Lead 90th Percentile Above the Action Level

Rank all 60 results from lowest to highest. The 90th percentile is the value at rank 0.9 x 60 = 54, counting from the lowest - equivalently, the 7th highest value.

Exactly seven results are above 0.015 mg/L, so the 7th highest is 0.017 mg/L, which is above the 0.015 mg/L action level. The action level is exceeded.

Two things follow from the arithmetic, and both are the point of the question:

• The 0.092 mg/L result, on its own, triggers nothing. A single high sample from one home is not an exceedance; it is the distribution of the whole set that matters.
• If only six results had exceeded, the 90th percentile would be the highest result at or below the action level and there would be no exceedance. The margin here is one sample.

That last fact is why the operator's instinct - drop two results and the 90th percentile falls to 0.014 - is not a compliance strategy. You may only invalidate samples for documented protocol failures or ineligible sites, and the state makes that call, not the operator. The correct path, if you believe sites were ineligible, is to document it and request approval to resample, not to delete results and quietly recompute.

Required actions:

• Public education within 60 days using the mandatory language: health effects of lead, sources of lead in drinking water, what the system is doing, and specific steps customers can take.
• Continued monitoring on the standard schedule (every six months while the exceedance stands).
• Corrosion control treatment. The state will designate optimal corrosion control treatment; with no OCC designated and pH sitting at 7.4, the system must have treatment in place on the state's schedule (typically within six months of the end of the monitoring period) and then demonstrate it is optimal.
• Water quality parameter monitoring at the entry points and in the distribution system to show OCC is holding.
• Source water and lead service line evaluation so you know whether the lead is coming from the service line, the plumbing, or both.

On the outlier argument: the method requires a first-draw sample after at least six hours of stagnation with the aerator removed. Water sitting overnight is exactly what the procedure calls for, so those two samples are more likely valid than invalid. The rest of the argument is also weak evidence: pH 7.4 with no designated OCC is a genuine treatment gap, and the data are consistent with the treatment being provided. You can request state review of specific sites, but you cannot decide on your own that they do not count.

Difference from an MCL: an action level is not a maximum contaminant level. It is a treatment-technique trigger - there is no "do not drink" and no Tier 1 notice, because the water is not unsafe in the way that nitrate or E. coli make it unsafe. But it is a real public health finding that obligates the system to inform people and to fix the corrosion chemistry.

What the public education must do, within 60 days:

• Explain the health effects, especially for pregnant women, infants, and young children.
• Say what the system is doing and when.
• Give concrete steps: run cold water before drinking or cooking, use only cold water for formula, remove aerators and flush them, and use a certified filter if the system is offering one.
• Reach everyone, not just bill payers - renters, schools, and community organizations serving at-risk populations. Deliver it through bill inserts and direct mail at minimum, and post it where people will see it.

Tone: accurate and non-alarmist. Contrast this deliberately with a true MCL violation (e.g., E. coli or nitrate), where the notice goes out within 24 hours because the water is unsafe now and customers must change behavior immediately. Here the message is exposure reduction plus a credible plan, and your credibility depends on not having to walk it back later.