Scholarlink

<p>Understanding land use and land cover (LULC) changes is essential for addressing challenges related to urban expansion, environmental degradation, and resource management. Despite the widespread application of LULC prediction models, many lack comprehensive evaluation frameworks that incorporate sensitivity analyses to assess the influence of driving factors. This study aims to analyze changes in land use in northwest Syria, which has undergone profound LULC transformations. These changes are influenced by socio-economic developments and environmental degradation, making this area a critical case for sustainable land management research. An integrated modeling approach was developed by combining a Multi-layer Perceptron-Artificial Neural Network (MLP-ANN) with Random Forest (RF), Classification and Regression Tree (CART), and Shapley Additive Explanations (SHAP) to perform sensitivity analysis. This hybrid methodology not only predicts future LULC patterns with high accuracy (83.89% agreement, Kappa = 0.72) but also quantifies the relative importance of spatial variables influencing land transitions. The findings reveal that topographic features are the primary drivers of LULC change, whereas proximity to roads has a minimal effect. The integration of SHAP values and Pearson correlation enhances model interpretability by highlighting key variables that shape land dynamics. These insights offer valuable guidance for sustainable urban planning and land conservation policies in vulnerable regions, aligning with global environmental goals such as the United Nations Sustainable Development Goals (SDGs).</p> Graphical Abstract <p>This study models land use and land cover (LULC) changes in Northwest Syria using a hybrid MLP-ANN approach combined with RF, CART, and SHAP for prediction and interpretability. Results show topography as the main driver of change, with flatter areas near infrastructure most at risk of urban expansion. Findings support sustainable land-use planning aligned with SDG goals.</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

Analysis and Prediction of Land Use Land Cover Dynamics for Northwest Syria Using Artificial Neural Network Model

  • 期刊论文

  • 发表日期：

    2025年6月23日

查看全文
  • Waseem Ahmad Ismaeel,
  • J. Satish Kumar

摘要

Understanding land use and land cover (LULC) changes is essential for addressing challenges related to urban expansion, environmental degradation, and resource management. Despite the widespread application of LULC prediction models, many lack comprehensive evaluation frameworks that incorporate sensitivity analyses to assess the influence of driving factors. This study aims to analyze changes in land use in northwest Syria, which has undergone profound LULC transformations. These changes are influenced by socio-economic developments and environmental degradation, making this area a critical case for sustainable land management research. An integrated modeling approach was developed by combining a Multi-layer Perceptron-Artificial Neural Network (MLP-ANN) with Random Forest (RF), Classification and Regression Tree (CART), and Shapley Additive Explanations (SHAP) to perform sensitivity analysis. This hybrid methodology not only predicts future LULC patterns with high accuracy (83.89% agreement, Kappa = 0.72) but also quantifies the relative importance of spatial variables influencing land transitions. The findings reveal that topographic features are the primary drivers of LULC change, whereas proximity to roads has a minimal effect. The integration of SHAP values and Pearson correlation enhances model interpretability by highlighting key variables that shape land dynamics. These insights offer valuable guidance for sustainable urban planning and land conservation policies in vulnerable regions, aligning with global environmental goals such as the United Nations Sustainable Development Goals (SDGs).

Graphical Abstract

This study models land use and land cover (LULC) changes in Northwest Syria using a hybrid MLP-ANN approach combined with RF, CART, and SHAP for prediction and interpretability. Results show topography as the main driver of change, with flatter areas near infrastructure most at risk of urban expansion. Findings support sustainable land-use planning aligned with SDG goals.