In nitrogen fertilisers, one upstream chemical — ammonia — quietly decides most of your CBAM bill, and that changes what “managing compliance” actually means for this sector.

In this article:

• The four fertiliser product categories CBAM actually covers, and which ones are excluded
• Why ammonia is the hidden lever behind nearly every nitrogen fertiliser’s carbon footprint
• Nitric acid’s distinct problem: N2O process emissions and mandatory continuous measurement
• A worked example showing precursors accounting for roughly 94% of a finished NPK blend’s emissions
• What producers and importers should actually do about it

If you produce, import, or trade nitrogen fertilisers into the EU, CBAM is no longer a future compliance project — it’s a live financial obligation as of the definitive period that started 1 January 2026. I’ve covered the general mechanics elsewhere on this blog, so just a quick refresher: CBAM makes importers pay a carbon cost comparable to what EU producers already pay under the EU Emissions Trading System, calculated per shipment from the actual (or default) embedded emissions of what’s imported.

Fertilisers are one of the six sectors CBAM covers, and they have a structural quirk that makes them different from steel or cement: almost the entire nitrogen fertiliser industry runs through one upstream chemical. Get that input’s emissions profile wrong, and everything downstream of it is wrong too. That’s what this post is about.

The Four Fertiliser Categories CBAM Covers

CBAM doesn’t regulate “fertilisers” as one blob. It groups nitrogen-containing fertiliser goods into four “aggregated goods categories,” each tied to specific customs (CN) codes:

The scope rule worth remembering: only nitrogen-containing fertilisers are in scope. Phosphorus-and-potassium-only blends (CN 3105 60 00) are explicitly excluded, because the EU guidance states plainly that products without nitrogen don’t carry significant embedded emissions. If your product line is straight P-K, you’re outside this part of CBAM entirely.

One more thing worth flagging early: ammonia and nitric acid are counted twice, in a sense. They’re finished CBAM goods you can import directly, and they’re also precursors — chemical building blocks — consumed inside other fertiliser production. That dual role is the whole story of this sector, so let’s get into it.

Why Ammonia Is the Single Biggest Lever in the Chain

Here’s the fact that should reshape how a producer or importer thinks about CBAM: ammonia isn’t just one of four product categories — it’s the near-universal precursor feeding almost every other category. Nitric acid is made by oxidising ammonia. Urea is made by reacting ammonia with CO2. Mixed fertilisers combine ammonia, nitric acid, urea, and other nitrogen salts. Trace the supply chain of almost any nitrogen fertiliser back far enough, and you land on an ammonia plant.

That matters because of how CBAM treats “complex goods.” A finished fertiliser doesn’t get its embedded emissions calculated from the final mixing or granulation step alone — the emissions of every precursor that went into it roll up into the final number. Import an NPK blend, and the emissions of the ammonia that eventually became part of it are baked into what you owe, even though the product looks nothing like ammonia by the time it reaches the EU.

Practically, this means the single highest-leverage decision in the whole value chain isn’t how a fertiliser is blended or granulated — it’s how the ammonia at the root of the chain was made.

Ammonia sits at the center of the nitrogen fertiliser value chain, feeding nitric acid, urea, and mixed fertiliser production — which is why its production route matters more than any other single decision in the chain.

Ammonia and, where used, urea together typically make up the overwhelming majority of a finished mixed fertiliser’s embedded emissions. Everything else — mixing, granulation, non-nitrogen inputs like potash — is a rounding error by comparison, as the numbers below show.

Ammonia’s Own Emissions: Where They Actually Come From

Ammonia is made through the Haber-Bosch process, and the emissions-heavy part is getting the hydrogen it needs. Two routes dominate today: steam reforming, which uses natural gas (or biogas) as the hydrogen source — highly energy-intensive, since the reaction is endothermic and needs continuous combustion of gas to drive it — and gasification, which uses coal or heavy refinery fuels instead. Either way, this is where nearly all the direct CO2 in the ammonia chain originates.

That CO2 stream is high-purity, which is why some plants pipe it directly to an on-site urea unit — urea synthesis actually consumes CO2 as a reactant. But under CBAM’s rules, transferred CO2 only avoids being counted at the sending installation if it’s “permanently chemically bound,” and CO2 used to make urea doesn’t meet that bar. It’s still counted — just at the urea step instead of the ammonia step. It doesn’t disappear from the ledger; it moves.

There’s a third route worth knowing about: producing hydrogen separately via electrolysis (using electricity to split water) rather than steam reforming. I’ve written more on how CBAM treats hydrogen in our post on the hydrogen sector — briefly, hydrogen is a “simple good” under CBAM with no precursors of its own, and where it’s produced as a genuinely separate, verifiable precursor feeding ammonia synthesis, the indirect electricity emissions behind it can be disregarded when calculating ammonia’s own footprint. That’s a real opening for green-ammonia supply chains, and a reason fertiliser teams should track the hydrogen sector too.

