once the preferred option has been implemented, the model assumptions, parameter values, and pathways (as appropriate) are revised in step 12 of the decision process. For this example, the area A source term is revised and the area B parameter va!ues associhted with soli and groundwater are modified based on the site data and the revised dose factors are obtained. To support the future request for license termination, the site survey results, soil maps, and methods used to revise Kd and dose factors are carefully documented. Table 3.5 lists the paratneters, information sources, and revised model parameter values. p February 2,1998 () Draft NUREG 1549 105
- Table 3.5 Revised Parameters and Supporting Information
(/ Symbol Parameters Discussic]
% Thickness of This exan,%e site is located in the lower Susquehanna river the basin in Cumberland County near Carlisle, Pennsylvania.
unsaturated General information about the lower Susquehanna river basin zone was obtained through two web sites supported by the USGS. Information associated with the National Water-Quality Assessment Program was obtained from http://www.rvares.er.usgs. gov /nawqa/nellsus/Isus factsheet.ht ml. Depth to water information was obtained from http://www.pah2o.er.usgs. gov /gw_rerd/. This site contains monthly information for observation wens in counties within the Susquehanna river basin. Each months data includes the minimum and maximum mean depth to water that has ever been recorded for the entire penod that the well has been monitored. For the C. ,berland county well, data i ave been recorded since 1951. As a first approximation, the licensee uses the minimum value that has ever been recorded for this well of 12.39 feet, or 3.78 meters. U, Ingestion rate This site does not support a pond, and therefore U,is set to 0. for fish from an on-site pond ' i, f,, fa Infiltration rate A sitt loam soil texture was determined to be representative of
& saturation the top 20 cm of soilin the study area, based on information ratios was obtained from the STATSGO data set. Based on Table 1 in the attached parameter discussion for infiltration rate, the mean saturated hydraulic conductivity (K ,)is 9.33E-05 cm/s.
This is equ; valent to an infiltration fraction of about 6%. Infiltration is estimated as follows: I = AR*lF, where AR is the application rate (precipitation plus irrigation) and IF is infiltration fraction. However, the infiltration rate used in the calculations is the lesser of the calculated rate and the saturated hydraulic conductivity, in this case, the calculated value for I is 3.0 inly, compared to a K , of 1.16E3 inly. Therefore, I is 3.0 in/v. 0 \ Draft NUREG-1549 106 February 2,1998
p Table 3.5 RevisaJ Pcrameters and Supporting Information ; U Symbol Parameters I Dicussion IR trrigation water Mean annual precipitation ranges from 38 to 44 inches in the application ' lower Susquehanna river basin (with 41 inches used as the rate best estimate for calculating infiltration). Based on the 1992 Census of Agriculture, the average acre-feet /y of water applied from wells for the Mid-Atlantic water resource area was 0.73. This is equivalent to an irrigydon rate of 1.37 acre-feet per acre, or i 14 L/m 2/d. Irrigation information obtained from the 1992 Census of Agriculture was downloaded from I http://www. census. gov /ftp/ pub / prod /1/agr/92fris/ l i , n,, pi, Porosities, soil Porosity was obtained for the study area from the STATSGO a2, P, bulk densities, data set, and has been set to Q3.1. Bulk density = (1 - and soil areal ,,orosity)*2.65 = 1.30 alem8 Soil areal density = 397.5*(1 - density of the porosity) = 195 kalm 2 surface plow layer DCFs for ICRP 68 dose Since 99% of the dose is from ingestion, the TEDE results from U238, conversion the model are modified by ?e ratio of the ICRP 68 ingestion U235, factors factor to the ICRP 30 ingestion factor. ICRP 30 and ICRP 68 U234 ingestion factors are as follows (Sv/Bq): ( U238: 6.88E-8,4.4E-8 U} U235: 7.19E-8,4.6E-8 U234: 7.66E-8, 4.9E-8 Second Iteration. Steo 4 The revised source term and parameter values are used in iteration 2 of the dose assessment in step 4. In this example, the licensee decides to leave the original default model assumptions and pathways unchangcd, and continues to use m e DandD software. [ Note that in other more complicated situations a licensee might seek to modify these assumptions and pathways. For example, if the groundwater pathway was more corrplex than could be handled by DandD, especially if the licensee needed to account for real transport or needed to better characterize the actual aquifer, a more complex groundwater model could be substituted within DandD. A detailed submittal discussing such changes would need to be developed). When the revised parameter values are input into the model, the result following remediation for area A (for 2 pCilg) is less than 5 mrem /y, and for area B (for 9.5 pCilg) the dose is less than 25 mremly. Second Iteration. Steo 5 & Steo 6 This brings the licensee back to step 5 and the question regarding whether the site can be released. Since the dose assessment result is less than or equal to 25 mrem /y, and the licensee can move on to consider any remaining survey and ALARA requirements. The licensee can document that best practice procedures were applied as part of its operational V Draft NUREG-1549 107 February 2,1998 w
program. ALARA was incorporated and documented in the options definition (step 8), analysis of options (step 9), and selection of the preferred option (step 10). Steal Based on the above, the license can be terminated and the site released. The licensee submits all required forms, including NRC Form 314, and documentation of the decision process, and the site is released for unrestricted use. u D Draft NUREG-1549 108 February 2,1998
O LETTER REPORT REVIEW OF PARAMETER DATA FOR THE NUREGICR-5512 RESIDENTIAL FARMER SCENARIO AND PROBABILITY DISTRIBUTIONS FOR THE DandD PARAMETER ANALYSIS i prepared by O- W. E. Beyeler,T. J. Brown, W. A. Hareland, S. Conrad, N. Olague, D. Brosseau,, E. Kalinina, D. P. Gallegos, and P. A. Davis Environmental Risk and Decision Analysis Department Sandia National Laboratories submitted to M. C, Dahf Office of Nuclear Regulator Research Radiation Protection and Health Effects Branch Letter Report for NRC Project JCN W6227 January 30,1998 O
PARAMETER ANALYSIS FOR THE NUREG/CR 5517 RESIDENTIAL FARMER SCENARIO (O) v
1.0 INTRODUCTION
This draft letter report presents the results of the parameter analysis for defining default dose conversion factors for the residential farmer scenario using the revised NUREG/CR 5512 models. Regulatory decisions on license termination will be based, in part, on the potential dose due residual radioactive contamination. Dose estimates are made using models for the transport and exposure processes that might occur at the site. The default models for the screening analyses are defined in NUREG/CR-5512, Volume 1, and have been implemented with some changes in DandD. Since this analysis is to determine the default dose conversion factors for the average annual dose for the residential farmer scenario, the values of the behavioral and metabolic parameters must be representative of the screening group, while the physical parameter values must represent all potential sites. All the parameters must represent an average annual exposure for the average member of this screening group. The residential farmer scenario is described in NUREG 1549. The screening g" >p is adult niales who live and work on a subsistence farm, producing and consuming a fraction of their diet from the site. They obtain all water required for drinking, domestic and agricultural use from an on-site well. Behavioral parameters represent the characteristics of the screening group and the parameter values chosen for this analysis represent the average behavior within that group. The screening group is used as a surrogate for potential exposures to contaminated surface c.si soil at all sites. Since this includes future sites, the physical parameters and combinations of those physical parameters must reflect the variability across the U.S.. (Y The modeling approach for deriving the default dose conversion factors is discussed in Section
- 2. Sections 3 through 5 present deta!!ed descriptions of the parameters, how they are used in the mooeling, how they influence the modeled dose, a summary of the data review, parameter variability and the probability distribution functions (PDFs) used to represent that variability in the parameters for this analysis. The parameters are classified as behavioral, metabolic or physical. The parameters are grouped in sections by class, behavioral parameters h section 3, metabolic parameters in section 4 and physical potameters in section 5. The results of the parameter analysis, dose modeling and the sensitvity of the results to the parameter variability are presented in section 6. Section 7 is an appendix with equations for the distribution functions.
Oh
\
l 2.0 Modeling Approach 2.1 Purpose The Level 1 (L1) screening dose assessment is designed to allow license termination decisions to be made without requiring site data. L1 dose assessment must therefore be " prudently conservative", meaning that the dose estimate is likely to decrease if more site information was included in the dose calculation. The purpose of this parameter analysis is to identify default dose conversion values using the DandD models and parameters which are prudently conservative or representative of this requirement. Ultimately, this type of analysis may be used to derive default values for the DandD model parameters. 2.2 Definition of prudently conservative parameter values A prudently conservative dose calculation is defined by considering the range of possible site-specific calculations that might be made, By designing the L1 screening calculations so that they tend to overestimate the possible site-specific calculations, the L1 dose assessment provides ? defensible basis for decision-making without site specific mode'5g. The L1 calculations are defined by the parameters values to be used when site-specific data is lacking. A prudently conservative dose calculation depends on prudently conservative parameter values for both the behavioral and physical parameter values. Different procedures were used to establish the values for these two categories of model parameters. For behavioral parameters, prudently conservative values were established by defining a generic screening group for the scenario. This screening group provides a reasonable upper bound on the behavior of the site-specific critical groups that might be defined in consideration of particular site features. Default values for the behavioral parameters represent the behavior of the average member of the screening group (AMSG), and are defined by the average value c' the parameter distribution. For physical parameters, probability distributions are defined which describe the parameter values that might be used if site-specific data werscollected. The range of possible parameter values leads to a range of possible doses resulting from exposure to a specified radionucide concentration. The probability distributions describing the ange of possible paramster values allow decisions to be made in the absence of such values by calculating the range of possible doses that might result if site data were collected and selecting a set of default dose conversion factors from that range of doses. 2.3 Procedure Overview Definition of default dose conversion values requires a characterization of parameter uncertainty, leading to a parameter probability distribution. Although distributions are used to define defaults for both behavioral and physical parameters, distributions for behavioral and physical parameters describe different types of variability, and serve different purposes in the analysis. The procedure used to develop these distributions, and to select default values based on these distributions, is summarized below: Residential Scenario 2-1 February 1,1998 9\!
- 1) For behavioral param;ters, d: fine distributions dcscnbing variability in th;se param;ter m values over individuals in the screening group. These distributions serve two purposes:
) the average values define the default behavioral parameter values, and the range of V values allow the range of doses to individual members of this group to be calculated.
The possible variability in dose to individuals provides assurance that the defined screening group satisfies the requirement of homogeneity
- 2) For physical parameters, define distributions describing variability in parameter values over sites. These distributions also represent uncertainty in the value at a particular site if no site-specific information is available about that parameter.
- 3) For each individual source nuclide, find the distribution of doses that might result from a site specific analysis given the range of possible physical parameter values that might be used in such an analysis, defined in Step 2. In the absence of site data, this distribution allows a prudently conservative screening dose value to be used in making license termination decisions. (This step produces the set of default dose conversion i factors presented in this report.)
- 4) Identify parameter values which, when used with any source nuclide, reproduce, as closely as possible, the screening dose value selected from the distribution. These parameter values provide a way of reproducing the screening dose values for all sources using a single DandD calculation, rather than the multiple calculations required ,
G to derive the dose distribution. (This step will be conducted later.) 2.4 Procedure O b The procedure outlined in Section 2.3 was applied to define default parameter values for the residential scenario. This application entails: classification of the parameters used in the dose - assessment model; review of information pertaining to possible parameter values; definition of probability distributions for each parameter based on this information; derivation of the dose distribution from the distributions defined for the physical parameters; and selection of default parameter values. The application of this procedure to define default values for the residential scenano is desenbed below. 2.4.1 Parameter Classification Different procedures are used to develop default values for behavioral and physical parameters. Any parameter whose value, for a given site and a given group of exposed individuals, would not change if a different group of individuals was considered is classified as a physical parameter. All other parameters are behavioral. Within behavioral parameters, a further distinction is made between parameters describing interaction with the site in the context of the scenario, and metabolic parameters. ICRP 43 paragraph 68 recommends that " metabolic parameters should be chosen to be typical of the age-group . , in the normal population rather than extreme values." Under this recommendation, values for metabolic parameters would not depend on site conditions nor on the composition of the generic screening group. Distributions were not defined for metabolic parameters. In the residential scenario model, the breathing rate parameters V,, V,, and V, were classified as metabolic. Residential Scenario 2-2 February 1,1998
2.4.2 Defining behavioral param ter distributions Distributions for behavioral parameters describe variability over individual members of the screening group, defined as adult male resident farmers (see Section 1.0). Published data from studies or surveys representative of the screening group are desired. This may require that da',a representative of the U.S. population be evaluated or filtered using the definition of the screening group. 1 2.4.3 Defining physical parameter distributions Distributions for physical parameters describe variability in site-specific values over current and future sites. Data sets describing variability over the US are most appropriate for defining these distributions. The following considerations were used in evaluating and selecting data sets from which parameter distributions were developed: Published data from diverse locations, or studies explicitly designed to characterize national variability are preferred; Data representative of the spatial and temporal scales of the model are desired, but are not available in all cases. Data sets used to develop parameter distributions were evaluated in consideratien of differences between the way the data were collected, and the use of the corresponding parameter in the model Probability distributions were defined by fitting to representative data, or by using reported data values to directly define an empirical distribution, if sufficient representative date was identified. For several behavioral parameters, distributional properties of the results of national surveys of a population representative of the screening group are reported in the literature. Physical constraints and causal connections among parameters were used to define correlations and functional relationships where appropriate. 2.4.4 Denving dose distnbutions The distributions defined for the physical parameters describe variability in possible site-specific parameter values over all sites to which the residential scenario model might be applied. At a particular site, if no information is available about the parameter value at the site (i.e. for an L1 calculation), this distribution also describes uncertainty about the parameter value for that site. Uncertainty about the parameter values leads to uncertainty about the dose that would result from a calculation that used site-specific parameter values. The distribution describing this uncertainty was derived by sampling from the defined parameter distributions, and calculating the dose for each combination of sampled parameter values using the residential scenano model. A separate dose distribution was derived for each of the 106 source nuclides with a half-life greater than 65 days. The resulting dose distribution allows a license termination decision to be based on a conservative selection of a dose value from this distribution: for a given dose value, the dose Residential Scenario 2-3 February 1,1998 9
distribution d; fin:s th3 probability that a site-specific dose would be isss than the selecttd value. If this probability is Ngh, then defining site-specific parameter values would be unlikely to modify a decision to terminate a license based on the selected dose value. For each source ( nuclide, a screening dose value can be defined by specifying the probability Pm that a site-specific dose would exceed the screening dose value. The dose distribution functions can be used to identify, for each nuclide, the screening dose value Dm corresponding to the selected probability Pm. 2.4.5. Find default values for the physical parameters The screening dose values Da provide can be used to define a set of prudently conservative default parameter values for the dose model. This procedure is described in detailin NUREG/CR 5512 Volume 3, but the requirements that such parameter values must satisfy can be easily summarized: License termination decisions based on the screening dose values will ! be prudently conservative, in that the probability that the site-specific dose is larger than the screening dose is less than the selected probability level PW similarly, license termination decisions based on a set of default parameter values will be prudently conservative provided those pt 'ameter values lead to a dose greater than or equal to the screering dose value for all source nuclides. 2.4.6 Evaluating the homogeneity of the screening group Using the set of default physical parameters, the distributions of the behavioral parameters were used to calculate the possible variability in dose to individual members of the screening group. This calculation was not used to set default values for the behavioral parameters, but to h insure that the definition of the screening group conforms to the principle of homogeneity as discussed in lCRP 43. Residential Scenario 2-4 February 1,1998 ( l
3,1 Behavirral P;r:m:t:rs with C:nstant Valu:s In this analysis the behavioral pcrameters that do not have significant variability or uncertainty ( for the defined screening group are held constant at the average value for the screening group. Other parameters are held constant by definition of the exposure scenario and specifically the exposure period of one year. Table 3.1.1 Lists the behavioral parameters that are held constant and the value used in the paranieter analysis. DIET is a behavioral parameter that represents the fraction of the diet of an individual at the site that is derived from the intake of home grown agricultural products. Kennedy and Strenge a [1992] define the parameter in Table 6.23 [NUREG/CR 5512, Volume 1) as the Fraction of Diet from Garden; however, the diet fraction pertains to all food products produced on-site for human consumption, including vegetables, fruits, grains, beef, poultry, milk, and eggs. The default value for DIET define in NUREG/CR-5512 is 0.25. As used in the residential scenario model, a single, common value for the DIET parameter is assumed to apply to all food products. This assumption requires, for example, that the fraction of domestically-produced beef in the diet equal the fraction of domestically produced leafy vegetables. This assumption is unlikely to be satisfied in poeral, and is not repree9tative of the screening group consisting of resident farmers. To better reflect the behavior of tne average member of the screening group, who is expected to produce different fractions of each food product domestically, the human consumption rates U, and U,(Section 3.9) are defined as the rate of consumption of food derived from on-site production rather than the rate of consumption in general. With this definition of consumption rates, the DIET parameter value is 1 in all cases. The remainder of the parameters in table 3.1.1 are set at the NUREG/CR 5512 Volume 1 defaults (Kennedy and Strenge,1992). The consumption periods for all foods is set equal to the total time in the exposure period ( 365.25 days) as determined by the assumptions in the screening scenario. No additional information on the holdup periods was gathered, and these are assumed to represent averages for the screening group. It is assumed for the screening analyses that all the animal feed is grown on-site, in contaminated soil, and that all of the animals water is from onsite sources (fraction of contaminated feed and water is 1). Table 3.1.1 Behavioral Parameters with Constant Values Parameter Descnotion Units Value 0.3J._______F_racliggA_ap_n_u_al f djeldppypp_frqg hqmp;g[qwg_fggdg 1 TT_R _ _ _ __T_o_tal tirne_i_n_t_h,e,1_-ye_a_r_ expo _sure period ______ d 365.25=.__ ISO (.1)______F_o_o_d_cgngump.tlo_n_pengd fgr_beeL___ e __ d _365.2_5_,,_ JR4(2) = Fo o _o_d_ggngumpjpi _n_ppggd fgtpoult_ry_____ _ _ _ _ _ . , _ _ d 365.25 JR6L3)...,,,_FpAd_gggggg1Atipaggggd fgt_m_ _ ilk ___________,_ _ _ ___ __ _ d 365.25 IE6@)_____. FpAd_gggsgr_ny.ti p aggggj_fgtpsgs___________________,,_ d 365.25 IEY(.13. ____Fpod_ggngumAtLo_n_pgngd_fgr!eafy_v_e_qetables.._ __ d 365_2_S.__ > IEYl2)_____ Food consumpjto_n_peggd_fgrpther veget_abjgs, __ __ __ _ d _ _365]_5,__ _ JCy(3) ,__ Food consumptongeriod L for fruits d 365.25 IEY@)_____ Food consumptipn,ggrigd fpt,qrain d 365.25 IE..... .__Dyig,k,igggater cons _umptignje_riod _____. ._____ d___ 365]_5.._ TF e Fish consumptiqr),penp_d d 365.25 Residential Scenario 3.1 - 1 January 31,1998 O 1
Table 3.1.1 Behaviorel Parameters with Constant Values O Parameter Description Units ' Value V Jh6l11...._hql.dypyjgd, fgtbpef,,,,__,,_,,,___,,,_____,,_,_ IE6l21.....hql.dypyjgd, fgtpgy[tgL,,,,,,,,, d d 20 ,,, 1
.T.H.A.(31...._H.o.l.d.uPE.r.io.d..fo.r. m.il.k... ...._ _._ _. ..... . _ _ _ _ _ _ ..._ _ _d.._ ....1. ....
THVlij Holdup.2eriod for leafg vegetables d 1 d 14 THV(2) HolduPEriod for otherve.aetables ..
.T.H_V_l31...._H_o_l_d.uPP.r.io.d. .fo.r.f.r.u.i t.s..........................._ _.d___.. . 14.....
d 14
.T.H.V.l4l...._Ho.l.d.uPPr.io.d..fo.r.9 rains
.T.T.G.... T_otal
=_ t_im_.e.i.n.Qa.r.d.e.n.i.n9.P_e_r_iod...................... ..d.._____9.0.....
.X..F111...... F.r.a.ct.io.n.o.f.c.o.n.ta_m.ina.te.d
. _ . . . _b_e_e_f c.a.tt.le..fo.ra1e_........___....._....1.....
.X.F12l_..... F_r_a_cbo..n.o.f.c.o.n.taminated 99.u.itE.fo.r_ age ,_,,,,,,,,,,_,,,_,, ....1.
,X,_F13},,, ___FJgggg_o.f.c_on,tamjggMd,mjik gqw.fp.n!ge - 1 f
X_FL41_,,,,,,,FJacjggA.coatgmjggled [alefAe!1 qr, f age. - 1
.X.G.l1l....___F_r_a_ct_io.n - -
- = of
- - contaminat.e.d..b.ee.f.ca..ttle_9_rai.n............__..._...__1_....
1
.X.G_l21... F.r.a.ct.io__n_o_f.c.o.n.ta..m.in.a.te.d . 99.u_itE9.re.i.n__........___.._.... .
.X.G.l31. ... F.r.a.ct.io.n.o.f
. conta.m.m.
. ._ ate _d_m.i.lk.c.o.w..Er_a.in....... ..... ... _...__1
.X..G_l4l... F_r_a.ct.io.n
.- of con.taminated..la1e_r.h.e.n.9.ra.i.n..__......____...._....1.....
.X.H.l1l....._F_r_a_ct_ ion of c.o.n.taminated
_ .----- - -- beef c_a_tt_le..h.aY..___ ..... . .___..1.__..
.X.H.l2l__ .._F.r.a.ct.io.n.o_f.c.o.n.ta.m.inat.e.d.99.u_itE_h_aY
. . .. ....___ ......._..___... 1 -
, .X.H.l3L.... Fra.ction.o_f_c_o_n_ta.m.ina_te.d__m_
.. _ .. _ .. .__........1.__..
ilk.c_o_w h_a1__..__... l XEl41....,_F,ractjgg,o_f co,n,tamjgate_d,lalerAen, hay,____,,,,, - 1 l
.X.W.{11.._.._.F.r.a.ct.io.n..o_f_c_o.n.ta.m.inate_d_.b.ee.f
. .._ . . ........._._..._-....1.....
.ca.tt.le..w_a.te.r M21______F_racpon_o.f,co,ntaminated n poultry,w_ater ______,,_,,,,_,__,___,j,,,,,
i l ,
.X.W.L3l_.... F.r.act.io.n..o.f.c.o.n.ta.m_ina.te_d..m.
. . . ilk.c.o.w..w.at.e.r....
....... ...._....1.....
t XW(4) Fraction of contaminated laver hen water - 1 V
References:
Kennedy, Jr., W.E., and D.L. Strenge,1992. " Residual Radioactive Contamination from Decommissioning: Technical Basis for Translating Contamination Levels to Annual Total Effective Dose Equivalent," NUREG/CR 5512, U.S. Nuclear Regulatory Commission, Washington, DC. Residential Scenario 3.12 January 31,1998
.. . . . _ . . .. l
3.2 Area cfland cultivat:d, A,(m')(behavisral) 3.2.1 Parameter Description A,is defined in the residential scenario as the area of land that is used for the production of agricultural products for both human and animal consumption. The default value for this parameter defined in NUREG/CR 5512 is 2500 m 2 The cultivated area is the area required to support that portion of the resident farmer's diet that derives from on-site production. Both food crops consumed directly by the resident farmer, and feed for animals raised by the farmer are produced on the cultivated area. As a behavioral parameter, the default value for cultivated area reflects the domestic crop production, and therefore the domestically-produced food consumption rates, for the average member of the screening group. A distributen for A,is not defined. Instead, A,is treated in this analysis as a function of the agricultural pathway parameters describing human consumption, animal consumption, and crop yields. The functional connection between these parameters and the cultivated area is described in this section. 3.2.2 Use of Parameter in Modeling A,is used to calculate the infiltration volume through the cultivated farmland area, V,,, The relationship between A, and V,,is described in NUREG/CR-SS12 Volume 1 (page 5.68) by the following equation: V,, = 1 A,1000
- 1 (3.2.1) where I is the infiltration rate (m/y), A, is the area of land under cultivation (m2 ),1000 is a unit conversion factor (Lim3), and 1 is the time period for infiltration and irrigation (y). In the parameter analysis, the cultivated area is also used to calculate the volume of water used for irrigation:
V,,, = IR A,1 (3.2.2) based on the specified annual average irrigation rate IR (L/m2 d). As discussed in NUREG/CR-1549, the definition of the area to which a receptor is exposed is closely related to the definition of the source concentration. Concentrations at a site generally vary in space, however a single value is used in the default dose model and may be used in other dose models. The appropriate source concentration for calculating dose due to exposure along a particular pathway is the average concentration, over the scenario exposure period, to which the receptor is exposed via the pathway under consideration. To properly reflect the actual spatial variability of concentrations over a site, the specified concentration should be the largest average concentration, over the a:aa to which the receptor is exposed, which is also consistent with available site data. For agricultural pathways, the " exposed" area is the area on which produce and animal feed are Residential Scenario 3.2-1 January 29,1998 O
grown for domestic consumption, A,. Th2 minimum cultivat;cs araa is that area r; quired to support the specified consur.1ption rates of an individual resident. This minimum required area is functionally related to other parameters of the agricultural pathways model, as described in O) ( the following section. 3.2.3 Area Required to Support Specified Consumption The area required to support the specified domestic consumption of the resident, A,, is given by: N. N, y A,=OlET [u +e1[d v.s , Y, (3.2.3) where: DIET' is the fraction of the resident's diet derived from domestic produce; N, is 'he number of food crops consider. 'n the diet; N, is the number of animal products considered i'1 the diet; U, is the ingestion rate of food crop type v by an individual (kg wet weightly); U, is the ingestion rate of animal product type a by an individual (amount /y); Y, is the crop yield for food crop iype v (kg wet-weight /m2y); Y,* is the animal product yield for animal product type a (amount /mty) The units of the animal product ingestion rates U, and the animal product yields Y,* may be CN different for different animal products, but must be consistent. In NUREGICR-5512, U,is i Q specified as kg wet-weight /y for meat, poultry, and eggs, and as L/y for milk. The animal product yield Y,* is the amount of consumable animal product produced through cultivation of 1 m2 of animal feed, and can be defined in terms of the yield and requirements of an individual animal: Y Y = 2 A ,, (3.2.4) where: Y,i is the annual product yield from an individual animal (amountly); A,i is the area required to supply the domestically-produced portion of 2 an individual animal's diet (m ) The cultivated area required to support an individual animal is related ;o the animal's consumption rate and the effective yield for the feed crops in the animal's diet: 1 As discussed in Section 3 4. the DIET parameter is assumed to be 1, and the human consumption rates u, and U, reflect the ingestion rates of domsshc produce in each food category Residential Scenario 3.2-2 January 29,1998 J
1 365.2500a , (3.2.5) n.s Y, u where: N, is the number of animal feed crops in the animal's diet; ' Q,, is the consumption rate of feed crop type k by animal type a (kg wet wt/d); Yo, is the effective crop yield for feed crop type k (kg wet-wt/m2/y); x,, is the fraction of feed crop type k consisting of domestic production in the diet of animal type a; 3.2.4 Parameters used to Calculate A, Equations 3.2.3 through 3.2.5 relate the model parameter A, to other agricultural parameters used in the residential scenario model. Two additional parameters, which are not required in the default dose model, are required to calculated A,: the individual animal product yields Y., i and the effective crop yields Y.. The individual animal product yields. Y,i , were assigned using data from the US Department of Agricultura. Annual data from the latest complete reported year were used in each case. Per-animal yields for beef were estimated from two data sets. The average dressed weight of federally inspected cattle in 1993 was 315 kg (USDA 1998a). Total red-meat yield from beef in 1993 was 10.4 billion kg (USDA 1998b) and the totai number of cattle slaughtered under federal inspection in that year was 33.3 million head, giving an average yield per head of 313 kg. The estimated values are consistent, and a median value of 314 kg per animal was assigned. The age at which beef cattle are slaughtered varies with the breed and with short term economic factors such as current beef and feed price, but is typically between 1 and 2 years (Bob Pate, oral communication). A representative age of 18 months gives an annual yield of 209 kglyear per head. The average per animal yield for poultry was estimated from the total net ready-to-cook production from young chickens of 11.3 million kg in 1995 (USDA 1998c) , and the total number of young chickens slauchtered in 1995,7.37 million (USDA 1998d), giving an average yield of 1.53 kg per chicken. Chickens are assumed to be no older than one year at slaughter. Average annual milk production per cow, using data described in the USDA source as coming from "22 major states", was 16,333 lbs in 1994 (USDA 1998e). Assuming a density equal to water, the average volume production was 7415 L per cow. The reporteu average production of table eggs in 1994 was 260.6 eggs per layer (USDA 1998f). The individual product yields for beef, poultry, milk, and eggs are summarized in Table 3.2.1 Residential Scenario 3.2-3 January 29,1998
~
9
Table 3.2.1 - Annual Animal Product Yields per Animal for the Four Animal Product Types Considered in the Residential Scenario Model v Animal Product Individual Animal Data Source . Type Product Yleid Beef 209 kgly 1994 average dressed weight; assumed age at slaughter of 18 months Poultry 1.53 kg// 1995 young chicken ready-to-cook production and number slaughtered Milk 7414 L/y 1994 average milk production; assumed density of 1 kg/L Eggs 260.6 eggs /y 1994 average table-egg production per layer hen The effective crop yield Ya, is the mass of consumau.s feed produced per unit cu.uvated area. For hay and fresh forage, this yield is assumed to be identical to the standing biomass yield. The standing biomass yield, Yu is a required parameter for the residential scenario model in NUREG/CR 5512 Vol.1 (see Sections 3.60 and 3.62). For grain, the effective crop yield was estimated from crop production figures for 1996 reported by the USDA (USDA 1997). The effective yield for " grain" was estimated from the reported average yield for three primary components of feed grain corn, sorghum, and oats. D\ Y *fcomYeom *IsommmYsomum *fontsY ws eonun (3.2.6)- where f is the fractional area planted with each grain type, and Y is the net feed yield per area for each grain type. Table 3.2.2 summarizes the fractional area and yield based on the reported national totals for 1996, giving ar. effective yield for grain of 0.73 kg wet wt/m2 , Table 3.2.2, Area Fractions and Net Yields for Feed Grains in 1996 Feed Grain Crop Fraction of Area Growing Yield (kg wet-wt/m2 ) Feed Grains Corn 0.834 0.798 Sorghum 0.136 0.424 Oats 0.030 0.207 The remaining parameters are required input for the residential scenario dose model. Table 3.2.3 summarizes the residential scerario model parameters used to calculate cultivated area, and the report sections defining values or distributions for these parameters. ( Residential Scenario 3.2-4 January 29,1998 v
l Table 3.2.3. Partm;t:rs of the Residential Sc; nano Model used to Calculate Cultivat::d Area i Parameter Description Section tv' umber DIET fraction of the resident's diet derived from domestic produce 3.1 U, ingestion rate of food crops 3.9 U, ingestion rate of animal products 3.9 Y, food crop yields 5.5 Q,, animal feed consumption rates 5.6 Y, i feed crop yields for hay and fresh forage 5.5 x,, fraction of domestically produced feed in animal diets 5.1 l 3.2,5 Alternative Parameter Values Equation 3.2.3 provides a cultivated area that is consistent with the consumption patterns of the receptor specified by the parameters of the agricultural pathway model. For the screening calculations, these parameters describe the average member of the screening group, and the default cultivated area is the corresponding area required to support their consumption. Site conditions may set physical limits on the area that can be cultivated: this I.mit in turn implies limits on one or more of the parameters Jescribing the agricultural pathway. The cultivated area may be modified to conform to site specific area restrictions by modifying these parameters. s Alternatively, the licensee may define a site-specific critical group. The behavioral parameters for the agricultural pathway model may be different for the average member of this group than for the AMSG, leading to a revised value of A, consistent with the behavior of the critical group.
REFERENCES:
USDA,1996. Agricultural Fact Book 1996. Office of Communications, U.S. Department of Agriculture NASS,1997. National Agricultural Statistics Service, U.S. Department of Agriculture http://www.usda. gov / news / pubs /factbook http://usda.mannlib. cornell.edu/ data sets / food /89015/ http://mann77.mannlib. cornell.edu/ reports /nassr/ Pate, Robert : Bernalillo County Cooperative Extension Service, OrrJ Communication, Jan 15, 1997. U S Department of Agriculture, file crop _ production _ annual _ summary _01.10.97 accessed through URL http:/.'mann77.mannlib. cornell.edu/ reports /nassr/ field /pcp-bban/ , Sept 22, Residential Scenario 3.2-5 January 29,1998 9
1997. O - U S Department of Agriculture, file fi_ctidw dta accessed t'arough URL http://www.mannlib. cornell.edu:80/usda/ data sets /Jvestock, Jan 15,1998. U S Department of Agriculture, file cm_bf mt.dta accessed through URL http://www.mannlib. cornell.edu:80/usda/ data sets / livestock, Jan 15,1998. - U S Department of Agriculture, file cm_ctlhd.dta accessed through URL http://www.mannlib. cornell.edu:80/usda/ data-sets / livestock, Jan 15,1998. U S Department of Agriculture, file table 073.wk1 accessed through URL http://www.mannlib.comell.edu:80/usda/ data-sets / livestock, Jan 15,1998. U S Department of Agriculture, file table 069.wk1 accessed through URL http://www.mannlib.comell.edu:80/usda/ data-sets / livestock, Jan 15,1998. U S Department of Agriculture, file dapdpc21.wk1 accessed t. .ough URL http://www.mannlib.comell.edu:80/usda/ data-sets / livestock, Jan 15,1998. U S Department of Agriculture, file m0qbt6ek.wk1 accessed through URL http://www.r:0 t nlib.comell.edu;80/usda/ data-sets / livestock, Jan 15,1998. O b Residential Scenario 3.2-6 January 29,1998
3.3 Exposure period: Indoors, t,, Outdrrs, t,, and Gardening, t, (dly) (beh;vi';rci) 3.3.1 Parameter Description The residential scenario model defines three distinct situations or contexts for potential exposure: indoor exposure, gardening exposure, and exposure- tdoors other than while gardening. These separate contexts are defined due to the distinctive pathways or transport rates that might apply to these situations. During the one year scenario period, the average member of the screening group (AMSG) is assumed to divide their on-site time among these three contexts. The three exposure periods ti, tx, and tg are behavioral parameters which specify the number of 24-hour days per year the AMSG spends indoors, outdoors (other then gardening), and gardening. The default values defined in NUREG/CR-5512, Volume 1 for the times spent indoors, outdoors, and gardeni'ig are 200 d/y,70.83 d/y, and 4.17 d/y, respectively. No reference is provided for these values. Default time allocations in RESRAD are based on-the assumption that 50% of a person's time is spent indoors, and 25% is spent outdoors in the contaminated area. The exposure periods are behavioral parameters. For the screersing calculat.. .is, the values for these parameters reflect the average member of the screening group, which consists of resident farmers. An estimate of the variability of exposure periods among individuals in this ' group is also required, to evaluate the homogeneity of the screening group. Current information on human activity patterns was reviewed to establish screening values for these parameters. Values representative of the screening group, consisting of adult resident farmers, were selected from this literature. For each of the three contexts, the average of these valuas is proposed as defining the behavior of the AMSG. A distribution was also identified to describe the potential variability in exposure time among individual members of the screening group. 3.3.2 Use of Parameter in Modeling The time allocation factors are used to calculate doses due to direct exposure and inhalation, as discussed in the following section. The rate of exposure differs in each environment due to differences in the physical characteristics of the environment (reflected in the shielding factors, dust loadings, and resuspension factors) and differences M behavior (reflected in environment-specific breathing rates). Within each erivironment, dose from each pathway varies linearly with the time spent in that environment. These parameters describe the time that the individual spends in various activities and are used to calculate extemal dose from exposure to radionuclide i in soils, DEXRj, and inhalation committed effective dose equivalent, DHRj, from exposure to radionuclide i during residential activity. Dose from external exposure is calculated as (see NUREG/CR-5512, page 5.53) DEXR, = [24(t/t y) SFO C, Ec,,uo S(A,g, ty} DFERJ
+ [24(t,/ty) SFO C Eo u,> S(A,g, t,} DFER)
+ (t,/t,) SFl C, En.u3 S{A,q, t,} DFERJ (3.3.1)
Residential Scenario 3.3-1 January 31,1998 O
l where DREF,is the external dose rate factor for radionuclide j for exposure to contami iation uniformly distributed in the top 15 cm of residential soil (mrem /h per pCilg), A,g is the l n' concentration factor for radionuclide j in soil at the beginning of the current annual exposure l V) ( period per initial unit concentration of parent radionuclide i in soil at time of site release (pCilg per pCi/g), tyis the gardening period (90 days per year), C., corresponds to the concentration of parent radionuclide i in soil at time of site release (pCilg dry weipt" soil), SFl and SFO are shielding factors by which external dose rate is reduced during periods of 1) indoor residence and 2) outdoor residence and gardening, respectively, J,is the number of explicit members of the decay chain for parent radionuclide I, S{A,y,t,}i is the time-integral operator used to develop the concentration time integral of radionuclide j for exposure over a 1 year period por unit initial concentration of parent radionuclide i in soil (pCi*d/g per pCi/g dry-weight soil), S{A,g,t,,} is the time-integral operator used to develop the concentration time integral of radionuclide j for exposure over one Cardening season during 1-year period per unit initial concentration of parent radionuclide iin soil (pCi*d/g per pCi/g dry weight soil), to is the time during the gardening period that the individual spends outdoors gardening (d for a year of residential scenario), t, and t, are time in the 1 year exposure period that the individual spends indoors and outdoors, other than gardening (d for a year of residential scenario), respectively,i t,is '4 total time in tt.J residential exposure period (d), and 24 is a un? conversion fac.or (h/d). Inhalation dose is given by (see NUREG/CR 5512, page 5.55): DHR, = (24V,(t,/t,,) CDG C, Eo un S(A,g,t i,}DFH,)
+ (24V,(t,/ty) CDO C., Ec.uo S{A,,,t,}DFH,)
+ [24V,(t/t,)
i (CDI + P RF,) Ec.ua S{A,y,t,}DFH,] (3.3.2) where V,, V,, and V, correspond to volumetric breathing rates for time spent gardening, indoors, and outdoors, respectively (m8 /h), tgis the time during the gardening period that the individual V spends outdoors gardening (d for a year of residential scenario), t, and t, are time in the 1-year exposure period that the individual spends indoors and outdoors, other than gardening (d for a year of residential scenario), respectively, t,is the total time in the residential exposure period (d), CDI and CDO are dust loading factors for indoor and outdoor exposure periods, respectively, (g/m8), CDG is the dust loading factor for gardening activities (g/m 8), C., corresponds to the concentration of parent radionuclide iin soil at time of site release (pCilg dry weight soil), J,is the number of explicit members of the aecay chain for parent raoionuctioe i, S{A,g,t,} is a time-irdegral operator used to develop the concentration time integral of radionuclide j for exposure over a 1-year period per unit initial concentration of parent radionuclide I in soil (pCi*d/g per pCilg dry-weight soil), S{A,y,t,} t is a time-integral operator used to develop the concentration time integral of radionuclide j for exposure over one gardening season during 1-year period per unit initial concentration of parent radionuclide i in soil (pCi*d/g per pCilg dry-weight soil), DFH, is the inhalation committed effective dose equivalent factor for radionuclide j for exposure to contaminated air (in units of mrem per pCiinhaled), Po is the indoor dust-loading on floors (g/m2 ), and RF,is the indoor resuspension factor (md ). 3.3.3 Information Reviewed to Define Expost e Periods The literature review conducted to support the Draft EPA Exposure Factors Handbook (EPA 1996) was adopted as the most current compilation of relevant literature. This document contains a review and summary of current time allocation studies, along with detailed results A i 1 Residential Scenario 3.3-2 January 31,1998 (J
from ssiscted studies. Time allocations are reported for a variety of activities and environmsnts. All revicwed studies minimally provide mean time allocations over the individuals surveyed. Defining ranges or distributions for the time allocation parameters of the residential scenario model, however, requires information on the variability of time allocation among individuals. In addition, time allocation data is required for the three environments considered in the residential scenario. Among the time allocation studies identified in the literature review, three primary sources were considered for the time allocation estimates in the three residential contexts. These sources are summarized below. Tsang and Klepeis [1996) is "the largest and most current human activity pattern survey available"[ EPA,1996). Over 9000 respondents provided minute by-minute 24-hour diaries between October 1992 and September 1994, and the responses weighted to produce results representative of the U. S. Population. Percentile values are reported for the distributions of time spent in a wide variety of activities for " doers" of those activities. These values describe the variability of day-to-day time allocation, and therefore cannot be used directly as estimates of annual average values. Among the activities and environments considered, reported values for " Minutes Spent Working ir ' Garden or Other Circumstances Working with Soil" [ EPA 1996 Table 14-60), " Minutes Spent at Home in the Yard or vther Areas Outside the House"[ EPA 1996 Table 14-118), and " Minutes Spent Indoors in a Residence (All Rooms)" [ EPA 1996 Table 14129) were used to estimate average values for the critical group of resident farmers, as well as distributions for individual members of this group, as described in Section 3.3.3. ] Hill [1985) also reports on individual variability in time allocation among a variety of activities. Data were collected in four waves, one per season, in 1975 and 1976. Weekly average values, and standard deviations of those weekly averages, are reported for various age and gender cohorts. Unlike other activity pattem studies (exemplified by Tsang and Klepeis) which provide data on daily time allocation, Hill's study provides information on the variability of longer term averages for individuals. Although the study period was also quite short in Hill [1985), observation periods were distributed throughout the year. The results of this study therefore appear to be the best basis for estimating the variability of annual average activity patterns among individuals. Hill provides time allocation information for a number of specific activities that are typically conducted at residences, including meal preparation and cleanup, indoor cleaning. washing / dressing, and reading Data on total time spent indoors, however, is not provided. While the mean value for time spent indoors, ,'ar examp;e, can be est;...ated from the mean values reported for activities typically conducted indoors, the variability in total indoor time among individuals cannot be estimated from the repcrted data without information on (or assumptions about) the correlation of time allocation among these component activities. Similarly, the time spent in a variety of outdoor activities is reported, however the total time spent outdoors at the residence is not. Among the outdoor activities, data on time spent in
" gardening / pet care"[ Hill 1985 Table 7.A.1) was considered in defining the distribution for tg , as discussed in Section 3.3.3.
Robinson and Thomas (1991) compare data from the 1987-1998 California Air Resources Board (CARB) time activity study and from a 1985 national study American's Use of Time. Reported values from the national study were assumed to be more representative of the screening group because of the b oader geographical basis. Time allocation data are reported for a number of activities, locations, and micro-environments. For each of these categories, Residential Scenario 3.3-3 January 31,1998 9
data ara sun mariz:d by th] av; rage timo spent, the standard crror of this av; rage, the average value for " doers", and the percentage of " doers"in the total sample. Among activities, / % locations, and micro-environments considered in this study, data on time spent outdoors at a () residence [ Robinson and Thomas 1991, Table 9-1) were considered in defining the distribution for t,. Data on time spent indoors are provided for two classifications: time spent in the kitchen, and time spent elsewhere indoors. As in the case cf the oata reported by Hill [1985). the average time spent indoors can be estimated by adding the average values for each classificrtion. Information or assumptions regarding the correlation between time spent in these two locations is required to estimate the variability in the total time spent indoors. Both Tsang and Klepeis (1996) and Hill [1985) report separate t;me allocation data for men and women, as well as aggregate time allocation data. There are signif; cant differences between the gender-specific time allocation values for some environments. For example, Tsang and Klepeis [1996) ieport an average time spent outdoors at the residence of 158 min /d for men, while women were found to spend and average of 115 min /d in the same environment. This difference presumably reflects a specialization of domestic roles which is relevant for characterizing the screening group for the residential scenario Because the screening group i Jefined as resident farr,'ers, data for men, who typically spend r, are time outuoors and gardening, but less time indoors, were used to estimate the three exposure time parameters. 3.3.3 Assumptions and Procedures Used to Derive Time Allocation Distributions A large amount of information on individual tirne allocation is available in the literature, however this information cannot be used to directly assign distributions for the exposure periods. In each of the three key studies discussed in Section 3.3.2, a number of assumptions and ('N inferences are required to derive parameter distributions from the reported data. These ( ,) assumptions and inferences are needed to supplement reported information, and to reconcile differences between the data reported and the model parameter values, in three areas:
- Time allocation values are " measured" over a single 24 hour period, while the model parameters reflect annual average values.
- Tsang and Klepeis [1996) provide detailed distributional information; in both Hill [1985) and Robinson and Thomas [1991), however, variability in time allocation among individuals is only characterized by the sample standard deviation. The form of the distribution is not avaliable from the latter two studies, and must be assumed.
- Robinson and Thomas [1991) do not directly report the standard deviation of time spent by " doers" This information must be derived from their reported values for the average times spent by " doers" and by all respondents, the standard deviation of time spent by all respondents, and the fraction of respondents considered " doers" in each area, the reported variability in time allocation does not directly correspond to the variability in annual average values among individuals in the screening group. The following sections describe the assumptions and procedures used to estimate the parameter distributions from the reported data. The average values over allindividuals can be estimated directly from the reported data. These averages do not depend on the assumptions and procedures which
/)) Residential Scenario 3.3-4 January 31,1998
cr3 r2quir:d to cstimat) the full distribution. 3.3.3.1 Estimating Annual Average Values from Daily Values l The time allocation studies found in the literature review use either diaries or retrospective questionnaires to measure individual's time allocation during a single day. Variabihty in these values represents both variability among individuals, and day-to-day variability of time allocation for a single individual. The time allocation parameters for the residential scenario should describe average behavior of an individual over one year, and the distributions for these parameters should describe variability in this annual average over individuals in the screening group. Because reported distributions generally describe variability of daily time allocation rather than annual average time allocation, they cannot be directly used to assign parameter distributions. Instead, estimating variability of annual average values from the reported distributions of daily values requires information or assumptions on the similarity of an individual's time allocation from one day to the next. The similarity of an individual's time allocations on successive days can be described by an autocorrelation function. Autocorrelation information is not available in the reviewed IL..ature: three alternative assumptions were therefore considered in order to define the effect of uncertainty in the autocorrelation of daily time allocation on the distribution of annual average time allocation. These alternative assumptions lead to alternative distributions for individual time allocation. The average time allocation over allindividuals, as discussed above, does not depend on these assumptions, and can be calculated directly from the reported data. Altemative assumptions will, however, lead to different estimates for variability in dose among members of the screening group. For a sing le individual, the correlation between the time spent in a given environment on one day was assumed to be positively correlated with the time spent on any subsequent day: individuals who report spending a large amount of time gardening on a single day, for example, are assumed to be likely to spend a large amount of time gardening on subsequent days. Given this assumption, the three attemative autocorrelations considered correspond to the two extreme limits on non-negative autocorrelation, and an intermediate degree of autocorrelation, The first case assumes perfect correlation in a single individual's time allocation from one day to the next. In this case, the time spent in each environmeN on each day is identicM to the time spent on any other day in the year. Under this assumption, the distribution of annual average time allocation values is identical to the distribution of daily values. This case produces the Irrgest variability in the estimated annual average values: all variability in the reported daily values is assumed to be due to variations among individuals. The resulting distributions are probably unrealistically broad: this assumption is used to illustrate the upper limit of variability of annual average time allocation values. The second case assumes no correlation in time allocation from one day to the next. In this case, an individual's annual average value for time allocation consists of 365 independent samples from the reported distnbution of daily time allocation. By the central limit theorem, the distribution of annual average values over individuals will be well approximated by a normal distribution, with a mean value equal to the mean daily value, and a variance equal to 1/365'th Residential Scenario 3.3-5 January 31,1998 9
of the vari:nce of the daily valu:s. This c:s3 produc2s th3 smallest variability in the cstimat:d annual average values' all variability in the reported daily values is assumed to be due to
/7
" random" day to-day variations which are the same for all individuals, and no variability is attributed to variations in individual habits. The resulting distributions are generally very narrow, and represent a lower limit on the variabiSty of annual average valu6s.
The third case assumes an intermediate degree of autocorrelation. A single individualis assumed to spend a constant amount of time in each environment for 30 successive days. The time spent in each environment is assumed to be independent from one 30-day period to the next. This assumed autocorrelation is not intended to be a realistic desenption of behavior: a realistic autocorrelation function might be expected to decay gradually with time, rather than to be limited to values of 1 and O. The simple autocorrelation function used in this case was designed to produce a plausible distributien of annual average values representing an intermediate degree of autocorrelation, and to simplify derivation of the distribution of annual average values. For a single individual, the annual average time allocation for each environment consists of the average of 12 independent samples from the reported distribution of daily values. 3.3.3.2 Assumed Distributions for Daily Values Reported by Hill [1985] and Robinson and Thomas [1991] Hill [1985] reports the mean time spent by individuals, and describes the variability among the sample population by the standard deviation; Robinson and Thomas [1991] report the mean, along with other information from which the sample standard deviation can be derived (see Section 3.3.3.3 bclow). No additionalinformation on the form of the distribution is provided in f either study. In each environment, and for any individual, the time spent is physically bounded ("]; i by 0 and 365.25 days / year. Without more specific information on the form of these l distributions, distributions were assigned using the principle of maximum entropy. As stated by Jaynes [1982), this principle requires that "when we make inferences based on incomplete information, we should draw from them that probability distribution that has the maximum entropy permitted by the information we do he.ve." In as much as the form of the exposure time distributions are unknown, the assumption of any specific distnbution is arbitrary, and likely to be wrong. Given this uncertainty, the maximum entropy distribution was juoged the most reasonable choice in that "most infccmation theorists have considereo it obvious that, in some sense, the possible distributions are concentrated strongly near the one of maximum entropy" (Jaynes,1982). Given the mean, standard deviation, and upper and lower limits, the maximum entropy distribution corresponds to a beta distribution. Beta distributions were therefore defined to describe the variability in individual time allocation based on these four pieces of information. 3.3.3.3 Calculating Standard Deviation in " Doer" Time from Data Reported in Robinson and Thomas [1991] Robinson and Thomas [1991] report the average and standard error for the time spent outdoors over allindividuals in the national survey American's Use of Time. This sample includes both individuals who regularly spend time outdoors (" doers"), as well as those who do not A separate average va'ue is reported for "dcers", as well as the number of individuals in the 7 Residential Scenario 3.3-6 January 31,1998 (Q
overall sample, and th3 fraction of the total sample classifi;d as " doers" individuals wao spend tim) outdoors ar] consid:r:d to be more r: pres 2ntative of th] scrG:ning group how:vsr th variability in time spent by this sub-group is not reported in Robinson and Thomas [1991). This variability can, however, be denved from the information presented. The standard error (SE) is related to the sample standard deviation (S) and the sample size (n) by: SE =ynh (3.3.3) while the sample standard deviation is (for large n): { (I -t)2 2 s S = n (3.3.4)
..i where t, .., the time spent by an individuali. The overall samp<e of size n un be divided into n, "non doer's" of the activity (all of whose time values are zero), and no " doers" with non-zero time vanes. The standard deviation of all time values in Equation 3.3.4 can ther, be expressed as the sum of two terms:
ai
,' aian S =[en n,+np 2
+ [ (f .,)2' (3.3.5)
, .,, . n n ,+ np The standard deviation of the sub-population of" doers"is defined as:
"o S*= (t' -I"f 3.3.6) o [
,-,,.s no which can t'e expressed in terms of the overall standard deviation, and the other quantities reported in Robinson and Thomas [1991), using Equation 3.3.5:
Sn 2 = "z"o(S _,2) + 2t3 (t-to) 2 (3.3.7) D Residential Scenario 3.3-7 January 31,1998 9 1 1
\
3.3.4 Pr:poxd Time Allocati:n Distributi:ns
/]
V Tsang and Klepeis (1996] provide data on daily time allocation for each of the three environments considered la the residential scenario. These data were used to estimate distributions for each of the three time allocation parameters. As discussed in Section 3.3.3.1, three alternative autocorrelation functions were considered to explors the effect of this unknown information on the derived distribution if individual annual average values. Robinson and Thomas [1991) report data on time spent outdoors at residences. Detailed distributional information is not provided however, and a beta distribution was assumed. Like the data from_Tsang and Klepeis [1996), these time allocations are daily values, and three alternative autocorrelation functions were used estimate the distribution of annual average values for t, from this data. For the distributions derived from the daih measurements reported by both Tsang and Klepeis [1996) and Robinson and Thomas [1991), ? ' distribution based on the intermediate degree (30 4 day period) of autocorrelation is recommens ;, although the bounding distributions (as well as other inti mediate distributions) are equally consistent with the data. Hill [1985) reports the average and standaid deviation of time spent gardening. Unlike the two other studies, each single time allocation value is an average of four separate reports from the same individual, taken in four seasonal" waves", As such, these values provide a more direct estimate of the annual average time allocation for each individual. The quality of this estimate is, however, uncertain, as it based on very limited data for each individual. A beta distribution was assumed based on the reported average, reported standard deviation, and the absolute l O physical upper and lower limits of 0 and 365.25 days / year. Note that although the beta l
'Q distribution fitted to the data from Hill [1985) has a theoretical upper limit of 365.25 days, this limit is not practically approached: 98% of the distribution values are less than 20 days.
3.3.4.1 Time SpendIndoors (t) Data describing the variability in daily values of total time spend indoors at a residence, reported by Tsang and Klepeis [1996), were used to define the distribution for t,. Table 3.3.1 reproduces the reported distribution of daily value for men, converted to units of 24-hour days / year. Figure 3.3.1 shows the distributions for indoor time resulting from the three assumed autocorrelation functions considered. There is considerable uncertainty in the distribution of annual average values due to uncertainty in the autocorrelation of daily values, although the bounding cases of no correlation and 365-day correlation can arguably be dismissed as unreasonable; the former shows very little variability in individual behavior around the common mean value of 266 days, while the latter shows nearly 5% of individuals spending less than 8 hours / day (approximately 120 24-hour days / year) indoors. 3.3.4.2 Time Spend Outdoors at the Residence (t) Data describing the variability i' daily values of time spent outdoors at a residence, reported by both Tsang and Klepeis [1996), and by Robinson and Thomas [1991) were considered in defining the distribution for t,. Table 3.3.2 reproduces the didribution reported by Tsang and Residential Scenario 3.3-8 January 31,1998
1 l Klepeis [1996) of daily valu:s for men. converted to units oi 24-nour days /y ar. Data for m:n w;re s;l;ct:d cs mora r;pr;s:ntative of the scrc;ning group. Figure 3.3.2 shows the distributions for outdoor time based on this data, resulting from tne three assumed autocorrelation functions considered . Table 3.3.1 - Distribution of Daily Values of Time Spent Indoors at a Residence (All Rooms)* Sample Size = 4269 Population Characteristic Value [24-hour days / year) Mean 240 Standard Deviation 69.4 Minimum 2.03 Maximum 365.25 Percentile Values: 0 05 137 0.25 190 0.5 228 0.75 294 0.9 342 0.95 363 0.98 365 0.99 365
*from Tsang and Klepeis [1996) cited in EPA [1996) Table 14-129, Data f'or Men I Table 3.3.2 - Distribution of Daily Values of Time Spent Outdoors at a Residence
- Sample Size = 1198 Population Characteristic Value [24 hour daystyear]
Mean 40.2 Standard Deviation 40.6 Minimum 0.3 Maximum 327 Percentile Values: 0.05 2.53 0.25 15.2 0.5 30.4 0.75 50.2 0.9 91.3 0.95 127 0.98 159 0.99 185 g
- from Tsang and Klepeis [1996) cited in EPA [1996) Table 14-118, Data for Men Residential Scenario 3.3-9 January 31,1998 9
1
Av;rega d:ily values reported by Robinson and Thomas [1991), and th3 sampl3 standard deviation derived from the reported standard error, average for " doers", and sample size (see
. Section 3.3.4), were used to define a beta distribution for daily values of outdoor time. Table v 3.3.3 summarizes the parameters of this distribution. Figure 3.3.3 shows the distributions for outdoor time resulting from the three assumed autocorre;ation functions considered.
Using either set of data, there is considerable uncertairity in the distribution of outdoor time due to uncertainty in autocorrelation of daily values. The distribution based on data from Tsang and Klepeis [1996) has a larger mean value (40 24-hour days / year) than the data from Robinson and Thomas [1991) (29 24 hour days / year). The former is recommended as the distribution for t, because of this conservative characteristic, and because the underlying distribution of daily time allocation values is more accurately defined. Table 3.3.3 - Distribution of Daily Values of Time I Spent Outdoors at a Residence
- Cample Size = 2762 Distribution Parameter l Value Reported Parameters Mean (All Subjects) (24-hour days / year) 12 Standard Error (All Subjects) [24-hour 0.8 days / year) l Mean (Doers)[24-hour days / year) 29.2 l Percent Doers 41 Derived Parameters for Doers
( Standard Deviation (Doers) [24-hour 58.3 days / year) Minimum [24-hour days / year] O Maximum [24-hour days / year) 365.25 Alpha 0.17 Beta 1.9
- from Robinson and Thomas [1991) Table 9-1, National Survey Data 3.3.4.3 Time Spent Gardening (t,)
Data describing the variability in daily values of time spent gardening, reported by both Tsang and Klepeis [1996), and by Hill [1985) were considered in defining the distribution for t,. Table 3.3.4 reproduces the distribution reported by Tsang and Klepeis [1996) of daily values for men _ of time spent working in a garden or other circumstances working with soil, converted to units of 24-hour days / year. Data for men were selected as more representative of the screening group. Residential Scenario 3.3-10 January 31,1998
Tcbi) 3.3.4 Distributi:n cf Daily Valu:s Cf Tim] Spent W:rking in a Gard:n cr Oth:r Circumstances Working with Soll* Sample Size = 2125 Population Characteristic Value [24 hour days / year) Mean 2.92 Standard Deviation 9.50 Minimum 0 Maximum 365.25 Percentile Values: 0.05 0 0.25 0 0.5 0 0.75 0.761 0.9 5.07 0.95 12.7 0.98 38.0 0.99 58.3
- from Tsang and Klepeis [1996) cited in EPA (1996) Table 14 60, Data for Men Gardening times reported by Hill [1985) were assumed to approximate annual average values.
A beta distribution for tg was developed directly from the reported mean and standard deviation, and the absolute physicallimits of 0 and 365.25 days / year. Unlike the results of Tsang and Klepeis (1996), reported mean values for men and women are quite similar: the overall average and standard deviation using both genders was therefore used to define the distribution. Table 3.3.5 summarizes the key parameters of this distribution. , Table 3.3,5 Distribution of Annual Values of Tirne Spent Gardening
- Sample Size = 971 Distribution Parameter l Value Reported Parameters Mean [24-hour drys / year) 2.1 Standard Deviation [2ejyur days / year) 5.4 Denved Parameters for Doers Minimum [24-hour days / year] O Maximum [24-hour days / year) 365.25 Alpha 0.17 Beta 29
- from Hill (1985) Tabie 7.A.1, Data for Men and Women Figure 3.3.4 shows the three distributions for gardening time based on the data of Tsang and Klepeis [1996) (using three altemar.ve autocorrelation functions), along with the beta distribution Residential Scenario 3.3-11 January 31,1998 9
based on the mean and standard devi.iion reported by Hill [1985). Although Hill s procedure s yields estimates of annual average time allocation (based on four daily measurernents of the
} same individual, distnbuted throughout the year), the fitted distribution is quite sim,iar to the
() distnbution of daily gardening times reported by Tsang and Klepeis [1996). Note that although the beta dmtnbution fitted to the data from Hill [1985) has a theoretical upper limit of 365 25 days, this !imit is not practically approached- 98% of the distnbution values are less than 20 days. Three considerations favor the distnbution based on Tsang and Klepeis [1996)(assuming a 30 day autocorrelation) our the distribution fitted to Hill [1985): the better definition of the distributional form provided by Tsang and Klepeis [1996); the similanty of the distribution based on Hill [1985) to the distribution of daily values reported by Tsang and Klepois [1996), suggesting that Hill's data are more representative of daily values than annual average values; and the small number of daily measurements on which Hill's annual average estimates are based. As in the case of annual average values for indoor time and outdoor time, there is considerable uncertainty in the distribution of gardening time due to uncedainty in autocorrelation of daily valuee 3.3.5 Summary The National Human Activity Patterns Survey analysis of Tsang and Klepeis [1996) was used to define exposure periods for the average mcmber of the screening group, and to estimate variability in axposure periods among individuals in the screening group. This study was preferred over available alternatives because of the large sample size, the availability of expoiure period data for micro-environments considered in the residential scenario, the availability of data for sub-populations approximating the screening group (i.e.
- doers" of Q gardening), and the availability of distributions of daily individual exposute time values, Mean values and dit,tnbutions for time indoors were developed from data in Robinson and Thomas
[1991), and Hill [1985) was used to estimate mean values and distnbutions for gardening time. These estimates a provided for comparison with the recommended values, but are not recommended for use in the residential scenario because of the lack of detailed distribution data from either study, and the difficulty in estimating exposure times for all three contexts from either stuoy alone. 3.3 5.1 Average &posure Time The exoosure time for the average member of the screening group were directly estimated by oaily time &! location values for men available in the literature. Tsang and Klepeis [1996] report average values for time spent indoors at a residence and outdoors at a residence. This study also provides quantile values for the distnbution of time spent gardening or working with soil. The average value was calculated from this distribution. Robinson and Thomas [1991) report an average for " doers" of time spent outdoors at a residence; Hill [1985) reports an average value for time spent gardening. Table 3.3.6 summarizes these reported average values. The average values given by Tsang and Klepeis [1996) have been adopted due to the large number of samples in the study, the availability of exposure time values for each of the three scenario contexts in a single study, and the availability of distributions of individual values for each context. d' Residential Scenario 3.3 12 January 31,1998
Table 3.3.6. Summary cf Av: rage Exposura Time Valu:s [24 h:ur dayslyear) Parometer Reponed Source Average (24-hour days per year) Indoor Time Q 240 Tsang and Klepeis [1996) Outdoor time (t,) 40.2 Tsang and Klepeia [1996) 29.2 Robinson and Thomas [1991) Gardening,,1;lME (I) 2.92 Tsang and Klepeis [1996) 2.1 Hill [1985) 3.3.5.2 Distribution of Exposure Times The recommended distnbutions for annual average time spern in each residential environinent are shown in Figures 3.3.5 and 3.3.6, and summary properties are listed in Table 3.3.7. Table 3.7.8 list quantile values of these distributions, which were generated by Monte Carlo sampling of the empirical distributions of daily time allocation reported in Tsang and K;epeis (1996)(see Tables 3.3.1,3.3.2, and 3.3.4 above). Each distribution is based on daily values reported in Tsang and Klepeis [1996). Among the three key studies considered, this survey presents the most complete definition of the distribution of daily values, from which the distributions of annual avei ge values were estimated. The estimated distribution of annual average values is based , on an assumed autocorrelat;on of 30 days. The autocorrelation of daily values is uncertain, and the assumed value is intermediate between the limiting values of no correlation between daily values, and perfect correlation between daily values. The spread of the time allocation distributions is sensitive to the assumed autocorrelation, however the mean value over all iMividuals does not depend on this assumption. 3.3.5 3 Correlations Among Exposure Times and Other Parame! cts The time that an ir$vidual spenos in a given context is constrained by the time scent in each of the other two conaxts. Some amount of negative correlation should therefore exist between each pair of time allocation distributions, however the size of this correlation is uncertain. The total time an individual spends on site (i.e. the sum of indoor time, outdoor time, and gardening time) was calculated using two limiting assumptions about this correlation: zero correlation, and a rark correlation coefficient of-0.5 between each pair of time categories. The latter correlation is the largest (negative) common correlation coefficient that still produces a positive-definite covariance matrix for the three time allocation parameters. Figure 3.3.7 shows the distribution of total on site time under these two assumptions. The distribution of total time is somewhat narrower when the component distributions are negatively correlated. For example, the 99'th percentile value for total time on site is 342 days assuming no correlation, but 325 days when a rank correlation coefficient of 0.5 is assumed. Because che distributions for the two limiting correlation assumptions are similar, uncertainty in Residential Scenario 3.3 13 January 31,1998 9
the appropriate corr:lation wbl not have a large influence on the estimated variability of dose over individuals in the screening group. A correlation coefficient of 0.5 is recommended O because it reflects the competition for an individual's time among indoor, outooor, and gardening activities. The amount of time spent gardening is also presumably related to the amount of food produced in the garden, although the magnitude of the correlation between these parameters is unknown. A correlation coefficien. of 1 between the gardening time and food production rate is assumed. Because the calculated dose is an increasing function of both gardening time and ingestion rate for domestic produce, this assumption conservatively bounds the potential variability in dose among members of the screening group. N+.3lther the assumed correlation among exposure times, nor the assumed correlation between gardening time and ingestion rate, affect the estimated mean values for these parameters. Table 3.3.9 lists the assumed rank correlation coefficients among the exposure times and other model parameters. Table 3.3.7 Summary Properties for Proposed Time Allocation Parameter Distributions Distnbution e n operties [24-hour days 4 ear) Parameter Mean Median 1" %ile 99* %IIe Indoer Time (t,) 240 238 189 285 l Outdoor time (t,) 40.2 40.9 20.1 75.8 Gardening TIME (to) 2.92 1.73 9.10 x 10 8 12.0 Table 3.3.8 - Quantile Values for Exposure Period Distributions y Probability t,(d/y) t,(d/y) t,(dfy) 0.00e+00 2.00e 02 1.74e+02 1.68e+01 1.00e-03 3.50e-02 1.74e+02 1.68e+01 1.10e-02 9.49e 02 1.90e+02 2.11e+01 5.10e 02 3.25e 01 2.02e+02 2.48e+01 j 1.01e 01 4.50e 01 2.08e+02 2.79e+01 2.0'e 01 7.20e-01 2.18e+02 3.25e+01 3.01e-01 1.03e+00 2.26e+02 - 3.54e+01 4.01e-01 1.35e+00 2.32e+02 3.83a+01 5.01e-01 1.74e+00 2.38e+02 4.09e+01 6.01e 01 2.56e+00 2.44e+02 4.43e+01 7.01e 01 3.58e+00 2.49e+02 4.80e+01 8.01e-01 5.21e+00 2.55e+02 5.23e+01 9.01e-01 7.07e+00 2.66e+02 5.81e+01 9.51e-01 8.44e+00 2.73e+02 6.34e+01 9.81e-01 1.10e+01 2.80e+02 6.99e+01 9.99e-01 1.67e+01 2.98e+02 8.43e+01 1.00e+00 1.70e+01 3.00e+02 9.00e+01 h v Residential Scenario 3.3 14 January 31,1998
Table 3.3.9 C:rrelati:ns Am:ng Eyposur] Times Parameters Rank Correlation Coefftctent t., t, -05 t,. t, 05 12 . t. 0,5 t,.Uv 1.0 PARAMETER UNCERTAINTY: The proposed distributions desenbing the variability of time allocation factors for individuals in the screening group rests on several assumptions which introduce uncertainty into the proposed distributions: (1) The screening group consists of resident fartners. Data from Tsang and Klepeis [1996) on
- Time Spent Gardenirg or Other Activities Working With Soil *, for the subset of individuals who engage in these activities, was assumed to be representative of this group. Data for time indoors and outdoors at a residence from this study were not available for this subset of the sample sbjects. Exposure periods for the latter two pareme'~s therefore include non-gardeners, and may overestimate the values for the screening group. Because gardening time represents a relatively small proportion of total time, the extent of overestimation would appear to be small.
(2) The majority of reported time allocation values reflect daily values rather than annual average values. The autocorrelation of daily values for individuals is required to estimate annual averages. This function is unknown, however bounding and intermediate approximations can be defined. Uncertainty in tnis function introduces considerable uncertainty in the variability of annual average time al;ocation over individual members of the screening group. The average value for this group does not depend on the assumed correlation. (3) In twn key studies, variability in time allocation is only chaiacterized by a standard deviation. The underlying distnbutions were assumed to follow a beta distribution defined by the reported mean and standard deviation, and by absoluts limiting values of 0 and 365.25 days / year. These limits represent theoretical bounds, and the effective range of the fitted distributions are
.maller than the theoretkal ranges in all cases.
ALTERNATIVE PARAMETER VA; UES: The exposure period parameters are behavioral parameters. Licensees may propose alternative values by defining a site specific critical group, as discussed in NUREG/CR-1549. If this screening group does not grow produce, gardening time (along with ingestion rates of domestic produce, cultivated area, and irrigation rate) for this group would be 0.
REFERENCES:
Kennedy, Jr., W.E., and D.L. Strenge,1992. " Residual Radioactive Contamination from Decommissioning: Technical Basis for Translating Contamination Levels to Annual Total Effective Dose Equivalent, "NUREG/CR 5512, U.S. Nuclear Regulatory Commission, Washington, DC. Residential Scenario 3.3 15 January 31,1998 9
Am:rican industrial H:alth Council (AlHC)(1994) Exposure Factors Sourcebook. AlHC, Washington, D.C. b California Air Resources Board (CARB)(1993) Measurement of Breathing Rate and Volume in Routinely Performed Daily Activities. Human Performance Lab, Contract No. A033 205. June 1993. National Council on Radiation Protection and Measurements (NCRP) (1984), Radiological Assesment: Predicting the Transport, Bioaccumulation, and Uptake by Men of Radionuclides Released to the Environment. Report No. 76. Hill, M. S. (1985) " Patterns of time use"in Time, goods, and well being, F. T. Juster and F. P. Stafford eds. University of Michigan, Survey Research Center, institue for Social Research, pp 133166 Ann Arbor, MI. Robinson, J. P. and J. Thomas (1991) Time spent in activities, locations, and microenvironments: a Califomin National Comparison project report. U. S. Environmental Protection Agency, Environmental Mori aring Systems Laboratory, '.as Vegas, NV, Tsang, A. M. and N. E. Klepois (1996) Results tables from a detallad analysis of the National Human Activity Pattern Survey (NHAPS) response. Draft Report prepared for the U. S. Environmental Protection Agency by Lockheed Martin, Contract No. 68-W6 001 Delivery Order No.13. U.S. Environmental Protection Agency (EPA) (1996), Exposure Factors Handbook, EPA Report No. EPA /600/P 06/002Ba. Draft of August 1996 U.S. Environmental Protection Agency (EPA)(1985), Development of Statistical Distributions or Ranges of Standard Factors used in Exposure Assessments. Washington, D.C.. Office of Health and Environmental Assestment, EPA Report No. EPA 600/8 85-010 Jaynes, E. T.,1982. "On the Rationale of Maximum Entropy Methods", Proceedings of the IEEE, vol. 70, no. 9,939 952. . Residential Scenario 3.3-16 Jar"su '31,1998
1 - - - - . - ~ - - - 09 . . . I
/
Of I os _
- q. ._, _
/. , ;
o, .f l l.,5w w...w ,eI ]_l f [ l
.,-~.m,-...,. , ,
L T.ang one K.t.4 re corrowie ! 04 f~ "
--~.
~ -
l
.4~
. _ _ !/ >
03 - - - . . - - - .- - - . . --_ __ -, - - _ . _ _ . . . - - . . - - . . . - - - . - . . _ _.l' Ot . .. . . . . - ...-. -,_.. . - . . - . - . , ,. , . - . - - . , . , . - _ _ _ . 01 . . - . . -
... ). - ,. . _ ,
/ l c -- >
0 to 100 150 200 22 300 350 400 Tien. Indoors jery) Figure 3.3.1 CDF of annual average time spent indoors based on daily data from Tsang and Klepeis [1996] for three assumed autocorrelations O
.. ; /. t . . ..__.L.-_ . . _ . _ _-
i
.. .I .
t
' ~~
/i *-i , .. . .. . ... ,C . r L--
,, ._ ,e .., .............4.... , _ , _ _
.. l . 1 ._
.. ..l ,: . . .
I 7 .,. . _ _ .
.. _j . .. .
,. J Figure 3.3 2 CDF of annual average time spent outdoors based on daily data from Tsang and Klepeis (1996) for three assumed autocorrelations Residential Scenario 3.3 17 February 1,1998 9
i_ _ e__. ._ es. - , 1 i !!~. .,.i ns,- ; 4 _- -p .- ..
- i. .. _ _ - __ .. . l
_7_._
.i _ . _ _ __ _ _ . - _ ,. _ _ _ . ._ _ - . -
. n m m m m m m a fim. Owneeees twyp Figure 3.3.3 PDF of annual average time spent outdoors based on daih data from Robinson and Thomas [1991) for three agsumed autocorrelations
_e,,,,. ~. -
;q l l / i 00 ..___J----. 7'" CL - .- - - _ - -. -- } - _ ._ - j _ -_ - - + .-- -- ---;---- i
. f,7 ,
I i [] O8 --
# - h- _ / . . . - . . . . _ p . -[ -- - j i
07 -
- - ---_ _ [ . -- . -
.. - - . -. --.4.---- ..-.--.--4 ll ,
4 06 _ . - - -- . -- _ . - j-~ . . _. j- - .--- -j - - . - ----- q l 7 ; . : ! i l Tsang ard Klepos 3654ay corteston
- /. - !
i i i 0$ : Tsang are k)epois. 304ay mrrotation " - - I- - - -- i Tsang and Klepas. no correlaton i !
'_ _f/ i 04
%m _._,_J e.
i i.7.__..-~.._.._ 7- p. -.-- - -- q I /. , 4
! t I
I ! { t I j 0.3 - . L. _ __ i _ _ __. _ _ . . ~ _ - . -._____4__a_- I
._[ __! 4 i l !
i 'I i i 021 ..-/..;i --.----.q-,_.----4.-- --.------------p -- e - 03 Q _.. ._-_. i
-._.. .. m . ._. n . ._ _1_ _._._ _;
0 J l
! ! l 0 2 4 6 8 10 12 14 16 18 20 meo=*res (en Figure 3.3.4 CDF of annual average time spent gardening based on data from Hili (1985) and dalh data from Tsang and Klepois [1996) for three assumed autocorrelations Residential Scenario 3.3 18 February 1,1998
i
;~ _._ ~.7_
./. . .. . .
-o......u.,...
s .. . ~ , u . v, .
," l.
j
. i. . . . . u . n. i j
.. . . . . . . . . a. . . ..
./ I. ; ,
i 4, - p . . -l. -: ,I'<:n ..,. l 4 :.-....
' ! { ;i!
o
.. _ _i-
! i
. . ', j q sI i
, . ql .j . i . _
._ . f.; __ .
j p
- i 1 _L ;
i ,
. apL
>i L
._ i l! J ii:.L -
41 i i J.4, u / ii
..;,i.. , m! !!
.: - ci- 2 _ _ . . _ ,, . _ . . -_ . , ,
t4 _.._ _ l l l 1il j ff I i l ' _ ___I._l J. l/ _l_ _....... , { l .--- - ! -. !!'Ull._ _ _._! ald,.
- s. 9 i
; ;,. t / ,! ii li!
I ! , iliI I j
., i i. u. io..
, .. . i .m Figure 3 3.5 Cumulative Probabikty Functions for Indoor Time (ti). Outdoor Time (tx). and Gardening Time (tg) i _ _. , _ . , ._ .- _
. m ...
r . r ' . - ' i,
! jl'jil,',
i i p ji l'!
. o,
. 4 I , ,
N.,c:i i!:' j j l'! ,
- m. ,
! +! ! ;
o I i i i ! -i; i ,'i;.! i . t jii
- , ..t_t:
i - o .i.. ., v . . n. i
+ s l j ll !l!
.i,-
I
" "' h
-- o . . .. . . , , .. . n e ,
' I** I'*'2"~ b'M$
,.....v.i.,
,r. ,- ,
, ;; l. i;; .
' ' I
! 4 , #
1 it ; , t 4
- !!1 ,
t ,i; 0 001 i i:! .
~!i i ,
is. 0i i 10 00 1000 t .. . i. w Figure 3.3.6 Probabikty Density Functions for Indoor Time (ti). Outdoor Time (tx), and Gardening Time (tg) Residential Scenario 3.3 19 February 1,1998 9
d __. - _ __ _ m . .m... - 5
- d t:xL -i. *. ; .;; ::t urt;td 1 !
. 4_:a
. .t : *d :. u :.::::L;_ ::
- r tris;ct.j , .l.V r:
tot;
* :; n ?.y-r: z : :: wt *1:1..t:fr't *
*t L. -d r : ; f t r _ :.;;1:* 1:211:r m..g-'r% +7i r:tlI-T ^152"-
0* :t r.b*rt:1 1: ::12: ---* 1 it:t xt r}.:t,':t--
- t::3.. :.:.: == +
n _ 27,1 + :. ::
-* 22 u*g:A;:: -;:t:gr t
- : ._ n
- :2 - : :
- :d:r L:tztn:r q-/dz-0.8 :
r :. .:htr r i: }=r x:: ::.:: 2: 3ITf,:r: :::
-+ =
- :m ' ;.. z :d_ : r:
- :. &- r :: :::::e :.: }:trr_t 4 -* - ~ - - - - -- -
- v- -
03 _
;;; 77 ._ : $=7 7 ;-_ _
;r 127 p ar:1, :::::; -
- :::: :=
= ::_ = _ ..= =. ._.=.:-=. ::= =.:_ .. ::_=_= = . .= _ . _=======_==
=
,, t_._ . . . _ .==*=_=_=_
. _l:_. :r : = _
=. = . .==_=. . . . = -.=,=.=.,=. =.=,=.._. =_ _._ = ..
._._ - . . ~. = .L.- ==__=.== . ,=__=.-,
== : =_..__- -== =
- r tc___
os : cr r r ::: xx : r rarmurarc======
- r=====1
=: ra g:
==
24: : :::x ant .t : cr : ::::::_ e == __ __ 04 - 15 E3: E*i5E3:- 3 : EEEEh5.I ,dhE*~51 1 MUENAlO IE E E~5
..a rr : :: : : crrr --
.:r.; ; No Correl. tion _
==== : : : == = === == =- a== = :_ ===
03 - r: 11 :: :: = -
- ::: : . . . c.,rei.ii.n no.wie r* .o e :-
- .L.::: : . g*=2 :::
' ~
.1 : 222: _- ._ - _ . . _ .
cm == 'ir~n:T.r : :}=' w:== =12 : = ::: :. r. +..::: --
- x= ::::::p:. . . _ _- :::
o2 : : : - r : : :: :trrar r tz; 2 22 L.rr.
- :::: r : g:
- r tp p -I>
d::4:g- 3: ::: ::::: : r=== I r: fr. T :r. ::: :r= :::; : :::=
- : : r q -
ry ,_t_Y1 212 .. ::: .=r+; M m; -- t a__ -
- 11:::-
o r-- ~ - + - A- -t+ -
+ -
200 220 240 aso 2:o 300 320 340 seo sto Tim e (dfy)- d Figure 3.3.7 CDF of Total On Site Time (ti + tx + tg) for Two Assumed Rank Correlation Coefficient Values Residential Scenario 3.3 20 February 1,1998
3.4 Indoor shielding fact:r, SFl The indoor shielding factor, SFI, as defined for NUREG/CR 5512. Volume 1 dose modeling, is a measure of the attenuation of gamma radiation by structural materials such as walls, floors, foundations, and support structures in buildings, and is defined as the ratio of equivalent dose behind the shield to that in front of the shield. The model uses a single, constant value for all radionuclides, and for all structural materials. SFl is classified as a behavioral parameter because its value depends on the type of construction of the residence. This report includes a brief discussion of the basis for the default value for SFl recommended in NUREG/CR 5512, Volume 1. Next, information general to the modeling of dose using SFl is presented. This is followed by a recommended method for estimating shielding factors based on a standard exposure model, Finally, indoor shielding factors are derived for a range of gamma energies and a combination of structural materials that are common in residential settings. DEFAULT VALUE USED IN NUREGICR 5512, VOLUME 1 The value of 0.33 for SFl was adopted as the default value in NUREG/CR 5512, Volume 1, and is based on information derived from studies on deposition of radioactive material from atmospheric plumes (Alrich,1978; Kocher,1978; Jensen 1985). The radiation sources considered in these models are fallout radioactivity deposited on roofs, outer walls, and ground surfaces. Although these models can be used to approximate shielding factors for contaminants deposited around and on buildings, they do not account for contaminants under structures, as required in DandD dose modeling. The RESRAD value for this parameter is 0.7. IMPORTANCE TO DOSE: SFl is directly related to dose. For a given concentration of a given radionuclide in soil, dose is proportional to SFl (i.e., the higher the value for SFI, the higher the total annual dose). USE OF PARAMETER IN MODELING: This parameter is used for calculating extemal dose from exposure to radionuclides in soils DEXRi, (mrem for a year of residential scenario) as described by the folbwing (Equation 5 69, page 5 53 in NUREG/CR 5512, Volume 1)- DEXR, = [24(t,/t,,) SFO C., L.,3 S{A,g, t,.1 DFER,)
+ [24(t,/ty ) SFO C., L.tu S{A,,, t,) DFER,)
+ [24(t/t,) SFI C., L.ty S{A,,, t,} DFER,) (3.4.1) where DFER,is the external dose rate factor for radionuclide j for exposure to contamination uniformly distributed in the top 15 cm of residential soil (mrem /h per pCilg); A.,is the concentration factor for radionuclide j in soil at the beginning of the current annual exposure period per initial unit concentration of parent radionuclide lin soil at time of site release (pCilg per pCilg); C.,is the concentration of parent radionuclide iin soil at time of site release (pCilg dry weight soil); SFl and SFO are shielding factors by which external dose rate is reduced during periods of indoor residence and outdoor residence, including gardening; J,is the number of explicit members of the decay chain for parent radionuclide i; S{A,g,t,} is a time integral operator used to develop the concentration time integral of radionuclide j for exposure over a 1-Residential Scenario 3.4 1 January 31,1998 O
year period per unit initial concentration of parent radionuclide iin soil (pCi*d/g per pCilg dry- l Weight soil); S{An.t,)is a time integral operator used to develop the concentration time integral l of radionuclide j for exposure over one gardening season during 1. year period per unit initial l
\
concentration of parent radionuclide iin soil (pCi'd/g per pCilg dry weight soil); t,, t,, and t, are q times in the 1 year exposure period that the individual spends gardening, indoors, and outdoors (excluding gardening); t,is the total time in the residential exposure period (d); and 24 is a unit conversion factor (h/d). The same shielding factor is used for all radionucl; des and is not dependent on the energy of the gamma radiation.
, INFORMATION REVIEWED TO DEFINE SFl References cited in NUREG/CR 5512, Volume 1, and more recent publications on radiation shielding were reviewed to determine if information was available to estimate shielding factors- -
for structures or buildings that were constructed or placed on contaminated land. (Jensen, l 1985) estimated shielding factors for a number of single family and multistory houses using the computer model, DEPSHIELD. [Leung,1992) calculated shielding factors for concrete and
- glass based on equivalent dose build-up factors in materials, and the shieldir- '7ctors were sed for estimating the protection against radioactive plumes. [Gcaf,1991] performed shielding calculations for 12 building types and compared the calculated factors with shielding factors denved from fallout measurements.
- The radiation sources considered in these models and studies are fallout radioactivity deposited <
- on roofs and outer walls of buildings and ground surfaces near buildings. These experimental conditions and models are inadequate for dose modeling in the residential scenario since NUREG/CR 5512 assumes that contaminated soilis also present beneath the structure.
Therefore, additional data and information derived from experimental measurements or appropriate models are needed to define a distribution for the indoor shielding factor. Shielding factors can be estimated for structures built or placed on contaminated soit using a MicroShield 4.20' The model simulates radiation levels inside a structure from external contamination beneath or adjacent to the structure for a wide range of structural materials and, 4 therefore, would approximate the conditions defined in NUREG/CR 5512, Volume 1. The shielding factor is determined from the following: SFl = e*' where p is the attenuation coefficient for the structural material (e.g., wood, concrete, gypsum) and x is the thickness of the material. p varies with energy of the incident gamma radiation and the type and density of the material. 'Other factors, such as source geometry and buildup (i.e., scattering of radiation to the detector), are included in MicroShield 4.208 Attenuation coefficients, buildup factors, and buildup factor coefficients are available from a library of reference data. The spatial distnbution of contaminants in soils, energy range of gamma i radiation, and physical characteristics and compositions of shielding materials are input parameters for MicroShield 4.20*.
- Residential Scenario 3.4-2 January 31,1998
DETER [lNATION OF PDF FOR SFl Estimates of shielding factors were based on the attenuation of external gamma radiation in a wood frame building with wood siding and either a wood or concrete floor. A wood frame structure assembled from common building materials was selected for these calculations because this type of structure would not overestimate or systematically bias the range and distribution of the snielding factor. a) Description of structure The structure used in this modelis a single-story wood frame building (1000 square feet) with a wood or concrete floor. The walls consitt of parallel 2" x 6" studs spaced 16" apart with gypsum wallboard (1/2
- thick) on the internal surface of the wall and external sheathing covered with cedar siding on the outside surface. Fiberglass insulation fills the void volume between the gypsum wallboard and external sheathing. The wood floor is constructed of 1" thick plywood sheathing over parallel 2" x 8" floor joists spaced 16" apart, with fiberglass insulation placed beneath the plywood sheathing and between the parallel floor joists. The thicknes. of the concrete floor was varied at increments to measure the effects of varying thicknesses of concrete on shielding. Gamma activity was calculated for a position at the center of the building at a height of 1 m above the contaminated soil surface as shown in Figurc 3.4.2 The model simulates the level of radiation through the floor and walls of the building from in infinite source uniformly distributed over the top 15 cm of soil and neglects shielding by floor Joists and studs. The input parameters for the model are identified ir' Table 3.4.1.
Table 3.4.1 Input Parameters for MicroShield 4.20 8 Factor Type or Value Remarks Wood floor 1" plywood (0.6 g/cm8) mobile homes, or manufactured houses, have no concrete slab foundation Concrete floor 3.5", 5.25", & 7" thick 3.5" is the minimum thickness for concrete slab allowed by the uniform building code Surface area of floor 1000 square feet Density of concrete 2.309 g/cm 3 Mark's Standard Handbook for Mechanical Engineers Windows, % 20% of total wall area Window thickness 3 mm, density 2.58 g/cm 8 Wall, gypsum %
- sheet rock,2.025 g/cm 8 Residential Scenario 3.4 3 January 31,1998 O
l
I l Table 3.4.1 Input Parameters fer MicroShield 4.205
'P \ Factor Type or Value Remarks Wall, glass fiber density 2 g/cm' Wall, sheathing 1 cm thick, density 0.35 - -
g/cm 3 Wall, external %
- cedar, density 0.35 . .
g/cm 3 Contaminated Soil Infinite slab,15 cm thick Assumed thickness of contaminated soil Gamma activity 0.037 d/sec/cm* d/sec/cm' = pCl/g Energy range 0.03 to 2.25 MeV Energy range established ' :m:
"0Ba (0.029L MeV); *Eu (2.27 MeV) b) Calculation of Shielding Factcts MicroShield 4.20' calculates the effective dose equivalent, EDE (mSv/h) with, and without, shielding. The shielding factor, SFI, is calculated as the ratio of the EDE rate for gamma
,- m. radiation at the center of the structure, EDEs, to the EDE rate for gamma radiation expected if no shielding were present, EDEu:
(U) SFl = EDEs/ECEu EDEs is the sum of the attenuated EDE rates attributed to gamma radiation shielding by the iLor, EDE,, and by the walls, EDEw EDEs.EDF + EDEw The energy range used in MicroShield 4.20' represents the range of energies for radionuclides identified in NUEG/CR 5512, Volume 1. Shielding factors were calculated for discrete gamma energies for wood and concrete floors, and the results are tabulated in Table 3.4.2 and presented in Figure 3.4 2. The range of gamma energies used in the model represents varW,ons across radionuclides and not uncertainty of the energies for single isotopes. The info mation in Table 3.4.2 can be used for estimating shielding factors for specific radionuclides based on their gamma energy spectrum. O
\v/
Residential Scenario 3.4-4 January 31,1998
Table 3.4.2 Shielding Fact:r as a Functi:n cf G:mma Energy Energy, MeV Shleiding Factor Wood 3.5" 5.25" 7" (Pier & beam) concrete concrete concrete 0.03 0.0967 0.00810 0.00810 0.0081 0 0.06 0.608 0 241 0.241 0.241 0.08 0.722 0.380 0.377 0.377 0.10 0.767 0.438 0.432 0.431 0.20 0.807 0.507 0.486 0.479 0.40 0.814 0.517 0.478 0.462 0.80 0.824 0.489 0.425 0.394 1.5 0.845 0.491 0.405 0.359 2.25 . .e7 0.514 0.422 0 369 c) Distnbution for SFI The distribution for SFl describes the variability in shielding factors over individual members of the screening group, which consists of resident f armers, and depends on the structural and material properties of the residence. Alternative assumptions about the residence corresponding to a range of current residential c-nstruction practices were used to define the variability of SFI over members of the screening group. The frequency distribution in Figure 3.4.3 was derived by selecting the maximum shielding factor for each of the four floor types in Tab!e 3.4.2 and assigning equal probabilities to each. PARAMETER UNCERTAINTY: The proposed distribution describing the uncertainty of the shielding factor over members of the screening nroup of resident farmers rests on several assumptions: A wocJ .'rame house was use in the model. This type of construction is typical of current practices, although other assumptions (e.o brick) are also consistent with screening group assumptions.
- Other structural materials that may contribute to shielding, such as stcel reinforcement, wall studs, and floor joists, are not included in the model calculations.
VARIABILITY ACROSS SITES: This parameter is expected to vary from site to site depending on the type and constructiors of buildings or structures.
REFERENCES:
Jensen, P. H.,1985. ' Shielding Factors for Gamma Radiation from Activity Deposited on Stractures and Ground Surfaces", Nuclear Technology 68 29 39. Residential Scenario 3.4 5 January 31,1998 9
U.S. Atomic Energy Commission,1962. The E//ects of Nuclear Weapons, Gamuel Glasstone, ed., Defense Atomic Support Agency of the Department of Defense. Davisson, C. M. and R. D. Evans,1952. Revs. Modern Phys. Vol. 24, p 79, Nuclear Engineering Handbook - U.S. DOE,1995. Housing Characteristics 1995, Report No. DOE /EIA (93), Washington, DC: U.S. Department of Energy Energy Information Administration. Halliday, D ,1950. Introductory Nuclear Physics, John Wiley & Sons, Inc., New York. Lueng, J. K, C.,1992.
- Application of Shielding Factors for Protection Against Gamma Radiations during a Nuclear Accident", IEEE Transactions on Nuclear Science 39(5),1512-l 1518.
I l Graf, O. and A. Bayer,1991. "Assestment of Gamma Fields inside Buildings in an Urban Area Resulting from External Radionuclide Deposition', Nuclear Te nnology,96,50 71. Marks' Standard Handbook for Mechanical Engineers,8* Edition, T Baumeister, E. Avallone, and T Baumeister lit, Eds., McGraw Hill. O
= . = = = = =
Residential Scenario 3.46 January 31,1998 j
1 I t .l x a -!, E ' a o i 3
! E to o
., G 1 3 i m i U-O l Fiberglass ensulation
- Estenor shaathing i
Ntetor pord 1 m tw cutanuW soil
\ .. - .
Gypsum watiboard Wood floor veth fibergfass insulatnyi er cm.7ete slab ! Cedar siding s ' i
\ -
! i $ u i.
?
N N ! O l Cortaminatad serl (15 cm track) i
- I
! i ! l . y- ~ :-; n q u e l I G m s ! E Figure 0.4.1 Cross section of building for calculating 2 gamma activity from contaminants in soil , w ! J' a O to (D ! 9 9 9 , i
l t k j t 1 Figure 3.4.2 Shielding Factor ; 5 a Function of Energy for Three Different Floors in Building , 1 Shielding factor: house (wood frame & siding, concrete foundation on grade), from infinite slab of contaminated , 1.00E+00 . _ _
.p_*0 k , , .. . , , , g ..
EEL F T- fE:q
/i ' r i u ,. M _
)
..~. _ ..
_ _.A _._ _ t 1.00E 01 gg _ . g gg g g. ,
! El .it M. iEE5i! Hide;Ejbri ;
, I, ).;j; p;;_ -. Q
, ; , ,j ,
(-- Y> 1.00E-02 h 4 [ .
}rm-hN{:r[1:;nl1-.M-.d3
- n . ! .. ,,,:
1
- - . . . ;rrt. r - . i . . u ;.. c ..._ ~-+r.a_5. . p. -c,13!!.!.
. .. ., 1 !. .. _ ..ct-
.. as....--....,.,.--_ . . ..,
j
.i il !!p;_ u Lit-i!]ill_i
. i ,i
.aJ4j L
j < l ! .!- L a J Lgl.. _ L .. w{j iiU, i 1.00E-03 4.. -1 .til Q&1 %
! 0.010 0.100 1.000 10.000 t
l l. Energy, MeV
.li _o 3.5 inch foundation + 7 inch foundation M H or P & B i
A 4 1 1 P Residential Scenario 3.48 January 31,1998 w.m e , - - , --w-.
, yww ---r y. T+--T+8 *r+"t- -wWw-*"'~'M* <-+re's--d'+W?e 3 -? 4 d'.-e'M-- P- --T'~'w- - - '***----'-
- - W=
i 5 o 6 t E u. k A $ $ E e B3m d O .E
> 2 M 2 h
% ii 8 m =5 ]
e .g s.e n.
