SpatioTemporal Asset Catalogs (STAC)

Summary

The STAC specification is a common language to describe geospatial information, so it can more easily be leveraged, indexed, and discovered. It is a standardized way to expose collections of spatiotemporal data and provides a common structure for describing and cataloging spatiotemporal assets. With wide adoption of this common standard, different catalogs can be compared and aggregated, tooling and libraries can be applied in the same manner against multiple catalogs and user interfaces can be reused.

Status

STAC is an approved NASA Earth Science Data Systems convention.

NASA Earth Science Community Recommendations for Use

Strengths

SpatioTemporal Asset Catalog (STAC) allows granule queries in the spatial and temporal domain with results in STAC format.  STAC has become, in effect, the standard for cloud-based catalog representation has already been adopted by NASA Earthdata search.  The STAC spec itself provides a lowest common denominator JSON format to wrap around any relevant data about the earth. The core GeoJSON object and related structures are designed for extension, so it can adapt to different domains.

Weaknesses

While STAC provides an interoperable way of discovering data, it does not force the data to be stored in any specific format(s).  This can be a positive from the perspective of a data producer as well as data discovery as STAC can search for the data in a cloud-friendly catalog.  However, this is limited to only data discovery and doesn’t address other file transformations such as format or reduction.  

Applicability

STAC is particularly well suited to data with an obvious time and location such as satellite imagery.  The stated goal of STAC is to enable a global index of all imagery (satellite, aerial, drone, etc.), derived data products and alternative geospatial captures (Lidar, SAR, Full Motion Video, Hyperspectral and beyond).  STAC is closely integrated with Cloud Optimized GeoTIFF (COG).

Limitations

STAC usage is currently limited by available low level client libraries and tools, this impacts a user’s ability to build custom search clients tailored to their specific needs.  As more data catalogs support searching via STAC, this issue should be resolved.