NSIDC Cryosphere Seminar
Utilizing historical satellite- and model-based snow data to address real-time challenges in water resource management with Dr. Noah Molotch
Seasonal snow covers over 30% of the Earth's land surface and provides the water supply for approximately one-sixth of the global population. In the western United States, meltwater from seasonal snowpack contributes 50% to 80% of annual runoff and provides the majority of water for municipal, agricultural, and industrial demands. Whereas many independent methods can be used to estimate snow water equivalent (SWE) and its spatial distribution and seasonal variability, a need exists for a systematic characterization of SWE in near-real-time. This seminar will present recent research aimed at fusing together satellite- and model-based historical reanalyses of snow water equivalent, ground-based snow measurements, and airborne data in order to derive real-time SWE estimates. This will be broken into three parts:
First, historical reanalyses of SWE can be used as information for real-time SWE estimation within statistical models. Hence, we conduct a multi-scale validation and comparison of five fine-scale SWE model-derived datasets in the Sierra Nevada Mountains, California, including two datasets from historical SWE reconstruction models, a SWE Reconstruction with Data Assimilation, and two operational SWE datasets from the U.S. National Weather Service, including the Snow Data Assimilation System (SNODAS) and the National Water Model (NWM-SWE).
Second, this study develops a statistically-based data-fusion framework to estimate SWE in real-time, which combines multi-source datasets including satellite-observed daily mean fractional snow-covered area, snow pillow SWE measurements, physiographic data, and historical SWE patterns (as noted above) into a linear regression model (LRM). We find that the statistical model explains 87% of the variance in the snow course SWE measurements with 0.1% PBIAS. This represents an improvement over other operational models such as SNODAS (73% of explained variance and -2.4% bias) and NWM-SWE (75% of explained variance and -15.9% bias. Additionally, the statistical model explained 85% of the median variance in the Airborne Snow Observatory SWE with -9.2% PBIAS, which is substantially better than SNODAS (64% and 28.2%, respectively) and NWM-SWE (33% and -30.1%, respectively).
Third, examples will be used to illustrate the value of the real-time SWE product, including the 2012-2016 drought, rain-on-snow flooding during the Lake Oroville Spillway disaster in 2017, and recent intense snowfall which have placed 13 California counties under a State of Emergency declaration.
Noah Molotch is an Associate Professor of Geography, a Fellow of INSTAAR at CU-Boulder, and a Research Scientist at NASA-JPL. His research and teaching interests are focused on mountain hydrology and fostering a diverse and equitable scientific community. Noah’s research projects utilize ground-based observations, remote sensing, and computational modeling to obtain comprehensive understanding of hydrological processes; in particular the distribution of snowmelt, soil moisture, and streamflow. He also conducts research focused on carbon cycling in montane forests. Noah mentors a diverse group of scientists with the goal of ensuring the sustainability of water resources and ecosystem services now, and into the future.