ARCHIVED REPORTS XR0011755 - Laserfiche WebLink

Home Browse Search

Evaluation of Natural.Attenuation: 7500 West Eleventh Street, Tracy, CA. Page 47 In, wells associated with its surrounding Delauney triangles. If that analysis shows that the concentrations of analytes of concern at a given location can be estimated, with a high degree of confidence, from the known measurements at nearby wells, the well being analyzed is selected as a candidate for removal from the well array because such action could be taken without degrading the database for the plume as a whole to a statistically significant degree. The results of the MAROS well redundancy analysis is shown in tabular form, "MAROS Sampling Location Optimization" on that sheet in Appendix A. Only one well, MWFP-1, is recommended for elimination from the monitoring well Array. This result demonstrates that the well array design implemented by SJC is well designed and provides statistically- reliable data for characterization of the conditions of the plume of contaminated groundwater. It does indicate that MWFP-1 may be redundant primarily because of its proximity to Monitoring Well MW-7; however, we do not recommend physical elimination of this well because it serves as a sentinel that can warn of any unanticipated migration of the LNAPL that has, from time to time, been detected in MW-7, which is located on the axis of the plume a short distance up-gradient from MWFP-1. - 8.2.2.4 Optimal Sampling Frequency To permit design of an efficient groundwater-quality monitoring program, the MAROS protocol computes optimal sampling frequencies for each well in the well array for each of the critical analytes of concern. The method employed by the software is based on the Cost-Effecting Sampling (CES) methodology developed by Lawrence Livermore National Laboratory(LLNL) (Ridley, et al 1995). The CES method employs three steps for determining sampling frequencies. In Step 1, the rates of change of concentrations of an analyte in recent samples from a given well are examined and the rates are divided into four categories. The lowest rates of change { are those that range from zero to 10 pg/L per year. The highest rates category is associated with a change of 30 pg/L per year. The rates of change between those two extremes are quantified by variability information, with higher variability requiring higher sampling frequency. Variability is characterized by a distribution-free version of the coefficient of variation, whereby the data range over the time period considered is divided by the mean concentration during that period with 1.0 as the cutoff. Step 2 adjusts the proposed sampling frequency based on the overall trend of the sampling data. If the long-term history of change is significantly greater than the recent change, the sampling frequency may be reduced by one level. If this is not the case, the proposed sampling frequency is left unchanged. Finally, in Step 3, sampling frequencies are further adjusted based on risk. Because not all compounds that may be present in groundwater are equally harmful, sampling frequency is reduced by one level if recent maximum concentrations of compounds of high risk (e.g., benzene) are less that one-half of the applicable maximum contaminant level (MCL). sic -1 , J