Extreme, abrupt Northern Hemisphere climate oscillations during the last glacial cycle (140,000 years ago to present) were modulated by changes in ocean circulation and atmospheric forcing. However, the variability of the Atlantic meridional overturning circulation (AMOC), which has a role in controlling heat transport from low to high latitudes and in ocean CO2 storage, is still poorly constrained beyond the Last Glacial Maximum. Here we show that a deep and vigorous overturning circulation mode has persisted for most of the last glacial cycle, dominating ocean circulation in the Atlantic, whereas a shallower glacial mode with southern-sourced waters filling the deep western North Atlantic prevailed during glacial maxima. Our results are based on a reconstruction of both the strength and the direction of the AMOC during the last glacial cycle from a highly resolved marine sedimentary record in the deep western North Atlantic. Parallel measurements of two independent chemical water tracers (the isotope ratios of (231)Pa/(230)Th and (143)Nd/(144)Nd), which are not directly affected by changes in the global cycle, reveal consistent responses of the AMOC during the last two glacial terminations. Any significant deviations from this configuration, resulting in slowdowns of the AMOC, were restricted to centennial-scale excursions during catastrophic iceberg discharges of the Heinrich stadials. Severe and multicentennial weakening of North Atlantic Deep Water formation occurred only during Heinrich stadials close to glacial maxima with increased ice coverage, probably as a result of increased fresh-water input. In contrast, the AMOC was relatively insensitive to submillennial meltwater pulses during warmer climate states, and an active AMOC prevailed during Dansgaard-Oeschger interstadials (Greenland warm periods).

No abstract

No abstract

No abstract

No abstract

Ga. In contrast, the U isotopic composition of MORB requires convective 1 stirring of recycled U throughout the upper mantle within 600 Ma. 2 3 'Recycling' of U from the surface to the Earth's deep interior can be monitored by a 4 decrease in the Th/U ratio of the mantle [2][3][4][5][6][7][8] . Thorium provides a valuable reference for 5 several reasons. Firstly, U and Th behave similarly as tetravalent species in the 6 mantle, such that they are difficult to fractionate significantly by melting processes. 7Only under more oxidised surface conditions do the elements show contrasting 8 behaviour, with Th remaining tetravalent and immobile during weathering, unlike 9 highly water-soluble hexavalent U species. Secondly, both Th and U are refractory, 10 lithophile elements and so the Th/U of the silicate Earth can be estimated from 11 measurements of meteorites. This planetary Th/U reference has recently been 12 refined 11 to a value of 3.876. The Th/U of the terrestrial upper mantle, as inferred 13 from analyses of MORB, is notably lower than this value; two global studies of 14 MORB yield a mean Th/U ~3.1 12,13 . The low Th/U of the upper mantle is attractively 15 explained by addition of significant recycled U from the surface and can be reconciled 16 with a surprisingly high time-integrated Th/U 14 , as gauged from 208 Pb/ 206 Pb ratios, if 17 the U recycling commenced in the latter half of Earth history 2-8 . This makes good 18 geological sense, as prior to the Great Oxidation Event (GOE) at ~2.4 Ga (e.g. see ref. 19 15) a reduced atmosphere inhibited the surface mobility of U and prevented U 20 recycling. 21 22Here, we test and extend this model of global U cycling using isotopic measurements 23 of U to complement the inferences from elemental Th/U. Recent work 16,17 has shown 24 that surface processes induce U isotopic variations (~1‰) significantly greater than 25 3 typical analytical precision (~0.05‰). Natural variations in 238 U/ 235 U are chiefly 1 linked to the reduction of U(VI) to U(IV) and the magnitudes of such fractionations 2 are inversely proportional to temperature 18,19 . So whilst U isotopic ratios can be 3 perturbed at the surface, the high temperatures and dominance of tetravalent U in the 4 mantle inhibit significant isotopic fractionations at depth. Any "exotic" 238 U/ 235 U 5 signatures, produced by low-temperature fractionation and transported into the mantle, 6should therefore provide a robust tracer of surface-processed U. 7 8To explore this potential, we have characterized, to high precision, the δ 238 U (the 9 parts per thousand difference in 238 U/ 235 U relative to a reference solution standard, 10 CRM 145) of a range of samples including: meteorites, mantle-derived basalts and the 11 inputs and outputs of an archetypal subduction zone. A summary of our results is 12 plotted in Fig 1 and for further discussion). Thus, we propose a planetary estimate based on the weighted 8 average of the two unaltered samples, with ( 234 U/ 238 U) within error of unity. This 9 yields δ 238 U = ...

No abstract

Isotopic data collected to date as part of the GEOTRACES and other programmes show that the oceanic dissolved pool is isotopically heavy relative to the inputs for zinc (Zn) and nickel (Ni). All Zn sinks measured until recently, and the only output yet measured for Ni, are isotopically heavier than the dissolved pool. This would require either a non-steady-state ocean or other unidentified sinks. Recently, isotopically light Zn has been measured in organic carbon-rich sediments from productive upwelling margins, providing a potential resolution of this issue, at least for Zn. However, the origin of the isotopically light sedimentary Zn signal is uncertain. Cellular uptake of isotopically light Zn followed by transfer to sediment does not appear to be a quantitatively important process. Here, we present Zn and Ni isotope data for the water column and sediments of the Black Sea. These data demonstrate that isotopically light Zn and Ni are extracted from the water column, probably through an equilibrium fractionation between different dissolved species followed by sequestration of light Zn and Ni in sulfide species to particulates and the sediment. We suggest that a similar, non-quantitative, process, operating in porewaters, explains the Zn data from organic carbon-rich sediments.This article is part of the themed issue 'Biological and climatic impacts of ocean trace element chemistry'.