Abstract
AbstractThe application of transfer functions on fossil assemblages to reconstruct past environments is fundamentally based on the assumption of stable environmental niches in both space and time. We quantitatively test this assumption for six dominant planktic foraminiferal species (Globigerinoides ruber (pink), G. ruber (white), Trilobatus sacculifer, Truncorotalia truncatulinoides, Globigerina bulloides, and Neogloboquadrina pachyderma) by contrasting reconstructions of species realized and optimum distributions in the modern and during the Last Glacial Maximum (LGM) using an ecological niche model (ENM; MaxEnt) and ordination framework. Global ecological niche models calibrated in the modern ocean have high predictive performance when projected to the LGM for subpolar and polar species, indicating that the environmental niches of these taxa are largely stable at the global scale across this interval. In contrast, ENMs had much poorer predictive performance for the optimal niche of tropical‐dwelling species, T. sacculifer and G. ruber (pink). This finding is supported by independent metrics of niche margin change, suggesting that niche stability in environmental space was greatest for (sub)polar species, with greatest expansion of the niche observed for tropical species. We find that globally calibrated ENMs showed good predictions of species occurrences globally, whereas models calibrated in either the Pacific or Atlantic Oceans only and then projected globally performed less well for T. sacculifer. Our results support the assumption of environmental niche stability over the last ~21,000 years for most of our focal planktic foraminiferal species and, thus, the application of transfer function techniques for palaeoenvironmental reconstruction during this interval. However, the lower observed niche stability for (sub)tropical taxa T. sacculifer and G. ruber (pink) suggests that (sub)tropical temperatures could be underestimated in the glacial ocean with the strongest effect in the equatorial Atlantic where both species are found today.
Highlights
The following supporting information include figures and tables which support the main article and include preliminary niche analyses
Testing the impact of preservation (Figure S3a-b) : To test ENM sensitivity to selective dissolution of species within core-top samples, we compared outputs from the full occurrence dataset spanning the entire range of water depths (~ 50 - 5500 m) against a subset of the data containing only sites above 3000 m water depth in the North Atlantic for Globigerinoides ruber and Neogloboquadrina pachyderma (Hemleben et al, 1989)
It is important to note that the MARGO data has been selected to avoid major dissolution biases and we likely will underestimate the potential impact of dissolution
Summary
The following supporting information include figures and tables which support the main article and include preliminary niche analyses. We applied three thresholds here: 0%, or the “least training presence threshold” (sensu, Pearson et al, 2007), 90% fossil occurrences included and 50% fossil occurrences included) The least training presence threshold uses the lowest suitability value associated with an occurrence record and assumes that taxa are restricted to locations at least as suitable as those at which the species are observed at, and that sites at or above that threshold are suitable for the species This defines a minimum area within which the organism can occur, but ensures that no presence records are incorrectly predicted.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
