SEARCH

SEARCH BY CITATION

Keywords:

  • Abrupt climate change;
  • decadal–centennial time-scales;
  • deep sea;
  • deepwater circulation;
  • Foraminifera;
  • macroecology;
  • Ostracoda;
  • palaeoecology;
  • species diversity;
  • temperature

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results and Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Aim

Little is known about how marine biodiversity responds to oceanographic and climatic changes over the decadal to centennial time-scales which are most relevant for predicted climate changes due to greenhouse gas forcing. This paper aims to reveal decadal–centennial scale deep-sea biodiversity dynamics for the last 20,000 years and then explore potential environmental drivers.

Location

The North Atlantic Ocean.

Methods

We investigated deep-sea benthic microfossil records to reveal biodiversity dynamics and subsequently applied comprehensive ecological modelling to test possible environmental factors (i.e. surface productivity, seasonality of productivity or deepwater circulation related to bottom-water temperature) that may have influenced deep-sea biodiversity over these time-scales.

Results

Deep-sea biodiversity changed synchronously with stadial–interstadial climate changes over the last 20,000 years across a large area of the North Atlantic in both ostracod crustaceans and foraminiferan protozoa (in spite of their different dispersal abilities). Species diversity rapidly increased during abrupt stadial events during the last deglacial and the Holocene interglacial periods. These include the well-known Heinrich 1, the Younger Dryas and the 8.2 ka events when the strength of Atlantic Meridional Overturning Circulation (AMOC) decreased. There is also evidence for quasi-cyclic changes in biodiversity at a c. 1500-year periodicity, consistent with the well-known ‘1500-year climatic cycle’. Statistical analyses revealed that AMOC variability (probably specifically the variability in AMOC-driven bottom-water temperature) is correlated with deep-sea biodiversity.

Main conclusions

Our finding of a significant AMOC–diversity relationship may indicate pervasive control of the diversity of deep-sea benthic species by rapidly changing climate, specifically bottom-water temperature, over decadal to centennial time-scales. Our results, based on highly resolved fossil records, may portend pervasive, synchronous and sudden ecosystem responses to human-induced changes to climate and ocean circulation in this century.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results and Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

There is growing evidence that deep-sea ecosystems respond to global climate change over seasonal to annual (Danovaro et al., 2004; Ruhl et al., 2008) and orbital (i.e. multi-millennial) time-scales (Cronin & Raymo, 1997; Hunt et al., 2005; Yasuhara & Cronin, 2008; Yasuhara et al., 2009, 2012a). However, we know little about changes to deep-sea or other marine ecosystems over decadal and centennial time-scales (Roy et al., 1996; Cannariato et al., 1999; Wollenburg et al., 2007; Yasuhara et al., 2008), scales that are most relevant for predicted climate changes due to anthropogenic greenhouse gas forcing. For example, decadal- to centennial-scale deep-sea benthic species diversity in the North Atlantic varied during the last deglaciation and the Holocene (Yasuhara et al., 2008), but it was unclear whether those biodiversity shifts were correlated with climatic events due to the lack of bottom-water palaeo-oceanographic proxy records in the study region. Potential environmental drivers of deep-sea biodiversity and ecosystem changes at these time-scales are poorly understood.

Since the Last Glacial Maximum 20,000 years ago, Earth's climate has been punctuated by several abrupt climate events including the Younger Dryas (11.7–12.9 ka) and the Heinrich 1 stadial (14.6–17 ka) events (Thornalley et al., 2010). For example, it is known that both the onset and the termination of the Younger Dryas were very rapid (within a 10–20-year period) (Mayewski et al., 1993). These abrupt events affected not only sea surface and atmospheric climate but also deep ocean circulation (Bond et al., 1997; Praetorius et al., 2008). Formation of North Atlantic Deep Water (NADW) in the subpolar North Atlantic, which is an essential component of the Atlantic Meridional Overturning Circulation (AMOC), and thus one of the most important components regulating the global climate system, was possibly shut down and certainly reduced during the Younger Dryas and the Heinrich 1 stadial (Praetorius et al., 2008; Thornalley et al., 2011).

Productivity and temperature-related mechanisms have emerged as likely drivers of large-scale marine diversity patterns (Hunt et al., 2005; Tittensor et al., 2010, 2011; Yasuhara et al., 2012b). In deep-sea biology, surface productivity and the resulting particulate organic carbon flux that serves as the only food source available to the deep-sea benthos (except for chemosynthetic systems such as hydrothermal vents and cold methane seeps) have been considered to be the most important factors affecting benthic biodiversity (Rex et al., 1993; Rex & Etter, 2010; Tittensor et al., 2011; McClain et al., 2012). Recent research has also emphasized the importance of seasonality of surface productivity for deep-sea biodiversity (Corliss et al., 2009), and increasing evidence suggests that bottom-water temperature is an important driver of deep-sea biodiversity (Danovaro et al., 2004; Hunt et al., 2005; Yasuhara et al., 2009). Therefore it is worth investigating potential environmental factors impacting deep-sea biodiversity.

Sediment core EW9302-23GGC (23GGC henceforth, 61.67705° N, 21.738° W, at a water depth of 1695 m; Fig. 1) was recovered from the Björn Drift along the Reykjanes Ridge in the north-eastern Atlantic by the R/V Maurice Ewing operated by the Lamont–Doherty Earth Observatory in 1993, and then stored at the Woods Hole Oceanographic Institution. The 23GGC site is situated directly under the path of the Iceland–Scotland Overflow Waters (Raymo et al., 2004), which are components of the NADW. The 520 cm long sediment core covers the past c. 20,000 years with an average sedimentation rate of c. 26 cm kyr–1. This high sedimentation rate enables us to resolve decadal–centennial patterns of faunal change.

figure

Figure 1. Core locations within the North Atlantic for 23GGC and other sites mentioned in the present study.

