Matthew D. Potts, PhD

Chief Science Officer

Contributed by Dr. Barbara Haya, PhD, director of the Berkeley Carbon Trading Project at the UC Berkeley Goldman School of Public Policy

Last month, I, along with nine other current and former members of Carbon Direct’s Science Team, co-authored an article in Frontiers in Forests and Global ChangeComprehensive Review of Carbon Quantification by Improved Forest Management Offset Protocols. Drawing on careful study of the scientific literature, and also informed by our earlier experience reviewing numerous improved forest management projects for Carbon Direct, we assess the accuracy and conservativeness of the eight improved forest management offset protocols that were active on the major voluntary registries as of the end of last year. We found that all of these improved forest management protocols, used by American Carbon Registry, California Air Resources Board, Climate Action Reserve, and Verra to issue carbon credits, diverge from good practice across three major factors affecting carbon credit quality—additionality/baselines, leakage, and durability.

Improved forest management has significant potential to remove and hold carbon from the atmosphere. Extending rotations in managed forests can increase forest carbon while maintaining or even increasing timber production; reduced-impact logging in tropical forests can reduce forest degradation and increase or preserve soil carbon; better management can make forests less susceptible to wildfire, drought, and pests; restoration activities in degraded forests can increase both carbon and biodiversity.

Each of the protocols defines methods for estimating the impact, and the carbon credits generated, from improved forest management activities including those described just above. Under these protocols, projects define their baseline scenario (what would likely have happened without the carbon financing) and measure the project impact as the difference between actual carbon stocks on the project lands and modeled stocks in the baseline scenario.

After measuring this difference, several deductions are taken. For projects that reduce timber harvesting, a deduction is made to cover the emissions associated with leakage, as any project that reduces timber harvesting is assumed to cause an increase in timber harvesting somewhere else to meet timber demand. A second deduction covers the risk of reversal, such as from fire, disease, and wind damage. This deduction is placed into an insurance buffer pool that can be used by any project to cover a reversal if one occurs. Finally, a deduction is taken for model uncertainty.

Our review found that most improved forest management projects claim that without carbon finance they would significantly increase the amount of timber harvesting, reducing on-site carbon stocks to a baseline scenario far below initial levels. Most credits to date reward forest owners for not doing so. We explain how this flexibility in baseline setting compounds with methods for assessing leakage and insuring against the risk of reversal that diverge from the published literature, raising the risk of significant over-crediting. Our paper provides specific recommendations for methodological improvements.

The baseline: The single most important change the protocols can make to improve quality and reduce the risk of over-crediting is to improve the way that baselines are established. The baselines commonly used by projects that are well below initial carbon stocks would accurately reflect project benefits if carbon finance truly helps preserve forest stocks that would otherwise have been reduced (except for over-stocked forests in high fire-risk areas (see this article in the same volume)). However, the true baseline is often highly uncertain. The flexibility protocols provide to choose baselines far below initial carbon stocks, when developers benefit financially from that choice, creates a high risk of over-crediting. Forest owners with high levels of carbon per acre can participate and earn credits from making an argument that they would have aggressively harvested, whether or not they really would have.

Protocols could mandate more conservative methods to establish project baselines, aligning baselines with initial or past carbon stocks, and net present value for lands managed primarily for profit from timber. We also suggest that dynamic baselines may provide accurate crediting in the future, although these are yet to be tested in practice. See this blog for a state of the art way to establish project baselines dynamically based on reference plots.

The Climate Action Reserve’s (CAR’s) Mexico improved forest management protocol uses initial stocks as the baseline, which eliminates potentially poor quality avoided reduction credits. We also note that since our article was submitted, a new version of the CAR Mexico improved forest management protocol resolves the most important quality issue we identify in the paper about that protocol; instead of allowing for crediting periods as short as one year, the minimum landowner commitment is now 30 years. This reduces the ability to choose crediting periods that match harvesting plans.

Leakage – California Air Resources Board and CAR can correct a contradiction in their baselines that credits avoidance of logging to the baseline in the first year of the project but deducts the leakage associated with that avoidance evenly over 100 years. This can be simply corrected by deducting leakage at the same time that avoided logging is credited. Protocols can require higher leakage rates to bring their methods into closer alignment with the existing, though sparse, literature and for conservativeness given high levels of uncertainty in true leakage rates.

Durability – All protocols should increase reversal risk estimates and allocation of credits to the insurance buffer pool to better cover current reversal risk and account for increased risk with climate change. This is especially true for parts of the world that have seen significant increases in wildfire frequency and severity over the past decade.

Please see our paper for more detailed discussions of each of the quality elements for all studied protocols.

Improved forest management offsets are the third most important offset project type on the major voluntary registries in terms of credits generated to date, and have generated 45% of credits from US-based projects. Given the importance of improved forest management, and nature-based solutions generally, to both climate mitigation and the integrity of the carbon offset market, the quality issues raised have consequences. To the extent that projects over-credit, purchases can take the place of more effective measures, and that over-crediting helps keep carbon credit prices too low to support real forest management changes that have real climate benefits.

As the registries develop new improved forest management protocols and evaluate existing ones, this synthesis of the literature and the articles on improved forest management offset quality which we cite and describe, should inform improved protocols that ensure quality. In the meantime, these insights can help project developers to design quality improved forest management offset projects, and credit buyers to identify them.

Barbara Haya, PhD is a researcher at the University of Berkeley and director of the Berkeley Carbon Trading Project at the UC Berkeley Goldman School of Public Policy.

Dr. Haya’s latest research includes a comprehensive assessment of all improved forest management protocols that was coauthored with nine other scientists, including Carbon Direct Science Advisory team members Letty Brown, Jacob Bukoski, Van Bustic, Bodie Cabiyo, Amber Kerr, Dan Sanchez, and Matthew Potts. Their findings were recently published in Frontiers in Forests and Global Change and covered by Bloomberg.