Plankton and benthic foraminiferal dataset for the study of the Eocene-Oligocene transition

The Eocene–Oligocene transition (EOT) was the crucial turning point when Earth’s climate shifted to its current icehouse state. Understanding how the marine biosphere responded during this transition is not well-constrained, appearing as a simple extinction pulse in low temporal resolution global compendia. Here we designed a novel AI-inspired metaheuristics algorithm to construct a high-resolution global species richness history across the EOT for the rich foraminifera fossil record with an imputed ~29,000-year resolution. The revealed diversity dynamics are complex and differ for each foraminiferal group with distinct ecology. Planktonic and shallow-water larger benthic foraminifera show steady diversity levels in the early phases of the transition in the latest Eocene after a long-term reduction, while the deeper-water small benthic foraminifera radiate remarkably and decline over the same interval. In the earliest Oligocene, the planktonic and larger foraminifera suffered major species losses coincident with the first continental-scale ice sheet forming on Antarctica, while small benthic foraminifera diversity held steady, followed by an accelerating lowering as the early Oligocene proceeded. These findings reveal complicated and ecologically differentiated environment-life processes, indicating the importance of high-resolution temporal data for dissecting out ecological responses to major environmental changes.

Methods
Data used in this study were manually collected from peer-reviewed publications. Each data record includes clear metadata linking back to the original source in the OneStratigraphy database. Any errors were corrected (e.g., spelling mistakes in species names), and missing information was filled in (e.g., latitude and longitude data). We selected sections/sites containing foraminifera occurrences from the Eocene to the Oligocene. The raw dataset contained 13,138 local bioevents records (i.e., first and last appearance records) and ~60,000 occurrences of 2,988 taxonomic units from 163 published stratigraphic sections, encompassing both calcareous and agglutinated foraminifera. These sections, including drill cores and outcrops, are widely distributed in the present oceans and continents such as Europe, Africa, and Asia. The dataset was first cleaned by excluding open nomenclature, such as sp./spp. (622), aff. (63), question marks for species names (6). Nevertheless, the conferring species (cf.; 175) and the group species (ex gr.; 25) were preserved and assigned to the referenced species. Taxonomic assignments below the species level (i.e., subspecies and variety) were mostly integrated to the species level. All non-foraminifera fossils were removed. The dataset after cleaning was thoroughly examined and verified against other independent data sources, including taxonomic atlases, foraminiferal databases (Mikrotax and WoRMS), and related taxonomic references, and further verified and resolved by a group of foraminiferal taxonomic experts for correctness and consistency: Bridget Wade (PF), Laia Alegret (SBF), Qinghai Zhang (LBF and SBF), and Peiyue Fang (PF, LBF and SBF).

In the present dataset, the drill core Hole 647A has been studied repeatedly, focusing on both SBF and PF for variable use, such as testing biotic response to EOT, studying high-latitude deep-water sedimentary sequence, and stratigraphic correlation. The three reports were integrated into one section by depth.

The final dataset after data cleaning and verification included 9,032 local first and last occurrence records of 1,269 species in 161 published sections.

The Constrained Optimization with an Evolutionary Algorithm (CONOP.EA) compositing method is used to integrate the local biostratigraphic data from all 161 sections/sites and to correct regional diachronism caused by migration, fossil preservation, and sampling biases.