GSA Annual Meeting in Indianapolis, Indiana, USA - 2018

Paper No. 116-8
Presentation Time: 9:00 AM-6:30 PM

OBSIDIAN PYROCLASTS: PRE-ERUPTIVE VANGUARD MAGMA, OR JUVENILE CLASTS FORMED AFTER FRAGMENTATION?


WIEJACZKA, Joshua1, GIACHETTI, Thomas1, WATKINS, James M.1, GARDNER, James E.2, TRAFTON, Kathy1 and NEEDHAM, Éamonn1, (1)Department of Earth Sciences, University of Oregon, Eugene, OR 97403, (2)University of Texas at Austin, Department of Geological Sciences, Austin, TX 78712

Obsidian pyroclasts are dense, glassy, partially degassed pieces of magma found in deposits of explosive eruptions. They were thought to be fragments of vanguard magma produced by permeable foam collapse before or during the explosive phase of an eruption. However, recent studies have suggested that obsidian pyroclasts can also be produced syn-eruptively by ash sintering above the fragmentation level. While samples of vanguard magma record the events that led up to an eruption, samples formed in situ may preserve information about the eruption itself, such as changes in the erupting magma’s gas composition in space and time, or the fragmentation depth. These two mechanisms of formation can be distinguished based on textural and chemical analysis. Obsidian pyroclasts formed by sintering are expected to exhibit complex vesicle shapes, textural heterogeneity, and H2O/CO2 gradients, depending on the time scale available for sintering, while vanguard magma is expected to be more texturally homogeneous and have higher crystallinity due to slower ascent. In this study, obsidian pyroclasts from the Cleetwood eruption of Mt. Mazama (~7700 BP) and the Big Obsidian Flow eruption of Newberry Volcano (~1300 BP) are compared. These eruptions are similar in that they are both highly silicic with an explosive phase followed by the emplacement of an obsidian flow. They differ in that the Cleetwood eruption was immediately followed by a caldera-forming eruption. Preliminary results show that most of the obsidians from the Cleetwood eruption exhibit different textural domains suggesting they sintered for a short period of time (few minutes) before being ejected. We will further analyze these samples by (1) measuring H2O and CO2 contents from obsidian pyroclasts and calculating diffusion time scales, from which time scales of formation can be estimated, and (2) quantifying the textures to calculate sintering time scales.