The retrievals of effective grain size and snow water equivalent from variationally-retrieved microwave surface emissivities

This study introduces a new technique for the estimation of the snow effective grain size and water equivalent based on the microwave surface emissivity spectra retrieved from a one-dimensional variational retrieval system and a microwave snow emissivity model. The microwave emissivity model is derived analytically from the dense media radiative transfer theory. The model snow physical parameters include the effective grain size, volume fraction and depth. The two-steps algorithm is based on variationally retrieving the emissivity spectrum from microwave remote sensing observations, followed by the estimation of the closest emissivity spectrum from a catalog to determine the snow water equivalent and the effective grain size. This catalog was computed off-line using the microwave emissivity model for realistic ranges of the effective snow parameters. Qualitative inspection of variationally retrieved emissivities and the snow parameters show large-scale consistency. The performance of this physically-based retrieval technique is quantitatively assessed against snow water equivalent measurements and against an empirical brightness temperature-based algorithm.