Most nitrate in a public water supply is an agriculture story. The part worth understanding is the mechanics: how it travels, why it lingers for decades, and why the speed of the trip decides what else survives it.
Nitrate is the stable, fully oxidized form of nitrogen. It is a normal part of nature, but it is also very soluble and very mobile — which is exactly why it travels from a field into groundwater so easily.
On this page
Plants and soil hold on to some nutrients and let others slip through. Nitrate slips through. It is a small, highly soluble ion that does not cling to soil particles the way ammonium or phosphate do, so when rain or irrigation soaks past the roots, the nitrate goes with it [4]. That single property is why nitrate is groundwater's most stubborn farm contaminant while phosphorus stays more of a surface-water problem. It also means ordinary treatment can't touch it: chlorinating water kills microbes and does nothing to nitrate [4].
From the root zone, nitrate takes one of two paths. In most places it percolates slowly downward to the water table. But across the tile-drained farmland of the Corn Belt, buried perforated pipes drain the fields so they can be worked — and those drains are a fast express lane, carrying leached nitrate straight to the nearest stream in days rather than years [4]. Which path dominates decides whether a place's nitrate problem shows up in its groundwater or in its rivers.
High nitrate in groundwater takes two things at once: a lot of nitrogen going in, and a landscape that lets it through without cleaning it up. National USGS modeling finds the highest predicted nitrate where heavy nitrogen loading meets high recharge, well-drained soils, fractured or porous rock, and little natural removal [1]. Where those line up — the Corn Belt and the High Plains above all, with California's Central Valley close behind — nitrate concentrates [2].
The rock matters as much as the farming. Cave-riddled karst, like far-northeast Iowa's, sends surface water straight underground through sinkholes with almost no filtering. Coarse sand and gravel, like the High Plains' Ogallala, is well-drained and heavily irrigated. Both run high [5]. Deep, confined aquifers hold older water that soaked in before the fertilizer era, so they tend to stay low — which is also why shallow wells run higher than deep ones drawing from the same ground [1].
There is a process that removes nitrate: denitrification, in which microbes convert it back to harmless nitrogen gas. It needs two conditions together — low oxygen and a source of organic carbon. Where wet, carbon-rich sediments provide both, nitrate can stay low even under heavy farming; where they don't, it accumulates [1]. This is why two counties with the same crops and the same fertilizer can have very different water.
The nitrate in a well today may be from fertilizer applied decades ago. It moves down through the unsaturated ground at often less than a meter a year, so where the water table is deep the trip can take more than fifty years [4]. Hydrologists call this stored, in-transit load legacy nitrate. It cuts both ways: pollution keeps arriving after practices improve, and improvements take years to show up in the water. It is the plain physical reason a watershed can work at conservation and see little change for a long time — a fact about travel time, not about effort [4].
Nitrate that reaches rivers doesn't stop being a problem; it changes into a different one. In surface water it feeds algae, and when the algae die and decompose they strip the oxygen from the water, creating low-oxygen "dead zones." Nitrate leached and tile-drained from the Corn Belt travels down the Mississippi and helps form one of the world's largest, in the Gulf of Mexico, each summer; USGS source-tracking assigns the basin's highest nitrogen yields to the farmland centered on Iowa and Indiana [3]. Same source, two different problems: a health concern at the tap, an ecological one in the water.
The home page puts it briefly: an elevated reading establishes the pathway, not the passengers. This is the longer version, because the distinction is the difference between a defensible claim and an overreach.
Nitrate is unusually mobile. It carries a negative charge, so it does not bind to soil particles the way ammonium or phosphate do, and in oxygen-rich ground nothing consumes it. It moves with the water at very nearly the speed of the water [4]. That is exactly why it is the signal worth reading: it survives a journey most contaminants do not.
The things that started out beside it behave differently. Many pesticides and herbicides sorb to organic matter and degrade over weeks to months. Bacteria, viruses and protozoa are physically strained out by fine-grained material and die off with time and distance. Each of them has its own travel time, its own half-life, and its own set of conditions under which it stops.
One thing is common to all of those defences. Sorption needs contact with soil. Degradation needs weeks. Straining needs fine material to pass through. Die-off needs distance. Not one of those is instant. Every one of them is a rate — and a rate only removes what it is given time to work on.
So ground that lets nitrate through quickly is not simply ground where nitrate arrives. It is ground where everything else gets less of every defence at once. Fractured rock, karst, coarse sand and gravel, a shallow water table, a tile drain: each is a shortcut, and a shortcut subtracts time from all of those processes simultaneously [1][5].
That is the sense in which an elevated reading is worth more than the nitrate in it. It does not measure the others and it does not imply them. What it reports is that this water took a route fast enough for a mobile ion to survive — and the speed that carried the nitrate is the same condition under which the rest is likeliest to survive too.
So the honest reading of an elevated nitrate result is this: nitrogen from the surface reached this water, and a transport pathway exists between the two. Whether anything less mobile completed the same trip is a separate question that nitrate cannot answer, and only a test for those specific things can.
Where the ground is vulnerable enough, co-occurrence does get measured. In the fractured dolomite of northeastern Wisconsin, where thin soil sits over creviced rock and the transit time is short, microbial source tracking found both cattle and human signatures in wells alongside the nitrate [6] The mechanism above is general; that measurement is not. It is one aquifer, sampled once, and it says what happened there rather than what happens anywhere the nitrate is elevated. It should not be read backwards into a reading somewhere else.
Sources are drawn from NitrateSmart's science references. DOIs are given where independently confirmed.
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.