Fertiliser Supply Disruption

Methodology and sources

CAP-A internal analysis · Global fertiliser supply and the 2026 Iran/Middle East conflict · Supply constraints last checked 13 July 2026 · Cumulative production losses calculated to 29 June 2026.

Methodology overview

This resource tracks how the 2026 Iran/Middle East conflict, which began on 28 February 2026, has disrupted the global supply of urea — the world's most widely traded nitrogen fertiliser. It measures three things: the plant-level capacity knocked offline, the cumulative production physically lost while plants stay down, and the scale of that loss set against what farmers actually use. It is a supply-side production analysis, not a price forecast.

Affected capacity is built from the bottom up. Each producing facility caught up in the conflict is entered individually with its annual urea capacity and its operating status, and these are summed into three groups: capacity offline, capacity at risk, and confirmed production cuts. Cumulative lost production is then calculated facility by facility from the day each plant went down, so the total grows for every day Iran and Qatar remain offline.

Every capacity figure is sourced wherever possible to the operator's own publication (the company or its parent), then cross-checked against independent specialist trade sources such as CRU, ICIS and Argus. The status of each plant — offline, curtailed, at risk or recovered — is verified separately against named, dated reports, preferring official statements and on-record analyst quotes. Each entry therefore rests on two distinct checks: one for capacity and one for status.

Where sources disagree, the analysis carries an explicit low–high range rather than a single point estimate, and every entry is given a confidence rating. Figures that are contested or contain a likely category error are identified clearly and excluded from the headline totals rather than hidden. The supply picture was last reviewed on 13 July 2026. The published cumulative production-loss estimate is calculated to 29 June 2026; it would continue to rise for as long as affected plants remain offline.

The status-check date records when the operating position of affected facilities was reviewed. The calculation date records the point to which cumulative lost production was calculated.

Key terms and definitions

The single most important distinction in this analysis is between urea product, nitrogen nutrient and urea-equivalent nitrogen demand. Because these are easily confused, they are defined first.

The three ways tonnes are counted

• Urea product — physical tonnes of urea, the finished granular fertiliser that is shipped and traded. Urea is about 46% nitrogen by mass. All of this analysis's supply-loss figures are in tonnes of urea product.

• Nitrogen nutrient (N) — the tonnes of actual nitrogen contained in a fertiliser. One tonne of urea contains about 0.46 tonnes of nitrogen. National fertiliser-use statistics (for example FAO data) are usually reported in nitrogen-nutrient tonnes.

• Urea-equivalent nitrogen demand — a nitrogen-nutrient figure converted into the quantity of urea that would contain the same amount of nitrogen. Sub-Saharan Africa's annual nitrogen use of approximately 2.7 Mt N is equivalent to approximately 5.9 Mt of urea.

In short, supply losses are measured in tonnes of urea product, while regional nitrogen demand is converted to a urea-equivalent quantity so that the comparison is like for like.

Capacity and status terms

• Plant-level capacity offline — annual urea capacity that is not producing, either because the plant was physically destroyed or because its upstream gas feedstock was destroyed. The headline offline figure is Iran, Qatar and (until repair) Bahrain GPIC.

• Production cuts — output reduced at plants that were idled by feedstock or logistics constraints rather than destroyed (for example Bangladesh and Pakistan Agritech). Tracked separately from destroyed/offline capacity and lower in confidence.

• Capacity at risk — plants that are intact and still producing but exposed to feedstock or export-route disruption (for example Egypt, Oman, SABIC). Counted separately and not part of the production-loss total.

• Back online / recovered — a plant whose outage has ended (for example Bahrain GPIC, repaired in mid-May 2026). Its lost output is frozen at the amount accrued during the outage and stops growing.

• Nameplate capacity — the rated maximum annual output of a plant or of the world (about 240 Mt/yr of urea globally). A conservative denominator.

• Actual production — what is really produced in normal operation (about 204.1 Mt/yr globally), lower than nameplate because plants do not run flat out. The operationally meaningful denominator.

• Realised production lost — the tonnes of urea actually not produced to date because of the outages, accumulated day by day. This is distinct from nameplate taken offline: Bangladesh, for instance, has 2.0–2.2 Mt of nameplate offline but a realised loss of only about 0.7–1.0 Mt because its plants ran well below capacity beforehand.

• Annualised loss / run rate — the forward loss if the current daily rate of lost production simply continued for a full year (about 12–15 Mt/yr while Iran and Qatar stay offline). A projection of the current rate, not a figure already realised.

• Urea-equivalent nitrogen demand — defined above; a nitrogen-demand figure re-expressed as urea so it can be compared directly with urea supply losses.