.c O E
I
- m O
h O es E
==== unm=9munu F l
e e o o m ao S S R R W o 7 w l Residential Scenario 3.4 9 January 31,1998 9
3.5 S:ll ing:sti:n transfer rate f:r the resid:ntial scenarl3, GR (g/d)(behayl:ral) 3.5.1 Parameter Description The soit ingestion transfer rate, GR, is a echavioral parameter that represents the average daily intake of soil by the Average Member of the Screening Group (AMSG) for the residential scenario. GR is the quantity of soilingested per day, averaged over the one year duration of the scenario, by inadvertent transfer from hands or other objects that have been in contact with a contaminated surface, such as food, cigarettes, etc., into the mouth. The default value for this parameter defined in NUREG/CR 5512, Volume 1 [ Kennedy and Strenge,1992), is 5x104 g/d. This value was defined based on published reports on soll ingestion studies . Nine references are listed for this data [ National Academy of Sciences, 1980; Lepow,1975; Hawley,1985; Binder,1986; Calabrese,1989; Davis,1990; Calabrese, 1990; Van Wijnen,1990; EPA,1991). Six of these studies focused on soilingestion by children. The screening group consists of adult resident farmers, and soilingestion rates for children are not representative of this group. The range of reported ingestion rates for *% adult wo .iers/ members of the public is 5x104 to 1x104 g/d. l The RESRAD value for the parameter is 1x104 g/d. 3.5.1 Use of Parameter in Modeling As detailed below, the dose from the ingestion pathway is directly proportional to GR. Overall dose will be sensitive to GR for those sources with sigreificant contributions of ingestion dose to total dose. The parameter GR is used to calculate CEDE for intemal ingestion dose, DSR,, resulting from inadvertent ingestion of soil and contaminants on surfaces. The relationship between GR and internal dose due to ingestion is defined in NUREG/CR 5512 Volume 1, (p 5.73) as: DSR, = GR C., L m DFG, S(Aq .t,} (3.5.1) where GR is effectiva transfer rate for ingestion of soil and dust transferred to the mouth (g/d), S{A .t,)is time integral operator used to develop the radionuclide j concentration in soil, over the residential exposure period for a unit initial concentration of parent radionuclide iin soil at the time of site release (pCi*d/g per pCilg for 1 year of residential scenr.rio), J,= number of explicit members of the decay chain for parent radionuclide 1, C ,is concentration of parent radionuclide lin soil at time of site release (pCilg dry weight soil), and DFG,is the ingestion CEDE factor for radionuclide j (mrem per pCi ingested), The resuiting internal dose is directly proportional to the soil ingestion rate. 3.5.2 Review of AdditionalInformation to Define PDF for GR A literature review was conducted to define a distribution for GR describing the variability in ingestion rate among members of the scraening group 'he average value of this distribution defines the ingestion rate for the AMSG. Residential Scenario 3.5 1 January 31,1998
l In general, soiling:stion is thrs in:dv:rt';nt oralintake of soil through a proc ss wh reby soil-l contaminated obj: cts (hands, cigarett:s, food, etc.) are plac d in the mouth. The average i value for the parameter GR represents the annual average quantity of soilingested per day by the AMSG, and the distnbution of this parameter desenbes the variability in annualingestion rate among individuals in the screening group. Most of the published measurements of soil ingestion found in the literature review penain to children. The screening group is defined as adult resident farmers, and the soilingestion rates of children are not representative of this group. NUREG/CR 5512, Volume 1 [ Kennedy and Strenge,1992) summarizes reported soilingestion rates published prior to 1992. These estimates were derived from limited studies on soil ingestion in adults, and postulates about mouthing behavior. Additional information was reviewed to determine if other data or approaches, preferably more recent than those cited in [ Kennedy and Strenge,1992), were available to provide a defensible basis for constructing a PDF for GR for use in the analysis. Additionalinformation reviewed included the Draft EPA Exposures Factors Handbook [1996), and the references cited therein, LaCoy (1987), Calabrese (1990), Gephr* (1004), and Stanek (1997) Soilingestion rates in adults have been estimated by: 1) analysis of selected tracer elements in human diets and comparing the dietary intake of these elements with tracer elements found in feces and urine of adult volunteers; and 2) observation of individual behavior pattern in adults under a range of environmental conditions and activities. Numerous studies on soilingestion have been conducted using a tracer method tBTM) deseloped by Binder (1986) [Stanek,1995; Sedman,1994; Calabrese,1995; Stanek,1997; and others). The following table (Table 3.5.1) and discussion provides a summary of observations and experimental data on soilingestion studies in adults from published sources. l Table 3.5.1 Sollingestion rates in adults Reference Soil Ingestion Rate (g/d) Comments Hawley,1985 6.1 x10 3 Ingestion rates and time activity pattems Calabrese,1987 1x10 8 1x10 ' Based on CDC estimates Krablin,1989 1x102 Arsenic studies, mouthing behavior, time activity patterns Gephart,1994 1x10'81x10 2 Estimate based on mass balance studies of soilingestion in adults Sheppard,1995 2x10-2 Intake of soil from non food sources Stanek,1995 6.4x 10 2 Revised estimate based on the measurement of four tracer elements in adults Residential Scenario 3.5 2 January 31,1998 9
Table 3.5,1. S:llIngesti:n rates in adults Reference Soil Ingestion Rate (g/d) Comments Stanek,1997 1x10 2 Mass balance studies on 10 adults over a period of 28 days Hawley (1985) reported a soilingestion rates of 4.8x10' g/d for outdoor activities and 5 6x10d to 1.1x104 g/d for indoor activities. The highest ingestion rates occurred for outdoor physical activities (e g., yard work, gardening, etc). The ingestion rates for indoor activities ranged over two orders of magnitude and included typical activities such as occupying a typicalliving space and working in uncleaned areas (e g., attic, utility room, garage). Based on an estimated duration for each activity, Hawley calculated an annual average soilingestion rate of 6,1x10-' g/d for an adult in a typical residential setting. Krablin (1989) estimated the soil ingestion rate in adults from urine arsenic epidemiological studies, mouthing behavior, and time activity patterns. He concluded from these studies that adults ingest 1x10-8 g of soil per day, Sheppard (1995) estimated the intake of soil from non food sources in adults based on indoor and outdoor activ: ties and exposure durations. Based on estimates of ext ,sure duration of 300 h/y ar:! a soilingestion rate of 2x102 g/h for gardening activities and an exposure duration of 5000 h/y and a soilingestion rate of 3x104g/h for indoor activit ,s, he calculated an average daily soil ingestion rate of 2x10 g/d. Stanek (1995) reviewed previous work and presented revised estimates on soilingestion in adults. Using data on four tracer elements, they calculated an average soilingestion rate of 6.4x10 g/d. Calabrese (1990) and Stanek (1995 and 1997) estimated soilingestion rates in adults based on O) ('# mass balance studes in which intake rates were estimated from concentrations of several trace elements in ingested foods and medicines, environmental dust and soil, and body excretions (feces and urine). These studies collected data over multiple 1 week per, Js, during which each subject ingested a controlled quantity of soil from their environment. This mass, along with soil mass ingested with food, was subtracted from the estimated mass that was derived from measured tracer elements in feces and urine. Although these studies draw on very limited data, the results are very consistent with previous studies mported in the literature Calabrese (1990) concluded from his evaluation, however, that the tracers used in his study failed to demonstrate adequate detection limits for assessing soil ingestion in adults. Using quantitative data on zirconium tracers from Calabrese (1990), Gephart (1994) estimated soilingestion rates in adults. Their analysis indicated that a soilingestion rate of 1x104 1x10 g/d is a very conservative estimate and recommended this range for purposes of risk assessments. Gephart (1994) derived a distribution of adult soilingestion by Monte Carlo simulation, however this distribution represents the variability in estimated daily ingestion values. These daily estimates were obtained from daily measurements of tracer concentrations in food and waste products. As discussed below, this procedure requires assumptions v . .cn create significant experimental error in the estimated daily rates. Because of the large measurement error, and because the distribution for GR should describe variations in average annual ingestion rate among individuals in the screening group, rather than day to-day p Residential Scenario 3.5 3 January 31,1998 b 1
v riati:ns in ing:stion rate, the distnbution present:d in Gephart (1994) is not appropriate for this analysis. The study by Stanek et al. (1997) included a larger number of subjects than the 1995 study (10 adults as opposed to 6) and incorporated methodological and interpretative improvements based on earlier studies. However, the expenmental approach used by Stanek [1997] relies on a number of idealizing assumptions of questionable validity. The resulting estimates of daily ingestion rate are highly uncertain, and are frequently less than O. For example, they neglected any absorption or metabolism of tracer substances in their studies, and they assumed that the transit time of the tracers in the intestinal tract was constant and consistent for all subjects in the study. The calculated soilingestion rates were predicated on the assumption that the ratio of tracer element to sollin the fecal sample is identical to the ratio of tracer element to sollin the local environment of the subject. As a result, their attempts to distinguish contributions from soil and house dust yielded conflicting results. 3.5.3 Proposed Distribution for GR Although there is very little empirical data representative of the screening gro"- the above
.,cudies provide a semiquantitative estimate of soilingestion rates in adults. According to studies on soil ingestion published between 1975 and 1997, soilingestion rates vary over a range of about 4 orders t.d magnitude. The variations observed in these studies have been cttributed to a number of factors, including the level of loose contaminants in the local environment, the nature of the individuars activity, the observed behavior of individuals in the studies, controls that are imposed, and the exposure time. Based on the data in Table 3.5.1, soilingestion rates range from a minimum of 0 g/d to a maximum of 1x104 g/d with a likely ingestion rate of 5x10 2 g/d. In the absence of a reliable quantitative estimate of variability in long term average rates among adult individuals, a triangular distribution for the parameter GR is recommended. Figure 3.5.1 shows the assigned cumulative distribution function, and Figure 3.5.2 shows the corresponding probability density function, using the minimum, maximum, and mode values cited above. The mean value of this distribution, representing the AMSG, is 5x10' 8
g/d. PARAMETER UNCERTAINTY: The proposed distnbution describing the variability in the soil ingestion rate among members of the screening group is based on several assumptions that contribute to uncertainty in the distribution: Empirical support for this parameter is very limitea. The most recent measurements of soilingestion in adults are subject to wide variability. Soilingestion has been studied in adults in residential settings using selected trace elements. Several assumptions were A ade ir * / experimental measurements: (1) The specific elements selected as tru. s ,or soil ingestion studies are not absorbed or retained in the digestive tract of the adult subjects or undergo any metabolic changes tbst would prevent excretion of the tracer elements. Residential Scenario 3.5-4 January 31,1998 9
(2) Tracer cl:m:nts in the body excret;ons originate exclusiv;ly from foods, medicines, and ingested soils. O (3) The quantity of soilingested is obtained from the ratio of the quantity of tracer excreted to the concentration of tracer in soil, with the assumption that the tracer element concentration is constant and distnbuted uniformly in soil and dust. ALTERNATIVE PARAMETER VALUES: The default parameter value is representative of the average member of the screening group of adult resident farmers. Alternative, site-specific critical groups may lead to a revised value for this parameter. Summary The proposed distnbution for the soit ingestion transfer rate, GR, for the NUREG/CR 5512 residential scenario is presented in Figures 3.5.1 and 3.5.2. This distribution was derived from limited measurements and estimates of soilingestion in adults. The distribution for the soil ingestion transfer rate ranges from 0 to 1x104 g/d with a mean of 5.0 x 104 pIri.
REFERENCES:
Kennedy, Jr., W. E. and D. L. Strenge,1992.
- Residual Radioactive Contamination from Decommission: Technical Basis for Translating Contamination Levels to Annual Total Effect Dose Equivalent," NUREG/CR 5512 Volume 1, U S. Nuclear Regulatory Commission, s Washington, DC.
p s National Academy of Sciences (NAS).1980. Leadin the Human Environment National Academy Press, Washington, D.C. Lepow, M. L., L Bruckman, M. Gillette, S. Varkowitx, R. Robino, and J. Kapish.1975.
" Investigations into Sources of Lead in the Environment of Urban Children." Environ. Res.
10:414-426. Hawley, J. K.1985. ' Assessment of Health Risk from E posure to Contaminated Soil? Risk Analysis 5(4) 289 302 Binder, S., D. Sokal, and D. Maughan.1986. " Estimating Soil ingestion: The Use of Tracer Elements in Estimating the Amount of SoilIngested by Young Children." Archives of Environ. Health 41:341345. Calabrese, E. J., R. M. Barnes, E. J. Stanek lit, H. Pastides, C. E. Gilbert, P. Veneman, X. Wang, A. Lasztity, and P. T. Kostecki.1989. How Much Soil Do Young Children ingest: An Epidemiologic Study." Regulatory Toxicology and Pharmacology 10:123-137. Davis, S., P. Waller, R. Buschbom, J. Ballou, and P. White.1990. " Quantitative Estimates of Soilingestion in Normal Children Between the Ages of 2 and 7 Years: Population Based Estimates Using Aluminum, Silicon and Titanium as Soil Tracer Elements." Arc. Environ. Health 45 (2):112122. (3
; 1 Residential Scenario 3.5 5 January 31,1998 LJ
Calabr:se, E. J., E. J. Stanck, C. E. Gilbert, and R. M. Barnes.1990. " Preliminary A'Jult Soil ingestion Estimates: R:sults of a Pilot Study." R:gulatory Toxicology and Pharmacology 12.88 93 Van Wijnen, J. H., P. Clausing, and B. Brunekreef.1990 " Estimated Soil Ingestion by Children." EnvironmentalResearch 51:147 62. U.S. Environmental Protection Agency (EPA) 1991. Risk Assessment Guidance for Superfund Volume I: Human Health Evaluation Manual Supplemental Guidance: Standard Default Exposure Factors. OSWER Directive 9285.6 03 (March 25,1991)Intenm Final, EPA Office of Emergency and Remedial Response, Washington, D.C. U.S. Environmental Protection Agency (EPA).1996. Exposure Factors Handbook. EPA /600/P 95. Office of Research and Development, Environmental Protection Agency, Washington, DC. (Current draft not citable) LaGoy, P. K.,1987. ' Estimated SoilIngestion Rates for use in Risk Assessment", Rio Anal. 7, 355 35 Sheppard, S. C.,1995.
- Parameter Values to Model the Soil Ingestion Pathway",
Environmental Monitoring and Assessment 34,27 44. , Calabrese, E. J. and E. J. Stanek,1994 ' Soil Ingestion Issues and Recommendations", J. ) Environ. Sci. Health, A29(3),517 530. l Stanek, E. J., E. J. Calabrese, R. Barnes, and P. Pekow,1997.
- Soil Ingestion in Adults:
Results of a Second Pilot Study , Ecotoxicology and Erwironmental Safety 36(3),249 257. Calabrese, E. J. and E. J. Stanek,1995. " Resolving Intertracer inconsistencies in Soil Ingestion Estimation", Environmental Health Perspectives 103(5),454-457. Gephart, L. A., J G. Tell, and L. R. Triemer,1994.
- Exposure Factors Manual", J. Soil Contamination 3(1),47-117.
Residential Scenario 3.5-6 January 31,1998 9
1 . - . . , , I -I l - i i og . . _ . _m m ;_ :- . . . . . _ . 4 i j , 4
. Os. -
- p. - . . . ~4 ;
4 07 L .. . , ; . 1- . . -4 . ! - - - l l 06 G os - l 04 k d - --. ~. L !_. .L.- I- ' 1 { 03 .- ., ~4- - _. ..-- ) d -. . . f . . - ! 02 - -- f- ___-. ( -- -
.... a i
1 01 _ . _ . . __ _ . . - --_-- _ ._-_i 0 I 0 0 01 0 02 0 03 0 04 0 05 0 06 0 07 0 08 0 09 0.1 SoilIngemon Rete (yd) Figure 3.5.1 - Est' mated Cumulative Probability Fun: tion fr GR l l I
'!9,-_._,.._._-,__.. . - . . , . . . _ _ . _ - I
~
I , > > i j i i __.__v' . ,._ yi 8 l ,t 4 t , . 1 ! i i ; , f_-,_4 _ ,._ __4. _' 16 - _ j___.__ l
~ t l . ._ _.. - , . .l , . l - - l _ . . _ _.l - . . .. l _.--
f14 _.-_v_. l 12 ___ _r + __ __ __ _, _ _ .{- . _{ 5 10 - -- !. I --- !_ _ i- 3-- b -- Ld I i , t e
; 8 . ..,_ .........,_ . . _ . _ _ . _ _ _ _ . _ _ . . . _ _ - - ,
i 6 - ..,-. .. . _
.--.-.q.-_..- ;
i i i e i 4 - __
.._..r..- _
.4 __ __
2- _ _ _. i _ ___1___ _ d ._ M. __.__.
-0 0 0 01 0 02 0 03 0 04 0 05 0 06 0 07 0 08 0 09 01 Soll ingestion Rate (g/d) ,
t i i t.__._ . . . . _ . _ . _ _ ~. ._.-- __ _ . _ _ . _ - . . . _ . _ _1 Figure 3.5.2 - Estimated Probability Density Function for GR Residential Scenario 3.5-7 January 31,1998
3,6 Drinking wat:r Ingesti:n rate, U,(lld) Drinking water ingestion rate, U, is the daily average human consumption rate of ground water from a well. The dose model uses a single, constant value for all contaminants. DEFAULT VALUE USED IN NUREGICR 5512, VOLUME 1 The default value for this parameter, as defined in NUREG/CR 5512 Volume 1,is 21/d. There was no justification or explanation for selection of this value. The RESRAD value for the parameter is 1.41/d. IMPORTANCE TO DOSE: Use of contaminated ground water f;r human contumption increases the dose from radionuclides present in ground water. The drinking water ingestion rate is used in calculating the dose due to consumption of cor.taminated ground water ard will depend to 6 large extent on the ages and dietary needs of irdividuals at the site. Therefore, U, is considered a behavioral parameter. USE OF PARAMETER IN MODELING: This pararneter is used in the irrigation and c' "' ting water c.sse model for calculating the ingestion dose from contaminated v.ater and may be used to calculate the volume in the aquifer. The drinking water ingest;on factor, AF,, is determined from the drinking water ingestion rate from the following (Equation 5.75 page 5.59 of NUREG/CR/5512 Volume 1): Afa, = U, DFG, t,(C,/C.,) (Equation 3.6) where U,is the daily intake of dnnking water (1/d), DFG,is the ingestion CEDE factor for radionuclide j (mrem per pCi ingested), tais the duration of water intake periocl (d for 1 year uf residential scenano), and C.,is the average annual concentration of radionuclide j in ground water. 3.6.1 Review of AdditionalInformation for Revised Parameter Value and Distribution for U, The 19771978 Nationwide Food Consumption Survey (NFCS) of the U.S. Department of Agricultum colle.ted information of food and beverage consumption from a random sample of the U.S. populauon, Survey results from 26,081 individua:s were analyzed, and a statistical analysis of the water intake rates were reported [Ershow,1989). [Roseberry,1992) fit lognormal distnbutions to NFCS data and developed distributions for use in public health risk assessments. The justification for applying these data to the screening group (i.e., adult males who garden and obtain drinking water from groundwater sources)is based on the assumption that the screening group would be represented by individuals in the group from 20 to 65 years of age. Although we do not have data specific to adult males or limited just to groups who garden, it is assumed that drinking water intake rates from these large populations would include members of the screening group. , Residential Scenario 3.6 1 January 31,1998 9
r.___.__ . . _ _ _ . _ _ - _ . _ _ o9 _ . - 4 .. L - [ 1.. 4 i; ._.4.&, !.4 es of .;.
.4
{'.)
,.f i
oe ,-1----. . ...u . . . . . , o6 _ - d --.1 -J f -[ d ' i . d- -+M 4 e 'DencR 6612 tef eat j
- o. __ 11 !
q
"'**'" iI os q._-. d. ._ _-
}
os ___ _ . ._u . _ _ y Q {gQj _;_q l
^
ot _ .. . . . .- o ^ o1 1 to we., wa. n n Figure 3.6.1 Proposed Cumulative Distribution Jr Drinking Water , Ingestion Rate (Ud) l 3.6.2 Proposed Distribution for Drinking Water Ingestion, Uw l The distribution for drinking waMr irg estion was determined for adults (20 to 65 years) from data reported by [Roseberry,1994. The intake rates for adults are lognormally distributed. Og The mean and sthndard c'eviation of the naturallog (drinking water intake rate (1/d)) are 0.1152 V and 0.489, respectively, for individuals in the age group from 20 to 65 years. The cumulative distribution for U,is shown in Figure 3.6.1 along with the NUREG/CR 5512 Volume 1, default and Resrad values. The distribution applies to the screening group by assuming that water intake rates in adults 20 65 yeart old are representative of adult male consumption. PARAMCTER UNCCRTAINTY: The distribution for the drinking water ingestion rate was based on a survey of 11,731 adults that were selected randomly nom the U.S. population. The individual survey data represents the average daily consumption of water over a 3-day period. Results from individual participants in the survey could be influenced by activities of individuals during the 3 day survey period and the season of the year. These factors, however, would be expected to balance since the 3-day survey periods were spread over the entire year. Drinking water ingestion rates could be less in females than in males. VARIABILITY ACROSS SITES: This parameter would be expected to vary from site to site due to uncertainty in the activities, dietary habits, and ages of individuals at the sito. Other factors such as the quality of the groundwater could have an influence on the ingestion rate (e.g., use of bottled water for drinking). SITE DATA COLLECTION: The licensee may collect information on water quality at the site and evaluate alternatives for groundwater use based on economic factors. For example, the Residential Scenario 3.6-2 January 31,1998 [ e
-. J
c:st for digging 9n on sit] w:ll may bo grcat:r than the cost for connection to a municipal or a rural wat:r syst:m. Wat:r quality may be very poor, r; quiring pretreatment of water suitable for drinking.
REFERENCES:
Ershow, A. G. and K. P. Cantor,1989. ' Total Water and Tap Water intake in the United States: Population Based Estimates of Quantities and Sources (Life Sciences Research Office. Federation of American Societies for Experimental Biology, Bethesda, May 1989). Roseberry, A. M. and D. E. Burmaster,1992. 'Lognormal Distributions for Water intake by Children and Adults", Risk Analysis 12(1),99 104. I 9i Residential Scenario 3.6-3 January 31,1998 9
3.7 Irrigati:n wat:r applicatirn rate, IR (Um2.d) and Volumn cf wat3r r:msv:d fr:m the aquifer for irrigation use, V (Ud) -(Behavioral) C The irrigation water application rate is the amount of water, from ground water, applied on a daily basis per unit area ofirrigated land. Parameter IR represents a long term average rate of watei application. The irrigation water application rate is used in the residential scenario to estimate the transfer of radionuclides from irrigation water to food crops. Use of contaminated water via irrigation systems deposits radionuclides on plant surfaces or directly on the soil, resulting in resuspension and plant uptake and transfer to edible parts of the plant. The default value for this parameter is 2.08 Um'-d (NUREG/CR 5512, Volume 1) and is based on an annual average irrigation rate of 76 cm/y, which was considered in the original analysis of Volume 1 a representative value sufficient to produce most crops. V,is ther volume of groundwater removed from the aquifer used for irrigation. NUREG/CR-5512, Volume 1, does not define a default value for V, Instead, V,is determined from the irrigation rate, IR, and the area of land cultivated, Ar , by assuming that the area defined by A,is irriga'.ed trom ground water at the site. Since the volume of water for irrigation use is a function of other 5 dependent and dependent parameters, a separate probability detribution function is not defined for V,. IMPORTANCE TO DOSE: The irrigation water application rate, IR, is used in calculating the dose due to consumptice of edible plants that are grown in land that is irrigated with contaminated ground water and the consumption of beef, milk, eggs, and poultry products from l animals that consume forage, hay, and grain crops that are grown on the irrigated land. 'q ( j V,is important in estimating the transport of radionuclides from contaminated irrigation water to soil and to edible plant and animal products in the residential scenario. The V, parameter is V used to calculate the total water volume in the aquifer, along with water volumes for domestic purposes and the surface-water pond. It is also used for deriving the fraction _of pumped water that is applied to the surface layer. The higher the irrigation water application rate, the higher will be the deposition rate of radionuclides to edible plants and soi:, and consequently the higher the dose due to ingestion of contaminated plants by humans and domesticated livestock. The concentration of contaminants in animals wi!! increase due to inges'uon of plant material and soil, and therefore dose to humans will also increase with consumption of animal products (i.e. meat, milk, eggs). USE OF. PARAMETERS IN MODELING: Irriaation Water Acolication Rate. IR The irrigation water application rate, IR, is used in nine different pathways in NUREGICR-5512
- or estimating the transfer of radionuclides from contaminated ground water to edible foods.
The equadons for each of the nine pathways can be found in Section 5.4.1, Food Crops Contaminated by irrigation Water, and Section 5.4.2, Animal Products Contaminated by Irrigation Water, in NUREG/CR-5512, Volume 1, and are summarized in the following: a) Irrigation water-plant-human pathway (Emation 5.22, page 5.27 of NUREGICR-5512, Volume 1) (
\
R.,, = IR r, T,/Y, [C,/C ] (3.7.1) where R,,,is the average deposition rate of radionuclide j to edible carts of plant v from ) application of irrigatir.1 water per unit average concentration of parent radionuclide i in water, IR is the average annual application rate of irrigation water, r,is the fraction of initial deposition (in water) retained on the plant, T,, is the t'anslocation factor for transfer of radionuclides from plant surfaces to edible parts of th o 1t, Y,is the yield of plant v, and C.,, and C., are the average annual concentration of uw clides j and i espectively, in irrigation water over the ! current annual period. l b) irrigation water soil-plant-human pathway (Equation 5.27, page 5.30 of NUREGICR.5512, Volume 1) R = IR/P [C /C ] (3.7.2) where R.,,is the average deposition rate of radionuclide j to soil from irrigation water applied onto the soil during the growing period fcr an average unit concentration of parent radionuclide i in water, anu P, is the areal :.1 density (kg/m^2). c) irrigation water-forage-animal-human pathway (Equation 5.37, page 5.36 of NUREGICR-5512, Volume 1) R,3, = IR r, T,/Y, [C,/C ] (3.'.3) where R,g,is the average deposition rate of parent radionuclide j to forage crop f from the application of irrigation water during the feeding period for an average unit concentration of parent ra sionuclide i in water, r,is the fraction of initial deposition of radionuclides in water retained on the plant, T,, is the translocation factor for trarafer of radionuclides from plant surfaces to edible parts of the plant, ard Y, is the yield of forage crop f. d) irrigation water-soil-forage-animal-human pathway (Equation 5.43, page 5.40 of NUREG/CR 5512. Volume 1) Roy, = IR/P, [C,/C.] (3.7.4) where R.,,,is the average deposition rate of radionuclide j to soil from irrigation water applied onto the soil during the feeding period for an average unit concentration of parent radionuclide i in wat- . e) irrigation water-stored hay-animal-human pathway (Equation 5.48, page 5.41 of NUREGICR-5512, Volume 1) R,n, = IR rn Tn/Yn [C,/C ] (3.7.5) where R,n,is the average deposition rate of radionuclide j to stored hay crop h from irrigation water application for an average unit concentration of parent radionuclide iin water, nr is the fraction of initial deposition of radionuclides in water retained on plant h, Tn is the translocation Residential Scenario 3.7-2 Janu?.ry 31,1998 O
factor for transf;r of radionucl'.d:s from plant surfaces to cdible parts of th3 plant, and Yn is ths yield of stored hay crop h. Q f) irrigation water soil-stored hay animal-human pathway (Equation 5.50, page 5.43 of NUREG/CR 5512, Volume 1) R,yg = IR/P, {C,/C,,) (3.7.6) where R.,a is the average deposition rate of radionuclide j to soil from irrigation water applied onto the soil during the growing period for an average unit concentration of parent radionuclide i in water, and P, is the areal soil density (kg/m^2). g) irrigation water-stored grain-animal-human pathway (Equation F.53, page 5.46 of NUREG/CR 5512, Volume 1) R.,,, = IR r, T,/Y, [C /C ) (3.7.7) where R is thr., average deposition rate of i Hionu .le j to stored grain crop g fam irrigation water application for an average unit concentration oi parent radionuclide i in water, r, is the fraction of initial deposition of radionuclides in water retained on grain plant g, T, is the translocation factor for transfer of radionuclides from plant surfaces to edible parts of grain plant g, and Y,is the yield of stored grain crop g. h) irrigation water-soil-stored grain-animal-human pathway (Equation 5.55, page 5.47 of NUREG/CR 5512, Volumu 1) O) R.,,, = IR/P, [C,/C ) (3.7.8) where R,y,is the average deposition rate of radionuclide j to soil from irrigation water applied onto the soil during the growing period for an average unit concentration of parent radionuclide i in water, and P,is the areal soil density (kg/m^2). i) irrigation water-soll anima!-human pathway (Equation 5.58, page 5.48 of NUREGICR-5512, Volume 1) Rmg:IR/P,[C/C] (3.7.9) where R.,g is the average deposition rate of radionuclide j to soil from irrigation water applied onto the soil during the feeding period for an average unit concentration of parent radionuclide i in water. IR is also used to calculate V,, along with the land area under cultivation, A,(m2 ), as shown in the following equation: < V, = IR*A, (3.7.10) Volume of Water for Irrioation. V, O V Residential Scenario 3.7-3 January 31,1998 J _ _ _ _ _ _ _ .
l The total w:ter voium3 in the aquif;r remains constant during tha yePr of simulation and is used as th3 dilution volum3 in d;t:rmining the average annual contaminant concentration in ground water. The total water volume is taken as the greater of the infiltration water volume or the sum of the water volumes used for irrigation, domestic purposes, and the surface-water pond. Thus, the total volume of water is evaluated as (from equation 5 88, page 5.68 of NUREG/CR-5512. Volume 1): V7, = greater of. V,i or V,,, + Va , + V , (3.7.11) where V7,is the total volume of water in the aquifer for dilution of activity over a 1 year period, Va,is the annual volume of water for domestic water use, and V,,is the volume of water in the surface water pond for growing fish during a 1 year period. The infiltration volume, V,,, is the sum of the annual infiltration (due to precipitation) and irrigation volume (V,,,) added to box 1 (the surface layer of soil). It is calculated (from equation 5.87, page 5.68, NUREG/CR 5512, Volume 1) as follows: V,, = l A,1000 1 (3.7.12) where I is the infiltration rate, A,is the area of land under cultivation,1000 is the area unit conversion factor, and 1 is the annual 1 year time period. Irrigation volume represents recycling of contaminant activity from the aquifer (box 3 of the water use model) to box 1. Note that irrigation is assumed to occur continuously during a year, even during non-growing periods. The fraction of irrigation water applied to the surface layer, F,, was calculated in NUREGICR-5512. Volume 1 (Equation 5.89, page 5.68) as follows: F, = v" (3.7.13) v,+ v,, where V,,,is the volume of water used for irrigation during a 1-year period (L/d) and Va ,is the volume of water used for domest.c purposes dunng a 1-year penod (L/d). Dunng ana!ycis and testing of the original methodology proposed in NUREG/CR-5512, it was found that the ground-water contamination models described in Volume 1 do not adequately account for natural discharge from the aquifer. The result was radionuclide build up in the aquifer box. Therefore, a water balance model was adied to the methodology to correct this problem. These changes are documented in Appendix A of NUREG/CR-5512, Volume 2. In Section A.3.6 of Appendix A (page A.10, equation 5.89m), equation Sfi 10 above was modified to the following: F, - v" (3.7.14) v,7 > Residential Scenario 3.7-4 January 31,1998 9
V,(and F,) repr ssnt the quantity of ground watsr removed for irrigation in the watsr use modst and is used to calculate the ' ate of change of the totbi activity of radionuclide j in box 1. (dCi /dt) as shown in the following (Equation 5.80, page 5.65 of NUREGICR 5512, Volume 1): dC,/dt = F, w, C3 , + N I r .,,.y d, C,, - (N + Lm) Cy (3.7.15) where F,is the fraction of water removed from box 3 in the residential three box water use model that is depraited on the surface layer (bN 1) by irrigation, w, is the removal rate constant for pumping of water from box 3 (dd ), C 3,is the total activity of radionuclide j in box 3 at time t, j is the index of the current chain-member position in the decay chain, n is the index of precursor chain members in the decay chain (n<j), C in is the total activity of the precursor radionuclide n in box 1 'at time t, N is the decay rate constant for decay of radionuclide j (dd), LM is the rate
~ constant for movement of radionuclide j from box 1 to box 2 (dd), and d g is the fraction of transitions of radionuclide n that result in production of radionuclide J.
It would appear from equations 3.7,14 and 3,7,15 that as V,,, increases, F, increases and will tend to increase the concentration in Layer 1. However, if the modeled infiltration rate (and thus V7,) is high, contaminants will be removed (flue ") quickly from Layer 1 into 'he aquifer. Therefore, if total aquifer volume is determined by infiltration water volume (and thus is large compared to removal flows), a high rate of contaminant flushing to the aquifer will occur. The , models were changed in Appendix A of Volume 2 to NUREG/CR 5512 to allow for natural 1 l discharge to prevent rapid buildup of aquifer concentration. At the same time, if outflows from the aquifer (for irrigation, for instance) dominate the determination of aquifer volume, the ground water water balance model was modified (in
- Volume 2 of NUREG/CR 5512) to allow for natural recharge, to prevent the aquifer volume from G decreasing dramatically and maintaining reasonable aquifer contaminant concentration levels.
3.7.1 Information Reviewed to Define the Distribution for IR - I The Farm and Ranch irrigation Survey (1994) [USDC,1994] provides the most recent and complete compilation _of irrigation practices for farms and ranches in the United States. The document contains detailed information_ on irrication, including farm size, total irrigated acres, and rstimateo quarnities of water applied by irrigation for individual states and water resource areas over the continental United States. Table 3.7.1 shows the irrigated land area and the
- quantities of water used for irrigation in twenty seven states. These states accounted for 98.22% of total irrigated land area for farms and ranches from which $1,000 or more of agricultural products were produced or sold. These data provide an estimate of long-term
-(annual) average irrigation rates across a variety of soils, crops : water quality and availability.
.The data may include surface water as well as ground water sources. As such, this data set provides an estimate of the irrigation rate for the screening group.
3.7.2 Proposed PDF for IR - The data from Table 3.7.1 were binned and fit to several distributions and the fitness to each distribution was evaluated with a Kolmogorov-Smirnov test. The data from regionalland areas Residential Scenario 3.7-5 January 31,1998
l (stat;s) w;ra ev:nly w;ight:d in d:v; loping tha distribution. The b:st fit was obtained with a log normal distribution. Distribution param:tsrs w:re p=0.67, o=0.87, and c=0.32. Figure 3.7.1 depicts the proposed probability distnbution for the imgation water application rate. This plot includes the corresponding data from Table 3.7.1 used to generate this fit. Figure 3.7.2 is the proposed cumulative distribution for the irngation water application rate, IR, Table 3.7.1 Irrigation of Farm and Ranch Land in the Conterminous U.S, [USDC,1994) State irrigated Area Water Applied Ave irrigation Rate Ave irrigation Rate (Acres) (Acre-feet /y) (Acre feet per acre) (L/m 2/ day) Arizona 752,019 3,310,159 4.40 3.67 Arkansas 2,853,929 3,196,019 1.12 0.93 California 7,245,487 22,474,499 3.10 2.59 Colorado 2,998,888 5,241,741 1.'5 1.46 Florida 1,416,019 1,922,166 1.36 1.13 Georgia 619,536 325,009 0.52 0.44 Idaho 3,183,733 6,023,644 1,u9 1.58 lilinois 271,725 168,518 0.62 0.52 Kansas 2,501,925 3,336,027 1.33 1.11 Louisiana 820,816 885,335 1.08 0.90 Michigan 305,481 165,843 0.54 0.45 Minnesota 326,781 185,034 0.57 0.47 Mississippi 646,761 684,643 1.06 0.88 Missouri 702,183 513,940 0.73 0.61 Montana 1,936,292 3.057,884 1.58 1.32 Nebraska 5,979,661 5,025.201 0.84 0.70 Nevada 519,507 1,138,138 2.19 1.83 New Mexico 685,695 1,630,390 2.38 1.98 North Dakota 157,426 138,954 0.88 0.74 Oklahoma 474,201 589,076 1.24 1.04 Oregon 1,587,152 2,946,868 1.86 1.55 South Dakota 304,454 302,997 1.00 0.83 Texas 5,100,979 7,605,827 1.49 1.24 Utah 1,085,083 2,412,250 2.22 1.86 Washington 1,434,800 3,125,619 2.18 1.82 Wisconsin 306,096 205,210 0.67 0.56 Wyoming 1,374,447 2,481,740 1.81 1.51 , Mean 1.25 Std Dev 0.735 PARAMETER UNCERTAINTY: The distribution for the irrigation water application rate, IR, was based on annual average irrigation rates throughout the United States. Since most farm and ranch land is irrigated only during the growing season, the data may underestimate the actual Residential Scenario 3.7-6 January 31,1998 9
I daily wat2r irrigation rate for some arsas of the country. The amount of water used for irrigation would be expected to vary from year to year, depending on the quantity of added moisture from
/' rainfall. Abnormallevels of rainfall could bias the survey data and skew the proposed j\ distribution. 4 V,is a dependent parameter derived from two separate and distinct parameters. Each of these parameters are uncertain and that uncertainty is represented by PDF's for those parameters.
VARIABILITY ACROSS SITES: The irrigation rate parameter, IR, would be expc -d to vary from site to site depending on local climatic conditions and seasonal changes at , e "te. For instance, in the arid west, higher values of irrigation would be expected. Whereat portions of the northwest, eastem and southeastern states, and humid coastal areas, rain' ' smounts-may allow valuation of this parameter at low values approaching zero. This can be seen in the 2 data for arid states like Arizona (3.67 Um -d) versus more humid states like Wisonsin (0.56 2 Um -d)in Table 3.7.1. Applican.a may elect to collect data at the site in an attempt to support lim'.s on IR Limiting values may be supported due to regional precipitation and soil moisture levels (as well as evapotranspiration rates, infiltration rates, etc.), regional soil properties, and data that support - al'ernative irrigation rates for certain forage crops or edible foods that may be supported due to p tvailing dietary pattems or land use patterns. V, may vary across sites due to differences in climatic conditions, the magnitude of seasonal changes, crops grown, soil hydraulic properties, ground water quality and quantity, and location and availability of surface water that may also be used for irrigation. Similar to the irrigation Q rate, IR, from which it is derived, the volume of water for irrigation can vary dramatically for sites in the arid west versus sites in more humid climates, such as the northwest, eastern or - southeastern regions, and along certain coastal areas. Its variability will depend upon the - variability associated with IR and the land area under cultivation. Licensees may also attempt to collect information to support alternative values for V,. Licensees may defend new site specific irrigation volumes due to regional precipitation and soil moisture levels, regional soil properties, and data that may support alterna;ive irrigation rates that may be appropriate due to prevailing dietary patterns or land use patterns.
REFERENCES:
[USDC,1994) 1992 Census of Agriculture, AC92-RS-1, Farm and Ranch irrigation Survey (1994), U.S. Department of Commerce, Economics and Statistics Administration, Bureau of the Census. Oo Residential Scenario 3.7-7 January 31,1998 (
5' s C/) 8
!. Probability Density 10 09 S" Data l l MLM Log Normal 07 06 - --
Probability / Frequency 05 - p 04 N 03 -- 02 - - 0.1 - 00 4 - 00 10 2.0 3.0 40 50 60 Irrigation Water Application Rate IT Q
- Figure 3.7.1 Calculated Probability Distribution for Irrigation Water Application Rate io 8
O O O
ll ll l1ll l ' l l ia m e r d g R n t o I e g u e t e t u a L - R
- n i
o
- t a
0 i c
~ 6 l p
p A . r
~ t e
~ a
~
0 5 W
~
n i t
~ t a
~ t e g a i r
r
~ R I 0 r y - 4 n o t
i - i o f s t a n _ n ~ c i o e i l t u D p p b i e r O 0 A t i v 3 r i s t a t e D l u a e v m W i t a u n l C 0 it o u 2 a m g u
/ i r C
//'
t I r d e s o f p r J 0 1 o r
/ P
/ /
6 2 1 7 3 e r
- 0 u 0 6 4 2 0 0 g i
1 0 0 o 0 F
)
( x F wa6 fe.tS. i li = _ W8$ .Q S1E. ll
3.8 Vcluma cf water r:m:v:d fr:m th2 aquif;r per y:ar f:r d:m:stic u::s, Va ,(L) Va,is the annual volume of groundwater removed from the aquifer for doinestic uses. This parameter, along with the annual volume of water used for irrigation, V,,,, is used for determining aquifer volume. Of the total volume for all domestic uses (showers, washing, etc.), a portion of this domestic use is directly ingested as dnnking water or in consumable products made from tt3 drinking water source. Other pathways, such as direct immersion while showering (and other purposes that use this source of contaminated water), are not included in the exposure calculations. In NUREG/CR-5512, Volume 1, the default value for this parameter was set to 91,250 liters. No basis is provided for this value and the variability of this parameter is not discussed in Volume 1. Of this amount, the daily drinking water ingestion rate was set to 2 L/d (this amount is represented by the parameter U,in the models). This rate was suggested by EPA (1989), and represents the 90* percentile daily drinking water ingestion rate as tap water, including uses in cooking and for beverages prepared using tap water (coffee, tea, etc.). It was concluded in Volume 1 that this value represented a conservative value for screening. sMPORTANCE TO DOSE: Va ,is used in estimating the transpon of radionuclides from contaminated ground water to humans in the residential scenario. That is, this parameter, along with the volume used for irrigation, establishes the total volume of the aquifer. USE OF PARAMETER IN MODELING: The total water volume in the aquifer (V7,) rem Ws constant during the year of simulation and is used as the dilution volume in determining the annual average water concentration. In the original analysis of NUREG/CR-5512, Volume 1, the total water volume is taken as the greater of the infiltration water volume or the sum of the water volumes used for irrigation, domestic purposes, and the surface-water pond. However, subsequent analysis by Sandia (discussed below) resulted in remcval of the surface water pond volume in the calculation of total aquifer volume. The contribution to the ingestion dose from the use of contaminated groundwater, DWR,, is l evaluated for drinking water and ingestion of irrigated foods as follows (equation 5.74, page 5.58, NUREG/CR-5512, Volume 1). [1. 1. DWR, = C, [ A n AF4 + DIET L A n AF 4 f=1 jai (3.8.1) where C., is the initial concentration of radionuclide I in soil at the time of site release, A, is the average concentration factor for radionuclide j in water over the current 1-year exposure period per initial unit concentration of parent radionuclide I in soil at the time of site release, AF o ,is the CEDE factor for the ingestion of drinking water per unit average concentration of radionuclide j in water, and AF,,,is the CEDE factor for radionuclide j per unit average concentration of radionuclide j in groundwater used for irrigation for the current 1-year period. The drinking water ingestion factor, AFa,, is calculated (equation 5.75, page 5.59, NUREC/CR-5512, Volume 1) as follows: Residential Scenario 3.8-1 February 1,1998 9
AFo, = U, DFG, to (C y / Cy ) (3.8.2) O .where_UJ is the daily intake of drir king water, DFG,is the ingestion CEDE factor for radionuclide j, and to is the duration of water intake period (1 year). The concentration ratio C y
/C yequal to 1 indicates normaliza lon to t. nit average concentration in water ovei the year of the residential scenario.
I The fraction of irrigation _ water appliec to the uurface layer, F,, is calculated in NUREGICR-
' 5512, Volume 1 (Equation 5 89, page 5.68) r.s follows:
V'" F, = (3.8.3) Vg V, where V,is the volume of water used for irrigation ouring a 1-year period (Ud) and Va,is the i volume of waier used for domestic purposes during a 1-year period (Ud). During ana ysis and testing of the originai methodology proposed in NUREG/CR 5512, it was found that the ground-
.ater contamination models described in Volume 1 do not adequiely account for natural discharge from the aquifer. The result was radionuclide build up in the aquifer box. Therefore,
. a water balance model was added to the methodology to correct this problem. These changes are documented as proposed Appendix A to Volume 2 of NUREG/CR-5512. In Sectior. A.3.6 of Appendix A (page A.10, equation 5.89m), equation 3.8.3 above was modified to the following:
V 5, = 2 (3.8.4) v,7 > F, is used in the water-use model mass balance equations for Box 1, the soil surface layer. ! In Volume 1 of NUREG/CR-5512 (equauon 5.86, page 5.67), Fractional Removal was defined as the fraction of total aquifer volume that is removed during a year, it was calculated as: Fractional Removal = V,,, c V* (3.8.5) Y Tr
- Once again, in order to correct this, equation 3.8.5 was modified in Appendix A to Volume 2 of NUREG/CR-5512 (section A.3.6, page A.10, equation 5.86m) as:
Fractional Removal = 1 -(3.8.6) Finally, in Volume 1 of NUREG/CR-5512 (equation 5.85, page 5.67), the fraction of the total aquifer volume that is removed during a year, w,, is given by the following equation:
, Fractional Removal y (3_g7y y ( 365.25 d, Residential Scenario 3.8-2 February 1,1998
whera 365.25 is the units conv;rsion factor (dly). This equation 3.8.7 was modifi:d in Appendix A to Voluma 2 of NUREG/CR 5512 (s:ction A.3.6, pag 3 A.10, equation 5.85m) as:
% " 365.25 d Information Reviewed to Define the Distribution for V ,
USGS water use data (USGS,1990 and USGS,1995) provide estimates of domestic water use in the United States by state. Per capita water use estimates were provided for both : elf-supplied as well as public-supplied delivery systems. Table 3.8.1 provides the origir.al per capita use of water, by state, from self-supplied water systems in gaHons per day. These quantities are converted to total liters / year by assuming a single resident in the household (for consistency with all other parameters in the residential scenario) and 365.25 days per year. The 50 values for annual per capita domestic water use by state were binned and fit tn %veral distribut ens. The fitne ss to each distribution was evaluated with a Kolmo,.;orov Smirnov test. The best fit was obtain ed from a log normal distribution. The frequency distribution and the corresponding PDF for innual domestic water removed from the aquifer, V ,, is shown in Figure 3.8.1. The correspondir g cumulative distribution for V ,is shown in Figure 3.8.1. l PARAMETER UNCERTAINTY: The values in Table 3.8.1 are based on estimates that depend on population estimates, inaccuracy in reported meter readings or other self-supplied means to measure water use, and differences in the definition of domestic use. Population estimates can l be a significant source of uncertainty when considering transient and non-resident users and may also depend on whether the approach used for estimation is consistent with the approach used for determining water use. Domestic water use that is measured and reported may l include different flow streams in different areas depending on distribution and metering system design, and what is considered domestic use, versus agricultural, etc.. We believe that uncertainty associated with determination of total annual domestic water use is small relative to the regional / climate variability discussed in the next section, and is captured in the proposed distribution. VARIABILITY ACROSS SITES: The greatest source of variabikty in the determination of total domestic water use at any given site is the region of the country and climate. The domestic water use figures given in Table 3,8.1 include water used for household purposes such as drinking, food preparation, b'?1, washing clothes and dishes, flushing toilets, car washing, and watering lawns and ga%s. Depending on the local climate, generally the largest indoor uses are for toilet flushing and bathing. Outdoor uses can range from near zero in humid areas to 60 percent of total domestic use in arid areas. The data reporte a Table 3.8.1 captures the variability of total domestic water use for the continental U.S. as .vell as Alaska and Hawaii. Residential Scenario 3.8-3 February 1,1998 O
Table 3.8.1 - Estimated Annual Domestic Water Use for US States (L) State Per Capita Total Use State Per Capita Total Use Gal /d L Galid L AL 75.1 103,824 MT 77.9 107.694 AK 39.7 54,884 NE 124.8 172,532 AZ 117.9 162,993 NV 119.8 165,623 AR 88.3 122,072 NH 65.0 89,861 CA 74.2 102,579 NJ 74.9 103,547 CO 75.9 104,930 NM 77.6 107,280 CT 75.0 103,685 NY 58.2 80,460 DE 79.2 109,492 NC 55.0 76,036 FL 175.1 242,071 ND 78.1 107,971 GA 75.2 103,962 OH 75.0 103,685 HI 188.8 261,011 OK 86.1 119,031 ID 199.8 276,218 OR 103.5 143,086 IL 84 1 116,266 PA 51.6 71,335 IN 76.0 105,068 RI 70.1 96,911 IA 66.6 92,073 SC 75.0 103.685 KS 99.5 137,556 SD 62.5 86,404 KY 49.8 68,847 TN 65.0 89,861 LA 82.7 114,330 1X 108.2 149,583 ME 90.0 124,422 UT 85.9 118,754 MD 82.9 114,607 VT 71.9 99,400 (] (,,/ MA MI 72.0 72.8 99,538 100,644 VA WA 75.0 115.5 103,685 159,675 MN 116.6 161,196 WV 80.0 110,598 MS 49.9 68,985 WI 60.7 83,916 MO 60.0 82,948 WY 75.0 103.685 SITE DATA COLLECTION: Licensees may wish to defend new values for the total annual comestic water volume aue to site specific consioerations impacting water use. Some of tnose considerations may include regional climate (terverature and humidity), rainfall and its impact on water use for outdoor requirements, local water rates and water use restrictions and other conservation efforts that may not be reflected in typical reported values of water use, and such. The sirr*st approach for site specific analysis is to select, as an alternative to the default value, the. value from Table 3.8.1 that corresponds to the location of the site. DEFINITION OF SITE DATA SOURCE: For the purpose of defining the distribution of total annual domestic water use, supporting data similar to that provided in this document are assumed to be available. For site-specific analysis, Table 3.8.1 already provides data at the state level that licensees could use to defend alternatives to the default value for Va ,. If required (but more difficult to defend), more detailed USGS data for all US counties is also available. NRC INTERPRETATION OF SITE-SPECIFIC VALUE: Licensees may attempt to defne site specific values for the annual domestic water use for their site under the constraints of the Residential Scenario 3.8-4 February 1,1998 G
residential farmer scenario. Those alternative values will need to be consistent with typical domestic water use in that region of the country, unless site characteristics, requirements, or use restrictions can be used to defend significant deviation from the representative state-specific values given in Table 3.8.1 and captured in the parameter distribution derived for V a,.
REFERENCES:
USGS,1990. " Estimated Use of Water in the United States in 1990," USGS National Circular 1081. USGS,1995. "1995 Water-Use Guidelines: Domestic Water Use," USGS taken from Web URL: http://h2o.usgs. gov /public/watuse/ guidelines /do.html E '
/
E 08 8 5 3 i Q I y 04 E a 02 0 0 50000 100000 150000 200000 250000 300000 Annual Domestic Water the (l) Figure 3.8.1 Proposed Cumulative Distribution for Annual Domestic Water Use, Var Residential Scenario 3.8-5 February 1,1998 9
I 3.9 Ingestion rates of home produced food, U,(kgly),0,(kgly) and U,(kgly)
\
The ingestion rates of homegrown produce, U,(kgly), and other home produced food, U,(kgly), (s and U,(kgly), as defined for NUREG/CR-5512, Volume 1 [ Kennedy and Strenge,1992), dose modeling, represent the consumption rate of specific contaminated food. The dose model uses different constant values of U, for 1eafy" vegetables, "other" vegetables, fruits and grains, different constant values of U, for beef, poultry, milk and eggs and a constant value of U, for fish. U,, U, and U, are behavioral parameters in that they represent the diet of the average member of the screening group (i.e., residential and light farmers). This section first includes brief discussions of the importance of U,, U, and U,with regard to the calculated dose and how U,,- U, and U, are specifically used in the dose model, Next, the default values for U,, U, and U, used in NUREG/CR 5512, Volume 1 are discussed. Lastly, U, and U, are redefined based on data reported by the Environmental Protection Agency (EPA) (EPA,1996) and distributions for U,, U,and U, are presented based on this data. U,, U, and U, are directly proportiona! M dose. Therefore, the higher the values for U,, U, and l U, the higher the calculated dose. More specifically, the inge.. ion rates, U, and U, are used in l the dose model to calculate the agricultural pathway transfer factors (PF). These factors are then used to calculate the annual dose from ingestion of home produced food. The mathematical expression to evaluate the PFs for unit average concentration of a parent radionuclide in soil is given in NUREG/CR 5512 (page 5.51) as: PF.,, = Et ,,,w U, PPTF,,,, + E c..m,3 U, PPTF,,, (3.9.1) O where PF.,, = the agricultural pathway transfer factors for radionuclide j as a progeny of
) radionuclide i per unit initial concentration of parent radionuclide in soil (pCi ingested per pCl/g dry weight soil for a year of residential scenario),
.U, = the ingestion rate for food crop type v by an individual (kg wet-weightly),
PPTF,,,, = the partial pathway transfer factor for food crop type v, racionuclide j as a progeny of radionuclide I, for unit average concentration of parent radionuclide I in soil (pCi y/kg dry weight food per pCilg dry-weight soil for > a year of residential scenario), U, = the ingestion rate of animal product type a by an individual (kg wet-weight /y), PPTF,y = the partial pathway transfer factor for animal product type a, radionuclide j as a progeny of radionuclide I, for unit average concentration of parent radionuclide I in soil (pCi ylkg wet-weight food per pCilg dry-weight soil for a year of residential scenario), N, = the number of animal products considered in the diet, and N, = the number of food crops considered in the diet. (O Residential Scenario 3.9 - 1 January 31,1998 _ - _ 1
The m them:tical expression to Ovaluate the PFs for unit av; rage conc;ntration of a parent radionuclide in irngation wat:r is giv:n in NUREGICR 5512 (page 5.52) as: P F ,y = E 3., u ,3 U, PPTF y,,, + E .4 3 y,> U, PPTF,,y (3.9.2) ! where PF,y = the agricultural pathway transfer factor for radionuclide j as a progeny of radionucliae i per unit initial concentration of parent radionuclide in irrigation water (pCiingested per pCi/L water for a year of residential scenario), U, = the ingestion rate for food crop type v by an individual (kg wet weightly), PPTF,,, = the partial pathway transfer factor for food crop type v, radionuclide j as a progeny of radionuclide I, for unit average concentration of parent radionuclide I in water (pCi y/kg wet weight food per pCi/L water for a year of residential scenar o), U. = the ingestion rate of a',imal product type a sy an individual (kg wet-weight /y), PPTF,,y = the partial pathwa', transfer factor for animal product type a, radionuclide j as a progeny of radionuclide I, for unit average concentration of parent radionuclide I in irrigation water (pCi y/kg wet-weight food per pCi/L water for a year of residential scenario), N, = the number of animal products considered in the diet, and N, = the number of food crops considered in the diet. The ingestion rate of fish, U,, is used in calculating the aquatic food ingestion factor (AF). AF is then used to calculate the annual dose from ingestion of aquatic foods. The mathematical expression for AF is given in NUREG/CR 5512, Volume 1 ( page 5.60) as: AF q= U, t, DFG, BA,, (C ,/C.,)/365.25 (39.3) where AF g = the aquatic pathway transfer fector for radionuclide j as a progeny of radionuclide I, per unit average concentration of radionuclide j in surface water (mrem per pCi/L for a year of the residential scenario), U, = the ingestion rate of aquatic foods produced in contaminated surface water, t, = We duration of fish consumption in days, DFG, = the ingestion CEDE factor for radionuclide j (mrem pr pCi ingested), Ba,, = the bioaccumulation factor for radionuclide j in aquatic foods, P.nd Residential Scenario 3.9 - 2 ,lanuary 31,1998 9
l C, a tha avsrage ann: ai cone:ntration of radionuclida j in wat:r (pCi/L). It should be noted that U, and U, are also used in the new proposed functional relationship to -V define the area of land cultivated parameter, A,. Section 3.2 provides a detailed description of this new function. The values used for U, and U,in NUREG/CR 5512, Volume 1, are based on food ingestion rates found in the 1977-78 Nationwide Food Consumption Survey [U.S. Department of Agriculture (USDA) 1983). The specific values are derived from mean values compiled by Higley and Strenge [1988) and Pao et al. [1985). These values are based on consumption data that repres.ents all food sources and not just home grown food. Using this data is not directly representative of the screening group. To compensate for this inaccuracy in the dose model, an additional parameter is used, DIET (see Section 3.1) The DIET parameter represents the fractiois cf the diet that an individual at the site eats that is from home produced food. A single, common value for the DIET parameter is assumed to apply to all food products. This assumption requires, for example, that the fraction of domestically produced beef in the diet equals the fraction of domcatically produced leafy vegetables. This assumption is unlikely to be satisfie..'in general, and is not representative of the screening group. An additional compensation in the model is the selection of the eight food groups that are thought to be representative of food products that would be produced on a farm for home consumption. The default value used in NUREGICR-5512, Volume 1, for U, was based on summary data presented by Rupp et al. [1980) The regional percentiles reported in Rupp et al. are based on the entire population,' including those individuals who eat no fish, which is not representative of the dose model screening group. To try to compensate for this inaccuracy, (i.e., Rupp et al. O reported that over 85% of the population eat no freshwater fish), the value for the highest i V regional rate reported by Rupp et al. was used as the default value in NUREGICR-5512, Volume 1. In the model, U, did not have a corresponding DIET parameter like U,and U,, which implies that it represents the consumption of domestically-produced fish. Table 3.9.1 displays the default values of ingestion rates for the eight food groups used in NUREG/CR-5512, Volume 1, dose modeling, Table 3.9.1 NURtii/CR-5512 U,, U, and U, Default Values Food Type Consumption Rate Leafy Vegetables (U,) 11(kgly) Other Vegetables (U,) 51 (kgly) Fruit (U,) 46 (kgly) Grain (U,) 69 (kgly)- Beef (U ) 59 (kgly) Residential Scenario 3.9 - 3 January 31,1998 a l
Tabb 3.9.1 NUREG/CR 5512 U,, U, cnd U, Default Valu;s Food Type Consumption Rate Poultry (U ) 9 (kgly) Milk (U,) 100(kgly) Eggs (U,) 10 (kgly) Fish (U,) 10 (kgly) The most recent Nationwide Food Consumption Survey (USDA,1993) was conducted in 1987-88 and is more reflective of long-term nationwide consumption trends compared to the 1977-78 survey data. Again, the individual survey data could not be used directly to measure consumption of home produced food because the source of the food item is not identified. However, EPA reports intal. ates for various home oroduced food items [ EPA,1996] based ori an analytical method that combined data from botn the household ar d individual 1987-88 USDA survey components. The data is reported in the form of cumulative probability distributions. Because o'the availability of this data set, U,, U, and the DIET parameter can be redefined for the dose mudet. U, and U, become the rate of consumption of food from on-I site production rather than the rate of consumption in general. With this definition of consumption rates, the DIET parameter value is 1 in all cases. These new definitions allow the fraction of domestically-produced food products to be specified for each crop and animal type, which is more representative of the screening group. In addition, redefining these parameters in this manner makes them consistent with the definition of U,. Despite directly representing the screening group, there are several assumptions that have to be made to match the dose modeling requirements with the data provided by EPA [1996). First, the eight food categories have to be defined with respect to the EPA data. EPA reports intake rates that directly match the "other" vegetables, fruits, beef, poultry, eggs and fish categories. For the " leafy" vegetables category, it is assumed that this category is equivalent to EPA's "expor d" vegetables category. EPA defines the " exposed" vegetables category as those vegetables that are grown above ground. Therefo", assuming that the ca" gory of " leafy" vagetables is equivalent to EPA's " exposed" vegetable category is reasonable given the fact that a!I leafy vegetables are grown above ground, although it may overestimate this category since not all vegetables that are grown above ground are leafy. For the food grain group it is assumed that the EPA data for corn is appropriate, given that corn is the only grain for which data was reported. This assumption is consistent with the study by McKone (1994), where he also used corn to represent the grain category. The milk category is assumed to be equivalent to the EPA's dairy category. Again, this assumption is reasonably conservative given the fact that milk is a dairy product. EPA notes that the survey data was taken during a week long period and therefore, may not be representative cf annual behavior (i.e., more home grown foods are typically eaten in the summer). EPA generated seasonally adjusted intake distributions for all meats. vegetables and Residential Scenario 3.9 - 4 January 31,1998 9
fruits by av:rrging the corr;sponding percentil;s of each of th3 four seasonalintake
,, distributions repcrted. This same approach was used to generate seasonally adjusted
( ) distributions for the eight food group categories required for the dose model. LJ The ingestion rates that EPA reports are indexed to the actual body weights of the survey respondents and are reported in units of mass ingested per time per respondent body weight. Although EPA does not recommend converting the intake rates into average ingestion rates of mass / time by multiplying by a single average body weight, they do indicate that if this is done, a weight of 60 kg should be used because the total survey population included children. The NUREG/CR-5512, Volume 1, dose model requires average ingestion rates. In order to convert the EPA data and maintain representation of the average member of the screening group, the seasonally adjusted data is scaled by the percentile average of the ratio of the 20-39 age data to the total population data and then converted by multiplying by the body weight, 70 kg, of the average member of the screening group. This data adjustment assumes that the data scales linearly. EPA does not provide any information about whether or not this assumption is valid, but it L a reasonable first approximation. The homegrown food ingestion rate distributions rep. .ed by EPA are based on ti e amount of food ' consumed" in an economic sense (i.e., food that has been brought into the house). EPA recommends converting these intake rates to reflect actual ingestion by decreasing the amounts by percent weight losses from preparing the foods. EPA provides percent weight losses for various meats, fruits and vegetables. Therefore, these losses were accounted for in deriving the dose model distributions for U,, U,and U, . However, losses were not reported for eggs and milk, so these losses were not accounted for in these two food categories.
/O Figure 3.9.1, Figure 3.9.2, and Figure 3.9.3 present the cumulative distribution functions based C on the derivations presented above for U,, U, and U,, respectively. Based on these cumulative distribution functions for U,, U, and U, probability distribution are derived. The characteristics of the resulting log normal distributions are presented in Table 3.36.2 along with the equivalent 5512 values (rates multiplied by the DIET parameter) and 1995 consumption rates for all foods not just homegrown [USDA,1997).
Trale ? 9.2 Statistical Characteristics of the Log Normal Distributions for U,, U, and U, Parameter Home Standard Equivalen 1995 U.S. Ratio of grown Deviation t 5512 Total Homegrown Ingestion Value Ingestion Mean to Rate (kgly) Rates 1995 Totals Mean (kgly) (kgly) U, - leafy 15 6.0 2.8 185 (total) (leafy + vegetables other)/ total = 0.25 o\ ld Residential Scenario 3.9 - 5 January 31,1998
. -- l
Table 3.9.2 Statistical Characteristics of the Log Normal Distributions for U,, U, and U, Parameter Home Standard Equivalen 1995 U.S. Ratio of grown Deviation t 5512 Total Homegrown Ingestion Value Ingestion Mean to Rate (kgly) Rates 1995 Totals Mean (kgly) (kgly) U, - other 32 6.2 13 185 (total) (leafy + vegetables other)/ total = 0.25 U, - fruits 34 7.2 12 128 0.25 U, - grain 12 4.3 17 87 0.14 U, - beef 32 4,3 15 29 1,1 U, - poultry 20 3.1 2.3 28.5 0.7 U, - milk 118 L/y 7.7 25 L/y 358 L/y 0.33 U, - eggs 18 3.6 2.5 12 1.5 U, - fish 16 7.0 10 7 2.29 Comparing the equivalent NUREG/CR 5512 de. fault parameters (i.e., consumption rate default parameters multipled by the default DIET parameter of 0.25) with the mean of the newly proposed distributions indicates that the mean of the new distributions are consistently higher than their 5512 equivalent, except for the grain category. However, given the differences in their derivations, especially the value for the DIET parameter, the means and default values are within reason when compared to the general consumption rates in the U.S. in 1995 as can be seen in Table 3.9.2. SITE VARIABILITY: The proposed distribution functions presented above represent the behavorial variability of the average member of the screening group and are not related to the physical characteristics of the specific site being considered. Therefore, these parameters are not expected to vary from site to site and it is very unlikely that a licensee would conduct any type of data collection activity to modify them.
REFERENCES:
Kennedy, Jr., W. E., ana D. L. Strenge,1992. " Residual Radioactive Contamination from Decommissioning: Technical Basis for Translating Contamination Levels to Annual Total Effective Dose Equivalent," NUREG/CR-5512, U.S. Nuclear Regulatory Commission, Washington, DC. U.S. Environmental Protection Agency (U.S. EPA),1996. -xposure Factors Handbook, Residential Scenario 19-6 January 31,1998 O
Vo,'uma II ofill- Food Ingesticn Factors, Update to Exposura Factors Handbook (Office of Research and Development. National Center for Environmental Assessment, EPA /600/P- [,') 95/002Bb,1996). \) U.S. Department of Agriculture (USDA),1983. Food intakes: Individuals in 48 States. Year 1977-78 Nationwide Food Consumption Survey 1977-1978. Report No. 77 I-1, USDA, Consurcar Nutrition Division, Hyattsville, Maryland. Higley, K.A. and D.L. Strenge,1988. "Use of a Monte Carlo Modeling Approach for Evaluating Risk and Environmental Compliance," Presented at the Fourth Annual DOE Model Conference, Oak Ridge, Tennessee, October 3-7,1988. PNL-SA-16062, Pacific Northwest Laboratory, Richland, Washington. Pao, E. M., K. H. Fleming, P. M. Guenther, and S. J. Mickle,1985. Foods Commonly Esten by Individuals: Amounts per Day and per Eating Occasion. Report No. 44, U.S. Department of Agriculture, Consumer Nutrition Division, Hyattsville, Maryland. Rupp, E. M., F. L. Miller, and C.F. Baes lit,1980. "Ss e Results of Recent Surve;3 of Fish and Shellfish Consumption by Age and Region of U.S. Residents,' . ,aalth Physics 39,165-176. U.S. Department of Agriculture (USDA),1D93. FoodIntakes: Individuals in 48 States, Year 1987-88 Nationwide Food Consumption Survey 1987-1988. Report No. 87-1-1, USDA, Consumer Nutrition Division, Hyattsville, Maryland. McKone, T. E.,1994. " Uncertainty and Variability in Human Exposures to Soil Contaminants i Through Home-Grown Food: A Monte Carlo Assessment", Risk Analysis 14(4),449-463. \_/' U.S. Department of Agriculture (USDA),1997. Agriculture Fact Book 1997, accessed through URL http://www.usda. gov / news / pubs /fbook97/ contents.html/. i (m) Residential Scenario 3.9 - 7 January 31,1998
\d a ______._______j
p' . Os
- t .
l ' Os
. _BPe l . i 02 0 ,F _
o ,e = = = = --
=
%aw Figure 3.9.1 Cumulative distribution of U, for exposed (leafy) vegetables, other vegetables, fruits, and corn (grain).
.u.w4, sq
- e*n .w %,
uc. 4F
, . f. t e s
. e .:,- s ~ . .
s,, O
, a %, f;: p'i),' :
g, '* , b . -
- : 7, A, , w, , *. , 4, r X. Kp"Am .e g 3p,.
- . *?-
4
.l
. . - . , , ..s y7cgig .,-,. ., , y. ,g
..s , , ~ . . .
lg; D$) ' *
, # 'l *
- 7 E' -
fid u r4*./f 5.C 1. 4.,'y . t(( r.f,' f.. vN t< ' h, ,# ..~ - .
,Y,~i .=,._
J .. ~l ,,.,. We em. y.q d, +,m><.;ce,d. g g 4 :y A.a,m--
, . ., , r 4 .e
,-..d.rs>... %:n
.. e s.
.am... 3 . , . ,
r)gcSM A. '.~)inp Q,ya , f: ;y, a,tc m:.n.n :5: ',y,-m. s.. .snifwW.# t
'?l%y,%
G .-
" q A. .a y
!., ,e, ,m# -
, g.* + >e:.- ,
.; ~ ,.-a, w j
;-; yuyp:a e,>c r ruh ma4.p ,Q- e a .a? ww .
t l, :3. , .:.. m., a. p .
- m. n 9.c. ,.- -7c- m
.u
,1 c r.+,+7
+ :,
$. .').7.,t2,: 4,,
u ..~ ' . in, .
, ."d W 4. 1. t a 9QiykWipt
./i.s. t,'[mifllM4,f9Nii. o-
. %%@M '
f
;%(%? E WVTW
+' -J
,a , -
. . . u
.~ m Figure 3.9.2 Cumulative distribution of U. for beef, poultry, dairy (milk) and eggs.
Residential Scenario 3.9 - 8 January 31,1998
O
.x
%;&: Qsissi;!.t W:t. % 1 2,'
{ :{ f .* * *) . t ' N , t%" t
.p, y +%
F~-R,8 w.J 3 a.y > .; {+ y;a * '+ i , r
;f.,,,
4
- r. . .. 6 5 ; cu ...3 , ., i
-,- e ~
00 9rs< ">,s u+,
< - y;s+** P 6.. %a e,r pv nhw* .: a'( ,wn .si e.nr &.n, 4-i- e n . J7y;; : .py
,j , 4MhWYNC,$dM,1-:gt.o* ,
E '.N$$3d.eo$'NdN t g sia,,. $RMY. N g = ?!A3/ WAp 1 LW2k%V*$pG)f4 "W@Q "'l#' e
%k!$a%. 6 ge or ;. th, *y;y %._ m
~
os h , e,,a . , a g% .. --, Y?$'.,3'gg;
. .n.; - . . . . ~ ,
$N0.v%pygg.g;g;
- . uy.WsFC . - ~ p;, cw.
on .a, . m.;g e., 44 n, 1w r a ry,a e u.yv
,,y.,w .
93
$3 . hw.6_ _ .e '
u ~ -
- 4. N vi
'# Mest p < ni P.L w, ..j.g.,.
*PG4DO c,t i4*y'
_ . . ar,a: scJug.9::Gc; o, M ar.m + war *
- Fan M4trg.,
-s
% A&;;if
% ;% Mi;thDC' Vjt? S i 2'l n 9 n_
W 2,;.w ,.
; ,;w ,;. cQt. r y .;L o a a m 4m a a a a a trgeston Rate (kgh)
Figure 3.9.3 Cumulative distribution of Ut for fish, N i (' Residential Scenario 3.9 - 9 January 31,1998 (s i
1 1 4,0 V: lum tric bre: thing rates (M: tab:lic), V,, V,, and V, (m 2/h) 4.0.1 Parameter Description The residential scenario defines three exposure situations or contexts for resident farmers; indoors, outdoors, and gardening. These exposure contexts are distinguished because the transport rates may differ significantly among them. The breathing rate parameters, in conjunction with the indoor resuspension factor, dust loadings, and isotope-specific inhalation CEDE factors, are used to calculate the average annual dose due to inhalation. The breathing rate parameters represent the annual average breathing rate of the average member of the screening group while indoors (V,), outdoors (V,) and gardening (V,). As described in Section 1.0, default values for metabolic parameters are established by the average value for adult males in the general population. The de'ault value defined for each of the three breathing rates in NUREG/CR-5512, Volume 1, 3 is 1.2 m /h. This value corresponds to an average for the 8 hour work day assuming light activity for a person, as sugreted in ICRP Publicaticq 23 [1975). Revised default values for these parametels were defined based on a review o, .orrent literature on breathir.g rate. 4,0.2 Use of Parameter in Modeling Within each of the three contexts defined for the residential scenario (indoors, outdoors, and l gardening), inhalation dose is directly proportional to breathing rate in that context. The overall importance of breathing rate in determining dose depends on the relative contribution of inhalation dose to total dose, wnich in turn depends on exposure rates via alternative pathways, anci en nuclide-specific dose factors. The breathing rate parameters are used to calculate the committed effective dose equivalent, CEDE, resulting from inhalation of resuspended surface contamination. The relationship between the volumetric breathing rates and internal dose due to inhalation (DHRi) is described by the following (see NUREG/CR 5512 page 5.55): DHRj = [24Vggt (t /t g) CDG Csi I 1.Ji) S{Ast).ttg}DFHj]
+ [24Vx(tx/t t r) GDO l Cs I j=1,Ji)(j=QA ct t j.t r}DFHj]
+ [24Vr(tj/tt r) (CDI d + P RF(r) i Cs I(j=1.Ji) t S{Ast),t r}DFHj](4.0.1) where Vgis the volumetric breathing rate for time spent gardening (m3/h), Vr is the volumetric breathing rate for time spent indoors (m3 /h ), Vx is the volumetric breathing rate for time spent outdoors (m3/h), tg si the time during the gardening period that the individual spends outdoors gardening (d for a year of residential scenario), tj is the time in the 1-year exposure period that the individual spends indoors (d for a year of residential scenario), tx is the time in the 1-year exposure period that the individual spends outdoors, other than gardening (d for a year of residential scenario), tt r is the total time in the residential exposure period (d), CDI, CDO, and CDG are the dust loading factors for indoor, outdoor, and gardening activities (g/m3),
respectively, Csi corresponds to the concentrction of parent radionuc!ide i in soil at time of site release (pCi/g dry-weight scil), Ji is the number of explicit members of the decay chain for parent radionuclide i, S{Astj,t tr} is a time-integral operator used to develop the concentration Residential Scenario 4.0-1 January 31,1998 9
tim) integral of radionuclid)J for cxposure over a 1 y3ar period per unit initial concentration of parent radionuclide iin soil (rOi"d/g per pCi/g dry weignt soil), S{Astj.ttg} is a time-integral O operator used to develop the concentration time integral of radionuclide j for exposure over one Q gardening season during 1 year period per unit initial concentration of parent radionuclide iin soil (pCi"d/g per pCi/g dry-weight soil), DFHj is the inhalation committed effective dose equivalent factor for radionuclide j for exposure to contaminated air (in units of mrem per pCi inhaled), Pdcorresponds to the indoor dust-loading on floors (g/m2), and RFr is the indoor resuspension factor (m"1). The resulting internal inhalation dose is directly proportional to the volumetric breathing rates for indoor, outdoor, and gardening activities. 4.0.3 Information Reviewed to Define Breathing Rate Distributions The literature review conducted to support the Draft EPA Exposure Factors Handbook (EPA 1996) was adopted for this study as the most current compilation of relevant literature. Eleven studies are reviewed and summarized in the Handbook. Five are identified as " key studies", and form the basis for inhalation values recommended there. The six remaining studies are considered " relevant", and contain supoorting information relating to inhalation rate. Separate breathing . ate estimates are not reported in any study for the 5. acific contexts defined for t"3 residential scenario instead, daily average values are reported, as well as breathing rates for individuals engaged in various levels of activity. These activity levels are descriptively defined, for example as " rest", " sedentary", " light", " moderate", and " heavy", Reported average daily values include a range and relative weighting of activities typical of an entire day: this range and weighting of activities is not representative of activ; ties specifically conducted indoors, outdcors, or while gardening. For this reason, reported average daily l O '-(~j values are not appropriate for these parameters. The three exposure contexts in the residential scenario can be distinguished by the types of activities that would typically take place in each: indoor activities would typically include sleeping and resting, for example, while outdoor activities would not. For this reason, breathing rates for each context have been anlgnet based on the range of activities that would occur in each context, and the reported average values for the corresponding activity levels (see Section 4.0.4 below). The summaries in the Handbook were used to evaluate the five " key' studies for the purpose of defining breathing rates for the average member of tne screening group. Each of the five " key" studies, and any resulting breathing rates that reflect the screening group, are summarized below. Layton (1993) presents a method for estimating breathing rate based on metabolic information: Ve = E x H x VQ (4.0.2) where: Ve is the ventilation rate E is the energy expenditure rate H is the volume of oxygen consumed in the production of 1 KJ of energy, and VQ is the ratio of intake volume to oxygen uptake O Residential Scenario 4.0-2 January 31,1998 U
i l ThrC3 cpproaches ara us;d to cstimate the en:rgy expenditure rate: annual calonc intake (corr:ct:d for r porting bias), ci;vation above basal m;tabolic rate (BMR) with BMR valu:s est; mated from body weight using a fitted regression model, and elevations above BMR using activity specific elevation factors and time allocation data. These methods are used to estimate average inhalation rates over various population subsets defined by age and gender. This study draws from comparatively large data sets, and provides information on the relative contributions of the diverse factors influencing inhalation rate, including general health, body weight, diet, activity level, age, and gender. The first two methods provide estimates of long-term average breathing rate, which is not specific to the residential exposure contexts. The third method provide estimates of breathing rate for different levels of activity. Average inhalation rates for adult males for five activity levels, estimated by the third method, are summarized in lable 4.0.1. Estimates for two sets of activity classifications are reported. For each set, activity level is characterized by a qualitative description as well as by a BMR value or range. Different sets of BMR values were used for each set. Table 4.0.1. Estimated Breathing Rates for Males from Layton (1993) for Two Sets of Five Activity Levels (m3/hr) Inhalation dates fnr Short Term Exposures' Activity Level Age (years) Rest Sedentary Light Mode ate Heavy BMR: 1 BMR:1.2 BMR: 1.5 - BMR 3 - 5 BMR: >5 - 20 2.5 18 - < 30 0.43 0.52 0.84 1.74 4.32 30 - < 60 0.4L 0.50 0.84 1.68 4.20 Activity-Dependent Inhalation Rates used to Estimate Daily inhalation Rate: Activity Level Age (years) Sleep . Light Moderate Hard Very Hard BMR; 1 BMR.1.5 BMR: 4 BMR: 6 BMR: 10 20-34 0.4 0.7 1.7 2.6 4.3 35-49 0.4 0.6 U 2.5 4.2 50-64 0.4 0.6 1.7 2.5 4.2
' Source: EPA (1996) Table 5 5 2 Source: EPA [1996] Table 5-6 Linn et al. (1992) estimates inhalation rates for "high-risk" sub-populations, including outdoor workers, elementary school students, high school students, asthmatic adults, young asthmatics, and construction workers. Of these sub-popu'ations, outdoor workers and construction workers approximate the screening group. The average breathing rate for healthy adults outdoor Residential Scenario 4.0-3 January 31,1998 9
I
work;rs, consisting of 15 wom;n and S man betw;;n the ag:s of 19 and 50, is r; port:d as 0.78; construction workers, consisting of 7 men between the ages of 26 and 34 have an (]/ [ average breathing rate of 1.50 m'/hr. Activity-dependent breathing rates are also reported for both subject groups at three activity levels, as shown in Table 4.0.2. Table 4.0.2. Estimated Breathing Rates from Linn [1992) for Two Panels of Healthy Adult Subjects' (m3/hr) Mean Self-Estimated Breathing Rates Subject Group Slow Medium Fast Outdoor Workers 0.72 1.02 3.06 Construction Workers 1.26 1.50 1.68
' Source: EPA [1996) Table 5-7 Linn et al (1993) reports breathing rates for 19 construction workers who P. form heavy outdoor labor both before and during a typical work shift. The subjects of this study approximate the screening group, although the number of subjects is small. A regression model relating breathing rate to heart rate was developed from data collected in a controlled laboratory protocol. Occupational breathing rates were estimated from measured heart rates using this regression model. Average breathing rates are reported for three self-estimated activity levels,
, as shown in Table 4.0.3. p Table 4.0.3. Estimated Breathing Rates from ("/ l Linn (1993) for Outdoor Workers' l (m3/hr) Mean Self-Estimated Breathing Rates Slow Medium Fast 1.44 1.86 2.04
' Source: EPA [1996) Table 5-9 Spier et al (1992) reports breathing rates for elementary and h,gh-school students. Although considered a
- key" study in the Handbook, this sub-population does not correspond to the screening group for the residential scenario. Results of this study were not used to establish values for the screening group.
The California Air Resources Board (CARB) (1993) reports breathing rates in routine daily activities for children and adults at various activity level classifications. The study included a laboratory protocol, in which ventilation rate, heart rate, breathing frequency, and oxygen consumption were measured during treadmill tests. Heart rate, ventilation rate, and breathing frequency were also measured during a 'fie'.d* protocol, which included (for adult males) driving and riding in cars, yard work, and mowing. Average breathing rates during the laboratory O
'LJ Residential Scenario 4.0-4 January 31,1998 1
protocol are r: port:d for five activity classifications. Average valu:s during th) field protocol ara r; port:d for thr:3 activity classific:tions. Tabla 4.0.4 summarizes tha rcport;d valu;s for adult males. Table 4.0.4. Average Inhalation Rates for Adult Males from CARB (1993) (m3/hr) Activity Level Resting Sedentary Light Moderate Heavy Laboratory 0.54 0.60 1.45 1.93 3 63 Protocols' Field 0.62 1.40 1.78 Protocols 2
~
' Source: EPA [1996] Table 5-13 2 Sourcw EPA [1996) Table 5-14 The six studies classified as " Relevant' provide supporting information, such as assessments of the quality of individual's subjective judgments of their breathing rate and activity level. These studies were not judged to provide information directly related to estimating breathing rates for the screening group. Three literature surveys are also classified as " Relevant". The U.S. EPA (1985) provides a summary of inhalation rates by age, gender, and activity level. This study compiles results of earlier investigations, and does not present information on the accuracy and methods used in these investigations. Reported breathing rates range from 0.7 to 4.8 m 3/hr for adult males depending on activity level. The International Commission on Radiological Protection (ICRP) (1981) presents ventilation estimates for reference adult males and females at two activity levels (" Resting" and " Light Activity") as well as daily inhalation ratas based on an assumed activity pattern during the day. For adult males , the respective rates are given as 3 8 0.45 m /hr,1.2 m /hr, and 22.8 m / day. The default values for V,, V, and V, defined in Volume i of NUREG/CR 5512 were based on the " Light Activity" breathing rate for males from this study. It was not considered a sufficient basis for defining default values for these parameters because of the availability of more current empirical data in four of the five " key" studies l
discussed above. The AlHC (1994) Exposure Factors Sourcebook recommends an average 3 adult inhalation rate of 18 m / day based on data presented in other studies. This report draws from information presented elsewhere, and does not present new data on breathing rate. 4.0.3 Average Breathing Rates for the Residential Scenario Contexts For the indoor, outdoor, and gardening contexts defined for the residential scenario, breathing rates of the average member of the screening group were estimated from the average breathing rates for adults discussed in Section 4.0.2. Where separate estimates are provided for males cred females, estimates for males were adopted as being more representh.ve of the screening group. Each context was first characterized by the range of activity levels for the activities that would typically occur in each. Indoor activities include sleeping, reading, watching television, kitchen Residential Scerwrio 4.0-5 January 31,1998 9 1
work and hous: work, and rep;ir and maint; nance. Such activities correspond to the *Risting*, l ' Sedentary",
- Light', and *M cderate' level classifications used by Layton [1993) and CARB f
(1993). Outdoor activities include yard work, recreation, and car and equipment repair and maintenance. Typical outdoor acttfil:es were therefore assumed to correspond to the
'Seoentary*," Light *, and
- Moderate
- categories of Layton [1993) and CARB 11993), and to tho
- Slow" and
- Medium" subjective breathing rate classifications used in Linn's studies of outdoor workers.
- Gardening activities include soil oreparation, planting, weeding, hoeing, and harvesting. These activities are assumed to correspond to the ' Light', *Moderete",
- Heavy", *Hard", and *Very Hard" levels adopted by Layton [1993) and by CARB [1993), and to lead to breathing rates subjectively classified as
- Medium" or ' Fast" by Linn's subjects.
For the outdoor and garder ng contexts, the reported average breathing rates for the activity levels typical of each context were identified. (For each of the two sets of values reported by Layton [1993), the median breathing rate over the individual age groups was selected as typical of adult males.) Table 4.u.5 lists the reported breathing rate values for activity levels expected occur outdoors, while Table 4.0.6 lists breathing rate values for ;ctivity level expected to occur while gardening. For both the outdoor and gardening contexts, estimated breathing rates cover a range of values due to ddferences among the studies, and to differencet in activity levels conducted in these contexts. An estimate of overall average breathing rate would require
-_information on time allccation among these activity levels. Because detailed time allocation information is not available, the median reported value was selected to characterize each context: 1,4 m'/hr for outdoor activities, and 1.7 m 8/hr for garoening activities.
Table 4.0.5. Reported Average Breathing Rates Corresponding to Activity Levels Typical of Outdoor Activities (Excluding Gardening) Breathing Refarence Study and Activity Level date 0.5 Layton (1993), Set 1: Median of ' Sedentary' values over adult age groups 0,6 - Layton [1993), Set 2: Median of .Jght" values over adult age groups Orr CARB [1993): ' Sedentary' value from laboretory protocol CARB [1E3): ' Sedentary
- value from field protocol 36 0.7 Linn [1992): ' Slow" value for outdoor workers 0.8 Layton [1993)l Set 1: Median of " Light" values o r adult age groups 1.0 Linn [1992):
- Medium" value for outdoor workers 1.3 Linn [1992): ' Slow
- value for construction workers 1,4 CARB [1993):
- Light' value from field protocol l Residential Scenario 4.0-6 January 31,1998
Tablo 4.0.5. R: ported Av: rage Breathing Rates Corresponding to Activity Levels Tynical of OutNor Activiti;s (ExJudMg Gard:ning) Breathing Reference Study and Activity Level Rate 1.4 Linn [1993): # Stow" value for outdoor workers 1.4 CARB [1993): " Light" value from laboratory protocol 1.5 Linn [1992): ' Medium' value for construction workers 1.7 Layton [1093), Set 1: Median of ' Moderate
- values over adult age groups 1.7 Layton [1993'. Set 2. Median of
- Moderate" values over adult age groups 1.8 CARB [1993):
- Moderate" value from field pictocol 1.9 Linn [1933):
- Medium" value for outdoor workers 1.9 CARB [1993): ' Moderate
- value from laboratory protoce' Table 4.0.6. Reported Average Breathing Rates Corresponding to Activity Levelc Typical of Gardening Activities Breathing Reference Study and Activity Level Rate 0.6 Layton [1993), Set 2: Median of ' Light" values over adult age groups 0.8 Layton [1993), Set 1: Median of " Light" values over adult age groups 1.0 Linn [1992): " Medium" value for outdoor workers 1.4 CARD [1993) " Light" value from field prviocol 1.4 CARB [1993): " Light" value from labore* ry protocol 1.5 Linn [1992): ' Medium
- value for construction workers 1.7 unn [1992): ' Fast" value for construction workers 1.7 Layton [1993), bet 1: Median of
- Moderate" values over adult age groups 1.7 Layton [1993), Set 2: Median of " Moderate" values over adult age groups 1.8 CARB [1993): " Moderate
- value from field protocol -
1.9 Linn [1993): " Medium' value for outdoor workers 1.9 CARB [1993): ' Moderate" value from laboratory protocol Residential Scenario 4.0 7 January 31,1998 9 A--_._____._- _ . _ _ _ _ - . -
Table 4.0.6. Reported Average Breathing Rates Correspond,ng to Activity Levels Typical of n Gardening Activities Breathing Reference Study and Activity Level Rate 2.0 Linn (1993):
- Fast" value for outdoor workers -
2.5 Layton (1993), Set 2: Median of *Hard' values over adult age groups 3.1 Linn (1992): " Fast" value for outdoor workers 3.6 CARB [1993):
- Heavy" value from laboratory protocol 4.3 Layton [1993), Se' ' Median of "Very Hard' values over adult age groups 4.3 Lays. J993), Set 1: Median of " Heavy" values over adult age groups As in the outdoor and gardening contexts, detailed time allocation information is not available for the variety of activities that might be conducted indoors. Time spent sleeping, however, is estimated in a number of activity surveys. Because a significant portion of indoor time is spent sleeping, and because of the low breathing rates characteristic of sleep, the average indoor breathing rate estimate distinguishes between the time spent sleeping and the time spent conducting other activities indoors:
O TgVg +T,V, l V V, = (4.0.3) where Ts and T4are the average time spent sleeping and awake indoors, and Vs and V, are the average breathing rates while asleep and awake indoors. Estimates for Ts and T4 are available from the National Hman Activity Pattern Somey (NHAPS)[Tsang ano Kepeis,1996)(see Section 3,7 for a discussion of time allocation studies). The EPA Exposure Factors Handbook desenbes Tsang and Klepeis [1996) as "the largest and most current human activity pattern survey available"[ EPA,1996). Over 9000 respondents provided minute-by minute 24 hour diaries between October 1992 and September 1994, and the responses weighted to produce results representative of the U. S. Population. Average time allocativ. . ms, as well as detailed distnbutionalinformation, is provided for a number of cohorts defined by age, race, gender, and other factors, however average values for adult males are not reported. The average time spent sleeping and napping by males of all ages is 523 minutes / day, while females spend an average of 529 minutes / day sleeping and napping. Adults of eithar gender between the ages of 18 and 64 spend an average of 497 minutes / day sleeping and napping. Because time spent sleeping depends on age more strongly than gender, a Ts value of 497 minutes / day was assumed for the screening group. The total time spent indoors (Ts + T4) by the average member of the screening group is 240 24-hour days / year, or 946 minutes / day (see Section 3.7). Residential Scenario 4.0-8 January 31,1998 m
The breathing rate while sl:eping, Vs, was estimat;d as the median of the values reported in Layton [1993) and from the CARB[1993) laboratory protocols,0 4 m3 /hr. Table 4 0.5 lists the reported brething rate values for activity levels expected to occur while awake indoors V4 was estimated as the median of these values,14 m 8/hr. The average indoor breathing rate was , then calculated from equation 4 0.3: 3
, 497 min / day 0 4m /hr -
449 min / day 14m3thr
=0.9m3 /hr (404) 946 min / day Table 4.0.8 summarizes the default breathing rate values for the three residential scenario exposure contexts. For comparison with breathing rate values recommended for other applications, the average long-term on site breathine; rate was also calculated using the average time spent in each context (see Section 3.7). The resulting long term breathing rate of 8
23 m / day is the same as that recommended for adult males in iCRP (1981), but larger than the 8 adult male breathing rate of 21.4 m / day based on EPA [1985)(see EPA [199A1 Table 5 20), and the more recent estimste f*om Layton (1993) of 17 m 8/ day. l Table 4.0.7. Reported Average Breathing Rates Correspcnding to Activity Levels Tyoical of I Waking Indoor Activities Breathing Reference Study and Activity Level , Rate j 0.5 Layton [1993), Set 1: Median of
- Sedentary' values over adult age groups 06 Layton [1993), Set 2: Median of ' Light" values over adult age groups 0.6 CARB [1993): " Sedentary' value from laboratory protocol 0.6 CARB [1993): ' Sedentary
- value from field protocol 0.8 Layton (1993). Set 1: Median of " Light' values over adult age groups 1.4 CARB [1993):
- Light' value from field protocci 1.4 CARB [1993): " Light' value from laboratory protocol 1.7 Layton [1993), Set 1: Median of " Moderate" values over adult age groups l
1.7 Layton [1993), Set 2: Median of ' Moderate' values over adult age groups 1.8 CARB [1993):
- Moderate
- value from field protocol 1.9 CARB [1993): " Moderate" value from laboratory protocol l
l l Residential Scenario 4.0-9 January 31,1998 9
Table 4.0.8. Default Breathing Rates for the Residantial Scenario Exposure Context / Breathing Rate Time Spent in Context' Parameter m3thr (dayslyear) Indoors V, 0.9 240 Outdoors V, 1.4 40.2 Gardening V, 1.7 2.92 Average On Site Rate' 23 m 8/ day
' See Section 3.7 8 Weightad by time spent in each context
REFERENCES:
Kenner.', Jr., W.E., and D.L. Strenge,1992.
- Residual Radioactive Conttmination froin Decommissioning: Technical Basis for Translating Contamination Levels to Annual Total Effective Dose Equivalent,' NUREGICR 5512, U.S. Nuclear Regulatory Commission, Washington, DC.
American Industrial Health Council (AlHC)(1994) Exposure Factors Sourcebook. AlHC, Washington, D.C.
'} California Air Resources Board (CARB)(1993) Measurement of Breathing Rate and Volume in Routinely Performed Daily Activities. Human Performance Lab, Contract No. A033 205.
June 1993. f International Commission on Radiological Protection (ICRP) (1981) Report of the Task Group on Reference Man, Pergammon Press, New York Layt_on, D. W. (1993) Metabolically Consistent Brualhing Rates for use in Dose Assessments. Health Physics, Vol 64 No 1,1993 M 23 36 Linn, W. S., D. A. Shamoo and J. D. Hackney (1992). Documentation of activity patterns in
'high-risk" groups exposed to ozone in the Los Angeles area in Proceedings of the Second EPA /AWMA Conference on Tropospheric Ozone, Atlanta. Nov 1991 pp 701-712. Air and Waste Management Assoc., Pittsburgh, PA.
Linn W. S., C. E. Spier and J. D. Hackney (1993) Activity Patterns in Ozone-exposed Construction Workers. J. Occ. Med. Tox.. Vol 2 No 1 pp 1 14 Spier,- C. E., D. E. Little, S. C. Trim, T. R. Johnson, W. S. Linn and J. D. Hackney (1992) Activity Patterns in Elementary and High School Students Exposed to Oxidant Pollution. J. Exp. Anal. Environ. Epid. Vol 2 No 3 pp 277 293 Residential Scenario 4.0-10 January 31,1998
%d l
U.S. Environmontal Protecton Agoney (EPA)(1996) Exposure Factors Handbook FPA RGport No. EPA /600/P 96/002Ba. Draft of August 1996 U.S. Environmental Protection Agency (EPA)(1985), Development of Statistical Distributions or Ranges of Standard Factors used in Exposure Assessments Washington. D.C. Office of Health and Environmental Assessment, EPA Report No. EPA 600/8 85 010 0 Residential Scenario 4.0-11 January 31,1998 9
5.1 Physical Parameters with C nstant Valu:s in this analysis physical parameters that do not have significant variability are held constant at the average value. Table 5.1.1 lists the physical parameters that are held constant and the value used in the parameter analysis. Additionalinformation was reviewed to determine the variability in the fraction of carbon in plante and animals. Although the data indicate little variebility in these parameters, the average values are slightly different than the initial default values. These data are presented in sections 5.1.1 and 5.1.2. The remainder of the parameters in table 5.1.1 are set at the NUREG/CR 5512 Volume 1 defaults (Kennedy and Strenge,1994). The plant concentration factors for the noble gases (BJAr,BjKr,BjRn & BjXe) and tritium (BjH) are set to zero because the gases are assumed not to accumulate in plant tissue and tritium is modeled separately. The outdoor shielding factor (SFO) is set to 1 because this scenario is evaluating surface soll contamination. Table 5.1.1 Constant Physical Parameters Parameters Desen ion i units i Value E T BTA7K..K.i.,.M.6.:X.6..".>d..ni.e.,6.if.i.tlo.s.f.ic.f.6.f.i.l57....a.".is.o.t:l.iD.. lo .l l
.. .. .. v ..
".(i.i.iE.'"..I'."
fc.a. ............ a.rb.o. .lr.a.. . i
.c i i
v .... 38 fc.a.] . . . . . .. . . . . . a. .r6. .o.t.t p .l f r . .tio.n. lo. .r.b.e.i e .c.a. t.t.e. ..................
....y...6..8.....
i 1 0.08.... . f c.a. 4. . , .a..rb . . . . . . . . . . . a.rb. .o.n. i lr.a. ct.i.o.n. l. . .r. .o. .
- e fc.a.
............. action o..r.m.. ilk.. cows .
. . i f.ed0 ra ct,o.n. lr.. . . . .a.n.s.......
ion . .a. of.ti.o.n.. c.ca rbon in s oil .
. . f o. .r,l . .. a1.e.r
... .. ........h. y .. ....... . .y..
.. . . . . . . _ . . .. . . . . . y . .
fc. i f.(i)" '". . ' ". ' '. .". .. . .. ". . alb. .i.ri .fi.s. .ct.ii. .ri.iii"~
. . lo.T. .itT. . l.6.r.i.
i..c. c.95 . . . . .a . . . . . . . . fo
. d. .r b.o.n.
.f . .a.n.im.a. . . . p.
r.a. . ct.i.o.n.
- l. 0o. .4......
.r.a.l.l .9 r.a.i n. c.o
- h. b a. . . . . . . . . . . . . . a.rb. .o.n. .f r.a. .r: tion ra!! ha i i fha.( .1). . . . . . . . . . . .. . d. action r ee cattle i i 0
. . . . r. . . R. y. .ro.g.e. n. .fr. . . 0, 1......
f h. a.( 2. ). . . . . . .a.c.t
. . i . .fo. . b. . . . . . . . .
fhaj 3)"""" "" '.d.r.o.g yd rog'e e.n. n .f.r. fra . . ct,io.n.
, . . . ion .fo .
for.p.o.u.l.tr.y .. . . . . . . . .11. . . . .. . . . . . . . . . . milk cows t . th. .a.(.............
< y.ii.rb. .g e.h. '.f.ri. .c'.t.id.6. 'f.o~i.'f "f.i.yE.
.~ Th'E h" " "
fhd.0... g . .. . .. .......g ra ctio.ngo.f h.,Ydro9.e n.,.in so,il . . . ... .. .. . . . . . ... . ....+p* .. ;..
.,......................+.......y.............
fh.p.,.f.h.............uA.y.......7,
.v.(4.). . . . . . . . . ; . .d.r.og.e.n.a.c.t.io.n.
~ .
.f.r. . . , . .f.o.r.a.
. . . . . . .ll.9 r a. in. . . . . . . . . . . . . . . . . . . . . . .!. . . . . . . . iv...00.68.....
fh.h. . . . . . . . . . . . . . . . . ,Hld.r99e.n.t.r. a. c.t.o.n. .fo. r. h. a Y . . . . . ............................,t.....0.1...... _ . . .. i i fhy(13 idrogen fracton for fruits and ve . 0.1
- k. i.H.,.i.e),V.e,~5.
. . - . P. i.,'s. n. ". . . y .a.rE.tio. .ni.d.i.f.fs.
. . . .. .. .. . y. i.e'rd.s.
y l5.i. H. ': .Y.e',V.i. ,g et
..H1 >
- - . . . . . . . . . . . . . . . . . T.h.ic.k.n.e.s.s. . . .o.f .s . u.rf.a.ce s.o.il l.a1.e.r. . . . . . . . . . . . . . . . . . . . . . 0 m i . .15. . . i. . . . .
.LA. .M.B. .D.W. . . . . . . . . . . . < . e.a.th.e.n.n.9 .ra. t.e. f.o.r.a.c.t.m.7.r.e.m.o.v.a.l.f.ro. m. . Et.a.n.ts. . . . . . . . . . . . . y . . . .
.. t i.d.. . i 4 95.E..02....
