4.7 Article

Threats of climate and land use change on future flood susceptibility

期刊

JOURNAL OF CLEANER PRODUCTION
卷 272, 期 -, 页码 -

出版社

ELSEVIER SCI LTD
DOI: 10.1016/j.jclepro.2020.122757

关键词

Natural disaster; SVM; RF; BBO; Flood prone areas

向作者/读者索取更多资源

Management of any unstable situation or conditions helps to return it to its natural state. In recent times, disaster management is required for sustainable management. Researchers, geographers, climatologists, regional planners in every country are trying to control and mitigate the effect of disasters. The Ajoy River basin of eastern India is facing the acute problem of flood in every monsoon period and this region is highly populated as well as agricultural productivity is very high. So, this region is taken into consideration as a study region through various natural and anthropogenic factors. The current research work is presenting flood susceptible areas in Ajoy River basin using Support Vector Machine (SVM), Random Forest (RF) and Biogeography Based Optimization (BBO) model in GIS environment. To establish the final result different flood susceptibility causal factors (topographical, hydrological, soil characteristics, environmental and geological) are envisaged. The models result was evaluated through Area Under Curve (AUC) values with Receiver Operating Characteristics (ROC). The condition or environment of flood is shown by AUC. When AUC values shows more than 0.95 then there will be strong probability of flood occurrence. Here the AUC values for the BBO, RF and SVM are 0.985, 0.925 and 0.896 respectively. In this study out of three models, BBO showed us best result (AUC 1/4 0.98) and made assurance that BBO is a prominent method for identifying the flood prone areas in eastern part of India, a monsoon dominated region and help to take managing step for regional mitigation as well as development planning. (c) 2020 Elsevier Ltd. All rights reserved.

作者

我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。

评论

主要评分

4.7
评分不足

次要评分

新颖性
-
重要性
-
科学严谨性
-
评价这篇论文

推荐

暂无数据
暂无数据