Aboveground biomass estimation of tropical peat swamp forests using SAR and optical data

Climate change mitigation mechanisms, such as REDD+, which aim at avoiding deforestation and forest degradation, require an accurate aboveground biomass (AGB) monitoring. In the present study, multi-temporal X-(TerraSAR-X) and L-band (ALOS PALSAR) SAR data and a multispectral RapidEye image were analyzed for their ability to estimate AGB in a tropical forested peatland area in Central Kalimantan on Borneo, Indonesia. Field inventory AGB data was used to calibrate regression models based on SAR backscatter values and spectral unmixed fractions of the RapidEye image. The independent validation indicated that the estimated AGB using optical data is more accurate (RMSE=44%) than the SAR estimated AGB (RMSE=82%). AGB derived from RapidEye data overestimates AGB on burned areas, but these estimations depict degradation through low impact selective logging. The SAR model estimated AGB accurately in lower biomass ranges and on burned scars.