# Tilth : full text for language models

License: CC BY 4.0 (attribution: Tilth (tilth.uk)). Prices are estimates in GBP/tonne, not offers.
Machine contract: https://intelligence.tilth.uk/openapi.json

## Current index values (live)

- AN_IMPORT: 500 GBP/t (band 483 to 517), as of 2026-10-12
- AN_UK: 497 GBP/t (band 482 to 512), as of 2026-10-12
- DAP: 794 GBP/t (band 780 to 808), as of 2026-10-12
- MOP: 363 GBP/t (band 356 to 370), as of 2026-10-12
- NITRATE_SULPHUR: 494 GBP/t (band 483 to 504), as of 2026-10-12
- POLYSULPHATE: 239 GBP/t (band 236 to 243), as of 2026-10-12
- TSP: 642 GBP/t (band 627 to 657), as of 2026-10-12
- UAN: 414 GBP/t (band 402 to 426), as of 2026-10-12
- UREA_GRANULAR: 499 GBP/t (band 484 to 514), as of 2026-10-12

## When to buy fertiliser in the UK

Nobody rings a bell at the bottom of the fertiliser market. But the UK fertiliser year does have a repeatable shape, and knowing it turns buying from a guess into a decision. This guide covers the seasonal pattern, what breaks it, and how to use a daily price index to act with your eyes open.

UK nitrogen selling typically restarts each early summer, when manufacturers publish new-season terms after the spring application season ends. Demand is thinnest through summer, builds through autumn as farms fill sheds, and peaks from January to early spring when the top-dressing season arrives and latecomers must buy whatever the market asks.

That demand curve is why the folk wisdom says buy early. Sellers price the thin summer market to get volume moving, and price the spring rush knowing buyers have no time left. Across many seasons the pattern holds often enough to matter, and in any single season it can fail completely.

The shape of the fertiliser year

Fertiliser is made from energy. When European gas jumps, as it did spectacularly in 2021 and 2022, the seasonal discount evaporates and early buyers can end up holding stock bought above the spring price. Currency moves, plant outages, import disruptions and policy changes do the same. Seasonality is a tendency, not a contract.

That is why this guide will not tell you a month to buy. The honest version is: the calendar gives you a starting bias (thin-market months have historically been kinder to buyers), and the current state of the drivers tells you whether this year is normal or not.

What breaks the pattern

Until recently a farmer checking the market had a weekly reference at best. Our index publishes a daily estimate for each major grade, produced from the drivers themselves: European gas, EU agri-food price indices, producer signals and currency, validated against weekly UK market prints.

The practical routine is simple. Decide your tonnage early. Watch the grade page for the fertiliser you buy: the current estimate, the one-month move, and where today sits in the 52-week range. If the price is in the lower half of its range during the thin months and the drivers are calm, waiting rarely pays. If the drivers are moving fast, the range tells you how quickly things can travel.

Set a price alert on your grade so a drop finds you, rather than you finding it after the season has turned.

Using a daily index instead of guessing

Most farms do not need to win the market, they need to avoid losing to it. Buying in two or three tranches across the buying window smooths the price you pay and caps the regret of any single decision. A first tranche on new-season terms, a second in autumn, and flexibility kept for spring is a common shape that works with the demand curve rather than against it.

Splitting the risk


## What drives UK fertiliser prices

Fertiliser prices look chaotic from the farm gate. They are not random: a small set of drivers explains most of the movement, and they are the same drivers our model reads every day to produce the published index. This guide walks through them in order of weight.

Nitrogen fertiliser is made by fixing nitrogen from the air using hydrogen from natural gas, and gas is typically 60 to 80 percent of a nitrogen plant's production cost. When European gas rises, every tonne of ammonia, urea and AN costs more to make, and the market reprices within weeks. The 2021 to 2022 gas shock, when European fertiliser plants curtailed production outright, remains the defining example.

Gas: the feedstock

The UK imports a large share of its fertiliser and manufactures the rest in a market priced against continental Europe. EU agri-food price indices and continental producer behaviour set the reference level a UK buyer ultimately pays, which is why our model reads European indices rather than starting from a UK quote sheet.

