CSIRO Veteran Warns Minns Koala Park Carbon Case Lowballs Forest Costs

The consultancy behind New South Wales’ $1.5 billion forest carbon revenue claim assumed public native forests could be managed for $15.90 per hectare per year, a third below what the Forestry Corporation spends and a sixth of what national parks have cost to run. That is according to former CSIRO chief research scientist John Raison and Forest and Wood Communities Australia chair Steve Dobbyns, whose open letter was sent yesterday to Premier Chris Minns, Climate Change and Energy Minister Chris Bowen, Environment and Water Minister Murray Watt, and Assistant Minister Josh Wilson. Wood Central has been granted permission to publish the open letter to Premier Minns, Ministers Bowen and Watt, and Assistant Minister Wilson in full:

Open letter to NSW Premier Minns, Federal Ministers Bowen and Watt, Assistant Minister Wilson

Carbon revenues and the long-term cost of managing INFM projects require greater scrutiny.

By Dr John Raison, former Chief Research Scientist, CSIRO and Steve Dobbyns, BSc (For), Forest & Wood Communities Australia.

Claims that ending timber harvesting in NSW public native forests could generate about $1.5 billion in gross carbon-credit revenue over a 15-year crediting period have become a central part of the case being advanced for the new Improved Native Forest Management in Multiple-use Public Native Forests carbon Method (INFM).

The figure originates from economic modelling by Mandala (2024) for the Australian Climate and Biodiversity Foundation. It is based on a claim that an average of 1.5 million Australian Carbon Credit Units (ACCUs) could be created annually over the 15-year crediting period and a central price scenario reaching $67 per ACCU by 2030. Here we examine whether those assumptions are realistic and whether the long-term costs of managing INFM projects have been fully allowed for in the Mandala study.

Mandala did not independently calculate the potential generation of ACCUs from ceasing harvesting in NSW. It relied on estimates in a September 2023 Barraband Consulting report commissioned by the Australian Climate and Biodiversity Foundation. The Barraband report identifies A. Macintosh as Barraband Consulting’s point of contact for correspondence. The modelling predates the final 2026 INFM Method and used assumptions including 5% leakage and a 5% reversal buffer. It did not include important substitution benefits of using wood products. Because the final Method differs significantly from the earlier modelling, ACCU estimates should be recalculated transparently before they are used to estimate revenues from INFM projects.

There are three critical issues that require strong scrutiny:

1. Are the ACCUs generated using the INFM method reliable, and how many could be created?
2. What is the likely revenue generated (the number of reliable ACCUs multiplied by their price)?
3. How much of the costs of the establishment and long-term management of an INFM project could be covered by ACCU revenues?

Issues related to the creation of reliable ACCUs

The key issue is not simply whether the statutory method generates ACCUs, but whether the credits calculated by the method represent an equivalent amount of real net climate abatement, and thus whether the revenue being promoted should be granted.

Assessing the climate consequences of changes in native forest management requires consideration of the entire forest and wood-products system through a full life-cycle assessment (LCA). The LCA needs to consider changes in forest carbon stocks; carbon stored in wood products in service and landfill; emissions from harvesting, transport and processing; use of residues for energy; substitution of timber for more emissions-intensive materials such as steel, aluminium and concrete; and the consequences of replacing Australian timber production with imported products. Research by Ximenes et al. (2016) and Raison (2024) demonstrates why these factors matter, and that when they are accurately taken account of, sustainable harvesting of native forests does not result in net carbon emissions.

The fact that the INFM accounting rules calculate positive abatement from ceasing harvesting does not mean that an equivalent reduction in net atmospheric emissions occurs. The INFM is incomplete in that it does not include the benefits of substituting wood for high emissions-intensive materials, and does not accurately model other important factors affecting the carbon balance of harvested forests.

Forestry Australia (FA), the professional body representing Australian forest scientists and foresters, has undertaken detailed technical analysis of the final method. Its forest-carbon experts concluded that the INFM could potentially over-credit projects by 40 to 100 per cent, even before the full effects of leakage and carbon stored in harvested wood products are taken into account (Forestry Australia). FA concerns include the treatment of forest carbon dynamics following selective harvesting, the application of FullCAM modelling and the accounting timeframes used in the method. A systematic modelling bias of the magnitude identified by FA should be addressed, not just dismissed.

