Non-Statistical Approaches....

Non-Statistical Approaches for Comparing Site and Background Data]

For most sites, a determination of whether site concentrations represent
background conditions can be made without using statistical tests. The basic approach is to define the upper end of the range of background concentrations as the lower of:
1) The maximum background concentration, or
2) Twice the mean background concentration.
The maximum concentration on site is compared with this upper limit on
background. If the maximum concentration found on site is less than or equal to this upper background limit, the chemical can be considered to be background and removed from further consideration in any risk assessment or site remediation decisions.
This approach has been used for decades and has widespread regulatory
acceptance. It is simple, conservative, and works with a limited number of background samples. When conducting this test, the following points apply:
• A minimum of seven background samples is needed (i.e., data from seven
different background locations).
• Both background and site samples should be discrete rather than composite
samples. Discrete samples are needed to identify the maximum background and
site concentrations, which are critical for this test.
• As noted above, comparisons should be made with equivalent soil horizons. In
general, data from different soil horizons should not be combined unless the
absence of concentration change with depth can be clearly demonstrated.
• For “non-detect” background samples, one-half the detection limit should be used in calculating the mean background concentration.
If site concentrations are above background, and background concentrations are
above risk-based criteria, cleanup to background levels may be warranted. In this
situation, the site-specific upper limit on background (i.e., the lower of the maximum or twice the mean background concentration) can be used as a not-to-exceed cleanup criterion. That is, removal or management of all concentrations above this value will be considered to have restored the site to background conditions with respect to this contaminant.
Another non-statistical approach involves a comparison of the 95% UCL of site
samples with the 95% UCL for background. This method has technical limitations that could, in theory, lead to misclassification of contaminants as background, or not background. The approach is accepted, however, while the Department evaluates its performance in practice. Consistent with requirements for using a 95% UCL in Chapter 62-780, F.A.C., a minimum of 10 samples is needed, of which seven must be above detection limits (or three above detection limits to use the Bounding Method). This minimum sample requirement applies for both site and background samples at each soil horizon. Also consistent with Chapter 62-780, F.A.C., the maximum concentration on the site should be less than, or equal to, 3-times the default CTL. The 95% UCL approach should not be used for chemicals with CTLs based on acute toxicity when exposure scenarios with children are plausible (e.g., residential land use, parks, schools).