ML20254A003
| ML20254A003 | |
| Person / Time | |
|---|---|
| Site: | PROJ0734 |
| Issue date: | 10/05/2020 |
| From: | Stephen Koenick Division of Decommissioning, Uranium Recovery and Waste Programs |
| To: | Folk J US Dept of Energy, Savannah River Operations Office |
| Felsher H | |
| References | |
| Download: ML20254A003 (15) | |
Text
UNITED STATES NUCLEAR REGULATORY COMMISSION WASHINGTON, D.C. 20555-0001 October 5, 2020 Mr. James L. Folk, Jr.
Assistant Manager Waste Disposition U.S. DOE Savannah River Operations Office P.O. Box A Aiken, SC 29802
SUBJECT:
PRELIMINARY REVIEW OF THE U.S. DEPARTMENT OF ENERGYS SUBMITTAL OF THE 2020 SAVANNAH RIVER SITE SALTSTONE DISPOSAL FACILITY PERFORMANCE ASSESSMENT
Dear Mr. Folk:
By letter dated July 6, 2020, the U.S. Department of Energy (DOE) submitted the 2020 DOE Savannah River Site (SRS) Saltstone Disposal Facility (SDF) Performance Assessment (PA) to the U.S. Nuclear Regulatory Commission (NRC) (NRCs Agencywide Documents Access and Management System [ADAMS] under Package Accession No. ML20190A055). That DOE Submittal to the NRC was under Section 3116(b) of the Ronald W. Reagan National Defense Authorization Act for Fiscal Year 2005 (NDAA) (i.e., NDAA Monitoring). By letter dated July 10, 2020 (ADAMS Accession No. ML20148M201), the NRC acknowledged receipt of that DOE Submittal.
The 2020 DOE SRS SDF PA and its supporting documents are used to demonstrate to the NRC and the NDAA-Covered State of South Carolina that the DOE disposal actions at the SRS SDF will satisfy the NDAA and that the waste can be managed as low-level waste.
The NRC has completed its Preliminary Review of the 2020 DOE SRS SDF PA. The NRC found that there is sufficient technical information in most areas in the DOE Submittal such that the NRC can begin the detailed technical review. However, the NRC requests that the DOE provide the supplemental information about the SDF found in the Enclosure and summarized below.
During the Preliminary Review, the NRC held a teleconference call with the DOE on September 2, 2020, and discussed the topical areas of Barrier Properties, Saltstone Properties, Near-Field Flow, and External Environment. Based on the 2017 NRC Technical Review Report (TRR) (ADAMS Accession No. ML17081A187) and the 2018 NRC/DOE interactions (ADAMS Accession Nos. ML18219B859 and ML19087A171), the NRC expected more information about those topical areas to be included in the 2020 DOE SRS SDF PA. Therefore, the NRC has identified the need for early supplemental information in these topical areas:
- Barrier Properties:
o analyses that showed the risk significance of the combined uncertainty in multiple barriers to flow; and
J. Folk o model support for the projected long-term performance of the sand drainage layers and composite barriers
- Saltstone Properties:
o analyses that showed the risk significance of alternative assumptions about saltstone degradation; and o technical bases for excluding several saltstone degradation mechanisms
- Near-Field Flow:
o analyses that showed the risk significance of moisture characteristic curves, if flow barriers do not perform as designed; and o complete water budgets
- External Environment:
o analysis of erosion adjacent to the SDF; and o technical basis for excluding receptor exposure to contaminated water from above the tan clay confining zone The NRC plans to have teleconference calls with the DOE and South Carolina throughout the technical review, as needed. The NRC plans to prepare TRRs on specific technical concerns and issue Request for Additional Information (RAI) Questions, if needed, when drafting those TRRs. Thus, the NRC expects that there will be multiple sets of RAI Questions. The NRC Technical Evaluation Report (TER) will reference all those TRRs and the TRRs will be issued at the same time as the TER.
Due to the need for the supplemental information about the SDF and uncertainty when the DOE will provide the supplemental information to the NRC, the NRC will not provide a review schedule at this time. After receiving the supplemental information, the NRC will develop a project schedule for the review and provide the schedule to the DOE and South Carolina. After the NRC provides the review schedule, if there are emergent complexities or challenges in our review that would cause changes to the schedule, then the reasons for the changes and the new schedule will be communicated to the DOE and South Carolina.
