Paleoecological Inferences for Turborotalita Nikolasi (Koutsoukos, 2014) Based on Stable Carbon and Oxygen Isotopes

The Cretaceous-Paleogene (K-Pg) boundary is characterized by one of the largest mass extinctions in the geological history of the Earth at the base of the Danian (Alvarez et al., 1980). Among the groups most affected by this event were planktic foraminifera (e.g., Luterbacher & Premoli-Silva, 1964; Koutsoukos, 1996, 2014; Arenillas et al., 2018; Molina, 2015; Huber et al., 2020) with the vast majority of larger and K-selected species becoming extinct at the end of the Cretaceous (Premoli-Silva & Sliter, 1999). The species that did survive (e.g., Guembelitria cretacea, Hedbergella holmdelensis, and H. monmouthensis) later gave rise to all Cenozoic species (Aze et al., 2011).

The species Turborotalita nikolasi (=Praemurica nikolasi) was described by Koutsoukos (2014) in the Campos Basin, South Atlantic Ocean, from lower Danian strata. It represents the earliest recorded Truncorotaloididae and the oldest record of a planktic foraminiferal species presenting a cancellate wall texture. Turborotalita nikolasi most likely evolved directly from Hedbergella monmouthensis within the early P0 Zone, a few thousand years after the K-Pg boundary event. Therefore, T. nikolasi probably gave rise to all subsequent cancellate lineages of the early Danian and the Cenozoic (Koutsoukos, 2014). Using new criteria for defining the genus, Pearson & Kučera (2018) reassigned Praemurica nikolasi to the genus Turborotalita, extending the stratigraphic range of Turborotalita across the entire Cenozoic. This taxonomic revision is further supported by the genetic study of T. quinqueloba (Natland) by Aurahs et al. (2009), who suggested a deep-time phylogenetic divergence for the Turborotalita lineage.

The significance of T. nikolasi for understanding Cenozoic planktic foraminiferal evolution cannot be overstated, as it is the likely progenitor of all cancellate Cenozoic species. However, its paleoecological preferences are currently unknown. Here we provide new paleoecological inferences for T. nikolasi based on a study of oxygen (δ¹⁸O) and carbon (δ¹³C) isotopes and compare this data with isotope records for other planktic and benthic foraminiferal species recovered from DSDP Site 356.

DSDP Site 356 is located on the southwestern São Paulo Plateau (28°17.22′S, 41°05.28′W; at 3.175 m water depth; Fig. 1a). Core recovery at Site 356 spanned the Albian to Recent (Perch-Nielsen et al., 1977), with the succession being assigned to seven lithological units. In this study, we focus on the Maastrichtian to lower Paleocene (Danian) interval (415.25 to 404.26 meters below seafloor - mbsf), which is characterized by pelagic carbonate sediments assigned to lithologic Unit 4, described as a nannofossil and nannofossil-foraminifer chalk (Perch-Nielsen et al., 1977). Specifically, the sediments recording the K-Pg boundary were deposited under oxic conditions, and the supply of terrigenous material was cut off during the late Maastrichtian (Perch-Nielsen et al., 1977). The biostratigraphic framework used here is based on the planktic foraminiferal biostratigraphy (Kochhann et al., 2014; Krahl et al., 2017).

A total of five samples were used in this study, taken from the interval that covers the base of the Danian at Site 356, between the Pα and P1a planktic foraminiferal biozones (Krahl et al., 2017). In order to evaluate the δ¹⁸O and δ¹³C signatures of T. nikolasi, we compared results with isotopic data of mixed-layer (Guembelitria cretacea) and thermocline (Subbotina trivialis and Chiloguembelina midwayensis) planktic foraminiferal species, as well as with δ¹⁸O and δ¹³C values of Nuttalides truempyi, a deep-water benthic foraminiferal species that is widely used in isotopic studies of the Danian (e.g., Coxall et al., 2006; Birch et al., 2016, 2021; Jehle et al., 2015, 2019).

Stable isotopes (δ¹³C and δ¹⁸O) were measured in a Thermo Finnigan MAT 253 Plus mass spectrometer, equipped with an automated carbonate preparation KIEL IV Carbonate device in the stable isotope facility at the Technological Institute for Paleoceanography and Climate Change (itt OCEANEON, UNISINOS University). For the analytical procedure, a calibration curve was constructed with the following standards: IAEA 603, IAEA-CO-8, and SHP2L (Crivellari et al., 2021); the latter was also used as an internal standard. Stable isotope results were reported in the Vienna PeeDee Belemnite (VPDB) scale, and analytical precision was better than 0.09‰ for δ¹⁸O and 0.04‰ for δ¹³C.

Selected foraminiferal specimens were recovered from the studied interval of DSDP Site 356 and used for δ¹⁸O and δ¹³C analyses (Fig. 2). The preservation of these specimens was considered glassy, as they displayed little or no taphonomic alteration (e.g., secondary crystallization). Measurements of δ¹⁸O and δ¹³C are presented in Appendix 1 and discussed individually below.

