Citation:

Singh, A. ESTIMATING THE EXPOSURE POINT CONCENTRATION TERM USING PROUCL, VERSION 3.0. Presented at Annual Meeting of the Society for Risk Analysis, Palm Springs, CA, December 15, 2004.

Impact/Purpose:

The overall objectives of this task are to: 1) provide ORD state-of-the-science technical support and assistance to Regional staff; 2) facilitate the evaluation and application of site characterization technologies at Superfund and RCRA sites; and 3) improve communication among Regions and ORD laboratories.

Description:

In superfund and RCRA Projects of the U.S. EPA, cleanup, exposure, and risk assessment decisions are often made based upon the mean concentrations of the contaminants of potential concern (COPC). A 95% upper confidence limit (UCL) of the population mean is used to estimate the exposure point concentrations (EPC) term, to determine the attainment of cleanup standards, to estimate background level contaminant concentrations, or to compare the soil concentrations with the site-specific soil screening levels. It is, therefore, important to compute an accurate and stable 95% UCL of the population mean from the available data. The formula for computing a UCL depends upon the data distribution. Typically, environmental data are positively skewed, and a lognormal distribution is often used to model such skewed data distributions. A positively skewed data set can quite often be modeled by lognormal or gamma distributions. However, due to computational ease, the lognormal distribution is used as a default model for positively skewed data sets. It is well known that the use of a lognormal model for an environmental data set unjustifiably inflates the minimum variance unbiased estimate of the mean and its UCL to levels that may not be applicable in practice. In this paper, we propose the use of gamma distribution to model positively skewed data sets. The objective of the present work is to study procedures which can be used to compute a stable and accurate UCL of the mean based upon a gamma distribution. Several nonparametric (e.g. the standard bootstrap, the bootstrap-t, Hall's bootstrap, and the Chebyshev inequality) methods of computing a UCL of an unknown population mean, , have also been considered. Monte Carol simulation experiments have been performed to compare the performances of those methods. A comparison of the various methods has been evaluated in terms of the coverage (confidence coefficient) probabilities achieved by the various UCLs. Based upon this study, recommendations have been made about the computation of a UCL of the mean for skewed data distributions originating from various environmental applications. Several parametric (e.g., based upon normal, lognormal, or gamma distributions) and nonparametric UCL computation methods with recommendations have been incorporated in to the EPA software, ProUCL Version 3.0. ProUCL will be used to demonstrate the computation of the various parametric and nonparametric UCL computation methods. Several data sets from real Superfund Sites will be considered.

Record Details: