ML20212G400
| ML20212G400 | |
| Person / Time | |
|---|---|
| Site: | 07000784 |
| Issue date: | 09/22/1999 |
| From: | Camper T NRC OFFICE OF NUCLEAR MATERIAL SAFETY & SAFEGUARDS (NMSS) |
| To: | Decker T NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION II) |
| References | |
| NUDOCS 9909290212 | |
| Download: ML20212G400 (19) | |
Text
c ,
a
, . ~,. ,
September' 22, 1999 MEMORANDUM TO: Tom Decker, Chief Materials Licensing / inspection Branch 1 Division of Nuclear Materials Safety, R ll )
i FROM: Larry W. Camper, Chief [0RIGINALSIGNEDBY:]
Decommissioning Branch Division of Waste Management, NMSS
SUBJECT:
COMMENTS ON UNION CARBIDE CORPORATION REMEDIATION PLAN Union Carbide Corporation (UCC) submitted its Decommissioning Plan (DP) to Region ll on August 19,1998. Region ll and Headquarters (HQ) staff reviewed the DP and identified those areas where additional information is required. In addition, the Dose Modeling Review Group evaluated UCC's dose-based guidelines for soil (Attachment C, Development of Dose-to-Source Factors and Dose-Based Guidelines for Soil). The Dose Modeling Review Group generated a report titled, " Union Carbide Corporation Derived Soil Concentration Limits" (see Attachment 1).
Attachment 2 is a compilation of comments generated by Region ll and HQ staff on the UCC DP. Region ll should contact UCC and request a response to these comments. Upon receiving a satisfactory response, HQ staff will finalize the Safety Evaluation Report (SER) and Environmental Assessment (EA) developed to support NRC's decision to approve the DP. Draft i versions of the SER and EA were e-mailed to Donna Moser/Rll on August 13,1999. Please submit the UCC comments to Headquarters via a Technical Assistance Request to facilitate tracking.
Docket No.70-784 License No. SNM-724, SMB-720 (Terminated)
Attachments: As stated .
Contact:
John Buckley, DWM f ') ,
(301) 415-6607 I For PUBLIC:
This doc t should / / ould not / / be made available to the PUBLIC: 9 / 19 9.
/]
ACNW: YES NO Delete file after distribution: Yes No b) f{
DISTRIBUTION:
(File Centerf JHolonich DCB r/f DWM r/f RJohnson j DMoser Rll BParker ,Rll RNelson
- SEE PREVIOUS CONCURRENC DOCUMENT NAME: s:dwm/ deb /jtb/ucar-com.wpd ,, ) )
lj OFC DCB* DCB* DCB* D / '
ql. NAME JBuckley/cv JCopeland TJohnson hk DATE 8/25/99 8/24/99 9/21/99. 9 4 MJ9 OFFICIAL RECORD COPY 290005 99o9290212 990922 q%9 PDR ADOCK 07 7y4 C
eun
[4 4 UNITED STATES s
) ( ,j NUCLEAR REGULATORY COMMISSION WASHINGTON, D.C. 20566-0001 q ,,,+ September 22, 1999 MEMORANDUM TO: Tom Decker, Chief Materials Licensing / Inspection Branch 1 Division of Nuclear Materia Saf , 11 J FROM: Larry W. Camper, C h. '-o.
Decommissioning Branch Division of Waste Managemert., NMSS
SUBJECT:
COMMENTS ON UNION CARBIDE CORPORATION'S REMEDIATION PLAN Union Carbide Corporation (UCC) submitted its Decommissioning Plan (DP) to Region ll on August 19,1998. Region 11 and Headquarters (HQ) staff reviewed the DP and identified those areas where additionalinformation is required. In addition, the Dose Modeling Review Group evaluatcd UCC's dose-based guidelines for soil (Attachment C, Development of Dose-to-Source Factors and Dose-Based Guidelines for Soil). The Dose Modeling Review Group generated a report titled, " Union Carbide Corporation Derived Soil Concentration Limits" (see Attachment 1).
Attachment 2 is a compilation of comments generated by Region il and HQ staff on the UCC DP. Region ll should contact UCC and request a response to these comments. Upon receiving a satisfactory response, HQ staff will finalize the Safety Evaluation Report (SER) and Environmental Assessment (EA) developed to support NRC's decision to approve the DP. Draft versions of the SER and EA were e-mailed to Donna Moser/Rll on August 13,1999. Please nibmit the UCC comments to Headquarters via a Technical Assistance Request to facilitate
> iracking.
