NGA vs GHG Protocol: Why Australian Carbon...

The most common question we get from finance teams sitting down to their first AASB S2 disclosure is a variation of the same theme. We already report under NGER. Why are our numbers different in the AASB S2 draft? Which one is right?

Both are right. They answer different questions. And the reason the numbers move between them is not a bug, it is the point.

Australia has two live carbon reporting frameworks running in parallel through 2026 and 2027. The National Greenhouse and Energy Reporting scheme, run by the Clean Energy Regulator under the NGER Act 2007, is a mandatory submission for corporations that cross the threshold. AASB S2 is the mandatory climate-related financial disclosure standard issued in September 2024, phased in for Group 1 entities from 1 January 2025 and expanding to Group 2 in FY starting 1 July 2026 and Group 3 in FY starting 1 July 2027. They share activity data. They do not share methodology.

The activity data lives in one place. The methodology diverges. If you build the ledger correctly, the same litre of diesel from the same Ampol invoice can render into an NGER submission and an AASB S2 disclosure without you touching it twice. If you build it incorrectly, you will double-count Scope 2, misclassify purchased goods, or worse, submit an NGER number that does not tie to your audited disclosure. That last one is the failure mode ASIC is watching for.

What NGA actually is

The National Greenhouse Accounts factors, published by the Department of Climate Change, Energy, the Environment and Water (DCCEEW), are an emission factor library. Nothing more, nothing less. The current edition, NGA 2025, contains around 193 factors covering fuels, electricity, transport, waste, fugitives and industrial processes. Of those, 21 are per-litre fuel factors sitting under Table 4 of the workbook: automotive diesel, unleaded petrol, aviation kerosene, LPG and the rest of the liquid fuel set that dominates Scope 1 in fleet-heavy industries.

Grid electricity factors sit under Table 1 (location-based) and Table 2 (market-based), state by state. NGA 2025 lists Victoria at 0.78 kg CO2-e/kWh, New South Wales and ACT at 0.64, Tasmania at 0.20. Those numbers matter because they will move each year as generation mixes decarbonise, and last year's factor cannot be silently overwritten in your ledger without breaking your audit trail. That factor-versioning problem is a whole separate discipline.

What NGA does not do is tell you which scope an emission belongs in, which reporting boundary you should draw, or how to categorise a Scope 3 transaction. Those questions sit elsewhere.

What GHG Protocol actually is

The GHG Protocol Corporate Standard, published jointly by the World Resources Institute and the World Business Council for Sustainable Development, is the categorisation and boundary framework. It defines Scope 1 as direct emissions from owned or controlled sources, Scope 2 as indirect emissions from purchased energy, and Scope 3 as everything else in the value chain. The companion Corporate Value Chain (Scope 3) Standard breaks Scope 3 into 15 categories, from purchased goods and services (Category 1) through end-of-life treatment of sold products (Category 12) and franchises (Category 14).

The GHG Protocol tells you where an emission belongs. It does not tell you what number to multiply the activity data by. It defers to jurisdictional factor libraries for that.

The two frameworks are complementary. NGA answers "what is the factor". GHG Protocol answers "what scope does it belong in". Confuse the two and you have made a category error that will surface at limited assurance.

Where NGER Measurement Determination sits

Between the activity data and the NGA factor library sits the NGER (Measurement) Determination 2008 (as amended). This is the legally binding methodology document that specifies how NGER-reporting corporations must calculate emissions. It uses NGA factors as its default (Method 1) and allows higher-tier methods (Method 2 and Method 3) that require sampling, analysis and Clean Energy Regulator approval.

The Determination is Australian statute, updated annually, and it uses IPCC AR5 (2013) global warming potentials for methane and nitrous oxide. Methane at 28. Nitrous oxide at 265. That is not the reporter's choice, it is written into the Determination.

If you are above the NGER threshold (25 kt CO2-e per facility or 50 kt CO2-e per corporate group, or 200 TJ energy consumption), you submit under this framework by 31 October each year. Records must be kept for five years from the end of the reporting year. Non-compliance carries criminal penalties up to two years imprisonment for dishonest or fraudulent behaviour, plus civil penalties calibrated in penalty units (currently $330 each from 7 November 2024).

Where AASB S2 sits

AASB S2 is the Australian implementation of IFRS S2 issued by the ISSB. It is a financial disclosure standard, sitting alongside the financial statements, and it references the GHG Protocol Corporate Standard as authoritative for scope categorisation (AASB S2 paragraph 29(a)(i)).

Paragraph 29(a)(v) requires Scope 2 to be reported using the location-based method, with market-based as voluntary supplementary disclosure. Paragraph 29(a)(vi) requires disaggregation of emissions by methodology used, which is where the reconciliation problem gets interesting.

The critical technical difference is the GWP set. IFRS S2 (and therefore AASB S2) requires the most recent GWPs from the IPCC assessment cycle, which is currently AR6 (2021). AR6 revises methane up to 29.8 for fossil sources and 27 for biogenic. Nitrous oxide moves to 273. That is not a small change for anyone with material methane exposure.

