ML25205A073

Assessment of Potential Variability in Dose Exceedance Distance Calculations

ML25205A073
Person / Time
Issue date:	07/30/2025
From:	Keith Compton, Salman Haq, Shockley S NRC/RES/DSA/AAB
To:
Shared Package
ML25205A069	List: ML25205A073
References
Download: ML25205A073 (1)

Text

ASSESSMENT OF POTENTIAL VARIABILITY IN DOSE EXCEEDANCE DISTANCE CALCULATIONS Revision 0 July 2025 Keith L. Compton, Steven Shockley, and Salman Haq Accident Analysis Branch Division of Systems Analysis Office of Nuclear Regulatory Research United States Nuclear Regulatory Commission

i TABLE OF CONTENTS Table of Contents...........................................................................................................................i List of Tables.................................................................................................................................ii List of Figures................................................................................................................................ii 1

INTRODUCTION.....................................................................................................................1 2

BACKGROUND.......................................................................................................................2 2.1 Uncertainty vs Variability.................................................................................................2 2.2 Review of Relevant Topical Areas..................................................................................3 3

SOURCES OF VARIABILITY IN DOSE EXCEEDANCE DISTANCE CALCULATIONS.........5 3.1 Source Term...................................................................................................................5 3.1.1 Initial Core Inventory................................................................................................5 3.1.2 Accident Progression...............................................................................................7 3.2 Meteorological Data........................................................................................................8 3.2.1 Intra-annual meteorological variability.....................................................................8 3.2.2 Inter-annual meteorological variability.....................................................................9 3.2.3 Meteorological data source....................................................................................10 3.3 Dispersion Modeling.....................................................................................................10 3.3.1 Dispersion modeling methods...............................................................................10 3.3.2 Dispersion parameters for Gaussian dispersion models.......................................10 3.3.3 Near-field effects (wake effects and plume buoyancy)..........................................11 3.4 Exposure Assessment..................................................................................................13 3.4.1 Breathing rate........................................................................................................13 3.4.2 Pathway-specific shielding factors.........................................................................14 3.5 Dosimetry......................................................................................................................15 3.5.1 Physical and chemical form...................................................................................15 3.5.2 Dosimetric figure of merit.......................................................................................16 3.6 Quantification................................................................................................................18 4

SUMMARY

AND CONCLUSIONS........................................................................................20 5

REFERENCES......................................................................................................................22

ii LIST OF TABLES Table 1 Figures-of-merit describing the weather data from the Sequoyah meteorological tower for years 2008 through 2012.........................................................................................................9 Table 2: Example of variability in conditional mean (over weather variability) individual LNT LCF risk in the 0-10-mile region using weather data for years 2008 through 2012 for a selected realization at a reference PWR.....................................................................................................9 Table 3: Reference ventilation rate values (m3/s) for a general Caucasian population at different levels of activity...........................................................................................................................13 Table 4: Variability in breathing rates..........................................................................................13 Table 5: Variability in normal activity shielding factors................................................................14 Table 6: Selected inhalation committed effective dose coefficients (Sv/Bq) for different absorption classes.......................................................................................................................16 Table 7: Selected inhalation dose coefficients (Sv/Bq) for different dosimetric measures..........17 Table 8: Ratio of selected MACCS peak dose outputs to MACCS L-ICRP60ED peak dose over a ten-mile radius assuming a sample source term......................................................................18 Table 9a: Mean (over all distances) geometric standard deviation for projected time integrated concentration of I-131 from a recent model intercomparison exercise [42].................................19 Table 9b: Mean (over all distances) geometric standard deviation for projected adult TEDE dose from a recent model intercomparison exercise [42]....................................................................20 LIST OF FIGURES Figure 1: Example of variation of decay heat with time in cycle and time after shutdown............6 Figure 2: Example of variation of initial core activity for a short-lived fission product (I-131) and a longer-lived fission product (Cs-137) with time in cycle................................................................6 Figure 3: Examples of variation in the time-dependent fraction of iodine core inventory released to the environment at a reference PWR with an ice-condenser containment for different times in the operating cycle........................................................................................................................7 Figure 4: Examples of variation in the time-dependent fraction of cesium core inventory released to the environment at a reference PWR with an ice-condenser containment for different times in the operating cycle........................................................................................................................8 Figure 5: Examples of variation in the time-dependent fraction of iodine and cesium core inventory released to the environment at a reference BWR..........................................................8 Figure 6: Ground-level, time-integrated /Q versus distance for Case 02 for AERMOD, ARCON96, QUIC compared with modified MACCS calculations................................................12

1 1

INTRODUCTION Regulatory Guide (RG) 1.242 provides methods acceptable to the NRC for meeting emergency planning (EP) requirements under 10 CFR 50.160 and 50.33(g). RG 1.242 was written to be risk-informed and technology inclusive and provides a framework for scoping the planning effort as opposed to defined methodologies. With the interest in microreactors, certain small modular reactors (SMRs), and advanced reactors, there are opportunities to provide refined guidance to ensure regulatory certainty and predictability in emergency planning zone (EPZ) determinations.

EPZs are a planning tool for EP and do not represent a boundary beyond which dose cannot exceed 10 mSv over 96 hours0.00111 days <br />0.0267 hours <br />1.587302e-4 weeks <br />3.6528e-5 months <br />. Many factors determine a site-specific dose-distance relationship for a radiological release, but the variability in results can appear amplified as EPZ distances get smaller.

Recent experience with applying the guidance for EPZ determinations in Appendix A of RG 1.242 has demonstrated the need for additional guidance on the acceptable variability in dose-distance calculations, when risk-informing the size of the EPZ. Some designers are addressing uncertainty by using different dose criteria depending on median or 95th percentile weather, but few are addressing the variability in distance beyond the EPZ boundary in the analysis and are establishing conservative criteria not to exceed dose beyond the EPZ for screened-in events.

As such, research is needed to support, in part, a basis for guidance for applicants and staff on acceptable variability in dose-distance curves.

The Office of Nuclear Security and Incident Response (NSIR) requested that the Office of Nuclear Regulatory Research (RES) identify and review published consequence analyses, including both light-water reactor (LWR) and non-LWR consequence analyses, to evaluate reasonable variability in the range at which high (e.g., >100 rad), moderate (e.g., 25 rem), and low (e.g., <10 rem) doses can extend as a function of reactor size (core thermal power) considering variability in accident source terms, meteorology, and dose assessment methods.

To the extent practicable, RES should also develop and carry out confirmatory calculations using MACCS to evaluate the range (and associated variability) at which high (>100 rad),

moderate (25 rem), and low (<10 rem) doses can extend as a function of reactor size (i.e., core thermal power) assuming relative release fractions comparable to selected representative LWR source terms. The results of these analyses are documented in this report.

2 2

BACKGROUND 2.1 Uncertainty vs Variability The notion of variability in a consequence assessment is closely related to the notion of uncertainty. The term uncertainty is defined in NUREG-2122 [1] as follows:

In defining uncertainty, there are two types: aleatory and epistemic. Aleatory uncertainty is based on the randomness of the nature of the events or phenomena and cannot be reduced by increasing the analysts knowledge of the systems being modeled.

Therefore, it is also known as random uncertainty or stochastic uncertainty. Epistemic uncertainty is the uncertainty related to the lack of knowledge or confidence about the system or model and is also known as state-of-knowledge uncertainty.

