Based on your work, at what point does drawdown significantly disrupt equilibrium and adversely impact sample results?

From a practical standpoint, the biggest concern generally tends to be dewatering the well screen, if that happens we are changing the conditions.  The worst possible thing you can do in terms of getting a groundwater sample is dewatering the well screen and then sampling on recovery. But if you pump the well with maintained pumping conditions, even if you dewater a portion of the well screen that water entering down below is essentially entering the well screen from the formation and will not be significantly altered from the formation water chemistry.  So as a practical limit, if the well screen is submerged try to keep it submerged.  But, if the water level starts out being just above the well screen or down within the well screen before you start, just try to minimize the drawdown as much as possible and get stabilized pumping water up as quickly as possible.  It’s really important to note something here: there is a carrot and there is a stick.  The stick approach would be to say this is your drawdown limit, it’s an arbitrary number but you can’t exceed it. Under those conditions you try to reduce your flow rates so you don’t exceed that number, but in low yield wells that may not be practical.  The carrot is that the quicker you get to a stabilized pumping water level the more quickly you get to stabilized indicative parameter conditions.  And then you will get samples much more rapidly.  In essence, there is really no numerical limit to drawdown. On a practical basis, you don’t want to dewater the well and wipe out all your sample volume and you try not to drawdown to the screen zone if you can avoid it.  But I am willing to accept a sample where drawdown goes below the top of the screen provided that pumping conditions are maintained all the way through the purging and sampling process.  We don’t want that well to recover and then sample the recovered water.