Seasonal Drought Prediction: The Changing Role of Snow in Water Supply Forecasting across the Western U.S.

Abstract:

In the western U.S., seasonal water supply predictions rely on knowledge of spring snow information, since for many systems, snow represents a greater seasonal storage of water than man-made reservoirs. There is a strong consensus that a warmer future will see less precipitation falling in the form of snow, which represents a critical challenge for predicting seasonal runoff in order to allocate resources—particularly critical during drought. A systematic evaluation will be presented into how this relationship is expected to evolve as temperatures rise through the end of this century. Preliminary results show that contribution of snowpack to predictive skill is greatly reduced over roughly two thirds of snow dominated regions, with greater reductions in maritime climates and lower elevation mountains, whereas predictive capabilities in cooler, higher elevation zones tend to show significant (p<0.01) resilience to warming. Ancillary predictors like accumulated precipitation and soil moisture can aid in recovering some fraction of the lost skill. Lastly, we present a set of idealized experiments to quantify the impact of non-stationary snow conditions on seasonal drought forecasts relative to the impacts from warm-season precipitation non-stationarity. Overall, this work seeks to understand strategies for enhancing drought prediction capabilities amidst a changing climate.