Scholarlink

<p>This study develops a cyclone hazard and risk mitigation framework for the Acapulco region in southwestern Mexico, integrating the Analytical Hierarchy Process (AHP) and machine learning techniques using the Random Forest (RF) model to classify land use/land cover (LULC). The framework evaluates three dimensions: hazard, exposure, and vulnerability. Hazard factors include cyclone severity (Category TS-H4), storm surge height, and precipitation levels. Exposure analysis considers terrain elevation, proximity to the coastline, LULC types, and population size and density, reflecting infrastructure risks. Vulnerability assessment focuses on socio-economic indicators, such as literacy rate, disabled population, gender composition, cyclone shelters, and healthcare access. The study area, spanning 3871.88 km<sup>2</sup>, is categorized into five levels: deficient to very high hazard, exposure, and vulnerability. Results show that 26.6% of the region faces moderate hazards, with 24.9% under very high hazards, and 27.63% of the area experiencing very high exposure, indicating significant risks to infrastructure and communities. Vulnerability analysis reveals that 28.5% of the region has very high vulnerability, necessitating socio-economic interventions. Integrating hazard, exposure, and vulnerability, the study identifies 25.66% of the area as very high risk and 26.49% as high risk. These findings stress the need for improved early warning systems, expanded cyclone shelters, enhanced healthcare infrastructure, and focused support for vulnerable populations. The RF model applied to LULC data improves accuracy in evaluating exposure and assists in predicting cyclone impacts. The proposed framework offers a robust methodology for cyclone risk assessment, balancing environmental, infrastructural, and socio-economic considerations to support disaster preparedness and community resilience. It provides actionable insights for policymakers, ensuring targeted interventions to mitigate risks from future cyclones.</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

Cyclone hazard evaluation and risk mitigation strategies using machine learning approaches for the Acapulco Region and coastal communities in southwestern Mexico

  • 期刊论文

  • 发表日期：

    2025年6月6日

查看全文
  • P Maniraj Kumar,
  • B Karthikeyan,
  • S Rajeswari,
  • M Ruba

摘要

This study develops a cyclone hazard and risk mitigation framework for the Acapulco region in southwestern Mexico, integrating the Analytical Hierarchy Process (AHP) and machine learning techniques using the Random Forest (RF) model to classify land use/land cover (LULC). The framework evaluates three dimensions: hazard, exposure, and vulnerability. Hazard factors include cyclone severity (Category TS-H4), storm surge height, and precipitation levels. Exposure analysis considers terrain elevation, proximity to the coastline, LULC types, and population size and density, reflecting infrastructure risks. Vulnerability assessment focuses on socio-economic indicators, such as literacy rate, disabled population, gender composition, cyclone shelters, and healthcare access. The study area, spanning 3871.88 km2, is categorized into five levels: deficient to very high hazard, exposure, and vulnerability. Results show that 26.6% of the region faces moderate hazards, with 24.9% under very high hazards, and 27.63% of the area experiencing very high exposure, indicating significant risks to infrastructure and communities. Vulnerability analysis reveals that 28.5% of the region has very high vulnerability, necessitating socio-economic interventions. Integrating hazard, exposure, and vulnerability, the study identifies 25.66% of the area as very high risk and 26.49% as high risk. These findings stress the need for improved early warning systems, expanded cyclone shelters, enhanced healthcare infrastructure, and focused support for vulnerable populations. The RF model applied to LULC data improves accuracy in evaluating exposure and assists in predicting cyclone impacts. The proposed framework offers a robust methodology for cyclone risk assessment, balancing environmental, infrastructural, and socio-economic considerations to support disaster preparedness and community resilience. It provides actionable insights for policymakers, ensuring targeted interventions to mitigate risks from future cyclones.