Laserfiche WebLink
<br />OJ1347 <br /> <br />18 <br /> <br />inputing selected rainfall events (from the extended regional rainfall <br />record) to the cal ibrated model. A flood frequency analysis of the <br />storm runoff events from each watershed is performed, and finally, the <br />storm runoff characteristics (peak storm runoff rate, dimensionless <br />runoff hydrograph, etc.) for a particular recurrence interval are re- <br />lated to the physiographic features of these watersheds by regression <br />analysis. <br />One of the major complaints regarding physically-based rainfall-" <br />runoff 1Ilodels is their poor estimation of antecedent moisture conditions. <br />The empirical equations used with some of these ~odels to approximate <br />watershed moisture conditions are simply not satisfactory. Continuous <br />hydrologic simulation models are physically-based models that solve this <br />problem by continuously accounting for all of the water (subsurface as <br />well as surface) within a particular watershed. This modeling approach <br />insures that the "exact" antecedent moisture conditions are simulated <br />when particular storm events occur. This "exactness" is not obtained <br />without cost, however; continuous simulation models require more and <br />significantly better input data than the non-continuous physically-based <br />rainfall-runoff models. <br />The most widely used continuous hydrologic simulation model is the <br />Stanford Watershed Model (or versions of the Stanford Model). James <br />(4ti) illustrates the application of this model. Like Dempster, he <br />develops regional flood-frequency-urbanization relationships (graphical <br />format) and ignores the development of relationships for the time <br />distribution of runoff -- relationships that could be generated with the" <br />output from the continuous simulation model. <br />