The AR5 vs AR6 GWP problem

Here is the gap. The same tonne of fugitive methane produces:

Under NGER (Method 1, AR5): 28 t CO2-e
Under AASB S2 (AR6, fossil): 29.8 t CO2-e

Same molecule, same measurement, two different disclosures. Multiply that across a gas distribution network, a landfill, or a coal mine ventilation shaft and the reconciliation gap gets material fast. AASB has published jurisdictional relief for NGER-covered portions (AASB S2025-1, December 2025), allowing NGER reporters to use AR5 GWPs for the NGER-covered subset of their emissions in AASB S2 without recalculation. That relief does not cover Scope 3 or non-NGER Scope 1 sources. The reconciliation is unavoidable at the boundary.

If your ledger stores only the final CO2-e number, you are stuck. If your ledger stores the activity quantity, the gas species, and applies the GWP at render time, you can produce both disclosures from the same source data. That architectural decision is the whole game.

The Scope 2 dual-render problem

Scope 2 is where most first-year reporters get double-counted. NGA gives you two factors per state, one for location-based (Table 1) and one for market-based (Table 2, the National Residual Mix Factor, currently 0.81 kg CO2-e/kWh). GHG Protocol Scope 2 guidance requires both to be disclosed where market-based instruments exist.

AASB S2 makes location-based mandatory and market-based voluntary. NGER accepts either method under recent amendments. Both frameworks require the same underlying kWh number.

The failure mode is straightforward. Reporter buys 10,000 MWh of GreenPower for a Victorian site. Reporter enters 10,000 MWh under Scope 2 with the location-based factor (0.78) and again under Scope 2 market-based with a zero-emissions factor. Now they have two Scope 2 records for the same electricity. If those two records both flow into the total, that is a double-count. If the reporter then also claims the LGCs under a separate voluntary Scope 2 reduction, that is a triple-count. We wrote about this failure pattern at length because it turns up in almost every first-time engagement we walk through.

The right architecture stores kWh once, then dual-renders. Location-based render uses Table 1 state factor. Market-based render adjusts for cancelled LGCs and GreenPower purchases against the residual mix. The reporter picks which is primary for their submission (AASB S2 says location-based) and the other becomes supplementary.

The Scope 3 categorisation problem

Scope 3 is where GHG Protocol and NGER diverge most sharply. NGER only mandates Scope 1 and Scope 2 for the corporate boundary, though scope 3 is tracked internally for some methodologies. GHG Protocol Category 1 is purchased goods and services, upstream of the reporting entity, and it is where most Australian companies find 60-80% of their footprint sits.

The mis-classification we see most often is this. A construction company treats subcontractor fuel as Scope 3 Category 4 (upstream transportation and distribution) when it is actually Category 1 (purchased goods and services) because the subcontractor delivered a construction service, not a transport service. The activity is the same. The category label is different. That label matters because Scope 3 categorisation drives how the disclosure is structured and what a limited assurance auditor will test against under ASSA 5010.

Where NGER touches Scope 3 (mostly for higher-tier methods and voluntary reporting), it uses NGA lifecycle factors for the same activities. But the boundary drawing is a GHG Protocol Corporate Value Chain Standard exercise, not an NGA lookup exercise.

How the reconciliation actually works in a defensible ledger

The reconciliation is architectural, not procedural. If you try to reconcile at the report-render step, you will lose. Here is the shape that works:

Store activity data at the source-document level. The BP Australia invoice records 4,850 litres of automotive diesel delivered to project site X on date Y. That is the primary record. Not a monthly aggregate. Not a spend line. The invoice itself, with its total, GST, supplier ABN, delivery address and product code, sits behind the emission record.

Apply NGA factors with version pinning. Each emission record carries the NGA edition it was calculated under (NGA 2024, NGA 2025) and the specific factor row used. When NGA 2026 is released next October, historic records are not silently overwritten. The factor version is part of the audit trail. This is not optional under ASSA 5010, which requires evidence of the factor source at the record level.

Render NGER with AR5 GWPs. For the NGER submission, gas-species records apply AR5 multipliers (CH4 at 28, N2O at 265). The NGER report generator writes to the Clean Energy Regulator's schema.

Render AASB S2 with AR6 GWPs. For the AASB S2 disclosure, the same gas-species records apply AR6 multipliers, with jurisdictional relief applied where AASB S2025-1 permits. The AASB S2 report generator writes to the disclosure schema.

Preserve methodology labels per record. Paragraph 29(a)(vi) requires disaggregation by methodology. Every record carries its methodology tag: activity-based, hybrid, supplier-specific, average-data, spend-based. When the auditor asks how much of the Scope 3 disclosure is spend-based, the answer is a query, not a spreadsheet exercise.