Docket No.70-784 License No. SNM 724, SMB-720 (Terminated)
Attachments: As stated
Contact:
John Buckley, DWM (301)415-6607 9
r MEMORANDUM TO: Tom Decker, Chief Materials Licensing / inspection Branch 1 Division of Nuclear Materials Safety, R 11 FROM: Larry W. Camper, Chief Decommissioning Branch Division of Waste Management, NMSS
SUBJECT:
COMMENTS ON UNION CARBIDE CORPORATION REME ATION PLAN Union Carbide Corporation (UCC) submitted its Decommissioning Plan (DP) Region 11 on August 19,1998. Region 11 and Headquarters (HQ) staff reviewed the DP d identified those areas where additional information is required. In addition, the Dose Mod ing Review Group evaluated UCC's dose-based guidelines for soil (Attachment C, Develo ent of Dose-to-Source Factors and Dose-Based Guidelines for Soil). The Dose Mod ng Review Group generated a report titled, " Union Carbide Corporation Derived Soil C centration Limits"(see Attachment 1).
Attachment 2 is a compilation ot comments generated by Regio 11 and HQ staff on the UCC DP. Region ll should contact UCC and request a response to ese comments. Upon receiving a satisfactory response, HQ staff will finalize the S fety Evaluation Report (SER) and Environmental Assessment (EA) developed to support NR 's decision to approve the DP.
Draft versions of the SER and EA were e-mailed to Don Moser/Ril on August 13,1999.
Contact:
John Buckley, DWM (301) 415-6607 Docket No.: 70-784 License No.: SNM-724, SMB-720 (Terminated Attachments: As stated DISTRIBUTION:
File Center J. Holonich DCB r/f DWM r/f RJohnson DMoser R 11 BParker R ll RNelson DOCUMENT NAME: s:dwm/ deb / /ucar-com.wpd M.M OFC DCB _ DC fk h DCS,f DCB NAME Jsackley pel d TJo n LCamper DATE 8/2(/99 !8/F/99 g/)//99 8/ /99
, OFFICIAL RECORD COPY ACNW: YES _ g Delete file after distribution: YesN No l
/m ( 9
- 1) This docum ( hould not be made available to the PUBLIC I/M/77IO) #
(Initials) (Date)
- 2) This doc ent i is no'tplated to the program. If it is related to HLW, it
! should/sh id not aced in the LSS. /# 6/'I Q7 '
,' (Initials) (Date)
Union Carbide Corporation Derived Soil Concentration Limits
. introduction Union Carbide Corporation (UCC) has submitted a report documenting their approach to developing soil concentration guidelines corresponding to a dose of 25 mrem / year.
Concentration guidelines for natural uranium, depleted uranium, enriched uranium, and Th-232 were derived from dose-to-source factors calculated using RESRAD version 5.781. Dose-to-source factors were calculated using unit soil concentrations and a combination of default parameter. values, PG-8-08'specified values, and site-related literature values. Three land-use scenarios (industrial worker, suburban resident, and resident farmer) were considered in deriving the concentration guidelines, with the resident farmer scenario providing the most restrictive guideline values.
Staff has reviewed the UCC assessment to ensure consistency with current approaches being used in dose assessments for demonstrating compliance with the license termination rule (LTR). DG-4006 2and NUREG-1549 lay out a general approach acceptable to the NRC for doing dose analyses in support of the LTR. Consistent with this guidance, the NRC requires licensees to provide appropriate justification for scenarios, moods, and parameter values used ;
in site-specific dose analyses. Accordingly, the UCC submittal was evaluated by staff with this aspect in mind. NRC staff's review was limited to information contained in UCC's Attachment C " Development of Dose-to-Source Factors and Dose-Based Guidelines for Soll."
Scenarios As previously stated, UCC derived concentration guidelines for three land-use scenarios, with the resident farmer scenario providing the most restrictive guideline values. UCC intends to use in their d0 commissioning, the more restrictive derived concentration limits based on the resident farmer scenario. Because the resident farmer scenario provides the most restrictive concentration limits and is the default scenario recommended in DG-4006 for screening analyses, no additional justification is needed for use of this scenario.
However, in developing parameter values for screening analyses, the NRC has developed a set of default behavioral parameters to represent the average member of the critical group, with the critical group being a resident farmer for soil contamination. These default parameter values
'U.S. Nuclear Regulatory Commission, " Scenarios for Assessing Potential Doses Associated with Residual Radioactivity," U.S. Nuclear Regulatory Commission, Washington, DC 20555, Policy and Guidance Directive PG-8-08, fev 1994.
8U.S. Nuclear Regulatory Commission," Demonstrating Compliance with the Radiological Criteriafor License Termination," U.S. Nuclear Regulatory Commission, Washington, DC 20555, Draft Regulatory Guide DG 4006, August 1998.
l U.S. Nuclear Regulatory Commission," Decision Methods for Dose Assessmmt to Comply with Radiological Criteria for License Termination - Draf t Report for Comment," U.S.
Nuclear Regulatory Commission, Washington, DC 20555, NUREG-1549, July 1998.
Page 1 AAUCAR,RM.WPD l l 999-07-30 l
Attachment 1 Y
i
p-
,. , 1 i I are different from most of the parameter values previously recommended in PG-8-08. 4 Accordingly, NRC staff has advocated that licensee no longer use the PG-8-08 values in screening analyses.
In looking at the specific UCC case, staff made comparison runs with RESRAD (Version 5.82) using both the PG-8-08 values and values taken from the NRC DandD screening code (see Attachment 2). The results are shown in Table 1 below.
Table 1. Comparison using DandD and PG-8-08 parameters.
Soll Concentration Limits (pCl/g)
High Retardation Low Retardation l Radionuclide PG-8-08 DandD PG-8-08 DandD j U-234 291.6 242.2 14.1 17.2 U-235 50.3 56.9 11.9 14.5 )
U-238 162.6 157.9 14.3 17.4 Th-232 1.7 3.1 1.9 2.3 1
Assuming the radionuclides are fairly immobile (i.e., highly retarded), the results are mixed.
The concentration limits for U-234 and U-233 are reduced (i.e., become more restrictive) with the use of the DandD parameter ve!ues. The concentration limits for U-235 and Th-232 are increased with the use of the DandD parameter values. Overall, the differences are fairly small, when the radionuclides are assumed to be highly retarded. When the radionuclides are assumed to be more mobile (i.e., only slightly retarded), the soil concentration limits resulting from using the DandD default parameters are less restrictive than those from using the PG 08 parameters. This is because PG-8-08 have higher values for the fraction of time 4 spent 4
outdoors (0.21 versus 0.1101), mass loading for inhalation (2x10 versus 3.14x10 g/m ), and drinking water intake (730 Uy versus 478.5 Uy) than the default values currently used in DandD. Based upon this comparison, UCC use of PG-8-08 parameter values would appear to be acceptable; however, as will be discussed later, the DandD parameters in combination with the UCC's plant uptake factors give more conservative soil limits. Therefore, it is advisable that the DandD def ault parameter values be used, where they are applicable, in any future analyses unless further justification as to why the PG-8-08 values are more representative of conditions at the site is provided by UCC. This will also ensure consistency with the assumed critical group.
! Models Page 2 A:\ucAR,RPT.WPo j l999-07-30
The UCC Attachment C report did not provide enough of a description of the features and processes at the site in order to construct a site-specific conceptual model. Therefore, no conclusions can be made as to whether or not the predefined conceptual models within RESRAD are consistent with or appropriate for analyzing the site.
Because UCC used an earlier version of RESRAD (i.e.,5.781), staff made a comparison run using the latest version of the code, using UCC's input, in order to see whether this would affect the results. Based on Argonne National Laboratory's (the RESRAD code ca gloper) list of updates
- to RESRAD from versions 5.781 to 5.82, little to no changes in the derived soil limits were expected; however, this comparison also enabled a confirmation of the UCC results and established a baseline for additional comparison runs. The results between the two versions, as shown in Table 2 below, are essentially the same.
(Table 2. Comparison of results from RESRAD Version 5.781 and 5.82. l Radionuclid RESRAD 5.781 Soll Limits (pCl/g) RESRAD 5.82 Soil Limits (pCl/g) e U-234 288.3 291.6 U-235 50.2 50.3 U-238 160.6 162.6 Th-232 2.6 2.6
\
Given that an upgrade to the latest version of RESRAD can be easily downloaded off the Internet (htto://www.ead.ani.aov/~resrad/ patches.htmit it is advisable that any future analyses be done using the latest version of the code.
As previously stated, the UCC Attachment C report does not provide much information on the site description. It does state that the contaminated area may be less than 1,000 m', although UCC assumed a contamination area of 10,000 m2 in deriving its soil concentration guidelines. '
Again using UCC's input, staff developed the following relationships for soil concentration limits as a function of contamination area:
l 1
l 1
1
'http://www.ead. ant.g ov/~ resrad/re sh stry.html
" Page 3 A;\UCAR,RPT.WPD l 1999-07-30
rT L !