As discussed in a recent consequence analysis [2],

It should be noted that there is a difference between uncertainty (i.e., where there is a lack of knowledge about the value of a parameter or which model is most appropriate) and sensitivity (which is a measure of how much an output measure will change given a change in an input). A parameter or model may be highly uncertain, but of low importance if the results are insensitive to that parameter or model. Conversely, a parameter may be relatively well known, but the model may be highly sensitive to that parameter (e.g., model parameters that could lead to cliff-edge effects), which would magnify the impact of the uncertainty of that input. Important sources of uncertainty are those that have a combination of uncertainty and sensitivity that is sufficient to lead to a potentially significant change in model results. It should also be noted that the sensitivity of the model is a function of the particular output measure; that is, some output measures may be relatively insensitive to certain model parameters, but highly sensitive to others. This being the case, the importance of a source of uncertainty will depend upon the particular output measure being analyzed.

In this report, the focus is on examining sources of variability in a calculation that may affect the distance at which a specified dose may be exceeded. These sources of variability may arise from either epistemic or aleatory uncertainties, particularly when there are alternative state-of-practice1 methods for quantifying or otherwise addressing that uncertainty. The approach taken is to

1) Identify sources of uncertainty by reviewing the technical elements of a consequence assessment; 1 As discussed in [2], State-of-practice methods, tools, and data refer to those that are routinely used by the NRC and licensees or have acceptance in the PRA technical community.

3

2) Identify reasonable alternative state-of-practice methods for quantifying or otherwise addressing those uncertainties;

3) Assess the importance of those uncertainties using published quantitative assessments where possible and using professional judgement2 where published quantitative assessments are unavailable.

2.2 Review of Relevant Topical Areas The Level 3 Probabilistic Risk Assessment (PRA) standard effort introduced, in a proposed standard for trial use and pilot application (TUPA) [3], a set of nine technical elements to organize technical review of offsite consequence assessments used in support of risk assessments. These technical elements, reproduced below, provide a convenient method for subdividing the subject matter of a consequence assessment.

Radionuclide Release Characterization for Level 3 Protective Action Parameters and Other Site Data Meteorological Data Atmospheric Transport and Dispersion Dosimetry Health Effects Economic Factors Conditional Consequence Quantification and Reporting Risk Estimation The radiological release characterization comprises the characteristics of the radionuclide release, including but not limited to the development of release categories, quantity of each radionuclide released to the environment, particle size distribution, the height and amount of energy associated with the release, the duration of the release, the time of the release after accident initiation, the warning time for evacuation, and the frequency of occurrence predicted for the release category. [3].

The protective action parameters and other site data technical element of a consequence assessment ensures that the protective actions are properly defined to enable calculation of the impact of mitigation strategies in the consequence analysis; and (ensures) that other site, local, and regional data are properly defined and developed to support the consequence analysis. [3].

In this analysis, the focus is on dose to a hypothetical individual in the absence of short-term protective actions.

2 The term professional judgment as used in this report is closely related to the concept of expert judgement. As discussed in [1], The ASME/ANS PRA Standard (Ref. 2) defines expert judgment as information provided by a technical expert, in the experts area of expertise, based on opinion, or on an interpretation based on reasoning that includes evaluations of theories, models, or experiments.

4 The objective of the meteorological data technical element is to locate sources of valid and representative meteorological data, which are input to atmospheric transport and dispersion (ATD) codes, that provide the basis for consequence analysis calculations. [3]. The meteorological data should be temporally representative and include characterization of wind direction and wind speed (including changes in wind direction and wind speed during the radiological release), atmospheric stability, and the amount and intensity of rainfall.

The ATD technical element account for the fate and transport of the radioactive plume as it is released and travels for many hours during which the meteorological conditions are very likely to change in both time and space. [3]. Phenomena typically considered include mixing in the turbulent wake of the buildings, plume rise due to buoyancy, the meandering shifts in wind direction under low-wind speed conditions, advection and turbulent dispersion of the plume as it travels downwind, vertical mixing of the plume throughout the planetary boundary layer, and depletion of the plume by radioactive decay or deposition to the vegetation, the ground surface, or other surface level obstacles. The objective of the atmospheric transport and dispersion technical element is to ensure that an appropriate dispersion methodology and meteorological data are used to determine the airborne concentration and ground deposition for input into dose models. [3].

The dosimetry technical element involves the estimation of radiation doses from short-and long-term exposures to offsite populations arising from airborne and deposited radioisotopes. It considers the dosimetric quantities (e.g., total effective dose equivalent, equivalent organ doses) to be assessed, identification of the short-and long-term exposure pathways, and the selection of recognized sources of pathway-specific dose coefficients. The objective for the dosimetry technical element is to ensure that appropriate dose conversion factors3 are used along with the computed isotopic concentrations and depositions to determine the doses received by the tissues and organs of interest due to exposure to radioactive material via each of the relevant dose pathways. [3].

The TUPA PRA standard [3] allows for quantification of consequences that may include not only radiological dose but also the health effects arising from that radiological dose and the economic impacts of protective actions taken to reduce or eliminate radiological dose. However, because this evaluation is limited to quantification of dose, consideration of these additional technical elements those involving health effects modeling and the consideration of economic factors are not considered in this report.

3 i.e., pathway-specific dose coefficients

5 3

SOURCES OF VARIABILITY IN DOSE EXCEEDANCE DISTANCE CALCULATIONS The discussion in Section 3 loosely follows the organization of the technical elements summarized in Section 2 to provide an evaluation of some of the factors that may vary due to either aleatory or epistemic uncertainty and therefore result in variability in computed dose estimates. In keeping with the organization of the technical elements, which also include consideration of consequence quantification and reporting and risk integration, it concludes with some examples of the effect that variability in one or more of the phenomena identified in Section 3 may have on the computed doses. It should be noted that the selection of phenomena discussed in Section 3, and the illustration of the effect of these phenomena on the radiological dose estimation, is not a comprehensive identification and evaluation of all potential sources of variability. However, this section does provide evidence for how dose estimates may reasonably vary and therefore provide some insights into the level of precision that can be expected in dose estimation.

3.1 Source Term 3.1.1 Initial Core Inventory The initial core inventory selected for the consequence assessment is a source of variability.

This can be affected by, for example, the time in the operating cycle assumed for the analysis.

The core inventory affects both the decay heat generation and the amount of radioactivity available for release. This is illustrated in Figures 1 and 2 below, showing how the decay heat (which has a significant effect on accident progression) and the inventory of a short-lived radionuclide (I-131) and a longer-lived fission produce (Cs-137) varied with time in cycle.

6 Figure 1: Example of variation of decay heat with time in cycle and time after shutdown Source: Figure 3-48 of [4]

Figure 2: Example of variation of initial core activity for a short-lived fission product (I-131) and a longer-lived fission product (Cs-137) with time in cycle Source: Figure 3-45 of [4]

7 3.1.2 Accident Progression The modeling of accident progression is a source of variability, even for a well-defined accident scenario. An example of this is provided in Figures 3 and 4 below, illustrating the level of source term variability exhibited in the uncertainty analyses for a short-term station blackout scenario at a reference pressurized water reactor (PWR) with an ice-condenser containment. A sample set of 600 realizations was analyzed for the short-term station blackout scenario. The figures below show release fraction variation, between the 5th 4 and 95th percentile, of over an order of magnitude.

