Flood Risk Assessment by Estimating Flood Depth Considering Uncertainties in Climate Change

Flood depth;Flood susceptibility;Probability distribution of precipitation;Extreme rainfall;Adaptation to flood;Spatial planning;712

농업생명과학대학 생태조경·지역시스템공학부 ;

University:서울대학교 대학원

关键词: Flood depth;    Flood susceptibility;    Probability distribution of precipitation;    Extreme rainfall;    Adaptation to flood;    Spatial planning;    712;   

Others  :  http://s-space.snu.ac.kr/bitstream/10371/125473/1/000000056884.pdf

美国|英语

来源: Seoul National University Open Repository

【 摘 要 】

A natural risk common in urban areas, flooding caused by localized heavy rain has shown an increase under climate change. Flood risk must be considered from the initial stages of urban planning because of limited expansion in developed areas. However, few studies have attempted to quantify flood risk under uncertainties for land use planning. In addition, most models used to derive flood depth are usually complicated and are not easily employed by land use planners. Therefore, the objectives of this study are to develop a predictive model for flood depth and assess flood risk considering the uncertainties of precipitation in a future climate scenario. Gimpo city for the study site is prone to flooding and includes large areas under development. This study is presented in four parts. First, Monte Carlo simulation is conducted to create ensemble Representative Concentration Pathway (RCP) scenarios for considering uncertainties of precipitation. Second, multi-regression analysis was performed to define the flood depth. Third, the flood susceptibility was estimated by using Maximum Entropy (MaxEnt) modeling software. Finally, a flood risk map is derived by superposing the flood depth map in 2050 onto the flood susceptibility map. The result of ensemble scenarios derived from Monte Carlo simulation indicated an increase in extreme rainfall in 2050. Moreover, the maximum precipitation of daily and accumulated precipitation are expected to increase as much as 106.76 mm and 55.66 mm, respectively, compared with current conditions. A predictive model for flood depth was derived with independent variables of daily precipitation, accumulated precipitation, relative elevation to the nearest stream, and Topographic Position Index (TPI). The flood depth caused by daily precipitation of extreme rainfall would be 0.01?0.05 m at the safest areas and 1.64?2.34 m at the most vulnerable areas. The flood depth would be significantly higher in the case of accumulated precipitation. Flood susceptibility was divided into eight classes representing areas vulnerable to floods according to environmental conditions. The areas of classes 5 to 8, flood susceptible areas, would account for 28% in Gimpo. Areas with flood depth over 1.0 m, Class 7, are expected to cover 7,200 m². Such areas may be vulnerable for all land use type and thus should not be developed.This study is significant because land use planners can use the results to establish decision-making criteria for reasonable planning. In the new paradigm of land use planning under climate change, this study can be used as a basic research guide for determining the installation of adaptation strategies to land use allocation.