GC045-13
Towards more Accurate Riverine Flood Forecasting over the Lower Mekong Basin: Assessment of the Rainfall Forecast CHIRPS-GEFS for the Mekong River Commission
Wednesday, 9 December 2020: 07:36
Virtual
Chinaporn Meechaiya1, Miguel Laverde1, Lam Hung Son2, Khem Sothea2, Arjen Haag3, Martijn Kwant3, Poortinga Ate4, Amanda (Weigel) Markert5, Shraddhanand Shukla6, Martin Francis Landsfeld6, Gregory J Husak6, Farrukh Chishtie4, Peeranan Towashiraporn1 and David S Saah7, (1)Asian Disaster Preparedness Center, SERVIR-Mekong, Bangkok, Thailand, (2)Mekong River Commission, Phnom Penh, Cambodia, (3)Deltares, Delft, Netherlands, (4)Spatial Informatics Group, LLC, Alameda, CA, United States, (5)University of Alabama in Huntsville, Earth System Science Center, Huntsville, AL, United States, (6)University of California Santa Barbara, Climate Hazards Group, Santa Barbara, CA, United States, (7)University of San Francisco, Geospatial Analysis Lab, San Francisco, CA, United States
Abstract:
The Mekong River is the longest river in Southeast Asia and flows through six countries: China, Myanmar, Thailand, Lao PDR, Cambodia and Vietnam. Countries in the Lower Mekong region are subject to seasonal floods caused by heavy monsoon and tropical storms. The majority of people who live in the region maintain water-related livelihoods and are, therefore, heavily impacted by floods. In contrast to many other disasters, floods can be predicted from weather data and warnings can be issued to ensure people’s safety and reduce damage from the flood. Since all countries in the Lower Mekong region rely on such flood early warnings and weather forecasts to take preparatory action, there is a major need to increase the lead-time of flood forecasts as well as its accuracy and spatial resolution. This will help overcome the region’s challenge to improve accuracy and spatio-temporal resolution for flood warning and emergency responses.
At the regional level, the Mekong River Commission (MRC) is mandated by its member countries to support riverine flood forecasting and other related environmental issues, providing early warnings and real-time updates via its information portal to support flood management. Through enhancing the MRC Regional Flood Forecasting Service, SERVIR-Mekong supports the MRC to use Climate Hazards Group InfraRed Precipitation with Station data - Global Ensemble Forecast System (CHIRPS-GEFS), a recent innovative forecast product which bias-corrects and downscales GEFS forecast with respect to CHIRPS. The objective is to improve the middle-term riverine forecasting by increasing both the spatial-temporal resolution and accuracy over the Lower Mekong Basin. This study is built upon MRC’s current use of the Global Forecast System (GFS). To evaluate the product, we compared GFS and CHIRPS-GEFS for the monsoon season in 2017-2019, using the bias-corrected GPM IMERG early version as reference. The results for the 6 days rainfall forecast show that CHIRPS-GEFS has a lower root-mean-square error (RMSE) and a higher Probability of Detection (POD) and Critical Success Index (CSI).This encouraged MRC to implement CHIRPS-GEFS into their Flood Forecasting System in an operational setting to improve the lead time, accuracy and spatial resolution of riverine flood Early Warning in the Lower Mekong Basin.
