HR: 1330h
AN: V12A-0550    [PDF]
TI: Insight Into the Magmatic Evolution of Fernandina Volcano, Galapagos, From Olivine- and Plagioclase-Hosted Melt Inclusions
AU: * Hedfield, E J
EM: hedf9207@uidaho.edu
AF: University of Idaho, Department of Geological Sciences, Moscow, ID 83844 United States
AU: Geist, D J
EM: dgeist@uidaho.edu
AF: University of Idaho, Department of Geological Sciences, Moscow, ID 83844 United States
AB: Lavas erupted subaerially from Fernandina Volcano have previously been characterized as being evolved, well-mixed, and homogeneous (Allan and Simkin 2000), obscuring details of earlier stages of magmatic evolution. In this study, phenocrysts and melt inclusions from recent submarine lava samples and representative subaerial lava samples are investigated to gain insight into earlier stages of Fernandina's magmatic evolution. Olivine is present as three populations: unzoned high-forsterite Type I olivine, unzoned low-forsterite Type II olivine, and normally-zoned Type III olivine with Type I cores and Type II rims. Melt inclusions were rehomogenized at 1180C and drop-quenched to remove the effects of post-entrapment crystal growth. Type I olivine contain highly variable melt inclusions (K$_{2}$O/TiO$_{2}$ from 0.03 to 0.29, compared to glass and whole rock values of 0.15 plus or minus 0.01). The variability of the inclusions can be produced by 1 to 2 percent fractional melting of a MORB-like mantle source. Plagioclase and Type II olivine contain more homogeneous melt inclusions that define a trend of fractional crystallization of olivine, plagioclase, and clinopyroxene from a parental melt represented by an average Type I melt inclusion. Type I olivine textures and olivine/host glass disequilibrium suggest that this olivine crystallized from primitive, compositionally-diverse magmas and is exotic. Type II olivine most likely grew in the evolving liquid prior to eruption but after homogenization by mixing. All olivine types consistently demonstrate an equilibrium relationship with their reheated melt inclusions. For Type I olivine, the olivine/inclusion equilibrium and the exotic origin suggest that olivine phenocrysts resided at a temperature close to the trapping/crystallization temperature until shortly before eruption. For Type II olivine, the equilibrium with its inclusions and its host glass indicates that crystal growth occurred shortly before eruption. Post-entrapment diffusive transport of potassium between plagioclase and its melt inclusions may have elevated K$_{2}$O, and therefore K$_{2}$O/TiO$_{2}$, in the inclusions. Diffusive transport may occur on a timescale of 10$^{2}$ to 10$^{4}$ years. These geochemical trends support a picture of early growth of Fo$_{86}$ to Fo$_{89}$ olivine in a dynamic environment, followed by homogenization of compositionally-variable melts and growth of plagioclase and lower-forsterite olivine in a well-mixed chamber. Fernandina's plagioclase-rich and olivine-poor subaerial lavas likely tap the upper part of the chamber, whereas olivine-rich submarine lavas likely tap the lower part of the chamber, where olivine has accumulated.
DE: 3620 Crystal chemistry
DE: 3630 Experimental mineralogy and petrology
DE: 3640 Igneous petrology
DE: 3655 Major element composition
SC: Volcanology, Geochemistry, Petrology [V]
MN: 2003 Fall Meeting