2025/08/28 by Sajana Maharjan, Kabir Uddin, Birendra Bajracharya +4 · 1 voice
Environmental Science · Social Sciences · #Rangeland Management and Livestock Ecology #Remote Sensing in Agriculture #Transboundary Water Resource Management
paper · doi:10.1016/j.envc.2025.101289
openalex publication_date 2025/08/28 · openalex created_date 2025/10/10 · openalex updated_date 2026/06/16
Desertification, conflict-driven degradation, and climate change increasingly threaten Afghanistan's landscapes, shaped by both natural processes and long-standing human-environment interactions. There is an urgent need to analyze land cover dynamics and methodological insights in the Hindu Kush Himalaya (HKH) region, particularly in Afghanistan, to guide landscape restoration and regeneration efforts. Addressing this gap, this study produces the first consistent, harmonized annual land cover dataset for Afghanistan from 2000 to 2018, using Google Earth Engine (GEE), the Random Forest algorithm, remote sensing techniques, and 30-meter resolution satellite images. Despite historical data constraints, the cloud-based approach enabled comprehensive national-scale mapping. In 2018, rangeland was the dominant land cover type (45.66%), followed by barren land (31.03%) and sand (7.71%). Over the 19-year period, Rangeland expanded by 1.08%, with notable expansions in built-up areas and sand-covered zones. Spatial patterns and fragmentation were assessed using five landscape metrics: greatest patch area, number of patches, overall core area, splitting index and, largest patch index. These analyses identified critical trends in urban expansion and rangeland fragmentation. The resulting annual land cover database and landscape metrics offer a robust evidence base to inform rural landscape planning, zoning, and restoration initiatives aligned with national and global sustainability goals.