Using mimics to model L-band SAR backscatter from a peat swamp forest

The study was aimed at improving understanding of radar backscatter from peat swamp forest using MIMICS-I model. Two satellite scenes of the Japanese Earth Resources Satellite (JERS-1) SAR, acquired in dry and wet seasons, covering Pekan peat swamp forest in the state of Pahang, Malaysia were used. MIMICS-I model results were generated for L-band radar with Horizontal-Horizontal (HH), Vertical-Vertical (W) and Horizontal-Vertical (HV) polarisations. The peat swamp forest floor was modelled either as an electromagnetically equivalent bare surface soil (dry season) or flooded with water (wet season). To understand the relative importance of the various parameters covering the soil, vegetation and environmental inputs on the forest backscatter, a sensitivity analysis was performed. Following that modelling was carried out to understand the backscatter characteristics of peat swamp forest. Results of the sensitivity analysis showed that parameters of the canopy layer contributed significantly to radar backscatter. The results of simulations showed that L-band microwave energy was capable of penetrating the canopy to the ground layer, as shown by significant trunk-ground contributions in both flooded and non-flooded forest conditions. This suggests that L-band SAR data can be used to extract forest stand information such as density and biomass from peat swamp forest.