Download figure to PowerPoint

Here we document deep-sea benthic species diversity and faunal composition for the last 20,000 years from site 23GGC, taking advantage of rich deep-sea fossil record of a major metazoan benthic group, the ostracod crustaceans. Members of this group occupy a wide range of benthic niches, representative of the benthic community as a whole (Yasuhara et al., 2009). Then, we consider the environmental parameters of surface productivity, seasonality of productivity and deepwater circulation related to bottom-water temperature variability (see Results and Discussion) to determine which factor is better related to deep-sea biodiversity. The 23GGC diversity record is then compared with other published high-resolution microfossil (both ostracod crustacean and foraminiferan protozoa) records in the North Atlantic Ocean.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results and Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Subsampling and ostracod analysis

One-centimetre thick subsamples of core 23GGC were examined at 1-cm intervals, yielding a sampling resolution for each interval of 35–155 years. The >150 μm size fraction was used to determine ostracod faunal diversity and composition. This size fraction is standard for deep-sea ostracod research and allows us to obtain adults and late moult stage juveniles for most species (Yasuhara et al., 2009). The number of specimens refers to valves. This counting method (i.e. counting one valve as one specimen and one articulated carapace as two specimens) has been commonly used in recent studies of North Atlantic deep-sea ostracods (Yasuhara et al., 2008, 2012a, 2009). Although several studies count both one valve and one carapace as one specimen (i.e. the number of specimens refers to valves and carapaces) (Yamaguchi et al., 2014), the difference between these two counting methods is slight here because very few articulated carapaces are usually recovered in Quaternary deep-sea sediments (i.e. most fossil ostracods occur as separate valves). More than 100 species and 13,000 specimens were identified in total.

We used rarefaction E(Sn), the expected number of species in n individuals (Hurlbert, 1971), for species diversity because it is widely used in deep-sea biodiversity studies enabling maximum comparability with published studies. The sampling thresholds, n, were 50 and 100, and both show similar trends (Fig. 2e). These thresholds are standard in deep-sea ecology and palaeoecology (Rex et al., 1993; Danovaro et al., 2004; Yasuhara et al., 2009, 2012a; Tittensor et al., 2011; McClain et al., 2012). Other diversity indices, Shannon H(S), Simpson 1 – D, Fisher's alpha, and raw species richness also show similar results (Fig. S1 in Supporting Information). Non-metric multidimensional scaling (NMDS) based on Bray–Curtis dissimilarity was used to explore ostracod faunal changes. We interpreted the first and second NMDS axes (NMDS1 and NMDS2) as capturing primary variation in faunal composition. The calculation of ostracod species diversity and NMDS was based on three-point moving sums of the census dataset because of relatively small sample size (c. 50 specimens per sample on average), though results were similar if samples were not lumped (results not shown). We performed Wisconsin double standardization and square root transformation of the census count data for NMDS.

figure

Figure 2. Deep-sea species diversity and climatic factors. Relationship between 23GGC ostracod species diversity E(S100) and (a) deepwater circulation proxy of ODP984C benthic δ13C (‰) (P < 0.0001, R2 = 0.2354), (b) warmer deepwater proxy of DS97-2P ‘Atlantic species group’ relative abundance (P = 0.0017, R2 = 0.0553), (c) surface productivity proxy of DS97-2P benthic foraminiferal accumulation rates [BFAR0.64 (1000 n cm−2 kyr–1)] (P = 0.0192, R2 = 0.0387), (d) DS97-2P foraminiferal species richness (P < 0.0001, R2 = 0.1550), and (e) 23GGC ostracod species diversity E(S50) (P < 0.0001, R2 = 0.9496). Linear (a, b, d, e) and second-order polynomial regression (c) lines are shown. (f) Result of spectral analysis for the 23GGC ostracod diversity E(S50) data.

Download figure to PowerPoint

Shallow-water taxa transported by downslope processes were minor constituents of the total assemblage (< 5% for most samples). These taxa, in addition to the non-benthic species Acetabulastoma arcticum (Cronin et al., 2010), were excluded from the faunal analysis.

Carbonate dissolution was not an influence on the results presented here because: (1) the ostracod specimens are well preserved; (2) the 23GGC site is shallower than the calcite lysocline; and (3) it is isolated from the influence of potentially corrosive Antarctic Bottom Water.

Chronology

Age control was established with nine accelerator mass spectrometer (AMS) radiocarbon dates and the Vedde Ash layer (Table S1). Seven radiocarbon ages are from Thornalley et al. (2013). Two additional radiocarbon ages were converted to calibrated calendar ages by using calib 6.0 (Stuiver & Reimer, 1993) based on the Marine09 dataset (Reimer et al., 2009) and a surface radiocarbon reservoir age of 400 years. The age model was based on linear interpolation between these radiocarbon dates and the ash layer. Although high variability in surface 14C reservoir ages has been suggested during the last deglaciation in this region (Waelbroeck et al., 2001; Thornalley et al., 2011), benthic δ18O patterns are fairly consistent among nearby cores (Fig. S2) and thus we considered that the age model used for the present study is reasonable and robust. The 23GGC δ18O data were acquired on a Finnigan MAT 252 mass spectrometer equipped with the ‘Kiel’ automated carbonate device (Table S2).

Environmental parameters

We used proxies to infer environmental conditions of the studied intervals for: (1) deepwater circulation (and related variability in bottom temperature), (2) surface productivity, and (3) seasonality of productivity. Deepwater circulation data are based on the well-established deepwater proxy of benthic foraminiferal δ13C and the warmer water indicator of relative abundance of an ‘Atlantic (benthic foraminiferal) species group’ (Rasmussen et al., 2003; Rasmussen & Thomsen, 2004), and are available for cores ODP984C (61°25′ N, 24°04′ W, 1650 m water depth; Fig. 1) (Praetorius et al., 2008) and DS97-2P (58°56.327′ N, 30°24.590′ W, 1685 m water depth; Fig. 1) (Rasmussen et al., 2003), respectively, which are nearby sites with a similar water depth to 23GGC and are suitable for comparison with the 23GGC ostracod data. Lower δ13C values indicate a weaker AMOC, and higher ‘Atlantic species group’ abundance indicates a higher temperature in the study region at intermediate water depths by enhanced flow of subsurface Atlantic water (see Results and Discussion). The δ13C and ‘Atlantic species group’ results are supported by measurements of Mg/Ca ratios in benthic foraminiferal species from intermediate water depths at other locations (Marcott et al., 2011; Cronin et al., 2012). Published surface productivity and seasonality of productivity records are based on benthic foraminiferal accumulation rates (BFAR; the number of individuals deposited per cm2 of ocean floor per thousand years) and the relative abundance of the ‘phytodetritus group’ of benthic foraminifera (Epistominella exigua and Alabaminella weddellensis), respectively, and are available from the nearby core DS97-2P (Rasmussen et al., 2003). All BFAR values were raised to the power of 0.64 because a calibration study (Herguera, 2000) found that the organic carbon flux to the sea floor was proportional to BFAR0.64. Both BFAR and relative abundance of the ‘phytodetritus group’ are widely used methods to reconstruct past productivity (Herguera, 2000; Yasuhara et al., 2012a) and seasonality of productivity (Sun et al., 2006), respectively.