In accordance with Title 10 of the Code of Federal Regulations (10 CFR) Section 2.390, Agency Rules of Practice and Procedure, a copy of this letter will be available electronically for public inspection in the NRC Public Document Room or from the Publicly Available Records component of ADAMS. ADAMS is accessible from the NRC Web site at http://www.nrc.gov/readingrm/adams.html
J. Folk If you have any questions, then please contact Mr. Harry Felsher of my staff at 301-415-6559 or by e-mail, harry.felsher@nrc.gov.
Sincerely, Stephen S. Koenick, Chief Low-Level Waste and Projects Branch Division of Decommissioning, Uranium Recovery, and Waste Programs Office of Nuclear Material Safety and Safeguards Docket No.: PROJ0734
Enclosure:
Request for Supplemental Information cc w/
Enclosure:
WIR ListServ
J. Folk
SUBJECT:
PRELIMINARY REVIEW OF THE U.S. DEPARTMENT OF ENERGYS SUBMITTAL OF THE 2020 SAVANNAH RIVER SITE SALTSTONE DISPOSAL FACILITY PERFORMANCE ASSESSMENT DATED: October 5, 2020 DISTRIBUTION:
GAlexander, NMSS HArlt, NMSS LDesotell, NMSS ADAMS Accession Number: ML20254A003 *via email OFFICE NMSS/DUWP NMSS/DUWP NMSS/DUWP NMSS/DUWP NAME HFelsher* ARidge* CMcKenney* SKoenick*
DATE 09/09/2020 09/18/2020 09/22/2020 10/5/2020 OFFICIAL RECORD COPY
Request for Supplemental Information for the 2020 U.S. Department of Energy Savannah River Site Saltstone Disposal Facility Performance Assessment INTRODUCTION:
The U.S. Nuclear Regulatory Commission (NRC) staff identified the following Request for Supplemental Information (RSI) Comments during the Preliminary Review of the 2020 U.S.
Department of Energy (DOE) Savannah River Site (SRS) Saltstone Disposal Facility (SDF)
Performance Assessment (PA).
RSI-1: Combined Uncertainty of Flow Barriers To evaluate the risk significance of flow barriers, the NRC staff needs information about the uncertainty in SDF performance caused by the combined uncertainty in the performance of closure cap, engineered barriers above the disposal structures, and saltstone wasteform.
Basis The NRC staff previously expressed the need for additional technical bases to support the modeled performance of the closure cap and the engineered barrier above the disposal structures (i.e., the lower lateral sand drainage layer (LLDL) and the underlying composite barrier made of high-density polyethylene (HDPE) and a geosynthetic clay liner (GCL)) in ADAMS Accession Nos. ML18219B859, ML19087A171, and ML17081A187). Specific concerns related to those barriers that were not addressed in the PA are described below in RSI-2 and RSI-3. Because those barriers both act to limit water flow through saltstone, the NRC staff needs information about the effects of the combined uncertainty in both barriers to evaluate the risk significance of those barriers and the projected radionuclide releases from saltstone. The NRC staff expects radionuclide releases to be sensitive to flow because increased water flow accelerates saltstone degradation (see RSI-4), which is expected to further increase flow, causing a feedback loop. That sensitivity to water flow makes it difficult to determine system performance by considering sensitivity analyses that evaluate barriers to flow individually.
A limited number of sensitivity cases in the PA considered changes in both the closure cap and the engineered barrier above the disposal structures. Although Section 5.8.2 of the PA included sensitivity analyses that tested the effects of degraded closure cap performance, those analyses assumed good performance of the engineered barriers above the disposal structures that did not address the concerns in RSI-2 and RSI-3. Although the analyses in Section 5.8.3 of the PA evaluated the effects of degraded performance of the sand layers in both the closure cap and the LLDL, no technical basis was provided for the limited range of sand hydraulic conductivities evaluated. In addition, those analyses assumed the remainder of the cap and the HDPE/GCL composite barriers above the disposal structures performed as designed, making the results difficult for the NRC staff to interpret.
Because of the significant uncertainties in the performance of both the closure cap and the engineered barriers above the disposal system, the NRC staff needs to consider the degraded performance in both layers to understand the uncertainty in the system performance. In addition, the closure cap and engineered barrier above the disposal structures contain common elements (e.g., HDPE/GCL composite barriers, sand drainage layers) that could affect the performance of both barriers.
