C41F-04
Operational Snow Modeling: A Look at the Current State and Future Challenges
Thursday, 17 December 2015: 08:45
3005 (Moscone West)
Adam H Winstral1, Tobias Jonas2, Danny G Marks3, Thomas H Painter4, Kat Bormann4, Jeffrey S Deems5, Scott Havens3, Andrew R Hedrick6, Nora Helbig1, Jan Magnusson1, Bruce J McGurk7 and McKenzie Skiles4, (1)WSL Institute for Snow and Avalanche Research SLF, Davos Dorf, Switzerland, (2)SLF / WSL, Davos Dorf, Switzerland, (3)USDA Agriculture Research Serv, Boise, ID, United States, (4)NASA Jet Propulsion Laboratory, Pasadena, CA, United States, (5)National Snow and Ice Data Center, Boulder, CO, United States, (6)USDA Agricultural Research Service New England Plant, Soil and Water Research Laboratory, East Wareham, MA, United States, (7)Self Employed, McGurk Hydrologic, Washington, DC, United States
Abstract:
Recent advances in distributed physically-based snow (and hydrological) models have moved these modeling tools into the operational arena. High resolution operational products that can account for the hydrologically-relevant heterogeneities in snow accumulation and melt (ca. 100m grid scale) are now being delivered to water managers in two select river basins of the American West. This comes at a time when simpler solutions based on historic trends are struggling to cope with modern-day weather scenarios that are quite different from those previously encountered. This marks a significant advancement in modeling capabilities and provides water managers with tools robust to climate and landscape changes. However, these models have higher data requirements, tend to be more sensitive to input data errors, and remain computationally intensive compared to simpler, parametrized approaches. In operational settings, where time is limited, the number of possible model runs is necessarily constrained. Data assimilation techniques and probabilistic forecasts though require numerous model realizations to establish the sound statistical foundations they are based upon. Whereas simpler solutions (e.g. conceptual, lumped, degree-day) are compatible with these latter ensemble procedures, the physically-based solutions currently are not. Recent work has modernized and enhanced the parameterized approaches as well and these are no longer the basic tools they once were. This research looks at the advantages and limitations of the most modern operational tools in use, and the research challenges that lie ahead. Operational models developed and applied by the WSL/SLF in Switzerland and the USDA-ARS NWRC in the western U.S. will be highlighted in this presentation.