Leakage may be much greater than the INFM method allows. Ceasing native forest harvesting does not stop demand for hardwood. Recent research by Venn and colleagues (Venn et al. 2026) found that 81% of the reduction in Australian native forest hardwood production has historically been substituted by imports. They also found that about 90% of that displaced demand came from countries with elevated risks associated with illegal harvesting. Imported wood carries risks of very high ‘production’ emissions (Ximenes et al. 2016; Venn 2023) resulting from unsustainable practices including forest degradation and disturbance of organic (peat) soils. It also has higher emissions from transport than local hardwood.

The INFM specifies a requirement for an independent assessment of indirect leakage, including emissions increases outside Australia. The method sets a cap on indirect leakage of 40% of the residual abatement calculated after other leakage deductions. This creates an important integrity question: if empirical evidence (Venn et al., 2026) indicates that 80% of reductions in Australian hardwood production are replaced by imports with potentially high production-related emissions, what evidence was used to support a maximum 40% indirect-leakage deduction?

There is also an unrealistic assumption in the INFM that an independent assessor can reliably estimate carbon emissions caused by wood imports, because the precise source/harvest location of imported hardwood is unknown.

Considerable doubt on the veracity of the ACCU estimate used by Mandala is provided by a comparison of it with the carbon contained in harvested logs from native forests in NSW. ABARES (2024) report harvest of 650,000 cubic metres of native forest logs from NSW public and private forests in 2022-23. Each cubic metre of wood approximately equates to 1 tonne of CO₂-equivalent (or 1 ACCU) and on average 70% of felled biomass is recovered as logs (Raison 2024). These numbers combine to provide an estimate of the total carbon contained in the felled biomass associated with that annual production (Raison et al. 2026) as approximately 0.93 Mt CO₂-e (650,000/0.7).

The 0.93 Mt CO₂-e is not necessarily an absolute upper limit on INFM credits because the method also credits modelled increases in forest carbon stocks over time in the project (unharvested forest) scenario. However, the 0.93 Mt estimate includes wood harvested from private forests and there is no deduction to account for carbon remaining in harvesting residues or stored in harvested wood products, so it is very likely that net carbon emissions from harvesting in public forests will be much less than this.

Mandala’s (2024) estimate of an average 1.5 million ACCUs a year is about 60% greater than the 0.93 Mt CO₂-e estimate for the felled biomass associated with annual NSW public and private native-forest log production. The comparison illustrates the importance of independently evaluating the projected ACCU quantity. The assumptions used when estimating ACCUs under the INFM should be publicly available and independently tested before they are used to create revenues.

Issues related to potential revenues derived from ACCUs

The forecast revenue is highly price sensitive. The Mandala (2024) central case assumed carbon prices rising to $67 by 2030 (NationBuilder). Generic ACCUs were trading at $38.65-$38.70 in early September 2026 (Demand Manager); Mandala’s assumption is 70% above the current price. At $38.65, 1.5 million ACCUs would have a gross value of about $58 million a year.

The total revenue over the 15-year crediting period is highly dependent on both the quantity of reliable credits generated and future carbon prices. The 1.5 million ACCU annual estimate has not been confirmed using the final (2026) Method, while the $67 price is a forecast that is much greater than the current market price. The forecast $1.519 billion gross revenue is thus an unverified modelled scenario, not a guaranteed return.

The INFM provides a crediting period of just 15 years, while requiring projects to have a 100-year permanence period. Put simply, the project can generate ACCU revenue for 15 years, while the cost of managing the land continues for a further 85 years after that ACCU revenue stream ends.

This has important implications for claims that INFM revenue can provide a long-term replacement for the economic activity associated with forestry and fund forest management for the next century. Rod Campbell (2026) of The Australia Institute has estimated that at an ACCU price around $40, carbon income attributable to the Great Koala National Park (GKNP) would cover only 5% of estimated park-management expenditure over 100 years. The estimate is illustrative rather than a detailed GKNP operating budget, but it highlights the mismatch between a short crediting period and the much longer costs of managing forests (Wood Central, 2026). Costs not met by ACCU revenue would need to be funded from other sources, including potentially future public funding.

