Wetland dynamics in the Indus River Delta: A Sentinel-2 and machine learning approach.

Journal: Journal of environmental management
Published Date: Aug 8, 2025

Abstract

Coastal wetlands of the Indus River Delta are vital ecological regions that have undergone significant transformations driven by anthropogenic activities and environmental stressors. This study assesses the dynamics of wetlands and reclamation in the deltaic region from 2018 to 2023, utilizing high-resolution Sentinel-2 imagery on the Google Earth Engine (GEE) platform. Random Forest (RF) machine learning algorithm with supervised technique was used to map land use land cover (LULC) classification. Their transformation was computed through change detection and land cover transfer matrix. A combined approach of normalized difference vegetation index (NDVI) and modified normalized difference water index (mNDWI) was applied for wetlands and vegetation extraction. An area-weighted centroid was used to determine the migration trend of wetlands. The results showed a substantial decline in waterbodies by -108.7 km, with the obvious loss occurring from 2018 to 2019. Built-up area expanded by 207.7 km, while cultivated land grew by 240.2 km, reflecting ongoing urbanization and agricultural intensification. Bareland and vegetative wetlands exhibited a decrease of -219.7 km and -119.5 km. This land use transformation marked by wetland loss is an indication of ecological pressure and potential consequences for biodiversity and ecosystem services. However, waterbody wetland recovery was detected by 8.47 km from 2022 to 2023, demonstrating the efficacy of restoration efforts. The findings of the study are critical for coastal ecosystem conservation and measures in the Indus Delta and similar coastal regions around the world.

Authors

  • Yaseen Laghari
    Key Laboratory of Remote Sensing and Digital Earth, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing, 100094, China; University of Chinese Academy of Sciences, Beijing, 100049, China.
  • Zhenguo Niu
    Key Laboratory of Remote Sensing and Digital Earth, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing, 100094, China; University of Chinese Academy of Sciences, Beijing, 100049, China. Electronic address: Niuzg@radi.ac.cn.
  • Shah Jahan Leghari
    College of Mechanical and Electronical Engineering, Northwest A&F University, Yangling, Shaanxi, 712100, China.
  • Muhammad Asgher Ali
    Key Laboratory of Remote Sensing and Digital Earth, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing, 100094, China; University of Chinese Academy of Sciences, Beijing, 100049, China.
  • Qingyu Li
    School of Mathematics, Physics and Data Science, Chongqing University of Science and Technology, Chongqing 401331, China.
  • Junkai Shi
    Key Laboratory of Remote Sensing and Digital Earth, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing, 100094, China; University of Chinese Academy of Sciences, Beijing, 100049, China.

Keywords

No keywords available for this article.

External Resources

Popular Topics
Recent Journals