Unlocking the secrets of slow slip in Cascadia using low-frequency earthquakes

Abstract

Recent discoveries in subduction zones worldwide--including here in Cascadia--have illuminated the once shrouded process of plate convergence below the seismogenic zone. Early geodetic [Dragert, et al., 2001] and seismic [Obara, 2002] signals were observed to correlate in space and time, and were associated with periodic episodes of deep slow slip, termed Episodic Tremor and Slip (ETS) [Rogers and Dragert, 2003]. In this dissertation, I present evidence further detailing the process of where, how, and how often deep slow slip occurs using several catalogs of low-frequency earthquakes (LFEs) as slow slip indicators. In the first section I compare four distinct LFE families that span the range of the ETS zone beneath western Washington State. I find that LFE behavior varies systematically with depth: LFE moments, swarm durations, and swarm recurrence intervals are all largest in the updip portion of the ETS zone, and smallest in the downdip portion. I interpret these systematic differences as a result of variation in fault strength on the subduction interface--with the strongest coupling found updip (near the seismogenic zone), and the weakest coupling found downdip. In the second section I look within individual LFE families and perform double-difference event relocations to map out the spatial extent of the LFE patch (or patches) responsible for LFE generation. I determine LFE locking efficiency from estimates of LFE density and released seismic moment. I also track LFE migrations over time in an effort to map the progression of slow slip fronts, rapid tremor reversals (RTRs), and other phenomena.