Yinan Wang , Juan Nie , Zhenxiang Xing , Zhenbo Wang , Chengdong Xu , Heng Li
{"title":"Assessing and mitigating dwelling collapse risk due to extreme precipitation: A comprehensive study using CNN-RF and GeoDetector","authors":"Yinan Wang , Juan Nie , Zhenxiang Xing , Zhenbo Wang , Chengdong Xu , Heng Li","doi":"10.1016/j.ijdrr.2024.104918","DOIUrl":null,"url":null,"abstract":"<div><div>Dragon boat rain, the most common extreme precipitation form in South China from May to June with more similar spatial distribution, caused serious loss of people's lives and property. The dwelling collapse is one of the main losses. Previous studies have paid little attention to the dwelling collapse risk caused by dragon boat rain (DCRDBR), the coupling model with CNN and RF applied to its assessment, and the influence of the precipitation process and interaction of natural and social factors on it. To fill these gaps, the CNN-RF was used to calculate the DCRDBR and the DCRDBR map was drawn. The Geodetctor was used to identify the main influencing factors and influencing factor interactions of DCRDBR, due to the spatial stratified heterogeneity of DCRDBR and the ability to obtain the determinant power of single factor and factor interaction. The results show that the F1 score and the AUC value of CNN-RF are 0.96 and 0.81, respectively. The spatial distribution of DCRDBR obtained by CNN-RF is high in the northeast and low in the southwest Guangdong Province. The total precipitation has the strongest determinant power (q = 0.54) followed by Slope (q = 0.52). The average determinant power of factors describing the precipitation process is 0.25. The combination of total precipitation and GDP/capita has the strongest determinant power of all combinations of natural and socio-economic factors (q = 0.72) followed by the total precipitation and ratio of urban population (q = 0.71). This study demonstrates the ability of CNN-RF applied to the DCRDBR assessment due to the integration of feature extraction and anti-overfitting ability, and identifies the influence of precipitation processes and the interaction of natural and socio-economic factors on the DCRDBR. It provides a solid scientific basis for crafting strategies to mitigate the impact of dragon boat rain and is conducive to the city's sustainable development.</div></div>","PeriodicalId":13915,"journal":{"name":"International journal of disaster risk reduction","volume":null,"pages":null},"PeriodicalIF":4.2000,"publicationDate":"2024-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"International journal of disaster risk reduction","FirstCategoryId":"89","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S2212420924006800","RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"GEOSCIENCES, MULTIDISCIPLINARY","Score":null,"Total":0}
引用次数: 0
Abstract
Dragon boat rain, the most common extreme precipitation form in South China from May to June with more similar spatial distribution, caused serious loss of people's lives and property. The dwelling collapse is one of the main losses. Previous studies have paid little attention to the dwelling collapse risk caused by dragon boat rain (DCRDBR), the coupling model with CNN and RF applied to its assessment, and the influence of the precipitation process and interaction of natural and social factors on it. To fill these gaps, the CNN-RF was used to calculate the DCRDBR and the DCRDBR map was drawn. The Geodetctor was used to identify the main influencing factors and influencing factor interactions of DCRDBR, due to the spatial stratified heterogeneity of DCRDBR and the ability to obtain the determinant power of single factor and factor interaction. The results show that the F1 score and the AUC value of CNN-RF are 0.96 and 0.81, respectively. The spatial distribution of DCRDBR obtained by CNN-RF is high in the northeast and low in the southwest Guangdong Province. The total precipitation has the strongest determinant power (q = 0.54) followed by Slope (q = 0.52). The average determinant power of factors describing the precipitation process is 0.25. The combination of total precipitation and GDP/capita has the strongest determinant power of all combinations of natural and socio-economic factors (q = 0.72) followed by the total precipitation and ratio of urban population (q = 0.71). This study demonstrates the ability of CNN-RF applied to the DCRDBR assessment due to the integration of feature extraction and anti-overfitting ability, and identifies the influence of precipitation processes and the interaction of natural and socio-economic factors on the DCRDBR. It provides a solid scientific basis for crafting strategies to mitigate the impact of dragon boat rain and is conducive to the city's sustainable development.
期刊介绍:
The International Journal of Disaster Risk Reduction (IJDRR) is the journal for researchers, policymakers and practitioners across diverse disciplines: earth sciences and their implications; environmental sciences; engineering; urban studies; geography; and the social sciences. IJDRR publishes fundamental and applied research, critical reviews, policy papers and case studies with a particular focus on multi-disciplinary research that aims to reduce the impact of natural, technological, social and intentional disasters. IJDRR stimulates exchange of ideas and knowledge transfer on disaster research, mitigation, adaptation, prevention and risk reduction at all geographical scales: local, national and international.
Key topics:-
-multifaceted disaster and cascading disasters
-the development of disaster risk reduction strategies and techniques
-discussion and development of effective warning and educational systems for risk management at all levels
-disasters associated with climate change
-vulnerability analysis and vulnerability trends
-emerging risks
-resilience against disasters.
The journal particularly encourages papers that approach risk from a multi-disciplinary perspective.