Weather Radar Data for Hydrological Modelling: An Application for South of Primorye Region, Russia

Abstract

This paper describes the experience of using weather radar data to simulate a catastrophic flood caused by intense rains that occurred in the period from August 5 to 8, 2017 in the south of Primorye, Russia. The Amba River (243 km²), where historical discharge peak was measured, was chosen as the test watershed of this study. Weather radar hourly precipitation fields with a spatial resolution of 1 km and a modified version of the kinematic-wave-based geomorphologic IUH model were used to estimate the hydrograph at the watershed outlet. The simulation was performed with an hourly time step; using the precipitation grids obtained from initial radar reflectivity data, and then using bias-correction based on precipitation measurement at the closest meteorological station. In the preliminary tests, the simulated discharge was found underestimated about two times in comparison with the observed discharges. After performing a precipitation bias-correction, the simulated and observed discharges showed good accordance.