Spatial and Temporal Variability of High Mountain Asia Snowmelt From Synthetic Aperture Radar

In glacier fed river systems, the timing and magnitude of peak runoff can vary substantially from year to year, challenging water resource managers and infrastructure. A lack of measurement due to rugged and remote terrain, as well as variable meteorological conditions due to complex atmospheric circulation patterns, limit runoff forecasting capabilities for glacial catchments in much of High Mountain Asia. This uncertainty is proliferated in efforts to anticipate cryospheric response to changes in climate. Many runoff models that incorporate glacier and snowmelt processes utilize optically based snow cover data sets, which are temporally useful but are often limited by spatial resolution and cloud cover issues in monsoonal Asia. Synthetic Aperture Radar (SAR) sensors acquire images regardless of cloud cover or sun illumination, at spatial resolutions useful for understanding heterogeneous snowpack processes. Due to radar sensitivity to dielectric properties, SAR sensors are often used to detect liquid water in snow, an important indicator of snowpack ripening prior to seasonal melt out. Here we utilize SAR imagery from the European Space Agency's (ESA) Sentinel-1 (S1) constellation to compare SAR derived wet snow maps with snow cover maps developed from multiple optical datasets. We select several glaciated locations along a longitudinal gradient through High Mountain Asia to compare snow cover datasets, and also explore the feasibility of SAR derived snow conditions as runoff model inputs in glaciated catchments of the Upper Indus Basin (UIB) of the Karakoram Mountains. SAR indications of snowmelt could offer useful constraints for physically-based model parameters, improving current runoff modeling as well as glacio-hydrologic models to anticipate future changes.