Large Earth Observation (EO) data archives are nowadays available on platforms that offer computing and storage resources (e.g., Amazon Web Services , Google Cloud , DIAS platforms ). The exploitation of EO data with a user-oriented focus can be realized by different interfaces (APIs) to ease processing and analysis tasks, such as xarray, Open Data Cube, OpenEO, or an implementation of the OGC (Open Geospatial Consortium) Application Deployment and Execution Service (ADES). Experiments on a Data Access and Processing API conducted as part of the OGC innovation program have shown benefits in simplifying data access and data analysis .
For analyzing EO data, user-oriented interfaces need to know what kind of data is available on the platform (e.g., different satellites and sensors). Additionally, the data need to be filtered by different parameters (e.g., spatial and temporal dimensions, cloud coverage). Thus, a metadata database containing the information needs to be queried either by the user or the API itself. EO data available on a platform are often registered in a metadata catalogue in order to be searchable by the users (e.g., CreoDIAS Finder application). However, specifications for such an EO metadata catalogue (e.g., OGC Catalogue Service for Web, OGC OpenSearch Extension for EO) previously focused mainly on data discovery rather than direct data access, which is needed for EO exploitation platforms. In most cases, such platform-dependent metadata catalogues were not connected to data analysis and visualization tools.
The SpatioTemporal Asset Catalog (STAC) specification  ushered in a new era, by not only making EO data discoverable, but accessible for data analysis as well as data visualization. The software ecosystem around STAC contains open source software for data discovery, data visualization, data catalogs, metadata creation, as well as integrations into already existing data analysis tools. As an example, the Open Data Cube software no longer needs its own database containing the available EO data. Instead, it can directly connect to a STAC API for data search and filtering. This allows a user to create an instance of the Open Data Cube with the results of a STAC API request on the fly and to use it for further analysis. Similar integrations exist for xarray (stackstac) and GRASS GIS (actinia-stac-plugin).
We present a STAC-oriented architecture for EO exploitation platforms, which includes technical solutions for user-oriented discovery, access, visualization, and analysis of EO data. This architecture is the basis for DLR’s EO Exploitation Platform terrabyte, which comprises 40 Petabyte online storage together with a large amount of computing resources. STAC metadata has been created initially for Sentinel-1 and Sentinel-2 data, which has been registered into a STAC API. Users are able to connect with Open Data Cube, stackstac or the GRASS GIS-based OpenEO interface to the STAC API for data analysis. The software titiler is used as a STAC-based visualization service.
From an EO data platform provider’s perspective this concept allows to focus on a performant STAC API service without needing to synchronize the metadata holdings for different additional services. In addition to the STAC API, services for data visualization and data access (e.g., sub-setting, time-series extraction) enable further on the fly exploitation of the provided geospatial data. In addition, users of the platform can create STAC metadata along with their processing and analysis results, which can then be used in all of the STAC-based platform services for on the fly data discovery, visualization, access and follow-up data analysis. Results of a comparison to traditional approaches and future capabilities and activities are shown and presented from a data provider’s perspective.
 Amazon Web Services, “Registry of Open Data on AWS,” https://registry.opendata.aws, accessed on 2021-02-26.
 Google, “Google Cloud Public Datasets,” https://cloud.google.com/public-datasets, accessed on 2021-02-26.
 Copernicus, “Data and Information Access Service,” https://www.copernicus.eu/en/access-data/dias, accessed on 2021-02-26.
 Vretanos, Panagiotis (2021), “OGC Testbed-16: Data Access and Processing Engineering Report,” Open Geospatial Consortium – OGC-20-016, https://docs.ogc.org/per/20-016.html
 Radiant Earth, “SpatioTemporal Asset Catalogs,” https://stacspec.org, accessed on 2021-02-26.