Soil Concentration Limits 350
" ~
~
300 __ _ _,,
250 U-235 g 200 --
.- l O
" m U-238 150 i 100 Th-232 50 -
l 0 , ,
0 200o 4000 6000 8000 10000 These Area plots show that the soil concentration limits decrease or become more restrictive 2
as the contaminated 1 l
area is increased. Based on these plots, use of an area of 10,000 m would appear to be appropriate as long as the contaminated area is s 10,000 m"; however, as will be pointed out below, it may be advisable for UCC to use the correct area of contamination in any future a
assessments. It should be noted that a land area of 1000 m is less than the area assumed in the NRC's default resident farmer scenario assumed for screening analyses.
RESRAD offers two approaches for modeling the ground-water pathway, the nondispersion (ND) and mass balance (MB) approaches. For their assessment, UCC used the nondispersion model for analyzing the ground-water pathway. The mass balance approach generally give more conservative estimates of impacts from the ground-water pathway than the ND approach because the hypothetical well is assumed to be located at the center of the contaminated area and generally little dilution is assumed to occur. Because of this conservatism, the developers of RESRAD recommend only using the MB approach for contaminated areas s 1000 m . For the ND method, the hypothetical well is assumed to be located at the down-gradient edge of the contaminated area. Because use of the MB approach is expected to give more conservative 2
soil limits, and given that the contaminated area may be actually s 1000 m , staff also made several comparison runs using the MB approach. The results are listed in Table 3.
I l
!
- Page 4 A:\ucAn_RPT.WPO l l 999-07-30
L . .
I i,
fTable 3 Comparison of results using the Mass Balance and Nondispersion Modeling 3 Approaches.
Soil Concentration Limits (pCi/g)
High Retardation Low Retardation Radionuclide - Nondispersion Mass Balance Nondispersion Mass Balance U-234 308.9 308.9 17.5 14.7 l U-235 52.5 52.5 14.2 12.0 I
U-238 171.3 171.3 17.6 14.4 Th-232 2.7 2.7 2.1 2.0
(
1 I
For all runs, a contamination area of 1000 m2 was assumed. If the radionuclides are assumed to be fairly immobile (i.e., highly retarded), the results between using the ND and MB approaches are the same. This is because the dose is primarily attributable to gamma radiation when the radionuclides are assumed to be immobile. On the other hand,if the radionuclides are assumed to be fairly mobile (i.e., only slightly retarded), use of the MB approach give lower or more restrictive soil concentration limits, especially for the uranium. For this reason, it is recommended that the MB approach be used in future assessments, if the contaminated area is indeed s 1000 m2 . In addition, this points to the need for using the correct contamination !
area, given its role in selecting the appropriate approach for modeling the ground-water pathway.
Parameters Consistent with the iterative approach to dose assessments described in NUREG 1549, the NRC recommends the use of sensitivity analyses to identify key parameters in the assessment used to demonstrate compliance with the LTR. It is then incumbent upon the licensea to provide appropriate justification for values used to represent these key parameters. Such justification can be site-specific data or information which shows that the selected parameter values are appropriate for the site or are expected to lead to a conservative estimate of the impacts. UCC included no sensitivity analyses in their Attachment C report. It is possible that sensitivity analyses were performed, and not included in the report, given that some " site- '
specific" parameter values were used. UCC used " site-specific" parameter values for soil distribution coefficients, plant uptake factors, and indoor gamma shielding factor.
Staff comparison runs, to this point, have generally looked at assuming the radionuclides are l
j either highly retarded or only slightly retarded. UCC in their assessment used distribution
- coefficients which essentially represent immobile conditions. This is a significant assumption
- Page 5 A:\UCAR,RPT.WPo l l 999-O7-30
because impacts from uranium are primarily associated with internal exposure (especially from ingestion of drinking water). Thus, use of high distribution coefficients for uranium and its daughter' products, which makes them essentially immobile, is expected to significantly increase the derived soil concentration limits.
UCC used geometric average distribution coefficients for each soil texture reported in Sheppard i
and Thibault (1990) to calculate a weighted average distribution coefficient for each radionuclide (and their daughters) based on the percentage of clay, sand, and organic content
_ in five soil samples at the site. Table 4 show calculated distribution coefficients that were assumed by UCC to be applicable for the contaminated soil zone, the unsaturated zone, and the saturated zone.
There are several potential concerns with Table 4. Distribution coefficients (Kos) used the approach taken by UCC in deriving by UCC.
_ distribution coefficients. First, no information Element K (cm'/g) is provided on the length of the soil samples.