How the main figures are calculated

Cumulative production lost to date. For each facility, lost production = annual capacity × utilisation × (days offline ÷ 365), summed across all affected plants and run to the calculation date (29 June 2026). This gives roughly 4.0–4.9 Mt of urea not produced since early March 2026 — about 6.1–7.5% of what those operations would normally have made over the same period. The estimate is cumulative. It would increase with each additional day that Iran and Qatar remain offline, so any published figure should state its calculation date.

Daily and monthly loss. With Iran and Qatar still offline, the ongoing loss runs at roughly 33–40 kt of urea per day, or about 1.0–1.2 Mt per month.

Forward annual loss (run rate). If nothing changes, the current daily rate annualises to about 12–15 Mt/yr (the model uses 12.1–14.5 Mt/yr for Iran plus Qatar). This is a run-rate projection, not production already lost.

Comparison with Sub-Saharan African nitrogen demand. To make the loss tangible, it is set against Sub-Saharan Africa's entire annual nitrogen use (FAO 2023 data for 48 countries). That demand, approximately 2.7 Mt of nitrogen, is converted to a urea-equivalent quantity of approximately 5.9 Mt. On this like-for-like basis, the 4.0–4.9 Mt lost so far is equivalent to roughly eight to ten months of the region's annual nitrogen use, and the forward annual loss is roughly two to two-and-a-half times it.

Low and high estimates. Wherever sources give different numbers (for example Iran's capacity, quoted between 6.5 and 8.9 Mt/yr), both bounds are carried through every calculation, producing a low–high range rather than a false single figure.

Principal assumptions

• Utilisation. The three fully offline plants (Iran, Qatar, Bahrain GPIC) are counted at 100% of nameplate, so the run-rate sits on the same basis as the capacity figures. Chronically under-utilised plants (notably Bangladesh at ~35%) are adjusted down to a realised-loss basis.

• Outage start dates. Iran is dated from 1 April 2026 (when output was reported already halted), Qatar from 4 March 2026, Bahrain GPIC from 5 April 2026 — conservative choices; earlier war-related curtailment would raise the figure.

• Recovery dates. Recovered plants stop accruing loss at their repair date (Bahrain GPIC, mid-May 2026). Iran and Qatar are assumed to remain offline on a multi-year rebuild absent confirmation of restart.

• Denominators. Global figures use 240 Mt/yr nameplate and 204.1 Mt/yr actual production; percentage-of-trade figures use seaborne urea trade of 54.7 Mt/yr.

Scope and limitations

What is included and excluded. The analysis counts urea capacity and urea production directly affected by the conflict. It excludes downstream effects and several contextual or disputed items from the headline totals: Russia’s Acron Dorogobuzh because it produces ammonium nitrate rather than urea; the questioned China export figure because it is inconsistent with reported export volumes; and Saudi Ma’aden phosphate because it concerns a different nutrient. European gas-exposed capacity is tracked separately as a latent risk and is not included in the headline loss.

Production loss versus trade disruption. These are kept strictly apart. Lost production is fertiliser that was never made because a plant stopped. Trade disruption is fertiliser that exists but cannot move — for example cargoes stranded by the closure of the Strait of Hormuz. The headline figure is lost production; the confirmed lost-trade figure (Argus: at least 2 Mt of urea) is reported separately.

Temporary outages and recovered facilities. A plant that comes back is not erased: its lost output during the outage is retained but frozen, so recovery halts further losses without deleting the tonnes already lost. Bahrain GPIC is treated this way.

Where evidence is lower confidence. Production cuts (Bangladesh, Pakistan, India) are lower confidence than the destroyed/offline plants and are shown separately. India's March cut was reframed to a realised one-month shortfall after output recovered. Egypt, Oman and SABIC are 'at risk' rather than confirmed losses. Ratings of Medium or Questioned flag exactly where the evidence is thinner.

Why falling prices do not recover lost production. Urea prices spiked after the conflict and then fell back by mid-2026 as the ceasefire and partial reopening of the Strait removed the war-risk premium. A lower price is welcome, but it recovers none of the tonnes already lost: the physical hole in production remains and keeps growing while Iran and Qatar stay offline. Price relief and supply repair are different things.

Public source register

Every key figure and status in this analysis is traceable to a named, dated, public source. The table below summarises the principal sources, capacity estimates, operating-status checks and relevant caveats used in the analysis.

Confidence ratings: High = capacity and status corroborated by authoritative sources, using primary sources where available; Medium = one or both need judgement or rest on fewer sources; Questioned = contested or contains a likely category error and is excluded from the headline totals.

Source: CAP-A internal analysis. Figures are urea product or the urea-equivalent of nitrogen demand; ranges show low–high estimates where sources disagree.