Auxiliary Material for
Effects of mid-latitude westerlies on the paleoproductivity at the Agulhas Bank slope
during the penultimate glacial cycle: evidence from coccolith Sr/Ca ratios
Luz María Mejía1*, Patrizia Ziveri2, Marilisa Cagnetti2, Clara Bolton1, Rainer Zahn,
Gianluca Marino2,3, Heather Stoll1
(1Dept. Geología, Universidad de Oviedo, Arias de Velasco s/n, 33005 Oviedo,
Asturias, Spain. 2 Institut de Ciència i Tecnologia Ambientals, Universitat Autònoma de
Barcelona, 08193 Bellaterra, Catalonia, Spain. 3 Now at Research School of Earth
Sciences, The Australian National University, Canberra 0200, Australia)
Paleoceanography, 2014
Introduction
The supplement illustrates benthic 18O records from core MD96-2080 in the Agulhas
Bank slope, GeoB-3603-2 in Cape Basin, MD02-2588 in the Agulhas Plateau and
MD96-2048 in the South Indian Gyre to show its structural fit for data comparison. The
supplement provides detailed information on the combination of Sr/Ca temperature
dependencies of the main carbonate contributors in the assemblage to a sole regression
applied to conduct temperature correction of measured Sr/Ca. It also shows the weak
correlation between temperature-corrected Sr/Ca and the carbonate contribution from
Sr-rich species in the assemblage (Calcidiscus leptoporus and Helicosphaera carteri),
used to isolate the productivity signal from the maximum possible contribution of
assemblage effects. The supplement illustrates a detailed picture of the temperature and
assemblage-corrected productivity signal, and shows its confidence intervals obtained
from the error propagation analysis conducted using the Monte-Carlo simulation
approach, which considered analytical temperature and Sr/Ca errors, quantification of
species and uncertainties of temperature and assemblage dependencies. This data also
illustrates the temporal variability of the offset between coccolith and Globigerina
bulloides 13C (coccolith) together with the productivity variability derived from coccolith
Sr/Ca as discussed in the main text section 4.1, both from core MD96-2080. Crossspectral analysis indicating coherence between obliquity and ice-rafted debris (IRD)
deposition in the Agulhas Plateau (core MD02-2588) during the last 265 kyr at the 95%
confidence interval in the obliquity band is also shown. Sr/Ca ratios and 13C from the
coccolith fraction were measured at the Universidad de Oviedo in 2009 and 2013,
respectively, while 13C from G. bulloides was measured on 2008 at the Universidad de
Barcelona. Assemblage identification and quantification was conducted at the
Universitat Autònoma de Barcelona in 2009. Benthic 18O records are given in figure
“fs01.pdf”, the combined Sr/Ca dependency on temperature obtained from individual
regressions of the main carbonate contributors are given in figure “fs02.pdf”, the weak
correlation between temperature-corrected Sr/Ca and carbonate contribution of Sr-rich
species is shown in “fs03.pdf”, the coccolith Sr/Ca productivity record including its
propagated errors as obtained from the Monte-Carlo simulation is given in “fs04.pdf”,
coccolith productivity and the offset between coccolith and G. bulloides 13C are given
in figure “fs05.pdf” and cross-spectral analysis from obliquity and IRD deposition are
given in “fs06.pdf”.
1. fs01.pdf Benthic 18O records from core MD96-2080 in the Agulhas Bank slope (red)
[Martínez-Méndez et al., 2008], core GeoB-3603-2 in Cape Basin (blue) [Peeters et al.,
2004], core MD02-2588 in the Agulhas Plateau (gray) [Ziegler et al., 2013], and core
MD96-2048 in the South Indian Gyre (yellow) [Caley et al., 2011]. The age scales of
published data from cores GeoB-3603-2 and MD96-2048 were slightly readjusted to
improve the alignment between the four different 18O records. Grey-shaded areas mark
the interglacial stages MIS 7 and MIS 5 and green vertical lines mark boundaries
between stages.
2. fs02.pdf Combined Sr/Ca dependency on temperature obtained from the regression of
the main carbonate contributors (C. leptoporus, H. carteri, G. oceanica and C.
braarudii) (black line), based on culture data of [Stoll et al., 2002; Müller et al., 2014]
after elimination of offsets in absolute Sr/Ca among species as described in the text of
appendix B.
3. fs03.pdf Weak positive correlation between temperature-corrected Sr/Ca and the
carbonate contribution of the Sr-richest species in the assemblage (C. leptoporus + H.
carteri) (r = 0.377; p = 5.7 x10-5; n = 108; equation: y = 0.0036 x + 1.844).
4. fs04.pdf Coccolith Sr/Ca productivity record from core MD96-2080 illustrating
confidence intervals derived from Monte-Carlo simulations as described in section 4.1.2
of the main text. Black data points illustrate direct application of the temperature and
assemblage correction to Sr/Ca measurements with no error propagation. Analytical
errors of temperature and Sr/Ca measurements, quantification of species and
uncertainties of the temperature and assemblage dependencies were included in the
Monte-Carlo analysis. Red line is the 50% confidence interval, blue lines include the
20-80% confidence intervals and the blue shaded area covers the 20-80% confidence
intervals when applying a 3 point running average to highlight the long term
productivity trend discussed in this paper.
