Carbon Accounting for Agriculture and Livestock in Australia

A single beef cow belches somewhere around 100–120 kg of methane per year. Multiply that across Australia's 30.4 million cattle (MLA Fast Facts, June 2024) and you start to understand why agriculture emissions reporting under NGER isn't something you can hand-wave away with a rough estimate.

Agriculture accounts for roughly 13% of Australia's national greenhouse gas emissions — about 55–60 Mt CO2-e annually according to the National Greenhouse Gas Inventory. Enteric fermentation alone (that's the methane produced during ruminant digestion) makes up around 71% of that. Add manure management, fertiliser use, and liming, and you've got a reporting obligation that looks nothing like tracking electricity bills in an office building.

We built Carbonly for utility bill extraction and energy-based emissions. But our team comes from enterprise data platforms at BHP and Rio Tinto, and we've spent enough time around primary industries to know this: agriculture emissions are the messiest reporting problem in Australia. Not because the science is unclear. Because the data collection is genuinely hard.

Why Most Farmers Don't Realise They're NGER Reporters

Here's the thing most agricultural operations get wrong. They assume NGER is for power stations and mines. It's not.

The NGER thresholds are 50,000 tonnes CO2-e or 200 TJ of energy for a corporate group, and 25,000 tonnes CO2-e or 100 TJ for an individual facility. A large beef cattle operation — say 40,000 head across multiple stations in Queensland — can easily produce 15,000–20,000 tonnes of CO2-e just from enteric fermentation. Add diesel for station vehicles and generators, natural gas, fertiliser application across cropping land, and the energy consumed in on-site processing, and you're closer to those thresholds than you think.

And if your corporate group includes feedlots, meat processing, and pastoral stations under a single controlling corporation? The aggregation rules under section 13 of the NGER Act pull everything together. We've written about how this catches companies off guard — agriculture is one of the sectors where it happens most.

Large feedlots are a clear example. A 50,000-head feedlot produces significant methane from both enteric fermentation and manure management. The facility also consumes substantial energy — feed milling, water pumping, effluent management. That combination of Scope 1 biological emissions and Scope 1/2 energy emissions can push a single facility past the 25 kt threshold on its own.

But even dispersed pastoral operations trigger reporting when they're consolidated under a corporate group. The NGER Act's operational control test (sections 11–11B) asks who has authority over operating, health and safety, and environmental policies. For vertically integrated agribusinesses — a company that runs cattle stations, a feedlot, and a processing plant — the answer is usually the parent entity. And that parent entity has to report for all of them.

The Six Emission Categories That Agriculture Reporters Face

Agriculture emissions reporting under NGER isn't one calculation. It's six distinct categories, each with its own methodology and data requirements. The NGER Measurement Determination specifies how each must be estimated.

Enteric fermentation is the big one. This is methane (CH4) produced during digestion in ruminant animals — cattle, sheep, and goats. Australia's National Greenhouse Gas Inventory uses a Tier 2 methodology for cattle, which means emission factors aren't just a flat rate per head. They depend on dry matter intake, feed quality, animal weight, and productivity class. A dairy cow producing 25 litres per day has a different emission factor than a rangeland beef steer eating dry tropical grass. For NGER reporting, you need to categorise your herd by animal class and apply the appropriate factors from the NGER Measurement Determination.

Manure management covers methane and nitrous oxide (N2O) from how animal waste is stored and treated. A beef cattle operation where animals graze on pasture (and manure decomposes aerobically in paddocks) has very different emissions than a feedlot using anaerobic lagoons. The methodology depends on the manure management system in place — pasture, dry lot, liquid systems, or composting all have different emission factors.

Agricultural soils contribute N2O emissions from nitrogen applied to soil — both synthetic fertiliser and animal dung and urine deposited during grazing. Nitrous oxide is roughly 265 times more potent than CO2 on a 100-year timeframe (AR5 GWP values, which NGER uses). A cropping operation applying 200 tonnes of urea across 5,000 hectares is generating reportable N2O emissions that have to be calculated using specific application rates and emission factors.

Urea application gets its own sub-category because urea releases CO2 when it breaks down in soil — separate from the N2O emissions counted under agricultural soils. If you're applying urea as a nitrogen fertiliser (and most Australian broadacre farmers are), you're reporting under both categories.

Liming — the application of calcium carbonate (lime) or dite to agricultural soils to manage pH — also releases CO2. It's a smaller category, making up about 2% of agricultural emissions nationally, but it's still a reporting line item you can't skip.

