Sedimentary Fe–Mn oxyhydroxides as paleoceanographic archives and the role of aeolian flux in regulating oceanic dissolved REE

In the marine environment, authigenic ferromanganese (Fe–Mn) oxides precipitate from seawater, incorporating many dissolved trace elements (such as Nd, Pb) whose isotopic composition is a direct proxy for ambient seawater. Thus, isotopic studies of the Fe–Mn oxyhydroxide component leached from sedimentary records can provide information on changes in erosional input or ocean circulation changes. Such studies may be complicated, however, by the presence of detrital material containing a Fe–Mn oxide component, which may mask the true ‘seawater’ signal. In addition, the formation of Fe–Mn oxyhydroxides in the marine environment remains poorly understood. In particular, the phase (carbonates, detrital particles) that controls the delivery of authigenic Fe–Mn oxides to the sediments has not yet been determined. In this study, we have analysed the REE and Nd isotopic compositions of Fe–Mn oxyhydroxide fractions dispersed in marine sediment cores from the Cape and Angola basins, in the SE Atlantic (cruise IMAGES II). For the Angola Basin deep-sea core, located at 1000 km south of the Congo River mouth, both the REE and Nd isotopic compositions (εNd) measured in recent Fe–Mn oxyhydroxide fractions are typical of Congo River-borne Fe–Mn oxides. This shows that ‘preformed’ Fe–Mn oxides associated with detrital fractions can locally contaminate the ‘seawater’ signal recorded by ‘authigenic’ oxyhydroxides. By contrast, examination of REE distributions, Nd isotope data and mass accumulation rates in Cape Basin sediments shows paradoxically that the flux of Fe–Mn oxides to sediments is controlled by aeolian particles from the nearby Namib Desert, even though their REE and εNd compositions point clearly to an ‘authigenic’ origin. It is proposed that partial dissolution of aeolian dust occurs in the water column, releasing into solution its easily leachable Fe–Mn component. The dissolved Fe2+ and Mn2+ would then fuel the reprecipitation of authigenic Fe–Mn oxyhydroxides, scavenging additional dissolved trace elements from the ocean, most probably from deep water masses. In this case, calculations suggest that aeolian deposition acts as a net sink for dissolved Nd in the Cape Basin, rather than a source, and allows us to estimate the rate of removal of dissolved Nd associated with atmospheric deposition, yielding a global oceanic residence time for Nd (τNd) of between 500 and 1400 years