Refrigerant Emissions: The Scope 1 Number Most Australian Companies Get Wrong

A typical mid-sized Australian retailer with 50 stores has somewhere between 200 and 600 individual refrigeration and air conditioning units across its sites. Most are running R-410A or R-404A. The first has a global warming potential of 2,088. The second sits at 3,922. A 5kg leak of R-404A is roughly 19.6 tonnes CO2-e. Across a 50-store portfolio with average 15% annual leak rates, the refrigerant emissions easily exceed the company's entire fleet.

Yet in most companies the refrigerant figure lives in a spreadsheet maintained by the facilities manager, calculated once a year using a back-of-envelope estimate. The sustainability team accepts the number and moves on. The auditor, when AASB S2 assurance kicks in, will not.

Refrigerants are the Scope 1 gap. Worse, the NGER reporting threshold for refrigerant emissions only captures equipment with a charge above 100 kg using gases with GWP above 1,000. That cutoff leaves most commercial HVAC and small refrigeration units invisible to the regulator, but the emissions are still real, still in your operational footprint, and still subject to AASB S2 disclosure.

What the NGER threshold actually leaves out

The NGER refrigerant reporting rule is narrow on purpose. It captures large industrial refrigeration and major commercial air conditioning. Specifically, you report under NGER only if:

• The unit is used in commercial air conditioning, commercial refrigeration, or industrial refrigeration
• The unit has a total refrigerant gas charge above 100 kg
• The refrigerant has a global warming potential above 1,000

Nearly every supermarket cold room and many shopping centre chiller plants exceed the 100 kg threshold. Most office air conditioning, smaller retail refrigeration, and almost all transport refrigeration units do not. The Ecolibrium analysis published mid-2025 estimated that Australia's refrigerant emissions are systematically underreported because of this threshold combined with reliance on estimated leak rates and the absence of a national leakage register.

For NGER purposes, you can ignore the small units. For AASB S2 purposes, you cannot. Paragraph 29 requires absolute Scope 1 emissions disclosure, and refrigerants sit firmly in Scope 1. The auditor will ask whether the small units are material. If they aren't material individually but are material in aggregate, they have to be in the disclosure.

The four sources of refrigerant emissions

There are four ways refrigerants enter your Scope 1 account, and each requires a different data source:

1. Annual operating leak. Refrigerants leak slowly from connections, valves, and compressors. For commercial refrigeration the DCCEEW guidance suggests typical leak rates of 10 to 25% per year. For chillers, 5 to 15%. For automotive air conditioning, 5 to 10%. The data source is service records: how much gas was added during routine maintenance.

2. Top-up after equipment failure. When a unit fails and is repaired, the recharge volume is the leak. Service records or invoices from refrigeration mechanics capture this.

3. End-of-life loss. When equipment is decommissioned, any residual refrigerant that wasn't properly recovered is an emission. Decommissioning records, if they exist, capture this. Often they don't exist.

4. Charge of new equipment. New equipment installation involves transfer of refrigerant from cylinder to system. There's typically a small operational loss in this process. Installation records capture this.

The cleanest single data source is the refrigeration mechanic's service invoices. They show the gas type, the kilograms added, the equipment serviced, and the date. If you process those invoices through your emissions ledger the way you process electricity bills, refrigerants stop being a once-a-year guess and become a continuous Scope 1 line item.

The GWP version problem

NGER reporting uses AR5 GWP values for refrigerants. AASB S2 paragraph 29(a) requires GWP values consistent with the latest IPCC assessment, which is currently AR6. The two are different.

For R-410A, AR5 puts the GWP at 2,088. AR6 puts it at 2,256. For R-404A, AR5 is 3,922; AR6 is 4,728. R-134a (still common in older equipment): AR5 is 1,300; AR6 is 1,526.

The same kilogram of leak produces a different CO2-e number depending on which framework's view you're producing. We've covered the AR5 vs AR6 GWP wrinkle in detail before. For refrigerants it matters more than for any other Scope 1 source, because the GWPs are so much higher than CO2.

The AASB S2025-1 amendments issued in December 2025 introduced limited jurisdictional relief: NGER reporters can use AR5 GWPs in the AASB S2 disclosure for the NGER-covered portion of their emissions. The relief does not extend to the non-NGER refrigerant emissions (the small units below the 100 kg threshold). For those, you have to use AR6.