Environmental proxies were not from the same core as the ostracod faunal data, accordingly we used linear interpolation of the proxy curves to assign environmental values to faunal samples on the basis of age. Revised chronologies were applied to ODP984C and DS97-2P by the following recent papers (Rasmussen & Thomsen, 2004; Thornalley et al., 2011).

Statistical modelling

We used regression models to test factors that might influence deep-sea species diversity and faunal composition, including deepwater circulation (related to bottom temperature), surface productivity and seasonality of productivity. Because hump-shaped diversity–productivity relationships have been reported frequently in previous studies (Tittensor et al., 2011; McClain et al., 2012), we included a quadratic term for surface productivity in the models. The environmental proxies ‘Atlantic species group’ and ‘phytodetritus group’ were log transformed to make their distributions more symmetric, and all environmental parameters were zero-centred for the regression modelling.

The Akaike information criterion corrected for small sample size (AICc) was used to measure model support in a way that balances goodness-of-fit and model complexity, and Akaike weights were used to summarize proportional support for all candidate models (Table 1) (Anderson et al., 2000). We also considered parameter estimates averaged over models, proportional to the support that each model receives (Table 2) (Anderson et al., 2000). This approach accounts for uncertainty in model selection and thus leads to appropriately broader confidence intervals than would be obtained by relying only on the single, best-supported model. The influences of the various predictor variables were measured as relative importance, which is the sum of the Akaike weights of models that include the variable in question (Burnham & Anderson, 2002). We computed pairwise correlations between predictor variables to assess whether multicollinearity was likely to influence the regression results. There is no strong correlation among the variables (all R2 < 0.17), indicating no serious multicollinearity. We accounted for the temporal autocorrelation of model residuals using generalized least squares by following the method of Hunt et al. (2005).

Table 1. Best three regression models of 23GGC species diversity [E(S100)] as a function of deepwater circulation, surface productivity and seasonality of surface productivity
ModelD Coef.P Coef.P2 Coef.SP Coef.R2AICcAW
  1. Datasets used are shown in Table S6.

  2. P, surface productivity; P2, quadratic term of surface productivity; SP, seasonality of productivity; D, deepwater circulation. The table shows the coefficient of each term (D Coef., P Coef., P2 Coef., SP Coef.): R2; the Akaike information criterion corrected for small sample size (AICc); and the Akaike weight (AW). Bold denotes significance at P < 0.05. Overall P is < 0.05 in all of the best three models. Null models, E(S100) ∼ 1, are shown for reference.

Ostracod E(S100) models with benthic δ13C as D       
1−4.9160.18620.70.404
2−5.299−0.5290.20622.70.151
3−5.471−2.1760.21622.70.15
Null624.80.051
Ostracod E(S100) models with % Atlantic species as D       
15.6890.14620.60.319
26.020−1.4490.15621.40.211
35.639−0.0920.14622.40.126
Null624.80.038
Table 2. Model-averaged parameter estimates and confidence intervals of core 23GGC species diversity
TermRICoefficientLower CIUpper CI
  1. P, surface productivity; P2, quadratic term of surface productivity; SP, seasonality of productivity; D, deepwater circulation; RI, relative importance; CI, confidence interval. Bold: CIs that exclude zero.

Ostracod E(S100) models with benthic δ13C as D    
D0.86−3.77−6.65−0.88
P0.380.26−0.450.97
SP0.27−0.63−3.722.45
P20.13−0.19−0.550.18
Ostracod E(S100) models with % Atlantic species as D    
D0.95.021.318.73
P0.420.31−0.411.03
SP0.38−1.82−5.201.56
P20.18−0.23−0.580.12

Analyses reported here (except spectral analysis) were implemented in the R programming language (R Core Team, 2013), using functions from the R packages vegan (Oksanen et al., 2013) for diversity and NMDS and MuMIn (Bartoń, 2013) to perform the model averaging. A maximum entropy spectral analysis was performed by using the software AnalySeries, version 2.0.4.2 (Paillard et al., 1996). Data are available at Dryad (http://doi.org/10.5061/dryad.53vs3).

Biodiversity data from other cores

The 23GGC ostracod species diversity record was compared with other published biodiversity records available with comparable time resolution and water depth (i.e. water mass, upper NADW). Benthic foraminiferal diversity data are available from the nearby core DS97-2P (Rasmussen et al., 2003) as raw species richness (i.e. total number of species per sample; sample size is > 200–300 specimens per sample) (Rasmussen et al., 2003). North-western Atlantic benthic ostracod diversity data are from site ODP1055 as E(S50). Further details will be discussed below.

