Scholarlink

<p>Accurate prediction of Land Use and Land Cover (LULC) change and its ecological impacts is critical for effective landscape management. However, many existing studies do not adequately incorporate dynamic environmental and socioeconomic drivers and often rely on limited validation approaches, reducing their applicability for planning. This study addresses these gaps by applying an integrated framework that combines a Cellular Automata Artificial Neural Network (CA–ANN) model with the Integrated Valuation of Ecosystem Services and Trade-offs (InVEST) model to simulate future LULC dynamics and assess habitat quality. Future LULC patterns were modeled using multiple drivers and validated using spatial metrics, including the Figure of Merit (FoM), Fuzzy Similarity Index (FSI), Shannon Diversity Index (SHDI), and Persistence Ratio (PR). The model demonstrated strong performance, with overall accuracy exceeding 93% and a Kappa coefficient above 0.92. Between 1992 and 2022, built-up areas more than doubled, agricultural land expanded, and forest cover declined. Habitat quality decreased from 81 to 60% and is projected to stabilize by 2092, with most forest areas classified as medium risk. The integrated framework provides a practical and reproducible approach for linking LULC prediction with ecosystem assessment, supporting long-term planning and sustainable resource 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

An integrated framework for mapping future land use and assessing habitat quality in a transboundary basin

  • 期刊论文

  • 发表日期：

    2026年7月9日

查看全文
  • Baragamba Stephano Ernest,
  • Ke Zhang,
  • Mohammed Abdallah,
  • Xu Yang,
  • Gazi Tawfiq Ezaz

摘要

Accurate prediction of Land Use and Land Cover (LULC) change and its ecological impacts is critical for effective landscape management. However, many existing studies do not adequately incorporate dynamic environmental and socioeconomic drivers and often rely on limited validation approaches, reducing their applicability for planning. This study addresses these gaps by applying an integrated framework that combines a Cellular Automata Artificial Neural Network (CA–ANN) model with the Integrated Valuation of Ecosystem Services and Trade-offs (InVEST) model to simulate future LULC dynamics and assess habitat quality. Future LULC patterns were modeled using multiple drivers and validated using spatial metrics, including the Figure of Merit (FoM), Fuzzy Similarity Index (FSI), Shannon Diversity Index (SHDI), and Persistence Ratio (PR). The model demonstrated strong performance, with overall accuracy exceeding 93% and a Kappa coefficient above 0.92. Between 1992 and 2022, built-up areas more than doubled, agricultural land expanded, and forest cover declined. Habitat quality decreased from 81 to 60% and is projected to stabilize by 2092, with most forest areas classified as medium risk. The integrated framework provides a practical and reproducible approach for linking LULC prediction with ecosystem assessment, supporting long-term planning and sustainable resource management.