River discharge simulation of a distributed hydrological model on global scale for the hazard quantification

Abstract

We introduce a global river discharge simulation system based on the BTOP models for individual river basins (Global BTOP) and apply Global BTOP to quantify flood and drought hazards globally. The BTOP model utilizes a modified topographic index and simulates runoff processes including snowmelt, overland flow, soil moisture in the root and unsaturated zones, sub-surface flow, and river discharge. The Global BTOP is constructed from available river network data on the 10-arcmin (about 20-km) grid in the current version. The BTOP model topographical features are obtained using river network upscaling algorithm that preserves 3-arcsec (about 90-m) HydroSHEDS and 30-arcsec (about 1-km) Hydro1K characteristics in order to reduce the impact of upscaled dataset to the discharge simulation. In the Global BTOP, the International Geosphere-Biosphere Programme (IGBP) land cover and the Food and Agriculture Organization of the United Nations (FAO) soil digital maps of the world are used for the root zone depth and soil properties, respectively. The long-term seasonal potential evapotranspiration are estimated by the Shuttleworth-Wallace model using climate forcing data CRU TS3.1 and a fortnightly Normalized Difference Vegetation Index (NDVI). The Global BTOP is run with globally available daily precipitation datasets and calibrated to the observed global and local river discharge data. From these preliminary calibration results, the Global BTOP has demonstrated good performance in selected river basins and can be utilized for many useful applications. For the hazard quantification, we utilized simulated daily river discharges of the Global BTOP to estimate flood peak discharges of the selected return periods for the past, present and future climates. In each 10-arcmin BTOP grid, the flood peak discharges of 10-, 25-, 50-, 100- and 200-year return periods were obtained using the Gumbel distribution with L-moments. We also utilized simulated runoff of the Global BTOP to estimate standardized runoff index (SRI) for quantifying hydrologic droughts. The climate change impact assessment is conducted using GCM outputs of MRI-AGCM3.2S after applying bias correction and evaluated using flood peak discharges. For the dynamic applications of present climate, the Global BTOP is run with JAXA GSMaP- NRT precipitation to simulate daily river discharges, which are utilized in a prototype near-real time discharge simulation system (GFAS-Streamflow). The GFAS-Streamflow is designed to issue flood hazard alerts globally using estimated flood peak discharges via web-interface and may also be used as a combined flood and drought hazard monitor. This may be particularly useful to close the gap between local and global scale hazard assessments under the International Flood Initiative (IFI) Flagship Project, which aims to support benchmarking flood risk reduction at global, national and local levels.