Mechanical and geometric controls on the structural evolution of pit crater and caldera subsidence

Abstract

[1] Pit craters and calderas are volcanic depressions produced by subsidence of a magma reservoir roof. To identify how geometric and mechanical factors may influence the structural evolution of this subsidence, we used two-dimensional distinct element method numerical models. The reservoir host rock was represented as an assemblage of bonded circular particles that interact according to elastic-frictional laws. Varying particle and bond properties produced a range of bulk material properties characteristic of natural rock masses. Fracturing results when bonds break, once their shear or tensile strength is exceeded. The magma reservoir was represented as a region of nonbonded low-friction particles. Withdrawal of magma was simulated by incrementally reducing the area of the reservoir particles. Resultant gravity-driven failure and subsidence of the reservoir roof were explicitly replicated. Interaction of the roof's strength, Young's modulus, thickness/diameter ratio (T/D), and the reservoir's shape yields a variety of model structures and subsidence styles. In conceptual terms, four end-member subsidence styles developed: (1) "central sagging" favored by low strength and low T/D; (2) "central snapping" favored by high strength, low T/D, and a sill-like reservoir shape; (3) "single central block" favored by low to intermediate strength, high Young's modulus, and intermediate T/D; and (4) "multiple central blocks" favored by high strength, low Young's modulus, and high T/D. Most model realizations incorporated some combination of each style, however. The models provide a geomechanical framework for understanding natural pit crater or caldera structures, as at Nindiri (Nicaragua), Fernandina (Galapagos), Dolomieu (La Reunion), and Miyakejima (Japan).

From mechanical modeling to seismic imaging of faults: A synthetic workflow to study the impact of faults on seismic

“…Also, contrary to FEM, it is not possible to specify a priori the stress-strain response of the assemblage. Therefore it is necessary to carry out calibration of particle properties to the emergent rock physical properties (Egholm et al, 2007;Holohan et al, 2011). Discrete element simulations of thrust-related structures have been carried out by Finch et al (2003), Strayer and Erickson (2004), Cardozo et al (2005), , Benesh et al (2007), ), Zhang et al (2013 and Gray et al (2014).…”

“…In 2D, we have the possibility to implement collapse and biaxial compression simulations. Such tests are performed on a subportion of the system in order to evaluate multiple sets of values of particle properties within acceptable runtimes (Finch et al, 2004;Holohan et al, 2011;paper 1). Collapse simulations allow the measure of the static angle of repose of the material, and biaxial tests at 24 several confining pressures allow the measure of the failure envelope of the material.…”

“…In addition, it allows localization and the formation and linkage of faults and fractures as a natural part of the numerical scheme. However, one disadvantage of the technique lies in the necessary, but tedious and time-consuming, calibration of particle parameters to the emergent rock physical properties (Egholm et al, 2007;Holohan et al, 2011). The selection of particle parameters ultimately controls the bulk mechanical behavior of the assembly under consideration, and has to be calibrated rather than directly input as in the case of finite element or finite difference methods.…”

“…The particle properties and the emergent bulk material properties are typically assessed through the use of angle of repose and biaxial tests. Such tests are performed on a subpart of the system in order to evaluate multiple sets of particle property values within acceptable runtimes (Oger et al, 1998;Finch et al, 2004;Holohan et al, 2011). Similarly, the sensitivity of models to subtle initial differences in, e.g., assembly packing is well known (Abe et al, 2011).…”

Morphological expressions of crater infill collapse: model simulations of Chaotic Terrains on Mars

“…DEM has been used in the past to investigate collapse phenomena (e.g., Bym et al, 2013;Holohan et al, 2011Holohan et al, , 2015. Bym et al (2013) show that DEM reproduces very well empirically observed subsidence profiles, while Holohan et al (2011Holohan et al ( , 2015 show the evolution of fractures and stress during the collapse of a volcanic caldera because of the withdrawal of a magmatic body.…”