Forest Carbon Diligence: Digital MRV for Jurisdictional and Voluntary Offsets Markets

Christopher Benjamin Anderson , Maxwell B. Joseph, Camile Söthe, Flávia De Souza Mendes, Thomas Maschler, Ryan C. McCarthy, Joseph Mascaro, Tara O'Shea, Amy Rosenthal, David C. Marvin

Forest Carbon Diligence is a digital MRV system for mapping forest structure and forest carbon over time. This manuscript describes the qualities of the Diligence datasets — annual maps of canopy cover, canopy height, aboveground carbon density, change detection, and their uncertainties — and quantifies performance across multiple MRV contexts.

Canopy height and canopy cover regression metrics were high (height r2=0.83, cover r2=0.79 over NEON), as were aboveground carbon metrics (GEDI L4A r2=0.64 at 30 m, GEDI L4B r2=0.82 at 1 km). Carbon intercomparisons found strong agreement among 8 independent datasets. Agreement was highest with GEDI, showing strong correlations at multiple scales (r=0.92 for 1 km comparisons, and r=0.98 for 1 degree, ecoregion, and national scales). Diligence and CCI were correlated in mid-latitude (r=0.8) and high-latitude ecoregions (r=0.88). Country-level agreement with FAO data was high (r=0.76), as was agreement with field- and project-based carbon estimates from FIA (r=0.86) and CARB (r=0.76). Correlations were lower in pixel-level comparisons at NEON field plots (r=0.58) and in Central Africa (r=0.71), identifying a negative bias in carbon-dense forests.

These results demonstrate how Forest Carbon Diligence can support policy and market needs by providing scientific, well-calibrated, and consistent estimates of aboveground forest carbon stocks.