Villa 2008 Marine-micropaleontology

Marine Micropaleontology 69 (2008) 173–192

Contents lists available at ScienceDirect

Marine Micropaleontology j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / m a r m i c r o

Middle Eocene–late Oligocene climate variability: Calcareous nannofossil response at Kerguelen Plateau, Site 748 G. Villa a,⁎, C. Fioroni b, L. Pea a, S. Bohaty c, D. Persico a a b c

Dipartimento di Scienze della Terra, Università di Parma, Viale Usberti, 157A, 43100 Parma, Italy Dipartimento di Scienze della Terra, Università di Modena e Reggio Emilia, L.go S. Eufemia,19, 41100 Modena, Italy Earth and Planetary Sciences Department, University of California—Santa Cruz, Santa Cruz, California, 95064, USA

a r t i c l e

i n f o

Article history: Received 27 February 2008 Received in revised form 23 July 2008 Accepted 25 July 2008 Keywords: Eocene Oligocene Nannofossils Stable isotopes Paleoclimatology

a b s t r a c t A major deterioration in global climate occurred through the Eocene–Oligocene time interval, characterized by long-term cooling in both terrestrial and marine environments. During this long-term cooling trend, however, recent studies have documented several short-lived warming and cooling phases. In order to further investigate high-latitude climate during these events, we developed a high-resolution calcareous nannofossil record from ODP Site 748 Hole B for the interval spanning the late middle Eocene to the late Oligocene (~ 42 to 26 Ma). The primary goals of this study were to construct a detailed biostratigraphic record and to use nannofossil assemblage variations to interpret short-term changes in surface-water temperature and nutrient conditions. The principal nannofossil assemblage variations are identified using a temperate-warm-water taxa index (Twwt), from which three warming and five cooling events are identified within the middle Eocene to the earliest Oligocene interval. Among these climatic trends, the cooling event at ~ 39 Ma (Cooling Event B) is recorded here for the first time. Variations in fine-fraction δ18O values at Site 748 are associated with changes in the Twwt index, supporting the idea that significant short-term variability in surface-water conditions occurred in the Kerguelen Plateau area during the middle and late Eocene. Furthermore, ODP Site 748 calcareous nannofossil paleoecology confirms the utility of these microfossils for biostratigraphic, paleoclimatic, and paleoceanographic reconstructions at Southern Ocean sites during the Paleogene. © 2008 Elsevier B.V. All rights reserved.

1. Introduction The Eocene Epoch (~55 to 34 Ma) was characterized by a dramatic transition in global climate from a warm, ice-free “greenhouse” world to a cool “icehouse” world with significant glaciation in the polar regions. The Eocene began with an extreme, rapid warming event during the Paleocene/Eocene Thermal Maximum (PETM) (e.g. Kennett and Stott, 1991), followed by a sustained period of global warmth in the earliest Eocene (~ 55 to 50 Ma), known as the Early Eocene Climatic Optimum (EECO) (Zachos et al., 2001). Following this early

⁎ Corresponding author. Tel.: +39 0521 905370; fax: +39 0521 905305. E-mail address: [email protected] (G. Villa). 0377-8398/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.marmicro.2008.07.006

Eocene greenhouse period, a long-term cooling trend ensued, culminating in the widespread glaciation of Antarctica at the Eocene/Oligocene boundary (~ 34 Ma) during the Oi-1 event (Miller et al., 1987; Zachos et al., 1996; Lear et al., 2000; Zachos et al., 2001). A substantial long-term decrease in global temperatures is interpreted through the middle and late Eocene, with cooling of up to 7 °C in deep waters and high-latitude surface waters (Miller et al., 1987; Zachos et al., 2001). Long-term Eocene cooling that occurred from the EECO to the Oi-1 event was not entirely monotonic or stepwise. Rather, intervals of both rapid warming and cooling have been documented in middle and late Eocene deep-sea records. Specifically, a significant warming anomaly was initially identified at approximately 41 Ma in high southern latitude drillcores (Barrera and Huber, 1993; Diester-Haass and Zahn,

174

G. Villa et al. / Marine Micropaleontology 69 (2008) 173–192

1996). Subsequent work has documented this transient warming event in both high-resolution fine-fraction and benthic foraminiferal stable isotope records from several Southern Ocean sites (ODP Site 689, 738 and 748) (Bohaty and Zachos, 2003). These records show a sharp decline in δ18O values (~ 1‰), corresponding to a temperature increase of 4 °C of both surface and intermediate deep waters on both the Kerguelen Plateau (Indian sector of the Southern Ocean) and Maud Rise (Atlantic sector of the Southern Ocean). This prominent event is designated as the Middle Eocene Climatic Optimum (MECO) and is interpreted to represent an important climatic reversal in the midst of long-term cooling through the middle to late Eocene (Bohaty and Zachos, 2003). Recent studies on Paleogene calcareous nannofossil paleoecology have been completed in several time intervals and key areas of the Southern Ocean (e.g. Wei and Wise, 1990a; Wei et al., 1992; Bralower, 2002; Persico and Villa, 2004; Villa and Persico, 2006) and mid-latitude oceans (Agnini et al., 2006; Gibbs et al., 2006). These studies have confirmed the important role of these microfossils for paleoclimatic and paleoceanographic reconstructions in the Paleogene. High-resolution calcareous nannofossil records, however, have not been generated within the middle and late Eocene interval at Southern Ocean sites. Therefore, in order to further investigate both the longterm paleoceanographic evolution through this interval, as

well as transient climate events, we have developed a nearcontinuous record of middle Eocene to late Oligocene calcareous nannofossil assemblages from ODP Hole 748B (Kerguelen Plateau) (Fig. 1). The primary aims of this study are to obtain a high-resolution biostratigraphic record (Fig. 2), and, most importantly, to evaluate diversity and abundance patterns of calcareous nannofossil assemblages at high latitudes through the greenhouse to icehouse transition. We have aimed to test whether episodes of climatic change are manifested in sea surface settings of the Southern Ocean and, therefore, reflected in calcareous nannofossil assemblage variations. In turn, we have utilized the quantitative record of assemblage fluctuations developed in this study for paleoclimatic reconstructions through both long-term and short-term climatic events within the late middle Eocene to late Oligocene interval (~ 42 to 26 Ma). 2. Materials and methods Site 748 was drilled during Ocean Drilling Program (ODP) Leg 120 (Wise et al., 1992) and is located on the Southern Kerguelen Plateau in the southern Indian Ocean (~58°S) at water depth of 1291 m (Fig. 1). The Eocene and Oligocene sediments at Site 748 were deposited at relatively shallow paleodepths (~ 600–900 m), well above the carbonate com-

Fig. 1. Location map of ODP Site748.

G. Villa et al. / Marine Micropaleontology 69 (2008) 173–192

175

Fig. 2. Calcareous nannofossil biostratigraphy of ODP Hole 748B from core 9 to core 20. Southern Ocean biozonation modified from Wei and Thierstein (1991) and additional bioevents proposed in this work, correlated with standard zonations of Martini (1971) and Okada and Bukry (1980). LO: Lowest Occurrence; LCO: Lowest Consistent Occurrence; HO: Highest Occurrence; HCO: Highest Consistent Occurrence.

176

G. Villa et al. / Marine Micropaleontology 69 (2008) 173–192

pensation depth. The predominant lithologies within the Eocene interval at this site are foraminiferal nannofossil oozes and diatom nannofossil oozes, and, as such, are well-suited for a calcareous microfossil study. We analyzed calcareous nannofossils from Cores 748B-9H to 20H (~66 to 181 mbsf) taking samples from u-channels studied by Roberts et al. (2003). Samples were taken every 8–10 cm, with the exception of two intervals: from 180.96 to 177.16 mbsf, every 20 cm, and from 104.79 to 95.30 mbsf, every 30 cm. From 133.88 to 130.71 mbsf the core was not sampled because of fragmentation of sediments in the u-channel in this short interval. All samples were prepared using the settling technique described by de Kaenel and Villa (1996), which assures a uniform and homogeneous distribution of the nannofossils. Calcareous nannofossils were examined under crossedpolarized light, transmitted light, and phase-contrast light at 1250× magnification. Quantitative analyses were performed by counting at least 500 specimens on each slide and converting the number of index species normalized to a prefixed area (1 mm2). This technique allowed a detailed evaluation of the biostratigraphic signal and conversion of the abundances to percentages, used to estimate the paleoecological significance of the assemblage variations. Specimens were counted only if more than one half of an individual was observed. Specimens of Zygrablithus bijugatus were broken in many samples, and, therefore, two halves of this taxon were counted as one specimen in the abundance totals. In this study, we present the combined range of results from ODP Site 748, joining the data set obtained in the present study with the late Oligocene results previously presented by Villa and Persico (2006). The combined data set includes results from a total of 957 samples, spanning a time interval from the middle Eocene to the late Oligocene (Appendix A). This long high-resolution dataset allows precise evaluation of the stratigraphic position of important bioevents and the paleoecological relationships between taxa, as well as a detailed examination of the assemblage response to paleoclimatic events. In the Eocene–Oligocene section of Hole 748B, calcareous nannofossil abundance varies from common to abundant (Fig. 3), and preservation ranges from good to moderate. Z. bijugatus, a nonresistant to dissolution species (Wind and Wise, 1978), is always present and well preserved, yet signs of dissolution were observed in the central area of some specimens of Chiasmolithus spp. and Reticulofenestra reticulata. Discoaster spp. are often characterized by diagenetic overgrowth, and, thus, most specimens within this genus could not be identified to species level. The calcareous nannofossil assemblage record from Hole 748B is compared and interpreted along with the fine-fraction stable isotope records (δ13C and δ18O) from the middle Eocene to the upper Oligocene interval of Hole 748B. Previously published results from Bohaty and Zachos (2003) are combined here with new high-resolution data (~10 cm sample spacing in most intervals) from Cores 748B-12H, 13H, 17H, 18H, and 20H. Additional low-resolution analyses were also carried out on samples from Cores 748B-8H, 9H, 10H, and 11H. This combined data set represents a nearly continuous, high-resolution record from ~43 to 29 Ma, with a low-resolution record for the upper Oligocene interval (~29 to 25 Ma). The complete fine-fraction stable isotope dataset for Hole 748B is compiled in Appendix B.

