Remote sensing-based river discharge estimation for a small river flowing over the high mountain regions of the Tibetan Plateau

摘要

River discharge, as one of the most essential climate variables, plays a vital role in the water cycle. Small-scale headwater catchments including high-mountain regions of Tibetan Plateau (TP) Rivers are mostly ungauged. Satellite technology shows its potential to fill this gap with a high correlation of satellite-derived effective river width and corresponding in-situ gauged discharge. This study is innovative in estimating daily river discharge using modified Manning equation (Model 1), Bjerklie equation (Model 2), and Rating curve approach (Model 3) by combining river surface hydraulic variables directly derived from remote-sensing datasets with other variables indirectly derived from empirical equations, which greatly contributes to the improvement of river flow measurement information especially over small rivers of TP. We extracted the effective width from Landsat image and flow depth via hydraulic geometry approach. All the input parameters directly or indirectly derived from remote sensing were combined and substituted into the fundamental flow equations/models to estimate discharges of Lhasa River. The validation of all three models' results against the in-situ discharge measurements shows a strong correlation (the Nash-Sutcliffe efficiency coefficient (NSE) and the coefficient of determination (R-2) values = 0.993), indicating the potentiality of the models in accurately estimating daily river discharges. Trends of an overestimation of discharge by Model 1 and underestimation by Model 2 are observed. The discharge estimation by using Model 3 outperforms Model 1 and Model 2 due to the uncertainties associated with the estimation of input parameters in the other two models. Generally, our discharge estimation methodology performs well and shows a superior result as compared with previously developed multivariate empirical equations and its application for other places globally can be the focus of upcoming studies.