The same architecture supports GHG Protocol reporting for CDP, GRI Standards for voluntary disclosure, and internal management reporting. Because the categorisation lives on the record, not on the report.

What Carbonly does at this reconciliation layer

We built the platform around this reconciliation problem because it is where the frameworks actually collide.

The NGA 2025 factor library is seeded and versioned. All 193 factors, including the 21 per-litre fuel factors, sit in the material library with edition tags. When NGA 2026 lands, the library takes the new edition without overwriting the 2025 records. Every emission record pins to the factor version that was current when the activity happened.

The GWP question sits at the reporting layer rather than in the stored numbers. The ledger calculates on AR5, which is what regulation 2.02 of the NGER Regulations requires and what AASB S2025-1 permits for the NGER-covered portion of an AASB S2 disclosure, so both documents run off the same activity data with no re-entry. Both GWP tables are held per gas, and an AR6 view of whatever the relief does not cover is designed capability rather than something the report generator produces today.

The 5-tier material matching maps supplier line items to NGA factors with a Match Provenance badge that surfaces one of eight states (exact match, alias, AI suggested, manual override, and so on). That badge is what an ASSA 5010 auditor will click on to trace the emission back to the invoice. It is also what makes the Auditor Workspace and Evidence Pack export functional as evidence, rather than a formatted PDF.

Methodology labels are captured per record. Activity-based diesel from an AGL invoice looks different in the ledger to spend-based Category 1 procurement from an ERP export, even when both roll up to the same total. That distinction survives into the paragraph 29(a)(vi) disaggregation.

Consultants leading first-year AASB S2 engagements use the platform as the evidence layer under their advisory work. The workflow surfaces the source document for every number in the disclosure, which is what turns an assurance conversation from a spreadsheet argument into a records review. External AI assistants like ChatGPT and Claude can connect directly to the emission ledger through Carbonly's MCP integration, so a finance team or consultant can ask questions like "show me the Scope 2 records where market-based and location-based disagree by more than 10 percent" and get an answer from live data.

None of this makes the framework difference go away. It just moves it from a manual reconciliation problem to a reporting-layer one, with the audit trail intact.

FAQ

Is NGA the same as the GHG Protocol?

No. NGA is an Australian emission factor library published by DCCEEW, currently in its 2025 edition, containing around 193 factors. The GHG Protocol is a global boundary and categorisation standard published by WRI and WBCSD. NGA tells you what number to multiply activity data by. GHG Protocol tells you which scope and category that emission sits in.

Does AASB S2 accept NGA emission factors?

Yes. AASB S2 does not mandate a specific factor library. It requires disclosure under the GHG Protocol Corporate Standard categorisation using the most recent GWPs (currently AR6). NGA factors are acceptable and appropriate for Australian activity, provided the AR6 GWPs are applied to gas-species emissions at render time. AASB S2025-1 permits AR5 GWPs for NGER-covered portions.

What is the difference between AR5 and AR6 for methane?

IPCC AR5 (2013) sets methane GWP at 28 over 100 years. IPCC AR6 (2021) revises this to 29.8 for fossil methane and 27 for biogenic methane. Nitrous oxide moves from 265 (AR5) to 273 (AR6). NGER uses AR5. AASB S2 requires AR6. The same activity data produces different disclosures under each framework.

Do I need to submit separately under NGER and AASB S2?

Yes. If you are above the NGER threshold (25 kt CO2-e facility or 50 kt CO2-e corporate group, or 200 TJ energy), you submit to the Clean Energy Regulator by 31 October each year using the NGER Measurement Determination methodology. If you are a Group 1, 2 or 3 entity under AASB S2, you also disclose climate-related financial information under the standard. Both submissions run in parallel with their own frameworks. NGER registration automatically pulls you into ASRS Group 2.

Can one platform produce both?

Yes, if the platform stores activity data with factor version pinning and applies GWPs at render time rather than at ingest. That architectural decision separates platforms that can reconcile NGER and AASB S2 from platforms that force the reporter to maintain two ledgers. Carbonly is built on that pattern, with per-gas storage and an AR5 committed calculation; ask us, or any vendor, which GWP views the reporting layer actually produces today rather than assuming the architecture and the feature are the same thing.

What to do next

If you are inside a first AASB S2 disclosure and the NGER numbers do not tie, do not adjust the outputs. Look at the ledger. If the activity data is stored as CO2-e totals, the reconciliation cannot be automated and you will fight this problem every year. If the activity data is stored at the source-document level with methodology labels and factor version pinning, the reconciliation is a render setting.

We keep a written explainer of the NGA library with the state-by-state factor tables and a walkthrough of the AR5/AR6 methane problem that shows the reconciliation numerically. The Scope 2 calculation guide covers the location vs market-based dual-render in more detail. If you are preparing a first-year disclosure, the CFO playbook for first AASB S2 walks the process end to end.

Pricing is per-project with a $100 per month workspace minimum. Get in touch at hello@carbonly.ai.