Assimilation of observed active and passive microwave signatures with a physical model for estimation of snow water equivalent

In this study we investigate techniques for improved passive and active microwave retrievals of SWE, by utilizing a state-of-the-art snow-evolution model (SnowModel). We examine two schemes: 1) Direct insertion of SnowModel outputs of snowpack properties to an inversion algorithm utilizing microwave emission and backscatter model (MEMLS) and 2) a feedback loop between MEMLS and the SnowModel to assimilate the SWE of the latter, and potentially improve the quality of the snow property outputs. Prior to this experiment we evaluate the performance of SnowModel by comparing it to field observations of boreal snow properties, including snow depth, density and grain size.

In situ snow properties and continuous time series of active and passive microwave observations are available from the NoSREx campaign. NoSREx took place in Northern Finland and it lasted for four complete snow seasons (2010-2013). The NoSREx campaign was designed to provide passive and active microwave observations of snow conditions representative of the boreal forest zone.