Null Island is in the Gulf of Guinea off western Africa, at latitude zero and longitude zero.
To learn more about Null Island I turned to GBIF. I downloaded the 1,431,391 occurrence records flagged “Zero Coordinate” as a Darwin Core archive.
From the verbatim.txt table I deleted 480,746 records with no entry at all in decimalLatitude or decimalLongitude, leaving 950,645 records.
Many different observers and institutions have shared their Null Island records with GBIF. Of particular note is Museu Botânico Municipal of Rio de Janeiro, Brasil, which contributed 215,775 records. Some other dataset contributors have added to the intellectual challenge of reviewing Null Island occurrences by shuffling data entries between Darwin Core fields. The resulting records are fascinating jigsaw puzzles.
Biologists have apparently been sampling Null Island from top to bottom. There are records up to 25000 m elevation and 3753 m depth. The earliest records date from 1004 and “predictive” records extend out to 9999-03-11.
It’s hard to generate flora and fauna lists for Null Island from GBIF data because so many of the taxon field entries are unfamiliar, but more than 50,000 of the records are from kingdom Plantae and ca 375,000 from Animalia. The flora of this tiny speck of land is fabulously diverse, with more than 1400 species. Even more impressive, there are some 14,800 animal species known from Null Island.
It’s also a little difficult to determine whether any type material of new species has come from Null Island, because the Darwin Core typeStatus field has numerous strange entries such as “J.F.Maxwell91-920”. However, there are indeed many Null Island types, including 1030 holotypes.
As a millipede specialist I was particularly interested to see which species had been recorded from Null Island. One fascinating record is for Metopidiothrix shelleyi Shear, 1994, previously known only from the Philippines but probably accidentally introduced by the pirates who are so active in the Gulf of Guinea.
Note: This will be my 51st and last post to the GBIF discussion forum. In December this year I will also be ending the BASHing data series of weekly posts on command-line data work, but the blog and the accompanying Darwin Core table checker and Data Cleaner’s Cookbook websites will remain online.
I’m happy to continue corresponding with biodiversity data workers about improving the quality of their datasets. If you are planning to share data with GBIF, I encourage you to have your dataset audited beforehand, to avoid adding more unusable records to the many millions already making life difficult for GBIF’s end-users.
Robert Mesibov (“datafixer”); mesibov@datafix.com.au