- .L.F.,.M.L.G. 4.M.L.H. t .M.L.V. . r .a.s.s. .lo.a.d.m. g.f.a.c.to.r.s.f.o.r .fo. r.a. ge.,.9r.a.in.,.h M M. . .
i i
- 00. (1). . . . . . . .. . . . . . . S_o.il. i.n.ta. k.e. .tr.a.c.to. .n.f.o.r.b.e.e.f.c.a.ttle. . . . . . . . . . . . . . . . . . . . . . . . . ,i. . . . . . . y . . . 0. 0 2. . . . . .
QD.(2). . . . . . . . . . . . . .w S.o. '. i.n.ta. k.e. .fr.a.c.to. .n. f.o.rP. .u.it.9. . . . . . . . . . . . . . . . . . . . . . . . . . . ,> . . ......y....01...... i
-O.D.(3). . . . . . . . . . . . . . : S.o.il. i.n.ta. k.e. .f r.a.c.to. .n.f.o.r.m. . ilk. .c. .o.w.s. . .........
Q%4).. . . . . . . . . .. . 4 S. oil.i.n.ta. k.e. .tr.a.cto. .n.f.o.r.la.1e. r. he. .n.s. . . . . . . w . i i
-.QH. (2). . . . . . . . . . . . . . >In. e. s.t.o.n. .r a. t.e. f.o.r.po.u.l.try.h.a.t . . . . . . i. . . . . . . . . . . . . . . . . . . . . y . .kg./d. . . y......0.......
i QH.(4).. . .. . .. . . . . . . in.9e. s.t.o.n. . rat.e.f.o.r.la1e. r .h.e.n.h.a.l . . . .. . . . . . . .. . . . . .. .. . . .. v . .k9.!d. . . * . . .. . 0. . . .. . QW11.L.............y:W y . .U. .d.. . y.... i 50...... QWl2. [ . . . . .:Wa.te.r.i.nge. . . . . . ..s.to. . ..n.r.a.te. e . .a.te.r.i.ng.e.s.to.
.fo.r. F. u. .tW. . . . . . . . . . . . . . . . .n.
. . . ..r.a.te. .fo.r.b.e. e. f. c. a. t.tl.e. . . . . . .
.....i 03.....-
QW13l . . . . . . . . . . . . . . . ate.r.in9'.ste. .n.r.a.te. 60.fo.r. m. .il.k.c.o.w. s i i QWL4F- .: Water ingeston ratt.for laJer hens ..................i..U..d...v..............i....0.3...... . O i Residential Scenario 5.1 1 January 29,1998 (
P ; D;senption . . . u. ni
.t.s. . . V. alue .
g .a a.m. . .t.e.r.s . . . .y..g.... .. .. ..
..P.............___...................,.........q.
_.......................fje..R......g 56!ac, , , , ,, .. , , , , , , , ,Speci appyttyp,qupaience . gforgliyestoj,k_
....... ,,,, ,,,,,,_,2_ . . ..
, , 1. , , , ,
salah ,,,,,,,, ,,,,,,,Trittum e,qujya[ence; anjmal producthntake ,,, , _ _ ,, , ,,, _ _ , ,, ,, _1, _ sag,h,,,,,,,,,,,,,,,T, ftt!um equjyaience pjant/ water ,,,, , , _ , _ , _ ,
;,_,,_ 1 , !.9. . . . . . . . . . . . . . . .Qutpppf Sh 9!d,ggaetof , , , , _ , _ , , , , , _ , , , , , _ , , , , , , , , , y, , _ , , _ , 1, , ,,, l Sh . . . . . . . . . . . . . .. . Abspjute, humyttrX . _ . _ . . . . . . _ . . _ . . . . . . . . . une3_ , _0 008, , , . l TL . ._ . . . . . . . . . . .T a! ns!qcaj9n fap,tet fpf,fpra ge , , , ,, ,, , , , , , . . . . , , , , , , , , , . ,, f,, , , , , , , . 1. , , , ,,
E.TJG TF,H u ,,,,,feedingpeflod for,all an!mals,., forage, gra[n & hay _,,,_,,,,, d ,,, L ,,,365 25,_ l T.FW ... ............ Water i.nges.tionper.iod..fo.r all anim.als.
. ......................d....i . . 365 25. ... .
............... g .....
.............................p.....g......
........... ................. .....$.......+..............
T.QG,,,,,,,,,,,,,,,Ngjgyg pfgMng ppppd,fpf,sjg[qq g!aip,,,,,,,,,,,,,,,,,, t,,,q,,, L ,,,,,99,,,,, T,QH_,,,,,,,,,,,,,ytnjgy!n gfgegg peppp,fpf,sjg[qq hay,,,,,,,,,,,,,,,,,,, g ,,,q ,,,;,,,,,{5,,,,,, T,Qy(1),,,,,,,,,,,,, giggyg g!gegg pefipp,fpf,leafagge(abjes ,,,,,,,,,,,,,, g,,,q,,, ,,,,,45,,,,, . T,Qy(2),,,,,,,,,,,,,ypjgyg g!gMgg peppp,fpf othp!ypggtables ,,,,,,,,,,,,,;,,,q,,,;,,,,,@Q,,,,,. TGy(3),,,,,,,, , ,,,, ygjgyg gfggng pe!pd fpf,f[ujs ,,, ,, ,,,,,,,,,,, ,,, , ,,,;,,,q ,,, ;;,,,,,? Q,,,,,.
.T.H. . . . . . . . . . . . . . . . . iT.r.a.n.sl.o.c.a.ti.o.n.f.a.c.
tor. f.o.r. ,h.a1. . .
....+.......
I.H,Ayj,,,, , ,, , ,,,,,[Holdyp,p,enod fg(pgg s,,,,,,,, ,,,,,,,,,,,,,,, ,,,,, ,,,,;,,,q,,, ; ,,,,
.T.V3 2. } . . . . . . . . . . . . . . .T.r.a.n.sloc.a.ti.o.n.f.a.c.to.r. f.o.r.o.t.h.e.r .ve9.e.ta.. b.l.e.s.
. . . . . . .. e. .. ... .. .. 0. .1.
TV,(3} ,, ,, , , , , , , , ,, , ,T!a n sip,cajpn f aptgt, for,f[uits, , , , , ,, , , , , , , ,, , , ,, , , ,, ,,,, , ,, , f,, , _ _ ,, ,,0 z ,1. ,,,, TV#1. . . . . . . . . . . . . . .Tra nsLqca39A f tactet of.g taas ,, ,,,,,,, . _ ,, _ ,,,, , ,,,,,,, ,,;,,,,, ,,,, 0,L ,,, . VSW Volume of water in surface water pond . L . 1.30E+06 WGi iill\lidi " " " "Wefd'ry"c6Me'si6Ff r acto 7f'ofsain' " ' " " ~ " " " " " " T " ~." ~ " Ts8 " " 5.1.1 Fraction of carbon in forage (f e,), stored grain (fca), and stored hay (fch) These parameters defines the mass fraction of elemental carbon in forage, stored grain and stored hay for livestock and is used in the agricultural pathway modelin the residential scensrio for calculating the dose from " C. The dose model assumes that the specific activity of " C in the animal product that is consumed by a human is equal to the specific activity of " C in the food the animal consumes. This section first includes brief discussions of the importance of fct ice and fcn with regard to the calculated dose and how fcs Ice and fen are specifically used in the dose model. Next the default values used for fcn Ice and fcn in NUREG/CR.5512, Volume 1, are discussed. Lastly, distributions for fen fce and fen are presented and values are proposed based on these distributions. The fraction of carbon in animal feed is important in estimating the dose from "C. The higher the value fcn Ica and icn the higher the total annual dose in the residential scenario. The default values for fen fe, and fen used in NUREGICR 5512, Volvne 1, dose modeling are 0.09. Additional information was reviewed to define the variability in fcn Ice and fen. The major Residential Scenano 5.1 2 January 29,1998 9
sourc:s of c rbon in foods are prot;lns, lipid., and carb )hydrat;s [Lehninger,1970). Therefore, the fraction of carbon in fntage, stored grein or stored hay can be determined based on the protein, lipid, and carbohydrate contents of the forage, stored grain or stored hay and the t fraction of carbon in proteins, lipids, and carbohydrates. The mathematical expression is given by: fe, = fp,(fe,) + f ,(fci) t + fc.(fce) (5.1.1) where fp,, ft ,, and fe, are the fraction of proteins, lipids and carbohydrates in the forage (x=f), stored grain (x=g) or stored hay (x=h), and fce. Ici and fee are the fraction of carbon in proteins, lipids and carbohydrates. Equation 5.1.1 is based only on the major sources of carbon in livestock feed and neglects minor carbon containing components such as vitamins and riucleic acids. Fraction of Carbon in Proteins. Linids and Carbohydrates Protein is a polyamino acid v.;.h a molecular weight ringe of t,,s00 to 40,000,000 and conssts of 50 to 340,000 amino acid monomer units. Proteina contain approximate.ly 50% carbon,7% i hydrogen,23% oxygen,16% nitrogen, and from 0 to 3% sulfur [Lehninger,1970). Lipids are esters of aliphatic acids and are composed of a hydrocarbon chain with a terminal carboxyl group linked to a acylglycerol moiety. The carbon composition of lipids varies slightly with the hydrocarbon chain length (14 to 24 carbon atoms in the fatty acid moiety) and the degree of saturation. Lipids contain approximately 76% carbon [Lehninger,1970). With the exception of milk, carbohydrates make up a very small portion of the total components v in animal products. Carbohydrates consist of carbon, hydrogen, and oxygen in the approximate CHO ratio of 1:2:1 and vary slightly in carbon content from 40% (simple sugars) to about 45% (storage and structural polysaccharides) (Lehninger,1970). Nutrient Comoosition of Forace. Stored Grain. and Stored Hav Tables 5.12, " 1.3 and 51.4 list common forage crops, stored grait, crops, and stored hay crops for livestock and the quantities of protein, l;r ds, and fibers in each. Fibers include structural polysaccharides and other carbohydrates. These three major components in the crops are readily digestible by livestock. There are, however, minor components of non-digestible proteins and fibers present in plant material [NRC,1985).
= _ _ _= _
. g
Tcble 5.1.2 Compositi:n cf Fresh Fer:ge Cr:ps [ NAP,1996] Forage Crop Protein Lipids Fiber Alfalfa 18.9 3.2 77.9 Bermuda grass 12.6 3.7 83.7 Bluegrass 17.4 3.5 79.1 Broome grass 21.3 4.0 74.7 Canary grass 17.0 4.1 78.9 Clover, Ladino 25.8 4.6 69.6 Clover, Red 14.6 2.9 82.5 Fescue 15.0 5.5 79.5 Orchard grass 12.8 3.7 83.5 Rye grass 17.9 4.1 78.0 Trefoil 20.6 4.0 75.4 Timothy 12.2 3.8 84 ^ Table 5.1.3 Composition cf Grain [ NAP,1996) Grain Frotein Lipids Fiber l Barley 13.2 2.2 84.6 l Canola 30.7 7.4 61.9 l Corn 9.8 4.1 86.1 Oats 13.6 5.2 81.2 Sorghum 12.6 3.0 84.4 Wheat 14.2 2.*J 83.5 Taple 5.1.4 Composition of Stored Hay [ NAP,1996) Hay Crop Ptotein Lipids Fiber Alfalfa 18.6 2.4 79.0 Bermuda grass 7.8 2.7 89.5 Broome grass 6.0 2.0 92.0 Canary grass 10.2 3.0 86.8 Clover. Ladino 22.4 2.7 74.9 Clover, Red 15.0 28 82.2 Corn w/ cob 2.8 0.6 96.6 Corn silage 8.7 3.1 88.2 Fescue 10.8 4.7 84.5 Residential Scenario 5.14 January 29,1998 9 l i
Orchard grass 12.8 2.9 G4.3 Sorghum silage 9.4 2.6 63.0 Wheat grass 8.7 2.2 89.1 Wheat silage 12.5 6.1 81.4 Trefoil 15 9 2.1 82.0 Timothy 10 8 2.8 )86.4 The largest variability in the fraction carbon parameters is due to the variety in the types of forage crops, stored grain crops and stored hay crops that livestock may eat. To cccount for this variability it is assumed that each type of potential feed is equally likely to be fed to livestock. Therefore, a uniform distribution representing each type is sampled from to obtain the specific crop being consumed by the livestock. Given the specific feed type, the amount of nutrients can be determined from Table 5.1.2, Table 5.1.3, and Table 51.4, combined with the specific fraction of carbon in the nutrients reported by Lehninger [1970) to calculate the mass fraction of carbon, using Equation 5.1.1. The variability in fe,, fce and fcn is not significant. In addition, Table 5.1.5 presents the default values used in NUREG/CR 5512, Volume 1, which are consistent with the distributions derived in this section. Given the lack of variability, the mean values will be used for the screening calculation. Table 5.1.5 Data Variability for fe,, fe, and fcn l Parameter Minimum Maximum Mean Standard NUREG/CR-Deviation 5512 default fe, 0.088 0.14 0.11 0.018 0.09 fe, 0.39 0.44 0.40 0.016 0.40 fen 0.020 0.12 0.07 0.031 0.09 These parameters would not be expected to vary froni site to site and it is very unlikely that a licensee would conddCt any type of data Collection activity to modify them. The one exception may be ic, because of the different forage crops that grow in different regions throughout the U.S. Therefore, there is a slight chance that a licensee may attempt to support alternative values for the fraction of carbon in forage based on regional data that supports specific forage crop growth. 5.1.2 Fraction of carbon in animal products, fe, This parameter defines the mass fraction of elemental carbon in mest (beef and poultry), milk, and eggs and is used in the agricultural pathway model in the residential scenario for calculating the dose from "C. The fraction of carbon in these animal products is a physical parameter because it is a function of the physical amount of "C in the specific animal product being considered. Residential Scenario 5.1 5 January 29,1998
The fraction of carbon in animal products is important in estimating the dose from "C. The high:r th2 value for fc.. the higher the total annual ducs in tha residential scenano. The default values for this parameter used in NUREG/CR 5512. Volume 1, dose modeling are; beef cattle 0.24; poultry,0.20; milk 0.07: and eggs. 0.15. The major sources of carbon in foods are proteins, lipids, and carbohydrates [Lehninger,1970). Therefore, the fraction of carbon in foods can be determined based on the protein. lipid. and carbohydrate contents of the food and the fraction of carbon in proteins, lipids, and carbohydrates. The mathematical expression is given by: fc " fe.(f co
) + f (fci) a + fc.(fee) (5.1.2) where fp,, fa, and fe, are the fraction of proteins, lipids and carbohydrates in food type 'a',
respectively, and fco fci and ice are the fraction of carbon in proteins, lipids and carbohydrates, respectively. Equation 5.1.2 is based only on the major sources of carbon in foods and neglects minor carbon containing components such as vitamins and nucleic acids. Fraction of Carbon in Poteins Licids and Carbohydrates . Protein is a polyamino acid with a molecular weight range of 6,000 to 40,000,000 and consists of 50 to 340,000 amino acid monomer units. Proteins contain approximately 50% carbon,7% hydrogen,23% oxygen,16% nitrogen, tnd from 0 to 3% sulfur by mass [Lehninger,1970). Lipids are esters of aliphatic acids and are composed of a hydrocarbon chain with a terminal carboxyl group linked to a acylglycerol moiety. The carbon composition of lipids vary with the hydrocarbon chain length (14 to 24 carbon atoms in the fatty acid moiety) and the degree of saturation. Lipids contain 73 to 79 % carbon and 11.6 to 12.8% hydrogen by mass [Lehninger, 1970). With the exception of milk, carbohydrates make up a very small portion of the total components in animal products. Carbohydrates consist of carcon, hydrogen, and oxygen in the approximate CHO ratio of 1:2:1 and vary in carbon content from 40% (simple sugars) to about 45% (storage and structural polysaccharides) by mass [Lehinger,1970). Nutrient Comoosition of Animal Produced Foods Table 5.1.6 lists the nutrient composition of products from beef cattle, poultry, milk cows, and layer hens. Residential Scenario 5.1 6 January 29,1998 9
Table 6.1.6 Compositi:n cf Animal Products [G:bhardt,1985) Product Protein Lipids Carbohydrates (g) Milk 3.3% 3.3% 4.5% Eggs 12 % 12 % 2.0% Beef 26 % 31 % 0 Pou!!ry 31 % 35% 0 The only uncertainty in the data is in the carbon content of lipids (73 79%) and the carbon content of carbohydrates (40 45%). Because there is no basis for any type of distribution of this uncertainty indicated by Lehninger (1970) these fractions are assumea to be uniformly distributed with the minimum and maximum values equal to the reporte range. Using these ! distributions and Equation 5.1.2, Table 5.1.7 presents the data for fc. for milk, eggs, beef and pouhry. along with the defat: values useo in NUREG/CR 5512. Volume 1. Table 5.1.7 Data on variat ility of fe, Product Minimum Maximum Mean of the Default Value. Range from 5512 Milk 0.0606 0.0626 0.06 0.07 Eggs 0.157 0.164 0.16 0.15 Beef 0.353 0.371 0.36 0.24 Poultry 0.182 0.185 0.18 0.20 As can be seen in Table 5.1.7, there is little variability in fc., especially given the accuracy of the data from which they are derived. Therefore, the average of the data will be used in the screening cal.lation.
REFERENCES:
Gebhardt, S. E. and R. Matthews,1985. ' Nutritive Value of the Edible Part of Foods', Home and Garden Bulletin No. 72, Human Nutrition and Information Services, U.S. Department of Agriculture. Kennedy, Jr., W. E., and D. L Strenge,1992 'ResMual Radioactive Contamination from Decommissioning: Technical Basis for Translating Contamination Levels to Annual Total Effective Dose Equivalent," NUREG/CR 5512, U.S. Nuclear Regulatory Commission, Washington, DC. Lehninger, A. L.1970. Biochemistry: The Molecular Basis for CeII Structure and Function, Worth Publishers, Inc. Residential Scenario 5.17 January 29,1998
National R:s arch Council (NRC),1985. Ruminant Nitrogen Usage. Washington, D.C.: National Acrd;my Pr:ss. NAP,1996. Nutrient Requirements of Beef Cattle: Seventh Revised Editic ,, Gubcommittee on Beef Cattle Nutntion, National Academy Press. O l Residential Scenano 5.1 8 January 29,1998 9
5.2 THICKNESS OF THE UNSATURATED ZONE, H,(m) p As defined in NUREG/CR 5512, Volume 1 [ Kennedy and Strenge,1992), H,is the thickness of the unsaturated zono for the three box ground-water model used in the residential scenario. The top box in the Tree box model represents a soillayer consisting of 15 cm of soil. The middle box represents the unsaturated zone. H2 is the thickness of this middle box. The bottom box represents the saturated zone or aquifer. H,is a physical parameter that is a characteristic of the specific site being assessed and is independent of the source term and group of exposed individuals. This section first included brief discussions of the importance of H, with regard to the calculated dose and how H,is specifically used in the three box ground water model. Next, the basis for the default value for H, recommended in NUREG/CR 5512, Volume 1 is discussed and general information regarding uncertainties associated with H, are presented. Lastly, a distribution of H,, representing the variability of unsaturated zone thickness throughout the U.S. to be used for the screening calculation is presented. The thickness of the unsaturated zone is important to dose because it is the distance radionuclides must travel to get into the saturated zone. Once in the saturated zone, the radionuclides contaminate drinking and irrigation water which results in a dose to man via several different possible pathways. A thick unsaturated zone compared to a thin unsaturated zone would provide a longer distance for radionuclides to be transported. This longer distance translates into a longer travel time and, with radioactive decay occurring, may result in a decrease in the amount of radionuclides reaching the saturated zone. Besides travel distance, the unsaturated zone has a wide variety of characteristics (e.g., adsorption rates, water content, hydraulic conductivity) that affect the transport of radionuclider Parameters that represent ( these characteristics combined with H, provide the basis for ostimating the total amount of N radioactivity that reaches the saturated zone in a given tiinc. For NUREG/CR 5512, Volume 1 dose modeling, the thickness of the unsaturated zorie is used in determining radionuclide leach rates from the unsaturated zone to the saturated zone in the three box ground water model. These teach rates are proportional to the amount of water that infdtrates into the unsaturated zone Onfiltration rate) and inversely proportional to the thickness of the unsaturated zone, the volumetric water ce"?nt of the unsaturated zone, and the radionuclide specific retardation factor (which is derived from adsorption coefficients). The mathematical relation between leach rate and unsaturated zone thickness is given in NUREG/CR 5512, Volume 1 ( page 4.9) as:
=
I L.n"
#20 2Rt2,365.25 5.2.1 where:
La = leach rate from the unsaturated zone to the saturated zone for radionuclide j (yd) l = Infiltration rate (myd ) H, = Unsaturated zone thickness (m) Q Residential Scenario 5.2 1 February 2,1998 Y] i
22 a Volumetric wat:r content of the unsaturated zone (dimension less) Rt2, a Retardation factor for movem;nt of radionuclide j from the unsaturated zone to the saturated zone (dimension less) The retardation factor is given in NUREG/CR 5512, Volume 1 (page 4.9) as: Rtn = 1 +
/"P'
- 5.2.2 where:
Kd,, = Partition coefficient for the jth radionuclide in the unsaturated zone D, = Bulk density of the unsaturated zone n, = Total porosity of the unsaturated zone The default value for H, defined in NUREG/CR 5512, Volume 1, is 1 m, which is representative of a thin unsaturated zone. It was argued that a thin unsaturated zone would be conservative because this would result in relatively fast travel times through the unsaturated zone which would allow for more radionuclides to reach the saturated zone. However, when contaminant transport is coupled with radioactive deczy, it is difficult to define a priori whether or not a thin unsaturated zone is conservative. For example, a short travel time through the unsaturated zone would not allow for ingrowth of a particularly toxic daughter product. Information concerning depth to the water table is a commonly reported quantity given the large number of observation wells located throughout the U S. For example, in New Mexico, there are 33,000 observation wells where data is regularly collected [USGS,1990). H: 'ver, there is no readily available summary digital database for the continental U.S. A report by USGS [USGS,1990), available on CD-ROM, does oresent State Water Data Reports from USGS observation wells throughout the continental U.S. This information is based on removing text from USGS open file reports. Therefore, there are inconsistencies in what data is reported and how it is reported from state to state. In addition, information from the western U.S. states is particularly sparse, especially compared to the dense coverage of the eastern U.S. states. For those areas whe'e data is especially sparse, additional references were found [ldaho Department of Water Resources,1998; USGS Colorado 1998, Wyomir g Water Resources Center,1997) on a state, groundwater region or local basis. The only groundwater region where specific well data could not be found was the Columbia Plateau. However, Guzowski et al. [1981) do provide summary water table depth information from this region, which was used to confirm that the resulting distribution included that range. Despite these problems with data availability, the combined data set is believed to be appropriate for representing the variability of unsaturated zone thickncss throughout the U.S. for the screening calculation. To use the water table depths to generate a probability distribution function of H2 from the referenced material, a 1.5 degree grid is overlayed onto a map of the continental U.S, which delineates the USGS groundwater regions [ Fetter,1988). The coarseness of the grid is chosen based on approximating the density of grid points per groundwater region to the areal density of the groundwater regions. The areal densities and grid point densities for the groundwater regions are presented in Table 5.2.1 and Table 5.2.2, respectively. Residential Scenario 5.2-2 February 2,1998 9
Table 5.2.1 USOS Ground Water Regions Areal Density (
\
Groundwater Region Area in Square Kilometers Percent of Total Area Alluvial Basins 1016791.19 13.06 Atlantic and Gulf Coastal 889928.98 11 43 Plain Colorado Plateau and 464019.23 5.96 Wyoming Basin Columbia Lava Plateau 369217.96 4.74 Glaciated Central Region 1253496.30 16.10 High Plains 382559.85 4.92 Nonglaciated Central Region 1859575.84 23.89 Northeast and Superior 379291.25 87 Uplands Piedmont and Blue Ridge 230726.81 2.96 Southeast Coastal Plain 194674.84 2.50 l Western Mountain Ranges 743214.91 9.55 O Table 5.2.2 USGS Ground Water Regions Gridded Sampling Point Density Groundwater Region Number of Grid Points Percent of Total Number of Points . Alluvial Basins 46 12.81 Atlantic and Gulf Coastal 38 10.58 Plain Colorado Plateau and 20 5.57 Wyoming Basin Columbia Lava Plateau 21 5.85 Glaciated Central Region 61 16.99 High Plains 16 4.46 Nonglaciated Central Region 89 24.79 Northeast and Superior 17 4.74 _ Uplands __ 1 Residential Scenario 5.2 3 February 2,1998 J l
Tcble 6.2.2 USCS Crcund Water Regi:ns Cridded S:mpilng P;lnt Density l Groundw;t:r R:gion Number of Grid Points Percent of Total Number of
, Points Piedmont and Blue Ridge l 10 l2.79 Southeast Coastal Plain 8 l 2.23 Western Mountain Ranges 33 l 9.01 To associate a water table depth with a grid point location, the closest well to the grid point is used to assign a value of the water table depth to the grid point. For the eastem states, wells are typically found within a 20 mile radius of the grid point. West of the Mississippi River, wells are typically found within a 5(
mile radius of the grid point. This process is chosen, as opposed to interpolation, in order to be consisten within a groundwater region (i.e., to avoid interpolating across groundwater regions) and because the resulting probability distribution is meant to represent the variability across the U.S. and not specific value at specific locations. The spsfic depth to water leva assigned to the specific grid point is an average of the highest and lowest water levels reported at the associated well, and therefore, re'"asents the long term a,erage of extremes. Values were not found for every grid point. l.. stead the search for values continued until the a representative number of values was found for each ground water region, based on the sampling point densities presented in Table 5.2.2. Figure 5.2.1 illustrates the 1.5 degree grid, along with the wells that were used to assign value to the nearest grid points. The exception to the data analys process defined above is for Wyoming, where the data that was obtained was a depth to water 2 dimensional surface. Therefore, the values at the surface that corresponded directly to the grid point locations were used. The resulting data set of H2ranged from a minimum of 0.3 meters in the High Plains ground water reg:on (a wellin north central Nebraska) to a maximum of 316 meters in the Alluvial Basins ground water region a woll on the south rim of the Grand Canyon), with an average depth to ground water of 22 meters for the continental United States. The proposed empirical probability distribution and cumulative probability distnbution of unsaturated zone thickness, H,, are shown in Figure 5.2.2 and Figure 5.2.3, respectively. An empincal distnbution was chosen due to lack of a statistical basis for choosing any other type of know distnbution. PARAMETER UNCERTAINTY: With regard to uncertainty, the thickness of the unsaturated zone is mor appropriately thought of as its equivalent, depth to the water table. Depth to the water tale often follows the general shape of the topography, although the water table relief is not as great as the topographic relief. In addition, water table depth is a function of time dependent variables such as seasonally variabit recharge rates and pumping rates. Information pertairang to water table depth are often available, cspecially at the state and city govemment level, because this data is critical to public water resource management and planning. Therefore, it is expected that a licensee would easily have this information cvailable and be able to apply a site specific value in the dose model.
REFERENCES:
Kennedy, Jr., W.E. and D. L. Strenge 1992. ' Residual Radioactive Contamination from Decommissionir Technical Basis fur Translating Contamination Levels to Annual Total Dose Equivalent," NUREG/CR 555 Volume 1, U.S. NRC, Washington, DC. Residential Scenario 5.2 4 February 2,1998 O
USGS,1990. 'Annu;l State Wat:r. Data Reports: A Digital Representation of the Hydrologic R cords of the United States,'Open File Report 624 H.119.76, USGS, Denver, Colorado. Idaho Department of Water Resources,1998. File well4.e00, accessed through URL ftp://ftp. state.id.us/ pub /gisdatalgwm/. USGS Colorado,1998.
- Water Resources Data for Colorado, Water Year 1996," accessed through URL http://webserver.cr.usgs. gov / publications /datareportfile/,
Wyoming Water Resources Center,1997.
- Depth to Initial Ground Water in Feet," accessed through URL http://www.sdvc.uwyo.edu/ clearinghouse /ddgw.html/.
Guzowski, R. V., Nimick, F. B., and Muller A. B.,1981, ' Repository Site Definition in Basalt: Pasco Basin, Washington,' NUREG/CR 2352, U.S. NRC, Washington, D.C. Fetter, C. W.,1988. Anolied Hydroneoloav, Macmillan Publisi:Ing Company, New York, New York.
\
_. - _ _ _ = _ _ = _ ___ f Residential Scenario 5.2 5 February 2,1998 I L
i.s p r onee.s % u,r .weie usos on,wie. % EW ,, ' ; P.- . ." e i
. . id l *
[: au L. .s.
- i. i i
.,g3:==:
~ ~ ~
M l U~'._._
~
Figure 5.2.1 Gridded sampling and well locations within the USGS ground water regions. l O l Probability Density l 00% , j Legend i 0 0J D et. j 003 om f(x) i 0 015 - l 0 01 - 0 005 - - OC ' '
.u)0 OO SO 0 100 0 150 0 300 300 0 3500 Unsat. rone thickness (m)
Figure 5.2.2 Proposed H, empirical probability distribution.
%sNential Scenario S.2 6 February 2,1998 O
o_ - -- -
Cumulative Density IO O8 f 06
,m
. 04 02 00 50 0 100 0 150 0 200 0 260 0 300 0 350 Unsat. rene thickness (m)
Figure t .2.3 Proposed H, empirical cumulative distribution function. l O Residential Scenario 5.2-7-- February 2,1998
5.3 Hydr: logic Parameters: S:ll Texture, Per:siti:s (ni, n ), R:lative Saturati:n (i ii .f ), infiltrati:n (I), Bulk Densiti:s ( pi, pa) and S:ll Ar:al D:nsity (Ps) Severalinput parameters represent charactenstics of the surface soil or the soil of the unsaturated layer. These parameters include porosity and saturation ratio. Rather than sample independently from distnbutions of these parameters, the dependence of these parameters is represented by first sampling soil texture then selecting an appropnate distnbution for porosity and saturation ratio for the sampled texture Soil densit.es are tied to the soil texture by a functional relationship to porosity, A common method of describing and quantifying soil texture is the USDA soil textural classification (National Soil Survey Handbook,1997). This classification was used by Meyer and others [1997] to represent the variability of a number of soil hydrologic properties that are related to porosity and saturation ratio. The USDA soil textural classification is also reported in a variety of available electronic data bases for the United States. Porosity (n)is a measure of the relative pore volume in the soil. It is the ratio of the ve' 9e of the void, to the total volume: n= , " " " =, "*",',- (5.3.1) anaal er, wear, ud , Soil bulk density (p) represents the ratio of the mass oi uaed soil to its total volume (solids and pores together): l Al wa Af wa '
=
P (5.3.2) l*,... l '.,,
- l '. .,,,
- l ',.a It is assumed that for each realization the porosities in the surface soil layer and in the unsaturated layer will be equivalent. The same holds true for the bulk densities. That is:
n i =n: (5.3.3) l Pi
- P: (5.3A)
Soil areal density of the surface plow layer is a measure of the mass of soil per square meter in the surface layer. The depth of this layer is assumed to be 0.15 m in the DandD model. The infiltration rate is measured as the volume of water per unit area per uriit time that percolates deeply beneath the root zone and becomes infiltration. It is the effective rate et which water moves through the surface soil layer and through the unsaturated layer, as well as the rate at which the aquifer receives recharge water. Its units are given as length / time. The saturation ratio (f) expresses the volume of water relative to the volume of the pore space. Residential Scenario 5.3 1 January 31,1998 1 0
l'" "' (5.3.5)
/= l'.n
- l'. ....
/= 0/n (5.3.6)
It is also a ratio of the moisture content (0) to the porosity, it is assumed that for each realization the saturation ratios in the surface soil layer and in the unsaturated layer will be equivalent. That is:
/, */ (5.3.7)
IMPORTANCE TO DOSE AND USE IN MODELING: The hydrologic parameters control the rate at which the contaminant is leached out of each layer and is transported into the next layer, Soil texture is not used directly in the modeling; it is used to determine the active distribution for the directly related parameters; porosity and saturation ratio. The following equation is a generic representation of the teaching model (NUREG/CR 5512 Volume 1, equations 4.7-4.12, pages 4.8 4.9). I (5.3.8) Laj _ H,OuRta365A Where L is the leach rate for layer k and contaminant j, H is the layer thickness, O is volumetric 'd moisture content, Rt, is the retardation factor 365.25 is a time unit conversion factor and I is the infiltration rate (m/y). The retardation coeficient is a function of the partition coefficient (Kd), porosity (n) and bulk density (p) and the volumetric moisture content ( A) is a function of the sampled relative saturation and the porosity: (6'3'I Rt. = 1 + - n (5.3.10) g _. fg The effect of the hydrologic parameters on the dose is uncertain due to uncertainty in the dominant exposure pathway. 5.3.1 Information Reviewed to Define the Distribution of Soil Yexture The CONUS SOIL database created and electroni< ally accessible through Pennsylvania State University (from http://www. esse.psu.edu) is a composite summary of detailed soi! databases (STATSGO databases) for states in the continental United States. This CONUS-SOIL database generalizes a variety of soils data, including the USDA soil texture, on a 1 Km grid ResidentialScenario 5.3 2 January 31,1998 {C
with c:nstant layering. The lay: ring consists of two 5 cm. thick layers near th) land surface follow;d by thr:010 cm. layers, thise 20 cm. lay:rs and finally thre) 50 cm. layers. In general, the total area of each texture class is fairly consistent from layer to layer with the clay content tending to increase slightly with depth. Since the uppermost soillayer in the DandD conceptualization is 15 crn. thick the 3 uppermost CONUS Soillayers were examined for uniformity and consistency. Approximately 85 percent of the area covered by matenals with USDA classified soil textures is a consistent texture for the 3 uppermost layers. Table 5.3.1 summarizes the areal distnbutions of textures for the 3 upper layers individually and the volume weighted distnbution of textures for the 3 layers combined. Table 5.3.1 CONUS SOIL Texture Summary USDA Layer 1 Layer 2 Layer 3 Volume Weighted % of 0-Soil (0 Scm) (510cm) [10 20cm) 20 cm Texture (% of area) (% of area) (% of area) silt 0.005 0.005 0.015 0.01 sandy clay 0.000 0.065 0.216 0.124 sandy clay loam 0.398 0.650 1.323 0.923 sitty clay 1.569 1.623 1.316 1.456 loamy sand 3.822 3.719 3.540 3.655 clay 3.525 3 845 5.766 4.726 clay loam 4.385 4.706 6.003 5.274 sitty clay loam 4.578 4.734 5.407 5.032 I ( sand 7.267 7.188 7.385 7.306 sandy ivam 23.541 22.673 21.792 22.450 sitt loam 25.339 25.336 24.424 24.881 loam 25.571 25 456 22.C13 24.163 PDF for Soll Texture The proposed probability distribution for soil texture is related to the volume weighted distnbution of soil texture for the first 3 layers of the CONUS SOIL database. The pr6bability of encountering a specific soil texture is equal to the percentage of the volume occupied by a this soil texture. For example, the probability of the site having a sitt learn soil texture is 24.881 percent. Information to Support PDF for Porosity Residential Scenario 5.3-3 January 31,1998 O
N:rmal distributions of porosities (assumsd to ba equivalent to saturat:d watzr cont nt) are given in Carsel and Parrish [1988). They are reported based on the 12 Soil Conservation r Service textural classifications and a compilation of data for each of the textural classes. These I, distributions are used in the parameter analysis. The means and standard deviations for these + normal distributions are given in Table 5.3.2. Table 6.3.2 Distributions for Porosity based on Soll Texture (after Carsel and Parrish,1988) Soil Type Mean Standard number of Deviation samples sand 0.43 0.06 246 , loamy sand OA1 0.09 315 4 sandy loam 0.41 0.09 1183 sandy clay loam 0.39 0.07 214 soam 0 43 0.10 735 it silt loam 0.45 0.08 1093 Silt 0.46 0.11 82 clay loam 0.41 0.09 364 silty clay loam 0.43 0.07 641 O sandy clay 0.38 0.05 46 silty clay 0.36 0.07 3/4 clay 0.38 0.09 400 Soil Bulk Density and Areal Density Bulk density is functionally related to porosity: p=(I-n)p, (5.3.11) where p is the soil bulk density (g/cm'), n is the porosity, and p,is the particle density (g/cm'). In most soils the mean particle density is very close to the density of quartz (2.65 g/cm2), typically the main com,nonent cf sandy soils. Clay minerals have a similar density. While the presence of heavy minerals such as iron oxides can increase the mean particle density or the presence of organic matter can lower it, as a practical matter mean particle density generally varies between 2.6 and 2.7 g/cm'[Hillel,1980) and can be represented as a constant of 2.65 g/cm'. With that, the bulk density becomes: ResidentialScenario 5.3-4 January 31,1998 O. ~
,+
p_-- gg--_, ,,---..w- _ , . ,.._e,.. ._p
- _ - _ y , 7_g.p.e._,_-..e. .wy9.- c
-w . ---vi. - . y - T
p =(1 -n) 2.65 (5.3.12) The soil areal density of the surface plow layer, P,(kg/m2), is a function of the bulk density (and hence the porosity). Actually, it amounts to nothing more than a conversion of units from the bulk density along with an assumption of a 0.15 m plowing depth. Mass is converted from grams to kilograms. Volume is converted from cubic centimeters to an area (in square mete.s) times an (implicit) depth of 0.15 meters: P, = 150 p (5.3.13) P, =3 97.5(1 -n) (5.3.14) Variability in infiltration rate Infiltration rate is a function of the amount of water applied to the land surface (either by precipitation or irrigation) and the soil hydraulic conductivity which controls the rate at which the soil is able to drain. To determine infiltration rate (I) we assume a modelin which the infiltration rate is the product of the epplication rate (AR) and the fraction of the applied water that will percolate deeply beneath the root zone and become infiltraiion. (The infiltration fraction is designated as IF.) The infiltration fraction is a function of the saturated hydrauSc conductivity (K,,). l =.4 R */F(K,,,) (5.3.15) O Distnbutions of saturated nycraulic conductivity are given in Carsel and Parnsh (1988). They are reported based on the 12 Soil Conservation Service textural classifications. Carsel and Parrish [1988) fitted distributions from a class of transtr med normal distributions. Meyer et al. [1997) refitted the distributions of Carsell and Parrish (1C) to distr.butional form:, : hat are more commonly used and more easily constructed either lognormal or beta. The lognormal distnbution is completely specified by the mean and standard deviation while the beta distribution is completely specified by mean, standard deviation, and range (upper and lower limits of the distribution). The distribution type and parameters for these distributions for each of the 12 soil types are given in Table 5.3.3. Residential Scenario 5.3 5 January 31,1998 9
Table 5.3.3 Saturated hydraulic conductivity distributions. Soil Type distribution Mean Standard bwer limit upper limit number of type (cm/s) Deviation samples sand beta 8.22E-03 4.49E 03 3.50E-04 1.86E-02 246 loamy beta 3.99E 03 3.17E-03 3.90E 05 1.34E 02 315 sand sandy lo,n$rmal 1.17E-03 1.37E 03 1183 loam sndy clay lognormal 3.23E 04 5.98E 04 214 le ra loam lognormal 2.92E-04 4.91E 04 735 slit loam lognormal 9.33E-05 2.24E-04 1093 silt lognormal 4.89E 05 2.76E 05 88 clay loam lognormal 9.93E 05 2.51E 04 345 sitty clay lognormal 1.54E 05 3.38E 05 592 loam sandy clay lognormal 3.55E 05 1.48E 04 46 V sitty clay lognormal 2.19E-06 4.08E 06 126 clay lognormal 3.65E-05 1.08E 04 114 The U.S. Bureau of Reclamation has developed an empirical relationship between soit permeability and the proportion of water that percolates beneath the root zone [USBR.1982) (shown in Figure SM.1 and in Table 5.3 4). Table 5.3.4. USBR relationship between soil permeability and Infiltration fraction. saturated hydraulic conductivity deep percolatien (inches /hr) (cm/sec) (percent) 0.05 3.53E 05 3 0.10 7.06E-05 5 0.20 1.41E-04 8 0.30 2.12E-04 10 Residential Scenario 5.3-6 January 31,1998
Table 5.3.4. USBR relationship between soll permeability and Infiltration fraction. . saturated hydraulic conductivity deep percolation (inches /hr) (cm/sec) (percent) 0.40 2.82E 04 12 0.50 3 53E-04 14 0.60 4.23E 04 16 0.70 A 94E-04 18 1.00 LO6E-04 20 1.25 8.82E-04 22 1.50 1.06E-03 24 2.00 1.41E 03 28 2.50 1.76E-03 31 3.00 2.12E-03 33 4.00 2.82E-03 37 1 1 Having now developed a relationship for the propensity of soil to drain based on its ability to transmit water, we now cunsider water application rates. Water applications at a particular site will always be equal to or exceed the annual precipitation (assuming negligible runoff). The distribution for precipitation is given in Figure 2. This distribution was derived by interpolating a precipitation surface using average precipitation data obtained from weather stations across the conten..Inous United States [ France,1992; Owenby and Ezell,1992). In humid regions of the country, precipitation supplies sufficient moisture to grow garden crops. In semi-arid or arid regions however, precipitation alone does not supply sufficient moisture to meet the risquirements of garden crops. This water deficit must be met through the application of irrigation water, in determining minimum water requirements, we considered crops grown in arid regions because data are available for irrigation rates and obtaining data for total application of water (irrigation plus precipitation) is more problematic. Under arid conditions, irrigation water alone is sufficient to meet or nearly meet the crop water requirements since the contribution of precipitation in meeting the crop water requirements will be small to negligible. For this exercise, we considered irrigation rates for idaho (USDC,1994). Idaho data was used for several reasons. Its main commercial crop, potatoes, has similar water requirements to small vegetables typically grown in a home garden. (In fact, potatoes are commonly grown in home gardens.) Its climate is arid such that the vast majority of water for crops is supplied by irrigation. And its position along the Northem border of the country give it a single-crop growing set. son. Idaho applies just under 2 acre-feet of irrigation water per acre per Residential Scenario 5.3-7 January 31,1998
year. As a compari n, water requrements for small vegetables, melons, and corn in New Mexico were also considared [USBR,1997). These requirements range from 17 to 30 inches of G water depending on the crop and the soil type, with an average requirement of about 24 inches (2 feet) of water, equivalent to the Idaho data. Based on this data, a cumulative distribution for application rate is presented in Figure 5.3.3 and Table 5.3.5. For all precipitation rates at or above the minimum crop requirement of 2 feet of water, the application rate is considered to be equal to the precipitation rate. For all arid and semi arid regions having precipitation rates of less than 24 inches, water application rates are assumed to be equal to 24 inches. An additional logical condition is that the sampled water application rate at a particular site should never be less than the irrigation rate. If the sampled application is less than the irrigation rate, then the application rate is set equal to the irrigation rate. If AP<lR, then AP =lR (5.3.16) Based on the preceding discussion, the steps to determining infiltration rate are as follows:
- 1. Sample soil type.
- 2. Sample a saturated hydraulic conductivity for that soil type.
3, Given the sampled hydraulic conductivity, use the USBR relationship relating soil conductivity to the infiltration fraction to determine the infiltration fraction. (Some interpolation or extrapolation may be required.)
- 4. Sample an application rate.(if AR<lR, then AR=lR.)
9 5. Calculate infiltration rate from the relationship, I = AR
- IF. In some cases, the pretence of low permeability soils will prevent infiltration to occur at ine calculated infiltration rate. The rate of water infiltration can be limited by the soil's abi'ity to transmit water. The most favorable conditions for transmitting water through soils occur under saturated conditions and a unit gradient. In this case, the rate at which water can be transmitted is equal to the soil's saturated hydraulic conductivity.
- 6. Compare (in consistent units) the infiltratica rate to the saturated hydraulic conductivity. Use tna lesser of the two as the infiltration rate.
- 7. Report infiltration rate in units of meters / year.
fable S 3 5 CDF for anolication rates Annual Precipitation percent (inches) <X
<24 0.00 24 46.24 25 47.63 30 54.04 35 S2.94 40 70.51 45] 80.39 Residential Scent.rio 5.3-8 January 31,1998
l Table 5.3.5 CDF for nonlication rates. Annual Precipitation percent (inches) <X 50 87.94 55 94.14 60 98.24 65 99.7F.
>65 100.0d Uncertainty in saturation ratios Campbell [1974] derived a relationship between unsaturated hydraulic conductivity, K(0) and saturation ratio, f:
K(0)=K,,/26 2 (5.3.17) where b is a curve fitting parameter related to pore size distribution. Under unit gradient, steady state conditions such as are assumed in the DandD model, the unsaturated hydraulic conductivity is equivalent to the infiltration rate determined above. Substitutirig infiltration rate for unsaturated hydraulic conductivity and rearranging to solve for the saturation ratio, results in: f= [ K,,,] 2* *' (5.3.18) Since infiltration rate and saturated hydraulic conductivity are known from above, all that remains is to determine a value for b. Meyer et al. [1997] derived a relationship for b using soil water reuntion parameters considered in Carsel and Parrish (1988). Using this relationship, Meyer et al. [1997] constructed distnbutions for b. They are reported based on the 12 Soil Conservation Service textural classifications. The distribution type and parameters for these distributions for each of the 12 soil types are given in Table 5.3.6. Table 5.3.6. Distributions for the parameter b. Soil Type distribution Mean Standard lower limit upper limit type Deviation sand lognormal 0.998 0.226 loamy sand lognormal 1.40 0.397 sandy loam lognormal 1.96 0.579 sandy clay loam lognormal 4.27 1.39 Residential Scenario 5.3-9 January 31,1998
O Table 5.3.6. Distributions for the parameter b. i \ (j Soil Type l distribution Mean Standard lower limit upper limit type Deviation loam lognormal 3.07 0.900 sitt loam lognormal 3.80 1.42 sitt lognormal 3.21 0.465 clay loam lognormal 5.97 2.37 silty clay loam lognormal 7.13 2.34 sandy clay lognormal 6.90 2.27 silty clay lognormal 10.2 2.96 clay beta 14.1 6.24 4.93 75.0 Meyer et al. [1997] also developed ,:orrelation matrices for parameters for each of the 12 soil types. There exists a moderate negative correlation between b and porosity as well as between b and saturated hydraulic conductivity. These correlations persist across all soil types. Summarizing the correlation matrievs given for all soils, a correlation of -0.35 for both relationships is e reasonable approximation. O) ( v Once 3 b value is samplod, the saturation ratio can be calculated using the above equation. PARAMETER UNCERTAINTY: The distc..ution for the soil texture was based on generalized soil textures throughout the contiriental United States. These textures omit bedrock, highly organic soils (peat, muck, etc.), water, and "other" tevtures ar.d should be representative of soil textures in most regions of the country. The distribu :en was selected to be most representative of surface soils (the upper 15 cm.). While deeper soils might tend to be slightly more clayey, this uncertainty is not expected to significantly affect the results of this analysis. VARIABILITY ACROSS SITES: Soil texture will vary from site to site and may vary over a site.
'Hhile soil texture is not an explicit parameter in the DandD analysis, knowing it for a site may
- table the applicant to refine the distributions of related parameters such as porosity and saturation ratio. For many sites, soil texture may evaluated by reviewing existing soil surveys available from state agencies or the US Department of Agriculture (USDA). For sites located in regions with highly variaNe soils, site data on soil texture are easily collecteri by routine sampling and particle-size analysis.
REFERENCES:
Campbell, G.S.1974. A simple method for determining unsaturated conductivity from moisture retention data. Soil Science, vol.117, pages 311-314 (3 V Residential Scenario 5.3-10 January 31,1998 1 1
l l Carsel, R.F., and R.S. Parrish.1988. Developing joint probability distributions of soil water retention characteristics. Water Resources Research, Vol. 24, No. 5, pages 755-769. France, Lewis,1992, CLIM81 196S90 Normals, TD-9641: U.S. Dept. of Commerce, National Oceanic and Atmospheric Administration, National Climatic Data Center, Asheville, N.C. Hillel, D.1980. Fundamentals of Soll Physics. Academic Press. Meyer, P.D., M.L. Rockhold, And G.W. Gee.1997. Uncertainty analysis of infiltration and subsurface flow and transport for SDMP sites. NUREG/CR-6565 Owenby, J.R. and Ezell, D.S.,1992 Climatography of the United States No. 81: monthly station normals of temperature, prec;pitation, and heating and cooling degree days, 1961-90: U.S. Dept. of Comrnerce, National Oceanic and Atmospheric Administration, National Climatic Data Center, Asheville, N.C. Soil Survey Staff, 'Nationti Soil Survey Handbook", title 430-VI, Natural Resources Conservation Service, Washington, D.C., U.S. Govemment Printing Office, December 1997. U.S. Bureau of Reclamation 1993. Drainaae Manual (revised repdnt) USDC,1994.1992 census of agriculture, AC92-RS-1, farm and ranch irrigation survey. U.S. Department of Commerce, Economics and Statistics Administration, Bureau of the Census. O 40 _._ __. g 30 . e 25 20 J 15 y 10 . E5
- 8. 0. +
0.00E+00 5 00E-04 1.00503 1.50503 2.00603 2.50603 3.00503 saturated bydraulle conductivity Figure 5.3.1 Percent Percolation as a Function of Ksat Residential Scenario 5.3-11 January 31,1998
m..._.__._._. _ . _ . . _ _ _ _ _ . - _ . - _ m 5 e sasase6 esee e e mzmze e mmoe m m eow s Figure 5.3.2 PDF for Precipitation m . _ _ __. _ m 7m . 1-
'a .
1 O m a m a e e e s e s en=kow Figure 5.3.3. CDF for Application Rate e Residential Scenario 5.3-12 January 31,1998
S.4 Dust loading: air dust loading outdoors, CDO, gardening CDG and Indoors CDI (g/m3); floor dust loading P,(g/m 8) and resuspension RF, (md ) The dust-loading factors are used to calculate the average annual dose resulting from inhalation of airborne contaminants. The dust-loading factors, CDO and CDG, are used to O calculate the inhalation dose due to activities occuring outdoors. CDO (g/m ) represents the mass concentration of contaminated airborne particles in air outdoors, as defined in the exposure model, and corresponds to the long-term average quantity of respirable particulate materialin outdoor air. CDG (g/m 8) represents the higher average mass loading of contaminated airbc,rne particles in air while the individual is gardening. The default values for these parameters defined in NUREG/CR-5512, Volume 1 [ Kennedy,1992), are 1 x 10d g/m8 for CDO and 5 x 10d g/m8for CDG. These va!ues were defined based on the review of literature from outdoor air pol!ution studies from the National Air Sampling Network and studies on suspended particles in the atmosphere in communities across the United States The indoor dust loading fa-tor, CDi, represents the process of infiltration of contaminated airborne particles into the here (mass-loading) as the mass of infi'trating particles per unit volume of air 'Inese particulates are distinguished hm contaminated soil that is (racked indoors and subsequently released into the air by resuspension. Since the source of contamination la the surface soil layer, CDI tmcomec a function of the outdoor dust loading factor (CDO). CDI is used to calculate the average annual dose resulting from inhalation of airbome contaminants that are represented by parent and daughter radionuclides. The default value for th!s parameter as defined in NUREG/CR-5512, Volume 1 [ Kennedy and Strenge, 1992, p. 6.10-6.11], is 5x104 g/m3 This value was selected based on a fraction (1/100 th) of the regulatory limit for total dust loading of respirable particulates in industrial settings (29 CFR 1910.1000,1990), considered rersentative of the long-term average concentration of contaminated respirable dust, e squivalent to 0.5 times the default CDO value. Po is a physical parameter that represents the long-term average mass of contaminated soil per unit area of floor inside the residence. Since it is a single parameter value for the entire time spent indoors, it is an average value for the entire house. The dust-loading on floo s is used to estimate the airborne particulate concentration due to resuspension of soil tracked into the house. The default value for this parameter defined in NUREGICR-5512, Volume 1, is 0.4 g/m2 , The resuspension factor, RF,, defined foi the NUREG/CR-5512 dose modeling, defines the ratio of the long-term average respirable contaminant concentration in air to the long-term average floor surface contaminant concentration due to contaminated soil tracked indoors. The default value for the resuspension factor recommended in NUREG/CR-5512, Volume 1, is 5 x 104 m4, based on recommendations from lAEA [lAEA,1970). The overall range of values 4 obtained from literature published from 1964 to 1990 is 2 x 10'" to 4 x 10-2 m . However, most data referenced are for outdoor conditions (wind stress and vegetation). Only two of the references cited in Volume 1 provide data for indoor resuspension. The first of these, an IAEA technical report [1970), reports a value of 5 x 104 m 4
, which has been obtained for operating nuclear facilities and may not provide a representative value for resuspension in a residential setting. The second of these two references, a review by Sehmel[1980), provides different resopension factors depending on the type of activity conducted within the rooms of the
< Residential Scenario 5.4 -1 January 31,1998
building (walking, vigorous sweeping, anc' fan) but does not differentiate between the FT resuspension of respirable and non-respirable particle sizes. The overall range cited by Sehmel ( ) is from 1 x 104 to 4 x 10-2 md which may over estimate the resuspension factor useo in this model because the data include non-respirable particles. IMPORTANCE TO DOSE: CDO , CDG, CDI, P, and Rf, are important to dose because, the higher the mass loading in air, the higher the total annual dose during the first year of the residential scenario. CDO also influences the dust mass loading indoors (CDI). As described below, the dose for the inhalation pathway is directly proportional to each of these parameters. USE OF PARAMETERS IN MODELING: These parameters are used for calculating the inhalation committed effective dose equivalent, DHR,, from contaminated indoor and outdoor air as described in the following formula (NUREG/CR 5512 Volume 1, page 5.55, equation 5.70): DHR, = [24V,(t,/ty ) CDG C., En.,a S{A,g,t,}DFHj
+ [24V,(t,/t,) CDO C., En.,2 S{A,g,t,}DFHj
+ [24V,(t/t,) (CDI + P.Ri',) C En.if S{A,g,t,}DFH) (5.4.1) where V,, V,, and V, correspond to the volumetric breathing rates for time spent gardening, indoors, and outdoors (m /h), respectively, tois the time during the gardening period that the individual spends outdoors gardening (d for a year of residential scenario), yt is the total time in one gardening period (d), t, and t, are the times in the 1-year exposure period that the individual spends indoors and outdoors (excluding gardening), respectively, t,is the total time in (a
'd
)
the residential exposure period (d), CDI, CDO, and CDG are dust loading factors for indoor, outdoor, and gardening activities (g/m3 ), respectively, C,is the concentration of parent radionuclide I in soil at time of site release (pCi/g dry-weight soil), J, corresponds to the number of explicit members of the decay chain for parent radionuclide I, S{A,g,t,} is a time-integral operator used to develop the concentration time integral of radionuclide j for exposure over a 1-year period per unit initial concentration of parent radionuclide i in soil (pCi*d/g per pCi/g dry-weight soil), S{A,g,ty )is a time-integral operator used to develop the concentration time integral of radionuclide j for exposure ov?r one gardening season during 1 year period per unit initial concentration of parent radionuclide I in soil (pCi*d/g per oCi/g dry weight soil), DFH,is the inhalation committed effective dose equivalent factor for radionuclide j for exposure to ' contaminated air (in units of mrem per pCi inhaled), P is the indoor dust-loading on floors (g/m2), and RF,is the indoor resuspension factor (md).The higher the value for each of the dust loading and resupension factors, the higher the dose. In this model, the concentration of contaminated particles in air indoors due to infiltration is some fraction (PF) of the outdoor air concentration. The long-term average outdoor air concentration (C% ) is estimated as the product of CDO and the contaminant concentration in soil (C ,). Co = CDO C., (5.4.2) Resulting in the following model of the concentration of indoor air due to infiltration: (~ {)j Residential Scenario 5.4 -2 January 31,1998
l C,,, = CDI
- C, = PF
- CDO
- C, (5.4.3)
The factor PF represents the fraction of airborne particulates that infiltrate the house and remain airbome. This factor will be a function of the abihty of the particulate matter to enter tha house (generally reported as a penetration factor) and remain suspended. There will be less resuspensien of particles indoors (due to cleaning, static electricity and tower wind speed (air disturbance)) which will lead to a net deposition or loss. 5.4.1 Review of Additional Information to Define PDFs for CDO and CDG Air concentrations are determined using mass loading factors and are converted to units of activity from the concentration of the source material. Fourteen references are listed in NUREG/CR 5512 volume i for this data (Hinton,1986; Hinton,1970; Stem,1968; HEW,1969; Magill,1956; Shinn,1989; Sehmel,1975; Sehmel,1977a; Sehmel,1984; Sehmel,1977b; Stewart,1964; Sinclair,1976; Soldat,1973; Anspaugh,1975). The outdoor air dust-loading 4 factors range from 1 x 10'5 to 2.3 x 10 g/m8 for all airbome padicles. Under extreme conditions, air dust-loading can be as high as 5 g/m*; however, these conditions persist for only very short periods of time. For particles less than 10 um diameter (the respirable fraction), air dust-loading factors range from 1 x 10-5 to 7 x 10d g/m*. Table 5.4.1 summarizes the experimental results on dust loading studies: Tablo 5.4.1. Total Dust Loading Reference Dust loading Anspaugh (1975) 9 x 104 - 7 x 10-8 g/m 3 Soldat (1973) 1 x 104 g/m 8 Shinn (1989) 2.1 x 10 5 g/m2 (background) 3.4 x 103 g/m 3(sea spray) Magi!!(1956) 1 x 10d - 2 x 10'sg/m8 HEW (1969) 1 x 10-5 g/m2 (ruralareas) 6 x 10'5 - 2.2 x 104 g/m3 (urban) Stern (1968) 9.8 x 10 5 g/m' (geometric mean) with ma:.imum of 1.7 x 10 8 g/m 3 Sehmel (1975; upper limit of 7 x 10d g/m2 (<10 um diameter) 1977a;1984) upper limit of 2.3 x 104 g/m8 (>10um diameter) Additional information was reviewed to determine if other data or approaches, preferably more recent than those cited in NUREG/CR-5512, Volume 1, were available to provide a defensible basis for constructing PDFs for CDO and CDG for use in this analysis. The outdoor dust-Residential Scenario 5.4 -3 January 31,1998 1 l
loading factor, CDO (g/m3), represents the long-term average quantity of respirable outdoor 9 dust, as defined in the exposure model. In order to define the parameter distribution, a detailed analysis of the factors that contribute to outdoor air-dust loading along with supporting experimental data on outdoor dust loading measurements is needed. In the absence of human activities that create or suspend airborne particulates, the major factor controlling the suspension and resulting particle concentration in air is wind speed. Higher dust loading due to human activity is represented by gardening. As shown by a number of authors, the particle concentration is an exponential function of the wind speed (e.g., Sehmel,1977b). Unfortunately, there is no reliable analytical relationship between the wind speed and dust loading factor that could be used in defining dust loading from the average wind speed. Moreover, it is not clear how to specify the function and determine the proportionality coefficient between the wind speed anri dust loading under different conditions. Another important factor influencing dust loading is soil moisture. As discussed in [Tegen, 1994), suspension of soit particles in air is only possible when the soil matric potential is greater than 104 J/kg. In other cases, no suspension will occur even under strong wird aonditions. t
.Aoreover, suspension is also influenced to a great extent oy veguation cover. High resuspension is common for areas without vegetation or with sparse vegetation, and low resuspension is common for areas of dense vegetative cover. Finally, dust loading is affected by soil type (composition). Some soils are easily eroded, while other soil types are resistant to erosion. Other less significant factors are: topography (surface roughness) and snow cover / surface soil freezing.
Since the wind speed, soil moisture (or amount of precipitation), and vegetation cover are G factors related to the climate, different categories could be defined based on different climatic cond',tions. More generally (including the other factors, such as soi! types, topography, and etc.), categories could be defined based on different environmental conditions. The usefulness of one or another category definition depends on the availability of information on dust loading factors measured under different climatic or environmental settings. A second, and equally important factor, is estimating the probability that a particular site is in a specific category. An extensive literature review was conducted to identify different categories of environmental conditions that could be reasonably defined based on published data. A summary of this review is presented in Table 5.4.2. The information allows us to evaluate single dust loading measurements and average values from a number of measurements, to distinguish between extreme conditions (dust loading during a dust storm) and normal conditions (dust loading under average wind conditions), and to compare environmental conditions specific to different sites. Most of the dust leading values available from the literature (Table 5.4.2) represent the total amount of dust resaspended in air. The dust loading factor, as defined in NUREG/CR-5512, corresponds to the quantity of contaminated, respirable airbome particulates. According to the EPA, the respirable particles are particles smaller than 10 pm. Various studies have been conducted to determine the relationship between the mass loading and particle sizes. Data from Hinton [1996] indicate the mass of respirable particles is C.5 to 2.5 orders of magnitude less than the total mass of airbome particles. These data are supported by other observations Residential Scenario 5.4-4 January 31,1998
[Sehmel, 1975; 1977a; 1984). Another factor that was not previously discussed is the difference in the particle mass resuspended in the air at different heights from the land surface. The dust loading factor defined in NUREG/CR 5512 should represent the air concentration at the respirable height. As discussed in [Sehmel,1977], the air concentration depends on the height. In other cases, the functional relationship ;s monotonic with higher concentrations near the land surface. In some cases, the concentrations near the land surface can be lower than at some distance from the surface (usually, below the respirable height) where it renches a maximum value. However, the concentrations of suspended particles in the air vary by about 20% for a height between 0.5 and 2.0 m. Sinco particles are measured near the land surface or at the reference height of 1 m, these small variation can be neglected when defining the dust loading factor ranges. The outdoor air-dust loading, CDO, varies with the particle size of the contaminant, quantity of loose particulate contaminants at the surface, and magnitude and types of extemal stresses. The concentration of dust in the atmosphere has been measured and modeled under a wide - range of conditions. Rognon (1991) conducted field measurements near the ground
- correlat.J the dust content with surrounding soils based on the compositicsn of the soil, state of the plant surface and ground cover, surface roughness, drag velocity, turbulence, wind velocity, and the atmospheric dust load and composition. The particle concentration varied from 1.6 x 104 to 1.25 x 10 5 g/m8. Tegen (1994) applied a model that takes into account the size distribution of the dust particles to estimate the distribution of atmospheric mineral dust. Tegen extended the model to calculate the atmospheric mineral aerosol load under conditions in which the soil surface is disrupted by agricultural activities or the soil surface is exposed to wind erosion through deforestation and shifting desert boundaries. Suspended particulate matter was monitored by the New York State Department of Environmental Conservation in residential and industrial sections of a small city. The concentration of particulate matter averaged 6.6 x 2
10-5 g/m over a 4-year period (New York State Air Quality Report). Table 5.4.2 summarizes the experimental data. These data are not specific to human activities. The residential farmer is likely to work under more extreme dust-loading conditions for short periods of time; however, dust loadings greater 3 than 4 x 10 g/m for an extended period of time has resulted in a significant increase in death rates [ Magill,1956). This information can be used to provide an upper bound on CDG if the time spent gardening is representative of the " extended periods of time"in the Magill study. Table 5.4.2 Outdoor Dust Loading Reference Dust Loading ( Site Description g/m*) Prospro (1976) 5 x 104- 2.5 x 104 Near large body of water (Summer) Residentia! Scenario 5.4 -5 January 31,1998
Table 5.4.2 Outdoor Dust Loading O Reference Dust Loading ( Site Description g/m 8) Sehmel(1977) 7.7 x 104 - 7.1 x 10d Hanford Site, arid climate, sparse vegetation, average annual wind 3.4 m/s,0.16 - 10 um particles (numerous long-term average values over a 4 year period) Sehmel(1977) 2 x 105- 2.3 x 104 Hanford Site,10 - 230 um particles (non-respirable, not included in data to support pdfs) Prospro (1981) 2 x 10 4 Near large body of water (Spring) Pye (1987) 1 x 10 6 x 104 Near ,arge body of water Hartmann (1989) 4.5 x 104 - 1.3 x 104 Humid climate, forni Gao (1992) 4 x 10 4 Near large body of water (Spring) Rognon (1991) 1.6 x 10'- 1.3 x 10-5 Desert region Zier (1991) 2.3 x 10-5 . 2 x 10 d Near surface air 4 d Friedrichs (1993) 8 x 10 - 1.6 x 10 Small indu. trial city 9 Tegen (1994) 1 x 10d Areas of high dust loading (deserts, eroding cultivated areas) Tegen (1994) 5 x 104- 2.5 x 10 5 Tropical climate, dense vegetation cover Tegen (1994) 6 x 10-5 Pacific Northwest Clausnitzer (1996) 3 x 10d - 1 x 10 2 Dust collector mounted 94 cm above disturbed soil on agricultural implement (<4 um dis meter particles). (Respirable fraction, but not representative of average conditions for exposure due to measurement conditions) Moulin (1997) 1 x 10^5 Tropical climate, dense vegetation cover (average over 30 year period) NY State (1981) 6.6 x 10 5 Annual average over 4-year period PDF for CDO The generic nature of these analyses the potentia! variability in site-specific conditions snd the large variability in the measured mass loading (orders of magnitude) prevent the use of a biased distribution for this analysis. The distribution of the dust loading is best represented by a log-uniform distribution with a lower limit of 1 x 10-7 g/m' and an upper limit of 1 x 10d g/m2. Residential Scenario 5.4 -6 January 31,1998
The range of mlues is defined by the range of average values for dust 'oading of respirable particles (<h, gm in size)in arid and humid climates. The use of a log uniform distribution ensures that the selection of a particular magnitude of CDO will be equally likely. In the absence of information on the fraction of sites in each of the two climatic categories, due to unknown location of future sites and the indistinct categories of arid and humid, an equal probability has been assumed. This distribution is also based on the assumption that all the respirable dust is contaminated at the same concentration as the soil. PDF for CDG Short-term gardening activitien are expected to produce localized, elevated levels of dust loadings. Based on the data presented in Tables 5.4.1 and 5.4.2, the upper limit on dust loading , for respirable particles is approximately 7 x 10" g/m3 (Sehmel, 1975; 1977a;1984). Higher dust loading of respirable particles has be' a measured (Clausnitzer,1996) but not under conditions reasonable for human exposure anc at levels that would :ause physical harm. For this analysis the gardening dust-loading factor is ussigned a uniform Jistribution with a lower limit of 1 x 104 g/m2and an upper limit of 7 x 104 g/m8 , based on b range of values from the literaturt Sr particulates less than 10 pm in diameter for higher dust loading activities (as cited in NUREG/CR-5512, Volume 1) and assumption that all the respirable dust is contaminated at the same concentration as the soil The lower limit for CDG corresponding to the upper limit of CDO, based on the intent of the gardening scenario to represent a higher level of activity while outdoors. This distribution for CDG will result in higher dust-loading during the time spent gardening.
~
5.4.2 Review of AdditionalInformation to Define PDF for CDI Additional information was reviewed to determine if other data or approaches to those presented in NUREG/CR-5512, Volume 1, were available to provide a defensible basis for constructing a PDF to represent the uncertainty and variability in CDI for residential settings at all current and future sites. The ratio of indoor to outdoor suspended particle matter has been reported from a number of studies. Whitby (1957) studied the properties of airbome dust indoors and outdoors at various locations and reported values ranging from 65 pg/m8 indoors to 93pg/m8 outdoors (a ratio of 0.70). Tel suspended particulate concentrations were monitored outdoors over a period of b about 4 years near an industrial site. Sterling (1977) compared indoor and outdoor suspended particulate concentrations and observed that the concentration of suspended particles indoors is 77 to 85% of the corresponding concentration outdoors. However these studies did not distinguish between infiltration of airborne particles and resuspension of contaminated soil tracked indoors. As a result, these studies can only provide an upper bound on the potential CDI for the specific conditions evaluated (i.e., by assuming the floor dust loading or resuspension factor indoors are negligible). A more recent study, with experimental data, modeling and evaluation of other published studies, by Thatcher and Layton (1995) discriminates between resuspension and infiltration of Residential Scenario 5.4-7 January 31,1998 0
particles, in their analysis. Thatcher and Layton's measurements and modeling support their conclusion that the difference in the indoor and outdoor air concentration due solcly to (( ]f infiltration (i.e., excluding resuspension) is a function of deposition indoors rather than the ability of the house to limit infiltration of particles. CDI represents the mass loading indoors of infiltrated particles and combines the effects of penetration and net deposition. As a result, studies that neglect deposition can be used to estimate the variability in PF (which can be viewed as the ratio of CDI to CDO). The studies cited by Thatcher and Layton and the results of Thatcher and Layton's studies are summarized in Table 5.4.3. Table 5.4.3 Reported Values for the Ratio of Indoor to Outdoor Dust Loading PF Reference Notes 0.2 - 0.6 Thatcher and Layton, 3-10 pm particle size range, assuming deposition 1995 negligible 0.4-0.6 Thatcher and Layton, 1-3 pm particle size range, assuming deposition 1995 neg);gible 0.7 Dockery and Spengler, respirable particles and sulfates 1981 0.4 Freed et al.,1983 sub-micron particles 0.2 Freed et al.,1983 super-micron particles 0.3 Alzona et al., '979 reported typical for Fe, Zn, Pb,Br and Ca b (g 0.45 Cohen ano Cohen,1980 sub-micron particles, reported average for Fe, Zn, Pb,Br and Ca in residential and industrial settings 0.2 Cohen and Cohen,1980 super-micron particles, reported average for Fe, Zn, Pb,Br and Ca in residential and industrial settings 0.7 Colome et al.,1992 <10 pm particle size, average for 35 Califomia homes (range 0.4 to 1.5, may neglect resuspension) 0.77- Sterling,1977 unknown size distribution, includes resuspension 0.85 therefore not used to estab;ish the pdf. CDI values for parameter analysis Based the studies summarized in table 5.4.3 it can be concluded that PF ranges from 0.2 to 0.7. This varisbility is due to a number of potential factors including the measurement technique, location within the house, and variability in the airbome particle size distribution. Given the generic nature of this analysis, limited number of studies and measurements to suppcrt a generic parameter value ano the uncertainty in the particle size distribution of the contaminated soil; the uncertainty in PF is best represented by a uniform distribution between the values of 0.2 and 0.7. CDI will not be sampled from a pdf, rather CDI will be a function of the values sampled for CDO f Residential Scenario 54-8 January 31,1998
.. _____________A
and PF (see equation 5.4.3). 5.4.3 Review of AdditionalInformation to Define PDF for P, Solomon (1976) measured floor dust in a number of residential settings. The floor dust loading ranged from 0.11 to 0.59 g/m2 based on 239 samples from 12 different dwellings. Similar results were reported from studies conducted by the New York State Department of Environmental Conservation (1982). In the absence of additionalinformation, a uniform distribution is proposed However, the results of these two studies are for total dust loading which may include non-soil components and soil from remote locations. As a result, these studies can be used to estimate an upper bound on Pa. Thatcher and Layton (1995) performed a detailed modeling and experimental study to quantify the sources of indoor air contamination. They report that the major component of floor dust is soil, but they do not present the results. Total dust loading in the two houses in the Thatcher and Layton study ranged from 0.06 g/m2 on linoleum to 43.4 g/m2 on a rug by the door. Dust loading on carpeted floors was significantly higher than on linoleum (0.58 to ? g/m 2) with the
,iigher values in high-traffic areas. Information on the area of flocd carpeted and the area covered with linoleum is not provided. If it is assumed that the floors are covered in equal parts linoleum and carpet and the area covered by the rug near the front door is negligible, then the average total dust load is on the order of 0.6 g/m 2.
A recent study by Rutz et al. (1997) wuated the average total dust loading on floors in two separate homes and estimated the fraction of dust that is from contaminated soil. The results of this analysis provide information necessary to estimate Pa for those two homes if it is assumed that the floors are covered in equal parts linoleum and carpet and that the dust loading in the ru] by the door is negligible when the dust density is averaged over the entire house. One house had an average total dust density of 0.4 g/m2 and an average of 30 percent of that dust is contaminated soil resulting in a Poo f 0.12 g/m2 . The other home had a lower average total dust density (0.1 g/m2) and an average of 20 percent of that dust is contaminated soil resulting in a Po of 0.02 g/m 2 Other studies on floor dust loading with contaminated soil cited by Rutz et al. (1997) indicate the dust is comprised of 31 to 50 percent contaminated soil (Calebrese and Stanck (1992) and Fergusson et al.( 1986)). PDF for P. Given this limited amount of information, the range of P values is 0.02 to 0.3 g/m2 and all values in that range are equally likely. A uniform distribution between 0.02 and 0.3 g/m2 should be used to represent the uncertainty in this par . meter. S.4.4 Review of AdditionalInformation to Define PDF for RF, An extensive literature review was conducted to identify any developments in the understanding of the resuspension process since the review reported in NUREG/CR-5512 in 1992, and to identify data or approaches that could be used to develop a probability distribution function for Residential Scenario 5.4 -9 January 31,1998
the indoor resuspension factor in the residential scenario. Older publications that were not l p referenced in NUREG/CR-5512, Volume 1, were also reviewed.
\" /
Resuspension factor values are reported in a number of studies published between 1964 and 1997. Reported values for resuspension factors vary over a wide range, from approximately 10-" m 4to approximately 10-2 m4. The review of some older publications indicate that the value of resuspension factor of 1 x 104 m4 was used in the development of general guidelines, and has been seen as a general value having a reasonable factor of safety for hazard , evaluation and design purposes (Brodsky,1980). This value was also recommended by the IAEA [1982 1986) and suggested as an average for Europe in Gartend [1982). These sources support (but were not cited to justify) the default parameter value for RF, in NUREG/CR-5512, Volume 1. Most studies, and all but one study not included in the review reported in NUREG/CR 5512 Volume 1, provide data on outdoor resuspension factors. These values are not directly relevant for the residential scenario model. Published estimates of resuspension factors and resuspension rates under indoor conditions, identified during the current literature review, are summanzed in Table 5.4.4. The results of this review are presented in the same format as the earlier review published in NUREG/CR-5512, Volume 1. The reported values from these sources range from 2 x 10 8 to 4 x 10-2 md . With one exception (Thatcher and Layton,1995), no recent information on indoor resuspension was found. Thatcher and Layton evaluated the sources of contamination inside a Califomia residence under controlled indoor conditions; however the results provided are estimates of resuspension rates of aerosols which cannot be directly translated into a resuspension factor for this analysis. The Thatcher and Layton study indicates that resuspension indoors is a function of the time individuals spend inside the home and that the two parameters are linearly correlated for the particular set of conditions analyzed. As discussed in the following text, the (Vn') var; ability from site to site in surface conditions, humidity, human activities and particle size distributions oroduces order-of-magnitude variations in RF, As a result, the uncertainty in the appropriate effective parameter value overwhelms the linear relationship between time spent indoors and RF, . Table 5.4.4 Reported Information for Indoor Resuspension Condition / Reference Range (md ) Comments Wind stress and mechanical 2x10 - 1.8x10d Resuspension of Pu-disturbances, (Jones,1964) -contaminate.d particles on various surfaces Wind stress and vehicular 1x104- 1.5x10-2 Resuspension of Pu - and mechanical dis- contaminated surfaces turbances, [Glauberman, Wind stress [Brunskill,1964) 2.5x10d - 3.9x10'2 Resuspension of radio-nuclide contaminants in change room Residential Scenario 5.4-10 (Ov) January 31,1998
Table 5.4.4 Reported Information for Indoor Resuspension Condition / Reference Range (md ) Comments Vigorous mechanical 1x10 4x10~8 Resuspension of BeO on (Asturbance (sweeping), contaminated wood floor [Mitchell,1964] Vigorous mechanical 9.4x104 - 7.1 x10d Redistribution of ZnS and disturbance (sweeping) CuO particles on tile floor [ Fish,1964} Various factors affecting resuspension, underlying the range of reported valuet, have been proposed. The effects of some factors are quantified in some studies, while other effects are discussed qualitatively. Although many studie= consider the factors affecting outdoor re-suspension, these factors have analogs in ine indoor transport pathway model. Such studies are therefore relevant for understanding potential variations in RF, across sites. The wiJe range of reported resuspension fcctor values is due to differences in measurement techniques and to variability in "hysical factors that affect resuspension both within and among studies. The common mt.. - sent techniques for determining indoor resuspension factors are: e direct measurement of contaminant concentrations on surfaces and in the air [ Jones,1964; Glauberman,1964; Brunskill,1964; Mitchell,1964) e redispersion of settled particulates [ Fish,1964] e recoil of " hot-atoms" during decay cf radionuc' ides [ Leonard,1995) in addition to differences in experimental techniques, measured values of resuspension factor may vary due to spatial variability of surface contaminant concentrations, variability of concentrations in air with location and with elevation, and spatial variations in surface texture leading to location-dependent resuspension. These variations can create uncertainty la the effective value of resuspension factor as estimated by the ratio of concentrations measured in air and on the contaminated surface. Infiltration of airborne particulate matter could increase the estimated resuspension factor. A large number of physical factors c a affect indcor resuspension. According to IAEA (1992], the major tactors affecting resuspension that might apply to indoor conditions include the following: e type of disturbance (air flow or mechanical) e intensity of disturbance (air flow speed, traffic intensity) e nature of surface (texture, composition, surface area) e surface moisture e particle size distribution e climatic conditions (temperature, humidity, wind) e type of deposition process (wet or dry) e chemical properties of the contaminant e surface chemistry Residential Scenario S.4 -11 January 31,1998
p The potential effects of some of these factors cn resuspension have been quantified, while only g qualitative characterizations are available for others. As discussed above, some studies discuss the effects cd these factors on outdoor resuspension factors. While values of outdoor resuspension fac ;. s are not appropriate for the residential scenario model, reported effects of variations in physical conditions (e.g. air flow) on relative resuspension factor values do provide usefulinformation about potential variations in indoor resuspension factor values due to variations in the resident's behavior or environment. Iyoe of disturbance f air flow or mechanicaD Among studies reporting indoor resuspension factors, the higher resuspension factors provided in Brunskill[1964), Glauberman [1964), and Mitchell[1964) were measured when disturbances significantly more severe than in normal operating conditions were applied to obtain measurable contaminant concentrations and when most of the surface contamination was a loose, easily removable, contamination (spills on the floor) and would be similar to the conditions assumed for the residential scenario (*l tracked indoors), l l Glauberman [1964] provides resuspension factors for a range of air-flow rates and mechanical 1 disturbances that may occur in occupational settings. They measured occupational exposure to airborne particulates in a operating facility by measuring the concentrations of particles in air (high efficiency particulate sampl.:r) and particles on surfaces (smear sampling), and reporiang the ratio as a resuspension factor. Airborne particle contaminants in this experiment may have originated from sources other than surfaces (e.g., processing equipment, etc), which would tend to increase estimated resuspension factor values. The reported values from Glauberman p) [1964] were retained for comparison in defining the distribution for RF,, but are judged to be (d highly uncertain and may overestimate the resuspension factor associated with the conditions described. Brunskill [1964] studied resuspension of contaminants from clothing under high air-flow conditions typical of a change room. Mitchell[1964] measured resuspension factors during vigorous mechanical disturbance of contamination on a wood floor. The experimental conditions were contrived to deliberately suspend loose contamination in order to produce measurable values of resuspension factor. These conditions are not necessarily representative of the long term average conditions in a residential setting. Fish [1964] reports a range in resuspension factor of 1.5 orders of magnitude due to different types of activities in the room. latensity of disturbance (air flow soeed. traffic intensitv) Anspaugh [1975] suggests that contaminant concentrations in the air are proportional to the power of the friction velocity which is, in turn, proportional to the speed of horizontal air flow (wind velocity). Consequer. y, the difference of 1 order in magnitude between the wind velocity may result in differences of a few orders of magnitude in resuspension factors. This power law relationship between the wind speed and resuspension factor is also demonstrated by Hollander [1994]. ('N t ) Residential Scenario 5.4 -12 January 31, i998 v
Among studies of indoor resuspension, Fish [1964) observed a power law relationship between the resuspension factor and the air velocity in the room, and Jones [1964) reports variations in resuspension factor due to different walking speeds. Nature of surface (texture. comoosition. surface area) j The magnitude of the influence of this factor on resuspension was not qusntified in the l literature. In a study of indoor re=uspension, Glauberman [1964) attributes a difference in resuspension factors of one order of magnitude to differences in room size. Particle size distribution It is suggested by Hinton [1995) that resuspension is gra.itest for particles with diameters smaller than 125 um, and it is suggested by the IAEA [1992] that resuspension factor increases with particle diameter in the range from 1 to 5 um. The resuspension factor is also correlated with the particle diameter. In Sehmel !1980), however, it is suggested that further studies are needed. In studies of indoor resuspension, Fish [1964] repos., a strong correlation with particle diameter and Thatcher and Layton [1995) report no indoor resuspension of particles less than 5 um, Chemical oronerties of the contaminant The difference between resuspension factors determined under the same conditions for different radionuclides is one order of magnitude, but could be significantly smaller as discussed by Hartmann [1989) and the IAEA [1992). Among studies reporting indoor resuspension factors, Jones [19S4) reports variation of the resuspension factor within one order of magnitude depending on the contaminant. Surface chemistrv Although cited by the IAEA [1992) as a factor influencing resuspension, no specific information on the effect of surface chemistry on resuspension factor was found in the literature. The main conclusions of this literature review are: e the new data on resuspension factors falls into the same range that was noted in NUREG/CR-5512, Volume 1; e no significantly new models of resuspension or methods of resuspension measurement were proposed since 1990; e additional information is available on resuspension factors datermined under indoor conditions; e the resuspension factor value of 1 x 10-e mdis the most frequently suggested and appears to represent some average of the experimental data; e data on probability distribution functions that could be used to reflect uncertainty and variability in resuspension factors are very limited a the range of the resuspension factor values measured under indoor conditions is around Residential Scenario 5.4 -13 JarJary 31,1998
four orders of magnitude [ Jones,1964). O ( l The published data indicate that resuspens> htor values vary over orders of magnitude
' depending on site specific conditions which include the nature and intensity of mechanical disturbance associated with activities in the home.
Grouping of Reported Resuspension Factors based on Experimental Conditions Table 5.4.5 summarizes the resuspension factors reported for experimental studies for various conditions (Jones,1964; and Fish,1964). The experiments by Jones [1964) provide average resuspension factors for a range of activities that are common in occupational settings. The measured resuspension factors reported by Jones (1964) are for four levels of activities conducted for 60 minute periods in a laboratory setting with different floor surfaces, using Pu(NO3 ), and PuO 2-contaminated particles (0.4 - 60 um diameter) and particulate air samplers positioned at 14-175 cm above the surface. The particle size distribution includes non-respirable components and the height above the floor surfaca is not necer.sarily representative of the exposure scenario. Fish (1964] provides average tr. suspension factors for a range of vigorous mechanicM disturbances of contamination on L tile floor based on 10 minutes of the reported activity. Tne values in Table 5.4.5 for this study are reported for four levels of disturbances. ~ Table 5.4.5 Resuspension Factors Measured Under Various Conditions \ Experimental Condition RF, (m") 7_s Reported by Jones [1964] l 1 Air circulation (no mechanical disturbance) 7.7 x 104 to 1.5 x 104 Walking (14 steps / min) 3 x 10#to 2 x 104 Walking (36 steps / min) 9.7 x 104 to 1.8 x 10" Walking (200 steps / min) with wind stress 8 x 104 to 1.5 x 10" (hair dryer directed toward floor) Reported by Fish [1964] Vigorous work activity, including sweeping 1.9 x 10' Vigorous walking 3.9 x 104 Light work activity 9A x 104 in order to develop a distribution that represents the average conditions in the residence, the average or effective activity level must be determined. Robinson and Thomas (1991) summarize the results of a national survey on time spent in activities. This survey, conducted in 1985, is based on averages from diaries kept by 1,980 adults (921 men) over a 2 month period. In this survey, adult men spent an average of 886 minutes per day at home,6 minutes per day cleaning the house (vigorous activity) and 483 minutes sleeping. Some of this time at home was spent in the yard or garage, using the uata presented for Califomia, the time spent at home outside is approximately 37 minutes per day, leaving approximately 849 minutes per day indoors. Of the time at home spent indoors approximately 0.7 percent is vigorous activity,57.2 percent sleeping (no activity), and the remaining 42.1 percent is spent in moderate to low activity. Given this estimate of how adult males time is spent indoors, the effective parameter n (Q' ) Residential Scenario 5.4 -14 January 31,1998 1 . .
value should be a time weighted average of the Rf, for each activity category. As can be seen in Table 5.4.6 resuspension during the low to moderate activities while awake will dominate. Table 5.4.6 Time Weighted Resuspension Factors O Activity Range of Rf,(md ) Fraction of Time Weighted Time Range RF, Sleeping 7.7 x 1040 to 1.5 x 0.572 4.4 x 1040 to 8.6 x 4 10-7 10 Awake (not sweeping) 3 x 10-7 to 1.8 x 0.421 1.3 x 10 7 to 7.6 x 10d 104 Awake (vigorous, 9.4 x 104 to 4x 0.007 6.6 x 104to 2.8 x sweeping) 10d 10 4 PDF for RF, The variability and uncertainty in the resuspension factor is best represented by a log-unifarm distribution with a lower limit of 1 x 107/m and an upper limit of 8 x 1041/m. The range of values is defined by the time weighted minimum and maximums of measured values for resuspension under low to moderate activities. The use of a log-uniform distribution ensures that the selection of a particular magnitude of RF,within this range will be equally likely. This distribution reflects the uncertainty in the effective model parameter value given limited data on the relative amount of time spent at different activity levels by adult males indoors at home. PARAMETER UNCERTAINTY: The proposed distributions describing the variability in the parameters representing indoor and outdoor dust-loading are determineo by the following assumptions that introd' ice uncertainty in the distributions:
. Respirable articles are less 10 um in diameter, as defined in the NUREG/CR-5512 exposure model; . there are an equal number of sites in each of the two climate categories (arid and humid). . airborne contaminated particles will have a distribution of sizes such that there is net deposition indoors, and - the long-term average PF is in the range 0.2 to 0.7 for all sites, indoor activities, outdoor activities and future houses, - resuspension of loose particles indoors occurs by a combination of wind stress from normal building ventilation and mechanical disturbances from walking and other activities (e.g.,
cooking, sweeping, running, playing, exercising, working, reading, watching television; and a resuspension factor values are reported to depend to some extent on a number of other factors, including surface texture and roughness (in this case the type of floor covering), particle size distribution, type of deposition, and chemicci oroperties of the contaminant and surface. These factors are assumed to produce site-to-site wiations in resuspension factor values. Residential Scenario 5.4-15 January 31,1996
,] , VARIABILITY ACROSS SITES:
The outdoor dust-loading factor would be expected to vary from site to site due to local climatic conditions, differences in the activities at the site, use of the property and activities that are likely to occur. The indoor dust-loading factor would be expected to vary from site to site due to differencee in the act,vities at the site resulting in an uncertain distribution of the airborne particle sizes and to a lesser degree to the ability of the house to filter and prevent infiltration. Po is a function of the amount of contaminated soil tracked indoors and stored on the floor surface. The average floor dust loading for an entire house will depend on the relative amount of smooth (wood, tile or linoleum) verses rough (carpet) floor covering, the construction style (number of stories) and to a lesser degree the cleaning habits of the occupant. There may be regional differences in the indoor dust-loading factors due to construction styles and climatic differences. The resuspension factor will vary across sites due to differences in the use of the properties, and due to factors unrelated to the use of the property such as surface chemistry and topography. Variations due to differences in radionuclides, surfaces, type of deposition, particle size, surface chemistry, and the nature of the surface are assumed to be unca*ollable by the IDensee. Several of the physical factors influencing dust loading and resuspension may be plausibly Nunded by characteristics of the site, or controlled by the licensee in an effort to support a site-specific values for these parameters.
REFERENCES:
. r l ( l 29 CFR 1910.1000.1990. U.S. Occupational Safety and Health Administration, Department of LJ Labor, " Occupational Safety and Health Standards - Subpart Z - Toxic and Hazardoas Substances." U.S. Code of Federal Regulations. Alzona, J., Cohen, B.L., Rudolph, H., Jow, H.N., and Frohliger, J.O.,1979. " Indoor-outdoor ! relationships for airborne particulate mater of outdoor origin," Atmospheric Environment 13,55-60. Anspaugh, L. R., J. H. Shinn, P. L. Phelps, and N. C. Kennedy.1975. "Resuspension and Redistribution of Plutonium in Soils." Health Physics 29:571-582. Brodsky, A. "Resuspension Factors and Probabilities of Intake of Material in Process (or AIS 10-6 a Magic Number in Health Physics?" Health Physics,39(4), pp. 992-1000, December 1980. Brunskill, R. T., The Relationship Between Surface and Airbome Contamination, in B. R. Fish ed., Surface Contamination Symposium Proceedings, pp. 93-105, June 1964, Gatlinburg, Tennessee, Pergamon Press, New York. Calabrese, E.J. and Stanck, E.J.,1992. "What proportion of household dust is derived from outdoor soil?,' Journal of Soil Contamination 1(3), 253 - 263. fm ( \ Residential Scenario 5.4 -16 January 31,1998 L) J l _ _ _ - _ _ _ _ _ _ _ _ _ _ . 1
Cohen, A.F. and Cohen, B.L.,1980. " Protection from being indoors against inhalation of suspert:ed matter of outdoor origin," Atmospheric Environment 14,183184. Colome, S.D., Kado, N.Y., Jaques, P., and Kleinman, M.,1992. " Indoor outdoor air pollution relation: particulate matter less than 10 pm in aerodynamic diameter (PM10) in the homes of asthmatics,' Atmospheric Environment 26a,2173-2178. Division of Air, New York State Air Quality Report. Continuous and Manual Air Monitoring Systems: AnnualReport 1981. (New York State Department of Environmental Conservation, 1982). Dockery, D.W. and Spengler, J.W.,1981. ' Indoor-outdoor relationships of respirable sulfates and particles," Atmospheric Environment 15,335-343. I Fergusson, J.E., Forbes, E.A., Schroeder, R.J. and Ryan, D.E.,1986. "The elemental composition and sources of house dust and street dust," Science and Total Environment 50, 217-221. Fish, B. R , Walker, R. L., Royster, G. W., and Thompson, J. L. "Redispersion of Settled Particles", in B. R. Fish ed., Surface Contamination Symposium Proceedings, pp.75-81, June 1964, Gatlinburg, Tennessee, Pergamon Press, New York. Freed, J.R., Chambers, T., Christie, W.N. and Carpenter, C.E.,1983. Methods for assessing exposure to chemical substance, EPA 560/5-83-015 Vol.2, pp.70-73, U.S. EPA Office of Toxic Substances. Garland , J. A. Resuspension of Particulate Material From Grass. Experimental Programme 1979-1980. London: HSMO;AERE-R10106; 1982. Glauberman, H., Bootmann, W. R., and Breslin, A. J., " Studies of the Significance of Surface contamination", in B. R. Fish ed., Surface Contamination Symposium Proceedings, pp. 169-178, June 1964, Gatlinburg, Tennessee, Pergamon Press, New York. Hagen, L. and N. O. Woodruff, (1973). " Air Pollution in the Great Plains *, Atmos. Env. 7,323-332. Hartmann, G., C. Thom, and K. Bachmann,1989. " Sources for Pu in Near Surface Air", Health Physics 56(1), 55-69. Hawley, J. K.1985. " Assessment of Health Risk from Exposure to Contaminated Soil." Risk Analysis 5(4),289-302. Hintcn. T. G., P. Kopp, S. Ibrahim, I. Bubryak, A. Syomov, L. Tobler, and C. Bell " Comparison of Technique used to Estimate the Amount of Resuspended Soil on Plant Surfaces," Health a Physics,68(4), pp. 523-531, April 1995. Hinton, D., J. Sune, J. Suggs, and W. Bamard.1986. Inhalable Particulate Network Report: Residential Scenario 5.4 -17 January 31,1998
Operation and Data Summary (Mass Concentrations Only). Vol.1-3, U.S. Environmental O Protection Agency, Washington, D.C. '\') l Hollander, W., *Resuspension Factors of 137Cs in Hannover After the Chemobyl Accident,' Aerosol Science, 25(5), pp. 789-792,1994. Homa, S. G. "The hotspot health physics codes," Health Physics,68(6 Supp):S59, June 1995. it.EA, " General Models and Parameters for Assessing the Environmental Transfer of Radionuclides from Routine Releases," Vienna: lAEA; Safety Series No. 57; 1982. I l lAEA, ' Derived Intervention Levels for Application in Controlling Radiation Doses to the Public in the Event of a Nuclear Accident or Radiological Emergency," Vienna: lAEA: Safety Series A No. 81; 1986. IAEA,'Modeling of Resuspension, Seasonalty and Losses During Food Processing," First Report of the VAMP Terrestrial Working Group, IAEA-TECDOC-6471,1992. IAEA," Validation of Models Using Chernobyl Fallout Data from Southem Finland. Scenario S," Second Report of the VAMP Multiple Pathways Assessment Working Group, IAEA TECDOC-904, September 1996. Jon.,s, l. S., and Pond, S. F., 'Some Experiments to Determine the Resuspension Factor of Plutonium from Various Surfaces,' in B. R. Fish ed., Surface Contamination Symposium m Proceedings, pp. 83-92, June 1964, Gatlinburg, Tennessee, Pergamon Press, New York. (d
) Kathren, R. L., in : Proc. Symp. On Radiological Protection of the Public in a Nuclear Mass Disaster, Interlaken, Swit.,26 May - 1 June,1968 (Bem: EDMZ).
Kennedy, Jr., W.E., arid D.L. Strenge,1992. " Residual Radioactive Contamination from Decommissioning: Technical Basis for Translating Contamination Levels to Annual Total Effective Dose Equivalent," NUREG/CR-5512, U.S. Nuclear Regulatory Commission, Washington, DC. Langham, W. H., USAEC Rept. USRL-50639 (Livermore: Lawrence Livermore Laboratory), 1969. Langham, W. H., in: Proc. Environmental Plutonium Symp., Los Alamos,4-5 August 1971, p. 3 (Los Alamos: Los Alamos Scientific Laboratory),19721. Leonard B. E. ARn-222 Progeny Surface Deposition and Resuspension Residential Materials.@, Health Physics,69(1), pp. 75-92, July 1995. Lillie, R. J.1970. Air Pollutants Affecting the Performance of Domestic Animals. U.S. Department of Agriculture, Washington, D.C. Magill, P. L., R. R. Holden, and C. Ackley, eds.1956. Air Pollution Handbook. McGraw Hill, New York OgD Residential Scenario 5.4-18 January 31,1998
Mitchell, R. N., and Eutsler, B. C., 'A Study of Beryllium Surface Contamination and Resuspension," in B. R. Fish ed., Surface Contamination Symposium Proceed:ngs, pp. 349 352, June 1964, Gatlinburg, Tennessee, Pergamon Press, New York. Moulin, M., C. E. Lambert, F. Dulac, and U. Dayan,1997. " Control of atmospheric export of dust from North Africa by the North Atlantic Oscillation", Nature 387(12),691-694. Nair, S. K., C. W. Miller, K. M. Thiessen, and E. K. Garger "Modeling the resuspension of radionuclides in Ukrainian regions impacted by Chemobyl fallout," Health Physics,68(6 Supp):S46, June 1995. Nair, S.K., Miller C. W., Thiessenn K. M., Garger E. K., Hoffman F. O. "Modeling the Resuspension of Radionuclides in Ukrainian Regions impacted by Chernobyl Fallout," Health Physics,72(1), pp. 77-85, January 1997. New York State Department of Environmental Consarvation (1982) (note, insert missing g reference) NRC, " Reactor Safety Study: An Assessment of Accident Risk in U.S. Commercial Nuclear Plants, Appendix VI. Calculation of Reactor Accident Consequences,' Rep. WASH-1400, 1975. Roberts, T. M., W. Gizyn, and T. C. Hutchinson,1974. ' Lead Contamination of Air, Soil, Vegetation and People in the Vicinity of Secondary Lead Smelters," Confer. Trace Subst. Environ. Health 8,155-166. Robinson, J.P. and Thomas, J.,1991. Time Spent in Activities, locations and Microenvironments: A Califomia - National Comparison, Project Report, EPAl600/4-91100, p. 83. Rognon, P.,1991. " Field measurements of dust near the ground, correlated with surrounding soils in the Sahara and Sahel", Z. Geomorph. N. F. 35(4),491-501. Rutz, E., J. Valentine, R. Eckard, and A. Yu,1997. " Pilot-Study to Determine Levels of Contamination in Indoor Dust Resulting from Contamination in Soils", J. Soil Contamination 6(5),525-536. Sehmel, G. A.1975. " Atmospheric Dust Si:e Distributions as a Function of Wind Speed." In Pacific Northwest Laboratory Annual Report for 1974, BNWL-1950-3, Pacific Norlhwest Laboratory, Richland, Washington. Sehmel, G. A.1977a. Transuranic and Tracer Simulant Resuspension. BNWL-SA-6236, Pacific Northwest Laboratory, Richland, IVashington. Sehmel, G. A.1984. " Deposition and Resuspension." In Atmospheric Science and Power Production. DOE / TIC-27601, U.S. Department of Energy, Washington, D.C. Residential Scenario 5.4 -19 January 31,1998
Sehmel, G. A.1977b. Radicactive Particle Resuspension Research Experiments on the Hanford Reservation. BNWL-2081, Pacific Northwest Laboratory, Richland, Washington.
