Khare et al. (2025) Estimates of global surface water dynamics harnessing near real-time land cover observations and open science geospatial capabilities

⚠️ Warning: This summary was generated from the abstract only, as the full text was not available.

Identification

• Journal: Environmental Research Letters
• Year: 2025
• Date: 2025-10-15
• Authors: Arushi Khare, Bikas C Gupta, Adnan Rajib, Melanie K. Vanderhoof, Qiusheng Wu
• DOI: 10.1088/1748-9326/ae137b

Research Groups

Not explicitly mentioned in the abstract.

Short Summary

This study developed the first 10 m resolution Global Surface Water Extents (GSWE) dataset for 2015–2023 using Sentinel-2 Dynamic World products, estimating 2.5 million km² of permanent and 8 million km² of seasonal waters globally, and providing an operational framework for actionable water information.

Objective

• To produce the first 10 m resolution Global Surface Water Extents (GSWE) dataset for the period 2015–2023 using Sentinel-2 based near real-time Dynamic World land cover products.

Study Configuration

• Spatial Scale: Global, with 10 m resolution. Specific regions of interest include Pantanal (South America), Haor (South Asia), Spain, Lake Urmia (Central Asia), and the Ganges–Brahmaputra confluence.
• Temporal Scale: 2015–2023, with a focus on time-varying, near real-time, and frequent monitoring capabilities.

Methodology and Data

• Models used: Dynamic World (DW) land cover products (a machine learning-based land cover classification system).
• Data sources: Remotely sensed Sentinel-2 satellite imagery. Comparisons and validation were performed against contemporary Landsat-based GSWE datasets, well-established observational products, and widely used GSWE datasets.

Main Results

• The dataset estimated 2.5 million km² of permanent waters and 8 million km² of seasonal waters worldwide.
• Compared to contemporary Landsat-based GSWE, the Sentinel-2 based data mapped less water within the >50% probability of occurrence range, suggesting a lower presence of open permanent water, especially in high latitudes.
• Statistical analysis against established observational products and widely used GSWE datasets confirmed the overall physical realism of the data in predicting global open surface water dynamics.
• The study demonstrated an operational capability through instant mapping of floods in Spain, drought in Lake Urmia, and frequent monitoring of river extent changes at the Ganges–Brahmaputra confluence.

Contributions

• Production of the first-of-its-kind 10 m resolution Global Surface Water Extents (GSWE) dataset for 2015–2023.
• Development of a prototype Open Science operational framework that extracts routinely available Dynamic World products, performs geospatial analytics, and generates actionable water information for various stakeholders.
• Demonstrated interoperability with other existing GSWE applications.

Funding

Not explicitly mentioned in the abstract.

Citation

@article{Khare2025Estimates,
  author = {Khare, Arushi and Gupta, Bikas C and Rajib, Adnan and Vanderhoof, Melanie K. and Wu, Qiusheng},
  title = {Estimates of global surface water dynamics harnessing near real-time land cover observations and open science geospatial capabilities},
  journal = {Environmental Research Letters},
  year = {2025},
  doi = {10.1088/1748-9326/ae137b},
  url = {https://doi.org/10.1088/1748-9326/ae137b}
}