Europe: the reference market

Imported tonnes are bought in euros or dollars and sold in pounds. A weaker pound raises the landed cost of imported AN and urea even when the world price stands still. Currency is the quiet driver: rarely dramatic week to week, always present in the level.

Currency: the landed cost

Listed fertiliser producers publish stock levels, production figures and outlooks that lead the market's direction. Layered on top is the seasonal demand curve covered in our timing guide: thin summer demand, spring urgency. Season shifts when a price move happens; the other drivers decide how big it is.

Producer signals and season

From 1 January 2027 the UK plans to apply its Carbon Border Adjustment Mechanism to imported fertiliser, pricing the carbon embedded in imports as the EU began doing in 2026. The direction of the effect is upward pressure on imported nitrogen costs; the size depends on carbon prices and implementation detail still being settled. Buyers planning spring 2027 nitrogen should watch it explicitly.

2027: the UK carbon border charge

Each day our model combines these drivers (European gas, EU agri-food price indices, producer signals and currency) into an estimate per grade, with an uncertainty band, validated against weekly UK market prints. When a grade's model has not yet earned trust against those prints, we stage it rather than publish a number we do not stand behind. The full methodology is on the data page.

How the index reads the drivers


## Urea vs ammonium nitrate: cost per kilogram of nitrogen

A tonne of granular urea nearly always costs more than a tonne of ammonium nitrate, and it is nearly always the cheaper way to buy nitrogen. Both things are true, because the tonne is the wrong unit. This guide shows the only comparison that matters: pounds per kilogram of nitrogen, and what the headline saving does not include.

Granular urea is 46% nitrogen: a tonne carries 460 kg of N. UK ammonium nitrate is 34.5% nitrogen: a tonne carries 345 kg. Divide each price per tonne by its kilograms of nitrogen and you get the true cost of the nutrient. Urea at 400 pounds per tonne is 0.87 pounds per kg N; AN at 350 pounds per tonne is 1.01 pounds per kg N. The dearer tonne is the cheaper nitrogen.

The break-even is fixed by the analysis: urea is the cheaper kilogram of nitrogen whenever its price per tonne is below 1.33 times the AN price. Above that ratio, AN wins on cost.

The arithmetic

Urea's nitrogen is not plant-ready. It must convert in the soil first, and during conversion some nitrogen can be lost to the air as ammonia, especially when urea is spread onto a warm, dry surface and no rain follows. Losses vary widely with weather and soil; the saving per kilogram bought is not the saving per kilogram the crop actually receives.

Regulation reflects this. In England, solid urea is subject to spreading restrictions under industry schemes: unprotected urea is for early-season use, and later applications are expected to use a urease inhibitor, which adds cost per tonne and narrows the gap to AN. Factor the inhibitor into your comparison if your window runs past early spring.

AN's nitrogen is half nitrate, available to the crop immediately and more predictable in cool conditions, which is why much of the UK trade still prices everything against it.

What the headline saving leaves out

The tiles below divide today's index estimates by each grade's nitrogen content. Both underlying figures are daily estimates produced from driver indices and validated against weekly UK market prints; follow the grade pages for the full picture including the uncertainty band.

Compare with live figures


## Fertiliser cost per hectare for winter wheat

A fertiliser quote is a price per tonne; a farm budget is £ per hectare. The bridge between them is three numbers you already know: application rate, nutrient content, and the going price. This guide gives you the arithmetic and worked examples you can rerun with today's index figures.

Cost per hectare equals nitrogen rate (kg N per hectare) divided by the product's kg of N per tonne, multiplied by the price per tonne. AN carries 345 kg N per tonne (34.5%), granular urea 460 kg (46%). A typical feed winter wheat nitrogen programme is in the region of 180 to 220 kg N per hectare (soil, yield potential and previous cropping move it; RB209 is the reference for your own number).

