F Test for Natural Attenuation in Groundwater: Application on Benzene

Background/Need: Chapter NR 726, Wis. Adm. Code, and, more recently, ch. NR 746, allow closure of a petroleum case where ch. NR 140 groundwater enforcement standards are exceeded upon the demonstration that natural attenuation (NA) is an effective remedial option at the site. However, the demonstration of NA’s effectiveness may rely on only a few monitoring wells that may not be optimally placed at a site. So closure decisions may be based on analysis of inadequate or inappropriate data, inappropriate especially because the contaminant concentration data from monitoring wells, which are the primary evidence on NA, can be compromised by the fluctuation of the water table. The history of the water-table fluctuations can track changes in important physical NA factors (e.g. recharge, groundwater flow direction), as well as non-NA factors like volume of water purged from the well before a sample is collected. When the contaminant concentration is affected by the rise or fall of the water table, any conclusion about rates of degradation (and hence NA’s effectiveness) may be premature. We have developed an F test to statistically look into the importance of fluctuation in the water table. We propose to apply the test to petroleum sites in Wisconsin with adequate benzene and water-elevation data for analysis. We will look into the utility of the technique to screen sites where the statistical procedures required by rule to demonstrate NA may be inappropriate.

Objectives: To investigate the utility of an F-test in analyzing the statistical significance of including both groundwater elevation and time as predictors of benzene concentration. The conclusions of the test would be compared to the conclusions from the Mann-Kendall and Mann-Whitney nonparametric tests.