A REVIEW OF APPLICATION ISSUES OF THE METROPOLITAN WATER-DISTRICT MAIN-WATER FORECASTING SYSTEM

This paper reviews the processes that occurred during an application o f the Metropolitan Water District (MWD)-MAIN water use forecasting sys tem for the City of Salinas, California. The review includes an analys is of sources of available data, methods for estimating input data, ca libration, and verification of the MWD-MAIN System, and an evaluation of the reliability of system output. We found that inexperienced users can have difficulty understanding the level of skill, knowledge, and amount of data that are required to produce reliable forecasts. Some o f the issues associated with application of the MWD-MAIN System includ e the following: All input data needed for accurate forecasts simply a re not available for many cities and towns. The data requirements are more extensive than many users anticipate. Substantial requirements fo r manipulation of input data produces opportunity for error that creat es major time demands in troubleshooting. Calibration and verification for specific uses can be substantially more difficult than is readily apparent from the guidance manual. Independent validity checks need t o be done to validate system output. If specified calibrating procedur es do not produce reasonable results, reestimating slope coefficients is an option, but this requires resources and expertise that can easil y exceed the limits of most users. These are problems typical of most complex models. Reviews such as this can help users to appreciate the level of data required, and to use the MWD-MAIN System in a more effec tive and efficient