Mapping biomass of a northern forest using multifrequency SAR data

The results of mapping standing biomass for a northern forest in Maine, using NASA/JPL AIRSAR polarimetric radar data, is presented. By examining the dependence of backscattering on standing biomass using backscatter modeling and aircraft data, it was determined, in agreement with other recent reports, that the cross-polarized (HV) data from longer wavelengths (L, P-band) were the best radar channels for mapping total above-ground forest biomass. The radar signal appeared to lose sensitivity to changes in biomass for dry biomass levels beyond about 15 kg/[m.sup.2] (150 Mton/Ha). The ratio of HV backscattering from two bands, a longer wavelength P (wavelength = 68 cm) or L band (24 cm) to a shorter wavelength C band (6 cm), enhanced the correlation of the image signature to standing biomass ([r.sup.2] = 0.83 for P/C and [r.sup.2] = 0.79 for L/C) and showed increased sensitivity to dry biomass beyond 15 kg/[m.sup.2].