Mg/Ca in the benthic foraminifera Cibicidoides wuellerstorfi and Cibicidoides mundulus: Temperature versus carbonate ion saturation

https://doi.org/10.1016/j.epsl.2008.09.015Get rights and content

Abstract

Magnesium/calcium (Mg/Ca) ratios in two widely used benthic foraminiferal species, Cibicidoides wuellerstorfi and Cibicidoides mundulus (Cibicidoides kullenbergi), picked from global ocean core-tops and from six cores in the North Atlantic Ocean have been measured to investigate Mg/Ca variability, cleaning effect, and influences from bottom water temperature (BWT) and deep water carbonate ion saturation (Δ[CO32−]). Replicate measurements of the two species reveal Mg/Ca variability of 0.07 ± 0.07 mmol/mol. Compared with Mg/Ca in samples cleaned by oxidative cleaning, Mg/Ca ratios are significantly lowered, by 0.10 ± 0.09 mmol/mol, in samples cleaned by reductive cleaning, likely due to preferential leaching during the reductive cleaning step. The cleaning influence is large relative to glacial–interglacial Mg/Ca changes (0.05–0.31 mmol/mol by this study). Regression of core-top data from a wide range of locations show that C. wuellerstorfi Mg/Ca is strongly affected by deep water Δ[CO32−] with a sensitivity of ~ 0.009 mmol/mol per µmol/kg. When BWT is included in regressions, the correlation is slightly more significant with a weak BWT sensitivity on Mg/Ca of between 0.03 ± 0.01 and 0.07 ± 0.02 mmol/mol per °C. Because Δ[CO32−] and BWT co-vary for most core-top samples, an alternate method of separating the two effects was made by comparing glacial to interglacial changes in Mg/Ca from a depth transect of North Atlantic cores where the BWT and carbonate ion histories since the last glacial period have been well constrained. Results from this transect suggest that the BWT effect is < 0.03 mmol/mol per °C, weaker than inferred from the core-top study. Therefore, core-top and down core data suggest that C. wuellerstorfi Mg/Ca is insensitive to BWT changes. The weak BWT effect, when compared with natural variability, complicates the use of Mg/Ca in this species to reconstruct past BWT. Taking all measured samples from various environments into account, C. mundulus Mg/Ca ratios show no correlation with BWT or deep water Δ[CO32−].

Introduction

The use of foraminiferal magnesium/calcium (Mg/Ca) ratios to estimate calcification temperatures from marine archives is now common. The exponential relationship between planktonic foraminiferal Mg/Ca and temperature has been well demonstrated by foraminifera culturing, sediment trap, and core-top studies (e.g., Mashiotta et al., 1999, Elderfield and Ganssen, 2000, Lea et al., 2000, Anand et al., 2003, Russell et al., 2004, McConnell and Thunell, 2005, Skinner and Elderfield, 2005). By comparison, variation of benthic foraminiferal Mg/Ca with bottom water temperature (BWT), largely investigated using core-top samples, is relatively poorly constrained. Benthic foraminiferal Mg/Ca, in general, shows positive correlations with BWT over large temperature ranges (− 1 to 19 °C) (Rosenthal et al., 1997, Lear et al., 2002, Martin et al., 2002), except in the Norwegian Sea where BWT is sub-zero but much elevated benthic Mg/Ca values are observed (Martin et al., 2002, Elderfield et al., 2006). These data from the Norwegian Sea together with observations of core-top Mg/Ca values higher than predicted from calibrations have led to suggestions that bottom water carbonate ion saturation (Δ[CO32−]) is an additional influence on benthic foraminiferal Mg/Ca (Elderfield et al., 2006, Rosenthal et al., 2006).

