Marine electromagnetic (EM) geophysics is an up-and-coming branch of the geosciences that is allowing for the advancement in our understanding of key properties of the oceanic lithosphere and subduction dynamics, particularly in how deformation manifests geophysically and how it evolves through time and under various conditions. This dissertation focuses on two unique marine EM data sets collected at the Hikurangi subduction zone, New Zealand, and on 33 Ma Pacific lithosphere. Analysis of the former, which constitutes the bulk of this dissertation, offers the first kilometer-scale characterization of offshore, margin-wide electrical resistivity variations at a subduction zone and provides an electrical framework for discussing the potential causes of along-strike differences in megathrust slip at the Hikurangi Margin. The latter data set is used to constrain electrical anisotropy of the shallow lithosphere, which enables an interpretation of the deformation history of normal oceanic lithosphere.
Chapter 2 of this dissertation gives a brief overview of the physical underpinnings of EM methods with attention given to the marine magnetotelluric (MT) and controlled-source electromagnetic (CSEM) methods. Maxwell's equations are reviewed and the relevant derivations leading to the temporal and spatial behavior of EM waves for the frequencies used in this dissertation (~0.001--0.1 Hz) are presented.
Chapter 3 focuses on the tectonic background of the Hikurangi Margin and on processing of the MT and CSEM data. Interest in the Hikurangi Margin has arisen both because of its proximity to the inhabitants of New Zealand and due to the recognition of several properties that vary along the strike of the margin. The most intriguing of those variations, and most concerning from a natural hazard perspective, are the along-strike change in interseismic coupling and slow slip event (SSE) occurrence, with stronger coupling and deeper, infrequent SSEs realized in the southern Hikurangi Margin and weaker coupling and shallower, more frequent SSEs in the north. Several proposed causes of these variations are cited, including differences in sediment thickness and roughness of the incoming plate, changes in the plate interface geometry, and the effect of geological terranes in the forearc on pore pressure. But the degree to which any or all of these factors affect interseismic coupling remains an open question. The remainder of Chapter 3 is devoted to detailing the steps involved in processing the marine MT and CSEM data. A workflow for optimizing MT response function estimation is presented and improvements to the marine CSEM processing scheme are described.
In Chapter 4 of this dissertation, inversions of the data collected at the southern Hikurangi Margin are presented, and these resistivity models are compared with co-located seismic data. Individual inversions of the CSEM and MT data along with joint inversion of the two data sets highlights the distinct sensitivities and resolving capabilities of each data type. A thick (4--6 km) sediment package covers the Hikurangi Plateau of the incoming plate. The plateau itself is evident as a dipping resistor (>10 Ω-m) that approximately corresponds with the seismically interpreted depth of the Hikurangi Plateau. Resistors in the shallow forearc are interpreted as free gas or gas hydrate, which is prevalent at the Hikurangi Margin. A resistive anomaly beneath one of two main ridges appears to comprise the footwall of a thrust fault, which potentially implies a high permeability system that allows for preferential dewatering of the footwall. Using available P-wave velocity data for this region, equations relating resistivity to velocity are derived.
The resistivity presented in Chapter 4 and Archie's law are used to derive porosity models of the southern Hikurangi profile in Chapter 5. Vertical compaction is shown to dominate trends in porosity. A reference compaction porosity model is approximated and removed from the resistivity-derived porosity model in order to identify porosity trends distinct from compaction. A deepening in the negative porosity anomaly of the shallow incoming plate sediments as they approach the trench suggests these sediments experience compression several kilometers seaward of the main frontal thrust. This could represent the early stages of protothrust zone development. An increasingly positive porosity anomaly observed in the sedimentary unit just above the Hikurangi Plateau as it nears the trench may indicate heightened fluid overpressures in an incipient décollement.
In Chapter 6 of this dissertation, inversions of the central Hikurangi Margin are shown and discussed. Compared to resistivity in the southern Hikurangi Margin, the forearc and incoming plate of the central Hikurangi Margin are more complex in their resistivity structure, possibly due to the impact of rougher seafloor. Extensive evidence for free gas or gas hydrates is found as shallow resistive anomalies in these models. Other anomalous resistors may correspond to exhumed terranes in the forearc. Anomalous forearc conductors could indicate sediment underplating or damage zones associated with subducting topography.
Chapter 7 shows the resistivity and porosity of the northern Hikurangi Margin and offers the first detailed electrical image of a seamount prior to and during subduction. The seamount on the incoming plate is shown to have a thin, resistive cap that traps a conductive matrix of porous volcaniclastics and altered material over a resistive core. Again applying Archie's law to estimate porosity from resistivity reveals that the seamount will allow ~3.2--4.7x more water than normal, unfaulted oceanic lithosphere to subduct with the seamount. In the forearc, a sharp, resistive peak on the slab is interpreted as the core of a subducting seamount. This cone of high resistivity lies directly beneath a prominent conductive anomaly in the upper plate. Burst-type repeating earthquakes and other seismicity from a recent SSE cluster in and around this conductive anomaly, which seems to implicate the subducting seamount in the generation of fluid-rich damage zones in the forearc. The interaction of the subducting topography with the upper plate will thus alter the effective normal stress at the plate interface by modulating fluid overpressure. The results in this chapter show that subducting topography can transport large volumes of water to the forearc and that such topography is able to severely modify the structure and physical conditions of the upper plate, which may influence the location and timing of SSEs.
Finally, Chapter 8 provides a robust constraint on the electrical azimuthal anisotropy of oceanic lithosphere. The data for this chapter were collected in a region of oceanic lithosphere removed from the influence of plate boundaries and intraplate volcanism. The survey design was chosen to maximize azimuthal coverage so as to constrain the directional dependence of resistivity. Inversions of the data resulted in an anisotropic resistivity model wherein the crust is ~18-36x more conductive in the paleo mid-ocean ridge direction than the perpendicular paleo-spreading direction. In the uppermost mantle conductivity is ~29x higher in the paleo-spreading direction. The crustal anisotropy is interpreted to result from sub-vertical porosity created by ridge parallel normal faulting during extension of the young crust and thermal stress-driven cracking from cooling of mature crust. Anisotropy in the uppermost mantle implies that shearing of mantle olivine during plate formation generates a strong electrical signal that is preserved as the plate ages. Reanalysis of EM data collected offshore Nicaragua suggests that the Pacific Plate electrical anisotropy is not a local anomaly but rather may be prevalent throughout oceanic lithosphere.
| Identifer | oai:union.ndltd.org:columbia.edu/oai:academiccommons.columbia.edu:10.7916/d8-2z8a-x818 |
| Date | January 2022 |
| Creators | Chesley, Christine Jessie |
| Source Sets | Columbia University |
| Language | English |
| Detected Language | English |
| Type | Theses |
Page generated in 0.0035 seconds