Magmatic intrusions are common to volcanoes worldwide. The emplacement of these intrusions disturbs the temperature and pressure conditions within the volcano and presents a potential heat source to develop and drive a hydrothermal system. Consequently, magmatic intrusions change the volume, topography, and properties of the volcano. In this thesis, I investigate the physical and mechanical effects a small shallow intrusion produces in its andesitic host rock.

The thesis opens with a preliminary study (Chapter 2) in which I thermally stress andesitic samples to demonstrate the geomechanical effects a rapidly emplaced and cooled magma body would have in hydrothermally altered andesitic lavas. Then, the thesis focuses the investigation (Chapters 3, 4, 5, and 6) to the primary field area, Pinnacle Ridge, Mt. Ruapehu, New Zealand, where a glacially dissected fossil hydrothermal system serves as a natural laboratory for studying the effects of an intrusion on the physical and mechanical rock properties of an andesitic host rock.

The opening study (Chapter 2) of this thesis considers the effects thermal stresses have on hydrothermally altered andesite from the Rotokawa Geothermal Field. The experiments demonstrate that short-duration thermal exposure (350 – 739 °C) in 20 MPa of H2Ofluid/vapour can increase permeability by over an order of magnitude (10-18 m2 to > 10-17 m2) as a result of an increase in porosity (< 1 % – > 7 %) driven by both chemical reaction and microfracturing. These results imply that intrusions may increase the porosity and permeability in the host rocks proximal to their emplacement, serving as targets for geothermal exploration.

Pinnacle Ridge is composed of several distinct geotechnical units surrounding small, irregularly shaped intrusions. I collected samples from the differing geotechnical units at varying distances to the largest intrusion (Chapter 3). Intact rock testing reveals the physical and mechanical (i.e. petrophysical and geomechanical) rock properties of one geotechnical unit, the brecciated lava margins, correlate with distance to the largest intrusion. Porosity and permeability of the brecciated lava margins decrease while strength increases approaching the intrusion. The study also finds a previously unclassified geotechnical unit (i.e. hydrothermal vein) for which I argue its own geotechnical classification on the grounds of its unique combination of low permeability and low uniaxial compressive strength.

The intact rock characterization from Pinnacle Ridge revealed a unique sample block with two distinct types of alteration (intermediate and advanced argillic). In Chapter 4, triaxial deformation testing of cores from this sample reveal that the alteration type changes the deformation mode (brittle to ductile) of andesite at low effective confining pressures (~10 MPa). The change in deformation behaviour is the result of advanced argillic alteration increasing the porosity through dissolution and partially replacing the primary mineral assemblage with weaker more ductile clay (i.e. kaolinite and smectite). I argue that this transition could explain the absence of volcano tectonic earthquakes at certain depths prior to volcanic eruptions.

When investigating the physical and mechanical effects of magmatic intrusion emplacement, scale requires consideration. In Chapter 5, I present my field observations from Pinnacle Ridge and consider them in conjunction with the lab data presented in Chapter 3. In general, discontinuities in altered rock masses are more abundant and have smoother surfaces than in unaltered rock masses. Numerical modeling demonstrates that the dichotomy between the discontinuity properties of the unaltered and altered rock masses creates the potential for altered brecciated lava margins to be weaker than unaltered brecciated lava margins despite the higher intact rock porosity and lower intact rock strength of the latter.

The numerical modeling in Chapter 5 demonstrates that outcrop scale properties must be measured in tandem with intact rock properties when considering rock mass strength. However, the modeling presented in Chapter 5 relied upon rock parameter assumptions for generic andesite, because the specialized data for altered andesite did not exist. In Chapter 6, I conduct triaxial deformation testing for all the geotechnical units of Pinnacle Ridge to complete the first full set of Generalised Hoek-Brown failure criterion parameters for an intrusion and its altered host rock. Using these data, I numerically model volcano stability with varying pore pressures within the geotechnical units. In doing so, I demonstrate that hydrothermal vein material is the weakest rock mass in the fossil hydrothermal system at Pinnacle Ridge. These results have implications for the importance of modelling even thin geotechnical units such as the hydrothermal vein when considering volcano stability.

In summary, I use a combination of lab- and field-based rock property testing, and finite element numerical modelling to examine the full physical and mechanical effects intrusions produce in volcanic host rock. I show that the effects of intrusions in volcanic (andesitic) rock varies with time (Chapter 2 versus Chapter 3), distance (Chapter 3), and scale (Chapter 3 versus Chapter 5; Chapter 6) as well as the environmental conditions (e.g. effective confining pressure) of the host rock (Chapters 3 and 6) and alteration type of the host rock (Chapter 4). Given the global and ubiquitous distribution of magmatic intrusions, these results have profound worldwide implications for the geothermal power industry (Chapters 2, 3, and 5), volcano monitoring (Chapters 4, 5, and 6), and volcano slope stability (Chapters 2, 3, 4, 5, and 6).