The fine-fraction material for stable isotope analysis was obtained by disaggregating a small piece of bulk sample (~ 0.2 cm3) in deionized water and wet-sieving through 10 μm nylon mesh. Microscopic examination of several smear slides of the fine-fraction residues indicates that the samples are predominantly composed of nannofossils b12 μm in diameter, although most samples contained a minor fraction (b5%) of non-nannofossil carbonate. The mass contribution of nonnannofossil material (e.g. foraminiferal fragments), however, is considered to be a very minor component. Stable isotope analysis of the samples was performed using VG Prism and Optima mass spectrometers in the light stable laboratory at the University of California, Santa Cruz. NBS-19 and Atlantis II standards, in addition to an in-house Carrara Marble standard, were included in all sample runs. All values are reported relative to the Vienna Pee Dee Belemnite (VPDB) standard, and analytical precision is estimated at 0.04‰ (1σ) for δ13C and 0.06‰ (1σ) for δ18O. Approximately 15% of the samples were replicated on separate sample runs. 3. Age model and nannofossil biostratigraphy The age model for the Oligocene section of Hole 748B was constructed primarily using the magnetostratigraphy and tie points identified by Roberts et al. (2003). A reliable magnetostratigraphy, however, is not available for the Eocene interval of Hole 748B (Roberts et al., 2003). Within this lower interval, the fine-fraction stable isotope records were correlated to Site 689 (Maud Rise, Atlantic sector of the Southern Ocean) using unique features of the stable isotope records that could be calibrated with the magnetostratigraphic record available at this site (Florindo and Roberts, 2005). The ages applied to the magnetostratigraphic (reversal) events at both Site 689 and Site 748 are taken from the astronomically-tuned calibrations of Pälike et al. (2006, Appendix). Age calibrations for important nannofossil bioevents recognized at Site 748 are listed in Table 1; age assignments inferred in previous studies are recalibrated with respect to the geomagnetic polarity timescale of Berggren et al. (1995), and listed in Table 2. Previous studies of calcareous nannofossils at Site 748 were performed on low-resolution sample sets by Wei et al. (1992) and Aubry (1992a). The high-resolution sampling performed in the present study allowed refinement of the position of the bioevents (Table 1) recognized in these previous studies, as well as several additional bioevents. The nannofossil biozonation for the Southern Ocean as defined by Wei and Wise (1990b) and Wei and Thierstein (1991) is applied to the Hole 748B section (Fig. 2). The study interval spans from the R. umbilicus (pars) Zone to the R. bisecta (pars) Zone, which corresponds to the CP 14a (pars) Zone to the CP 19 (pars) Zone of Okada and Bukry (1980), respectively (Fig. 2). In this section, we identified 22 bioevents (Table 1) delineated as Lowest Occurrence (LO), Highest Occurrence (HO), Lowest Consistent Occurrence (LCO) and Highest Consistent Occurrence (HCO) of index species, according to Raffi et al. (2006). In the following a brief discussion of 14 key biohorizons, which we regard as most essential for the subdivision of the Eocene– Oligocene interval in the Southern Ocean, is given: (1) LO of R. reticulata. The LO of R. reticulata is identified at 171.15 mbsf (Fig. 3), at about the same position indicated

G. Villa et al. / Marine Micropaleontology 69 (2008) 173–192 Fig. 3. Calcareous nannofossil total abundance and abundance patterns of selected nannofossil species at ODP Hole 748B, expressed as number of specimens per mm2, are plotted against depth, chronostratigraphy, and magnetostratigraphy (Roberts et al., 2003) correlated to the geochronological time scale (Pälike et al., 2006). Biostratigraphic events are indicated with arrows. The grey area indicates an interval with no biostratigraphic data.

177

178

G. Villa et al. / Marine Micropaleontology 69 (2008) 173–192

Table 1 Summary of positions of calcareous nannofossil biohorizons at ODP Hole 748B Event

Depth (mbsf)

Core, section, interval (cm)

Ma (Berggren Ma (Pälike et al., 1995) et al., 2006)

HO C. altus HO C. oamaruensis HCO R. umbilicus HO I. recurvus LO C. altus HO R. oamaruensis LO R. oamaruensis LCO I. recurvus HO R. reticulata HO C. cf. altus HO N. dubius LO R. bisecta LO I. recurvus LCO C. oamaruensis HO C. solitus HCO C. solitus LO C. oamaruensis LO C. cf. altus HCO E. formosa HCO Discoaster spp. HO R. clatrata LO R. reticulata

71.60 104.84 105.94 109.01 112.95 115.86 125.20 127.48 128.35 130.35 134.34 143.90 148.25 149.25 149.45 151.35 155.01 155.81 157.11 165.99 170.01 171.15

9-4, 53–54 13-1, 29–30 13-1, 139–140 13-3, 149–150 13-6, 80–81 14-2, 21–22 15-2, 1–2 15-3, 80–81 15-4, 30–31 15-5, 80–81 16-1, 129–130 17-1, 120–121 17-4, 110–111 17-5, 60–61 17-5, 90–91 17-6, 140–141 18-2, 136–138 18-3, 66–68 18-4, 56–58 19-3, 143–145 19-6, 95–97 19-CC, 7–9

26.2 31.21 31.52 32.46 33.30

25.58 31.17 31.51 32.49 33.31 33.97 35.54 35.77 35.92 36.23 36.70 37.59 37.98 38.06 38.08 38.34 38.80 38.88 39.02 40.08 40.56 40.69 (hiatus)

LO = Lowest Occurrence, HO = Highest Occurrence, LCO = Lowest Consistent Occurrence, HCO = Highest Consistent Occurrence.

by Wei et al. (1992). The presence of a hiatus is inferred at 171.16 mbsf, at the core break between Cores 748B-19H and 20H, recognizable from the distribution of several nannofossil taxa and from abrupt changes in the δ18O and δ13C records. Therefore, the age of this bioevent (40.69 Ma) should be considered with a degree of caution (Table 1). (2) HO of R. clatrata. This event is recorded at 170.01 mbsf (Fig. 3) with a calibrated age of 40.56 Ma (Table 1). The range of this species at Site 748 corresponds to the range of R. onusta in Wei et al. (1992). Because the species recognized here show the features of R. clatrata, as described by Müller (1970), we therefore believe that

our R. clatrata corresponds to R. onusta of Wei et al. (1992). (3) HCO of Discoaster spp. The abrupt drop in abundance of Discoaster spp. at 166.00 mbsf is interpreted as a Highest Common Occurrence (HCO) datum for this genus (Fig. 3). This event has an age of 40.08 Ma (Table 1). Compared to the distribution of Eocene rosette-shaped discoasters at lower latitudes, their exclusion at Site 748 occurs much earlier and is here thought to be indicative of a paleoclimatic signal (i.e. an environmentally-controlled local disappearance). Wei and Wise (1990a), Persico and Villa (2004), and Arney and Wise (2003) recorded the absence of discoasters in the upper Eocene sediments at Maud Rise and Kerguelen Plateau; the latter authors also noted that discoaster abundance and diversity at Leg 183 are dramatically reduced during the middle Eocene. (4) HCO of Ericsonia formosa. As observed in the discoaster group, a similar middle to high-latitude diachroneity is noted for the HCO of E. formosa. This event occurs at 157.11 mbsf in the middle Eocene section, below the HO of Chiasmolithus solitus, and not in the earliest Oligocene (Fig. 3) as documented at middle latitude sites (Martini, 1971). Wei and Wise (1990b) noted a similar earlier extinction of E. formosa at high-latitude Sites 689 and 690 on Maud Rise. As suggested by Aubry (1992a) and Berggren et al. (1995), this bioevent is clearly diachronous between mid and high-latitude sites. The HCOs of both discoasters and E. formosa are also in agreement with the data documented at Site 738 (unpublished data) and are considered regional paleoecological bioevents. The age assigned here to HCO of E. formosa at Site 748 is 39.02 Ma (Table 1). (5) HO of C. solitus and LO of C. oamaruensis. In the biozonation of Wei and Thierstein (1991), the HO of C. solitus is reported to occur below the LO of C. oamaruensis. However, at some sites this relative position is not observed, and diachrony between high and mid-latitude sites has also been identified

Table 2 Biostratigraphic events published ages Event

Wei and Thierstein (1991) Site 744

Wei and Wise (1992)

HO C. altus HCO R. umbilicus HO C. oamaruensis LO C. abisectus HO I. recurvus LO C. altus HO R. oamaruensis LO R. oamaruensis HO R. reticulata LO R. bisecta LO I. recurvus HO C. solitus LO C. oamaruensis HCO E. formosa LO R. reticulata

26.1 31.3

25.8 31.3

late C12r

31.3 31.8

32.3

Late C12r Early C12r

33.7 35.4 36.1

33.7 35.4

36.0

36.0 38.4 38.0 41.2

Marino and Flores (2002a,b) Site 1090

34.0 35.36 36.1 C18r (40.13–41.257) 36.0 Reported as not reliable early C17r 32.8 41.2

Wei (2004)

Persico and Villa (2004) Sites 689 744

26.1 32.3

26.1 31.3

26.15 31.40–31.4 33.91–33.97

32.6

31.1 32.3

Mc Gonigal and Di Stefano (2002) Sites 1123–1124

35.0 38.0 36.0 40.3 37.0 32.8

Revised ages for nannofossil datums with respect to the geomagnetic polarity timescale of Berggren et al. (1995).

