>> and at that level, no health impacts have been demonstrated

But again, absence of evidence is not evidence of absence.

The fact is, our ability to detect harmful health effects, particularly over the long term, is still in its absolute infancy. There is absolutely no reason to declare things safe until proven unsafe, particularly, particularly when you look with a historical lens.

The damage of a chemical will always be greater than what we can measure, because science (measuring blood levels of tens of thousands of people with mass spectometry and correlating against long-term health events like cancer) is incredibly hard, expensive, and slow.

As a thought experiment, suppose a chemical released in the water increased emotional irritability by 1%, or ADD by 2%, or reduced sperm count by 3%? What are the odds that we'd be able to conduct a study sensitive enough to prove the link? And what are the odds that anybody would happen to guess that correlation existed and choose to look in the first place?