Straight answers about what a nitrate number means, what it doesn't, and what you can do — sourced, and careful about the difference between the legal limit and the evidence.
NitrateSmart reads nitrate levels for United States community water systems from the systems' own compliance reporting, standardizes them to one unit, and sorts each into a plain five-band scale. It is an interpretation layer over public data — not a laboratory, not a regulator, and not a substitute for testing your own tap. Its focus is the range a simple “over or under the legal limit” ignores.
Ten milligrams per liter, measured as nitrogen, is the enforceable United States maximum contaminant level — the level a community water system may not exceed. It was set in 1962 to prevent an acute infant illness (methemoglobinemia), not around a lifetime of lower-dose exposure. A result below 10 is within the legal standard; it does not mean there is nothing else to consider. More on the 10 mg/L break.
0.14 milligrams per liter as nitrogen is a protective reference point proposed by Temkin and colleagues, tied to a modeled one-in-a-million cancer risk — about 70 times below the legal limit. It is an evidence-derived reference point, not a proven-harm threshold: a value above 0.14 is not “proven unsafe.” It also sits within concentrations that occur naturally, so it cannot separate natural nitrate from human-caused. Why the span from 0.14 to 10 is the point.
NitrateSmart does not label water “safe” or “unsafe” — that is a personal decision, best made with a clinician if a pregnancy or an infant is involved. Below 10 mg/L means within the legal limit. Still below it, two long-term Iowa studies reported higher thyroid and bladder cancer among women whose supply stayed above 5 mg/L as N for years — 5 is a level that matters when it persists, and associations are reported at lower levels too. So a legal result can still sit in the range the health studies examined. What 5 mg/L means.
No. 0.14 mg/L is a modeled reference point, not a line above which harm is established. The health studies at higher levels are observational — they report associations, not proven cause, and the difference for any one person is small. Most of the country's water is above 0.14; that fact alone says nothing about proven harm.
Your system's most recent reported result places it in one of five bands: Natural background (0 to 0.99 mg/L as N), Human impact likely (1–2.99), Human impact clear (3–4.99), Health-study range (5–9.99), and Above the federal limit (10 and above). The bands describe how strongly human activity has marked the water — and, at or above 5, that it sits in the range where health studies report associations. They are not a safety verdict.
It is the same molecule in different company, and that context is the whole point. In vegetables, nitrate arrives alongside antioxidants such as vitamin C that steer the body's chemistry toward beneficial nitric oxide; in drinking water it arrives without them. So “vegetable nitrate looks healthy” and “water nitrate is studied for harm” are not a contradiction — the evidence is about nitrate in the water context, not a blanket verdict on the molecule.
This is the circumstance the research concentrates on. Infants under about six months are uniquely vulnerable to nitrate — it is what the federal limit was set around — and pregnancy is the other focus of the literature. If either applies in your household, take your system's number to a clinician, and do not boil the water to reduce nitrate.
No — boiling concentrates it. Boiling drives off water as steam and leaves the nitrate behind, so the level rises. Boiling is the right move for bacteria; for nitrate it makes the number worse.
Almost certainly not. Standard carbon filters — the media in most pitchers and refrigerator dispensers — do not remove nitrate. What does: reverse osmosis, ion exchange with a nitrate-selective resin, or distillation. Confirm a filter states nitrate reduction specifically. What actually removes nitrate.
Nitrite is a separate contaminant with its own federal limit (1 mg/L as N), and it is more acutely toxic than nitrate. In aerated tap water it is almost always a non-detect, because it oxidizes to nitrate — which is why nitrate is the signal worth reading.
Usually one of a few reasons, all about making systems comparable. The report may use nitrate as the ion while NitrateSmart uses nitrate as nitrogen (about 4.4 times smaller); it may show an annual average while NitrateSmart shows the most recent result plus the three-year range; or it may present a detection differently. If a number looks off, the units and the analyte are the first things to check.
At the interpretive level: every result is converted to milligrams per liter as nitrogen; a non-detect is treated as evidence, not as zero; the band reflects the most recent qualifying result, with the full three-year range shown alongside; and where a result reports a range, the high end is used. The reason for each choice is public; the machinery that applies them at scale is not.
It means the system has no nitrate result inside the current three-year window. The active backbone is 47,581 community water systems in the lower 48; 38,762 have an in-window result and are classified. Of the rest, 65 pages show their last reported result with its date — for reference, counted in no statistic — and the remainder show “no current data” rather than a guess.
Confidence describes how much testing stands behind a system's classification — how many results, and how direct they are — not how good or bad the water is. A high-confidence result rests on many finished-water tests; a low-confidence one may rest on a single reading or a transcribed summary. It is a measure of the record, not the water.
It is what your community water system reported for its supply, standardized — not a measurement at your specific tap. Population figures count people served by a system, not individual taps. For your own tap, or a private well, a certified-laboratory test is the only way to know.
Your address is used once, to find the community water system whose service area covers it. One deliberate record is kept — the address you typed, the address it matched, the system found, and the time — for 30 days, and then it is erased automatically. It is never connected to advertising, never built into a profile, and never sent to the AI that answers questions. The details are on the Privacy page. The result you see is the system's, not yours: everyone matched to the same system sees the same page. Matching is not perfect. Service-area boundaries are approximate in places, and an address near the edge of two systems can be matched to the wrong one. If the system named here does not match the utility on your water bill, trust the bill and look the utility up by name instead.
NitrateSmart is maintained as a standing interpretation of public compliance data, refreshed on a rolling basis and corrected when a record or a method is found wanting, with a dated change log. The classifications, boundaries, and sources are public and cited; the data-acquisition and normalization machinery is proprietary. For licensing or bulk data-access inquiries, get in touch.
Nitrate in community water systems, from the utilities’ own compliance reporting.
Results reaching back to July 2023 · current through August 9, 2026.
Educational information, not medical or personal safety advice.