Multiscale Modeling of High and Low Reynolds Number in Small and Large Volume Volcanic Events
Loading...
Date
2024-12-19
Authors
Kubo, Allison
Journal Title
Journal ISSN
Volume Title
Publisher
University of Oregon
Abstract
Worldwide, 500 million people live on volcanoes and face their hazards. At the high Reynolds number end of the spectrum are pyroclastic density currents (PDCs), which are granular flows ranging from bed load to dilute suspension. Due to their mobility, high velocities, and high temperatures, these turbulent events are particularly sensitive to topography and channelization into drainage basins. Understanding the flow transition initiated by overspill from valley-confined PDCs to unconfined PDCs is necessary to mitigate their damage. We present three-dimensional multiphase models using the National Energy Technology Laboratory's Multiphase with Interphase Exchange (MFiX) to model channelized PDCs and establish a link between the generation of overspill currents and channel geometry (width, depth, and curvature). We show that the main overspill mechanism can include significant portions of the insulated underflow layer from the channel, leading to a dangerously hot overspilled current. In all sinuous channel simulations, the underflow of the current becomes superelevated when it encounters bends, potentially overwhelming channel walls. Superelevation of the current increases with channel curvature and decreasing channel width, but is underestimated using traditional estimates of superelevation due to the lack of a free surface in these flows.Meanwhile, in the low Reynolds number regime, dikes transport magma from reservoirs to the surface. Large Igneous Provinces are the largest known magmatic events and are associated with strong climate perturbations and mass extinctions. Huge amounts of magma and gases are transported by a crust-spanning dike system. However, spatial complexity, protracted emplacement history, and uneven surface exposure typically make it difficult to quantify patterns in dike swarms on different scales. First, we address this challenge using the Hough Transform to objectively link dissected dike segments and analyze multiscale spatial structure in dike swarms. We show that for both the Columbia River Flood Basalts and Deccan Traps, a single radial or circumferential swarm does not accurately characterize the data. Finally, apply state-of-the-art computer modeling to cooling dikes using Idaho National Lab's Multiphase Object-Oriented Simulation Environment (MOOSE) porous flow and Navier-Stokes modules. We investigate the effect of advective heat transport in the partially molten dike. We calculate the effective heat conductivity with a hydrothermal system over time.
This dissertation includes previously published co-authored material.
Description
Keywords
dike, HPC, multiphase, pyroclastic