Chiloguembelina midwayensis δ¹⁸O values (ranging from −1.23 to 0.43‰) and Subbotina trivialis δ¹⁸O values (ranging from −1.38 to −0.77‰) depict parallel trends (Fig. 3a). Turborotalita nikolasi records lower δ¹⁸O values, varying between −1.25 to −2.07‰, close to those of Guembelitria cretacea, which range from −2.47 to −1.25‰. Turborotalita nikolasi δ¹⁸O values were overall slightly higher than those of G. cretacea, except for sample 29-1 (145–148 cm) at the top of the Pα biozone (Fig. 3a).

The distinctive δ¹⁸O signatures recognized for the different early Danian species at Site 356 imply that they inhabited different depths in the water column. High δ¹⁸O values observed for the benthic species N. truempyi are typical of cold bottom waters (e.g., Coxall et al., 2006; Birch et al., 2012, 2016). The species Chiloguembelina midwayensis and Subbotina trivialis record intermediate δ¹⁸O values, between those of N. truempyi and G. cretacea, suggesting that they inhabited the thermocline during the early Danian (Fig. 3a). This observation is in line with paleoecological interpretations for S. trivialis by Coxall et al. (2000). The closely related species Chiloguembelina morsei probably inhabited relatively deep waters (Boersma & Premoli Silva, 1983; D'Hondt & Zachos, 1993), supporting our interpretation that C. midwayensis lived within the thermocline.

The recorded isotope data indicates that T. nikolasi most likely lived in the mixed-layer, due to its relatively low δ¹⁸O values, similar to those of Guembelitria cretacea (= mixed-layer and/or near-surface dweller; Boersma et al., 1979; Boersma, 1984; D'Hondt & Zachos, 1993; Birch et al., 2012, 2016). According to Koutsoukos (2014), T. nikolasi is practically homeomorphic with Turborotalita quinqueloba (Natland, 1938), whose biostratigraphic range spans from the Oligocene–Recent, and is remarkably (morphologically) similar to the late Eocene species Turborotalita praequinquelobaHemleben & Olsson, 2006 (in Pearson et al., 2006). The extant species T. quinqueloba is supposed to thrive at the upper thermocline or the mixed layer (Pearson & Wade, 2009). Additionally, plankton tow studies indicate that T. quinqueloba inhabits the upper water column (e.g., Meilland et al., 2020), supporting our interpretation of a mixed layer habitat for T. nikolasi.

The ranges of planktic foraminiferal δ¹³C values for the studied interval at Site 356 varied as follows: 0.4 to −0.39‰ for G. cretacea, 0.54 to −0.14‰ for C. midwayensis, 1.59 to 0.38‰ for S. trivialis, and 1.69 to 0.54‰ for T. nikolasi (Fig. 3b). Nuttalides truempyi exhibits higher δ¹³C values, ranging from 2.28 to 1.15‰ (Fig. 3b).

The recorded N. truempyi δ¹³C values at Site 356 are higher than the planktic foraminiferal values. This pattern is unusual but similar to what was observed by Birch et al. (2016) for N. truempyi and G. cretacea δ¹³C values immediately after the K-Pg boundary at ODP Site 1262 (Walvis Ridge; South Atlantic). Usually, high δ¹³C values occur at surface waters (planktic foraminifera = high), and low δ¹³C values occur at bottom waters (benthic foraminifera = low) as a consequence of the organic carbon biological pump (e.g., Birch et al., 2021). Therefore, we assume that higher-than-planktic benthic δ¹³C may reflect the inefficiency of the biological pump in transferring ¹²C to the ocean floor in the aftermath of the K-Pg boundary event, probably related with a near-collapse of surface ocean productivity (e.g., Birch et al., 2021). According to Birch et al. (2016, 2021), low δ¹³C gradients between surface and bottom waters persisted for ∼300 kyrs after the K-Pg boundary event, which is in accordance with the high N. truempyi δ¹³C values within the Pα and P1a biozones observed at Site 356.

The δ¹⁸O and δ¹³C isotope signatures of foraminiferal species that lived in the mixed-layer, thermocline, and deep waters during the earliest Danian at DSDP Site 356, São Paulo Plateau, was assessed. Our main goal was to characterize the paleoecology of T. nikolasi (=Praemurica nikolasi). Turborotalita nikolasi exhibits δ¹⁸O values similar to those of G. cretacea, suggesting that both species cohabited the mixed layer, while S. trivialis and C. midwayensis inhabited the thermocline. The high δ¹³C values of T. nikolasi compared with those of other planktic foraminiferal species, including G. cretacea, suggest that it most likely engaged in photosymbiosis, and thus might represent the earliest record of photosynthesis for Cenozoic planktic foraminifera. In addition, N. truempyi shows higher-than-planktic δ¹³C values, suggesting reduced transfer of ¹²C to the ocean floor, which may have been caused by inefficiency of the organic carbon biological pump after the K-Pg extinction event.

We would like to thank the International Ocean Discovery Program (IODP) for providing the studied samples, as well as to the support of technicians at itt OCEANEON (UNISINOS) during geochemical analyses. We thank Mark Leckie and an anonymous reviewer for suggestions and comments that helped us to significantly improve the manuscript. This research is part of the Project IODP/Capes #88887.091703/2014-1. G.K. and M.H.H.B. would also like to thank the Coordination for the Improvement of Higher Education Personal (CAPES) for the scholarships provided. G.F. is a research fellow of CNPq (grant 308087/2019-4).