Changes in Intensity-Duration-Frequency Curves with an Ensemble of EA-CORDEX over South Korea

An ensemble of ten regional climate model (RCM) simulations, forced by two global climate models (GCM) such as HadGEM2-AO, MPI-ESM-LR, and GFDL2M, at 25km spatial resolution from the CORDEX-EA Phase-2 is explored to assess the changes in rainfall intensity-duration-frequency (IDF), commonly employed in the hydrologic study, in a changing climate. This study first constructs a probability density function (PDF) for the observed precipitation. The log-likelihood for the modeled precipitation is then estimated from the PDF to rank the RCMs. Ensemble construction is further performed based on these rankings. The temporal downscaling approach employed in this study is based on a conditional copula function method developed by So et al. (2018), which incorporates a quantile mapping approach for bias correction. The proposed ensemble modeling framework for constructing future IDF relationships could provide a better representation of the uncertainty associated with climate models. A detailed discussion of the potential application of the ensemble approach in extreme analysis is further provided.