LocalContexts – GBIF Pilot

Scope

This pilot aims to promote the integration of LocalContexts Notices and Labels within the GBIF data publishing framework. This work focuses on engaging existing GBIF publishers to publish datasets in which the underlying data is explicitly connected to Traditional Knowledge (TK), Cultural Heritage, or other Indigenous-related contexts linked to geographic scope, species identification, specimen collection, and data use conditions.

The main objective of this pilot is to demonstrate how a subset of GBIF publishers can enrich their data with LocalContexts identifiers (Notices, Labels, and Projects), and to explore how these identifiers can be obtained and visualized through the GBIF API, IPT (Integrated Publishing Toolkit), and GBIF.org data portal.

It is important to note that the publishing workflow described for this pilot should not be considered the definitive or standardized mechanism for publishing TK- or Indigenous-linked data in GBIF. To establish long-term practices, GBIF has convened an Indigenous Data Governance Task Group. This group will provide formal recommendations on embedding LocalContexts identifiers into GBIF’s publishing architecture and aligning with FAIR and CARE principles (Findable, Accessible, Interoperable, Reusable / Collective benefit, Authority to control, Responsibility, Ethics).

Pilot Participants

This pilot is a collaboration between Local Contexts and the GBIF Secretariat, with guidance for the GBIF Indigenous Data Governance task group. This pilot engages a small subset of GBIF data publishers who are either already known to be associated with LocalContexts projects or are considered likely candidates. The coordinating group will work directly with these publishers to annotate selected records and datasets with LocalContexts identifiers (Project IDs).

It is important to note that the effort of annotating data is undertaken solely for the purposes of this pilot. The annotations will serve as a proof of concept to test technical approaches, evaluate workflows, and demonstrate how LocalContexts metadata can be surfaced within the GBIF API and data portal. These practices should not be understood as a final or permanent requirement for GBIF publishers.

Pilot Tasks

This pilot will prioritize the following activities:

  1. Engagement and Capacity Building

    • Contact and support GBIF publishers in understanding and applying LocalContexts Notices and Labels in their dataset metadata.

    • Provide guidance on integrating the use of LocalContexts identifiers within the GBIF publishing workflow (e.g., Darwin Core fields, Datasets metadata).

  2. Technical Implementation

    • Define and prototype a simple mechanism for exposing LocalContexts identifiers through the GBIF API and related infrastructure.

    • Ensure alignment with metadata standards and explore potential mappings within Darwin Core and GBIF’s registry schema.

  3. Demonstration and Visualization

    • Develop a Hosted Portal to showcase datasets and occurrence records annotated with LocalContexts identifiers.

    • Position this portal as both a demonstrator of technical feasibility and a central communication hub for the pilot, including documentation, FAQs, and updates on joint GBIF–Local Contexts efforts.

Technical Implementation

Data Publishing

The proposed implementation is to use the Darwin Core term dynamicProperties as a temporary mechanism for publishing LocalContexts identifiers. This term allows publishers to include additional measurements, facts, or assertions about a record that are not explicitly covered by other DwC terms.

The recommended (and expected) practice is to encode values using a key:value schema in a data interchange format such as JSON. For the purposes of this pilot, and to maintain simplicity:

  • Only JSON-formatted values will be interpreted.

  • Only the local_contexts_project_id will initially be recognized by the GBIF data platform, though additional properties may be supported in the future. This property may point either to the LocalContexts Hub API endpoint or to the corresponding project landing page, also the project id without any url schema can be provided. In all cases, the GBIF portal will use the LocalContexts API to extract the relevant project metadata and provide a direct link to the project detail page in the LocalContexts Web Portal. Examples:

    1. Value using LocalContexts API endpoint

{
  "local_contexts_project_id": "https://sandbox.localcontextshub.org/api/v2/projects/b23048f6-b832-4379-982a-7fe39a425ae8/"
}

  1. Value using a LocalContexts Project landing page

{
  ...
  "local_contexts_project_id": "https://sandbox.localcontextshub.org/projects/b23048f6-b832-4379-982a-7fe39a425ae8"
  ...
}

  1. Value using a LocalContexts Project Id

{
  ...
  "local_contexts_project_id": "b23048f6-b832-4379-982a-7fe39a425ae8"
  ...
}

  • Additionally, datasets can be manually tagged in the GBIF Registry in cases where the publisher prefers to link the LocalContexts project identifier at the dataset level rather than annotating individual occurrence records. This tagging will be manually applied by the GBIF Data Products team upon request from the data publisher. Example values when tagging a dataset through the MachineTag API:

{
  "namespace": "localcontexts",
  "name": "local_contexts_project_id",
  "value": "https://sandbox.localcontextshub.org/api/v2/projects/b23048f6-b832-4379-982a-7fe39a425ae8/"
}

GBIF Pilot Hosted Portal on Indigenous Data Governance

As a mechanism to promote and showcase the relevant information and mediated data, the coordinating group decided on the implementation of a GBIF Hosted Portal with, initially, the following components:

Data sections

  • Dataset view: lists the datasets related to this pilot. Datasets that have been tagged using machine tags will display the Notices and Labels, along with a link to the associated project landing page.

  • Occurrence records view: shows all records published through those datasets. Similar to datasets, occurrence records will provide a visualization of Notices and Labels, and a link to the associated project landing page.

  • Publishers: a list view of the data publishers that participated in this pilot.

  • Literature (optional): a list view of literature (scientific papers, articles, etc.) that cite the datasets involved in this pilot.