Field burning of agricultural residues is the final category. Burning stubble or crop residues after harvest produces CH4, N2O, and CO2. This category has declined in significance as more farmers adopt stubble retention, but it's still relevant — particularly for sugarcane operations in Queensland and some cereal crop regions.

The Data Collection Problem Nobody Talks About

Here's where we need to be honest about the limits of software — including ours.

Agricultural emissions data is fundamentally different from energy data. When you're tracking Scope 2 electricity emissions, you've got a utility bill with kWh clearly printed on it. We've built an AI pipeline that extracts this data automatically. Farming doesn't work that way.

To estimate enteric fermentation for a cattle operation, you need to know: how many animals you had, in what categories (breeding cows, steers, heifers, calves, bulls), at what point in the year, and ideally what they were eating. Rangeland cattle in the Northern Territory eating native pasture in the dry season have different dry matter intake (and therefore different methane output) than feedlot cattle on a high-grain ration. The NGER Measurement Determination requires you to account for this.

The ABS tried to estimate Australia's cattle herd using traditional survey methods for years and recently discovered they'd been undercounting by about 3.4 million head — a 14% error. If the national statistics agency can't count cattle accurately using surveys, imagine what happens when a station manager tries to estimate average annual head counts across three properties totalling 500,000 hectares.

Seasonal variation makes everything harder. Cattle numbers on a property change throughout the year as animals are mustered, sold, born, or moved between paddocks and outstations. A snapshot count on 30 June doesn't capture the reality of a herd that peaked at 12,000 head in March but dropped to 8,500 by October after a dry spell forced destocking. The NGER methodology expects an annual average — getting to that number honestly requires stock records that many operations don't keep with the precision a regulator would expect.

We're not sure any software fully solves the livestock data collection problem yet. There are property management systems (like Maia Technology or AgriWebb) that track individual animal movements, and NLIS (the National Livestock Identification System) records movements through saleyards and processing plants. But connecting these data sources into a format that maps cleanly to NGER emission categories — that's still a largely manual exercise on most farms.

What software can do well is the energy side: diesel consumption from fuel records, electricity from utility bills, natural gas from invoices. These are the same document types we process through Carbonly's extraction pipeline every day. Carbonly's AI Document Processing handles eight input formats — PDF, CSV, multi-sheet Excel, Word, PPT, RTF, and even photographed or scanned images — which matters when your data arrives as a photographed fuel docket from a station hand, a CSV export from your fuel card provider, and a PDF electricity bill from the retailer, all in the same week. The system's five-tier material matching automatically maps each line item to the correct NGA emission factor, including the livestock-specific factors in the National Greenhouse Accounts for enteric fermentation and manure management, with a confidence score so you can see exactly how each match was made.

For operations running multiple properties or stations, Carbonly's multi-project structure lets you track each property as its own project — separate facilities, separate data feeds, separate emission totals — while rolling everything up to the corporate group level for NGER reporting. Email ingestion means each property can have its own forwarding address: invoices and fuel dockets sent to that address are automatically processed and assigned to the correct project without anyone manually uploading files.

And when it comes to modelling herd management changes — shifting calving times, improving feed quality, earlier turn-off weights — Carbonly's Carbon Planning module lets you build scenarios with cost-benefit analysis against your baseline. You can model how a change in feed ration affects methane intensity per head, estimate the abatement in tonnes CO2-e, and weigh that against the cost of supplementary feeding. It's not a replacement for agronomic advice, but it gives you the numbers to back up decisions with data rather than intuition.

For agriculture operations, this energy data often represents 20–30% of total reportable emissions — significant, and much easier to automate than the biological emissions.

The Safeguard Mechanism: Agriculture Gets a Pass (For Now)

The Safeguard Mechanism applies to facilities emitting more than 100,000 tonnes CO2-e of Scope 1 emissions per year, with baselines declining at 4.9% annually to 2030. It currently covers around 220 facilities across mining, manufacturing, oil and gas, transport, and waste.

Agriculture is not covered. Full stop.

That's worth understanding clearly because it's a source of confusion. An agricultural facility can be an NGER reporter (because it crosses the 50 kt corporate group or 25 kt facility threshold) without being a Safeguard Mechanism facility. The two schemes have different thresholds and different sectoral coverage. Even a very large cattle operation — say 100,000 head producing 30,000+ tonnes CO2-e — wouldn't trigger the Safeguard Mechanism's 100 kt facility threshold from enteric fermentation alone.

