Liquid water content in a seasonal snowpack: a comparison between satellite products and model simulations

The spatial and temporal quantification of the liquid water content (LWC) of the snowpack in alpine regions provides information on the short-term availability of water, which could eventually lead to wet snow avalanches or river floods. The monitoring and forecasting of snow wetness are hence of paramount importance in many fields, from operational avalanche forecasting to hydropower production and flood prediction, when combined with hydrological models. 
Remote sensing is an essential tool for snow monitoring as it offers observations of the snowpack's physical properties. For instance, Sentinel-1 satellites provide C-band synthetic aperture radar (SAR) data at high temporal and spatial resolutions and are capable to detect the presence of wet snow.  On the other side, many snow models were built in literature to simulate snowpack mass dynamics in space and time (see e.g., Crocus and HyS model) and can provide predictions of variables of interest in snow hydrology, such as the LWC. 
In the present work, we aim at identifying and quantifying the differences between satellite products and model snow estimates. In particular, the comparison is led among (1) Sentinel-1-based wet-snow products, (2) HSAF products, coming from the processing of data from Earth observation satellites and revealing the wet or dry status of the snow mantle, and (3) simulations of the liquid water content from HyS model, a temperature-index model, leveraged in both its one-layer and two-layer version. The case study is the Mallero basin, a middle-size alpine basin, whose flow regime is strongly influenced by snow melting and glacier ablation in the spring and summer seasons. 
The comparison returns a good agreement between Sentinel-1 products and HyS simulations. The short period of mismatches between the two outputs is analyzed to identify the physical processes that the model is not able to reproduce. On the other side HSAF products have a coarser resolution if compared to Sentinel-1 products and for this, they can just provide a qualitative overview of the snow mantle status, over a middle-size basin. Moreover, such products are also limited by the effect of cloud covering that makes it impossible to have information on the snow wetness when it is present.