1. A composite test is said to have been done using 30 samples from around the site. In the letter from Agresource, a test result for one particular sample that had a high level of something is mentioned, but no individual test results are provided, only this composite.

2. Federal limits on PCBs are at levels higher than fractions of parts per trillion. It does not seem reasonable that a composite result of 30 samples would be the correct method for identifying any PCB hot spots within the composting area, and in fact would seem to prove insufficiently sensitive if less than most/all of the samples had PCB contamination. So a good question to ask the lab would be, to what level of sensitivity can your lab detect PCBs? Using this level of sensitivity, one can guess how many of the samples MIGHT have contamination AT the maximum acceptable level and still end up with a composite sample that showed background levels of PCBs but no problem.

3. I am a gardener and I do soil tests every year. The instructions on a soil test are: if you have any area that is atypical of the other areas, test it separately. In land disturbed by typical grading operations used to create suburban landscapes, I have seen tests come back with enormous acidity difference that in a normal, undisturbed soil would be highly unusual (7.4 pH vs. 4.8 in the same yard). If a composite had been done of these two types, the soil might well appear to be a typical soil from the area, or perhaps slighly more acid than typical. The composite, for purposes of remediation, would have been meaningless.

The point is: when you do sampling, you have to know what your goal is. If your goal is to identify and remediate problems, you do not do composite samples of an entire area. The correct method of sampling would be done by a specialist in environmental hazard remediation, not by an employee of a firm with a financial interest in finding the compost clean (since it might be liable if contaminated, as it re-sells Stamford compost).

One method of remediating lead, for example, is simply to raise the pH and add lots of organic matter. Thus, if several samples had unacceptably high lead levels, the mere addition of the other organic matter samples would dilute that lead level to an acceptable level.

Legal levels of a particular contaminant appear to be separate and apart from whether those contaminants e.g. have any phytotoxic effect on plants. I found one discussion of phytotoxicity from dicamba that was at a much lower level than the legally acceptable rate ppm of dicamba in the compost.

State levels and federal levels also may be distinct. Federal level limits for lead, for example, are higher than state levels in CT.