Figure 3: Examples of variation in the time-dependent fraction of iodine core inventory released to the environment at a reference PWR with an ice-condenser containment for different times in the operating cycle Source: Figure 4-91 of [4]

Figure 4: Examples of variation in the time-dependent fraction of cesium core inventory released to the environment at a reference PWR with an ice-condenser containment for different times in the operating cycle Source: Figure 4-83 of [4]

4 Note that the 5th percentile curves are below 1.0E-5 and are not visible on the figure.

8 This level of source term variability is consistent with the uncertainty analyses for a single accident scenario representing an unmitigated long-term station blackout for at a reference boiling water reactor (BWR), illustrated in Figure 5. Each light grey line represents a single realization out of 865 sample realizations. The blue and red lines represent the 5th and 95th percentile values, illustrating over an order of magnitude possible variation in the release magnitude of a single accident scenario.

a. iodine

b. cesium Figure 5: Examples of variation in the time-dependent fraction of iodine and cesium core inventory released to the environment at a reference BWR Source: Figures 6.1-1 and 6.1-2 of [5]

3.2 Meteorological Data 3.2.1 Intra-annual meteorological variability Intra-annual meteorological variability refers to the variability in weather conditions that occurs over the course of a year. Over the course of the year, weather conditions (such as wind speed, wind direction, atmospheric stability, and precipitation) vary over time scales from seconds to months. Dispersion calculations can account for this variation by either averaging over a suitable time interval or by selecting a single weather condition from the range of observed weather conditions. In the MACCS code, the intra-annual variability in weather conditions is treated by sampling weather conditions from a set of hourly observations.

Some insights into the effect of this variability can be gained by examination of Appendix C of

[6]. That appendix tabulated the mean (across all 200 parameter samples), median, and 95th percentile exceedance range for three different sites and two different source terms. It was noted that the the 95th percentile distance, representing the range at which a particular dose level would be exceeded in no more than 5 percent of all weather trials, is noticeably (on the order of a factor of 2-3) larger than the median exceedance distance, i.e., the range at which a particular dose level would be exceeded in 50 percent of all weather trials. This is consistent with an examination of intermediate outputs from a recent Level 3 PRA of a reference PWR [7].

Examination of the variability in the distance to which an early-phase effective dose of 50 rem to

9 the non-evacuating cohort could be exceeded suggested that the 90-percentile range (5th-95th) could vary from 30% of the mean range at the 5th percentile to almost twice the mean range at the 95th percentile. The interquartile range varied from on the order of 60% of the mean range at the 25th percentile to 150% of the mean range at the 75th percentile.

3.2.2 Inter-annual meteorological variability Inter-annual meteorological variability refers to the variation in annual weather conditions that occurs from year to year. At a given site, subject to the same climactic conditions, year-over-year variability in annual weather conditions is subject to less variability than the variability naturally occurring over the course of a day or a year. The effect of this inter-annual variability was examined in Section 6.5.3 of [4]. An example of the variability across five annual meteorological sets, shown in Table 1.

Table 1 Figures-of-merit describing the weather data from the Sequoyah meteorological tower for years 2008 through 2012 Precipitation Total (inches)

Hours of Observed Precipitation Average Wind Speed Frequency Weighted 10-mile population exposed to plume Frequency Weighted 50-mile population exposed to plume 39.7 508.0 3.2 1425 22448 47.9 635.0 2.6 1440 23142 25.5 426.0 3.7 1544 24334 52.9 507.0 3.4 1356 22185 39.3 526.0 2.4 1414 22805 Source: Table 6-25 of [4]

The effect of this variability on consequence in that analysis, the lifetime risk to an individual of a fatal cancer within a specific geographic region is illustrated in Table 2.

Table 2: Example of variability in conditional mean (over weather variability) individual LNT LCF risk in the 0-10-mile region using weather data for years 2008 through 2012 for a selected realization at a reference PWR Accident Phase 2008 2009 2010 2011 2012 Mean Emergency 6.0E-04 5.7E-04 5.8E-04 5.4E-04 5.3E-04 (5.6+/-0.30) E-04 Long-Term 5.8E-04 6.2E-04 6.1E-04 5.7E-04 5.9E-04 (6.0+/-0.20) E-04 Total Risk 1.2E-03 1.2E-03 1.2E-03 1.1E-03 1.1E-03 (1.2+/-0.04) E-03 Source: Extracted from Table 6-24 of [4]

As expected, the annual dataset has only a limited effect on the mean value over interannual variability, with less than 10% variation across the mean. It may be noted that greater variability may exist when the statistic under consideration reflects upper bound values, such as the 95th percentile.

10 3.2.3 Meteorological data source Meteorological data may be obtained from a variety of sources. Traditionally, nuclear power plants licensed by the NRC maintain onsite meteorological monitoring programs that can provide high quality observational data [8], and meteorological guidance for nuclear facilities generally is provided in [9]. However, both the draft Level 3 PRA standard (Supporting Requirement ME-A1 of Reference [3], Capability Category I) and Section 6.7 of U.S.

Environmental Protection Agency meteorological monitoring guidance [10] recognize that meteorological data from other regional sources such as surface weather observations collected in support of various NWS and Federal Aviation Administration (FAA) programs may be used in modeling. In addition, extensive datasets of gridded weather data for much of the continental United States is available from the National Oceanic and Atmospheric Administration (NOAA) National Centers for Environmental Information (NCEI). A recent analysis [11] compared the use of observational data using ASOS meteorological data sources and a synthetic data source interpolated from the NAM12 dataset to compare a variety of weather-averaged consequence measures. That comparison found that Using ASOS compared to NAM12 data resulted in a less than a factor of 2 change in all ATD and consequence outputs evaluated.

3.3 Dispersion Modeling 3.3.1 Dispersion modeling methods Dispersion modeling has traditionally relied on simple Gaussian plume models parameterized using discrete stability classes [12]. These types of models have been considered to yield results that are accurate to within a factor of two to ten under realistic modeling conditions, with the accuracy decreasing under circumstances such as dispersion in building wakes, buoyant flows, flows over irregular terrain surfaces, dispersion in very low or very high stability conditions, or dispersion at long distances (greater than about 10 km) [13]. More advanced dispersion modeling methods, such as Lagrangian particle tracking methods (e.g., as used in the HYSPLIT code [14]) or methods that rely on computational fluid dynamics [15], are often used to address such modeling circumstances. However, these more advanced methods are often much more demanding in terms of input data and computational time. Model comparisons

( [16], [17]) have shown that the weather-averaged results from Gaussian plume segment models such as MACCS can compare well (within a factor of 2-3) with the results from higher fidelity models when configured appropriately.

3.3.2 Dispersion parameters for Gaussian dispersion models A variety of sources for the development of dispersion curves are summarized in Section 3.6.1 of [18]. Many dispersion parameterizations used for short distances (over a distance of a few kilometers) are based on the Pasquill-Gifford (PG) dispersion curves [19] [20], which were developed from field data including the 1956 Prairie Grass diffusion experiments [21]. An example of the impact of several different dispersion parameterizations based in whole or in part

11 on the PG curves is provided in Appendix A of [22], which compares what they term the Briggs Open Country, EPA, and NRC parameterizations. These are described as follows:

The Briggs open country parameterization is appropriate for more rural environments, like SRS, and is based on the combined results of several diffusion experiments (Pasquill [1961], Brookhaven National Laboratory [Smith 1961], and the Tennessee Valley Authority [Carpenter et al. 1971]). The Environmental Protection Agencys (EPAs) Industrial Source Complex (ISC3) model (EPA 1995) rural mode option implements a parameterization that approximately fits the original PG dispersion curves (Turner, 1970). The dispersion parameterization typically used by the Nuclear Regulatory Commission (NRC) is attributed to Eimutis and Konicek (1972).

