Close popup
Close popup
Carbon Ratings

Understanding what’s happening between the forests and the trees: Updating how we rate REDD projects

January 11, 2024

Currently, the world spends less than half the amount of funding needed by 2025 on nature-based solutions. In the rapidly changing landscape of the voluntary carbon market, REDD projects play a crucial role in financing the protection of large areas from deforestation and forest degradation caused by human activities around the world. 

To help organizations ensure they're making the most effective investments, Sylvera builds software that automates the evaluation of carbon projects that capture, remove, or avoid emissions, such as REDD. We leverage the latest technology and climate science to provide our customers with the most robust data and ratings about carbon projects. 

To deliver best-in-class accuracy and depth, we recently updated our REDD rating methodology to incorporate new technology and data into our ratings assessments. With these enhancements, Sylvera’s customers drive funding to the most effective projects on the voluntary carbon market and make real climate impact. 

Even more accurate project baseline assessment

Our approach to measuring baselines has always been rooted in rigorous analysis. Over the past few years, there has been extensive debate around the effectiveness of REDD projects, highlighting serious issues with projects exaggerating baseline carbon loss and inflating their claimed emissions reductions as a result. To help  further combat some of the issues, our REDD update takes our baseline analysis further. We’ve incorporated cutting-edge data and methodologies to allow us to assess over-crediting risk more accurately. 

The relative contribution of each comparative area to Sylvera’s over-crediting risk score is dependent on the appropriateness of the project’s reference area as a proxy. With this update, Sylvera measures deforestation rates both within the project-selected reference area selected and in Sylvera-selected comparative area(s). In selecting comparative areas, we leverage the most relevant local political and economic context for the project’s location, such as transport and market accessibility, forest structure, agricultural pressures and extractive mineral prevalence, as well as geospatial data. 

Refining deforestation detection in drylands

We developed deforestation detection models bespoke to different regions and forest types, such as dense tropical forests, drylands and mangroves and incorporated these models into our REDD ratings framework. Not only is it important to correctly estimate the level of deforestation pressure for REDD projects, appropriate data must also be used to detect the deforestation itself across the world's dynamic forest ecosystems. 

For dry woodlands located primarily in Sub-saharan Africa, we improved the consistency of our Synthetic Aperture Radar (SAR)-based model and incorporated this new data into our latest framework. The model was trained on 4000 points interpreted from high-resolution satellite imagery of Tanzania and was validated against aerial LiDAR data from Miombo woodland in Mozambique collected by Sylvera. Studies show that lower biomass areas generally have higher deforestation rates than high biomass areas (McNicol et al, 2018), so to assess the baselines of Miombo forest projects, Sylvera also ensures that Sylvera-selected reference areas also match the project area biomass at the project start date.

Unparalleled carbon stock accuracy

To keep our carbon stock calculations and over-crediting risk assessments as precise and reliable as possible, we implemented cutting-edge multi-scale lidar technology and leading methods of carbon measurement. 

Sylvera field scientists have visited forests around the world, scanning them from the ground and air using our proprietary multi-scale lidar (MSL) methods. For each REDD project, we have introduced MSL-driven biomass maps used to estimate biomass density, carbon stocks, and their confidence intervals. We then compare our carbon stock data with project-reported carbon stock data and assess the likelihood of over-crediting associated with these values.

This advanced technology accurately measures forest biomass with a margin error as low as 3% (Burt et al, 2021), whereas other methods relying on allometries tend to see errors ranging from 15% to 30%. 

’The reference data being collected by Sylvera follows the gold standard for biomass, as determined by the international biomass remote sensing community, including representatives from NASA, ESA, and other space agencies. Very few datasets are publicly available with this level of accuracy, largely because they are very expensive and time consuming to collect. These are absolutely critical for mapping forest carbon stocks and fluxes, because we can never directly measure forest carbon stocks from space, and all wide area carbon maps ultimately rely on the quality of field and airborne reference data.'  Dr Laura Duncanson, GEDI Science Team, University of Maryland

Our MSL data enables us to train our models with the best reference biomass data across different biomes, from drylands to tropical rainforests. Using Synthetic Aperture Radar (SAR) data, we upscale our biomass estimations over large areas to “see” through clouds. 

Using the data and this level of accuracy keeps both over-crediting risk assessments as precise and reliable as possible.

“Accurate site-based estimates of carbon stocks are crucial in any discussion of the benefits or otherwise of carbon projects, but getting those estimates is really difficult. The only direct measurement of carbon stocks involves destructive harvesting - cutting down and weighing trees. But that's counter-productive for obvious reasons! The next best way to do it is via remote sensing of forest structure, particularly using lidar and radar, which provide accurate estimates of timber volume and mass. Sylvera's multiscale lidar approach can provide very high-quality benchmark data to provide unique insight into the accuracy of other estimates of forest carbon.”
Prof. Mat Disney, University College London, NERC National Centre for Earth Observation (NCEO)

Making it easy to navigate the ever-changing REDD landscape 

In an ever-evolving market, Sylvera strives to give its customers an understanding of upcoming risks and uncertainties for all VCM projects. To ensure our customers receive analysis tailored to all the latest REDD developments, we now have forward-looking insights in our new Credit Outlook section.

There are a slew of new programs related to REDD, from jurisdictional programs under the LEAF Coalition or ART TREES and the World Bank's Forest Carbon Partnership Facility (FCPF), to Verra's new consolidated methodology for individual REDD pivoting towards projects nested in jurisdictional baselines, and government-sponsored REDD+ nested projects. All of these evolving programs create too many developments for carbon market participants to track effectively–but our latest REDD framework makes it easy. 

Credit Outlooks tracks jurisdictional developments as well as intricate local patterns, trends, and developer engagement, offering a nuanced perspective and unlocking expected upcoming changes in issuance volumes. Together, these enable customers to stay up-to-date and make informed decisions in an ever-changing landscape.

Implementing enhanced tests

Finally, we have also introduced a comprehensive array of tests, incorporating a wealth of primary data inputs, to understand the evolving quality of projects at a more granular level. These tests encompass changes in the physical and political context of projects over time, including: 

  • More uniform, science-supported approaches to assessing the additionality of a project’s activities, over-crediting risk (including leakage) and biodiversity, as a result of incorporating more nuanced tests and robust geospatial datasets across.
  • Integrating a proprietary database of policy- and country-level analyses, land tenure disputes, and reputational risks in order to provide a holistic view of each project's regulatory landscape and community engagements.
With our REDD framework updates, we’re providing the most comprehensive, data-driven ratings to increase certainty and confidence in REDD investments and the impact they’re driving. To see all these enhanced updates, head to the Sylvera app now.
Get up to speed with "Unlocking Carbon"
Subscribe to our newsletter to get fresh insights and news on all things carbon markets.
Thank you!
Oops! Something went wrong while submitting the form.
Get up to speed with "Unlocking Carbon"

Sign up to our newsletter for the latest carbon insights.

Thank you!
Oops! Something went wrong while submitting the form.
About the author
VP Ratings

Jess Roberts: Jess is the VP of Ratings at Sylvera. She has specialised in geospatial data, with an MSc in Geographical Information Science and a BSc in Geology & Physical Geography. After her studies she worked on GIS projects with a marine conservation NGO in Greece and Ecometrica in Edinburgh. Prior to Sylvera, she was a Geospatial Advisor with Resilience Constellation working on designing data products solving key business sustainability questions.