Therefore, there is no basis to conclude that Ac 2603 these samples are representative of the geologic stratum from the soil surface down Pa 2983 to the water table. Given that no information Pb 299 is provided on the unsaturated and saturated zones, no credit should be Ra 8000 allowed for retardation in these zones.
Table 5 shcw that the derived soit limits are Th 13990 significantly reduced when no credit is taken a for retardation in the unsaturated and U 1403 i saturated zones even with the fairly high (, __ /
retardation still assumed for the i contaminated zone.
Table 5. Comparison using UCC's Kas for all three zones against using UCC's Kg only for the contaminated zone. Second, there is no conclude Soil Concentration Limits (pC1/g) fhat es No retardation in UZ and _ geometric average Radionuclide UCC SZ distribution ,
coefficients taken U 234 288.3 95.4 from the literature are representative !
U-235' 50.2 37.2 of the average l conditions at the i U-238 160.6 77.7 site. Distribution l Th-232 2.6 2.5 coefficients are l
( ) known to be '
affected by a number of factors that can vary from one site to another or even among locations at a specific Page 6 A:\UCAR,RPT.WPo l 1999-07-30
U. v.
o V .
site. Thus, distribution coefficients commonly vary by several orders of magnitude within j- reported literature. The authors of the literature cited by UCC (Sheppard and Thibault)
. specificall'y state that "modelers must be cautioned that the use of generic K3 values can lead to gross error if used to predict absolute impacts of a disposal facility; site-soecific K,. values are !
? essential." Further, the geometric average reported by Sheppard and Thibault (1990) for Ac, Pa, and Pb are actually calculated from a regression equation using a soil-to-plant concentration ratio. If we reduce' UCC's soil distribution coefficients by a factor of 10 and take no credit for retardation in the unsaturated and saturated zones, the derived soil limits for uranium are reduced even further as shown in Table 6.
' h (Table 6. Comparison of results using UCC's Kas against reduced retardation in the contaminated zone and no retardation in the unsaturated and saturated zones.
Soil Concentration Limits (pCl/g} ,
Reduced retardation in CZ and No retardation in UZ Radionuclide UCC and SZ U-234 288.3 14.1 U-235 50.2 11.9 U-238 160.6 14.3 k )
in addition to modifying the distribution coefficients, UCC also modified the plant uptake 5 factors by calculating a weighted average from soil-to-plant factors listed in NUREG/CR-5512 . The percentage of the assumed intake for leafy vegetables, rooted vegetables, f ruits, and grain, taken from PG-8-08, were used as weights. The reported uptake factors, which are on a dry- '
weight basis, were adjusted using dry to-wet factors listed in NUREG/CR-5512. UCC does not explain why they consider their calculated uptake factors to be more appropriate than the default values in the RESRAD code. With the exception of using a weighted average and the use of revised dry to-wet factors, these are the plant uptake factors currently recommended for use in the DandD code. Accordingly, they should be acceptable to the NRC.
Table 7 show the following plant uptake factors were used by UCC in their assessment.
5 Kennedy, W.E., and D.L. Strenge, " Residual Radioactive Contamination from Decommissioning - Technical Basis for Translating Contamination Levels to Annual Total f
1 Effective Dose Equivalent, Final Report," U.S. Nuclear Regulatory Commission, Washington, DC 20555, NUREG/CR-5512, Vol.1, October 1992.
- - Page 7 A:\UCAR,RPT,WPo l t 999-07-30 l I
\
l I
Table 7. Plant uptake factors used in the assessment.
UCC's Factors RESRAD Default s
Element (d/kg) (d/kg) 4 Ac 2.1x10 2.5x10' Pa 1.5x10 4 1.0x10 2 Pb 2.4x10 1.0x10 2 Ra 1.9x10 4 4.0x10-2 4
Th 1.1 x10 1.0x10' U 1.9x10 2.5x10~
J The UCC derived factors are generally an order of magnitude less than the default f actors in RESRAD.
Staff calculated uptake factors using more recent dry-to-wet factors and intake values currently being used in the DandD code to see what effect these would have on the calculated uptake factors. Staff also looked at the effect of using no weights in calculating the average. As shown in Table 8, in general, these had only nominal effect on the calculated factors.
A (Table 8. Plant uptake factors.
Updated Weighted Updated Straight Element UCC Factor Average Average Ac 2.1x10 4 1.7x10 4
2.3x10 4 Pa 1.5x10 4 1.2x10 4
1.7x10 4 Pb 2.4x10~ 1.8x10~ 2x10' Ra 1.9x10~3 2.6x10~ 3.3x10 8 Th 1.1x10 d
1.6x10 4
2.4x10 4 U 1.9x10~' 1.7x10'8 1.7x10~
\ -
- Page 8 A:\UCAR_RPT.WPo l 1999-07-30
r7 - .
i
)
.. j The updated factors were derived using information listed in Table 9, taken from DandD V,ersion 1.0:
i -..
