A weighted, multi-method approach for accurate basin-wide streamflow estimation in an ungauged watershed

Abstract

River discharge is a commonly measured hydrologic variable; however, estimate uncertainty is often higher than acceptable limits. To quantify method limitations and spatiotemporal variability, a multi-year hydrologic flow partitioning investigation was completed under monsoonal conditions in the ungauged complex terrain of the Haean Catchment, South Korea. Our results indicate that sediment transport from a single annual monsoonal event can significantly modify the channel cross-sectional area resulting in inaccurate stage-discharge rating curves. We compare six discharge measurement methods at 13 locations that vary in slope from 1% to 80%, with discharge ranging up to four orders in magnitude, which enabled us to weight the accuracy of each method over a specific range in discharge. The most accurate discharge estimation methods are the weir, the acoustic Doppler current profiler, and the in-stream velocity area method; however, under certain conditions each of these methods is less desirable than other methods. The uncertainty in the three methods is on average 0.4%, 4.7%, and 6.1% of the total discharge, respectively. The accuracy of the discharge estimates has a direct influence on the characterization of basin-wide hydrologic partitioning, which can lead to significant variability in sediment erosion rates and nutrient fate and transport. © 2013.