Scholarlink

<p>The monitoring and assessment of the water environment are essential for the accurate demarcation of flood-prone areas, biodiversity hotspots, and wetland inventories. Although presently available remote sensing-based indices heavily rely upon thresholding for the precise demarcation of water-spread areas, dynamic nature, environmental complexities, problems related to mixed pixels, and the effect of shadow from surrounding features affect their outcome. Further, challenges also arise due to the high sensitivity of shortwave infrared bands to water in waterlogged vegetative areas and wetlands infested by aquatic plants. This study presents a new robust water index without thresholding requirement—the ‘Advanced Multi-Spectral Water Index’ (AMSWI)—using visible and near infra-red bands of 10&#xa0;m resolution derived from Sentinel 2 with enhanced robustness from the errors related to spectral and spatial variability in the effective assessment of open water bodies. Eight wetlands located in different environmental and climatic conditions of the Indian subcontinent have been chosen for the testing and validation of the index with the three widely used water indices. The AMSWI performed best among all the indices, with the highest F1 score of 0.98. In contrast, AWEI obtained an overall accuracy of 71.4% and 83.2% in Sundarbans mangroves and the high-altitude wetlands of Tso Moriri, respectively. At the same time, MNDWI achieved an overall accuracy of 36.1% in Sundarbans mangroves and 84.5% in Tso Moriri. Moreover, the robustness of AMSWI is also evident from the stability of the optimal threshold value of zero for different sites during different seasons.</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

Assessment of surface area changes of open water bodies in wetlands using a new water index

  • 期刊论文

  • 发表日期：

    2025年5月2日

查看全文
  • Prema Somanathan Smitha,
  • Kesavan Maya,
  • K. P. Sudheer,
  • Vijayalekshmi Muraleedharan Bindhu,
  • Krishnan Sreelash,
  • Damodharan Padmalal

摘要

The monitoring and assessment of the water environment are essential for the accurate demarcation of flood-prone areas, biodiversity hotspots, and wetland inventories. Although presently available remote sensing-based indices heavily rely upon thresholding for the precise demarcation of water-spread areas, dynamic nature, environmental complexities, problems related to mixed pixels, and the effect of shadow from surrounding features affect their outcome. Further, challenges also arise due to the high sensitivity of shortwave infrared bands to water in waterlogged vegetative areas and wetlands infested by aquatic plants. This study presents a new robust water index without thresholding requirement—the ‘Advanced Multi-Spectral Water Index’ (AMSWI)—using visible and near infra-red bands of 10 m resolution derived from Sentinel 2 with enhanced robustness from the errors related to spectral and spatial variability in the effective assessment of open water bodies. Eight wetlands located in different environmental and climatic conditions of the Indian subcontinent have been chosen for the testing and validation of the index with the three widely used water indices. The AMSWI performed best among all the indices, with the highest F1 score of 0.98. In contrast, AWEI obtained an overall accuracy of 71.4% and 83.2% in Sundarbans mangroves and the high-altitude wetlands of Tso Moriri, respectively. At the same time, MNDWI achieved an overall accuracy of 36.1% in Sundarbans mangroves and 84.5% in Tso Moriri. Moreover, the robustness of AMSWI is also evident from the stability of the optimal threshold value of zero for different sites during different seasons.