부산대학교 도서관

The risk of flood and drought have been increasing due to climate change. The major rivers catchments in Korea are relatively well maintained, however, most of mid-small sized ones, which are remained as ungauged, are still in difficulties in water resources management. Therefore, a reliable estimation of discharge is needed for flood risk management in mid-small sized catchment. This study is aimed to find the most suitable Rainfall-Runoff(RR) model for the regionalisation in upper Kemgang region. 9 RR models, which are combined form 3 Soil Moisture Accounting models and 3 Routing models in Rainfall-Runoff Modeling Toolkit(RRMT), are applied to 7 sub-catchments in Miho catchment. The performance of continuous simulation of these models is investigated with Monte-Carlo Analysis Toolkit(MCAT). The period of 2004~2007 and 2008~2009 are used for calibration and validation respectively. The Nash Surcliffe Efficiency($NSE^*$=1-NSE) measures is used as objective function. The results show that CWI-2PMP(0.09-0.39 in $NSE^*$) and PDM-2PMP(0.16-0.37 in $NSE^*$) are promising models for further regionalisation research in Upper Kumgang region. The routing components (2PMP) in RR models shows that the fast response of catchments are dominant hydrological characteristics in upper Keumgang region. Further research will increase the number of catchments to draw a concrete conclusion.