Author(s): Kenichiro Kobayashi
Linked Author(s): Kenichiro Kobayashi
Keywords: DRRFI; D4PDF; River basin; Flooding
Abstract: Effect of climate change on the extreme flood was investigated using mega ensemble climate change projections called d4PDF and a basin-scale distributed rainfall-runoff/flood-inundation (hereafter DRRFI) model. d4PDF stands for “database for Policy Decision making for Future climate change”. d4PDF is a data set of the large ensemble climate simulations with a 60 km atmospheric general circulation model for the entire globe and dynamical downscaling with a 20 km regional climate model around Japan. 3000 year data exist for September 1950 – August 2011 (60 years x 50 members = 3000 years) . Likewise, 5400 year data exist for September 2050 – August 2111 (60 years x 90 members = 5400 years). 90 members consist of 6 different SST simulations x 15 perturbations. Using this data set, the annual maximum 24 hour rainfall – return level was estimated by statistical analysis using Generalized Extreme Value (GEV) distribution. Such mega ensemble climate projections enabled to draw detailed probabilistic rainfall – return level plot. Then, using the result of the statistical analysis, possible maximum flooding of the Yodogawa River basin (catchment area: approx. 8400 km2) was investigated using DRRFI. DRRFI can simulate discharge, water level of the rivers as well as the inundation depths of the entire basin. As a result, it was observed that some members of the ensemble discharges became larger than the design flood level (before regulation) of the basin.