|
R. Michael Turnipseed, Director Allen Biaggi, Administrator (775) 687-4670 TDD 687-4678
Administration |
STATE OF NEVADA KENNY C. GUINN Governor 
|
Waste Management Corrective Actions Federal Facilities Facsimile 687-6396
Air Quality |
|
|
Division of Environmental Protection 333 W. Nye Lane, Room 138 Carson City, Nevada 89706-0851 |
|||
|
Ms. Runore C. Wycoff, Director Environmental Restoration Division National Nuclear Security Administration Nevada Operations Office P.O. Box 98593-8518 Las Vegas, Nevada 89193-8518 RE: Evaluation of the "Data Decision Analysis: Central Nevada Test Area, Corrective Action Unit 443, submitted by the National Nuclear Security Administration, March 7, 2001 Dear Ms. Wycoff: The Nevada Division of Environmental Protection (NDEP) has reviewed the National Nuclear Security Administration Nevada Operations Office's (NNSA/NV) submittal of the Data Decision Analysis: Central Nevada Test Area, received March 7, 2001. The Data Decision Analysis (DDA) was prepared to address NDEP concerns over the uncertainty associated with predictions made using the current CNTA flow and transport computer model. A report on the model was submitted to NDEP in the September 1999 Desert Research Institute publication, "Evaluation of Groundwater Flow and Transport at the Faultless Underground Nuclear Test Area, Publication 45165". The purpose of the DDA was to quantify model uncertainty and assess which site characterization activities would effectively reduce uncertainty in a cost-effective manner. The DDA used a statistical approach to quantify model uncertainty, which began with the selection of the six most uncertain parameters as identified by a sensitivity analysis. Parameter uncertainty was described by calculating prior statistical distributions for each parameter. Ms. Runore C. Wycoff, Director Prediction uncertainty was determined by sampling from these distributions and running predictive simulations using a two dimensional version of the model. Total uncertainty in the predicted contaminant boundary radius was calculated by performing over 200 realizations using the prior distributions. The results suggest that although there is considerable uncertainty in the input parameters, the uncertainty in the prediction of the contaminant boundary, within a 1000-year time frame, is relatively low, with the prediction error calculated at less than 100 meters. Through successive simulations, large changes in the six uncertain input parameters result in only small changes in the location of the predicted contaminant boundary. The report stipulates that this is due primarily to low transport velocities in the immediate vicinity of the contamination source. The presence of low hydraulic conductivity volcanic rocks surrounding the Faultless cavity is thought by NNSA/NV and its contractors to play a major role in the limited movement of radionulcides away from the site. NDEP understands the conclusions presented in the DDA, however, NDEP has some concerns. Though the DDA constitutes a partial verification of the flow and transport model, NNSA/NV has not presented a model that has been fully verified by an independent set of data. Based on the information presented to date, NDEP will concur with the model, conditioned on a commitment for future model verification. This will allow for the CAI process to move forward to the "Define/Negotiate CAU Boundaries" task. Available data and information support model acceptance at this point in the investigation and provide a margin of confidence for allowing the process to continue. Supporting factors include: 1) CAU 443 involves only a single detonation and this CAU encompasses a small withdrawal area. 2) The hydrogeologic conceptual model is simple. Site-specific information suggests that the higher hydraulic conductivity welded tuff unit is not present in the immediate region surrounding the shot cavity. 3) Model uncertainty, a fundamental concern of NDEP, has been investigated and quantified in the DDA in terms of the model-predicted tritium contaminant boundary. As stated above, the results suggest that though input parameter uncertainty is large, the range of model prediction uncertainty is small. 4) Detonation-specific data from cavity infilling measurements taken during a period of over 20 years, strongly suggests slow advective movement potential away from the shot cavity. Water level recovery has been recorded in post shot well UC-1-P-2SR on a regular basis since shortly after the detonation. The slow recovery rate suggests that in the near field, low hydraulic conductivity (K) materials surround the shot cavity. These low K tuffaceous sediments will likely limit the rate of movement of radionuclides away from the shot cavity, once chimney infilling has ceased (currently estimated at 80 to 100 years post detonation). Cavity inflow is further discussed in Section 2.5.3 and Appendix 3 of the modeling report. 5) Though true risk is difficult to assess, in the near term, there is relatively no additional risk posed by making a decision at this point to proceed with the corrective action process. Additional data from within the reasonable vicinity of the shot cavity, which is expected to be collected during validation in the not-too-distant future, which contradicts the current thinking, would immediately trigger a revisit of the model and revision of conceptual model assumptions. This fact, combined with the observations presented in #4 above, provide an acceptable level of risk in making this regulatory decision to move forward. 6) This CAU presents relatively low risk to human health and the environment at this isolated site with gradients to the north and no surface water bodies nearby. NDEP will authorize the corrective action process to move forward, with an accelerated emphasis on the formulation and execution of a model validation plan. NDEP hereby approves of the Corrective Action Unit Model. NNSA/NV can proceed to proposing contaminant boundaries, pursuant to the following conditions: 1. NNSA must create a validation plan, which addresses conditions downgradient of the Faultless cavity, north of ground zero. The validation plan may be developed in conjunction with the monitoring plan. If the Corrective Action Decision Document concludes that monitoring is the appropriate course of action, dual-purpose monitoring wells (model validation and long-term monitoring) should be thoroughly evaluated. 2. The validation plan must include clearly defined trigger mechanisms for revisiting the model; such as the presence of tritium above background levels in downgradient areas or a significant thickness of the high hydraulic conductivity densely welded tuff at shot cavity depths. Key elements of the validation plan, including the particulars of trigger mechanisms, should be discussed with NDEP as development of the plan progresses. NDEP expects NNSA/NV to begin the process of defining the CAU contaminant boundary by performing the necessary model runs. Prior to engaging in the formal "Define/Negotiate CAU Boundaries" task, NDEP expects NNSA/NV to submit predictions of contaminant migration in the 5, 10, 20, 50, and 100-year time periods. This interim deliverable will assist both NDEP and NNSA/NV to prepare for the negotiation phase of this task with a realistic frame of reference. Following this submittal, NNSA/NV can then prepare its best estimate of the contaminant boundary. NNSA/NV will also begin concurrent preparation of a combined CADD/CAP, which will include a validation plan and monitoring plan. The CAU 443 CADD deadline had been held in abeyance until the groundwater model was reviewed. The CAP deadline was to be determined after the CADD was completed (letter Liebendorfer to Wycoff, September 10, 1999). At this point, a new deadline must be established for the CADD. Within 45 days NNSA/NV needs to submit, 1) a proposed milestone for the submittal of the contaminant boundary and 2) a proposed milestone for submission of the CADD/CAP. If you have questions regarding this matter please contact Sigurd Jaunarajs at (775) 687-4670 (ext. 3030), or me at (775) 687-4670 (ext. 3039). PJL/KKB/REN/SJ/js cc: Peter Sanders, ERD, NNSA/NV Patti Hall, EM, NNSA/NV Ken Hoar, ESHD, NNSA/NV L.J. Gabriel, DTRA Frank Di Sanza, WMD, NNSA/NV Michael McKinnon, NDEP/LV Matthew DeBurle, NDEP/CC |
|||
