SITE INFORMATION AND CORRESPONDENCE_FILE 2

Home Browse Search

Mr.Russell Chapin 1766 W.Monte Diablo Avenue Page 3 of 7 Point 4: The EHD had questioned the use of 0.0012 foot/foot as the hydraulic gradient in the model as opposed to the 0.015 to 0.020 ft/ft hydraulic gradient determined through a pumping test conducted by AGE. ATC chose the 0.0012 ft/ft based on matching model output with observed contaminant concentration distribution, and discounted the 0.015 and 0.020 ft/ft as possibly being influenced by the higher conductivity of pea gravel backfill in the former excavations. According to the Response Letter, ATC estimated the hydraulic conductivity and the recharge flux concentration, and calibrated the flow and transport model, by matching model output to the observed contaminant distribution and concentrations that were also used in determining the predominant flow direction. The EHD concedes that a well-defined model dependent on seven variables, for example, can match predicted output to observed data to tightly constrain or solve for one variable, if the other six are well characterized. The more variables there are that must be solved or estimated through matching output to observed data, the less precise each solution must be until there is potential for circular reasoning. Intuitively, the predominant flow direction appears to be constrained by the presence or absence of dissolved contaminants. In regard to the second point presented by ATC, soil sample depth records indicate that the former excavations locally were as much as 14 feet deep and, at the time of the pumping test, the depth to groundwater on the site was approximately 10 to 12 feet; therefore, water was probably present in some pea gravel. The boring logs for MW-4 and VE-6 show fine-grained soil from near surface to approximately 10 feet bsg and silty sand from 10 or 15 feet bsg to 25 or 30 feet bsg; no pea gravel was encountered. If the saturated pea gravel were supplying sufficient volumes of water to the silty sand between the two wells, then the hydraulic conductivity could be overestimated, and ATC may be correct. However, ATC has not put forward a way to quantify the effect, if any, of saturated pea gravel on the pumping test results. Therefore, the EHD believes the model should be run with the pumping test-derived hydraulic conductivity to evaluate the effects of the higher conductivity on the model response, in the event that the contribution from the saturated pea gravel was minimal. This would provide a test of the model sensitivity to this variable. Point 5: As stated in the 04 October 2004 letter, the EHD believes that engineered remediation has been the dominant contaminant mass reduction mechanism on the site. ATC responded by citing apparent contaminant concentration reductions in monitoring wells MW-3, MW-4 and MW- 5 that ATC interprets to indicate that natural attenuation has been a significant factor at the site. The EHD believes there may be several mechanisms acting to cause the observed reduction. MW-4 is likely to have been influenced by SVE and GWE; MW-3 could be influenced by source reduction by SVE and GWE and by changes of groundwater flow direction; MW-5, the most distal well, would be the most sensitive well to small changes of flow direction —an unresolved problem for the model as discussed above. Therefore, the EHD does not consider contaminant trends in these three wells as definitive evidence for natural attenuation. Regarding this point, please note that the EHD has not maintained that natural attenuation was not occurring at the site, but has pointed out that active remediation has had an obviously discernable effect on contaminant concentrations, while the portion of the contaminant reduction attributable to natural attenuation has not been demonstrated with existing data. In neither the closure report nor the response letter, has ATC clearly demonstrated that natural attenuation is active, nor, more importantly, has the rate at which natural attenuation is reducing the contaminant mass been estimated.