Appendix A of [22] concludes that all three parameterizations give essentially the same result, with the exception of the highly unstable PG stability class A. Use of the parameterizations in an actual dispersion model (e.g. MACCS2), however, would result in more comparable results than this simple example indicates, because the model would limit vertical plume dispersion to the depth of the boundary layer, which would tend to increase the concentrations shown for class A at longer distances. Furthermore, the more restrictive dispersion conditions (i.e., PG stability classes D, E, and F) are often of most concern in safety assessments, and the various parameterizations agree to within a few percent out to distances of 10 to 12 km, beyond which the Briggs open country parameterization results in increasingly higher estimates of relative concentration.

A broader, but still plausible, set of dispersion parameterizations was considered in the expert elicitations documented in [23]. This broader set is discussed in Section 5.9.6 of [4], which noted that historic values from Vogt (1977), Dobbins (1979), and Panitz (1989) [54] were used by multiple experts in constructing their distributions, showing they continue to be regarded as reasonable bases to estimate dispersion. The discussion in Section 5.9.6 of [4] went on to state that it was beneficial to only sample one of the two constants to simplify the specification of uncertainty. Bixler et al. [78] characterized the uncertainty as an uncertainty in CYSIGA and this characterization is used here.. The authors of [4] then concluded that The expert data indicate about one order-of-magnitude uncertainty within the 90-percent confidence interval and about two orders of magnitude at the 100-percent confidence interval.

3.3.3 Near-field effects (wake effects and plume buoyancy)

As noted in [13], buoyant flows that result in plume rise or flows in the wakes of structures can be complex relative to the steady-state one dimensional flow conditions used to derive the Gaussian plume equations. A summary of approaches for modeling these near-field effects is provided in [15], which states Near-field pollutant dispersion, involving the interaction of a

12 plume and the flow field perturbed by building obstacles, is an element of outdoor air pollution that is particularly complex to predict. The modeling approaches are categorized into field measurements, laboratory (wind and water tunnel) experiments, (semi-) empirical models, and computational fluid dynamics (CFD) models. Each of these approaches has advantages and disadvantages. Examples of the differences in results that can arise from different methods to account for these effects are examined in [24] and [25]. The range in results is illustrated in Figure 6 (reproduced from Figure 5-6 of [24], showing that at distances less than a kilometer the results from several different models can vary by up to an order of magnitude. Figure 6 also illustrates that treating the plume as originating from a non-buoyant, ground level, point source can be quite conservative in relation to models that include the more complex wake flows.

Figure 6: Ground-level, time-integrated /Q versus distance for Case 02 for AERMOD, ARCON96, QUIC compared with modified MACCS calculations Source: Reproduced from Figure 5-6 of [24]

3.4 Exposure Assessment Sources of variability in the exposure assessment include variability in parameters such as the breathing rate and the pathway-specific shielding factors.

13 3.4.1 Breathing rate5 Inhalation dose, which can be a significant contributor to the total dose received during the early phase of an accident, is proportional to breathing rate. Breathing rate can vary by age, gender, and activity, as illustrated in Table 3.

Table 3: Reference ventilation rate values (m3/s) for a general Caucasian population at different levels of activity Age/Gender Resting (sleeping)

Sitting Awake Light Exercise Heavy Exercise 3 mo 2.5E-05 N/A 5.3E-05 N/A 1 Y 4.2E-05 6.1E-05 9.7E-05 N/A 5 Y 6.7E-05 8.9E-05 1.6E-04 N/A 10 Y 8.6E-05 1.1E-04 3.1E-04 6.2E-04 (M) 5.1E-04 (F) 15Y Male 1.2E-04 1.3E-04 3.8E-04 8.1E-04 15Y Female 9.7E-05 1.1E-04 3.6E-04 7.1E-04 Adult Male 1.3E-04 1.5E-04 4.2E-04 8.3E-04 Adult Female 8.9E-05 1.1E-04 3.5E-04 7.5E-04 Source: adapted from Table 8 of [26]

Despite this variability, for many analyses an age, gender, and activity-averaged breathing rate is used for consistency with the inhalation dose factors (which also commonly assume an age and gender-averaged population). A value of 2.66E-04 m3/s was recommended by Sprung et al. [27]

and is consistent with a population-and activity-averaged breathing rate. Breathing rate was also examined in the NRC-CEC expert elicitation on internal dosimetry [28] and was postprocessed by Gregory et al. [29] to obtain the distribution shown in Table 4.

Table 4: Variability in breathing rates Quantile Breathing Rate (m3/s) 5%

1.38E-04 50%

2.19E-04 95%

3.93E-04 Source: adapted from Table 4.11 of [29]

Regulatory Guides 1.195 [30] and 1.183 [31] suggest a recommended value of 3.5 x 10-4 m3/s for the breathing rate during the first eight hours of an accident, a recommended value of 1.8 x 10-4 m3/s from 8 to 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> following the accident, and a recommended value of 2.3 x 10-4 m3/s thereafter. In contrast, the FRMAC Assessment Manual [32] recommends a value of 4.2 x 10-4 m3/s based on the adult male light exercise breathing rate for assessing doses from the passing cloud, with a note stating that This calculation uses the light exercise breathing rate rather than the activity averaged breathing rate (BRAA) because it is assumed 5 This discussion is adapted from Section 4.2.1.3 of NUREG/CR-7270 [18].

14 that the individual will be actively seeking to exit the plume. For longer term resuspension exposures, the Federal Radiological Monitoring and Assessment Center (FRMAC) recommended value for estimating resuspension inhalation exposures is based on the activity-averaged breathing rate of 2.56 x 10-4 m3/s for an adult male based on Table B.16B of

[26]. Additional compilations of data on activity, age, and gender distributions of breathing rates include [26] as well as Chapter 6 of [33].

It can be seen from inspection of the 5th - 95th percentile range in Table 4 that reasonable breathing rates could plausibly vary by a factor on the order of about two, with the factors recommended by Regulatory Guides 1.195 [30] 1.183 [31] and the FRMAC Assessment Manual [32] lying in the upper range of the distribution of plausible breathing rates.

3.4.2 Pathway-specific shielding factors When using MACCS, pathway-specific shielding factors are dimensionless quantities used to reduce the computed radiation dose to account for the dose reduction associated with a given activity (normal activity, active sheltering, and evacuation) for a given exposure pathway (cloudshine, inhalation, groundshine, and skin deposition). It may be noted that even in the absence of protective actions such as active sheltering or evacuation, a degree of protection is afforded by normal activities. For example, individuals are often located indoors at home or at work over the course of a normal day. Even if an individual is located outdoors, obstacles such as nearby vegetation or structure, variations in terrain height, and irregular ground surfaces on which the material is deposited can result in some degree of shielding relative to the flat, smooth, infinite plain used to develop the external dose coefficients. The variability in pathway-specific shielding factors for normal activity is illustrated by the range of shielding factor values used by Gregory et al. [29] in an uncertainty analysis of early exposure results. For the groundshine and inhalation pathways, these values were based on post-processing the results of an NRC-CEC expert elicitation on internal dosimetry [28]. For the cloudshine pathway, Gregory et al. [29] relied on an uncertainty analysis documented in Helton et al. [34]. Some of the quantiles used in that uncertainty analysis are shown In Table 5.

