60 Myr records of major elements and Pb-Nd isotopes from hydrogenous ferromanganese crusts: Reconstruction of seawater paleochemistry
We compare the time series of major element geochemical and Pb- and Nd-isotopic composition obtained for seven hydrogenous ferromanganese crusts from the Atlantic, Indian, and Pacific Oceans which cover the last 60 Myr. Average crust growth rates and age-depth relationships were determined directly for the last about 10 Myr using 10Be/9Be profiles. In the absence of other information these were extrapolated to the base of the crusts assuming constant growth rates and constant initial 10Be/9Be ratios due to the lack of additional information. Co contents have also been used previously to estimate growth rates in Co-rich Pacific and Atlantic seamount crusts (Puteanus and Halbach, 1988). A comparison of 10Be/9Be- and Co-based dating of three Co-rich crusts supports the validity of this approach and confirms the earlier chronologies derived from extrapolated 10Be/9Be-based growth rates back to 60 Ma. Our data show that the flux of Co into Co-poor crusts has been considerably lower. The relationship between growth rate and Co content for the Co-poor crusts developed from these data is in good agreement with a previous study of a wider range of marine deposits (Manheim, 1986). The results suggest that the Co content provides detailed information on the growth history of ferromanganese crusts, particularly prior to 10-12 Ma where the 10Be-based method is not applicable. The distributions of Pb and Nd isotopes in the deep oceans over the last 60 Myr are expected to be controlled by two main factors: (a) variations of oceanic mixing patterns and flow paths of water masses with distinct isotopic signatures related to major paleogeographic changes and (b) variability of supply rates or provenance of detrital material delivered to the ocean, linked to climate change (glaciations) or major tectonic uplift. The major element profiles of crusts in this study show neither systematic features which are common to crusts with similar isotope records nor do they generally show coherent relationships to the isotope records within a single crust. Consequently, any interpretation of time series of major element concentrations of a single crust in terms of paleoceanographic variations must be considered with caution. This is because local processes appear to have dominated over more basin-wide paleoceanographic effects. In this study Co is the only element which shows a relationship to Pb and Nd isotopes in Pacific crusts. A possible link to changes of Pacific deep water properties associated with an enhanced northward advection of Antarctic bottom water from about 14 Ma is consistent with the Pb but not with the Nd isotopic results. The self-consistent profiles of the Pb and Nd isotopes suggest that postdepositional diagenetic processes in hydrogenous crusts, including phosphatization events, have been insignificant for particle reactive elements such as Pb, Be, and Nd. Isotope time series of Pb and Nd show no systematic relationships with major element contents of the crusts, which supports their use as tracers of paleo-seawater isotopic composition.