통계적 분석기법을 통한 하천 정비사업의 효과도 평가

Abstract

To prevent and prepare for natural disasters, the government is implementing Hazard Mitigation Projects(H.M.P) for Natural Disaster Risk Reduction Districts. This study is to develop a water level prediction model(W.L.P.M) using multiple linear regression analysis based on the measurement data of rainfall and water level to evaluate the effect of the H.M.P on the prevention of river disasters. For this, a point where the effect of the H.M.P in the Mihocheon Basin can be reviewed was selected, and the time scale of the independent variable was determined through autocorrelation analysis. The regression coefficient for each model was calculated using the stepwise selection method, and the highest-accuracy W.L.P.M was selected using statistical indicators. To evaluate the effect of the H.M.P by comparing the water level prediction results using the W.L.P.M and the actual water level in the rainfall events that occurred after the H.M.P was implemented. After the H.M.P was implemented, the water level increased compared to the W.L.P.M, and the water level before and after the peak flow also decreased. Based on the results of this study, it is judged that the W.L.P.M can be used to evaluate the effectiveness of H.M.