Results and Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results and Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Decadal–centennial-scale deep-sea biodiversity and faunal variability

The 23GGC species diversity as measured by the expected number of species in 50 individuals, E(S50), generally increased during the Heinrich 1, the Younger Dryas and the 8.2 ka stadial events recorded in the NGRIP Greenland ice core δ18O data (North Greenland Ice Core Project members, 2004). At the onset of the Younger Dryas, species diversity rapidly increased within about a century. During the latter half of Heinrich 1, E(S50) increases abruptly from 12 to 18 within about a century (Fig. 3). In addition, the 23GGC diversity data showed shorter-term variability within each event interval (Fig. 3). In the Henrich 1 event in particular, the ostracod diversity shows distinct double ‘high-diversity’ peaks at c. 14.5 ka and c. 16 ka, respectively (note that y-axis is upside down in Fig. 3b).

figure

Figure 3. Changes in deep-sea species diversity for the last 20,000 years with proxy records of deepwater circulation and atmospheric climate. (a) Greenland ice core NGRIP δ18O proxy for atmospheric temperature (North Greenland Ice Core Project members, 2004). (b) North-eastern Atlantic core 23GGC ostracod species diversity E(S50). (c) North-eastern Atlantic core DS97-2P foraminiferal species richness (Rasmussen et al., 2003) with the revised chronology (Rasmussen & Thomsen, 2004). Deepwater circulation proxies of (d) benthic foraminiferal δ13C in north-eastern Atlantic cores ODP984C (Praetorius et al., 2008) with the revised chronology (Thornalley et al., 2011), (e) relative abundance of ‘Atlantic (benthic foraminiferal) species group’, an indicator of warmer deep water, in core DS97-2P (Rasmussen et al., 2003; Rasmussen & Thomsen, 2004), and (f) north-eastern Atlantic deepwater 14C ventilation age (Thornalley et al., 2011). Lower δ13C value and higher ‘Atlantic species group’ abundance and 14C ventilation age values mean weaker deepwater circulation and resulting higher bottom water temperature at this water depth (i.e. intermediate depth) for this region (see the main text). (g) North-western Atlantic core ODP1055 ostracod species diversity E(S50) (Yasuhara et al., 2008). The 8.2 ka (8.2), Younger Dryas (YD) and Heinrich 1 (H1) stadial events are indicated.

Download figure to PowerPoint

Benthic foraminiferal diversity from the nearby core DS97-2P (58°56.327′ N, 30°24.590′ W, 1685 m water depth; Fig. 1) tightly co-varies with the 23GGC ostracod diversity and has distinct Heinrich 1 and Younger Dryas signals (Figs 2d & 3). The ostracod diversity record from the 5000-km distant north-western Atlantic core ODP1055 (32.78402° N, 76.28632° W, 1798 m water depth; this core has similar water depth to 23GGC; Fig. 1) displays similar trends (Fig. 3). Notably, the diversity in ODP1055 shows similar double peaks to 23GGC during the Heinrich 1 event, which may be related to short-term deepwater variability within this event, because a deepwater proxy of 14C ventilation age shows a similar trend (Fig. 3f). Diversity records from these cores reveal that deep-sea biodiversity changes synchronously with stadial–interstadial climate changes across a large area of the North Atlantic, despite the different dispersal abilities of the two benthic groups. Ostracods lack a dispersal stage and most do not swim, but foraminifera are known to disperse widely (Brandt et al., 2007; Yasuhara et al., 2012a, 2012c). In addition, cores 23GGC and ODP1055 both show distinct ostracod diversity peaks corresponding to many of brief Holocene stadial events of the rapid climate changes (RCCs) (Mayewski et al., 2004) recorded in Greenland ice core GISP2 potassium and north-western Atlantic core KNR158-4-21MC/22GGC detrital CaCO3 palaeoclimatic proxies (Fig. 4). Spectral analysis of the 23GGC record shows a c. 1500-year cycle in species diversity (Fig. 2f). This cyclicity is almost identical to the 1500-year climatic cycle evident in some palaeoclimate records (Bond et al., 1997), suggesting a quasi-cyclic climatic control of deep-sea biodiversity for the last 20,000 years.

figure

Figure 4. Variability of Holocene species diversity with proxy records of atmospheric climate. (a) North-eastern Atlantic core 23GGC ostracod species diversity E(S50). (b) North-western Atlantic core ODP1055 ostracod species diversity E(S50) (Yasuhara et al., 2008). (c) The Greenland ice core GISP2 potassium ion (K+: 60-point moving average) proxy for the Siberian High (O'Brien et al., 1995; Meeker & Mayewski, 2002). (d) North-western Atlantic core KNR158-4-21MC/22GGC percentage detrital CaCO3 proxy for ice-rafted debris (i.e. stadial) events (Bond et al., 2001). Holocene RCC (rapid climate changes) stadial events (Mayewski et al., 2004) are indicated by grey bars.

Download figure to PowerPoint

The 23GGC NMDS results show pronounced faunal change at the beginning of the Heinrich 1 event (Fig. 5). The NMDS1 score is stable during the Last Glacial Maximum, and then rapidly starts to decrease at the beginning of Heinrich 1 event. The score continues to decrease to the end of Heinrich 1 event, and then show relatively stable value throughout the last 14,000 years. In other words, there are distinct faunas before and after the Heinrich 1 event. The NMDS2 score shows strong peak during the Heinrich 1 event, suggesting the existence of a third distinct fauna. Here we call these faunal assemblages the Last Glacial, Holocene and Deglacial faunas, respectively (Fig. S3). The Last Glacial fauna is characterized by Krithe trinidadensis (KriT), Legitimocythere acanthoderma (Leg) and a few other species; the Holocene fauna is characterized by species of Argilloecia (such as Arg1, Arg2, Arg4), Henryhowella asperrima, Pennyella rexi and many other species; and the Deglacial fauna is characterized by Polycope species (PolO, PolS, PolV, Pol2) (Fig. S3). These faunal changes related to the Heinrich 1 event suggest that the deep-sea ecosystem is strongly forced by abrupt climate changes.

figure

Figure 5. Deep-sea faunal composition changes for the last 20,000 years with proxy records of deepwater circulation and atmospheric climate. (a) Greenland ice core NGRIP δ18O record (North Greenland Ice Core Project members, 2004). (b) North-eastern Atlantic core 23GGC ostracod faunal composition NMDS1. (c) Benthic foraminiferal δ13C in north-eastern Atlantic cores ODP984C (Praetorius et al., 2008) with the revised chronology (Thornalley et al., 2011), deepwater proxy. (d) North-eastern Atlantic core 23GGC ostracod faunal composition NMDS2.

