Deep water provenance and dynamics of the (de)glacial Atlantic meridional overturning circulation
Introduction
Today, the Atlantic Meridional Overturning Circulation (AMOC) accounts for ∼50% of the planetary poleward advection of heat and contributes to the removal of CO2 from the surface ocean through the formation of nutrient-depleted North Atlantic Deep Water (NADW) (Sabine et al., 2004). A change in the relative volumetric contribution of North Atlantic- versus Southern Ocean-sourced waters during the last deglaciation is considered to have significantly impacted the efficiency of the marine soft tissue pump and associated alkalinity feedbacks, thereby directly altering the partitioning of CO2 between the ocean interior and the atmosphere (Jaccard et al., 2016).
Studies using the non-conservative nutrient-tracers and Cd/Ca in benthic foraminifera or radiocarbon ventilation ages have provided a wealth of evidence that the water mass distribution in the glacial North Atlantic was significantly different from the modern geometry (e.g. Curry and Oppo, 2005). Today NADW occupies most of the water column in the western North Atlantic. During the Last Glacial Maximum (LGM) waters below about 2500 to 3000 m were dominated by nutrient-rich deep waters likely originating from the Southern Ocean (Southern Component Water, SCW), while the upper limb was bathed by a relatively fresher water mass referred to as Glacial North Atlantic Intermediate Water (GNAIW) (Lynch-Stieglitz and Fairbanks, 1994). During the deglaciation the GNAIW depth distribution evolved into the modern situation of NADW, dominating the North Atlantic down to ∼5000 m water depth. However, reconstruction of the exact spatial and temporal sequence of events characterising deglacial AMOC mode changes is not straightforward and remains a matter of debate.
Our current understanding of the glacial, deglacial and Holocene evolution of Atlantic Ocean circulation patterns is mainly based on paleoceanographic reconstructions gleaned from carbon isotopes, in particular benthic foraminifera . While benthic foraminifera-derived provide useful information about the geometry of subsurface circulation patterns, it only provides incomplete constraints about flow rates. Furthermore, the benthic signal is influenced by a number of factors, complicating its interpretation. The average of the global ocean is a function of changes in the amount of carbon stored on land and is affected by changes in the sinks and sources of carbon reservoirs. Furthermore, may be sensitive to the balance of photosynthesis versus respiration, microhabitat conditions and CO2 air–sea gas exchange, because the equilibration timescale in the surface ocean is proportional to the ratio of dissolved inorganic carbon to CO2 (Galbraith et al., 2015).
Seawater-derived 231Pa/230Th and Nd isotopes provide two independent, yet complementary proxies of past ocean circulation, which are insensitive to variations in the global carbon cycle. While the first proxy, based on the sedimentary distribution of protactinium/thorium isotopes (231Pa/230Th), provides quantitative information about the strength and the dynamics of overturning circulation (McManus et al., 2004), the second proxy, based on the authigenic neodymium isotopic composition (143Nd/144Nd), allows the fingerprinting of water mass provenance and therefore constraining flow paths throughout the deep ocean (e.g. Frank, 2002, Piotrowski et al., 2004).
Although, the spatial and temporal coverage of and 231Pa/230Th data sets from the last glacial cycle is continuously improving, the lack of direct comparison between proxies of circulation strength, water mass provenance and nutrients still does not provide a consistent picture of glacial and deglacial AMOC changes. There is a coherent picture of a glacial deep Atlantic dominated by SCW, which was gradually replaced by NCW during the deglacial and the Holocene (e.g. Gutjahr et al., 2008, Piotrowski et al., 2004) indicated by Nd-isotope signatures in agreement with findings based on carbon isotopes (e.g. Curry and Oppo, 2005, Sarnthein et al., 1994). Compilations of 231Pa/230Th data sets (Bradtmiller et al., 2014, Gherardi et al., 2009, Lippold et al., 2012b) found generally higher values in the deep Atlantic and unchanged or lower values in shallower waters compared to the Holocene. These studies suggest that the shallower circulation cell exhibited stronger overturning than the deeper cell during the LGM and are also consistent with the interpretation of a still active but reduced net ocean overturning during Heinrich Stadial 1 (HS1). The first publication presenting combined 231Pa/230Th (McManus et al., 2004) and (Roberts et al., 2010) from two neighbouring sediment cores from the Bermuda Rise (GGC5 and GGC6) revealed a high degree of synchronicity between the changes in water mass provenance and circulation strength across the last glacial termination. More recently this approach was extended, in another adjacent Bermuda Rise sediment core from a similar water depth (ODP1063, Table 1), covering the last 140 ka (Böhm et al., 2015). These studies found a gradual transition from a SCW dominated deep North-West Atlantic during the LGM into a Northern Component Water (NCW) dominated vigorous circulation state. This active circulation prevailed for most of the last glacial cycle with a continuous strong contribution of NCW in the deep Atlantic as indicated from Nd isotope signatures, while the domination of sluggishly ventilated SCW was restricted to time ranges around the peak glacial. Although temporally highly resolved, these reconstructions were derived from a single location and thus may be subject to local obfuscating factors. Here we present combined measurements of 231Pa/230Th and 143Nd/144Nd from two published and six new data sets, spanning a latitudinal transect across the entire Atlantic, to infer past changes in both the strength and geometry of the AMOC back to the LGM. Particular attention is paid to consistencies and discrepancies between the combined proxy records and the presence or absence of gradients between neighbouring core sites. Our results thus aim to provide an updated illustration of the potential of this combined multi-core 231Pa/230Th and isotope approach.
