SPATIAL MODELING OF SNOW WATER EQUIVALENT USING COVARIANCES ESTIMATEDFROM SPATIAL AND GEOMORPHIC ATTRIBUTES

As the demand for water in the USA rapidly approaches the total availa ble water supply, it is essential that water resources be accurately m onitored. Consequently, the National Weather Service (NWS) maintains a set of conceptual, continuous, hydrologic simulation models used to g enerate extended streamflow predictions, water supply outlooks, and fl ood forecasts. To obtain accurate predictions and forecasts, it is nec essary, periodically throughout the snow season, to estimate the snow water equivalent in river basins throughout the USA. The estimates are obtained using a geostatistical model and snow course, SNOTEL, and ai rborne snow data. in this research, we develop a positive-definite spa tial covariance function that allows researchers to incorporate geomor phic site attributes when snow water equivalent estimates are obtained . We illustrate our approach using snow course and SNOTEL data collect ed in the North Fork Clearwater River basin. Our results indicate that by incorporating elevation into the covariance model used for the Nor th Fork Clearwater River basin we are able to improve substantially th e accuracy of the snow water equivalent estimates.