Get Started

Overview

Leverage our platform for impact science and academia

Scientific integrity, excellence, and innovation are part of our core values at Descartes Labs. We believe that advancing the understanding of human activity and natural processes throughout the world is in the interest of all stakeholders. As a result, we are willing to put our platform and resources behind efforts to demonstrate this.

For the first time, we are granting researchers access to petabytes of data from numerous satellite constellations in a consistent and standardized format through a simple API—pre-processed into either top or bottom-of-atmosphere reflectance.

Selection criteria

We look for impact science and academic partners. Specific considerations include the proposed project and the composition of the technical team. Partners should be willing and able to:

  • Provide feedback based on their experience with the platform, helping Descartes Labs improve the platform for future research teams
  • Share certain details and outcomes of the research in Descartes Labs’ marketing efforts, including co-branded social media posts and conversations with the media

Example use cases

Our simple but powerful programmatic interface enables a rich range of capabilities and use cases. Here are a few:

  • Create a single composite image for an arbitrary geographic area from your chosen date range of imagery
  • Retrieve a time stack of imagery and analyze how your area of interest changes over time
  • Run machine learning classifiers to segment your data (e.g. to create land cover classifications)
  • Run computer vision models to identify objects in imagery such as wastewater treatment plants or solar panels
  • Display before/after imagery of deforestation due to illegal harvesting or mining operations
  • Construct a time series of statistics from imagery such as NDVI for Iowa or total area of rice paddy fields in Vietnam

Functionality

Our platform exposes over 13 petabytes of imagery alongside other remotely sense and geographic data by means of python APIs that enable quick access to data that has been processed to be analysis-ready. We’ve developed a cross-sensor atmospheric compensation algorithm as well as a cross-sensor cloud mask and are continuing to invest in improving the experience of working with our data. We’ve also developed a scalable compute capability that lets users easily run analytics on multiple nodes. Our web GUIs allow real-time viewing and analysis of any of our geospatial data, including any analyses that you create on our platform.

All partners will receive full access to:

  • Airbus OneAtlas
  • Landsat 4/5/7/8
  • Sentinel 1/2/3/5P
  • MODIS
  • Shuttle Radar Topography Mission
  • NAIP
  • Texas Orthoimagery Program
  • Cropland Data Layer

Additional products may be added upon request.