Section snippets
Overview of core locations, measurements and age controls
The eight sediment core locations cover a meridional transect through the North and South Atlantic Ocean with a wide range of water depths and hydrological settings from 42°N to 40°S (Table 1). The six new combined profiles are measured from sediment cores featuring well-established, published age models (supplementary material, S4), with the exception of GeoB1523-1, for which a chronology was determined by comparison of its down-core record to the adjacent 14C dated core GeoB1515-1 (Vidal
Results and discussion
Six new combined 231Pa/230Th and down-core profiles represent a considerable improvement in the available data bases of both proxies. We therefore present a short outline of the new results from each core before discussing what can be deduced from a comprehensive compilation of all data sets along with the literature records.
In order to provide support of the applications 231Pa/230Th results are also presented as a function of opal concentrations in section 3.6. For Nd isotopes the
Conclusions
We observe no basin wide correlation of the 231Pa/230Th with the abundance of opal, with the exception of the southernmost core (ODP Site 1089). For all the northern locations the generally robust correlation of with opal concentrations suggests that opal is better preserved at core locations bathed by SCW. Hence, co-variation of 231Pa/230Th with opal does not necessarily indicate a causal link (i.e., high opal causes high 231Pa/230Th) but instead results from a change in the water mass
Acknowledgments
This project was funded by DFG grant Li1815/2. J. Lippold was further supported by the FP7-PEOPLE-2013-IEF, Marie Curie proposal 622483. We thank the ODP core repository and the MARUM GeoB Core Repository in Bremen for dedicated support. Stefan Mulitza was supported through the DFG Research Center/Cluster of Excellence EXC 309 “The Ocean in the Earth System”. E. Böhm acknowledges support from the European Research Council grant ACCLIMATE/no 339108. S.L. Jaccard and O. Cartapanis were supported
References (60)
- et al.
Variations of biogenic particle flux in the southern Atlantic section of the Subantarctic zone during the late Quaternary: evidence from sedimentary 231Paex and 230Thex
Mar. Geol.
(1999) - et al.
The influence of particle composition and particle flux on scavenging of Th, Pa and Be in the ocean
Earth Planet. Sci. Lett.
(2002) - et al.
Protactinium-231, a new radionuclide for AMS
Nucl. Instrum. Methods Phys. Res., Sect. B, Beam Interact. Mater. Atoms
(2007) - et al.
231Pa/230Th: a proxy for upwelling off the coast of West Africa
Nucl. Instrum. Methods Phys. Res., Sect. B, Beam Interact. Mater. Atoms
(2010) - et al.
Neodymium isotopic composition of deep-sea corals from the NE Atlantic: implications for past hydrological changes during the Holocene
Quat. Sci. Rev.
(2010) - et al.
Progress in determination of 227-Ac in sea water
Mar. Chem.
(2008) - et al.
Reliable extraction of a deepwater trace metal isotope signal from Fe–Mn oxyhydroxide coatings of marine sediments
Chem. Geol.
(2007) - et al.
Tracing the Nd isotope evolution of North Atlantic Deep and Intermediate Waters in the western North Atlantic since the LGM from Blake Ridge sediments
Earth Planet. Sci. Lett.
(2008) - et al.
Intensity of Th and Pa scavenging partitioned by particle chemistry in the North Atlantic Ocean
Mar. Chem.
(2015) Concentration and isotopic composition of neodymium in the South Atlantic Ocean
Earth Planet. Sci. Lett.
(1993)