Forest Aboveground Biomass Estimation Using a Combination of Sentinel-1 and Sentinel-2 Data

摘要

Estimation of forest aboveground biomass is crucial for carbon accounting and forest management. In this paper we present the estimation of forest above ground biomass derived from a combination of Sentinel-1 and Sentinel-2 data using the Random Forest Regression approach. The research was conducted over a temperate forest in Poland, where the forest biomass is ranging from 1 to 300 t/ha. The results revealed a saturation effect around 200 t/ha. The biomass model tends to overestimate small values of aboveground biomass and underestimate larger values of biomass (greater than 250 t/ha). The overall RMSE value is equal to 60 t/ha, however it varies for biomass ranges. The independent validation performed at the forest stand level showed the best agreement for biomass ranging for 100-200 t/ha. This research was conducted as part of the ESA funded DUE GlobBiomass project.