This creates the awkward situation where a single shopping centre might have:

• A 200kg chiller plant reported under NGER using AR5 (relief applies)
• 30 small split systems reported under AASB S2 only, using AR6
• All of them sitting in the same scope 1 disclosure with two different GWP versions

The disclosure has to make this clear. The methodology note has to explain it. The auditor will check both.

What the data system needs to capture

To produce an audit-grade refrigerant emissions number, your system needs:

The hardest field is "annual top-ups". For a 50-store retailer, that's potentially 50 service contractors, each issuing 4 to 12 invoices per year, often as PDFs, often with the kilograms buried in line item detail rather than called out on the invoice header.

This is exactly the pattern we built our document engine for. Service invoices look like utility bills: structured PDFs with quantitative data buried in unstructured layouts. Reading them programmatically and extracting the gas type, the kilograms, the equipment, and the date is a tractable problem if you've solved it for 200 different retailer formats.

What the auditor will check

When ASSA 5010 assurance kicks in for your AASB S2 disclosure, the auditor will run a specific test on refrigerants:

1. Asset coverage. Does the disclosure cover all refrigeration and air conditioning equipment, or just the NGER-reportable items? If just NGER-reportable, what's the materiality argument for excluding the rest?
2. Leak rate basis. Is the leak rate based on actual top-up data, or on default assumptions? Default assumptions are penalised under reasonable assurance.
3. GWP version. Is the version stated, and is it consistent with the methodology note? Mixed AR5/AR6 has to be explained.
4. End-of-life treatment. Is decommissioning loss accounted for? If equipment was retired during the period, where did the refrigerant go?
5. Reconciliation to NGER. Does the AASB S2 refrigerant figure reconcile to the NGER refrigerant figure for the NGER-reportable subset?

Failure on any of these is restatement territory.

The DCCEEW GWP limit proposal

The Department of Climate Change, Energy, the Environment and Water has proposed phased GWP limits on refrigerants used in new equipment, aligned with the Kigali Amendment. The direction is toward lower-GWP alternatives like R-32 (GWP around 675), R-454B (around 466), and natural refrigerants like CO2 and ammonia.

For most companies the implication is operational rather than reporting. As old R-404A and R-410A equipment is replaced with lower-GWP alternatives, the same physical leak generates dramatically less CO2-e. A retailer that systematically replaces R-404A with CO2 transcritical systems can reduce refrigerant emissions by more than 95%, equipment by equipment. That's a measurable Scope 1 reduction that shows up cleanly in transition plan disclosures.

Practical sequence for getting refrigerants right

If your refrigerant number is currently a once-a-year estimate from facilities, here's how to fix it:

1. Build the asset register. Every unit, every site, refrigerant type, charge size. This is the spine.
2. Centralise service invoices. Either via email forwarding to a single inbox or via OneDrive sync from each contractor.
3. Process invoices through your emissions ledger. Extract the kilograms added per service, tagged to the unit.
4. Calculate two GWP views. AR5 for NGER, AR6 (or hybrid) for AASB S2.
5. Set anomaly thresholds. A 25 kg refill on a 30 kg unit means something failed. Flag it.
6. Document the methodology. Write down which equipment is in scope, which leak rates are observed vs assumed, and how the calculation works.

The first time you do this you'll find emissions that weren't on the books. Often the increase is 20 to 40% of the previously-reported Scope 1 total. That's not a problem. It's a corrected baseline. The companies that get caught are the ones that pretend the previous number was right and then have to restate when the auditor finds the gap.

The bottom line

Refrigerants are the most concentrated Scope 1 emissions source most companies have, and the worst-tracked. NGER's narrow threshold doesn't help. AASB S2 closes the gap at the disclosure end. The data discipline lives in service invoices nobody has been processing systematically.

If your sustainability team is producing refrigerant numbers from a back-of-envelope estimate and you know AASB S2 assurance is coming, email hello@carbonly.ai or join the waitlist. We'll show you what an audit-grade refrigerant ledger looks like and how it integrates with the rest of your Scope 1 reporting.