Download figure to PowerPoint

Controlling factor of deep-sea biodiversity

We compared the 23GGC biodiversity record with two proxies of AMOC variability (i.e. proxies of deepwater circulation variability around this intermediate depth site), benthic foraminiferal δ13C and ‘Atlantic species group’ benthic foraminiferal relative abundance (Rasmussen et al., 2003), and with proxies for surface ocean productivity and its seasonality (BFAR and ‘phytodetritus group’ relative abundance, respectively) from nearby sites (see Methods). There is a significant correlation between biodiversity patterns at 23GGC and benthic δ13C and ‘Atlantic species group’ abundance (Figs 2 & 3), with high species diversity characterizing periods with low δ13C and high ‘Atlantic species group’ abundance values, which signify a reduction or shutdown of AMOC (and warmer deep water; see below). For example, both the largest increases in diversity and the largest reductions in AMOC are recorded during the Younger Dryas and Heinrich 1 events, though there are certain differences in AMOC trends among different proxies (Fig. 3d–f). In contrast to the AMOC proxies (especially benthic δ13C), the proxies of surface productivity and seasonality do not seem to show a strong correspondence to the biodiversity curve (Figs 2 & 6). However, more rigorous tests are needed to confirm this interpretation.

figure

Figure 6. Deep-sea species diversity changes for the last 20,000 years with proxy records of surface productivity and seasonality of productivity. (a) Greenland ice core NGRIP δ18O record (North Greenland Ice Core Project members, 2004). (b) North-eastern Atlantic core 23GGC ostracod species diversity E(S50). (c) North-eastern Atlantic core DS97-2P foraminiferal species richness (Rasmussen et al., 2003). (d) Benthic foraminiferal accumulation rates (BFAR) (Rasmussen et al., 2003), surface productivity proxy. (e) Relative abundance of phytodetritus species (Epistominella exigua and Alabaminella weddellensis) (Rasmussen et al., 2003), as a proxy of seasonality of productivity.

Download figure to PowerPoint

Thus, we used regression models to further test the relationships between species diversity and AMOC, surface productivity and seasonality of productivity, using a higher sampling threshold, E(S100), to reduce possible noise (Tables 1 & 2). The three best regressions consistently indicate that AMOC (i.e. deepwater circulation) is a significant predictor of species diversity (Table 1). The model averaging results support the significance of the AMOC–diversity relationship, showing that the relative importance is always higher for the deepwater circulation term than the other terms, and that the coefficient for the deepwater circulation term is always significantly different from zero (Table 2). Only the deepwater circulation term receives consistent support between both of the modelling results using two different deepwater proxies (Tables 1 & 2). These results suggest an important role for deepwater circulation controlling diversity. We showed a significant deepwater–diversity correlation, assuming that the ODP984C benthic δ13C record actually represents deepwater changes in this area (i.e. subpolar North Atlantic, south of Iceland and along the Reykjanes Ridge, at intermediate water depth under the path of the Iceland–Scotland Overflow Waters; Raymo et al., 2004). This assumption is supported by many deepwater proxies (e.g. 14C ventilation age, Cd/Ca ratio, sortable silt grain size and ‘Atlantic species group’ abundance) in this area (including RAPiD, NEAP4K, DS97-2P, and other cores; Fig. 1; Rasmussen & Thomsen, 2004; Rickaby & Elderfield, 2005; Thornalley et al., 2011) which show a similar trend to the ODP984C benthic δ13C record (Praetorius et al., 2008; Thornalley et al., 2011), reflecting changes in the Iceland–Scotland Overflow Waters. However, intercore comparison (i.e. comparison of palaeo-proxy data among different cores) always involves some uncertainty due to local differences of palaeoclimatic and palaeo-oceanographic changes, differences in age–model robustness, possible sedimentary gaps (hiatuses), etc. Thus, further investigation using a single core with all of the diversity and environmental proxies in identical sample horizons is needed to further assess our conclusion.

Recent studies have shown that variability in AMOC is accompanied in the study region by changes at intermediate water depths in the relative strength of colder NADW from the Nordic seas and warmer subsurface Atlantic water from the subtropics (Rasmussen & Thomsen, 2009; Marcott et al., 2011). Warmer subsurface Atlantic water is enhanced and colder NADW is weakened during the stadial events characterized by AMOC reduction, so that a warmer bottom-water temperature characterizes periods with a weaker AMOC (i.e. stadial events). This interpretation is supported by increased relative abundance of the ‘Atlantic species group’, a group of indicator foraminiferal species for warmer temperature in the subpolar North Atlantic Ocean and Nordic seas during the stadial periods (Rasmussen et al., 2003; Rasmussen & Thomsen, 2004). Thus, the significant deepwater–diversity relationship shown in the present study suggests a positive temperature control of species diversity, and thus a significant role of deep-sea temperature in shaping large-scale diversity patterns. This interpretation involving the ‘Atlantic species group’ and species diversity is not circular, because we compared the ‘Atlantic species group’ abundance of foraminifera with the species diversity of ostracods, a different taxonomic group.

Although our modelling results indicated that among the environmental proxies used here only the deepwater proxies seem important for diversity, the effect sizes are relatively small even for the deepwater proxies. In deepwater–diversity relationships the R2 values are c. 0.24 for the benthic δ13C and much less for the ‘Atlantic species group’ (Fig. 2), suggesting that deepwater variability explains only c. 24% or less of the total diversity variation. These small effect sizes suggest: (1) the importance of other factors not included in the present study, and/or (2) above-mentioned uncertainties involving inter-core comparison. If the former is correct, future studies should focus on identifying and measuring other environmental and biotic parameters that may influence biodiversity changes. For the latter case, we need the ‘single core with all proxies’ approach as mentioned above. Both approaches may be fruitful future directions for research to better understand the driving forces of deep-sea biodiversity.

