COMPARATIVE STUDY OF SYNTHETIC UNIT HYDRO GRAPH METHODS

Abstract

The synthetic unit hydrograph (SUH) methods are widely used for estimating design flood for unguaged catchments, and for partial data availability conditions, as in developing countries for majority of small watersheds are unguaged. Several SUH methods are available in literature. Most of them involve manual, subjective fitting of a hydrograph through few data points. Because it is difficult, the generated unit hydrograph (UH) is often left unadjusted for unit runoff volume. Therefore, probability distribution function (pdf) based methods are favored to derive an SUH more conveniently and accurately than the traditional methods of Snyder and Soil conservation Service. pdfs as SUH is accepted because of the similarity between pdf of a distribution with area under the pdf curve and a conventional UH being unity, an important feature of a pdf. This study explores the potential and suitability of the parametric expressions of geomorphological instantaneous unit hydrograph (GIUH)-coupled probability models of One-parameter Chi-square (1PCSD), Two-parameter Frechet (2PFD), Two-parameter Inverse Gamma (2PIGD) and Two-parameter Gamma (2PGD) (Rosso, 1989) distributions for limited data availability condition compared with the widely used traditional methods of Snyder and SCS for SUH derivation. UH features, peak discharge, time to peak, etc. are derived using Horton order ratios given by Rodriguez-Iturbe and Valdes (1979). Geomorphologic characteristics of study watershed have been extracted from ASTER data (resolution 30 m) using ILWIS (version 3.31) GIS software. Analytical procedure is proposed for estimation of parameters of the used four distributions. The applicability of the four pdf models-coupled with GIUH and traditional methods of Snyder and SCS for SUH derivation is tested on both text and field data. Their suitability is compared with the observed UH of the four study catchments (area = 27.93 km2 - 4000 km2) located in different parts of the country. Finally, their performance is evaluated based on STDER and RE in peak flow rate. 2PFD performed marginally better than 2PIGD, followed by 2PGD and IPCSD models for mid-sized study catchments (Myntdu-Leska and Gagas). However, for the large size catchment (Burhner) as well as for the smallest (Kothuwatari), the 2PGD model performed better than 2PIGD, followed by 2PFD and 1PCSD models. The four density functions coupled with GIUH except 1PCSD compared well with the observed UH, indicating good potentiality for SUH derivation.