Skip to main content

Seismic Anisotropy and Magma Systems

Final Report Summary - SAMS (Seismic Anisotropy and Magma Systems)

This project was to develop a method of using Finite Element Models (FEMs) to calculate seismic velocity variation, strain and stress due to magma movement at active volcanoes to resolve sub-surface geometry, magma volumes and pressures. In particular, to understand the relationship between anisotropic seismic velocities measured using shear wave splitting, and pressurization of magma reservoirs by inverting the shear wave splitting data for anisotropy parameters, including crack geometry and pore content. The models make use of multidisciplinary constraints such as ground deformation, petrology and seismicity, all of which provide evidence for depths, pressures and timing of magma movement at different scales. The use of a good deformation record is key as the pressure sources, inverted for using geodetic data, are needed to model the stress field. The methods have been applied to real volcanoes in order to improve our understanding of specific systems, as well as recognising any general trends.

Seismic monitoring of active volcanoes typically relies on observing changes in the number or type of events, but further information about the magnitude and orientation of stress changes is contained within focal mechanisms, seismic velocities and anisotropy. Seismic anisotropy is an indicator of geometric ordering in a material, where features smaller than the seismic wavelength (e.g. crystals, cracks, pores, layers or inclusions) have a dominant alignment. This alignment leads to a directional variation of elastic wave speed. A shear wave, generated by a local earthquake, travelling through an anisotropic medium will be split into a fast and slow component. An applied stress field can cause micro-cracks to preferentially open parallel to the maximum compressive stress, causing the medium to become seismically anisotropic.

The measurement of shear wave splitting has been found to be a proxy for determining the direction of maximum horizontal stress in the crust. The mechanism of aligned micro-cracks is thought to be the only one that allows seismic anisotropy to vary on observable time scales, and temporal changes are traditionally interpreted as stemming from variations in the stress field due to large earthquakes or magmatic intrusions. There is mounting evidence, however, that the content of the micro-cracks; such as water, steam or other fluids; can have a dramatic effect, not only on the SWS, but on the way the system as a whole reacts to increased pressure. The effects of liquids and gasses have been investigated experimentally and theoretically and have been found to be an important factor. In a volcanic environment, liquid magma, volcanic gasses and hydrothermal are common and so their contributions to SWS and other parameters indicative of stress and strain need to be understood.

During this project, automatic shear wave splitting analysis has been conducted on available data from Kilauea (USA), Tungurahua (Ecuador), and Upptyppingar (Iceland) Volcanoes.

Shear wave splitting analysis of seismic data from a magmatic intrusion in Iceland has revealed that there is a two-layer system of anisotropic media. Comparison of my data with other published results suggests that this phenomenon may occur throughout the island. The consequence of this significant observation is that any future work in this location needs to be interpreted in this context. This work was carried out in collaboration with colleagues at the University of Cambridge.

Analysis of seismic data at Tungurahua Volcano in Ecuador has shown that the routine locations of earthquakes carried out by the Tungurahua Volcano Observatory have significant errors associated with them. Shear wave splitting analysis of this data was calculated using a method developed at the University of Bristol and one developed at Victoria University of Wellington (NZ). The results were compared and were found to be the same to within a 95% confidence interval. This is important to validify both of these methods. The shear wave splitting results themselves show that there is significant azimuthal effect, indicating that the region is extremely heterogeneous and that temporal changes in seismic anisotropy must be scrutinised.

Inversion of shear wave splitting data has been carried out at Kilauea Volcano and has highlighted a region of known magma storage. The results from the inversion correlate with deformation and gas emission data. These results are important because (a) they demonstrate that the inversion technique is an appropriate tool for identifying regions of magma storage and (b) corroborates the existence of the hypothesised magma reservoir, which has recently also been identified with petrological methods.

Finite element modelling of Kilauea Volcano has illuminated the shortfalls of traditional modelling techniques when the pressurised magma reservoir is shallow. In particular, tilt measurements close to extreme topography show anomalous results. I have discovered the reason for these anomalies and am currently in the process of devising formulae relating the geometry of the system to the anomaly. The outcomes of this significant finding are (a) that these formulae will enable better design of monitoring networks and (b) that these apparently anomalous data can be used to constrain the geometry of the subsurface.