ESTABLISHMENT OF RATING CURVE AT AN UNGAUGED RIVER SITE

Abstract

In most of the engineering activities related with hydrologic design,- flood forecasting, flood control, water resources planning and management etc. we need to know the discharge in a river or channel at different locations and for different elevations of water surface, measured in terms of gauge or stage. Discharge is determined directly through either discharge measurement or an established stage-discharge rating curve or simply a rating curve. To develop a rating curve at a gauging site, in a conventional manner, a number of concurrent measurements of stage and discharge are made and these- measurements are used for the development of a rating curve based on graphical or some statistical curve fitting procedures. Using this developed rating curve, one can obtain discharge corresponding to any measured stage. However, in case of ungauged river sites it is not possible to develop a rating curve due to non-availability or scantily available stage and discharge data. If some structure (or project) is to come up at an ungauged location in a river, the first information we need is the discharges of the ,stream at various stages. Direct measurement of discharges involves a great deal of labour and money, and is risky at time of high floods in difficult terrains. However, it is relatively easy to make stage measurements using electronic sensors, not only in different terrains of rivers, but also at night times. Using a stage routing technique, it is envisaged to estimate the Manning's roughness coefficient for the routing reach, which can then be used for estimating the discharge at the downstream or upstream or at both the locations. Using stage measurements and the corresponding estimated discharges the rating curve can then be developed. An approximate variable parameter stage-routing method proposed by Perumal and Ranga Raju (1998a,b) is employed for routing the known stage hydrograph in a given reach characterized by a two- stage compound cross-section comprising of a trapezoidal main channel section and a trapezoidal floodplain section. In this method one set of measured input iii and corresponding output stage hydrographs of a flood event is assumed to be available at the inlet and outlet of the considered river reach. This approach of developing a rating curve is verified for a number of hypothetical data, sets (stage hydrographs at the inlet and outlet of a considered channel reach) obtained using the U.S. National Weather Service DAMBRK model, which is used as the benchmark model in this study.