摘要
[目的]识别灾害的敏感性是防灾减灾的重要前提,复合灾害系统敏感性评估是复合灾害系统风险评估的重要组成部分.[方法]首先根据暴雨—滑坡灾害链特征构建复合灾害系统敏感性评估指标体系,进而采用卷积神经网络(CNN)和参数最优地理探测器—层次分析法耦合模型(OPGD-AHP)对粤港澳大湾区的暴雨—滑坡灾害链敏感性进行评估.通过对暴雨和滑坡单灾种的敏感性分析,分别得到粤港澳大湾区暴雨和滑坡的敏感性分级图,并由统计分析得出暴雨诱发滑坡的阈值及其转换概率,以此对暴雨-滑坡灾害链敏感性进行分析,最终得到粤港澳大湾区暴雨—滑坡灾害链敏感性分级图.[结果](1)粤港澳地区暴雨-滑坡灾害链高敏感区域主要分布在广宁县、怀集县等西北部地区和惠阳区、惠东县等东部地区;(2)次高敏感性和中敏感性区域主要分布在封开、德庆、从化、龙门、博罗等区域;(3)低敏感区和次低敏感区主要分布在南海、荔湾、禅城、顺德、南沙、番禺、三水、白云等中部地区.[结论]研究结果可为粤港澳大湾区的防灾减灾、增强灾害链风险抵抗力和恢复力提供科学依据和技术支持.
Abstract
[Objective]Identifying the sensitivity of disasters is an important prerequisite for disaster prevention and reduction,and sensitivity assessment of complex disaster systems is an important component of risk assessment of composite disaster systems.[Methods]Firstly,the sensitivity assessment index system of the composite disaster system was constructed according to the characteristics of the rainstorm landslide disaster chain,and then the susceptibility of the rainstorm-landslide disaster chain in Guangdong-Hong Kong-Macao Greater Bay Area was evaluated by using Convolutional Neural Network(CNN)and Optimal Parameters-based Geographical Detectors-Analytic Hierarchy Process(OPGD-AHP).Through the sensitivity analysis of rainstorm and landslide,the sensitivity classification maps of rainstorm and landslide in the Greater Bay Area were obtained respectively.And the threshold value and conversion probability of landslide induced by rainstorm were obtained through statistical analysis,so as to analyze the sensitivity of rainstorm landslide disaster chain and finally the sensitivity classification map of the rainstorm-landslide disaster chain in the Greater Bay Area was obtained.[Results](1)The highly sensitive areas of the rainstorm landslide disaster chain in Guangdong,Hong Kong and Macao were mainly distributed in the northwest of Guangning County and Huaiji County,and in the east of Huiyang District and Huidong County.(2)The sub high sensitivity and medium sensitivity areas were mainly distributed in areas such as Fengkai,Deqing,Conghua,Longmen,Boluo,etc.(3)Low sensitivity areas and sub low sensitivity areas were mainly distributed in central areas such as South China Sea,Liwan,Chancheng,Shunde,Nansha,Panyu,Sanshui,Baiyun,etc.[Conclusion]The research results could provide scientific basis and technical support for disaster prevention and reduction,as well as enhancing the resistance and resilience of the disaster chain in Guangdong-Hong Kong-Macao Greater Bay Area.
基金项目
国家重点研发计划项目(2021YFC3001002)
国家自然科学基金(71974052)
维普
万方数据