FIELD DOCUMENTS - Laserfiche WebLink

Home Browse Search

Quarterly Monitoring Report- Third Quarter 2015 Page 7 ' OAK RIDGE WINERY WKA No. 6364.08 August 25, 2015 20415(e) (8) (A)-(E). The purpose of the statistical analysis is to determine whether there has been a "measurably significant' change in the constituents of concern versus background ' concentrations in groundwater. ' The following presents the results of the statistical analysis using Analysis of Variation (ANOVA) as specified in the referenced section of CCR Title 27. Parametric ANOVA evaluation of the data was conducted using the Single Factor ANOVA function built into the Microsoft' Excel° spreadsheet software package. The input data and output results of each evaluation are included in Appendix E. Table E-1 of Appendix E presents the input values and Tables E-2 through E-9 present the output results from the evaluation. In general, the parametric ANOVA analysis indicates there are significant differences in the ' means of four of the six data sets. The ANOVA evaluation of pH and nitrate indicate there is no measurably significant variation in the means. It should be noted that the nitrate significance test is trivial because the input values are based predominantly on one-half the laboratory ' reporting limits. Nitrate was not detected at a concentration exceeding its laboratory reporting limit in each sample analyzed. The table below presents F-critical and F, which when compared, indicate whether or not significant (non-random) difference in the means of the data exists. ' NH3 K NO3 pH TKN TDS Log Log [TDS] [TKN] ' Data F-crit 2.65 2.65 2.65 2.65 2.64 2.65 2.64 2.65 Sets F 18.03 68.14 2.21 1.55 44.51 15.39 200.49 59.94 ' Notes: NH3 —Ammonia pH — log [H+] ' K— Potassium TKN —Total Kjeldahl Nitrogen NO3 - Nitrate TDS —Total Dissolved Solids ' Where F is greater than F-crit, there likely is a significant difference in the means. As shown, F is greater than F-crit in six of the eight evaluations. F does not exceed F-crit in two of the data ' set evaluations (Nitrate and pH). Because groundwater constituent concentrations typically are not normally distributed, the data ' sets for TKN and TDS were also transformed using a logarithm (logio) function and evaluated using ANOVA. The transformed data do not, upon inspection, appear to have been significantly affected by the transformation; therefore, the untransformed data appear to be sufficiently close ' to a normally distributed data set for a parametric ANOVA evaluation.