GSA Annual Meeting in Indianapolis, Indiana, USA - 2018

Paper No. 276-3
Presentation Time: 2:10 PM

DOES SIZE MATTER? WHAT VARIABLES CONTROL THE THERMAL HISTORY OF MAGMAS? (Invited Presentation)


COOPER, Kari M.1, SCHRECENGOST, Kevin L.1, SCHLIEDER, Tyler1, BRADSHAW, Richard W.2, KENT, Adam J.R.3 and HUBER, Christian4, (1)Earth and Planetary Sciences, UC Davis, 1 Shields Ave, Davis, CA 95616, (2)College of Earth, Ocean & Atmospheric Sciences, Oregon State University, 104 Wilkinson Hall, Corvallis, OR 97331, (3)College of Earth, Ocean, and Atmospheric Sciences, Oregon State University, Corvallis, OR 97331, (4)Department of Earth, Environmental & Planetary Science, Brown University, 324 Brook St., Box 1846, Providence, RI 02912

Magma temperature is one of the main variables controlling crystallinity, which in turn exerts a major influence on bulk viscosity and therefore mobility. A long-standing debate centers on the thermal conditions of magma storage in the crust. Some data suggest storage of erupted magmas over tens to hundreds of kyr at temperatures where the magma is melt-dominated (“warm storage”), whereas other data suggest that magmas are stored largely in a crystal-dominated state and remobilized in the decades to centuries prior to eruption (“cold storage”). Therefore, while it is clear that some magmas (or some components in magmas) are stored cold, the question therefore becomes which magmas have experienced dominantly cold storage, and which variables (e.g., erupted volume, composition) exert the primary control on magma thermal histories.

We present U-series ages of plagioclase combined with durations of intracrystalline diffusion of Sr in plagioclase from the same crystal populations from dacitic eruptions ranging in size from <1 km3 to >10 km3, including Lassen Volcanic Center, Mount Hood, Mount St Helens, Volcán Quizapu, Chile, Mount Pinatubo and Huaynaputina, Chile. Pre-eruptive crystal residence ages for plagioclase in each of these eruptions are 103-104 y. In contrast, modeled diffusion durations at 750°C are decades to centuries (up to a few millenia for Lassen samples). In most cases, these modeled diffusion durations are a factor of 10-100 shorter than the crystal residence times, clearly indicating that some crystals in this suite of eruptions were derived from regions of the magma reservoirs that were stored cold (i.e, <750°C) for >90% of their history. These storage temperatures imply a crystal-dominated mush rather than a liquid-dominated magma like those erupted. Although mobilization of magma is complex and can occur at a range of crystallinities, this pattern suggests that the conditions of storage and the time scales of assembly of erupted dacitic magma bodies are similar across a broad range in eruptive volumes typical of arc systems. However, very large eruptions (>100 km3) and/or eruptions of different composition may show different behaviour.