Arney and Wise (2003) Site 1138

35.4

33.7 35.8 35.9

38.0 36.0 37.9 37.0

36.0 38.2

41.2

32.34–32.55 33.18–33.19 33.91–33.71 35.78

36.1–35.8

G. Villa et al. / Marine Micropaleontology 69 (2008) 173–192

(Aubry, 1992a; Wei et al., 1992; Marino and Flores, 2002a) (Table 2). In addition, poor preservation of the central area and taxonomic problems (Marino and Flores, 2002a) may lead to misinterpretation of the stratigraphic range of these species. At Site 748, the HO of C. solitus (149.45 mbsf, 38.08 Ma) clearly occurs above the LO of C. oamaruensis (155.01 mbsf, 38.80 Ma) (Fig. 3 and Table 1). One possible reason for this discrepancy may be linked to the difference in the quantitative methods used in the different studies. In the present study, a detailed quantitative analysis reveals the presence of an interval containing rare specimens of C. oamaruensis below its LCO; consequently, the LO of this taxon is placed at the base of this interval of rare occurrence. Additionally, rare specimens of C. solitus are also consistently noted above the HCO of this taxon. As such, the abundance patterns of these two species (Fig. 3) clearly show that the LCO of C. oamaruensis (149.25 mbsf) and HCO of C. solitus (151.35 mbsf) likely correspond to the events originally identified in the previous biozonations. The ages of these events at Site 748 are calibrated at 38.06 Ma and 38.34 Ma, respectively (Table 1). Therefore, considering the HCO of C. solitus and LCO of C. oamaruensis at Site 748 as the bottom and the top of the Discoaster saipanensis Zone respectively, we can recognize this biozone, although it is very reduced in thickness (Fig. 2) compared to previous studies (e.g. Wei and Thierstein, 1991). In addition, the zonal boundary (top of the Discoaster saipanensis Zone) is located below the middle/late Eocene boundary; this situation has been demonstrated also at Sites 690 and 689 (Florindo and Roberts, 2005) and Site 738 (unpublished data). (6) LO and LCO of Isthmolithus recurvus. At Sites 689 and 744, Persico and Villa (2004) detected the LO of I. recurvus at 36.10 Ma and at 35.80 Ma, respectively, above the HO of R. reticulata, in general agreement with previous age assignments and stratigraphic relationships for these bioevents (Table 2). In contrast to these studies, we detected rare, but incontestable specimens of I. recurvus in Hole 748B at 148.25 mbsf (Fig. 4), below the HO of R. reticulata (at 128.35 mbsf, with an age of 35.92 Ma). The LO of I. recurvus at Site 748 has an approximate age of 37.98 Ma (Table 1). A similar early occurrence was indicated in Chron C17n by Backman (1987) at Site 523 in the Southern Atlantic Ocean, but it was interpreted as downhole contamination. Recent data on Eocene Alpine Italian sections provide a similar result (Rio D., pers. com., 2006). This early occurrence in such different latitudinal and oceanographic settings cannot be explained by means of contamination at both sites. The LCO of I. recurvus occurs at 127.48 mbsf in Hole 748B and is used here to mark the base of the I. recurvus Zone. If the LCO is time-trangressive, the base of the biozone could be diachronous at different latitudes. The age of this event at Site 748 has been estimated at 35.77 Ma (Table 1). These observations underline the importance of high-resolution quantitative analyses.

179

(7) LO of R. bisecta. The LO of R. bisecta occurs at 143.90 mbsf in Hole 748B (Fig. 4) with an age of 37.59 Ma (Table 1). In accordance with the original description ((Hay et al., 1966) Roth, 1970), we only include in this taxon specimens b10 μm, distinguishing it from R. stavensis (N10 μm). Thus, it is difficult to compare our data with the LO reported by Jovane et al. (2007) at 38 Ma, which includes specimens of R. bisecta N10 μm ( = R. stavensis). The age assignments for this event reported by the different authors are given in Table 2. (8) HO of Neococcolithes dubius. The HO of N. dubius occurs in the Hole 748B section at 134.34 mbsf (Fig. 4), with a calibrated age of 36.70 Ma (Table 1). We consider this datum as a possible additional bioevent within the late Eocene, in general agreement with the results of Marino and Flores (2002b), Wei and Wise (1990a) and Madile and Monechi (1991) who have reported this event within CP 15a Zone. (9) LO and HO of R. oamaruensis. The LO (125.20 mbsf) and the HO (115.86 mbsf) of R. oamaruensis in Hole 748B (Fig. 4) are important bioevents in the biozonation used for the Southern Ocean, due to low abundance or complete absence of discoasters in the late Eocene at all Southern Ocean sites. The LO of R. oamaruensis is calibrated at 35.54 Ma (Table 1). The HO of R. oamaruensis is used to identify the Eocene/Oligocene boundary, and usually occurs at the top of Chron C13r (Table 2). Although the 748B section is most likely condensed within an interval containing ice-rafted debris and the magnetostratigraphy is unreliable below C13n (Roberts et al., 2003), an approximate age of 33.97 Ma is assigned to the HO of R. oamaruensis at this site (Table 1). (10) HCO of R. umbilicus. The HCO of R. umbilicus occurs at 105.94 mbsf (Fig. 4), calibrated at 31.51 Ma, which is comparable with previous age assignments for this bioevent (Table 2). (11) HO of C. altus. This bioevent has been matter of discussion between several authors, with reported ages ranging from the middle Eocene (see Firth and Wise, 1992) to the early Oligocene (Perch-Nielsen, 1985; de Kaenel and Villa, 1996). In Hole 748B, we identify the LO of C. altus at 112.95 mbsf within the lowermost Oligocene, and the age of this bioevent is calibrated at 33.31 Ma. We believe that the different proposed ages for the LO of C. altus may be due to the occurrence of similar species that could be confused with C. altus. In fact, in a restricted upper Eocene interval of Hole 748B (from 155.81 to 130.35 mbsf; Fig. 4), we detected specimens with morphological features similar to C. altus, marked here as cf., but distinct enough to suggest that this form represents a species different from C. altus. Therefore, a detailed study on the biometry of the C. solitus– oamaruensis–altus group was undertaken to determine the real stratigraphic distribution of C. altus and a possible phylogenesis of C. altus from C. oamaruensis, C. solitus, or C. expansus (Persico and Villa, 2008). 4. Paleoecology Although there is uncertainty in assigning environmental preferences to extinct nannofossil taxa, there is a general

180 G. Villa et al. / Marine Micropaleontology 69 (2008) 173–192 Fig. 4. Abundance patterns of selected nannofossil species at ODP Hole 748B, expressed as number of specimens per mm2, are plotted against depth, chronostratigraphy, and magnetostratigraphy (Roberts et al., 2003) correlated to the geochronological time scale (Pälike et al., 2006). Biostratigraphic events are indicated with arrows. The grey area indicates an interval with no biostratigraphic data.

G. Villa et al. / Marine Micropaleontology 69 (2008) 173–192

consensus that some nannofossil species can be interpreted as reflecting distinct paleoecological conditions, thus being indicative of climatic and oceanographic changes. For the Paleogene, numerous papers have focused on calcareous nannofossil paleoecology, in particular across the Paleocene– Eocene boundary and within the earliest Oligocene interval. For the Paleocene–Eocene boundary interval, several authors have proposed a link between selected nannofossil species abundance and oceanographic changes occurred during the PETM (e.g. Bralower, 2002; Kahn and Aubry, 2004; Tremolada and Bralower, 2004; Gibbs et al., 2006; Agnini et al., 2006; Jiang and Wise, 2006, 2007). The earliest Oligocene interval, which is interpreted as a time of widespread glaciation of Antarctica, is another crucial time for climatic changes and has been investigated for the response of nannofossil assemblage variations (Wei and Wise, 1990a; Persico and Villa, 2004), resulting in a profound change at high latitudes from temperate to cool dominated assemblages. In this work, we consider an extended middle Eocene to late Oligocene interval from ~42 to 26 Ma at Site 748. Our approach is to consider paleoecological preferences of the recorded species according to literature (Table 3) and then directly compare the nannofossil abundance patterns with fine-fraction stable isotope records (Figs. 5–7). The main nannofossil abundance variations are assumed to represent a paleoecological response and an adaptation to paleoclimatic and/or trophic changes. In the following discussion, a brief explanation of the most paleoecologically-indicative taxa is given, and a summary overview of previous paleoecological assignments is provided in Table 3. 4.1. Discoaster Discoaster spp. are usually considered warm-water taxa and adapted to oligotrophic conditions (Table 3). In Hole 748B, the discoaster HCO falls in the sample immediately above the minimum δ18O values within the middle Eocene section (Fig. 5), considered the interval of maximum temperature during the MECO event. This event is interpreted as the warmest interval of the entire late middle to late Eocene interval (Bohaty and Zachos, 2003). Following this event, the δ18O record shows a general tendency toward positive values, indicating a long-term cooling trend persisting up to the early late Eocene (Fig. 5). It is noteworthy that at Site 748 the discoaster group disappears in the middle Eocene (40.08 Ma), as previously reported in other Southern Ocean sites (Wei and Wise, 1990a, Wei et al., 1992; Arney and Wise, 2003). This stratigraphic distribution is also confirmed at Site 738 (unpublished data), where the discoaster HCO is detected (at 94.00 mbsf) close to the peak of the MECO event, i.e. much earlier than the extinction of rosette-shaped discoasters that occurs near the E/O boundary (~ 34 Ma) in lower latitude sections (Miller et al., 2008; Pearson et al., 2008). This disappearance may have been influenced by a SST cooling trend, as indicated by the δ18O data, and thus represents the biogeographic exclusion of rosette-shaped discoasters from Southern Ocean following the MECO event. In support of this hypothesis, discoasters were not detected in the late Eocene interval at Sites 689 and 744 (Persico and Villa, 2004). On the

