<p>This study investigates and aims to restore mangrove ecosystems in the Indus Delta, Pakistan, from 2000 to 2020 by utilizing RS and GIS techniques, integrated with the AHP and FAHP models. A multidimensional approach was employed by integrating physical parameters such as the TWI, SPI, curvature, slope, flow direction, and environmental parameters like the NDVI, NDMI, NDTI, NDSI, and rainfall alongside assessments of LULC. The primary focus was on the comprehensive assessment of LULC and the factors affecting it, to ensure the efficient management of the mangrove ecosystem. This study addresses multiple Sustainable Development Goals (SDGs) as part of the 2030 Agenda for Sustainable Development, and the commitments proposed under the Living Indus Initiative program of the government of Pakistan, which promotes ecosystem conservation and restoration efforts. The findings reveal that 56.3% of the study area is suitable for mangrove plantations based on the AHP model, while 62.5% is suitable according to the FAPH model. The 2020 LULC map shows that 30.68% was already covered with mangroves leaving 25.62% and 31.82% of the area is suitable for further restoration basedon AHP and FAHP, respectively. The identification of suitable zones for restoration was significantly influenced by factors such as LULC, NDTI, and curvature; moderately influenced by NDSI, slope, flow direction, and NDMI; but weakly influenced by other factors while using the AHP model. In the case of the FAHP model, the suitability mapping is highly dependent on LULC, NDTI, and NDSI; moderately dependent on slope, curvature, NDMI, and NDVI; while minimally dependent on other parameters. The FAHP model yielded more optimistic results, demonstrating its superiority in capturing interdependencies and handling uncertainty in decision-making. This approach is suggested as an effective decision-making tool for managing and restoring mangrove-covered areas, offering valuable insights for future efforts.</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 mangroves land suitability using machine learning and MCDM in the Indus Delta, Pakistan

  • Sher Shah Hassan,
  • Muhammad Ajmal,
  • Muhammad Arif Goheer,
  • Zainab Ali,
  • Muhammad Suleman

摘要

This study investigates and aims to restore mangrove ecosystems in the Indus Delta, Pakistan, from 2000 to 2020 by utilizing RS and GIS techniques, integrated with the AHP and FAHP models. A multidimensional approach was employed by integrating physical parameters such as the TWI, SPI, curvature, slope, flow direction, and environmental parameters like the NDVI, NDMI, NDTI, NDSI, and rainfall alongside assessments of LULC. The primary focus was on the comprehensive assessment of LULC and the factors affecting it, to ensure the efficient management of the mangrove ecosystem. This study addresses multiple Sustainable Development Goals (SDGs) as part of the 2030 Agenda for Sustainable Development, and the commitments proposed under the Living Indus Initiative program of the government of Pakistan, which promotes ecosystem conservation and restoration efforts. The findings reveal that 56.3% of the study area is suitable for mangrove plantations based on the AHP model, while 62.5% is suitable according to the FAPH model. The 2020 LULC map shows that 30.68% was already covered with mangroves leaving 25.62% and 31.82% of the area is suitable for further restoration basedon AHP and FAHP, respectively. The identification of suitable zones for restoration was significantly influenced by factors such as LULC, NDTI, and curvature; moderately influenced by NDSI, slope, flow direction, and NDMI; but weakly influenced by other factors while using the AHP model. In the case of the FAHP model, the suitability mapping is highly dependent on LULC, NDTI, and NDSI; moderately dependent on slope, curvature, NDMI, and NDVI; while minimally dependent on other parameters. The FAHP model yielded more optimistic results, demonstrating its superiority in capturing interdependencies and handling uncertainty in decision-making. This approach is suggested as an effective decision-making tool for managing and restoring mangrove-covered areas, offering valuable insights for future efforts.