ARCHIVED REPORTS_XR0013078 - Laserfiche WebLink

Home Browse Search

a� LW 15 June 2009 \.W AGE-NC Project No.05-1306 Page 16 of 33 closure request. Tests are limited to those wells that contained measurable contaminant levels and therefore not ` conducted on wells indicating historical non-detect levels.In addition,the tests were performed only on wells that had the minimum number of consecutive rounds of detectable contaminants to properly perform each test. Well MW-2 was removed and therefore was not included in the statistical tests. Tests were not completed for wells including MW-1,MW-2r,MW-10,MW-11,and MW-12.Wells with an acceptable amount of data to perform the statistical test include MW-3, MW-4, MW-5, MW-6, and MW-7. These wells represent both on and off site locations. It should be noted that this site generally lacks BTEX constituents such that these compounds have been rarely detected as a whole, and only in select monitoring wells. Therefore, statistical analysis �- could not be performed relative to these constituents on all monitoring well data sets,which serves to further highlight this site as a low-risk when considering the constituents and levels of those constituents that remain at residual levels. A brief discussion of the Mann-Kendall analysis procedure is described in Section 6.1 and the Mann-Whitney test procedure is described in Section 6.2. Input data and modeling assumptions are '' L■,r described in Sections 6.1.1 and 6.2.1 respectively. The Results are presented with a brief discussion in Sections 6.1.2 and 6.2.2 respectively. V 6.1. MANN-KENDALL TEST LM The Mann-Kendall test is a non-parametric test for identifying trends in time series data and compares the relative magnitudes of sample data rather than the data values. One of the benefits of .L. utilizing this test is that the data do not need to conform to any particular distribution. Also, data reported as non-detects can be included by assigning them a common value that is smaller than the smallest measured value within the data set. For purposes of the Mann-Kendall test, all non-detect values were assigned a single value that is less than the detection limit and we arbitrarily used a value that is '/z of the detection limit consistently when applying data to this test. A— The Mann-Kendall Test can be used with a minimum of four rounds of sampling results; however, the Mann-Kendall Test is not for data that exhibit seasonal behavior. The data values are evaluated as an ordered time series with each data value being compared to all subsequent data values. The �- initial value of the Mann-Kendall statistic, S,is assumed to be 0(e.g.,no trend).If a data value from a later time period is higher than a data value from an earlier time period, S is incremented by 1. If the data value from a later time period is lower than a data value sampled earlier, S is decremented by 1. The net result of all such increments and decrements yields the final value of S. A very high value of S is an indicator of an increasing trend,and a very low negative value indicates a decreasing v Advanced GeoEnvironmental,Inc. i..