Forest parameter retrieval from SAR data using an estimation algorithm applied to regrowing forest stands in Queensland, Australia

The use of a non-linear estimation algorithm for retrieving the biomass and structure of vegetation from polarimetric Synthetic Aperture Radar (SAR) data is demonstrated for woody regrowth in Queensland, Australia dominated by Acacia harpophylla (Brigalow). By varying the size and density of trees and associated woody components (branches and trunks), multiple simulations of the backscattering coefficient (σ⁰) were performed based on the SAR simulation model of. Functions relating σ⁰ to these variables were subsequently used to generate spatial estimates from NASA JPL airborne SAR (AIRSAR) data. Above ground biomass was estimated from stem density and size measurements using available allometric relationships. The study demonstrates potential for retrieval of regrowth structure and biomass through nonlinear estimation.