But don't get comfortable. The policy direction is towards expanding coverage, and agriculture's exclusion from the Safeguard Mechanism doesn't mean exclusion from reporting. If your operation crosses NGER thresholds, you report. And if you're an NGER reporter, you're now automatically pulled into ASRS Group 2 mandatory climate disclosure obligations. That means AASB S2 reporting, assurance requirements, and director liability for the accuracy of your Scope 1 and 2 numbers.

For a beef producer who's been treating NGER as an administrative nuisance, that escalation changes everything.

The Opportunity Side: ACCUs and Carbon Farming

Agriculture isn't only a source of emissions. It's also one of the few sectors that can generate Australian Carbon Credit Units (ACCUs) through the ACCU Scheme administered by the Clean Energy Regulator.

Three method categories matter for agricultural operations.

Soil carbon sequestration has been the fastest-growing ACCU method by project registrations. As of late 2023, 525 soil carbon projects had been registered — though only nine had progressed to receiving credits, which tells you something about how difficult measurement and verification is in practice. The method requires demonstrating an increase in soil organic carbon through changed management practices (rotational grazing, cover cropping, reduced tillage). Projects must elect either a 25-year or 100-year permanence period. Most elect 25 years.

Savanna fire management is the most established agricultural-adjacent ACCU method, particularly for northern Australia. Indigenous-led savanna burning projects now cover over 22 million hectares and generate more than 1 million tonnes of emissions reductions annually, worth roughly $50–60 million per year to Indigenous organisations. The method works by replacing destructive late dry season fires with controlled early dry season burns, reducing methane and N2O from wildfire. More than 9 million ACCUs have been issued for savanna fire projects to date, making it one of the scheme's most successful methods.

Beef cattle herd management was a method that allowed producers to earn ACCUs by reducing emissions intensity per kilogram of liveweight gain. It's now closed to new projects, but it's worth noting because it demonstrated that herd management changes (better genetics, improved nutrition, earlier turn-off) can measurably reduce enteric methane. The data infrastructure required to prove those reductions, though, was substantial — and that's partly why uptake was limited.

The honest reality is that ACCU generation from agriculture requires serious data collection, long-term commitment (25-year permanence periods aren't trivial for a farming business), and tolerance for the slow pace of verification. It's an opportunity, but it's not quick money.

What This Means for Agribusiness in Practice

If you're running an agricultural operation that's anywhere near NGER thresholds — or if you're a corporate agribusiness that consolidated multiple properties through acquisition — here are the things that actually matter.

Start with your energy data. Diesel, electricity, gas — these are the emissions sources where data quality is highest and automation is easiest. Get your fuel purchase records, utility bills, and gas invoices into a system that can calculate emissions automatically using NGA emission factors. This won't cover your enteric fermentation, but it gives you a defensible baseline for energy-related emissions and takes the easy wins off the table.

Sort out your stock records. The NGER Measurement Determination requires you to report livestock by category and estimate annual averages. If your property management system can generate a report showing monthly head counts by animal class, you're in reasonable shape. If your records consist of a notebook in the station office and your manager's memory, you've got a problem that no software can fix until the underlying data exists.

Understand the GWP values. NGER uses AR5 Global Warming Potential values: methane at 28 times CO2, nitrous oxide at 265 times CO2. If you're also reporting under AASB S2 for ASRS, note that AASB S2 requires AR6 values — where methane is 27.2 and nitrous oxide is 273. It's a small difference but it means you might need to run two sets of calculations. We've covered this NGER-ASRS mismatch in detail elsewhere.

Don't confuse the Safeguard Mechanism with NGER. Agriculture isn't covered by the Safeguard Mechanism, but that doesn't mean you're exempt from emissions reporting. The NGER thresholds are the ones that matter, and the consequence of crossing them now includes ASRS Group 2 climate disclosure obligations.

And if you're thinking about ACCUs — soil carbon, savanna burning, vegetation — start the conversation early. The permanence commitments are long, the measurement is complex, and the crediting timelines are slow. But the revenue potential is real, and the Carbon Farming Outreach Program (now administered by DAFF with $27.8 million in funding from 2024-25) exists specifically to help agricultural operations navigate these methods.

Agriculture emissions reporting is genuinely hard. The biological nature of the emissions, the dispersed geography of Australian farming, and the difficulty of counting animals accurately across vast properties make this one of the most challenging sectors for carbon accounting. We won't pretend otherwise. But the regulatory requirements aren't going away — and for agribusinesses near the NGER thresholds, getting the data infrastructure right now is cheaper than scrambling when the Clean Energy Regulator comes asking questions.

Related Reading: