|dc.description.abstract||Calderas are 1 - 100 km diameter, volcanic depressions that form primarily through m-scale to km–scale subsidence of a magma reservoir roof into an underlying magma reservoir. This study addresses several fundamental geometric and kinematic aspects of the structure of caldera volcanoes via a series of scaled physical (or analogue) modelling experiments. Results of these models are combined in detail with observations from natural calderas recorded in the literature, in order to provide as strong an element of ‘ground-truthing’ for the experimental results as possible.
The basic experimental materials and methods are common to all the results chapters (2-6), and so for convenience are summarised together in Chapter 1. An exception to some extent is the experiment set described is Chapter 5; these were made with different materials and apparatus, the particulars of which are described in that chapter. Each of the results chapters is otherwise self-contained, with its own specific introduction to the problems addressed, summary of data, discussion, and conclusions. The contents of Chapters 1, 5, and 6 have been published in two peer-reviewed articles. The remaining chapters are presently unpublished, although Chapter 3 and part of Chapter 4 have been submitted for peer review, whilst the rest of Chapter 4 and a part of Chapter 2 will be submitted for publication at a later date. Given the assistance of co-authors in the preparation of manuscripts now integrated into the thesis, my contribution to the individual chapters as presented here may be summarised as: Chapter 1: 85%, Chapter 2: 90%, Chapter 3: 75%, Chapter 4: 80%, Chapter 5: 50%, Chapter 5: 75%. The structure of the thesis, the major aspects of caldera evolution investigated, and the main results are as follows:
Chapter 1: Introduction and Methodology - outlines the current understanding of caldera structure as achieved to date by previous authors and supplies the rationale for using physical models to help unravel caldera evolution. This section also summarises the insights for how these. Details of the experimental methodology employed in this study, as well as scaling considerations for the application of small-scale models to the study of large-scale deformation associated with caldera subsidence in nature, are also provided.
Chapter 2: The Geometry and Evolution of Circular Caldera Collapse Structures - details the results of experimental subsidence of mechanically isotropic reservoir roofs that are circular in plan view. This baseline experiment set illustrates the generalised kinematic evolution of caldera subsidence. Structures resolving radial and concentric strains are documented, the latter for the first time. The control of the roof’s thickness/diameter ratio upon these structures’ development is also highlighted. The kinematics of caldera ring faults, the generation of polygonal caldera outlines and the development of piecemeal-type subsidence are also discussed in light of the evidence from experiment and nature.
Chapter 3: The Caldera Collapse Continuum - systematically integrates results of simple circular collapses in Chapter 2 with those from past experimental, numerical and field studies to produce a ‘state-of-the-art’ synthesis of the mechanics and kinematics of caldera collapse. This synthesis sheds new light on the long-standing but poorly-defined continuum between end-member caldera collapse styles. Structural elements characteristic of each end-member are shown to reflect progressively overlapping structural sub-processes in a single collapse event. The continuum between end-member collapse styles is thus defined primarily as a progressive evolution related to increasing subsidence, although the regulatory role of roof thickness/diameter and caldera subsidence/diameter ratios in the development of structural elements of each collapse end-member is also highlighted.
Chapter 4: The Role of Magma Chamber Ellipticity in Caldera Collapse – documents the results of experimental collapse of mechanically isotropic reservoir roofs that are varyingly elliptical in plan view. Physical models demonstrate that an elliptical reservoir roof predictably fails first at the ends of its shorter principal axis, and show that lateral propagation (‘unzipping’) patterns vary systematically with ellipticity. The models also highlight a characteristic geometric arrangement of elliptical ring structures and provide a first comprehensive view of the elliptical caldera structural architecture in 3D.
Chapters 5 & 6: Regional tectonic influences on the evolution of caldera collapse – outlines the results of experimental collapse of reservoir roofs that contain pre-existing mechanical anisotropies in the form of regional-tectonic faults and collapse under regional tectonic stress. Although disregarded in most previous physical and numerical models of calderas, regional faults have long been thought to play a major role in caldera development, principally through their reactivation during collapse. In Chapter 5, caldera formation in areas of orthogonal regional deformation is examined; in Chapter 6, caldera collapse in strike-slip regimes is investigated. In physical models, regional faults reactivated during collapse where they coincided with and ran parallel to the magma reservoir margins. Here, where syn-collapse strain usually tends to localise, pre-collapse faults are optimally orientated for reactivation. Contrary to inferences from some previous field studies, differential movement on pre-collapse faults was not observed in central parts of the subsiding reservoir roof. Physical models thus confirm that regional fault reactivation is an important process, and help clarify conditions favouring it.
Chapter 7: Summary of Main Conclusions and Outlook - provides a brief overview of the main findings of this thesis, notes their general implications for our understanding of calderas, and outlines some potential avenues for future work.
An electronic appendix (DVD) is attached to the back of the thesis. This appendix contains photos of all experiments (in jpeg format), animations of the experiments (in MS PowerPoint files), and copies of the two published articles (in PDF format).||