Prediction of Total Load Transport of an Indian Alluvial River to Estimate Unmeasured Bed Load through an Alternative Approach

Abstract

Predicting sediment transport in a natural stream is essential to adequately design different hydraulic structures like bridge piers, dam, causeway, etc., having a long service life. The prediction of sediment transport is a challenging task keeping in view the dynamic conditions of stream flow, which in turn depends upon a number of continuously and randomly changing flow parameters, channel parameters and fluid properties and thus no uniform mathematical or physical relationship can be adopted for prediction of sediment transport. The available empirical solutions, based mostly on regression, vary largely from one site condition to other. In India the bed load data is rarely measured and thus the availability of total load data for Indian alluvial river is virtually nonexistent and therefore a true empirical relationship cannot be developed for predicting total load in Indian streams. The present study aims to bridge this gap through a three-prong approach to predict the total load of an alluvial river (Shetrunji River). The unavailable (unmeasured) bed load data is computed using firstly, selected bed load transport equations and secondly, using Maddock's estimation. These computed total load (computed bed load plus observed suspended load) are compared with the total load transport predicted using Yang' 1973 and Yang' 1979 Unit Stream Power (USP) equations. It was found that the best prediction of total load is obtained for Yang' 1973 equation, when Shields (1936) bed load formula is used to compute bed load or when bed load is taken as 5% of observed suspended load. This methodology can be applied to predict the total load of rivers with reasonably good accuracy even in the absence of unmeasured bed load.