CHERRY PICKING VERSUS THE WHOLE PICTURE

Often a dishonest informant can pick individual unrepresentative examples to try and prove a falsehood. For example hypothetically a dictatorship murders 50 political opponents and critics per year. The dictatorship is overthrown and after the new democratic government is installed politically motivated murders plummet to four over the course of the next year. A dishonest critic could scream to the heavens that the new government are routinely involved in murdering critics/opponents and yet an honest appraisal would clearly show a cause and effect whereby murders are dramatically reduced. Four murders need to be investigated absolutely but to routinely assign responsibility to the new government as with the old is inaccurate.

In matters scientific or hydrogeologic the expression is cherry picking whereby either a proponent or a critic carefully choose individual samples from a large group and then assign conclusions incorrectly to the whole group based upon a very small sample. Last week I picked three wells with concentrations of MBT over a twenty year period in Chemtura's south-west corner. All three wells showed increases and decreases but very little if any overall downward trend. This was not cherrypicking for two reasons. Firstly I expressly stated that there were numerous wells that did have decreasing trends. Secondly my point is not to assign the results of these three wells to all the others either in the south-west corner or on the whole site. My point was to clearly indicate the different behaviour of these wells to those with decreasing concentration trends.
It is my understanding and belief that downgradient groundwater contaminant concentrations are not homogeneous. Downgradient wells are screened both in different aquifers as well as at different heights within the same aquifers. Furthermore DNAPL chemicals do not automatically sink in their entirety through aquifers and then congregate in one single pool sitting on the surface of the low permeability surface or aquitard. As a result multiple separate groundwater plumes can theoretically result from one DNAPL chemical being released at the surface. Finally in the case of a couple of pools of free phase DNAPL connected by a trail(tail) of residual DNAPL; the residual DNAPL which is simply a tiny quantity of DNAPL within the pore spaces between the grains of sand or gravel will dissolve much faster than the pools of DNAPL. Hence for example a plume from a higher elevation of residual DNAPL will have decreasing concentrations much sooner than a plume emanating from a pool of free phase DNAPL. This is the point which I wished to send in my posting last week. Within the same geographic area such as Retention Pond West 6 (RPW6) one can have wells with decreasing concentration trends over time as well as wells which do not. One can not assign the behaviour of the wells with decreasing trends to all the others by saying "aha there isn't free phase DNAPL present". On the other hand it is legitimate to say that probably there isn't free phase DNAPL up gradient of those wells whereas there probably is free phase DNAPL upgradient of these other wells without decreasing concentration trends.