Table 5: Variability in normal activity shielding factors Quantile Cloudshine Shielding Factor1 Groundshine Shielding Factor2 Inhalation Shielding Factor2 5%

0.62 0.10 0.15 50%

0.78 0.22 0.69 95%

0.93 0.42 1.00 Sources: 1 - estimated from Table 3.1 of [29] based on information in [34]; 2 - adapted from Table 4-11 of [29]

It can be seen from inspection of the 5th - 95th percentile range in Table 5 that reasonable shielding factors for normal activity, particularly for the inhalation and groundshine exposure pathways, could plausibly vary by well over a factor of two. Likewise, the decision to credit structural shielding associated with building occupancy as opposed to making the

15 conservative assumption that the individual is located outside during plume passage could similarly result in dose reduction by a factor of around two to almost an order of magnitude.

3.5 Dosimetry Sources of variability in dosimetry include consideration of physical and chemical form used to select the inhalation and ingestion dose coefficients and the selection of the dosimetric figure of merit.

3.5.1 Physical and chemical form From a dosimetric perspective, the chemical and physical form affects the biokinetic behavior of a radionuclide following intake. The biokinetic properties of an inhaled radionuclide are characterized by the absorption class and gastrointestinal update fraction (f1) assigned to the nuclide. Reference [35] provides risk estimates for a wide range of chemical and physical forms.

As stated in Chapter 4 of [35], Although the model permits consideration of compound-specific dissolution rates, a particulate is generally assigned to one of three default absorption types:

Type F (fast dissolution and a high level of absorption to blood), Type M (an intermediate rate of dissolution and an intermediate level of absorption to blood), and Type S (slow dissolution and a low level of absorption to blood). As noted in the introduction to Table 2.4 of [35], For consideration of inhaled gases or vapors, it was assumed that deposition is complete and that absorption to blood is rapid and complete. Although the tables in [35] provide risk coefficients rather than dose coefficients, information on the potential variability in dose due to differences in chemical form of a given radionuclide is available from examination of the CD supplement to Federal Guidance Report (FGR)-13 [36], which provides dose coefficients6 for a variety of absorption classes. Examination of the variation in the dose coefficients for a selected set of nuclides typically considered in LWR assessments suggests that the chemical and physical form of a radionuclide could, for the subset examined, change the dose from inhalation of that nuclide by over an order of magnitude. This is illustrated by examination of the variation of the adult inhalation committed effective dose coefficient for selected isotopes of ruthenium, tellurium, iodine, and plutonium shown in Table 6 below. Note that this table shows only variability due to differences in absorption class; variability due to difference in aerosol particle size, which affects the deposition of particles in the respiratory tract, is not shown in this table.

6 Selected dose coefficients were obtained by examination of the data in \fgr13pak\FGR13INH.HDB, which contains committed absorbed dose coefficients for 23 organs and the effective dose for intakes by inhalation for six age groups.

16 Table 6: Selected inhalation committed effective dose coefficients (Sv/Bq) for different absorption classes Isotope MACCS 4.2 Default Dose Coefficient1 FGR-13 Default Class2 Vapor3 Particulate Class F, 1 µm AMAD3,4 Particulate Class M, 1 µm AMAD3,4 Particulate Class S, 1 µm AMAD3,4 Ru-103 3.0E-09 M

1.1E-09 4.8E-10 2.4E-09 2.9E-09 Ru-105 1.8E-10 M

1.8E-10 6.6E-11 1.7E-10 1.8E-10 Ru-106 6.6E-08 M

1.9E-08 8.0E-09 2.8E-08 6.6E-08 Te-127 1.3E-10 M

7.7E-11 3.9E-11 1.3E-10 1.4E-10 Te-127m 7.4E-09 M

4.6E-09 1.5E-09 7.4E-09 9.8E-09 Te-129 3.7E-11 M

3.7E-11 1.6E-11 3.7E-11 3.9E-11 I-131 7.4E-09 F

2.0E-08 7.4E-09 2.4E-09 1.6E-09 I-132 9.4E-11 F

3.1E-10 9.4E-11 1.1E-10 1.1E-10 I-133 1.5E-09 F

4.0E-09 1.5E-09 5.5E-10 4.3E-10 I-134 4.5E-11 F

1.5E-10 4.5E-11 5.5E-11 5.6E-11 I-135 3.2E-10 F

9.2E-10 3.2E-10 2.4E-10 2.2E-10 Pu-238 1.6E-05 M

n/a 1.1E-04 4.6E-05 1.6E-05 Pu-239 1.6E-05 M

n/a 1.2E-04 5.0E-05 1.6E-05 Pu-240 1.6E-05 M

n/a 1.2E-04 5.0E-05 1.6E-05 Pu-241 1.7E-07 M

n/a 2.3E-06 9.0E-07 1.7E-07

1. Source: MACCS DCF file FGR13GyEquiv_TEDE_v2.inp, column labeled INHALED CHRONIC

2. Source: Table 4.1 of [35], which provides recommended default absorption type when no specific information is available

3. Source: FGR13INH.HDB, adult values, column labeled e_50

4. A value of 1 µm AMAD represents the particle size recommended by the ICRP for consideration of environmental exposures in the absence of specific information about the physical characteristics of the aerosol [35]

3.5.2 Dosimetric figure of merit The term dosimetric figure of merit refers to the specific dosimetric quantity that is quantified in a dose assessment. For example, MACCS 4.2 allows consideration of two different measures of effective dose. The MACCS L-TEDE pseudo-organ uses the committed effective dose equivalent, as defined in the 1977 recommendations of the ICRP [37], and is based on the older dosimetry used in FGR 11 [38]. The MACCS L-ICRP60ED pseudo-organ uses the committed effective dose, as defined in the 1990 recommendations of the ICRP [39], and is based on the dosimetry used by FGR 13 [35]. Both the effective dose equivalent and the effective dose involve a weighted sum of absorbed organ doses, where the organ doses are weighted by a tissue-specific factor based on the probability of stochastic effects from irradiation of that tissue.

Both effective dose quantities used in MACCS reflect a committed dose from internal exposures, in which the doses are integrated over a fifty-year period to reflect the cumulative dose received from intakes of radioactive material. In contrast, because the probability of deterministic health effects such early fatality from hematopoietic syndrome, pulmonary syndrome, or gastrointestinal syndrome are strongly dependent on the both the dose and dose rate i.e., high doses delivered over a matter of hours or days pose a much greater biological hazard than the same dose delivered at lower dose rates over months or years MACCS

17 acute dose coefficients for the inhalation pathway are computed with a time-weighting factor that accounts for the lessened effectiveness of dose delivered at low dose rates over long time periods (dose protraction) (Section 1.2.6 of [40]). The difference between the MACCS effective dose coefficients (L-ICRP60ED and L-TEDE) and the MACCS acute inhalation organ dose coefficients (e.g., A-RED MARR, A-LUNGS, and A-STOMACH) can be substantial, as seen in Table 7 below.

