Estimation of boreal forest biomass from multi-temporal INSAR data by inverting an empirical backscattering-coherence model

The applicability of INSAR coherence observations for stem volume (biomass) retrieval is investigated by applying coherence data from 14 ERS-1 and ERS-2 C-band SAR image pairs. A novel technique for stem volume retrieval is developed based on the inversion of a non-linear empirical forest coherence model. The seasonal behavior of interferometric coherence, the accuracy of coherence modeling as well as the performance of stem volume retrieval is tested. The data set enables the study of stem volume retrieval performance under varying conditions and as a function of the number of images. The results indicate that the applicability of winter images with snow-covered terrain is superior to that of images obtained under summer conditions. The highest correlation coefficient between the estimated stem volume and the ground truth stem volume shows values as high as r=0.89 (obtained with two optimum images using separate training and testing data sets) and a percentage RMSE level of 48%.