(Table 9. Updated percent consurr.ption and dry-to-wet factors in DandD. '
% !ntake Dry-to-Wet Factor Leafy vegetables 16 0.13 l Rooting vegetables 33 0.16 !
Fruits 40 0.28 Grain 11 0.88
(......_... .. . . ... . ... ,
l To gage the effects of UCC modifying the plant uptake factors, several comparison runs were made to compare the derived soillimits calculated using UCC's plant uptake factors against the soil limits that would have been obtained if the RESRAD default uptake factors had been used.
Table 10 provides the comparative results.
Table 10. Comparison using UCC's plant uptake factors and RESRAD default plant uptake factors.
Soil Concentration Limits (pCl/g)
High Retardation Low Retardation PG-8-08 DandD PG-8-08 DandD Radionuclide UCC Default UCC Default UCC Default UCC Default U-234 291.6 247.2 242.2 168.3 14.1 14.1 17.2 17.1 U 235 50.3 46.7 56.9 30 11.8 11.8 14.5 14.2 U-238 162.6 148.9 157.9 157.1 14.3 14.2 17.4 17.3 Th-232 2.6 1.6 3.1 1.5 1.9 1.3 2.3 1.3 x J If the radionuclides are assumed to be immobile, the use of the modified plant uptake factors does allow an increase in the calculated soil concentration limits, especially for U-234 and U-235. Because of the higher assumed leafy vegetable and milk consumption, thic difference is magnified with the use of the DandD parameters. On the other hand, if the radionuclides are assumed to be fairly mobile, changes in the plant uptake factor have very little effect on the l calculated soillimits.
I a Page 9 A;\UCAR_RPT.WPo l t 999-O7 30
\.
F::
i :,
- Conclusion NRC staff has used a number of comparison runs with the RESRAD code to gage the relative importance of the UCC assessment. From these comparative analyses,it is clear that the derived soil limits (especially for uranium) are highly sensitive to the distribution coefficients
, ~used in the_ assessment. Further, based upon the information contained in UCC's Attachment C report, the distribution coefficients used by UCC are neither demonstrated to be conservative nor appropriate; therefore, there is no basis for accepting the values used by UCC.
Accordingly, it can be concluded that there is an insufficient basis for accepting the soil -
concentration limits currently proposed by UCC for their decommissioning. If UCC intends to account for retardation in their assessment, it is recommended that a more defensible basis be provided for accepting the specific Ko values used in the assessment. This can be accomplished through either the use of Ko values that are demonstrated as being conservative or Ko values that have been derived specifically from the site material. In addition, it is recommended that additional sensitivity analyses be performed to determine whether there are other significant parameters. For example, analyses with less optimistic distribution coefficients show that the ground-water pathway is an important pathway for establishing the limits for uranium. Sensitivity analyses using less optimistic distribution coefficients may show that other parameters important to the ground-water pathway may be also important (e.g.,
hydraulic conductivity or porosity).
