2015 GSA Annual Meeting in Baltimore, Maryland, USA (1-4 November 2015)

Paper No. 76-4
Presentation Time: 2:15 PM

OLIVINE-MELT EQUILIBRATION AND REDOX RELATIONSHIPS IN SCORIA FROM THE 1959 ERUPTION OF KīLAUEA  VOLCANO, HAWAI‘I


HELZ, Rosalind T., US Geological Survey, M.S. 926A, Reston, VA 20192, COTTRELL, E., Mineral Sciences, Smithsonian Institution, PO Box 37012, National Museum of Natural History MRC-119, Washington, DC 20013-7012, KELLEY, Katherine A., Graduate School of Oceanography, University of Rhode Island, 215 South Ferry Road, Narragansett, RI 02882 and BROUNCE, M.N., California Institute of Technology, Division of Geological and Planetary Sciences, Pasadena, CA 91125, rhelz@usgs.gov

The 1959 Kīlauea summit eruption, with its high lava fountains of gas-rich, primitive magma, is unique in Kīlauea’s historical record. The 1959 scoria contain abundant olivine ± chromian spinel in highly vesicular brown glass. In 8 samples from phases 1-2 of the eruption, euhedral olivine phenocrysts (≥1mm in length, and which appear to be in textural equilibrium with the enclosing glass), have rim compositions that vary from 88.2 to 86.0 mol % forsterite (Fo) within a thin section (and hence at constant major-element melt composition). Compositions of microphenocrystic and groundmass olivine in the same samples range from 88.0 to 82.0 mol %Fo, again within thin sections. These ranges do not track the large shifts in melt composition that occur with time in phases 1-2. By contrast, in the latest scoria of the eruption (from phases 10-16), the compositional range of phenocryst rims and groundmass olivines alike is only 85.5 to 84.0 mol %Fo. This narrower range, plus the overlap in ranges for large and small olivines, suggests that the exchange of Fe+2and Mg between olivine and melt in the later samples was approaching chemical equilibrium.

Eight 1959 scoria samples (five from phase 1, and one each from phases 3,10, and 15) were investigated by µ-XANES to determine directly the in situ Fe+3/total iron for glasses next to selected olivines. This was done to test the hypothesis that the observed variable Fo contents might reflect variation in ferric iron in the melt, as variation in glass MgO and total FeO is minor within each sample. Fe+3/total iron varies from 0.185 to 0.124 in various scoria glasses, with higher values in locally oxidized glass edges; this range explains some but not all of the variation in Fo content. The remaining variation in olivine compositions appears to result from mixing of thermally disparate magmas, which were erupted and quenched before internal equilibrium was achieved.

Data for 24 olivine-melt pairs in the well-equilibrated late samples include 7 pairs of phenocryst rims + glass, one melt inclusion and 16 groundmass olivines + glass. Using the µXANES data to correct for Fe+3, the resulting values are KD = 0.28±0.03 for the exchange of Fe+2 and Mg between olivine and melt, somewhat displaced from the widely accepted value of 0.30±0.03 of Roeder and Emslie (1970).