A Δ[CO32−] effect on benthic foraminiferal Mg/Ca significantly complicates the use of benthic Mg/Ca for BWT reconstructions. In order to extract the BWT signature from benthic foraminiferal Mg/Ca, sensitivities on benthic Mg/Ca from BWT and Δ[CO32−] should be well quantified. However, some complications may be associated with literature benthic Mg/Ca studies. Many published calibrations are regional, apart from a few studies (e.g., Lear et al., 2002, Martin et al., 2002, Elderfield et al., 2006). Because changes in benthic Mg/Ca with temperature are small at low temperatures, effects of different cleaning methods (e.g., Rosenthal et al., 2004, Yu et al., 2007) and perhaps from inter-laboratory offsets of analytical standards (e.g., Lear et al., 2002, Martin et al., 2002) need to be evaluated. Inter-species differences in Mg/Ca also complicate calibrations based on mixed benthic species (Elderfield et al., 2006, Rosenthal et al., 1997). Therefore, controlling factors on benthic Mg/Ca warrant further evaluation.

In this study, we have expanded the dataset from the previous work (Elderfield et al., 2006) and present global core-top Mg/Ca data in two widely used benthic foraminifera, Cibicidoides wuellerstorfi and Cibicidoides mundulus. Additionally, we show down core benthic Mg/Ca ratios back to the last 40 kyr for a suite of cores located at water depths of 1–4 km in the North Atlantic. Past deep water Δ[CO32−] for these down cores have evolved distinctly and been quantitatively estimated using an independent proxy boron/calcium (B/Ca) (Yu et al., 2008), facilitating evaluation of Δ[CO32−] effects on benthic Mg/Ca. We have assessed influences on Mg/Ca in these two species from Mg/Ca variability, different cleaning methods, and changes in BWT and Δ[CO32−]. Our core-top and down core data suggest a very weak temperature effect on Mg/Ca in C. wuellerstorfi and no temperature influence on Mg/Ca in C. mundulus, restricting the use of their Mg/Ca for past BWT reconstructions.

Section snippets

Samples

We have measured Mg/Ca ratios in two widely distributed benthic foraminiferal species, C. wuellerstorfi and C. mundulus (also called C. kullenbergi), in core-top sediments from the global oceans (Appendix, Table A1). Sediments were verified to be Holocene in age using published benthic and planktonic foraminiferal δ18O and radiocarbon dates (for details, refer to Yu and Elderfield, 2007). Down core Mg/Ca ratios were measured in C. mundulus from BOFS 17K and in C. wuellerstorfi from BOFS 11K,

Mg/Ca variability and cleaning effect

Replicate measurements of Mg/Ca ratios in C. wuellerstorfi and C. mundulus picked from the same core-top sediments (subject to the same cleaning method), based on 2–4 Mg/Ca replicates and starting shell weights of ~ 400 μg, show a standard deviation (SD) range of 0–0.3 mmol/mol, with the average of 0.07 ± 0.07 mmol/mol (Fig. 1A). This is significantly larger than the analytical error of 0.02 mmol/mol, suggestive of heterogeneous distributions of Mg in and/or between benthic foraminiferal shells.

BWT effect

Fig. 2A, B shows Mg/Ca in C. wuellerstorfi, cleaned using two different methods, from global oceans spanning a BWT range of ~  1–4.5 °C. At 0.5–4.0 °C, core-top C. wuellerstorfi Mg/Ca increases with BWT, displaying a much higher temperature sensitivity than observed for C. pachyderma living in warm conditions (Elderfield et al., 2006, Lear et al., 2002, Martin et al., 2002) and for inorganically precipitated carbonates (Oomori et al., 1987). As previously documented (Elderfield et al., 2006,

Mg/Ca in C. mundulus

Considering the entire core-top and down core dataset obtained in this study, C. mundulus Mg/Ca shows no correlation with BWT or bottom water Δ[CO32−] (Fig. 8), in contrast with early studies (Elderfield et al., 2006, Healey et al., 2008). The higher glacial C. mundulus Mg/Ca in BOFS 17K than Holocene provide strong evidence arguing against BWT as the sole control on C. mundulus Mg/Ca (Fig. 1, Fig. 4). If only samples (except for one data from T90 15B; see Table A1.2) from < ~ 50 µmol/kg are

Conclusions

In this study, we have investigated the influences of Mg/Ca variability, cleaning effects, BWT, and Δ[CO32−] on Mg/Ca in two widely used benthic foraminiferal species, C. wuellerstorfi and C. mundulus, using core-top samples from the global oceans and down core samples from six down cores at 1–4 km water depths from the North Atlantic Ocean. Mg/Ca in the two species shows sizable variability (0.07 ± 0.07 mmol/mol) and is significantly affected by different cleaning procedures (0.10 ± 0.09 mmol/mol)