~k Shinn, J. H., D. N. Homan, and W. L. Robinson.1989. Resuspension Studies at Bikin.' Atoll.
! 4JCID-18538-Rev.1, Lawrence Livermore National Laboratory, Livermore, California. Sinclair, P, C.1976. " Vertical Transport of Deser1 Particulates by Dust Devils and Clear Thermals." In Atmosphere-Surface Exchange of Particulate and Gaseous Pollutants. CONF-740921, U.S. Department of Energy, Richland, Washington. Soldat, J. K., J. G. Droppo, Jr., W. H. Rickard, and L. G. f-aust.1973. Assessment of the EnvironmentalImpact of the Retrievable Surface Storage Facility. BNWL-B-313, Pacific Northwest Laboratory, Richland, Washington. Solomon, R. L and J. W. Hartford,1976. ' Lead and Cadmium in Dusts and Soils in a Small Urban Community," Environ. Sci. Technol. 10,773-777. 4.. ling, T. D. and D. M. Kobayashi,1977. " Exposure to Pollutar..s in Enclosed 'Living Spaces'.' Environ. Res.13,1-35. Stem, A. C., ed.1968. Air Pollution. 2nd ed. Acadernic Press New York. Stewart, K.1964. "The Resuspension of Particulate Material from Surfaces." In Proceedings of the Surface Contamination Symposium. B. R. Fish ed. Pergamon Press, New York. C'\i t Tegen, l. and I. Fung,1994. 'Modeling of mineral dust in the atmosphere: Sources, transport, V and optical thickness', J. Geophysica! Research 99(11),22,897 - 22,914. Thatcher, T.L. and Layton,D.W.,1995. ' Deposition, Resuspension, and Penetration of Particles Within a Residence,' Atmospheric Environment 29(13),1487-1497. U.S. Department of Health, Education, and Welfare (HEW).1969. Air Quality Criteria for Particulate Matter. HEW, Washington, D.C. Whitby, K. T., A. B. Algren, R. C. Jordan, and J. C. Annis,1957. "The ASHRAE Air-Bome Dust Survey,' Heating, Piping and Air Conditioning 29, Nov.185-192. g Residential Scenario 5.4-20 January 31,1998 s
l 5.5 Crop yields for vegetables, fruits, and grains consumed by humans, Y,, ar.d fh forage, Y,, stored grain, Y,, and stored hay, Yn, consumed by beef cattle, poultry, V milk cows, and layer hens (kg/m') The crop yields represent the average annual yields of garden produce (vegetables, fruit, grain) and livestock feed (hay, forage, and grain) that are grown on contaminated land and consumed by individuals and livestock at the site. Importance to Dose: The crop yielde are needed for determining the optake and transport of radionuclides in: 1) irrigation water-plant human pathway; 2) Irrigation water-forage-anima'- human pathway; 3) irrigation water stored grain-animal-human pathway; and 4) irrigation water-stored hay-animal-human pathway, ano the parameters are used to calculate the cultivated area,A,. 5.5,1 Crop yields for vegetables, fruits, and grains, Y, (kg/m') Crop yields for vegetables, fruits, and grains, Y,, estimate the amounts of garden prodt., : grown per unit area of cultivated land at the site. The model proposes using different values of Y, for vegetables (leafy), vegetables (other than leafy), fruits, and grains. The default values of 2.0 kg/m2 (leafy vegetables),4.0 kg/m' (other vegetables),2.0 kg/m2(fruits), and 1.0 kg/m2 (grains) were adopted as the default values in NUREG/CR 5512, Volume 1, and are based on information published by [Shor,1982), [Strenge,1987), and [Napier,1988). 5.5.1.1 Use of parameter in modeling: Y,is used in determining the average deposition rate p of radionuclide j to edible parts of plant v from application of irrigation water per unit average ( concentration of parent radionuclide i in water (pCi/d kg wet weight plant per pCl/L water), R,, as shown by the following (Equation 5.22, page 5.27 of NUREG/CR 5512, Volume 1): R.,,, = IR r, T/Y, [C,/C,j (5.5.1) 2 where IR is the average annual application rate of irrigation water (L/m d), r,is the fraction of initial deposition (in water) retained on the plant (pCi retained per pCi deposited), T,is the translocation factor for transfer of radionuclides from plant surfaces to edible parts of the plant (pCi in edible plant part per pCi retained), Y,is the yield of plant v (kg wet-weight plant /m2 ), Cy is the average annual concentration of radionuclide j in irrigation water over the current annual period (pCi/L water), and C., is the average annual concentration of parent radionuclide i in irrigation water over the current annual period (pCi/L water). 5.5.1.2 Additional Information Reviewed to Define Revised Values for Y,. Estimate 9 of the yields for vegetables, fruits, and grains were obtained from USDA crop reports collected during the period from 1994 to 1996. Distributions for the individual crops for the residential scenario were determined from the annual average yields and the fraction of total crop area that is devoted to each crop. Tables 5.5.1.1 thru 5.5.1.4 list the individual crops in each of the four categories (vegetables [ leafy], vegetables [other], fruits, and grains), the total land area (averaged over three years) for production of each crop, and the average annual yield (kg/m2), V Residential Scenario 5.5-1 January 31,1998
~ ~ ~ ~ ~
Tabie15.13I6 duction oiUegetable Crops (leafyiinI993~1996' __ Crop _ Area, acres Fraction Std Dev Yi_ eld (kg/m') Std Dev Artienokes ,8633,_ 0.0143. 0.0005 _ _ 1.182 _0.240 Broccoli ,____ _ 119333 _ 0.1978 _ 0.0045 _ _ 1.355 0.039
'Brus.s_el sprouts 3400 0.0056 0.0001 1.926 0.086
. Cabbage 81273 0.1348 _ 0_._0022 3.811' O.211j
, Cauliflower : 50317 3 0.0834: 0.0057 1.500 0.085 pelery !
27833] 0.0461! 0.0010: 7.401! 0.5271
; Head lettuce 204237i 0.3385' O.0095 3.582! 0.056 Leaf lettuce .
39300 0.0652I 0.0038: 2.546 0.014! Romain_e_lettucei 30813. 0.0511l0.0080 3.116, 0.015;
@pinach i 38030: 0.0630; 0.0011! 1.543 0.048
- Source: " Crop Production Annual Survey", National Agricultural Statistics Service (NASS), Agricultural Statistics Board, U.S. Dc, tment of Agriculture, Janucey 1997.
Table 5.5.1.2 Production of Vegetable Crops (other than leafy vegetables)in 19941996* _ Crop Area, acres Fraction .Std DeviYleid (kg/m2 ) .Std Dev
. Asparagus 74217l 0.02675' O.00087: 0.31278; 0.00895 s
Beans, Lima j 53767; 0.01937i0.00093 031310! 0.00687! _ Bean _s, snap __ 2942_80, 0.10599;0.00266 0 /3221! 0.04381! _ Beets 10217' O.00368 0.00015 3.11675' O.32048 _Cantalcups 103447. 0.03729.~ 0.00113 2.15530: 0.20594' Carrots _ 108323 0.03909;0.00437: 3.71015: 0.12441!
; Corn 713270 0.2570110.00669 1.43156 0.02909~
Cun'mbers 171103_ 0.06163:0.00146: 1.44046; 0.03415. Eggplant 3067l 0.00110' O.00015i 2.46063; 0.27061i 0
' Escarole 3613 0.00130:0.00007' 1.72619. 0.01439' Garlic 28667. 0.01035 0.00101! 1.90709l 0.05609 Honeydews 26000' O.00938 0.00068' 1.98313 0.18518
' Onions 161653 0.05826 0.00130 4.37844; 0.07288' Peas 280203 0.10084'O.00844 0.37263' O.00802
_ Bell Peppers ! 66700' O.02405:0.00133. 2.60024) 0.17790'
; Tomatoes ! 470387! 0.16949'O.00043' 6.26320' O.12603 katermelon 206423 0.07441;0.00224- 2.25624' O.11714j
- Source: " Crop Production Annual Survey", National Agricultural Statistics Service (NASS), Agricultural Statistics Board, U.S. Department of Agriculture, January 1997.
Residential Scenario 5.5-2 January 31,1998
Table 5.5.1.3 Production of Fruit Crops in 19941996* 9 _ _ _ _ _ _ _ _ . _ _ _
.. -._ . C rop __ Are_a, acres Fra_gtloniStd DevJield_(kg/m Apples _ _ 459703 0.1540 0.00312_
2
)Jtd Dev:
2.6400 0.1442 Aprigots 21423 0.0072 0.00009 1.0261 0.5190 Avo_cado_s _ 67670 0.0227 0.00221 0.6163 0.0718 Cherries, s_weet _ _ _47347, 0.0159 _0.0_0011: 0.8339, 0.1438' Cherries, tart i 44950 0.0151: 0.00132 0.8062 0.1452 Cranberries _32467; 0.0109 0.000201 1.5563 0.1283
. Dates _ 5127; 0.0017LO.00017; 1.0553l 0.1678 Figs j 14767L 0.0049LO.00004! 0.7606l 0.1160j
. grapes 759833- 0.2545! 0.00521! 1.7269 0.0965.
i 733! C.0002; 0.00001i 2.5727i 0.1236' [guavesK5vifruit [ 6700 0.0022) 0.00010. 1.2146 0.1251l
, Nectarines 31633' O.0106 0.00065 1.57351 0.3354I LOlives _ 33133! 0.0111! 0.00016: 0.73741 0.3214!
, Papayas 2157 0.0007. 0.00011' 2.6849 0.4294' 173072; 0.0580 0.00132;
.Peache_s 1.4875, 0.1403j
, Pears __ 70510 0.0236' O.00047i 2.9766 0.3599 Plums 41633 0.0139' O.00034' 1.0672' O.3566 I
Prunes 79300 0.0266' O.00037! 1.7867I 0.1588 G Strawberries
, Oranges Grapefruit 48610 0.0163' O.00044 763757, 0.2556LO.01114~
165297 3.7544, 0.0259 3.2780; 0.0408: 0.0553' O.00218_ 3.7573, 0.2316, Lemons 61133 0.0205 0.00054 3.5158' O.2279
' Limes 1933 0.0006 0.00001' 1.2714: 0.2656l
,Taggelos 12133_ 0.0041i 0.00017- 2.4964! 0.5210;
, Tangerines 34300_ 0.011E 0.00124 2.0941l 0.2632j
. Temples 6700 0.0022! 0.0000/: 3.4799; 0.2464l
- Source: http://mannlib. cornell.edu Table 5.5.1.4 Production of Grain Crops in 1994-1996* !
Grain Area, acres Fraction Std Dev tYleid(kg/m )2 Std Dev: C_orn 66,434,000 0.3092 0.0136; 0.7282; 0.0976
, Sorghum 9,960,000! 0.0464; 0.0056: 0.4021! 0.0401!
pats 4,802,000 0.0225; 0.0073; 0.1967, 0.0243 Barley ! 7,562,000' O.0354; 0.0059 0.2927; 0.0370. Bye i 443,000 0.00211 0.0006 0.1697 0.0103 Wheat 60,927,000 0.2838 0.0140 0.2460 0.0155 Residential Scenario 5.5-3 January 31,1998
~ '
Table 5.5.1.4 Production of Grain Crops in 19941996*
~ ' ~ ~
~Gralti[Are[acreslractiori Std Dev_.Ylald (kglm'),Std Dev Rice __ _ _,_2,870.0_00_ _0.0134_ _0.0012_._0.6398 0.0230 Flax __ _221,000 0.001_1_ _ _0.00_0_5, _ 0.0952_ 0.0276 Sunflowers. _;.485,000_ 0.0111_0.002_8,._0.1370_0.0216
. Soybeans 59,008,000 0.2764 0.0111 0.2329. 0.0258
' Source: ' Crop Production Annual Survey *, National Agricultural Statistics Service (NASS), Agricultural Statistics Board, U.S. Department of Agnwulture, January 1997.
5.5.1.3 Proposed distribution for crop yleids for vegetables, fruit, and gialn Distributions for crop yields were determined from the annual average yields for individual crops and the fraction of land for production of these crops from to the following equation: Y, = En.,,n3 Fn*Yn (3.9.2) where Y,is the total crop yield for various classifications of produce (i.e., vegetables, fruits, and grains), J corresponds to a particular crop, Fn is the fraction of the totalland area for production of crop c, and Y, is the reported yield of crop c. Figures 5.5.1.1 thru 5.5.1.8 show the PDFs and CDFs for each of the edible crops identified in NUREG/CR 5512. The mean and range of crop yields for vegetable, fruit, and grain crops are summarized in Table 5.5.1.5. Table 5.5.1.5 Average Yields and Distribution for Edible Crcps O Crop Average Yield (kg/m') Range (kg/m') Vegetables (leafy) 2.9 2.7 - 3.2 Vegetables (other) 2.4 2.32.5 Fruits >2.4 2.22.6 Grains 0.40 0.28 0.52 5.5.2 Crop yield for forage, Y,(kona') Crop yield for forage, Y,, represents the quantity of forage produced per unit area of cultivated land. The model proposes using different values of Y, for forage crops grown for consumption by beef cattle, poultry, milk cows, and layer hens. The crop yields are defined by standing biomass in NUREG/CR 5512, Volume 1 [ Kennedy and Strenge,1992), and have the following default values: beef cattle,1.5 kg/m2 ; poultry,1.0 kg/m'; milk cows,1.5 kg/mr; layer hens,1.0 kg/m'. These value were based on information published by [Shor,1982), [Strenge,1987), and [Napier,1988). Residential Scenario 5.5-4 January 31,1908
5.5.2.1 Use of parameter in roodeling: Y,is used to calculate the average deposition rate of O radionuclide j to forage crop f from application of irrigation water during the feeding period for an average unit concentration of parent radionuclide iin water (pCl/d kg wet weight plant per pCl/L water), R4. The relationship between Y,and R4 si described by the following: R4, = IR r, T/Y, (C,/C.,) (5.5.2) where IR la the annual average application rate of irrigation water (L/mr d), r,is the fraction of initial deposition of radionuclides in water retained on plant h (pCi retained per pCl deposited), T, is the translocation factor for transfer of radionuclides from plant surfaces to edible parts of the plant (pCl in edible plant parts per pCl retalnad), Y,is the yield of the forage crop f (kg wet-weight plant /m'), C.,is the average concentration of radionuclide j in irrigation water over the current annual period (pCl/L water), and C,is the average concentration of parent radionuclide I in irrigation water over the current annual period (pCl/L water). 5.5.2.2 Additional information Reviewed to Define Revised Values for Y,. stimates of the crop yields for forage were obtained from information compiled by the U.S. Department of Agriculture (USDA,1997), and the data are summarized in Table 5.5.2.1. Table 5.5.2.1 Crop Yleids for Forage Crops (USDA,1997] Year Yield (kg dry-weight /m') 1987 0.484 O 1988 1989 0.383 0.456 1990 0.473 1991 0.486 1992 0,492 1993 0.486 1994 0.503 1995 0.511 1996 0.484 5.5.2.3 Proposed distribution for crop yleids for forage The distribution for W, was besed on the average annual yield of forage crops. The binned data from Table 5.5.2.1 were fit to several functions and evaluateo. The best fit was obtained with a beta function. The distribution parameters are shown in Table 5.5.2.2. O V Residential Scenario 5.5-5 January 31,1998
Table 5.5.2.2 Distribution Parameters for Crop Yields for Forage Parameter . Value a, 2.36439 a, 1.259357 d, 0.3702 6, 0.5238 The frequency distribution and corresponding PDF for average annual yield of forage crops in Table 5.5.2.1 are shown in Figure 5.5.2.1. The calculated and observed cumulative distributions for the crop yields for forage, V., -= shown in Figure 5.5.2.2. 5.5.3 Crop yield for stored grain, Y,(kg ' ; Crop yield for stored grain, Y,, is the quantity of grain produced per unit area of cultivr'. d land. The mouel uses e single, constant value for the yield of grain crops grown for consumption by beef cattle, poultry, milk cows, and layer hens. The crop yields have the following default values: beef cattle,1.0 kg/m'; poultry,1.0 kg/m2; milk cows,1.0 kg/m'; layer hens,1.0 kg/m'. These values were based on information published by [Shor,1982), [Strenge,1987), and [Napier,1988). 5.5.3.1 Use of parameterin modeling: Y, is used to calculate the average deposition rate of radionuclide j to stored grain from irrigation water application unit concentration of parent radionuclido iin water (pCl/d kg wet weight plant per pCi/L water), R,,, The relationship between Y, and R ,is described by the following (Equation 5.53, page 5,46 if NUREG/CR-5512, Volume 1): R,, = IR r, T,/Y, [C,/C,J (5.5.3) where IR is the annual average application rate of irrigation water (Um2 d), r, is the fraction of initial deposition of radionuclides in water retained on plant h (pCi retained per pCi deposited), T, is the translocation factor for transfer of radionuclides from plant surfaces to edible parts of the plant (pCiin edible plant parts per pCl retained), Y,is the yield of stored grain g (kg wet-weight plant /m'of land), C,is the average concentration of radionuclidej in irrigation water over the current annual period (pCl/t. water), and C,is the average concentration of parent radionuclide i in irrigation water over the current annual period (pCl/L water). 5.5.3.2 AdditionalInformation Reviewed to Define Revised Values for Y,. An estimate of the crop yield for grain was obtained from USDA crop reports collected across the U.S. Tables 5.5.3.1 5.5 3.3 show the total acres harvested and the quantities aad yields of com, sorghum, and oats during the ten-year perbd beginning in 1987 [USDA,1997). Residential Scenario 5.5-6 January 31,1998
Table 5.5.3.1 Annual Production of Corn in the U.S. l Year Acres Fraction Bushels Yield (kg/m^2) ; 1987 59,505,000 0.773556 7,131,300,000 0.753
; 1988 58,250,000 0,799896 4,928,681,000 _ 0.532 1989 64,783,000 0.782706 7,531,953,000 0.730 4 1990 66,952,000 0.816607 7,934,028,000 0.744 1
1991 68,822,000 0.824137 7,474,765,000 0.682
; 1992 72,077,000 0.813299 9,476,698,000 0.826
- 1993 62,921,000 0.831848 6,336,470,000 0.633 1 1994 72,917,000 0.84936 10,102,735,000 0.871
[ 1995 64,995,000 0.853681 7,373,876,000 0.713 1996 73,147,000 0.833727 9,293,435,000 0.798 J Mean 0.817882 0.728 i Std. Dev. 0.02647 0.098 l ) Table 5.5.3.2 Annual Production of Sorghum in the U.S. Year Acres Fraction Bushels /1000 Yield (kg/m') 1987 10,531,000 0.136901 730,809,000 0.436 1988 9,042,000- 0.124166 576,686,000 0.401 1989 11,103,000 0.134146 615,420,000 0.348 1990 9,089,000 0.110858 .573,303,000 0.396 , 1991 9,870,000 0.118192 584,860,000 0.372 1992 12,050.000 0.135969 875,022,000 0.456 1993 8,916,000 0.117874 534,172,000 0.376 1994 ~ 8,917,000 0.103911 649,206.000 0.457 1995 8,178,000 0.107414 460,373,000 0.354 1996 11,901,000 0.135647 802,974,000 0.424 Mean 0.122508 0.402 Std. Dev. 0.012687 0.0401 m f 4 o mes _ .,e - e.5 7 2. _ ,31.,99e ?
, , ~.... _ __ _
Table 5.5.3.3 Annual Production of Oats in the U.S. Year Acres Fraction Bushels Yleid (kg/m') 1987 6,888,000 0.089543 373.713,000 0.195 1988 5,530,000 0.075939 217,375,000 0.141 1989 6,882,000 0.083148 373,587,000 0.195 1990 5,947,000 0.072535 357,654,000 0.216 1991 4,816,000 0.057671 243,851,000 0.182 1992 4,496,000 0.050732 294,229,000 0.235 1993 3,803,000 0.050278 206,770,000 0.195 1994 4,010,000 0.046729 229,008,000 0.205 1995 2,962,000 0.038905 162,027,000 0.196 19CS 2,687,000 0.030626 155,225,000 0.207 Mean 0.05961 0.197 Std. Dev. 0.019687 0.0243 Table 5.5.3.4 Weighted Average Annual Yield of Grain Crops Year Yield (kg dry weight /m') 1987 0.581 1988 0.428 1989 0.559 1990 0.588 1991 0.543 1992 0.657 1993 0.510 1994 0.701 1995 0.576 1996 0.642 5.5.3.3 Proposed distribution for crop yloids for grain The distribution for W, was determined from the average annual yields of grain crops. The binned data from Table 5.5.3.4 were fit to several functions and evaluated. The best fit was obtained with a normal function. The distribution parameters are shown in Table 5.5.3.5. Residential Scenario 5.5-8 January 31,1998
t Table 5.5.3.5 Distribution Parameters for Crop Yields for Grain Parameter Value p 0.57818729 0 0.077651595 The frequency distribution and corresponding PDF for average annual yield of grain crops in Table 5.5.3.4 are shown in Figure 5.5.3.1. The calculated and observed cumulative distributions for the crop yields for grain, Y,, are shown in Figure 5.5.3.2. 5.5.4 Crop yield for stored hay, Yn (kg/m8) Crop yield for stored hay, Yn, represeds the quantity of hay produced per unit area of cultivated land. The model proposes using different values of Yn for hay crops grown for consumptic,1 by beef cattle, poultry, milk cows, and layer hens. The crop yields are defined by standing biomass in NUREG/CR-5512, Volume 1 [ Kennedy and Strenge,1992), and have the following default values: beef cattle,1.5 kg/m8 ; poultry,1.0 kg/m8 ; milk cows,1.5 kg/m'; layer hens,1.0 kg/m'. These values were based on information published by (Shor,1982), [Strenge,1987), and (Napier,1988). p a 5.5.4.1 Use of parameter in modeling: The average deposition rate of radionuclide j to the e stored hay crop from irrigation water, Rg, is calculated as follows (Equation 5.48, page 5.41 of V NUREG/CR 5512, Volume 1): Rg = IR r Tn/Yn n (C,/C.) (5.5.4) where IR is the annual average application rate of irrigation water (Um2 -d), rn is the fraction of initial ciepositien of radionuclides in water retained on plant h (pCl retained per pCl deposited), Tn is the translocation factor for transfer of radionuclides from plant surfaces to edible parts of the plant (pCi!n edible plant parts per pCi retained), Y3 is the yield of the stored hay crop h (kg t wet weight plant /m ), Cy is the average concentration of radionuclidej in irrigation water over the current annual period (pCl/L water), and C,is the average concentration of parent radionuclide i in irrigation water over the current annual period (pCl/L water). 5.5.4.2 AdditionalInformation Reviewed to Define Revised Values for Yn. Estimates of the crop yieldt, for hay were obtained from data compiled by the U.S. Department of Agriculture (USDA,1997), and the data are summarized in Table 5.5.4.1. O Q ResidentialScenario 5.5-9 January 31,1998
Table 5.5.4.1 Crop Yleids for Hay Crops [USDA,1997] Year Yleid (kg dry welght/m') 1987 0.484 1988 0.383 1989 0.456 1990 0.473 1991 0.486 1992 0.492 1993 0.486 1994 0.503 1995 0.511 1996 0.464 5.5.4.3 Proposed distribution for crop yleids for stored hay The distribution for Wn was determined from the average annual yields of hay crops. The binned data from Table 5.5.4.1 were fit to several functions and evaluated. The best fit was obtained with a beta function. The distribution parametert are shown in Table 5.5.4.2. Table 5.5.4.2 Distribution Parameters for Crop Yields for Hay Parameter Value a, 2.36439 a, 1.259357 6, 0.3702 6, 0.5238 The frequency distribution and corresponding PDF for average annual yield of hay crops in Table 5.5.4.1 are shown in Figure 5.5.4.1. The calculated and observed cumulative distributions for the crop yields for stored hay, Yn are shown in Figure 5.5.4.2. 5.5.5 Parameter Uncertainty: The distributions for each of the individual crops are based on the production of all crops in direct proportion to the produs.;n across the United States. It is unlikely, however, that the applicant in the residential scenario would attempt to grow all crops. 5.5.6 Variability Across Sites: Crop yields can vary from site to site depending on the location, climatic conditions, and soil type. 5.5.7 References Kennedy, Jr., W.E., and D.L. Strenge,1992. " Residual Radioactive Contamination from Decommissioning: Tecnnical Basis for Translating Contamination Levels to Annual Total Effective Dose Equivalent," NUREG/CR 5512, U.S. Nuclear Regulatory Commission, Residential Scenario 5.5-10 January 31,1998
, .n . ~ _
f i i Washington, DC. USDA,1997. ' Crop Production Annual Survey", National Agricultural Statistics Service (NASS), Agricultural Statistics Board, U.S. Department of Agriculture, January 1997. l i I f d 1 4 T i V ResidentialScenario 5.5 11 January 31,1998
5.6 Animal feed intake rate., for forage (Q,), stored grain (Q,), and stored hay (Qn) consumed by beef cattle, poultry, milk cows, and layer hens (kg/d)-(Physical) The animal feed intake rates represent the average daily quantities of on site produced foods consumed by livestock in the residential scenario. The feed intake rates for beef cattle, poultry, milk cows, and layer hens are used in the agricultural pathway to determine the total dose due to consumption of animal products. The animal feed consumption rates are combined with the fraction of food consumed that is contaminated and plant concentiation factors to determine animal product concentration factors of radionuclides in a given quantity of product consumed by humans over the time period of interest. The default values presented in NUREGICR 5512, Volume 1 (Table 6.8, page 6.19) for foods consumed by beef cattle, poultry, milk cows, and layer hens are shown in Table 5.6.1. Note that ' intake media" is the animal product consumed by humans that transfers radionuclides as a result of the feeding habits of the domesticated animals raised by the resident Grmer; dairy < cattle produce contaminated milk; laying hens produce contaminated eggs. The transfer of radionuclides to humans from animal products also includes the direct ingestion of soil by animals while consuming fresh forage. The default value for intake rate of soil for cattle (beef and milk cows) was set to 5 percent of dry-matter intake. For poultry and egg-laying hens, the defauit intake value of soil was set to 10 percent of dry-matter intake. As summarized and concluded in Section 5.1 on the soil intake fractioi , Q,. these default values will continue to be used in the models. Table 5.6.1 Animal feed intake rates from NUREG /CR 5512, Volume 1 Intake rate (ka wet-weiaht/d) Intake media Beef Poultry Milk Eggs Fresh forage (Q,) 27 (14) 0.13 36 0.13 Stored hay (On) 14 (27) 0 29 0 Stored grain (Q,) 3 0 09 2 0.09 Determination of the wet weight intake rates reported in Table 5.6.1 was performed using the , dry weight intake rate, the percent intake by feed type, and the percent water content in the feed of interest for the animal type (from NUREG/CR 5512, Volume 1, equation 6.12, page 6.19) as follows: (Wet Weignt Intake Rate) , (Dry weight Intake Rate)(Percent Intake) (5.6.1) (100-Percent Water Content) Derivation of the default values in Table 5.6.1 assumed that the intake rate for beef cattle is based on a total daily intake of 12 kg (dry-weight), with 25% in the form of fresh forage,50% as stored hay, and 25% as stored grain. A water content of 78% is used in converting stored hay Residential Scenario 5.6 - 1 January 31,1998
_ . - - - - - . = _ _ _ _ . - - _ _ - . - _ __ _- _ __- - - and forage (dry weight) to a corresponding wet weight basis. The stored grain has a water A content of 9%. When the default values were calculated for fresh forage and stored hay for beef using equation 5.6.1, we found the corresponding values were transposed in Table 6.8 of NUREG/CR 5512, Volume 1. The corrected values are shown in parenthesis in Table 5.6.1. The intake rate for milk cows is based on a total daily intake of 16 kg (dry weight), with 50% in the form of fresh forage,40% as stored hay, anu 10% as stored grain. For poultry, the intake rates are based on a total daily intake of 0.11 kg, with 25% as fresh forage and 75% as stored graln. It is assumed that poultry do not consume stored hay or any products made from stored I hay in the residential scenario. IMPORTANCE TO DOSE: The animal feed intakes rates are used in the calculation of partial pathway transfer factors, PPTFs, for plant and animal products contaminated by soll. For a given concentration of contaminants in foods consumed by animels, the greater the animal feed intake rate, the higher the dose to humans via consumption of animal products. Default values ! assume the total annual diet for animals is derived from on site contaminated sources. 1 However, in sito specific ane";=es, this quantity can be adjusted by the fraction of foods the animals are assumed to eat from contaminated vers ~ non-contaminated areas. USE OF PARAMETER IN MODELING: 4 The animal feed intake rates, Q,, Q,, and On, are used to calculate the concentrations of radionuclides in beef, milk producing cows, egg laying hens, and meat producing poultry that consume fresh forage, grain, or hay raised in contaminated soil irrigated with contaminated water, fhose contaminated animal products are assumed to be raised and consumed on site pI t by humans. While grazing fresh forage, the transfer of contaminants from soil to animal _ ' products occurs in two different processes: 1) ingestion of contaminated plant matter (through resuspension and root uptake from soil to plants) by animals, and 2) ingestion of contaminated soil by animals during grazing. For ingestion of stored grain or stort.d hay, tne transfer of contaminants from soil to stored grain occurs by resuspension and root uptake from soil to the grain crop. imals consume the contaminated plant matter which is then converted to animal products consumed by humans. The following equations taken from NUREG/CR 5512, Volume 1, are those for fresh forage and therefore include the subscript T, Unless noted, identical equations are used for stored grain (subscript 'g') and stored hay (subscript 'h'). Note that some of the parameters in the equations have somewhat different definitions, primarily with respect to the timing of events. The references to the equations for stored hay and stored grain are also given in the following discussion. The concentration of radionuclides in fresh forage consumed by the animal (at any time) is evaluated as follows (equation 5.13, page 5.19, NUREG/CR 5512, Volume 1): C,n, = 1000 (ML, + Bf) W, A(C,3, t}/C,(0) (5.6.2) where C,ni i s the concentration factor for radionuclide j in fresh forage crop f at time t, from an initial unit concentration of parent radionuclide i in soil, ML,is the plant soil mass-loading factor
; Residential Scenario 5.0 - 2 January 31,1998
for resuspension of soil onto the forage plant f, Bp is the concentration factor for uptake of radionuclide j from the sollin fresh forage crop f, W,is the dry weight to wet weight conversion factor 'or fresh forage crop f, A{C y,1) denotes concentration of radionuclide j in soil at time t during the feeding period for fresh forage crop f, t b any point in time during the fresh forage feeding period, and C,(0) is the initial concentration of parent radionuclide iin soil at the fert of the growing period. For stored grain and stored hay, the NUREG/CR 5512, Volume 1 referenes are equations 5.12 and 5.11, respectively. For fresh forage only, the average concentration of raolonuclides in forage over the feeding period, to, is evaluated as (from equation 5.15, page 5.21, NUREG/CR 5512, Volume 1): C,9 = 1000 (ML, + Bf ) W, S{Cy , t,)/[t, C,(0)) (5.6.3) where C.,is the average concentration factor for radioriuclide j in fresh forage crop i over the leeding period at time of animal consumption of forage from an initial unit concentration of parent radionuclide i in soil, S{C , t,) is the concentration time integral factor for radionuclide j in soil over the feeding period, and t,is the feeding period for forage crop f. The concentration factor for animal products a, over the time period of feeding on fresh forage for radionuclide j for an initial unit concentration of parent radionuclide in soil, C ,, is given by (Equation 5.18, page 5.22 of NUREG/CR 5512, Volume 1): C,y, = F, Q, x, C,3 (5.6.4) where F,is the transfer coefficient that relates daily intake in animal feed and ingested soil to the concentration of radionuclide j ir an animal product a, Q,is the consumption rate of fresh forage by the animal, x,is the fraction of animal forage intake that is contaminated, and C,9 i s the average concentration factor for radionuclide j in fresh forage crop f, over the feeding period, at the time of animal consumption of forage from an initial unit concentration of parent radionuclide iin soil. For stored grain and stored hay, the NUREG/CR 5512, Volume 1 referenes are equations 5.17 and 5.16, respectively. While ingesting fresh forage only, the amcunt of soilingested while grazing is a function of the fresh forage intake rate. The average concentration factor for animal product a, over the fresh forage feeding period for radionuclide j for initial unit concentration of parent radionuclide i in soil, Cog, is given by the following (Equation 5.19, page 5.22 of NUREG/CR 5512, Volume 1): Cog = 1000 F, Q, W, Q, x, S{Cy , t,)/[t, C,(0)) (5.6.5) where Q,is the soil intake as a fraction of forage intake for the animal, W,is the dry to wet-weight conversion factor for fresh forage, ${Cy , t,)is the concentration time-integral factor for radionuclide j in f, ash forage crop i over the feeding period, t,, t, is the feeding period for the forage crop, and C,(0)ic the hitial concentration of parent radionuclide in soil at the start of the growing period. Finally, the ingestion dose from agricultural products grown in contaminated soil, secondary ingestion of soil, and ingestion of animal products is given by the following (equation 5.71, page Residential Scenario 5.6-3 January 31,1998
5.56, NUREG/CR 5512, Volume 1): A DGR, =y C DIET fg AAr g (5.6.6) 11 where DGR,is the annual dose from intake of home-grown food and animal products, Cu is the initial concentration of parent radionuclide in soil at the time of release of the she (i.e. the start of the growing season for the first year), DIET is the fraction cf annual diet duived from home-grown foods, Aqis the concentration factor for radionuclide j in soil at the baginning of the current annual exposure period per initial unit concentration of parent radionuclide 1 :,, soi' it time of site release, and AFyis the dose factor for ingestion of agricultural product per unit concentration of radionuclide j in soil at the beginning of the growing season. Revised Parameter Distribution for intake of Fresh Forage, Stored Grain, and Stored ' ay Information on the consumption of forage, grain and hay crops by beef and dairy cattle, poultry, and layer hans was obtained from National Research Council publications on the nutrient requirements of livestock [NRC,1996, and referetices cited therein). This new information both includes and supercedes the original references (such as IAEA,1982 and Till and Meyer,1983) provided in NUREG/CR-5512, Volume 1 for determining the default values for animal food intake. In the following four subsections that summarize food consumption by bof cattle, dairy cattle, i poultry, and layer hens, a consistent approach was followed for developing distributions for dry-and wet-weight matter intake for animals. The NRC publications provide average values from a number of studies for
- dry matter intake" (DMI). Those reported averages include a 12 percent moisture content.
in the following subsections, the DMI values are provided in tables and reduced to actual dry matter by backing out the 12 percent moisture content as reported. The actual dry matter data, Qw , are then used to develop distributions for the respective animal feed intake rates as dry matter. The distributions are corrected (shifted) to account for the percentage intake of food products by each animal as originally reported in NUREG/CR 5512 and as summarized above in the discussies of the default values, in Section 5.9, the distributions for W,, W,, and Wn, the wet-to-dry-weight conversion factors for forage, stored grain, and stored hay, are determined based on the following equation (usin0 the subscript T for fresh forage as an 3xample): W, = (100 - Percent Moisture, Forage)/100 (5.6.7) The dry intake rate distributions, Og, are sampled in the parameter estimation calculations, along with samples of the wet to-dry conversion factor, to derive the distributions for Q (Q,, Q,, or Qn for forage, grain, or hay, respectively) on a wet-weight basis. This calculation is based on the following (again, using the example for fresh forage): Residential Scenario 5.6-4 January 31,1998
0, = O" x Fraction of Intake (5.6S) where Fraction of Intake is tne Percent intake (equation 5.6.1) divided by 100. Therefore, Onx Pcecont Intake O= f (5.6.9) 100 - Percent Moisture of Forago This section will not report the actual distributions for Q,, Q,, or 0,,, but, rather, the distribution forQ,y only. A. Frosh Forago, Stored Grain, and Stored Hay Consumed by Beef Cattle The dominant factors that determine dry matter intake of beef cattle are physiological demand (based on body weight and age) for maintenance and potential for production, differences among breeds of beef cattle, and gastrointestinal capacity limits, in this analysis, we assume that the nutrient value of fresh forage (as well as stored grain and stored hay) is the same as the dry matter documented here. We also assume, consistent with NUREG/CR-5512, Volume 1 that fresh forage provides 25 percant of the total nutrient requirements for beef cattle, and stored grain and stored hay provide 25 percent and 50 percent, respectively, of totalintake requirements. Researchers referenced by the National Research Council (NRC,1996) i developed equations and relationships to predict and estimate dry matter intake requirements for beef cattle. One of these, Thomton et. al. (1985) reported results on 119,482 yearling British breed cattle over a 12-month period. The data in Table 5.6.2 show 14-day averages for actual daily intake of dry matter as fed to cattle (includes 12% moisture assumed by Thornton). The daily intake for cattle is a function of size and weight. The distribution for dry forage, stored grain, or stored hay, Qay, consumed by beef cattle was developed from data in Table 5.6.2 by backing out the moisture content and equally weighting the average daily dry intake rate for each age category. This distribution represents the variability of the daily ints,ke of food. The binned data were fit to several distributions and the fitness to each distribution was evaluated with a Kolmogorov-Smirnov test. The best fit was obtained with a beta distribution. Table 5.6.3 provides the beta distribution parameters for fresh forage, stored grain, and stored hay consumed by beef cattle. The frequency distribution and the corresponding PDF for the intake rate for forage by beef cattle, Qay, is shown in Figure 5.6.1. Similar PDF's for stored grain and stored hay are represented in Figures 5.6.2 and 5.6.3. The corresponding cumulative distributions for Qay for fresh forage, stored grain, and stored hay are shown in Figures 5.6.4, 5.6.5, and 5.6.6. Residential Scenario 5.6-5 January 31,1998
_. _ .- - - - - - - . . - ~ - _ - . - - - . --. - -.-_- - - Table 5.6.2 Dry Matter intake (DMI) for Beef Cattle [NRC,1996) Age (days) Weight (kg) Actual Average Dry Matter Intake (kald) DMI (No moisture) 0-14 321 7.91 6.96 15-28 329 9.91 8.72 29-42 352 9.96 8.76 43 56 374 10.04 8.84 57 70 394 10.13 8.91 71 84 415 10.18 8.96 85-98 433 10.13 8.91
- 99 112 451 9.95 8.76 113 126 468 9.50 8.36 127 140 485 8.95 7.88 Table 5.6.3 Beta Distribution Paramett,rs for Fresh Forage, Stored Grain, and SWed Hay Parameter Fresh Forage Stored Grain Stored Hay ai 1.99 1.99 1.99 a, 0.91 0.91 0.91 6, 1,69 1.69 3.38 6, 2.29 2.29 4.58 B. Forage, Stored Gmin, and Stored HayConsumed by Dairy Cattle Table 5.6.4 showt, dry matter intake for dairy cattle by body weight and milk production (NRC, 1996). Estimates of dr/ matter intake for dairy cattle are complicated by milk production rates, lactation periods, environmental factors, feed quality, body weight, and other physiological factors. Many researchers quoted in the NRC reports have proposed equations and apprnaches for predicting and estimating feeding rates. Odwongo and Conrad (1983) developed equations for predicting daily DMI for dairy cattle as shown in Table 5.6.4.
As noted above, these DMI values were corrected to actual dry matter intake, Qw, by backing out the 12 percent moisture content that was reported and correcting for the percentage of forage, stored grain, or stored hay intake for dairy cattle, in this case, dairy cattle are assumed , to derive 50 percent of total nutrient requirements from fresh forage,40 percent from stored hay, and 10 percent from stored grain. The binned data from the table were then fit to several distributions and the fitness to each distribution was evaluated with a Kolmogorov-Smimov test. The best fit for fresh forage and stored hay was obtained with a gamma distribution. For stored grain, the best fit was represented by a normal distribution. Tabic 5.6.5 provides the gamma and normal distribution parameters for fresh forage, stored grain, and stored hay consumed by dairy cattle. The frequency distribution and the corresponding PDF for the intake rate for forage for dairy cattle, Residential Scenario 5.6-6 January 31,1998 u , - . - _ _ - _ - _ - . _ . . _ ,
On, is shuwn in Figure 5.6.7. Similar PDF's for stored grain and stored hay are represented in Figures 5.6.8 and 5.6.9. The corresponding cuinulative dittnw.hos for On for fresh forage, sured grain, and stored hay dor dairy cattle are shown in Figures u.6.10,5.6.11, and 5.6.12. Table 5.6.4 Predicted Dry Matter intake (DMI)in Dairy Cattle (kg/d)(NRC,1996) BodyMfticht (ko) Milk Production 400 450 500 550 600 650 700 800 (Kg/d) DMI (ka/d) 15 14.7 15.7 16.8 17.7 18.7 19.6 20.5 22.1 20 14.9 16.0 17.1 18.0 19.0 20.0 20.9 20.5 25 14.7 15.8 16.8 17.8 18.8 19.7 20.5 22.7 30 14.5 15.6 16.6 17.6 18.5 19.4 20.3 22.2 35 16.4 17.5 18.5 19.5 20.4 21.3 22.0 40 *
- 18.3 19.4 20.4 21.4 22.4 28.6 29.0 45 20.2 21.2 22.2 23.2 ,
29.7 55 19.9 .0 22.0 23.0
- amount of feed computed was in excess of the amount that cows would be expected to eat l
Table 5.6.5 Distribution Parameters for Forage, Stored Grain, and Stored Hay Parameter Fresh Forage Stored Grain Stored Hay
~
Gamma K 2.74 2.74 A 1.15 1.43 c 6.26 5.01 Normal p 1.71 a 0.26 C. Fresh Forage and Stored Grain Consumed by Poultry Waldroup et.al.(1976), Hurwitz et.al.(1978), and the NRC (1981) derived equations and estimetes of dry matter intake based on energy needs of a growing broiler chick. Table 5.6.6 summarizes these estimates in terms of the estimated average daily dry ma"" 5take rate for poultry derived from their estimated energy needs based on age. In poultry (i. in), feeding rate generally increases with age and body weight. The published values incluoed a 12 percent moisture content which was factored into the DMI values given in the table. As above, this moisture content was then backed out to derive the intake of actual ory matter in broilers. Consistent w',th NUREG/CR 5512, Volume 1 poultry are assumed to derive 25 percent of their
- total nutrient reqerements from fresh forage and 75 percent from stored grain.
Residential Scenario 5.6-7 January 31,1998
i The binned data from the table were converted to consistent units (kg/d), corrected for the p percentage of forage or grain intake for poultry, and were then fit to several distributions and g the fitness to each distribution was evaluated with a Kolmogorov Smirnov test. The best fit was obtained with a beta distribution. Table 5.6.7 provides the beta distribution parameters for fresh forage and stored grain consumed by poultry. The frequency distribution and the corresponding PDF for the intake rates for forage and store grain for poultry, O n , are shown in Figures 5.6.13 and 5.6.14. The corresponding cumulative distributions for poultry are shown in Figures 5.6.15 and 5.6.16. Table 5.6.6 Predicted Dry Matter intake for Brollers at Different Ages [NRC,1996) Age (days) BW (g) Daily Gain (g) Est. Energy Needs DMI (g/d) Dry Matter (kcal/d) No moisture 7 100 27 102.7 28.3 24.9 14 320 34 155.6 42.8 37.7 21 560 43 212.5 58.4 51.4 28 860 56 279.9 77.0 67.8 35 1250 63 340.5 93.6 82.4 42 1690 59 378.8 104.2 91.7 49 2100 60 420.6 115.6 101.7 3 Table 5.6.7 Bets Distribution Parameters for Fresh Forage and Stored Grain - Poultry -(' Parameter Fresh Forage Stored Grain ai 1.509 1.509
- a, 1.412 1,412 5, 0.004 0.011 5, 0.028 0.085 O. Fresh Forage and Stored Grain Consumed by Layer Hens Table 5.6.8 provides estimates of the average daily dry matter intake rate for egg laying hens at different times in the egg production process. Laying hens generally attain a steady state of feed consumption once peak egg production has occurred. Byerly et. al. (1980) and Hurwitz et.
al. (1978) developed equations that characterized observed feeding behav!or of laying hens. Those equations were used to derive the dry matter intake rates given in Table 5.6.8 which confirm the steady state feeding rate when egg production stabilizes in mature hens. The published values included a 12 percent moisture content which was factored into the DMI values given in the table. Once again, this moisture content was then backed out to derive the intake of actual d_ry matter in laying hens. As with poultry, layer hens also derive 25 percent of their total nutrient requirements from fresh forage and 75 percent from stored grain. Residential Scenario 5.6-8 January 31,1998
Based on the average sisady state dry matter intake rate for mature hens, the data were converted to consistent units (kg/d) and corrected for the percentage of forage and grain intake for laying hens. The data were then binned and iit to several distributions and the fitness to each distribution was evaluated with a Kolmogorov Smirnov test. The best fit was obtained with a beta distribution. Table 5.6.9 provides the beta distnbution parameters for fresh forage and stored grain consumed by laying hens. The frequency distribution and the corresponding PDF for the intake rates for forage and store grain for laying hens, Qw, are shown in Figures 5.6.17 and 5.6.18. The corresponding cumulative distributions for laying hens are shown in Figures 5.6.19 and 5.6.20. Table 5.6.8 Predicted Dry Matter intake for Laying Hens at Different Stages of Egg Production [NRC,1996)
~
hge (weeks) Egg production (%) BW (g) Dry matter intake (g/d) Dry matter (No moisture) 20 5 1317 60.2 53.0 56.0 49.3 59.7 52.5 61.9 54.5 24 62 1513 82.2 72.3 78.2 68.8 , 81.5 71.7 83.9 73.8 28 91 1663 98.0 80.2 94.1 82.8 96.7 85.1 99.3 87.4 32 89 1737 93.2 82.0 89.4 78.7 94.6 83.2 97.2 85.5 36 87 1821 92.6 81.5 88.8 78.1 95.1 83.7 97.9 86.2 40 85 1877 88.5 77.9 84.9 74.7 98.0 86.2 95.8 84.3 Residential Scenario 5.6-9 January 31,1998
-_- - _ - - ~ _ - - . . . - . _ - - _ - -. - _ - . . . - - = . . . _ .
Table 6 " Distribution Parameters for Fresh Forage and Stored Grain Laying Hens
~~
Parameter Fresh Forage Stored Grain a, 1,43 1.43 at 0.79 0.79 5, 0.01 0.04 5, 0.02 0.07 PARAMETER UNCERTAINTY: The information in the NRC reports was based on selected models that we believe arc representative of the data required and reported. However, those models do include uncertainty. We believe the data and models as presented in the NRC reports incorporated all of the sources of uncertainty noted in the reports. VARIABILITY ACROSS SITC:#: These parameters are expected to vary to P .* mall degree iom site to site. The distributions for animal feed intake rates are established cased on averge d daily intake rates that depend on factors such as the breed of animal, the age and size of the animal, physiological response, environmental factors (particularly temperature and huridity), diet water content, quantity and quality of food stocks fed to the animals, feed process'ing methods, use of anabolic stimulants and other feed additives, timing of feeding, and production rates. All of these factors introduce variability in establishing default average daily intake rates that are applied across all sites, variability that is captured in the data and proposed parameter distributions. 7 (' Applicants may attempt to support attemative values for animal feed intake rates based on regional / seasonal variations in food availability, animal breeds, different varieties of forage and feeds available and intended for animal consumption, and intended production and use of the animal products for human consumption.
REFERENCES:
NRC,1996. " Predicting Feed Intake of Food-Producing Animals', Subcommittee on Feed intake, Committee on Animal Nutrition, Board on Agriculture, National Research Council. Nutrient Requirements of Poultry, NAS-NRC Publication 1345 (5* Edition,1066). e ,
i i i I i 1 1 1 1 Probability Density 60 .i I Legend
*i o --- M D*
. MLM Beta 1 i 40 - - - - - - - - - { f(X) 30 --- I 2.0 - l l l 1.0 l 00 1.7 1.8 1.9 2.0 2.1 2.2 2.3 l Dry forage intake (beef), kg/d i i Figure 5.6.1 Calculated Probability Distribution for Forage Consumed by Beef Cattle i 9 9 ,.- - . _ - - - - , , , , . - . . . , - . _ - . - . , , . _ 9
, I j . i l i k 1 1 i l Probability Density 60 Legend D*t * } 50 - ] MLM Beta 40 - - f(x) 30 - - l i 20 - - 1.0 - ! 00 i' 1.7 1.8 1.9 2.0 2.1 22 23 Dry grain intake (beef), kg/d l 4 ! I j Figure 5.6.2 Calculated Probability Distribution for Stored Grain Consumed by Beef Cattle j I 1
mummu im O e g SI m i A llll 1
%l l I. Im g
i t
= * - N k ~ . a a
o
?D ?o o a e
$ E$ k
- m I
= , ,
- 5
.o 1
ii U R E A $ 5 $ 8" hm
?
7 W O
. . . . _ . .. . - - . . . . - _ . . . = . . - . - . . - - . . . . . . . . . . . . . _ . . . . . . . - -
i l i l
~
t t ! f' f, ! Cumulative Density t 10 t
/ Lg% l l
/ 1
/ l l DWa )
08 r
/ i
/ --- MLM Beta ;
/ \
/
4
/
06 F(x) / s 04 J
/ (
/ t
/ /
\
/ !
i UU j.7 1.8 1.9 20 2.1 2.2 U i
- Dry forage intake (beef), kgfd
- I l
4
?
i Figure 5.6.4 Proposed Cumulative Distribution for Forage Consumed by Beef Cattle ! t l
?
I t
Cumulative Density I 1.0
/ Legend j
/ I l Dat.
08 # f __ , yty g,,, ;
/
/ t
/ t 06 ,'
F(x) /
/ ;
o4 / s
;2 .V ,
f 00 1.7 1.8 1.9 2.0 2.1 2.2 23 Dry grain Intake (beef), kg/d t f r l Figure 5.6.5 Proposed Cumulative Distribution for Stored Grain Consumed by Beef Cattle ! t O O O
i i i I l I. i
?
1 I Cumulative Density ; 1.0
/ LN l l
/ l l Data I I
' O8 # l
/ --- MLM Beta ,
)
i / 06 ,# F(x) / ! (
/ i
- 04 # < t
/ i
/ t
,/ l 02 ^ !
00 3 25 35 3 75 40 4 25 45 4 75 Dry hay intake (beef), @ i l 4 i i I i 4 f t 4 ;
; Figure 5.6.6 Proposed Cumulative Distribution for Stored Hay Consumed by Beef Cattle t
I
l.-.... 1 i i i a l Probability Density 0" t g 0 35 - Data . 03 - --- hm Ganna 0 25 i qx) j 02 . ! 0.15 L ! 0.1 - l 5 N OO 60 70 80 90 10 0 11.0 12.0 13 0 14 0 Dry forege intake (mNk),kg/d i 1 i Figure 5.6.7 Calculated Probability Distributior. for Forage Consumed by Deiry Cattle I m i ! O O O
i O O O l I Probability Density 1.75 Legend 1.5 - Data i MLM norma 1 25 10 f(X) 0 75 ! 05 l 2s > g 00 j 1.0 125 1.5 1.75 2.0 2.25 2.5 2.75 Dry grain intake (m!!k), kg/d I i I ! Figure 5.6.8 Calculated Probability Distribution for Stored Grain Consumed by Dairy Cattle l i I
i l i i i i i Frobability Density l 05 , l Legend k # Data OA MOM Gamma l l 03 - I f(X) l 02 - i 01 00 50 60 7.0 80 90 10 0 11 0 Dry hay intake (milk), kg/d Figure 5.6.9 Calculated Probability Distribution for Stcred Hay Consumed by Dairy Cattle O O O
i J l i i i t i i i i t Cumulative Density 10 t I b
// --
OB /
#- l t Data
__- Mou Gamma
/ ~
i 06 A F(x) f i
/
/ i G" // ;
J/ t t ) * >
/ l i
/I
/ -
/
/ '
00 : 60 7.0 80 90 10 0 11 0 12.0 13 0 14 0 Dry forage intake (mNk), kgfd ; t s I i t L Figure 5.6.10 Proposed Cumulative Distribution for Forara Consumed by Dairy Cattle l l I
e.c . . g i l l Cumulative Density 10 s -- f-y Legend [ l I Data
--- MLM Normal F
F(x) 0.4 0.2
~~~ ~
OO 1.0 1.25 1.5 1.75 20 2.25 2.5 2.75 Dry grain intake (milk), kg/d Figure 5.6.11 Proposed Cumulative Distribution for Stored Grain Cousumed by Dairy Cattle O O O
, l!lltj 1I \lIl O a m
m a G e a M l t a O t t DM a l C y _ i r _ a l D 0 y
- 1 1 b
- d
- e m
u s n o 7_, 0 0 1 C H y a
, d e
r o . d t 0 9 g
/ S y r t
i k f o s ) n n kI o e Wn i t D u O ( b e , 0 e i r v 8 ka t i t a g, t i D s l u i n e y v m a i t u h l a C r 0 y u j r m
/
7 D u C
,r d
/ e
, s o
' p o
0 r A, 6 P 2
/ 1
, 6 5
e r
- 0 u g
0 8 6 4 2 0 5 i 1 0 O 0 0 0 F
)
( x F O ll
l 1J 1 I l ! Il t B a e e d n a M t e a L gD M e y r L t l u o 3 0 P 0 y b d e m 5 u 2 s l 0 n 0 o C d
/
e g g a h r 2 , o y l 0 ) F t 0 y r i r o s t l f n u n e o o D ( p i t y 5 u t l 1 e b i l 0 k i r 0 a t o i b t s a i n i D b e y o r g it P 1 a r l i b 0 a 0 *o b y c r r D P d e
=
- = = t- = 5 0
0 l t a 0 u c l a C 3 1 O 6 0 O 0 0 0 0 0 0 O 0 D J 0 0 0 0 0 0 5 8 I b 5 4 3 2 1 e r u g i F I D e I I
l ljj ll ill l O . t a e d B n a M e a t goM y e t l r u LM 9 P o 0 y 0 b d e 0 m 0 u s
- 0 n
o C 7 i n
' 0 a 0 /d r g G k
y , d
) e t
i s ' myr r o n o t l u S t e o r D p f o ( y 5 n t i 0 e o l 0 k i O t i b t a u a i n b i r b o ' 4 0 n t s r i 0 ar i P D g y t y r i l i 3 b 0 D a 0 b or P 2 d 0 0 t e l a u l c l 1 0 C a 0 0 0 0 0 0 0 0 5 0 5 0 5 4 3 2 2 1 1 1 6 5 e r u g i I F N O
~
{
l ll!l1 1 !j \ ' l O t a . e d e n a Mt e a L gD M y eI - r t L - l u I - o 3 P 0 y 0 b d e m u s
/ /
5 2 0 n o 0 C
' e j g
" a i
d o
/
g F r 2 0 k o y 0 )
, f t
y n i s t r o n i l D e /, u o p b i t u ( r t e
/ 5 e s v
O 1 i i
- 0 k D t 0 at a e l
u i n i v m e g t a u l
/ a u C '
r o m 1 0 f u 0 y C r d D e s
/ o p
/ 5 0
0 0 P 5 1 6 or 5 e r u 0 g i 0 s 6 4 2 00 F 1 o 0 0 0 0
)
( x F 9 (l1l
O e I
}e=G5 r
.a A 8
i __ l E 8 il 6 a
.\ s a
\ o e
\ g
\\k \
g e i e a i a E
'As s( .
3 \ O
. A 8 o 3 e ~
e i , - 2 E c % 3 31 3 u g o
& E
\
o D $ i 8
's a 2
\
\< B $
\ ui
\ m i
E I E E
- ; a a E 7
A O
l!fi I , O t a . e d B n a M t s e gDa ML n e e H r LM 3 e y a 2 0 L . 0 y 2 2 b d 0 0 e m 1 u 2 s l 0 n . 0 o C 2 0 /d e g 0 g k a r y 9 ) o t 1 0 r F i s 0 e r n y a f o e 8 l n D 1 0 ( e o y 0 k i t t i a u l i 71 t b b 0 n i r O i t a 0 e s b g i D o r 6 1 a y 0 r P - 0 of t i l i 5 y b 1 r a 0 0 D b o r 4 1 P 0 d 0 t e 3 l a 1 0 u 0 l c a 2 1 C 0 7 0 0 0 0 0 0 C 0 0 0 0 0 0 1 0 5 0 5 0 5 3 2 2 1 1 6 5 er u g i F I N O il l I
llj 1l! l O - t a e d B s n n e h ML e gD aM H r e e y LE 7 L a y 0 b 0 d
-- - - - - - - 5 u
e m 6 s 0 n 0 o C i n a r 6 0 /d G 0 g k d y , e r t i s
)
r t o n e S 5 y e 5 0 a r o D 0 l ( f y e n t i k o l i a i t t u O b 5 n b a 0 i i r b 0 n o r i a i t, r r D P g y 5 y t 4 r i l 0 D i b
- 0 a
0 0 4 b P d t o r e l a u 0 0 0 0
- 0 0 0 0 5
3 0 l C c a 0 0 0 0 0 0 9 7 6 5 4 2 0 2 0 1 8 1 6 5 e r u g i
) F
( X f O i 1l !!
O W W c = m e E = mO-9 ' 9 Z
.J l w G
- i A h '
O h J3
% h s
sg " C 8 0 Nx\ a O V r O h h E p e i
\
$f Q g C CD o ,t a=
Q # a S T
- r
> \ 5
% % 5E 5 s
\
- -m 2 u \
i
= 2 Q o
.I=
1 i O o o
\ 5
\ o I.
e a
\
3 Q O
\ Q m @
- va O
m U C O O w N Qo @ w Q Q O O O g E O
g 5.7 Vegetation Concentration Factors For Uptake, Bp (unitless) ( w The concentration factors for uptake by vegetation, By, as defined for NUREG/CR-5512, Volume 1 dose modeling, estimate the amount of radionuclide uptake by plants grown in contaminated soil for both human consumption and as forage and feed for animals. The model uses a single, constant value for each contaminant for each of the following plant types: vegetsbles (" leafy
- and " root"), fruits, and grains. Each value should, thus, represent the average uptake for each of these cultivar groups.
This section first includes information general to the modeling of dose using By . Following this is a bricf discussion of the default values for By. Also included is discussion of the many factors that contribute to uncertainty in By. Following this, potential revisions to the default uptake concentration factors are discussed. IMPORTANCE TO DOSE The concentration factor for uptake is important to modeling dose since the higher the value for By, the higher the CEDE value for ingestion via the agricultural pathway (i.e., soil-plant-human and soil plant animal-human). USE OF CONCENTRATION FACTOR FOR UPTAKE IN MODELING The concentration factor for uptake (By )is used to calculate the concentration factor (C.,) for a radionuclide in a plant at harvest from an initial soil concentration of parent radionuclide. The g mathematical relation between By and C., is given in NUREG/CR-5512, Volume 1 (Equation Q 5.5, page 5.12) by: C.,p = 1000 (ML, + By) W, A{C y, t,,,)/ C.,(0), where - (5.7.1) C. , = concentration factor for radionuclide j in plant v at harvest from an initial unit concentration of parent radionuclide i in soil (pCi/kg wet-weight plant per pCi/kg dry-weight soil), By = concentration factor for uptake of radiauclide j from the soil ir. plant v (pCi/kg dry-weight plant per pCi/kg dry-weight soil), ML, = plant soil mass-loading factor for resuspension of soil to plant type v (pCi/kg dry-weight plant per pCi/kg dry-weight soil), ) W, = dry-weight-to-wet-weight conversion factor for plant v (kg dry-weight plant J per kg wet-weight plant), A{C y, t,y} = decay operator notation used to develop the concentration of radionuclide j in soil at the end cf the crop-growing period, t,,(pCilg dry-weight soi!), Cy = concentration of radionuclide j in soil during the growing priod (pCi/g dry-weight soil), C ,(0) = initial concentration of parent radionuclide i in soil (pCi/g dry-weight soil), t,, = growing period for food crop v (d), and 1000 = unit conversion factor (g/kg). The units of radionuclide activity are not always in pCl. However, as long as the units of activity
\ Residential Scenario 5.7-1 January 31,1998
for the plant and the soil are the same, the ratio of plant to soil concennation is presers ed and can be used to compare data from different sources. DEFAULT UPTAKE CONCENTRATION FACTORS Default soil to-plant concentration factors are given for leafy vegetables, root vegetables, fruits, and grains (NUREG/CR 5512, Table 6.16, repeated here as Table 5.7-1). Leafy vegetables are part of the ' vegetative' portion of plants, while all the other categories are considered
" reproductive" portions of plants. Therefore, there are values for yB for four vtgetation categories and 82 elements, for a total of 328 values. However, for nearly all the elements, there is one value given for hafy vegetables and one value that is given for all the reproductive crop types, reducing the number of values for By from 328 to approximately 164.
All but a few of the default values were obtained from Baes et al. (1984). The remainder come from a compilation of the Intemational Union of Radioecologists (IUR 1989), except for the element californiura, for which default values were taken from Strenge et al. (1987). Most of the default values taken from Baes et al. (1984) are the geometric means of data distributions. For many elements Baes et al. (1984) also provide the geometric standard deviation. The range between two standard deviations from the mean for a single element often exceeds two orders of magnitude. Soil-to-plant concentration factor distributions with ranges several orders of magnitude apart are not uncommon (Arkhipov et al 1975, Dahlman et al.1976, Whicker 1978, and Sheppard and Evenden 1988). The variability iri concentration factors is the result of numerous and complex underlying processes such as climate, grewing conditions, plant metabolism, plant rooting traits, soil type, soil moisture, soil texture, and soil pH. It is unlikely that a single, average value adequately captures the variability and uncertainty in climate, location, soil conditions, and plant physiology inherent in the concentration factor. Rather, it seems reasonable to provide distributions, when possible, for concentration factors. INFORMATION REVIEWED TO DEFINE ELEMdNT-SPECIFIC PDFs FOR B y A lognormal distribution is consistently proposed as the most appropriate distribution fer concentration factors (Gilbert and Simpson 1985, Shermard and Evenden 1988, Sheppard and Evenden 1990, and Murphy and Tuckfield 1992). Because By is the product of several variables, the lognormal distribution helps to solve for the uncertainty in By (Sheppard and Evenden 1988). The lognormal distribution bounds By by zero and allows Byto go to infinity at probabilities approaching zero. At some level of contaminant concentration for each plant and each element, Byis bound by a toxicity limit. Rarely are these limits observed experimentally. Lacking this information, the mean plus two standard deviations is used here to set the upper bound on each distribution. This assumes the measured, experimental variability (as expressed by the stanoard deviation) adequately encompasses the actual uncertainty that exists in By. This also modifies the distributions from lognormal, which extend to infinity, tt 1 ma distributions, which can be bound by an upper limit. Residential Scenario 5.7-2 January 31,1998
It is not likely that site-specific informatior, can reduce the uncertainty in concentration factors. There are simply too roany factors affecting B y, factors that vary non- nnearly in time and across ( locations, even to determine which ones might be the most important to predicting By (and thus, reducing uncertainty) at a particular site. It is known that the inclusion of environmental variables, such as soil texture and pH, reduces the variability in concentration factors only marginally (Sheppard and Evenden 1990). Thus, there is no benefit in correlating By to site-specific parameters such as precipitation or soil properties. Distribution parameters were taken from Ng et al. (1982 and Bt.4 et al. (1984) (Table 5.7 2). For the elements reponed in Ng et al. (1982), the geometric means and geometric standard deviations (GSD) were taken directly from the text. For data given in Baes et al. (1984) geometric means are provioed in the text, but the GSDs are provided only graphically and only for some elemeras. In lieu of visual estimation of the GSD for an element, a " generic" GSD proposed by Sheppard and Evenden (1990) was used. This GSD (2.47) was Jetermined from a pool of 23 elements and more than 1,250 values for By . Sheppard and Evenden (1990) demonstrate that the variance of Byis unrelated to site or element characteristics, suggesting that a generic GSD is appropriate for stochastic modeling of plant uptake. Bewise Ng et al. l ,1982) includes more detailed informrtion on distribution parameurs of By than Baes et al. l (1984), Ng et al. (1982) was used as the primary source for B y values. No revisions were required to the distributions of By as Dey encompassed concentratiori factors found in other reports. l All the data from Baes et al. (1984) are given in units of pCi plant dry-weight per pCi soil dry-i weight. Ng et al. (1982) give the data for leafy vegetation in units of pCi plant dry-weight per p pCi soil diy-weight and for reproductive vegetation in units of pCi plant wet-weight per pCi soil
;O I dry-weight. In developing the input for this paramter for the DandD coje, the distribution of dry-to-wet weight conversion factors will be sampled from and the selected conversion factor applied to the values of By to estimate plant radionuclide concentrations. Note that 1-(dry-to-wet weight conversion factor) must be used to calculate a wet-to-dry weight conversion factor.
Table 5.7-1. Default soil-to-plant concentration factors from NUREGICR-5512 (Table 6.16, pages 6.25-6.27), pCi/kg dry weight per DCi/kg soil. Element Leafy veoetables Root Veaetables Fruit Grain _ H *
- Be 1.0E-2 1.5E-3 1.5E-3 1.5E-3 C 7.0E-1 7.0E-1 7.0E-1 7.0E-1 N 3.0E+1 3.0E+1 3.0E+1 3.0E+1 F 6.0E-2 6.0E-3 6.0E-3 ___ 6.0E-3 Na 7.5E-2 5.5E-2 5.5E-2 5.5E-2 Mg _ 1.0E+0 5.5E-1 5.5E-1 5.5E-1 Si 3.5E-1 7.0E-2 7.0E-2 7.0E-2 P 3.5E+0 3.5E+0 3.5E+0 3.5E+0 m
( Residential Scenario 5.7-3 January 31,1998
Table 5.71. Default soil-to plant concentration factors from NUREG/CR-5512 (Table 6.16, pages 6.25-6.27), pCi/kg dry weight per pCi/kg soil. Leafy veaetables Root Veaetables Fruit flement Grain 3 1.5E+0 1.5E+0 1.5E+0 1.5E+0 Cl 7.0E+1 7.0E+1 7.0E+1 7.0E+1 Ar " " " K 1.0E+0 5.5E-1 5.5E-1 5.5E-1 Ca 3.5E+0 3.5E-1 3.5E-1 3.5E-1 Sc 6.0E-3 1.0E-3 1.0E-3 1.0E-3 Cr 7.5E-3 4.5E-3 4.5E-3 4.5E-3 Mn 5.6E-1 1.5E-1 5.0E-2 2.9E-1 Fe 4.0E-3 1.0E-3 1.0E-3 1.0E-3 La 8.1 E-2 4.0E-2 7.0E-3 3.7E-3 , Ni 2.8E-1 6.0E-2 6.0E-2 3.0E-2 Cu 4.0E-1 2.5E-1 2.5E-1 2.5E-1 Zn 1.4E+0 5.9E-1 9.0E-1 1.3E+0 Ga 4.0E-3 4.0E-4 4.0E-4 4.0E-4 As 4.0E-2 6.0E-3 6.0E-3 6.0E-3 Se 2.5E-2 2.5E-2 2.5E-2 2.5E-2 Br 1.5E+0 1.5E+0 1.5E+0 1.5E+0 Kr " " " Rb 1.5E-1 7.0E-2 ". 3 E-2 7.0E-2
~
Sr 1.6E+0 8.1 E-1 1.7E-1 1.3E-1 Y 1.5E-2 6.0E-3 6.0E-3 6.0E-3 Zr 2.0E-3 5.0E-4 5.0E-4 5.0E-4 Nb 2.0E 2 5.0E-3 5.0E-3 5.0E-3 Mo 2.5E-1 6.0E-2 6.0E-2 6.0E-2 Tc 4.4E+1 1.1 E+0 1.5E+0 7.3E-1 P.u 5.2E-1 2.0E-2 2.0E-2 5.0E-3 Rh 1.5E-1 4.0E-2 4.0E 2 4.0E-2 Pd 1.5E-1 4.0E-2 4.0E-2 4.0E-2 Ag 2.7E-4 1.3E-3 8.0E-4 1.0E-1 Cd 5.5E-1 1.5E-1 1.5E-1 1.5E-1 in 4.0E-3 4.0E-4 4.0E-4 4.0E-4 > Sn 3.0E-2 6.0E-3 6.0E-3 6.0E-3 Sb 1.3F-4 5.6E-4 8.0E-5 3.0E-2 Te 2.5E-2 4.0E 3 4.0E-3 4.0E-3 Residential Scenario 5.7-4 January 31,1998 1 \
l Table 5.71. Default soil-to-plant concentration factors from NUREG/CR-O 5512 (Table 6.16, pages 6.25-6.27), pCi/kg dry weight per pCi/kg soil. V r Element Leafy veoetables Root Veoetables Fruit Grain 1 3.4E-3 5.0E-2 5.0E 2 5.0E-2 Xe Cs 1.3E-1 4.9E-2 2.2E-1 2.6E-2 Ba 1.5E-; 1.5E-2 1.5E-2 1.5E-2 La 5.7E-4 6.4E-4 4.0E-3 4.0E-3 Ce 1.0E-2 4.0E-3 4.0E ': 4.0E-3 Pr 1.0E-2 4.0E-3 4.0E-3 4.0E-3 ' Nd 1.0E-2 4.0E-3 4.0E-3 4.0E-3 Pm 1.0E 2 4.0E-3 4.0E-3 4.0E-3 Sm 1.0E-2 4.0E-3 4.0E-3 4.0E-3 Eu 1.0E-2 4.0E-3 4.0E-3 4.0E-3 Gd 1.0E-2 4.0E-3 4.0E-3 4.0E-3 Tb 1.0E-2 4.0E-3 4.0E-3 4.0E-3 Dy i 1.0E-2 4.0E-3 4.0E-3 4.0E-3 Ho 1.0E-2 4.0E-3 4.0E-3 4.0E-3 Er 1.0E-2 4.0E-3 4.0E-3 4.0E-3 Hf 3.5E-3 8.5E-4 8.5E-4 8.5E-4 Ta 1.0E-2 2.5E-3 2.5E-3 2.5E-3 W 4.5E-2 1.0E-2 1.0E-2 1.0E-2 Re 1.5E+0 3.5E-1 3.5E-1 3.5E-1 Os 1.5E-2 3.5E-3 3.5E-3 3.5E-3 Ir 5.5E-2 1.5E-2 1.5E-2 1.5E-2 Au 4.0E-1 1.0E-1 1.0E-1 1.0E-1 Hg 9.0E-1 2.0E-1 2.0F-1 2.0E-1 Tl 4.0E-3 4.0E-4 4.0E-4 4.0E-4 , Pb 5.8E-3 3.2E-3 9.0E-3 4.7E-3 Bi 3.5E-2 5.0E-3 5.0E-3 5.0E-3 Po 2.5E-3 9.0E-3 4.0E-4 4.0E-4 Rn Ra 7.5E-2 3.2E-3 6.1 E-3 1.2E-3 ~ Ac 3.5E-3 3.5E-4 3.5E-4 3.5E-4 ~ Th 6.6E-3 1.2E-4 8.5E-5 3.4E-5 Pa 2.5E-3 2.5E-4 2.5E-4 2.5E-4 U 1.7E-2 1.4E-2 4.0E-3 1.3E-3 g Residential Scenario 5.7-5 January 31,1998
Table 5.7-1. Default soil-to-plant concEWion factors from NUREG/CR-5512 (Table 6.16, pages 6.25-6.27), pCi/kg dry weight per pCi/kg soil. L Element Leafy veoetables Root Veoetables Eruit Grain Np 1.3E-2 9.4E-3 1.b5-2 2.7E-3 Pu 3.9E-4 2.0E-4 4.5E-5 2.6E-5 Am 5.8E-4 4. IE-4 2.5E-4 5.9E-5 i Cm 3.0E-4 2.4E-4 1.5E-5 2.1 E-S ) l Cf 1.0E-2 1.0E-2 1.0E-2 1.0E-2
- Concentration factors for tntium are not needed because a special model is used to deermine tritium uptake in plants.
" Noble gases are not assumed to be taken up by plants.
Table 5.7 2. Revised snil-to-plant concentration factors. Leafy (non-reproductive) Vegetation (pCi dry plant mass /pCi dry soil mass) Reproductive Vegetation
- Geometnc Geometnc Element Geometric Standard High Data Geometric Standard High Data Mean Deviation Value" Source
- Mean Deviatian Value* Source' H *
? * * * *
- Be 1.0E-2 2.47E+0 4.95E+0 2 1.5E-3 2.47E+0 4.94 E+0 2 C 7.0E 1 2.47E+0 5.64E+0 3 7.0E 1 2.47E +0 5.64E+0 3 N 3.0E+ 1 2.47E +0 3.49E+1 2 3.0E+1 2.47E+0 3.49E+1 2 F 6.0E 2 2.47E+0 5.00E+0 2 6.0E-3 2.47E+0 4.95E+0 2 Na 7.4E-? 2 47E+0 5.01E+0 4.6E 3 1 4.10E+0 8.2E+0 1 Mg 1.0E+0 2.47E+0 5.94E+0 2 5.5E-1 2 47E+0 5.49E+0 2 Si 3.5E-1 2.47E+0 5.29E+0 2 7.0E-2 2.47E+0 5.01E+0 2 P 3.5E+0 2.47E+0 8.44E+0 2 3.5E+0 2.47E +0 8.44E+0 2 S 1.5E+0 2.47E+0 6.44E+0 2 1.5E+0 2.47E+0 6.44E+0 2 Cl 7.0E+1 2.47E+0 7.49E+1 2 7.0E+1 2.47E+0 7.49E+1 2 Ar " " " " " " " "
K 1.0E+0 2.47E+0 5 94E+0 2 5.fC-1 2.47E+0 5.49E+0 2 Ca 3.5E+0 2.47E+0 8.44E+0 2 3.5E-1 2.47E+0 5.29E+0 2 Sc 6.0E-3 2.47E+0 4.95E+0 2 1.0E-3 2.47E+0 4.94E+0 2 Cr 2.2E-2 2.20E+0 4.42E+0 1.3E-2 1 2.00E+0 4.0E+0 1 Mn 3.3E-1 7.60E+0 1.55E+1 1 1.2 E-1 4.90E+0 9.9E+0 *. Fe 5.6E-3 3.80E+0 7.61E+0 4.2E-4 1 3.50E+0 7.0E+0 1 Co 8.8E-2 4.70E+0 9.49E+0 1.5E-2 1 3.30E+0 6.6E+0 1 Ni 3.4E-2 3.20E+0 6.43E+0 1 2.1E-2 2.50E+0 5.0E+0 1 Cu 4.9E-1 2.60E+0 5.69E+0 4.3E-2 1 1.00E+1 2.0E+1 1 Zn 5.8E-1 2.60E+0 5.78E+0 1 1.1 E-1 3.90E+0 7.9E+0 1 Ga 4.0E-3 2.47E+0 4 94E+0 2 4.0E-4 2.47E+0 4.94E+0 2 As 4 OE-2 2 47E-+0 4.98E+0 2 6 OE-3 2.47E+0 4.95E+0 t 2 Residential Scenario 5.7-6 January 31,1998 1
[\. Table 6.7 2. Revised son-to-plant concentration factors. {
\,
Leafy (non-reproductive) Vegetation (pCi dry plant mass /pCi dry soil mass) Reproductive Vegetation' Geometric Geometnc Element Geometric Standard High Data Geometric Standard High Data Mean Deviation Value' Source
- Mean Deviation Value' Source' Se 2.5E-2 2.47E+0 4.97E+0 2 2.5E 2 2.47E+0 4.97E+0 2 Br 1.5E+0 2.47E+0 6.44E+0 2 1.5E+0 2.47E+0 6.44E+0 2 Kr " "
Rb 8.1E 1 3.60E+0 8.01E+0 1 7.0E-2 2.47E+0 5.01 E+0 2 Sr 1.8E+0 3.80E+0 9.40E+0 1 7.5E-2 3.80E +0 7.7E+0 1 Y 1.5E 2 2.47E+0 4.96E+0 2 6.0E-3 2.47E+0 4.95E+0 2 Zr 7.22 2 2.00E+0 4.07E+0 1 7.72-4 9.50E+0 1.9E+1 1 Nt' 2.0E-2 2.47E+0 4.96E+0 2 5.0E-3 2.47E+0 4.95E+0 2 Mo 2.2E+0 3.30E *0 8.80E+0 1 6.0E-2 2.47E+0 5.00E+0 2 Tc 9.5E+0 2.47E+0 4.4E+1 2 1.5E+0 2.47E+0 6.44E+0 2 Ru 6.2E-2 4.80E+0 9.66E+0 1 1.4E-3 4. .,0E +0 9.8E+0 1 Rh 1.5E 1 2.47E+0 5.09E+0 2 4.0E-2 2 47E+0 4.98E+0 2 Pd 1.5E 1 2.47E+0 5.09E+0 2 4.0E-2 2.47E+0 4.98E+0 2 Ag 4.0E-1 2.47E+0 4.94E+0 2 1.0E 1 2.47E+0 4.94E+0 2 Cd 5.5E 1 2.47E+0 5.49E+0 2 1.5E-1 2 47E+0 5.09E+0 2
. 4.0E 3 2.47E+0 4 94E+0 2 4.0E-4 2.47E+0 4.94E+0 2 Sn 3.0E 2 2.47E+0 4.97E+0 2 6.0E 3 2.47E+0 4.95E+0 2 / Sb 2.0E-1 2.47E+0 5.14E+0 2 3.0E-2 2.47E+0 4.94E+0 2
( Te 2.5E-2 2.47E+0 4.97E+0 2 4.0E-3 2.47E+0 4.94E+0 2 1 1.6E 1 3.50E+0 7.16E+0 1 4.5E-3 4.90E+0 9.8E+0 1 Xe " " " " " Cs 4.1E-2 3.50E+0 7.04E+0 1 5.0E 3 4.10E+0 8.2E+0 1 Ba 3.9E-2 2.90E+0 5.84E+0 1 1.3E-3 3.10E+0 6.2E+0 1 La 1.0E-2 2.47E+0 4.94E+0 2 4.0E 3 2.47E+0 4.94E+0 2 l Ce 2.1E-2 4 30E+0 8.62E+0 1 7.3E-4 6.20E+0 1.2 E+1 1 Pr 1.OE-2 2.47E+0 4.95E+0 2 4.0E-3 2.47E+0 4.94E+0 2 Nd 1.0E-2 2.47E+0 4.95E+0 2 4.0E 3 2.47E+0 4.94E+0 2 Pm 1.0E-2 2.47E+0 4.95E+0 2 4.0E-3 2.47E+0 4.94E+0 2 Sm 1.0E-2 2.47E+0 4.95E+0 2 4.0E-3 2.47E+0 4.94E+0 2 Eu 1.0E-2 2.47E+0 4.95E+0 2 4.02-3 2.47E+0 4.94E+0 2 Gd 1.0E 2 2.47E+0 4.95E+0 2 4.0E-3 2.47E+0 4.94E+0 2 Tb 1.0E 2 2.47E+0 4.95E+0 2 4.0E 3 2.47E+0 4.94E+0 2 Dy 1.0E-2 2.47E+0 4.95E+0 2 4.0E 3 2.47E+0 4.94E+0 2 He 1.0E-2 2.47E+ 0 4.95E +0 2 4.0E 3 2.47E+0 4.9*E+0 2 Er 1.0E-2 2.47E+0 4.95E+0 2 4.0E 3 2.47E+0 4.94E+0 2 Hf 3.5E-3 2.47E+0 4.94E+0 ; 8.5E-4 2.47E+0 4.94E+0 2 Ta 1.0E 2 2.47E+r 4.95E+0 2.5E-3 2.47E+0 4.94d+0 2 W 4.5E-2 2 47E+0 4 99E+0 2 1.0E-2 2 47E+0 4.95E+0 2 Residential Scenario (Ve\ 5.7-7 January 31,1998 1
1 l l Table 5.7 2. Revised soil-to-plant concentration factors. Leafy (non-reproductive) Vegetation (pCi dry plant mass /pCi dry soil mase Reproductive Vegetation' Geometnc Geometnc Element Geometric Standard High Data Geometric Standard High Data Mean Deviation Valueb Source' Mean Deviation Value' Source' Re 1.5E+0 2.47E+0 6 44E+0 2 3.5E-1 2.47E+0 5.29E+0 2 Os 1.5E-2 2.47E+0 4.96E+0 2 3.5E 3 2.47E+0 4.94E+0 2 tr 5.5E-2 2.47E +0 5.00E+0 2 1.5E-2 2.47E+0 4.96E+0 2 Au 4.0E 1 2.47E+0 5.34E+0 2 1.0E 1 2.47E+0 5.04 E+0 2 Hg 9.0E-1 2.47E+0 5.84E+0 2 2.0E-1 2.47E+0 5.14E+0 2 Tl 4.0E 3 2 47E+0 4.94E+0 2 4.0E-4 2/7E+0 4.94E+0 2 Pb 4.5E-2 2.47E+0 4.95E+0 2 9.0E-3 2.%7E *0 4.94E+0 2 Bi 3.5E-2 2.47E+0 4.98E+0 2 5.0E-3 2.47E+0 4.95E+0 2 Po 2.6E-3 2.47E+0 4.94E+0 2 4.0E-4 2.47E+0 4.95E+0 2 Rn "
, Ra 1.5E-2 2.47E+0 5.02E+0 2 1.5E 3 2.47L +0 4.94E+0 2 l Ac 3.5E-3 2.47E+0 4.94E+0 2 3.5E-4 2 47E+0 4 94E+0 2 Th 8.5E-4 2.47E+0 4.95E+0 2 8.5E-5 2.47E+0 4.94E+0 2 Pa 2.5E-3 2.47E+0 4.94E+0 2 2.5E-4 2.470+0 4.94E+0 2 U 8.5F-3 2.47E+0 4.96E+0 2 4.0E-3 2.47E+0 4.95E+0 2 Np 1.1 E +0 4 90E+0 1.09E+1 1 6.0E-2 3.00E+0 6.1E+0 1 Pu 4.5E-4 2.47E+0 4.94E+0 2 4.5E 5 2.473+0 4.94E+0 2 Am 5.5E 3 2.47E+0 4 94E+0 2 2.5E-4 2.47E+0 4.94E+0 2 Cm 8.5E-4 2.47E+0 4.94 E+0 2 1.5E 5 2.47E+0 4.94E+0 2 Cf 1.0E-2 2.47E+0 4.95E+0 3 1.0E 2 2.47E+0 4.95E+0 3
- Concentration factors for intium are not needed because a special modelis used to determine tntium uptake in plants.
** Noble gases are not assumed to be taken up by plants.
- Data Source 1 (pCi wet plant mass /pCl dry soil mass), indicaten with bold font; Data Sources 2 and 3 (pCi dry plant mass /pCi dry soil mass).
- geometric mean+2(GSD).
*1 = Ng et al. (1982); 2 = Baes et al. g1984); 3 = NUREG/CR-5512.
REFERENCES Arkhipov, N. P., Ye. A. Fedorov, R. M. Aleksakhin, P. F. Bondar', T. L. Kozhevnikova, and V. V.
) Suslova.1975. Soil chemistry and root accumulation of artificial radionuclides in the crop harvest. Soil Science 11:690-711.
Baes, C. F. Ill, R. D. Sharp, A. L. Sjoreen, and R. W. Shor.1984. A Review and Analysis of
' Parameters for Assessing Transport of Environmentally Relea:ed Radionuclides Through Agriculture. Oak Ridge National Laboratory Report. ORNL-5786.
Dahlman, R. C., E. A. Conditti, and L. D. Eyman.1976. Ciological pathways and chemical behavior of plutonium and other actinids in the environment, in A. M. Friedman (ed.), Actinides Residential Scenario 5.7-8 January 31,1998 1
in the Environment, ACS Symposium Series, American Chemical Society, Washington, D.C., pp. 47 80. M Gilbert, R. O. and J. C. Simpson 1985. Comparing computing formulas for estimating concentration ratios. Environ. Antl.11:25-47. Intsmational Union of Radioecologists (IUR).: 1989. Sixth report of the working group on soil-to-plant transfer factors.E RIVM, Bilthoven, The Netherlands.
- Murphy, C. E., Jr.,' and J. C. Tuckfield; 1992. Transuranic element uptake and cycling in a forest established over an old burial ground (U). Westinghouse Savannah River Company.
WSRC-MS-92-110.- Ng, Y. C.; C. S. Colsher, and S. E. Thompson. 1982. Soil-to-plant concentration factors for radiological assessments. NUREG/CR-2975. [ Sheppard, S. C. and W. G. Evenden. .1988. Critical compilation and review of plant / soil
. concentration ratios for uranium, thorium and lead. J. Environ. Radioactivity 8:255-285.- ,
- Sheppard, S. C. and W. G; Evenden; 1990. Characteristics of plant concentration ratios -
assessed in a 64-site field survey of 23 elements."J. Environ. Radioactivity 11:15-36. Strenge, D. L., T. J. Bander, and J. K. Soldat.1987. GASPAR ll-Technical Reference and User Guide. NUREGICR-4653, PNL-5907, U.S. Nuclear Regulatory Commission, Washington, D, C. ~
. Whicker, F. W,1978. Biologicalinteractions and reclamation of uranium mill tailings.
Symposium on Uranium Mil! Tailings Management, Fort Collins, Colorado, November 20-21.
.o s
b Residential Scenario 5.7-9 January 31,1998
S.8 Interception Fraction for Vegetafion, r, (unitiess) The interception fraction for vegetation, r,, as defined for NUREG/CR-5512 Volume 1 dose modeling, estimates the fraction of deposited contamination retained on various cultivars grown for food and animal feed after above ground irr;gation with contaminated water, Conceptually, the mode! proposes itsing different values of r, for plants grown both for direct human consumption: " leafy" vegetables, "other" vegetables, fruits, and grains and for indirect human consumption as animal feed: ferage plants (e.g., grass and alfalfa), grain, and hay. Mathematically, the model uses a single, constant value for all contaminants and all plant types. Thus, this value should represent the average fraction of all contaminants retained on edible plant sudaces after irrigation of all cultivars grown by a critical group member who performs both residential and light farming activities, growing for both food and animal feed. This section first includes a brief discussion of the basis for the default value for r, recommended in NUREG/CR-5512, Volume 1. Next, information general to the modeling of dose using r, is presented. Following this, potential revisions to the default retention factor based on plant type are discussed. Lastly, the applicability of using a single value (or pdf) for r, for all contaminants is discussed. The default value of 0.25, used for all plant types, is based on recommendations by Baker et al. (1976); 0.25 is also adopted as a default value in Regulatory Guide 1.109. Baker et al. (1976) provide no explanation or justification of this value. As such, the only way to evaluate the appropriateness of this value is by comparison to existing experimental data. IMPORTANCE TO DOSE The interception fraction is important to modeling dose since the higher the value for r,, the higher the CEDE value for ingestion via the agricultural pathway (i.e., irrigation water-plant-human and irrigation water-plant-animal-human). USE OF RETENTION FACTOR IN MODELING The interception fraction is used to calculate the constant, .Sveiage rate of deposition vf a contaminant to plants. The mat.kematical relation between deposition and retention is given in NUREG/CR-5512, Volume 1 (Equation 5.22, page 5.27) by: R,,,, = IR r,1, /Y, [C,,3 /C.), where (5.8.1) R ,, = average deposition rate of radionuclide j to edible parts of plant v from application of dirrigation water per unit average concentration of parent radionuclide i in water (pCi/d kg wet weight plant per pCi/L water), IR = average annual application rate of irrigation water (L/m2 dd ), r, = fraction of initial deposition (in water) retained on plant v (pCi retained per pC! deposited), T, = translocation factor for transfer of radionuclides from plant surfaces to edible parts of plant v (pCi in edible plant part per pCi retained), Y, = yield of plant v (kg wet weight /m2), Residential Scenario 5.8-1 Februcry 1,1998
2 C, = average annual concentration of parent radionuclide i in irrigation water over the f] current annual period (pCi/L water), and V C, = average annual concentration of radionuclide j in irrigation water over the current annual period (pCi/L water). Rm represents the activity retained on the edible portions of a plant after a single deposition event, even though some of the equation is dependent on annual averages (irrigation rate, and concentrations of parent and daughter radionuclides in irrigation water). Dose calculations require an estimate of the average, annual amount of a contaminant retained on a plant. In the irrigation water-plant-human pathway dose calculations, this is expressed as the amount of concentration received throughout the growing period and retained on the plant at the time of harvest (Equation 5.23, Volume 1, page 5.28): Cyn = R, [R,, t,], where (5.8.2) Cyn = concentration f.-dor for radionuclide j in plant v at harvest from deposition onto plant surfaces for an average unit concentratw.) of parent radionuclide i in water (pCi/kg wet weight plant per pCi/L water), t, = growing period for plant v (d), and R. [R,] = deposition, accumulation operator used to develop the concentration factor of radionuclide j in plant v at harvest from deposition onto plant surfaces for an average unit concentration of parent radionuclide in water (pCi/kg wet weight plant per pCl/L water). p] Mathematicallimits require that 0 s r, s 1. Because r, represents the fraction of a contaminant (.x in irrigation water that is retained on the surface of a plant, r, cannot exceed one. Neither can r, be a negative number, with a plant losing more of a contaminant than it is exposed to by irrigation. 1 ( INFORMATION REVIEWED TO DEFINE r, FOR DlFFERENT PLANT T(PES Fxperimental rasults from an interception study using contaminated, simulated rain (Hoffman et dl.1992) !ndicate that biomass density is more important than vegetation type in affecting retention; when the data are normalized for biomass, differences in vegetation type, while statistically significant, are never major controlling variables for retention. Hoffman et al. (1992) also report sim3ar results for a variety of herbaceous and woody plant types. Because dose calculations using r, account for biomass yield (Y,), the units for which are given as density (kg 2 wet weight /m ), a separate retention factor for different plant types is not necessary. Given this, the assumption that retention factors apply equally to all plant types, like the default values proposed in NUREG/CR-5512, seems an appropriate assumption, supportable by experimental evidence. INFORMATION REVIEWED TO DEFINE RADIONUCLIDE-SPECIFIC PDFs FOR r, The same experiment by Hoffman et al. (1992) provides good evidence for grouping contaminants by their ionic charge and/or solubility. The study found that anions are essentially O V Pesidential Scenario 5.8-2 February 1,1998
removed with the water once the vegetation surface becomes saturated, that cations are ceadily adsorbed to the plant surface, and that insoluble particles readily settle out on the plant surface. For cations, insoluble particles, and anions at irrigation rates comparable to those being proposed (Section 3.7), the adsorptien and settling rates are comparable, resulting in similar values of retention. Therefore, it is unnecessary to separate r,into categories based on solubility or ionic charge. This approsch is also impractical because the modeling does not require detailed knowledge of grounduter geochemistry; because it is unknown what chemical forms contamin ats might take, this leve: of detailin the r, parameter is unnecessary. The adsorption (retention) of cations and int,oluble particles on vegetation is similar, though the underlying processes differ. For cations, retention appears to be controlled by chemical adsorption to cation exchange sites in the leaf cuticle, while for insoluble materials, retention is controlled by the rapid settling out of particles from rain droplets and their consequent adsorption on the plant surface. Interception fractions for cations and insoluble particles as reported by Hoffman et al. (1992) generally range from 0.1 to 0.6 with gemetric means ranging from 0.15 to 0.37. The mean of the geometric means is 0.28. Given this, the default value of .,.25 recommended in NUREG/CR 5512 seems appropriate as an average value for the retention of contaminants on plants for this particular group of contaminants. The data provide practicallimits for r y, suggesting that the mean value of r, can be increased or decreased by a factor of two and still remain within experimenta"y-derived limits of r, Default values for r, from NUREGICR-5512 and the proposed revised values for r, are provided in Table 5.8-1. The probability distribution function of r, given three values (minimum, maximum, and mean) is modeled with a uniform distribution (Figure 5.8-1). The nterception fraction for anions, as measured with '3'l by Hoffman et al. (1992) is dependent on the amount of irrigation applied. " Low" irrigation amounts from Hoffman et al. (1982) are approximately 1-15 mm do and are de only rates applicable here, as the average irrigction rate being proposed is approximately 0.7 mm dd (Section 3.7). At low irrigation levels the average r, for anions is approximately 0.3; as with cations and insoluble particles, the defanit value of 0.25 recommended in NUREG/CR-5512 is slightly lower than that average. The data provide pract,sallimits for r y, with a range of 0.15 to 0.6, suggesting that the mean value of r, can be increased or decreased by a factor of two and still remain within experimentally-derived limits of ry . Thus, the range given for cations and insoluble particles (0.1 to 0.6) bpplies also to anions, and no further revisions to the pdf for r, are recommended. The default and revised values of r, are given in Table 5.8-1 and the pdf is shown in Figure 5.8-1. Tabie 5.81. Volume I default and pdfs for r . PDF of r, (uniform distribution) NUREGICR-Vegetation Type 5512 Default Minimum Maximum Mean leafy vegetable 0.25 0.10 0.60 0.35 Residential 3cenario 5.8-3 February 1,1998
Table 5.8-1. Volume I default and pdfs for r,. [h \ NUREGICR-PDF of r, (uniform distribution) Vegetation Type 5512 Default Minimum Maximum Mean other vegetable 0.25 0.10 0.60 0.35 fruit 0.25 0.10 0.60 0.35 grain consumed by humans 0.25 0.10 0.60 0.35 forage consumed by beef cattle 0.25 0.10 0.60 0.35 forage consumed by poultry 0.25 0.10 0.60 0.35 forage consumed by milk cows 0.25 0.10 0.60 0.35 forage consumed by layer hens 0.25 0.10 0.60 0.35 stored grain consumed by beef cattle 0.25 0.10 0.60 0.35 stored grain consumed by poultry ' O.25 0.10 0.60 0.35 stored grain consumed by milk cows 0.25 0.10 0.60 0.35 stored grain consumed by layer hens 0.25 0.10 n 60 0.35 stored hay consu'ned by beef cattle 0.25 0.10 0 60 0.35 stored hay consumed by poultry 0.25 0.10 0.60 0.35 stored hay consumed by milk cows 0.25 0.10 0.60 0.35 stored hay consumed by layer hens 0.25 0.10 0.60 0.35 ( ) () For all contaminant categories, retention is positively correlated with the total amount of biomass. This is explicitly accounted for in the model, since the modeling of dose using r,(i.e., Equation 5.22) increases with increasing amounts of biomass (Y,). The limits of r, are not likely to change with site-specific data because r,is independent of vegetation type. The effect on r, due to variability in the amount of vegetation at a site is solved for mathematically (as discussed above) and does not require site-specificity. Also, r, was measured over a broad range of irrigation conditions, assumed here to be broad enough to encor pass the expected range of variability in irrigation intensity and amount from one site to anott Jr. Hoffman et al. (1992) also demonstrate that contaminants that have dried on plant surfa :es after an irrigation event are not lost with subsequent washing; given this, r, should not be di.Jted (reduced) due to uncontaminated rain that may fall at a site after irrigation. REFERENCES Baker, D. A., G. R. Hoenes, and J. K. Soldat.1976. FOOD: an interactive cde to calculate internal radiation doses from contaminated food products. BNWL-SA-5523,8 p. Hoffman, F. O., K. M. Theissen, M. L. Frank, and B. G. Blaylock.1992. Quantification of the interception and initial retention of radioactive contaminants deposited on pasture grass by simulated rain. Atmospheric Environment 26A(18):3313-3321. (n) i l Residential Scenario 5.8-4 February 1,1998
( ' I $ 07 06 e% 0$ 04 03
'2
<} .., ~
0 01 02 0.3 04 06 06 07 retention factor Figure 5.8-1. Retention factor cumulative probability distribution function. O Residential Scenario 5.8-5 February 1,1998
l 5.9 Wet to-dry weight conversion factors for vegetables, fruits, and grains cc.nsumed p by humans, W,, and forage, W,, stored grain, W,, and stored hay, W3 , consumed by beef catt'e, poultry, milk cows, and layer hens (kg dry weight /kg wet-weight) The wet to-dry weight conversion factors for garden produce and animal feed, as defined for ( NUREG/CR 5512, Volume 1, dose modeling, estimate the dry weight of edible plants for human and animal consumption and represent the average concentration of dry matter in plants, importance to Dose: The conversion factors are needed to correct for the moisture content in edible parts of plants since both dry-weight and wet-weight factors are used in dose modeling. The soil-to-plant concentration factors for individual radionuclides are defined in terms of the dry weight of plants while the agricultural models require wet weight of plants. 5.9.1 Wet-to-dry-weight conversion factors for vegetables, fruits, and grains, W, The four wet to-dry weight conversion factors, W,(vegetables, leafy), W, (vegetables, other than leafy), W, (fruit), and W, (grain), represent the fractions of dry matter in garden p*uce grown o., contaminated land. Table 5.9.1.1 lists the plant types and the carresponding default conversion factors used in NUREG/CR 5512, Volume 1. The conversion factors were taken from [Till,1983). Table 5.9.1.1 Default Values for Wet to-Dry-Weight Conversion Factors for Vegetables, Fruits, and Grains from NUREG/CR 5512, Volume 1 g Plant type Conversion factor g (kg dry-weight /kg wet-weight) Vegetables, leafy 0.2 Vegetables, other 0.25 Fruit 0.18 Grain 0.91 5.9.1.1 Use of parameter in modeling: The wet to-dry-weight conversion factors convert the weight of the garden produce et harvest to the corresponding or equivalent dry weight. Thes7 factors are required in two pathways: 1) soil-plant-numan pathway to calculate the concentration factor for radionuclide j in plant v at harvest from an initial unit concentration of parent radionuclide i in soil, C ,3n, and 2) irrigation water soil-plant-human pathway to calculate the concentration factor for radionuclide j in plant v at time of harvest resulting from resuspension and root uptake for an average unit concentration of parent radionuclide i in water, C,,,n. C.,,n is calculated from the following equation (Equation 5.5, page 5.12 of NUREG/CR-5512, Volume 1): C.,3n = 1000 (ML, + By ) W, A{C,,.t,,}/C,(0) (5.9.1.1) Residential Scenario 5.9-1 January 31,1998
where ML,is the plant soil mass-loading factor for resuspension of soil to plant type v By is the concentration factor for uptake of radionuclide j from the soil in plant v, W,is the wet to-dry-weight conversion factor for plant v. A{C.,,to ,)is the decay operator notation used to develop the concentration of radionuclide j in soil at the end of the crop-growing period, t,,is the growing period for food crop v, and C.,(0) is the initial concentration of parent radionuclide i in soil. C,,,n, is caiculated from the following equation (Equation 5.31, page 5.31 of NUREG/CR-5512, Volume 1): Cyn = (ML, + yB ) W, C,ym (5.9.1.2) where ML, is the plant mass-loading factor for resuspension of soil to edible plant parts for plant v, Byis the concentation factor for uptake of radionuclide j from soilin plant v, and C,y,is the concentration factor for radionuclide j in soil at harvest time for plant v for an average unit concentration of parent radionuclide iin water. 5.9.1.2 Review of additional information to define the distribution for W, The Human Nutrition and Information Service of the w3DA compiled information un the nutritive value of over G00 foods food products, and beverages [Gebhardt,1985]. The data included water contents of vegetables, fruits, and grains, which are summarized in Table 5.9.1.2. The wet-to-dry-weight conversion factor is calculated from the fo!!owing equation: W, = (100 - % water)/100 (5.9.1.3) Table 5.9.1.2 Moisture Content of Farm and Garden Produce [Gebhardt,1985] Garden Produce Water (% by wt.) l Vegetables, leafy Lettuce 96 Broccoli 91 Cauliflower 92 Celery 95 Parsley 88 Spinach 92 Cabbage 92 Vegetables, other Carrots 68 Radishm 95 Potatoes 77 Tomatoes 94 Peppers 93 ResidentialScenario 5.9-2 January 31,1998
Fruit O Apples 84
) Apricots 86 Blueberries 85 Cherries 90 Grapefruit 91 Grapes -81 l Cantaloupe 90 Oranges 87 Peaches 88 Pears 84 Plums 85 Strawberries 92 Watermelon 92-Grain Wheat 12 Corn 12 Barley 11 Rice 12 S.9.1.3 Proposed distributions for wet-to-dry-weight conversion factors for vegetables, ,
n fruit, and grain The moisture content varies from 77 to 96% in vegetables and fruits and from 88 to 89% in grains; Because of the similarity in the moisture content _in vegetables and fruits, the distribution for W, (vegetables) and W, (fruits) were combined. The frequency distribution and corresponding PDF_(Figure 5.9.1.1) for W, (vegetables & fruits) and curnulative distribution (Figure 5.9.1.2) were deterrained from data in Table 5.9.1.2. The PDF is defined by a gamma
- function with a mean of 0.1088 and lower and upper limits of 0.04 and 0.23. The calculated parameters for the gamma distribution are shown in Table 5.9.1.3. Since W,(grains) varies only slightly, a fixed value of 0.88 is recommended.'
