Modeling the Spatial Distribution of Snow Water Equivalence
and Snowmelt in Mountain Basins
Don Cline1 and Kelly Elder2
Two models based on very different methodologies are used to investigate the spatial distribution of
snow water equivalence (SWE) in two mountain basins. The first model (SWETREE) uses binary
decision trees (regression trees) to relate physically based independent variables (net solar radiation,
topography, soil and vegetation cover types) to ground measurements of SWE in order to interpolate
SWE across a gridded domain. The second model (SNODIS) estimates the spatial and temporal
distributions of both SWE and snowmelt post facto at the end of the snowmelt season using a coup-
led remote sensing/spatial energy balance approach. Gridded fields of meteorological variables
needed to compute the snow surface energy balance are estimated from micrometeorological data
collected within each basin, and time series of remotely sensed snow cover are used to determine
how long snow remains on each grid cell through the melt season. In this study, the two models are
applied in the Green Lakes Valley and the Loch Vale Watershed, both located in the Colorado Front
Range of the Rocky Mountains. Extensive field observations of SWE in these basins are used to
evaluate the model-estimated SWE distributions. The SWE results from the two models are com-
pared to help understand relative strengths and weaknesses of these different approaches to estimat-
ing SWE, and to identify how the two models could potentially be used in combination for opera-
tional forecasting purposes.
1 National Operational Hydrologic Remote Sensing Center, National Weather Service, 1735 Lake Drive West,
Chanhassen, Minnesota 55317, USA
2 Department
of Earth Resources, Colorado State University, Fort Collins, Colorado 80523, USA
83
Previous Page
