Investigation of source spectral models and scaling for slow slip processes in subduction zones

点击翻译按钮获取中文摘要

This award is funded under the American Recovery and Reinvestment Act of 2009 (Public Law 111-5)Recently, a new mode of coupled deformation and weak seismic radiation has been discovered in numerous subduction zones globally. Slow slip and weak tremor occur episodically down-dip from the locked portion of the megathrust interface at quasi-regular intervals and have been termed Episodic Tremor and Slip (ETS). The likely interaction of ETS with the locked and hazardous portions of subduction zones mandate better understanding of this intriguing phenomenon.The researcher will develop appropriate models to relate the different sizes and durations of these slow slip processes. The models will facilitate comparison of different instances of these processes, as well as comparisons of available data at various frequency ranges. This is an avenue to understanding how the physics of slow slip processes differs from the better-understood physics of earthquakes.Broader impacts include furthering the education of one or more graduate students, and possibly undergraduates. This work will elucidate a new mode of fault slip, and could have a transformative impact on studies in seismology, geodesy, friction and rock mechanics. A better understanding of the relationship between smaller and larger slow slip events in subduction zones may help forecast the occurrence of large earthquakes there, and thus could be directly relevant to seismic hazard. As well, episodic tremor and slip events provide a valuable opportunity to advance public awareness of the regional seismic hazard. At present we lack a fundamental understanding of the nature and basic physics of the ETS source process. Nor do we understand the relationship between the growing menagerie of recent observations of slow slip processes, which include tiny low frequency earthquakes, somewhat larger very low frequency earthquakes, slow slip events, episodic tremor and slip, and large silent earthquakes. These slow slip processes are characterized by emergent beginnings, weak amplitudes, and long durations. Their spectra are depleted in high-frequency energy compared with regular earthquakes of similar moment.One way in which seismologists understand regular earthquakes is through models of the source spectrum such as the omega-squared model. For regular earthquakes, a source spectral model represents our understanding of earthquakes as a scale-invariant process that is self-similar over a very large range of seismic moments. Such a unifying model is lacking for slow slip phenomena. I propose to develop appropriate source scaling models for slow slip processes and compare with available data at various frequency ranges to constrain possible models of the source spectrum of slow slip events. Constructing such models involves determining relations among and constraints on spectral slope, corner frequency, stress drop, and propagation velocity.Burgeoning numbers of observations of slow slip processes from subduction zones, particularly Cascadia and Japan render this a timely proposal. An appropriate spectral model will facilitate comparison of data from different subduction zones, frequency bands, and magnitudes of ETS events. In some cases these data should be compared in terms of absolute moment rate spectra. In other cases only a relative comparison can be made, but that may still place strong constraints on potential source spectral models. Whereas regular earthquakes have a roughly constant rupture velocity underpinning their self-similar nature, for slow slip events there is evidence that rupture propagation velocity is size-dependent. Similarly, stress drop, although constant over a very large range of sizes of regular earthquakes, may be size-dependent for slow slip processes. Constraining valid source spectral models will reveal relationships between important physical properties such as stress drop, rupture velocity, and moment.The proposed source spectral model construction and comparison with various types of data will address many key questions about slow slip processes and further elucidate the physics of the ETS process. Understanding whether and how slow slip processes scale with size would represent a fundamental advance, and as such has the potential to be truly transformative.