A parameterized multiple-scattering model for microwave emission from dry snow

Snow water equivalent (SWE) is one of the key parameters for many applications in climatology, hydrology, and water resource planning and management. Satellite-based passive microwave sensors have provided global, long-term observations that are sensitive to SWE. However, the complexity of the snowpack makes modeling the microwave emission and inversion of a model to retrieve SWE difficult, with the consequence that retrievals are sometimes incorrect. Here we develop a parameterized dry snow emission model for analyzing passive microwave data, including those from the Advanced Microwave Scanning Radiometer-Earth Observing System (AMSR-E) at 10.65 GHz, 18.7 GHz, and 36.5 GHz for SWE estimation. We first evaluate a multiple-scattering microwave emission model that consists of a single snow layer over a rough surface by comparing model calculations with data from two field measurements, from the Cold Land Process Experiment (CLPX) in 2003 and from Switzerland in 1995. This model uses the matrix doubling approach to include incoherent multiple-scattering in the snow, and the model combines the Dense Media Radiative Transfer Model (DMRT) for snow volume scattering and emission with the Advanced Integral Equation Model (AIEM) for the randomly rough snow/ground interface to calculate dry snow emission signals. The combined model agrees well with experimental measurements. With this confirmation, we develop a parameterized emission model, much faster computationally, using a database that the more physical multiple-scattering model generates. For a wide range of snow and soil properties, this parameterized modelʹs results are within 0.013 of those from the multiple-scattering model. This simplified model can be applied to the simulation of the microwave emission signal and to developing algorithms for SWE retrieval.