HR: 08:15h
AN: V41F-02    [Abstracts]
TI: Why do magmas stall? Insights from petrologic and geodetic data
AU: * Zimmer, M M
EM: mzimmer@bu.edu
AF: Dept. Earth Sciences, Boston University, Boston, MA 02215, United States
AU: Plank, T
EM: tplank@bu.edu
AF: Dept. Earth Sciences, Boston University, Boston, MA 02215, United States
AU: Freymueller, J
EM: jfreymue@gi.alaska.edu
AF: Geophysical Institute, Alaska Volcano Observatory, University of Alaska Fairbanks, Fairbanks, AK 99775, United States
AU: Hauri, E H
EM: hauri@dtm.ciw.edu
AF: Department of Terrestrial Magnetism, Carnegie Institution of Washington, Washington, DC 20015, United States
AU: Larsen, J F
EM: faust@gi.alaska.edu
AF: Geophysical Institute, Alaska Volcano Observatory, University of Alaska Fairbanks, Fairbanks, AK 99775, United States
AU: Nye, C J
EM: cnye@giseis.alaska.edu
AF: Alaska Volcano Observatory, Alaska Divison of Geological and Geophysical Survey, Fairbanks, AK 99709, United States
AB: Magmas stall at various depths in the crust due to their internal properties (magma viscosity, buoyancy) and external crustal controls (local stress regime, wallrock strength). Annen et al. (JPet 2006) propose a petrological model in which buoyant magma ascends through the crust until the depth of water saturation, after which it crystallizes catastrophically and stalls due to the large increase in magma viscosity. Magmas may erupt from this storage region, or viscous death may result in pluton formation. In order to test this model, and constrain magma storage depths, we combine petrological and geodetic data for several active volcanoes along the Aleutian-Alaska arc. We analyzed glassy, primarily olivine-hosted melt inclusions by SIMS in tephra samples for their pre-eruptive volatile contents, which can be related to the depth of entrapment via pressure-dependent H₂O-CO₂ solubility models (e.g., VolatileCalc). Melt inclusions are not in equilibrium with pure water vapor (all will contain S and C species), but >50% of the inclusion population are in equilibrium with a vapor containing >85% H₂O. Geodetic data (InSAR, GPS) record surface deformation related to volcano inflation/deflation, and can be inverted to solve for the depths of volume change (magma storage) in the crust. In the Aleutians, we find that the maximum melt inclusion trapping depths and geodetic depths correlate, suggesting both techniques record crustal magma storage and crystallization. Melt inclusions from the 1997 Okmok eruption are trapped at ≤3 km; deformation during the eruption and subsequent inflation occurred at 3±0.5 km (Miyagi et al., EPSL 2004; Lu & Masterlark, JGR 2005). At Akutan, melt inclusions and GPS data indicate magma storage at ~5-7 km. Inclusions from flank cones of Makushin yield depths of 7 km, similar to inflation observed beneath the main edifice (6.8 km, Lu et al., JGR 2002). Pleistocene inclusions from Augustine volcano indicate magma storage at 10-18 km, in accord with a deep magma source proposed for the 2006 eruption. Melt inclusions from Shishaldin are trapped at depths up to 4 km, coincident with the base of the conduit (Vergnoille & Caplan Auerbach, BVolc 2006). Other volcanoes record similar depths of melt inclusion entrapment and deformation, including Mt. St. Helens, Irazú, Soufriere Hills, Vesuvius, and Etna. Clearly, crystallization will occur where magmas stall, cool, and degas, so it may not be surprising that the depths of deformation correlate with the depths of melt inclusion entrapment. But the question of why magmas stall at various depths remains. In the Aleutians, maximum H₂O contents of melt inclusions (from 2 wt% at Shishaldin to 7 wt% at Augustine) negatively correlate with measures of the degree of mantle melting (Ti_6.0 and Y_6.0), which is expected if water drives mantle melting beneath arcs (e.g. Kelley et al. JGR 2006; Portnyagin et al EPSL 2007). Thus, if magmas stall near the depths where they reach H₂O-saturation, as predicted by Annen et al. and observed here, then magma chamber and pluton depths may ultimately be controlled by the primary magmatic water contents set in the mantle.
DE: 8185 Volcanic arcs
DE: 8413 Subduction zone processes (1031, 3060, 3613, 8170)
DE: 8439 Physics and chemistry of magma bodies
SC: Volcanology, Geochemistry, Petrology [V]
MN: 2007 Fall Meeting