Site Characterization – How Fair Has Your Site Been Sampled?

This is a critical question because everything that follows: remediation, funding, tax credits, legal ramifications, and schedule are all driven by what sampling finds. If not done properly at the outset, the shocks from an improper site characterization can reverberate throughout the life of your project. “A hand played badly is difficult to undo.”
New York State Department of Environmental Conservation Program Policy DER-10/Technical Guidance for Site Investigation and Remediation (DER-10) states that Site Characterization/Remediation sampling should be “biased toward locations expected to be contaminated . . . .” While this is helpful as an initial screening procedure, following this guidance exclusively during site characterization can lead to a skewed—and distorted—picture of the contaminant distribution. In short, you could end up spending a lot of money to remediate areas that are actually clean, or are already below the cleanup standards. Is this a sound foundation for undertaking costly remediation and deploying precious resources?
A better approach is to strategically include some random sampling in the site characterization plan. Why? Random sampling isolates contaminated areas from non-contaminated areas and protects against lumping both together, which would mean having to remediate uncontaminated material. Strictly following regulatory guidelines leads to sampling bias. Less contaminated areas—where the bulk of the contamination may occur—are underrepresented or not sampled at all. Only to be found unexpectedly when remediation is underway and you realize that the majority of your “contaminated area” is actually clean accept for localized areas where high concentrations exist.

The figure illustrates the effect of focusing only on the most contaminated samples and on delineation samples.
Delineating the contaminant limit incorporates only the samples below the cleanup goal (left of the left blue dashed line). Sampling the most stained soils with the highest PID measurements incorporates only 5% of the samples (to the right of the right dashed line). The remaining 30 percent of the samples—representing a large portion of the site— are excluded! No information on about a third of the site!
Do you want to base everything on 5% of the most highly contaminated samples and samples where there are no impacts? At FLS, we can help you with this most critical project component. We can develop site characterization plans that accurately assess your site and pave the way for remediation that is lower cost with less uncertainty. A few more samples upfront will serve as an investment that will pay for itself many times over down the project line. You would not purchase a car with just a photo of the glove compartment. Why would you base a costly remediation on only a fraction of the data that are biased high (or low)?