Potential carbon revenue should also be evaluated in the context of the existing hardwood economy. Ernst & Young (2023) estimated that in 2021-22 the NSW hardwood industry generated approximately $2.9 billion in gross revenue, $1.1 billion in Gross Value Added and supported about 8,900 full-time-equivalent jobs. Northeast NSW accounted for approximately $1.84 billion in revenue, $700 million GVA and 5,700 FTE jobs. EY reported that more than three quarters of NSW hardwood timber supply came from native forests and that State forests supplied about 67% of hardwood harvesting across the study regions. Those gross economic-contribution figures are not estimates of the amount that would automatically be forgone if harvesting ceased. Unlike INFM credits, however, the existing hardwood economy is not limited to a 15-year crediting period.

Any credible policy assessment must consider the carbon project, timber substitution, imports, downstream manufacturing, regional employment and future taxpayer costs together, not treat carbon-credit income in isolation.

A comparison of the long-term costs of managing unharvested and wood production forests

Mandala (2024) assumed a forest management cost of $15.90/ha/year across about 1.33 million hectares, giving a forest-management cost of about $21.1 million a year, or $317 million over the 15-year crediting period. That $317 million was deducted, along with monitoring and audit costs, from the central $1.5 billion revenue from ACCUs, resulting in $1.136 billion net income.

Rod Campbell (2026) considered the forest management cost of $15.90/ha/yr to be unrealistically low. Our analysis summarised below confirms his concerns.

Forestry Corporation NSW’s published Native Forest Land Management figures over the five years prior to 2024-25 imply average gross management expenditure of approximately $48 million a year. Across the managed estate of 2 million hectares, this is about $24/ha/year. Individual annual figures fluctuate and recent years include exceptional expenditure such as flood-related infrastructure repairs, so the five-year average is a more useful medium-term cost. The observed average cost of managing the public native State forest estate is about 50% higher than Mandala’s $15.90/ha assumption.

We also reconstructed audited National Parks and Wildlife Service (NPWS) service-group expenditure for five consecutive years from 2012-13 to 2016-17. Expressed in June 2025 dollars, that historical expenditure is approximately $91-$98/ha/year. This is broader than Mandala’s narrowly defined INFM costs, but it is directly relevant where State forests are transferred into the national park estate. For the proposed GKNP, the NSW Government says around 100 new NPWS staff will be required for enhanced fire management, feral animal and weed control, ecological rehabilitation and visitor management; most existing recreational activities are expected to continue; and new visitor infrastructure, enhanced campgrounds and walking trails are planned. An INFM project would add carbon-accounting, monitoring, reporting, verification and permanence obligations on top of those park-management functions and costs.

Applying the above cost estimates to Mandala’s 1.33 million hectares provides a useful comparison:

If Mandala’s $15.90/ha assumption is replaced with the approximately $24/ha FCNSW five-year average, and leaving all its other assumptions unchanged, the 15-year management cost rises by about $162 million. This reduces Mandala’s central claimed net income from approximately $1.136 billion to about $974 million.

The NPWS figures are relevant where the intended future tenure is national park, although they should not be used as a precise forecast. Applied to Mandala’s 1.33 million ha management area, the historical NPWS management costs equate to approximately $121-$130 million a year, or about $1.82-$1.96 billion over 15 years. Those costs would exceed Mandala’s estimated central $1.519 billion gross ACCU revenue even before separate costs of meeting INFM obligations are added.

Our FCNSW analysis showed that the majority (81%) of the Hardwood Division’s reported FY 2024-25 deficit was attributable to Native Forest Land Management costs rather than to costs of producing Timber & Extractive Resources. Ceasing harvesting does not make those public-land-management obligations disappear. The costs of fire management, roads, pests, weeds, environmental management, public access and much of the associated staffing and infrastructure remain, while timber revenue and its contribution towards those fixed costs disappear. The appropriate comparison for taxpayers is the net cost of managing state forest for timber production with associated commercial revenue versus the full cost of managing the alternative tenure after that revenue is removed. That full counterfactual does not appear to be what Mandala modelled.

