Multi-scale remote sensing and modeling for estimating liquid water content and LAPs on snow in the European Alps

The spectral reflectance of snow and ice varies widely depending on several quantities related (1) to the local environmental variables, such as the solar zenith angle and the surface slope, (2) to the physical properties of the snow, such as the grain size and the snow liquid content, and (3) to the presence of light-absorbing particles (LAPs).  Different absorption features are displayed in snow spectra. In particular, the absorption at 1030 nm has been exploited for estimating the grain effective radius of snow both from remote and proximal sensing data (Dozier et al., 2009, Garzonio et al., 2018). This absorption feature has been also used for the retrieval of the liquid water content (LWC) of surface snow since it is characterized by a shift toward shorter wavelengths when LWC increases (Green et al., 2006). Taking benefit of this spectral shift of the absorption feature, we applied a continuum removal approach to obtain both the grain equivalent radius and the LWC value. Furthermore, the accumulation of LAPs, such as dust, black carbon, volcanic ash, and pigmented snow algae on the snowpack albedo increases the absorption of solar radiation and induces a positive surface radiative forcing, enhancing the surface melting.

In this contribution, we show a retrieval algorithm to estimate the variables of snow (i.e., snow grain size, snow water equivalent, LAPs concentration) by using the openly available radiative transfer model BioSnicar (Bio-optical Snow, Ice, and Aerosol Radiative model) to simulate the spectral albedo of snow and the absorption of solar light in the snowpack. We present data from two experimental sites located in the Eastern Alps (Stelvio Pass and Brenta Dolomites) collected using a Spectral Evolution spectroradiometer. Measured variables of snow with a Snow Sensor device were compared with those estimated from BioSnicar simulations. Moreover, the impurities content in snow samples collected will be analyzed in a laboratory to better constrain modeling results. Remote sensing is a fundamental tool for characterizing snow cover properties, from the accumulation of LAPs to the wet/dry state of the snow, and the use of satellite sensors (e.g. PRISMA) opens the possibility for monitoring their spatial and temporal variability. This may have an important impact on snow hydrology studies, mainly for monitoring snow melting and improving the management of freshwater resources in the Alpine environment.