Hydrologic uncertainties in climate change from IPCC AR4 GCM simulations of the Chungju Basin, Korea

This study attempts to analyze the effects of hydrological models and potential evapotranspiration (PET) computation methods on climate change impact assessment of water resources by using Intergovernmental Panel on Climate Change (IPCC) Forth Assessment Report (AR4) General Circulation Model (GCM) simulations. Three semi-distributed hydrological models (PRMS, SLURP and SWAT) and seven different PET computation methods (Hamon and Jensen–Haise methods for PRMS, Penman–Monteith, Granger and Spittlehous-Black for SLURP, Penman–Monteith, Priestley-Taylor and Hargreaves for SWAT) are used for comparing differences of response to climate change in the Chungju Dam basin, Korea. For future climate change projections, the 13 GCM outputs with three greenhouse gas (GHG) emission scenarios are downscaled for the regional-scale hydrological model inputs by using a stochastic weather generator, WXGEN. Our results show that the hydrological models and PET methods can induce major differences in runoff change under the same climate change simulations, and that those differences are greater for 2071–2100 than for 2011–2040. The different sensitivities of PET methods to climate simulations greatly increase the range of projected runoff changes. Additionally, the differences in modeled runoff changes are smaller for the wet period (May–October) than for the dry period (November–April). This result indicates that the runoff projections for the dry season could be highly uncertain due to hydrologic models and PET methods, indicating that more caution will be needed to assess future changes in the risk of low flows and droughts.