Table 5.9.1.1 Distribution Parameters for Wet to-Dry-Weight Conversion Factor for Vegetables and Fruits Parameter Value K- 2.6803735 A 35.083423 c 0.0324 (~ Residential Scenario 5.9-3 January 31,1998 t N l l l
5.9.2 W:t t:-dry-weight c:nycrci:n fact:rs f:r f rago c:nsum;d by Se:f cattle, pruttry, milk cows, and layer hens, We The wet-to-dry-weight conversion factors for forage, W,, as defined for NUREG/CR-5512, Volume 1 dose modeling, estimates the fraction of dry matter in forage consumed by beef cattle, poultry, milk cows, and layer hens. The model uses a single, constant value for W,for all contaminarts. Thus, this value represents the average concentration of dry matter in all forage crops consumed by livestock in the residential scenario. The value of 0.22 for W, was adopted as the default value in NUREG/CR-5512, Volume 1, and is based on recommendations by Till (1983). 5.9.2.1 Use of p ameter in modeling: The wet to dry-weight conversion factor converts the weight of forage to the corresponding weight of dry matter. This factor is required in the soil-forage faed-animal-human pathway for calculating 1) the concentration factor for radionuclide j in fresh forage crop i at the time, t, from in initial unit concentration of parent radionuclide i in soil, C,3,(Equation 5.13, page 5.19 of NUREG/CR-5512, Volume 1), 2) the average concentration factor for radionuclide j in fresh forage crop f over the feeding period at the time of animal consumption of forage from an initial unit concentratien of parent radionuclide i in soil, C,3e (Equation 5.15, page 5.21 of NUREG/CR-5512, Volume , and 3) the average concentration facter for animal produc; a over the fresh forae i feeding period for soil ingestion by animals for radionuclide j for initial unit concentration of p ent radionuclide in toil, C og (Equation 5.19, page 5.22 of NUREG/CR-5512, Volume 1) according to the foilowing equations: C., = 1000 (ML, + Bg ) W, A(Cy ,t}/C (0) (5.9.2.1) where ML,is the plant soil mass-loading factor for resuspension of soil onto forage plant f, Bf is the concentra' ion factor for uptake of radionuclide j from the soil in fresh forage crop f, W,is the dry to wet-weight conversion factor for fresh forage, A{Cy,t} is the decay operator notation used to develop the concentration of radionuc:ide J in soil at time t durirg the feeding period for fresh forage crop f, and C.,(0) is the initial concentration of parent radionuclide i in soil at the start of the growing period; C,ge = 1000 (ML, + Bf ) W, S{Cy ,t,}/[t, C ,(0)] (5.9.2.2) . where S{C y,tn} is a concentratiori time-integral factor for radionuclide j in soil over the feeding period for crop forage, tn; and Cog = 1000 F, Q, W, Q, x, S{C y,tn}/[t, C.,(0)) (5.9.2.3) where Qa is the soilintake as a fraction of forage intake for the animal. 5.9.2.2 Review of AdditionalInformation to Define the Distribution for W, The National Research Council (NRC) published detailed information on nutrients in forage, hay and grain crops for livestock. Since livestock feed intake is based on dry-matter intake, and the conesponding nutrient content in dry matter, the NRC data included moisture content. Table 5 9.2.1 lists common types of grasses and the fraction of dry matter (NRC.1996). Residential scenario 5.9-4 January 31,1998
Tcbi) 5.9.2.1 Maisture Cent:nt in Foraga Creps (NRC,1996)
./~N Hay Crop Dry Matter (kg dry-weight /kg wet weight)
Alfalfa 0.234 Bermuda grass 0.303 Bluegrass 0.308 Broome grass 0.261 Canary grass 0.228 Clover, Ladino 0.193 Clover, Red 'O.262 Fescue 0.313 < Orchard grass 0.235 l Rye grass 0.226 Trefoil 0.193 Timothy 0.267 5.9.2.3 Proposed distributions for wet to-dry-weight conversion factors for forage The uncertainty and variability of W,were determined from the average dry matter content over the twelve hay crops in Table 5.9.2.1. Since the type of forage crop consumed by livestock is uncertain, each of the crops was weighted equally. The distribution for the wet-to-dry we;ght conversion factor was determined by fitting a beta functicn to the observed frequency
.( distribution for dry matter in Table 5.9.2.1. The distribution parameters for the beta distribution
(' are shown in Table 5.9.2.1.- The frequency distribution and corresponding PDF are shown in Figure 5.9.2.1. The PDF has a mean of 0.2519 and lower and upper limits of 0.193 and 0.313. } The cumulative distribution for W,is shown in Figure 5.9.2.2. Table 5.9.2.1 Distribution Parameters for Wet to-Dry Weight Conversion Fac+9r for Forage Parameter Value - a, 1,154211 a2 1.1796847 6, 0.183 62 0.323 Residential Scenario 5.9-5 January 31,1998
6.9.3 Wst to dry wsight conversion factors for stored grain coneumed by beef cattle, poultry, milk cows, and layer hens, W, The wet to dry weight conversion factor, W,, correspond to the fraction of dry matter in stored grains. The quantity of moisture in grain varies with thts type of grain and physical conditions under which the grain is store 1(e.g., dew point). The default value for this parameter as defined in NUREG/CR 5512, Volume 1, is 0.91 [Till,1983). 5.9.3.1 USE OF PARAMETER IN MODELING: The wet to dry weight convercion factor converts the weight of the as stored grain to a coiresponding weight of dry matter. This factor is required in the soil-stored grain animal-human pathway to determine the quantity of contaminated Drain consumed by livestock and is used in the cr.iculation of the concentration factor for radionuclide j ;n stored grain crop g at the time of initial feeding to animals from an initial unit concentration of parent radionuclide iin soil, C.,, as shown ,n the following equation (Equation 512, paqe 5.18 of NUREG/CR 5512, Volume 1): C., = 1000 (ML, + B,,) W, A{C.,,t,,}/C.,(0) '5.9.3.1) where ML,is the plant soil mass lo. . sing factor for resuspension of soil onto grain plant g, B,, is I the concentration factor for uptake of radionuclide j from the soil into stored grain vrop g, W,is l the wet to dry weight conversion factor for stored grain crop g, A{C.,,t,)is the decay operator l notation used to devlop the concentration of ' .dionuclide j in soil at the end of the crop-growing season,t,c c the growing period for stored grain crop g, and C.,(0) is the initial concentration of parent radionuclide iin soil at the start of the growing period 5.9.3.2 Review of AdditionalInformation to Define the Distribution for W, Grain crops provide the major dietary needs for poultry and layer hens and supplement of diets of ruminbnt animals in agricultural operations. The dry matter content of common grain crops for livestock consumption were taken from data compiled by the National Research Council [NRC,1996) and are shown in Table 5.9.3.1. Table 5.9.3.1 Molsture Content in Stored Grain [NRC,1996) Grain Crop D.y Matter (kg dry weight /kg wet weight) Barley 0,881 Canob 0.922 Corn 0.000 Oats 0.892 Sorghum 0.900 Wheat 0.902 Residential Scenario 5.9-6 January 31,1998 B I
.- .-- . - . . .-_ _ _ - _ - . _ - . - - _ - - _ _ - - - - - . _ - . ~
5.9.3.3 Pr: posed distributi:n f:r wetit: dry weight c:nversi:n fact:rs f:r st r:d grain The distnbution for the wet to-dry weight conversion factor was determined by fitting a log Q normal functim to the observed frequency distrioution for dry matter in Table 5.9.3.1. The distribution parameters for the log normal distribution are shown in Table 5.9.3.1. The frequency distribution and corresponding PDF are shotyn in Figure 5.9.3.1. The PDF has a mean of 0.8995 and lower and upper limits of 0.881 and 0.922. The cumulative distribution for W,is shown in Figure 5.9.3.2. Table 5.9.3.1 Distribution Parameters for Wet to Dry Weight Conversion Factor for Stored Grain Parameter Value p 0.022356333 0 0.50002201 i c 0,87416667 5.9.4 Wet to-dry weight conversion factors for stored hay consumed by beef cattle, poultry, milk cows, and layer hens, Wn The wet-to-dry weight conversion factor for stored hay consurr . i by beef cattle, poultry, milk , O,s cows, and layer hens, as defined for NUREG/CR 5512, Volume 1, dose modeling, converts the weight of the as-cut plant to a corresponding dry weight, and the factor is a measure of the dry matter content in hay crops. The model uses a single, constant value for all stored hay crops. The value of 0.22 for Wn was adopted as the default value in NUREG/CR 5512, Volume 1, based on studies by [Till,1983). 5.9.4.1 use of parameter in modeling: The wet to drv. weight conversion factor converts the
.veight of the as stored t ay to a corresponding weight of dry matter. This factor is required iri the soil stored hay animal human pathway to determine the quantity of contaminated hay consumed by livestock. Wn is applied m the calculation of the concentration factor for radionuclide j in stored hay h at the time of initial feeding to animals from an initial unit concentration of parent radionuclide iin soil, C,ng, accon ing to the following equation (Equation 5 li, page 5.18 of NUREG/CR 5512, Volume 1):
C,ny = 1000 (MLn + B,n) Wn A{Cy ,1,n}/C,(0) (59.4.1) where MLn is the plant svil mass-loading factor for resuspension of soil onto hay plant li, B,n is the concentration factor f uptake of radionuclide j from the soil into stored hay crop h, Wn is the wet to dry weight conversion facar for stored hay crop h, A{C,j,t,n)is tne decay operator i O Residential Scenario 5.97 January 31,1998
not: tion used to dev;l p the concentration of radionuclide j in soil at th] end of th3 crop-growing season, tpis the growing period for stored hay crop h, and C,(0) is the initial concentration of parent radionuclide iin soil at the start of the growing period. 5.9.4.2 Review of AdditionalInformation to Define the Distribution for Wn Hay crops provide the major dietary needs for ruminant animals in agricultural operations. These hay crops are identical to the forage crops listed in Table 5.9.2.1 except in the manner in which the crops are harvest, stored, and subsequently fed to livestock. Since the wet to dry-weight conversion factor is equal to the dry matter content of the hay crop, Wn and W, are equal. 5.9.5 Parameter uncertainty: The distributions fur wet-to-dry weight conversion factors are established based on the average moisture content in a wide range of garden produce and forage, grain, and grain crops. Among the factors that affect the moisture content are the type of crop and environmon'al conditions under which the crops are grown (e.g temperature, humidity, length of growing season). 5.9.6 Variabillay across sites: This parameter wm =. ely vary from site to site depending on the local growing conditions (i.e., some crops may not be suitable for growing because of soil and weather). 5.9.7
References:
National Research Council,1996.
- Predicting Fead Intake of Food Producing Animals",
Nutrient Requirements of Beef Cattle Seventh Revised Edition, National Academy Press Till, J. E. and H. R. Meyer, eds.,1983. Radiological Assessment. NUREG/CR 3332, ORNL-5968, U.S. Nuclear Regulatory Commission, Washington, D.C. Gebhardi, S. E. and R. Matthews,1985.
- Nutritive Value of the Edible Part of Foods", U.S.
Department of Agriculture, Human Nutrition and Information Service, Home and Garden Bulletin No.72 Residential Scenario ' 5.9-8 .anuary 31,1998
r I l if j E g E 17 a 8-Probability Density 12.5 Legend
' ^ ' '
10 0 MOM Ga nma j 7.5 -- l f(X) g 50 - 4 e , 25 - 00 00 0 05 01 0 :5 02 0 25 03 WV (vegetables and fruits) i g ! $ Figure 3.63.1 Frequency Distribution and Proposed PDF for W. 1 a i M i
<=
I
- n a
f a
~_
W a 8-Cumulative Density 10
. p " ---
Legend
/ I I Data 08 f --- MOM Gamma
/
/
/
06 / F(x) /
/ ,
04 / us
# l 8 l f
jV
/
00 00 0 95 01 0 15 02 0 25 03 , WV (vegetables and fruits) LT E a 4 Figure 3.63.3 Proposed Cumulative Distribution for Wet-to-Dry-Weight Conversion Factor, W. 2 8 O O O
1 ' i s x
! 2 I a E
i j E B i 8-a Probability Density 3 l 12 5 ' l
; Legend i
1 10 0 M Data MtM Beta 1 i 7.5 l i f(X) e 50 -
?
25 l I OO O175 02 0225 0 25 0275 03 0 325 l WI(forage) 1 I t l Figure 3.64.1 Frequency Di::tribution and PDF for Wet-to-Dry-Weight Conversion ' g Factor for Forage Consumed by Live. sock E
- =
. Q 1
.y i
e i 1
lil ll1llllll !!ll l t a a c O d n aa t e u t gD A el - L - l - 5 2 3 _ 0 3 0 l 5 7 2 y 0 t i s n e / ) e D g e a i t v /V 5 2 0 f( r o Wr e l a u m u s
-'s f
W f it b o n o u O . C 5 i r
/ 2 2
t s 0 iD
/' i t
e v a
, tu m
u
/ 2 0
C d
# e s
/ o p
o r P 2 5 4 7 6 1 0 8 6 a 2 3 1 0 0 o 0 00 0 e r
) u x ig
( F F F . d'( f'Sae $=8 wYG kEO 3L m 6 1l1:
l l .l i\1ll I\lllllll I s w r d o n e t, ga r. e a o L o r. t M 5 0 9 0
- 4 t
h i g e 9 0 W-y r D-t o t 3 9 e y 0 Wr o t ) i n f n s Fi a n 2 i a Dr PG e lI 9 r g D 0 dd ee y d s r t e oo t r pS i l i b - 1 t o o-P e r a 9 s( f 0 d r b g no o at c r W na P oF u n it a 9 o 0 ibir s r t iv s e Dn - yo 9 cC 8 n 0 e u q 0 0
- 0
/
0 0 0U 8 E F 1 6 6 e r 0 0 0 0 3 5 ' 3 2 1 e r u g i F w x3WBE E9a vw w W a I~ Mi a =
, {
l i 3' E a E W 8 g. Cumulative Density t,g,,, 08 4
--- Mou tog norma
/
06
/
F(x) f u, ! p 04 #
$ l
/
f 02 /
/l 00 0 88 0 89 09 0 91 0 92 0 93 0 94 0 95 Wg (stored grain)
I t Figure 3.66.2 Proposed Cumulative Distribution for We E E 2 ! .u M l
= ;
t O O O !
5,10 R:di:nuclide Partiti:n Coeffici:nts, KD,,,, The radionuclide partition coefficients define the ratio between radionuclide solid concentrations (radionuclide quantity adsorbed on the soil / rock particles) and radionuclide liquid concentrations (radionuclide quantity dissolved in the soil / rock pore water) under equilibrium conditions and are expressed in volume per mass units (DandD units are mL/g) . Use of Parameter in Modeling Partition coefficients for the i* radionuclide are used to calculate radionuc ide retardation in the soil layer (Rt,,) and unsaturated zone (Rt,,) as follows (Volume 1, p.49, equation 4.9 and 4.12): Rt i, = 1 + Kd,,* p,/ n, (5.10.1) Rt,, = 1 + Kd,,* p,/ n, (5.10.2) In Volume 1 of NUREG/CR 5512, [ Kennedy and Strenge,1992]it is assumed that partition coefficients for the i* radionuclide in the unsaturated layer (Kd,,) are the same as partition coefficients of the soillayer (Kd i ,); bulk density of the soillayer (pi) is thr, same as the k density vf the unsaturated layer (p,): and total porosity of the soillayet (r,i) is the same as total porosity of the unsaturated layer (n,). These assumptions lead to an assumption that radionuclide retardation in soil layer is the same as in the unsaturated layer (Rtii = Rta,). The retardation coefficients define the radionuclide transport velocities within the soil layer (vi,) and within the unsaturated layer (ve ,) as follows (vo is introduced for the sake of simplicity and is l not a Volume 1 notation): vi, = II (R,,
- 0,) (5.10.3) v,, = 1/ (R,,
- 02 ) (5.10.4) where I is infiltration rate and 0, and 0, are volumetric water contents of the soillayer and unsaturated zone respectively, in Volume 1, fully saturated conditions were assumed for both the soil and unsaturated layers; therefore,0, = 0, = 1. As a result, the transport velocities of a specific radionuclide were the same within the soil layer and unsaturated layairs. In this analysis, the volumetric water content is also assumed to be the same for the soil and unsaturated layers, but instead of using constant value of one (1.0), a probability distnbution function is defined for this parameter to be less than or equal to one (1.0).
The differences in the transport velocities of the different elements is due soley to the differences in partition coefficients. The transport velocities determine the radionuclide leaching rates from the soillayer (L r,) i and from the unsaturated layer (Ln.) which, in turn, are the parameters of the system of the ordinary differential equations that describes the time. dependent distribution of mass among the soil layer, unsaturated layer, and aquifer layer, importance to Dose Partition coefficients can noticeably affect doses because they may significantly influence the
._ mass transfer rates between the soil, unsaturated zone, and the aquifer and, consequently, the radionuclide concentrations in soil, drinking water consumed by the humans, water consumed Residential Scenario 5.10 1 January 28,1998
by animals, wat:r us:d for irrigation, and wattr in the surface pond. This aff; cts th] tim > dependent distribution of the contaminant mass among all the contaminant pathways includec: ( in residential scenario (partial pathway transfer factors, PPTFs, in Volume 1 terminology) and. as a result, the pathway doses and the total effective dose equivalent (TEDE). The influence of the partition coefficient on the total dose should be greater in the case when the leaching rates L,,, and L 2> are comparable to or greater than the radioactive ' decay constant. Default Parameter Values from NUREGICR 6512 Volume 1 The partition coefficient values in Volume 1 are listed in Table 5.10.1. Of the total (73 elements) four elements in this table (H Kr, Xe, and Rn) have partition coeffictents equal to zero, since they only are transported in gaseous phase. This leaves 69 elements of interest with respect to the partition coefficient. The partition coefficient values for the remaining 69 elements reprecent either the minimum values (the most mobile conditions) of the experimentally derived values provided in Sheppard and Thibault (1990) and Sheppard, Sheppard, and Amiro [1991)(25 partition coefficients), or values estimated from soil to-plant concentration ratios (43 partition coefficients) using the following formula: In(Kdo)=2.11-0.56tn(B,/4) (5.10,5) where B,,is concentration ratio for vegetative parts of the plant v (dry weight basis) for the i$ radionuclide,4 is a dry weight to wet weight conversion factor, and 2.11 and 0.56 are correlation coefficients proposed by Thibault, Sheppard, and Smith [1990] for sandy soil.These carrelation coefficients are representative of sandy soil to provide lower values for the estimated partition coefficients. B,, values were based on concentration ratios for leafy vegetables from the following sources: International Union of Radioecologists [lUR,1989), [Baes et al.,1984), and (Strenge, Banderf, and Soldat,1987) The concentration ratio based estimates of the partition coefficient were used in the absence of experimental data. , Table 5.10,1, Default Values of the Radionuclide Partition Coefficients in mUg from NUREGICR 5512, Volume 1, (Table 6,7 in Volume 1, p. 6,18), Element Partition Basis
- Element Partition Basis
- Coefficient Coefficient H 0.0E+0 M Sb 4.5E+ 1 E Be 2.4E+2 R Te 1.4E+2 R C G.7E+0 C I 1.0E+0 E F 8.i E+ 1 R Xe 0.0E+0 M Na 7.6E+1 R Cs 2.7E+2 E P 8.9E+0 R Ba 5.2E+1 R S 1.4E+ 1 R La 1.2E+3 R Residential Scenaf.o 5.10 2 January 28,1998
Table 6.10.1. Default Values of the Radionuclide Partitlod Coefficients in mUg f rom NUREGICR 6612, Volume 1, (Table 6.7 in Volume 1, p. 6.18). Element Partition Basis
- Element Partition Basis
- Coefficient Coefficient Cl 1.7E+0 R Ce 5.0E+2 E K 1.8E+1 R Pr 2.4E+2 R Ca 8.9E+0 R Nd 2.4E4 2 R Sc 3.1 E+2 R Pm 2.4E+2 R Cr 3.0E+1 E Sm 2.4E+2 R Mn 5.0E+1 E Eu 2.4E+2 R Fe 1.6E+2 E Gd 2.4E+2 R Co 6.0E+1 E Tb 2.4E+2 R Ni 4.0E+2 E Ho 2.4E+2 R Cu 3.0E+1 R W 1.0E+2 R Zn 2.0E+2 E Re 1.4E+1 R l As 1.1 E+2 R Os 1.9E+2 R Se 1.4E+2 R Ir 9.1 E+1 R Br 1.4E+1 R Au 3.0E+1 R Kr 0.0E+0 M Hg 1.9E+1 R Rb 5.2E+ 1 R Tl 3.9E+2 -R Sr 1.5E+ 1 E Pb 2.7E+2 E Y 1.9E+2 R Bi 1.2E+2 R Zr 5.8E+2 R Po 1.5E+2 E Nb 1.6E+2 R Rn 0.0F+0 M Mo - 1.0E+ 1 E Ra 5.0E+2 E Tc 1.0E-1 E Ac 4.2E+2 R Ru 5.5E+1 E Th 3.2E+3 E Rh 5.2E+1 R Pa 5.1 Et2 R
=_
Pd 5.2E+1 R U 1.5E+1 E Residential Scenario 5.10 3 January 28,1998
.LJ us
Tau 10,1. Default Values of the Radionuclide Partition Coefficients in ml.Jg from NLIREG/CR 5512, Volume 1,(Table 6,7 in Volume 1, p,6.18). Element Partition Basis
- Element Partition Basis
- Coefficient Coefficient Ag 9.0E + 1 E Np 5.0E+0 E Cd 4.0E+ 1 E Pu 5.5E+2 E In 3.9E+2 R Am 1.9E+3 E Sn 1.3E+2 R Cm 4.0E+3 E Cf 51E+2 R
- Values for partition coefficients are based on: M - Assumed to be mobile; R - Calculated from concentratiun ratios ; C - Experimental data from Shepoard, Sheppard, and Amiro [1991);
or E - Experimental data from Sheppard and Thibault I1900). Additional Data Reviewed to Develop PDFs for Partition Coefficients Data to support the development of pdfs describing the variability in partition coefficient values are based on the following: Individual measurements of partition r efficients obtained from experiments ara p eferable to mean or best-estimate values, a Variability based on experimental measurements [Thibault et al.,1990; Sheppard and Thibault,1990) represents small scale spatial variability and may not sufficiently describe the variability in effective Kd values ovar a large soil volume. Given the potentia! scale-dependant variabiltiy, best estimates of emall-scale Kd values derived from Thibault et al. [1990) should be compared to the best estimates of the large scale Kd values. Estimates oflarge-scale Kd values are available from McKinkley and Schottis [1991). McKinkley and Scholtis [1991] presented a summary of Kd databases used in repository performance assessment. These date do no' provide information on ranges, number of samples, and other statistics and, thus, cannot be used for developing empirical distributions; however, they provide best estimater values that can be evaluated against smaller scale best estimates to guage the scale effects. A significant source of information on partition coefficient values (approximately 11,000 experiment-based partition coefficient values), that was not used in the Volume 1 analysis, is the Nuclear Energy Agency (NEA) data base [NEA,1989). This and other (published after 1992) sources of data should be reviewed. If there are additional data, these data should be used in developing pdfs to describe the variability in the partition , coerficients. A large number of experimental data on partition coefficients is availaty.e from NEA sorption data base (SDB) [NEA,1989). The SDB incorporates the information previously contained in the International Sorption Informatinn Retrieval System (ISIRS) and additional data compiled by NEA. The data base contains approximately 11,000 values of partition coefficient for different Residential Scenario 5.10 4 January 28,1998
clem:nts. Most of th3 data cro from static batch sorption experim:nts, some are from column (dynamic) experiments, and a few data are from retardation (dynamic) stu' a. When available, the data base provides information on the reference source, mw used, solution 1 phase, initial contaminant concentration, type of solid material used, reducingloxidizing conditions, experiment duration, and a number of other details. The SDB data base was searched to extract all the data available for the 69 elements of interest from experiments using unconsolidated and consolidated deposits. The unconsolidated deposits are described in SDB in general terms such as: clay, fine sand, sand, soil, and loam. This differs from the classification used in Shoppard and Thibault [1990), where four different types of soils are specified based on the particle size distribution and organic material quantity. Additional data are provided for consolidated deposits, including dolomite, gypsum, sandstone, shale, limestone, rock of unspecified mineral composition and sediment. Data from SDB for unconsolidated and consolidated deposits were obtained for the following 19 radionuclides: C, Mn, Co, Ni, Zn, Sr, Y, Tc, Pd, Ag, l, Cs, Ce, Eu, Ra, U, Np, Pu, and Am. Note, that the experimental data for Pd and Y that are available from SDB are not available from Thibault et al. [1990) or Sheppard ard Thibault [1990]. Data in the SDB ~e combined with data from Thibault et al. [1990) for this antlysis. Data Analysis The primary goals of the data analysis conducted here were: l to determi,$e if there is a strong correlation between the composition of the unconsolidated deposits and their ability to sorb different radiontNides; l to develop radionuclide partition coefficient probability distributions that provide the best ( fitting to all experimental data available for unconsolidated deposits; and, to provide more justification where possible for the radionuclide partition coefficient probability distributions that do not have individual measurement data.
.Qgrrelation between Partition Coefficie,st Values and ComDosition of the Unconsolidated Deoosits Thibault et al. [1991) provided data on partition coefficient values along with the information on the corresponding composition of the unconsolidated deposits used in the experiment. The data on deposit composition are expressed as percentage of clay particles, sitt particles, sand particles, and organic material of the sample. These data were used to generate scatter plots of Kd versus composition (expressed in percent composition), and the degree of correlation was analyzed, quantitatively and qualitatively. When available, the partition coefficients were plotted against the percent of clay, silt, sand, and organic material. Table 5.10.2 provides information on radionuclide analyzed and, qualitatively, the correlation observed. An example showing a lack of correlation of partition coefficients with the percentage of the particles of differerit sizes for Po is shown in Figure 5.10.1.
n N ResidentialScenario 5.10 5 January 28,1998
~ u
Tabl; 5.10.2. Currelati.n between Partiti.n Coefficient Valu.s and Composition of the Unconsolldated Deposits Description of Correlation Element % Clay % Salt % Sand % Organic l indistinguishable indistinguishable weak insignificant Pb indistinguishable NA NA indis:inguishaole _ Ru NA NA NA weak Ni weak indistinguishable weak indistinguishable Fe indistinguishable weak weak weak Po indistingu.shable indistinguishable Indistinguishable indistinguishable U indistinguishable indistinguishable indistinguishable NA Tc indistinguishable indistinguishable indistinguishable weak Co indistinguishable indistinguirheible indistinguishable anoistinguishable Sr indistinguishable indisting yt.. ale weak indistinguishable Cd indistinguishable indistinguishable indistinguishable indistinguishable Cs indistinguishable weak weak indistinguishable Ra indistinguishable indistinguishable indistinguishable weak Mn indistinguishable ind.:nguishable indistinguishable indistinguishable Np indistinguishable indistinguishabie indist...guishable indistinguisEJle Se weak NA NA indistinguishable Th indistinguishable NA NA NA Zn indistinguishable indistinguishable indistinguishable indistinguishable Cm indistinguishable indistinguishable indistinguishable indistinguishable Cr NA NA NA indistinguishable Ce weak indistinouishable indistinouishable indistinouishable As it can be seen from Table 5.10.2, most of the radionuclides analyzed show an absence of correlation with the percentage of the panicles of different sizes: 16 of the 19 radionuclides show no correlation to percentage of clay; 14 of the 16 radionuclides show no correlation to percentage of silt; 11 of the 16 radionuclides show no correlation to percentage of sand; and 15 of the 10 radionuclides show no correlation to percentage of organic material. Some of the radionuclide partition coefficient values show weak correlation; however, this does not provide a basis to justify any functional relationship. The data from NEA [1989) combined with the data from Thibault et al. (1990) were used to analyze correlation between the radionuclide partition coefficient values and composition of deposits. The partition coefficient values of a few radionuclides were plotted for the different unconsolidated deposit types (clay, sand, and loam) and for the different consolidated deposit types (g/psum, dolcmite, sandstone, limestone, and shale). There was no discernible correlation or trends for the partition coefficient values across different types of unconsolidated deposits for Pu , Am, and Se. Pu ano Am exhibited similar partition coefficients between unconsolidated and consolidated deposits. The partition coefficients typical of unconsolidated ' deposits for Se were significantly lower than the partition coefficients in consolidated deposits. Based on this analysis, it was concluded that no reliable correlations could be developed for the _ radionuclides of interest. The absenets of a distinguishable correlation between the composition of the unconsolidated deposits and their partition coefficients allow us to develop one probability Residential Scenario 5.10 4 January 28,1998 3
distnbution function for each cism nt bas d on all data availabla, rather than using separate probability distributions for each element and soil tyoe. y EDLtition Coefficient Probability Distributions Based on boetimental Data for Unconsolidated Deposits Experimental data on partition coefficients for unconsolidated deposits are available for 34 of the 69 elements of interest. The experimental data from Thibault et al. [1990) were used to develop probability distributions for 15 radionuclides. The experimental data from NEA SDB [1989) were used for 2 radionuclieds. The experimental data from Thibault et al. were combined with the experimental data from the NEA SDB to deveiop probability distributions fo-the 17 remaining elements. Information on data sources and number of samples available for each element is provided in Table 5.10.3. The computer code C FIT (Center for Engineering Research Inc.,1996) was used to develop radionuclide probability distribution functions based on the experimental data. C FIT provides three different optimization techniques (method of moments, maximum likelihH method, and
. ast squares method) to fit experimental data into16 different powible probability distribution functions. The decision on which distribution provides the best fit can be made either visually based on the companson of the experimental data histogram and different probability distribution functions and/or btsed on the results of the goodness of fit tests. Two test are 3 available with the softwsi,e: chi-square test anc Kolmogorov Smirnov test. Both tests calculate g significance levels corresponding to the hypothesis that experimental data are sampled from a L specified distribution. The higher the significance level, the higher the probability that the experimental data are from this distribution.
The analysis of data for each of 34 radionuclides consisted of plotting the histograms of partition coefficients and logarithms of partition coefficients, and comparing tnem with the different theoretical distributions. In most of the cases developing distributions for partition co9fficients using C FIT was not succcssfulin that the significance levels from both statistical tests were very low. This is due in part to the variability in the partition coefficient values over mar y orders of magnitude. To reduce the spread, distributions were fit the log transformed partition coefficient data. Using log transformed data allowed development of histograms with smallor ranges and distributions with higher signihance levels. All three optimization methods were used to search for the best fit. Both statistical goodness-of fit tests were performed for each run. However, it was found that chi square test does not calculate an acceptable significance level even in the cases where the experimental data appear to be in good agreement with the theoretical distribution. Conversely, the Kolmogorov-Smirnov test results were in good agreement with visual analysis of the results. The results of the Kolmogorov Smirnov test were used to evaluate the goodness of fit The summary of the analysis is presented in Table 5.10.3. This table provides information on type of distnbu* ion obtained, parameters that characterize the distribution, the fitting method , that provided the h5 hest significanca level, and the significance level from the Kolmogorov-Smirnov test. In addition to this information Table 5.10.3 provides the corresponding default values from NUREGICR-5512 Volume 1 and the best estimatet of the partition coefficients ! Residential Scenario 5.10 7 January 28,1998
(log nthmically conv;rt;d) from th) r;pository performanca ass;ssment studi;s comphed in McKinkley [1991) obtained for soil and surface deposits. Seven of the 34 elements analyzed (Y, Ba, Eu, Cu, Ca, As, and Sb) did not have enough data (fifteen or fewer samples) to develop distributions fit to the data. The uncertainty in these Kd values is represented using normal distributions with mean values based on the mean of the experimental data and a standard deviation based on the larger of the standard deviation in the data for that element or the standard deviation in the d6ta for all elements For 21 of the 34 biements, the logarithms of the partition coefficients fit a normal distribution. An example of the actual data and C-FIT calculations is shown on Figure 5.10.2 for Po. The mean values of these distributions vary from 0.66 (Kd= 4.6 mL/g) for I to 3.83 ( Kd= 6761 mUg) for Cm with an average value of 2.37 (Kd= 234.4 mUg). The mean standard deviation of the normal distributions obtained is 1.09. However, some ! distributions have much lower standard deviations (standard deviation for Se is 0.25) and some distributions have much higher standard deviations (standard deviation for Zn is 1.93) m J mean %.Jes for Pd, Tc, and Se lay outside of the range of the best estimmed values provided in McKinkley [1991). In the cases of Pd and Se this may be related to the sma!! size of the populations considered (nine samples for Pd and 22 samples for Se). In the case of Tc, the size of the population appears to be representative (206 samples) and the observed difference may be related to experiment scale. As it was noticed above, the McKinkley [1991) data are from large scale observations as opposed to Thibault et al. [1990), where the data are from , sr'all scale experiments. The other radionuclide mean values (I, Sr, Cs, U, Ni, Am, Pu, and l Th) are within the McKinkley [1991) range. l For three of the 34 elements (Mn, Ag arJ C), the logarithmic values of the partition coefficients demonstrated the best fit wi;h a log normal distribution. The log-normal distribution better describes the shift of the logarithms of the experimental data to the lower values. The mean values vary from 0.14 to 2.04. The standard deviation varies from 0.52 to 1.17. An example of the actual experimental data and C-FIT calculations for Mn is presented on Figure 5.10.3. The data from McKinkley [1991) are available only for C. The mean value obtained for C is within the best estimate range. For three other elements (Co, Fe, and Np), the logarithmic values of the partition coefficients demonstrated the best fit with the Gumbel distribution (Gumbel minimum for Co and Fe and Gumbel maximum fc ? Np). The Gumbel distribution better describes the shift of the logarithms of the experimental data to the higher values. In all cases the population sizes (292 samples I for Co,44 samples for Fe, and 262 samples for Np) appear to be representative enough to justify these values. The standard deviation varies from 0.52 to 1.17. An example of the actual experimental data and C-FIT calculations for Fe is precented on Figure 5.10.4. l l Eartition Coefficient Probability Distnbutions for Elements without da's l j The remaining 35 of the 69 elements of interest have no data on partition coefficient. In l Volume 1, partition coefficients for these and other elements were defined based on plant to-soil concentration ratio model. A different approach is taken in this analysis because of the ! Residential Scenano 5.10-8 January 28,1998
pot:ntial inconsist:nci:s anri un int:ntional sample corrstations in cstimating unc:rtrinty in partition coefficient values. These difficulties arise in estimating the partition coefficient based [' on plant uptake, because the conentration in plants is modeled as a function of the I concentration ratio and the total soil concentration (which is c function of the partition coefficient) . A major assumption of the approach taken in this analysis is that the variability, in the loganthms of the partition coefficients for elements without experimental data, wili b1 normally distributed. This is based on the cbservation that majority of the distributions fit to experimental data are normally distributed (see Table 5.10.3). In addition, we have assumed that the standard deviation of these normal distributions will be the same as the standard deviation derived from a distribution of allthe expeilmental observations in Table 5.10.-? . To obtain the mean standard deviation, all the experimental data available for all the radionuclides were combined and analyzed. The histogram of these data and probability function plots are presented on Figure 5.10.5. The resulting distribution is normal with the mean equal to 2.2 and the standard deviation equal to 1.4. An attempt was made to define mean v41ues for distributions based on review of additional literature. The additianal information was found in Thibault et u.' [1990) for; Be, P, Br, Te, sm, Ho, Re, Rb, Zr, Nb, Sn, Bi, Ac, and Pa. In Thibault et al. [1990), the mean values of the experimental data are presented for each these 14 radionuclidos for each of four types of soil (sand, clay, silt, and organic). Based on these deta, the mean values for all soils were assumed to be the mean partition coefficients of each soil type and were used as the mean of the corresponding normal distributions having standard deviation of 1.4 (Table 5.10.3). O Eight elements were assumed to behave similarly to iodine: K, Na, F, S Cl, La, Gd, and Tb [McKinley and Scholtis,1991). These elements are known to have low sorption capabilities, similar to I, and are assumed to have partition coefficients similar to iodine. The distribution of the variability in the log transformed partition coefficients for these elements is assumed to have the same mean ( 0.7) as I, but a higher standard deviation of 1.4 to account for the potential differences (Table 5.10.3). No additionalinformation was found for the partition coefficients of tne remaining 13 elements: Pm, Sc, Pr, Nd, W, Os, it, Au, Rh, in, Hg, TI, and Cf. The partition coefficier" probability distributions for these elements were assumed to have mean equal to 2.2 and standard deviation equal to 1.4, based on the mean and standard deviation of all experimental data (Table 5.10.3).
Table EW Proposed Radionuclide Fartition Coefficient Distributions, Logarithmic Values in mUg Element Data Number of Distnbutson Fattmg Significance Distnbution Parameters Volumet PA Study Source (*) Samples Type Method Levelr) mean l stand.dev. other variance Default Range (~) Sr 1. 2 539 normal LS 0 10 1.50 0 91 0 85 1.18 10 to 2 0 f 1. 2 109 norm 4 LS 0.37 0 66 0 95 0 9P O 00 .
- to 2.0 Cs 1. 2 564 normal MLM 0 06 2 65 1.01 1.02 2.43 2.0 to 4 0 Tc 1, 2 206 normal LS O 65 0 87 1.33 1.77 -1 00 -ce to 0 7 Ra 1. 2 53 normal MLM 0.52 3.55 0.74 0 55 2.70 U 1,2 60 normal MLM 0 64 2.10 1.36 1.85 1.18 1.3 to 3 2 NJ 1. 2 52 normal LS O 23 1.57 1.48 2.19 2.60 1.0 to 3 o Po 1 50 normal LS 0.97 226 0.73 0 53 2.18 Pb 1 18 normal MOM 0.96 3 38 1.20 1.44 2.43 Ru 1 47 W LS O 30 3 20 1.36 1.85 1.74 Cd 1 87 normal LS 0.22 1.53 1.30 1.69 1.60 Am 1. 2 219 normal LS 0.53 3.16 1.37 1.i;8 3.28 2 0 to 5 0 Pu 1. 2 205 no. mal MLM 0.75 2.98 0.82 0 67 2.74 2.5 to 5 0 Pd 2 9 normal LS O 92 227 1.37 1 88 1.72 0 6 to 2.0
~
Ce 1. 2 29 normal LS 0.55 1.93 0 43 0.18 2.m Mo 1 24 normal LS 1.00 1.42 0 75 0.56 1.00 Th 1 26 normal MLM 1.00 3.77 1.57 2.46 3.51 2.9 to 4 8 Cr 1 22 norma 6 LS 0.94 2.01 12J 1.44 1.48 Cm 1 23 normal LS 0.90 3 83 0.79 0 62 3 60 Zn 1, 2 98 normal MLM 0.18 3.03 1.93 3.72 2.30 Se 1 22 normal MOM 1.00 2.06 0.25 0.06 2.15 0 0 to 1.7 Y 2 15 normal 2.90 1.4 2.28 Mn 1. 2 127 log. normal MLM 0.50 1.15 0.70 1.70 Ag 1. 2 27 log-norme MOM 0.75 2.04 0.52 1.95 Eu 1, 2 14 normal 2.9ti 1.74 2.38 Ba 1 9 normal 1.65 3.53 1.72 ResidentialScenario 5.10-10 Janu: ry 28,1998 O O O
p_ /~\ p i Table 5.10.3 Proposed Radionuclide Parti *lon Coefficient Distributions, Loga.ithr.iic Values in mUg Element Data Number of Destnbution Fitting Segnsficance Destnbubon Parameters Volumet PA Study Source (*) Samples Type Method Level (") mean stand.dev. other ! variance Default Range (*") C 1. 2 66 log <a mat MLM 0 02 1.32 0.79 0 83 -~ to 2.0 Co 1, 2 292 Gumbei Min MOM O 59 3 00 1.18 1.78 j Fe 1 44 Gumbel Man MLM O 97 2.P5 1 65 221
#p 1, 2 262 Gumbet Max MLM 0 29 0 85 128 0.70 10 to 3 0 Cu 1 4 normai 225 1.40 1.48 !
Ca 1 4 normal 3 17 1.40 0.95 As 1 4 normat 2.06 1.40 1 04 Sb 1 4 non.W 224 1.40 1.65 Be 3 2 97 1.40 2.38 F 0.70 1.40 1.94 P 3 1 41 1.40 0.95 S 2.0G 1.40 1.15 Cf 0.70 1.40 023 -~to 2 0 Sc 220 1.40 2.49 go j 23 Br 3 1.75 1.40 1.15 To 3 1 74 1.40 2.15 -~to 12 La 0.70 1.40 3.08 Pr 220 1.40 2.38 Nd 2.20 1.40 2.38 Pm 3 r 3 70 1.40 2.38 3 to 4 Sm 3 2 97 1.40 2.38 0 to 3 7 Gd 0.70 1.40 2.38 -1.5 to 3 0 Tb 220 1.40 2.38 0 8 to 2.9 Ho 3 2.97 1.40 2.38 2.4 to 3 4 w 220 1.40 2.00 Re 3 1.64 1.40 1.15 Residential Scenario 5.10-11 January 28,1998
4 Table 5.10.3 Proposed Radionuclide Partition Coefficient Distributions, Logarit..mic Values in mUg Element Data Number of Distnbutson Frttmg Segneficance Distnbutson Parameturs Voesset PA Study l Source (*) Samples Type Method Level (**) mean stand.t= -- other variance - Default Range *"*) Os 220 1.40 228 fr 220 1.40 1.96 Au 220 1.40 1.48 Rb 3 2.31 1.40 1.72 -1 to 22 Zr 3 3.38 1.40 2.76 1.0 to 3.9 Nb 3 2.80 1.40 220 0 to 3.7 Rh ; 220 1.40 1.72 in 220 1.40 2.59 Sn 3 2.70 1.40 2.11 1.7 to 2.9 Hg 220 1.40 128 TF 220 1.40 2.59 Bf 3 2.65 1.40 2 08 12 to 22 Ac 3 324 1.40 2.62 1.0 to 3 7 Pa 3 3.31 1.40 2.71 Cf 220 1.40 2.71 Na 0 70 1.40 1.88 K 0.70 1.40 0.10 (*)- 1 = Thibault et at [1990]; 2 = Sorpton Data Base (SDB) NEA[1989). 3 = Sheppard and Thibault (1990] (") - Qnifkance level from Kolmogorov-Smimov goodness of fitness test (***) - best estimate value range from the reposstory performance assessment study McKinkley [1991) i ResidentialScenario 5.10-12 Januasy 28,1998 O O O
REFERENCES
/3 i i Baes, Ill, C.F..lli, R.D. Sharp, A.L. Sjoreen, and R.W. Shor,1984. A review and Analysis of Parameters for Assessing Transport of Environmentally Released radionuclirles Trough Agriculture. ORNL 5786, Oak Ridge Nationallaboratory, Oak Ridge, Tennessee.
4 Center for Engineering Research Inc., C FIT, Probability Distribution Fitting Software,1996. 1 Chow, Ven Te, editor : Handbook of Applied Hydrology, McGraw Hut,1964. i I international Union of Radioecologists (IUR).1989. Sixth Report of the Working Group on Soil-to-Plant Transfer Factors.RIVM, Bilthoven, The Netherlands. Kennedy, W.E., Jr. and D. L. Strenge: Residua / Radioactive Contamination From Decommissioning, NUREG/CR 5512, PNL 7994, Vol.1,1992. McKinley, l. G. and A. Scholo-* Comoilation and Corr"3arison of Radionuclide Sorotion Databases Usea in Recent Performance Assessmetim in Radionuclide Sorption fism the Safety l Evaluation Perspective, Proceedings of an NEA Workshop, October 1618,1991, 'nterlaken J Switzerland. ' NEA (Nuclear Energy Agency, OECD),1989. Sorption Data Base. Nuclear Regulatory Commission: Site Decommissioning Management Plan, NUREG-1444, 1993.
\'
Oztunali, O. l., G.C. Re, P. M. Moskowitz, E.D. Picazo, C. J. Pitt: Data Bese for Radioactive Waste Management, NUREG/CR-1759 Vol. 3,1981. Sheppard, M. l.: Radionuclide Partitionina Coefficizats in Soils and Plants and Their Correlation, Health Physics, Vol. 49, No., i pp.106111, July,1985. Sheppard, M. I and D. H. Thibault: Default Soil /Liouid Partition Coefficients. K s. for Four Maior Soil Tvoes A Comoendium. Health Physics, Vol. 59, No. 4 pp. 471-482, October 1990. Sheppard, M.I., G.C. Sheppard, and B.D.Amiro,1991. Mobility and plant uptake of inorganic 14C and 14C-Labelled PCB in Soils of High and Low Retention. Health Physics 61(4):481-492, Smyth, J, D., E. Bresler, G. W. Gee, C. T. Kincaid: Developmeat of an infiltration Evaluation Methodology for Low Level Waste Shallow Land Buria,' Sites, NUREGICR 5523, PNL-7356, 1990. Strenge, D.L., T.J. Bander, and J.K Soldat,1987. GASPAR-Il - Thechnical Reference and user Guide. NUREG/CR-4653. PNL-5907, U.S. Nuclear Regulatory Commission, Washington, D.C. Thibault D. H., M. l. Sheppard, P. A. Smith: A Critical Compilation and Review of Default Soil p Residential Scenario 5.10 13 January 28,1998
Solid / Liquid Pa:1ition Co:ffici:nts, K,. for Use in Environmental Assessments, AECL 10125, Atomic Energy of Canada Limited,1990. U. S. Geological Survey: Estimated use of water in the United States in 1990 Domestic Water Use, URL <http://h2o.er.usgs gov /public/watuse/ tables /dotab.st.html>, 2/15/95. O e....a, S_. s.s e .. ,_ ,2e.1,ee g
8,0 Results cf P rameter An: lysis C3 ( 1 6.1 Summary of Parameter Type, Variability, Means and input PDFs A list of all input parameters, their description, units, classification (behvioral B, Physical P, or metabolic M), how the parameter is represented (constant c, sampled from the distribution s, or a function of other parameters f), mean values and correlations to other model parameters are presented in table 6.1.1. The distribution functions representing the variability of each sampled paramter are summarized in table 6.1.2, Information contained in this table includes: the parameter I abbreviation, the parameter values for the distribution and the distribution type. Tcble 6.1.1 Summary of Parameter Types, Mean Values and Correlations
)
Parameter Desenplon Units T s Mean Sgma Related Parameterts) y C P F I Tl Expo *** pemd ind=2 at B S 240 TX,TG i TX E**** I* nod outdars at B S 40.2 Tl?G l TG E8Pa** 8**d gammng arv B S 2 92 TI,TX,UV l TTR Tot *I timo la the i y'*' **po$u'e Penod d B C 365 25 l SFl indwr sNeeng tsetor - B S 552E 01 1 SFO outdar Sh**ng ' actor - P C1 PD rioor dusiioady g/ma2 P S 0.16 RFR Resuspenson factor for indoor dust 1/m P S f 's CDI u dumeadog inders g **3 p F SFO.PF ( ) CD0 h dust **dino outda's 9**3 a P S
'd CDG h dustdoadog garoenog gem3 P S 4.00E 04 VR ereathing rue inom m*1% M C 09 VX BreatNngrate outdoors m'3h -
C 1,4 VG BmatNno ram gaenog m a lt iv, :, 1 GR seingeston transic ram o'd B b.. UW omaco name ogeston rate Ud B S 1.31E+00 H1 inckren v surta.w wyer m P C 0.15 H2 inckress v uncturated rone m P X IDH2 Ni Peosty of swfam sod - P F SCSST,NDEV N2 Porosty of unsaturated tone - P 1 N1 F1 Saturation rato for the setased mye P F BDEV,KSDEV,1 F2 saturaton rato for tre unsaturated-son - P l F1 layer VDR Vo'um* of *ater for domeste uses L B X 1.18E+05 IDVDR VSW veume of wucin surfowater pond L P C 1,30E+06 l infattaton ram rnty P F KSDEV, AP,IR Ait Ares v und evitnrsted m*2 B F 2.40E+03 IR imgaton rate Um*24 0 X 1.29E+00 SH'R 0 3204 PS soa ealdroity of swin pkie mye krm'2 P F RH01,H1 DIET Fracton of annual det dernredrom hone. - S C1 gree,11uods UV(1) Hwnan det ofleah vegetables kaY B S 2.14E+01 UV(2) Hwnan det of other vegetates ka4 B S 4 46E+01 UV(3) Henan diet of fruits kat B S 53E+01
/^'_ Residential Scenario 6.1 1 February 1,1998
Table 6.1.1 Surn. nary of Pararneter Types, Mean Values and Correlations Pa'ameter Descr'plon Uncs S Mean Sgma Related Pa'ameter(s) UV(4j Han cet of gram kg'r B S 144E+01 UA(1) Humaidet of beef kg'y B S 3 98E+01 UA(2) Human det of poultry k91 B S 253E+01 UA(3) F man def o' mA W B S 2 33E+02 UA(4) Human det of eggs kot B S 1 91E+01 UF Human det of feh kgt B S 2.06E+01 Food consumpbon penod for lea'r 8 1CV(1) vegetabw B C 3%25 TCV(2) Fad ans*Non M*d fo' oN' 8 8 C ?6525 vegetabies Fe immumpton nenod for fruns TCV(3) d B C 36525 Fad amapten penod for oram o TCV(4) B C 365 25 Food consumpton penod for beef d TCA(1) B C 365.25 Food unsumpton penod for poultry d TCA(2) B C 365.25 Fad amason pomd b a'$ TCA(3) d B C 36u TCAM) Food consumpton pened for eggs 8 8 C 365 a THV(1) Hotfup pmd b leah vegetabien d B C1 THV(2) Hoidup penod for otner vegetables d B C 14 Holdup penod for fruits THV(3) d B C 14 THV(4) Ho' dup f* nod for gesins 8 B C 14 Holdup penod for beef THA(1) d B C 20 THA(2) Hoktup Med la po*y 8 B C1 HSoup Mnod for milk THA(3) d B C1 THA(4) Holdup pomd for eggs d p Cj TGV(1) Mmimum o'o*2no penod for 'ea'y d P C 45 Wegetables TGV(2) Mamum 9"no Demd 23' oth*' d vegetaba P C 90 unmum growing pemd for fruns i TGV(3) e P C 90 i TGV(4) unmum growing pened for grains d P C 90 I ' TGF(1) Mame growing pmd w wage d P C 30 consumed by vcau TGF(2) unmum growing pe,od for forage o P C 30 consumed bv pouary TGF(3) uame owng Mmd for bage d P C 30 consumed by milk cow TGF(4) Mamum growing pened for Isage d P C 30 consumed by layer hen: TGG(1) unemm owng aned b stored grain d P C 90 consumed by beef caw TGG(2) unmum awng Mmd b stomd gram o consumed by pouty P C 90 TGG(3) une gwng semd fu stomo grain d P C 90 consumed by mA cows l ' TGG(4) Mamum o*as M*d b St* 9eam d consumed by layer hens P C 90 i ' TGH(1) uce growog mmd fa stored hay a P C 45 consumed by beef carne TGH(2) uom gwog pmd w stored hay d P C 45 consumed by pounty TGH(3) Mm e 9 "no Med b8tomdh8'I d conseed by ma cows P C 45 TGH(4) umm growog ened for stored hay 8 P C 45 conseed bylayer hens Residential Scenario 6.1-2 February 1,1998
Table 6.1.1 Summary of Parameter Types, Mean Values and Correlations 7% Parameter Desenpton Unr.s T S Mean Sgma (Wed Pavneleis) ( y c 'sg) p F RV(1) inteapton fracten br wy ve9etab*$ - P S 0.35 RV(2) l*=pton #xton b et'er egetaa - P S 0 35 inteapton fracten b fn*
- P S 0.35 RV(3)
RV(4) I*apton '*on 85 0'*ns - P S 0.35 RF(1) leapton fmeten b taf can bage - r S 0 35 RF(2) wrceptenimete u geuw bge - P i 0 35 RF(1) RF(3) leapton kacten br m* ma bage - P l 0 35 rtF(1) inteapton fracion b wyc hen brage - P l 0.35 RF(4) RF(1) RG(1) tenteon exten b taf catt* omm - P S 0 35 RG(2) iercepton freton b pww gen - P l 0 35 RG(1) RG(3) laapten kxton b am ma emm - P 1 035 RG(1) RG(4) leapte heta b wy* ten o*n - P l 0 35 RG(1) RH(1) leapton tacton for taf cattw hay - P I 0.35 RF(1) RH(2) inapton exton b pah hay - P i 0 35 RF(1) RH(3) imacten tacton baa ma hay - P I 0.35 RF(1) RH(q weapton hacion b layw ten hay - P l 0 35 RF(1) TV(1) imnsecaten faer b wafy nge4*: - P C1 TV(2) Transecaten fetw b otrer vegeta*: - P C 0.1 TV(3) Transecaten facts b frun - P C 0.1 TV(4) Transbcaton factor tw 9*ns - P C 0.1 TF(1) Trwoecaten factor br teef catte bge - P C1 TF(2) Transecaten txts b paw bege - P C1 TF(3) Trusecaton txtor b tr* ** lorage - P C1 Transecaton factor b byer hen forage - P C1 (j \' g TF(4) TG(1) Tm=$caten factor b test catte gram - P C 0.1 TG(2) Tanskcaton fxt* br smu% oren - P C 0.1 TG(3) Transiocaten fxta b mm co. omm - P C 0.1 TG(4) Tmnsecate tutor b wyw hen ome - P C 0.1 TH(1) Tensecaton factor b taf can hay - P C1 TH(2) Tmnsecate factw or pa% har - P C1 TH(3) Transecaen facts be mm cow hav - P C1 TH(4) Trruecaton faeor trlayw Way P C1 XF(1) Fracten of contamnated teef caed - B C1 forage XF(2) Fracton of contammated pww brge - B C1 XF(3) Fmeten or contamnmed enA con twMe - B C1 XF(4) Fracten of entamnated myw ten bage - B C1 XG(1) Fracts of contammated teef cat
- O'an - B Cj XG(2) Fracton of contamma:ed poutry own -
B C1 XG(3) Fmetenorantammatedamco.omm - B C1 XG(4) Fracten et mniammmed layer ten omm - B C1 XH(1) Fruton of contammated taf can hay
- B C1 XH(2) Fracten of contammated poutry hay -
B C1 XH(3) Fracten of contammatea mm cow hay - B C1 XH(4) Fracten of contammated layer tea toy - B C1 XW(1) Fracten of cTtarmated teef cattle water - B C1 XW(2) Fracton of contannated pouw wate - B C1 XW(3) Fracton of contammmed am cow water - B C1 [] Residential Scenario 6.1-3 February 1,1998 ()
Table 6.1.1 Summary of Parameter Types, Mean Values and Correlations Pararneter Descripton Units $ Mean $g'na Related Pammeterts) Fracton of contaminated layer hen water - XW(4) B C1 Crop ye4 br 1889 ve9etabies YV(1) kaw2 P S YV(2) Crop yew for otter vegetates kgw2 P S Crop yee b fruns kgw2 YV(3) P S yy(4) Crop yeid for grans how2 P $ YF(1) C'op y*d b '*ef carta %e kaw2 P F DYF(1) WF(1) YF(2) crop yew b po*y bage kow2 P 1 YF(1) YF(3) Crop y*e b ma cow forage how2 P l YF(1) YF(4) Crop y** b isfer ten forage how2 P ! YF(1) YG(1) crop yew bteefcan gram ko w 2 P F Crop ye bpowy pram DYG(1) WG(1) YG(2) kaw2 P i Crop reid b rna me aran YG(1) YG(3) kaw2 P I YG(1) YG(4) cropyed b laywfenomm 60w2 P ! YG(1) YH(1) C'op yed b teef catue her kow2 P 1 YF(1) YH(2) Crop Y*id b po*y hat how2 P i Y. ,o v'H(3) Crop yeid b rna con rey kow2 P l YF(3) yy(4) crop yeid for layer hen hay hw2 P l YF(4) WV(1) Wevdry conv. mon factor for lea 4 - F X vegetables DWV(1) 0.0324 Wy(2) wevery convemon factor b cher - P X vegetables DWV(2) 0.0324 Wet /d'y conversen tactor br fruits WV(3) P X DWV(3) ').0324 WV(4) Wet /d'y convemen factor for grains - P C 0fB wndry unvemon factw b teef atte - WF(1) forage P S 0.252 3.84E 02 WF(2) WW*y convemon factor br powy - P l forage WF(1) WF(3) WW*y converson factor for rna cow - P , lorage WF(1) WF(4) WWory conarson factor b larer hen - P 1 brage WF(1) l WG(1) wecy convemon factor b teef catte - P C 0,88 gram DWG(1) 0.87417 WG(2) W**y mavenen fac* b' paltry omm - P t WG(1) WG(3) Weu*y convecon fanor for trum mw - P i gran WG(1) WG(4) Wet / dry con orson factor for byer hen - P l gSe WG(1) WH(1) wevey converson factor fet te f catta - P l WF(1) hay WH(2) wwey convwson fe:w br powy hay - P 1 WF(2) WH(3) Ww*y convenen facte tema cow hay - P i WF(3) WH(4) WeVory converson factor b tayer hen - P i hay WF(4) QF(1) ingeston rate b teef can bage kg'd P F DQF(1) QF(2) Wuton rate b powy bmge ko'd P F DQF(2) QF(3) Inge$ ten ute b rna m. brage kg'd P F DOF(3) 6 2584 QF(4) Westen rate forlayer hen forage ka'd P F DOF(4) QG(1) Ingesten rate b teef can gram kg1 P F DOG (1) QG(2) ingeston rate b po*y o'aa 'g'd P F DQG(2) QG(3) Watoa Ste b rna co* o'.m ko'd P F DQG(3) OG(4) inguten rate br ierer hen gram koM P F DOG (4) Residential Scenario 6.1-4 February 1,1998
j
- l l
Table 6.1.1 Summary of Parameter Types. Mean Values and Correlations l , Pryneve Onenpten Una T S Mean $gma Related Prometerts) y C D F OH(1) _ ino*So rate b teef came har ko'd _P F DOH(1) , OH(2) Insesten tem for pou% har kold P .C 0 QH(3) inonton rate for rna coe hay kg/d P.F DOH(3) 5.00672 OH(4) Ino*eten raie forioyw hen hay ko'd P C0 QW(1) war monten r* b tad cam Ud P C 50 QW(2) **' p9 den r* b prey V P C 0.3 QW(3) Wm'inomien r*e b ma cows Ud P C 60 QW(4) W*' *v.eten rm b in w hens Ud P C 0.3 OD(1) sos m keen b teef catt* - P C 002 0D(2) Sd Hake keen h po*y - P C 0.1 0D(3) Sd in'ad kdan b ** cows -
- P C 0.02 00(4) su reake tecten for mye hws .
P C 0.1 i - MLV(1) Ma**bedmg teor b levy nomises 91 P C 0.1 MLV(2) Ma***d"o'Mo'bathernosten et P C 0.1 Maatseng beior = trun 91 P C 0.1 MLV(3) MLV(4) Maa***no '8c80' 80' oroms o'o P O 0.1 LAMBDW W**t*nne r** b act'v4 'unovel frorn Ud P C 4 956 02 , pine RHO 1 Sudac* Sd Den $ry ou P F N1 RHO 2 Unsaturmo Zw= Sd DwmW ou P t RHO 1 TTG Totalinne e gardenco renad d B C 90 TF Fah consumpton pomd d B C 36525 , TD Drr*9-**r consumpton penod d B C 365.25 2 MLF(1) uassung tacts b teelcane base 91 P C 0.1
- MLF(2) Maseadmg facts W paw wage 01 P C 0.1 MLF(3) Ma** Moo factor W ak co* Image et P C 0.1 MLF(4) Ma** Wing facts for mye ten wage sig P C 0.1 -
MLG(1) M*uwog beer b tow cans gram et P C 0.1 - MLG(2) Ma* Woo hem' b po*y own 91 P C 0.1 MLG(3) usesee$ng factor tw am cow own n P C _ 0.1 MLG(4) Ma*Was betr
- Myv hen oma o1 P C _0.1 MLH(1) - Masstamng factor for teef cane hay 91 P C 0.1 MLH(2) Masswing factor for po*y hay 91 P C 0.1 MLH(3) M'a#8&no factor lo'ma cow her . et P C 0.1 MLh(4)
Ma*** dmo factor Wlays hen hay sig P , C 0.1 TFF(1) Fad' ped b ted cane boos ' d P. C 365.25 TFF(2) F**d'no p*no8 lor pov% woo. 8 P C 365.25 TFF(3) Feedmg pened for nm coo km o P C 365.25 TFF(4) F**d no panod lo'leyw tenIwase d P C 365.25 TFG(1) Faeno penod 80' teef catie grain 8 p C 36525 1 Fadmo :=md W pa% own d P C 365 25
- TFG(2)
Fad *g pwed for ma co. gen d P C 365.25 TFG(3) . TFG(4) F** des penod for layer h*a gran d P C 365.25
- Fudog p.md w teef can hay 8 P C 365.25 TFH(1)
Fades p*ed W pou% toy 8 P C 365.25
- TFH(2)
' TFH(3)
Fad"? 'emd is mm cow hay d P C 365.25 _ F**d*o penod 'o*yer hen hay P C 365.25 d TFH(4) W*' *onten pened b t d cett* d P C 365 25 TFW(1)
; D Residential Scenario 6.15 February 1,1998
,-.,yw-m--w.-- . . _ _ , , ,.7 --,--v .- ,ow-, , , - ,-,, --
Table 6.1.1 Summary of Parameter Types, Mean Values and Larrelations Parameter Desenption Uruts T S Mean Sgm, Related Parameter (s) - y C p F e Water Westion oenoo for povitry TFW(2) a P C 365.25 watergesten pem : for ma cows d TFW(3) P C 365 25 Wateringesten penoi. .oriayer nens TFW(4) d P C 365.25 H Partition coemcent mW P C0 Be mW P X L10KdBe C mW P X L10KdC F mV9 P X L10r(dF Na mV9 P X L10KdNa P mW P X L10KdP S mW P X linKdS Cl mLt P X '1Cl K mug p x ,cg Ca mug P X +1 dCa . Sc mV9 P X MdSc Cr mU9 P ' c10KdCr Mn mueP L10KdMn Fe mug F DL10KdFe Co mug P ), DL10KdCo Ni mug F X L10KdNi Cu mug P X L10KdCu Zn mV9 P X L10KdZn As mUr P X L10KdAs Se mus P X L10KdSe Br mug P X L10KdBr Kr mus P C0 Rb mu9 P X L10KdRb Sr mus P X L10KdSr Y mW P X L10KdY Zr mug P X L10KdZr Nb mW P X L10KdNb Mo mug P X L10KdMo Tc mV9 P X L10KdTc Re mug P ". L10KdRu Rh mug P X L10KdRh Pd mU9 P X L10KdPd Ag mV9 P X L10KdAg Cd mV9 P X L10KdCd in mug P X L10Kdin Sn mug P X L10KdSn Sb mV9 P X L10KdSb Te mug P X L10KdTe i mug P X L10Kdi Xe mug P C0 Cs mu9 P X L10KdCs Ba mug P X L10KdBa La mug P X L10Kdla Ce mug P X L10KdCe Residential Scenario 6.1-6 February 1,1998
Table 6.1.1 Summery of Parannter Types, Mean Values and Correlations f^s Parameter Desenpbon Units T S Mean Sgma Related Parameter (s) y \, )} l C Pr- mus P X L10KdPr Nd mug P X L10KdNd Pm mug P X L10KdPm Sm mV9 P X L10KdSm Eu mug P X L10KdEu 1 Gd mW P X L10KdGd j Tb mug P X L10KdTb Ho mV9 P X L10KriHo W mus P X L10KdW Re mug P X L10KdRe Os mus P X L10KdOs tr mUs P X L10Kdtr Au mus P X L10KdAu Hg mW P X L10KdHg mus P X L10Kc m mVD P X- L10KdPb Bi mUp P X L10KdBi Po mVD P X L10KdPo Rn mug P C0 Ra t'f>9 P X L10KdRa Ac mW P X L10KdAc Th . mVU P X L10KdTh Pa rnug P X L10KdPa (] ('~y U Np
.nuo P X mug P X L10Kdu DL10KaNp Pu mug P X L10KdPu Am mW P X L10KdAm Cm mV9 P X L10KdCm Cf mug P X L10KdCf fea(1) carton frr.cten fr beef catue P C 0.36 fea(2) es:t,n fracton kr poumy -
P C 0.18 fca(3) carten frrten for mfk em P o 0,06 fea(4) carton fracton forlayer hens - P C 016 icf(1) carten fracten b teef cattle scrage - P C 0.11 e fcf(2) carten fracten b poultry wage - P C 0.11 fcf(3) carten fracten for ek cow forage - P C 0.11 fcf(4) carte fracten b tayer men forage P C 0.11 fch(a) carten fracten b beef catre har - P C 0.07 fch(a) carton fracten b poumy hay P C 0.07 fch(a; . carbon fracten b mdk ccw hay - P C 0.07 fch(a) carton fracton forlayer hen hay P C 0.07 fcg(a) r ten h for oeef came con - P C 0.4 fcg(a) aten fracron b poultry gen - P C 0.4 fcg(a) certon fraccon b mdk cow gen - P C 0.4 fcg(a) cartonfractonforlay rhengrain - P C 0.4 fedO5 Fraccon of czten m soil - P C 0.03 satac specific actu.ty equivmerce b westrk - P C1
,q Reeidential Scenario 6.1-7 February 1,1998
]
Table 6.1.1 Summary of Parameter Types, Maan Values and Correlations Pa'ameter Descepton Units T s Mean Sgma Related Parametefts) y C p F e fha (1) Hydrogen kacton W t>eef catDe - P C 0.1 tha(2) Hydrogen fracton for pou% - P C 0.1 tha(3) Hydmge kactu W sk cows P C 0.11 tha(4) Hydmgen kacten br layer hens - P C 0.11 fhv(1) .Hydege fracton b leafy vegeta*s P C 0.1 fhv(2) Hydrogen tacton for other vegeta*s - P C 0.1 fhv(3) Hydrogen fradon for fnds P C 0.1 fhv(4) Hydrogen fracten for grens - P C 0.068 fhf(1) Hydrogen fracten for beef catue forage - P C 0.1 fhf(2) Hydrogen kacten W poultry bre - P C 0.1 fhf(3) HY1mga kacten w mk cow forage - P C 0.1 thf(4) Hydrogen frauon iorlayer hen forage - P C 0.1 thh(1) Hydmgen hacten for tieeicatue hay - P C 0.1 fhh(2) Hydegen fracton b poultry hay - P C 0.1 thh(3) Hydrogen fracten for mk cow hay - P C 0.1 fhh(4) Hydrogen fracton forlayer hen hay - P C 0.1 fhg(1) Hydmga l'acton w taf catte gran - P C 0.068 thg(2) Hydrogen fracten for poultry gram - P C 0.068 thg(3) Hydogen fracten for ok cow grain - P C 0068 fhg(4) Hydmgan fracten for layer hen grain - p C Q.068 fhdO16 Fraction of nyosogen in sod P F 0.011 SH sasvh Treumeouivasnce paresod - P C1 sawvh Treummuvaience plant! water - P C1 satah Treummuvaence anreiproductantake - P C1 sh umsture content of sod um a3 P F 0.1 N1, RHO 1.F1 BIH concerraten tactor wafy P C0 B1Be concentraten factoc inafy - p 3 B1C concetraten factor leafy p >; B1N concetmon factx leafy p s B1F concentmen factoc ieafy p 3 B1Na Concat**a 'ar'ar *a'Y - P S B1Mg raentraten factx leafy - p s BISi concentraton 'actac2a4 p s BIP concentraton fa:tx leafy - p s BIS cor.:entrcton factx ieah p s B1CI conceaaten factor 'efy - p s B1Ar ccncentraten factor leafy P C0 B1x concentraton factor leafy p s B1Ca concent aten factor u, p s B1Sc cor.centraton facts leafy - p s B1Cr concetmon facte leafy - p s B1Mn cecentmon factor leafy p 3 B1Fe cmcentraten factx ieafy p s B1Co concentmen factx teafy - p s B1Ni concentration factor leafy - p s CICu cecccaton facts ieafy p s B1Zn concentraten factor l-, - p s Residential Scenario 6.1-8 February 1,1998 l l l l I
- Tabb 6.1.1 Summary of Parameter Types, Mean Values and Correlations Parametu Desenpton Unds T S Mean- Sgma Related Pa ameter(s) y C O- -
p F p s BIGa coramaton factx wsh . B1As- concenramn factx % - - p s B1Se _ corantraen factx % . p_s
- BIBr conce rmen factor. %- . p s B1Kr concenvaten factx we . .
P C0 BIRb coraemon facix % . .p s BISr concenteen factor we = . ps B1Y coraemon faciot M . ps BIZr canemen metr % . p s B1Nb Cacemmen factx % . P S B1W - comemon taca mah . p s B1Tc caammen m he . p s B1Ru concommen hew w . p s-B1Rh ca n e menfactx % . p s , B1Pd concommen facte w . p s B1Ag Can estontaca M . p s B1Cd cerammen tactor wah . p s Biln conceeston hcw % - _p_s l B1Sn caammatecw w . P S l BISb concecumn taca me . p s I- B1Te coraremen facir nah . p s Bil concewmmn factor % . p s B1Xe coraremen factor % . P C0 B1Cs concommon facw w . p s g B1Ba concommon facte wah - . p s BILa conce**en factoc % . p S Bice conc emenfactorwah . p- 3 BIPr concremen factx % . p s BIND caammen factoc % . p 3 B1Pm concecumn face wary . p s BISm conecemen facir we - p s BIEu cora e amnfactx % p s ,
-B1Gd conceemon bcier % . p ,,
B1Tb conceemen tactor. mah - p s B1Dy canemen factx % . p s B1Ho- concessen tactx w . p s-B1Er concemmontectorwah . p s B1Hf concommen faa % . ps B1Ta concentraten tectoc led - p s 31W can eamnfaca % . p s B1Re cuaesen factne we - p s B10s concemmen tactet kah . p s B1tr concecaentectocisah - - p s B.Aa concemmen taciocioah . p 3
?+; conceween tectoc hah -
p s
., rt i cweeremen faciociesty - .
p 3 B1Pb - caammenfactx % . ps e Residential Scenario 6.1-9 February 1,1998 (s (
Table 6.1.1 Summary of Parameter Types, Mean Values and Correlations Parameter Desenpten Units T S Mean Sgma Related Psrameter(s) y C p F e BIBi cmmaten factx ha9 - P S BIPo cecemten factx wa9 P S BIRn cmcewaten facer ha9 - P C0 BIRa coxmaten factor lea 9 - P S BIAc cecennten factx ha9 - P S BITh concentraten factor lea 9 - P 3 B1Pa cmcewaten factx ka9 - P S B1U comemm factx ha9 - P S BINp concerwen factor lea 9 - P S BIPu cmceemen factx ka9 - P S BIAm conceenten faca kaY - P S B1Cm ceceesten factor lea 9 - P S B1Cf conceemen tactor leah - P S 82H concentraen factor r4 - P C0 828e concentraten factor rmt - P S B2C concematen factor rut P S B2N cecematen tactx rut - P S B2F concentraen factor rmt - P S B2Na concentraten factor rmt - P F WB2Na B2Mg concemmen factor rmt - P S B2Si concentranon factor rmt - P S B2P cecantraten factx root P S B2S concentraten factor ret - P S B2Cl concentraten factx rmt - P S B2Ar concentraten factor root - P C0 B2K conceemen factx rmt P S B2Ca concentraten factx root - P S B2Sc concetraten factor rmt - r S B2Cr emeaten factx rmt - P F WB2Cr B2Mn commaton factx root - P F WB2Mn B2Fe concentraten factx rat - P F WB2Fe B2Co concentra=n factor fut - P F WB2Co B2Ni Cecewaten factor r*t - P F WB2Ni B2Cu concentra=n factor rmt P F WB2Cu B22n concemaen factx rmt P F WB22n B2Ga conceeamn factor rmt - P S B2As concentraten factor trot P S B2Se contecaten factor rmt P S 52Br concentraten factor rout P S B2Kr concentraten factor root - P C0 B2Rb concentraten factor root - p s B2Sr concecaen facmr rmt - P F WB2Sr B2Y concentraen factx rmi p s B22r concemamn tactx rout - P F WB22r B2Nb cmemnon factor root - p s 82Mo concewaten tactor rmt - p c B2Tc concernen factx rmt - p s Residential Scenario 6.1-10 February 1,1993
Table 6.1.1 Summary of Parameter Types, Mean Values and Correlations c Paremmer Desenphon Units T S - Mean -Sgma' Related Parameter (s)
/ y C
( p F B2Ru . Commann for ret . P F WB2Ru B2Rh- countraen facte root - - p s B2Pd. cececate factx mot - p s B2Ag conceemen factoc root . p s B2Cd comeann factor rmt . ps B2in . co==am facior rat - p s 82Sn Cace*senfacirrat . p s
- 82Sb Co*ntr8mn factm 'oct -
P S
~B2Te comesa tecex mot .
P S B21 cmemenen facme mot . P F WB21 B2Xe concentraen facir root . P C0 l- B2Cs comemen tector root - . P.F WB2Cs - B2Ba ceceemen teciac mot .
-P F. WB2Ba
! B2La cmecesten factor rat . p s B2Ce cmessem tecion runt _ . P F WB2Ce
-B2Pr cmememen facior mot . p s B2Nd cmceemenlocac rut - p 's B2Pm com sectoc rat - - p 3.
B2Sm : concemehm factor rat . p -3 B2Eu. cecesma tecior rat .
.p s B2Gd ceceemm factx rooi .
p s B2Tb cmcemann facioc rat - - p-s 82Dy . concuenten facmem . p s B2Ho- *e factor rat . . p 3 8
,\ M-- B2Er cmceemen factor rat - .
os B2Hf ' conceemen facie me: - . .p s
-B2Ta caceamen facme mot . p .s B2W- ceceema facioc root . p s
-B2Re comemen fator rmt . p s B20s cecewaten factor u - - p-s --
B2tr cmemesa facme rmt . .p s
'B2Au conceemen facmcie. .
p s u2Hg ~^ m tectorro.' . p s B2TI cecewann tector rmt . p s B2Pb caceram tacion rat - . p 3 B2Bi cmeccehon factor rat . . p- s B2Po conceemontecme rom . p s b- B2Rn cacewman fuoc rat . P C0 B2Ra cecewaten tactoc rat . p s B2Ac concocatenlectoc mot . p S B2Th- cmecemm factor mot . p s
-B2Pa- Ceceemon factoc mot .
p 3 B2U cecewmenfactx mot . p s B2Np Cece* men facix root . P F WB2Np B2'%3 cecentreonsectoc rat . p s B2Am cmeceam tecix mot . p s B2Cm - cececaten facioc ret . p s Residential Scenario 6.1-11 February 1,1998
'\
~I i
Table 6.9.1 Summary of Parameter Types, Mean Values and Correlations Parameter Desenpton Units T S Mea. Sgma Rela:ed Parameterts) y C p F e B2Cf cmcemmen factor mot p s B3H concentraton factor inat P C0 B3Be concentraton factor wt - p s B3C concentraton factor ht p s B3N comenton factor inat - p s 83F concentraton factor M - p s B3Na cmcevramn factor fnat - P F WB3Na B3Mg commratra factor fndt p s B3Si concewate factx Ni - p s B3a concentramn factor ha p s B3S commann factx fndt - p s 63Cl concentraten factx ha - p s B3Ar ccncecaten factx M - P C0 B3K conceem factor h4 - P S B3Ca concentramn factor inat - P S B3Sc countru factor enat - P S B3Cr concentramn factx fnnt - P F WB3Ct B3Mn concentramn factor fnnt P F WB3Mn < B3Fe concentraten factor fnat P F WB3Fe B3Co concentramn factor Nt P F WB3Co ' B3Ni concentraten factor bt P F WB3Ni B3Cu cmcentramn factor indt P F WB3Cu B32n concentraten factor Nt - P F WB32n B3Ga concentrawn tactor M P S B3As concentramn tactor M - P S B3Se concentramn factor fndt P S 83Br concentraen factx fnat - P S B3Kr concentraton factx fnat P C0 C3Rb cmcemamn factor fnat P S B3Sr concewamn factx indt - P F WB3St B3Y Cecaeamn W 6t P S B3Zr concentraton fact:c fnat - p , WB3Zr E'Nb conce- m factor tnat - P S B3Mo concentraton factx fnat P S B3Tc comentraen factor Nt P S B3Ru concentramn factor fnat P F WB3Ru B3Rh concentrawn factor fnet P S B3Pd cmcentramn factx ha P S B3Ag concemamn factor h4 - P S B3Cd concontramn factor h4 - P S B31n concentraten factor bt P S B3Sn countraten factor innt - P S B3Sb concentraton factor fnat P S B3Te concentramn factor Inst - P S B31 cmcentraton factor fnat P F WB31 B3Xe concentratm factor ht P C0 B3Cs concentraton factor u - P F WB3Cs ResMential Scenario 6.1-12 February 1,1998
Table 6.1.1 Summary of Parameter Types, Mean Values and Correlations Parameter Descr$ ton Units T S Mer Sgma Related Pammeterts) y C 9 B3Ba B3La ceceesten factor fut comeenten factx fut o F e P p F s WB3Ba B3Ce coramten facts fut - P F WB3Ce B3Pr cecenten factor fut . p s B3Nd cecewaten fetx fut - p G B3Pm comentaten factx fut - p S B3sm conceenten factor fut - p s B3Eu emceesten tactx tna . p s B3Gd cecentaten factor fut . p s B3Tb concentaten tactor fut - p s B30y concetraten factor fut , p s B3Ho concewaten factx M - p s , B3Er ceceesten factor In* - p s B3Hf coramaton factor fut - p s B3Ta cmcemen factor fut - p s B3W conewaten factx fna - p s B3Re conewaten tactx tut . p s B30s ceceente factx fut - p s 83lr concemmen factor fnm . p s B3Au coweten factx fut - p S B3Hg heten factor fut - p $ B3Tl cometraten factx fut - p s t B3Pb concewaten factor fut p S Esi concentratm factor fut . p s B3Po concentraten factx fnd p s O i B3Rn cmcewaten factor fut - P C0 B3Ra cmcewaton factor fut p s B3Ac cecematon factx fut p s B3Th coramten factx fut p s B3Pa comentrau factx fut p S B3U corantraten te fut . p s 53Np co<c mon facy fnut P F WB3Np B3Pu cmcentaten facar fut . P s B3Am concewaten factor fut P s B3Cm concewaten factor fut - p s B3Cf cec itratenfacerfnnt - p s B4H concentraten factor gen P C0 B4Be comeenten factor gum - p s B4C concentrauen factor gran p s B4N concentraten factor gr:m p s B4F concentraten factor grain - p S B4Na cecewaten factor gen . P F WB4Na B4Mg comeeaten facw son - p s B4si concentraten factor gran - p s B4P comewaomlactor gen - p s B4s coramaten factor gram - p s B4Cl ceramaten facer gam - p s Residential Scenario 6.1-13 February 1,1998 9 1 3 _ _ ..
Prameter Tabla 6.t1 Summary of Parameter Types. Mean Valu Desenpton Una T s Mean Sgma Re!ated Paameteqs) p F B4Ar e concentraten far'or ?$n - B4K B4Ca concentraton facte een P C0 concentraten factor grain P s B4Sc concewaen factx gmin P s B4Cr cmcersaron faer een PS B4Mn concennten fxtor grain P F B4Fe concecaten factor g*n P F WB4Cr B4Co F F WB4Mn concentrat.on factor vam - B4Ni concemaenfacmrsm* P F WB4Fe B4Cu concennten factor gen P F WB4Co B4Zn concetman factx g*n P F WB4Ni B4Ga concewaten factor g am P F WB4Cu B4As ceceee tactx gram - p s WB42n Base concentrae , actor som p s B4Br concecaten factx gram p s B4Kr concennten fa-tor grain - p s 84Rb concewamn fxtx gen P C0 B4Sr concentraten fac or grain p s B4Y concentraten factor gen P F B4Zr p 3 WB4Sr concentraen factor gmn B4Nb cecemaen factor g am P F 84Mo concentraton factor grain P s WB4Zr B4Tc concennten fxtor gram P S BARu concentrate, tactor grain P s B4Rh P F concentraen fxtor g%n B4Pd concent aton factor gen -P WB4Ru S B4Ag concentrawn fxtor grain - PS B4Cd concentreten factor gen - PS B4ln conceesten factor gen P S B4Sn concentraten factor gen PS B4Sb concentraten factor P S B4Te ceceesen 'aer ein , gen - P s B41 concennten t ctor gram P S B4Xe concentrate-faer gran P F B4Cs B4Ba concecamn factor gram - P C0 WB41 concentraen facte gram - P F B4La cacentraen factor gen P F WB4Cs B4Ce concentraten factor gram - p s WB4Ba B4Pr concentraten factor gram P F B4Nd concentraten factor grain - p s WB4Ce B4Pm concentraron face r gem . p s B4Sm concentraen facte gen p s B4Eu concentraton t.ctor gran - p s B4Gd concewaen factx gem - p s B4Tb p S cmcemamn factor gran - B4Dy cc,ceeaan factor gen - p s B4Ho conceeamn factor grain - p s p s Residential Scenario 6.1-14 February 1,1998
Table 6.1.1 Summary of Parameter Types, Mean Values and Correlations [ Parameter Desenpten Units T S Mean sgrna - Related Panneter(s)
~( '
y- c p F B4Er excemann facw gen - . p s B4Hf concentraen factx gen . p s 84Ta concentraen factx g3n . p s B4W cawtraen factx gen . p s B4Re . Conceesen factor gran - p s B40s concentranon face gran . p s i l B4lr cwentracon facte gen . p S ! B4Au concentraen factor gen - - p s B4Hg concesesen factor gree p 3 l- - B4TI conceesmaniactx gen . p s B4Pb concemanon factor gen - p s B4Bi concewanonsaccor een - p s B4Po conce*abon fac80c g*n - P S
-B4Rn conewescon factor gran .
P C0 - BtRa concewatm facer gen . p s B4Ac conemeanon facex gen . p s B4Th concecuen ta:tx g=n . p s
-B4Pa concentramn factor gran .
p's B4U concevaan factx gen - - p s-B4Np concentrater w gran - P F WB4Np B4Pu concentracon facier gran concentraten factoe gran os B4Am .
, s B4Cm cxcemmen tactor gren .
