Farm nitrogen balance

Why must the nitrogen that does not leave in the crop turn up somewhere unwelcome?

A farmer spreads nitrogen fertiliser and harvests a crop. Ask him where the nitrogen went and the natural answer is "into the grain". Ask what happened to the rest and the answers get vaguer: it was used up, it was taken up by soil life, it broke down.

But nitrogen is an element. It has no "used up" state. Whatever fraction did not leave the gate inside the harvest is still, atom for atom, somewhere — and the question is not whether it went somewhere but where. What kind of reasoning lets you say something confident about a substance you cannot see, on land you cannot dig up, without measuring anything?

Begin with the only premise we really need: nitrogen atoms are not created or destroyed by farming. Chemistry rearranges them — nitrate to gas, ammonium to protein — but never abolishes them. So over any period and any boundary you choose to draw, what came in must equal what went out plus what accumulated inside.

Draw the boundary at the farm gate and the accounting becomes concrete. What comes in? Fertiliser, manure and feed brought onto the farm, nitrogen fixed biologically by legumes, a modest amount deposited from the atmosphere, a little dissolved in irrigation water. What leaves in a way anyone is happy about? Grain, milk, meat, fibre — the products that are sold.

Now ask the awkward question: is the difference zero? Almost never. Across the world's cropland, the share of applied nitrogen that ends up in the harvested crop — nitrogen use efficiency — has commonly been estimated at around half, and in many intensive systems considerably less. So typically half the input is unaccounted for by the harvest. Where can it be?

Only two places are logically available: still inside the boundary, or across it by some route other than the truck. Take the first. Soil organic matter really can hold nitrogen, and a farm that is building organic matter genuinely is storing some. But notice the constraint that imposes: soil organic matter has a fairly stable ratio of carbon to nitrogen, so you cannot bank nitrogen indefinitely without banking carbon alongside it. A soil at steady state — which most long-cultivated soils approximately are — has nowhere left to put it.

So the remainder crosses the boundary. By what routes? Nitrate is highly soluble and, unlike phosphate, is not held by soil particles, so water moving downwards carries it to groundwater and water moving sideways carries it to streams. Some is converted by soil microbes back to gases — mostly inert dinitrogen, but a fraction as nitrous oxide, a greenhouse gas with a hundred-year warming potential around 273 times that of carbon dioxide, and the dominant remaining threat to stratospheric ozone. Some volatilises from manure and urea as ammonia, drifts, and deposits downwind. Some leaves attached to eroded soil.

Do you see that we derived the list of environmental problems without ever discussing environmental problems? We asked only "the atoms are somewhere — where can they be?" and the exits enumerated themselves. That is the whole force of a conservation argument: it converts a vague worry into an accounting identity with a small number of terms.

One more step, and it is the one that matters most in practice. Does the surplus have to appear now? No. The balance says the atoms exist; it says nothing about their speed. Nitrate can sit in the unsaturated zone beneath a field, moving down at the pace of drainage, and reach a well decades later. Hydrologists call that legacy nitrogen, and it is why a catchment can cut fertiliser sharply and see stream nitrate barely move for years.

Two things are easy to overstate. The first is that a balance identifies a culprit. It does not. It bounds the total and names the possible exits; deciding how much went to leaching rather than denitrification takes measurement, and those partitions vary enormously with rainfall, soil texture, drainage and timing. A budget is a constraint on the answer, not the answer.

The second is that a zero surplus is the goal. It is not quite. Some losses are unavoidable, and a deficit is its own problem — a farm harvesting more nitrogen than it brings in is mining soil organic matter, which shows up later as declining fertility. What the balance really supports is a directional judgement: a persistent, large surplus is a reliable warning, and shrinking it is one of the few interventions that improves water quality, air quality and the farm's own input bill at once.

The practical levers follow from the exits rather than from the total. Applying nitrogen close to when the crop can take it up shortens the window in which rain can move it. Cover crops hold nitrate in living tissue over winter, when there is drainage but no cash crop. Incorporating manure rather than leaving it on the surface reduces ammonia loss. None of these creates or destroys nitrogen — they only change which exit it takes and how fast.

• Why does phosphorus, which the same conservation logic governs, behave so differently in soil and travel by such different routes?
• What is legacy nitrogen doing to the lag between policy change and measured water quality in a catchment?
• How do denitrifying wetlands and bioreactors intercept nitrate after it has left the field, and what do they cost per kilogram removed?

Every term the collection defines is gathered in the glossary.