I'm finding it interesting to reason about this subject, not being a geologist and what not. It seems that there is a relatively stable rate at which the farms output nitrates regardless of how much water Amazon pumps from the aquifer. The issue RollingStone seems to have identified is that as Amazon pulls more water, it effectively concentrates the solutes by evaporating some of the water. As a result, less water going into wastewater treatment, so the pollutants are more highly concentrated. But the total amount of contaminants is unchanged, right? Unless Amazon's water consumption somehow increases the solutes output by the farms. Not sure how that would work, but it doesn't seem impossible.
Do you two think I've framed this issue correctly?