Last Glacial Maximum
Enqing Huang,1, 2 Luke C. Skinner,3 Stefan Mulitza,1 André Paul,1 and Michael Schulz1
With 3 Figures
Although the volume and the distribution of water masses in the glacial Atlantic has been well established using benthic foraminiferal δ13C and Cd/Ca, no robust results have been obtained regarding the ventilation age of glacial deep Atlantic. The 14C age of a water mass has been suggested as promising proxy for the deep ocean circulation rate. The preformed
14C of deep waters would decay at a known rate once they are isolated from the atmosphere and being transferred into the ocean interior. In practice, however, the application of B–P age (14C age difference between paired planktonic and benthic foraminifera) or deep water Δ14C for reconstructing past ocean circulation age has long been hindered for several reasons.
First, surface reservoir ages at core locations could differ from those at deep water formation areas, and both kinds of reservoir ages are temporally variable. Second, deep water at a given location and depth in the Atlantic is usually a mixture of water masses from several sources with different initial 14C ages, and the mixing ratio is also temporally variable. Only when wa-ter-mass sources and their initial ages can be reconstructed, and the mixing ratio of different water masses at a given location and water depth can be estimated by using an independent conservative water-mass tracer, the ocean circulation age could then be inferred.
We combine existing results with newly measured 14C ages of paired benthic and plank-tonic foraminifera from 28 sediment cores (Fig. 1) to map the seawater Δ14C distribution at the Last Glacial Maximum (LGM, 24 –18 ka BP) in the Atlantic Ocean. We further attempt to calculate the circulation age of the glacial Atlantic through assuming seawater δ13C as a largely pseudo-conservative tracer in the LGM Atlantic.
In the upper 1500 m the mean ΔΔ14C value (the Δ14C difference between deep water Δ14C and the contemptuous atmospheric Δ14C) of all LGM reconstructions is –170 ‰, which is 65 ‰ less than the pre-bomb mean value (Fig. 2). Below 1500 m, the majority of the LGM ΔΔ14C reconstructions range between –220 ‰ and –310 ‰. The mean value of all LGM ΔΔ14C reconstructions below 1500 m is –275 ‰, nearly 190 ‰ less than the pre-bomb value.
Taken together, the LGM ΔΔ14C above 1500 m and below 1500 m show a much stronger gra-dient relative to that of the pre-bomb era.
1 MARUM – Center for Marine Environmental Sciences and Faculty of Geosciences, University of Bremen, Bre-men, Germany.
2 School of Ocean and Earth Science, Tongji University, Shanghai (China).
3 Godwin Laboratory for Palaeoclimate Research, Department of Earth Sciences, University of Cambridge, Cam-bridge (UK).
Enqing Huang, Luke C. Skinner, Stefan Mulitza, André Paul, and Michael Schulz
102 Nova Acta Leopoldina NF 121, Nr. 408, 101–105 (2015)
Fig. 1 Vertical distribution of the Atlantic sediment cores with the LGM seawater radiocarbon reconstructions. Me-ridional sections of seawater Δ14C (after correction of the bomb effect) along the western and the eastern Atlantic are derived from the gridded GLODAP dataset. The position of the two sections is indicated in enclosed panels.
Circulation ages of the LGM Atlantic are estimated based on three assumptions. First, the ma-jority of the LGM Atlantic was filled with waters from three sources, i.e. glacial North Atlan-tic Deep Water (GNADW), glacial AntarcAtlan-tic Intermediate Water (GAAIW) and glacial Ant-arctic Bottom Water (GAABW). Second, δ13C and ΔΔ14C end-member values of GNADW, GAAIW and GAABW were constant or showed limited changes over a few thousand years prior to the LGM and within the LGM. The initial 14C age of GNADW is taken from the sim-ulation output (Butzin et al. 2005, Franke et al. 2008), while that of GAAIW is based on deep water coral reconstructions from the Drake Passage (Burke and Robinson 2012). The initial 14C age of GAABW is the mean value of reconstructions from cores MD07-3076 and TNO57-21 (Barker et al. 2010, Skinner et al. 2010). Third, we assume that seawater δ13C can be considered as largely pseudo-conservative tracer in the Atlantic during the LGM. This assumption is only justified if the replenishment of deep waters was accomplished relatively rapidly, such that the remineralization of sinking organic particles did not significantly alter the seawater δ13C signature.
When plotting all results in a δ13C-ΔΔ14C diagram (Fig. 3), we find that the glacial AAIW values happen to lie on the mixing line between the GNADW and GAABW. Except for a few outliers, the vast majority of reconstructions fall along the mixing line within uncertainties.
