Subduction zone fault processes range from tsunami-generating megathrust events to aseismic creep along the deeper portions of the fault. Episodic tremor and slow slip (ETS) represents the transition between these two regimes, where slip occurs at semi-regular recurrence intervals of months-to-years. These events are also accompanied by low frequency earthquakes, referred to as tremor. The study of ETS in Cascadia has been made possible by the enhancement of large-scale seismic and geodetic networks. In this dissertation, I use a range of geodetic and seismic observations at sub-daily to decadal time scales to investigate the kinematic behavior of individual ETS events, as well as the long-term behavior of the ETS zone and its relationship with the updip seismogenic zone.
In Cascadia, current seismic hazard maps use the ETS zone as the downdip limit of rupture during future megathrust events. In Chapter II, I utilize uplift rates derived from 80 years of leveling measurements to explore the possibility that long-term strain accumulation exists near the ETS zone. The uplift rates are consistent with a region of 10-20% locking on the updip side of the ETS zone. The lack of associated topography indicates that the accumulated strain must be released during the megathrust cycle. The correlation of tremor and slip in Cascadia suggests there is an inherent relationship between the two. In Chapter III, I develop a method for using tremor as a proxy for slip to assess the spatial relationship of tremor and slip. I compare predictions of tremor-derived slip models to results from static inversions of GPS offsets by modeling slip based on the density of tremor. These comparisons suggest that the correlation of tremor and slip is variable along strike and along dip. In Chapter IV, I explore how borehole strainmeters can improve our resolution of slip on the plate interface. I incorporate strainmeters into joint, time-dependent kinematic inversions with GPS data. The temporal resolution of strainmeters provides improved constraints when deriving time-dependent slip estimates during slow slip events, allowing us to better image the kinematics of slow slip.
This dissertation includes previously published and unpublished material.
Identifer | oai:union.ndltd.org:uoregon.edu/oai:scholarsbank.uoregon.edu:1794/20720 |
Date | 21 November 2016 |
Creators | Krogstad, Randy |
Contributors | Humphreys, Eugene |
Publisher | University of Oregon |
Source Sets | University of Oregon |
Language | en_US |
Detected Language | English |
Type | Electronic Thesis or Dissertation |
Rights | All Rights Reserved. |
Page generated in 0.009 seconds