Controlling factors of deep-sea faunal composition

NMDS1 shows a general glacial-high and Holocene-low trend, similar to surface climatic trends (represented by the Greenland ice-core δ18O record), though the Younger Dryas and Heinrich 1 events are indistinct in NMDS1 (Fig. 5). In contrast, NMDS2 shows a very similar trend to AMOC proxy benthic δ13C (Fig. 5). These NMDS1 and -2 trends suggest that the deep-sea faunal assemblage is affected not only by deepwater-related changes (such as bottom temperature) but also by surface-climate-related changes (e.g. quantity, quality, composition and/or seasonality of surface production, as the primary food source for the deep-sea benthos). Our modelling results support the significance of the AMOC–NMDS2 relationship (Tables S3 & S4), suggesting an impact of AMOC on deep-sea faunal composition. However, our modelling does not show consistent support for the productivity–NMDS1 or seasonality–NMDS1 relationship: for example, the best three models include the null model (Table S3); and the model averaging result shows no significance for any environmental parameters (Table S4). These non-significant results imply that deep-sea faunal composition (specifically NMDS1) is impacted by a surface-climate-related factor that is not included in our modelling. One possibility is the quality or composition of primary production. The composition of the pelagic plankton assemblage is distinct between interglacial (including Holocene) and glacial periods and shows regular glacial–interglacial variability related to change in surface climate (e.g. Kandiano et al., 2004). Such glacial–interglacial compositional difference or resulting quality difference of surface production (i.e. food for the deep-sea benthos) may explain the glacial–Holocene NMDS1 trend.

High-resolution palaeoecology and global diversity patterns: broader implications

The rapid (decadal–centennial) and widely synchronous (between cores that are 5000 km apart) diversity changes accompanying changes in AMOC shown in the present study suggest that diversity can be rapidly reorganized, most likely through shifts in the bathymetric or geographic ranges of species (Roy et al., 1996; Yasuhara et al., 2009, 2012b). They therefore provide strong evidence for a dominant role of ecological processes in shaping and regulating modern-day global diversity patterns, rather than the evolutionary dynamics of speciation and extinction (Jablonski et al., 2006) that rarely occur at these time-scales. Species temperature tolerances are probably responsible for such range shifts, supporting the physiological tolerance hypothesis (Currie et al., 2004). Because climatic modelling studies indicate that human-induced global warming will alter AMOC (Hu et al., 2009), our results imply that future climate changes may involve abrupt reorganization of marine ecosystems and global diversity patterns. Exceptionally highly resolved fossil records help us to understand past, present and future ecosystems by bridging the gap between biological and palaeontological time-scales.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results and Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

We thank D. W. Oppo for providing core samples, radiocarbon dates and isotope data; J. Dyszynski for sample processing; and D. J. Currie, M. Rex and three anonymous referees for helpful comments and constructive criticisms. This work was supported by a Smithsonian Postdoctoral Fellowship, a Smithsonian Marine Science Network Postdoctoral Fellowship and the Japan Society for the Promotion of Science Postdoctoral Fellowships for Research Abroad (to M.Y.).

