UC Riverside

Slow earthquakes essentially represent a gray area of the slip spectrum, where regular fast slip earthquakes and aseismic creep constitute the spectrum’s end members. Since their discovery in 2003 [Rogers and Dragert, 2003], increasing seismic observations, in addition to models, geodetic observations, and laboratory studies, have helped elucidate the nature of these events that vary quite vastly amongst themselves. Empirically, however, they share a linear moment rate scaling law that distinguishes them from regular earthquakes [Ide et al., 2007]. Slow earthquakes include, but are not limited to, low frequency earthquakes (LFEs), tremor, very low frequency earthquakes (VLFEs), slow slip events (SSEs), and episodic tremor and slip (ETS) events. Each type of these events has distinctive properties that makes their detection challenging. Furthermore, their relationships to each other remains unclear, particularly in terms of source properties. They may be important in the context of regular earthquakes and seismic hazard assessment as they appear to have a correlation with large events. Some slow earthquakes have been observed spatiotemporally preceding regular earthquakes such as the 2011 Mw 9.0 Tohoku earthquake [Kato et al., 2012] and the 2014 Mw 8.1 Iqueique earthquake [Ruiz et al., 2014]. Additionally, a region near the trench known to produce VLFEs ruptured coseismically during the Tohoku earthquake producing a larger earthquake and tsunami than was previously anticipated [Ide et al., 2012]. Ultimately, more observations and studies are required to understand the mechanical properties of slow earthquake source physics and to understand their greater role in the earthquake cycle.

The research shared herein describes a variety of observations of slow earthquakes in discrete tectonic settings. The first section of this thesis examines a locked section of the San Jacinto Fault – a non-plate boundary transform fault – called the Anza Gap. The study begins in chapter 2 by introducing the detection of ambient tectonic tremor [Hutchison and Ghosh, 2017], likely acting as a seismic manifestation of slow-slip or deep creep. The latter chapter within this section, chapter 3, examines a series of teleseismically triggered tremor and small earthquakes leading up to the June 10, 2016 Mw 5.2 Borrego earthquake. This cascade of tremor and foreshocks is interpreted to signify seismic manifestations of deep creep triggered by the energy from two distant earthquakes that occurred earlier in the day. The second section of this thesis focuses on VLFEs in the Cascadia subduction zone, a plate boundary where the Juan de Fuca plate is subducting beneath the North American plate. The first chapter in this section (chapter 4) uses grid search centroid moment tensor inversion during the 2014 ETS event to detect VLFEs. Unlike previous studies where VLFEs and tremor are spatiotemporally coincident, tremor and VLFE are only quasicoincidentally occurring during this ETS event, indicating for the first time that VLFE and tremor may have discrete sources. Chapter 5 successfully employs matched filter analysis for VLFEs detection in Cascadia, indicating that VLFEs are repeating events, generating a high resolution temporal catalog for the 2011 and 2014 ETS events. These data are compared to SSE data in both ETS events. VLFEs acts as an accurate proxy for both SSEs, even when tremor does not. This further indicates that tremor and VLFEs may have discrete source mechanics, though they may both be related through slow-slip. In addition, an inter-ETS VLFE catalog shows sustained background VLFE activity throughout the inter-ETS period. Finally, in Chapter 6, a grid-search centroid moment tensor inversion VLFE catalog is created between the 2011 and 2014 ETS events. This allows for further investigation of VLFE behavior, particularly during inter-ETS periods. These new observations further underscore a quasi-spatiotemporal relationship between VLFE and tremor, suggesting an underlying process that uniting these events. Through these observations, I propose that there are ongoing small, and largely undetectable SSEs that are manifested through inter-ETS slow earthquake seismic signatures, "mini-ETS" events, and that these events would fill the gap in the slow earthquake scaling law [Ide et al., 2007]. These observations support the Colella et al., [2011] rate- and state- dependent friction models that suggest a non-linear scaling relationship between moment and duration. I also propose slow-slip is the driving mechanism uniting all slow earthquake activity, or the gray area, of the slip spectrum [e.g. Wech and Bartlow, 2014, Wech et al., 2009, Bartlow et al., 2011].