Scholarlink

<p>Landslides are a recurrent hazard in the Western Ghats, India, driven by steep terrain, intense monsoonal rainfall, and increasing anthropogenic disturbances. This study delineates landslide-prone areas in Sindhudurg district, western Maharashtra, by integrating stakeholder perception with geospatial susceptibility modelling. Perception data are collected from 322 stakeholders using a structured questionnaire based on a criterion-based purposive sampling approach targeting landslide-prone villages. Statistical analyses are used to identify dominant triggering factors, while eleven causative factors are integrated using the analytical hierarchy process (AHP) within a Geographic Information System (GIS) framework. Relative weights are derived through pairwise comparison with acceptable consistency ratio (CR = 0.08). The resulting landslide susceptibility map (LSM) indicates that approximately 40% of the district falls under high to very high susceptibility, concentrated along the eastern and southern escarpments, particularly around <i>Phondaghat</i>, <i>Amboli</i>, <i>Kankavli</i>, and <i>Sawantwadi</i>. A total of 282 landslide inventory points is compiled, of which 30% are used for validation. The model yielded a receiver operating characteristic–area under the curve (ROC-AUC) value of 0.753, indicating good predictive capability. The perception-based landslide risk index (PLRI) shows a moderate positive correlation with modelled susceptibility (<i>ρ</i> = 0.34, <i>p</i> &lt; 0.05), indicating partial correspondence between physically derived susceptibility and community-perceived risk. The integration of geospatial modelling with stakeholder perception provides a practical framework for identifying and prioritizing vulnerable areas in data-limited regions. The findings support targeted land-use planning, community-based mitigation, and improved risk communication, highlighting the value of combining scientific modelling with local knowledge for landslide risk management in the Western Ghats.</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

Stakeholder-integrated landslide susceptibility mapping in the Western Ghats, India: integrating community perception with AHP-based GIS modelling

  • 期刊论文

  • 发表日期：

    2026年7月24日

查看全文
  • Amit Vijaykumar Shirke,
  • Maria Asim,
  • Samiksha Gurav,
  • Viraj Koparkar,
  • Akanksha Shrikant Kashikar,
  • Bhavana Naren Umrikar

摘要

Landslides are a recurrent hazard in the Western Ghats, India, driven by steep terrain, intense monsoonal rainfall, and increasing anthropogenic disturbances. This study delineates landslide-prone areas in Sindhudurg district, western Maharashtra, by integrating stakeholder perception with geospatial susceptibility modelling. Perception data are collected from 322 stakeholders using a structured questionnaire based on a criterion-based purposive sampling approach targeting landslide-prone villages. Statistical analyses are used to identify dominant triggering factors, while eleven causative factors are integrated using the analytical hierarchy process (AHP) within a Geographic Information System (GIS) framework. Relative weights are derived through pairwise comparison with acceptable consistency ratio (CR = 0.08). The resulting landslide susceptibility map (LSM) indicates that approximately 40% of the district falls under high to very high susceptibility, concentrated along the eastern and southern escarpments, particularly around Phondaghat, Amboli, Kankavli, and Sawantwadi. A total of 282 landslide inventory points is compiled, of which 30% are used for validation. The model yielded a receiver operating characteristic–area under the curve (ROC-AUC) value of 0.753, indicating good predictive capability. The perception-based landslide risk index (PLRI) shows a moderate positive correlation with modelled susceptibility (ρ = 0.34, p < 0.05), indicating partial correspondence between physically derived susceptibility and community-perceived risk. The integration of geospatial modelling with stakeholder perception provides a practical framework for identifying and prioritizing vulnerable areas in data-limited regions. The findings support targeted land-use planning, community-based mitigation, and improved risk communication, highlighting the value of combining scientific modelling with local knowledge for landslide risk management in the Western Ghats.