5. fs05.pdf a) Coccolith Sr/Ca productivity record from core MD96-2080, as illustrated
in Figure 3f. b) Difference of 13C of the coccolith fraction and G. bulloides (coccolith) in
core MD96-2080 during the period of study showing increasing differences towards
cold glacial periods and decreasing values during interglacials. 5 point running average
is shown. Grey-shaded areas and green vertical lines as in Figure A1. Color coding of
the sample points corresponds to the % carbonate contribution from larger coccoliths C.
leptoporus and H. carteri.
6. fs06.pdf Cross-spectral analysis between obliquity and ln(IRD) at core site MD022588, over the interval 0-265 kyr, performed using ARAND software [Howell,
http://www.ncdc.noaa.gov/paleo/softlib/arand/arand.html, 2001]. IRD data are
originally published in Marino et al. [2013] and Simon et al. [2013]. Because of the
non-Gaussian distribution of the IRD data, we normalized the IRD record for crossspectral analysis by taking its natural logarithm. IRD data were then interpolated to an
even age scale (delta t = 0.712 kyr). Obliquity was computed using Analyseries
(Paillard et al. [1996], with astronomical solutions of Laskar et al. [2004]), and was
multiplied by -1 so that obliquity minima coincide with ln(IRD) maxima. Obliquity and
ln(IRD) are coherent at the 95% confidence interval in the obliquity band, with
obliquity minima leading ln(IRD) maxima by 2.9 (±2.8) kyr.
REFERENCES
Caley, T. et al. (2011), High-latitude obliquity as a dominant forcing in the Agulhas
current system, Climate of the Past, 7(4), 1285–1296, doi:10.5194/cp-7-12852011.
Laskar, J., P. Robutel, F. Joutel, M. Gastineau, A. C. M. Correia, and B. Levrard (2004),
A long-term numerical solution for the insolation quantities of the Earth,
Astronomy and Astrophysics, 428(1), 261–285, doi:10.1051/0004-6361:20041335.
Marino, G., R. Zahn, M. Ziegler, C. Purcell, G. Knorr, I. R. Hall, P. Ziveri, and H.
Elderfield (2013), Agulhas salt-leakage oscillations during abrupt climate changes
of the Late Pleistocene, Paleoceanography, 28, 1–8, doi:10.1002/palo.20038.
Martínez-Méndez, G., R. Zahn, I. R. Hall, L. D. Pena, and I. Cacho (2008), 345,000year-long multi-proxy records off South Africa document variable contributions of
Northern versus Southern Component Water to the Deep South Atlantic, Earth and
Planetary Science Letters, 267(1), 309–321, doi:10.1016/j.epsl.2007.11.050.
Müller, M. N., M. Lebrato, U. Riebesell, J. Barcelos e Ramos, K. G. Schulz, S. BlancoAmeijeiras, S. Sett, A. Eisenhauer, and H. M. Stoll (2014), Influence of
temperature and CO2 on the strontium and magnesium composition of
coccolithophore calcite, Biogeosciences, 11(4), 1065–1075, doi:10.5194/bg-111065-2014.
Paillard, D., L. Labeyrie, and P. Yiou (1996), Macintosh Program performs time-series
analysis, Eos, Transactions American Geophysical Union, 77(39), 379,
doi:10.1029/96EO00259.
Peeters, F. J. C., R. Acheson, G.-J. A. Brummer, W. P. M. De Ruijter, R. R. Schneider,
G. M. Ganssen, E. Ufkes, and D. Kroon (2004), Vigorous exchange between the
Indian and Atlantic oceans at the end of the past five glacial periods., Nature,
430(7000), 661–5, doi:10.1038/nature02785.
Simon, M. H., K. L. Arthur, I. R. Hall, F. J. C. Peeters, B. R. Loveday, S. Barker, M.
Ziegler, and R. Zahn (2013), Millennial-scale Agulhas Current variability and its
implications for salt-leakage through the Indian–Atlantic Ocean Gateway, Earth
and Planetary Science Letters, 383, 101–112, doi:10.1016/j.epsl.2013.09.035.
Stoll, H. M., C. M. Klaas, I. Probert, J. Ruiz Encinar, and J. I. Garcia Alonso (2002),
Calcification rate and temperature effects on Sr partitioning in coccoliths of
multiple species of coccolithophorids in culture, Global and Planetary Change,
34(3), 153–171, doi:10.1016/S0921-8181(02)00112-1.
Ziegler, M., P. Diz, I. R. Hall, and R. Zahn (2013), Millennial-scale changes in
atmospheric CO2 levels linked to the Southern Ocean carbon isotope gradient and
dust flux, Nature Geoscience, 6(6), 457–461, doi:10.1038/ngeo1782.