Table 7: Selected inhalation dose coefficients (Sv/Bq) for different dosimetric measures Isotope L-ICRP60ED L-TEDE A-RED MARR A-LUNGS A-STOMACH Ru-103 3.00E-09 2.40E-09 6.10E-11 2.00E-09 6.10E-11 Ru-105 1.80E-10 1.20E-10 7.70E-12 6.20E-10 9.70E-11 Ru-106 6.60E-08 1.30E-07 3.40E-11 1.20E-08 3.60E-10 Te-127 1.30E-10 8.60E-11 1.70E-12 7.50E-10 4.60E-11 Te-127m 7.40E-09 5.80E-09 9.40E-11 3.80E-09 2.90E-11 Te-129 3.70E-11 2.10E-11 3.50E-13 1.50E-10 6.40E-11 Te-129m 6.60E-09 6.50E-09 1.70E-10 4.30E-09 1.10E-10 Te-131 2.90E-11 1.20E-10 4.70E-13 7.70E-11 4.30E-11 Te-131m 1.10E-09 1.70E-09 6.50E-11 1.80E-09 1.50E-10 Te-132 2.10E-09 2.60E-09 1.40E-10 2.70E-09 1.80E-10 I-131 7.40E-09 8.90E-09 2.20E-11 3.40E-11 3.40E-11 I-132 9.40E-11 1.00E-10 1.20E-11 3.60E-11 6.50E-11 I-133 1.50E-09 1.60E-09 1.90E-11 4.20E-11 6.20E-11 I-134 4.50E-11 3.60E-11 5.50E-12 3.00E-11 5.10E-11 I-135 3.20E-10 3.30E-10 1.70E-11 4.00E-11 6.00E-11 Pu-238 1.60E-05 7.80E-05 3.00E-10 1.50E-06 1.40E-10 Pu-239 1.60E-05 8.30E-05 2.80E-10 1.30E-06 1.30E-10 Pu-240 1.60E-05 8.30E-05 2.80E-10 1.30E-06 1.30E-10 Pu-241 1.70E-07 1.30E-06 5.50E-14 2.10E-10 1.30E-12 Source: MACCS DCF file FGR13GyEquiv_TEDE_v2.inp. Measures with prefix L-are taken from column labeled INHALED CHRONIC and measures with pre A-are taken from column labeled INHALED ACUTE It is challenging to provide any universally applicable quantitative comparison of the differences between these dosimetric measures, because they are dependent on both the exposure pathway and the specific nuclide. However, an illustration of the potential impact can be provided by examination of the early phase peak dose values to the non-evacuating cohort computed using the MACCS 4.2 sample problem. The results are given in Table 8 below, showing that for this source term and considering only early phase exposures to the airborne plume, the total effective dose equivalent (TEDE) can be noticeably higher than the total effective dose (TED, quantified using the MACCS L-ICRP60ED pseudo-organ), but acute inhalation doses suitable for evaluation of early fatality from hematopoietic syndrome, pulmonary syndrome, or gastrointestinal syndrome can be considerably lower than the committed effective doses

18 Table 8: Ratio of selected MACCS peak dose outputs to MACCS L-ICRP60ED peak dose over a ten-mile radius assuming a sample source term Distance (km)

L-ICRP60ED L-TEDE A-RED MARR A-LUNGS A-STOMACH 0.08 1.0 1.7 0.10 0.21 0.09 0.34 1.0 1.6 0.11 0.20 0.10 0.865 1.0 1.6 0.13 0.23 0.11 1.41 1.0 1.6 0.14 0.25 0.13 1.87 1.0 1.7 0.15 0.25 0.13 2.675 1.0 1.7 0.16 0.26 0.14 3.62 1.0 1.7 0.16 0.26 0.14 4.425 1.0 1.7 0.16 0.26 0.14 5.23 1.0 1.7 0.16 0.26 0.14 6.84 1.0 1.6 0.15 0.25 0.14 9.66 1.0 1.6 0.15 0.24 0.13 13.68 1.0 1.6 0.14 0.23 0.13 Source: Reanalysis of MACCS 4.2 sample problem Point Estimates LNT with additional output statements 3.6 Quantification The previous sections provided examples of reasonable variability in selected dose assessment inputs that can range by a factor of two to ten or more. The effect of this such variability on dose projections has been examined in a study [41] of the the sensitivity of dose exceedance distances to variations in user input parameters. The study examined the sensitivity of the distance to which a variety of dose thresholds could be exceeded due to variability in the source term, the initial core inventory, the source of the meteorological data, the selection of a plume meander model, the surface roughness, the assumed breathing rate, and the use of TED vs TEDE as a dosimetric quantity. Inspection of the results for the mean (over all weather trials) exceedance distances suggests that dose exceedance distances could vary by a factor of two or more, with larger variation observed for shorter dose exceedance distances and smaller dose quantities. Variations in the initial core inventory and the source term generally had the largest effect, highlighting the significance of uncertainties in the source terms. The authors concluded that variations in these input parameters lead to substantial differences in dose exceedance distances for the examined dose thresholds. [41]

A practical example of the reasonable variability that can arise in dose projections is provided by a recent examination of dose prognosis capabilities [42] in the context of incident response. The study provides an understanding of why the results are, or may be, different and analyses how the identified differences can be understood and taken into account by neighboring countries or territories for their respective situational interpretation and decisions. Recommendations about what could be considered good general agreements on the outputs are also considered. The study compared dose projections produced by several national dose projection organizations for

19 three different meteorological scenarios and with different levels of shared information. The meteorological scenarios, referred to in the exercise as drills, each comprised a distinct meteorological scenario with specific weather conditions: 1) steady wind; 2) low wind velocity and turning wind directions; and 3) frontal passage with precipitation. [42]. For each drill, the exercise was conducted in three rounds with increasing levels of shared information. For Round 0, only location and date of the accident, the basic information of the reactor, and a short description of the accident sequence at the facility were shared between participants. To eliminate the influence of source term variability on variability in dose projections, all participants in Round 1 used a shared detailed time-dependent source term. Finally, to eliminate the influence of variability in the meteorological data, Round 2 used both a shared source term and a shared set of meteorological input files from the European Centre for Medium-Range Weather Forecasts.

The results of the model intercomparison exercises support the range of variability suggested by the review in the previous sections. This is seen in Table 9a, showing the mean (over all modeled distances) geometric standard deviation (across all project participants) for the projected time-integrated air concentration of I-131, and Table 9b, showing the mean geometric standard deviation of the projected adult TEDE dose. Table 9 clearly illustrates the effect of variability in the source term (by comparing the results of Round 1 to round 0), and the combined effect of source term variability and meteorological variability (by comparing the results of Round 2 to Round 0. Even for Round 2, which was designed to minimize variability due to variability in the source term and meteorological data, the geometric standard deviation ranged between a factor of 2-3.

Table 9a: Mean (over all distances) geometric standard deviation for projected time integrated concentration of I-131 from a recent model intercomparison exercise [42]

Round Drill 1* - Steady Wind Drill 2** - High Pressure System Drill 3*** -

Frontal Passage Round 0 - Initial assessment, basic information only 5.8 20.8 6.7 Round 1 - Shared source term 2.3 2.6 2.6 Round 2 -Shared source term and same meteorological data 2.0 2.6

2.5 Sources

• Table 3-2 of [42]; **: Table 3-4 of [42]; ***: Table 3-6 of [42]

20 Table 9b: Mean (over all distances) geometric standard deviation for projected adult TEDE dose from a recent model intercomparison exercise [42]

Round Drill 1* - Steady Wind Drill 2** - High Pressure System Drill 3*** -

Frontal Passage Round 0 - Initial assessment, basic information only 8.0 25.9 3.9 Round 1 - Shared source term 2.8 2.6 1.9 Round 2 -Shared source term and same meteorological data 2.0 2.4

1.7 Sources

• Table 3-2 of [42]; **: Table 3-4 of [42]; ***: Table 3-6 of [42]

Some of the key findings from the report, considering all drills, were that the differences were larger close to the source (< 5 km) and at larger distances (> 50 km) the higher value of GSD at small distances was attributed to many factors. Basically, the plume is the narrowest, which meant that even small differences in the modelling of atmospheric transport and dose impacts could have a significant impact on the results due to the steep concentration gradient. These differences (choice of the height of the ground layer or others) were gradually smoothed when the distance grew because of the diffusion of airborne activity.;

that the largest differences to the [geometric standard deviation] were added by the modelling of the source term; and that Dose calculations had only a minor impact on the differences. This was consistent with the expectations considering that the usual dose evaluation process followed well accepted approaches with similar parameters. [42].

