Unfortunately my thesis is a massive file due to the lots of big jpeg files needed to show my results so I can't put the whole thing on the web. If you are interested in this stuff send me an email Ben@eps.mcgill.ca
The Nature and Origin of Caldera structure and Morphology, using results from analogue modeling.
ABSTRACT
Calderas illustrate a variety of different styles which are controlled by their internal structure and morphology. The internal structure of many calderas is not exposed; as a result, the calderas frequently are interpreted as simple pistons. The results and examples from this thesis indicate that caldera structure is often more complex than this and that caldera formation consists of several stages controlled by complex interactions of many variables. Internal processes and parameters include rock properties (shear strength, planes of weakness, vertical and horizontal variations), dimensions and internal pressure of the associated magma chamber, styles of tumescence and resurgence, and the size of the eruption. External processes and parameters are also important, such as the regional stress regime (e.g., extensional and pull-apart basins), pre-existing topography, and pre-existing structures (e.g., regional faults, basement grain).
Scaled physical models of caldera formation were carried out to investigate the effects of some of these variables on the temporal development of calderas. Dry sand contained in a 1 m-diameter cylinder was used as an analogue of crustal rocks, and a water-filled 60 cm-diameter rubber bladder was used as an analogue for a magma chamber. Scaling parameters used were a length ratio (L*) of 2.5 x 10-5 and a stress ratio (s *) of 2.0 x 10-5.The collapse process was initiated by withdrawing water from the bladder at a rate of 1600 cm3 min-1. The depth of the rubber bladder then was varied from 6.0 to 24 cm to simulate different magma chamber depths (2.4-9.6 km). The pressure within the bladder also was varied by changing the initial volume of water in the bladder from 40 to 45 liters before evacuation. The effects of different surface topography prior to collapse also were investigated by building scaled ridges, mountain ranges, and stratocone volcanoes, and positioning them in different positions and orientations above the bladder. Experiments also were carried out with the rubber bladder tilted to promote asymmetric collapse.
Generally, deformation began with broad sagging, then an arcuate or linear outward dipping fault formed on one side of the caldera. This fault propagated laterally around the caldera in both directions and sometimes joined up with other faults, forming a polygonal caldera. As subsidence continued, the caldera grew incrementally outwards, progressively forming a series of concentric subsidence-controlling faults. Lastly, a peripheral zone of downsagging would develop, bounded by an inward dipping outer fault related to extension. As the depth of the bladder was increased, (1) the volume of the caldera decreased, (2) the area of faulting decreased, (3) the symmetry of the caldera was affected and (4) the coherence of the subsiding block decreased. With greater topographic relief, (1) the volume of the resultant caldera increased, (2) single ring fractures became larger and more coherent, and (3) the caldera was more prone to slumping. The location of the topographic relief in relation to the bladder did not appear to affect the symmetry of subsidence. As the initial volume and internal pressure of the bladder was decreased, (1) the resultant caldera had a larger volume, and (2) faults were initiated earlier. When the bladder was tilted, subsidence was highly asymmetric; faults formed first where the bladder was shallowest. Subsidence then shifted rapidly to where the bladder was deepest, producing an elongate trapdoor caldera which was deepest where the bladder was deepest.
Experimental setup
Conclusions
The observations from our experiments attempt to illustrate some important structural details that could occur during caldera collapse. The experiments also show that piston-style collapse occurs only under certain conditions, and that many other collapse geometries may be common. These details help provide useful information when mapping young calderas where the internal structure is hidden. Structural details of faulting relationships and fault interactions also may provide information concerning hydrothermal pathways. Understanding controls on these pathways is very useful to the mining industry, as volcanogenic massive sulfide and epithermal mineralization may be controlled by these structures within calderas. Hydrothermal pathways also have economic interest for the geothermal power industry. Our observations show that the structural style of faulting varies considerably between the different caldera types. This may be one reason why hydrothermal activity is not always restricted to the area of the ring fault.
The major conclusions from this set of experimental results are the following. (1) Trapdoor-style calderas form as a result of both sagging and faulting, with faults usually propagating laterally from the area of maximum subsidence. (2) Concentric annular faults form from progressive outward growth of the caldera and may be the result of the convex nature of the magma chamber roof. (3) The aspect ratio of the subsiding block affects the collapse style of the caldera. (4) Pre-existing topography controls the position of fault initiation at the surface, promotes early fault initiation and piston style calderas, but does not affect the cross-sectional symmetry of the caldera. (5) Ring structures may be more complicated than previously thought, showing linear portions and crossing faults which produce a complex polygonal form. (6) Tilted and overpressured magma chambers also control the collapse style of the caldera.