How partner organizations’ data reach the Observatory and the world

Interoperability makes it possible for datasets produced by different organizations, with different methods, scales, and tools, to be integrated, interpreted, and reused together. In the Amazon Waters Alliance it is addressed across four dimensions: legal (licenses, terms of use, and agreements), organizational (roles and responsibilities), semantic (shared standard, controlled vocabularies, and metadata), and technical (open formats, persistent identifiers, and programmatic access). The model does not centralize the information or homogenize field practices: each organization retains ownership and control over its data, and its contribution is traced throughout the cycle.

Information cycle: from the partner organization to GBIF

Every record published in the Observatory goes through six stages:

1. Share. The partner organization submits the set structured in the Darwin Core template, with its metadata and the declaration of license, restrictions, and sensitive information.
2. Steward. The Technical Secretariat checks the formal validity of the contribution and keeps the original unchanged, with restricted access (level 0).
3. Integrate. The contribution is incorporated into the cumulative internal set, retaining the contribution identifier, the source organization, and the date of receipt.
4. Validate. Profiling, normalization, taxonomic, geographic, and temporal validation, and duplicate detection are applied; each transformation is documented.
5. Protect. Personal data are anonymized and location is generalized to the sub-basin level, producing controlled-access versions (levels 1 and 2).
6. Publish. With authorization from the Technical Secretariat, the open-access version (level 3) is generated.

Alliance partner organizations share their data with the ICTIO system through the Technical Secretariat. The process has four steps:

1. Data preparation. The organization fills in the biological records template, which follows the Darwin Core standard, and the metadata template, which describes who generated the data, how, where, when, and under what terms of use.
2. Submission through the single channel. Both files are attached via the Observatory upload mechanism. Each submission corresponds to a single dataset with a single metadata and terms declaration; distinct campaigns or methodologies are submitted separately.
3. Receipt confirmation. The Technical Secretariat acknowledges receipt and checks that the documentation is complete and the structure is valid.
4. Verification outcome. The contribution may be accepted, accepted with observations, or returned with concrete instructions for correction. Returns are framed as support, not as rejection.

Each organization retains ownership of the data it shares and may request a copy of what was contributed at any time.