Satlas

Satlas, developed by the Allen Institute for AI (AI2), is a pioneering platform that provides open and AI-generated geospatial data to the public. Its core mission is to leverage state-of-the-art artificial intelligence to analyze vast amounts of satellite imagery, making it possible to monitor our planet's changes with unprecedented detail and frequency. By processing public domain imagery from the European Space Agency’s Sentinel-2 satellites, Satlas generates high-accuracy, global datasets that are updated monthly, revealing trends in deforestation, renewable energy expansion, and marine infrastructure development.

The platform is built upon a foundation of powerful computer vision models trained on SatlasPretrain, a massive proprietary dataset with over 30 TB of imagery and 300 million labels. This allows Satlas to identify and classify features like wind turbines, solar farms, and tree cover with remarkable precision. The entire project, including the data, AI models, and training code, is made freely available to foster research, transparency, and innovation in environmental science and beyond.

How to use Satlas

Using Satlas is straightforward, catering to both casual observers and data professionals:

1. Explore the Interactive Map: The primary way to interact with Satlas is through its global map. Users can navigate to any location on Earth to view the data.
2. Select Data Layers: On the map interface, you can toggle different data layers, including Marine Infrastructure (offshore wind turbines, platforms), Renewable Energy (solar farms, onshore wind turbines), and Tree Cover. This allows you to focus on specific areas of interest.
3. Analyze Change Over Time: Satlas features powerful time-lapse tools. You can use a time slider or select different dates in a side-by-side or split-screen view to visually compare how a region has changed between months or years.
4. Leverage Super-Resolution: For a more detailed view, you can activate the Super-Resolution layer. This AI-enhanced imagery transforms standard Sentinel-2 images into high-resolution visuals, revealing details like individual structures or small-scale land changes that are otherwise invisible.
5. Download Data for Offline Analysis: For researchers, journalists, and analysts, all AI-generated geospatial data is available for download in standard GeoJSON or GeoTIFF formats from the 'Data' section of the website.
6. Access AI Models: For advanced users, the underlying AI models and source code are available on GitHub, allowing for custom analysis or integration into other projects.

Core Features of Satlas

• AI-Generated Geospatial Datasets: Provides high-accuracy, global, and monthly updated data on marine infrastructure, renewable energy, and tree cover.
• Super-Resolution Imagery: An AI model (adapted from ESRGAN) that enhances 10m resolution Sentinel-2 imagery to produce high-resolution, cloud-free views, often more up-to-date than commercial mapping services.
• Interactive Time-Lapse Visualization: An intuitive map interface with single, side-by-side, and split-screen layouts to easily track and visualize environmental and infrastructural changes over time.
• Completely Open and Accessible: All data is released under the open ODC-BY license, and the AI models and code are open-sourced, removing barriers to access for researchers, NGOs, and the public.
• Global Coverage and High Frequency: Offers a worldwide perspective on critical trends with monthly updates, enabling timely analysis and response.

Use Cases for Satlas

Satlas is a versatile tool with applications across various fields:

• Environmental Monitoring: Researchers and conservation groups can track Amazon rainforest deforestation, monitor the impact of logging activities, and assess changes in global tree cover.
• Renewable Energy Tracking: Governments and energy analysts can quantify the growth of solar and wind farms globally, helping to verify renewable energy goals and plan future developments.
• Infrastructure and Urban Planning: Planners and engineers can observe the development of large-scale projects like dams, ports, and offshore platforms, using up-to-date imagery for monitoring and assessment.
• Journalism and Reporting: Journalists can use the compelling visual evidence from Satlas to create data-driven stories about climate change, global development, and environmental issues.
• Disaster Response: The high-resolution and frequently updated imagery can be invaluable for assessing building damage and landscape changes after natural disasters like floods or wildfires.

Advantages of Satlas

The primary advantage of Satlas is its commitment to democratizing access to crucial Earth observation data. By making its high-quality, AI-generated datasets and tools completely free, it empowers a global community of users who may not have access to expensive commercial data. Its use of cutting-edge AI ensures high accuracy and the unique Super-Resolution feature provides insights that are not available from other free platforms. Furthermore, its transparency through open-sourcing models and code builds trust and encourages collaborative scientific advancement.

Pricing and Plans

Satlas is a non-profit project from the Allen Institute for AI and is completely free to use. All AI-generated geospatial data, super-resolution imagery, and the underlying AI models are publicly available at no cost. The data is released under the permissive ODC-BY license, which allows for sharing, use, and adaptation for any purpose, including commercial applications, with proper attribution.