These selected key findings illustrate the significance of the variability in near-field dose estimates, the significance of source term variability, and the significance of prescriptive methods in reducing variability in dose estimates.

4

SUMMARY

AND CONCLUSIONS The results of this survey suggest that reasonable quantitative dose projections may vary by a factor on the order of two to ten and possibly more depending upon the degree to which the variability in the inputs to the dose projection are constrained. This observation is supported by a review of sources of variability in inputs to dose assessments along with formal model comparison exercises designed to ascertain the differences in dose projections conducted by competent experts. The results also support the observation that dose projections in the near field may be particularly variable, because of reasonable variability in modeling of near-field effects including but not limited to effects such as initial release elevation, plume rise, building wake effects, or plume meander.

It may be noted that while constraining the variability of the dose projection inputs by providing prescriptive guidance for the dose analysis may reduce the variability in the results of the analysis, such prescriptive constraints may not be consistent with the goals of emergency

21 planning. Emergency planning is focused on plans to provide dose savings for a spectrum of accidents. As noted in [42],

The definition of a good agreement has to be considered in the context of the individual working with the results of the dose projection tools. Here, the perspective of two groups should be considered: (i) the modelling and assessment experts and (ii) the decisionmakers. For decision-makers in charge of protecting people and the environment, good agreement in assessments is likely to correspond to a situation where the assessments lead to similar protective actions implemented within the affected countries and territories. The good agreement in this sense is defined by the end product, the protective actions as a result of a combination of the dose assessment chain and the countrys regulatory and policy framework. If each countrys protective actions agree well across the borders, the decisionmaker can assume a good agreement. It should be noted that good agreement for decision-makers does not necessarily imply good agreement at the expert level.

It may be noted that the converse may also be true i.e., good agreement at the expert level does not necessarily imply good agreement for decision-makers. In other words, the objective of a dose assessment whether it is to inform an emergency planning process or whether it is to inform emergency response decisions for an actual event needs to be considered when assessing the implications of a particular quantitative result.

It is therefore recommended that the analyst be aware of the potential for effects in which a relatively small change in dose assessment inputs might result in a significantly different decision regarding emergency planning. For example, dose assessments for emergency planning involve planning for a spectrum of accidents. The source terms associated with the spectrum of accidents used in the dose assessments used to scope the planning effort may not exactly mirror those of real accidents. Because of this, accounting for uncertainty in the radiological source terms may be of considerable importance when interpreting the results of a quantitative analysis.

22 5

REFERENCES

[1] M. Drouin, M. Gonzalez, S. Herrick, J. S. Hyslop, D. Stroup, J. Lehner, T. Pratt, M. Dennis, J. LaChance and T. Wheeler, "Glossary of Risk-Related Terms in Support of Risk-Informed Decisionmaking (NUREG-2122)," U.S. Nuclear Regulatory Commission, Washington, DC, 2013.

[2] U.S. Nuclear Regulatory Commission (NRC), "U.S. NRC Level 3 Probabilistic Risk Assessment (PRA) Project, Volume 3d: Reactor, At-Power, Level 3 PRA for Internal Events and Floods (ML22067A215)," U.S. Nuclear Regulatory Commission, Washington, DC, 2022.

[3] American Society of Mechanical Engineers & American Nuclear Society, "Standard for Radiological Accident Offsite Consequence Analysis (Level 3 PRA) to Support Nuclear Installation Applications (ASME/ANS RA-S-1.3-2017)," American Nuclear Society, La Grange Park, IL, 2017.

[4] Sandia National Laboratories, "State-of-the-Art Reactor Consequence Analyses Project:

Sequoyah Integrated Deterministic and Uncertainty Analyses (NUREG/CR-7245)," U.S.

Nuclear Regulatory Commission, Washington DC, 2019.

[5] Sandia National Laboratories, "State-of-the-Art Reactor Consequence Analyses Project:

Uncertainty Analysis of the Unmitigated Long-Term Station Blackout of the Peach Bottom Atomic Power Station (NUREG/CR-7155)," Sandia National Laboratories, Albuquerque, NM, 2016.

[6] K. Compton and A. Hathaway, "Analyses Informing Emergency Planning Zone Size Determinations: Identification of Parameter Sensitivities (ML19338E460)," U.S. Nuclear Regulatory Commission, Washington, DC, 2019.

[7] U.S. Nuclear Regulatory Commission, "U.S. NRC Level 3 Probabilistic Risk Assessment (PRA) Project, Volume 3d: Reactor, At-Power, Level 3 PRA for Internal Events and Floods (ML22067A215)," U.S. Nuclear Regulatory Commission, Washington, DC, 2022.

[8] U.S. Nuclear Regulatory Commission, "Meteorological Monitoring Programs for Nuclear Power Plants, Rev 1 (Regulatory Guide 1.23)," U.S. Nuclear Regulatory Commission, Washington D.C., 2007.

[9] American National Standards Institute/American Nuclear Society, "Determining Meteorological Information at Nuclear Facilities (ANSI/ANS-3.11-2015)," American National Standards Institute / American Nuclear Society, LaGrange Park, IL, 2015.

[10] U.S. Environmental Protection Agency, "Meteorological Monitoring Guidance for Regulatory Modeling Applications (EPA-454/R-99-005)," U.S. Environmental Protection Agency, Office of Air Quality Planning and Standards, Research Triangle Park, NC, 2000.

[11] K. D. Weiksnar, M. L. Garcia, A. T. Nguyen and D. J. Clayton, "Use of National Centers for Environmental Prediction (NCEP) Data to Support Severe Accident Consequence Analysis at Locations Without Onsite Meteorological Data (SAND2024-12539) (ML25049A013),"

Sandia National Laboratories, Albuquerque, NM, 2024.

23

[12] D. B. Turner, "The Long Lifetime of the Dispersion Methods of Pasquill in U.S. Regulatory Air Modeling," Journal of Applied Meteorology and Climatology, vol. 36, no. 8, p. 1016-1020, 1997.

[13] AMS 1977 Committee on Atmospheric Turbulence and Dispersion, "Accuracy of dispersion models," Bulletin of the American Meteorological Society, vol. 59, no. 8, pp.

1025-1026, 1978.

[14] A. F. Stein, R. R. Draxler, G. D. Rolph, J. B. Stunder, M. D. Cohen and F. Ngan, "NOAAS HYSPLIT Atmospheric Transport And Dispersion Modeling System," Bulletin of the American Meteorological Society, vol. 96, no. 12, pp. 2059-2077, 2015.

[15] Y. Tominaga and T. Stathopoulos, "Ten questions concerning modeling of near-field pollutant dispersion in the build environment," Building and Environment, vol. 105, pp. 390-402, 2016.

