1
Introduction It has been recently shown that thallium (Tl) isotopes are a reliable proxy to track redox (reduction-oxidation) processes of the overlying oceanic water column due to their affinity to be absorbed in ferrous- manganese sediments. Manganese oxides (MnO) found in sediments are sensitive indicators for the presence of oxygen. Thus when present, oxygen will bond with Mn to form MnO, which Tl adsorbs to readily, but if the above water becomes oxygen free, anoxic, then MnO is re- dissolved. Previous studies have found that the global Tl budget of seawater is not affected by the inputs (dust, riverine and hydrothermal fluxes) because they all have very similar isotopic values. Only the two outputs create a large isotopic fractionation, manganese oxide absorption and oceanic crust alteration, which makes Tl useful in tracking the behavior of the sediment water interface. New data suggests that sediments from the Cariaco Basin and the Black Sea, the two largest anoxic and sulfidic basins in the modern ocean, faithfully record the overlying seawater value. Thus we can use Tl isotopes from sediments and rocks deposited in similar environments to track changes in ferrous-manganese sediments, which is due to variations in bottom water oxygenation. Here we have measured the Tl isotope composition of ancient Cariaco sediments from the ocean drilling program 165, which captures transitions of glacial-interglacial cycles. It has been hypothesized that bottom water oxygenation varies during these natural cycles largely due to variable primary production as measured by atmospheric CO 2 contents. The low-resolution dataset suggests there is large variability in Tl isotopes, which could be due to changes in global bottom water oxygen contents. It should be noted that the Cariaco basin provides an incomplete record of recent glacial- interglacial variability as the basin becomes isolated from the global ocean due to drastic sea level drop during major glaciations. Thus, we have measured reducing but non-sulfidic California Margin sediments to explore the possibility of capturing seawater Tl isotopes to extend the utility of this new and developing proxy. Materials and methods •Samples heated at 65°C to remove water and were powdered to original grain size using a mortar and pestle. •Weighed ~50 "g of sample and leached in solution of 2M HNO 3 at 130°C for ~16 hours •Centrifuged for 10 minutes at 2000 rpm 3 times to separate leachate from silicate faction. •Particles dissolved using standard HNO 3 and HCl and any organics eliminated using H 2 O 2 •Chromatography used to separate naturally existing Pb from Tl using Br, HNO 3 and HCl solutions; Tl from sample was eluted using HCl-SO 2 solution •Organics from resin and HCl-SO 2 dissolved using HNO 3 •Collected Tl prepared for Multi-Collector Inductively-Coupled-Plasma Mass-Spectrometry Acknowledgments This research was funded by DMR1157490. Analysis was funded by NSF and NASA to JDO. Results Conclusions •Initial data from Cariaco Basin suggests further and higher resolution data is needed to fingerprint glacial-interglacial bottom water oxygen. •New data suggest permanently anoxic sediments archive seawater Tl isotopes which allows a wider amount of ancient sediments to be used to record climate oxygen variability Mary E. Powell and Jeremy D. Owens National High Magnetic Field Laboratory Literature cited Nielsen, S.G., et al., (2011) Thallium isotopes in early diagenetic pyrite – A paleoredox proxy? Geochimica et Cosmochimica Acta, 75, 6690-6704. Owens, J.D., et al., (submitted) Thallium isotopes: a new redox proxy tracking the global burial of manganese oxide burial Chemical Geology. Future Work •Additional samples from oxygen minimum zones such as Peru-Chili Margin and Namibia Coastline to confirm their utility for tracking seawater Tl isotopes •Analyze Santa Barbra Basin samples for trace metal geochemistry, to better understand local redox conditions •Obtain a deep core from the Santa Barbara Basin that goes through glacial-interglacial cycles Current mass balance Santa Barbra Basin Cariaco Basin •A euxinic basin, meaning it is anoxic and has sulfide in the water column •The large shifts in ε 205 Tl is possibly due to global and local shifts in dissolved oxygen • Slightly heavier value in the Younger Dryas (the last occurrence of enhanced glaciation) suggests less global bottom water oxygen as seen through a decreased amount of MnO in sediments • ε 205 Tl during oxic restriction records the local MnO burial of +4. 0 0.2 0.4 0.6 0.8 1 1.2 1.4 -7 -6 -5 -4 Depth (mbsf) 205 Tl (ε) G6C-47 Figure 6 (left): ε 205 Tl values for deep core G6C-47 samples going back ~1,000 years. The dotted line represents the average global seawater value of ε 205 Tl. Figure 8 (below): Map of Santa Barbara Basin of the coast of California The numbers refer to the sampling sites. Figure 5: ε 205 Tl values in relation to depth; -6 is the current global value for seawater Figure 7 (left): ε 205 Tl values for box core samples of the top 5 cm. The dotted line represents the average global seawater value. Site 2 •In the center of the Santa Barbra Basin, the amount of dissolve oxygen at the water- sediment interface is zero. •Surface sediment at this location records ε 205 Tl values of the seawater. •The ε 205 Tl leach of each samples seems to track seawater throughout the core. •The variation of values throughout the core might suggest the record of temporal shifts in the global bottom water dissolved oxygen. Sites 1, 3, 4, and 5 • None of the cores seem to record the seawater value of each site. •While there is a variability in the ε 205 Tl values found, sites with similar depths have consistent values. Figures 1-3 (counter clockwise): Sample dissolution (1); Ion chromatography (2); Instrumental analysis (3) Interlab standard comparison Figure 4: Map of Cariaco Basin off of Venezuela. The red marks where the samples were gathered. 0 0.01 0.02 0.03 0.04 0.05 0.06 -8 -6 -4 -2 0 Depth (mbsf) 205 Tl (ε) Site 2 BC-42A (425 m) Site 3 BC-52 (918 m) Site 3 BC-53B (846 m) Site 4 BC-57 (59 m) Site 5 BC-62A (60 m) Sample Institute ε 205 Tl (average) 2-Standard Deviation (average) Sample Size SCO-1 WHOI -2.94 0.18 9 SCO-1 FSU -2.93 0.43 3 1 2 3
