GLACIAL LAKE OUTBURST FLOOD FREQUENCY ANALYSIS IN THE SAPTA KOSHI RIVER BASIN OF NEPAL

Abstract

In recent years, Glacial Lake Outburst Floods (GLOF) has become a huge challenge for any infrastructural development in the Sapta Koshi basin, the most flood prone river of Nepal. This study presents the empirical and statistical analysis of GLOF discharges to estimate the event magnitude and frequency of its occurrence. The empirical approach requires comparison of discharge calculated using the four established methods for GLOF peak estimation with the measured peak at gauging stations. The statistical analysis involves estimation of design flood using Gumbel (EV1) distribution and the standard error associated with it. For the GLOF frequency analysis, instantaneous maximum and daily maximum discharges of the day of the event were obtained from literature and from the Department of Hydrology and Meteorology (DHM) of Nepal. The result based on empirical analysis shows that among the all four methods compared, the calculated discharge values using the Walder and Costa method are closest to the measured GLOF values. The probability plot shows that datasets follow the EV1 distribution quite well. Further, the flood frequency analysis shows that designed values obtained using the EV1 model and published design flood peaks of the Koshi basin lies within 95% confidence Interval. However, higher calculated design flood values in comparison to published values highlights the need of GLOF and Seasonal High Floods (SHF) to be analyzed individually. For more accurate estimation establishment of more gauging stations and regular study on GLOF is strongly recommended.