Cloud Native Geospatial Ecosystem Community Releases STAC Specification version 1.0.0 to Connect Remote Sensing Data into a Network of Information about the Earth

The SpatioTemporal Asset Catalog (STAC) specification provides a common language to describe a range of geospatial information, so it can more easily be indexed and discovered

Washington, District of Columbia, UNITED STATES

WASHINGTON, June 10, 2021 (GLOBE NEWSWIRE) -- The Cloud Native Geospatial Ecosystem Community announces the release of the SpatioTemporal Asset Catalog (STAC) specification version 1.0.0. The STAC specification is an open metadata standard that systemically describes remotely sensed data of the Earth. The specification supports emerging cloud-based geoprocessing engines by allowing spatial data to be indexed and discovered more efficiently. This feature is fundamental when building artificial intelligence applications using Earth imagery.

The Cloud Native Geospatial Ecosystem Community comprises Earth observation data providers from the private and public sectors, developers working with geospatial data, and consumers of spatial datasets. These organizations and individuals from all industries worldwide came together throughout the last three and a half years to increase the interoperability of searching for and discovering satellite images of the Earth.

STAC provides a uniform index that describes a range of geospatial information, allowing users to discover spatial data across time, space, and cloud storage locations. This global indexing of Earth imagery and derived products works similar to how search engines use specific HTML tags to index web pages. For data providers, it makes it easier to expose and share Earth imagery across different systems. Developers can manage spatial data without having to write additional code every time a new dataset is released. And data users need to build their pipelines for ingesting spatial data into their system only once.

STAC is entirely community-driven, from the initial draft specification in 2017 to its adoption today as the standard for cataloging geospatial data. Dozens of major geospatial organizations and individuals have adopted STAC or built tools to support version 1.0.0. CBERS, one of the early adopters, and NASA Avris data were the first to update their catalog to 1.0.0. Google Earth Engine's STAC catalog was also updated to the 1.0.0 specification. Several tools and utilities, including STAC Browser, a client to create an interactive website for STAC catalogs, have also been released with support for STAC 1.0.0.

Radiant Earth Foundation, a nonprofit focused on making Earth observations and machine learning (ML) insights more accessible for global nonprofits, humanitarian organizations, and emerging economy governments, has been the driving entity behind STAC. They sponsored the first coding sprint in 2017 that kickstarted the specification and funded a Technical Fellowship program since then to help advance the specification. As a neutral entity, Radiant Earth also managed sponsorships from an extensive range of organizations – such as Microsoft, Planet, Maxar, Azavea, Element 84, Sparkgeo, Arturo, and many others – that supported STAC.

"I would like to thank all the myriad of sponsors and developers that helped to nurture this specification to maturity," said Dr. Hamed Alemohammad, Executive Director and Chief Data Scientist at Radiant Earth. "This is a game-changer for the Earth observation community that will lead to more efficient workflows, especially as it relates to building and maintaining geospatial data pipelines for artificial intelligence applications. This specification embodies what Radiant Earth stands for: open, innovative, and collaborative. We are proud to be part of this historic milestone."

Chris Holmes, a Radiant Earth Foundation Technology Fellow and a Fellow and VP of Product and Strategy at Planet, led the development and overall coordination of this community effort. In a recent blog post, Holmes highlighted the journey to STAC 1.0.0, listing the key organizations, individuals, and tools that led to this stable specification. "It was important from the outset that we provide a foundational layer for the geospatial architectures that were being built for the cloud," said Holmes. "Geospatial data processing in the cloud is the future. And we had an opportunity to get the metadata right for it. Now, data providers have a simple, common language to describe their holdings."

In the next couple of months, Holmes will write a series of articles that will delve deeper into the specification. Follow Radiant Earth Insights to receive notifications of these updates.

Radiant Earth Foundation is a nonprofit corporation working to empower organizations and individuals with open ML and Earth observation data, standards, and tools to address the world’s most critical international development challenges. Radiant Earth fosters collaboration through a cloud-based open geospatial training data library, Radiant MLHub. Radiant also supports an ecosystem of practitioners to develop standards, expand interoperability around ML on Earth observation, and provide information and training to help advance the capacity of those working in the global sector using ML and Earth observation. Visit us on Twitter, LinkedIn, Medium, and GitHub.

Louisa Nakanuku-Diggs
202 – 596 – 3603