181

other hand, discoasters are also thought to have been influenced by the nutrient regime of surface-waters, preferring oligotrophic conditions (Table 3). In fact, they are more abundant from 180 to 172 mbsf, where fine-fraction δ13C values are low (Fig. 5), possibly suggesting a link to long-term change in nutrient conditions through the late middle Eocene interval. Above this level, discoasters decrease as δ13C values increase (Fig. 5), which suggests that increased fine-fraction δ13C values may be indicative of augmented nutrient availability in surface waters. However, although nutrient conditions may have played a role in discoaster abundance in the middle Eocene, we believe that the progressive decrease in SSTs in the Southern Ocean during this interval was most likely the primary factor in the local disappearance of the discoaster group at Southern Ocean sites in the late middle Eocene. 4.2. Sphenolithus moriformis Sphenolithus moriformis is the only representative of the genus Sphenolithus observed in the nannofossil assemblage record from Hole 748B. Its abundance is higher during the MECO interval, and it broadly mirrors the long-term profile of the fine-fraction δ18O curve, decreasing toward the Eocene– Oligocene boundary and being nearly absent through most of the Oligocene following the Oi-1 event (Fig. 5). It re-occurs again in the late Oligocene, during a time interval interpreted as characterized by warmer surface waters (Villa and Persico, 2006; Pekar et al., 2006; Zachos et al., 2001). Sphenolithus is considered to be an indicator of oligotrophic, warm-water conditions (Aubry, 1998; Bralower, 2002; Gibbs et al., 2004). Gibbs et al. (2006) and Agnini et al. (2006) infer a major nutrient control over temperature during the PETM, and conclude that paleofertility is the primary factor controlling the distribution and abundance of this taxon. Therefore, the longterm decline of Sphenolithus at Site 748 may indicate a change from oligotrophic to eutrophic or mesotrophic conditions through the late Eocene to the early Oligocene interval. 4.3. Ericsonia formosa The genus Ericsonia is thought to have thrived in warmwaters in the Paleogene (Haq and Lohmann, 1976; Wei and Wise, 1990a; Aubry, 1992b; Kelly et al., 1996; Bralower, 2002). Agnini et al. (2006), studying the Venetian Pre-Alps, recognized an acme of Ericsonia during the PETM, and consider it as a warm and eutrophic taxon. In Hole 748B, E. formosa is abundant in the middle Eocene interval from 180 to 160 mbsf prior to and during the MECO interval (Fig. 5). Therefore, we also consider this species as a warm-water indicator. As discussed above for the discoaster HCO, the HCO of E. formosa in the middle Eocene (Fig. 3) can be considered an ecological bioevent in the Southern Ocean, influenced by a gradual SST decrease that occurred at the end of the middle Eocene (~ 39 to 37 Ma). 4.4. Neococcolithes dubius Neococcolithes dubius is here considered a temperate-water indicator (Fig. 6). This taxon is abundant within the MECO interval in Hole 748B but also persists with high abundance after the MECO interval. This stratigraphic distribution may

182

Table 3 Published paleoecological preferences of selected nannofossil species Species

Authors Bukry (1973)

Aubry (1992a,b)

Firth and Wise (1992)

Wei et al. (1992)

Chiasmolithus Temperatespp. cool

Cool-cold

Eutrophic cold

Cold

Cool

E. formosa

Warm

C. pelagicus C. floridanus

Warmtemperate

Warm (Large Ch.), cold (Small Ch.)

Warm

Warm oligotrophic

Monechi et al. (2000)

Bralower (2002)

Eutrophic (C. oamaruensis)

Cool eutrophic

Warm

Warm oligotrophic

Kahn and Aubry (2004)

Eutrophic temperatecold Oligotrophic warm

I. recurvus R. bisecta

Warmtemperate

R. daviesi R. oamaruensis R. reticulata Warmtemperate R. samodurovi

Cold

Temperate

R. umbilicus

Temperate

Not warm

S. moriformis Oligotrophic

Warm mesotrophic Oligotrophic– mesotrophic warm Oligotrophic– mesotrophic warm Warm

Warm oligotrophic

Oligotrophic

Tremolada Agnini and Bralower et al. (2004) (2006)

Cool

Cool eutrophic

Gibbs et al. (2006)

Warm oligotrophic

Villa and Persico (2006)

Present work

Cool

Cool

Warm oligotrophic

Temperate No-temp affinity

Eutrophic

Warm

Persico and Villa (2004)

Warm, probably oligotrophic

Warm Temperatecool

Discoaster spp.

Z. bijugatus

Warm

Kelly et al. (1996)

Warm oligotrophic

Temperate

Warm, probably oligotrophic

Warm oligotrophic

Temperate

Warm oligotrophic

Cool Temperate

Cool Warm

Temperate Temperate

Temperate

Temperate

Cool

Cool Cool

Cool

Cool

Cool

Cool

Oligotrophic

Cool (R. hillae) Warm

Warm Near-shore

Warm oligotrophic Warm oligotrophic

Temperate

Temperate

Temperate

Temperate

Warm Warm oligotrophic

Cool eutrophic

Warm oligotrophic Oligotrophic

Warm oligotrophic Warm

Warm

Warm oligotrophic Temperate eutrophic

G. Villa et al. / Marine Micropaleontology 69 (2008) 173–192

Wei and Wise Wei and (1990a,b) Thierstein (1991)

G. Villa et al. / Marine Micropaleontology 69 (2008) 173–192 Fig. 5. Percentage distribution of warm-water taxa, δ18O and δ13C data are plotted against depth, magnetostratigraphy (Roberts et al., 2003) and chronostratigraphy and correlated to the geochronological time scale (Pälike et al., 2006). The grey area indicates an interval with no biostratigraphic data.

183

184 G. Villa et al. / Marine Micropaleontology 69 (2008) 173–192 Fig. 6. Percentage distribution of temperate-water taxa, δ18O and δ13C data are plotted against depth, magnetostratigraphy (Roberts et al., 2003) and chronostratigraphy and correlated to the geochronological time scale (Pälike et al., 2006). The grey area indicates an interval with no biostratigraphic data.

G. Villa et al. / Marine Micropaleontology 69 (2008) 173–192 Fig. 7. Percentage distribution of cool-water taxa (R. daviesi and Chiasmolithus spp.) plotted against depth, δ18O record (left). R. reticulata plotted against δ13C record (right). On the left: chronostratigraphy and magnetostratigraphy (Roberts et al., 2003) of ODP Hole 748B correlated to the geochronological time scale (Pälike et al., 2006).The grey area indicates an interval with no biostratigraphic data.

185

186

G. Villa et al. / Marine Micropaleontology 69 (2008) 173–192

Fig. 8. Distribution of temperate-water taxa, cool-water taxa and warm-water taxa plotted against depth, chronostratigraphy, magnetostratigraphy (Roberts et al., 2003) correlated to the geochronological time scale (Pälike et al., 2006) and fine-fraction δ18O record, from ODP Hole 748B.

suggest that SST cooling following the MECO was insufficient to exclude this taxon from Southern Ocean waters until ~ 37 Ma, when a stronger cooling occurred (see Section 5.5). The abundance distribution of N. dubius in Hole 748B may also reflect a nutrient control on its occurrence. This idea is supported by the generally high values observed in the δ13C curve in the interval corresponding to the higher abundance of N. dubius (174–152 mbsf). 4.5. Zygrablithus bijugatus There is currently disagreement about the main factors controlling Z. bijugatus abundance during the Paleogene. This taxon has been interpreted as both a warm (Bralower, 2002; Kahn and Aubry, 2004) and cool-water taxon (Tremolada and Bralower, 2004), thriving in shallow (Monechi et al., 2000) or deep photic habitats (Aubry, 1998; Bralower, 2002; Stoll et al., 2007), with preference for eutrophic (Tremolada and Bralower, 2004) or oligotrophic conditions (Aubry, 1998; Bralower, 2002; Agnini et al., 2006; Gibbs et al., 2006). In the Hole 748B section, this taxon shows a general decrease in abundance from the base

of the studied interval up to the E/O boundary (Fig. 6), suggesting a temperate-water preference for this species. 4.6. Reticulofenestra umbilicus group This group, which includes R. umbilicus and R. samodurovi, has been associated with temperate-water conditions in previous studies (Table 3). This ecological preference is generally supported by the observed decline in abundance through the Eocene–Oligocene boundary interval at Site 748 (Fig. 6). A peak in abundance is observed at about 120 mbsf (34.8 Ma) during the late Eocene, when alternating rapid SST fluctuations have been recognized at Sites 689 and 744 (Persico and Villa, 2004). 4.7. Coccolithus pelagicus group This group includes both Coccolithus pelagicus and Coccolithus eopelagicus and shows a marked decrease in abundance at the Eocene–Oligocene boundary. We consider this group having a temperate-water preference in the Paleogene of the Southern Ocean (Fig. 6), as previously suggested by other

G. Villa et al. / Marine Micropaleontology 69 (2008) 173–192

187

authors (Table 3). This ecological association stands in contrast to the cold-temperature preference of the living specimens (Ziveri et al., 2004). Haq and Lohmann (1976) infer an evolution in its ecological preference through time to explain this discrepancy.

gical group in this study. Previous studies have interpreted a temperature dependence for this taxon, with a preference towards cool (Wei et al., 1992; Monechi et al., 2000) or temperate waters (Persico and Villa, 2004; Villa and Persico, 2006).

4.8. Reticulofenestra bisecta group

5. Discussion

This group, which includes Reticulofenestra bisecta and Reticulofenestra stavensis, decreases in abundance through the Eocene–Oligocene boundary interval and increases again in the late Oligocene. As such, it is regarded as a temperate-water group (Fig. 6; Table 3). An interval of increased abundance is also recorded between ~ 137 and 130 mbsf (~ 36.9 to 36.1 Ma) in the upper Eocene section of Hole 784B, which corresponds to an interval of lower δ18O values, thus suggesting an interval of warmer SSTs.

Based on the Hole 748B results discussed above, it is evident that calcareous nannoplankton experienced significant assemblage fluctuations in response to climatic variations that occurred during the middle Eocene to late Oligocene in the Indian sector of the Southern Ocean. In order to further assess paleoceanographic variability through this time interval, we have subdivided the nannofossil taxa into major paleoecological groups. Taxa with similar inferred SST preferences have been grouped, thus obtaining warm-water taxa (Discoaster spp., E. formosa, S. moriformis), temperate-water taxa (R. umbilicus group, C. pelagicus group, R. bisecta group, N. dubius, Z. bijugatus), and cool-water taxa (R. daviesi, Chiasmolithus spp.) curves (Fig. 8). A curve based on the abundance of R. reticulata is also plotted to indicate possible trophic variations, where increased abundance of this taxon may correspond to lower nutrient levels (Fig. 7). Additionally, a temperate-warm-water taxa index (Twwt) calculated as [(temperate + warm) / (temperate + warm + cool)] ⁎ 100 is plotted against the isotopic curves (Fig. 9). During the middle Eocene to the late Oligocene interval, several paleoclimatic events have been previously identified within both paleontological and geochemical data sets, including the MECO warming event at ~ 40 Ma (Bohaty and Zachos, 2003; Jovane et al., 2007), the late Eocene warming interval at ~ 36 Ma (Bohaty and Zachos, 2003), the late Eocene cooling event at ~35 Ma (Vonhof et al., 2000; Bohaty and Zachos, 2003), the Oi-1 event at ~ 34 Ma (Miller et al., 1991; Zachos et al., 1996; Coxall et al., 2005; Wei et al., 1992; Aubry, 1992b; Persico and Villa, 2004), and a warming episode in the late Oligocene at ~ 26 Ma (Miller et al., 1987; Zachos et al., 2001; Villa and Persico, 2006; Pekar et al., 2006). The long record of nannofossil data collected at high resolution from Hole 748B allows us to examine each of these events in detail, as well as infer broad climatic trends. Overall, the nannofossil record shows a gradual stepwise cooling trend along the studied time interval. This trend, however, is punctuated by several important short-term climatic warming events. Starting from the base of the section we describe below these events in detail (Fig. 9). In addition to the shortterm warming episodes, the prominent cooling events are also highlighted and labelled with letters A, B, and C.

4.9. Reticulofenestra daviesi group The R. daviesi group includes Reticulofenestra spp. between 5 and 8 µm and has been considered a cool-water taxon in all previous studies (Table 3). In Hole 748B, the abundance curve of R. daviesi is noticeably in phase and inversely correlated with the fine-fraction δ18O record (Fig. 7), confirming the previous paleoecological preference assignment. A significant increase in abundance of this species over a short interval near the E/O boundary indicates a direct response to the initiation of the Oi-1 event, as observed by Persico and Villa (2004). Above this level, R. daviesi becomes a major component of the assemblage throughout most of the Oligocene interval (Fig. 7). 4.10. Chiasmolithus spp. The Chiasmolithus spp. group (C. altus, C. oamaruensis, C. solitus, C. expansus, Chiasmolithus sp.) abundance curve shows a gradual increase from the late middle Eocene to the late Oligocene, with a clear decrease during the MECO event and an increase in correspondence to the Oi-1 event. Therefore, we consider this group as indicative of cool-water conditions (Fig. 7), in agreement with previous ecological assignments (Table 3). 4.11. Reticulofenestra reticulata Reticulofenestra reticulata has been considered as dependent on both surface-water temperature and fertility conditions (Table 3). The abundance curve of R. reticulata obtained in this study from Hole 748B shows an increase between 160 and 135 mbsf (Fig. 7), which roughly corresponds to an interval of decreased fine-fraction δ13C values, suggesting an oligotrophic preference for this species. Wei et al. (1992), based on the comparison of the distribution of this species between high and low latitudes, attribute a preference for cool waters, as it is more abundant at high-latitude sites. 4.12. Isthmolithus recurvus In Hole 748B, I. recurvus is rare and its distribution does not allow any paleoecological consideration. Therefore, I. recurvus has not been included in any specific paleoecolo-

5.1. Eocene cooling event A From 41.6 to 41.3 Ma (174–172 mbsf), prior to the MECO event and immediately before a short hiatus, the Twwt index suggests a sharp decrease in surface-water temperature. This brief cooling event is also recognized in the fine-fraction δ18O record but only the initial phase of the cooling episode is observed because the section is truncated by a hiatus at 171.16 mbsf. Tripati et al. (2005) recognized a cooling episode at this time and attributed it to an early glacial phase in the late middle Eocene, which has been considered synchronous in

188

G. Villa et al. / Marine Micropaleontology 69 (2008) 173–192

both hemispheres by Tripati et al. (2008). Edgar et al. (2007) also recognized, at ~41.6 Ma, a strong positive shift in δ18O isotope curve from Site ODP 1260 (Demerara Rise), but they suggest, however, that most of this signal is linked to a cooling of surface and bottom waters of the Atlantic Ocean, and not to ice growth, thus excluding the presence of large ice sheets in the Northern Hemisphere at this time. Our nannofossil data from Site 748 support this interpretation, indicating that a significant component of the positive δ18O shift is related to cooling of Southern Ocean surface waters. 5.2. The MECO warming event In general, there is evidence of an interval of warmer surface waters from ~40.7 to 39.1 Ma at Site 748, within the late middle Eocene interval that includes the MECO event. The MECO event itself is associated with a higher percentage of temperate-water taxa and an increase of warm-water taxa. Nevertheless, the latter (e.g. discoasters) do not show a positive increase exactly at the peak of the MECO, but just below it. Immediately above the peak of the MECO event, however, a sharp decrease in the abundance of warm-water taxa occurs (Figs. 8 and 9).

The rapidity and magnitude of warming phase during the MECO imply that this event affected the Southern Ocean biological communities. In spite of this, our data show that during the warming peak indicated by the δ18O (i.e. at ~ 40.0 Ma), the temperate-water-taxa are dominant and only minor variations of the other groups are recognizable. In comparison, the nannofossil response to the MECO event is not as marked as the profound turnover described during the PETM (Bralower, 2002; Agnini et al., 2006; Gibbs et al., 2006) or in association with the Oi-1 event (Persico and Villa, 2004). The relative lack of assemblage variation during the peak warming of the MECO event could possibly be a function of the absolute range and magnitude of warming, interpreted from the δ18O record as indicative of ~ 4 °C temperature increase, i.e. from 10° to 14 °C through the entire MECO event (Bohaty and Zachos, 2003). It is possible that the nannofossil taxa present at Site 748 were not sensitive to this range of SST variation, which is outside of the critical temperature range defined between 2 °C and 8 °C (Persico et al., 2006). Although we suggest that nannofossil behavior during the MECO might be the result of SST warming, the possibility that other factors, such as nutrient availability, water mass stratification and/or surface current changes, may have also

Fig. 9. Summary of the main paleoclimatic events evidenced by the Twwt index ([(temperate + warm) / (temperate + warm + cool)]⁎100) and fine-fraction δ18O record, plotted against age (Pälike et al., 2006) and chronostratigraphy. Dark grey bands indicate warmer events, light grey bands show cooler events. On the left, the enlarged area of the interval around the MECO event, showing that calcareous nannofossils register the warming event slightly before the δ18O record, followed by the cooling event B.

G. Villa et al. / Marine Micropaleontology 69 (2008) 173–192

influenced the assemblage variation cannot be ruled out. As we have previously discussed, for example, the decline of Discoaster spp. abundance through the MECO interval could indicate a change from an oligotrophic to a more eutrophic regime through the course of the warming event. 5.2.1. The age of the MECO The age of this event was proposed by Bohaty and Zachos (2003) at 41.5 Ma on the basis of the biochronology derived time scale of the LO of R. reticulata, which has a calibrated age of 42 Ma (Berggren et al., 1995). However, the Southern Ocean sites studied by Bohaty and Zachos (2003) do not have a good magnetostratigraphic record in this interval, so direct or precise calibration of the bioevents is not possible. Recently, Jovane et al. (2007) revised the magnetostratigraphy of the Contessa section (Italy), where they precisely dated a positive shift of the bulk δ13C curve and correlated it to the MECO event, occurring in Chron C18n2n, at about 40 Ma, using the Berggren et al. (1995) time scale. They documented the LO of R. reticulata at a level 5 m below the positive shift in δ13C, in C18r at about 41 Ma; they therefore suggest that the bioevent is diachronous at different latitudes. In Hole 748B, we recorded an analogous stratigraphic position of the LO of R. reticulata to that observed in the Contessa section, occurring 7 m below the positive shift of the δ13C. Assuming an accumulation rate of ~9 m/m.y. in the MECO interval of Site 748, the LO of R. reticulata has an age of ~ 40.7 Ma. We thus suggest that the LO of R. reticulata has similar ages at high latitudes and in the northern mid-latitude Contessa section. This age assignment sheds doubts on the reliability of the magnetostratigraphic signal in the Kerguelen Plateau sites for this time interval, as mentioned by Jovane et al. (2007) and Roberts et al. (2003). Therefore, the original age calibration of the LO of R. reticulata is most likely misinterpreted (Pospichal and Wise, 1990).

189

5.5. Late Eocene warming interval From 36.26 to 36.0 Ma, an increase in the Twwt index corresponds to a decrease in fine-fraction δ18O values (Fig. 9). This event corresponds to the late Eocene warming event interpreted by Bohaty and Zachos (2003). 5.6. Vonhof cooling event This cooling event, which was previously-recognized at Site 689B (Vonhof et al., 2000) and subsequently confirmed by Bohaty and Zachos (2003), is recorded in the nannofossil assemblage record from Hole 748B. The cooling signal recorded by oxygen isotopes between 35.5 and 35.25 Ma corresponds to a sharp decrease in the Twwt index (Fig. 9). 5.7. Latest Eocene climate instability From 35.5 to 34.1 Ma the Twwt index shows evidence of a dynamic and cyclic signal (Fig. 9). During this time interval, the same trend of the Twwt index has been clearly identified at Sites 689 and 744 (Persico and Villa, 2004), suggesting climate instability, with alternations of relatively cool and relatively temperate conditions, preceding the Oi-1 event. Within this same interval, Jovane et al. (2006) documented a cyclic signal in the magnetic susceptibility record from the mid-latitude Massignano section (Italy), interpreted to be related to orbital forcing. Further studies on calcareous nannofossils at Southern Ocean Sites 748, 689 and 744, focusing in detail on this time interval, could possibly confirm a similar astronomical pacing and correlation to the Massignano section. 5.8. Earliest Oligocene cooling and the Oi-1 event

Immediately following the MECO event a cooling episode at ~ 39 Ma is indicated by a drop in the Twwt index (Fig. 9), which marks the end of the late middle Eocene warming phase. This pronounced cooling event detected in the Twwt index, recorded here for the first time, does not correspond to a related increase in fine-fraction δ18O values, which show a more gradual, long-term trend toward positive values. In general, the δ18O record between ~ 39 and 37 Ma indicates gradual cooling, while the Twwt index shows several prominent cycles. The nannofossil assemblage variation in this interval is most likely related to instability in Southern Ocean SST conditions, but it is unclear why there is no direct relationship to changes in the fine-fraction δ18O record.

Just above the Eocene/Oligocene boundary (33.79 Ma) at the base of Chron C13n (33.705 Ma), a remarkable change in the Twwt index marks a sudden and profound cooling event. This dramatic change in nannofossil assemblages occurs exactly at the same level as the rapid positive shift of the δ18O curve, confirming the significance of this paleoclimatic event. The same age was deduced from the Twt index at ODP Site 689 at Maud Rise and Site 744 in Kerguelen Plateau (Persico and Villa, 2004). This shift of the Twwt index is much more evident than at the MECO warming event and is considered to reflect the response to a pronounced decrease in SST below 8°C (within the nannoplankton critical temperature interval) and increased nutrient levels. The augmented nutrient availability in the earliest Oligocene in the Southern Ocean could possibly have triggered an increase in primary productivity, also observable in the total abundance curve (Fig. 3), inducing a decrease in atmospheric CO2, which likely turned out to be a positive feedback for the cooling event.

5.4. Middle/late Eocene cooling event C

5.9. Oligocene

The brief decrease of the Twwt index at ~37 Ma is interpreted as a cooling episode that occurred near the middle/ late Eocene boundary. It coincides with a rapid positive shift in fine-fraction δ18O values (Fig. 9), which represents the peak of a gradual cooling that follows the MECO warming event.

From the has early to the early late Oligocene (~34.0 to 26.3 Ma), nannofossil assemblages at Site 748 indicate cool surface waters, with the coolest conditions in the early late Oligocene interval from ~ 28.5 to 26.3 Ma. Following this cool phase, an increase in temperate-water taxa begins at ~ 26 Ma (73 mbsf) (Fig. 8), indicating a SST warming episode in the

5.3. Eocene cooling event B

190

G. Villa et al. / Marine Micropaleontology 69 (2008) 173–192

Southern Ocean during the late Oligocene. As discussed by Villa and Persico (2006), it is still matter of debate whether the decrease in deep-sea benthic δ18O values in the latest Oligocene (Zachos et al., 2001) is tied to warm saline water masses originating in the Atlantic Ocean and flowing in the whole Southern Ocean (Pekar et al., 2006), or if warming, coupled with the collapse of Antarctic ice sheet, globally affected deep water masses (Zachos et al., 2001). The nannofossil assemblage records from both Sites 689 and 748 indirectly contribute to this debate and suggest that the surface waters in the southern high latitudes warmed significantly during the latest Oligocene. This observation may further suggest that at least some component of the observed decrease in benthic δ18O values may in fact be due to deep-sea warming. 6. Conclusions In this study, detailed quantitative nannofossil assemblage analysis of ODP Hole 748B sediments has enabled an improved Southern Ocean biostratigraphic scheme for the middle Eocene to late Oligocene interval. The ages of previouslyrecognized events are refined, and several additional bioevents are identified. These calibrations will be useful for further chronostratigraphic correlations between different sites in the circum-Antarctic region. Within the improved biostratigraphic framework, we have assigned the nannofossil species to different paleoecological groups, with respect to the SST and nutrient availability, deduced also by comparison with the δ18O and δ13C data, and created a temperate-warm-water taxa (Twwt) index. Variations in the Twwt index through time are inferred to be primarily indicative of SST variations in response to climate changes. The fine-fraction δ18O and the nannofossil assemblage records give a similar picture for the long-term climatic trends from the middle Eocene to the late Oligocene, allowing us to recognize five cooling events within this time interval (Fig. 9). Among them, the cooling event at about 39 Ma, indicated as cooling event B, is reported here for the first time. The MECO event at 40 Ma represents the last major warming event of the Eocene. The termination of the MECO event is associated with the regional exclusion of rosette-shaped discoasters from the Southern Ocean, most likely due to rapid cooling following the warming event. In the long interval following the MECO event, we register a continuous progressive cooling trend, which intensified at 37 Ma (event C) and at 35 Ma (Vonhof event). Brief interruptions in this overall cooling trend are noted during two intervals of SST instability in the late middle Eocene, and in the latest Eocene, respectively. Further investigation of the apparent cyclic signal in the latest Eocene interval, as previously suggested by Jovane et al. (2006) for the Massignano section, is recommendable. In the early Oligocene, changes in calcareous nannofossil assemblages are closely associated with Oi-1 event recorded in the δ18O records, indicating cooling of Southern Ocean surface waters in conjunction with growth of the East Antarctic ice sheet. This event is followed by a general indication of cool surface waters throughout the Oligocene from nannofossil assemblages. This long-term cooling pattern ends abruptly during a distinct late Oligocene warming phase from 26.5 Ma to the top of the section, con-