Enclosure
Therefore, including degraded performance of both layers in a single analysis is needed to assess the effects of potential common cause failures and does not represent a worst case scenario. Furthermore, because increased flow is expected to accelerate saltstone degradation, which in turn is expected to increase water flow, the uncertainty in the hydraulic performance of saltstone is physically related to the performance of the other flow barriers.
Therefore, consideration of increased saltstone degradation (see RSI-4) with decreased performance of the sand drainage layers (see RSI-2) and HDPE/GCL composite barriers (see RSI-3) is physically reasonable and does not represent a worst case scenario.
Probabilistic analyses can project the combined uncertainty based on the uncertainty in interrelated parts of a complex system. Therefore, the NRC staff evaluated Sections 4.5 and 5.7 of the PA to determine if the probabilistic analysis described in those sections provided information about the uncertainty in SDF performance that results from the combined uncertainty in the flow barriers. However, that information was not available in the PA because the LLDL, HDPE/GCL above the disposal structures, and moisture characteristic curves (MCCs) were not included in the probabilistic analysis. Furthermore, although the cementitious degradation rate (including saltstone degradation) was included in the probabilistic analysis, the averaging method used and the degradation mechanisms considered significantly limited the utility of that aspect of the analysis (see RSI-4).
The NRC staff could not independently test the effects of uncertainty in flow barriers with the DOE SDF GoldSim model because those barriers were not explicitly represented in the GoldSim model. Instead, the DOE used a hybrid modeling approach in which each different set of near-field flow conditions was represented by a set of PORFLOW flow fields that were input into the GoldSim model. That approach limited the scope of the uncertainty information provided by the GoldSim model to those features varied in the set of PORFLOW analyses used as GoldSim inputs. In the PA, four variables were included in the set of PORFLOW cases used in the probabilistic analysis: (1) infiltration rate; (2) cementitious degradation rate; (3) saturated hydraulic conductivity of backfill; and (4) initial saltstone hydraulic conductivity. The DOE did not provide an explanation for why other features that were expected to be significant barriers to flow, such as the LLDL or HDPE/GCL composite barriers above the disposal structures, were not included in the probabilistic analysis.
Path Forward Provide an analysis that demonstrates the effects of the combined uncertainties of the sand drainage layers, HDPE, HDPE/GCL composite barriers, saltstone degradation, and MCCs. One method for providing that information would be to include those features and variables in a probabilistic analysis of the SDF performance and to provide results similar to the results provided in Sections 5.7.3 through 5.7.5 of the PA. The input ranges used in an analysis should be consistent with the responses to NRC RSI Comments about the sand drainage layers (see RSI-2); HDPE, GCL, and HDPE/GCL composite barriers (see RSI-3); saltstone degradation (see RSI-4); and MCCs (see RSI-5).
RSI-2: Sand Drainage Layers To evaluate the risk significance and the projected performance of the sand drainage layers, the NRC staff needs supplemental information about the uncertainty associated with the LLDL and the upper lateral drainage layer (ULDL).
Basis Based on the analysis of the results for the various modeling cases presented in the PA, the most significant barrier with respect to overall performance is the engineered closure cap, which significantly limits infiltration. Infiltration rates have dramatically decreased in the PA compared to the 2016 DOE SDF Special Analysis Document (ADAMS Accession No. ML18081A262)
(e.g., in the year 560, decreasing from 26 to 0.091 millimeters/year), which is 285 times less water moving towards the disposal structures each year during the latter part of the DOE compliance period. The sand drainage layers are key system components in significantly reducing simulated infiltration rates. Much of the information that the DOE relied upon for the closure cap and for the lateral sand drainage layers in particular, is in the April 23, 2018, Report, "Predicting Long-Term Percolation from the SDF Closure Cap" (ADAMS Accession No. ML18170A244) (April 2018 Report).
Previously, the NRC staff asked the DOE for information that would strengthen the DOE technical basis for the lateral sand drainage performance. The NRC staff expressed concerns about the assumed performance of the two sand drainage layers (i.e., LLDL, ULDL) in the:
(1) April 12, 2017, NRC Technical Review Report (TRR), Performance of the High Density Polyethylene Layer, High Density Polyethylene/Geosynthetic Clay Liner Composite Layer, and the Lower Lateral Drainage Layer (ADAMS Accession No. ML17081A187); (2) November 1, 2018, Onsite Observation Visit (OOV) Report for the July 9-11, 2018, OOV (ADAMS Accession No. ML18219B859), and (3) July 2018 NRC staff Preliminary Comments to the DOE on the April 2018 Report (ADAMS Accession No. ML19087A171).
