Login

    Requirements for data deliverables of BID projects

    The BID program aims to enhance the capacity for effective data mobilization in Africa, the Latin America and the Caribbean region, and the Pacific islands.

    BID-funded projects must produce data deliverables as datasets published on GBIF. There are several classes of datasets supported by GBIF that start simple and become progressively richer, more structured and more complex (see #1-4 below). We encourage all BID applicants and participants to aim making their datasets as rich as possible to support wider use of the data. For some more informal information about dataset types on GBIF, please consult this blog article.

    Dataset metadata requirements

    All datasets delivered in the context of the BID-funded projects must comply with the following requirements:

    • The datasets must be published and successfully indexed on GBIF.org
    • The datasets must have extensive metadata which should give context for the associated data (answer the “What? Where? When? How? Why?” questions about the data). This entails filling the metadata fields relating to the following sections:
      • Title, general description and license information
      • Taxonomic coverage
      • Geographic coverage
      • Temporal coverage
      • (Sampling) methods
      • Purpose
    • The datasets must be linked to their respective project page by using the related project identifier as issued by GBIF.

    1. Description of biodiversity data holding

    The simplest type of dataset that can be shared on GBIF are metadata-only datasets. They allow participants to highlight and describe data that may not be available online yet. Metadata-only datasets can be particularly useful in regions where there is a need to identify possible sources of biodiversity data. They can describe for example, undigitized or untranscribed biodiversity-related documents.

    Metadata-only dataset must comply with the GBIF dataset requirements described above.

    2. Species or taxon lists (checklists)

    A richer type of dataset available on GBIF is checklist datasets. They can help capture and share information at the taxon level, such as taxonomy,  or properties (e.g. invasive, vernacular names, threatened, etc.) of a given list of taxa. In addition to the GBIF dataset requirements, checklists must also contain individual taxon records and their relevant associated fields. The minimum information needed is the following:

    3. Occurrences

    Occurrence datasets contain records of observations or collections of a given set of organisms at a given time and place. In addition to the GBIF dataset requirements, occurrence datasets must also contain individual occurrence records with the following information (“What? Where? When?”):

    Note that occurrence datasets can be enriched with extensions.

    4. Sampling-events, survey, monitoring

    The richest type of datasets available on GBIF are sampling-event datasets.

    Sampling-event data describes species occurrences in time and space together with details of sampling effort. Such data is available from thousands of environmental, ecological, and natural resource investigations. These can be one-off studies or monitoring programs. Such data are usually quantitative, calibrated, and follow certain protocols so that changes and trends of populations can be detected.

    All sampling event datasets contain occurrences and must comply to the same requirements as occurrence datasets. In addition to those, they must also contain records of events and information about the sampling context (“how?”). These can be captured in event fields (like sampling protocol, sample size value and unit, sampling efforts) or in the Humbolt extension for ecological repositories.

    Please contact the BID support team if you have any question or wish additional information at BID@gbif.org.