FLOOD02234 - Laserfiche WebLink

Home Browse Search

series of empiricisms selected to provide a mathematical continuum from ridge top to watershed outlet in terms of input readily available to the analyst. The restriction of input information to that "readily available" to the analyst appears contrary to the development of a model truly cap- able of simulating the various subsystems of the hydrologic cycle. 28. The point here is that a deterministic streamflow model will surely require a detailed quantitative description of the drainage basin. In fact, the required description can be conjectured to consist of two fundamental parts: (a) a stream channel network description, somewhat after the pattern of that produced for the European Waterways Study; and (b) a description of the interfluvial terrains designed in terms of all of the factors that influence runoff, infiltration, storage, and transpiration. It seems clear that a system for describimg a watershed in these terms must be developed prior to or concurrently with the model which it is intended to feed. Almost surely the descriptors will be environmental factors not now "readily available" to the analyst; it is therefore mandatory that hydrologists face the cruel fact that new data acquisition procedures, perhaps similar to those used in the European Waterways Study, will have to be developed and used. 29. The development of a physically based simulation model of all the various subsystems of the hydrologic cycle appears to be a monumental effort. Although many hydrologists6,7 feel that such a model can be devel- oped, it is generally agreed that relating the various subsystems to spe- cific environmental descriptors is a prerequisite to the completion of the general model. Once the relevant descriptors are identified and watersheds can be described in terms of the descriptors, the completed model would pro- vide deterministic predictions of streamflow. The Rowlesburg Lake Model System 30. Having arrived at the conclusion that the quantitative terrain description part of the problem of creating deterministic models can be solved, an example of such a model is in order. The following discussion concerns a model developed by WES for the Pittsburgh District to assist in a study of possible alternates to the construction of a large multipurpose dam on the Cheat River at Rowlesburg, West Virginia (fig. 6). 31. The Rowlesburg Lake was planned as the last of a system of reser- voirs in the Ohio River Valley system of flood control. Although the proj- ect had been authorized since the 1940's, construction was not planned to begin until 1970 or 1971. The initial environmental impact statement sub- , mitted by the Pittsburgh District was deemed lacking in detail and compre- hensiveness by the Environmental Protection Agency (EPA). A major problem I at Pittsburgh in performing impact studies was the development of capability 21