[16] C. Molenkamp, N. Bixler, C. Morrow, J. Ramsdell and J. Mitchell, "Comparison of Average Transport and Dispersion Among a Gaussian, a Two-Dimensional, and a Three-Dimensional Model (NUREG/CR-6853)," Lawrence Livermore National Laboratory, Sandia National Laboratories, and Pacific Northwest National Laboraratory, Livermore, CA, Albuquerque, NM, and Richland, WA, 2004.

[17] D. J. Clayton, N. E. Bixler and K. L. Compton, "HYSPLIT/MACCS Atmospheric Dispersion Model Technical Documentation and Benchmark Analysis (SAND2022-5515)," Sandia National Laboratories, Albuquerque, NM, 2022.

[18] N. Bixler, K. Compton, M. Dennis, L. Eubanks, R. Haaker, J. Jones, M. Kimura, K.

McFadden, A. Nosek, A. Outkin and F. Walton, "Technical Bases for Consequence Analyses Using MACCS (NUREG/CR-7270; SAND2022-12166 R) (ML22294A091),"

Sandia National Laboratories, Albuquerque, NM, 2022.

[19] F. Pasquill, "The estimation of the dispersion of windborne material," Meteorological Magazine, pp. 90:33-49, 1961.

[20] F. A. Gifford, "Turbulent diffusion typing schemes - a review," Nuclear Safety, pp. 17:68-86, 1976.

[21] M. L. Barad and D. A. Haugen, "Project Prairie Grass, a Field Program in Diffusion (Geophysical Research Papers No. 59)," Air Force Cambridge Research Center, Geophysics Research Directorate, Bedford, MA, 1959.

[22] B. A. Napier, J. P. Rishell and N. E. Bixler, "Final Review of Safety Assessment Issues at Savannah River Site (PNNL-20990)," Pacific Northwest National Laboratory, Richland, WA, 2011.

[23] F. T. Harper, S. C. Hora, M. L. Young, L. A. Miller, C. H. Lui, M. D. McKay, J. C. Helton, L.

H. J. Goossens,, R. M. Cooke, J. Pasler-Sauer, B. Kraan and J. A. Jones, "Probabilistic Accident Consequence Uncertainty Analysis, Dispersion and Deposition Uncertainty Assessment, Volumes 1 and 2 (NUREG/CR-6244)," Commission of the European Communities and U.S. Nuclear Regulatory Commission, Brussels, Belgium, and Washington, DC, 1995.

24

[24] D. J. Clayton and N. E. Bixler, "Assessment of the MACCS Code Applicability for Nearfield Consequence Analysis (SAND2020-2609)," Sandia National Laboratories, Albuquerque, NM, 2020.

[25] U.S. Department of Energy, "Technical Report for Calculations of Atmospheric Dispersion at Onsite Locations for Department of Energy Nuclear Facilities (NSRD-2015-TD01)," U.S.

Department of Energy, 2015.

[26] International Commission on Radiological Protection, "Human Respiratory Tract Model for Radiological Protection (ICRP Publication 66)," International Commission on Radiological Protection, 1994.

[27] J. L. Sprung, J. A. Rollstin, J. C. Helton and H.-N. Jow, "Evaluation of Severe Accident Risks: Volume 2, Rev. 1, Part 7: Quantification of Major Input Parameters, MACCS Input (NUREG/CR-4551)," U.S. Nuclear Regulatory Commission, Washington DC, 1990.

[28] L. Goossens, J. K. B. Harrison, R. Cooke, F. Harper and S. Hora, "Probabilistic Accident Consequence Uncertainty Analysis, Uncertainty Assessment for Internal Dosimetry, Volumes 1 and 2 (NUREG/CR-6571)," Commission of the European Communities and U.S. Nuclear Regulatory Commission, Brussels, Belgium, and Washington, DC, 1998.

[29] J. Gregory, D. Whitehead, C. Ottinger and T. Brown, "Task 5 Letter Report: MACCS Uncertainty Analysis of EARLY Exposure Results," Sandia National Laboratories, Albuquerque, NM, 2000.

[30] U.S. Nuclear Regulatory Commission, "Methods and Assumptions for Evaluating Radiological Consequences of Design Basis Accidents at Light-Water Nuclear Power Reactors (Regulatory Guide 1.195)," U.S. Nuclear Regulatory Commission, Washington, DC, 2003.

[31] U.S. Nuclear Regulatory Commission, "Alternative Radiological Source Terms for Evaluating Design Basis Accidents at Nuclear Power Reactors (Regulatory Guide 1.183),"

U.S. Nuclear Regulatory Commission, Washington, DC, 2000.

[32] Sandia National Laboratories, "FRMAC Assessment Manual, Volume 1: Overview and Methods (SAND2019-0247 R)," Sandia National Laboratories, Albuquerque, NM, 2019.

[33] U.S. Environmental Protection Agency, "Exposure Factors Handbook (EPA-600-R-090/052F)," U.S. Environmental Protection Agency, Washington, DC, 2011.

[34] J. C. Helton, J. D. Johnson, M. D. McKay, A. W. Shiver and J. L. Sprung, "Uncertainty and Sensitivity Analysis of Early Exposure Results with the MACCS Reactor Accident Consequence Model (NUREG/CR-6135 / SAND93-2371)," Sandia National Laboratories, Albuquerque, NM, 1995b.

[35] U.S. Environmental Protection Agency, "Federal Guidance Report 13: Cancer Risk Coefficients for Environmental Exposure to Radionuclides (EPA-402-R-99-001)," U.S.

Environmental Protection Agency, Washington, D.C, 1999.

[36] U.S. Environmental Protection Agency, "Federal Guidance Report 13: Cancer Risk Coefficients for Environmental Exposure to Radionuclides, Updates and Supplements, CD Supplement, Rev 1 (EPA 402-C-99-001)," Oak Ridge National Laboratory, Oak Ridge, TN, 2002.

25

[37] International Commission on Radiological Protection, "Recommendations of the ICRP (ICRP Publication 26)," International Commission on Radiological Protection, 1977.

[38] K. F. Eckerman, A. B. Wolbarst and A. C. B. Richardson, "Limiting Values for Radionuclide Intake and Air Concentration and Dose Conversion Factors for Inhalation, Submersion, and Ingestion (Federal Guidance Report 11, EPA 520/1-88-020, second printing with corrections)," Environmental Protection Agency, Washington, DC., 1989.

[39] International Commission on Radiological Protection, "1990 Recommendations of the International Commission on Radiological Protection (ICRP Publication 60)," International Commission on Radiological Protection, 1991.

[40] H.-N. Jow, J. Sprung, J. Rollstin, L. Ritchie and D. Chanin, "MELCOR Accident Consequence Code System (MACCS): Volume 2, Model Description (NUREG/CR-4691),"

Sandia National Laboratories, Albuquerque, NM, 1990.

[41] M. L. Garcia and K. A. Clavier, "Dose Exceedance Distance Sensitivity Based on Parametric Uncertainty (SAND-2025-08913)," Sandia National Laboratories, Albuquerque, NM, July 2025.

[42] Organisation for Economic Co-Operation and Development / Nuclear Energy Agency (NEA), " Understanding Dose Prognosis in Nuclear and Radiological Emergencies:

Comparison of National Assessment Chains and Practical Guidance for Better Information Exchange and Co-ordination of Protective Actions," OECD Publishing, Paris, 2024.

Memo ML25205A073 OFFICE RES/DSA/AAB RES/DSA/AAB RES/DSA/AAB RES/DSA/AAB NAME SShockley SHaq KCompton ASharp DATE Jul 24, 2025 Jul 24, 2025 Jul 24, 2025 Jul 29, 2025