Descartes Labs added imagery from Airbus OneAtlas catalog to the platform it is developing for global-scale predictive analytics. The Airbus catalog offers 50-centimeters-per-pixel imagery for the largest 2,600 cities. This is an Airbus image of Toulouse, France. Credit: Descartes Labs

PHOENIX – Descartes Labs, a startup focused on applying machine learning to geospatial datasets, announced Aug. 15 a partnership with Airbus to obtain high resolution global imagery. At the same time, the Santa Fe, New Mexico-based startup announced it completed beta testing of the Descartes Labs Platform, a cloud-based “data refinery,” and added weather data.

Descartes Labs began beta testing 15 months ago with a library that included Earth imagery from NASA’s Landsat and the European Space Agency’s Sentinel constellations. Adding the entire Airbus OneAtlas global Basemap catalog allows “customers to, for the first time, build models and do their own machine-learning modeling directly on Airbus’ OneAtlas data at-scale,” Mark Johnson, Descartes Labs chief executive and co-founder, said by email. “We have ingested all of the Airbus OneAtlas data onto our platform so customers don’t need to know or test specific areas. They can look for signals across the entire planet.”

The Airbus data has a resolution of 1.5 meters per pixel globally and 50 centimeters per pixel for the largest 2,600 cities.

Descartes Labs added weather data to its geospatial data platform because that information is “critical in understanding and tracking how weather conditions impact plant health, the shipping and transportation of goods across the planet and energy pipeline health,” Johnson added.

For geospatial analysts, the holy grail is not simply detecting change that has occurred but creating models that can predict change or future events. That requires massive amounts of data and massive computing capabilities.

Descartes Labs has processed more than 11 petabytes of compressed data into its platform. The company processes nine additional terabytes daily, Johnson said.

“Our data refinery pulls in raw data, cleans it up, fuses data from disparate sources and adds tools on top of it for easier analysis so when customers go in, they can instantly experiment and build models on our platform,” he added. “That is what differentiates us from pure analytics and satellite companies.”

Debra Werner is a correspondent for SpaceNews based in San Francisco. Debra earned a bachelor’s degree in communications from the University of California, Berkeley, and a master’s degree in Journalism from Northwestern University. She...