Acknowledgements

Discussion with David Thornalley proved helpful. We are very grateful to Gerald Ganssen for access to foraminifera from the APNAP project and Heather Johnstone for the Caribbean Sea and Ceara Rise samples. Samples from the NEAPACC project were collected with NERC funding to Nick McCave, Nick Shackleton, and H.E. We thank Johannes Simstich, Frank Bassinot, Babette Hoogeker, and the Ocean Drilling Program (ODP) for core samples, Jason Day and Mervyn Greaves for laboratory assistance, and Linda

References (58)

  • MartinW.R. et al.

    CaCO3 dissolution in sediments of the Ceara Rise, western equatorial Atlantic

    Geochim. Cosmochim. Acta

    (1996)
  • MartinW.R. et al.

    Organic matter oxidation in deep-sea sediments: distribution in the sediment column and implications for calcite dissolution

    Deep-Sea Res. II

    (2006)
  • MartinP.A. et al.

    Quaternary deep sea temperature histories derived from benthic foraminiferal Mg/Ca

    Earth Planet. Sci. Lett.

    (2002)
  • MashiottaT.A. et al.

    Glacial–interglacial changes in subAntarctic sea surface temperature and δ18Owater using foraminiferal Mg

    Earth Planet. Sci. Lett.

    (1999)
  • MilleroF.J.

    The thermodynamics of the carbonate system in seawater

    Geochim. Cosmochim. Acta

    (1979)
  • MilleroF.J.

    Thermodynamics of the carbon dioxide system in the oceans

    Geochemica et Cosmochemica Acta

    (1995)
  • OomoriT. et al.

    Distribution coefficient of Mg2+ ions between calcite and solution at 10–15 °C

    Mar. Chem.

    (1987)
  • RosenthalY. et al.

    Temperature control on the incorporation of magnesium, strontium, fluorine, and cadmium into benthic foraminiferal shells from Little Bahama Bank: prospects for thermocline paleoceanography

    Geochim. Cosmochim. Acta

    (1997)
  • RussellA.D. et al.

    Effects of seawater carbonate ion concentration and temperature on shell U, Mg, and Sr in cultured planktonic foraminifera

    Geochim. Cosmochim. Acta

    (2004)
  • SchragD.P. et al.

    The oxygen isotopic composition of seawater during the Last Glacial Maximum

    Quat. Sci. Rev.

    (2002)
  • YuJ.M. et al.

    Benthic foraminiferal B/Ca ratios reflect deep water carbonate saturation state

    Earth Planet. Sci. Lett.

    (2007)
  • YuJ.M. et al.

    Seawater carbonate ion-δ13C systematics and application to glacial–interglacial North Atlantic Ocean circulation

    Earth Planet. Sci. Lett

    (2008)
  • AdkinsJ.F. et al.

    The salinity, temperature, and δ18O of the glacial deep ocean

    Science

    (2002)
  • AnandP. et al.

    Calibration of Mg/Ca thermometry in planktonic foraminifera from a sediment trap time series

    Paleoceanogr.

    (2003)
  • BarkerS. et al.

    Foraminiferal calcification response to glacial–interglacial changes in atmospheric CO2

    Science

    (2002)
  • BarkerS. et al.

    A study of cleaning procedures used for foraminiferal Mg/Ca paleothermometry

    Geochem. Geophys. Geosyst.

    (2003)
  • BarkerS. et al.

    Temporal changes in North Atlantic circulation constrained by planktonic foraminiferal shell weights

    Paleoceanogr.

    (2004)
  • BentovS. et al.

    Impact of biomineralization processes on the Mg content of foraminiferal shells: a biological perspective

    Geochem. Geophys. Geosyst.

    (2006)
  • BertramC.J. et al.

    Cadmium/calcium and carbon isotope reconstructions of the glacial Northeast Atlantic Ocean

    Paleoceanogr.

    (1995)
  • Cited by (95)

    View all citing articles on Scopus
    View full text