Metadata is information about data, i.e. the description of a data set including any info which is necessary for the understanding of the data: who, what, when, where and how were the data produced? The metadata fields describe the factual data in the context of sampling, measurement and publication and also make it citable. The size/composition of a data set (and consequently the amount of required metadata) depends on the definition of its granularity.
The following fields are composed as the metadata description which is given as the header of each data set (mandatory fields in bold):
- Event with geographical position (for georeferenced data), elevation, area, comment, devices; defined in event table.
- Expedition/cruise/campaign/leg information (label, ship, station, chief scientist, area); defined in campaign table.
- PI (principle investigator/originator) for each data series (full name with email); defined in staff table.
- Author(s) of data set (if in addition to PI); defined in staff table.
- Data set title (phrasing should be similar to the title of a publication)
- Reference(s) (one if data are a supplement to a publication or several if data are a compilation from various sources); defined in reference table.
- Method for each data series (with URI or description as pdf), defined in method table.
- Comment for each data series; free text, may include URIs.
- Comment/description related to the data set as free text in the comment field or under
- Further details linked via a persistent identifier, mostly as pdf-file.
- Abstract (if data are a supplement to a publication or the data set is a data publication)
- Keyword(s) as provided through the thesaurus.
- Project(s) full name and acronym; defined in the project table.
- Cross-references to other versions (URI).
!> Pangaea should not be used for the management of metadata solely, e.g. as reference database or just listing campaigns/events of a project. In principle metadata should be archived only, if the related factual data are also imported, following the principle:
- No metadata without data - no data without metadata!