2026-11-03 –, Gardenia
We present a demo of STAC Search, a search engine for geospatial datasets. STAC Search addresses the lack of discoverability of public STAC catalogs by indexing catalogs and making them searchable in a single place.
There are currently a large number of public STAC catalogs available from many dataset providers. While STAC catalogs provide a convenient way for dataset users to browse through dataset metadata, the usefulness of STAC catalogs is limited by their lack of discoverability. Dataset users must manually traverse multiple STAC catalogs from multiple organizations to find datasets that are relevant to their use cases.
To address the discoverability problem, we introduce STAC Search, a search engine for geospatial datasets. STAC Search indexes public STAC catalogs and makes them searchable in a single place. It provides a web interface that allows searching by spatial and temporal extents, keywords and themes.
In this talk, we will demonstrate the use of STAC Search and briefly discuss its technical design. STAC Search consists of three parts: (1) a crawler implemented in Python that mirrors metadata from public STAC catalogs to a central storage system, (2) a service that ingests the crawled metadata into a geospatial database and provides a search API, and (3) a web application that provides the user-facing search interface.
At the time of proposal submission, STAC Search is still in development and has not been deployed. We plan to make STAC Search publicly available before the conference.
Yifan Mai is an independent researcher interested in geospatial tools for climate change and sustainability. They have over a decade of professional experience in software engineering, including cloud infrastructure, machine learning infrastructure, and distributed data processing.