A simple model for the estimation of biomass density of regenerating tropical forest using JERS-1 SAR and its application to Amazon region image mosaics

Regenerating tropical forests are important to the regional carbon balance because they represent areas of high carbon flux from the atmosphere to the land. These forests are complex in species distribution and structure so generalised models of radar backscatter are difficult to develop. However, empirical studies suggest a sensitivity of L-band SAR to the biomass density in these forests which may be employed to map and measure their growth. This paper investigates a simple empirical model representing the response of the JERS-1 SAR to the biomass density of regenerating forest in Brazil. This model was verified using independent data sources and employed to map biomass density over large regions of Amazonia from large scale JERS-1 image mosaics provided by NASDA.