Volcanic eruptions can have significant impact on life, infrastructure, commerce and climate. Erupting magma is a mixture of hot rock fragments (formerly magma), such as volcanic ash, and volcanic gases that are transported to surface as bubbles within the erupting magma. Bubbles may both, confine magmatic gases potentially resulting in large excess gas pressures, or they may facilitate syneruptive gas loss by forming permeable networks during bubble coalescence. Consequently, bubbles play a crucial role in a feedback with magma ascent rate and affect the style and explosive intensity of eruptions.
This project will further our quantitative understanding of the various processes that affect bubbles and magmatic gases during volcanic eruptions. We will develop a new bubble-scale numerical model for bubble growth, deformation and coalescence. The model will be calibrated to laboratory experiments that are specifically designed to facilitate the integration of empirical data. The bubble-scale model will then be coupled with a multiphase conduit flow model to address problems such as (1) the spatiotemporal evolution of gas pressure in an erupting magma with a polydisperse bubble size distribution; (2) the implications for magma fragmentation; (3) bubble coalescence during shear deformation of the erupting magma; (4) open-system gas loss by permeable flow through coalesced bubbles; and (5) shear localization within the ascending magma.
The overall goal of the project was to achieve a more realistic representation of how bubbles of magmatic gases affect volcanic eruptions. Specifically, the project was focused on the coalescence of bubbles during magma flow within the volcanic conduit. The collaboration between Rice University and the Georgia Institute of Technology has allowed us to integrate laboratory fluid dynamics experiments (Rice), measurements of bubble interconnectivity due to coalescence (Rice), with numerical modeling of bubble-scale processes (Georgia Tech), and eruptive magma ascent (Rice). The outcomes of the project include a new empirical scaling by which the coalescence time of bubbles in magmas can be better predicted, as well as the empirically validated modeling of bubble coalescence, using a new state of the art numerical approach called the Lattice Boltzmann Method. We refer the reader to the corresponding Georgia Tech Project Outcomes Report for a more thorough discussion of the Lattice Boltzmann Method models. The new empirical scaling for bubble coalescence suggests that previoius predictions of the bubble coalescence rate in magmas may have underestimated coalescence rates by up to several orders of magnitude. The modeling of eruptive magma ascent incorporated the effect of gas escape due to interconnected coalesced bubbles, making use of the new measurements of pyroclast porosity and permeability made as part of this project, indicates that such open-system gas escape does not significantly affect sustained explosive volcanic eruptions. The erupton modeling was focused on the largest explosive eruption of the 20th century, the 1912 eruption of Novarupta volcano, Alaska, which erupted 13 cubic kilometers of volcanic material and produced the 'Valley of Ten Thousand Smokes' in Katmai National Park. We modeled one of the three Plinian episodes, during which magma erupted continuously for approximately 10-26 hours, producing a tens of kilometers tall column of volcanic gas and ash, which ultimately was transported and deposited for thousands of kilometers across parts of North America. We find that the open-system gas escape alone appears insufficient to have caused cessation of explosive activity during each of the three Plinian epsisodes, as well as the transition to magma effusion of Episodes 4 and 5 of the eruption. Rather, the change from explosive to effusive activity was likely a consequence of high versus low magma ascent rates, perhaps controlled by magma chamber processes.
