Combination of SPOT-5 and ALOS PALSAR images in estimating aboveground biomass of lowland Dipterocarp forest

Abstract

Aboveground biomass (AGB) of forests is the one of the key parameters for carbon accounting. However, estimating AGB by using remote sensing approach has been challenging as it is constrained by various limitations, especially in a complex tropical forest ecosystem. Optical or radar system has its potential in retrieving AGB but issues such as cloud cover, complex forest ecosystem and saturation at certain biomass levels remain unanswered and are continuously being studied. The study was conducted to investigate the possibility of combining both optical and radar to improve the accuracy of AGB estimation in lowland dipterocarp forest. SPOT-5 and ALOS PALSAR data were used and regression models were developed between the measured AGB and variables derived from both satellite images. The study found that the best performing model was from the multivariate regression from incorporating both normalized difference fraction index (NDFI) with HV-polarized backscatter with R2 of 0.803 and RMSE of 32.6 Mg ha−1. The study found that the combination of optical and radar images can counter limitations of each other and has improved slightly the estimate.