Estimation of critical streamflow discharge level using quantile regression approach

Effective flood risk measures are important tools for river catchments' management. Among others, flood hazard mitigation programs and insurance companies can actively use these measures for reservoir operations, relief effort planning and premium setting, respectively. This work develops a nonparametric quantile regression approach for the estimation of flood quantiles conditional on past strearnflow discharges. The resulting estimator of conditional quantiles is found to be consistent and asymptotically normal. Real data from a streamflow gauging station in the Nyando Basin, Western Kenya, is used to demonstrate the potential of the approach.