All the available data sets containing diatom counts were compiled. These data sets were the 166 database used by Pichon et al. (1992), new data gathered separately by Dr Pichon (1994) and the author (L. Armand-1997), and Abbott's (1973) southeast Indian Ocean surface sample data set.
Since each of the data sets were developed over periods of years or were subject to differences in processing and taxonomy, primary analysis was concerned with aligning current taxonomic advances and knowledge against the record of species identified in each data. With exception only to the Abbott southeast Indian Ocean data set (which could no longer be located for verification), the remaining data sets were combined and subjected to taxonomic reform.
Following taxonomic reform a core-top diatom database was compiled using sample-selection criteria and through the removal of samples representing unique environmental regimes (i.e. Antarctic Peninsula region). A total of 115 samples from an original 194 remained available for use, and is identified as the Antarctic Diatom Database (ADB).