Earthquake Stress and Strain through High Resolution Data Analysis: Aftershock Trirggering, Afterslip, and a Novel Borehole Strainmeter Array

Abstract

Studies of earthquake deformation continually push the bounds of geophysical data resolution to better constrain societally relevant fault hazards. In three chapters, I explore high-resolution datasets that enable the study of earthquake processes in further detail—testing long-standing hypotheses and investigating little-studied gaps in our knowledge of earthquake triggering, fault slip, and key observations afforded by new instruments. In the first chapter (Appendix A), co-authors and I pursue an uncertainty analysis for the common static stress transfer hypothesis, whereby coseismic static stress change is postulated as the primary mechanism for triggering aftershocks. Retrospective analysis of this mechanism depends on aftershock rupture plane orientations and locations, which are not always known a priori. Instead of adopting the common assumption that aftershocks occur on planes that are optimally oriented in the stress field, we utilize an unusually large dataset comprising hundreds to thousands of focal mechanisms following three earthquake sequences: 1997 Umbria-Marche (Italy), 2009 L’Aquila (Italy), and 2019 Ridgecrest (California, US). We perturb the focal mechanism solutions within their range of uncertainty to produce a distribution of Coulomb failure stress calculations from the largest earthquakes in each sequence. Results indicate that nearly any aftershock could have been promoted within uncertainty; but, nominally, only ~50-60% are primed for failure. This improves with distances > 5 km from the causative rupture planes, to 63-87%. Furthermore, the classic assumption of optimal plane orientations overpredicts the number of promoted aftershocks, and fails to capture their apparent tendency to occur on inherited tectonic structure. In the second chapter (Appendix B), co-authors and I pursue another process related to postseismic deformation following the Mw 7.1 2019 Ridgecrest earthquake: A detailed geodetic study of afterslip and triggered fault creep, which we contextualize in the stress-driven, frictional rate-dependent framework. Considerable evidence supports this hypothesized framework, but consensus for its validity is not yet achieved in the geophysical community because of differences in modeling approach that lead to variable fault slip distributions, and a lack of detailed slip evolution through time. This becomes more apparent in the relatively uninvestigated span from minutes to months following the coseismic rupture, when common geodetic resolution (e.g. GNSS and InSAR analyses) limit modeling capability. Three borehole strainmeters (BSM) recorded the Ridgecrest earthquake with sub-nanostrain precision, permitting a detailed analysis of afterslip in this time frame. We jointly invert BSM strains and GNSS displacements for coseismic and postseismic slip in 8 independent periods spanning hours to months. The model explains nonmonotonic transient deformation observed at borehole station B921 < 2 km from the rupture plane, and well-approximates main-fault slip in the rate-dependent framework. Slip on subsidiary and neighboring structures, including triggered Garlock fault creep, highlights an active fault network that expands the footprint of postseismic slip, critical to characterizing evolving hazards. Finally, in the third chapter (Appendix C), I calibrate and characterize the recently installed (2021-2022) Alto Tiberina Near Fault Observatory Strainmeter Array (TABOO-STAR), affording tectonic observations that demonstrate expanded measurement capability of the existing superb geophysical network. We test the tidal calibration results against environmental signals and signals from coseismic deformation associated with polarized body waves and static offsets originating from immediate to 1000s km distances. While characterizing the amplitude uncertainty would benefit from future detailed analysis, the orientation resolution achieves sub-nanostrain precision for capturing earthquake-related deformation. One station may further exhibit influence from dynamically-triggered near-field fracture slip and fluid flow. Overall, the analyses demonstrate enhanced capability for detecting near and far field strains beyond the existing geodetic and seismologic network. In culmination, this dissertation fills observational gaps in our knowledge of data capabilities and fault processes, aiming to enhance our understanding of global active tectonic hazards.