Mandala’s headline $1.136 billion net-income estimate is very sensitive to the forest management cost assumption it used. Its $15.90/ha figure is about one-third below the approximately $24/ha FCNSW five-year average and around one-sixth of the historical audited NPWS benchmark. The future management context will include the higher costs of national park management as well as additional INFM project obligations. Further, Mandala’s model deducts management costs over the 15-year crediting period, while land management costs will continue throughout the 100-year permanence period.

Conclusions

Climate policy and carbon markets depend on integrity. If one ACCU is used by another party to offset one tonne of emissions, that credit should represent a genuine tonne of additional abatement. The available evidence raises serious doubt whether the volume of ACCUs promoted for NSW will correspond to an equivalent quantity of additional net carbon abatement, and whether the claimed net revenues adequately account for long-term forest management costs. Those concerns have not received independent scientific and financial scrutiny.

The modelling underpinning the INFM and claimed benefits needs to be transparent to enable public scrutiny: showing precisely how integrity concerns are addressed when estimating ACCUs, as well as forest-management budgets and funding arrangements for the full 100-year permanence period. This is required before projected INFM revenues are relied upon to justify major changes in the management of public native forests in NSW.

References and further supporting material

• ABARES. 2024. Australian forest and wood products statistics. Department of Agriculture, Fisheries and Forestry.
• Australian Bureau of Statistics. 2025. Consumer Price Index, Australia, June Quarter 2025.
• Australian Government Department of Climate Change, Energy, the Environment and Water. Improved Native Forest Management (INFM) method.
• Barraband Consulting. 2023. Improved Native Forest Management (INFM): Abatement and Carbon Credit Estimates from Cessation of Native Forest Harvesting. Unpublished report, 20 pp. No publicly available copy located, but referenced in the following.
• Campbell, R. 21 August 2026. The Point.
• Carbon Credits (Carbon Farming Initiative-Improved Forest Management in Multiple-use Public Native Forests) Methodology Determination 2026.
• Demand Manager. Certificate Prices – ACCU spot market data (September 2026).
• Ernst & Young. 2023. Economic Contribution Study of the NSW Hardwood Timber Industry. Forest & Wood Products Australia / North East NSW Regional Forestry Hub.
• Forestry Australia. 2026. The maths does not add up: forest carbon method risks millions of low-integrity credits. 30 July 2026.
• Forestry Corporation of NSW. Annual Report 2024-25 and annual report archive.
• Mandala. 2024/2025. Implementing High Integrity Forest Carbon Projects in NSW: Costs, returns, and income reinvestment scenarios. Unpublished report commissioned by the Australian Climate and Biodiversity Foundation.
• NSW Government. Great Koala National Park.
• NSW Office of Environment and Heritage. Annual reports archive.
• NSW Department of Premier and Cabinet. Annual Report 2013-14.
• Raison, R.J. 2024. A review of the impacts of sustainable harvesting, non-harvest management and wildfire on net carbon emissions from Australian native forests. Australian Forestry 87(4):176-194.
• Raison, R.J., Nambiar, E.K.S., Kile, G.A. & Bren, L.J. 2026. Australia’s native forests can be sustainably managed for wood production together with other important forest values. Australian Forestry.
• Venn, T.J. 2023. Reconciling timber harvesting, biodiversity conservation and carbon sequestration in Queensland, Australia. Forest Policy and Economics 152:102979.
• Venn, T.J., Huang, Z. & Lachhwani, J. 2026. An Analysis of Australia’s Solid Wood Timber Production and Imports. North East NSW Regional Forestry Hub.
• Wood Central. 11 September 2026. NSW’s $1.5 Billion Carbon Claim Outstrips Every Tree the State Fells.
• Ximenes, F., Roxburgh, S., Cameron, N., Coburn, R. & Bi, H. 2016. Carbon stocks and flows in native forests and harvested wood products in SE Australia. Forest & Wood Products Australia.