The formula

Take 200 kg N per hectare from AN at £350 per tonne. Product needed: 200 divided by 345 is 0.58 tonnes per hectare. Cost: 0.58 times 350 is roughly £203 per hectare. The same nitrogen from urea at £400 per tonne needs 200 divided by 460, or 0.43 tonnes, costing about £174 per hectare before any inhibitor or application-loss adjustment (see the urea vs AN guide).

Every £10 per tonne the AN price moves changes that 200 kg N programme by about £5.80 per hectare. That single sensitivity is the most useful number in this guide: it converts any index move you see into budget terms instantly.

Worked example: AN at £350 per tonne

Nitrogen is the volatile line, but a full programme also carries phosphate, potash and increasingly sulphur. Maintenance dressings priced from DAP, muriate of potash or polysulphate typically add a further meaningful share of the bill, and their prices move on their own drivers. Each grade has its own live page on the index.

Do not forget P, K and sulphur

The worked examples above freeze a price to show the method; your budget should not. Check the current estimate for your grade, rerun the two-line calculation, and set a price alert so a market move updates your plan rather than surprising it.

Keeping the budget current


## How accurate is a fertiliser price forecast?

A daily fertiliser price forecast is only worth your attention if you can tell how often it is right. This guide explains how our index is tested, what the shaded band on every chart really means, and how to read the live scorecard below, which shows how each grade's model is performing right now against real UK market prints.

Fertiliser prices are persistent: tomorrow usually looks a lot like today. That makes one very dumb strategy surprisingly hard to beat: assume the price simply stays where it was last printed. Forecasters call this the flat-carry baseline, and any model that cannot beat it is adding noise, not information.

So that is the bar we set. A grade earns the right to publish a modelled figure only by beating flat-carry by a clear margin, at least ten percent lower error, on months it has never seen. Where a grade cannot clear that bar yet, we say so on the chart rather than dress a flat line up as a forecast.

The baseline any forecast must beat

Accuracy measured on the data a model was trained on is fiction. Every grade is therefore tested walk-forward: the model is fitted only on months that came before the month it must predict, one month at a time, with at least three years of history behind the first test. The errors that come out of that process are the honest kind, out-of-sample, with no future information leaking backwards. The same discipline applies to changing the model itself: a candidate driver or fitting method runs the same walk-forward gauntlet against the live model, and ships only if the improvement survives statistical significance testing.

Tested only on months it has never seen

Not every grade has enough history to pass the test. Newer or thinner grades publish with an explicit early stage label carried in the data itself, and a deliberately reduced response, so the model moves them gently until the evidence firms up. The test re-runs with every calibration: a grade that accumulates enough evidence is promoted automatically, and a grade whose edge decays steps back down. Nothing about the label is editorial; it is the test's own verdict, and the API carries it on every figure it publishes.

Grades earn their label, and can lose it

The band around each line is computed from that grade's own recent volatility, at a fixed one-week horizon. A calm grade wears a tight band; a jumpy grade wears a wide one. It is the model stating how far it trusts its own figure, and it is policed: a daily move larger than three standard deviations is held rather than published, and derived figures such as price per kilogram of nitrogen publish only while the latest UK print falls inside the band. The band-hits column in the scorecard shows how honest that band has recently been, grade by grade.

What the shaded band really tells you

A forecast validated once and then left alone rots quietly. Ours is recalibrated and re-scored continuously: every daily estimate is compared against the next realised weekly print, the walk-forward test re-runs on the growing record, and candidate improvements are trialled against the live model under the same rules. Most candidates lose, and the failures are recorded; the few that win with statistical significance ship. The scorecard above is the running output of that loop, not a snapshot taken the day this guide was written.

The model has to keep earning its place

Read the band before the line: the line is the centre of a range, not a promise. Use the direction and the band together to time purchases inside your own window, check the status label so you know how much validation stands behind the grade you buy, and compare any quote you receive against the going range rather than a single number. Used this way a forecast is a negotiating instrument; used as gospel it is a disappointment waiting to happen.

How to use a forecast well