V23B-4795:
Understanding Magma Storage Conditions that Produce Highly Explosive Monogenetic Basaltic Eruptions Using Olivine-Hosted Melt Inclusions from Sunset Crater, AZ

Tuesday, 16 December 2014
Chelsea M Allison, Kurt Roggensack and Amanda B Clarke, Arizona State University, Tempe, AZ, United States
Abstract:
To investigate mechanisms of explosive basaltic volcanism, we studied the ca. 1085 AD Sunset Crater eruption in the San Francisco Volcanic Field (SFVF) of northern Arizona. This eruption, the youngest in the SFVF, first featured fissure eruptions (explosive phases 1-2) and a small lava flow, and then activity narrowed to a central vent producing explosive phases 3-8 and two additional lava flows. While the first two phases were Strombolian-style explosions, middle phases (3-5) were subplinian in character and produced an anomalously large tephra deposit. The final phases (6-8) are poorly characterized at this stage. The total erupted volume of lava and tephra is >0.7 km3 DRE of alkali olivine basalt with a large proportion of crystal-free glass and low phenocryst content.

We studied 82 primary melt inclusions (MIs) in the largest tephra units (explosive phases 3, 4) to investigate magma volatiles and storage conditions. To prioritize primary volatile contents, we picked rapidly quenched free olivine crystals (Fo 81-85; 0.5-2 mm) and selected large volume MIs (50-180 µm) located near crystal cores for analysis. We observed vapor bubbles in all MIs and also noted rare occurrences of CO2-rich gas inclusions. MIs show little major element variability suggesting little crystal fractionation (K2O 0.8-1.1 wt.%). Post-entrapment crystallization is also minor (2-9%). The MI compositions from the two phases largely overlap, with phase 4 skewed to slightly higher K2O. FTIR spectroscopy shows that the MIs are relatively dry and CO2-rich. Water abundances vary 0.8-1.6 wt.% with a median of 1.25 wt.%, while most MIs have CO2 abundances 1,600-3,400 ppm. Phases 3 and 4 are essentially identical in water content. CO2 contents of phases 3 and 4 show considerable overlap, however the phase 4 MIs are skewed toward high CO2 (>2,500 ppm). These results require a minimum MI entrapment depth of ~11 km from fluid saturation constraints. Overall, the MIs indicate a largely homogeneous magma stored at mid-crustal depths. These two highly explosive phases (3 & 4) are quite similar, though the latter phase tapped a slightly deeper portion of the magma storage zone.