Description: The strength of the North Atlantic Meridional Overturning Circulation during climatically highly variable Marine Isotope Stage (MIS) 3 has attracted much attention in recent years. Here we present highresolution Nd isotope compositions of past seawater derived from authigenic Fe-Mn oxyhydroxides recovered from drift sediments on the Blake Ridge in the deep western North Atlantic (ODP Leg 172, Site 1060, 3481 m water depth). The data cover the period from 45 to 35 ka BP, tracing circulation changes during major Heinrich iceberg discharge event 4 (H4, ~40-39kaBP), The Nd isotope record suggests that there was no northern source water (NSW) mass like modern NADW at the deeper part of Blake Ridge at any time between 45 and 35 ka. This is fundamentally different from the hydrographic situation during the Holocene where NADW extends below 4500 m at this location. The εNd of past deep water recorded in the Blake Ridge sediments was least radiogenic during Dansgaard/Oeschger (D/O) Interstadial (IS) 8 (εNd = -11.3)and most radiogenic immediately preceding IS 9 (εNd = -9.8). More radiogenic compositions were also recorded during H4 (-10.2 ≤ εNd ≤ 9.9). The Nd isotope variability in MIS 3 matches that of a physical bottom current strength reconstruction from the same location. Neither record follows the pattern of Northern Hemisphere D/O climatic cycles. In our record, reduced mixing with northern source waters started in stadial 12 and lasted until after H4 in stadial 9, followed by a rapid increase in NSW contribution thereafter. This major change in the Nd isotope record predates the iceberg discharge event Heinrich 4 by more than 3 ka indicating a shallowing of the water mass boundary between Glacial North Atlantic Intermediate Water and Southern Source Water beneath. This early change in bottom water properties at the deep Blake Ridge suggests that North Atlantic deep water advection may already have decreased several thousand years before the actual iceberg discharge event and associated freshening of the surface waters in the North Atlantic. The change can thus not be attributed to climatic events in the North Atlantic but may be related to changes in flux of deep water from the South.

Description: Formation of Labrador Sea Water proper commenced about 7000 years ago during the Holocene interglacial. To test whether fresher surface water conditions may have inhibited Labrador Sea Water convection during the early Holocene we measured planktonic foraminiferal (Globigerina bulloides) oxygen isotopes (d18O) and Mg/Ca ratios at Orphan Knoll (cores HU91-045-093 and MD95-2024, 3488 m) in the Labrador Sea to reconstruct shallow subsurface summer conditions (temperature and seawater d18O). Lighter foraminiferal d18O values are recorded during the early Holocene between 11000 and 7000 years ago. Part of these lighter foraminiferal d18O values can be explained by increased calcification temperatures. Reconstructed seawater d18O values were, however, still on average 0.5 per mil lighter compared with those of recent times, confirming that fresher surface waters in the Labrador Sea were probably a limiting factor in Labrador Sea Water formation during the early Holocene.

Description: Planktic and benthic foraminifera iodine (I) to calcium (Ca) molar ratios have been proposed as an exciting new proxy to assess subsurface and bottom water oxygenation in the past. Compared to trace metals, the analysis of iodine in foraminifera calcite is more challenging, as iodine is volatile in acid solution. Here we compare previous analyses that use tertiary amine with alternative analyses using tetramethylammonium hydroxide (TMAH) and ammonium hydroxide (NH4OH). In addition, we assess the effect of sample size and cleaning on planktic and benthic foraminiferal I/Ca. Our stabilization experiments with TMAH and NH4OH show similar trends as those using tertiary amine, giving relatively low I/Ca ratios for planktic and benthic foraminifera samples from locations with either poorly oxygenated subsurface waters or bottom waters, and high ratios for locations characterized by well oxygenated subsurface or bottom waters. This suggests that both alternative methods are suitable to stabilize iodine initially dissolved in acid. In terms of number of planktic foraminifera analyzed, our results show a wide spread in I/Ca when samples contain 5 to 10 specimens. Samples containing 20 specimens or more show I/Ca values centered around the average of the smaller samples, indicated that the larger sample sizes are more representative of the average planktic foraminifera sample community. The impact of cleaning on planktic and benthic foraminifera I/Ca ratios is very similar to Mg/Ca, with the largest effect occurring during the clay removal step. Largest iodine contaminations were recorded at locations characterized by moderate to high organic carbon contents. We recommend doubling the oxidative cleaning step (4 instead of 2 steps) to ensure that all organic material is removed.

Description: High resolution flow speed reconstructions of two core sites located on Gardar Drift in the northeast Atlantic Basin and Orphan Knoll in the northwest Atlantic Basin reveal a long-term decrease in flow speed of Northeast Atlantic Deep Water (NEADW) after 6,500 years. Benthic foraminiferal oxygen isotopes of sites currently bathed in NEADW show a 0.2per mil depletion after 6,500 years, shortly after the start of the development of a carbon isotope gradient between NEADW and Norwegian Sea Deep Water. We consider these changes in near-bottom flow vigor and benthic foraminiferal isotope records to mark a significant reorganization of the Holocene deep ocean circulation, and attribute the changes to a weakening of NEADW flow during the mid to late Holocene that allowed the shoaling of Lower Deep Water and deeper eastward advection of Labrador Sea Water into the northeast Atlantic Basin.

Description: During the last and penultimate glacial maxima, atmospheric CO2 concentrations were lower than present, possibly in part because of increased storage of respired carbon in the deep oceans (Sigman and Boyle, 2000, doi:10.1038/35038000). The amount of respired carbon present in a water mass can be calculated from its oxygen content through apparent oxygen utilization; the oxygen content can in turn be calculated from the carbon isotope gradient within the sediment column (McCorkle and Emerson, 1988, doi:10.1016/0016-7037(88)90270-0). Here we analyse the shells of benthic foraminifera occurring at the sediment surface and the oxic/anoxic interface on the Portuguese Margin to reconstruct the carbon isotope gradient and hence bottom-water oxygenation over the past 150,000 years. We find that bottom-water oxygen concentrations were 45 and 65 µmol/kg lower than present during the last and penultimate glacial maxima, respectively. We calculate that concentrations of remineralized organic carbon were at least twice as high as today during the glacial maxima. We attribute these changes to decreased ventilation linked to a reorganization of ocean circulation (McManus et al., 2004, doi:10.1038/nature02494) and a strengthened global biological pump (Kohfeld et al., 2005, doi:10.1126/science.1105375). If the respired carbon pool was of a similar size throughout the entire glacial deep Atlantic basin, then this sink could account for 15 and 20 per cent of the glacial PCO2 drawdown during the last and penultimate glacial maxima.