Scholarlink

<p>This study develops a Coastal Erosion Vulnerability Map (CEM) for the Puri coast of Odisha, India, focusing on its ecologically fragile coastal and marine environments. The increasing frequency and intensity of coastal erosion events along the Puri coast, driven by both natural and anthropogenic factors, necessitated a booming, data-driven vulnerability assessment framework. The lack of high-resolution, spatially explicit vulnerability maps for Odisha’s coastline leads to the integration of geospatial analysis with machine learning to address this critical knowledge gap. Integrating geospatial analysis and machine learning techniques, it examines shoreline changes from Landsat satellite images (2000–2020) to assess coastal erosion vulnerability. Using Random Forest algorithms, the research identifies 13 Coastal Erosion Vulnerability Factors (CEVFs), including land surface elevation, wave characteristics, and land use/land cover, offering a detailed understanding of vulnerability dynamics. The Digital Shoreline Analysis System (DSAS) maps erosion, accretion, and stability across 694 transects. The study reveals a mean shoreline change of 0.12&#xa0;m, with significant regional variability. Vulnerability levels are categorized into five classes, with areas like Puri Sadar exhibiting high vulnerability and Kakatpur showing resilience. The study underscores the importance of wave characteristics and environmental factors in determining vulnerability, proposing targeted mitigation strategies like shoreline stabilization and vegetation restoration, alongside integrated coastal zone management. The main challenge faced in this study is integrating diverse geospatial data and machine learning techniques to accurately assess and predict coastal erosion vulnerability in a dynamic and ecologically sensitive region.</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

Geospatial and machine learning-based coastal vulnerability assessment of Puri, Odisha

  • 期刊论文

  • 发表日期：

    2025年10月25日

查看全文
  • Nihar Ranjan Parida,
  • Subhasmita Parida,
  • Pinaki Samal,
  • Abisha P

摘要

This study develops a Coastal Erosion Vulnerability Map (CEM) for the Puri coast of Odisha, India, focusing on its ecologically fragile coastal and marine environments. The increasing frequency and intensity of coastal erosion events along the Puri coast, driven by both natural and anthropogenic factors, necessitated a booming, data-driven vulnerability assessment framework. The lack of high-resolution, spatially explicit vulnerability maps for Odisha’s coastline leads to the integration of geospatial analysis with machine learning to address this critical knowledge gap. Integrating geospatial analysis and machine learning techniques, it examines shoreline changes from Landsat satellite images (2000–2020) to assess coastal erosion vulnerability. Using Random Forest algorithms, the research identifies 13 Coastal Erosion Vulnerability Factors (CEVFs), including land surface elevation, wave characteristics, and land use/land cover, offering a detailed understanding of vulnerability dynamics. The Digital Shoreline Analysis System (DSAS) maps erosion, accretion, and stability across 694 transects. The study reveals a mean shoreline change of 0.12 m, with significant regional variability. Vulnerability levels are categorized into five classes, with areas like Puri Sadar exhibiting high vulnerability and Kakatpur showing resilience. The study underscores the importance of wave characteristics and environmental factors in determining vulnerability, proposing targeted mitigation strategies like shoreline stabilization and vegetation restoration, alongside integrated coastal zone management. The main challenge faced in this study is integrating diverse geospatial data and machine learning techniques to accurately assess and predict coastal erosion vulnerability in a dynamic and ecologically sensitive region.