Pumice: a post-fragmentation product?

University Of Oregon Eugene, Eugene OR

Investigators

Abstract

Plinian eruptions are among the most explosive volcanic eruptions and one of the principal geological hazards along volcanic arcs. In the US, several volcanoes have produced, and will again one day, devastating Plinian eruptions. Products from a Plinian eruption are ejected into the atmosphere and can travel up to thousands of kilometers. Their fallout can seriously damage infrastructure, agriculture, and eventually encircle the globe, disrupting aviation and affecting climate. Typically, more than 70% of the ejected Plinian material consist of light weight fragments of fresh magma called pumice. The overall formation of pumice seems mostly understood; during its ascent to the surface, magma foams as gas bubbles nucleate and grow. This process continues until gas pressure inside the bubbles causes the magma to break up into porous fragments that are ejected in the atmosphere. The texture of pumice is thought to be “frozen in” at the time magma fragments, whereas the size of fragments (microns to centimeters) is reduced upon inter-particle collisions during transport in the conduit, the eruptive column and/or the pyroclastic flows. Past research has used the texture and size distribution of pumice to put quantitative constraints on the eruptive processes occurring inside the conduit, especially on magma fragmentation, even though the initial fragment size distribution remains largely unknown. This study is driven by new textural observations made on pumice from Medicine Lake Volcano (CA) that suggest some pumices are in fact made of numerous smaller pieces of magma that collided, agglomerated and partly sintered inside the conduit. This project seeks to (1) assess whether these newly documented textures vary with pumice size and between eruptions of diverse intensity and conduit shape, (2) further investigate the implications of this discovery in terms of the conditions for magma fragmentation and characteristics of its products, (3) determine whether the observed textures can be quantitatively reproduced by sintering and decompression experiments in laboratory. This study will support an early career scientist, a PhD student, and undergraduate students who will gain experience with a wide variety of field methods, experimental approaches, and analytical tools. The link between research and education will include the development of two new First-year Interest Group courses that expose students to volcanology and the links between volcanology and other disciplines (anthropology, archeology and journalism). Magma fragmentation is a process of fundamental importance to volcanology. Beyond that, it has relevance to volcanic hazard assessment, as it discriminates an explosive volcanic eruption from an effusive one. In particular, the evolution of the size distribution of pyroclasts inherited from magma fragmentation is an important input of plume development and tephra dispersion models. Thus, a better understanding of fragmentation and its products is important not only for basic science, but also for practical reasons. To test the aforementioned hypotheses, the research team will first quantify the size, shape and textures, both in 3D and in 2D, of hundreds of porous pyroclasts from five rhyolitic eruptions that spanning more than three orders of magnitude in eruptive volume. These eruptions occurred at Newberry, Medicine Lake, Crater Lake and Long Valley volcanoes and through conduits of different shape (sub-circular, dike, or unknown shape). Second, the researchers will decipher the implications, in terms of magma fragmentation, of the new observations by individualizing the particles that make up pumice clasts and characterizing their size and shape distributions. Finally, experiments will be carried out to attempt to reproduce the observed textures. Rhyolitic glass will be hydrated with different amounts of H2O, then crushed, and finally sintered during decompression experiments that reproduce the conditions of magma ascent between the fragmentation level and the Earth’s surface. Preliminary observations, analyses and experiments show that (1) agglomerate textures occur in pumices from several Plinian eruptions at different volcanoes, (2) the size distribution of individual particles making individual pumice clast is consistent with the former being derived from primary fragmentation in the conduit, and (3) it is possible to reproduce the observed textures experimentally under certain conditions of initial particle size, water content and decompression rate. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.