[
Page 10 A:\UCAR,RPT.WPo l 1999-07-30
4
- ATTACHMENT 1 COMPARATIVE ANALYSES USING RESRAD PLANT U-234 U 235 U-238 TH-232 PARAMETERS UPTAKE Ko GW MODEL AREA (pCUg) (pCUg) (pCUg) (pCUg)
PG-8-08 UCC UCC ND 10,000 291.60 50.30 162.60 2.60 DindD UCC UCC ND 10,000 242.20 56.95 157.90 3.14 RESRAD PG-8-08 DEFAULT UCC ND 10,000 247.20 46.66 148.90 1.65 PG-8-08 UCC UCC MB 1,000 308.90 52.51 171.30 2.74 PG-8-08 UCC MODIFIED ND 10,000 14.10 11.85 14.26 1.88 DindD UCC MODIFIED ND 10,000 17.23 14.47 17.41 2.30 PG-8-08 UCC MIXED ND 10,000 95.35 37.22 77.74 2.50 PG-8-08 UCC MODIFIED MB 1,000 14.17 11.99 14.35 1.96 RESRAD PG-8-08 DEFAULT UCC ND 10,000 247.20 46.66 148.90 1.65 RESRAD DandD DEFAULT UCC ND 10,000 168.30 29.96 157.10 1.52 RESRAD PG-8-08 DEFAULT MODIFIED ND 10,000 14.05 11.76 14.21 1.33 PG-8-08 UCC MODIFIED b!D 1,000 17.53 14.22 17.59 2.08 RESRAD D ndD DEFAULT MODIFIED ND 10,000 17.08 14.22 17.26 1.28 i
Page 11 A:\UCAR_RPT.WPo l l 999-07-30
. ATTACHMENT 2 DandD Parameters Used in RESRAD Parameter Value Units Inhalation rate 1.169e+04 m'/y Mass loading for inhalation 3.14e-06 g/m 8 Shielding factor for extemal gamma radiation 0.5512 Fraction of time spent indoors 0.6571 Fraction of time spent outdoors 0.1101 Fruits, vegetables, and grain consurnption 112 kg/y Leafy vegetable consumption 21.4 kgly Milk consumption 233 Uy Meat and poultry consumption 65.1 kg/y Fish consumption 20.6 kg/y Soilingestion 18.26 g/y Drinking water intake 478.5 Uy Contamination fraction of drinking water 1 Contamination fraction of livestock water 1 Contamination fraction of irrigation water 1 Contamination fraction of aquatic food 1 Contamination fraction of plant food 1 Contamination fraction of meat 1 Contamination fraction of milk 1 Livestock fodder intake for meat 27.1 kg/d Livestock fodder intake for milk 63.25 kg/d Livestock water intake for meat 50 Ud Livestock water intake for milk 60 Ud Growing season for non leafy vegetables 0.25 y Growing season for leafy vegetables 0.123 y Growing season for fodder 0.1 "2 y Storage time for fruits, non-leafy veg., and grain 14 d Storage time for leafy vegetables 1 d Page 12 A:\UCAR,RPT.WPo l 1999-07-30 I'
+. Storage time for milk 1 d Btorage time for meat and poultry 20 d Storage time for livestock fodder 0 d Fraction of grain in beef cattle feed 0.0743 Fraction of grain in milk cow feed 0.0308 Well pumping rate 118 m'/y irrigation rate 0.5 m/y
\
l l l
1 Page 13 A:\UCAR_RPT.WPo l l 999-07-30 l
l
nz , ,
- e. , ,4 t.
'h I COMMENTS REGARDING THE REMEDIATION (DECOMMISSIONING) PLAN FOR THE
- FORMERLY LICENSED UNION CARBIDE CORPORATION FACILITY (UCC) IN
"" LAWRENCEBURG, TN-11.~ Please identify the remediation contractor'and provide qualifications for the personnel who i will perform remediation. Will remediation be conducted under a State of Tennessee license; if
- so, provide the license number.
2.1 The Remediation Plan (RP)'does not include the Health and Safety Plan (HSP). A copy of the HSP should be provided for NRC review, and indicate whether de plan is associated with an NRC license or a State'of Tennessee license. ,
- 3. L Section 1.3, Page' 6: It states that " . . . a review of records (for the discharge point o' f the 4
, second impoundment) indicated that none of the samples exceeded 7 x 10 uCi/ml which was
, ' nearly an order of magnitude lower than permissible limits for natural uranium and thorium." To Lwhat ' permissible limits' does this text refer? 10 CFR Part 20, Appendix B, Table 2, Column 2 lists 3 x 104uCi/ml as the effluent release limits for uranium 235 and natural uranium.
4.1Section 2.1,'Page 10: It states that the guideline value for general exposure rate
- measurements is 10 uR/hr above background at one meter above the surface being measured.
This exposure rate is applicable for outdoor areas. The exposure rate guideline value for building surfaces is 5 uR/hr.
5.l Section 3.1.3, Page 25: The location of the impoundment / canal / ponds should be shown on Figure 1-1 or an additional figure.
6.D Section 3.1.3, Page 39: Although a core sample taken in Room 133 of Building 10 resulted in a total uranium concentration of 824.4 pCi/g, the RP does not propose further investigation or remediation of potential contamination below Building 10. The RP should provide the rationale for not evaluating the potential for contamination below Building 10.
- 7. Section 3.1.3.5, Page 42: The plan leaves open whether the incinerator pad will be remediated or whether additional modeling will be conducted. State whether additional
' modeling will be conducted, and, if so, provide a description of the model and its conclusions.
- 8. Section 3.1.3.7, Page 44:; Sediment sample collected during the characterization survey
- from 3 manholes around Building 10 exceeded the derived concentration guideline limit of 274'pCi/g (i.e.,2465,4287, and 513 pCi/g). UCAR proposes to either obtain additional sampling or conduct sophisticated modeling. UCAR should provide the method to be used for evaluating whether the effluent piping meets the criteria for release for unrestricted use.
