Radionuclides measured in surface water samples from the South Atlantic

In the nutrient-rich Southern Ocean, Fe is a vital constituent controlling the growth of phytoplankton. Despite much effort, the origin and transport of Fe to the oceans are not well understood. In this study we address the issue with geochemical data and Nd isotopic compositions of suspended particle samples collected from 1997 to 1999 in the South Atlantic Sector of the Southern Ocean. Al, Th, and rare earth element (REE) concentrations as well as 143Nd/144Nd isotopic ratios in acetic acid-leached particle samples representing the lithogenic fraction delineate three major sources: (1) Patagonia and the Antarctic Peninsula provide material with eNd > -4 that is transported toward the east with the polar and subpolar front jets, (2) the south African shelf, although its influence is limited by the circumpolar circulation and wind direction, can account for material with eNd of -12 to -14 adjacent to South Africa, and (3) East Antarctica provides material with eNd of -10 to -15 to the eastern Weddell Sea and adjacent Antarctic Circumpolar Current. For this region we interpret the Nd isotopic evidence in combination with oceanographic/atmospheric constraints as evidence for supply of significant amounts of terrigenous detritus by icebergs.

Beryllium concentrations and deposition rates in ODP holes from the north Atlantic

Increases in the production rate of cosmogenic radionuclides associated with geomagnetic excursions have been used as global tie-points for correlation between records of past climate from marine and terrestrial archives. We have investigated the relative timing of variations in 10Be production rate and the corresponding palaeomagnetic signal during one of the largest Pleistocene excursions, the Iceland Basin (IB) event (ca. 190 kyr), as recorded in two marine sediment cores (ODP Sites 1063 and 983) with high sedimentation rates. Variations in 10Be production rate during the excursion were estimated by use of 230Thxs normalized 10Be deposition rates and authigenic 10Be/9Be. Resulting 10Be production rates are compared with high-resolution records of geomagnetic field behaviour acquired from the same discrete samples. We find no evidence for a significant lock-in depth of the palaeomagnetic signal in these high sedimentation-rate cores. Apparent lock-in depths in other cores may sometimes be the result of lower sample resolution. Our results also indicate that the period of increased 10Be production during the IB excursion lasted longer and, most likely, started earlier than the corresponding palaeomagnetic anomaly, in accordance with previous observations that polarity transitions occur after periods of reduced geomagnetic field intensity prior to the transition. The lack of evidence in this study for a significant palaeomagnetic lock-in depth suggests that there is no systematic offset between the 10Be signal and palaeomagnetic anomalies associated with excursions and reversals, with significance for the global correlation of climate records from different archives.

Observation of manganese nodules and crusts recovered during the CHAIN 115 cruise from Nov 1973 until June 1974 in the Atlantic Ocean

Investigations of piston cores from the Vema Channel and lower flanks of the Rio Grande Rise suggest the presence of episodic flow of deep and bottom water during the Late Pleistocene. Cores from below the present-day foraminiferal lysocline (at ~4000 m) contain an incomplete depositional record consisting of Mn nodules and encrustations, hemipelagic clay, displaced high-latitude diatoms, and poorly preserved heterogeneous microfossil assemblages. Cores from the depth range between 2900 m and 4000 m contain an essentially complete Late Pleistocene record, and consist of well-defined carbonate dissolution cycles with periodicities of ~100,000 years. Low carbonate content and increased dissolution correspond to glacial episodes, as interpreted by oxygen isotopic analysis of bulk foraminiferal assemblages. The absence of diagnostic high-latitude indicators (Antarctic diatoms) within the dissolution cyclss, however, suggests that AABW may not have extended to significantly shallower elevations on the lower flanks of the Rio Grande Rise during the Late Pleistocene. Therefore episodic AABW flow may not necessarily be the mechanism responsible for producing these cyclic events. This interpretation is also supported by the presence of an apparently complete Brunhes depositional record in the same cores, suggesting current velocities insufficient for significant erosion. Fluctuations in the properties and flow characteristics of another water mass, such as NADW, may be involved. The geologic evidence in core-top samples near the present-day AABW/NADW transition zone is consistent with either of two possible interpretations of the upper limit of AABW on the east flank of the channel. The foraminiferal lysocline, at ~4000 m, is near the top of the benthic thermocline and nepheloid layer, and may therefore correspond to the upper limit of relatively corrosive AABW. On the other hand, the carbonate compensation depth (CDD) at ~4250 m, which corresponds to the maximum gradient in the benthic thermocline, is characterized by rapid deposition of relatively fine-grained sediment. Such a zone of convergence and preferential sediment accumulation would be expected near the level of no motion in the AABW/NADW transition zone as a consequence of Ekman-layer veering of the mean velocity vector in the bottom boundary layer. It is possible that both of these interpretations are in part correct. The "level of no motion'' may in fact correspond to the CCD, while at the same time relatively corrosive water of Antarctic origin may mix with overlying NADW and therefore elevate the foraminifera] lysocline to depths above the level of no motion. Closely spaced observations of the hydrography and flow characteristics within the benthic thermocline will be required in order to use sediment parameters as more precise indicators of paleo-circulation.