HR: 1330h
AN: V32A-0993    [PDF]
TI: Platinum Group Element (PGE) Abundances in Lava Flows Generated by the Hawaiian Plume: Insights into Plume Evolution
AU: * Shafer, J T
EM: jshafer@nd.edu
AF: University of Notre Dame, Dept. Civil Eng. and Geol. Sci 156 Fitzpatrick Hall, Notre Dame, In 46556
AU: Neal, C R
EM: Neal.1@nd.edu
AF: University of Notre Dame, Dept. Civil Eng. and Geol. Sci 156 Fitzpatrick Hall, Notre Dame, In 46556
AB: Picritic and high-MgO (7.7-24 wt.%) basalt samples from Detroit (/sim81-76 Ma) and Koko (/sim48 Ma) Seamounts along the ESC have been analyzed for PGEs (Ru, Rh, Pd, Ir, and Pt) allowing an examination of how the PGEs in lavas from the Hawaiian plume have changed over time. Major and trace element (including the PGEs) concentrations were quantified by ICP methods at the University of Notre Dame. See Ely et al. (1999, Chem. Geol. 157:219) for the PGE analytical method. Bennett et al. (2000) analyzed Hawaiian picrites and found PGE abundances slightly greater than average MORB and comparable to the low-PGE basaltic komatiites. These authors modeled the PGE abundances of these picrites by using variable amounts of residual sulfide during melting, such that Koolau (low PGE contents) formed from a relatively sulfide-rich source and Loihi (high PGEs) from a sulfide-poor source. Our PGE data from Detroit Seamount show slightly higher PGE abundances than Loihi and Kilauea, suggesting these picrites formed from a source lacking residual sulfide. These results suggest that, if the model of Bennett et al. (2000) is correct, the dilution of plume lava with MORB source, as hypothesized on the basis of depleted isotope ratios and lower trace element abundances than modern Hawaii (Keller et al., 2000, Nature 405:603; Kinman \& Neal, 2002, Eos 83:F1282; Regelous et al., 2003, JPet 44:113), was not the controlling factor in PGE abundances. However, since MORB PGE concentrations are not substantially different than low-PGE Hawaiian picrites, incorporation of MORB material within the Hawaiian plume at Detroit Seamount would not have drastically reduced the PGE abundances. Koko Seamount has relatively high PGE concentrations (/sim3-12 times greater than those from Detroit lavas). This may be the result of a lack of residual sulfide facilitated by higher degrees of partial melting. Although our initial data are consistent with variable degrees of partial melting and/or source heterogeneity over the life of the Hawaiian plume, the data from Detroit Seamount can be modeled by, for example, magma mixing between Koko-type "PGE-rich" plume and MORB end members (cf. Kinman \& Neal, 2002). The Pt/Ir ratios and PGE abundances of picrites from Detroit and Koko Seamounts and from Hawaii (as analyzed by Bennett et al., 2000) increase in the order: Hawaii (4.8), Detroit (5.8), Koko (8.1). Bennett et al. argued that if more sulfide was retained in the source the PGE profile would be more fractionated and abundances would be lower. Our data suggest the opposite is true. For Koko Seamount to have PGE abundances approximately 3-12 times greater than the high-PGE picrites from Hawaii and yet have a more fractionated profile, the source of the Hawaiian plume must have been relatively PGE-enriched at 48 Ma than it is currently. In addition, the more fractionated profile of Detroit Seamount is consistent with the incorporation of MORB material (Pt/Ir /sim 25.9), thereby raising its Pt/Ir ratio.
DE: 1749 Volcanology, geochemistry, and petrology
SC: Volcanology, Geochemistry, Petrology [V]
MN: 2003 Fall Meeting