Although the NRC staff did not expect a direct DOE response to the July 2018 NRC staff Preliminary Comments, the NRC staff did expect to see the topics of the Preliminary Comments described in either the PA or its supporting documents, which was expected to strengthen the DOE technical bases for those significant barriers. However, that information was not included in either the PA or its supporting documents.
In the July 2018 NRC staff Preliminary Comments, the NRC staff stated that reliance or decisions about model projections of infiltration at thousands of years after site closure should have multiple lines of reasoning to support the conclusions, which was especially true if most of the reasons were based on observations in much younger (i.e., decades old or less) engineered systems. The NRC staff was aware that the Giroud-Houlihan analytical solution and the hydraulic conductivity value of the ULDL were expected to be risk-significant and that the NRC staff would review those areas in more detail. The following were observations and comments pertaining to the ULDL and the LLDL:
- The NRC staff noted that examples in the December 2011 NUREG/CR-7028, Engineered Covers for Waste Containment: Changes in Engineering Properties and Implications for Long-Term Performance Assessment (ADAMS Accession No. ML12005A110) were not ideal long-term analogues of fine-over-coarse interfaces at depth over very long periods of time due to the 5-to-15 years cover life-span of those examples. However, even after the relatively short service lives of those covers, NUREG/CR-7028 described the discovery of fine particles within the exhumed geotextiles, geonets, and geosynthetic drainage layers that apparently were transported there due to the migration of those fines.
- The NRC staff noted that the Kyushu burial mound site in Japan used alternating layers of clay and loam, which was different than the current SDF Cover design (i.e., drainage layer made of sand). In addition, it was not clear if the coarse-grained layers of both long-term, man-made analog sites (i.e., Kyushu, Tu-Dun) ever experienced the volumetric flow predicted for sand drainage layers at the SDF. Also, it was not clear if those two analog sites were representative of all known or unknown/destroyed analog sites.
As stated above, the NRC staff had previously asked the DOE for information that would strengthen the DOE technical basis for the lateral sand drainage performance. The PA did not include additional information that could have strengthened that technical basis and the full range of uncertainty was not represented. Therefore, the previous DOE conceptual models of sand drainage layer degradation and performance remain plausible, including the conceptual model and parameter values associated with the sand drainage layers from the 2009 DOE SDF PA (ML101590008) (e.g., the ULDL and the resulting infiltration rates). In the January 31, 2018, NRC TRR, Hydraulic Performance and Erosion Control of the Planned SDF Closure Cap and Adjacent Area (ML18002A545), the NRC staff commented on the DOE analysis of the closure cap described in the December 2007 F-Tank Farm Closure Cap Concept and Infiltration Estimates, WSRC-STI-2007-00184, Rev. 2 (ADAMS Accession No. ML111240597), which presented an alternative degradation conceptual model of the lateral drainage performance different from that of the conceptual model considered for the SDF Cover. The DOE has not presented robust technical bases to exclude those additional conceptual models from consideration. Therefore, the NRC staff needs information about the system performance based on those previously documented DOE conceptual models and their parameter values.
Path Forward Provide ranges of saturated vertical and horizontal hydraulic conductivity for the sand drainage layers (i.e., both LLDL and ULDL) that reflect the sources of parametric and model uncertainty (see Discussion above). The response should address both initial and degraded values.
Provide infiltration rates that result from considering those sources of degradation and uncertainty in the ULDL.
RSI-3: High Density Polyethylene / Geosynthetic Clay Liner Composite Barriers To evaluate the risk significance and the projected performance of flow barriers, the NRC staff needs supplemental information about the uncertainty associated with the HDPE, the GCL, and the combined HDPE/GCL composite barriers in the closure cap, above the disposal structure roofs, and between the mud mats.
