Checklist Bank stores names and information about names. This information may come from one of several resource types:
- Published as taxonomic data using Taxon Concept Schema or as a taxon object using the Darwin Core terminology adopted by GBIF for the Global Names Architecture
- Names and classification data published to GBIF with primary occurrence data (specimens, observations, etc.) via Darwin Core or Access to Biological Collections Data (ABCD)
- Names originating in Really Simple Syndication (RSS) feeds, publications and documents indexed by GBIF through direct or partnered publication indexing.
Indexing name usages
The atomic unit of Checklist Bank is a ‘name usage’. This refers to the use of some sort of name (generally a scientific name) in association with a data or content object, such as a specimen, observation, a publication, or a member of a list. GBIF maintains very large indices of names associated with specimen, observation and publication information; Checklist Bank maintains a distinct list of names found within these indices.
Reconciling orthography
ChecklistBank uses tools and interfaces to group different spellings of the same name into common lexical groups. This enables services to be built that ensure a search for information on a species returns all data even if it is not spelled exactly the same. Checklist Bank stores this grouped information and relies on algorithms, such as Taxamatch [link to SVN] and manual interfaces to maintain these groups.
One goal of the Global Names Architecture is to reconcile all names to a lexical group that has at least one member tied to a name usage originating in a authoritative taxonomic or nomenclatural database.
Resolving nomenclature
Checklist Bank explicitly seeks to index and collate nomenclatural groups in order to provide services that can suggest nomenclatural synonymy to expand a search. Nomenclatural synonyms are names that share a common type and typically occur when a species is moved from one genus to another. For example the species ‘Francolinus afer (Müller)’ was originally placed in the genus Tetrao, creating the nomenclatural synonym, ‘Tetrao afer Müller’. In many cases, a search using the first term might also include a search using the second term.
In Checklist Bank, a nomenclatural group is derived from two or more lexical groups.
One goal of the Global Names Architecture is to resolve at least one name usage within all nomenclatural groups to an original nomenclatural act, or to identify the name as a member of a ’blacklist’ of names that no longer carry standing within nomenclature.
Distinguishing homonyms
Checklist Bank seeks to provide the means to quantify the complete set of all primary and secondary homonyms. See the GBIF Names Management Hierarchy, below.
Indexing published checklists
Checklist Bank will accurately represent checklist data published within the GBIF network. Checklist data may range in detail from simple lists of species organized around a semantic theme (e.g. Venomous Reptiles of the US Southwest) to detail taxonomic monographs (e.g. Catalogue of Living Whales). These data will be consistently presented via web service interfaces and users will be easily redirected to the source databases for access to the most up-to-date and complete information. The Global Names Architecture provides the means to keep consistently up-to-date with these sources.
Maintaining GBIF names management hierarchy
One specialized use of Checklist Bank is to provide the raw materials to organise all data published through the GBIF network that is indexed in the GBIF data portal into a common taxonomic framework. Individual checklists can be used as the basis for organising and accessing data within the GBIF index, but no single checklist captures all the names within it. So an additional structure is created by intelligently merging all the data into a taxonomically-informed tree that ensures all the data are tied to some higher taxonomic group.
