Heterogeneity of Contaminants in Solid Waste Materials Sampled for Compliance Testing

Abstract

In legal environmental control, solid wastes are sampled to determine whether the mean concentrations of contaminants meet the regulatory standard. Current sampling and statistical testing procedures for solid wastes often rely on assumptions about the heterogeneity of the contaminants and on the assumption that the frequency distribution of the data is (log)normal. To investigate the reliability of these assumptions, we analyzed the data of nineteen waste materials. Our results provide evidence that the variability of contaminant concentrations in solid wastes is large and difficult to predict. The concentrations followed different types of frequency distributions, and outliers occurred frequently. Large differences in concentration between the sub-lots were observed, and the concentration in a composite sample sometimes deviated considerably from the mean of the individual samples. The contaminant type did not seem to have a large effect on the coefficient of variation, the bias of the composite, or the frequency distribution type. Sampling standards and compliance criteria are often not adapted to the quality of the data encountered in practice. Regulatory standards should provide some prescribed level of statistical assurance that the standard is actually being met or violated. For compliance testing it seems better to measure uncertainty directly from sample results than to rely on assumptions.