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results and Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information
  • Anderson, D.R., Burnham, K.P. & Thompson, W.L. (2000) Null hypothesis testing: problems, prevalence, and an alternative. Journal of Wildlife Management, 64, 912923.
  • Bartoń, K. (2013) MuMIn: Multi-model inference. R package version 1.9.13. Available at: http://CRAN.R-project.org/package=MuMIn (accessed 30 March 2014).
  • Bond, G.C., Showers, W., Cheseby, M., Lotti, R., Almasi, P., deMenocal, P., Priore, P., Cullen, H., Hajdas, I. & Bonani, G. (1997) A pervasive millennial-scale cycle in North Atlantic Holocene and glacial climates. Science, 278, 12571266.
  • Bond, G.C., Kromer, B., Beer, J., Muscheler, R., Evans, M.N., Showers, W., Hoffmann, S., Lotti-Bond, R., Hajdas, I. & Bonani, G. (2001) Persistent solar influence on North Atlantic climate during the Holocene. Science, 294, 21302136.
  • Brandt, A., Gooday, A.J., Brandão, S.N. et al. (2007) First insights into the biodiversity and biogeography of the Southern Ocean deep sea. Nature, 447, 307311.
  • Burnham, K.P. & Anderson, D.R. (2002) Model selection and multimodel inference: a practical information-theoretic approach. Springer, New York.
  • Cannariato, K.G., Kennett, J.P. & Behl, R.J. (1999) Biotic response to late Quaternary rapid climate switches in Santa Barbara Basin: ecological and evolutionary implications. Geology, 27, 6366.
  • Corliss, B.H., Brown, C.W., Sun, X. & Showers, W.J. (2009) Deep-sea benthic diversity linked to seasonality of pelagic productivity. Deep-Sea Research I, 56, 835841.
  • Cronin, T.M. & Raymo, M.E. (1997) Orbital forcing of deep-sea benthic species diversity. Nature, 385, 624627.
  • Cronin, T.M., Gemery, L., Briggs, W.M., Jakobsson, M., Polyak, L. & Brouwers, E.M. (2010) Quaternary sea-ice history in the Arctic Ocean based on a new ostracode sea-ice proxy. Quaternary Science Reviews, 29, 34153429.
  • Cronin, T.M., Dwyer, G.S., Farmer, J., Bauch, H.A., Spielhagen, R.F., Jakobsson, M., Nilsson, J., Briggs, W.M. & Stepanova, A. (2012) Deep Arctic Ocean warming during the last glacial cycle. Nature Geoscience, 5, 631634.
  • Currie, D.J., Mittelbach, G.G., Cornell, H.V., Field, R., Guegan, J.F., Hawkins, B.A., Kaufman, D.M., Kerr, J.T., Oberdorff, T., O'Brien, E. & Turner, J.R.G. (2004) Predictions and tests of climate-based hypotheses of broad-scale variation in taxonomic richness. Ecology Letters, 7, 11211134.
  • Danovaro, R., Dell'Anno, A. & Pusceddu, A. (2004) Biodiversity response to climate change in a warm deep sea. Ecology Letters, 7, 821828.
  • Herguera, J.C. (2000) Last glacial paleoproductivity patterns in the eastern equatorial Pacific: benthic foraminifera records. Marine Micropaleontology, 40, 259275.
  • Hu, A., Meehl, G.A., Han, W. & Yin, J. (2009) Transient response of the MOC and climate to potential melting of the Greenland Ice Sheet in the 21st century. Geophysical Research Letters, 36, L10707, doi: 10.1029/2009GL037998.
  • Hunt, G., Cronin, T.M. & Roy, K. (2005) Species–energy relationship in the deep sea: a test using the Quaternary fossil record. Ecology Letters, 8, 739747.
  • Hurlbert, S.H. (1971) The nonconcept of species diversity: a critique and alternative parameters. Ecology, 52, 577586.
  • Jablonski, D., Roy, K. & Valentine, J.W. (2006) Out of the tropics: evolutionary dynamics of the latitudinal diversity gradient. Science, 314, 102106.
  • Kandiano, E.S., Bauch, H.A. & Müller, A. (2004) Sea surface temperature variability in the North Atlantic during the last two glacial–interglacial cycles: comparison of faunal, oxygen isotopic, and Mg/Ca-derived records. Palaeogeography, Palaeoclimatology, Palaeoecology, 204, 145164.
  • McClain, C.R., Allen, A.P., Tittensor, D.P. & Rex, M.A. (2012) Energetics of life on the deep seafloor. Proceedings of the National Academy of Sciences of the USA, 109, 1536615371.
  • Marcott, S.A., Clark, P.U., Padman, L., Klinkhammer, G.P., Springer, S.R., Liu, Z., Otto-Bliesner, B.L., Carlson, A.E., Ungerer, A., Padman, J., He, F., Cheng, J. & Schmittner, A. (2011) Ice-shelf collapse from subsurface warming as a trigger for Heinrich events. Proceedings of the National Academy of Sciences of the USA, 108, 1341513419.
  • Mayewski, P.A., Meeker, L.D., Whitlow, S., Twickler, M.S., Morrison, M.C., Alley, R.B., Bloomfield, P. & Taylor, K. (1993) The atmosphere during the Younger Dryas. Science, 261, 195197.
  • Mayewski, P.A., Rohling, E.E., Stager, J.C., Karlén, W., Maasch, K.A., Meeker, L.D., Meyerson, E.A., Gasse, F., van Kreveld, S., Holmgren, K., Lee-Thorp, J., Rosqvist, G., Rack, F., Staubwasser, M., Schneider, R.R. & Steig, E.J. (2004) Holocene climate variability. Quaternary Research, 62, 243255.
  • Meeker, L.D. & Mayewski, P.A. (2002) A 1400-year high-resolution record of atmospheric circulation over the North Atlantic and Asia. The Holocene, 12, 257266.
  • North Greenland Ice Core Project members (2004) High-resolution record of northern hemisphere climate extending into the last interglacial period. Nature, 431, 147151.
  • O'Brien, S.R., Mayewski, P.A., Meeker, L.D., Meese, D.A., Twickler, M.S. & Whitlow, S.I. (1995) Complexity of Holocene climate as reconstructed from a Greenland ice core. Science, 270, 19621964.
  • Oksanen, J., Blanchet, F.G., Kindt, R., Legendre, P., Minchin, P.R., O'Hara, R.B., Simpson, G.L., Solymos, P., Stevens, M.H.H. & Wagner, H. (2013) vegan: Community Ecology Package. R package version 2.0-10. Available at: http://CRAN.R-project.org/package=vegan (accessed 30 March 2014).
  • Paillard, D., Labeyrie, L. & Yiou, P. (1996) Macintosh program performs time-series analysis. EOS Transactions, AGU, 77, 379.
  • Praetorius, S.K., McManus, J.F., Oppo, D.W. & Curry, W.B. (2008) Episodic reductions in bottom-water currents since the last ice age. Nature Geoscience, 1, 449452.
  • R Core Team (2013) R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. Available at: http://www.R-project.org (accessed 30 March 2014).
  • Rasmussen, T.L. & Thomsen, E. (2004) The role of the North Atlantic Drift in the millennial timescale glacial climate fluctuations. Palaeogeography, Palaeoclimatology, Palaeoecology, 210, 101116.
  • Rasmussen, T.L. & Thomsen, E. (2009) Ventilation changes in intermediate water on millennial time scales in the SE Nordic seas, 65–14 kyr BP. Geophysical Research Letters, 36, L01601, doi: 10.1029/2008GL036563.
  • Rasmussen, T.L., Thomsen, E., Troelstra, S.R., Kuijpers, A. & Prins, M.A. (2003) Millennial-scale glacial variability versus Holocene stability: changes in planktic and benthic foraminifera faunas and ocean circulation in the North Atlantic during the last 60000 years. Marine Micropaleontology, 47, 143176.
  • Raymo, M.E., Oppo, D.W., Flower, B.P., Hodell, D.A., McManus, J.F., Venz, K.A., Kleiven, K.F. & McIntyre, K. (2004) Stability of North Atlantic water masses in face of pronounced climate variability during the Pleistocene. Paleoceanography, 19, PA2008, doi: 10.1029/2003PA000921.
  • Reimer, P.J., Baillie, M.G.L., Bard, E. et al. (2009) Intcal09 and Marine09 radiocarbon age calibration curves, 0–50,000 years cal BP. Radiocarbon, 51, 11111150.
  • Rex, M.A. & Etter, R.J. (2010) Deep-sea biodiversity: pattern and scale. Harvard University Press, Cambridge, MA.
  • Rex, M.A., Stuart, C.T., Hessler, R.R., Allen, J.A., Sanders, H.L. & Wilson, G.D.F. (1993) Global-scale latitudinal patterns of species diversity in the deep-sea benthos. Nature, 365, 636639.
  • Rickaby, R.E.M. & Elderfield, H. (2005) Evidence from the high-latitude North Atlantic for variations in Antarctic intermediate water flow during the last deglaciation. Geochemistry, Geophysics, Geosystems, 6, Q05001, doi: 10.1029/2004GC000858.
  • Roy, K., Valentine, J.W., Jablonski, D. & Kidwell, S.M. (1996) Scales of climatic variability and time averaging in Pleistocene biotas: implications for ecology and evolution. Trends in Ecology and Evolution, 11, 458463.
  • Ruhl, H.A., Ellena, J.A. & Smith, K.L., Jr (2008) Connections between climate, food limitation, and carbon cycling in abyssal sediment communities. Proceedings of the National Academy of Sciences of the USA, 105, 1700617011.
  • Stuiver, M. & Reimer, P.J. (1993) Extended 14C database and revised CALIB radiocarbon calibration program. Radiocarbon, 35, 215230.
  • Sun, X., Corliss, B.H., Brown, C.W. & Showers, W.J. (2006) The effect of primary productivity and seasonality on the distribution of deep-sea benthic foraminifera in the North Atlantic. Deep-Sea Research I, 53, 2847.
  • Thornalley, D.J.R., Elderfield, H. & McCave, I.N. (2010) Intermediate and deep water paleoceanography of the northern North Atlantic over the past 21,000 years. Paleoceanography, 25, PA1211, doi: 10.1029/2009PA001833.
  • Thornalley, D.J.R., Barker, S., Broecker, W.S., Elderfield, H. & McCave, I.N. (2011) The deglacial evolution of North Atlantic deep convection. Science, 331, 202205.
  • Thornalley, D.J.R., Blascheck, M., Davies, F.J., Praetorius, S.K., Oppo, D.W., McManus, J., Hall, I.R., Kleiven, H., Renssen, H. & McCave, I.N. (2013) Long-term variations in Iceland–Scotland overflow strength during the Holocene. Climate of the Past, 9, 20732084.
  • Tittensor, D.P., Mora, C., Jetz, W., Lotze, H.K., Ricard, D., Berghe, E.V. & Worm, B. (2010) Global patterns and predictors of marine biodiversity across taxa. Nature, 466, 10981101.
  • Tittensor, D.P., Rex, M.A., Stuart, C.T., McClain, C.R. & Smith, C.R. (2011) Species–energy relationships in deep-sea molluscs. Biology Letters, 7, 718722.
  • Waelbroeck, C., Duplessy, J.C., Michel, E., Labeyrie, L., Paillard, D. & Duprat, J. (2001) The timing of the last deglaciation in North Atlantic climate records. Nature, 412, 724727.
  • Wollenburg, J.E., Mackensen, A. & Kuhnt, W. (2007) Benthic foraminiferal biodiversity response to a changing Arctic palaeoclimate in the last 24.000 years. Palaeogeography, Palaeoclimatology, Palaeoecology, 255, 195222.
  • Yamaguchi, T., Norris, R.D. & Dockert, D.T., III (2014) Shallow-marine ostracode turnover during the Eocene–Oligocene transition in Mississippi, the Gulf Coast Plain, USA. Marine Micropaleontology, 106, 1021.
  • Yasuhara, M. & Cronin, T.M. (2008) Climatic influences on deep-sea ostracode (Crustacea) diversity for the last three million years. Ecology, 89, S52S65.
  • Yasuhara, M., Cronin, T.M., deMenocal, P.B., Okahashi, H. & Linsley, B.K. (2008) Abrupt climate change and collapse of deep-sea ecosystems. Proceedings of the National Academy of Sciences of the USA, 105, 15561560.
  • Yasuhara, M., Hunt, G., Cronin, T.M. & Okahashi, H. (2009) Temporal latitudinal-gradient dynamics and tropical instability of deep-sea species diversity. Proceedings of the National Academy of Sciences of the USA, 106, 2171721720.
  • Yasuhara, M., Hunt, G., Cronin, T.M., Hokanishi, N., Kawahata, H., Tsujimoto, A. & Ishitake, M. (2012a) Climatic forcing of Quaternary deep-sea benthic communities in the North Pacific Ocean. Paleobiology, 38, 162179.
  • Yasuhara, M., Hunt, G., Dowsett, H.J., Robinson, M.M. & Stoll, D.K. (2012b) Latitudinal species diversity gradient of marine zooplankton for the last three million years. Ecology Letters, 15, 11741179.
  • Yasuhara, M., Hunt, G., van Dijken, G., Arrigo, K.R., Cronin, T.M. & Wollenburg, J.E. (2012c) Patterns and controlling factors of species diversity in the Arctic Ocean. Journal of Biogeography, 39, 20812088.

Moriaki Yasuhara is an assistant professor of environmental science in the School of Biological Sciences, the Swire Institute of Marine Science and the Department of Earth Sciences at the University of Hong Kong. He has broad interests in marine palaeoecology and macroecology using highly resolved micropalaeontological records. His recent research has focused on the spatio-temporal dynamics of large-scale biodiversity patterns, the impact of climate on species diversity and the controlling factor(s) of biodiversity pattern/change in deep-sea, shallow-marine and pelagic ecosystems. He is also interested in microfossil-based conservation palaeobiology and palaeontology of the Ostracoda in general.

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results and Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information
FilenameFormatSizeDescription
geb12178-sup-0001-fig_s1.pdf169K

Figure S1 Comparison of different species diversity measures.

geb12178-sup-0002-fig_s2.pdf92K

Figure S2 Benthic δ18O records of core 23GGC and nearby cores.

geb12178-sup-0003-fig_s3.pdf533K

Figure S3 Non-metric multidimensional scaling ordination.

geb12178-sup-0004-table_s1.pdf44K

Table S1 Accelerator mass spectrometer radiocarbon dates and Vedde Ash layer for site 23GGC.

geb12178-sup-0005-table_s2.pdf46K

Table S2 23GGC benthic δ18O data.

geb12178-sup-0006-table_s3.pdf70K

Table S3 Best three regression models of 23GGC faunal composition non-metric multidimensional scaling 1 and 2.

geb12178-sup-0007-table_s4.pdf59K

Table S4 Model-averaged parameter estimates of 23GGC faunal composition non-metric multidimensional scaling 1 and 2.

geb12178-sup-0008-table_s5.pdf52K

Table S5 Ostracod species name abbreviation for non-metric multidimensional scaling ordination.

geb12178-sup-0009-table_s6.pdf54K

Table S6 Dataset used for diversity modelling.

geb12178-sup-0010-table_s7.pdf75K

Table S7 23GGC ostracod diversity data.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.