Investigations toward understanding the spatial representativeness of SNOTEL measurements of snow water equivalence

The spatial distribution of snow water equivalence (SWE) is important for accurately predicting water availability, particularly in semi-arid, high elevation regions typical of the Western United States where as much as 75 of stream flow originates as snowmelt. The Natural Resource Conservation Service (NRCS), formerly the Soil Conservation Service, provides much of the data currently available for operational snow hydrology in the Western United States through its Snow Telemetry (SNOTEL) and snow course network. While this network provides a valuable index to seasonal runoff, little is known about how well these stations represent the local snow distribution. This is of utmost importance as snow hydrologists look for data sources for ground truth of remote sensing applications and move towards physically based snowmelt models. Intensive field campaigns in the Colorado Rockies and the Sierra Nevada have been conducted to address these issues on the scales of 30m x 30m and 1km2. Both grid sizes are commonly used in Digital Elevation Models (DEM) and satellite remote sensing.Our results indicate that the measured SNOTEL sites tend to over represent SWE in their local areas.