Nitric Acid’s Different Problem: N2O, Not Just CO2

Nitric acid production looks different from ammonia production, and CBAM treats it differently as a result. The process — called the Ostwald process — oxidises ammonia to nitrogen oxide, oxidises that further, then absorbs it in water. It’s exothermic, releasing heat rather than consuming fuel, so direct CO2 combustion emissions here are comparatively minor.

The real issue is a different gas entirely: nitrous oxide, N2O, released as a direct process by-product of catalytic oxidation (or from the abatement equipment meant to control it). N2O has a global warming potential of 265 — one tonne of N2O carries the same climate impact as 265 tonnes of CO2. A small volume of N2O leakage carries an outsized footprint.

Because the stakes of an N2O measurement error are so high, CBAM doesn’t let producers simply calculate this figure from theoretical process yields. It requires continuous emission measurement — a CEMS (Continuous Emission Measurement System) monitoring process N2O directly at the relevant point in the plant, with separate measurement and combined results wherever there’s more than one emission point. That’s a stricter, hardware-based requirement that doesn’t apply the same way elsewhere in the fertiliser chain. One helpful boundary: combustion-related N2O is explicitly excluded — only process N2O counts.

The practical takeaway for anyone sourcing nitric acid, or a mixed fertiliser made with it, is that the abatement technology installed at the producing plant isn’t a minor technical detail. Given the 265x multiplier, it’s the difference between a modest N2O contribution and a very large one.

The Worked Example: Precursors Are ~94% of the Footprint

The clearest way to see why precursor tracing matters more than the blending step is to look at an actual number. The EU’s own worked example calculates the embedded emissions of a standard NPK 15-15-15 mixed fertiliser, granulated from purchased ammonia and urea plus non-nitrogen inputs (potash, phosphates, sulphuric acid — none of which carry embedded emissions under CBAM).

Add up the two purchased precursors — ammonia and urea — and they account for roughly 94% of the finished product’s total embedded emissions (0.458 out of 0.489 t CO2/t product). The producer’s own granulation step contributes only about 6%.

Scaled to a shipment, this is what turns into an actual CBAM bill: 100 tonnes of this NPK product imported into the EU works out to roughly 48.9 tonnes of embedded CO2 — the figure that determines how many CBAM certificates the importer must surrender. (CBAM expresses fertiliser emissions per kilogram of nitrogen content, not per tonne of product, so this also reports as about 0.00326 t CO2e per kg N — worth knowing when comparing products at different nitrogen concentrations.)

Roughly 94% of this NPK blend’s embedded emissions come from the purchased ammonia and urea precursors — the producer’s own granulation step is a small slice of the total.

That example is illustrative, not a market-share claim — the EU’s guidance doesn’t name real countries or trade volumes for this sector; the installation and precursor sourcing are placeholders chosen only to walk through the math.

What This Means in Practice

A few things follow directly from the mechanics above, whether you’re making fertiliser or importing it into the EU.

If you’re a producer, the highest-leverage decision isn’t in your blending or granulation operation — it’s the production route behind your ammonia. Steam reforming and gasification carry large direct CO2 emissions into every downstream product; verified ammonia made from electrolysis-based hydrogen has a meaningfully different profile under CBAM’s own rules. If you’re not integrated back to ammonia production, the figures your ammonia and urea suppliers give you do 90%+ of the work in your final CBAM number — so who you buy from, and whether they supply verified data instead of forcing you onto conservative defaults, is a commercial decision with a real cost attached.

If you produce nitric acid, or buy fertilisers made with it, N2O abatement technology at the producing plant is worth asking about directly — given the 265x multiplier, well-abated versus poorly-abated production shows up as a large swing in embedded emissions.

If you’re an importer or trader, the implication is a data dependency running back through however many precursor tiers sit behind your shipment. A non-integrated producer needs specific emissions data from whoever supplied its ammonia, nitric acid, and urea — an uncooperative supplier means falling back on a conservative default that will generally cost more than the real figure. Getting ahead of that data request, rather than scrambling for it at declaration time, is the single most useful thing an importer can do this year.

How Sprih Helps

The pattern in this sector is the same one we see across CBAM generally, just more concentrated: compliance is fundamentally a supply-chain emissions-data problem, not a paperwork problem. In fertilisers specifically, that’s true in an unusually literal way — with precursors routinely accounting for the vast majority of a finished product’s footprint, getting the ammonia and urea numbers right matters more than almost anything a fertiliser producer does on its own factory floor.

Sprih’s core product, SustainSense, is an AI agent layer that indexes data across 120,000+ companies and 400,000+ sustainability reports — infrastructure built to make exactly this kind of precursor-chain visibility tractable at scale, the way Bloomberg made financial data tractable for finance. We’re also building CBAM-specific tooling to help importers and their suppliers calculate and manage embedded-emissions data, including precursor-chain tracing that matters enormously for sectors like this one. If your fertiliser supply chain runs through multiple tiers of ammonia, nitric acid, and urea suppliers and you need a clearer picture of where your actual CBAM exposure sits, that’s exactly the problem we work on.

Sprih is an AI-native sustainability intelligence company helping enterprises turn climate and compliance data into competitive advantage.