Scholarlink

<p>Coastal regions of the Global South, particularly rapidly urbanizing areas like Chattogram, Bangladesh, face significant ecological deterioration requiring long term geospatial assessment of ecosystem service value (ESV) dynamics. This study employed an integrated methodology combining random forest classification of Landsat imagery (2000 to 2020), the Patch-generating Land Use Simulation (PLUS) model for prediction (2040 and 2060), and the Geographical Detector Model (GDM) to analyze land use and land cover change drivers. Historical classifications demonstrated high accuracy with Kappa Coefficients ranging from 0.81 to 0.88. The PLUS model showed strong predictive robustness with low error metrics for waterbody (RMSE 0.08) and vegetation (RMSE 0.09). Findings reveal intense urbanization: settlement areas rose from 5.35% in 2000 to 25.07% in 2020. Under business-as-usual trajectories, settlements are projected to reach 33.65% by 2060, driving vegetation decline from 73.52 to 41.28%. This transformation correlates with a projected 43.85% decline in vegetation ESV. Although settlement ESV is projected to increase to $1397.75 million by 2060, this compensates for only three service indicators compared to 21 provided by vegetation. Analysis identified slope (Power of Determinant 0.512), proximity to railway (0.480), and Digital Elevation Model (0.390) as significant ESV drivers. Sensitivity analysis confirmed ESV reliability, and Monte Carlo simulations bounded overall ESV decline from $5206 million (2000) to $4821 million (2060). These results underscore critical trade-offs between urbanization and ecosystem services, demanding immediate policy revisions to implement mandatory green buffer zones along railway corridors and strict protection zoning for hills to achieve sustainable urban planning and ecosystem management.</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 simulation of land use change and ecosystem services in Chattogram, Bangladesh using PLUS model (2000–2060)

  • 期刊论文

  • 发表日期：

    2025年11月26日

查看全文
  • T. Ahmed,
  • I. Jahan

摘要

Coastal regions of the Global South, particularly rapidly urbanizing areas like Chattogram, Bangladesh, face significant ecological deterioration requiring long term geospatial assessment of ecosystem service value (ESV) dynamics. This study employed an integrated methodology combining random forest classification of Landsat imagery (2000 to 2020), the Patch-generating Land Use Simulation (PLUS) model for prediction (2040 and 2060), and the Geographical Detector Model (GDM) to analyze land use and land cover change drivers. Historical classifications demonstrated high accuracy with Kappa Coefficients ranging from 0.81 to 0.88. The PLUS model showed strong predictive robustness with low error metrics for waterbody (RMSE 0.08) and vegetation (RMSE 0.09). Findings reveal intense urbanization: settlement areas rose from 5.35% in 2000 to 25.07% in 2020. Under business-as-usual trajectories, settlements are projected to reach 33.65% by 2060, driving vegetation decline from 73.52 to 41.28%. This transformation correlates with a projected 43.85% decline in vegetation ESV. Although settlement ESV is projected to increase to $1397.75 million by 2060, this compensates for only three service indicators compared to 21 provided by vegetation. Analysis identified slope (Power of Determinant 0.512), proximity to railway (0.480), and Digital Elevation Model (0.390) as significant ESV drivers. Sensitivity analysis confirmed ESV reliability, and Monte Carlo simulations bounded overall ESV decline from $5206 million (2000) to $4821 million (2060). These results underscore critical trade-offs between urbanization and ecosystem services, demanding immediate policy revisions to implement mandatory green buffer zones along railway corridors and strict protection zoning for hills to achieve sustainable urban planning and ecosystem management.