Abstract:
Since its inception, the unit hydrograph has been used as a key concept in estimation of storm runoff from a catchment. Significant. contribution to the unit hydrograph theory was given by Clark (1945), Nash (1957) and SCS (1972).in the form of development of synthetic unit hydrographs. Present study was taken up to determine the most suitable model for simulation of storm runoff for Jolarpet watershed having outlet at railway bridge No. 719. A digital elevation model has been developed for the simulation of rainfall-runoff process. The HEC-GeoHMS interface has been used for terrain processing to prepare input files for use in HEC- HMS model. The storm transformation options for SCS unit hydrograph and Clark unit graph of HEC-HMS have been used. Computations for Nash model have been done using Excel worksheets. In all six storm events have been used for judging performance of different models. In addition a comparison between two procedures, method of moments and multi-storm processing, for calculating basin average parameters for Nash model has also been done. Performance of different models used is evaluated using visual comparison and comparison based on statistical criteria such as Nash-Sutcliffe efficiency (NSE), peak weighted root mean square error (PWRMSE) and standard error (SE). Visually all models are showing approximate similar results but statistically the difference is clear. Among three models, based on average value of statistical indices used, Clark model with average NSE 83.32 is adjudged best followed by the Nash model with average NSE 79.45. The SCS Dimensionless unit hydrograph model performed poorest.. For derivation of representative values of Nash model parameters, method of
moments give slightly better results compared to multi-storm processing.
