Seasonal and spatial distribution of wet snow on three volcanoes in Western Washington mapped with synthetic aperture radar

In order to constrain the effects of climate change on snowpack in mountain watersheds, an understanding of the spatial extent and timing of snowmelt is necessary. However, lack of field records and difficulty modeling in remote and complex terrain lead to uncertainty in precipitation measurements across many mountainous regions. Fortunately, recent advances in remote sensing technology and data accessibility now provide valuable information that can be used to fill gaps in our understanding of mountain hydrology. In this study, remote sensing data from the European Space Agency (ESA) Sentinel-1 synthetic aperture radar (SAR) satellite is used to map wet snow over water year 2016 on three prominent peaks in western Washington, USA: Mount Rainier, Mount Adams, and Mount Baker. These stratovolcanoes were chosen as natural laboratories because they have relatively symmetrical topography at all aspects and significant area above tree line that is perennially covered by snow. We apply a standard method for mapping wet snow with SAR data using high-resolution Sentinel-1 imagery. Wet snow maps show expected seasonal changes in the large-scale pattern of wet-snowcovered area, along with nuances that occur due to small-scale variations in surface energy balance terms. Maps of wet-snow-covered area were also compared to temperature data recorded at the highest meteorological station in each study area and extrapolated across the region with a constant, assumed lapse rate. This temperature extrapolation was chosen to mimic records often used to force snowmelt models. Results iv show that SAR maps provide detailed information about wet-snow-covered area that could not be captured by simple snowmelt models forced with limited temperature data. This paper demonstrates the potential of Sentinel-1 imagery to provide high-quality information about the spatial and temporal distribution of wet snow, which could be used operationally in snowmelt forecast models or academically to validate snowpack energy balance models.