ARCHIVED REPORTS_2013_7 - Laserfiche WebLink

Home Browse Search

APPENDIX D STATISTICAL ANALYSIS METHODS FOR THE FORWARD LANDFILL Statistical analysis of water quality monitoring data collected at the Forward Landfill was conducted by Geo-Logic Associates (GLA) in accordance with protocols outlined in WDR Order No. R5-2003-0049 issued by the Central Valley RWQCB and Code of Federal Regulations, Title 40, Part 258 (40 CFR 258) and California Code of Regulations (CCR) Title 27, Section 20415. The statistical analyses were performed with historical statistical algorithms that recognize that groundwater quality data may be correlated through time (in contrast, non-historical statistics rely on the data not being correlated). The statistical analyses described below were performed with the Sanitas° statistical software program, which performs user-determined statistical analyses using methods recommended by the California State Water Resources Control Board (SWRCB) and the United States Environmental Protection Agency (EPA). TOLERANCE LIMITS In accordance with WDR Order No. R5-2003-0049, it was determined that an intrawell statistical comparison was most appropriate for all groundwater monitoring wells at the Forward Unit, while interwell statistical comparison were used for DMP wells at the Austin Unit. For the statistical comparisons, the tolerance limits method described in the USEPA's 2009 Unified Guidance Document, "Statistical Analysis of Ground Water Monitoring Data at RCRA Facilities" was used to establish concentration limits (CLs) for the Forward and Austin Units. A tolerance limit is a statistical method used to compare a single observation to a group of observations. For groundwater monitoring, a tolerance limit may be used to make a comparison between background and compliance data. The tolerance limit is calculated to contain all future observations, within a specified confidence. If any future observation exceeds the tolerance limit, this is statistically significant evidence of contamination. For the Forward Unit, intrawell tolerance limits were developed based on a 95% confidence (with a false positive rate of 5 percent)that future observations will fall below the limit. For the Austin Unit DMP wells, interwell tolerance limits were developed based on a 95% confidence (with a false positive rate of 5 percent)that future observations will fall below the limit. Statistical calculations are based on the procedures described in the USEPA's 2009 Unified Guidance Document(USEPA, 2009). The tolerance limit is defined as: PL=X+KS Where, the mean background observations (X) and standard deviation(S) of background observations are calculated for the raw or transformed background data and a one-sided (two- sided for pH)tolerance factor(K) is specified at the 95% confidence level. C:\2013-0021WppendixDStats.doc D-1 Geo-Logic Associates