Implications of New VCS REDD Methodology on Existing Avoided Deforestation Projects
Exciting news from our team: Sujit Gosh and Durga Prasad have published a peer-reviewed article on their latest VCS REDD+ methodology analysis. Their work offers valuable insights into REDD+ future. This is a must-read for those in the field of carbon reduction strategies.
- The paper clearly demonstrates that most of the projects will have an 80% reduction in credits under the new methodology.
- The introduction of jurisdictional reference areas and independent data service providers are welcome additions to the new redd+ methodology. This will result in a cap of credits for a jurisdiction.
- Change in the historical reference period from 10 years to 6 years will result in more up-to-date baselines.
- The adopted risk modeling approach is based on factors such as proximity to forest edges and historical deforestation patterns, which is better than current artificially inflated models though it fails to capture the local drivers and their behavior. The risk of gerrymandering and the selection of areas with high perceived risks are other potential loopholes.
Time is running out, and the slow progress we witnessed at COP28 and similar global efforts is increasing the risks to our planet. On the other hand, voluntary carbon market mechanisms are evolving. Recent events related to REDD+ projects have drawn widespread attention to the issues within the Voluntary Carbon Market (VCM), leading to a substantial decline in trading volumes for REDD+ projects.
VERRA has been working on the development of the next generation of REDD programs for the past three years. They have launched the VM0048 methodology to address some of the existing inefficiencies in the system.
The approach adopted under VM0048 will have significant implications for REDD+ projects within VERRA. To better understand these implications, we have conducted a research project on the draft version of VM0048 and published our findings in a research paper. This article is intended to provide a discussion of those findings.
The primary goal is to evaluate the effects of the new VCS REDD methodology on the baseline measurements of current REDD projects. To achieve this, we selected four existing REDD+ projects spanning four continents, each representing two major forest types. Our methodology included several steps: generating activity data for land cover changes, creating Forest Cover Benchmark Maps (FCBMs), developing Risk maps, and allocating jurisdictional activity data to the project area. We then compared these new ex-ante estimates to the existing baseline estimates of the projects using the old methodologies.
Shorter Historical Reference Period
The previous requirement of having a 10-year historical reference period has been modified. The new methodology suggests shorter historical reference periods, recognizing that deforestation rates can rapidly change due to policy implementations. This adjustment ensures that baseline emissions remain relevant throughout the project. The methodology now adopts 6-year baselines, but there's potential for even more ambitious baselines in the near future, thanks to advances in open data and AI/ML models.
The length of the historical reference period to be selected for past deforestation observation varied in every project. It ranges from 8 years for VCS1689 to 15 years for VCS1686. The fixed 6-year historical reference period brings standardization.
Independent Data Service Providers
The addition of Data Service Providers (DSPs) to the program represents a significant change. In the past, project developers were responsible for both the project data and models, creating a conflict of interest. Under VM0048, independent DSPs will supply the underlying data to VERRA, with expert reviews and an emphasis on collaboration and engagement with local and national governments.
Caps Through Jurisdictional Reference Areas
The introduction of jurisdictional reference areas is a welcome addition to the new redd+ methodology. The reference region has huge implications for the resulting deforestation rate; when a larger area was taken into account for establishing the baseline, the deforestation rate either decreased or remained at a similar level. For instance, consider the reference region of VCS1622, which covers 1,635,426 hectares. In our study, we selected a jurisdictional level encompassing an area of 5,506,799 hectares. As a result, the deforestation rate decreased from 3.41% to 1.24% in this case.
In contrast, for VCS1689, the jurisdictional area was smaller than the reference region for the project. Consequently, the deforestation rate increased from 1.59% to 2.06% for VCS1689.
It is clear that the clear definition of reference regions will result in a cap of credits for a jurisdiction. The paper clearly demonstrates that most of the projects will have a huge reduction in credits under the new methodology.
Standardized Risk Model
In terms of Deforestation Estimation and Risk Mapping, the new approach relies on a risk modeling method based on proximity to past deforestation. This standardized approach will be applied to all projects, reducing the risk of projects using overly inflated models. However, it's worth noting that this approach has its challenges, including its simplicity and the fact that it doesn't consider local drivers. In the final version of the risk mapping tool VERRA might address those challenges. Additionally, there's a risk of gerrymandering and selective project locations with high perceived risk.
In the new methodology, three out of the four projects examined in the study experienced a significant reduction in the number of issued credits compared to previous methodologies. This indicates a shift toward a more realistic and stringent approach to modeling emissions reductions.
Substantial Lower Projections From New Methodology
To illustrate this, let's look at the allocated emission values for the 2021–2026 period compared to baseline emissions estimated using the old methodology:
VCS1686: Baseline emission numbers underwent a substantial reduction. The old baseline allocated over 2.7 million tCO2e, whereas the new methodology reduced this to 0.54 million tCO2e, which represents only 20% of the initial baseline.
VCS1622: This project showed the most significant reduction in baseline emission allocation. The old baseline estimated emissions at around 11 million tCO2e, but according to the new methodology, it decreased to 0.7 million tCO2e.
VCS1689: In this case, the reduction in baseline emissions was not as dramatic. The old baseline allocated emissions at 2.8 million tCO2e, whereas the new methodology reduced it to 1.9 million tCO2e.
VCS1168: Interestingly, this project saw an increase in baseline emissions when adopting the new methodology. Baseline emissions grew from 3.1 million to 5.1 million.
It's essential to note that these numbers are ex-ante modeled figures for both the individual projects and the study itself. These figures should not be confused with issued credits.
Overall, the new methodology establishes a standardized approach to accounting and data generation, reducing the risk of diverse interpretations and making it easier to identify bad actors in the system.
