Verification of Water Supply Forecast Tools
Abstract
Water supply forecasting has a lengthy history in the semi arid western United States where water is so valuable and central to many aspects of the economy, the landscape, and the life that calls the west home. Historically, forecast agencies have largely relied on two forecast approaches: statistical models relating snow measurements and other predictor variables to runoff and simulation models that to varying degrees account for physical processes and weather forecasts important in producing runoff. As snow and land surface modeling capabilities and weather and climate forecasting have improved, the potential for applying these tools to realize forecast improvement has grown. In order to support leveraging these capabilities into water supply forecasting, this paper compares the skill and reliability of these two forecast tools in the Colorado and eastern Great Basins. We find that the two forecast tools are generally comparable in forecast skill and reliability except late in the forecast season when ensemble techniques with the Colorado Basin River Forecast Center (CBRFC) simulation model fails to adequately capture the range of observed outcomes. However, we find that statistical post adjustment techniques have the potential to address this deficiency. By validating the current water supply forecasting tools, this study aims to inform management decisions regarding forecasting tools and techniques that may be required to fulfill forecasts requirements in the future.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2013
- Bibcode:
- 2013AGUFM.H32G..05N
- Keywords:
-
- 1884 HYDROLOGY Water supply