Department of Civil and Environmental Engineering
University of Washington
Research

Forests and Snow

Intermittent Snow and Process Dynamics

Snow Surface Temperature and Snow Depth in the Tuolumne Watershed

OLYMPEX


Orographic Precipitation

Mapping temperature in complex terrain

Spatial patterns of snow-fed streamflow

Rain vs. Snow

How meadow ecology relates to snow and climate

Intercomparison of Meteorological Forcing Data from Empirical and Mesoscale Model Sources

Silvicliture to maximize snow retention

Remote sensing of radiation to improve snow modeling

Wildflowers and Snow

Remote sensing of clouds and radiation to improve snow modeling

Net radiation is the dominant driver of snowmelt, but is generally only estimated from other observations. Can satellite data improve these estimates?

Working with Laura Hinkelman (Research Professor at JISAO, UW), we are examining how CERES and MODIS products can improve both shortwave and longwave radiation inputs to snow models and aid in hydrologic forecasts. This research is sponsored by NASA.

The greatest potential sources of error in simulating snowmelt rates and timing are inaccurate solar and longwave radiation inputs. Ground-based measurements of radiation are not widely available, and attenuation of solar radiation by clouds is particularly difficult to account for. Most hydrologic models estimate solar inputs from the position of the sun and the local diurnal temperature range. This can lead to errors of up to 50% in snowmelt rates. Fortunately, there is another, underexploited source of surface solar radiation data, namely satellite measurements.

In this project, we examine the benefit of using solar and longwave surface flux data from NASA satellites in place of the rough parameterizations found in current snow models. The performance of a range of snow models of varying complexity will be evaluated with and without application of surface fluxes from the CERES SYN product. In addition, a novel 5 km resolution MODIS-based data set will be used to test the importance of surface flux spatial resolution to snow model accuracy. Since MODIS data is available only twice daily, the results from this experiment will be contrasted with the results computed using the CERES SYN product, which is produced on a one-degree spatial grid at hourly intervals. Experiments with these two data sets will allow us to determine the combination of spatial and temporal sampling that provides optimum model performance. We anticipate that use of the satellite data will greatly improve the accuracy of snow models, providing substantial benefits to our understanding of snow melt processes and to the snow model stakeholder community.