Wu et al. (2026) Efficient large-scale land cover change detection using Google Earth Engine: Climate-driven vegetation dynamics in Asian drylands (2001–2022)

Identification

• Journal: PLoS ONE
• Year: 2026
• Date: 2026-04-01
• Authors: Jianfeng Wu, Shengtao Wei, Haichao Hao, Meng Chen, Sadaf Ismail
• DOI: 10.1371/journal.pone.0344835

Research Groups

• Guizhou Provincial Key Laboratory of Geographic State Monitoring of Watershed, School of Geography and Resources, Guizhou Education University, Guiyang, China
• Key Laboratory of Geographic Information Science (Ministry of Education), School of Geographic Sciences, East China Normal University, Shanghai, China
• School of Life Science, Shanxi University, Taiyuan, China
• Graduate School of Advanced Integrated Studies in Human Survivability, Kyoto University, Kyoto, Japan

Short Summary

This study analyzed land cover dynamics and climate-driven vegetation changes in Asian drylands from 2001 to 2022 using MODIS, TerraClimate, and Google Earth Engine. It found significant land cover changes, including grassland and cropland expansion, primarily influenced by increasing temperatures, soil moisture, and vapor pressure, coupled with decreasing precipitation and drought indices.

Objective

• To systematically analyze land cover dynamics, vegetation type transitions, and their climatic drivers across Asian drylands from 2001 to 2022.
• To quantify spatiotemporal patterns of major land cover types, identify transition pathways, evaluate vegetation dynamics and stability, and examine the relationships between climate change and different vegetation change zones.

Study Configuration

• Spatial Scale: Asian drylands (46°-127°E, 31°-56°N), covering over 10 million square kilometers. Data harmonized to a 4 km (1/24°) grid.
• Temporal Scale: 2001–2022 for land cover dynamics and vegetation responses. Climate trend analysis for 2001–2020. Future climate projections for 2015–2100.

Methodology and Data

• Models used: Google Earth Engine (GEE) platform, ArcGIS. Methodologies include Land Cover Dynamic Indices (Temporal Dynamic Index - TDI, Spatial Dynamic Index - SDI), Transition Probability, Transfer Matrix Analysis, Grid-based Statistical Approach, and Vapor Saturated Pressure (VSP) calculation.
• Data sources:
  • MODIS MCD12Q1 land cover product (annual, 500 m spatial resolution, IGBP scheme).
  • TerraClimate climate reanalysis data (monthly, approximately 4 km (1/24°) spatial resolution). Variables: maximum temperature (TMMX), minimum temperature (TMMN), precipitation (PRE), climatic water deficit (DEF), Palmer Drought Severity Index (PDSI), soil moisture (SOIL), surface shortwave radiation (SRAD), potential evapotranspiration (PET), actual evapotranspiration (AET), vapor saturated pressure (VSP), actual vapor pressure (VAP), and vapor pressure deficit (VPD).
  • SRTM Digital Elevation Model (DEM) data (30 m spatial resolution).
  • CMIP6 climate projection datasets (SSP2–4.5 and SSP5–8.5 scenarios, 0.25° spatial resolution).

Main Results

• Pronounced land cover changes occurred across Asian drylands from 2001 to 2022, characterized by expansions of grasslands (GRA), savannas (SAV), croplands (CRO), and water, snow, and ice (WSI).
• Contractions were observed in shrublands (SH), mixed forests (MF), permanent wetlands (WET), and barren land (BAR).
• The most prominent land cover transition pathways were from barren land to grasslands (BAR to GRA) and from grasslands to croplands (GRA to CRO).
• Vegetation dynamics across different stability zones exhibited distinct responses to long-term climate trends.
• Increasing maximum temperature (TMMX), soil moisture (SOIL), and vapor-related variables (VSP, VAP) jointly shaped vegetation persistence, expansion, or degradation.
• Declining precipitation (PRE), drought indices (PDSI, DEF), and surface radiation (SRAD) also played differentiated roles in vegetation dynamics.
• Future climate projections (CMIP6) indicate significant warming (up to 0.0312 °C/year under SSP5-8.5) and increased precipitation (up to 0.5175 mm/year under SSP5-8.5), with complex regional impacts on vegetation, potentially intensifying evaporation and limiting vegetation expansion despite increased rainfall.

Contributions

• Provides a scalable and reproducible framework for monitoring land cover change and vegetation stability in arid and semi-arid regions using cloud-based geospatial computing and multi-source long-term datasets.
• Enhances the understanding of dryland ecosystem dynamics under climate change by systematically linking land cover transitions, vegetation stability indicators, and climate drivers at a continental scale.
• Offers methodological support and scientific evidence for large-scale ecological assessment and adaptive land management in data-scarce environments.

Funding

• Guizhou Provincial Key Technology R&D Program (No. [2023]226)
• National Natural Science Foundation of China (U24A20579)
• Engineering Research Center of Colleges and Universities of Guizhou Province (No. [2023]039)
• Guizhou Provincial Department of Education “Revealing the List and Appointing the Leader” Project (Document No. [2025]019)

Citation

@article{Wu2026Efficient,
  author = {Wu, Jianfeng and Wei, Shengtao and Hao, Haichao and Chen, Meng and Ismail, Sadaf},
  title = {Efficient large-scale land cover change detection using Google Earth Engine: Climate-driven vegetation dynamics in Asian drylands (2001–2022)},
  journal = {PLoS ONE},
  year = {2026},
  doi = {10.1371/journal.pone.0344835},
  url = {https://doi.org/10.1371/journal.pone.0344835}
}