Speaker
Description
Carbon markets, operating on the principles of emission trading and carbon offsetting, offer market-based incentives to reduce emissions and to adopt low-carbon practices. New rules from leading registries such as Verra, Gold Standard, and Isometric are advancing digital Measurement, Reporting, and Verification (dMRV) systems to enhance transparency and reduce monitoring costs in carbon offset projects by utilizing Earth Observation data to track carbon sequestration progress. Yet, the capabilities offered by remote sensing to support carbon markets have yet to be fully understood. Existing research data products are usually developed not considering the specific requirements of carbon markets. Not being tailored to the local site conditions, they are typically affected by inaccuracies, undermining their reliability.
The ESA-funded project CarbonEye aims at the development of transparent EO-based approaches for monitoring the progress of Verified Carbon Standard (VCS) projects by integrating information from localized ground data into an established and fully transparent above-ground biomass (AGB) mapping algorithm using decadal timeseries (years 2015 to 2025) of dual-polarization C- and L-band radar backscatter observations from the European Sentinel-1 and the Japanese ALOS-2 PALSAR-2 missions to deliver accurate time series of forest biomass maps. The CCI Biomass approach (Santoro et al., 2024) has been selected because of its physical model basis and the flexibility in adjusting its components to different forest structures while minimizing the need for contemporary, frequently updated in-situ data. Here, we discuss the performance of the modified CCI Biomass approach for seven VCS afforestation projects in Ghana, Sierra Leone, Uganda, Paraguay, Uruguay, and the Baltic states, which were selected because characterized by open datasets of carbon stocks in verification and validation reports in the Verra registry (the world’s largest carbon-offset registry).
A validation of AGB retrieval results obtained with the adapted CCI Biomass approach when using different combinations of multi-temporal C- and L-band co- and/or cross-polarization backscatter observations with plot- or stand-level AGB ground data indicated the reliability of the approach to map carbon stocks. Yet, the AGB mapping tends to require solutions that are somewhat tailored for each individual plantation site. Complicating factors were that i) projects participating in the carbon market often plant non-native, fast-growing tree species with structural properties that differ substantially from indigenous forests in the region, and ii) notable differences in the backscatter to AGB relationships could be observed also between species that are commonly planted (e.g., different types of Eucalyptus, Pine, Teak, or Acacia). Species related effects in the backscatter to AGB relationships did not affect all frequencies and polarizations equally. At several of the project sites, a combination of both frequencies and polarizations helped to overcome limitations (e.g., biased estimates) associated with a single frequency/polarization. At other project sites, instead, unbiased AGB estimates could only be achieved with a single frequency and polarization (e.g., L-band HV polarization). Different combinations of the available time series of C- and L-band backscatter observations therefore have to be considered when producing a decadal stack of AGB maps for each project area.
The trajectories of AGB and carbon stocks are to be compared to numbers released by the projects in verification and validation reports to assess the true potential of the suggested approach as an independent EO-based monitoring tool. First results of the map validation will be presented at the conference.