Wow- that has to be one of the very best pieces of science journalism I’ve read in a very, very long time. My hat is off to you, Kevin Folta- your analysis of the likely causes is spot on, and your analysis into the motivations and meaning of the over-arching problem this kind of study represents is well argued and very credible.
As someone who did groundwater and wastewater analysis in an effort to develop new water treatment technologies in an earlier part of my career, I can tell you that you’re dead right about the difficulties in making meaningful measurements like this. The tendency to read beyond the accuracy of the analytical data, or below what SHOULD really be considered the detection limit, is very strong indeed. We always said that the compound we had the hardest time treating was the analytical artifact- compounds that just weren’t in the water at all, or were there but in such minute concentrations that no reliable trend in the noise could be obtained.