SITE INFORMATION AND CORRESPONDENCE

Home Browse Search

PDDR • -3- • 31 March 1995 DDRW; Tracy and the Depot Area on a relatively smaller scale (Figure 4-3). However, it is unclear how discretization was established at the boundaries and how the scale of the discretization was refined in the area of interest during model development to reflect available field data. Apparently, the area of interest was discretized into rectangular cells in horizontal planes which were applied in each of the model layers in the same manner. It is unclear whether or not this approach was appropriate and adequate. The PDDR must substantiate and illustrate discretization of the area of interest with respect to field data and describe boundary condition effects. In addition, the PDDR must include a map illustrating the boundary and initial conditions in the area of interest with respect to ground water flow in each horizon. This map should be provided on a relatively small scale and used to illustrate the initial grid system and it's refinement. 4. Sensitivity Analysis. The PDDR provides a general discussion of a sensitivity analysis that was apparently conducted after the model was calibrated. As discussed in our 22 September 1993 comment letter- specific comment 3, sensitivity analyses should have been performed during calibration of the model to quantify the uncertainties associated with the model parameters. The PDDR should provide an appendix of the details from the calibration history in a table, including any sensitivity analyses, used to compare these uncertainties to the model and available field data. EVALUATION OF THE REMEDIAL ALTERNATIVES 1. Specific Objectives. Section 5.1 of the PDDR presents the design objectives used to evaluate four modeling scenarios. Four modeling scenarios were simulated with the intent of developing potential remedial design alternatives to meet these design objectives. However, these objectives are not consistent with the requirements to demonstrate whether or not it is technically and economically feasible to clean up the contaminated aquifer on and off base to background concentrations, specifically to non-detect (ND) levels for TCE and PCE. The design objectives, as presented, appear to focus on providing hydraulic containment versus capture and cleanup of the plume. 2. Model Scenarios. The model scenarios appear to assume certain physical constraints associated with potential locations for remedial wells. However, the model scenarios should be used to optimize the locations and pumping capacities of these wells. This optimization should then be used as a tool to assist in selecting the remedial well field. 3. Costs Associated with Hydraulic Capture versus Time Required for Clean Up. The technical and economic feasibility associated with hydraulic capture of the plume to background levels, versus levels greater than background, or MCLS (capital costs) must be evaluated. A separate analysis should be performed to evaluate the feasibility and time required to clean up ground water within the capture zone(s) (operation costs) to background levels versus levels greater than background, or MCLS. 4. Model Limitations. The limitations of the ground water models, as discussed in Sections 4.6 and 10.0 of the PDDR, to accurately predict cleanup times, do not automatically become effective when MCLS are achieved. These limitations should be ". . . considered in decision-making based on the model predictions." as indicated in Section 4.6.0.5 of the PDDR. The uncertainties associated with the model predictions may be equally applicable to all model predictions. Therefore, it seems inappropriate to extend the clean up time for cleanup alternative 4B (alternative focused on containing and treating the combined TCE/PCE plume to detection limits) by some calculated amount for the