Interpreting snowpack radiometry using currently existing microwave radiative transfer models

A radiative transfer model (RTM) to calculate a snow brightness temperature (Tb) is a critical element to retrieve terrestrial snow from microwave remote sensing observations. The RTM simulates the Tb based on a layered snow by solving a set of microwave radiative transfer formulas. Even with the same snow physical inputs used for the RTM, currently existing models such as Microwave Emission Model of Layered Snowpacks (MEMLS), Dense Media Radiative Transfer (DMRT-Tsang), and Helsinki University of Technology (HUT) models produce different Tb responses. To backwardly invert snow physical properties from the Tb, the differences from the RTMs are to be quantitatively explained. To this end, the paper evaluates the sources of perturbations in the RTMs, and reveals the equations where the variations are made among three models. Investigations are conducted by providing the same but gradual changes in snow physical inputs such as snow grain size, and snow density to the 3 RTMs. Simulations are done with the frequencies consistent with the Advanced Microwave Scanning Radiometer-E (AMSR-E) at 6.9, 10.7, 18.7, 23.8, 36.5, and 89.0 GHz. For realistic simulations, the 3 RTMs are simultaneously driven by the same snow physics model with the meteorological forcing datasets and are validated from the snow core samplings from the CLPX (Cold Land Processes Field Experiment) 2002-2003, and NoSREx (Nordic Snow Radar Experiment) 2009-2010.