Leveraging Climate Prediction in Water Supply Forecasts

The Colorado Basin River Forecast Center (CBRFC) is one of thirteen River Forecast Centers throughout the National Oceanic and Atmospheric Administration’s National Weather Service that provide advanced hydrologic prediction. The CBRFC’s area of responsibility spans the entire Colorado River Basin and the eastern portion of the Great Basin. Forecasts of seasonal (April-July) water supply conditions are a primary focus of the CBRFC. This process uses Ensemble Streamflow Prediction (ESP) to develop a forecast ensemble of 30 traces based on historical temperature and precipitation data over the calibration period (1981-2010). It is important for the CBRFC to understand if a forecast ensemble of 30 traces sufficiently captures the hydrologic variability of a region, and if that variability is subject to change based on projected climate information. Climate information is not yet considered in these predictions but as the impacts of climate change continue to be realized, there are questions as to whether past historical data still appropriately represents future conditions. In an attempt to incorporate climate forecast information and reduce reliability on historical data, a Stochastic Weather Generator (SWG) developed by the University of Colorado was used. The SWG is able to simulate future weather conditions based on historical data and probabilistic climate outlooks. In this study, probabilistic climate outlooks issued by the Climate Prediction Center (CPC) are used to develop future weather conditions that are independent of historical sequencing. The probabilities refer to how likely an area will undergo warmer, normal, or cooler temperatures and drier, normal, or wetter precipitation conditions.

In this project, the Bear River Basin was considered, which is a sub-basin of the Great Basin. The Bear River is the largest tributary to the Great Salt Lake and an important source of irrigation for Utah based agriculture. It is further shown that weighted climate simulations provide additional information towards forecast development and improvement.