A headline that drinking-water PFAS may be underestimated invites an easy reaction: the number on the lab report is too low. That reaction may be right, but the mechanism matters before anyone revises a table. Circle of Blue coverage of a new study, dated October 23, 2026, points to concentrations that "fall below laboratory reporting limits". Results in that range are hard to read as a true zero, so the useful question is not only what the lab reported but what the lab was able to see, and what it checked along the way.
Two kinds of low
A reporting limit is a laboratory's stated floor for a reliably quantified value. A result under that floor may appear as a non-detect, a qualified estimate, or a number carrying a flag, depending on the method and the laboratory's reporting convention. None of those labels says the sample contained nothing. They say the laboratory did not quantify the compound at the level it committed to report. That is a reporting decision layered on top of a measurement decision, and the two should be read separately.
Sample-specific limits add another layer. When a sample is diluted to bring an interfering matrix into range, the effective reporting limit rises for that sample. A utility comparing a dated result against a published limit may be comparing two different floors without realizing it. The sample report should state the limit that applies to each compound in that specific sample, not only the method limit printed on the front page.
What recovery tells you
Recovery is the fraction of a known spike that the method returns after the sample has passed through extraction, cleanup, and analysis. Low recovery for a compound means some of it was lost along the way. Losses can occur at container walls, on solid-phase cartridges, during incomplete elution, or through suppression in the instrument. A low recovery makes a non-detect less trustworthy, because any signal that would have appeared may have been removed before measurement.
The practical step is to request the spike recovery data for the batch that produced the result, not just the summary table. Recovery acceptance ranges vary by method and by laboratory, so the report should state both the observed recovery and the range the laboratory applied. A result that passed a broad acceptance window may still warrant a closer look if it sits near the edge for a compound of concern.
What blanks tell you
Field blanks, trip blanks, and laboratory method blanks test for contamination introduced by sampling equipment, shipping, or the laboratory itself. Some sampling materials and laboratory consumables have been identified as possible sources of background signal, so blank results matter in both directions. A detection in a blank can justify a qualifier on nearby results. A clean blank supports the sample, but only for the materials, steps, and dates the blank actually covers.
Ask which blank types were run, how frequently, and whether each was matched to the sampling event on the same date. A single method blank for a large batch says less about a particular field visit than a field blank collected on that visit does. The record should make those links visible.
Build the request before the number arrives
A reader or a utility reviewing a low PFAS result can set the request list in advance:
- The reporting limit for each compound, and whether it is the method limit or a sample-specific limit adjusted for dilution.
- Spike recovery values and the acceptance range the laboratory applied to the batch.
- Blank types, frequency, and results matched to the sampling date.
- Whether qualified estimates are counted as detections or excluded from any published total.
Setting these terms before the result arrives keeps the review tied to the record. It also prevents a common shortcut, where a single headline total is treated as a measurement of a specific tap.
Where the study headline stops
A headline about underestimation indicates a direction, not a size. Whether a given sample is underestimated, by how much, and for which compounds depends on the methods and quality records behind it. Trade readers should treat a national claim as a reason to check their own file, not as a correction to apply to every result. The claim should be kept to what the study measured and what the source reporting states, and no further.