TN-72 : Review of methodologies and software for flood forecasting

Abstract

Flood forecasting is required for a number of purposes, e.g. , flood control, flood damage reduction, and reservoir operation. Significant improvements in operation of water resources systems can be achieved if reliable forecasts are available with sufficient lead time.

The models which are used for real time flood forecasting can be classified in three categories - black box models, conceptual models, and physically based models. The first category includes models which are mostly based on Unit hydro-graph theory. The physically based models are still in production stage and the computational requirements vis-à-vis facilities available at a typical installation make them unsuitable for real time application, at least for next several years. The conceptual models provide a framework which is theoretically more sound than the black box models. Further the fact that a basin is not a random assembly of different parts but a geomorphological system whose parts are related to each other by a long common history, encourages the hope that simplified concepts may be found adequate to describe the operation of the basin in converting rainfall to runoff.

This present report critically reviews the available real time forecasting models. Special emphasis is placed on application of conceptual models like NWSRFS, HBV model and NAM model. The attempts made in India in this area have also been reviewed.