Basis Based on an NRC staff analysis of the results for the various modeling cases presented in the PA, the most significant barrier with respect to overall performance is the engineered closure cap, which significantly limits infiltration. Section 4.4.1.3.6 of the PA indicates that the HDPE is an extremely important barrier relative to the performance of the SDF. The HDPE, the GCL, and the composite barrier are key system components in significantly reducing simulated infiltration rates. According to the DOE, the differences in modeling assumptions for the composite barrier in the closure cap and the composite barrier above the disposal structures were driven by the different purposes of the two models. It is not clear to the NRC staff what the technical justification was for that modeling decision. The NRC staff believes that much of the information relied upon for the closure cap and for the composite barrier in particular is found in the April 2018 Report.
As described above in RSI-2, the NRC staff previously asked for additional technical bases for aspects of the SDF closure cap modeling. Although Table 2 in the April 2018 Report provided and represented some measure of uncertainty, it did not represent the full range of uncertainty associated with the long-term performance of engineered surface covers (e.g., uncertainty in the frequency of defects in the HDPE at thousands of years after site closure, uncertainty in long-term parameter values for a manmade drainage layer). The April 2018 Report did not represent uncertainty in the size and spacing of defects nor the number of potential pinholes resulting from manufacturing flaws, such as polymerization deficiencies. The April 2018 Report and the PA listed one source for the number and size of the holes, although several sources were found in the literature. Although NUREG/CR-7028 was frequently referenced in the April 2018 Report, the GCL hydraulic conductivity values in Table 2 of the Report were based on a 14-year old Barnwell Cover sample that resulted in a percolation rate to the layer below of less than one-fifth of that predicted for a GCL with the NUREG/CR-7028 recommended value.
The November 1, 2018, OOV Report summarized both the technical discussion of the DOE information on the closure cap and the presentation by Dr. Craig Benson on the April 2018 Report. The OOV Report stated that the NRC staff indicated that the level of confidence that was expressed in the service life of HDPE in the April 2018 Report would need significant technical support. For example, the NRC staff expressed interest in the technical basis for the estimated service life given in the April 2018 Report exceeding 1,900 years because NUREG/CR-7028 indicated that the service life of HDPE in cover systems could be expected to be 50-to-100 years. The NRC staff did not know of any other project that has claimed significant HDPE performance for more than a few hundred years combined with extremely low infiltration rates in such a humid climate.
The November 1, 2018, OOV Report stated that the DOE and NRC staffs agreed that the performance of the seam welds will strongly depend on the quality of the installation. The PA described the quality of the future installation as being reasonably good. However, the issue of the potential Saltstone Disposal Structure (SDS) 3A HDPE breach, as described in the April 12, 2017, NRC TRR, was not described in the PA or its supporting documents. It is not clear that the DOE had taken that type of breach into consideration for the current technical basis supporting the expected performance of the HDPE layer and composite barrier. The NRC staff previously suggested that the DOE document the basis for why the DOE does not expect that breach to occur again and why the that type of breach was not applicable to other areas of the SDF. However, the NRC staff is not aware of any such DOE documentation. Although that breach may have occurred in an area with a three-dimensional HDPE layout, the disposal structure roofs will also have slightly three-dimensional cone shapes, which will require more seams to be welded as the installation gets closer to the top of a roof. The DOE, NRC staff, and geomembrane literature all agree that a significant portion of the geomembrane performance is usually a function of the installation quality. Guaranteeing that quality is difficult to do considering the size and number of geomembrane panels to be welded and numerous other variables. Narrowly bounding the uncertainty range in this area also presents significant difficulties. For example, the pictures of the composite barrier installation between the SDS 7 mud mats, from the October 2018 Pictures from Construction of SDS 7: Excavation/lower Mud Mat/HDPE-GCL/Upper Mud Mat, SRR-CWDA-2018-00078, Rev. 1 (ADAMS Accession No. ML18338A189) showed many waves and wrinkles in the sheets of HDPE, which would require extra care to straighten out as the cementitious material was being poured.
As stated above, the NRC staff had previously asked the DOE for information that would strengthen the DOE technical basis for the composite barrier performance. The PA did not include additional information that could have strengthened that technical basis. Therefore, the previous DOE conceptual models of HDPE and GCL degradation and performance remain plausible, including the conceptual model and parameter values associated with the composite barrier layers from the 2009 PA. The DOE has not presented a robust technical basis to exclude that additional conceptual model from consideration. Therefore, the NRC staff needs information about the projected system performance based on the previously documented DOE conceptual model and its parameter values.
