STREAMFLOW FORECASTING USING THE MODULAR MODELING SYSTEM AND AN OBJECT-USER INTERFACE

Extreme deviations in hydroclimatic conditions are a source of considerable error in statistical water supply forecast models. Much attention has been given over the past years to the relationship between snowpack, precipitation and streamflow (Martinec, 1975, Hawley, et al. 1980, McCuen, 1993). These relationships tend to vary in strength, but in large part are satisfactory for water supply forecasting purposes. Many other hydroclimatic variables such as soil moisture are implicit in these statistical relationships. As long as these variables (soil moisture) remain proportional to the independent variables (snowpack, precipitation) in the forecasting relationship, then the model will be stable. If there is some amount of disproportion, then the model will most likely produce significant error. Such a case in northern Utah is presented along with an explanation of a small- scale implementation of soil moisture monitoring of two small watersheds in the Wasatch Mountains near Salt Lake City.