Radiocarbon Distribution and Radiocarbon-Based Circulation Age of the Atlantic Ocean
Nova Acta Leopoldina NF 121, Nr. 408, 101–105 (2015) 103
This strongly suggests that the reconstructed ΔΔ14C signals hold significant information on mixing but only little information on circulation ages. The current dataset suggest that circu-lation ages of the majority of the deep Atlantic (> 1500 m) are less than 400 years during the LGM (Fig. 3), which is comparable to or even slightly less than pre-bomb value of 200 – 400 years. This implies a fast glacial deep water circulation rate, which also supports our assump-tion of taking δ13C as a pseudo-conservative tracer.
A strong ocean circulation during the LGM could actually explain a large number of observations. The chemical gradient between the upper and the lower Atlantic, shown by the vertical distribution of seawater δ13C, Cd/Ca and ΔΔ14C data, is the most robust characteristic of the glacial Atlantic. To sustain such a strong gradient, besides that end-member values of the LGM AABW and NADW are significantly different, either a strong export of the LGM AABW and NADW or a very limited mixing across the NADW/AABW boundary or a com-bination of both is also required. Indeed, the transport to vertical diffusivity ratio of AABW in the southwestern Atlantic is estimated to be an order of magnitude larger during the LGM than the Holocene (Hoffman and Lund 2012). Given that the vertical mixing in the LGM Ocean interior is probably strengthened due to the generally enhanced wind field and the greater tidal mixing in deep basins (Wunsch 2003), a strengthening of the water mass export is a more likely explanation for the presence of the chemical gradient.
Fig. 2 Comparison of the pre-bomb ΔΔ14C values (A and C) with the LGM reconstructions (B and D) at a set of Atlantic core locations. The LGM ΔΔ14C value at each other is the mean result of several samples, and the associated uncertainty represents ±1 sigma standard error. Question marks indicate data excluded from the final analyses. The LGM water depths of all sediment cores are adjusted by –120 m.
Enqing Huang, Luke C. Skinner, Stefan Mulitza, André Paul, and Michael Schulz
104 Nova Acta Leopoldina NF 121, Nr. 408, 101–105 (2015)
References
Barker, S., Knorr, G., Vautravers, M. J., Diz, P., and Skinner, L. C.: Extreme deepening of the Atlantic over-turning circulation during deglaciation. Nature Geosci. 3, 567–571 (2010)
Burke, A., and Robinson, L. F.: The Southern Ocean’s role in carbon exchange during the last deglaciation. Science 335, 557–561 (2012)
Butzin, M., Prange, M., and Lohmann, G.: Radiocarbon simulations for the glacial ocean: The effects of wind stress, Southern Ocean sea ice and Heinrich events. Earth Planet. Sci. Lett. 235, 45 – 61 (2005)
Franke, J., Paul, A., and Schulz, M.: Modeling variations of marine reservoir ages during the last 45000 years.
Clim. Past 4, 125 –136 (2008)
Fig. 3 The δ13C-ΔΔ14C diagram of the LGM bottom waters at a set of Atlantic sediment cores and water masses from three end members. Blue and red diamonds are data from the western and the eastern Atlantic, respectively.
Black diamonds denote sediment cores located within the ventilated thermocline, where bottom waters are possibly influenced by subducted waters from other sources besides the GNADW and GAAIW. Error bars represent ±1 sigma standard deviation for δ13C and ±1 sigma standard error for ΔΔ14C. The green line is the mixing line between the GNADW and GAABW. Dashed lines represent contours with the constant radiocarbon age offset from the mixing line when assuming that δ13C is a conservative tracer.
Radiocarbon Distribution and Radiocarbon-Based Circulation Age of the Atlantic Ocean
Nova Acta Leopoldina NF 121, Nr. 408, 101–105 (2015) 105
Hoffman, J. L., and Lund, D. C.: Refining the stable isotope budget for Antarctic Bottom Water: New foraminife-ral data from the abyssal southwest Atlantic. Paleoceanography 27, PA1213; doi:10.1029/2011PA002216 (2012) Skinner, L. C., Fallon, S., Waelbroeck, C., Michel, E., and Barker, S.: Ventilation of the deep southern ocean
and deglacial CO2 rise. Science 328, 1147–1151 (2010)
Wunsch, C.: Determining paleoceanographic circulations, with emphasis on the Last Glacial Maximum. Quat. Sci.
Rev. 22, 371–385 (2003)
Enqing Huang
Room 316
School of Ocean and Earth Science Tongji University
1239 Siping Road 200092, Shanghai China
E-Mail: ehuang@tongji.edu.cn
Nova Acta Leopoldina NF 121, Nr. 408, 107–110 (2015)
107