P S b B4Cf concemmen facioc gen - - p s Table 6.1.2 Summary of 'ampled Distributions Parar,eter Symbol DirtType Parameters Description SoilTexture SCSST DISCRETE CUMULATIVE 12 #
- sitt 1 1.00E 04 .#
s::..dy day 2 1.34E-03 # sandy dayloam 3 1.06E-02 # silty clay 4 2.51E-02 # loamy sand 5 6.17E-02 .# day 6 1.09E-01 # day loam . 7 1.62E 01 # silty dayloam 8 2.12E-01 # sand 9 2.85E 01 # sandyloam 10 5.10E-01 # sittloam 11 7.58E-01 # loam 12 1
; Exposure period: indoors TI CONTINUOUS 1.INEAR 17 #
174 0 # 174.12351 0.001 # 190.200'.5 0.011 # 201.56782. 0.051 # O G' Residential Scenario 6.1-15 February 1,1998
Table 6.1.2 Summary of Sampled Distributions Parameter Symbol D:st Type Parameters Desenption 209 20749 0.101 # 218 39387 0.201 # 226.26713 0 301 a 232.06768 0 401 # 238 49259 0.501 # 243 90886 0.601 # 249.02371 0.701 0 255.31197 0.801 # 266 28312 0.901 # 273.17426 0.951 # 279.68321 0.981 # 297.96814 0.999 # 3% 1 Exposure penod: outdoors TX CONTINUOUS UNEAR 17 # 16.8 0 # 16.810221 0.001 # 21.108236 0.011 # 24.760913 0.051 # 27.866839 0.131 # 32.480083 0.201 # 35.372987 0.301 # 38.321517 0.401 # 7 40.876787 0.501 # 44.327105 0.601 # 48.013569 0301 # 32.277452 0.801 0 58.049585 0.901 # 63.358368 0.951 # 69 873059 0.981 # 84.319197 0.999 # 90 1 Exposure penod gardaning TG CONTINUOUS UNEAR 17 # 0.02 0 # 0.0350267 0.001 # 00943619 0.011 # 0.3248655 0.051 # 0 4503669 0.101 # 0.7201827 0.201 # 1.0312204 0.301 # g 1.3469017 0.401 # 1.742766 0.501 # 2.5641202 0.601 # 3.5792239 0.701 # 5.2068126 0.801 # 7.0728595 0.901 # c 8.4409701 0.951 # 10.993619 0.981 # 16.686302 0.999 # Resiciential Scenario 6.1-16 February 1,1998 l 1
Table 6.1.2 Sumrary of Sampled Distnbutions Parameter Symbol Dist Type Paramete a Desenpton 17 1 Indoor shielding factor SFl DISCRETE CUMULATIVE 4 # 0.479 0 25 # 0 486 0.5 # 0.517 0.75 # 0.857 1 Floot dust-loading PD UNIFORM 0.02 0.3 Resuspension factor for indoor dust RFR LOGUNIFORM 1.00E-07 8.00E-05 Indoor! Outdoor pentrabon factor PF UNIFORM 0.2 0.7 Air dust-loadir.g outdoors CDO LOGUNIFORM 1.00E 07 1.00E 04 Air dustsoading gardening COG UNIFORM 1.00E-04 7.00E 04 Soilingeston transfer rate GR TRIANGULAR 0 0.05 Dnnking wateringeshon rate UW LOGNORMAL N 0.15245 0.489 index of thickness of unsaturated IDH2 UNIFORM 0.5 211.5 zone Porosity pobabilitylevel NDEV UNIFORM 0 1 Ksat probabilitylevel KcDEV NORMAL 0 1 Ksat sand KSSAND BETA 3.50E 04 1.86E 02 Ksatloamy sand KSLMS BETA 3.90E 05 1.34E 02 Kr 'b' parameter: lognormal deviate BDEV1 NORMAL 0 1 , Kr *b' paranater: lognormal deviate BCLAY B2TA 4.93 75 Volume of wa 'or domestic uses: IDVOR UNIFORM 0.5 50.5 index Potential applicahon rate AP CONTINUOUS LINEAR 12 # 23.9 0 # 24 4.62E-01 # 25 4.76E 01 # 30 5.40E 01 # 35 6.29E-01 # 40 7.05E 01 # 45 8.04E-01 # SO 8.79E-01 # 55 9.41E 01 # 60 9.82E-01 # 65 9.98EI.1 # 7u 1 Shifted irngaton rate SHIR !.OGNORMAL N -0.40048 0.8698 Human (et ofleafv vegetables CONTINUCUS LINEAR 11 # UV(1) 0 0 # 1 0.01 # 1.0419771 0.05 # 2.40428d2 0.1 # 5.8935704 0.25 # 11.678991 0.5 # 24.577 % 1 0.75 # 46.2664 % ,3 # 66.028747 0.95 # 135.51767 0.99 # ResidentialSctnario 6.1-17 February 1,1998 9
Table 6.1.2 Summary of Sampled Distributio:1s Parameter Symbol Dist Type Parameters Desenpbon 222.94958 1 Human diet of oher vegetables UV(2) CONTINUOUS LINEAR 11 # 0 0 # 2.227247 0.01 # 4.1f20783 0.05 # - 5.9511124 0.1 # 11.27494 0.25 # 2*.637421 0.5 # 55.573628 0.75 # 77.073484 0.9 # 145.5 % 79 0.95 # 301.48693 0.99 # 384.02519 1 Human diet of fruits UV(3) CONTINUOUS LINEAR 11 # 0 0 # 1.9304561 0.01 # 3.6383867 0.05 # 5.0829004 0.1 # 9.4811069 0.25 # 20.479001 0.5 # 45.360963 0.75 # 125.95513 0.9 # 190 05007 0.95 # 460.83695 0.99 # 573.56751 1 Human diet cf grain UV(4) CONTINUOUS LINEAR 11 # 0 0 # 1.4059247 0.01 # 2.2237757 0.05 # 3.2207749 0.1 # 4.8292151 0.25 # 8.2018766 0.5 # 15.803996 0.75 # $ 31.779351 03 # 44 00817 0.95 # 84.783882 0.90 # 99,466253 1 Human diet of beef UA(1) CONTINUOUS LINEAR 11 # 0 0 # 2.4236457 0.01 # 7.034812 0.05 # 8.1970853 0.1 #
- 13.258072 0.25 #
28.791752 0.5 # 48.407663 0.75 # 76.750818 0.9 # 105.7057 0.95 # 220.05707 0.99 # Residential Sce;.ario 6.1-18 February 1,1998
Table 6.1.2 Summary of Samp!:d Distributions n Parameter Symbol Dist Type Parameters Descri; bon {s 222.74866 ,1 :
- Human det of poultry UA(2) CONTINUOUS LINEAR 11 #
O O # 3.8466442 0.01 # 4.1841562 0.05 # l 5.9362405 0.1 # 9.569458 0.25 # 19.853647 0.5 # 1 38.218271 0.75 # 50.825337 0.9 # 58.518625 0.95 # 72.813251- 0.09 # 72.9 1
- Human diet of milk UA(3) CONT!NUOUS LINEAR 11 #
0 0 # a 6.5899315 0.01 6.8579431 0.05 # 7.672488 0.1 #. 58.626217 0.25 '# 148.56249 0.5 # 294.81273 0.75 # 554.9416 0.9 # 721.00367 ' O.95 # 1210.7817 0.99 # 1211 1 (v Human diet of eggs UA(4) CONTINUOL 4R it # 0 0 # 2.7958156 0.01 # 4.5018568 0.05 # 5.3004593 0.1 #
. 8.2333802 _ 0.25 #
12.360671 0.5 # 21.35025 0 'S # -+ 35.901778 0.9 # 47.35077 0.95 # 120.79913 0.99 # 121- 1
- Human diet of Esh - UF CONTINUOUS LINEAR - 11
#=
0 0 # 1.8477969 0.01 # 1.9170239 0.05 # 2.8419286 0.1 # 3.6786896 0.25 # 7.7675824 0.5 # 16.139546 0.75 # 39.081046 0.9 # 79.047552 0.95 # 111 81583 0.99- # Residential Scenario 6.1-19 February 1,1998 Ci
Tabb 6.1.2 Summary of SampM Distributions Parameter Symbo! Dist Type Parameters Desenption 852 05534 1 Interception fraction for leafy RV(1) UNIFORM 0.1 0.6 vegetables interception frachon for other RV(2) UNIFORM 0.1 06 vegetables Interception fraction for fruits UNIFORM RV(3) 0.1 06 Interception fraction for grains RV(4) UNIFORM 0.1 0.6 Interception fraction for beef cattle RF(1) UNIFORM 0.1 0.6 l farage ! Interception fraction for beef cattle RG(1) UNIFORM 0.1 0.6 ) grain Crop vield forleafy vegetables YV(1) CONTINUOUS LINEAR 22 # 2.7 0 # 2.713438 0.0016 # 2.735421 0.006 # 2.757405 0.0176 # 2.779388 0.0436 # 2.801372 0.0848 # 2.823355 0.156 # 2.845338 0.2568 # 2.867322 0.3636 # 2.889305 0.5004 # 2.911288 0.6392 # 2.933272 0.7456 # 2.955256 0.8416 # - 2.977239 0.9092 # 2.999222 0.9604 # 3.021206 0.984 # 3.043189 0.9936 # 3.065172 0.9972 # 3.C87156 0.9988 # 3.109139 0.9996 # 3 131123 0 9998 8 3.153106 1 Crop yield for other vegetables YV(2) CONTINUOUS LINEAR 22 # 2.26 0 # 2.28714 0.0008 # 2.299984 0.0012 # 2.312828 0.0064 # 2.325672 0.0152 # 2.338516 0.0328 # 2.35136 0 0744 # 2.364204 0.14 # 2.377048 0.2492 # 2.389892 0.38 # 2.402736 0.53 # 2.41558 0.6612 # 2.428424 0.7876 # ; Residential Scenario 6.1-20 February 1,1998 i7 _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
Table 6.1.2 Summary of Sampled Distributions Parameter Symbol Dist Type Parameters G Desenpbon 2 41268 2 454112 0 8856 0.9416 2 466956 0.9748 # 2.4798 0.9884 # 2 492644 0.996 # 2.505488 0 9972 # 2.518332 0.9992 # 2.531176 0.99 % # 2.54402 1 Crop yield for fruits YV(3) CONTINUOUS LINEAR 22 # 2.17 0 # 2.194561 0.0012 # , 2.212967 0.'A24 # 2 231374 0.0068 #
, 2.24978 0.018 #
2.268186 0.0436 2.286592 0.0764 # 2.304999 0.138 # 2.323405 0.2136 # 2.341811 0.3272 # 2.360218 0.45 # 2.378624 0.5764 # 2.39703 0.6868 # 2.415437 0.7876 # , G 2.433843 0.868 # 2.452249 0.9248 # 2.470655 0.9604 # 2 489062 0.9808 # 2.507468 0.9916 # 2.525874 0.998 # 2.544281 0.9996 # 2.562687 1 Crop yield for grains YV(4) CONTINUOUS LINEAR 22 # 0.285 0 # 0.2897287 0.0006 # 0.3017949 0.0028 # 0.313861 0.0094 # 0.3259272 0.0214 # 0.3379934 0.0542 # 0.3500596 0.1082 # 0.3621257 0.2022 # 0.3741919 0.3146 # 0.3862581 0.4504 # 0.3980243 0.5916 # 0.4103904 0.7202 # 0.4224566 0.8256 # 0.43452?? 0.903 # 0.4465889 0.551 # Residential Scenario 6.1-21 February 1,1998 G
-um mm -
,. .-m.m -
Table 6.1.2 Summary of Sampled DistribArs Parameter Symbol Dist Type Parameters Desenption 0.4586551 0.9768 # 0.4707212 0.9912 # 0 4827874 0.9958 # 0.4948536 0 999 # 0.5069197 0.9996 # 0.5189859 0.9998 # 0.5310521 1 Crop yield for beef cattle forage, dry DYF(1) BETA 0.3702 0.5238 Crop yield for beef cattle grain, dry DYG(1) NORMAL 0.57818729 0.077651553 Wet / dry conversion factor for leafy DWV(1) GAMMA 2.68 35.08 vegetables Wet / dry conversion factor for other DWV(2) GAMMA 2.68 35.08 vegetables Wet / dry conversion factor for fruits DWVG) GAMMA 2.68 35.08 Wet /ory conversivn factor for beef WF(1) BETA 0.183
- 223 cattle foras, Wet / dry conversion factor for beef DWG(1) LOGNORMAL N -3.80E+00 1.15135 cattle grain Dry matter ingestion rate for beef DOF(1) BETA 1.690L 2.28954 cattle forage Dry matter ingestion rate for poultry DOF(2) BETA 3.48E 03 2.82E-02 forage Dry matter ingestion rate 6r milk cow DOF(3) GAMMA 2.7434341 1.1473046 forage Dry matter ingeston rate for layer DOF(4) BETA 1,19E-02 2.22E-02 hen forage Dry matter ingestion rate for beef DQG(1) BEL 1.69026 2.28954 cattle grain 3 Dry matter ingestion rate for poultry DOG (2) BETA 104E-02 845E 02 grain Dry matter ingestion rate er milk cow DOG (3) NORMAL 1.7135556 0.26199017 grain Dry matter ingestion rate for layer DQG(4) BETA 3.58E 02 6.67E-02 hen grain Dry matter ingestion rate for beef DQH(1) BETA 3.38052 4.57908 cattle hay Dry maNet ingestion rate er milk cow DOH(3) GAMM2 2.7434347 1.4341307 hay Log 10(Kd) L10KdBe NORMAL 2.97 1.4 Log 10(Kd) L10KdC LOGNORMAL N 2.769E-01 7.850E-01 Log 10(Kd) L10KdF NORMAL 0.7 1.4 Log 10(Kd) L10KdNa NORMAL 0.7 1.4 Log 10(Kd) L10KdP NORMAL 1.41 1.4 Log 10(Kd) L10KdS NORMAL 2 1.4 Log 10(Kd) L10KdCl NORMAL 0.7 1.4 Log 10(Kd) L10KdK NORMAL 0.7 1.4 Log 10(Kd) L10KdCa NOPMAL 3.17 1.4 Log 10(Kd) L10KdSc NORMAL 2.2 1.4 Residential Scenario 6.1-22 February 1,1998 i
- Tab!e 6.1.2 Summary of Sampled Distributions
( ( Parameter Description Symbol Dist Type - Parameters Log 10(Kd) L10KdCr NORMAL 2.01 1.2 P
- .og10(Kd) L10KdMn LOGNORMAL N 1.398E-01 6 990E-01 Log 10(Kd) OL10KdFe UNIFORM 0 1 i Log 10(Kd) DL10KdCo UNIFORM - 0 1 Log 10(Kd) L10KdNi NORMAL. 1.57 1.48 Log 10(Kd) L10KdCu NORMAL 2.25 1.4 Log 10(Kd) L10KdZn NORMAL 3.03 1.93 Log 10(Kd) Lit;KdAs NORMAL 2.06 1.4 Log 10(Kd) L10KdSe- NORMAL 2.06 0.25 -
Log 10(Kd) L10KdBr NORMAL 1.75. 1.4 Log 10(Kd) L10KdRb NORMAL 2.31 1.4 Log 10(Kd)- L10KdSr NORMAL 1.5 _ 0.92 Log 10(Kd) L10KdY NORMAL 2.900E+00 1.400E+00 Log 10(Kd) L10KdZr NORMAL 3 38 1.4 Log 10(Kd) L10KdNb NORMAL 2.8 1.4
- Log 10(Kd) L10KdMo NORMAL 1.42 0.75 -
og10(Kd) L10KdTc NORMAL 0.87 1.33 l Log 10(Kd) L10KdRu NORMAL 3.2 1.36 Log 10(Kd) L10KdRh NORMAL 2.2 1.4 Log 10(Kd) L10KdPd NORMAL . 2.27 - . 1.37
- Log 10(Kd) L10KdAg LOGNORMAL N 7.129E-01 5.19CE 01 Log 10(Kd) - L10KdCd NORMAL 1.53 1.3 Log 10(Kd; L10Kdin NORMAL 2.2 1.4 Log 10(Kd) L10KdSn NORMAL ?' 1.4 Log 10(Kd) L10KdSb NORMAL .1.4
[ gy ' Log 10(Kd) Log 10(Kd) L10KdTe L10Kdi NORMAL NORMAL 2.74 0.66' 1.4 O.95 Log 10(Kd)- L10KdCs NORMAL 2.35 1.01
- Log 10(Kd) L10KdBa NORMAL 1.650E+00 3.530E+00 Log 10(Kd) L10KdLa NORMAL 0.7 1.4 Log 10(Kd) L10KdCe NORMAL .1.93 0.43 Log 10(Kd) L10KdPr NORMAL 2.2 1.4 Log 10(Kd) L10KdNd NORMAL 2.2 - 1.4 Log 10(Kd) L10KdPm NORMAL 3.7 1.4 Log 10(Kd) L10KdSm NORMAL 2.97 1.4 Log 10(Kd) L10KdEu NORMAL 2.980E+00 1.740E+00 Log 10(Kd) L10KdGd NORMAL 0.7 1.4 Log 10(Kd) L10KdTb NORMAL 2.2 1.4 Log 10(Kd) L10KdHo NORMAL 2.97 1.4 Log 10(Kd) L10KdW NORMAL 2.2 1.4 Log 10(Kd) L10KdRe NORMAL 1.64 : 1.4
' Log 10(Kd) L10KdOs NORMAL 2.2 1.4 Log 10(Kd) L10Kdir NORMAL 2.2 1.4 Log 10(Kd) L10KdAu NORMAL 2.2 1.4 Log 10(Kd) L10KdHg NORMAL 2.2 1.4 Log 10(Kd) L10KdTI NORMAL 2.2 1.4 Log 10(Kd) L10KdPb NORMAL 3.38 1.2 Log 10(KC) L10KdBi NORMAL 2.65 ,4 Residential Scenario 6.1-23 February 1,1998
(
\
Tabb 6.1.2 Summary of Sampled Distributions ( Parameter Symbol Dist Type Parameters Desenpbon Log 10(Kd) L10KdPo NORMAL 2.26 Log 10(Kd) L10KdRa NORMAL 3 55 0.74 Log 10(Kd) L10KdAc NORMAL 3 24 1.4 Log 10(KC) L10KdTh NORMAL 3.77 1.57 Log 10(Kd) L10KdPa NORMAL 3 31 1.4 Log 10(Kd) L10Kdu NORMAL 2.1 1.36 Log 10(Kd) D!.10KdNp UNIFORM 0 1 Log 10(Kd) L10KdPu NORMAL 2.98 0.82 Log 10(Kd) L10KdAm NORMAL 3.16 1.37
- Log 10(Kd) L10hdCm NORMAL 3.83 0.79 Log 10(Kd) L10KdCf NORMAL 2.2 1.4 Concentrabon ratio BIBe LOGNORMAL N -4.605E+00 9.042E-01 Concentration rabo B1C LOGNORMAL N 3.567E 01 9.042E 01 Concentrabon rabo B1N LOGNORMAL N 3.401E+00 9h42E 01 Concentrabon rabo B1F LOGNORMAL-N 2.813E+00 9.042E-01 Concentration rabo B1Na LOGNORMAl-N 2.604E+00 9N2E-01 Concentra..,:n ratio 31Mg LOCNORMAL-N ;00E+00 9.04 ":-01 Concentrabon rabo 81Si LOGNORMAL-N -1.050E+00 9.042E 01 Concentrabon rabo B1P LOGNORMAL-N 1.253E+00 9.042E-01 Concentration rabo B1S LOGNOR'AAL-N 4.055E 01 9.042E-01 -
Concentrabon rabo 81C1 LOGNORMAL-N 4.248E+00 9.042E 01 Concentrabon rabc B1K LOGNORMAL N 0.000E+00 9.042E-01 Concentrabon rabo B1Ca LOGNORMAL N 1.25?E+00 9.042E 01 Concentrabon rabo B1Sc LOGNORMAL-N 5.1%E+00 9.042E-01 Concentrabon ratio B1Cr LOGNORMAL-N 3.817E+00 7.88SE-01 Concentrabon rabo B1Mn LOGNORMAL-N 1.109E+00 2.028E+00 , Concentrabon rabo B1Fe LOGNORMAL N -5185E+00 1.335E+00 Concentrabon ratie 21Co LOGNORMAL-N -2.430E+00 1.548E+00 Concentration rabo B1Ni LOGNORMAs-N -3.381E+00 1.163E+00 Concentrabon ratio B1Cu LOGNORMAL-N 7.133E 01 9 555E-01 Conantration abo B1Zn LOGNORMAL-N -5.447E-01 9.555E 01 Concentraton ratio B1Ga LOGNORMAL-N -5.521E+00 9.042E-01 Concentrabon rabo B1As LOGNORMAL-N 3.219E+00 9.042E-01 Concentration rabo B1Se LOGNORMAL N -3.689E+00 9.042E-01 Concentration rabo B1Br LOGNORMAL N 4.055E-01 9.042E-01 Concentrabon rubo B1Rb LOGNORMAL-N -2.107E-01 1.281E+00 Concentration ratio B1Sr UX3 NORMAL N 5.878E-01 1.335E+00 Concentrabon ratio B1Y LOGNORMAL N 4.200E+00 9.042E-01 Concentrationifo B1Zr LOGNORMAL N -2.651E+00 6.931E-01 Concentration rabo B1Nb LOGNORMAL N -3.912E+00 9.042E-01 Concentrabon rabo B1Mo LOGNORMAL-N 7.885E-01 1.194E+00 Concentration ratio B1Tc LOGNORMAL-N 2.251E+00 9.042E-01 Concentrabon ratio BIRu LOGNORMAL-N -2.781E+00 1.569E+00 Concentrabon ratio B1Rh LOGNORMAL-N 1.897E+00 9.042E-01 Concentrabon rabo B1Pd LOGNORMAL-N -1.897E+00 9.042E-01 Concentration rabo B1Ag LOGNORMAL N -9.163E-01 9042E 31 Concentrabon ratio B1Cd LOGNORMAL-N -5.978E-01 9.042E-01 Concentrabon ratio Biln LOGNORMAL-N -5.521E+00 9.042E-01 Residential Scenario 6.1-24 February 1,1998 i
Table 6.1.2 Summary of Sampled Distribu%ns O Parameter Symbol Dist Type Parameters Desenpbon Concentrat on ratio B1Sn LOGNORMAL-N 3 507E+00 9042E 01 Cone.entrat on rabo BISb LOGNORMAL-N 1.609E+00 9.042E 01 Concentration rabo B1Te LOGNORMAL-N 3 689E+00 9 042E 01 Concentration rabo B11 LOGNORMAL N 1.833E+00 1253E+00 Concentraton ratio B1Cs LOGNORMAL N -3194E+00 1.253P n0 Concentrabon rato B1Ba i OGNORMAL N 3.244E+00 1.065c.+00 Concentrabon rabo Bila LOGWORMAL N -4.605E+00 9.042ti 01 Concentrabon rato BICe LOGNORMAL-N 3.863E+00 1.459E+00 Concentration rato B1Pr LOGNORMAL-N -4 605E+00 9.042E-01 ConcMtratoq rabc. BIND LOGNORMAL-N -4 605E+00 9.042E 01 Concentration rabo B1Pm LOGNORMAL-N -4.605E+00 9.042E 01 Concentration rabo BISm LOGNORMAL-N -4.605E+00 9.042E 01 Concentration rabo BIEu LOGNORMAL N -4.605E+00 9.042E-01 Concentraton rabo BIGd LOGNORMAL-N -4.605E+00 9.042E 01 Concentrabon rato BITb LOGNORMALN t605E+00 9 042E-01 Concentrabon rabo B1Dy LOGNORMAL-N -4. J5E+00 9.042E41 Concentrabon rabo B1Ho LOGNORMAL-N -4.60SE+00 9.042E-01 Concent :: ion rabo B1Er LOGNORMAL-N -4.605E+00 9.042E 01 Concentration ratio B1Hf LOGNORMAL-N -5 655E+00 9.042E-01 Concentrabon rato BITa LOGNORMAL N -4.605E+00 9.042E 01
' oncentration rabo B1W LOGNORMAL-N 3.10iE+00 9.042E 01 Concentrabon ratio BIRe LOGNORMAL-N 4.053E 01 9.042E 01 Concentration ratio B10s LOGNORMAL-N -4.200E+00 9.042E 01 Concentration rato B1lt LOGNORMAL-N 2.900E+00 9.042E 01 S Concentrationrabo Concentraton ratio Concentrabon rato Concentrabon ratio B1Au B1Hg BITI BiPb LOGNORMAL-N LOGNORMAL-N LOGNORMAL N LOGNORMAL-N 9.163E-01
-1.054E-01
-5.521E+00 9.042E 01 9.042E 01 9.042E 01
-3.101E+00 9.042E-01 Concentration rabo B1Bi LOGNORMAL-N 3.352E+00 9.042E 01 Concentrabon ratio B1Po LOGNORMAL-N 5 991E+00 9.042E 01 Concentration ratio BIRa LOGNORMAL-N -4.200E+00 9.042E-01 Concentration rabo B1Ac LOGNORMAL-N -5.655E+00 9.042E 01 Concentrabe, ratio B1Th LOGNORMAL-N 7 070E+03 9.042E 01 Concentrabon rabo S1Pa LOGNORM/ -N 5.991E+00 9.042E-01 Concentrabon rabo B1U L'ENORMAL-N -4.768E+00 9.042E 01 Concentation rat B1Np LO3 NORMAL-N 9.531E-02 1.589E+00 Concentrabon ratio B1Pu LOGNORMAL-N -7.706E+00 9.042E-01 Concentration ratio B1Am LOGNORMAL N 5.203E+00 9.042E 01 Concentraton ratio B1Cm LOGNORMAL N -7.070E+00 9.042E-01 Concentraton rabo B1Cf LOGNORMAL N 4.605E+00 9.0'2E-01 Concentration rato B2Be LOGNORMAL N -6.502E+00 9.042E 01 Concentrabon rato B2C LOGNORMAL-N -3.567E-01 9042E 01 Concentration ratio B2N LOGNORMAL-N 3 401E+00 9M2E-01 Concentration rt' o 92F LOGNORMAL-N -5.116E+00 9.042E-01 Concentraton ratio WB2Na LOGNORMAL-N -5.382E+00 1.411E+00
)
Concentrabon ratio B2Mg LOGNC' MAL-N 5.978E-01 9.042E-01 ConcentrJon ratio B2Si LOGNORMAL-N -2.659E+00 9.042E 01 Concentraton rato B2P LOGNORMAL-N 1.253E+00 9.042E-01 Residential Scenario 6.1-25 February 1,1998
Table 6.1.2 Summary of Sampled Distributions Parameter Symbol Dist Type Parameters Desenpbon Concentrabon rato B2S LOGNORMAL-N 4 055E 01 9 042E-01 Concentrabon rabo B2Ci LOGNORMAL N 4.248E+00 9.042E 01 Ccacentrabon rabo B2K LOGNORMAL h -5.978E 01 9.042E 01 Concentration rato B2Ca LOGNORMAL-N 1.050E+00 9.042E-01 Concentrabon rabo B2Sc LOGNORMAL N -6.908E+00 9.042E-01 Concentrabon rato WB2Cr LOGNORMAL-N -4.343E+00 6.931E 01 Concentrabon rabo WB2Mn LOGNORMAL N 2,120E+00 1.589E+00 Concentrabon rato WB2Fe LOGNORMAL N 7.775E+00 1.253E+00 Concentrabon ratio WB2Co LOGNORMAL-N -4.200E+00 1.194E+00 Concentrabon rabo WB2Ni LOGNORMAL-N 3.862E+00 9.163E-01 Concentrabon rabo WB2Cu LOGNORMAL-N -3.147E+00 2.303E+00 Concentration ratio WB22n LOGNORMAL N -2.207E+00 1.361E+00 Concentrabon rabo B2Ga LOGNORMAL-N 7.824E+00 9.042E-01 rnneentration rabo 92As LOGNORMAL N -5.116E+00 9.042E-01 Cancentration rabo 82Se LOGNORMAL-N 3.689E+00 9.042E-01
.ancentrabon ratio B2Br LOGNORMAL-N 4.05561 9.042E-01 Cor.centrabon rabo B2Rb LOGNORMAL-N 2.659E+00 9.042E-01 Concentration ratio WB2Sr LOGNORM&N -2.590E+00 1.335E+00 Concentrabon rabo B2Y LOGNORMAL-N 5.116E+00 9.042E-01 Cor.centrabon ratio WB2Zr LOGNORMAL-N 7.169E+00 2.251E+00 Concentration ratio B2Nb LOGNORMAL-N -5.298E+00 9.042E 01 Concentraton rabo B2Mo LOGNORMAL-N -2.813E+00 9.042E-01 Concentration ratio B2Tc LOGNORMAL-N 4.055E-01 9.042E-01 Concentration rabo WB2Ru LOGNORMAL-N -6 571E+00 1.589E+00 Concentrabon ratio B2Rh LOGNORMAL-N -3.219E+00 9.042E-01 Concentrabon rato B2Pd LOGNORMAL-N -3.219E+00 9.N2E-01 Concentraboa rabo B2Ag LOGNORMAL-N -2.303E+00 9.042E-01 Concentration rabo B2Cd LOGNORMAL-N 1.897E+00 9.042E-01 Concentrabon rabo 82in LOGNORMAL N 7.824E+00 9.042E-01 Concentrabon rato B2Sn LOGNORMAL N 5.116E+00 9.042E-01 l Concentrabon ratio B2Sb LOGNOPMAL-N -3.507E+00 9.042E-01 Concentration rabo B2Te LOGNORMAL-N 5.521E+00 9.042E-01 Concentration rabo WB21 LOGNORMA'.-N 5.404E+00 1.589E+00 l Concentrabon rabo WB2Cs LOGNORMAL-N -5.298E+00 1.411E+00 l Concentrabon rabo WB2Ba LOGNORM&N -6.645E+00 1.131E+00 Concentration rabo 82La LOGNORMAL-N -5.521E+00 9.042E 01
! Concentraton rabo WB2Ce LOGNORMAL-N 7.222E+00 1.825E+00 Concentrabon rabo B2Pr LOGNORMAL N 5.521E+00 9.042E-01 Concenvation ratio B2Nd LOGNORMAL-N -5.521E+00 9.042E-01 Concentrabon rato B2Pm LOGNORMAL N 5.521E+00 9.042E 01 Concentrabcn rato B2Sm LOGNORMAL-N 5.521E+00 9.042E-01 Concentration rabo B2Eu LOGNORM L N -5.521C+00 9.042E-01 l Concentration ratio B2Gd LOGNORMAL-N -5 521E+00 9.042E-01 Concentration rabo B2Tb LOGNORMAL N 5.521E+00 9.042E 01 Concentration rato B2Dy LOGNORMAL-N -5.521E+00 9.042E-01 Concentration rato B2Ho LOGNORMEN -5 521E+00 9.042E-01 Concentrabon rato B2Er LOGNORMAL-N -5.521E+00 9.042E-01 Concentrabon rabo B2Hf LOGNORMAL-N -7.070E+00 9.042E-01 Residential Scenario 6.1-26 February 1,1998 d l
Tabb 6.1.2 Summary of Sampled Distnbutions Parameter Symbol Dist Type Parameters
- j Descriphon V Concentrabon rabo B2Ta LOGNORMAL-N 5991E+00 9 042E 01 Concentrabon rato B2W LOGNORMAL N -4.605E+00 9042E 01 Concentration ratio B2Re LOGNORMAL-N 1.050E+00 9.042E 01 Concentrabon ratio B20s LOGNORMAL N -5 655E+00 9.042E 01 Concentration rabo B2tr LOGNORMAL-N 4.200E+00 9.042E 01 Ccocentration rabo B2Au LOGNORMAL N 2.303E+00 9.042E 01 Concentrabon rabo B2Hg LOGNORMAlot 1.609E+00 9.042E 01 Conceatrabon rabo B2TI LOGNORMAL-N 7.824E+00 9.042E-01 Concentrabon rabo B2Pb LOGNORMAL N -4.711 E+00 9.042E-01 Concentration rata B2Bi LOGNORMAbN 5.298E+00 9.042E-01 Concentrabon rabo B2Po LOGNORMAL N 7.824E+00 9.042E-01 Concentrabon rabo B2Ra LOGNORMAL N -6.502E+00 9.042E C1 Concentration rato B2Ac LOGNORMAL N 7.958E+00 9.042E-01 Concentration rabo B2Th LOGNORMAL N 9.373E+00 9.042E 01 Concentration rabo B2r'a LOGNORMAL-N -8.294E+00 9.042E 01 Concentrabon ratio B2U LOGNORMAL N -5.521E+00 9.042E 01 Concentrabon rabo WB2Np LOGNORMAL-N 4.v13E+00 1.099L r00 Concentrabon rabo B2Pu ' OGNORMAL N
. 1.001E+01 9,04'E 01 Concentration rabo B2Am LOGNORMAL-N 8.294E+00 9.042E 01 Concentrabon ratio B2Cm LOGNORMAL-N -1.111E+01 9.042E 01 Concentration ratio B2Cf LOGNORMAL-N -4.605E+00 9.042E-01 Concentration rabo B3Be LOGNORMAL N 6.502E+00 9.042E 01 Concentration rato B3C LOGNORMAL-N 3.567E 01 9.042E 01 Concentration rato B3N LOGNORMAL N 3 401E+0C 9.042E 01 /~N Concentrabon ratio B3, LOGNORMAL-N 5.116E+00 9.042E-01
( Concentrabon rabo WB3Na LOGNORMAL N 5.382E+00 1.411E+00
'x Concentraton rato B3Mg LOGNORMAL N -b.978E-01 9.042E-01 Concentration rato B3Si LOGNORMAL-N -2.659E+00 9.042E 01 ConcenTabon ratio B3P LOGNORMAbN 1.253E+00 9.042E 01 Concentrabon rabo B3S LOGNORMAL N 4.055E-01 9.C 3-01 Concentration rabo B3Cl LOGNORMAL-N 4.248E+00 9.042E-01 Concentrabon rabo B3K LOGNORMAL-N -5.978E-01 9.042E-01 Concentration ratio B3Ca LOGNORMALN 1.050E+00 9.042E-01 Concentration ratio B3Sc LOGNORMAL-N -6.908E+00 9.042E-01 Concentration ratio WB3Cr LOGNORMAL-N 4.043E+00 6.931E 01 Concentrabon ratio WB3Mn LOGNORMAL-N 2.120E+00 1.589E+00 Concentrabon rato WB3Fe LOGNORMAL-N 7.775E+00 1.253E+00 Concentraton rabo WB3Co LOGNORMAL-N -4.200E+00 1.194E+00 Concentrabon rato WB3Ni LOGNORMAL N 3.863E+00 9.163E 01 Concentration rabo WD3Cu LOGNORMAL-N -3.147E+00 2.303E+00 Concentration rabo WB3Zn LOGNORMAL-N 2.207E+00 1.361E+00 Concentrabon ratio B3Ga LOGNORMAL-N -7.824E+00 9.042E-01 Concentration rato B3As LOGNORMAL-N 5.116E+00 9.042E-01 Concentration rabo B3Se LOGNORMAL-N -3.689E+00 9.042E 01 Concentration rato B3Br LOGNORMAL-N 4.055E-01 9.042E-01 Concentraton rabo B3Rb LOGNORMAL-N -2.659E+00 9.042E 01 Concentration rabo WB3Sr LOGNORMAbN 2.590E+00 1.335E+00 Concentration ratio B3Y LOGNORMAL N -5.116E+00 9.042E 01 p Residential Scenario 6.1-27 February 1,1998 b
Table 6.1.2 Summary of Sarnpied Distrit.utions Parameter Symbol Dist Type Parameters Descripton Concentraton rato WB32r LOGNORMAL-N 7.169E+00 2 251E+00 Concentraton rabo B3Nb LOGNORMAL-N 5.298E+00 9.042E 01 Concentrabon rato B3Mo LOGNORMAL N 2.813E+00 9.042E 01 Concentraton rabo B3Tc LOGNORMALN 4 055E-01 9 042E-Ci Concentabon rabo WB3Ru LOGNORMAL N 6.571E+00 1.589E+00 Concentraton rabo B3Rh LOGNORMAL N 3.219E+00 9.042E-01 Concentrabon ratio B3Pd LOGNORMAL-N 3 219E+00 9.042E-01 Concentrabon rabo B3Ag LOGNORMAL N 2.303E+00 9.042E-01 Concentration rato B3Cd LOGNORMAL N -1.897E+00 9.042E-01 Concentrabon rabo B3ln LOGNORMAL N -7.824E+00 9.042E 01 Cnncentration rabo B3Sn LOGNORMAL-N -5.116E+00 9.042E-01 Concentrabon rabo B3Sb LOGNORMAL N 3.507E+00 9.042E-01 Concentration ratio B3Te LOGNG.NL N 5.521E+00 9.042E-01 Concentration ratio WB31 LOGNORMAL N 5.404E+00 1.589E+00 Concentrabon ratio WB3Cs LOGNORMAL N -5.298E+00 1.411E+00 Concentration rabo W93Ba LOGNORMAL-N -6.645E+00 1.131E+00 Concentration ram 83La LOGNORMAL : -5.521E+00 9.042E-01 Concentrabon rato WB3Ce LOGNORMAL-N 7.222E+00 1.825E+00 Concentrabon ratio B3Pr LOGNORMAL-N 5.521E+00 9.042E 01 Concentration rabo B3Nd LOGNORMAL-N 5.521E+00 9.042E-01 Concentration rabo B3Pm LOGNORMAL-N -5.521E+00 9.042E-01 Concentration ratio B3Sm LOGNORMAL N -5.521E+00 9.042E 01 Concentrabon ratio B3Eu LOGNORMAL-N -5.521E+00 9.042E-01 Concentration rabo B3Gd LOGNORMAL-N 5.521E+00 9.042E-01 Concentrabon rabo B3Tb LOGNCRMAL N -5.521E+00 9.042E-01 Concentration rato B3Dy LOGNORMAL N 5.521E+00 9.042E 01 Concentraton rabo B3Ho LOGNORMAL-N 5.521E+00 9.042E-01 Concentrabon rato B3Er LOGNORMAL N 5.521E+00 9.042E-01 Concentration ratio B3Hf LOGNORMAL N 7.070E+00 9.042E-01 Concentraton rabo B3Ta LOGNORMAL N -5.991E+00 9.042E-01 Concentration rabo B3W LOGNORMAL N -4.605E+00 9.042E 01 Concentration rato B3ke LOGNORMAL-N ' L0E+00 9.042E-01 Co.1centrabon rabo B30s LOGNORM" N -5.655E+00 9.042E-01 Concentration rabo 83lr LOGNORMAL-N -4.200E+00 9.042E-01 Concentration rabo 83Au LOGNORMAL N -2.303E+00 9.042E 01 Concentration rabo B3Hg LOGNORMAL N 1.609D00 9.042E-01 Concentrabon rabo B3Tl LOGNORMAL N -7.824E+00 9.042E-01 Concentration rabo B3PD LOGNORMAL-N -4.711E+00 9.042E-01 Concentration ratio B3Bi LOGNORMAL-N 5.298E+00 9.042E-01 Concentration rabo B3Po LOGNORMAL N 7.824E+00 9.042E 01 Concentration ratio B3Ra LOGNORMAL-N -6.502E+00 9.042E.01 Concentration ratio B3Ac LOGNORMAL N -7.958E+00 9.042E-01 Concentration rato B3Th LOGNORMAL-N -9.373E+00 9.042E 01 Concentration rabo B3Pa LOGNORMAL N -8.294E+00 9.042E 01 Concentration rabo B3U LOGNORMAL-N -5.521E+00 9.042E-0t Concentrabon ratio WB3Np LOGNORMAL-N 2.813E+00 1.099E+00 Concentration re*2o B3Pu LOGNORMAL-N -1.001E+01 9.042E 01 Concentrabon ratio B3Am LOGNORMAL-N -8.294E+00 9.042E-01 Residential Scenario 6.1-26 February 1,1998
Table 6.1.2 Summary of Sampbd Distributions rN Parameter Symbol Dist Type Parametes V) i DesenpGon Concentrabon rato foncentrabon rabo B3Cm B3Cf LOGNORMAL N LOGNORMAL N 1.111E+01
-4.605E+00 9.042E-01 9.042E 01 Concentrabon rabo B4Be LOGNORMAL N 6.502E+00 9.042E 01 ConcentrMion rabo B4C LOGNORMAL N 3.567E 01 9.042E 01 Concentrabon rabo B4N LOGNORMAL N 3 401E+00 9 042E-01 Concentrabon rat B4F LOGNORMAL N 4.116E+00 9.042E 01 Conantrabon rabo WB4Na LOGNORMAL N 5.382E+00 1.411E+00 Concentrabon ratio B4Mg LOGNORMAL N -5.978E-01 9 042E-01 Concentrabon rabo B4Si LOGNORMAL-N 2.659E+00 9.042E-01 Concentration rabo B4P LOGNORMAL N 1.253E+00 9.042E-01 Concentrabon rabo B4S LOGNORMAL-N 4.055E-01 9.042E 01 Concentration ratio B4Cl LOGNORMAL-N 4.248E+00 9.042E-01 Concentrabon rabo B4K LOGNORMAL-N 5.978E 01 9.042E-01 Concentrabon rabo B4Ca LOGNORMAL-N 1.050E+00 9.042E-01 Concentrabon sabo eP: LOGNORMAL-N -6.908E+00 9.042E 01 Concentration rato - WB4Cr LOGNORMAL N -4.343E+00 6.931E 01 Concentration ratio WB4Mn LC4 NORMAL-N 2.120E+00 1.589E+00 Concentration ratio WB4Fe LOGNORMAL-N 7.775E+00 1.253E+00 Concentration rabo WB4Co LOGNORMAL-N 4.200E+00 1.194E+00 Concentratm rato WB4Ni LOGNORMAL-N 3.863E+00 9.163E 01 Concentrabon rabo WB4Cu LOGNORMAL-N 3.147E+00 2.303E+00 Concentrabon rabo WB42n LOGNORMAL-N 2.207E+00 1.361E+00 Concentrabon rabo B4Ga LOGNORMAL N 7.824E+00 9.042E-01 Concentrabon rabo B4As LOGNORMAL N 5.116E+00 9.042E-01
]
j Concentrabon rabo Concentrabon rabo B%e B4Br LOGNORMAL-N LOGNORMAL-N 3.689E+00 4.055E 01 9.042E-01 9.042E 01
.V Concentration rabo B4Rb LOGNORMAL-N - 2.659E+00 9.042E 01 Concentrabon rabo WB4Sr LOGNORVAL N 2.590E+00 1.335E+00 Concentrabon rabo B4Y LOGNORMAL-N 5.116E+00 9.042E-01 ,
Concentrabon ratio WB4Zr LOGNORMAL N 7,169E+00 2.251E+00 Concentrabon rabo B4Nb LOGNORMAL N 5.298E+W 9.042E 01 Concentrayon rabo B4Mo LOGNORMAL N 2.813E+00 9.042E 01 0:ncentrabon ratio 84Tc LOGNORMAL N 4.055E-01 9.042E-01 Concentrabon rabo WB4Ru LOGNORMAL-N -6.571E+00 1.569E+00 Concentrabon ratio B4Rh LOGNORMAL-N 3.219E+00 9.042E-01 Concentrabon rabo 04Pd LOGNORMAL N 3.219E+00 9.042E-01 Concentration ratio B4Ag LOGNORMAL-N 2.303E+00 9.042E 01 Concentration ratio B4Cd LOGNORMAL-N 1.897E+00 9.042E 01 Concentration ratio B4ln LOGNORMAL-N 7824E+00 9 042E-01 Concentrabon rabn B4Sn LOGNORMAL-N -5.116E+00 9.042E 01 Concentrabon rabo B4Sb LOGNORMAL N 3.507E+00 9.042E-01 Concentrabon rato B4Te LOGNORMAL-N -5.521E+00 9.042E 01 Concentrabon ratio WB41 LOGNORMAL N 5.404E+00 1.589E+00 Concentrabon ratio WB4Cs LOGNORMAL-N 5.290E+00 1.411E+00 Concentration rabo WB4Ba LOGNORMAL N -6.645E+00 1.131E+00 Concentration ratio B4La LOGNORM/1.-N -5 621E+00 9.042E-01 Concentrabon rabo WB4Ce LOGNORMAL-N -7.222E+00 1.825E+00 ConcentratM ratio ' B4Pr LOGNORMAL N -5.521E+00 9.042E-01 A Residential Scenario 6.1-29 February 1,1998
\v}
Table 6.12 Summary of Sampled Distributions Parameter Symbo! D:st Type Parameters Desenption Concentrabon ratio B4Nd LOGNORMAL-N -5521E+00 9 042E-01 Concentrabon rabo B4Pm LOGNORMAL-N 5.521E+00 9.042E 01 Concentrabon rabo B4Sm LOGNORMALN 5 521E+00 9.042E-01 Concentrabon rato B4Eu LOGNORMAL-N 5.521E+00 9.042E-01 Concentrabon ratio B4Gd LOGNORMAL-N 5521E+00 9.042E 01 Concentrabon reo B4Tb LOGNORMAL-N 5.521E+00 9.042E-01 Concentration ratio B4Dy LOGNORMAL-N 5 521E+00 9.042E-01 Concentration rabo B4Ho LOGNORMAL-N 5.521E+00 9.042E-01 Concentration raho B4Er LOGNORMAL-N 5 521E+00 9.042E-01 l Concentrabon ratio B4Hf LOGNORMAL N 7.070E+00 9.042E 01 l Concentrabon ratio B4Ta LOGNORMAL-N 5 991E+00 9.042E-01 Concentrabon rabo B4W LOGNORMAL-N -4.605E+00 9 042E-01 Concentrabon rabo B4Re LOGNORMAL-N 1.050E+00 9.042E 01 Concentration ratio B40s LOGNORMAL-N 5.655E+00 9 042E-01 : Concentrabon rabo B41r LOGNORMAL N -4.200E+00 9.042E-01 Concentrabon ratio EMAu LOGNORMAL-N 2.<d3E+00 9.042E 01 Concentrabon rabo B4Hg LOGNORMAL-N 1.609E+00 9.042E-01 Concentrabon rabo B4Tl LOGNORMAL-N 7.824E+00 9.042E-01 Crncentrabon rabo B4Pb LOGNORMAL N -4.711 E+00 9.042E-01 Concentrabon rabo B4Bi LOGNORMAL ;4 -5.298E+00 9.042E-01 Concentration rabo B4Po LOGNORMAL-N 7.824E+00 9 042E 01 Concentrabon rabo B4Ra LOGNORMAL-N -6.502E+00 9.042E-01 Concentration rabo B4Ac LOGNORMAL-N 7.958E+00 9.042E-01 Concentrabon rabo B4Th LOGNORMAL N 9.373E+00 9.042E 01 Concentrabon rabo B4Pa LOGNORMAL N -8.294E+00 9.042E-01 Concentraton rabo B4U LOGNORMAL N 5.52IE+00 9.042E 01 Concentrabon ratio WB4Np LOGNORMAL N -2.813E+00 1.099E+00 Concentrabon reio B4Pu LOGNORMAL-N 1.001E+01 9.042E-01 Concentration ratio B4Am LOGNORMAL N -8.294E+00 9.042E-01 Concentrabon ratio B4Cm LOGNORMAL N 1.111E+01 9.042E-01 Concentraton ratio B4Cf LOGNORMAL N -4.605E+00 9.042E-01 9
j 6.2 Parameter Sample Distributions O The results of the parameter sampling are illustrated in figures 6.2.1 through 6.2.28. These figures show the
, cumulative frequency of the sampled parameter values and the default value from NUREG/CR 5512 Volume 1 for wmparison. Two of the parameters, partition coefficient and vegetation concentranon factor, have a distribution for each element resulting in 69 distributions for each of those parameters. Those distributions are summarized in table
- format (table 6.1.1) with example figures provided in this section.
1 0.9
> 0.8 '
O.7 l l [ 0.6 ; E 0.5 '
. 0.4 ,
'l 0.3 u 02 0.1
' [
O O $0 100 150 200 250 300 350 i_ Thickness of Unsaturated Zone (m) f _ Figure 6.2.1 Cumuladve Frequency of Sampled H2 values (NUREG/CR 5512 v.1 default shown _ as vertical dashed line). f w 1 _. 0.9 ' O.8 ,' O.7 l
~ 0.6 j 0% l I !4 g 0.3
{ 0.2 l 0.1 l 0 0.25 0.3 0.35 0.4 0.45 0.5 0.55 Porosity of Surface Soll Figure 6.2.2 Cumulative Frequency of Sampled N1 values (NUREGICR 5512 v.1 default shown as vertical dashed line). 'h'C/ ResidentialScenario 6.2 1 January 30,1998
1 0.9 0.7 E cr 0.6 I> 0.5 50.4
?
E 0.3 o 0.2 0.1 0
- 0.4 0.6 0.8 1 Saturation Ratio for the Surface Soll Layer Figure 6.2.3 Cumulative Frequency of Sampled 11 values (NUREG/CR 5512 v.1 default shown as verticaldashed line).
r---- - _ _ _ _ _ ~_ 1 f 0.9 , - l * !
' j 0.8 ,
l
, [ 0.7
. l 1
, . * . i l Er* 0.6
- f f >0.5 i
4 I y 0.4 i s ' l
! o3 ,
0.2 . 0.1 e f f 4
\ * +
0 : 0.1 0.2 0.3 0.4 0.5 0.6 Infiltration Rate (m/y) l e Figure 6.2.4 Cumulative Frequency of Sampled IR values (NUREGICR 5512 v.1 default shown as verticaldashed line). Residential Scenario 6.2 2 January 30,1998
. . _ _ . . _ _ . . . _ _ . _ _ _ _ . . _ . . _ . . _ . _ . . _ . _ _ . _ . _ . ~ . - - _ _ . _ . _ . _ . _ _ . _ . _ . _ . _
1 0.9 *
' i 0.8 i-4 0.7 ,'
06 e
- 0.5 i 0.4 e
., 0.3 .
- 0 0.2 i .
. 0.1 .
i 0 170 190 210 230- 250 270 Soil Areal Density of Surface Plow Layer (kg/m') I Figure 6.2.5 Cumulative Frequency of Sampled Ps values (NUREG/CR 5512 v,1 default shown as vertical dashed line).
\
, 1 _ h, 0.9 l O.8 l
- 0.7 l 1 , I i 0.6 ,
I 6 . 0.5 I
- 0.4 ,
0.3 l u ' 0.2 , I i- 0.1 . 0 1.2 1.3 1.4 1.5 1.6 1.7 q Surface Soil Density (g/mL) _.J Figure 6.2.6 Cumulative Frequency of Sampled p Values (NUREG/Cr.; 5512 V.1 default shown as verticaldashedline) Residential Scenario 6.2 - 3 January 30,1998 J i
- + , _ m -
0.9 0.8 * , h0.7 ' s ' { 0.6
- O.5 I
j 0.4 ' s u [ 0.3 l 0.2 , 0.1 l 0' O.0 E+ 00 2.0 E- 0 5 4.0E 05 6.0 E- 05 8.0 E- 05 1.0E-0* 1.2 E- 04 Air Dust-Loading Outdoors (g/m3) 9 1 0.9 e 0.8 ' h0.7 f0.6 ,
' O.5 j 0.4 o
E 0.3 l u , 0.2 , 0.1 l 0 J 0.0 E+ 00 1.0 E- 0 5 2.0 E- 05 3.0 E- 05 4.0 E- 0 5 5.0 E- 05 6.0 E- 05 Air Dust Loading Indoors (g/m*) Figure 6.2.7 Cumulative Frequency of a) sampled CD0 values and b) resulting CDI values l (NUREGICR 5512 v.1 default shown as vertical dashed line). I
. . . . . - . -. . , . . . . . . . - - . . - ~ . - - . . . _ . _ _ _ - _ . . _ . - . - . . . - - . . . . ~ . .
1------....=.._..- _,
-l
! b (s 0.9 i i i 08 . , I 0.7 l i 06 l j' . 05- l 04 l i
'o 0.3 l
' 02 , 0.1 l $ 0 1.0E-04 2.0E 04 3.0E 04 4.0E 04- 5.0E 04 6.0E-04 7.0E-04 8 } Air Dust Loading Gardening (g/m ) F4ure 6.2.8 Cumulative Frequency Of Sampled CDG values (NUREG/CR 5512 v.1 default shown i as verticaldashed line). 1 k 1 . 0.9 l 0.8 . 1 . 4, 5 0.7 i
~
0.6
- i 0.5 ,'
O.4 r O.3 o . 1 0.2 . 0.1 . 0 0 0.1 0.2 0.3 0.4 2 Floor Dust Loading (gtm8 ). Figure 6.2.9 Cumulative Frequency of Sampled Pd values (NUREG/CR 5512 v.1 default shown as vertical dashed line). 4 " 4 a
1 1 0.9 . ; e O.8 *
? 0.7 E ,
3 0 .6 i F : g 0.5 e a 0.4 .' 3 m . u l 0.3
- 0.2 '
0.1 l 0 0,0 E+ 00 2.0 E- 0 5 4.0E 05 6.0 E- 0 5 8.0E 05 Resuspension Factor for Indoor Dust ..n) Figure 6.2.10 Cumulative Frequency of Sampled RFr values (NUREG/CR 5512 v.1 default shown as vertical dashed line). O l l O
. . _ _ . . . . . _ . _ _ _ ~ . . . . . _ _ . . _ . _ _ _ . _ . . - _ _ _ _ _ _ . . _ . . . . . . . . . . _ . . _ _ _ _ . _ _ .
8 1
-1 , ,
0.9 .
; l \
l
.l 0.8 ,
! l 0.7 l i-0.6 , ;
. 1 0.4 l O.3 l 0.2 0.1 .
0 O.0001 0.001 0 01 0.1 1 Concentra*lon Factor Leafy Be [ Figure 6.2.18 Cumulative Frequency of Sampled Bjv for Be in Leafy Vegetables (dashed vertical line = NUREG/CR 5512 Volume 1 default) ) ^ 1 0.9 ' a 4 0.8 . , x M 0.7 ' I 0 ; E .8
- 0.5 e e
f 0.4 4 m l 0.3 , u 0.2 . 4 0.1 . 8 3 0.01 0,1 1 10 100
- Concentration Factor - Leafy - Sr Figure 6.2.19 Cumulative Frequency of Sampled Bjv for Sr in Leafy Vegetables (dashed vertical line = NUREG/CR 5512 Volume 1 default)
ResidentialScenario 62 9 January 30,1998 i N d
- l 4
1 0.9 0.8 [0.7 . S 0 .6 . E . j 0.5 . fs 0.4 e g 0.3 . u 0.2 l 0.1 l 0 _ 0.0001 0.001 0.01 0.1 1 ij Concentration Factor - Leafy - Cs Figure 6.2.20 Cumulative Frequency of Sampled Bjv for Cs in Leafy Vegetables (dashed stical line = NUREG/CR 5512 Volume 1 default) I 1 _ i 0.9 l l 0.8 ! f0.7 l h0.6 l j 2 . ! j 0.5 ,
' j o' ,
l g 0.3 , I 0.2 l 0.1 . 0 0.0001 0.001 0.01 0.1 1 L l Concentration Factor Leafy U j
! _ . . . _ _ _ _ - - _ . _ _ _ _ . . . _ _ . _ _ _ _ _ _ I Figure 6.2.21 Cumulative Frequency of Sampled Bjv for U in Leafy Vegetables (dashed vertical line
= NUREG/CR 5512 Volume 1 default)
Residential Scenario 6.2 21 January 30,1998
0. l 0.8
- h0.7 '
g0.6 i
' O.5 i 0.4 f,0.3 U
0.2 0.1 . l 0 O1 0.2 0.3 0.4 0.5 0.6 Interception Fraction for Leafy Vegetables Figure 6.2.22 Cumulative Frequency of Sampled r, for Leafy Vegetables (dashed vertical line = NUREG/CR 5512 Volume 1 default) 1 0,9 . 0.8
- 0.7 0.6 .
6 0.5 . t .
- 0.4 .
1 . g 0.3 . o 0.2 0.1 0 O C.05 0.1 0.15 0.2 0.25 0.3 0.35' Wet / Dry conversion Factor for Leafy Vegatables Figure 6.2.23 Cumulative Frequency of Sampled Wv for Leafy Vegetables (dashed verticalline = NUREG/CR 5512 Volume i default) O ResidentialScenario 6.2 22 January 30,1998
1 0.9 0.8 i O.7 , 0.6 I l f 0.5 i 30.4 3 i E 0.3 e u 0.2 i 0.1 i 0 0.15 0.2 0.25 0.3 0.35 Wet / Dry Conversion Factor for Beef t attle Forage Figure 6.2.24 Cumulative Frequency of Sampled Wf for Forage (dashed verticalline = NUREG/CR 5512 Volume 1 default) I ! 0.9 0.6 l [0.7 l l h0.6 l 0.5 I !
% 0.4 3 .
E 0.3 , 0.2 , 0.1 l 0 -
- 1. E- 01 1.E+ 0 0 1.E+ 01 1.E+ O2 1.E+ 0 3 1.E+ 04 1.E+ 0 5 1.E+ 06 Partition Coefficient Cs (mUg)
Figure 6.2.25 Cumulative Frequency of Sampled Kd for Cs (dashed verticalline = NUREGICR 5512 Volume 1 default) Residential Scenario 6.2 - 23 January 30,1998
1 0.9 0.8 i 5 0.7 . 0.6 . S 0.5 O.4 0.3 u 0.2 0.1 , 0
- E+ 01 1.E+ O2 1.E+ 03 . E+ 0 4 1.E+ 0 5 1J'06 PartitionCoefficient Ra (mUg)
Figure 6.2.26 Cumulative Frequency of Simpled Kd for Ra (dashed vedcal line = NUREG/CR 5512 Volume 1 default) l O 0.9 i 0.8 ' O.7 '- E"
- 0.5 f:: :
l 0.3
$ 0.2
- 0.1 .
0 ! 1.E- 0 3 1.E- 01 1.E+ 01 1.E+ 03 1.E+ 0 5 Partition Coefficient Ni (mUg) Figure 6.2.27 Cumulative Frequency of Sampled Kd for Ni(dashed verticalline = NUREG/CR 5512 Volume 1 default)- Residential Scenario 6.2 24 January 30,1998
1 e l () .9 () . 8 . h (1.7 , l I . g 0.6 j 0.5 ,' 50.4 3 e 0.3 e u 0.2 . 0.1 , O M 1.E- 0 2 1.E+ 00 1.E+O2 1,E+ 04 1 E+ 06 l l Partition Coemelent.U (mUg) Figure 6.2.28 Cumulative Frequency of Samp',ed Kd for U (dashed verticalline = NUREG/CR 5512 Volume 1 default) O i l Residential Scenario 6.2 25 January 30,1998
0.9 0.8 , 0.7 O.6 S 0.5 ' O.4 0.3 0 0.2 a 0.1 ,' O 1.9 2.1 z.3 2.5 27 2.9 3.1 Crop Yield for Leafy Vegetables (kg/m . Figure 6.2.11 Cum"lative Frequency of Sampled Yv values (NUREG/CR 5512 v.i default shown as verticaldashedline). 1
- r 0.9 0.8 '
{0.7 l 0.6 l
' O.5 .
0.4 ' 0.3 ' u ' O.2 ' 0.1 O _ g 2 2.5 3 3.5 4 4.5 Crop Yield for Other Vegetables (kg/m8 ) Figure 6.2.12 Cumulative Frequeng of Sampled Yv (other) values (NUREGICR 5512 v.1 default shown as verticaldashed line). ResidentialScenario 6.2 - 7 January 30,1998
I e , 0.9 ' i e
- 0.8 ' '
h0.7 ' g0.6 t 0.5 e k0.4
's i E 0.3 i 0
0.2 { . 0.1 i 0 1.9 .: 2.1 2.2 2.3 2.4 2.5 2.6 Crop Yield for Fruits (kg/m') Figure 6.2.13 Cumulative Frequency of Sampled Yv (fruit) values (NUREGlCR 5512 v.1 default shown as verticaldashed line). l 0.9 ', l 0.8 ' f0.7
- 0.6
{ 0.5
- l 5
$ v.4 '
E 0.3 . U 0.2 . 0.1 l 0 O.25 . 0.45 0.65 0.85 .1.05 Cf op Yield for Grains (kg/m') Figure 6.2.14 Cumulative Frequency of Sampled Yg values (NUREG/CR 5512 v.1 default shown as verticaldashed line). Residential Scenario 6.2 8 January 30,1998 1
4 I 0 ' 0.8 '
' )
0.7 e 0.6 '
~
' O.5 '
0.4
- 0.3 e u 0.2 0.1 0.5 1 1.5 2 2.5 3 Crop ileid for Beef Cattle Forage (kghn')
Figure 6.2.15 Cumulative Frequency of Sampled Yf values (NUREG/CR 5512 v.1 default shown as verticaldashed line). 0.9 0.8 ,' O.7 l 0.6 , j 0.5 . 0.4 e
$'O.3
- u '
0.2 , 0.1 l 0 O.3 0.5 0.7 0.9 1,1 Crop Yield for Beef Cattle Grain (kgtm') Figure 6.2.16 Cumulative Frequency of Sampled Yg values (NUREGlCR 5512 v.1 default shown as verticaldashedline). Residential Scenario 6.29 January 30,1998 _a
0.9 . , 0.8 . (0.7 . ' I 0 .6 . k . j 0.5 . k0.4
-i .
g 0.3 .
" 0.2 l 0.1 .'
0 S 10 15 20 25 30 Ingestion Rate for Beef Cattle Forage (kg/d) Figure 6.2.17 Cumulative Frequency of Sampled Of values (NUREG/CR 5512 v.1 default shown as vertical dashed line). O l Residerdial Scenario 6.2-10 January 30,1998 9
6.3.1 Do:c Modsling Rzsults The parameter distributions defined in Sections 3 and 5 were used to derivs screening dose values for unit concentrations of each of the 106 potential source radionuclides having half-lives greater than 65 days (see Table 6.3.1). The general procedure for establishing these dose values is described in Section 2.0. The application of this procedure to the residential scenario, a.id the resulting screening dose values, are summarized below. 6.3.1 Behavioral parameter values for the screening group The screening group is a generic critical group suitable for making decisions at any site without site specific information about, or constraints on, potential resident behaviur. For the residential scenario, the screening group is defined as adult male resident farmers. The behavioral parameter values for the average member of the screening group (AMSG) are defined by the mean values of the respective parameter distributions, described in Section 3.0. For the behavioral parameters, the mean values in Table 6.1.1 summarize the default va!ues characterizing the AMSG. 6.2 Calculation of Screening Dose Values As described in Section 2.0, screening dose values are calculated by deriving the distribution of pcssible dose values over all sites from the distributions defined for the physical parameters, given the behavioral parameter values defining the AMSG. Screening dose values are then defined by selecting, for each source nuclide, a dose value near the upper end of the resulting dose distribution. In general, this calculation entails: sampling the distributions for the scenario parameters characterizing the physical properties of the sites; using the scenario model to calculate the dose to the AMSG for each set of sampled values of the physical parametert; assembling the dose distribution from the resulting individual dose calculations; and identifying the dose value at the selected quantile of this distribution. The residential scenario model has 435 parameters characterizing the physical properties of the site for which distributions were defined. Five hundred and aighty samples from these distributions were generated using stratified Monte-Carlo (LHS) sampling. Figures in Section 6.7 show the cumulative frequency distributions for the sampled values of these parameters. Because of the large number of element-specific parameters, four representative distributions were selected for the partition coefficients and concentration factors. For each set of sampled parameter valuet dose to the AMSG was calculated for unit concentrations of each of the 106 possible source nuclides. For each source, the distribution describing possible doses to the AMSG was then constructed from these calculated doses. Due to the large number of calculations required by this analysis, the mixing cell model described in NUREGICR-5512 Volume 1 was used to represent the groundwater pathway. This model results in faster execution time then the more accurate numerical transport model, but introduces some amount of numerical dispersio,1. A selective sensitivity analysis was conducted to investigate differences between implementing Residential Scenario 6.3.1-1 January 31,1998
th) unsaturat;d zona analytic l mixing cell modsl and the numerical model. Using the mean values for all model parameters, the TEDE for all 106 isotopes was calculated using brih the l mixing cell and numerical models. The TEDE for every nuclide was the same to six significant l figures except for 3H, 75 Se, " Mo, '2' I, 22 era, 22 era +C, 230Th. 230Th+C, 233U, 23eU+C, 2'5Cm, and 2"Cm. For these radicnuclides, the TEDE's from the mixirg cell and numerical models were equivalent to three significant digits. l l Analyzing these results indicated that the mean radionuclide partition coefficients, most of which are larger than the NUREG/CR 5512 defaults', are producing the effect of retaining the radionuclides in the unsaturated zone and therefore, decreasing the importance of the dose ' from the ground water pathways, which in tum decreases the sensitivity of using the mixing celi or numerical model for unsaturated zone transport. To bound the potential error associated with using the mixing cell modelinstead of the numerical model, calculations were conducted assuming no sorption and a relatively thick unsaturated zone. Results of these analyses for selecwd radionuclides are shown in Table 6.3.2. For each of these radionuclices, the maximum error (numerical solution TEDE minus mixing cell solution TEDE divided by the numerical solution TEDE) is less than 12%. "E tunds to .e overestimated by the mixing f.;eh model. Possible screening dose values were selected from the derived dose distributions by stipulating a tolerance for underestimating dose (i.e. Peni). For three alternative values of Peni, and for each source nuclide, a screening dose value was identified such that the fraction of doses larger than i I the screening dose was equal to Peni. These values correspond to the (1-Pen,) quantiles of the calculated dose distributions. Table 6.3.3 lists these screening dose values for each of the source nuclides, and for the three alternative values for Pcrit. As a measure of the spread of the dose distributior"), Table 6.3.3 also shows the ratio of dose at the 99'th percentile to the median (50'th percentile) dose. The derived dose distribution functions can also be used to test or formulate more complex decision criteria. As an example, the dose value at the 95'th percentile of the dose distribution can be identified by stipulating the dose value at some other quantile of the dose distr;bution. Table 6.3.4 lists, for each of the three Pen, values, the dose value at the 95'tn percentile, given that the dose at the (1-Pen,) quantile is 25 mrem. Dose values at the selected qt'antiles can also be used to calculate th3 source concentration equivalent to a dose of 25 mrem. Table 6.3.5 summanzes these concentration values.
' The mean of the pdf for Kd for 60 of the 69 elements is greater than the volume 1 default Residential Scenario 6.3.1-2 January 31,1998
Table 6.3.1 - Source Nuclides used in the Parameter O Analysis O Source Source Source ' Source Source Source ID ID ID 1 3H 87 126Sn+C 180 232Th 2 10Be 89 125Sb 181 232Th+C 3 14C 93 123mTe 183 231PA 5 22Na 95 127mTe 184 231Pa+C 9 35S 106 1291 187 232U 10 36Cl 114 134Cs 188 232U+C 11 40K 115 135Cs 189 233U 12 41Ca 117 137Cs 190 233U+C 13 45Ca 128 144Ce 191 234U 14 463c 132 147Pm 192 235U 16 54Mn 137 147Sm 193 235U+C 18 55Fe 138 151Sm 194 236U 20 57Co 140 152Eu 196 238U 21 58Co 141 154Eu 197 238U+C 22 60Co 142 155Eu 199 237Np 23 59Ni 144 153Gd 200 237Np+C 24 63Ni 145 160Tb 203 236Pu ps, 27 652n 146 166mHo 205 238Pu
\ j 31 75Se 147 181W 206 239Pu 32 798e 148 185W 207 240Pu 41 90Sr 150 187Re 208 241Pu 48 93Zr 151 1850s 209 242Pu 49 93Zr+C 153 192lr 211 244Pu 52 93mNb 156 210Pb 212 241Am 53 94Nb 160 210Po 213 242mAm 58 93Mo 165 226Ra 215 243Am 61 99Tc 166 226Ra+C 216 242Cm 65 106Ru 167 228Ra 217 243Cm 69 107Pd 169 227Ac 218 244Cm 71 110 mag 170 227Ac+C 219 245Cm 73 109Cd 173 228Th 220 246Cm 74 113mCd 174 228Th+C 221 247Cm 81 119mSn 175 229Th 222 248Cm l 82 121mSn 176 229Th+d' 223 252Cf 84 123Sn '77 230Th 86 126Sn 178 230Th+C D ! Residential Scenario 6.3.1-3 January 31,1998 l .
ma4 s. A h.4 mmwh ahm a.ed.m g paa m w em,em u . A m.44m am%A um aw as u.a p a al.aar. wwm.m.p_-mM a.m 4.paam .a w. e e-4 sear e m.a.s m44em m-- -- m we , i 1 J I i O ! I m; u e i l O' I i i I l 9 1 1 I
Table 6.3.3 - Selected quantiles of unit-concentrationTEDE O distributions for the residential scenario (mrem) i i source Pcnt=0.25 Pcnt=0.10 Pcnt=0.05 Dose @ Pcnt=0.01
/
1 ( Dose @ Pcrit=0.50 652n 184E+30 2 32E+00 2 80E+00 3 38 75Se 4 24E-01 4 29E 01 4 32E 01 1.05 79Se 105E-01 121 E-01 135E-01 1 92 903r 8 80E+00 146E +01 2 05E+01 8 42 932r 182E-02 2 32E-02 3 86E-02 13.38 93Zr+C 9 84E 03 133E-02 2 01E-02 12 60 93mNb 124E-02 1.38E 02 1.67E-02 7 68 94Nb 4 30E+00 4 32E+00 4 34E+00 1.03 93Mo 5 94E 02 1.17E-01 1.67E-01 11 03 99Tc 8 57E-01 1.34 E +00 168E+00 5 63 106Ru 4 73E-01 4 94E-01 518E-01 1.29 107Pd 2.76E-03 3 89E 03 611E 03 12 35 110 mag 4 93E+00 5 08E+0L 523E+00 1.20 109Cd 163E-01 2.35E-01 3 46E-01 10 77 113mCd 2 84E+00 5 05E+00 9 07E+00 1952 119mSn 6 95E-03 810E-03 1.10E 02 14 6* 121mSn 183E-02 4 39E-02 194E-01 61.15 123Sn 2 86E-02 3 24E 02 4 06E-02 5.76 126Sn 5 3rE+00 5 32E+00 536E+00 2.13 126Sn+C 2 48E+00 2 49E+00 2.53E+00 2.13 125Sb 9 71E 01 9 76E-01 9 82E-01 1.16 123mTe 134E-01 1.35E 01 1.36E-01 1.22 127mTe 164E 02 1.75E 02 188E-02 3 54 (g 1291 147E+01 4 65E+01 1.01 E+02 49 83 l
'v} g 134Cs 135Cs 137Cs 418E+00 8 946-02 2 06E+00 4 40E+00 136E-01 2 27E+00 4 66E+00 2.18E 01 1.92 29 74 2.54E+00 5 67 144Ce 129E-01 1.36E-01 144E 01 1.36 147Pm 2.75E 03 3 05E-03 3 24E 03 1 92 147Sm 6 07E-01 6 91E-01 8 66E-01 6 61 151Sm 124E-03 142E 03 167E-03 6 26 152Eu 2 88E+00 2 88E +00 2.89E+00 1.01 154Eu 312E+00 312E+00 312E+00 1.01 if5Eu 8 75E 02 8 80E-02 8.86E 02 1.07 153Gd 7.66E-02 7 93E-02 8 83E-02 1 66 160Tb 8 29E-01 8 29E-01 8.29E-01 1.02 166mHo 449E+00 4 49E+00 4 50E+00 1A1 181W 164E 02 1.66E 02 1.77E-02 1 95 185W 187E 03 2 43E-03 5 51E 03 27.86 187Re 4 09E 04 5 95E 04 8.25E-04 7.15 1850s 6 48E 01 6 49E-01 6 50E-01 1.03 192ir 6 04E-01 6 05E-01 6 05E-01 1.01 210Pb 2 63E+01 2 956+01 317E+01 5 37 210Po 2.64E+00 2 82E+00 2 97E+00 1.69 226Ra 3 22E+01 3 60E+01 3 86E+01 5 60 226Ra+C 415E+00 4 58E+00 4 85E+00 5 24 228Ra 649E+00 6 84E+00 7 05E+00 1.24 227Ac 422E+01 4 70E+01 516E+01 8 85 /"T Residential Scenario 6.3.1-5 January 31,1998 N.si
Table 6.3.3 - Selected quanti,es of unit.concentrationTEDE distributions for the residential scenario (mrem) Eurce Pcnt=0.25 Pcnt=0.10 Pcnt=0.05 Dose @ Pcnt=0.01i Dose @ Perit=0.50 227Ac+C 5 2BE+00 5 88E+00 6 43E+00 8 83 228Th 512E+00 529E+00 543E+00 1.15 22BTh+C / 37E 01 7 62E-01 7 81E-01 1 15 229Th 1.22E + 01 135E+01 146E+01 5 99 229Th+C 1.53E+00 169E+00 183E +00 5 98 230Th 1.19E+01 136E +01 151E+01 9.18 230Th+C 3 88E+00 4 33E+00 4.67E+00 6 32 232Th 2 05E+01 2.21 E +01 2.32E+01 3 01 232Th+C 212E+00 2 27E+00 2 40E+00 315 231Pa 6 82E+01 7.66E +01 9 01E+01 .9 231Pa+C 824E+00 9 38E+00 106E+01 7.36 232U 101E+01 1.28E+01 4.25E+01 17.79 232U+C 1.43E+00 1.72E+00 5 21E+00 15.34 233U 1.71E+00 2.74E+00 6 76E+00 21.20 233O+C 153E+00 1.79E+00 2 55E+00 6 90 234U 1.12E+00 1.89E +00 6 62E+00 22.71 235U 2.22E+00 311E+00 7 47E+00 20 57 235U+C 6 99E+00 7.91 E+00 9 09E+00 7.19 236U 1.06E+00 179E+00 6 27E+00 22 81 238U 111E+00 1.80E+00 6 33E+00 21.80 238U+C 3 04E+00 3 51E+00 4 59E+00 7.03 237P'o 141E+02 2.72E+02 4 30E+02 11 45
~37Np+C 1.36E+01 2 55E+01 4 35E+01 10 12 236Pu 2 74E+00 3 06E+00 3 35E+00 2.87 238Pu 8 88E+00 9 83E+00 1.05E +01 1.93 239Pu 9 88E+00 1.09E+01 1.17E +01 2 47 240Pu 9 88E+00 109E+01 1.17E +01 2 46 241Pu 3 02F-01 3 49E 01 5 81E 01 11.58 242Pu 9 38E+00 104E+01 1.11E+01 2 47 244Pu 103E+01 1.13E+01 1.21 E +01 2 23 241 A'm 105E+01 1.20E +01 165E+01 10.28 242mAm 24 ~,Am 109E+01 1.24E+01 1.68E+01 9 96 l
242Cm 1.38E 01 153E-01 1 6' ' 01 1 43 243Cm 7.15E+00 7.82E+00 8.2bd+00 L41 244Cm 5 46E+00 6 OnE+00 8 34E+00 1.42 245Cm 153E+01 181E +01 2.12E+01 3.93 l 246Cm 103E+01 1.145+01 1.23E+01 1 42 247Cm 107E+01 1.18E+01 124E+01 1.35 248Cm 3 80E+01 ,418E+01 4 41E+01 1 42
?M7Ff 11?F+00 1 CAF+nn ad?F+nn 9 7 Td Residential Scenario 6.3.1-6 January 31,1998
(___ l Tcble 6.3A - 95'th perc:ntil] d:Co values f:r 25mr;m dO o valu;s at P em [ ( < Source 3H Pcrit=0.25 5 49E+01 Pcrit=0.10 3 34E+D1 Pcrit=0.05 2S Source 166mHo Perit=0.25 2 50E+01 Pcrit=0.10 2 50E+01 Pcrit=0.05
?S 10Be 318E+01 2 84E+01 2S 181W 2 70E+01 2 67E+01
14C 1.58E+02 4 47E+01 25 185W 7.36E+01 5 66E+01 & 22Na 312E+01 2 91E+01 26 187Re 5 0$E+01 3 47E+01 >$ 35S 4 64E+01 3 24E+01 25 ibSOs 2 51E+01 2 51E+01 25 36CI 479E+01 3 09E+01 25 192tr 2 50E+01 250E+01 25 40K 1.35E+02 5 34E+01 25 210Pb 3 01E+01 2 68E+01 75 a 41Cs 533E+01 3.22E+01 25 210Po 2 81E+01 2 64E+01 ?S 45Ca 6 43E+01 3 31E+01 25 226Ra 3 00E+01 2 68E'+01 25 46Sc 2 50E+01 2.50E+01 25 226Ra+C 2 92E+01 2 64E+01 25 54Mn 2 81E+01 2 66E+01 25 228Ra 2 72E+01 2.58E+01 25 55Fe 3 Oy+01 2.75E+01 25 227Ac 3 05E+01 2.74E+01 57Co 2 62E+01 2.57E+01 25 _ 25 227Ac+C 3 0$E+01 2.74E+01 25 58Co 2 52E+01 2 51E+01 25 228Th 2.65E+01 2 56E+01 25 60Co 2 61E+01 2 57E+01 25 228Th+C 2.65E+01 2 56E+C1 25 59Ns 1.63E+02 7 4SE+01 25 229Th 2.99E+01 70E+01 25 63r4 154E+02 7.36E+0t 25 229Th+C 2 99E+01 2.71 E+01 25 652n 3 80E+01 3 01E+01 25 230Th 317E+01 2 77E+01 25 75Se 2 55E+01 2 52E+01 25 230Th+C 3 01E+0! 2 69E+01 25 79Se 3 23E+01 2.79E+01 25 23:Th 2 82E+01 2 62E+01 25 90Sr 5 83E+01 3 52E+01 75 232Th+C 2 83E+01 2 64E+01 25 93Zr 5 31E+01 417E +01 25 231Pa 3 30E+01 2 94E+01 25 932r+C 510E+01 3 78E+01 25 231Pa+C 3 22E+01 2.83E +01 2h __ 93mNb 3 38E+01 3 03E+01 25 232U 1.05E+02 8 32E+01 25 94Nb 2.52E+01 2.51 E +01 25 2320+C
,/m\ 93Mo 7.03E+01 3.57E+01 25 233U 9 09E+01 9 91E+01 7.60E+01 616E+01 2$
25
-( 99Tc 506Ru 4 89E+01 2 74E+01 314E+01 2 62E+01 2S 25 233U+C 234U 415E +01 14BE+02 3 5EE+01 -
8 76E+01 2S 25 107Pd 5 54E+01 3 93E+01 ?S 235U 8 41E+01 6 00E+01 25 110 mag 2 65E+01 2.57E + 01 25 2350+C 3.2SE+01 2.87E+01 2S 109Cd 5 32E+01 3 68E+01 ?!i 236U 148E+02 8 76E+01 25 113mCd 7.98E+01 4 49E+01 2' 238U 143E+02 8 78E+01 25 119mSn 3 97E+01 3d1E+01 25 238U+C 3 77E+01 3 28E+01 25 121mSn 2 65E+02 1.10E+02 5 237No 7.63E+0' 3 95E+01 25 123Sn 3 55E+01 313E+01 25 237No+C 6 01E+01 4 26E+01 25 176Sn 2 53E+01 2 52E+01 25 236Pu 3 06E+01 2 74E+01 2!i . 126Sn+C 2.55E+01 2.53E+01 25 238Pu 2.94 E + 01 2 66E+01 2S 125Sb 2 53E+01 2 51E+01 25 239Pu 2.95E+01 - 2 6SE+01 25 123mTe 2 53E+01 2 52E+01 25 240Pu 2 95E+01 2.66E+01 25 127mTe 2 86E+01 2 68E+01 25 241Pu 481E+01 416E+01 25 1291 1.72E+02 5 44E+01 2S 242Pu 2 95E+01 2.66E+01 2S 134Cs 2.79E+01 2 65E+01 25 244Pu 2 92E+01 2 67E+01 25 135Cs 610E+01 4 00E+01 2S 241Am 3 94E+01 3 42E+01 25 13Es, 3 09E+01 2.81d+01 25 243Am 3 86E+01 3 38E+01 25 M~~ 2.78E+01 2 65E+01 25 243mAm 25 147Pm 2 94E+01 2.66E+01 25 242Cm 2 91E+01 2 63E+01 25 147Sm 3 57E+01 313E+01 25 243Cm 2 89E+01 2 64E+01 25 151Sm 3.36E+01 2 94E+01 25 244Cm 2 90E+01 2 64E+01 2S _152Eu 2 50E+01 2 50E+01 "5 245Cm 3 46E+01 2 94E+01 % p) ( v Residential Scenario 6.3.1-7 January 31,1998
Tcbl] 6.3.4 - 95'th perc:ntile d 03 valu:s f r 25mr:m d;s3 v:lu:s at Pc,,, ~5ource Pcrit=0.25 Pcrit=0.10 Pcrit=0.05 Source Perit=0.25 Perit=DE Pcrit=0.0T 154Eu 2 51E+01 2 50E+01 2S 245Cm 2 90E+01 264E+0i ')!i 155Eu 2 53E+01 2 52E+01 ?fi 247Cm 2 63E+01 75
?_40E+01 153Gd 288E+01 2 79E+01 95 24BCm 2 90E+01 2 64E+01 ?S 160Tb 2 506+01 2 50E+01 75 252Cf 3 54E+01 3 03E+01 m O
Residential Scenario 6.3.1-8 January 31,1998
Table 6.3.5 Concentration (pCilg) equivalent to 25 mremly f^ for the speCified value of Pm ( ' Source Pcrit=0.25 Perst=0.10 Pertt=0.05 l Source Pcrit=0.25 Perit=0.10 Perst=0.05 3H 177E+02 108E+02 8 06E+01 166mHo 5 57E+00 5 56E+00 5 SSE+00 10Be 169E+03 151E+03 133E+03 181W 152E+G, 151E+03 141E +03 14C 410E+01 1.16E+01 6 50E+00 185W 1 34E+04 103E+04 454E+03 22Na 455E+00 4 25E+00 3 65E+00 187Re 612E+04 4 20E+04 3 03E+04 35S 3 87E+02 2 70E+02 2 08E+02 1850s 3 86E+01 3 85E+01 3 85E+01 36CI 5 61E-01 3 62E-01 2 93E 01 192ir 414E +01 413E +01 413E+01 40K 913E+00 3 60E+00 169E +00 210Pb 9 50E 01 8 46E 0* 7.90E-01 41Ca 1.10E+02 6 63E+01 515E+01 210Po 9 46E+00 8 87E+00 8 41E+00 45Ca 9 29E+01 5 67E+01 4 28E+01 226Ra 7.77E 01 6 94E 01 6 48E 01 46Sc 147E+01 147E+01 147E+01 226Ra+C 6 03E+00 545E+01 516E+00 54Mn 1.57E+01 1.48E+01 139E+01 228Ra 3 85E+00 3 65E+00 3 54E+00 55Fe i 13E+04 103E+04 9 35E+03 227Ac 5 92E 01 5 31E-01 4 85E-01 57Co 151E+02 148E+02 144E+02 227Ac*C 4 74E+00 4 25E+00 3 89E+00 58Co 3 49E+01 3 47E+01 3 45E+01 228Th 4 89E+00 4 73E+00 4 61E+00 60Co 3 85E+00 3 79E+00 3 68E+00 , 228Th+C 3 39E+01 3 28E+01 3 20E+01 59Ni 1.21 E +04 5 5 tE+03 1.85E+03 22PTh 2.04 E +00 18f S00 1.71 E +w 63Ni 4 43E+03 211E+03 ' 7.17E +02 229Th+C 1.63E +01 148E +01 1.36E+01 65Zn 1.36E+01 108E+01 8 93E+00 230Th 210E+00 183E +00 165E+00 75Se 589E+01 5811:+01 5 78E+01 230Th+C 6 44E+00 5.78E+00 5 36E+00 79Se 2 39E+02 247t:+02 185E+02 232Th 122E+00 1.13E+00 108E+00 90Sr 2 84E+00 1.72E +00 122E+00 232Th+C 1 1oE +01 1 10E+01 104E +01 932r 1.38E+03 108E+03 A 48E+02 231Pa 3 66E 01 3 27E 01 2.77E 01 93Zr+C 2 54E+03 188E+03 125E+03 231Pa+C 3 UJE+00 2.67E+00 2.36E+00 93mNb 2 02E+03 181F+03 149E+03 232U 2 47E+00 1 %E+00 5 88E 01 (j) i l ( 94Nb 5b1E+00 579E+00 576E+00 232U+C C 174 E+01 146E+01 4 80E+00 93Mo 421E+02 213E+02 149E+02 233U 147E+01 911E+00 270E+00 l 99Tc 2 92E+01 1 8'E +01 149E+01 233U+C 1.63E+01 140E +01 E81E+00 l 106Ru 5 28E+01 5 06E+01 4 83E+01 234U 2 23E+01 132E+01 3 78E+00 _107Pd 9 07E+03 6 43E+03 4 09E+03 235U 113E+01 604E+00 3.35E+00 110 mag 507E+00 4 92E+00 4 78E'00 235U+C 3 585+00 316E+ 00 2.75E+00 109Cd 154E+02 106E+02 7 23E+'.; , 236U 2 36E+01 140E+01 3 99E+00 113mCd 8 80E+00 4 95E+00 2 76E+001 - 238U 2 26E+01 1.39E+01 3 95E+00 119mSn 3 60E+03 3 09E+03 2.268!+03 23U+C 8 21E+00 7.13E+00 544E+00 _121mSn 137E+03 5.70E+02 129E+02 237No 1.77E 01 918E-02 5.81 E-02 _ 123Sn 8 74E+02 7 71E+02 616E+02 237Np+C 184E+00 9 81E-01 5 75E-01 126Sn 472E+00 4 70E*00 4 66E+00 236Pu 911E+00 817E+00 7 45E+00 126Sn+C 1.01 E + 01 100E+01 9 89E+00 238Pu 2 81E+00 2 54E+00 2.39E+00 125Sb 2 57E+01 2 56E+01 2 55E+01 239Pu 2 53E+00 2 28E+00 215E+00 123mTe 186E+02 185E+02 184E+02 240P., 2 53E+00 2.28E+00 2.16E+00 127mTe 152E+03 143E+03 133E+03 241Pu 8.28E+01 7.16E +01 4 3JE+01 1291 170E+00 5 38E-01 2 47E-01 242Pu 2 66E+00 2 41E,00 2.26E+00 134Cs 5 98E+00 5 68E+00 5 36E+00 244Pu 2 42E+00 2.22E+0C 2.07E+00 135Cs 2 80E+02 183E+02 1.15E+02 241Am 2 39E+00 2 08E+00 1.52E+00 137Cs 1.22E+01 1.10E+01 983E+00 243Am 2.30E +00 2 01E+00 1.49E+00 144Ce 1,93E+02 1.84E+02 1.74E+02 243mAm 147Pm 9 08E+03 8 20E+03 7.71 E+03 242Cm 181 E+02 164E+02 1.56E+02 147Sm 412E+01 3 62E+01 2.89E+01 243Cm 3 50E+00 3.20E+00 3 03E+00 151Sm 2 01E+04 176E+04 150E+04 244Cm 458E+00 417E+00 3 94E+00
/O Residential Scenario 6.3.19 January 31,1998 (v )
l
Table 6.3.5 - Concentration (pCilg) equival:nt 13 25 mr;mly for the specified value of P t,,,
~ ~ Source Pent =0.26 Pcnt=0.10 Pcrit=0.05 Source Perst=0. 5 Perit=0.10 Perit=0.05' 152Eu 8 68E+00 8 67E+00 8 66E+00 245Cm 163E+00 138E+ 00 1 1BE+ D0 154Eu 8 02E+00 8 01E+00 246Cm 2 42E+00 2 20F+00 2 09E+00 8 00E+0_0 155Eu 2 86E+02 2 B4E+02 2 82E+02 247Cm 2 33E+00 212E+00 2 02E+00 153Gd 3 27E+02 315E +02 2 83E+02 248Cm 6 57E 01 5 98E-01 5 67E 01 1ANTh _
1 n?p,nt 3nyp,g9 q l l l 1 1
)
I 9 l r l Residential Scenario 6.3.1-10 January 31,1998
6A S:nsitivity Analysis I r ,~ (' The results of the Monte Carlo dose calculations were processed to identify parameters control! sng TEDE for each source. The dependence of TEDE on the model parameter values is potentially complex: total dose may depend non monotonically on the parameter value, or may be sensitive to the parameter value only within certain limits, or only in conjunction with certain ranges of values for other parameters. Because of these complexities, a linear regression analysis was not used to identify sensitive parameters. Instead, a robust test which does not rely on monotonicity was employed. For each source nuclide, sensitive parameters were identified by dividing the sample vectors into two groups with equal numbers of samples: vectors having doses above the median dose, and vectors with doses below the median dose. For each parametc. 9e Kolmogorov Smirnov (K S) test was used to assess the significance of the differences in the distributions of parameter values between these two groups. Parameters whose distributions differed at a significance level of 0.001 were selected. A restrictive value of the significance levelis appropriate in this analysis because of the large number of tests performed (580 vectors x 435 sampled parameters), and the correspond;ngly high prospect of producing low K S statetic values by random chance. t
'g For each parameter selected, the strength of the dependence of TEDE on the parameter value was calculated by segregating the sample vectors on the basis of the parameter value. This segregation defines two groups of sample vectors: vectors having values for the selected parameterless than a chosen quantile; and vectors having parameter values greater han the chosen quantik Within each group, the TEDE distrioution was estimated using only vectors in that group. The 95'th percentile of this distributio.. was then compared to the 95'th percentile of the original TEDE distribution using all sample vectors. The ratio of the 95'th percentile TEDE value from the segregater' sample to the 95'th percentile TEDE value trom the original sample measures the strength of the relationship between the TEDE and the parameter. This measure of the strength of dependence of dose on parameter value provides a direct indication of the potential for new p arameter information (expressed as a revised limit on the parameter value) !
+ to change the screening dose value. Finally, those paraw tors with 'signif. cant" potential to modify the screening dose value were selected basr .ne calculated strength measure. A threshold value of 0.52 for *significant" f m Residential Se ario 6.4-1 February 1,1998
reduction of the 95'th percenti'e of the dose distnbution was selected. Paramet:rs hrving strength measures Mss than this threshold (i e. with the potential to ef'ect a greater reduction in the 95'th percentile) were considered to be strongly and significantly correlated with dose. The threshold strength measure value of 0.52 was selected by noting the spunous associations between parameter values and TEDE that emerged. The indoor shielding factor SFI was identified as significant by the K S test, and had a , associated strength measure of 0.52. This parameter, however, was not used in the calculation, and the reported strength measure is an artifact of sampling error. Strength measure values less than this threshold were assumed to be significant. The output of the LHS sampling was not used directly as input to the residential scenario model. Several of the model parameter (for example cultivated area, porosity, and infiltration rate) wero functionally defined in terms of other model pararr ers, or in terms of parameMrs not used directly in the model (soil classification, for example). The resident'al scenario input was produced by imposing these functional connections using an 1.HS sampling of the distinct random variables defined for the analysis. ! Separate sensitivity analyses were conducted for the two sets of sample vectors: the first set consisting of samples of the fundamental parameters produced by LHS, and the second consisting of the resulting samp'is of modelinput parameters based on these fundamei,w, parameters. Table 6.4.1 lists, for each source nuclide, the identifiers of the sampled parameters identified as l having a strong significant relationship to dose due to that nuclide. Table 6.4.2 lists the l corresponding results for the modelinput parameters, Some relationships in Table 6.4.2 are an artifact of the functional connection among soil properties and soil type. The fraction of hydrogen in soil, for example, is only used in the tritium model. It appears as a significant l parameter for dose due to 1291, however, because of the functional connection between the i hydrogen fraction and the soil satura. ion fraction, F1. 1 l For many source nuclides, no significant controlling parameters were identified. The small range of the dose distribution for some nuclides may make the relationship between parameter values and dose difficult to distinguish from sampung error. Residential Scenario 6.4 2 February 1,1998 l l
i Table 6.4.1 8:mpled par;m;ters having significani strong correlations with TEDE bource Parameter $trength 3H Soil class 0 44 14C KdC 0.18 H2 0.19 Soil class 0.38 36Cl Concentration factor: grainCL 0.48 IOK H2 0.20 KdK 0.22 Soil class 0.47 41Ca ConcentreLn factor: leafy CA 0 25 45Ca Conce' trat;on factor: leafy CA 0 21 59Ni KdNI 0.19 Concentration factor: root NI 0.48 33Ni .KdNI 0.20 Concentration factor: root NI 0.48 33Mo Concentration factor: leafy $10 0.28 99Tc Concentration factor: leafy TC 0.33 107Pd Concentration factor: leafy PD 0.50 113mCd KdCD 0.41 121mSn KdSN 0.10 1291 Kdl 0.14 H2 0.15 Soil class 0.45 185W KdW 0.39 Soil class 0.43 Wet / dry conversion: nonleafy 0.43 232U H2 0.25 Wet / dry cor /ersion: nonleafy 0.26 KdU 0.27 Soil class 0.31 232U+C Wet / dry conversion: nonleafy 0.29 Soil class 0.34 233U H2 0.26 Wet / dry conversion: nonleafy 0.30 Kdu 0.33 Soil class 0.41 234U KdU 0.19 Wet / dry conversion: nonleafy 0.25 l Soil class 0.38 ( 235U Wet / dry conversion: nonleafy 0.36 l Residential Scenario 6.4-3 February 1,1998 l
KdU _ 0.37 Soil class 0 46 236U KdU 0 19 Wet / dry conversion: nonleafy 0 25 Soil class 0 38 238U KdU 0 20 238U Wet / dry conversion: nonleafy 0.25 [3RL' Soil c!nts 0 38 O e.. ..
. 1,., ec. .,,, e. - ,.,m. ,,.,eee
~I khhhkhm hb ,s ' able 6.4.2 T Model parameters having significant strong
- orrelations with TEDE Bource Parameter Min RDCF 3H Fraction of hydrogen in soil 0 22 Moisture content of soil 0 22 F1 0 27 F2 0 27 1 0.29 N1 0 40 N2 0 40 PS 0 40 RHO 1 0 40 RHO 2 0 40 14 0 F1 0.14 F2 0.14 Fraction of hydrogen in soil 0.15 Moisture content of soil 0.15 KdC 0.18 1 0.18 H2 0.19 3GCl Concentration factor. grain Cl 0 48 e 40K H2 0 20
( KdK 0.22 41Ca Concentration factor: leafy Ca 0.25 45Cs Concentration factor: leafy Ca 0.21 59Ni 1 0.17 KdNi 0.19 33Ni KdNi 0.20 93Mo Concentration factor. leafy Mo 0.28 99Tc Concentration factor: leafy Tc 0.33 107Pd Concentration factor: leafy Pd 0.50 113mCd KdCd 0.41 121mSn KdSn 0.10 1291 F1 0.10 F2 0.10 Fraction of hydrogen in soil 0.11 Moisture content of soil 0.11 Kdi 0.14 H2 0.15 1 0.16 185W KdW 0 39 ('S Residential Scenario 6.4 5 February 1,1998 N
WeVdry convorsion nontafy 0 43 232U H2 0 25 I WeVdry conversion: nonleafy 0 26 l U 0.27 232U+C WeVdry conversion: nonleafy 0 29 233U H2 0 26 Weydry conversion nonleafy 0 30 U 0 33 234U U 0.19 Wevdry conversion: nonleafy 0.25 235U WeVdry conversion: nonleafy 0.36 U 0.37 236U i 0.18 U 0.19 WeVdry conversion: nonleafy 0.25 238U U 0. 2 Wet / dry conversion: nonienfv 0 25 O Residential Scenario 6.4-6 February 1,1998
Secti:n 7.0. Equati:ns f:r Distributi:n Functi:ns and LHS Calculati:ns 7.1 C Fit Program PDF Equations . The following equations and definitions were taken from the C Fit' software (C Fit,1996) that was used in the development of probability and cumulative distnbution fits for residential . scenario parameters based on data acquired to support the estimates of parameter values and ' ranges. Among the many distribution types included in C Fit, we used normal, log normal, beta, gamma, and Gumbel distributions. When using the software, selection of fitted distributions is based on either the Chi square or Kolmogorov Smirnov goodness of fitness tests. When a l distribution is selected, graphical plots can be generated as well as full reports of basic statistics and the distribution parameters. PDF charts display the sample data histogram along with the plot of the selected fx(x) distribution. The CDF charts display the sample data as an area chart corresponding to the histogram plotted in the PDF, The cumulative distribution function, Fx(x), is plotted as a continuous line representing the probability that the random variable X w;ll be less than or equal to x: Fx(x) = P(Xsx) for all x. ,' The CDF functioc. Fx(x) also has the properties that 'x(-=) = 0; Fx(+=) = 1.0, 2; 3.0 s Fx(x) s 1 1.0, and is nondecreasing, and 3) Fx(x) is continuous for any real value of x. Following are the equations for the distribution types used in our analysis, from the C-Fit User Guide. Normal Distribution ( f(r:p,0) = 1 c h3iEf (7.1.1) , @o where p and o are the mean and standard deviation of the variable and are defined as: i Mean p,n = - o I 1,f.s (7.1.2) l l i j and Standard Deviation is: I (7.1.3) i o, = h n - 1,f (x,- p,)2 l O V Residential 71 February 1,1998
Loa Normal
~ bi'"Y f(x;p,o,t) = ' (714)
(x -t)o dii l l Gumbel (Extreme Value Tvoe l Max ) l f(x;p,a) = ne h - " " ") (7.1,5) Gamma f(x;K,A,c) = 1[A(x -t)7e D""1 (7.1.6) f(K) where: K = A= (7,j,7) and the statistical gamma function: P(K) = t"'t ~ ' dd (7.1.8) Bela l
*i -i ' a-i l ' , _ 3, ' , _3, i i 1
t 6* - 6' ' ' 6, - 6 ' ' (7.1.9) f(x;a,,a,,6,,6,) = (5, - 6,) B(a,, a,)
- where the coefficient of variation (COV)is
Residential 72 February 1,1998
6,
- o' (7.1.10) s H,
and the statistical beta function: a(x,y) = I'I*)I'D'I = t O t)' ' 9s (7 m I(x +y) , 7.2 LHS Distribution Equations The LHS program was used for sampling calculations required to generate output distributions of dose to humans for the residential scenario. Distribution input files are deva %oed for all of
.e defined PDF's for the residential scenatio parameters. This s.ction provides the equations -
for the distributions used to define the parameter characteristics. 1 Unbounded Normal Distribution There are two input parameters required when defining a normal distribution: the mean and the standard deviation or variance. The mean may be any real value; however, the standard deviation must oe strictly positive. The normal distribution is defined in terms of the mean p and varianca c2by the following density function: O I
, (s p).
f(x) = e -=<x<. (7.2.1) 05 Unbounded Lognormal Distribution A lognormal distribution is simply the logarithm of a normal distribution, defined by the density function:
, Os t p)8 I 2#
fo-) = e y>0 (7.2.2) 105N where the mean, variance and median are, respectively; (} E b')
- e d Residential 7-3 February 1,1998
r y) , 2 a (, . I ) (7.2.4) Median = e ' (7.2.5) O Th;J distribution requires an error factor parameter, interpreted as the ratio of the value at the 95% quantile of the median; it is also the ratio of the median 'o the 5% quantile. This distribution operates by first converting the input mean and error factor into the mean and standard deviation of the underlying normal distribution by the following relations, sampling a normal distribution with these parameters, then taking the exponential of the resultant distribution: y, In(trror factor) 1.645 l p = In(Input mean) - c' (7.2.7) Loanormal-N Distribution This produces a lognormal distribution sampled over all quantiles. Input required in LHS are the mean and standard deviation of the underlying normal distribution. l Uniform Distribution - Uniform Intervals This distribution samples values uniformly between two specified intervals A and B. It is defined by the following density function: I
/(x) = , AsxsB (7.2.5)
B-A The mean and variance are: E(x) = ^
- 8 and V(x) = I8 ~ (7.2.9) 2 12 Residential 7-4 February 1,1998
Loguniform Distribution Un: form Intervals A loguniform distribution is the exponential of a uniform distribution. This distribution samples uniformly on the log base 10 of the specified end points A and B where A and B are both >0. The first step in the implementation of this routine is to find the base 10 loganthms of each of the input values. Next, a uniform distribution is generated on that interval. Finally, the antilog of each of these latter values is determined (i.e.10'). This process allows uniform sampling on a logarithmic scale. The following equations are stated in terms of natural logarithms to simplify the presentation. The density function for this distribution is:
/(x) = f(In A -In8), A<x<8 (7.2.10)
The me n, variance, and median (respectively) are as follows:
#~^
E(2) = (7.2.11)
, , , , ,, j V(2) * (8 - A)('" "^" ~( ~^'
(7.2.12) 2(In B -In A)2 i le 8
- In 4 Median = e 8
=/Is (7.2.13)
Triangular Distribution The triangular distribution is required when specific data are absent. The lower a and upper c parameters provide bounds beyond which sampling is not to occur. The most likely value is specified by the b parameter. The case we used was where a < b < c. The density functions for this triangular distribution are: f(x) = 2(x a asxsb (7.2.14) and, _= , . _ = s - Residential 75 February 1,1998 ___o
2 I# ~ *) f(x) - , bsxsc (7.2.15) (c - a)(c - 6) The mean, variance, and median (respectively) are as foliows: a E(x) = #
- b ^ # (7.2.16) 3
,,{,) , a(a - b) + b(b -c) + c(c - a) 7 2M7) 18 (c -a)(6 - a) , b2 (7210)
Afedian = a + h 2 2 I'~#N# , b s # ## (7219) Afedian = c h 2 2 O Beta Distribution A beta distribution is used as a rough modelin the absence of data. It is sampled from endpoints A to B, and includes shape parameters p and a The following conditions must be satisfied: p, q 2 0.001 0sA<B. The beta distribution is defined with the following density functions: (7.2.20) iD A Residential 7-6 February 1,1998
. -- - - -. -- _ . - . - _ ~ - - . . - . . ~ . . . - - - ~ . - - _ . - - - . - .
User Defined Cumulative Continuous Distribution with Linear Interoolation ' f A continuous distribution is used when the user knows certain values that the data will take, and I linearly interpolates between those values. It is most commonly used for production of discrete < j p 2r4(1 -1)r 8 (7.2.21) items, values of sales, and good approximations of irregular distributions. The user must specify n, an integer (n>1) number of ordered pairs to be read in, followed by the n ordered pairs. Within the ordered pairs, the first number is the value of the distribution; the second number if the cumulative probability associated with the value. The probabilities in the ordered , pairs must increase monotonically starting with 0.0 and ending with 1.0. The values must also l increase monotonically, t.HS then performs a linear interpolation ot this distribution function. If only two points are specified, a uniform distribution is generated between the two points. l User Defined Discrete Cumulative Distribution A discrete cumulative distribution 's used when the user has a discrete number of pessibilities
- that may occur. The user must specify an integer, n>1, which signifies the number of ordered pairs to be read in. The next n ordered pairs consist of the discrete value of the distribution with the cumulative probability associated with that value. The probabilities in the orriered pairs l must increase monotonically starting with a value greater than 0.0 and ending with 1.0. The values must also increase monotonically.
t-References C Fit User Guide, Version 1,
- Probability Distribution Fitting Software," Centre for Engineering .
