Untitled DocumentHydrologic modeling of SWE and snowmelt
Macro Theme Area:
Basin Scale Water Balance [Project ID: B02]
PI:
Roger Bales
CO-PI(s):
Kevin Dressler, George Leavesley, Steven Fassnacht
Basin focus:
Colorado, Rio Grande
Specific area in
basin /
field sites:
White, Black, Upper Rio Grande basins
Summary/Goals: Seeks to improve forecasts of snow runoff through a combined approach of remote sensing and ground-based data. Future studies will define where remote sensing may be lacking and how combining the remote effort with ground-based validation may be improved or changed to optimize the accuracy of streamflow prediction.
Activities and outcomes during past year:
Three science questions were pursued for this program. First, uncertainty in the spatial distribution of snow-water equivalent (SWE) was explored by evaluating two distinct data sets from snowcourse and SNOTEL. The point values in these datasets were spatially distributed using a hyposometric technique developed by Fassnacht et al. [2003] from previous SAHRA work. While SNOTEL were implemented to replace the snowcourses, measurements at co-located sites indicate generally small differences in SWE at the point scale, but these differences translate to large interpolated volume differences at the basin scale. Trade-offs in usability of either dataset must be considered in modeling, as the impact of assimilating snow information for water supply forecasting is increased with higher frequency observations in locations representative of elevation and associated SWE amount. Second, a comparative evaluation of SCA and SWE products based on blending satellite and ground-based measurements versus a modeled snowpack (estimated from temperature and precipitation) in two headwater basins. We used the Precipitation Runoff Modeling System due to its minimal forcing data requirements, compatibility with available data and also because of previous success in simulating snow packs in the study region. Differences between modeled and measured fields are evaluated in time and space and in the context of simulated discharge from those different fields. It was determined that the satellite-derived products were systematically lower than model estimates with underestimates increasing with elevation. Satellite derived estimates of SCA and SWE were most different than modeled in the complex terrain of the Upper Rio Grande relative to the homogeneous terrain in the Black River, AZ. Third, the influence of vegetation in simulating snowmelt runoff using satellite was tested by applying a physically-based canopy correction to the satellite-derived product and comparing to the modeled snowpack. The canopy correction reduced differences between the measured and modeled results and improved estimations of streamflow over base model runs in PRMS for the topographically homogeneous basin (Black River, AZ). Results from the second and third questions are in publication in press at Hydrological Processes.
Plans for the upcoming year:
We will be working on understanding hydrologic processes at the headwater catchments in the Upper Rio Grande using the Distributed Hydrology, Soil and Vegetation Model. Our primary objective is to improve our understanding of water balance at basin scale.
