How Deep Do Ruptures Penetrate in Large Earthquakes?

Columbia University, New York NY

Investigators

Abstract

This award is funded under the American Recovery and Reinvestment Act of 2009 (Public Law 111-5). Technical Description: In the shallow crust, most of the strain at plate boundaries is relieved during sudden rare large earthquakes. Below the shallow locked portion, strain has often been thought as being relieved through steady creep, or, in the case of layers just below the locked portions, slow decaying afterslip. In the last few years, observations of Episodic Tremor and Slip in subduction zones has caused some rethinking of this simple picture. An even bigger challenge to this picture has come from new theoretical work suggesting rapid coseismic slip during large earthquakes might penetrate much deeper below the locked fault than previously thought. This possibility, that rapid coseismic deep slip may have been, not missed, but rather misplaced, inferred erroneously to be happening at shallower locked depths, has significant implications for our picture of how strain relief occurs at plate boundaries, and for seismic hazard estimates. This work seeks to further explore and constrain this possibility. Two sets of projects are proposed. The first set focuses on a newly identified quantification of the deep slip behavior found to have significant impacts on seismic hazard estimates: the ratio of seismogenic moment to total moment (or seismogenic moment fraction). This ratio is argued to impact magnitude-area scaling, moment balanced estimates of repeat times of large events, and high frequency shaking hazard. Further re-examing the implications of this ratio, and constraining its values forms a core of this set of projects. Examining this ratio in a forward modeling context to capture its robustness to friction and other uncertainties is a central proposed project. Specifically, variations in friction in the seismogenic layer, and frictional forms in the deeper stably sliding layer will be examined, and robustness of this dimensionless ratio explored. Additional projects studying the seismogenic moment fraction and its impact on quantities relevant to seismic hazard will be carried out. In particular, we will examine how well the estimated seismogenic moment fraction corrections indeed match magnitude area scaling and large event repeat times. The second set of projects focuses on measurements which have the potential to provide direct observational constraints of the phenomena of deep slip below the locked portion of faults. Taking advantage of the geometry of dipping faults, where depth and horizontal location are associated, one project will examine the potential to detect deep slip at the surface in dipping fault geometries. In the case of subduction zones, the project aims to see whether slip propagation into zones where Episodic Tremor and Slip occur might be observable during large earthquakes. A final project examines magnitude-area scaling on dipping faults, in light of our new scaling laws. Broader Significance: In addition to the broad areas of earth science touched by this research, including geological, seismological, geodetic, and laboratory observations, there are significant potential practical benefits to society. Current methods for estimating seismic hazard rely on a series of of parameterizations of ruptures to construct probabilities of earthquake occurrence. Uncertainties in the parameterizations lead to uncertainties in the estimates, and added uncertainties mean added costs, when trying to manage risk. It has become increasingly clear that a more physics-based approach would provide better constraints on parameterizations and better hazard estimates. This work has important implications for a number of key parameterizations used in seismic hazard estimates. These seismic hazard estimates have big societal impacts, and are used for setting building codes and earthquake insurance rates. Improving estimates help in better allocating resources to mitigate these hazards, ultimately saving lives and property.