Scholarlink

<p>Developing multihazard susceptibility maps is a practical approach for improving environmental planning and supporting crisis management efforts. In this study, the Khuzestan Province in Iran was selected as the case area to evaluate susceptibility to two major environmental hazards: dust storms and drought. First, a dust storm susceptibility map was generated using four machine-learning algorithms: support vector machine (SVM), boosted regression tree (BRT), random forest (RF), and maximum entropy (MaxEnt). Second, a drought susceptibility map was created based on the standardized precipitation index (SPI). These two maps were then combined to produce a multihazard susceptibility map that reflected the spatial overlap of both hazards. Among the models, the RF algorithm yielded the highest predictive performance with an area under the curve (AUC) value of 0.94. The results showed that the southern and southwestern areas of the province were the most susceptible to dust storms. For drought, the SPI results indicated that the southern, southwestern, and western parts of the province experienced drought conditions, whereas the eastern and northwestern regions were more humid. In the final stage, the susceptibility of urban infrastructure and agricultural land to these two hazards was assessed. The findings revealed that 29.88% of industrial zones and 8.89% of residential areas fell within the “very high” susceptibility class for dust storms. The outcomes of this study offer valuable input for crisis planning and land management, providing a province-scale multihazard framework aimed at minimizing potential environmental stress on infrastructure.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Mapping dust storms and drought susceptibility: a multihazard approach for reducing infrastructure risk

  • 期刊论文

  • 发表日期：

    2025年9月10日

查看全文
  • Soheila Pouyan,
  • Mojgan Bordbar,
  • Hamid Reza Pourghasemi

摘要

Developing multihazard susceptibility maps is a practical approach for improving environmental planning and supporting crisis management efforts. In this study, the Khuzestan Province in Iran was selected as the case area to evaluate susceptibility to two major environmental hazards: dust storms and drought. First, a dust storm susceptibility map was generated using four machine-learning algorithms: support vector machine (SVM), boosted regression tree (BRT), random forest (RF), and maximum entropy (MaxEnt). Second, a drought susceptibility map was created based on the standardized precipitation index (SPI). These two maps were then combined to produce a multihazard susceptibility map that reflected the spatial overlap of both hazards. Among the models, the RF algorithm yielded the highest predictive performance with an area under the curve (AUC) value of 0.94. The results showed that the southern and southwestern areas of the province were the most susceptible to dust storms. For drought, the SPI results indicated that the southern, southwestern, and western parts of the province experienced drought conditions, whereas the eastern and northwestern regions were more humid. In the final stage, the susceptibility of urban infrastructure and agricultural land to these two hazards was assessed. The findings revealed that 29.88% of industrial zones and 8.89% of residential areas fell within the “very high” susceptibility class for dust storms. The outcomes of this study offer valuable input for crisis planning and land management, providing a province-scale multihazard framework aimed at minimizing potential environmental stress on infrastructure.