About Dynamic World

The real world is as dynamic as the people and natural processes that shape it. Dynamic World is a near realtime 10m resolution global land use land cover dataset, produced using deep learning, freely available and openly licensed. It is the result of a partnership between Google and the World Resources Institute, to produce a dynamic dataset of the physical material on the surface of the Earth. Dynamic World is intended to be used as a data product for users to add custom rules with which to assign final class values, producing derivative land cover maps.

Key innovations of Dynamic World

  1. Near realtime data. Over 5000 Dynamic World image are produced every day, whereas traditional approaches to building land cover data can take months or years to produce. As a result of leveraging a novel deep learning approach, based on Sentinel-2 Top of Atmosphere, Dynamic World offers global land cover updating every 2-5 days depending on location.
  2. Per-pixel probabilities across 9 land cover classes. A major benefit of an AI-powered approach is the model looks at an incoming Sentinel-2 satellite image and, for every pixel in the image, estimates the degree of tree cover, how built up a particular area is, or snow coverage if there’s been a recent snowstorm, for example.
  3. Ten meter resolution. As a result of the European Commission’s Copernicus Programme making European Space Agency Sentinel data freely and openly available, products like Dynamic World are able to offer 10m resolution land cover data. This is important because quantifying data in higher resolution produces more accurate results for what’s really on the surface of the Earth.

Dynamic World is produced using the Google Earth Engine and AI Platform. We developed the Dynamic World dataset in alignment with Google's AI Principles.

Explore Dynamic World in Resource Watch by the World Resources Institute, and learn more about WRI’s Land & Carbon Lab.

Read the Nature Scientific Data publication: Dynamic World, Near real-time global 10m land use land cover mapping.

GoogleWorld Resources Institute

Data Access and Availability

View Dynamic World in the Google Earth Engine public data catalog. Developers can access Dynamic World data by signing up for Google Earth Engine.

The Dynamic World dataset has been made available as an Earth Engine Image Collection under "GOOGLE/DYNAMICWORLD/V1". This is referenced in either the Earth Engine Python or JavaScript client library with: ee.ImageCollection('GOOGLE/DYNAMICWORLD/V1').

For deeper exploration of the dataset visit Explore Dynamic World Time Series in EE Apps.

The Dynamic World model has been run for all historic Sentinel-2 TOA imagery, and is being run for newly acquired Sentinel-2 TOA imagery; users are therefore encouraged to work with outputs available in the near realtime Image Collection available on Earth Engine.

Learn more about how to use Dynamic World:
Introduction to Dynamic World (Part 1): Visualization and Creating Composites
Introduction to Dynamic World (Part 2): Calculating Statistics of a Region
Introduction to Dynamic World (Part 3): Exploring Time Series

Dynamic World Data employed principles of open data to advance scientific understanding of planetary processes. When using Dynamic World data in your research, please include the attribution for the data products.

Training Data

Dynamic World was developed using National Geographic Society’s Dynamic World training data. This data is available under a Creative Commons BY-4.0 license and requires the following attribution: This dataset is produced for the Dynamic World Project by National Geographic Society in partnership with Google and the World Resources Institute.

Dynamic World data

This data is available under a Creative Commons BY-4.0 license and requires the following attribution: This dataset is produced for the Dynamic World Project by Google in partnership with National Geographic Society and the World Resources Institute.

We have made the trained model, example code for running inference, and additional information on the model architecture available at: