HR: 16:30h
AN: V32H-03 [PDF]
TI: Chemical Stratigraphy of Basalts From the 5000' Borehole NPR-E/WO-2, Eastern Snake River Plain, Idaho:
Evidence for Mixed Asthenosphere-Lithosphere Sources.
AU: * Shervais, J W
EM: shervais@cc.usu.edu
AF: Dept of Geology, Utah State University, Logan, UT 84322
AU: Hanan, B B
EM: bhanan@mail.sdsu.edu
AF: Dept Geological Sciences, San Diego State University, San Diego, CA 92182
AU: Vetter, S K
EM: svetter@centenary.edu
AF: Geology Dept, Centenary College, Shreveport, LA 71104
AB:
Basaltic volcanism in the eastern Snake River Plain presents a fundamental conundrum: the major and trace element
concentrations of these basalts suggest derivation from an asthenospheric source, consisting of depleted mantle and possibly
a Yellowstone plume component, whereas isotopes suggest derivation from an enriched lithospheric source similar to that which
underlies the Archean Wyoming province. We have analyzed 59 whole rock samples from 38 basalt flow groups from a deep bore
hole pair in the eastern Snake River Plain. All of the basalts are LREE-enriched, with La ranging from 18 to 142 x chrondrite
and (La/Lu)N = 2.4 to 9.9. The data generally define smooth trends on MgO-variation plots, suggesting crystal fractionation
as a dominant factor in the major and trace element variations, a conclusion which is consistent with the overall low mg\#s
of 58-40. There is an increase in La/Lu with fractionation, suggesting crustal or lithospheric assimilation. There are three
distinct depth-correlated trends in major and trace elements: first, below 2500 feet there are wide fluctuations in major and
trace elements that are unrelated to fractionation; second, there are five mega-cycles of upward fractionation that occur
between, and in some cases across, major sediment-bounding units, defined by a decrease in mg\#s up section, and by increases
in Ti, P, Zr, La, and La/Lu up section; and third, there are two super-cycles in which Cr and Ni concentrations increase-up
section. The first trend is apparently the result of variable assimilation of crust and/or lithospheric mantle material. This
is supported by the Pb, Sr, and Nd isotope data that shows the lowest basalts have higher $^{208}$Pb/$^{204}$Pb for a given
$^{206}$Pb/$^{204}$Pb, and higher $^{87}$Sr/$^{86}$Sr and lower $^{143}$Nd/$^{144}$Nd than basalts from higher in the
section. The second trend may result from the fractionation of large batches of primitive magma that were stored in the
crust, while the third trend represents increased melt fractions or repeated melting of a refractory source (possibly due to
continued melting of a single source region). K2O/P2O5 ratios also vary widely below 2500 feet, but show a persistent upward
increase above that depth, suggesting an upward increase in crustal or lithospheric assimilation. Forward modeling of crystal
fractionation shows that the observed trends cannot form by low pressure crystal fractionation, even with repeated mixing
and assimilation events; high pressure (6-8 kb) pyroxene fractionation is needed, but mixing and assimilation are also
required. Trace element partial melting models require a plume-related, E-MORB-like source for the basalts, with 7-12%
partial melting of a spinel or garnet-poor lherzolite source, consistent with estimated melting pressures of 12-18 kb.
Calculations suggest mixing about 5% highly enriched lithospheric melt with 95% mantle melt will generate basalts with the
chemical and isotopic characteristics observed in these core samples. We suggest that this mixing occurred in the lithosphere
as the asthenospheric melts percolated upward to intrude the crust, creating a hybrid magma with plume-like major and trace
element characteristics, but with lithospheric isotopic compositions. Further assimilation may have occurred during storage
and fractionation in the middle crust.
DE: 1020 Composition of the crust
DE: 1040 Isotopic composition/chemistry
DE: 1065 Trace elements (3670)
DE: 8439 Physics and chemistry of magma bodies
SC: Volcanology, Geochemistry, Petrology [V]
MN: 2003 Fall Meeting