Scholarlink

<p>International trade hubs are vital for post-conflict economic recovery, particularly in landlocked and developing nations. This study examines how changes in land use and land cover (LULC) reflect the progress of reconstruction in Afghanistan’s three major trading hubs, Hairatan, Sherkhan Bandar, and Torghundi, between 2016 and 2023, along the international borders. Harmonized Sentinel-2 imagery (10&#xa0;m resolution) is classified into five LULC classes: built-up, vegetation, swamp, barren land, and water using Artificial Neural Network (ANN) and Random Forest (RF) classifiers on Google Earth Engine and Google Colab to map and assess spatial changes. The RF classifier achieved higher accuracies (89–99%) than ANN across all towns. Built-up areas increased in all sites, with Hairatan expanding from 2.23&#xa0;km² (15%) in 2016 to 3.22&#xa0;km² (22%) in 2023. Torghundi showed a similar rise from 3.78&#xa0;km² to 5.38&#xa0;km², accompanied by a decline in vegetation from 2.54&#xa0;km² to 1.10&#xa0;km². Sherkhan Bandar recorded moderate built-up growth alongside seasonal increases in vegetation. A predictive RF model projects a further 53% built-up expansion along the Torghundi railway corridor by 2030. These findings demonstrate that LULC dynamics can serve as measurable indicators of post-conflict recovery while also revealing environmental trade-offs. The study highlights the role of remote sensing for monitoring reconstruction and emphasizes the need to integrate infrastructure development with sustainable land management.</p> Graphical Abstract <p></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

The impact of land use dynamics at the trading ports on post-conflict economic development in Afghanistan

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年6月24日

查看全文
  • Vineet Chaturvedi,
  • Walter de Vries

摘要

International trade hubs are vital for post-conflict economic recovery, particularly in landlocked and developing nations. This study examines how changes in land use and land cover (LULC) reflect the progress of reconstruction in Afghanistan’s three major trading hubs, Hairatan, Sherkhan Bandar, and Torghundi, between 2016 and 2023, along the international borders. Harmonized Sentinel-2 imagery (10 m resolution) is classified into five LULC classes: built-up, vegetation, swamp, barren land, and water using Artificial Neural Network (ANN) and Random Forest (RF) classifiers on Google Earth Engine and Google Colab to map and assess spatial changes. The RF classifier achieved higher accuracies (89–99%) than ANN across all towns. Built-up areas increased in all sites, with Hairatan expanding from 2.23 km² (15%) in 2016 to 3.22 km² (22%) in 2023. Torghundi showed a similar rise from 3.78 km² to 5.38 km², accompanied by a decline in vegetation from 2.54 km² to 1.10 km². Sherkhan Bandar recorded moderate built-up growth alongside seasonal increases in vegetation. A predictive RF model projects a further 53% built-up expansion along the Torghundi railway corridor by 2030. These findings demonstrate that LULC dynamics can serve as measurable indicators of post-conflict recovery while also revealing environmental trade-offs. The study highlights the role of remote sensing for monitoring reconstruction and emphasizes the need to integrate infrastructure development with sustainable land management.

Graphical Abstract