Research Inc., Edmonton, Alberta, Canada,1996. J i 'l. 1 Residential 77 February 1,1998
DRAFT LETTER REPORT REVIEW OF PARAMETER DATA FOR THE NUREG/CR-5512 BUILDING OCCUPANCY SCENARIO AND REVISED PROPOSED PDFS FOR THE DandD PARAMETER ANALYSIS l l prepared by W. A. Hareland, F. A. Duren, E. Kalinina, W. E. Beyeler, D. P. Gallegos, and P. A. Davis Environmental Risk and Decision Analysis Department Sandia National Laboratories submitted to M. C. Daily Office of Nu; lear Regulator Research Radiation Protection and Health Effects Branch A Draft Letter Report for NRC Project JCN W6227 August 11,1997 O
1.0 INTRODUCTION
This draft letter report presents a review of parameter data for the NUREG/CR 5512 building occupancy scenaric and includes revised proposed probability distributions functions (PDFs) to be used in the second iteration of the DandD parameter analysis to define default parameter values. This introduction provides a discussion on the general process for dose assessment, the uncertainty in dose assessments, and regulatory decision making based on dose estimates. Regulatory decisions to terminate the license for a site will be based in part on estimated dose from residual radioactive contamination. These dose estimates are made using models for the transport and exposure processes that might occur at the site. Estimating dose at a particular site is a process that involves several key steps including the data collection, model development, parameter estimation, and dose calculation. Data collection entails finding and documenting information about the site relevant for estimating dose. This information might include direct measurements taken at the site, regionalized data from publicly available sources, and published literature aboid the site, the L .., rounding region, or ge ':erally applicable information about tranvort and exposure processt , that might occur. Data collection produces a body of site data containing allinfcrmation relevant to dose estimation. Defining default parameter values for a generic analysis requires collection of data relevant for estimating dose at the population of sites that will be using the generic analysis. For a generic anelysis of an individual site using the NUREG/CR 5512 default models and default parameter values, the only site data required are source term data. Attematively, other site data for developing site specific parameter values along with the source term data can be used with the default models for a more detailed analysis. Model development draws on the site data to define the transport and exposure processes that might occur at the site. Model development leads to a procedure for calculating dose based on these processes. For the NUREG/CR 5512 modeling, this step has been completed. The default models for the generic analyses and for analyses that may incorporate site specific parameter values were initially defined in NUREG/CR 5512, Volume 1, and have been implemented with some changes in DandD. Parameter estimation uses information in the site data to define values for parameters of the transport and exposure modela in addition to the numerical values found in the site data, the amount and type of available information, and the use of the calculated dose in regulatory decision-making, are important considerations in defining parameter values. The parameter values must be accepted during the NRC review process based on the site data provided. Dose calculation uses the model and its parameter values to estimate the dose that would result from residual contamination. This estimated dose is the basis for comparing site performance against the defined regulatory dose limit. This dose estimation proceso produces an estimate for the dose that would result from residual contamination. The estimated dose will generally differ from the real dose, by some unknown amount, because of the inevitable errors and approximations associated with the model. These errors and approximations create uncertainty about the real dose value. This uncertainty exists for severel reasons. In general, we always have incomplete knowledge about both current and future conditions at the site. 2 9
Assumptions are made in developing the scenarios, pathways, and models to represent the p transport and exposure processes that might occur at the site. Some processes may be only 1 I ( approximately represented, or may be omitted altogether. The model may rest on assumptions , that are not strictly satisfied at the site. For example physical properties may be assumed to be ' spatially uniform and constant, Or the model may be based on scenarios that describe stereotyped conditions rather than the actual site conoitions. The site data may not include information on all relevant physical properties. Some of the site data may be subject to measurement error. Too few measurements may be provided to completely characterize the variations in physical properties over the site or over time. Some data may have been collected from nearby or similar locations, rather than from the site being modeled. Differences between the idealized structure of the model(e.g., spatially uniform properties) and real site conditions (e g., spatial variations in measured values) create ambiguity about the appropriate parameter value to use. Errors in menurement can also introduce errors in parameter values. Also, many model parameters are not directly measured, but instead are estimated from the measure data. Errors may be introduced in this estimation process. If the available information about a specific parameter value is limited, this information may be cons! stent with a wide range of possible parameter values. Within the context of this process for dose assessment, this report reviews parameter data used for the NUREG/CR 5512 building occupancy scenario. Section 2.0 of this letter report presents an overview of the assumptions, parameters, and modeling for the building occupancy scenario. Section 3,0 presents discussions of each parameter in detail. The definition of the parameter value in NUREG/CR 5512, Volume 1 [ Kennedy and Strenge,1992), is presented, and values for corresponding parameters in RESRAD are noted. Then the PDF defined for the fG first iteration of the parameter analysis is discussed. For the second iteration of the parameter analysis, additional data and information are being considered to evaluate and revise, if necessary, the PDFs for the analysis. Defining appropriate PDFs for the DandD parameter analysis requires consideration of several pieces of information for each parameter, including the following: Importance to Dose: It is expected that all parameters are important for calculating dose. A general statement is made here with regard to the relationship of the parameter to the calculated dose.
. Use of the Parameter in the Modeling: How is the parameter used and how should it be represented in the model? . Parameter Uncertainty: What is the expected uncertainty for the parameter? . Variability Across Sites: Is it expected that the parameter value will vary across sites? . Site Data Collection: Is it expected that a licensee will conduct additional data collection activities to determine and support a site specific vane for the parameter.
L)' 3
Definition of the Site Data Source: What are the possible data sources to support the betermination of a site specific value? Should expected site specific or more general regional data be used to define the input distribution? NRC Interpretation of Site Specific Value. Do or wiil enteria exist that NRC will use as a bat is for accepting or rejecting a site specific value? ' For each parameter, this information is discussed. The first two items are included in the general background introductory discussion for each parameter. Additional data and
- information for revising the PDF are reviewed, and the use and uncertainty of these data and the apprcach for developing the respective input PDFs for the parameter analysis calculations are discussed. The last five items above are included as part of the discussions of the proposed PDFs. References are cited for each individual parameter. Finally, Section 4.0 provides a summary of the proposed PDFs for each parameter.
] l 2.0 ASSUMPTIONS AND PARAMETERS FOR THE BUILDING OCCUPANCY SCENARIO IN NUREGICR 5512 1 2.1 Scenario Assumptions i The following assumptions were made for the building occupancy scenario as defined in i NUREG/CR 5512 Volume 1 [ Kennedy and Strenge,1992), and implemented in interim Release 1.0 M DandD [Wernig et. al,1996): Radioactive dose results from exposure via three major exposure pathways: (1) external exposure to penetrating radiation from surface sources, (2) inhalation of resuspended surface contamination, and (3) inadvertent ingestion of surface contamination Four other potential exposure pathways are not included in the analysis: (1) external exposure during submersion in airborne radioactive dust, (2) It'ternal contamination from puncture wounds infected by contaminated surfaces, (3) dermal absorption of radionuclides, and (4) inhalation of indoor radon aerosol The building will be commercially used after decommissioning. The occupancy of the building will occur immediately after its release. The residual contamination will be represented by a thin surface layer left on the inner budding surfaces. The exposure type will be a long term chronic exposure to lo.v radioactive level contamination since major contamination will be cleaned up prior to decommissioning. 4
2.2 Scenario Parameters The modeling for the building occupancy scenario includes eight parameters:
. Extemal dose rate factor for exposure from contamination uniformly distnbuted on surfaces, DFES,(mrom/h per dpm/100 cm')
Inhalation CEDE factor, DFH, (mrem /pCl inhaled) ,
. _ ingestion CEDE factor, DFG,(mrem /pCl ingetted)
. Length of the occupancy period, te (d)
. Time that exposure occurs during the 1. year building occupancy period, to (d)
. Resuspension factor for surface contamination. RF,(md)
. Volumetric breathing rate, Vo (m'/h)
. Effecilve transfer rate for ingestion of removable surface contamination from surfaces to hands, from hands to mouth, GO (m2/h)
The first three parameters are radionuclide specific and are, in fact, dose rate conversion . factors used to translate medium cone:atration values into doses values. These parameter values are not asowed to be variable. Thus, the values specifLJ for each radionuclide are .o be constant for all the sites. However, a number of simplified assumptions (some of them are conservative, some of them are prudently conservative) were used to derive the corresponding values. The five re,maining parameters are not radionuclide specific and are defined in Volume 1 of NUREG/CR 5512 as parameters that may be varied in the analysis. 2.3 Overview of Modeling for the Occupancy Scenario The annual TEDE for a parent radionuclide in the building occupancy scenario TEDEO,is calculated as a sum of; e external dose resulting from external exposure to penetrating radiation from the surface cources represented by the parent and daughter (if any) radionuclides, DEXO,; e CEDE for inhalation resulting from inhalation of resuspended surface contamination represented by the parent and deughter (if any) rartionuclides, DHO,: and
. CEDE for ingestion resulting from inadvertent ingestion of surface contamination represented by the parent and daughter (if any) radionuclides, DGO, The mathematical formulation of the above is:
TEDEO, =DEXO,+DHO,+ DGO, (2.1) The calculation nf DEXR. DHO,, and DGO,is based on the estimation of the t;verage annual surface activity per unit area of the parent, C,, and daughter radionuctious, C,, during the first year of the bunding occupancy scenario. Since radionuclide activities diminish with time, the maximum annual TEDE will always occur during the first year of the scenario (assuming, of course, that the parent and daughter radionuclidos in a decay chain are in secular equilibrium), and there is no need in considering the following years. The average annual activity is b \ 5
determined as an integral of the radionuclide activities cunng the first yea. after tht,3 adit,g release over the length of the occupancy period, to, divided by the average tirn6, t,,, which is equal to one year (365 25 days). The release of the building is conservatively assumed to occur at time zero, and building occupancy is conservatively assumed to be at least one year (default value for teis 365 25 days), The mathematical formulation is as follows: t,o C.,,= 1/t,, f C,(t)dt = A,,*E in.,,, K,n[(1. exp(A,n'te)/A,n) (2.2) K,n(n.%,) = [ i,.n,.y [d,,*A,,*K,n]/(A yA,n) K, = C,(0)/A,-[in,y y K,, where A,,is the radioactive decay constant of radionuclide j, dp ,is the decay fraction, and C,(0)'s the initial activity of radionuclide J. In the above equation, C.,,= S{Co,,t io}/t.,, corresponding to '"JREG/CR 5512 Volume i notations. The expressions for DEXP, DHO,, anu UGO, are as follows: DEXO,= 24*to*[ o.u> DFES,'C ,, (2.3) DHO =45.05*24*to' RFo*Vo*[ o.u> DFHj'C,,, (2.4) DGO,=45.05*24*to*GO'[ o.u) DFG,'C.,, (2.5) where J,, RFo, Vo, GO, DFES,, DFH,, DFGj correspond to the number of radionuclides iri chain I, resuspension factor, volumetric breathing rate, effective transfer rate factor, extemal dose rate factor, inhalation CEDE factor, and ingestion dose fac' , respectively. Substituting Equations (2 3), (2 4), and (2.5) in (2.1), the annual TEDE can be expressed as: TEDEO,=24/365.25*io*E o.u>[C.,,'(DFES,+45.05*RFo*Vo*DFH,+45.05'GO'DFG,)) (2.6) As Equation (2.6) indicates, TEDE is directly proportional to the parameter to. The larger the time that exposure occurs during the building occupancy period, the higher the total dose. The same is also true for the total dose from a source represented by any combination of radionuclides. The total dose is not in direct proportion to the other parameters. However, increasing these parameter values will result in some (significant or not) increase in the total dose. The significance of the default parameters that are not radionuclide specife, such as RFo, Vo, and GO, will be different for different radionuclides and will depend on ratios of DFH, and DFES,, DFG, and DFES,, and DFH, and DFG, for a!! radionuclides in the chain. For example, if DFES,is s O
significantly larger that DFH, and DFG, for all radionuclides in the chain, then TEDEO, v,on't be
/m, sensitive to RFo , V , or GO.
v) 3,0 PARAMETER INFORMATION Section 3.0 provides a discussion of each of the parameters used to calculate the dose for the NUREG/CR E512 building occupancy scenario. The first four parameters will not be evaluated in the second iteration of the parameter analysis, so only discussion about the basis for these values is included. For the remaining four parameters, additional information has been reviewed to determine if attemative data or approaches can provide a defensible basis for the parameter PDFs, and revised PFDs are proposed. 3,1 F.xternal don rate factor for exposure from contamination uniformly distributed on surfaces, OFES,(mremlh per dpm/100 cm') The radionuclide specific external dase rates conversion factors are defined as suggested in the EPA Federal Guidance report No.12 [Eckerman and Ryman,1992). These factor- ~ ovide the extelual effective dose equivalent by summing the product of individut, organ doses and organ weighting factors over the body organs. For the building occupancy scenario, these factors are defined for an ir* nite surface (thin-layer) source condition. This source condition assumes the approximation of the non uniform residual contamination on building walls, ceilings, and floors by the uniform surface contamination on the floor of the room having infinite surface. This assumption is based on the earlier sensitivity study by Kennedy and Peloquin [1990). Relative dose rates obtained for rooms of different volumes with uniform and sonie
/ T non-uniform sources of contamination were compared with the dose rates obtained using
(,/ infinite flat uniform source. It was concluded that infinite flat uniform source provides a conservative estimate for the small rooms (less than 200 m') and prudently conservative estimate (about 15% smaller rates) for the larger rooms. However, the sensitivity study was performed using one radionuclide only (Co-60). A constant distance between floor and ceiling (3 m) was assumed. It also appears that only one or a few of possible non uniform distributions were considered. Although a number uf assumptions underlie the values defined for the external dose conversion l factors, these values have been obtained from a standardized dosimetry data base and have been determined to be appropriate for use in the NUREG/CR 5512 modeling. These values will not bo evaluated in the parameter analysis. l lMPORTANCE TO DOSE: Radionuclide specific, the sensitivity of this parameter will depend on values of DFH,, DFG,, RFe.Vo, and GO. The higher tha value ef DFES, for each of the radionuclides in the chain, the higher the total dose. USE OF PARAMETER IN MODELING: This parameter is used to calculate the external dose, i DEXO,, resulting from extemal exposure to penetrating radiation from an infinite surface source. The relationship between DFES, and external dose is described by the following formula: DEXO,= 24*to*[ pm DFES/C,y (3.1) I n [V \ 7
where J,is the number of radionuclides in chain I,13is the time that exposure occurs during the building occupancy period, and C,y is the average annual activity of the radionuclide j during first year of the building occupancy scenario. The higher the value of DFES,for each of the r1dionuclides in the chain, the higher the resulting external dose. ) PARAMETER UNCERTAINTY: Given the expected assumptions made to approximate these types of factors, it is expected thet the values defined for these conversion factors have a large amount of uncertainty. This uncertainty may be characterized in the EPA report [Eckerman and Ryman,1992), but has not been explicitly addressed for the NUREGICR 5512 modeling. VARIABILITY ACROSS SITES: These factors are radionuclide specific and would not vary from site to site. Variability may be related to differences in contaminant distribution on building surfaces. Different types of industrial activities at different buildings / sites could result in different contaminant distributions. For example, in some cases (predominantly gaseous releases of condensible materials), contaminants could be distributed uniformly over all surfaces while liquid contaminants would be on the floor. SITE DATA COLLECTION: It is unlikely that a liceri .a would conduct the types .I studies required to collect additional data to modify these conversion factors. DEFINITION OF SITE DATA SOURCE: Not applicable. NRC INTERPRETATION OF SITE SPECIFIC VALUE: Not applicable.
REFERENCES:
Eckerman, K.F., and J.C. Ryn:an,1992. " Dose Coefficients for External Exposure to Radionuclides Distributed in Air, Water and Soil," Federal Guidance Report No.12, U.S. Environmer,tal Protection Agency, Washington, DC. Kennedy, Jr., W.E., and R.A. Peloquin,1990. " Residual Radioactive Contamination from Decommissioning: Technical Basir for Translating Contamination Levels to Annual Total Effective Dose Equivalent," Draft NUREG/CR 5512, Volume 1, U.S. Nuclear Regulatory Commission, Washington, DC. 3,2 inhalation CEDE factor, DFHj (mrem /pCIInhaled) The radionuclide-specific internal inhalation dose rate conversion factors are defined as suggested in the EPA Federal Guidance report No.11 [Eckerman et al.,1988). These factors are intended for general use in assessing average individual committed doses for inhalation of radioactive rrwterials in any population that can be characterized by Reference Man. Although a number of assumptions underlie the values defined for the internalinhalation dose conversion factors, these values have been obtained from a standardized dosimetry data base and have been determined to be appropriate for use in the NUREG/CR 5512 modeling. These values will not be evaluated in the parameter analysis. 8
3 IMPORTANCE TO DOSE: Radionuclide specific, the sensitivity of this parameter will depend [ on values of DFES,, DFG,, R%, Vo, and GO. The higher the value of DFH, for each of the ,
\ radionuclides in the chain, the higher the total dose.
USE OF PARAMETER IN MODELING: This parameter is used to calculate CEDE for DHO, resulting from inhalation of resuspended surface contamination. The relationship between DFH3 and internal dose due to inhalation is desenbed by the following formula: DHO,=45.05*24*to' RFo*Vo*E g.m DFH,'C,4 (3.2) where J,is the number of radionuclides in chain I, to is the time that exposure occurs during the building occupancy period, C,y is the averags annual activity of the radionuclide j during first year of the building occupancy scenario, RFo is the resuspension factor, and Vo is the volumetric breathing rate. The higher the value of DFH,for each of the radionuclides in the chain, the higher the resulting internal itihalation dose. PARAMETER UNCERTAINTY: Given 'he expected assumptions made to approximate these i types of factors, it is expected that the values defined for thest. Jonversions factors have a large amount of uncertainty. This uncertainty may be characterized in the EPA report [Eckerman et al.,1988), but has not been explicitly addressed for the NUREG/CR 5512 modeling. VARIABILITY ACROSS SITES: These factors are radionuclide specific and would not vary from site to site. SITE DATA COLLECTION: It is unlikely that a licensee would conduct the types of studies required to collect additional data to modify these conversion factors. DEFINITION OF SITE DATA SOURCE: Not applicable. NRC INTERPRETATION OF SITE SPECIFIC VALUE: Not applicable.
REFERENCES:
Eckerman, K.F., A.B. Wolbarst, and A.C.B. Richardson,1988. ' Limiting Values of Radionuclide intake and Air Concentration and Dose Conversion Factors for Inhalation, Submersion, and ingestion," Federal Guidance Report No.11 EPA-520/188 020, U.S. Environmental Protection Agency, Washington, DC. Kennedy, Jr., W,E., and D.L. Strenge,1992. " Residual Radioactive Contamination from Decommissioning: Technical Basis for Translating Contamination Levels to Annual Total Effective Dose Equivalent," NUREG/CR 5512, Volume 1, U.S. Nuclear Regulatory Commission, Washingto' , DC. O Lj e
3.3 Ingestion CEDE factor, DFGj (mrem /pClingested) The radionuclide specific internal ingestion dose rate conversion factors are defined as suggested in the EPA Federal Guidance report No.11 [Eckerman et al.,1988). These factors ero intended for general use in assessing average individual committed doses for inhalation cf radioactive materiais in any population that can be characterized by Reference Man. Although a number of assumptions underlie the values defined for the internalingestion dose conversion factors, these values have been obtained from a standardized dosimetry data base and have been determined to be appropriate for use in the NUREG/CR 5512 modeling. These values will not be evaluated in the parameter analysis. IMPORTANCE TO DOSE: Radionuclide specific, the sensitivity of this parameter will depend on values of DFES,, DFH,, RFo, Vo, and GO. The higher the value of DFG, for each of the radionuclides in the chain, the higher the total dose. USE OF PARAMETER IN MODELING: This parameter is used to calculate CFDE for DGO, i Julting from inadverten: ingestion of surface contamination. The elationship vetween DFG, and intemal dose due to ingestion is described by the following formula: DGO,=45.05*24*to*GO[ o.mDFG/C.,, (3.3) where J,is the number of radionuclides in chain I, to is the time that exposure occurs during th's building occupancy period, C.,,is the average annual activity of the radionuclide j during first l year of the building occupancy scenario, and GO is the effective transfer factor. The higher the l value of DFG,for each of the radionuclides in the chain, the higher the resulting internal ingestion dose. PARAMETER UNCERTAINTY: Given the expected assumptions made to approximate these types of factors, it is expected that these parameter values have a large amount of uncertainty. This "acertainty may be characterized in the EPA Report [Eckerman et al.,1988), but has not been explicitly addressed for the NUREG/CR 5512 modeling. VARIABILITY ACROSS SITES: These factors are radionuclide specific and would not vary from site to site. SITE DATA COLLECTION: It is unlikely that a licensee would conduct the types of studies required to collect additional data to modify these conversion factors. DEFINITION OF SITE DATA SOURCE: Not applicable. NRC INTERPRETATION OF EITE SPECIFIC VALUE: Not applicable.
REFERENCES:
Eckerman, K.F., A.B. Wolbarst, and A.C.B. Richardson,1988. ' Limiting Values of Radionuclide intake and Air Concentration and Dose Conversion Factors for Inhalation, Submersion, 10
i - j and Ingestion," Federal Guidance Report No.11, EPA 520/183 020, U.S. i Environmental Protection Agency, Washington, DC. Kennedy, Jr., W.E., and D.L. Strenge,1992.
- Residual Radioactive Contamination from Decommissioning: Technical Basis for Translating Contamination Levels to Annual Total Effective Dose Equivalent," NUREG/CR 5512, Volume 1, U.S. Nuclear Regulatory ,
Commission, Washington, DC. 3.4 Length of the occupancy period, t,(d) The time parameter tto is used to determine the time integral of activity over the building occupancy perk >d, which is turn is used to determine the mean activity level of each radionuclide or decay chain. The default value for this parameter defined in NUREG/CR 5512, Volume 1, is 365.25 d or 1 year. Using 365.25 days in a year accounts for a leap year. This represents continuous use of a building for 100% of the calendar year so that, as stated in the regulatory criterion, annual TEDE is calculated. The RE6.dD value for the same parameter is 365.0 d. For the first iteration of the parameter analysis, this parameter was assumed to always be at its maximum possible value,365.25 d, because of the regulatory criterion to calculate annual TEDE. Because of the regulatory basis for defining this parameter, it will not be evaluated in the parameter analysis. IMPORTANCE TO DOSE: The 1" '*r the building occupancy period (this period can vary from ( 0 to 365.25 days), the higher tt m annual dose during the first year of the scenario. . USE OF PARAMETER IN MODELING: This parameter is used to calculate the average annual surface activity of radionuclide j per unit area C,4 during the first year of the building occupancy scenario. The relationship between C,y and to is described by the following formula:
.In C,y=1/t., f C,(t)dt = N*E cn.w K,n[(1. exp(6*t )/4) (3.4a) 0 K,n(n.mg)= [ <p.nfy[d,%,*K,nl/(N-() (3.4b)
K,=C,(0)/N-[ cnqu K g. f3.4c) where t., is the average period for dose calculation (365.25 days), N is the radioac ' te decay constant of radionuclide j, dgis the decay fraction, and C,(0) is the initial activity of radionuclide J. The larger the value for te, the larger the radionuclide aversgo annual activities. PARAEETER UNCERTAINTY: In general, uncertainty will exist because of lack of complete knowledge of actual building occupancy. However, the value for this parameter is defined to calculate dose based on an exposure duration of one year. ( 11
VARIABILITY ACROSS SITES: This parameter would not be expected to vary from site to site. 1 SITE DATA COLLEC . ON: it is very unlikely that a licensee would conduct any type of data colbetion activity data to modify this parameter. DEFINITION OF SITE DATA SOURCE: May not be applicable. NRC INTERPRETATION OF SITE SPECIFIC VALUE: May not be applicable.
REFERENCES:
Kennedy, Jr., W.E., and D.L. Strenge,1992. " Residual Radioactive Contamination from Decommissloning: Technical Basis for Translating Contamination Levels to Annual Total Effective Dose Equivalent,' NUREG/CR 5512, Volume 1, U.S. Nuclear Regulatory Commission Washington, DC. O
,2 e
i 3.5 Time that exposure occurs during the 1 year building occupancy period, t (d) (O) The time parameter, to, is the actual time spent on the job during the one year duration of the building occupancy scenario. The default value for this parameter defined in NUREG/CR 5512. Volume 1,is 83.33 effective 24 h days. This is calculated assuming that the actual time on the job is 100% of a work year during which a person spends 2000 h/y working in t! ) building (40 h work week for 50 working weeks with 2 weeks of vacation / sick leave /any other leave) The RESRAD value for the same parameter is 183 d [REF). The basis for this value requires further investigation for comparison to the value to be defined for DandD. For the first iteration of the parameter analysis, this parameter was assigned a beta distribution with lower and upper limits of 41.67 and 125 effective 24 h days, respectively, with a median value of 83. The lower limit is based on a person working half time or 20 h/wk, and the upper end is based on a person working 60 h/wk. The median value is based on a 40 h work week. l A beta distribution was selected so that specific endpoints could defined and therefore the tails ! for values less than 20 h/wk and more than 60 h/wk would not be considered. A plot of the beta l distribution !s shown in Figure 3.5.1. i l D ets { 0 03 o - I N 0 016 1 5 1 I 0 01 ! k l 1 0 006 0 40 60 60 70 00 00 100 110 120 f o ld) Figure 3.5.1. Beta distribution used for tofor first iteration of the parameter analysis. IMPORTANCE TO DOSE: The total dose is directly proportional to the time of exposure during the building occupancy period. USE OF PARAMETER IN MODELING: Definition of this parameter and its use in the modeling depends on the specific definition of the " Average Member of Critical Group." This parameter is () 3.5.1
1 l used to calculate the total dose, TEDEO,, from parent radionuclide I and its daughters due to external exposure to surface contamination, inhalation of resuspended surface contamination, and inadvertent ingestion of surface contamination dunng the first year of the building occupancy scenano. The relationship between TEDEO, and to is described by the following formula: l TEDEO,=24/365.25*to*T n.mC.,'(DFESp45.05*RFo*Vo*DFHp45.05'GO*DFG,) (3.5) where J,is the number of radionuclides in chain I, C,g is the average annual activity of the radionuclide j during first year of the building occupancy scenario, RFo is the resuspension factor, Vo is the volumetric breathing rate, GO is the effective transfer rate factor, DFES,is the external dose rate factor, DFH is the inhalation CEDE factor, and DFG) is the ingestian dose factor. An increase in the to vafue results in a proportionalincrease in the annual total dose value. 3.5.1 Additionalinformation Reviewed to Define A Revised PDF for te
- ,r the second iteration cf the parameter analysis, additionalinfor .iation was reviewed to determine if other data and approaches were available to provide a defensible basis for constructing a PDF for to for use in the analysis. For t ,o the additional information reviewed included Bureau of Labor Statistics (BLS) data on hours worked [BLS,1996a; BLS,1996b; and BLS,1997) and relevant references cited in the Draft EPA Exposures Factors Handbook [1996]
that discussed studies of human activity pattern. The following sections summarize the data and information available from these sources. 3.5.1.1 BLS Data on Hours Worked in June,1996, as part of the work to follow up on NRC review of the parameter analysis report, data from the BLS Current Population Survey (CPS) were obtained from the BLS website. The CPS is a monthly survey and during 1995 was sent out to approximately 50,000 households a month, and was used to obtain informaten for about 94,000 persons ages 16 years and older [BLS,1996a). During 1996, the approximately 56,000 household units were surveyed, and information was obtained for about 107,000 persons ages 16 and older [BLS,1997). Annual averages from the CPS are also published in January issues of Employment and Eamings [BLS,1996a; BLS 1997), The CPS is used to determine ' Characteristics of the Employed" statistics, including hours worked. Current data for
- Characteristics of the Employed" can be accessed from the BLS home page for " Labor Force Stat!stics from the Current Population Survey
- at the website http:// stats.bts. gov /cpshome.htm. The specific page for the data listings on ' Characteristics of the Employed" is located at http:// stats. bis gov:80/cpsaatab.h#charemp, and can be accessed from CPS home page. In June 1996, the data for persons at work in agriculture and non agriculturalindustries by hours of work for 1995 were downloaded and are presented in Table 3.5.1, These data are also published in the January 1996 issue of Employment and Eamings. The reported data range is from 1 to 4 h/wk of work to 60 h/wk and over. The 1995 overall annual average reported is 39.3 h/wk. In Aprii 1997, the data from the 1996 Annual Average Tables (BLS,1997) were also reviewed. The actual numbers were slightly different, 3.5.2
but the percentages for each range did not differ by more than two tenths of a percent. The other available BLS data are from the National Current Employment Statistics (CES). These statistics are determined from a industry survey of employers that report man hours, number of employees, and payrollinformation, but do not report anything about part time or
; full time employees. The website location for these statistics is http:// stats.bls. gov:80/cgi bin /surveymost?ee. In June 1996, several series of data related to i national employment, hours, and eamings, were downloaded and reviewed, including the following:
. -Total Private Average Weekly Hours of Production Workers Seasonally Adjusted
. Total Private Indexes of Aggregate Weekly Hours - Seasonally Adjusted
. Total Private Average Weekly Hours of Production Workers - Not Seasonally Adjusted 4 . Goods producing Average Weekly Hours of Production Workers Seasonally Adjusted
- e Goods producing indexes of Aggregate Weekly Hours Seasonally Adjusted .
. - Mining Average Weekly Hours of Production Workers Seasonally Adjusted j . Manufacturing Average Weekly Hours of Production Workers - Seasonally Adjusted i e Manl.acturing Average Weekly Overtime of Production Workers Semanally Adjusted l
. Manufacturing indexes of Aggregate Weekly Hours - Seasonally Adjuste:f
- . Private Service producing Average Weekly Hours of Production Workers - Seasonally Adjusted Privsta Service producing indexes of Aggregate Weekly Hours - Seasonally Adjusted
. Transporu tion and Public Utilities Average Weekly Hours of Production Workers -
Seasonnlly Adjusted (
\
Wholesale Trade Average Weekly Hours of Production Workers - Seasonally Adjusted Retail Trade Average Weekly Hours of Production Workers Seasonally Adjusted
- These data include the average weekly hours for each month during a year. These data for all years available, which varied for each category, were also reviewed. During 1995, the average weekly hours by month for Total Private Industry - Seasonally Adjusted ranged from 34.2 to 34.7 h/wk, Historically (1964 through May 1996), the range is 34.2 to 38.9 h/wk. The 1995 average weekly hwrs by month for Manufacturing - Seasonally Adjusted ranged from 41.2 to 42.2 h/wk. Historically (1932 through May 1996), the range is 34.2 to 42.2 h/wk. The 1995 average weekly hours by month for Transportation and Public Utilities - Seasonally Adjusted ranged from 39.1 to 39.8 h/wk.- Historically (1964 through May 19%), the range is 37.8 to 41.5 h/wk. Other historical CES data for annual average hours of production and non supervisory workers on non farm payrolls for 1986 through 1994 report ranges from 34.3 to 34.8 h/wk for the private sector,40.7 to 41.4 for manufacturing, and 38.6 to 39.9 for transportation and public utilities (BLS,1996b].
% 3.5.3
Table 3.5.1 1995 data for " Persons at work in agriculture and nonagricultural industries by hou.s of work,"(BLS,1996a) Thousands of Persons Percent Distnbution Hou.s of Work All Agriculture Non- All Agriculture Nonagricultural Industries agncultural industries Industries Industries Total Persons at 119,318 3,247 116,071 100.0 100.0 100.0 Work,16 years and over 1 to 34 hours 30,664 1,051 29,613 25.7 32.4 25.5 1 to 4 hours 1,297 83 1,214 1.1 2.6 1.0 5 to 14 hours 4,943 262 4,681 4.1 8.1 4.0 15 to 29 hours 15,120 476 14,6*, 12.7 14.7 12.6 30 to 34 hours 9,304 229 9,075 7.8 7.1 7.8 35 hours and 88,654 2,196 86,458 74.3 67.6 74.5 over 35 to 39 hours 8,783 173 8,610 7.4 5.3 7.4 40 hours 42,228 635 41,592 35.4 19.6 35.8 41 hours and 37,643 1,388 36,255 31.5 42.7 31.2 over 41 to 48 hours 13,958 250 13,708 11.7 11.8 7.7 l 49 to 59 hours 13,591 388 13,203 11.4 11.9 11.4 00 hours and 10,094 750 9.344 8.5 23.1 8.1 over Average hours, 39.3 42.2 39.2 - - - total at work Average hours, 43.4 49.7 43.2 - - - persons who usually work full time 3.5 A
3.5.1.2 Data from Studies on Human Activity Patterns (% ( ) The EPA Draft Exposure Factors Handbook icPA,1996] includes a summary of several studies on human activity patterns, including some information that may be relevant for estimating oc:upancy duration. The discussion for each cited study outlines the methodology anci type of data collected, and discusses the strengths and limitations of each. The following is a discussion of four studies and the data relevant to occupancy duration. Although, several of the actual references for the four studies discussed have not been directly reviewed, it appears that many of the tables directly from each of these studies have been reprinted in the EPA handbook. Additional work will continue on reviewing the references for these studies so that they may be directly cited for the parameter an 'ysis report. in each of these studies, data on time use was recoroed for either the preceding 24 hours, based on recall (telephone surveys) or for the succeeding 24 hours (based on diaries). In some studies (e.g. [ Hill,1985]) the same respondents appear to have been polled in successive
" waves" throughout the year. This follow-up aside, the data cited in these studies provides information on the variability over tho cample population of time spent during a single day. The occupancy duration parameter would presumaoly be based o average behavior of indivic' Jals over the year. This average is of course not the same as the average of daily behavior over a population of individuals. To estimate the former, information on the variability in daily activities for single individuals may also be required. The studies that include follow-up surveys may have collected this information, however it does not appear to have been published.
Robinson and Thomas [1991] report population averages for time spent performing various O activities (e.g. " Paid Work", " Household Work") and in various microenvironments (e.g. ('j " Restaurant /Bar'," Work / Study nonresidence"). Data from Californians (1762 respondents ages 12 and older collected between October 1987 and August 1988) and from a National sample (5,000 respondents across the US ages 12 and older collected in during January through December 1985) are disting(shed, as are gender subpopulations. Separste statistics are also reported for " Doers" of an activity as distinct from the general population (for example time spent cooking by people who actually cook). Population statistics are not reported, however the standard error of the mean is given in some cases. Mean time spent in paid work for ages 18-64 years ranged from 150 m/d (34.31 effective 24-h d/y) for women in the national survey to 346 m/d (62.47 effective 24-h d/y) for men in the California study. These numbers correspond to 15.83 h/wk and 28.83 h/wk, respectively. (Note: Other than the Volume default value, effective 24-h d/y are calculated based on 52 wk/y; weekly hours are based on a 5-day work week.) Given the age range for the survey, a significant portion of the survey population must be part-time workers. The mean time for " doers" spent in the work / study-other microenvironment in the total population (ages 12 years and older) ranged from 383 to 450 m/d (69.15 to 81.25 effective 24-h d/y); during the weekday, ranged from 401 to 415 m/d (72.40 to 74.93 effective 24-h d/y); and for ages 24-64 ranged from 410 to 429 m/d (74.02 to 77.46 effective 24-h d/y), respectively. The range of all these values (383 to 450 m/d) corresponds to 31.92 to 37.50 h/wk. With further investigation, some of the information on " doers" may provide additional support for in defining the range for a PDF. Tsang and Klepeis [1996] contains information from the largest and most current human activity pattem survey available. The survey was conducted by the EPA. Data from 9,386 respondents V 3.5.5
in the 48 contiguous states were collected via minute-by-minute 24 h diaries betweer October 1992 and September 1994. The editors of the Handbook appear to regard this study as the most reliable source of information. Distributions are reported for the number of minutes spent I working for pay, the number of minutes spent in a " main job," the number of minutes spent indoors at work, the number of minutes spent in at a plant / factory / warehouse, and the numb 3r , of minutes spent in an o# ice or factory. Distributions are provided for the entire sample populations, as well as sub-populations defin'd by gender, race, employment status, region, season, and other factors. The mean 24-h cumulative number of minutes in a main job for full-time employees is 504.350 m/d (standard deviation = 164.818), which corresponds to 91.06 effective 24-h d/y and 42.03 h/wk. 1 Robinson (1977) compares average time spent in " Work for Pay" in 1965 and 1975. Averages are reported by gender, employment status, age, race, and education. These data are not as ' current as the two previous sources. For four age categories spanning 25-65 years of age, these averages ranged from 29.2 to 35.9 h/wk in 1965 and 20.4 to 34.4 h/wk in 1975. Hill [1985] reports average time spent at " Market Work" from data collected during the jd-1970s for subpopulations defined by gender, region, day of 11.: week, ano season. l ! Distributions are not provided, however sample standard deviations are given for some ! quantities. Weighted mean hours per week (standard deviation) for married men and women working full-time were 47.84 (16.54) and 38.55 (16.87), respectively. Data on seasonal variations appear to have been obtained by re-sampling the same population. If so, some limited information on variability of annual individual activity might be found in Hill's original data. 3.5.3 Preposed Distribution for to The data used to develop the PDF for the second iteration of the parameter analysis is based l on the BLS CPS 1995 data [BLS,1996a] for hours worked by persons at work in nonagricultural industries presented in column 4 of Table 3.5.1 of this report and is presented in Figure 3.5.4. These data are representative of annual estimates for the entire US worker population and are determined from the largest sample of data that has been collected and processed in a standardized manner for almost 40 years. This survey represents hours worked by persons working 1 to over 60 hours per week and was chosen over the survey data representing hours worked by persons working from 15 to over 60 hours per week [BLS,1996b) because it represents a broader range of workers. However, a PDF for the BLS,1996b data is presented I as an option in Figure 3.5.5 in the event NRC believes that the 15 to greater than 60 hour work l week is more representative of the critical group. I j Histograms for the 1995 [BLS,1996a] data are presented in Figures 3.5.2 and 3.5.3. The data l representing nonagricultural workers (Finure 3.5.3) are used to define the PDF that is associated with the average member of critical group. As indicated in Table 3.5.1, significant l portions of the working population in nonagricultural industries work less than or more than 40 h/wk. Only 35.8% of workers in nonagricultural industries work 40 h/wk; 27.8% work 15 to 39 h/wk and 23.2% work 41 to 59 h/wk. From Table 3.5.1, the 1995 weekly average for persons who usually work full time for nonagricultural industries is 43.2 h/wk. 3.5.6
)
l l l l
Given that the values in Tab'e 3.5.1 are reported for ranges of hours worked, an empirbal piece-wise uniform distribution across each intervals has been constructed. For example, the v range of effective 24 h d/y for 1 to 4 hours is 2.17 to 8.67 across three 1-h intervals. Then, the probability density for this range is the fraction of the workers who work this range of hours divided by the number of intervals for this range of hours (for nonagriculturalindustries, this number is 0.0015 ((1214/116071)/3). The densities for the other ranges are calculated in the same manner. As mentioned previously, two distributions for these data (in terms of effective 24 h dly) are presented, one in Figure 3.5.4 for a range of 1 to 60+ hours /wk (2.17 to 130 d/y) and another in Figure 3.5.5 for a range of 15 to 60+ hours /wk (32.5 to 130 d/y). The mean for the first distribution for the full range of the BLS data is 84.48 d/y (38.99 h/wk). The mean for the second distribution for the narrower range is 88.95 d/y (41.05 h/wk). As stated previously, the first distribution (Figure 3.5.4) is recommended because it represents the widest range of potential works. Again however, NRC may chose the second PDF (Figure 3.5.5) is they believe that it is more representative of the critical group. PARAMETER UNCERTAINTY: In general, uncertainty about this parameter exists because of a lack of complete knowledge about the hours worked by workers throughout the entire populatic.. of regulated sites. In addition, uncertainty in parameter exits beause no complete survey of the annual average number of hours worked by all workers has or can be performed.
- However, the BLS data used for the PDF are representative of annua l estimates for the entire US worker population and are determined from the largest sample of data available that has been collected and processed in a standardized manner for almost 40 years. The BLS CPS covers about 92% of the decennial census population. Also, a sample rotation scheme allows
/ \
u.m u.a. . u.m u.m u.m i .. _ i i u..a _ u.m _ MM -- -
' . " . ' . 'E
, .'.lI .
u 'II .
.'!I .'.!I ."
Figure 3.5.2. Histogram of 1995 hours worked for allindustries [BLS,1996a] 3.5.7
e 3. 1
... T .t n. g 1
i ..
}.. -
.c I-muut -
.. . .. n n .. . .. ..
u n u n .. .. n .... Figure 3.5.3 Histogram of 1995 hours worked for nonagricultural industries [BLS, 1996a). O O 40 . - - - - - - ~ - - - - - - - - - - - - - - - - - - - - - 0 35 I 0 30 ,,
# 0 25 h
p 0 20 ff 5 ti I "" !! o 0 10 ' {l O 05 l' O 8 2 3 2 2 0 8 0 0 S R
" O N A 0 S 5 : y 5 Time in Building, to (dty)
Figure 3.5.4 PDF based for hours worked (1 to 60+) from BLS CPS [BLS,1996a). 3.5.8
A 0.40 0 O.35 0.30 , fe 0.25 5 g 0.20 2 0.15 O.10 0.05 4
- 0. 00 **** * * * ** * * * * *
- 8 E D 8 9 S S E D 8
& !:i # 8 $ d $ Bi d Si $
Time in Building, to (dly) Figure 3.5.5. PDF based on hours worked (15 to 60+) from BLS CPS [BLS,1996a). for 50% of the sample to be common from year to year. Thus, the uncertainty of the data in terms of sampling and non sampling error and historical comparability is minimal and well-characterized (BLS.1996a). VARIABILITY ACROSS SITES: The PDF proposed in Figure 3.5.4. represents the variability of worker hours across different industries and different regions of the country, in support of this data, BLS does present data for a wide range of industries (See Section 3.5.1.1). i SITE DATA COLLECTION: For this parameter, other BLS or similar data sets may provide the basis for a licensee to develop a different distribution of hours worked for a site-specific critical group. For example, a licensee may propose that the primary use of the building following license termination wi!! be for manufacturing. Then, the licensee may use the BLS data to define the range of expected hours for the dose assessment. However, the licensee will need to provide the NRC with the assurance that the building will only be used for manufacturing over the regulated time period. 3.5.9
DEFINITION OF SITE DATA SOURCE: The discussions above about the BLS data and the information from the EPA Exposure Factors Handbook may provide a basis to define the site data source. NRC INTERPRETATION OF SITE SPECIFIC VALUE: May not be applicable.
REFERENCES:
Bureau of Labor Statistics,1996a. " Annual Household Data," in Employment and Eamings, Vol. 43, No.1, January 1996. Data also downloaded in June 1996 from BLS internet site http:// stats. bis.gove:80/cpsaatab.htm#charemp. Bureau of Labor Statistics,1996b. " Current Labor Statistics," in Monthly Labor Review, Vol. 119, Nos.1 and 2, January / February 1996. Bureau of Labor Statistics,,1997. " Annual Household Data," in Employment and Eamings, Vol. 44, No.1, January 1993. Data also downloaded in April *997 frm BLS inte iet site http:// stats.bls.gove:80/cpsaatab.htm#charemp. EPA,1996. Exposure Factors Handbook. EPA /600/P 95. Office of Research and Development, Environmental Protection Agency, Washington, DC. (Current draft not citable.) Hill, M.S.,1985. " Patterns of time use," in Juster, F.T., and F.P. Stafford, eds., " Time, goods, and well being," University of Michigan, Survey Research Center, institute for Social Research, Ann Arbor, MI, pp. 133-166. Kennedy, Jr., W.E., and D.L. Strenge,1992. " Residual Radioactive Contamination from Decommissioning: Technical Basis for Translating Contamination Levels to Annual Total Effective Dose Equivalent," NUREG/CR 5512, Volume 1, U.S. Nuclear Regulatory Commission, Washington, DC. Robinson, J.P.,1977. " Changes in Americans' use of time: 1965-1975. A progress report." Cleveland State University, Communication Research Center, Cleveland, OH. Robinson, J.P., and J. Thomas,1991. " Time spent in activities, locations, and microenvironments: A California-National Comparison Project Report, U.S. Environmental Protection Agency, Environmental Monitoring Systems Laboratory, Las Vegas, NV. Tsang, A.M., and N.E. Klepeis,1996. "Results tables from a detailed analysis of the National Human Activity Pattern Survey (NSAPS) response," Draft report prepared for the U.S. Environmental Protection Agency by Lockheed Martin, Contract No. 68-W6-001, Delivery Order No.13. 3.5.10
3.6 Resuspension factor for surface contamination, RFo (m') () The resuspension factor, RF., defined for the NUREG/CR-5512 dose modeling defines the ratio of contaminant concentration in inhaled air to surface contamination concentrations. The model uses a single, constant (time invariant) value. This value should therefore represent the effective value for the average member of the critical group over the one year duration of the building occupancy scenario. The default value for the resuspension factor recommended in NUREG/CR 5512, Volume 1, is 1x10* m', based on a literature analysis of studies published from 1964 through 1990. The overall range of values obtained from these literature sources is 2x10" to 4x10' m'. However, most data referenced are not for indoor conditions (wind stress and vegetation). Only two of the references cited in Volume 1 provide data for indoor resuspension. The first of these, an lAEA technical report [1970), reports a value of 5x104 m' which has been obtained for operating nuclear facilities. The second of these two references, a review by Sehmel [1980], provides different resuspension factors depending on the type of activity conducted within the rooms of the building (walking, vigorous sweeping, and fan). The overall range cited by Sehmel is from 1x104 to 4x108 m4 The lower end of this range is suggested as a default based on the fact that surfaces are assumed to be cleaned of eseily removabit contamination at the moment of license termination. For the initialiteration of the parameter analysis, the resuspension factor was assigned a log-uniform distribution with a lower limit of 1x10*m' and an upper limit of 1x10* m4 Following decommissioning, removable surface contamination is assumed to be insignificant. RF,is intended to represent a long term average related to normal activities in a commercial facility. Any contamination is assumed to be uniformly distributed over surfaces, and this range appears p) i representative of values cited in NUREG/CR-5512, Volume 1, given the preceding assumptions. A loguniform distribution was selected to emphasize the default value defined in Volume 1. The range of the distribution (two orders of magnitude) was thought to be representative of the values cited in NUREG/CR-5512, Volume 1, for resuspension caused by mechanical disturbances and resuspension measured in indoor conditions. A plot of the PDF is presented in Figure 3.6.1. s g.i i g .. e L"
., .. . 4. 4. . .
t ..., n ,% Figure 3.6.1 PDF for resuspension factor for first iteration of the parameter s analysis. U 3.6.1
IMPORTANCE TO DOSE: Resuspension is important to dose because the higher the value for RF , the higher the total annual dose during the first year of the building occupancy scenario. USE OF PARAMETER IN MODELING: This parameter is used to calculate CEDE for DHO, resulting from inhalation of resuspended surface contamination. The relationship between RF, and internal dose due to inhalation is described by the folluwing formula: , DHO,=45.05*24*t,' RF,*V,* 3 y DFH,*C , (1) 1 I where J,is the number of radionuclides in chain i, to is the time that exposure occurs during the building occupancy period, C,is the average annual activity of the radionuclide j during first year of the building occupancy scenario, DFHj is the inhalation CEDE factor, and V,is the volu-metric breathing rate. The resulting internalinhalation dose is directly proportional to the re-suspension factor. 3.6.1 Additional Information Reviewed to Define A Revised PDF for RF, The parameter analysis requires a distribution descrio...g the variability of site-spec..ic values for this parameter over licensed sites. To define this distribution, an applicant is assumed to have detailed information about (or control over) factors effecting resuspension at their site, such as the activities of om.cupants. This information would be used to define a critical group for the site by selecting a sub-set of occupants exposed to a relatively high concentration of resuspended contaminants. RF, would then be defined as the time-weighted average resuspension factor for this group over the one-year scenario duration. For the second iteration of the parameter analysis, an extensive literature review was conducted to identify any developments in the understanding of the resuspension process since the review reported in NUREGICR 5512 in 1992, and to identify data or approaches that could be used to develop a probability distribution function for the indoor resuspension factor. Older publications that were not referenced in NURFG/CR-5512, Volume 1, were also reviewed for the same purpose. Resuspension factor values are reported in a number of studies published between 1964 and 1997. Reported values for resuspension fcetors vary over a wide range, f om approximately 10" m' to appremmately 108 m' The review of some older publications indicate that the value of resuspension factor of 1x10* m' was used in the development of general guidelines, and has been seen as a general value having a reasonable factor of safety for hazard evaluation and design purposes (Brodsky,1980). This value was also recommended by the IAEA [1982; 1986) and suggested as an average for Europe in Garland [1982). These sources support (but were not cited to justify) the default parameter value for RF,in NUREG/CR-5512, Volume 1. Most studies, and all but one study not included in the review reported in NUREG/CR-5512 Volume 1, provide data on outdoor resuspension factors. These values are not directly relevant for the occupancy scenario model. Additionally, most reported resuspension factor values were measured or inferred under conditions that would not reasonably be sustained during the one-year exposure period. The different time scales of the experimental conditions and the scenario model must be considered in determining site-specific values for RF.. Published estimates of resuspension factors and resuspension rates under indoor conditions, identified during the current literature review, are summarized in Table 3.6.1. The results of this review are presented in the same format as the earlier review published in NUREGICR-5512, Volume 1. The reported values from these sources range from 2x10' to 4x108 m-' With one 3.6.2
exception ([ Thatcher,1995]), no recent information on indoor resuspension was found. This f] most recent study provides estimates of resuspension rates of aerosols measured in a (j California residence under controlled indoor conditions. However, these rates cannot be directly translated into resuspension factor values. Table 3.6.1, Reported Information for Indoor Resuspension Condition / Reference Range (m 1) Comments Wind stress and mechanical 2x10~' - 5x10-5 Resuspension of loose Pu-disturbances, [ Jones,1964) -nitrate particles deposited on various surfaces Wind stress and vehicular 1x10.s - 1.5x104 Resuspension from Pu-and mechanical dis- -contaminated surfaces; 0.2% to turbances, [Glauberman, 10% removable by smear 1964) sampling Wind sness (Brunskill,1964) 2.5x104 - 3.9x10'3 Rt Jspension of radionuCulide contaminants from clothing in change room Vigorous mechanical 1x10 2 - 4x104 Resuspension of BeO on disturbance (sweeping), contaminated wood floor; -4% (Mitchell,1964) removable by smear sampling (,m) Vigorous mechanical 9.4x10 - 7.1 x104 Redistribution of loose thorium (/ disturbance (sweeping) oxide and thorium metal aerosol [ Fish,1964) particles, ZnS and CuO particles on stainless steel surfaces Indoor Residence 1.2x10" - 1.0x104 Resuspension in a California [ Thatcher,1995] see d residence (Note: These values cannot be directly translated to resuspension factors.) Various factors affecting resuspension, underlying the range of reported values, have been proposed. The effects of some factors are quantified in some studies, while other effects are discussed qualitatively. Although many studies consider the factors affecting outdoor re-suspension, these factors have analogs in the indoor transport pathway model. Such studies are therefore relevant for understanding potential variations in RF, across sites. The wide range of reported resuspension factor values is due to differences in measurement techniques and to variability in physical factors that affect resuspension both witt'in and among studies. These sources of variability in reported resuspension factor value are described in more detail below. The common measurement techniques for determining indoor resuspension factors are: e direct measurement of contaminant concentrations on surfaces and in the air (Jones,1964; Glauberman,1964; Brunskill,1964; Mitchell,1964)
- redispersion of settled particulates [ Fish,1964]
h' V 3.6.3
e recoil of " hot-atoms" during decay of radionuclides [Leonaro,1995) In addition to differences in experimental technique, measured values of resuspension factor may vary due to spatial variability of surface contaminant concentrations, variability of concentrations in air with location and with elevation, and spatial variations in surface texture leading to location-dependent resuspension. These variations can create uncertainty in the effective value of resuspension factor as estimated by the ratio of concentrations measured in air and on the contaminated surface. A large number of physical factors can affect resuspension. According to IAEA [1992), the major factors are the following: e time since disposal e type of disturbance (air flow or mechanical) e inter 3ity of disturbance (air flow speed, traffic intensity) e nature of surface (texture, composition, surface area) e surface moisture particle size distribution e climatic conditions (temperature, humidity, wind) e type of deposition process (wet or dry) e chemical properties of the contaminant e surface chemistry e topographic features The potential effects of some of these factors on resuspension have been quantified, while only qualitative characterizations are available for others. As discussed above, some studies discuss the effects of these factors on outdoor resuspension factors. While values of outdoor resuspension factors are not appropriate for the occupancy scenario model, reported effects of variations in physical conditions (e.g. air flow) on relative resuspension factor values do provide useful information about potential variations in indoor resuspension factor values due to variations in the occupant's behavior or environment. Surface moisture and climatic conditions are factors that may influence resuspension in outdoor conditions but are assumed to be irrelevant for indoor resuspension. These factors are therefore not considered in the following discussion. For the other factors listed above, the studies cited in NUREG/CR-5512, Volume 1, and Fish (1964), Jones, [1964), Brunskill [1964), Glauberman [1964), and Mitchell [ 1964) were reviewed to better understand the factors controlling resuspension factors. The following i discussion considers both outdoor and indoor conditions, but consideration of these factors especially as they are reported for indoor conditions is emphasized. Time since disoosal The parameter RF is constant with time, however several studies model variations of re-suspension factor with time, including Kathren [1968), Langham [1969,1970), NRC [1975), IAEA [1982,1986), Garland [1982), and Nair [1997). All of these models produce a decrease in resuspension factor with time, reflecting the experimentally observed decrease in contaminant air concentrations with time over contaminated areas. The difference between the initial re-suspension factor and the resuspension factor at a later time according to these models. Rather than a decrease in resuspension factor per se, this observed decrease in air concentrations may instead oe due to overall depletion of surface contamination (e.g., 3.6.4
l downward migration of contaminants, downwind transport of resuspended contaminanti, and other removal processes). The observed decrease might also be due to preferential depletion []v of easily suspended contaminants. All discussions of reduction of resuspension factor with time found in the literature survey pertain to outdoor resuspension. No information on the potential time-variation of indoor resuspension factors was found. Tyoe of disturbance (air flow or mechanical) Resuspension factors determined under conditions with mechanical disturbance can be at least one order of magnitude higher than resuspension factors determined under conditions where only wind resuspension occurred (Nair,1997; Mihaila,1994; Stewart,1994 ; Thatcher,1995: Nicholson,1989; and IAEA,1992). Among studies reporting indoor resuspension factors, the higher resuspension factors provided in Brunskill[1964), Glauberman [1964), and Mitchall[1964) were measured when distmbances significat.Jy more severe than in normal operating conditions were applied to obtain measurable contaminant concentrations and when most of the surface contamination was a loose, easily removable, contamination (spills on the floor). Fish (1964] reports a difference in resuspension factor of 1.5 orders of magnitude due to the type of activities in the room. Intensity of disturbance (air flow soeed traffic intensitv) Anspaugh [1975) suggests that contaminant concentrations in the air are proportional to the ( ' power of the friction velocity which is, in turn, proportional to the horizontal wind velocity. Consequently, the difference of 1 order in magnitude between the wind speed may result in difference of a few orders of magnitude in resuspension factors. The power law relationship between the wind speed and resuspension factor is also demonstrated by Hollander [1994).
' Among studies of indoor resuspension, Fish (1964] observed a power law relationship between the resuspension factor and the air velocity in the room, and Jones 1.1964] reports variations in resuspension factor due to different walking speeds.
Nature of surface (texture. comoosition. surface areal The magnitude of the influence of this factor on resuspension was not quantified in the literature. In a study of indoor resuspension, Glauberman [1964] attributes a difference in resuspension factors of one order of magnitude to differences in roor aize. Particle size distribution it is suggested by Hinton [1995] that resuspension is greatest for particles with diameter smaller than 125 :m and it is suggested by the IAEA [1992] that resuspension factor increases with particle diameter in the range from 1 to 5 :m. The resuspension factor is also correlated to the particle diameter. In Sehmel [1980), however, it is suggested that further studies are needed. In a study of indoor resuspension, Fish [1964) reports a strong correlation with particle diameter. O m 3.6.5
Tvoe of deoosition orocess (s.t or drv) Among studies reporting indoor resuspension factors, the higher resuspension factors provided in Brunskill[1964), Glauberman [1964), and Mitchell[1964) were measured when disturbances significantly more severe than in normal operating conditions were applied to obtain measurable contaminant concentrations and when most of the surface contamination was a loose, easily removable, contamination (spills on the floor). Chemical orocerties of the contaminant The difference between resuspension factors determined in the same conditions for different radionuclides.is one order of magnitude, but could be significantly smaller as discussed by Hartmann [1989) and the IAEA [1992). Among studies reporting indoor resuspension factors, Jones [1964) reports variation of the resuspension factor within one order of magnitude depending on the contaminant. Surface chemistry Although cited by the IAEA [1992) as a factor influencing resuspension, no specific information on the effect of surface chemistry on resuspension factor was found in the literature. Toooorachic features No specific information on the effect of topography on resuspension factor was found in the literature. For outdoor resuspension, topographic variations would presumably create variations in near-surface wind speed, leading to variations in the effective resuspension factor. An analogous e'fect might occur for indoor resuspension due to the placement of ventilation ductwork and furniture. The main conclusions of this literature review are: e the new data on resuspension factors falls into the same range that was noted in NUREG/CR 5512, Volume 1; however, the low end of the range is two orders of magnitude higher (1x10-8 to 1x10-2 m-1); e no significantly new models of resuspension and methods of resuspension measurement were proposed since 1990; e additional information is available on resuspension factors determined under indoor conditions e the resuspension factor value of 1x'10-6 m-1 is the most frequently suggested and appears to represent some average of the experimental data; e data on probability distribution functions that could be used to reflect uncertainty and variability in resuspension factors is very limited; however, it may be possible to derive the distribution for RFo based from experimental data on resuspension; cnd a the range of the resuspension factor values measured under indoor conditions is around four orders of magnitude [ Jones,1964] 3.6.6 Ol
3,6.3 Estimating RFo from Site information For a particular site applying the building occupancy scenario model, an applicant might seek to defend a specific value for RFo based on the physical fea'ures of the site that influence resuspension directly, or based on expectations about, ce nstrictions on, the behavior of occupants which may affect resuspension. In view of the reported decrease in resuspension factor with time discussed above, a constant value of RFo reflecting the initial resuspension factor is assumed to be appropriate for assessing regulatory compliance. It is useful to express the resuspension factor used for the building occupancy dose calculation I as the product of two separate parameters: the resuspension factor for " loose" contamination, and the fraction of the total contaminant that is
- loose." " Loose" contamination refers to contamination that is available for transport via resuspension, and excludes any contamination that adheres to, is absorbed into, or is covered by exposed surfaces. This decomposition allows a more direct use of many reported values of resuspension factor given the underlying experimental conditions, and provides a physically plausible mechanism for linking the values of resuspension factor and secondary incastion rate used in the dose calculation.
Based on the analysis of the literature data, the initial resuspension factor values can differ at least by a few orders of magnitude depending on site specific conditions which depend on the use of the property (i.e. the nature and intensity of mechanical disturbance associated with activities of the critical group), by an order of magnitude depending on radionuclide, and an order of magnitude depen 6ng on modeling approach used. Variations due to differences in radionuclides, topography, type of deposition, particle size, surface chemistry, and the nature of
/7 I the surface are assumed to be uncontrollable by the licensee.
(
\ '"}
Several of the physical factors discussed in Section 3.6.2 influencing resuspension may be plausibly bounded by characteristics of the site, or controlled by the licensee in an effort to support a site-specific value for RFo. Other factors do not appear amenable to characterization or control. Site-to-site variations in these factors create variations among site-specific values of RFo, but would presumably not be controllable by the licensee. Considerations of these factors are presented in the following bulleted paragraphs. e Time since disoosal - Because RFo is constant with time, the potential for resuspension factor to decrease with time is assumed to be conservatively disregarded, as discussed above.
- Tvoe of disturbance - Mechanical disturbance significantly increases the observed resuspension factor. Lower values of RFo may be appropriate if surface contamination is undisturbed by sweeping or walking. In addition, the effective (time averaged) resuspension factor may be reduced if the contaminated area is subject to brief intermittent disturbance rather than continuous disturbance.
e Intensity of disturbance - Large air-flow rates and vigorous mechanical disturbance lead to increased resuspension factors. Demonstration of limits on intensity, or of intermittence of periods of intense disturbance, may affect the value of RFo.
- Nature of surface - Little quantitative information on the effect of this factor was found in the literature. Available information is therefore assumed to be insufficient to support alternative values for RFo based on site-specific information about this factor.
(Jn) 3.6.7
e PE1[gle Size Distributi&D - Particle size is generally regarded as influencing resuspension factor. Charactenzing or controlling particle size does not appear to be a practical option for applicants. e Tyoe of deoosition orocen - Reported resuspension factor values are higher for loose, easily removable contamination than for contamination that is bound to, or absorbed into, the surface. Applicants are assumed to have removed most loose contamination prior to de-commissioning. This assumption can be reflected in the occupancy scenario calculations in two ways: measured resuspension values for loose contamination may be excluded in defining RFo, or RFo may be initially defined for loose contamination, and the applicant may later reduce this value based on the fraction of loose contamination at their site. Excluding measurements on loose contamination in defining RFo assumes that a/Iloose contamination has been removed, and that no mechanism will loosen contamination during the occupancy period. This assumption does not appear to be justifiable in all cases. The second approach, which decomposes the resuspension factor used in the occupancy scenario model into a resuspension factor for loose contamination, and a fraction of contamination that is loose (~ .:. available for resuspe3sion) allows uncertainty in the fra: tion c' toose conth,nination to be explicitly addressed. This approach also provides a convenient mechanism for connecting the values for resuspension factor and secondary ingestion rate by using a common value for the fraction of loose contamination. The parameter RFo is therefore assumed to describe loose (resuspendable) contamination, and the applicant is assumed to be able to reduce this value by demonstrating an upper limit, over the occupancy period, on the fraction of loose contamination.
- Chemical orocerties of the contaminant - The potential effect of chemical properties on resuspension factor is estimated to be one order of magnitude or less. Although a source of site-to-site variability in RFo, applicants are not assumed to base site specific values for RFo on chemical property arguments due to the relatively small size of this effect, and the potential difficulty in collecting supporting data, e Toooaraohv - No quantitative information on the effect of this factor was found in the literature. Available information is therefore assumed to be insufficient to support alternative values for RFo based on site-specific topographic information.
Of the factors influencing resuspension discussed above, site-specific values for RFo might be supported by information about the nature and intensity of disturbances likely to occur during the occupancy period. The effect of the remaining factors on resuspension is either relatively small (an order of magnitude or less), or is insufficiently defined in the literature for the applicant to defensibly derive a site-specific value of RFo from information about these factors. Variations in these factors from site to site introduce variations in RFo which are not expected be controllable by the applicant by restricting the use of the property. In addition to the nature and intensity of disturbance, the fraction of loose contamination will also control resuspension, and may be estimated from site data. As discussed above, any site-specific estimate for this fraction is assumed to be used to scale RFo, while RFo is assumed to describe resuspension of loose contamination. 3.6.8 O
Variations in the site-specific values for RFo were estimated using published experimental data > Q that were measured under a variety of activities and conditions. The procedure is summarized (] below, followed by a description of the application and results. (1) Reported values for resuspension factor were categorized according to similanty in the descriptions of the experimental conditions regarding the nature and intensity of disturbance. As discussed above, variations in resuspension factor due to variations in mechanO:..I disturbance may be plausibly controlled by the behavior of the critical group. (2) For each category, a range of acute resuspension factors was defined based on the reponed values in each category. Within each category, variations in reported values are assumed to reflect variations due to factors other than the nature and intensity of surface disturbance, such as surface chemistry, surface topography, and particle size distribution. Variability in these factors among sites will also produce variability in site-specific values for RFo, however the e~ *ts of these factors on resuspension would not depend on the activities of occupai Instead, such variations are modeled as random variations among sites, independent of the use of the property. (3) For each category, a range of chronic resuspension factors was defined using the range of reported resuspension factor values for that category. In general, the reported values for resuspension factor correspond to activities that would be performed at intervals in an occupational setting, and performed only for a limited neriod of time. RFo represents a chronic (year-long) effective value, and should therefore reflect the mixture and duration of activities performed by members of the critical group during a typical year. The range of p chronic resuspension factor values is based on the observed range in reported resuspension factor values in consideration of uncertainties in time allocation estimates , y x and in the estimated range of acute resuspension factor values. (4) For a range of possible property uses, the occupation of the entical group at these ! properties was associated with one of the categories defined by the nature and intensity of disturbance in Step (1). This assignment reflects the most severe occupational conditions l to which a member of the critical group is expected to be exposed at such properties. Due
- to the limited number of measurements, only two categories were used to describe the l potential occupational environments for members of the critical group. These categories l are distinguished by the presence or absence of high air-flow rates.
(5) A distribution describing the variability of RFo over sites was constructed based on: the estimated fraction of sites whose critical group is associated with each surface disturbance category defined in Step (4); and the distribution of chronic resuspension factor values associated with each category, defined in Step (3). In estimating the fraction of sites in each category, both the current use of the property, and the potential conversion of the property to other uses were considered. Grouping of Reported Resuspension Factors based on Experimental Conditions Table 3.6.2 summarizes the resuspension factors reported for experimental studies for various conditions [ Jones,1964; Glauberman,1964; Mitchell,1964; and Fish,1964). Brunskill [1964] studied resuspension from contaminated clothing in the high air-flow conditions typical of a change room, in the occupancy scenario, contamination is assumed to occur on building (V') 3.6.9
surfaces. Resuspension from clothing was not assumed to be representative of resuspension from these surfaces: values reported by Brunskill were therefore not considered in defining a distribution for RFo. The experiments by Jones [1964] provide resuspension factors for a range of activities that are common in occupational settings. The measured resuspension factors reported by Jones d [1964) are for four levels of activities using PuO2 contaminated particles (0.4 - 60 m diameter) and particulate air samplers positioned at 14-175 cm above the surface. Glauberman [1964) provides resuspension factors for a range of air-flow rates and mechanical disturbances that may occur in occupational settings. The values for this study reported in Table 3.6.2 show the relatively narrow =nge of resuspension factors observeo for four experimental conditions. Glauberraan measured occupational exposure to airborne particulates in a operating facility by meast.ing the concentrations of particles in air (high efficiency particulate sampler) and particles on surfaces (smear sampling), and reporting the ratio as a resuspension factor. Airbome particle contaminants in this experiment may have originated from sources other than surfam (e.g., processing equipment, etc), which would tend to increase estimateu resuspension factor values. The .,.arted values from Glauber. '.an [1964) were retained for comparison in defining the distribution for RFo, but are judged to be highly uncertain and to overestimate the resuspension factor associated with the conditions described. Table 3.6.2, Resuspension Facters Measured Under Various Conditions Experimental Condition RFo (m ') Reported by Jones [1964) Normal room ventilation 3.3x10' Walking (14 steps / min) 9.1 x10
- Walking (36 steps! min) 6.9x105 Walking (100 steps / min) with wind stress 1.5x104 (hair dryer directed toward floor)
Reported by Glauberman [1964] Undisturbed 1.5x10 5 to 3.6x10" Fans on 3.4x10'5 to 1.6x10-Vibration (dolly) 1.2x10d to 1.9x10" Fans + vibration 1.2x10d to 1.5x10 2 Reported by Mitchell [1964) Vigorous sweeping by 2 workmen 1.02x10'2 to 4.2x10 2 Reported by Fish [1964) Vigorous work activity, including sweeping 1.9x10" Vigorous walking 3.9x10-5 Light work activity 9.4x10 4 Rapid air circulation 7.1 x10" Mitchell [1964) measured resuspension factors during vigorous mechanical disturbance of contamination on a wood floor. The experimental conditions were contrived to deliberately 3.6.10
suspend loose contamination in order to produce measurable values of resuspension factor.- These conditions are not considered to be representative of conditions that would occur in an
-\ ~ _ occupational setting. The reported values were therefore not included in defining a distribution for RFo. Fish (1964) provides resuspension factors for a range of vigorous mechanical-disturbances of contamination on a tile floor, and for high air-flow rates. The values in Table 1
3.6.2 for this study are reported for four types of disturbance.
'. in order to separate the effects of occupation related factors from uncontrollable factors on
- resuspension, the resuspension factor values reported in Table 3.6.2 were grouped according ,
i to the nature and extent of surface disturbance. The presence or absence of high air flow rates t was first used to define two groups. For measurements made in the absence of high air flow,
- i. the descriptions of mechanical disturbance of the surface were used to classify each reported value in to one of two categories based on the presence or absence of mechanical disturbance.-
For high air-flow conditions, too few values are available to support a distinction based on l~ mechanical disturbance. Table 3.6.3 shows the values assigned to each of the three resulting i categories. As discussed above, values reported by Glauberman [1964] are assumed to overestimate resuspension by at least m order of magnitude. . Trends among categories in Table 3.6.3 are generally consistent with expectations about resuspension: values tend to increase when mechanical disturbance or high air flow rates are present. Within each category, however, the range of reported values is generally large. This range is assumed to reflect variability in factors other than the nature and intensity of disturbance, such as surface chemistry and topography, particle size.
- Table 3.6.3. Reported Resuspension Factor Values Grouped by Experimental Conditions -
l Mechanical . ,
~ Air Flow Stress Reference Rfo (m 1)
! LowlNone Absent - Jones [1964] : Normal room ventilation 3.3x10'8 ! Glauberman [1964] : Undisturbed 1.5x10'8 to 3.6x104 i~ ! ~ LowlNone Present Jones [1964] : Walking (14 ctops' min) ' 9.1x10 e {; Fish [1964] : Light work activity 9.4x10 Jones [1964] : Walking (36 steps / min) 6.9x10-5 Glauberman [1964] : Vibration (dolly) 1.2x104 to 1.9x10d Fish [1964] : Vigorous work activity, 1.9x10d j including sweeping Fish [1954] : Vigorous walking 3.9x10-5 i High ' Glauberman [1964] : Fans on 3.4x10 5. to 1.6x10 8 i Fish [1964] : Rapid air circulation 7.1x10d i Jones [1964] : Walking (100 steps / min) 1.5x10d 1 with wind stress (hair dryer directed toward 2 floor) Glauberman [1964] : Fans + vibration 1.2x104 to 1.5x10-2 3.6.11 i 4 4 a , , , - , 7 , . , . - - ,
Ranges of Resuspension Factors for Various Stress Conditions Ranges of resuspension factor values were defined using the infornietion in Table 3.6.3. Table 3.6.4 defines the ranges of resuspension factor values corresponding to each category. Estimates of the upper and lower limits, along with the source of these estimates, are provided for each category. As discussed above, values reported by Glauberman (1964) are subject to special uncertainty, and are assumed to overestimate resuspension. Table 3.6.4, Ranges of Resuspension Factor Values for Categories of Surface Stress Conditions Mechanical Value Category Air Flow Stress Limit (m 1) Source A Lowl Absent Lower 3.3x10' Jones [1964) : Normal room None ventilation Upper <3.6x10-* Glauberman'[1964) : UMisturbed user limit B Lowl Present Lower 9.1 x10' Jones [1964) : Walking (14 None steps / min) Upper 1.9x10
- Fish [1964) : Vigorous work activity, including swesping C High Lower 1.5x10
- Jones [1964) : Walking with wind stress Upper 7.1 x10
- Fish [1964) : Rapid air circulation Ranges of Chronic Resuspension Factors for Various Stress Conditions As discussed above, the values in Table 3.6.4 are based on the range of reported acute resuspension factor values for distinct conditions of surface disturbance. The particular activities of occupants at a given site will entail characteristic disturbance conditions, and therefore control the effective resuspension factor values appropriate for those occupants.
Within each category, the range of reported values is assumed to reflect the effects of factors specific to the site but unrelated to occupation, such as surface topography and chemistry, and particle size. The value for RFo used in the dose calculation should reflect the time-averaged value of condition-specific resuspension factors over the one year duration of the occupancy scenario. This time average (chronic) value will generally differ from the acute values in Table 3.6.4 due to variations in the occupant's behavior over time. In addition, the larger resuspension factor values given in Table 3.6.4 for high air-flow conditions imply significant depletion of the source over the one year period of the scenario. The effects of these factors on the proposed distribution for RFo are described in the following sections. 3.6.12 O
A) Acute vs. Chronic Resusfension Factor Values i V For a given indivV tal, the resuspension factor will vary win timo because their activities vary with time. (deally, an estimate of the chronic (time averaged) resuspension factor value therefore requires an estimate of the time spent in activities corresponding to each category. The chronic resuspension factors would then be calculated as the sum of the acute resuspension factors for each category, weighied by the amount of time spent in each category. As a result of this averaging process, the range of chronic values for occupants who tend to spend their time in activities in a given category will be narrower than the range of acute values expecienced by the occupant over time. A formal estimate of chronic resuspension factor values would require estimates of the time spent on each category, and of the acute resuspension factor for each category. For this analysis, the results of such a process would be subject to two important and counteracting sources of uncertainty, First, estimates of time aSocation for particular occupations that might occur at licenser' propertiea would be highly uncertain. Although, as discussed above, the range of chronic values for occupants would be narrower than the range of acute values due to the effect of time averaging, the location of the range of chronic values within the larger range of acute values would be subject to considerable uncertainty due to uncertainty in the estimated time allocation. Second, the ranges of possible acute resuspension factor values corresponding to distinct stress conditions is uncertain. Although ranges for the categories defined in Table 3.6.4 were /] V defined by the limits of reported values, very few observations are available for each category. As a result, the potential range of acute resuspension factor values corresponding to distinct stress conditions is expected to be wider than the range in reported values due to limited sampling of these conditions by published experimental results. In arder to formally calculate chronic resuspension factor values, estimates of the true range of acute resuspension factors, developed in consideration of the limited number of samples available in each category, would be required. These estimates would also be subject to considerable uncertainty, Rather than attempting a formal calculation of chronic resuspension factor values, the ranges of values in Table 3.6.4 were directly adopted as estimates of the ranges of potential chronic values for occupants typically exposed to conditions defined by each category. This approach does not require assumptions regarding time allocation for various occupations nor assumptions about the actual range of potential acute values given the range in reported measurements. As discussed above, these assumptions would introduce considerable uncertainty in the calculated chronic values. The ranges in Table 3.6.4 are also uncertain as estimates of chronic resuspension factors, however the two primary sources of uncertainty discussed above tend to have counteracting effects: time averaging of acute values would result in chronic values that are narrower than the range of acute values, however the actual range of acute values is wider than the range of observed values due to the limited number of samples. B) Source Mass Conservation p g j Based on the above considerations, the ranges of reported acute resuspension factor values in Table 3.6.4 were assumed to define the ranges in potential annual average resuspension factor v 3.6.13
values. For h;gh air floa cond.tions, however, the annual average resuspension factor value may also be limited by the total source mass. Because the occupancy scenario model does not include source mass foss via resuspension, resuspension factor values which imply subt tantial l depletion of source contaminants willlead to overestimates of dose The effect of source oepletion by resuspension in the presence of high air flow can be included in one of two ways: the occupancy scenario model can be revised to include source mass conservation, or an effective resuspension factor can be derived which includes the effect of source mass loss during the one-year scenario period. The latter approach was adopted, as described below, to calculate an effective chronic resuspension factor value from the potential chronic values in Table 3.6.4. This effective value incorporates the influence of source depletion, which is not modeled in the defau't occupancy scenario model as defined in NUREGICR-5512 Volume 1. The resulting resuspension factor values are not appropriate for models which explicitly include source mass loss via resuspension. Under conditions of high air-flow, any resuspended materialis assumed to be removed as a potential source. Under this Petumption, the rate of source depletion is equal to the resuspension rate. dC,(t)
= - Ar ., C,(t) dt (2) where C,(t) is the source concentration, and A,,, is the resuspension rate, in equation (2),
all source mass loss is assumed to occur through resuspension, and mass loss due to other process is assumed to be negligible over the one-year performance period. During this period, the amount of resuspended material is calculated from a constant specified source C,(0) and the specified resuspension factor value RF,, . Mass depletion implied by equation (2) may ba approximatel" included via RF,, by requiring that the resuspended mass, calculated using RF,, and C,(0) ,is equal to the average resuspended mass calculated using the potential chronic resuspension factor RF, and the depleting source C,(t) : RF,,C,(0) = A ' RF,C,(t)dt = Af ' RF,C,(0)e *"dt Tfo ' TJo
$ _, 4..r ) (3)
= RF, C,(0)
\
hres T , so that the effective and potential resuspension factors are related by: 3.6.14 O
l l 0 pp'" ,pp* 1 -e *"* (4) q A,,,T , in Equation (4), T is the length of time during which source mass loss occurs, which was assumed to correspond to a standard working year of 250 eight hour days. The resuspension rate can be estimated from the room geometry, the ventilation rate, and the resuspension factor: f i A,,, RF,A y (5) rA,, , where V is the room volume, A is the room area, and A y is the ventilation rate. V The ratio - typically ranges from approximately 0.5 m for small rooms to approximately 1 m A for large rooms. Ventilation rates corresponding to "high" air flow rates were estimated using _ the Versar [1990] Database of PFT Ventilation Measurements, as summarized in the EPA Exposure Factors Handbook. The database compiles results from a number of separate
; studies, each study reporting a number of measurements taken at different residences or
\ during different seasons. These measurements were made in residential rather than occupational settings, and cannot be used directly to estimate ventilation rates for high air-flow conditions. Across the summarized studies, the 90'th percentile ventilation rates range from 0.38 to 5.89 hd. A ventilation rate of 5 h4 was therefore chosen to represent high air-flow-conditions in an occupational setting.
3.6.4 Proposed Distribution for RFo For a given site, a site-specific value for RFo should reflect conditions experienced by the average member of the critical group. The critical group at a given site is in turn assumed to be defined by the occupation associated with the largest effective resuspension factor value. The resuspension factor value used to calculate dose should also consider alternative uses of the property potentially creating more severe surface stress conditions. A distribution function describing the variability of site-specific values for RFo was calculated as the weighted sum of the distributions for the surface stress categories defined in Table 3.6.4. Weights for each category represent the fraction of sites having critical groups which are chronically exposed to the type of surface disturbance characterizing each category. Within each category, site-to-site variability in topography, chemistry, particle size, and other factors unrelated to occupation were assumed to produce a log-uniform distribution of values between the lower and upper limits for that category. For resuspension in the presence of high
, air-flow, effective resuspension factor values were calculated from the potential resuspension
( factor values (Category C of Table 3.6.4) using Equations 4 and 5. 3.6.15 4
r Assuming that any property might be devoted, at some future time, to light industry, ro applicant would be able to exclude the possibility of mechanical disturbance chronically occurring at their site. The fraction of sites having entical groups exposed to Category A is therefore assumed to be 0. Many applicants would, however, be able to exclude the possibility of high air flow rates, as such rates seem likely to be associated with customized ventilation systems, large openings such as bay doors, or other structural features which the building may lack. The fraction of sites containing such features was estimated as the fraction of non-service enterprises devoted to manufacturing in 1993 (approximately 9.8%), as reported by the U. S. Census Bureau. The remaining sites (90.2%) are assumed to have resuspension f actor values from Category B. In summary,90.2% of sites are assumed to have resuspension factor values between 9.1x10.e and 1.9x104 md, with a log-uniform distribution assumed between these limits. The remaining 9.8% of sites are assumed to have structural features that might create high air flow conditions, d d and therefore have potential resuspension factor values ranging from 1.5x10 to 7.1x104 m, with a log-uniform didribution assumed between thete limla. High resuspension factor values, however, in conjunction with high air-flow conditions, imply substantial depletion of source mass t . ring the one-year perfctmance period. This depletion is include :in the resuspension factor value by calculating an effective annual average value from the potential resuspension factor 4 value using Equations 4 and 5. In this calculation, the ventilation rate is assumed to be 5 hd, while the volume / area ratio was assumed to be uniformly distributed between 0.5 m and 1.0 m. Figure 3.6.2 shows the resulting cumulative distribution functions for RFo, while figure 3.6.3 shows the corresponding probability density function. The proposed distribution for RFo ranges from 9.1x10* m4 to 1.9x108 m*, with a median value of 5.0x10'. Although the resuspension factors for various experimental conditions ranges over several orders of magnitude, values of resuspension factor for the critical group are biased towards the upper end of reported values based on the range of surface stress conditions assumed for the occupants. The critical grour for a given site is assumed to consist of those workers experiencing the largest chronic resuspension factors. 3.6.16 O 1
_ . . _ _ _ _ _ _ _ . . . _ . . _ . . . _ _ _ . . _ . _ _ . _ . _ _ . . _ . _ _ _ _ _ _ . . . . _ . . . _ . _ . _ . . . _ . . _ . . . _ ~ . _ _ _ __ 4 s l 1 , ; TTTT i ;
, ; 'Tn : !
l 0e . _ J . L I_ _! [. !._ .J_,.., ,n.
' ,I ;i .:;J p! '
l Oe l' f l -l ll ! ii l f..f ! Or
......-.- ( -._' ! l _,! !#_! -_- .
.,__.l l
.p .. _ _ . _ . . _ . . .
..LJ J .._q l, !
- Oe ..- . _ ._._,_ ,J . . _ _ _ _.. .
06 ...-. j O4 ... . . ._ , ..- U 0.3 -_.-, .; O2 -.- -. ..--. . o$ . _.] 0 i t l-l I i-100E 06 1.00E 05 100E 44 1.00E 03 RF o (m *.1) Figure 3.6.2. Proposed Cumulative Probability Function for RFo 3.50E+04 .. . 3 00E+04 - __ - -
- p. !
l l ! 2.50E+04 -- - .._. .. - _i 2.00E+04 - 5 1.50E+04 A 1.00E+04 -~ t I 5.00E+03 d_ -d I 0.00E+00 !lll 1 1.00E-06 1.00E-05 1.00E-04 1.00E-03 RFo (ma 1) g Figure 3.6.3. Proposed Probability Density Function for RFo 3.6.17
- PARAMETER UNCERTAINTY
- The proposed distribution describing thc, variability in the l resuspension factor is based on several assumptions, leading to uncertainty in this distnbution as an estimate of the potential variability of RFo over sites:
l (1) Resuspension of loose particles in a building occurs by a combination of wind stress from normal building ventilation and mechanical disturbances from walking and l vehicular traffic. Other than in manufacturing establishments, persistent high air-flow
- conditions are assumed to be unlikely.
(2) Resuspension factor values are reported to depend to some extent on a number of other factors, including surface texture and topography, particle size distribution, type of deposition, and chemical properties of the contaminant and surface. These factors are assumed to produce site-to-site variations in recuspension factor values which are unrelated to the occupation of the critical group. i (3) The reported ranges of resuspension factors, measured under experimental conditions corresponding to ac. % occupational activitics, were adopted as the range of possibb chronic resuspension factor values for occusnts typically engaged in tL:se activities. This assumption was made in consideration of the uncertainties associated with estimated time allocations for occupations, the tendency for time averaged values to have lower variability than the true acute values, and the expectation that the limited number of measurements of resuspension underestimate the true v&riability in acute resuspension factor values. (4) The combination of high air-flow rates with high air flow conditions implies substantial depletion of source mass during the one year performance period. Because the default occupancy scenario model does not include source mass loss through resuspension, effective (annual average) resuspension factor values, which approximate the effect of source depletion, were developed for these conditions. This approximation assumes that resuspension is the primary mechanism of source depletion, and that mass loss through other processes (such as radioactive decay) are comparatively small during the performance period. These effective resuspension factor v910es are not suitable for use in models that explicitly include source mass loss by resuspension. (5) The data on resuspension of reported by Glauberman [1964) are regarded as uncertain over-estimates. (6) U.S. Census data on the numbers and types of industrial divisions in the United States reflect the variability in property uses over the licensed sites. (7) Alternative future property uses would be considered in establishing a site-specific resuspension factor value. Such uses might increase mechanical surface disturbance. Increases in air flow are as'" . 'd to require extensive modifications to existing , structures, i VARIABILITY ACROSS SITES: The resuspension factor will vary across sites due to ! differences in the use of the properties, and due to factors unrelated to the use of the property l such as surface chemistry and topography. l 3.6.18
SITE DATA COLLECTION: Applicants are not expected to col lect resuspension data. However, the applicant may attempt to support limits on RFo based on the intended use of the ('] (j property or provide site-specific data regarding fixed vs. removable contamination. DEFINITION OF SITE DATA SOURCE: Not applicable. NRC INTERPRETATION OF SITE-SPECIFIC VALUE: Applicants may attempt to support alternative values for RFo based on the intended use of the property. NRC may require legal assurances from the applicant. REFERENCES Anspaugh, L. R., J. H. Shinn, P. L. Phelps, N. C. Kennedy "Resuspension and Redistribution of Plutonium in Soils", Health Physics,29(4), pp. 57182, October 1975. Brodsky, A. "Resuspension Factors and Probabilities of Intake of Material in Process (or AIS 10-6 a Magic Number in Health Physics?* Health Physics,39(4), pp. 992-1000, December 1980. Brunskill, R. T., The Relationship Between Surface and Airborne Contamination, in C'. R. Fish ed., Surface Contamination Symposium Proceedings, pp. 93-105, June 1964, Gatlinburg, Tennessee, Pergamon Press, New York. Fish, B. R., Walker, R. L., Royster, G. W., and Thompson, J. L. "Redispersion of Settled fQ Particles", in B. R. Fish ed., Surface Contamination Symposium Proceedings, pp.75-81, i w) June 1964, Gatlinburg, Tennessee, Pergamon Press, New York. Garland , J. A. Resuspension of Particulate Material From Grass. Experimental Programme 1979-1980. London: HSMO;AERE-R10106; 1982. Glauberman, H., Bootmann, W. R., and Breslin, A. J., " Studies of the L;, .ificance of Surface contamination", in B. R. Fish ed., Surface Contamination Symposium Proceedings, pp. 169-178, June 1964, Gatlinburg, Tennessee, Pergamon Press, New York. Hartmann, G., C. Thom, and K. Bachmann " Sources for Pu in Near Surface Air", Health Physics, 57(1), pp. 55-69, January 1989. Hinton, T. G., P. Kopp, S. Ibrahim, I. Bubryak, A. Syomov, L. Tobler, and C, Bell " Comparison of Technique used to Estimate the Amount of Resuspended Soil on Plant Surfaces," Health Physics, 68(d), pp. 523-531, April 1995. Hollander, W., "Resuspension Factors of 137Cs in Hannover After the Chernobyl Accident," Aerosol Science, 25(5), pp. 789-79^.,1994. Homa, S. G. "The hotspot health physics codes," Health Physics,68(6 Supp):S59, June 1995. IAEA, " General Models and Parameters for Assessing the Environmental Transfer of t n i Radionuclides from Routine Releases," Vienna: lAEA: Safety Series No. 57; 1982. _.) i 3.6.19 1 ___ o
IAEA, " Derived intervention Levels for Application in Controlling Radiation Doses to the Public in the Event of a Nuclear Accident cr Radiological Emergency," Vienna: lAEA: Safety Series No. 81; 1986. lAEA, *Modeling of Resuspension, Seasonality and Losses During Food Processing," First Report of the VAMP Terrestrial Working Group, IA: A-TECDOC-6471,1992. IAEA, " Validation of Models Using Onernobyl Fallout Data from Southern Finland. Scenario S " Second Repor1 of the VAMP Multiple Pathways Assessment Working Group, IAEA-TECDOC-904, September 1996. Jonea, I. S., and Pond, S. F., "Some Experiments to Determine the Resuspension Factor of s Plutonium from Various Surfaces," in B. R. Fish ed., Surface Contamination Symposium Proceedings, pp. 83 92, June 1964, Gatlinburg, Tenn ssee, Pergamon Press, New York. Kathren, R. L., in Proc. Symp. On Radiological Protectu of the Public in a Nuclear Mass Disaster, Interlaken, Swit.,26 May - 1 June,1968 (Bern: EDMZ). L. .gham, W. H., USAEC Rept. USRL-50639 (Livermore: Lawrt ice Livermore Laboratory). 1969. Langham, W. H., in: Proc. Environmental Plutonium Symp., Los Alamos,4 5 August 1971, p. 3 (Los Alamos: Los Alamos Scientific Laboratory),19721. Leonard B. E. ara-222 Progeny Surface Deposition and Resuspension Residential Materials.@, Health Physics,69(1), pp. 75-92, July 1995. Mitchell, R. N., and Eutsler, B. C., "A Study of Beryllium Surface Contamination and Resuspension," in B. R. Fish ed., Scrface Contamination Symposium Proceedings, pp. 349-352, June 1964, Gatlinburg, Tennessee, Pergamon Press, New York.
- Nair, S. K., C. W. Miller, K. M. Thiessen, and E. K. Garger "Modeling the resuspension of radionuclidE 5 in Ukrainian regions impacted by Chernobyl fallout," Health Physics, 68(6 Supp).S46, June 1995.
Nair, S.K., Miller C. W., Thiessenn K. M., Garger E. K., Hoffman F. O. "Modeling the Resuspension of Radionuclides in Ukrainian Regions impacted by Chernobyl Fallout," Health Physics,72(1), pp. 77-85, January 1997. NRC, " Reactor Safely Study: An Assessment of Accident Risk in U.S. Commercial Nuclear Plants, Appendix VI. Calculation of Rer: tor Accident Consequences," Rep. WASH-1400, 1975. Thatcher, T L., Layton, D. W. " Deposition, Resuspension, and Penetration of Particles within a Residence," Atmospheric Environment,29(13): 1487-1497. Versar " Database of PFT Ventilation Measurements: Description and User's Manual" USEPA Contract No 68-02-4254, Task No. 39, Washington, D.C. U.S. Environmental Protsetion Agency, Office of Toxic Substances. 3.6.20 O 1
3.7 Volumetric breathing rate, V (m 8/h)
/ -Q The breathing rate parameter (Vo), in conjunction with the resuspension fac:or and isotope-specific inhalation CEDE factors, is used in calculating '.he average annual dose due to inhalation. The default value for this parameter defineJ in NUREG/CR 5512, Volume 1, is 1.2 m8/h. This value corresponds to an average for the 8 nour work day assuming light activity for a person as defined in ICRP Publication 23 [1975).
j The RESRAD value for the same parameter is 0.96 m3 /h. Foi the first iteration of the parameter analysis, this parameter was assigned a uniform distribution with lower and upper limits of 0.39 knd 1.5 m3/h, respectively, for lack of better information. The limits were taken from the range in ICRP Publication 66 provided for workers and adult general population at various activity levels: rest (0.39-0.54), light exercise (1.25-1.5), heavy exercise (2.7 3.0). The EPA anthropometric data report [ EPA,1985) contains different ranges for rest (0.3-0.7), light activity (0.5-0.8), and heavy activity (2.9-4.8). A plot of the PDF
!s presented in Figure 3.7.1.
IMPORTANCE TO DOSE: The higher the value for Vo, the higher the total annual dose during the first year of the building occupancy scenario. Inhalation dose is linearly proportional to Vo. 4 O 1 0.9 08 os j o. 0.6 04 3 0.3 02 0* l 0 0 02 04 08 08 1 1.2 14 1.8 18 2 Vo(m*3mr) Figure 3.7.1 PDF for breathing rate for first iteration of parameter analysis. O h 3.7.1
USE OF PARAMETER IN MODELING: Vo is used to calculate CEDE for DHO, resulting from inhalation of resuspended surface contsmination. The relationship between Vo and internal dose due to inhalation is desenbed by the following formula: DHO,=45.05'24*t/ RF/V/ L,m DFHlC,,, (3.7.1) where J,is the number of radionuclides in chain 1, to is the time that exposure occurs dunng the building occupancy period, C.,,is the average annual activity of the radionuclide j during first year of the building occupancy scenario, DFH,is the inhalation CEDE factor, and RFo is the suspension factor. The resulting internal inhalation dose is directly proportional to the volumetric breathing rate. 3.7.1 Additionallnformation Reviewed to Define Revised PDF for Vo For the second iteration of the parameter analysis, additionalinformation was reviewed to determine if other data or approaches, preferably more recent than those cited in NUREG/CR-5512, Volume 1 [ Kennedy and Strenge,1992), were available to provide a defensible basis for constructing a PDC for Vo for use in the analysis. In t" Sontext of the dose exposre model, Vo represents the breathing rate, averaged over the period of occupancy, of the average member of the critical group. A distribution describing the variability of site specific values for this parameter, among sites employing the NUREG/CR 5512 dose assessment,is required of use in the secono iteration of parameter analysis. To define this distribution, a licensee is ideally assumed to have detailed information on the breathing rate of all occupants during the first year of occupancy. This information would be used to select a subset of occupants as the site-specific critical group for the property. Vo would then be determined as the average, over this subset, of time averagsd bresthing rate during the period of occupancy. The literatura citee in the Draft EPA Exposure Factors Handbook [ EPA,1996) was adopted as the most curient compilation of relevant literature. Of the eleven studies reviewed in the Handbook, five ([Layton,1993), [Linn et al.,1992), [Linn et al.,1993), [ Spier et al.,1992), and [CARB,1993)) are identified as
- key studies,* and form the basis for recommended inhalation values. The remaining studies are considered
- relevant," and contain supporting information relating to inhalation rate. Summaries of the eight studies most relevant for the purpose of derining a distribwion for Vo, are d'scussed in the following paragraphs.
Layton [1993) presents a muthod for estimating breathing rate based on metabolic information: Vt= E x H x VQ (3.7.2) where: Vg is the ventilation rate E is the energy expenditure rate H is the volume of oxygen consumed in the production of 1 KJ of energy, and VQ is the ratio of intake volume to oxygen uptake 3.7.2 l l
Three approaches are used to estimate the energy expenditure rate: annual caloric intake ( ) (corrected for reporting bias), elevation above basal metabolic rate (BMR) with BMR values C/ estimated from body weight using a fitted regression model, and elevations above BMR using activity specific elevation factors and time allocation data. These methods are used to estimate average inhalation rates over various population subsets defined by age and gender. This study draws from comparatively large data sets, and provides information on the relative contributions of the diverbe factors influencing inhalation rate, including general health, body weight, diet, activity level, age, and gender, For these reasons, it was used as the primary basis for defining the proposed breathing rate distribution for the second iteration of the parameter analysis, as detailed below in Section 3.7.3. Linn et al. [1992) estimates inhalation rates for *high risk
- subpopulations, including outdoor workers, elementary school students, high school students, asthmatic adults, young asthmatics, and construction workers. Because of its focus on selected sub populations, this study is not sufficiently comprehensive to define breathing rate distributions for this analysis. The study does provide information on the average breathing rate and breathing rate variability among individuals. The average breathing rak for healthy adults is reported as 0.78 m'/hr with a 99th percentile value of 2.46 m8 /hr; while construction workers are t,aracterized by an average , ate of 1.50 m8/hr with a 99th percentile value of 4.26 m8 /hr. Mean and 99th percentile values for asthmatic adults are given as 1.20 and 2.40 m8/hr.
Linn et al. (1993) reports breathing rates for construction workers both before and during a typical work shift. The restricted focus of this study again limits is usefulness for defining a distribution of Vo over till sites, however the distributionalinformation for the subject population O is a valuable basis for comparison. Over the subjects 'he average breathing rate is given as
!d 1.68 m8 /hr, with a 99th percentile value of 3.90 m8/hr.
