HR: 0800h
AN: V31B-0612    [Abstracts]
TI: Depth Variations in Redox State and Fluid Mobile Element Enrichments in the Mantle Wedge Beneath the Cascades
AU: * Agranier, A
EM: arnaud@rice.edu
AF: Rice University, MS-126, 6100 Main St., Houston, Tx 77005 United States
AU: Lee, C
EM: ctlee@rice.edu
AF: Rice University, MS-126, 6100 Main St., Houston, Tx 77005 United States
AU: Leeman, W
EM: wleeman@nsf.gov
AF: National Science Fondation, 4201 Wilson Boulevard, Arlington, Va 22230 United States
AB: In order to further elucidate the nature of sources and mechanisms involved in magma genesis beneath the Cascades, we selected 9 samples of mostly primitive (MgO contents ranging from 7.7 to 10.8%, Ni>65ppm and Cr>280 ppm) Quaternary basalts from the southern Washington and the northern Oregon and analyzed them for their trace element compositions. Primitive mantle normalized spidergrams illustrate the existence, in this region, of two geochemically distinct groups of magmas. On the one hand, the two easternmost samples, with the lowest Mg# (59.3 and 62) show flattened MORB-like spidergram spectra (La/Yb lower than 3.5) without significant anomalies (group-1). On the other hand, all other samples, which have Mg# ranging from 64 to 71, have trace element patterns typical of calcalkalic magmas with steep slopes (La/Yb up to 26), strong enrichment in Ba (up 100 x P.M.) and LILE and significant negative anomalies in Nb, Ta and Ti (group-2). Moreover, samples from group-1 have fluid mobile element concentrations very close to asthenospheric like values (Li/Y<0.35 and Ba/Nb<18) whereas group-2 basalts show systematically more fluid enriched signatures (Li/Y up to 0.85 and Ba/Nb up to 125). These observations together indicate a very low influence of slab derived material in the group-1 basalt source region and a larger contribution to group-2 lava source region. We also present B concentrations of the samples in order to amplify our investigation on fluid mobile elements. This slab derived enrichment of group-2 lava source by comparison to group-1 is problematic considering that, from petrological constraints (Leeman et al., 2005), group-1 melts are likely to be generated deeper than those of group-2 (50-70 km and 30-50 km respectively) and thus closer to the slab-mantle interface. Leeman et al (2005) propose that the two groups of magmas tap two different and geochemically distinct, stratigraphic levels of the mantle wedge. We propose to use V/Sc ratios as proxy to define fO2 conditions in the group-1 and group-2 sources of magmas. V/Sc ratio is a choice index to evaluate redox conditions in sources of these lavas because V is redox sensitive whereas Sc is not. From our dataset, group-1 magmas have uniform V/Sc ratios which are significantly lower than those of Group-2 (5.7 vs.7.7 to 9.1, respectively). Considering these basalts as the product of about 10% of partial melting, we can estimate (after Lee et al., 2005) that fO2 in the sources of group-1 and group-2 lavas were different (about FMQ-1 and FMQ respectively). The influence of garnet retention in the source could have had on the V/Sc of basalts has been explored but does not seems to be significant in this case.
DE: 1031 Subduction zone processes (3060, 3613, 8170, 8413)
DE: 1065 Major and trace element geochemistry
SC: Volcanology, Geochemistry, Petrology [V]
MN: Fall Meeting 2005