- 9. Section 3.3.6, Page 50: It states that there will be no liquid effluents. Why are no liquid
= ' effluents anticipated, and how will rain water in excavations be handled?
10.1 Section 4.2, Page 55: it states that the proposed survey design follows guidance provided
_. in MARSSIM or NUREG/CR-5949. An " unaffected area" survey apprnch was used during the
- characterization survey to identify affected areas. The RP does not propose to conduct i <
Attachment 2
I affected area surveys (as described in NUREG/CR- 5849) of those areas / rooms that are designated as affected. Instead of performing an affected area survey of an affected area, the RP' proposes to " survey [only) suspect floor grids and all contiguous grids" or only the susoect areas that were identified by an unaffected survey method. This section should be revised to be consistent with NUREG/CR-5849.
- 11. Section 4.2.1, Page 56-57: It states the uranium guidelines will be applied separately and independently to both alpha and beta / gamma measurements. We agree that the guidelines should be applied separately and independently, however, the Table 4.1 values for beta / gamma need to be adjusted to reflect uranium enrichment.
t 12. Section 4.2.1, Page 56: This section should also include a discussion of the elevated level criteria.-
- 13. Section 4.2.3, Page 59: It states that where survey data confirms that release criteria are met, the area will be classified as unaffected. Further, only those areas classified as affected will be sampled during the final status survey. This area classification method is inconsistent with the process described in NUREG/CR-5849. UCAR must identify survey units .in accordance with NUREG/CR-5849.
i
- 14. Section 4.2.4, Page 59: This section references Figure 3-1 which shows a 10 f t x 10 ft grid.
Does UCAR intend to survey on 10 ft square grids or 10 m square grids?
- 15. Section 4.2.5, Page 59: It says that 10 samples will be obtained from the 10,000 fta area <
surrounding the incinerator pad. This approach is inconsistent with NUREG/CR-5849. . UCAR must use the approach described in NUREG/CR-5849, or provide statii st cal basis for 10 !
j samples.
i
- 16. Section - Executive Summary, Vol. 2, Page i: It states, " Trace levels of thorium material l were also identified but have little or no impact in determining whether contamination uceeds the applicable guideline." Since thorium and depleted uranium were also used at this facility under AEC License No. SMB-720, the concentration limits for unrestricted release of 10 pCi/g for thorium and 35 pCi/g for depleted uranium also apply. The limit for thorium is clearly the j most restrictive. The is also true for the facility surface survey. Therefore, those facilities in which thorium was used must meet the unrestricted release criteria for thorium.
- 17. Section 3.2 Vol 2, Page 10: What is the depth of the groundwater monitoring wells, and did these wells sample the uppermost (watertable) aquifer?
118. Section 3.2.2 Vol. 2, Page 11: The RP indicates that only one soil coring and one surface sample were collected from the Shoal Creek outfall area which spans a distance of approximately 500-1000 yards. Only three soil samples were obtained of the interconnecting canal which covers an 580 m x 20 m area. It is not clear that any type of surface scan was performed of the Shoal Creek outfall area or of the interconnecting canal. Considering the size of these two outdoor areas, the characterization survey should be supplemented with additional data (i.e., more complete surface scan or additional samples) to conclude that these areas meet the criteria for release for unrestricted use.
2
l' SPECIFIC COMMENTS ON ATT CHMENT C - DEVELOPMENT OF DOSE-TO-SOURCE FACTORS AND DOSE-BASED GUIDELINES FOR SOIL -
- 19. Attachment C does not provide sufficient details of the site features and processes to I L construct a site specific conceptual model. Therefore, no conclusions can be made as to j t
whether or not the predefined conceptual models within RESRAD are consistent with or appropriate for analyzing the site.
- 20. UCC should use the latest version of RESRAD for its analyses. The latest version of RESRAD is available on the internet at http://www.ead.ani.aov/~resrad/oatches.html
- 21. It appears that UCC did not perform sensitivity analyses to identify key parameters in the {
assessment used to demonstrate compliance with the license termination rule. Sensitivity '
analyses should be used to provide appropriate justification for values used to represent key parameters.
- 22. The distribution coefficients used in the analyses are neither demonstrated to be conservative nor appropriate, and therefore, there is no basis for accepting the values used by UCC. UCC should provide a more defensible basis for the specific Ka values used in the l
- assessment. This can be accomplished by using Kovalues that are demonstrated to be l l
conservative or by using Ko values that have derived specifically from the site material.
)
j 4 3
)
- l