Spier et al. (1992) reports breathing rates for elementary and high-school students. This sub-population is not contemplated in dose assessments for the occupancy scenario. The California Air Resources Board (CARB) (1993) reports breathing rates in routine daily activities for children and adults at various activity level classifications. Breathing rates are Msed on short term data and may not represent long term averages required to define Vo. Breathing rates for adults are reported for five activity level classifications from
- Resting
- to
' Heavy'. These levels are defined in terms of routine domestic and recreational activities, such as housework and jogging. Reported breathing rates for adult females range from 0.43 to 2.96 m'/hr, and from 0.54 to 3.63 m 8/hr for adult males.
The studies classified as ' Relevant
- provide supporting information, such as assessments of the quality of individual's subjective judgments of their breathing rate and activity level. These studies were not judged to previde informaon directly related to estimating distributions of Vo.
Three of these studies are literature surveys. The U.S. EPA [1985) provides a summary of inhalation rates by age, gender, and activity level. This study compiles results of earlier investigations, and does not present information on the accuracy and methods used in these investigations. Reported breathing rates range from 0.3 to 2.9 m*/hr for adult females, and from 0.7 to 4.8 m8/hr for adult males depending on activity level. The Intemational Commission on Radiological Protection (ICRP) [1981) presents ventilation estimates for reference adult O 3.7.3
males and females at two activity levels (* Resting" and
- Light Activity') as well as dailj inhalation rates based on an assumed activity pattern dunng the day. For adult females, the respective rates are given as 0.36 m8 /hr,1.14 m 8/hr, and 21.1 m /3 day, while the corresponding rates for adult males are 0.45 m8 /hr,1.2 m'/hr, and 22.8 m /3 day. The default value for Vo defined in Volume 1 of NUREG/CR 5512 was based on the
- Light Activity" breathing rate for males from this study. It was not considered a sufficient basis for defining distributions of Vo because of currency, and the lack of information on individual variability of breathing rate. The American Industrial Health Council (AlHC) [1904) Exposure Factors Sourcebook recommends an average adult inhalation rate of 18 m8/ day based on data presented in other studies. This report draws from information presented elsewhere, and does not consider the recent work of Layton [1993).
3,7.2 Overview of Breathing Rate Estimation Methods from Layton [1993) As discussed in Section 3.7.1, Layton [1993) was considered the most comprehensive basis for defining distributions for Va because of currency, data-set size, focus on long term inhalation rates, population composition, and the explicit connection of breathing rate to underlying F/slological parameters 'brough a metabolic model. Equation (3.7.2) estimates the breathing rate Vg required to sustain energy expenditure rate E bascd on the metabolic f. actors H and VQ. H is the amount of oxygen consumed in the expenditure of 1 kJ of esrgy. Layton provides estimates of H based on oxygen requirements for metabolizing food, and the resu' ting energy released by ingested food. The consequent oxygen requirements per unit energy vary within a narrow range with the type of food ingested. l Citing James et al. [1989) and McLean and Tobin [1987), Layton estimates oxygen l requirements of 0.0476,0.0508, and 0.0529 L O/kJ for carbohydrates, fat, and protein respectively. Based on the dietary information from the U.S. Department of Agriculture (USDA) [1984) and the U.S. Department of Health and Human Services [1983), Layton estimates an effective value of 0.05 L O/kJ. The assumption that food digestion deteimines oxygen requirements is evidently consistent with the defir;ition of Va as a chronic breathing rate. The effectiva value of H is evidently insensitive to dietary variations among individuals, due to the simdarity of values across food categories. Information on individual variations in the amount of oxygen required to metabolize foods of different categories, and on the energy yields of food, are not discussed in Layton, and the cited sources were not reviewed. These variations are assumed to be small, and the value of H derived by Layton is regarded as a constant. The ventilatory equivalent (VQ)is the ratio of ventilation rate to oxygen uptake rate. According to Layton, VQ varies from individual to individual reflecting variations in oxygen uptake I efficiency, lung physiology and metabolic efficiency. Layton reports literature values in the ( range of 25 to 30, and develops a log normal distribution from 159 measurements on 75 adults assembled from the literature. Most subjects were male, and many were athletes. This distribution has a geometric mean of 27, and a geometric standard deviation of 1.18. Layton uses three methods to estimate energy expenditure E, deriving chronic breathing rate estimates for cohorts defined by age and gender. The first method uses average food energy 3.7.4
intakes reported in the 1977-1978 USDA survey, corrected by a factor of 1.2 to adjust fr.r observed bias in self reporting of food intake. The second and third methods estimate energy expenditure via the basal metabolic rate (BMR). The BMR represents a minimum energy expenditure rate in the absence of physical activity. Layton summarizes data presented by Shofield [1985) to estimate BMR as a function of body mass for several age and gender cohorts. Linear regression equations are reported for each cohort, with model regression coefficients *r" ranging from 0.6 to 0.73 over the adult cohorts. The underlying sample populations are also summarized by average and standard deviation for each cohort. The second method for estimating energy expenditure is based on reported ratios of daily energy intake (dcnoted EFD) to BMR. Several studies were reviewed to define average values for this ratio over the age and gender cohorts considered. Some reported EFD values were corrected for bias due to under reporting of food intake. In general, mean values for this ratio over cohorts are reported. The summary of Basiotis et al. [1989) includes the mean and an error range for the 16 female and 13 male subjects: 1.38 (+/ 0.24) and 1.59 (+/ 0.33) respectively. Prentice et al. [1985) used doubly-labeled water to estimate energy expenditure rates for 12 women aged 23 to 40, yielding a EFD/BMR ratio of 1.38(+/ 0.16), while Rimallo [1989) appled the same approach to six males engaged in light agriculturi activities to produce sn estimate of 1.78 (+/ 0.12). In summarizing population variability for these studies, cited error ranges are not explicitly defined. Average EFD/BMR ratios for population cohorts are multiplied by average BMR values for each cohort (estimated from average body mass and the regression model) to produce an average energy expenditure rate. The third methou uses a ' factorial' approach to estimate energy expenditures based on a pattern of daily activity, and relative energy expenditure rates associated with various activities. ( Time activity data from Sallis [1985) were used to characterize the behavior of average individuals in each of several age / gender cohorts. Daily energy exponditure was estimated from the time spent at each activity level and the metabolic equivalent (MET) value associated with each activity level. The MET value represents relative energy expenditure with respect to BMR, and ranged from 1 (* Sleep") to 10 ("Very Hard") for the activity classifications considered. 4 3.7.3- Estimates of Individual Breathing Rates based on Layton [1993) Equation (3.7.2) can be used to estimate chronic breathing rates for individuals given values for E, H and VQ characterizing a single individual. As discussed above, little variability in H (0.0476 to 0.0529 L 0,/kJ for different types of food)is expected across individuals. Variations in individual values of VQ, reflecting individual variations in physiological factors, have been described using a lognormal distribution with a mean value of 27 and a geometric standard deviation of 1.18. As discussed by Layton, the population described by this distribution (men and/or ath!etes) does not represent a random sample from the general population. Individual variations in energy expenditure E can be estimated using a number of methods, including the three methods discussed in Section 3.7.2. Although variations in individual energy intake are not discussed in Layton, this information might be obtained from other references. This approach could provide estimates of average breathing rates for individuals over extended time periods (e.g. days or greater) assuming no change in body weight. This average rate would reflect total energy expenditure, including time spent sleeping, and would not directly O) q 3.,.s
reflect occupational behavior. Separate estimates can be made for inactive (sleep) and active (awake) periods, as was done in Layton, using data or assumptions about time allocated to sleep and assumptions about BMR. For the purpose of the parameter analysis, it is more helpful to explicitly relate breathing rate to particular activities via the BMR, as in the third method discussed above. Doing this, Equation 3.7.2 becomes: Vt= MET x BMR x H x VQ (3.7.3) where MET is the average metabolic equivalent of an individual's activities dunng the period of occupancy. This relationship makes the dependence of breathing rate on occupation explicit, and permits breathing rates to be estimated for groups defined by occupational activities. As discussed above, data on individual variability of BMR is summarized in Layton from Schofield (1985) by gender and age cohorts. These data are presented in Table 3.7.1. Layton, citing Durnin and Passmore [1967] and Saltin and Astrand [1967), provides MET values for various activities. Table 3.7.2 summarizes values for activities which might be considered under the occuparty scenario. No information is pro'/"'-d about the potential individual variability of MET values, in using Equation (3.7.3) to estimate individualinhalation rates, physiological variations among individuals are assumed to be reflected in variations in VQ and BMR, while MET measures the level of effort inherently required by the activity. This assumption is consistent with the definition of MET as an intrinsic characteristic of a particular level of activity. Using Equation (3.7.3) limiting values for an individual's breathing rate based on the parameter information contained in Layton [1993) can be estimated. As discussed above, H appears to be relatively insensitive to dietary composition, and is assumed to be 0.05 L 0, kJ4 for all individuals. The fitted lognormal distribution for VQ has 1 and 99 percentile values of approximately 19 and 40 and a mean of 27. Sample means and standard deviations for BMR are similar for the two younger cohorts, but differ between men an women (Table 3.7.1). Bounding values can be estimated by adopting the 30 to 60 year cohort as a reference, and selecting two standard deviations as the extent of variation among individuals, The low, central, and high values for BMR for females are 4.36,5.62, and 0.88, respectively and for males,5.01, 6.7o, and 8.49, respectively. A nominal MET value of 3, characteristic of moderate activity, is selected with the low and high values ranging from 1.5 to 10. These parameter ranges and the resulting ranges for breathing rate calculated using Equation (3.7.3) are presented in Table 3.7.3. The range of breathing rate calculated for alllow, all central, and all high values in Equation (3.7.3) are reported for females (0.26,0.95,5.73 m'/hr, respectively) and males (0.30, 1.14,7.08 m 3/hr, respectively)in the last row of the table. The relative contributions of each of the three variable factors in Equation (3.7.3) to variations in individual breathing rate can be calculated in turn using 3.7.6
Table 3.7.1. Individual Variability of Basal Metabolic Rate (BMR) by Gender and Age i s 1 Gender / age Basal Metabolic Rate (MJ/d) I (y) Mean Std. Dev. ' Males 18 to <30 0.87 0.843 30 to <60 6.75 0.872 60+ 5.59 0.928 3 Females 4 18 to <30 5.33 0,721 I 30 to <60 5.62 0.630
! 60+ 4.85 0 605 l
l Table 3.7.2 Metabolic Equivalent (MET) Values ; i for Activities in Building occupancy l l Activity MET l Light i Standing 1.5 l
- Floor Sweeping 1.8 Office Work 1.8
. Moderate Carpet Sweeping 2,7 Cooking 2.7 > Dish Washing 2,7
- Light industry 3.1 d
Walking 3.5 Vacuuming 3.8 Heavy Heavy Industrial 8.0 Work 1 -l 2 3.7.7 J
-------,-c-- e-- ,_ 7,- . , , , - - , , . . . , , , - - -
,,.,,,,y --y-, ,- - - -- - - , ,- - .~,,
Table 3.7.3 Estimated Individual Breathing Rates from Equation (3.7.3) using Central and Bounding Parameter Values f3reathing Rate (m8 /hr) Females Males Parameter Low Central High Low Central High Low Central High VQ 19 27 40 0.67 0.95 1.41 0.80 1.14 1.69 BMR (Female) 4.36 5.62 6.88 0.74 0.95 1.16 (Male) 5.01 6.75 8.49 0.85 1.1dt 1.43 MET 1.5 3.0 10 0.47 0.95 3.16 0.57 1.14 3.80 All 0.26 0.95 5.73 , 0.30 1.14 7.08 a bounding value for one parameter along with central values for the remaining variable pmameters. For females, the relative contribution of VQ results in a range of 0 7 to 1.41 m lhr. The relative contributions for each parameter for males and remales are summarized in Table 3.7.3. Comparing the ranges in Table 3.7.3, it is evident that variability in occupational behavior, reflected in variability in MET values, is the largest source of variability in individual breathing rate. The range calculated with the low and high MET values is 0.47 to 3.16 m 8/hr for females and 0.57 to 3.8 m3/hr for males, compared to somewhat narrower ranges for low and high variations of OVQ and BMR. In general, the breathing rate values are consistent with reported values in the literature, as summarized in Section 3.7.2, however the ranges of values in Table 3.7.3 are somewhat wider than the ranges of literature values. This is to be expected as the values in Table 3.7.3 are estimates for individual behavior, while values reported in the literature are typically average values over population cohorts. 3.7.4 Deriving V Based on MET Values for Establishment Classifications Distributions for individual breathing rates can be derived from Equation (3.7.3) using distributions for the parameters VQ, BMR, and MET describing the variability of these parameters over individuals. Layton [1993] provides information which can be used to estimate distributions for VQ and BMR. There is, however, some uncertainty in these distributions as representations of the general population. A distribution of individual activity levels, in terms of MET values, would also be required. The breathing rate parameter Vo used in the occupancy scenario dose model is intended to describe the behavior of the average member of the entical group over the duration of occupancy during the first year of occupancy. The critical group, for a given site, is the group of individuals reasonably expected to receive the greatest exposure to residual radioactivity given the circumstances under which the analysis would be carried out. The distribution of Vo is not identical to the distribution of individual breathing rates, but instead reflects an averaging over some sub-population of individuals. Because of the tendency of activity level to dominate 3.7.8
variability in breathing rate (Table 3.7.3), it is reasonable to suppose that the critical group (N would be defined by the on site occupation associated with the most strenuous activity level. (V ) Because of the tendency of male BMR values to be larger than female BMR values, the critical group would further be restricted to males engaged in the most strenuous occupation. The average breathing rate over the individuals having the occupation with the most strenuous activity level would tend to reflect average physiological properties, so that individual variations in VQ and BMR would tend to be suppressed in the distribution of Ve. The tendency toward average values for these parameters would be determined by the independence of physiological characteristics from occupational activity, and by the number of individuals engaged in the
- critical' occupation. The independence of these factors, and the number of individuals in the critical group, are uncertain.
To estimate the distribution for Vo, the critical group is assumed to be large enough so that the average values of VQ and BMR appropriately describe the average member of the entical group at every site. Variability in Vo across sites is then determined by variability in the MET values for the critical group across sites. Variability in MET values is in tum assumed to be controlled by variability in the types of activities performed by the critical gmup across (. wen sites. The time scale associated with the MET values in Table 3.7.2 is uncertain, and has been inferred from the activity descriptions. Office Work and Light Industry ere assumed to refer to a mix of activities typical of a normal working day. The associated MET values are therefore assumed to be appropriate for estimating chronic exposure over one year. Individuals engaged N in Heavy Industrial Work are unlikely to be continuously engaged in such work during a typical s L} day. A chronic MET value for heavy industriallaborers was estimated as 8.0 assuming that 6 hours per day are devoted to work characterized by an MET value of 10 and, that 2 hours per day are spent in activities characterized by an MET value of 1.6, such as coffee breaks. The breathing rate for the critical group at a given site is assumed to be the rate associated with the most strenuous activity performed at the site. Data from the U. S. Census Bureau [1997) describing the number of business establishmenss in various general classifications was used to estimate the variability over sites of the most strenuous activity that would occur at decommissioned sites. Table 3.7.4 lists the business classification, the total number of establishments in each classification in 1993, and the 2-digit Standard Industrial Classification (SIC) codes associated with each classification. The descriptions of the SIC codes wm then considered, along with the activity descriptions in Table 3.7.2, to estimate a range of ME T values for each establishment classification as described in the following bulleted paragraphs. l (h 3.7.9
Table 3,7,4 Number of Estab!Ishments by Industrial Division for 1993 Number of Standard Industnal Classification Establishments Classificatiot Codes Agricultural Production 100,685 07.08,09 Mining 28,570 10,12,13,14 Construction 600,299 15,16,17 Manufacturing 387,337 20 39 Transportation, Communication, 267,175 41,42,44-49 Public Utilities Wholesale Trade 509,604 50,51 Retail Trade 1,554,437 52 59 Finance, insurance, and Real 609,492 60-65,67 Estate Service 2,294 559 70-89 Unclassified 49,075 99
. Agricultural Production - Conversion to agricultural production appears to be inconsistent with the occupancy scenario. Existing structures are assumed to be unsuitable for I specialized agricultural purposes (e.g. equipment storage, processing, barns). This classification was therefore not considered in estimating activity distributions. . Mining - Use of properties by mining establishments is assumed to exclude mining at the site, and therefore to exclude the use of existing buildings to support on site mining l operations. Instead, properties are assumed to be used for off site support functions such as office space for administration or technical work. As such, activity levels for the critical group would be typical of the most strenuous activities in offices. From Table 3.7.2, the MET value associated with office work is 1.8. This value is assumed to be appropriate for the mix of activities performed by a typical office worker, including reading and writing at a desk and waming around the office. More strenuous activities may occur in office bulidings, however. Mail clerks and file clerks, for example, would plausibly have chronic MET values typical of walking (3.5), and cleaning and maintenance staff would also have larger MET values typical of walking or vacuuming (3.8). Such individuals would plausibly comprise the entical group in an office large enough to require these services full time, if these more strenuous activities are instead performed by part time employees, or are distributed among office staff, the critical group would consist of the relatively homogenous office workers having lower chronic activity levels. The MET value for the critical group in an office is therefore assumed to depend on the number of employees. The critical group for 'small' offices, with fewer than 100 employees, is assumed to consist of office workers with MET values of 1.8, while *large' offices are assumed to support a critical group devoted to more strenuous activities with MET values of 3.6.
i 3.7.10
l
. Construction Construction establishments, like mining establishments, would seem most l p
t likely to use decommissioned properties for support functions related to construction, and therefore be characterized by activity levels typical of office work. Some existing structures, however, might be used as warehouses or for prefabrication, and therefore be characteriod by activities associated with light to heavy industry, with MET values between 3.1 and 8. Half of the construction establishments are assumed to be devoted to office work, and half to more vigorous construction related work characterized by MET values uniformly distributed between 3.1 and 8.
. Manufacturina Properties used for manufacturing enterprises might also be devoted to office work, but are assumed to be more likely to support the more strenuous activit!es typical of light to heavy industry. Chronic MET values for the critical groups at these properties are assumed to be uniformly distributed between 3.1 and 8.
. Transoortation. Communication. Public Utilities Like construction, many of the more strenuous activities associated with transportation, communications, and public utilities (e.g.
cargo loading / unloading, installation and maintenance of infrastructure) would be expected to be ccnducted off site rather than in the buildings belongie to such establishments, Some fraction of the properties used by such establishments would be devoted to administration and support (e g. management, accounting, reservations), and therefore be characterized by office activity. Other properties might be used for vehicle maintenance, as warehouses, or in component manufacturing, and therefore be characterized by light industrial activities. Heavy industrial work is assumed to be very unlikely at these establishments. Half of these establishments are assumed to be devoted to office work, and half to activities with MET values uniformly distributed between 3.1 and 5.
. Wholesale Trade - Properties devoted to wholesale trade are assumed to be primarily used for warehousing, repackaging, and distribution, and therefore to be characterized by moderate to heavy activities with MET values uniformly distributed between 5 and 8.
. Retail Trade Properties used for retail trade might be used for management and administrative support (especially for large retail businesses), but are assumed to be primarily devoted to display, sales, and distribution of goods. Such properties would -
therefore be charactesized by the moderately strenuous activities associated with sales, and with unpacking and stocking goods. A chronic MET value typical of walking (3.5)is assumed to characterize the critical group at these establishments. Larger retail establishments (with 100 or more employees) would be expected to support full time dock workers and stockers who chronically engage in more strenuous activities. An MET value of 5 is assumed to characterize the critical group at these larger establishments.
. Finance. Insyrance. and Real Estate - Establishments classified under finance, insurance, and real estate include banks, brokerages, insurance agencies, and property management firms. They are assumed to support activities typical of office work. The MET values at these establishments is therefore assumed to depend on the number of employees, as described above.
3.7.I 1 j
. . ~ . . . - - ~ _ .-
e Semcn The service classification includes a broad range of businesses and asscciated activity levels for critical groups ranging from computer programming and tax preparation (presumably characterized by activities typical of office work) to linen supply, car repair shops, and dance studios (characterized by moderate to heavy activities). The number of establishments in each of these sub-categories can be requested from the U.S. Census Bureau, but was not available for this study. Because service establishments tend to be small(with 82% having 20 or fewer employees), a entical group engaged in chronic heavy industry is assumed to be unlikely. Considering the diversity in covered occupations, the MET values for the entical group are assumed to be uniformly distnbuted between 1.8 and 5 across these establishments. Table 3.7.5 summarizes the establishment classifications and sub-classifications discussed above, and the corresponding MET values or value ranges assumed for each establishment classification. Corresponding values for breathing rates associated with each establishment l classification can be calculated using Equation (3.7.3). 3.7.5 Proposed Distribution for V. Based on the discussion in the previous sections, the proposed distribution for Vo for the second iteration of the parameter analysis was derived using Equation (3.7.3). Average physiological parameters for males aged 30 to 60 years were assumed. BMR = 6.75 MJ/d; H
= 0.05 L O/kJ; VQ = 27. These parameters are treated as constants because variability in individual breathing rates is more strongly controlled by variations in activity levels than by l variations in physiological characteristics (Table 3.7.3), and because variations in physiology l among members of the critical group will tend to be suppressed when calculating the breathing l rate for the average member. The variability in MET values at licensed sites is assumed to be adequately approximated by the variability of the values assigned to general commercial properties, as shown in Table 3.7.5. Using the variables defined for Equation (3.7.3) and the MET values assigned in Table 3.7.5, corresponding values for breathing rates can be calculated. For example, an MET value of 1.8 corresponds to a breathing rate of 0.68 m8 /hr
((1.8 x 6.75 x 0.05 x 27)/(24 h/d)). For increasing values of the breathing rate, the cumulative frequency is the fraction of sites characterized by breathing rates less than that value. The cumulative frequency for a particular value of breathing rate is calculated by summing the fraction of total establishments (column 3 of Table 3.7.5) with values less than that breathing rate. For example, the cumulative frequency for a breathing rate of 0.68 m3 /hr is 0.14, corresponding to the fraction of L,es with an MET value of 1.8 (0.0036 + 0.0473 + 0.0158 + 0.0720). The data to construct the distribution for V, are given in Table 3.7.6. Figures 3.7.2 and 3.7.3, respectively, show the cumulative probability function and the probability density function for the proposed distribution. The probability density function is defined as the derivative of the cumulative distribution function. It was calculated by dividing the change in cumulative frequency between successive values of breathing rates by the change in breathing rate. Some of the establishment classifications are characterized by constant MET values, and therefore constant breathing rates; other have ranges of MET values assigned. The resulting cumulative distribution shows abrupt changes around, and linear changes between 3.7.12
Table 3.7.5 Estimsted Critical Group MET Values by industrial Division Number of Fraction of Total MET Value Classification Estabhshments Establishments or Range Mining Small offices
- 22,323 0.0036 1.8 Large offices" 6.247 0.010 3.6 Coi. truction Offices *" 300,150 Small Offices
- 295,547 0.0473 1.8 Large Offices" 4,603 0.0007 3.6 Prefabrication /
Warehousing /etc."' 300,150 0.0480 3.1 to 8 Manufacturing 387,337 0.0620 3.1 to 8 Transportation, Communication, & Utilities Offices"* 133,588 Small Offices
- 98,668 0.0158 1.8 Large Offices" 34,920 0.0056 3.6 Maintenance /
Production /etc.*" 133,588 0.0214 3.1 to 5 Wholesale Trade 509,604 0.0815 5 to 8 Retail Trade Small* 1,156,160 0.1849 3.5 O Large** Finance, insurance & Real Estate 398,277 0.0637 5 Small Offices
- 450,346 0.0720 1.8 Large Offices" 159,146 0.0255 3.6 Services 2,294,559 0.3670 1.8 to 5 Fewer than 100 employees
~ 100 or more employees
~ Half of the establishments in this classification assumed these constant values. For breathing rates where the cumulative frequency changes abruptly (e.g.,0.68 m 8/h and 1.33 m 8/h), the derivative is undefined. The probability density function at these points is indicated by arrows on Figure 3.7.3 denoting impulse functions, The numbers in parentheses are the weights, or integrals, of these impulse functions.
The distribution used in the first iteration of the parameter analysis is included in Figure 3.7.3 for 8 comparison. The revised distribution has a median value of 1.33 m /hr, with upper and lower 8 limits of 0.68 and 3.04 m /hr. /D g 3.7.13
Table 3.7.6 Estimated Cumulativo Distribution of V, for the Occupancy Scenario Breathing Rate Cumulative (m8/hr) Frequency 0.68. 0.0 0.68 0.14 1.18 0.29 1.33 0.34 1.33 0.53 1.37 0.54 1.37 0.57 1.90 0.78 1.90 0.84 3.04 1.00 The proposed distribution generally leads to larger values for Vo than the original distribution. The low end of the original distribution (0.39 m8 /hr) is characteristic of individuals at rest. The critical group at a given site is assumed to be composed of workers engaged in the most strenuous activities at the site, and would therefore exclude such individuals. The original upper limit of 1.5 m'/hr is characteristic of
- light activity" and is consistent with recent estimates of the average breathing rate for construction workers cited in Linn et al. [1992). In deriving the proposed distribution, some number of sites are expected to have critical groups regularly engaged in more strenuous work.
PARAMETER UNCERTAINTY: The proposed distribution describing the variability of breathing rate for the critical group rests on several assumptions which introduce uncertainty into the proposed distribution: . The entical aroup is assumed to be defined with reforence to the most strenuous occupation at a given site. . Physiological parameters are assumed to be independent of occupation. . The critical group is assumed to be sufficiently large so that individual variations in physiological parameters are suppressed in the average value for the group; . The relative frequency of most strenuous occupations is unkr.own, and has been estimated from the total number of establishments in various industrial classifications in 1993, and the assoc:sted descriptions of those classifications, reported by the U.S. Census Bureau. 3.7.14
i i 1- ,
!) i- Ij i, '1 ,
. )
i 0.9 p ]. -}.-j l ' ,-, 4- --
. . 7 .. . , ; i , .. , ,
08 .J1. .j ,ll1lj!!! ! 4 ... .i.. ......... i ' ~ I-g' , ;, . g 01 . .e. .
.; q;. , ; , g .
7 l ;l-.j , Proposed 06 _, .
.el4 p
.l= b g' 3 Dstrbution !
O.S E - . 4 - -... L - i 4 First teraton 1))- l l j 04 0.3 -. -. 1 IZE 1_ _--- _ . o 02 .
-[ . - . -
0.1 . O 0 0.5 1 1.5 2 ?* 3 3.5 Vo (in ^3/hr) Figure 3.7.2 Proposed Cumulative Probability Function for Breathing Rate (V ) i t w 0.5 0 45 !0.19)j j 0.4 R 0.35 (0.14)A . I fp o25 0.3 g(0,06) 0.2 0.15 (0.03) 0.1 d k 0.05 0 0 0.5 1 1.5 2 2.5 3 35 Vo (m *3/hr) Figure 3.7.3 Proposed Probability Density Function for Breathing Rate (V,) 3.7.15
+ Individual variations in the amount of oxygen required to metabolize food were est umeo to be small. . The MET values characterizing activities other than Heavy Industnal Labor are assumed to reflect long term average behavier over one year. Workers in a critical group engaged in heavy industriallabor are assumed to have a chronic MET value of 8.0.
VARIABILITY ACROSS SITES: The annual average breathing rate for the average mernber of the critical group will vary across sites due to differences in the use of the properties, differences in the composition of the critical group, and variations in physiological factors. Variability in site use appears to be the most important factor. SITE DATA COLLECTION: Applicants are not expected te collect breathing rate data to support alternative values for this parameter. Instead, applicants may attempt to support upper limits on breathing rate based on the intended use of the property. DEFINITION OF SITE DATA SOURCE: For the purpose of defining the distribution of Vo, the r qual average breathing rate for individual occupants was assurrad to be ava.. dole. NRC INTERPRETATION OF SITE SPECIFIC VALUE: As discussed above, applicants may att3mpt to support alternative values for Vo based on the projected use of the property, if so, the NRC may require legal assurances that the property will be used in a manner consistent with the proposed parameter value.
REFERENCES:
Kennedy, Jr., W.E., and D.L. Strenge,1992.
- Residual Radioactive Contamination from Decommissioning: Technical Basis for Translating Contamination Levels to Annual Total
- Effective Dose Equivalent," NUREG/CR 5512, Volume 1, U.S. Nuclear Regulatory Commission, Washington, DC.
American Industrial Health Council (AlHC) (1994) Exposure Factors Sourcebook. AlHC, Washington, DC. i l Basiotis, P, P., R. G. Thomas, J. L. Kelsay and W. Mertz (1989) Source of Vanation in Energy j Intake by Man and Women as Determined from One Year s Daily Dietary Records. Am, J. Clin. Nutr. Vol 50 pp 448-453. I California Air Resources Board (CARB)(1993) Measurement of Breathing Rate and Volume in i Routinely Performed Daily Activities. Human Performance Lab, Contract No. A033 235. June 1993. l Durnin, J. V. G. A. and R. Passmore (1967), Energy, Work and Leisure. Heinemann l Educational Books Ltd., London. International Commission on Radiological Protection (ICRP) (1931). Report of the Task Group on Reference Man, Pergammon Press, New York. 3.7.16 l
James. W. P. T, A. Ralph an') A. Ferro Luzzi(1989) Er.ergy Needs of Elderly, A New Approach
/n Munro, H. N., D. E. Danford eds., Nutrition, Aging, and the Elderly, Plenum Press, O New York., pp 129 151.
Layton, D. W. (1993) Metabolically Consistent Broathing Rates for use in Dose Assessments. Health Physics, Vol 64 No 1,1993 pp 23 36. Linn, W. S., D. A. Shamoo and J. D. Hw , ey (1992). Documentation of activity patterns in
*high risk" groups exposed to ozone in the hs Angeles area in Proceedings of the Second EPAIAWMA Conference on Tropospheric Ozone, Atlanta. I!ov 1991 pp 701 712. Air and Waste Management Assoc., Pittsburgh, PA.
Linn, W. S., C. E. Spier and J. D. Hackney (1993) Activity Pattems in Ozone-exposed Construction Workers, J. Occ. Med. Tox... Vol 2 No 1 pp 1 14. McLean, J, A. and G. Tobin (1987) Animal and Human Calorimetry, Cambridge University Press, Cambridge, MA. Prenticc, A. M., H. L Davies, A. E. Black, J. Ashford, W. A. Coward, P. R. Murgatroyd, G. R. Goldberg, M. Sawyer, and R. G. Whitehead (1985) Unexpectedly low Levels of Energy Expenditure in Heally Women. Lancet 1 pp 14191422. Riumallo, J. A., D, Schoeller, G. Barrera, V. Gattas, R. Uauy (1989) Energy Expenditure in Underweight Free-living Adults: Impact of Energy Supplementation as Determined by
' Doubly Ubeled Water and Indirect Calorimetry. Am. J. Clin. Nutt. Vol 49 pp 239 246.
O Sallis, J. F., W. L Haskell, P. D. Wood, S. P. F, nann, T. Rogers, S. N. Blair, and R. S. Paffenbarger Jr. (1985) Physical Actwily ssessment Methodology in the Five-City Project. Am. J. Epidemiol. Vol 121 pp 91 106. Saltin, B. and P. O. Astrand (1967) Maximal Oxygen Uptake in Athletes. J. Appl. Physio. Vol 23 pp 353 358. Schofield, W. N. (1985) Predicting Basal Metabolic Rate, New Standards and Review of Previcus Work. Human Nutr, Clin. Nutr. 39C Suppl 1 pp5-41. Spier, C. E., D. E. Little, S. C. Trim, T. R. Johnson, W. S. Linn and J. D. Hackney (1992) Activity Pattems in Elementary and High School Students Exposed to Oxida;- Pollution. J. Exp. Anal. Environ. Epid. Vol 2 No 3 pp 277 293. U.S. Census Bureau (1997) United States Firms, Establishments, Employment, Annual Payroll, and Estimated Receipts by industrial Division and Enterpr;se Employment for 1993; URL www. census. gov /eped/wwwlsbOO1.html accessed on 5/12/1997, U. S. Department of Agriculture (USDA) (1984) Nutrient intakes: Individuals in the United States, Year 19771978, NFCS 1977-1978. USDA Human Nutrition Information Service, Report No 12, Washington, DC. 3.7,17
U.S. Department of Health and Human Services (DHHS) (1983) Dietary intakes Source Data: United States,19761980. National Center for Health Statistics, DHHS Publication No (PHS) 831681, Hyattsville. MD. U.S. Environmental Protection Agency (EPA) (1996). Exposure Factors Handbook, EPA Report No. EPA /600/P 96/002Ba. Draft of August 1996. U.S. Environmental Protection Agency (EPA) (1985), Development of Statistical Distributions or Ranges of Standard Factors used in Exposure Assessments. Washington. D.C.. Office of Health and Environmental Assessment, EPA Report No. EPA 600/8 85 010, l 9 4 e
3.8 Effective transfer rate for Ingestion of removable surface contamination from surfaces to hands, from hands to mouth, GO (m 8/h) Ingestion of removable surface contamination inside buildings that is transferred from contaminated surfaces via hands, food, and other items to the mouth is referred to as secondary ir.gestion. The parameter GO is defined as the effective transfer rate and provides a mechanism for calculating the quantity of secondary ingestion. The effective transfar rate is described as the surface area contacted per unit time, the contents of which are ultimately transferred to the mouth by inadvertent fingering of the mouth or placing contaminated objects, such as food, cigarettes, pencils, etc., that had been in contact with a contaminated surface, into the mouth. The default value for GO is defined in NUREGICR 5512, Volume 1, to be 1x10-
' m'/h. This value was defined based on the literature analysis of surface contamination Ingestion data. Eight references are listed for this data (Dunster,1962; Gibson,1979; Healy, 1971; Kennedy,1981; Sayre,1974; Lepow,1975; Walter,1980; and Gallacher,1984),
Half of these studies focused on intake by children of surface contamination. These estimates 8 tend to be larger than the correspondino estimates for adults (i.e., greater than 1x10'8 m/h). The range cf ingestion rates for the adult worker / members of i' a public is 4x10-6 to 1x10-8 9'/h. The default value of 1x10d m'/h is consistent with the range for adults. For the initial iteration of the parameter analysis, the effective transfer rate wns assigned a loguniform distribution with a lower limit of 4x10 m'/h and an upper limit of 3x10 8 mr /h. These values were based on the range of values from the literature cited in Table 6.5 of NUREG/CR-5512, Volume 1. A plot of this PDF is presented in Figure 3.8.1. The verticalline in Figure 3.8.1 represents the default value chosen. y/ (.., y., 6, ' n ., e ,e se se u o Figure 3 8.1 PDF for secondary ingestion for the first teration of the parafneter analysis. IMPORTANCE TO DOSE: As described below, the dose for the ingestion pathway is directly proportional to GO GO is therefore an important parameter for situations in which a significant proportion of the total dose is received through ingestion. USE OF PARAMETER IN MODELING: The parameter GO is used to calculate CEDE for internal ingestion dose (DGO) resulting from inadvertent ingestion of surface contamination. 7- y The relationship between GO and intemal dose due to ingestion is described by the following v) formula: 3.8.1
DGOp45.05'24't,'GO' ,,,DFG,'C,, (3.8.1) where J,is the number of radionuclides in chain 1, t,is the time that exposure occurs during the building occupancy period, C.,is the overage annual activity of the radionuclide j during the first year of the building occupancy scenario, and DFG,is the ingestion dose factor for radionuclidc J. The reculting internal ingestion dose is directly proportional to the effective transfer rate. As discussed above, GO measures the tendency for occupants to ingest surface contamination an a surface area per unit time. Ingestion is mediated by touching contaminated surfaces with the hands or other objects, and placing contaminated objects in the mouth. As such, GO is a summary measure of chronic behavioral patterns for the average member of the entical group in an occupatbnal setting. In Equation 3.8.1, all surface contamination is assumed to be available for ingestion by this mechanism, and the concentration of ingested material is assumed to be equal to the source concentration C,y. Tne overallingestion rate may be lower if the amount of
- loose" contamination (i.e. contamination available for transport by this mechanism) is less than the ts.al amount of contamination or if the ingested dust or soil;s on'y ,.artially composed of '
contaminated material. Equation 3.8.1 can be generalized to include the fraction of " loose' contamination and the fraction of contacted surfaces that are contaminated by scaling the available concentration: DGO745.05'24*t,'GO' ,,, DFG,* F.* F,'C., (3.8.1) where F,is the fraction of ' loose" contamination and F. is the fraction of the total surface area contacted by the receptor that is contaminated. This scaling is equivalent to defining an effective secondary ingestion trensfer factor as: GO = F,'F,'GO (3.8.2) and by replacing GO in Equation 3.8.1 by the effective rate GO'. This decomposition preserves the definition of GO as a measure of behavior (the area accessed per unit time), allows the effective parameter value used in the dose calculation (GO-) to properly account for imbedded or absorbed contamination, and functionally relates the secondary ingestion rate used in the dose calculation to the effective value of the resuspension factor (see Section 3.6) in a physically plausible way. 3,8,1 Review of Information Related to Secondary Ingestion Kennedy and Strenge [1992) summarize estimates of GO published prior to 1992. In general, these estimates derive from postulates about behavior or from measured rates of ingestion. Information on ingestion by adults is especially sparse, and no direct measurements of adult ingestion rates are cited as a basis for GO. In addition, most theoretical estimates cited for GO or for adult ingestion rates found in the literature (Dunster [1962), Gibson and Wrixon [1979), 2 Hawley (1985]) derive from the supposition by Dunster that 10 cm of surface area would be accessed by a typical adult in a typical day Hawley [1985), in calculating adult ingestion rates, assumed that arjults working outdoors would transfer contamination from the inside surface of the fingers twice during a typical day of outdoor work, implying a secondary ingestion transfer rate of 137 cm2 in an 8-hour day. This estimate, however, is speculative, and was proposed in 3.8.2
4 l the absence of empirical data on adult ingestion or behavior. For the second iteration of the parameter analysis, additionalinformation was reviewed to determine if other data or approaches, preferably more recent than those cited in NUREG/CR-5512 Volume 1 [ Kennedy and Strenge,1992), were available to provide a defensible basis for constructing a PDF for GO. Recent publications, including references cited in the Draft EPA Erposure Factors Handbook (1996) were reviewed to identify and evaluate data related to secondary ingestion transfer rate. The goal of most studies was to estimate rates of soil ingestion as a mass per unit time, isther than to estimate a transfer factor analogous to GO. In addition, most of the recent literature continues to focus on children. Because children are excluded from the critical group, and because children are presumably exposed to higher densities of dust and soil, and to inger.t dust and soll at greater rates for a given density, estimated ingestion rates for children we not considered to be directly relevant for estimating i GO. Several studies on soil ingestion have been published since 1990. Ingestion rates for adults have been measured or estimated by a number of techniques and under a variety of conditions. Sheppar(* (1995) summarized the literature and oescribed a basic model(" soil ingest,cn that included food consumption and other activities, such as mouthing and ingestion of non food items, concentration enrichment, and the bioavailability of contaminants in soil. He recommended the use of simple models, rather than explicit use of empirical data, for estimating sollingestion in humans. Reported values for soilingestion rates by normal adults, summarized by Sheppard(1995) from other studies, range from 1 to 65 mg/d. Soilingestion rates in adults have been estimated by 1) analysis of selected tracer elements in ( human diets and comparing the dietary intake of tracer elements with tracer elements in feces V and 2) observations of individual behavior patterns under a range of environmental conditions and activities, Recently, numerous studies on soilingestion rates have been conducted using a tracer method (BTM) developed by Binder (1986) [ Stanek,1995; Sedman,1994; Calabrese, 1995; Stonek et al.,1997 and others). Stanek (1995 and 1997) estimated soilingestion rates in adults based on mass balance studies in which intake rates were estimated from concentrations of several trace elements in foods, medicines, environmental dust and soil, and . feces. Both studies collected data over multiple 1-week periods,' during which each subject Ingested a controlled quantity of soll from their environment. This mass, along with soil mass ingested with food, was subtracted from the estimated mass in feces to estimate the daily amount of inadvertent ingestion. These studies are the only published measurements of adult ingestion found in the literature review, and are therefore the ra.y empirical basis for defining a distribution for GO. Two types of published data related to the secondary ingestion transfer factor were found: direct estimates of the area of skin surface (and therefore area of contaminated surface) contacted by mouth in a given time, and measurements or estimates of the rate of soilingestion by adults. No studies report actual measurements of contacted area: the two primary sources for direct area estimates are Dunster's [1962) proposal that 'in order to arrive at some indication of the magnitude of the problem, it is assumed here that a person may ingest all the contamination from 10 cm' of contaminated skin every day,' and Hawley's [1985) assumption that adults working outdoors would transfer contamination from the inside surface of the fingt rs twice during a typical day of outdoor work, implying a secondary ingestion transfer rMe of 137 cm'in an 8-hour day. Both estimates, while plausible, have no empirical support. 3.8.3
3.8.2 Inferring GO from Ingestion Rates Estiniates of inadvertent soil ingestion rates by adults provide indirect information on secondary l ingestion transfer rates. The rate of soilingestion by an individual can be related to the l Individual's behavior (reflected in the secondary ingestion transfer rate for the individual), and to the environmental conditions (reflected in the average dust or soil loading experienced by the individual) using the following simple model: Sie,=GOe,
- DLe, (3.8.3) where Sl is the inadvertent soil ingestion rate [mg/hr), GO is the transfer factor (m^2/hr), and DL .
l is the average surface density of dust or soil which is ingested. The suffix C,1 denotes chronic (annual average) values for individual subjects. Equation 3.8 3 is consistent with the exposure model used in dose assessment (Equation 3.8.1). In the absence of direct mear" aments of transfer facior, this model was used to derive a distribution of individual transfer factor values from es. .iated distrbutions for soil 17estion rate and cstimated distributions for soil densities. The chronic behavior of individuals, characterized by goc,, is assumed to be independent of their environment, characterized by DLe,, so that E(log (SI )) = E(log (goc,)) + E(log (DLe,)) (3.8.4a) and Var (log (Sle,)) = Var [ log (GOe,)) + Var [ log (DLe,)) (3.8.4b) where E[X) and Var [X) deriote the expected value and variance over the population of individuals. Equations 3.8.5 allow distributional properties of GOe, to be inferred from distributional properties of Sle, and DLet This procedure requires a distribution for Sle,, describing the variability of soilingestion rate over iMividuals, and a distribution for DLe,, describing the variability in the soil density on skin corresponding to the conditions under which SI was measurmi or estimated. Defining a distribution for GO entails four main steps: A. Estimating distributional properties for individual chron;c soil ingestion rates (Sle,) from available literature. As discussed in Section 3.8.1, there are few published estimates of adult ingestion rates. The summaries of acute (daily)individualingestion rates provided N Stanek [1997] provide the most recent experimental basis for estimating adult soil ing6 Nt This study is therefora considered in some detail. B. Estimating distributional properties for the individual chronic soil densities (DL ) corresponding to the expenmental situation in which the ingestion rates were measured or estimated. C. Deriving distributional properties for the individual chronic transfer factor (GO ) from the distributional properties of soil ingdion rate and soil density, assuming that the variations in transfer factor and soit density among individuals are independent. D. Deriving the distribution for the transfer factor for the average member of the critical group from the distribution of individual transfer factors. 3.8.4
l l I 1 l Section 3.8.3 below describes the application of this procedure to derive a distobution for GO. l (] t "j A number of intermediate assumptions and inferences are required, which create a large I degree of uncertainty in the derived distribution. These assumptions are summarized below, e By using this model to estimate transfer factors for individuals from measurements or estimates of soilingestion rate, allinadvertent ingestion (i.e. excluding ingestion through food and medicine) is assumed to occurs through transfer from surficial sources: other potential sources, such as swallowed wind borne soil, are neglected.
- Measured inadvertent ingestion Sl includes dust and soil ingestion from any surfaces in the subject's environment, in the occupancy scenario model, surface contamination is assumed to occur only on walls and floors. As a result, secondary ingestion transf er factors inferred from measured ingestion rates will overestimate transfer factors from the contaminated surfaces considered in the scenario. Using the effective transfer factor GO' defined in Equation 3.8.2, F. = 1 for all measured ingestion rates, while F,is expected to be less than i based on the source location assumed in the scenario definition.
. Most of the few available estimates of adult ingestion rates are for residential environments, while the parameter GO characterizes occupational environments. In Equation 3.8.3, ingestion rate is decomposed into a behavioral component GO,, and an environmental component DLe,. Both components will differ between residential and occupational settings, although the extent and direction of this difference is uncertain. Individuals are possibly more likely to engage in habitual mouthing behavior in a domestic setting, however higher Q stress in an occupational setting may also lead to increased frequency of
- nervous" habits such as nail biting. Transfer rates based on soilingestion in a residential setting are assumed to be representative of transfer rates in an occupationa! setting. Under this assumption, soil ingestion rates in residences would be higher than ingestion rates in occupational settings solely due to the higher soil density in residences.
. in deriving equations 3.8.4, GOe,and DL., are assumed to be independent. Individuals who tend to behave in ways leading to large (small) transfer factors are not preferentially exposed to environments with high (low) dust densities. This assumption is plausible, but cannot be tested with available information.
- The parameter analysis requires a distribution describing the variability among licensed sites of the transfer factor for the average member of the critical group. The distribution of GO,,, however, describes the variability of transfer factors among individuals. At a given site, the critical group is assumed to consist of those individuals characterized by large values of GO,,. Deriving a distribution for GO from the inferred distribution of goc, requires assumptions about the factors potentially influencing individual transfer rates, about how those factors might be affected by different occupational situations, and about the relative frequency with which those situations would occur over the range of licensed sites. These assumptions are discussed in detailin Section 3.8.3.
- The available information c n adult soil ingestion rates is quite limited, and is not sufficient to determine the distribution c4 41... Similarly, the distribution of DL , corresponding to the reported ingestion rates is highly uncertain. For both soilingestion rate and soll density, the p)
( v mean, minimum, and maximum values of these distributions were estimated as detailed 3.8.5
4 below in Section 3.8.3. Without more specific information on the form of these dit tributions, distributions were assigned using the principle of maximum entropy. As stated by Jaynes [1982), this principle requires that "When we make inferences based on incomplete information we should draw from them that probability distribution that has the maximum entropy permitted by the information we do have." In as much as the form of the secondary ingestion rate and dust loading distributions are unknown, the assumption of any specific distribution is arbitrary, and likely to be wrong. Given this uncertainty, the maximum entropy distribution was judged the most reasonable choice in that "most information thet ists have considered it obvious that, in some sense, the possible distributions are concentrated strongly near the one of maximum *npy" (Jaynes,1982). With a specified mean value, lower limit, and upper limit, the maximum entropy distribution corresponds to a truncated exponential distribution. 3.8.3 Derivation of a Distribution for GO The procedure described in Section 3.8.2 was used to develop a distribution for GO. Details and intermediate results are presented below. A. Distributional Properties of Chronic IndividualIngestion Rate A.1 Mean IndividualIngestion Rate Sheppard (1995) provides a summary of current literature on soilingestion, and cited soil ingestion rates by normal adults ranging from 1 to 65 mg/d. These estimates include the theoretical calculations by Hawley [1985), based on assumed transfer rates and soil densities, as well as the estimates based on tracer measurements reported by Calabrese [1989,1990). Stanek [1997) describes a more recent application of the "best tracer" method to estimate adult soilingestion, which drew from a larger number of subjects and a longer measurement period than the earlier work of Calabrese (1989). Individualingestion rates reported by Stanek appear to be the strongest available experimenta' basis for estimating adult soil ingestion. This study is therefore considered in some detail. Soil ingestion rates were estimated by Stanek et al. [1997) for each of 10 adult subjects on each of 28 days. The measurement period was divided into four periods of 7 days each During each period, a known mass of soil was ingested by each participant. This mass, along with the estimated soil mass ingested with food, was subtracted from the total estimated ingested mass, yielding 280 values for daily individualinadvertent ingestion. Total ingested mass on a given day was estimated as the mass of dust and soilin feces on the subsequent day. Soil and dust masses in meals and feces were in turn estimated from measured concentrations of 8 trace elements found in soil and dust (Al, Ce, La, Nd, Si, Ti, Y, and Zr). Resulting f stimates of daily soilingestion, and daily dust ingestion are summarized in Stanek et al. [1997). This summary describes the distribution of daily individualingestion rate estimates over the entire study period, and over each of the four 7-day intervals. There is considerable variability in these estimates, as illustrated in Figure 3.8.2 Many negative values are reported, suggesting that a large amount of the variability in reported values is due to experimental error rather than to variability in ingestion rate among individuals, or to variability over time. Daily estimates for a single individual over a one-week period (Stanek et al.1997 Table 8) suggest that estimates of chronic ingestion rate may be considerably more stable than daily values, 3.8.6
however chronic rates cannot be derived for allindividuals from the summaries presented in the t report. Overallingestion rates, averaged over both time and individuals, are provided, and
\ have been used to estimate the potential variability in chronic dust ingestion over individuals, as discussed below, t
-- ._ _ .L _. - -.~ _ _ _ __
I 09i i
! t .
07 f 06 06 S 0 .
; o i '3- ,
l 01 i n 5000 500 0 600 1000 it,on i Ostly Duet ingestion Rete (mg/ day) { c i t
- Figure 3.8.2. Distribution of 5stimated Daily Dust ingestion Rates for 10 Adults and 28 Days U Based on the Median Value from 4 Tracers (data from Stanek et al.,1997 Table 10)
Table 3.8.1 shows the average dust ingestion rate over the 10 subjects for the entire study duration, and for each of the four time periods. Standard errors for the average are also reported, calculated from the sample standard deviations provided in Stanek et al. [1997). Table 3,8,1. Average Estimated Daily Dust ingestion Rates over 10 Individuals and Four One-week Observation Periode using Median Dally Values from the Four Best Tracer Elements (from Stanek et al.1997) 1 Period Average Dust Ingestion Standard Error' ' ' Rate (mg/d) (mg/d) Week 1 (0 mg/ day capsule ingestion) 139 52 Week 2 (20 mg/ day capsule ingestion) 73 22 Week 3 (100 mg/ day capsule ingestion) 129 32 Week 4 (500 mg/ day capsule ingestion) 225 32 All 4 Weeks 29 20
- Calculated from reported sample standard deviations The overall average ingestion rate of 29 mg/d is an estimate of the mean of the distribution of individual acute (daily) soilingestion rates over time and over individuals. The mean of this distribution is identical to the mean of chronic ingestion rates over individuals, Ster Due to the Q
3.8.7
large variation in individual daily values, there is considerable uncertainty in the estimate of the mesn, as indicated by the large standard error. Using two standard errors as an indication of this uncertainty, the experimental results are consistent with a mean ingestion rate between 0 and 69 mg/d. For comparison, Stanek and Calabese (1995) re analyzed results of their previous study of adult soll ingestion (Calabrese et al.,1990) using the best tracer method to rank the reliability of estimated rates based on individual tracers. The resulting average , ingestion rate over 6 adults and 3 weeks was 64 mg/ day. Available expenmental data appear to be consistent with mean ingestion rates for adults i between 0 and 70 mg/ day. The large variability in estimates of daily ingestion rate (e.g. Figure 3.8 2) lead to large uncertainty in the estimate of average chronic ingestion rate. Ingestion l l rates typically recommended for adults (e.g. 50 mg/ day in EPA (1996]) appear to reflect the l detection limit associated with current experimental practice. A.2 Upper and Lower Limits for IndividualIngestion Rate The minimum chronic individual soilingestion rate is evidently O. An upper limit for chronic adult soilingestio/ rate is more difficult to establish,! ever the experimental res"'ts summarized in Table 3.8.1 can be used, along with other information, to assign a plausible upper bound. For a particular subject, the chronic soilingestion rate (over the 250 day period relevant for the occupancy scenario) would be calculated as the average of 250 daily estimates for that subject. Average values for individual subjects are not avai!able in Stanek et al. (1997), 1 however the data in Table 3.8.1 indicate that the average ingestion rate over 210 subject days (that is the average over 10 subjects and 21 days) can be as large as 114 mg/ day, taking the average value over the three weekly periods having the largest weekly averages, or can be as small as O considering the three weeks having the lowest weekly averages. Soil ingestion by children has been much more extensively studied than adult soit ingestion. Children's soil ingestion rates tend to be larger than reported adult ingestion rates, presumably due to their more frequent exposure to soil, and to a higher rate of hand to mouth transfer. Ingestion rates for :nildren are therefore not appropriate as estimates for adults, but may provide information about reasonable upper limits for adults. A number of recent studies report measurements of soilingestion rates for children using tho tracer mass balance approach described above (Stanek and Calabrese,1995, Binder et al.,1986, Clausing et al.,1987, van Wijnen et al.1990, Davis et al.1990). The EPA Exposure Factors Handbook (EPA,1997) provides summaries and evaluations of these studies, leading to a recommended average ingestion rate for children of 100 mg/ day. This rate represents an average over individuals and over the various study periods, however the study periods were typically short (days or weeks), and were typically conducted in the summer when ingestion rates are expected to be higher than during other times of the year. An upper percentile (unquantified) of 400 mg/ day is also recommended in the EPA Exposure Factors Handbook. however low confidence is assigned to this estimate in view of the limited study period. An upper limit for the individual chronic adult soil ingestion rate of 200 mg/ day was adopted for this analysis based on the above information. This limit is consistent with the averages of daily rates from the limited sample reported by Stanek et al. (1997). The adopted upper limit for adults is larger than the average value recommended for children in the EPA Exposure Factors Handbook, however the latter value represents an average over individuals, while the former represents extreme behavior of a single individual, 3.8.8
_ _ . _ _ _ _ _ . _ . _ _ . _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ ~ B. Distributional Properties of Chronic Dust Loading B.1 Mean Dust Loading AJult ingestion rates from Etanek et al. [1997) and Calabrese et al. [1990) were muasured in a ! residential environment, and other published values for adult ingestion rate (e.g. Sheppard ; [1993)) typically describe residential conditions. Dust densities used to infer secondary ingestiori transfer rates from Equation 3,8.3, using ingestion rates measured or estimated for a residential environment, should therefore represent chronic values that may be encountered in this environment. Hawley [1985) discusses ranges of dust densities found inside residences. Citing Solomon and , Hartford [1976), he reports average dust densities for 239 floor dust samples taken from 12 homes of 320 mg/m' and 290 mg/m' based on concentrations of Pb and Cd, respectively. The larger number was adopted as an astimate of the average chronic dust concentration DL.,. B.2 Upper and Lower Limits for Dust Loading l< A sower limit on DL., was established based on the range of reported indoor dust fall rates discussed in Hawley (1985), and assuming daily remeval of accumulated dust. In a sample of suburban homes with closed windows, Shaefer et al. [1972, cited in Hawley,1985) measured a mean dust fall rate of 20 mg/m'/ day. This dust fall rate is the lowest cited by Hawley, and with the assuniption of daily cleaning, corresponds to a chronic average density of 10 mg/m' as a , lower limit in residential environments. .i ingestion rates in a residential setting may include ingestion while outdoors, where the subject's hands may become heavily solled. The burface soil density to which the individualis exposed in outdoor settings is assumed to be limited by the density of soil retained on the hands. Sheppard [1994) summarizes measured and estimated soilloads on hands for a variety of sod types and conditions, reproduced as Table 3.8.2. An upper limit of DL., of 5000 mg/m' was . ; assumed on the basis of these estimates. This density is generally consistent with reported > densitiet for soiled hands, with the notable exception of Hawley's theoretical value of 3.5 mg/cm'. Sheppard [1994] proposes that soll loads higher than 1 mg/cm' would prompt cleaning, and that higher densities would therefore not be associated with chronic ingestion. Table 3.8.2, Measurements and Estimates of Soll Load on Hands for Freshly Solled or Partially Cleaned Hands from Sheppard 1994, Table til Reference Load Conditions (mg/cm') Driver et al. (1989) 0.2-0.9 Dry whole soil, no cleaning 0.8 2 Dry sieved soil, < 150 pm diameter Hawley (1985) 3.5 Estimate assuming 50 pm thick covering Lepow et al. (1975) 0.5 Children.- sampled with adhesive film - _ Que Hee et al. (1985) 0.5 House dust adhering to palm Sheppart and Evenden (1994) 0.06 2 Dry soil, brushed clean, adhesive film sample s 4 3.8.9
Table 3.8.2. Measurements and Estimates of Soll Load on Hands for Freshly Solled or Partially Cleaned Hands from Sheppard 1994, Table lli
-8 Reference Load Conditions (mgicm')
0.3 0.5 Moist soil, brushed clean, adhesive film sample 0.40.8 Wet soil, brushed clean, adhesive film sample
<1 Visually clean, adhesive film sample C. Estimated Distribution for Chronic Individual Transfer Rato goc, The variatior, among individuals in chronic values of soilingestion, and of surface soil densities corresponding to the conditions in which that ingestion occurs, have been characterized by a mean value, an upper limit, and a lower limit. Without additional information to define the distributions for soil ingestion rate and surface soit density, a maximum entropy distribution was at ~ gned for both variabler With a specified mean value, lower lirr", and uppe, . unit, the maximum entropy distribution corresponds to a truncated exponential distribution. Figures r
TTiti!i" !Ti][iiii 'i 'iiij!!i , r!;["- [qqji{,!;;g. ! 09 4 . ....,.4 . . , - , -
, -i c.
7 .- 7 06 ... . . . .. ... . . .... .g.
- ,-,.,;.m '
07 ., sl(Mien s 50) .,; , ; ,,, {, [ , ,;,, , I !
. . . si <ueen = 0 s) j> H n -
; De . . : p.:. . .: :1 ::" .;;
,, ,, l_ , , I ,f. . h .
. ,_!ll !'p%p-.ly:,y i i lOs { .[}lId-.-l!P I!d
! :d la i .I[hh {.{![iL
! !jf . -_
M' h l ' di j j ;tM-. j}i! i H i 04 .' ..lil l! - , q' ! U ,
..y 03 , . o,p . , . , . , - ........ . . . . . . . -. -..i,...
i 02 ij +il ! jly' ,
- +
i l '! .!f
!: i: j .4j :ip?
i!'% I'='I in!!bb.jl iliiW - f
. u!L i l01 djf k jI i - I_ h .b I !
l I 0-..IIluib . - tzt.
, :! @i' ' '
_ Ub i ! W' l!
' l il l} !U l O001 0 01 01 1 10 100 1000 Dust ingestion Rats (mgid)
Figure 3.8.3. Estimated Distributior, of Chronic Dust ingestion Rates l Based on Two Alternative Mean Ingestion Rates 3.8.3 and 3.8.4 show the assigned distributions for Sle, and DLc,, respectively, i I As discussed in Section 3.8.3. (A) above, there is considerable uncertainty in the estimate of mean ingestion rate due to the large variability in daily ingestion estimates. Available data are consistent with mean ingestion rates between 0 and 70 mg/ day. To illustrate the effect of this uncertainty, two alternative distributions for Sie, denoted SI',, and Slve ,, based on alternative mean ingestion rates of 0.5 mg/ day and 50 mg/ day, are shown in Figure 3.8.3. 3.8.10 O l l
m iiT.it i . . .
. > i. '
09 - - . i . H i j h ,. j h ...-
. l b . 2 j p-- . . . j7 v Os m.
t4!' u . ..; I I,i lf l i* , j-07 . . . , . , - - De .. i }_;241 4 _; .-..- .r__ <
- i - . -- , J . .
Os .._ _ 1 [ d _. !
]-jj '! . ! 'n !
04 ___ _ - . 4._ . , _ - . _ _ . yj _y_. _, 03 _._ ._ . _ _ _. L1 _ __. ._i .J, 02 - .-. _ . . . . .. . . . - - . . - -. . .
~_;
.(
0.1 __ _. - .___. _ _ . _ _ _ ._ _ _ . ._. q 0- J ! I 10 100 1000 10000 bust Deneity (mel.n'2) Figure 3.8.4. Estimated Distribution of Chronic Surface Dust Densities ! The mean and variance of the logarithm of chronic individual transfer rate GO , was calculated using Equations 3.8.4, given the mean and variance of the logarithms of SI,, and DL,,. The alternative distributions for dust ingestion, Sl',, and Sl% were each used to evaluate the effect of uncertainty in mean ingestion rate on the inferred distribution of trantfsr rate, producing the corresponding transfer rate distributions GO'e, and GO% Table 3.8.3 summarizes tne i properties of these disthbutions. Loguniform distributions were then defined for GO',, and
% goo... based on the calculated mean and varirna:e of log (GO',,) and log (GO%) from Table 3.8.3 Figure 3.8.5 shows the dsrived distributions for GO ,. In converting the units of GO,, y from m*/ day to m'/hr, measured dust ingestion was assumed to occur over a 16-hour period.
This period corresponds to the period during which the reported soil ingestion rates would typically be operative. Table 3.8.3. Distributional Properties for Chronic Individual Dust Ingestion Rate (SI.,) , Dust Density (DL.,) and Transfer Factor (GO..,) Parameter Mean Lower Limit Upper Limit Mean of Variance of Log,. Log,. ) SI',, [mg/d] 0.50 0 200 -0.55 0.30 Sl% [mg/d] 50 0 200 1.47 0.29 DL , [mg/m^2) 320 10 5000 2.29 0.22 GO',., [m ^ 2/d] 1.8E-3 4.4E-4 4.6E 3 2.85 0.09 GO% [m^2/d] 1.CE 1 5.1E 2 4.3E 1 -0.82 0.07 3.8.11
1 - . - - , , . , n ,,. , , , , , , j
; 09 l !l ! f l!!
- -j { - [ , j . .a . .
'll I 06 -d I !!!; e .[ $ . .i: ! ! .i i 07 - ; . ! .j - J i a e-.
,,i
" 3 !!!!<* ; ij; e - ,' d s 06 --, 4 i l g e -- -,- r4 4 4 4
. -- i - + . -. p
!il i '
jilo i li, u 1 0$ _ _ _ . - l -- - - - ~4 - - - - L - i -- ,- -- 7 04 __ w.! n q-+-- p ,1---+- .I- - - ! -+ - h i 03
~' ~ ~ - -~ '
Mean ingestio' n Rate a $0 mg/d 02 -
. . . Mean ingestion Rate = 0 $0 mg/d , _ _ _ , _ _
}4 . _ _ . _ . ,
.. y ,. )
01 - . _ _ _ .8 -
- -h ~} - - - - - -
h 0 * - 1E.05 - '4 1E-03 1E42 1 E.01 Secondary Ingestion Transfer t . tor (m*2thr) Figure 3.8.5. Estimated Distributions of the Chronic Individual Secondary h Ingestion Transfer Factor goc, Corresponding to Alternative Mean Ingestion Rates } D. Estimated Distribution for Transfer Rate for the Average Member of the Critical Group The secondary ingestion transfer rate for individuals goc., describes the tendency for ind.sidmW to transfer mass from surfaces in their immediate environment to the mouth. This tendency may be influenced by a number of factors. Individual habits and mannerisms, such as smoking, snacking, chewing gum, nail biting, wiping the mouth, or covering the mouth with the hands @ may lead to elevated transfer rates. Different occupations may also induce behavior leading to % elevated transfer rates, however all occupations would appear to present opportunities for high h transfer rates when performed by individuals with common mannerisms. In administrative, professional, or cierical work, for example, wetting fingers to turn pages or chewing on pens or pencils would plausibly lead to high values of goc,. In manuallabor, wiping the mouth or face or holding small items in the mouth while wors.ing with hands would also lead to elevated transfer rates. p Because there is no empirical basis to differentiate transfer ratas by occupation, and no r compelling argument for proposing distinct ranges of rates for different occupations, the variability of individual transfer rates among occupants is assumed to be independent of occupation, and to be the same as the variability among individuals in general, approximated by the distributions in Figure 3.8.5. At a particular site, an applicant would ideally have information on the individual transfer factors for all future occupants. A critical group would then be defined as those occupants with high transfer factors. Applying the concept of critical group homogeneity, this group would be defined in such a way that the variations in individual transfer factor values within the group w as no greater than one order or magnitude. The site-specific transfer factor value would then be 3.8.12
._ _ _ . . _ . - . - _ _ . _. . _ ._ _ - . . . m . _ . _ . . _ _ . _ _ - the average of the ind!vidua ' transfer factor values over this group. (' ( Because the estimtad varit, tion in transfer factor values among individuals is less than an order of magnitude (Hou e 3.8.5), tu expected value of GO for the average member of the critical group is simply *,he expected value of the distribution of individual transfer factors. Because the distributian of individual values is assumed to be the same at all sites, variations in GO from site to site, calculated by averaging values for individuals at that site, would arise only from random variations among individuals in the critical group. Provided the number of individuals in the critical group is large, the distribudon of GO over sites will be very narrow, and centered around the average value for individuals. Figure 3.8.6 shows the var l ability of the average individual transfer rate for critical group sizes of 10,100, and 1000, assuming GO% for the individual transfer rates. Even for the smallest critical group size of 10, calculated average . values differ by less than a factor of 4. Using the central limit theorem, these distributioJs can be closely approximated by normal distributions with mean values equal to the mean individual ingestion rate, and variances equal to the variance of the individual rate divided by the critical group size.- 1 - .,_ .. 7 _. 7.
~~ ~ -~10 occupants"~
06 -. 100 Occupants t._. d. jj
. . 1000 Occupants i i ,
l l j 07 H ~CTMvh- i l_. 1 , b{ 06 ..---- - . . , - _. - . - 0$ y_. - -
\
l N] O4 -..._ _ 0.3 - . - - - - - - - -_.. - - . - .-- - i I 0.2 1
--t 1 01 - ~.h-j- , +.--
I I 0 1E.03 ' 1E-02 1E 01 secondary ingestion Transfer Factor (m*2thr) Figure 3.8.6. Distributions of the Secondary ingestion Transfer Factor for the Critical Group from Three Critical Group Sizes From the above argument, the variability in GO over sites is expected to be re!atively small, with most values clustered around the average individual transfer rate. The uncertainty in this average, however, which arises from uncertainty in the average adult ingestion rate, is relatively large. Figure 3.8.7 shows two possible distributions for GO, corresponding to average adult ingestion rates of 50 and 0.5 mg/ day. Both values are consistent with available data on adult ingestion, and the larger value is the approximate detection limit of current experimental procedures. v 3.8.13 i
[ 09
, j i ii <+
; ; ; 1 { Ii i i
.s .
t<< , 06 lill; ; I {li ! {!j,i!' ! ' '
-7
+
+ qj;++ -b+b+++t' +
]4 +;;'; * * ' " ' *
- 1 '
, o7 .
!!ll' I '
i ll' ;i ,,
-'~
t-{t]+ f ---+- t- t-[4 titj
. . - - , 4-l- 06 tf { t- itl{y 05 n-- ~
f 04 4 -- + ' Me Ingeston E 50 $ d f 03 _ _ _ _-
-- Man maa"=a o 50 mS'8 _E -y l l 02 - - - - - --- --- --
h- - --- ~ 01 --- - l - - 0 1E 05 1 E-04 1E.03 1E.0; . -01 Secondary ingestion Transfer Fa ctor (m*2/hr)
. _ _ _ _ _ ___ ___ _ - _ _ . _ _ . _ _ _ _ . . _ _ _ . _._5 Figure 3.8.7. Distributions of GO Corresponding to Two Alternative Ingestion Rates Assuming a Critical Group Size of 10 Uncertainty in the mean value of Sic, creates large uncertainty in GO relative to the estimated variability of GO over sites, as can be seen in Figure 3.8.7. Given this uncertainty, a conservative assignment of the distribution for GO (i.e. a distribution corresponding to large average ingestion) would insure that decisions made using existing information would be unlikely to change if additional information were available. Additionalinformation (such as a more accurate determination of mean ingestion rate, or demonstration of a " surface factor" (F.)
value less than 1) would reduce existing uncertainty in the transfer rate, and therefore reduce the values of a conservatively assigned transfer rate. Other information might also be used to reduce uncertainty in the distribution of GO. For example, measurements of the frequency of various mouthing behaviors among adults might be used to estimate the surface area potentially accessed through such behavior, as well as the fraction of this surface area consisting of walls and floors. No studies of this kind were identified in the literature review, however some transfer rates consistent with measured ingestion may be judged implausible from consideration of adult behavior in an occupational setting. A transfer rate of 108 milhr, for example, implies mouthing an area equivalent to the inner surface of the hand once each hour. A rate of 10' milhr implies transfer from an area roughly equivalent to two postage stamps each hour. The behavior implied by the latter rate is arguably a plausible upper limit for individuals in an occupational setting, and distributions having higher rates may be rejected on the basis of this judgment. The resulting distribution for GO would not be conservative with respect to uncertainty in the average ingestion rate given existing measurements, however the likelihood that additional information would lead to higher transfer rates would still be assumed to be small, in view of the behavioralimplications of these higher 3.8.14 l
rates. kA) Among the possible distributions of GO consistent with measured ingestion rates, the two distributions shown in Figure 3.8.7 are both recommended for use in the parameter analysis. This approach will allow the large uncertainty in GO (relative to the expected variability) to be reflected in the resulting default parameter values. The distribution centered around of 108 m'/hr (corresponding to a mean ingestion rate of 50 mg/ day)is proposed as a conservative bound: the ingestion rate from which this distribution was derived reflects the apparent detection limit of current experimental practice. The distribution centered around 104 m'/hr (corresponding to a mean ingestion rate of 0.5 mg/ day) includes plausible reductions from the conservative distribdion in consideration of two factors, each of which is assumed to reduce the transfer factor by an order of magnitude: the stipulation that an individual transfer rate of 108 corresponds to unreasonable behavior in an occupational setting; and the assumption that walls and floors are much less likely to be contacted than other surfaces, such Ls tables and desks. As discussed in Section 3.8, the actual amount of contamination ingested will also depend on other factors, including the fraction F, of the total source term that is
- loose *, and therefore available f" ingestion. The fraction of loose contamination (F,) is expected M be estimau or bounded ui.ng data collected prior to decommissioning.
For the upper distribution, the revised range for GO extends from 5.5x108 to 1.8x108 m>/h with a mean value of 1.1x108 man, and a median value of 1.1x10' m>/h. For the lower distribution, the range extends from 5.0x10' to 1.9x10' m'/h with a mean value of 1.1x104 m>/h, and a median value of 1.1x104 m>/h. From the central limit theorem, these distributions are well approximated O by normal distributions. The standard deviation of the upper distribution is 2.11x10'3 m'/h, while the standard deviation of the lower is 2.28x104 m'/h. PARAMETER UNCERTAINTY: The proposed distribution describing the variability of the secondary ingestion effective transfer rate rests on a number of assumptions, introducing a large amount of uncertainty in the assigned distribution. Because the uncertainty in the distribution is much larger than the anticipated variability of the transfer factor across sites, two alternative distributions are proposed. (1) Empirical support for this parameter is very limited. The most recent measurements of soil ingestion in adults are subject to wide variability, and are consistent with average ingestion rates ranging from 0 to 70 mg/ day. The proposed alternative values of 0.5 mg/ day and 50 mg/ day are consistent with available information. The larger value is represents the apparent detection limit of current experimental practhe The smaller value is not conservative with respect to current uncertainties in adult ingenen rate, but was established in consideration of judgments about 1) the plausibility of the behavior associated with higher rates, and 2) the fraction of the total contacted surface area consisting of walls and floors. (2) Ingestion rates have been measured for adults in residential settings. Transfer factors in occupMional settings, representing behavioral characteristics of individuals, are assumed to be similar to those in residential environments. Higher ingestion rates in residences are therefore assumed to be due to exposure to higher soil density, rather than to distinctive behavior, n (3) Surface dust and soil densities associated with available measurements of adult ingestion i rates are unknown, and have been estimated from independent studies of dust densities and (d 3.8.15
dust fall rates in residences, and soit densities on soiled bands (4) Transfer rates for individuals are assumed to be independent of occupation, and the critical group is assumed to be composed of those individuals at a given site having the largest transfer rates. The transfer rate for the entical group at a given site therefore depends on the number of members of the critical group. Based on this assumption, the variability among sites is quite small compared to the uncertainty in the average value. VARIABILITY ACROSS SITES: The secondary ingestion effective transfer rate may vary across sites depending on the behavior patterns of the site occupants and any imposed restrictions (i.e-, no smoking or eating on site). Behavior patterns are assumed to be unrelated to the potential uses of the site, and the applicant is assumed to be unable to guarantee the effectiveness of imposed restrictions for the entire performance period. The site specific average for the critical group at any site is therefore expected to be close to the average individual transfer factor. Little variability of this average among sites is expected. SITE DATA COLLECTION: Acolicants would not be expected to collect additional data on the secondary ingesticq effective transfer rate. DEFINITION OF SITE DATA SOURCE: For the purpose of defining the distribution of GO, additional information on individual occupants may be obtained from the EPA Exposure Factors Handbook. NRC INTERPRETATION OF SITE-SPECIFIC VALUE: Because transfer factor values for the critical group are not clearly related to specific occupations or activities, applicants are not expected to be able to provide information supporting site-specific values for this parameter.
REFERENCES:
International Commission on Radiological Protection (ICRP),1975. " Report of tne Task Group on Reference Man," ICRP Publication 23, Pergammon Press, New York. Kennedy, Jr., W.E., and D L. Strenge,1992.
- Residual Radioactive Contamination from Decommissioning: Technical Basis for Translating Contamination Levels to Annual Total Effective Dose Equivalent," NUREGICR-5512, Volume 1, U.S. Nuclear Regulatory Commission, Washington, DC.
Calabrese, E. J., E. J. Stanek, and R. Bames,1996, " Methodology to Estimate the Amount and Particle-Size of SoilIngested by Children: Implications for Exposure Assessment at Waste Sites", Regulatory Toxicology and Pharmacology 24(3),264-68. Calabrese, E. J. and E. J. Stanek,1995. " Resolving Intertracer inconsistencies in Soil Ingestion Estimation", Environmental Health Perspectives 103(5),454-57. Sheppard, S.C.,1995. " Parameter Values to Model the Soil Ingestion Pathway", Environmental Monitoring and Assessment 34(1),27-44. Stanek, E. J. and E. J. Calabrese,1995. " Soil Ingestion Estimates for Use in Site Evaluation Based on the Best Tracer Method", Human and Ecological Risk assessment 1,133-56. 3.8.16 l l l
, Calabrese, E. J. and E. J. Stanek,1994. " Soil Ingestion issues and Recommendations", J.
( Environ. Sci. and Health, Part A-Environ. Sci. and Engr. 29(3),517 30. Sedman, R. M. and R. J. Isahmood,1994. " Soil Ingestion by Children and Adults Reconsidered Using the Results of Recent Tracer Studies", J. Air & Waste Manag. Assoc. 44(2),14144. Calabrese, E. J. and E. J. Stanek,1993. "An improved Method for Estimating Soil Ingestion in Children and Adults", J. Environ. Sci.& Health, Pt. A-Environ. Sci.& Engr. 28(2),363 71. Calabrese, E, J. and E. J. Stanek,1991. "A Guide to Interpreting Soil ingestion Studies: 1) Qualitative and Quantitative Evidence of Soil Ingestion", Reg. Toxicol. & Pharm.13(3), 278 92. Stanek, E. J. and E. J. Calabrese,1991. "A Guide to Interpreting Soil Ingestion Studies: 2) Development of a Model to Estimate the Soil Ingestion Detection Level of Soil Ingestion Studies", Reg. Toxicol. & Pharm. 13(3),263-77. l Calabrese, E. J., E. J. Stanek, C.E. Gilbert, and R. M. Barnes,1990. " Preliminary Adult Soil l Ingestion Estimates: Results of a Pilot Study", Reguistory Toxicology and Pharmacology 12(1),88-95. Stanek, E. J,, E. J. Calabrese, R. M. Barnes, and P. Pekow,1997. '? Soil Ingestion in Adults - Results of a Second Pilot-Study", Ecotoxicology and Environmental Safety 36,249-257. D) t. Barnes, R. M., A. Lasztity, M. Viczian, and X. R. Wang,1990. " Analysis of Environmental and Biological-Materials by Inductively Coupled Plasma Emission and Mass Spectrometry", Chemia Analityczna 35(1-3),91-98. Binder, S., D. Sokal, and D. Maughan,1986. " Estimating Soil Ingestion: The Use of Trace Elements in Estimating the Amount of Soil Ingested by Young Children", Arch. Environ. Health 41(6), 341-45. Solomon, R. L. and J. W. Hartford,1976 " Lead and Cadmium in Dusts and Soils in a Small Urban Community," Environ. Sci. Technol., vol 10, 773-777 Schaefer, V. J., V. A. Mohnen, and V. R. Veirs,1972. " Air Quality of American Homes," Science vof 175,173-175. Hawley, J. K.,1985. " Assessment of Health Risks from Exposure to Contaminated Soil", Risk Anal. 5, 289-302. Jaynes, E. T.,1982. "On the Rationale of Maximum-Entropy Methods", Proceedings of the IEEE, vol. 70, no. 9,939-952. Dunster, H. J.1962. Maximum Permissible Levels of Skin Contamination. AHSB(RP)R28, United Kingdom Atomic Energy Authority, London. p ( Gibson, J. A. B. and A. D. Wrixon.1979. " Methods for the Calculation of Derived Working 3.8.17
Limits for Surface Contamination by Low Toxicity Radionuclides." Health Phyt,cs (36)3:311-321. Healy, J. W.1971. Surface Contamination: Decision Levels. LA-4558-MS. Los Alamos Scientific Laboratory, Los Alamos, New Mexico. Kennedy, Jr., W. E., E. C. Watson, D. W. Murphy, B. J. Harrer, R. Harty, and J. M. Aldrich. 1981. A Review of Removable Surface Contamination on Radioactive Materials Transportation Containers. NUREG/CR-1859, PNL-3666, U.S. Nuclear Regulatory Commission, Washington, D.C. Sayre, J. W. E. Charney, J. Vostal, and I. B. Pless.1974. ' House and Hand Dust as a Potential Source of Childhood Lead Exposure." Am. J. of Dis. Chil. 127:167-170. Lepow, M. L., L. Bruckman, M. Gillette, S. Varkowitx, R. Robino, and J. Kapish.1975.
' Investigations into Sources of Leadin the Environment of Urban Children.' Environ.
Res.10.414-426. Walter, S. D., A. J. Yankel, and I. H. Von Lindem.1980. ' Age-Specific Risk Factors for Lead Absorption in Children.' Archives of Environ. Health 53(1):53-58. Gallacher, J. E. J., P. C. Elwood, K. M. Phillips, B. E. Davies, and D. T. Jones.1984. " Relation Between Pica and Blood Lead in Areas of Differing Lead Exposure." Archives of Disease in Childhood 59:40-44. EPA,1996. Exposure Factors Handbook. EPA /600/P-95. Office of Research and Development, Environmental Protection Agency, Washington, D.C. (Current draft not cita'ble). U.S. Census Bureau (1997) United States Firms, Establishments, Employment, Annual Payroll, and Estimated Receipts by Industrial Division and Enterprise Employment for 1994; URL www. census. gov /eped/www/sbOO1.html accessed on 9/2/1997. 3.8.18 O i l
i i ) Building Occupancy d Default Dose and Concentration Tables 4 i Tables constructed by using the mean of the parameter distributions for behavioral parameters
- (breathing rate, occupancy time, and inadvertent ingestion) and the full distribution for the physical parameter (resuspension). The distribution for resuspension used in the calculations is 10% of the full distribution, on the assumption that the licensee will demonstrate that removable contamination is no more than 10% of the total surface contamination.
The first table presents the quantiles of the dose distribution by radionuclide, in units of mrem /y. , The second table presents the radionuclide-specific concentration (in units of dpm/100cm 2) equivalent to 25 mrom/y, J i
-w w ,p v p ,- a,,,, --,-r , , - - , -- +- -r--. g n ., ,, .g ,
Building Occup:ncy Radionuct.ds-Specific Dosa (Assuming 10% Remov:bla) 12/16/97 3.57 F M
-._ _ _ _ _ _ Juantile of tl.1 Dose Distributlon_ _ _ _ _ _ _ __ _
S:urce 0.01 0.05 0.1 0.25 0.5 0.75 0.9 0.95 0.99 3H 5.22E-08 5 35E-08 5 55E 08 6 34E 08 9.06E-08 1.45E-07 1,79E 07 1.98E 07 219E-07 g 10Be 5.50E-05 6 21E-05 7.35E-05 1.19E-04 2 74E-04 5 81E-04 7.77E 04 8 86E 04 1.01E 03 181E45 4 87E-06 6 02E-06 6.66E 06 7.37E-06 14C 1.77E 06 1.88E-06 2.15E 06 3 06E-06 22Na 2.60E-03 2 60E-03 2.60E-03 2.60E-03 2 61E-03 2.61E-03 2.62E-03 2.02E 03 2 62E 03 35S 2 90E 07 3.06E-07 3 32E-07 4 35E-07 7.89E-07 149E-06 1.93E-06 2.18E-06 2 46E-06 36Cl 6'21E-06 6 65E-06 7.36E-06 1.02E-05 1.98E-05 3.88E-05 5 09E-05 5.76E-05 6.51E-05 40K 2.19E-04 2.20E-04 2.20E-04 2.22E-04 2.27E 04 2.38E 04 2.45E-04 2.48E-04 2.52E-04 41Ca 1.08E-06 1.11E-06 1.15E-06 1.32E-06 1.91E-06 3.08E-06 3.82E-06 4.24E-06 4.69E 06 45Ca 1.63E-06 1.70E-06 1.81 E-06 2.24E-06 3.71 E-06 6.62E-06 8.48E-06 9.51 E-06 1.06E-05 46Sc 6 52E-04 8.52E-04 8.52E-04 8.53E-04 8.57E-04 8.66E-04 8.71E-04 8.74E-04 8.77E-04 54Mn 7.83E-04 7.83E 04 7.83E-04 7.84E-04 7.86E-04 7.90E-04 7.92E-04 7.93E-04 7.95E-04 55Fe 7.14E-07 7.61E-07 8.37E-07 1.14 E-06 2.18E-06 4.23E-06 5.54 E-06 6.27E-06 7.07E-06 57Co if7E-04 1.07E-04 1.075~04 1.05E-04 1.16E-04 1.15E-04 1.19s-04 1.20s!64 1.22E-64~ 55Co 3635'04 3.65E%4 3 63s-04 3.64E-04 3.65s-04 3.68 -04 3.69E-04 3.70E-04 3.7TE 04 60Co 3.14E-03 3.1G03 3.15E-03 3.17E-03 3.2C' '3 3.44E-03 3.55E-03 3.62E-03 3.69E-03 59Ni ~ 5.36E-07 ~ 5.90E-07 6.77E-07 ~ 1.02E-06 2.21 E-06 4.55E-06 6.05E-06 6.88E-06 7.79E-06 ~ 63Ni 1.3DE 06
~ ~
1.43E-06 163E-06 2 43E 06 5.i7E-06 1.06s-05 1.41E!65 160 EMS
~
1.81E 65 65Zn 4 9iE-04 4 915~04 4 92E-04 4 93s-04 4.99E-04 5.10E-04 5.17s-04 5 21 s-04 5.25E!04 75Se 2.24E-04 2.24E 04 2.24E-04 2.24E-04 2.26E-04 2.29E-04 2 31E-04 2.32E-04 2.33E-04 79Se 7.49E-06 7.68E-06 8 00E_06 9 26E_-06 1.36E_-05 2.21_E-05 2.75E-05 3.06E-05 3.39_E-05 90Sr 2.99E-04 3.25E-04 3 67E-04 5 31E-04 1.10E-03 2.22E-03 2.93E-03 3.33E-03 3 76E-03 93Zr 4.76E-05 5.40E-05 6 44E-05 1.05E-04 2.46E-04 5.25E-04 7.02E-04 8.01 E-04 9.10E 04 93Zr+C 5 34E-05~ 6.04E-05 ~ 716E-05
~ -
1.16E-04 2.70E-04 5.73E-04 7.66E-04 8.74E-04 9.92E-04
~ ~
93mNb 575E-06 C32E-66 75dE[06 1769E-05 233ET65 4 8dE-05 6.38E-05 7.26Ee05 8 2'2E 05 94Nb 2.21E-03 2.22E-03 2.23E-03 2.28E 03 2.46E-03 2.82E-03 3.05E-03 3.18E-03 3.32E-03 93Mo 1.27E-05 1.33E-05 1.42E-05 1.79E-05 3.07E-05 5.59E-05 7.20E-05 8.10E-05 9.08E-05 99Tc 2.33E-06 2.49E-06 2.76E-06 3.82E-06 ~ 7.47E-06 1.47E-05 1.93E-05 2.18E-05 2.47E-05 106Ru 2.75E-04 2.85E-04 2 56s 04 3T40s-04 4.925 % 4 7.91 E-04 9.815 % 4 1.09E-03 1.20s!03 107Pd 1.95E-06 2.20E-06 2 61E-06 4.24E-06 9.83E-06 2.09E-05 2.79E-05 3.19E-05 3.62E-05 110 mag 2.35E-03 2.35E-03 2.35E-03 2.36E-03 2.38E 03 2.43E-03 2.45E 03 2.47E-03 2.49E-03 109Cd 4 40E-05 4.58E_-05 4.86E 05 5.99E-05 9.84E-05 1.75E_-04 2.24E-04 2.51 E-04 2.80E-04 113mCd 3.24E-04 3.54E-04 4.02E-04 5.92E-04 1.24E-03 2.54E-03 3.36E-03 3.82E-03 4.32E-03 119mSn 1.1_1 E-05 1.11 E-05 1.13E-05 1.18E-05 1.36E-05 1.73E-05 1.96E-05 2.09E-05 2.23E-0_5 121mSn 1.02E 1.04E-05 1.08E-05 1.23E-05 1.75E-05 2.77E-05 3.43E-05 3.79E-05 4.19E-05 123Sn 9.75E-06 1.00E-05 1.05E-05 1.23E-05 1.85E-05 3.09E-05 3.88E-05 4.31E-05 4.79E-05 i~2 5Sn 2.75E-03 2.75E 03 2.76E-03 2.775-03 2.82E-03 2.90E-03 2.96E-03 2.99E-03 3.025!63 125Sn+C 2.78E-03 2.78E-03 2.78E-03 2.80E-03 2.84E-03 2.93E-03 2.99E-03 3.02E-03 3.05E !O3 525Sb 5.39E-04 5.39E-04 5 40E-04 5.418-04 5 47E-04 5.57E-04 5.64E-04 5.67E-04 5.71s!64 1 23mTe 8.55E-05 8.56E-05 8.57E-05 8.63E-05 8.82E-05 9.20E-05 9.45E-05 9.58E-05 9.73_E_-05 127mTe 1.25E-05 1.27E-05 1.30E-05 1.40E-05 1.77E-05 2.51E-05 2.98E-05 3.24E-05 3.52E-05 1291 2.53E-04 2.56E-04 2.62E-04 2.84E-04 3.60E-04 5.10d 04 6.06E-04 6.60E-04 7.18E-04 134Cs 1.86E-03 1.86E-03 1.86E-03 1.86E-03 1.88E-03 1.91 E-03 1.94E-03 1.95E-03 1.96E-03 135Cs 5.61E-06 5.70E-06 5.85E-06 ts43E-06 8.42E-06 1.24E-05 1.49E-05 1.63E-05 1.78E 05 137Cs 8.08E-04 8.08E 04 8.09E-04 8.14 E-04 8.27E-04 8.55E 8.72E-04 ' 8.82E-04 8.93E-04 144Ce 9.96E-05 1.05E-04 1.13E-04 1.44E-04 2.52E-04 4.67E-04 6.03E-04 6.79E-04 7.63E-04 147Pm 5.66E-06 6.35E-06 7.46E-06 1.19E-05 2.69E-05 5.68E-05 7.59E-05 8.64E-05 9.81E-05 Page 1 9
. . _ _ - - - _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ - _ _ _ . _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ _ - - -~__.)
Budding Occupancy Ra&onuchde-Specific Dos)(Assuming 10% R imov!bb) 12/16/97 3.57 PM Cuantile of the Dose Distribution
/~N. Source
! ! 0.01 0.05 0.1 0.25 0.5 0.75 0.9 0.95 0.99 147Sm 1.09E 02 1.24 E-02 1.48E-02 2.445-02 5.71 E-02 1.22E-01 1.63E-01 186E-01 2.11 E-01 151Sm 4.59E-06 5.18E-06 615E-06 9 95E-06 2.30E-05 4 89E 05 6 54E-05 7 46E-05 8.47E-05 152Eu 1.54E-03 1."SE 03 1.56E 03 158E-03 1.68E-03 1.86E-03 1.98E-03 2.05E 03 2.12E-03 154Eu 1.64E-03 1.64E-03 1.65E-03 1.69E-03 1.81E-03 2.05E-03 2.20E-03 2.28E-03 2.38E-03
~ ~
iS'5Eu 839535 8T45E 05 8 59hT05 9.d8$! 5 1.d5s-04 iiiE-04 1!65E-04 1T75E-04 1.88E-04 ~ iS3Gd 9.46E-05 9795705 9.54E-05 9.75E-05 1.0iE-04 1.16E-04 1.258-04 1.25E-04 1.34 E-04 I6 T'b 4.225-04 4.255-04 4.22E-04 4.2ie-04 4.26E 04 4.32E-64 4.35E-04 4.38h-04 4 40E 65 i66mHo 2.50E-03 2.52533 2.545763 2.64$-03 2T98hT03 3 65E-03 4.08E-03~ 4.32E-03 4.58ET O ~3 d1W 2.32E-05 2.32E-05 2.32E-05 2.32E-05 2.32E 05 2.33E 05 2.33E-05 2.33E-05 2.33E 05 IB5W ' 5.01E-07; 5.05E-07 5.12E-07 5.40E-07 6.34E-07 8.20E-07 9.39E ' 1.01 E-06 1.08E-06 187Re - 1.45E-08 1.55E-08 1.73E-08 2.42E-08 4.80E-08 9.52E-08 1.25E-07 1.42E-07 1.60E-6) 1850s ! 3 42E-04: 3.42E-04 3.43E-04. 3.43E-04 : 3.45E-04 3 48E-04 3.50E-04 3.51 E-04 3.52E-04 1921f 3.21E-04 3.21 E-04 ; 3.21 E-04, 3.22E 3.25E 04 3.32E-04 i 3.37E-04 ' 3.39E-04 3.42E-04 210Pb 7.03E 7.39E 03 7.99E-03 1.ME-02 1.84E-02 3.44E-02 4.46E-02 5.03E-02 5 65E-02 210Po 1.23E 03 1.31E 03 1.45E 03 2.00E-03 3.88E-03 7. ,lE-03 9.99E-03' 1.13E ,2 1.28E-02 226Ra~ 4.56E-03 4.74E-03 5.02E-03 6.15E 1.00E-02 1.77E 02 2.26E 02 2.53E-02 2.83sW2 226Ra+C 1.30E-02 1.37E-02 1.47E-02 1.88E-02 3.28E-02 6.06E-02 7.84E-02 8.82E-02 9.91E-02 228Ra 1.09E-02 1.21E-02 1.40E-02 2.14E-02 4.68E 02 9.71E-02 1.29E-01 1.47E-01 1.67E 01 227Ac 9.658-01 1.10E+00 1.31E+00 2.15E+00 5.05E+00 1.08E+01 1.44 E+01 1.65E+01 1.87E+01 227Ac+C 9 66E%1 1.10ET00 1.31ET60 2.15E+00 5.05s+00 1.088+01 1.44 E+01 1.65E+01 1.87E+01 228Th 4 39E-02 4.96E-02 5.90E 02 9.59E-02 2.22E-01 4.73E-01 6.33E-01 7.22E-01 8.20E-01 228Th+C 4.39E-02 4.97E-02 5.90E-02 9.59E-02 2.23E 01 4.73E-01 6.33E-01 7.22E-01 8.20E-01 229Th 3.15E-01 3 59E-01 4.28E-01 7.04E 01 1.65E+00 3.52E+00 4.72E+00 5.39E+00 6.12E+00
/
D 1 229Th+C 3.16E-01 3.59E-01 4.29E-01 7.05E-01 1.65E+00 3.53E+00 4.72E+00 5.39E+00 6.12E+00 i /236Th 4.74E-02 5.39E-02 6.44E-02 1.06E 01 2.48E-01 5.31E 01 7.11E 01 8.11 E-01 9.21 E-01 ~ 235Th+C 6.06E-02 6.77E-02 7.93E-02 1.25E-01 2.82E-01 5.92E-01 7.9 5'01 9.00E-01 1.02E+05 232Th
~
2.39E 2.72E-01 3.25E 01 5.34E-01 1.25E+00 2.67E+00 3.58E+00 ' 4.09E+00 4.64E+00 232Th+C 2.94E-01 3.355-01 3.98E-01 SliE-65~ 1.52E+00 3.25E+00 4.35E+00 4.96E+00 - 5.63E+00 f3 ipa 2.08E-01 2.36E-01 2.81E31 4.5PE-01 1.07E+00 2T21h+00 3.04E+00 3.47E+00 3.94E+00 231Pa+C 1.17E+00 1.33E+00 1.59E+00 2.61E+00 6.12E+00 1.31 E+01 1.75E+01 1.99E+01 2.27E+01 232U 1.04E-01 ' 1.18E 01 1.41E-01 2.32E-01 5 43E 1.16E+00 1.55E+00 ' 1.77E+00 2.01E+00 232U+C 1.49E-01 ' 1.69E-01 2.02E-01 3.30E-01 7.71 E-01 ' 1.64 E+00 2.20E+00 - 2.51E+00 2.85E+00 233U 1.98E 02- 2.25E-02 2.68E-02 4.41 E-02 1.03E-01 2.21E-01 2.96E-01 ' 3.37E-01 3.83E-01 233U+C 3.51 E-01 ' 3.99E-01 4.76E-01 7.83E-01' 1.84E+00 3.92E+00 5.25E+00 ' 5.99E+00 6.80E+00 234U 1.93E 02 2.20E 2.62E-02 4.31E 02 1.01 E-01 2.16E-01 2.89E-01: 3.30E-01 3.75E-01 235U 1.82E 2.06E-02 2.46E-02 4.02E-02 9.40E-02 2.00E-01 2.68E 3.06E-01 3.48E-01 235U+C 1.19E+00 1.35E+00 1.62E+00 2.65E+00 6.21E+00 1.33E+01 1.77E+01 2.02E+01 2.30E+01 236U 1.83E-02 2.08E-02 2.48E-02 4.08E-02 9.57E-02 2.04E-01 2.74E-01. 3.12E-01 3.55E-01 238U 1.73E-02 1.97E-02 2.35E-02 3.86E-02 9.05ES2 1.93E-01 + 2.59E-01 2.95E-01 : 3.35E-01 238U+C 9.69E 02! 1.09E-01 1.29E-01 2.06E-01 4.73E-01 1.00E+00 - 1.3e E+00, 1.52E+00 1.73E+00 237Np 8.14 E-02 ' 9.22E-02 1.10E 01 1.7BE-01 4.15E-01 8.83E 1.18E+00 1.35E+00 1.53E+00 237Np+C ' 4 45E 5.05E-01 6.03E-01 9.90E 2.32E+00 4.94E+00 6.62E+00 7.55E+00' 8.585+00_ 236Pu 1.97E-02 2.23E-02 2.66E-02 4.33E-02 1.01 E 2.15E-01 2.88E-01 ' 3.28E-01 3.72E-01 238Pu 5.86E 6.64E-02 7.90E-02 1.29E 3.00E-01 : 6.38E-01 8.54E 9.74E-01 1.11 E+00 239Pu 6.44E-02 7.30E-02 8.68E-02 1.42E-01; 3.29E-01 7.01E-01 9.39E-01 ' 1.07E+00 1.22E+00 240Pu i 6.44 E-02 ' 7.30E-02 8.68E 1.42E-01 3.29E-01 ' 7.01 E-01 ' 9.39E-01 1.07E+00: 1.22E+00 Page 2 m
w s om, .> . w r .. . Building Occupancy Radionuclid:-Specific Dos)(Assuming 10% R:moviool}}