ISM provides an estimate of the mean contaminant concentration in a defined volume (area and depth) of soil. ISM is particularly useful when practical constraints (e.g., budgets) limit the number of discrete samples that can be collected and therefore limit the precision with which the mean concentration in heterogeneous matrices may be estimated. As discussed in Hyperlink 1, most action levels are derived from risk-based receptor models which assume a specific exposure area (i.e., exposure unit). Therefore, estimates of mean concentrations in volumes of media are generally the appropriate statistic to compare to action levels.

However, it is important to match the project objectives with the type of sampling employed. For some objectives, discrete sampling is appropriate (when sufficient numbers of discrete samples are used); for other,s ISM sampling is the best option. In certain situations, sampling designs consisting of combinations of discrete and ISM samples may be advantageous. For example, discrete samples might be used to make decisions on obviously contaminated volumes of soil in which contaminant concentrations are very likely to exceed action levels. Even though contaminant concentrations in this situation may be highly variable, this variation would not result in decision errors since any possible sample collected from the volume will likely have contaminant concentrations above the action level. Discrete samples may also be used to estimate the variability within a DU prior to ISM sampling. When field analytical methods (or other cost-saving analytical approaches) are available, sufficient numbers of discrete samples may be used to characterize some contaminants or DUs, while ISM may be appropriate for those contaminants for which these analytical approaches are not available.