Path Forward Provide ranges of hydraulic property values that reflect the sources of parametric and model uncertainty discussed in this RSI for the HDPE, the GCL, and the combined HDPE/GCL composite barriers in the closure cap, above the disposal structure roofs, and between the mud mats. The response should address both initial and degraded values. Provide infiltration rates that result from considering those sources of uncertainty in the closure cap.
RSI-4: Saltstone Degradation To evaluate the risk significance of the hydraulic performance of saltstone, the NRC staff needs information about how the following issues affect saltstone degradation: (1) additional and coupled saltstone degradation mechanisms; (2) arithmetic averaging of hydraulic conductivity for degraded and intact saltstone; and (3) uncertainty in flow through the closure cap and engineered barriers above the disposal structures.
Basis In the PA, the DOE assumed that the saltstone grout would degrade by decalcification when water flows through saltstone. With extremely limited water flow through the closure cap (i.e., infiltration is reduced by approximately three orders of magnitude from natural infiltration),
the LLDL, and underlying composite barrier, the saltstone is projected not to degrade appreciably in the timeframes analyzed in the PA. That is because of the combination of the long time to the projected complete degradation of saltstone (i.e., 17 million years) and the use of a geometric average to calculate the effective hydraulic conductivity of saltstone.
The NRC staff is concerned that: (1) decalcification could occur more quickly than projected due to potentially greater-than-assumed infiltration (see both RSI-1 and RSI-2 above);
(2) additional and coupled degradation mechanisms could result in more rapid degradation of saltstone than the DOE assumed in the PA; and (3) the use of a geometric average is not adequately supported and could significantly underestimate the effective hydraulic conductivity of degraded saltstone.
For saltstone grout, the only degradation mechanism carried past the screening process into the PORFLOW model supporting the PA was advection-controlled decalcification. The assumptions about saltstone from Section 2.7.6 of the PA stated:
- The saltstone will be completely encapsulated within the concrete [disposal structures].
As such, no significant mechanical degradation is expected to influence the performance of saltstone. Similarly, due to the chemical characteristics of saltstone, it is not subject to sulfate attack or microbial induced degradation and, because saltstone has no rebar or steel embedded within it, it is also not subject to carbonation. Therefore, it is reasonable to assume that decalcification (i.e., dissolution and chemical leaching of calcium) is the primary mechanism of saltstone degradation.
The NRC staff evaluated those characteristics in Appendix A of the May 18, 2019, NRC TRR, Saltstone Waste Form Physical Degradation (ADAMS Accession No. ML19031B221) and provided information on potential additional saltstone degradation mechanisms. In that TRR, the NRC staff described that mechanical degradation can still affect saltstone due to mechanisms such as: long-term drying shrinkage, settlement, and loading. With respect to chemical degradation, the NRC staff described the potential for expansive phase formation and the lack of the DOE support for excluding sulfate attack. The NRC staff also described an observation of microbial activity on cast stone, which is similar to saltstone. The alkalinity and high pH of saltstone do not appear to preclude microbial degradation, especially with successive pore volume flushes decreasing the alkalinity and salt content. Thermal degradation due to temporal and spatial thermal gradients also could occur. Furthermore, feedback between multiple degradation mechanisms could further increase the rate of degradation. The PA and its supporting documents did not provide information to: (1) refute other plausible degradation mechanisms; (2) demonstrate that the assumed degradation rate due to decalcification represented or exceeded the potential rate of degradation due to additional and coupled degradation mechanisms; or (3) demonstrate the risk significance of saltstone degradation.
The DOE use of geometric averaging is not adequately supported because it depends on the assumption that flow is perpendicular to degraded layers. If saltstone does not degrade uniformly from top-to-bottom creating a uniform, horizontal layer of degraded saltstone, then the geometric average will not yield a reasonable effective hydraulic conductivity. Degradation of saltstone is likely to be non-uniform and may be caused by formation of preferential flow paths and localized decalcification or degradation caused by other mechanisms. Under a more-typical, non-uniform degradation front, flow would tend to be parallel to the path of flow, which is better represented by an arithmetic average. Using the arithmetic average is consistent with what the DOE previously used in the 2014 DOE SDF Special Analysis Document (ADAMS Accession No. ML15097A366), which stated:
- This [2014 DOE SDF Special Analysis Document] applies the more conservative approach of linear averaging, in part to compensate for departures from flow and transport perpendicular to the uniform degradation front.
Although the DOE presented results for an accelerated cementitious materials degradation case in Section 5.8.2.4 of the PA, that case does not appear to have any appreciable impact on saltstone grout within 100,000 years and therefore does not provide any insight into what could happen if there is more water flow through saltstone. The DOE also tested the effects of a higher initial hydraulic conductivity for saltstone in Section 5.8.2.4 of the PA; but, the range of conductivities tested in that case is within the range of observed hydraulic conductivity values for intact saltstone, not degraded saltstone.
Path Forward Provide an analysis that: (1) provides risk insight into the effects of additional and coupled saltstone degradation mechanisms; (2) uses a more realistic arithmetic average for degraded and intact saltstone; and (3) is consistent with the response to RSI-1, RSI-2, and RSI-3 regarding uncertainty in flow through the closure cap and engineered barriers above the disposal structures. It would be useful if that analysis would include the volumetric flow rates, as shown in Figure 7.1.1 of the PA, for key materials (e.g., saltstone, roof, walls, joints, fast flow paths).
RSI-5: Moisture Characteristic Curves To evaluate the risk significance of MCCs for cementitious materials and the gravel for the fast flow paths, the NRC staff needs information about the effect of the MCCs on near-field flow when the closure cap and engineered barriers above the disposal structures do not perform as designed.
Basis Section 5.8.8.3 of the PA indicated that the MCCs do not affect the dose because the flow through the system was limited by infiltration rather than flow through the cementitious materials. However, as described above in both RSI-2 and RSI-3, the NRC staff is concerned that the PA did not consider the full range of uncertainty in the performance of the closure cap or the engineered barriers above the disposal structures. The PA did not provide information on the risk significance of MCCs if the flow through the system was limited by the flow through the cementitious materials.
Path Forward Provide an analysis that provides insight into the risk significance of MCCs of cementitious materials and fast pathways through the disposal structures if the closure cap, LLDL, and composite barrier above the disposal structures do not perform as expected. That information could include a new sensitivity case consistent with the DOE responses to RSI-1, RSI-2, RSI-3, and RSI-4, which would include a run with the MCCs assumed in the PA and a run with the relative permeability equal to 1.
RSI-6: Complete Water Budgets To evaluate the implementation of the conceptual model for near-field flow, the NRC staff needs supplemental information about the internal flow components of the entire disposal system with in- and outflow- components having the same units of rate or flow.
Basis Although partial water budgets, also called water balances, were in Sections 3.2.6, 4.4.4, and 7.1.1 of the PA, the PA did not include complete water budgets of the entire disposal system for specific time intervals of any of the sensitivity analyses or the three sets of the Central Scenario.
Complete flow balances would include volumetric flow from lateral and vertical drainage.
In the July 15, 2016, OOV Report for the April 19-21, 2016, OOV at the SDF (ADAMS Accession No. (ML16147A197), the NRC staff described water budgets consisting of the outflow components (e.g., evaporation, evapotranspiration, runoff, lateral or vertical outflow to the model boundaries) and separating the inflow components of surface recharge from lateral or vertical inflow from the model boundaries. Such information would provide more understanding of the magnitudes of internal water flow rates and the location of significant features and processes.
During the OOV, the NRC staff and the DOE discussed that the DOE might provide that information in the PA.
Water budgets of the entire system allow greater insights into the workings of the system, showing strengths and weaknesses of the system and helping to determine which components of the disposal structures are significant to performance and to identifying potential simulation errors. Such water budgets are standard features for documenting the results of groundwater models and, although the PA is not comparable to a groundwater model, the insights gained by that technique would be the same.
Path Forward Provide complete water budgets of the disposal system in its entirety for: (1) the compliance case; (2) the second sensitivity case from Section 5.8.2.3 in the PA; (3) one of the three soil-only closure cap cases in Section 5.8.3 of the PA; and (4) the three cases to be performed for RSI-1 that resulted in the greatest dose, and, if they are not the same, then the three cases that resulted in the greatest amount of water flowing through the saltstone. That information should include volumetric flow rates (or a comparable unit) coming from the various components of the disposal system, as shown in the cross section drawing below, and should be presented as figures and as tables as follows: (1) tables presenting the quantified flow components for the 100-, 1000-, and 10,000-year timesteps; and (2) water budget figures with a time axis and volumetric flow axis (or a comparable unit) similar to Figure 7.1-1 or 4.4-72 in the PA. The figures may require additional zoom-in figures due to the scale of flow in the disposal structure components.
