C31B-06
Are existing snow microwave emission models so different ?
Wednesday, 16 December 2015: 09:15
3007 (Moscone West)
Ghislain Picard, UJF/LGGE Laboratoire de Glaciologie et Géophysique de l’Environnement, Grenoble, France, Henning Loewe, WSL Institute for Snow and Avalanche Research SLF, Davos Dorf, Switzerland, Melody J Sandells, University of Reading, Reading, United Kingdom, Michael T Durand, Ohio St Univ-Earth Sciences, Columbus, OH, United States, Jinmei Pan, Ohio State University Main Campus, Columbus, OH, United States, Alain Royer, University of Sherbrooke, Sherbrooke, QC, Canada, Christian Mätzler, Gamma Remote Sensing, Guemligen, Switzerland and Nicolas Floury, European Space Agency, ESTEC, Netherlands
Abstract:
Several models to compute the thermal emission in the microwave domain of snow-covered areas have been developed in the past few decades and have become very popular, like HUT, MEMLS and a few others based on the DMRT theory. These models have many differences between each other. Numerous studies have exploited one of them and have drawn conclusions about their skills or used them to retrieve snow properties or assimilate observations, etc. Nevertheless the portability of this knowledge to the other models is a concern. An increasing number of studies have also compared pairs of these models to show, in specific conditions, the differences between simulations and skills with respect to observations. There is a consensus on the fact that none of these models performs always (or sufficiently often) better than all the others. Nevertheless, the strategies developed to perform these comparisons are so dependent on how the model differences are handled, that the scope of numerically-based comparisons is limited. The diversity of investigated snow conditions and the ground truth uncertainties are also limiting factors.
With the raising need to consolidate its findings, the snow passive microwave remote sensing community is eager to understand the profound similarities and differences of their models and to this end has started to re-cast the different model formulations in a common framework. In this work, we address this question for a few specific points of the models: the description of the micro-structure (grain size parameters and underlying assumption) and the solution method of the radiative transfer theory (multi-stream versus N-stream and phase function representation). We show that existing models are not so different from each other, especially when consistent parameters (or representations) are used. At last, we present propositions to build a new-generation model that would be sufficiently modular to encompass the (small) diversity of current models.