sidered to fall within the latest Oligocene (Villa and Persico, 2006). In this work, we confirm that calcareous nannofossils are a valuable tool for paleoclimatic reconstructions in highlatitude settings during the Paleogene. At Site 748, nannofossil assemblage variation, interpreted from a paleoecological perspective, shows similar long-term climatic trends as the fine-fraction δ18O data within the middle Eocene to late Oligocene interval. There are, however, some inconsistencies in the finer-scale details, particularly with regard to the peak of the MECO event and the cooling event B, in which the two records seem to present a slightly different behaviour. The reason for these discrepancies may be in part related to an enhanced response of nannoplankton to changes in SST, nutrient conditions or stratification. Alternatively, the temperature signal deduced from the δ18O record is obscured by ice-volume or local salinity changes. Numerous short-term fluctuations in the Twwt index within the cooling events at ~ 37 and ~ 35 Ma are also not evident in the fine-fraction δ18O record (Fig. 9). These intervals of surface-water instability interpreted from the nannofossil assemblage data may represent a finer response of a more complex picture that is not clearly identified in the stable isotope records alone. These events should be investigated in future high-resolution work at other sites. Acknowledgements We are grateful to A. Roberts for kindly making available the samples from u-channels at the National Oceanography Centre, Southampton, UK. Careful reviews by S.W. Wise and an anonymous reviewer helped to improve the manuscript. This research was supported by COFIN 2005 to I. Premoli Silva and NSF Polar Programs grant OPP-0338337 to J.C. Zachos and M.L. Delaney. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.marmicro.2008.07.006. References Agnini, C., Fornaciari, E., Rio, D., Tateo, F., Backman, J., Giusberti, L., 2006. Responses of calcareous nannofossil assemblages, mineralogy and geochemistry to the environmental perturbations across the Paleocene/ Eocene boundary in the Venetian Pre-Alps. Mar. Micropaleontol. 63, 19–38. doi:10.1016/j.marmicro.2006.10.002. Arney, J.E., Wise Jr., S.W., 2003. Paleocene–Eocene nannofossil biostratigraphy of ODP Leg 183, Kerguelen Plateau. In: Frey, F.A., Coffin, M.F., Wallace, P.J., Quilty, P.G. (Eds.), Proc. ODP. Sci. Results, vol. 183, pp. 1–59 ([Online]. Available from World Wide Web: bhttp://www-odp.tamu.edu/publications/ 183_SR/VOLUME/CHAPTERS/014.PDF). Aubry, M.P., 1992a. Paleogene Calcareous nannofossils from the Kerguesen Plateau, Leg 120. In: Wise Jr., S.W., Schlich, R., et al. (Eds.), Proc. ODP. Sci. Results, vol. 120, pp. 471–491. Aubry, M.P., 1992b. Late Paleogene nannoplankton evolution: a tale of climatic deterioration. In: Prothero, D.R., Berggren, W.A. (Eds.), Eocene–Oligocene Climatic and Biotic Evolution. Princeton Univ. Press, Princeton, pp. 272–309. Aubry, M.P., 1998. Early Paleogene calcareous nannoplankton evolution: a tale of climatic amelioration. In: Aubry, M.-P., Lucas, S., Berggren, W.A. (Eds.), Late Paleocene and Early Eocene Climatic and Biotic Evolution. Columbia Univ. Press, New York, pp. 158–203. Backman, J., 1987. Quantitative Calcareous nannofossils biochronology of Middle Eocene through Early Oligocene sediment from DSDP sites 522 and 523. Abh. Geol. Bundesanst.-A 39, 21–31.

G. Villa et al. / Marine Micropaleontology 69 (2008) 173–192 Barrera, E., Huber, B.T., 1993. Eocene and Oligocene oceanography and temperatures in the Antarctic Indian Ocean. In: Kennett, J.P., Warnke, D.A. (Eds.), The Antarctic Paleoenvironment: a Perspective on Global Change. American Geophysical Union Antarctic Research Series, vol. 60, pp. 49–65. Berggren, W.A., Kent, D.V., Swisher III, C.C., Aubry, M.-P., 1995. A revised Cenozoic geochronology and chronostratigraphy. In: Berggren, W.A., Kent, D.V., Aubry, M.-P., Hardenbol, J. (Eds.), Geochronology, Time Scales and Global Stratigraphic Correlation. Spec. Publ.-SEPM (Soc. Sediment. Geol.), vol. 54, pp. 129–212. Bohaty, S.M., Zachos, J.C., 2003. A significant Southern Ocean warming event in the late middle Eocene. Geology 31, 1017–1020. Bralower, T.J., 2002. Evidence of surface water oligotrophy during the Paleocene–Eocene thermal maximum: nannofossil assemblage data from Ocean Drilling Program Site 690, Maud Rise, Weddel Sea. Paleoceanography 17, 13.1–13.13. Bukry, D., 1973. Low-latitude coccolith biostratigraphic zonation. In: Edgar, N.T., Saunders, J.B. (Eds.), Initial Reports of the Deep Sea Drilling Project, vol. 15, pp. 685–704. Coxall, H.K., Wilson, P.A., Pälike, H., Lear, C.H., Backman, J., 2005. Rapid stepwise onset of Antarctic glaciation and deeper calcite compensation in the Pacific Ocean. Nature 433, 53–57. doi:10.1038/nature 03135. de Kaenel, E., Villa, G., 1996. Oligocene/Miocene calcareous nannofossil biostratigraphy and paleoecology from the Iberia Abyssal Plain, Northeastern Atlantic. In: Whitmarsh, R.B., Sawyer, D.S., Klaus, A., Masson, D.G. (Eds.), Proc. ODP Sci. Results, vol. 149, pp. 79–145. Diester-Haass, L., Zahn, R., 1996. Eocene–Oligocene transition in the Southern Ocean: history of water mass circulation and biological productivity. Geology 24, 163–166. Edgar, K.M., Wilson, P.A., Sexton, P.F., Suganuma, Y., 2007. No extreme bipolar glaciation during the main Eocene calcite compensation shift. Nature 448, 908–911. doi:10.1038/nature06053. Firth, J.V., Wise Jr., S.W., 1992. A preliminary study of the evolution of Chiasmolithus in the middle Eocene to Oligocene of Sites 647 and 748. In: Wise Jr., S.W., Schlich, R., et al. (Eds.), Proc. ODP. Sci. Results, vol. 120, pp. 493–508. doi:10.2973.odp.proc.sr.120.157.1992. Florindo, F., Roberts, A.P., 2005. Eocene–Oligocene magnetobiochronology of ODP Sites 689 and 690, Maud Rise, Weddell Sea, Antarctica. Geol. Soc. Am. Bull. 117, 46–66. doi:10.1130/B25541.1. Gibbs, S., Shackleton, N.J., Young, J., 2004. Orbitally forced climate signals in mid-Pliocene nannofossil assemblages. Mar. Micropaleontol. 51, 39–56. doi:10.1016/j.marmicro.2003.09.002. Gibbs, S.J., Bralower, T.J., Bown, P.R., Zachos, J.C., Bybell, L.M., 2006. Shelf and open-ocean calcareous phytoplankton assemblages across the Paleocene–Eocene Thermal Maximum: implications for global productivity gradients. Geology 34, 233–236. doi:10.1130/G22381.1. Hay, H.H., Mohler, H., Wade, M.E., 1966. Calcareous Nannofossils from Nal'chik (Northwest Caucasus). Ecl. Geol. Helv. 59, 379–399. Haq, B.U., Lohmann, G.P., 1976. Early Cenozoic calcareous nannoplankton biogeography of the Atlantic Ocean. Mar. Micropaleontol. 1, 119–194. doi:10.1016/0377-8398(76)90008-6. Jiang, S., Wise, S.W., 2006. Surface-water chemistry and fertility variations in the tropical Atlantic across the Paleocene/Eocene thermal maximum as evidenced by calcareous nannoflora from ODP Leg 207, Hole 1259B. Rev. Micropaleontol. 49 (4), 227–244. doi:10.1016/j.revmic.2006.10.002. Jiang, S., Wise, S.W., 2007. Upper Cenozoic calcareous nannofossil biostratigraphy, sedimentation rates, and paleoceanography of ODP Leg 206, East Pacific Rise. Proceedings of Ocean Drilling Program Scientific Results, vol. 206, pp. 1–25. doi:10.2973/odp.proc.sr.206.008.2007. Jovane, L., Florindo, F., Sprovieri, M., Pälike, H., 2006. Astronomic calibration of the late Eocene/early Oligocene Massignano section (central Italy). Geochem. Geophys. Geosyst. 7, Q07012. doi:10.1029/2005GC001195. Jovane, L., Florindo, F., Coccioni, R., Dinares-Turell, J., Marsili, A., Monechi, S., Roberts, A.P., Sprovieri, M., 2007. The middle Eocene climatic optimum event in the Contessa Highway section, Umbrian Apennines, Italy. Geol. Soc. Amer. Bull. 119 (3), 413–427. doi:10.1130/B25917.1. Kahn, A., Aubry, M.-P., 2004. Provincialism associated with the Paleocene/ Eocene thermal maximum: temporal constraint. Mar. Micropaleontol. 52, 117–131. doi:10.1016/j.marmicro.2004.04.003. Kelly, D.C., Bralower, T.J., Zachos, J., Premoli Silva, I., Thomas, E., 1996. Rapid diversification of planktonic foraminifera in the tropical Pacific (ODP Site 865) during the late Paleocene thermal maximum. Geology 24, 423–426. Kennett, J.P., Stott, L.D., 1991. Proteus and proto-oceanus: ancestral Paleogene oceans as revealed from Antarctic stable isotopic results; ODP 113. In: Barker, P.F., Kennett, J.P., et al. (Eds.), Proc. ODP Sci. Results, vol. 113, pp. 865–880. Lear, C.H., Elderfield, H., Wilson, P.A., 2000. Cenozoic deep-sea temperatures and global ice volumes from Mg/Ca in benthic foraminiferal calcite. Science 287, 269–272. Madile, M., Monechi, S., 1991. Late Eocene to early Oligocene calcareous nannofossil assemblages from Sites 699 and 703, subantarctic South

