The future of snow science: Spatial data assimilation and modeling in mountain regions

Snow science and modeling are in transition from interpolation of point samples to high resolution spatial measurement of snow properties from airborne and ground-based Lidar (ALS and TLS), digital photogrammetry and ground penetrating radar (GPR). New methodology has provided time-series of detailed high-resolution (sub-meter scale) measurements of snow depth and distribution over complex mountain regions from Lidar and photogrammetrically-derived structure from motion (SfM). Density and SWE can be derived from GPR, however GPR is more difficult to apply from an airborne or drone platform. The initial phase of this new paradigm in snow science has been conducted over a large region of the southern Sierra Nevada and Colorado River Basin for the past 7 snow seasons (2013 - 2019) as a component of the NASA JPL Airborne Snow Observatory. Repeat interval airborne lidar flown to provide 3 m snow depth measurements over the region, provided the first measurements of snow distribution over a large mountain region. The lidar-derived snow depth field was assimilated at 50 m resolution as a state variable into the iSnobal energy balance snow model. Output from this effort was provided to water managers across the region and has been used as the foundation for more refined and improved water supply forecasts. This initial 7-year test and evaluation of the new snow science paradigm validates the approach, provides very unique scientific data for the community, and illustrates just how critically important high-resolution spatial data and modeling are to understanding the development and ablation of the seasonal snow cover.