- a. Where P = ET (from root zone) + lateral drainage (from SR + UB + MB + ULDL +
LB + LLDL + R + W including bearing pads + F + UMM + LMM) + vertical drainage (from LLM); and
- b. Where R HDPE/GCL vertical drainage = lateral drainage (from R + W including bearing pads) + vertical drainage (from W + SS + C).
- c. Component abbreviation:
C (column)
CAP HDPE/GCL (composite barrier in the cap)
EB (erosion barrier)
ET (evapotranspiration)
F (floor)
F HDPE/GCL (composite barrier in the floor)
LB (lower backfill)
LLDL (lower lateral drainage layer)
LMM (lower mud mat)
P (precipitation)
R (roof)
R HDPE/GCL (composite barrier in the roof)
RZ (root zone)
SR (surface runoff)
SS (saltstone)
UB (upper backfill)
ULDL (upper lateral drainage layer)
UMM (upper mud mat)
W (wall)
RSI-7: Erosion in Adjacent Area to the SDF To evaluate the stability of the SDF, the NRC staff needs information about potential future erosion in the adjacent area surrounding the SDF that may affect stability of the SDF.
Basis The PA did not include an assessment of erosion in the adjacent area surrounding the SDF, as associated with Monitoring Factor 2.02 (Erosion Control of the SDF Engineered Surface Cover and Adjacent Area) as described in the January 31, 2018, NRC TRR and later incorporated into the current NRC SDF Monitoring Plan (ADAMS Accession No. ML13100A113) by the June 29, 2018, NRC Letter (ADAMS Accession No. ML18107A161). Although there is no evidence of significant erosion currently in the area surrounding the SDF, existing gullies will be within 1,000 feet (305 meters) of the planned SDF and current conditions are representative only of the present and the near past. Different potential future changes to vegetation, infiltration, and erosion could impact future SDF performance differently and additional DOE analyses could reduce scenario and conceptual model uncertainty.
Path Forward Provide an evaluation of erosion in the adjacent area surrounding the SDF and the planned closure cap system. The evaluation should demonstrate that future precipitation rates and climate conditions will not endanger the structural integrity of the SDF due to future erosion.
RSI-8: Upper Three Runs Aquifer-Upper Aquifer Zone Lateral Flow Analysis To evaluate the effects of plausible alternative conceptual models, the NRC staff needs information about: (1) lateral flow in the Upper Three Runs Aquifer-Upper Aquifer Zone (UTRA-UAZ); (2) contaminant flow and transport in the UTRA-UAZ on top of the Tan Clay Confining Zone (TCCZ); and (3) the projected dose to a human receptor who uses water from the UTRA-UAZ.
Basis It appears that in the PA the DOE simulated contaminated water flow having a strong vertical component through the TCCZ and minimal lateral flow; however, recent information indicates that saturated flow near the SDF has a strong lateral component. The May 17, 2018, NRC TRR, Groundwater Monitoring at and Near the Planned SDF (ADAMS Accession No. ML18117A494) evaluated the SRS Z-Area groundwater characterization studies, including the DOE document SRNS-RP-2015-00902, Z-Area Groundwater Characterization Data Report (ADAMS Accession No. ML16057A135) that documented the lengthy flow and transport occurring on top of the TCCZ. The May 17, 2018, TRR also stated that the DOE should consider including that alternative conceptual model in the next revision to the PA.
In the July 15, 2016, OOV Report, the DOE confirmed that that the SDF saturated zone transport model simulated lateral flow and transport on top of the TCCZ. In the November 1, 2018, OOV Report, the DOE indicated that it had begun to consider the calibration of groundwater models of the SRS Z-Area to the SDS 4 plume, as was suggested in the May 17, 2018, NRC TRR.
Figure 5-4 in the DOE document SRR-CWDA-2018-00036, Evaluation of Soil and Groundwater Contamination from SDU 4 (ADAMS Accession No. ML20206L238) (see below) showed a proposed generalized conceptual model of migration of contamination from SDS 4 to Well ZBG 2. Most of the contaminated water moved laterally within the UTRA-UAZ and into the TCCZ.
Path Forward Provide an alternative conceptual model of contaminant flow and transport in the UTRA-UAZ on top of the TCCZ that includes either an assessment with the UTRA-UAZ being the source of water for the human receptor or a demonstration that water from such a source is implausible for a member of the public.