191

Atlantic Ocean. In: Ciesielski, P.F., Kristoffersen, Y., Clement, B., Moore, T.C. (Eds.), Proceedings of the Ocean Drilling Program. Scientific Results, vol. 114, pp. 179–192. Marino, M., Flores, J.A., 2002a. Middle Eocene to Early Oligocene calcareous nannofossil stratigraphy at Leg 177 Site 1090. Mar. Micropaleontol. 45, 383–398. Marino, M., Flores, J.A., 2002b. Data report: calcareous nannofossil data from the Eocene to Oligocene, Leg 177, Hole 1090B. In: Gersonde, R., Hodell, D.A., Blum, P. (Eds.), Proceedings of the Ocean Drilling Program. Scientific Results, vol. 177, pp. 1–9. Martini, E., 1971. Standard Tertiary and Quaternary calcareous nannoplankton zonation. In: Farinacci, A. (Ed.), Proc. 2nd Int. Conf. Planktonic Microfossils Roma, vol. 2, pp. 739–785. Mc Gonigal, K., Di Stefano, A., 2002. Calcareous nannofossil biostratigraphy of the Eocene–Oligocene transition, ODP Sites 1123 and 1124. In: Richter, C. (Ed.), Proc. ODP, Sci. Results, vol. 181, pp. 1–22. Miller, K.G., Browning, J.V., Aubry, M.-P., Wade, B.S., Katz, M.E., Kulpecz, A.A., Wright, J.D., 2008. Eocene–Oligocene global climate and sea-level changes: St. Stephens Quarry, Alabama. Geol. Soc. Amer. Bull. 120, 34–53. Miller, K.G., Fairbanks, R.G., Mountain, G.S., 1987. Tertiary oxygen isotope synthesis, sea-level history and continental margin erosion. Paleoceanography 2, 1–19. Miller, K.G., Wright, J.D., Fairbanks, R.G.,1991. Unlocking the ice house: Oligocene– Miocene oxygen isotopes, eustasy, and margin erosion. J. Geophys. Res. 96, 6829–6848. Monechi, S., Buccianti, A., Gardin, S., 2000. Biotic signals from nannoflora across the iridium anomaly in the upper Eocene of the Massignano section: evidence from statistical analysis. Mar. Micropaleontol. 39, 219–237. Müller, C., 1970. Nannoplankton-Zonen der Unteren meeresmolasse Bayerns. Geol. Bavarica 63, 107–118. Okada, H., Bukry, D., 1980. Supplementary modification and introduction of code numbers to the low latitude coccolith biostratigraphic zonation (Bukry 1973, 1975). Mar. Micropaleontol. 5, 321–325. doi:10.1016/03778398(80)90016-X. Pälike, H., Norris, R.D., Herrle, J.O., Wilson, P.A., Coxall, H.K., Lear, C.H., Shackleton, N.J., Tripati, A.K., Wade, B.S., 2006. The heartbeat of the Oligocene climate system. Science 314 (5807), 1894–1898. doi:10.1126/ science.1133822. Pearson, Paul N., McMillan, Ian K., Wade, Bridget S., Jones, Tom Dunkley, Coxall, Helen K., Bown, Paul R., Lear, Caroline H., 2008. Extinction and environmental change across the Eocene–Oligocene boundary in Tanzania. Geology 36, 179–182. Pekar, S.F., DeConto, R.M., Haarwood, D.M., 2006. Resolving a late Oligocene conundrum: deep-sea warming and Antarctic glaciation. Palaeogeogr. Palaeoclimatol. Palaeoecol. 231, 29–40. Perch-Nielsen, K., 1985. Cenozoic calcareous nannofossils. In: Bolli, H.M., Saunders, J.B., Perch-Nielsen, K. (Eds.), Plankton Stratigraphy. Cambridge University Press, pp. 427–554. Persico, D., Villa, G., 2004. Eocene–Oligocene calcareous nannofossils from Maud Rise and Kerguelen Plateau /Antarctica): paleoecological and paleoceanographic implications. Mar. Micropaleontol. 52, 153–179. doi:10.1016/j.marmicro.2004.05.002. Persico, D., Villa, G. 2008. Middle Eocene–Oligocene Chiasmolithus evolution: evidence of phyletic gradualism. J. Nannoplankton Res. 30 (1), 23–33. Persico, D., Villa, G., Fioroni, C., Pea, L., 2006. Response of Eocene–Oligocene to Southern Ocean temperature variations. Abstract 11th International Nannoplankton Association Conference, Lincoln, September, 2006. Pospichal, J.J., Wise Jr., S.W., 1990. Paleocene to middle Eocene calcareous nannofossils of ODP Sites 689 and 690, Maud Rise, Weddell Sea. In: Barker, P.F., Kennet, J.P., et al. (Eds.), Proc. ODP. Sci. Results, vol. 113, pp. 613–638. Raffi, I., Backman, J., Fornaciari, E., Pälike, H., Rio, D., Lourens, L., Hilgen, F., 2006. A review of calcareous nannofossil astrobiochronology encompassing the past 25 million years. Quat. Sci. Rev. 25, 3113–3137. doi:10.1016/j. quascirev.2006.07.007. Roberts, A.P., Bicknell, S.J., Byatt, J., Bohaty, S.M., Florindo, F., Harwood, D.M., 2003. Magnetostratigraphic calibration of Southern Ocean diatom datums from the Eocene–Oligocene of Kerguelen Plateau (Ocean Drilling Program sites 744 and 748). Palaeogeogr. Paleoclimatol. Palaeoecol. 198, 145–168. doi:10.1016/S0031-0182(03)00397-3. Roth, P.H., 1970. Oligocene calcareous nannoplankton biostratigraphy. Eclogae Geol. Helv. 63, 799–881. Stoll, H.M., Shimizu, N., Archer, D., Ziveri, P., 2007. Coccolithophore productivity response to greenhouse event of the Paleocene–Eocene Thermal Maximum. Earth Planet. Sci. Lett 258 (1-2), 192–206. doi:10.1016/j.epsl.2007.03.037. Tremolada, F., Bralower, T.J., 2004. Nannofossil assemblage fluctuations during the Paleocene–Eocene Thermal Maximum at Sites 213 (Indian Ocean) and 401 (North Atlantic Ocean): palaeoceanographic implications. Mar. Micropaleontol. 52, 107–116. doi:10.1016/j.marmicro.2004.04.002.

192

G. Villa et al. / Marine Micropaleontology 69 (2008) 173–192

Tripati, A., Backman, J., Elderfield, H., Ferretti, P., 2005. Eocene bipolar glaciation associated with global carbon cycle changes. Nature 436, 341–346. Tripati, A.K., Eagle, R.A., Morton, A., Dowdeswell, J.A., Atkinson, K.L., Bahé, Y., Dawber, C.F., Khadun, E., Shaw, R.M.H., Shorttle, O., Thanabalasundaram, L., 2008. Evidence for glaciation in the Northern Hemisphere back to 44 Ma from ice-rafted debris in the Greenland Sea. Earth Planet Sci. Lett. 265, 112–122. doi:10.1016/j.epsl.2007.09.045. Villa, G., Persico, D., 2006. Late Oligocene climatic changes: evidence from calcareous nannofossils at Kerguelen Plateau Site 748 (Southern Ocean). Palaeogeogr. Palaeoclimatol. Palaeoecol. 231, 110–119. doi:10.1016/j. palaeo.2005.07.028. Vonhof, H.B., Smit, J., Brinkhuis, H., Montanari, A., Nederbragt, A.J., 2000. Global cooling accelerated by early late Eocene impacts? Geology 28, 687–690. Wei, W., 2004. Opening of the Australia–Antarctica Gateway as dated by nannofossils. Mar. Micropaleontol. 52, 133–152. doi:10.1016/j. marmicro.2004.04.008. Wei, W., Thierstein, H.R., 1991. Upper Cretaceous and Cenozoic calcareous nannofossils of the Kerguelen Plateau (southern Indian Ocean) and Prydz Bay (East Antarctica). In: Barron, J.A., Larsen, B., et al. (Eds.), Proc. ODP. Sci Results, vol. 119, pp. 467–493. Wei, W., Wise Jr., S.W., 1990a. Biogeographic gradients of middle Eocene– Oligocene calcareous nannoplankton in the South Atlantic Ocean. Palaeogeogr. Palaeoclimatol. Palaeoecol. 79, 29–61. doi:10.1016/00310182(90)90104-F.

Wei, W., Wise Jr., S.W., 1990b. Middle Eocene to Pleistocene calcareous nannofossils recovered by Ocean Drilling Program Leg 113 from the Weddell Sea. In: Barker, P.F., Kennet, J.P., et al. (Eds.), Proc. ODP. Sci. Results, vol. 113, pp. 639–666. Wei, W., Wise Jr., S.W., 1992. Eocene–Oligocene calcareous nannofossil magnetobiochronology of the Southern Ocean. Newsl. Stratigr. 26, 119–132. Wei, W., Villa, G., Wise Jr., S.W., 1992. Paleoceanographic implications of Eocene–Oligocene calcareous nannofossils from Sites 711 and 748 in the Indian Ocean. In: Wise Jr., S.W., Schlich, R., et al. (Eds.), Proc. ODP. Sci. Results, vol. 120, pp. 979–999. Wind, F.H., Wise Jr., S.W., 1978. Mesozoic holococcoliths Geology 6, 140–142. Wise Jr., S.W., Schlich, R., et al., 1992. Proc. ODP Sci. Results, vol. 120. Zachos, J.C., Quinn, T.M., Salamy, K.A., 1996. High-resolution (104 years) deepsea foraminiferal stable isotope records of the Eocene–Oligocene climate transition. Paleoceanography 11, 251–256. Zachos, J.C., Pagani, M., Sloan, L., Thomas, E., Billups, K., 2001. Trends, rhythms, and aberrations in global climate 65 Ma to Present. Science 292, 686–693. Ziveri, P., Baumann, K.H., Böckel, B., Bollmann, J., Young, J.R., 2004. Biogeography of selected Holocene coccoliths in the Atlantic Ocean. In: Thierstein, H.R., Young, J.R. (Eds.), Coccolithophores: from Molecular Processes to Global Impact. Springer, pp. 403–428.