Entropy-Based Flow and Sediment Routing in Data Deficit River Networks

The reliable estimate of the sediment load and streamflow is essential for water resources and flood management. In this study, the entropy-based technique and HEC-RAS are used for flow routing followed by sediment routing in HEC-RAS. The paper’s novelty is its application to data-deficit river networks, where observed sediment load and flow on tributaries are absent. The proposed method accommodates the flow and sediment contribution from the tributaries to the downstream station on a reach, despite unavailable observed data on it. The adopted flow routing techniques are applied to predict downstream flow on three different reaches (on the Mahanadi and the Godavari River). The prediction accuracy is evaluated using three statistical indices ‒ Nash–Sutcliffe efficiency (NSE), relative error (RE), and Coefficient of determination (R 2 ). Both flow routing techniques showed good performance for all three reaches (with or without tributaries), having NSE, R 2 > 0.8, and RE < 13%. Despite the comparable performance, the entropy-based routing is suggested for natural rivers with or without tributary as it avoids the iterative calibration process to determine the roughness coefficient. Further, the sediment routing is performed on the data-deficit reach of the Mahanadi River to obtain the best-suited sediment transport function. The simulated sediment load using the Yang transport function matched satisfactorily with the observed data with NSE, R 2 > 0.85, and RE < –27%. Subsequently, the Yang transport function and entropy-based flow routing are utilized for the sediment and flow estimation at an ungauged station on the Mahanadi river.