Vegetation Water Content Retrieval by Means of Multifrequency Microwave Acquisitions From AMSR2

摘要: In this study, two retrieval algorithms for estimating the water content of vegetation (VWC) in range 0–8 kg/m2 from multifrequency microwave radiometric data have been implemented, with purpose contributing to development an all-weather VWC product satellite AMSR2 radiometer JAXA (Japan Aerospace Exploration Agency) GCOM-W mission. The first algorithm estimates through a semi-empirical combination polarization index (PI) acquired at X and Ku bands, while second one endeavors improve accuracy adding C-band using artificial neural network (ANN) method. sensitivity biomass PIs various frequencies, which was already pointed out previous literature, has further evaluated support well-known tau-omega solution radiative transfer model experimental data. Two years collected on wide portion Africa, includes large variety types biomasses, considered implementing testing both algorithms. reference same temporal spatial coverage AMSR2, needed validating outputs, derived SPOT4 normalized difference (NDVI), downsampled ground resolution. test results provided correlation coefficient R > 0.88 root mean square error (RMSE) = 0.98 RMSE 0.82 around 1.5 algorithm, 0.86 ANN. This study demonstrated that indices can be legitimately used produce maps global scale by separating several levels biomass, without any need information other sensors.