Understanding Marine Magnetotellurics

Abstract

The theory of plate tectonics proposes that the Earth’s lithosphere is separated into rigid plates which are capable of motion through interactions with the underlying asthenosphere. Following its adoption in the 1960’s, it has become the prominent theory used to understand geodynamic processes within the Earth. The mechanisms of tectonism are of particular interest due to their implications regarding the formation of economic resources. Despite numerous studies attempting to characterise these mechanisms, the lithosphere-asthenosphere rheological contrast (LARC) remains an enigmatic component of plate tectonic theory. The magnetotelluric (MT) method is of particular interest when investigating the upper mantle as it is primarily sensitive to electrical conductivity. From electrical conductivity, conclusions regarding the temperature, pressure, physical and chemical state, porosity, and permeability of rocks can be inferred. This thesis examines laboratory conductivity measurements with ocean-bottom MT data collected from the Pacific Ocean. From these data, I create an upper mantle reference model for electrical conductivity and propose a hybrid MT impedance which improves the bandwidth and confidence intervals of ocean-bottom MT data. Between 50 km and 100 km depth, MT data predicts variable electrical conductivity structures. My reference model is able to encapsulate this variability with a function which varies according to hydration, partial melt, and the age of the overlying oceanic lithospheric. For 200 km to 400 km, the hydrated end-member of this reference model predicts the convergence of conductivity structures with increasing depth observable in published conductivity structures. As such, this reference model constrains the presence of hydration and partial melt within the LARC for a range of lithospheric ages and is a representative model of the Earth’s oceanic lithosphere and asthenosphere. Following an analysis of ocean-bottom EM field observations, I observe the attenuation of magnetic field variations by the conductive ocean water. This attenuation results in MT impedances which are difficult to interpret using available modelling algorithms. In contrast, the calculation of a hybrid MT impedance using ocean-bottom electric and continental magnetic fields is observed to improve the signal-to-noise ratio. This improved signal-to-noise ratio extends the usable bandwidth of ocean-bottom MT data from just over one decade to approximately four decades. It is important to note that this impedance represents the normalisation of ocean-bottom electric fields using continental magnetic fields. As a consequence, alterations must be made to modelling algorithms before attempting to reproduce hybrid impedances. Finally, a case study is conducted to assess the validity of my reference model and hybrid impedance. To do so, structurally simple forward models of both standard and hybrid impedances are calculated. The conductivity structure of this model was constrained by a 3-Dimensional inversion of continental MT data, controlled source EM (CSEM) data, and my reference model. From this model, hybrid impedances are observed to reproduces four decades of data measured by numerous receivers. From this evidence, I conclude that my reference model constrains the LARC and that it realistically represents the upper mantle. Additionally, I conclude that my hybrid impedance is a useful alternative to traditional MT impedance when conducting oceanbottom MT studies.