Cenozoic macroperforate planktonic foraminifera phylogeny of Aze & others (2011). Relational database for TimeScale Creator Evolutionary Tree. Corrected Version, July 2018

The TSCEvolTree_Aze&2011_CorrJul2018 database is based upon the phylogeny of Aze & others [1] (the "2011" study/paper/etc.) and comprises two main tables, one for morphospecies, MorphospeciesAze_TableS3, the other for lineages, BiospeciesAze_aL.

The morphospecies table [following 1, Appendix S1, Table S3] compiles stratigraphic ranges and indications of morphology, ecology, and geography from stated sources, as well as stratigraphic analyses made using the Neptune database. The main focus of the morphospecies table for this transfer to TimeScale Creator is upon the ranges. This includes augmenting the table to record corrections or amendments to some dates (always within the provision not to revise but to better represent the intensions of the 2011 study) and, for a few ranges, to add questionable or conjectured range extensions. To enable the ranges to be recalibrated for subsequent time scales, the date sources are here linked to an ancillary table (MorphospeciesAze_TableS3DateRef) that explicitly indicates time scales for these sources. This measure also is needed to support the transfer because corrections or amendments made to Ma values from the 2011 study during the transfer are not necessarily against the time scale employed by that study (i.e., that of Wade & others [2], calibrated to Cande & Kent [3]). The measure also facilitates clear differentiation in the database where more than one time scale may, in effect, be applied by or to the source. Also added via the morphospecies table are indications of levels of evidential support for the ancestor–descendant proposals; these indications are generalized and not meant to suggest new information but rather to highlight a small number of demonstrably less or better supported proposals, and also to provide future capability for more deliberate attention to this aspect.

The main feature provided by the lineage table which augments the 2011 study is to embed in the table any links which may have been exercised manually in the 2011 study between the stratigraphic range of a lineage and that of an associated morphospecies; this applied to approximately 40 % of start dates and almost all end dates of lineages. This is implemented in the database by assigning a morphospecies range point (start or end) to each lineage range point (start or end) and employing a field to turn on or off the link between the lineage and morphospecies range points; if this field is turned on, the lineage point adopts the Ma of the morphospecies range point; if turned off, the original Ma given in the 2011 study to the lineage point (or its replacement if corrected or amended) is retained but the time scale of the morphospecies point is employed for calibration of the Ma. This measure enables correspondences between the timing of the morphospecies and lineage trees to be made transparent but also to easily retain these linkages if morphospecies ranges are changed. A similar embedding feature has been added which allows the database to work out the lineage memberships of each morphospecies, thus obviating the need to manually construct these lists (and inevitably make human errors).

Another key table (Wade & others, 2011 Datum) compiles the datums from Wade & others [2]. Here their tables are augmented to separately depict the zonations of Berggren & others (1995) [4] and Berggren & Pearson (2005) [5] in addition to the Wade & others zonation in its two versions (calibrated to Cande & Kent, 1995 [3], and to Gradstein & others, 2004 [6]). The remaining tables in the database include those (“global”) shared with other databases to provide, amongst other information, species and genus nomenclature, links to portals, and TimeScale Creator time units and datums.

From the tables: derived data and information, and datapacks

Much of the essential data needed for tree construction — names, dates, ancestors — are tabular or relational (or nearly so) and so mostly amounts to employing SQL queries to recast or combine elements from the database tables. This includes, from the main tables, the selection of ancestors, dates, and sources and associated commentary where, for instance, there are multiple options (e.g., corrected or amended entries), and then linking with accessory tables to add key determinants such as zonation time scales, accessory information such as taxonomic and grouping details, and paraphernalia such as colours.

Time-scale calibration of morphospecies and lineage dates employed a nested series of queries progressively recalibrating from those dates against the earlier published zonations to those against subsequent schemes, based on the augmented compilation of datums of Wade & others [2] (database table Wade & others, 2011 Datum). Proportional calibration between the zonations of Berggren & others (1995) [4], Berggren & Pearson (2005) [5], and finally Wade & others, calibrated to Cande & Kent (1995) [3], employed zonal index (marker) datums only. So these calibrations, based only on sometimes relatively coarsely spaced planktonic-foraminiferal events, should be considered minimally adequate. For the jump to GTS2004, that is, from the scales of Cande & Kent (1995) to Gradstein & others (2004) [6], all datums of Wade & others were used, providing a finely tuned conversion, at least in terms of planktonic-foraminiferal events. Later GTS conversions, from GTS2004 to GTS2012 and finally GTS2016, employed planktonic-foraminiferal zonal index datums using tables augmented from TimeScale Creator spreadsheet data.

Programming, coded in the database’s Visual Basic, is then used to integrate tables and queries into derived tables from which TimeScale Creator datapacks can be formulated. The programming includes procedural programming which is especially needed to generate the textural information provided in pop-ups as this is mostly nonrelational. Although the programming was developed inexpertly “in-house” and is code-intensive, it had the advantage of being able to be developed specifically to purpose and, being nonproprietary, able to be made available for scrutiny.

1. Aze T, Ezard TH, Purvis A, Coxall HK, Stewart DR, Wade BS, et al. A phylogeny of Cenozoic macroperforate planktonic foraminifera from fossil data. Biological Reviews of the Cambridge Philosophical Society. 2011;86(4):900-27. doi: 10.1111/j.1469-185X.2011.00178.x.
2. Wade BS, Pearson PN, Berggren WA, Pälike H. Review and revision of Cenozoic tropical planktonic foraminiferal biostratigraphy and calibration to the geomagnetic polarity and astronomical time scale. Earth-Science Reviews. 2011;104(1-3):111-42. doi: 10.1016/j.earscirev.2010.09.003.
3. Cande SC, Kent DV. Revised calibration of the geomagnetic polarity timescale for the Late Cretaceous and Cenozoic. Journal of Geophysical Research: Solid Earth. 1995;100(B4):6093-5. doi: 10.1029/94JB03098.
4. Berggren WA, Kent DV, Swisher CC, III, Aubry M-P. A revised Cenozoic geochronology and chronostratigraphy. In: Berggren WA, Kent DV, Aubry M-P, Hardenbol J, editors. Geochronology, time scales and global stratigraphic correlations. Tulsa (Oklahoma, USA): Society for Sedimentary Geology, Special Publication. No. 54.; 1995. p. 129-212.
5. Berggren WA, Pearson PN. A revised tropical to subtropical Paleogene planktonic foraminiferal zonation. Journal of Foraminiferal Research. 2005;35(4):279-98. doi: 10.2113/35.4.279.
6. Gradstein FM, Ogg JG, Smith AG, editors. A Geologic Time Scale 2004. Cambridge: Cambridge University Press; 2004.