A data product version is an identifier that distinguishes one release of a data product from another, for example releases produced with different processing algorithms or updated calibration parameters.1
In NASA dataset metadata, the version field serves this purpose, and users are encouraged to use the latest version of a data product.1 That advice comes from one catalogue’s guidance. A newer release is not automatically the right choice for every analysis, because comparability with earlier work may matter and the source does not discuss that trade-off.1
SWEET places version beneath its iteration class, which reflects the idea of a product being produced repeatedly with changes.2 The ontology gives only that hierarchy, so it does not set numbering schemes, compatibility rules or how versions relate to a persistent identifier.2
A version is not the same as a processing state or a processing level. A product at a given level may exist in several versions, and the sources reviewed do not describe how any particular centre numbers them.
Footnotes
-
NASA Earthdata, “Earth Observation Data Basics”. https://www.earthdata.nasa.gov/learn/earth-observation-data-basics ↩ ↩2 ↩3
-
ESIP, SWEET ontology, reprDataProduct module (Turtle source). https://raw.githubusercontent.com/ESIPFed/sweet/master/src/reprDataProduct.ttl ↩ ↩2