ALBERT

Description: A prototype earthquake early warning (EEW) system is currently in development in the Pacific Northwest. We have taken a two-stage approach to EEW: (1) detection and initial characterization using strong-motion data with the Earthquake Alarm Systems (ElarmS) seismic early warning package and (2) the triggering of geodetic modeling modules using Global Navigation Satellite Systems data that help provide robust estimates of large-magnitude earthquakes. In this article we demonstrate the performance of the latter, the Geodetic First Approximation of Size and Time (G-FAST) geodetic early warning system, using simulated displacements for the 2001 M w  6.8 Nisqually earthquake. We test the timing and performance of the two G-FAST source characterization modules, peak ground displacement scaling, and Centroid Moment Tensor-driven finite-fault-slip modeling under ideal, latent, noisy, and incomplete data conditions. We show good agreement between source parameters computed by G-FAST with previously published and postprocessed seismic and geodetic results for all test cases and modeling modules, and we discuss the challenges with integration into the U.S. Geological Survey’s ShakeAlert EEW system.

Description: Great earthquakes rarely occur within active accretionary prisms, despite the intense long-term deformation associated with the formation of these geologic structures. This paucity of earthquakes is often attributed to partitioning of deformation across multiple structures as well as aseismic deformation within and at the base of the prism ( Davis et al. , 1983 ). We use teleseismic data and satellite optical and radar imaging of the 2013 M w  7.7 earthquake that occurred on the southeastern edge of the Makran plate boundary zone to study this unexpected earthquake. We first compute a multiple point-source solution from W -phase waveforms to estimate fault geometry and rupture duration and timing. We then derive the distribution of subsurface fault slip from geodetic coseismic offsets. We sample for the slip posterior probability density function using a Bayesian approach, including a full description of the data covariance and accounting for errors in the elastic structure of the crust. The rupture nucleated on a subvertical segment, branching out of the Chaman fault system, and grew into a major earthquake along a 50° north-dipping thrust fault with significant along-strike curvature. Fault slip propagated at an average speed of 3.0 km/s for about 180 km and is concentrated in the top 10 km with no displacement on the underlying décollement. This earthquake does not exhibit significant slip deficit near the surface, nor is there significant segmentation of the rupture. We propose that complex interaction between the subduction accommodating the Arabia–Eurasia convergence to the south and the Ornach Nal fault plate boundary between India and Eurasia resulted in the significant strain gradient observed prior to this earthquake. Convergence in this region is accommodated both along the subduction megathrust and as internal deformation of the accretionary wedge. Online Material: Figures showing waveform fits, focal mechanism, root mean square misfit, variation of the shear modulus, covariance functions, and model predictions.

Description: This study lays the groundwork for a new generation of earthquake source models based on a general formalism that rigorously quantifies and incorporates the impact of uncertainties in fault slip inverse problems. We distinguish two sources of uncertainty when considering the discrepancy between data and forward model predictions. The first class of error is induced by imperfect measurements and is often referred to as observational error. The second source of uncertainty is generally neglected and corresponds to the prediction error, that is the uncertainty due to imperfect forward modelling. Yet the prediction error can be shown to scale approximately with the size of earthquakes and thus can dwarf the observational error, particularly for large events. Both sources of uncertainty can be formulated using the misfit covariance matrix, C , which combines a covariance matrix for observation errors, C d and a covariance matrix for prediction errors, C p , associated with inaccurate model predictions. We develop a physically based stochastic forward model to treat the model prediction uncertainty and show how C p can be constructed to explicitly account for some of the inaccuracies in the earth model. Based on a first-order perturbation approach, our formalism relates C p to uncertainties on the elastic parameters of different regions (e.g. crust, mantle, etc.). We demonstrate the importance of including C p using a simple example of an infinite strike-slip fault in the quasi-static approximation. In this toy model, we treat only uncertainties in the 1-D depth distribution of the shear modulus. We discuss how this can be extended to general 3-D cases and applied to other parameters (e.g. fault geometry) using our formalism for C p . The improved modelling of C p is expected to lead to more reliable images of the earthquake rupture, that are more resistant to overfitting of data and include more realistic estimates of uncertainty on inferred model parameters.

Description: Inverting geophysical data has provided fundamental information about the behavior of earthquake rupture. However, inferring kinematic source model parameters for finite-fault ruptures is an intrinsically underdetermined problem (the problem of nonuniqueness), because we are restricted to finite noisy observations. Although many studies use least-squares techniques to make the finite-fault problem tractable, these methods generally lack the ability to apply non-Gaussian error analysis and the imposition of nonlinear constraints. However, the Bayesian approach can be employed to find a Gaussian or non-Gaussian distribution of all probable model parameters, while utilizing nonlinear constraints. We present case studies to quantify the resolving power and associated uncertainties using only teleseismic body waves in a Bayesian framework to infer the slip history for a synthetic case and two earthquakes: the 2011 M w  7.1 Van, east Turkey, earthquake and the 2010 M w  7.2 El Mayor–Cucapah, Baja California, earthquake. In implementing the Bayesian method, we further present two distinct solutions to investigate the uncertainties by performing the inversion with and without velocity structure perturbations. We find that the posterior ensemble becomes broader when including velocity structure variability and introduces a spatial smearing of slip. Using the Bayesian framework solely on teleseismic body waves, we find rake is poorly constrained by the observations and rise time is poorly resolved when slip amplitude is low. Electronic Supplement: Figures of histograms of slip and rise time, waveform comparisons between data and synthetics, and slip velocity along the fault plane for a synthetic case, as well as for the 2011 M w  7.1 Van, east Turkey, earthquake, and the 2010 M w  7.2 El Mayor–Cucapah, Baja California, earthquake.

Description: Earthquake-related fault slip in the upper hundreds of meters of Earth’s surface has remained largely unstudied because of challenges measuring deformation in the near field of a fault rupture. We analyze centimeter-scale accuracy mobile laser scanning (MLS) data of deformed vine rows within ±300 m of the principal surface expression of the M (magnitude) 6.0 2014 South Napa earthquake. Rather than assuming surface displacement equivalence to fault slip, we invert the near-field data with a model that allows for, but does not require, the fault to be buried below the surface. The inversion maps the position on a preexisting fault plane of a slip front that terminates ~3 to 25 m below the surface coseismically and within a few hours postseismically. The lack of surface-breaching fault slip is verified by two trenches. We estimate near-surface slip ranging from ~0.5 to 1.25 m. Surface displacement can underestimate fault slip by as much as 30%. This implies that similar biases could be present in short-term geologic slip rates used in seismic hazard analyses. Along strike and downdip, we find deficits in slip: The along-strike deficit is erased after ~1 month by afterslip. We find no evidence of off-fault deformation and conclude that the downdip shallow slip deficit for this event is likely an artifact. As near-field geodetic data rapidly proliferate and will become commonplace, we suggest that analyses of near-surface fault rupture should also use more sophisticated mechanical models and subsurface geomechanical tests.

Description: We present a fully Bayesian inversion of kinematic rupture parameters for the 2011 M w 9 Tohoku-oki, Japan earthquake. Albeit computationally expensive, this approach to kinematic source modelling has the advantage of producing an ensemble of slip models that are consistent with physical a priori constraints, realistic data uncertainties, and realistic but simplistic uncertainties in the physics of the kinematic forward model, all without being biased by non-physical regularization constraints. Combining 1 Hz kinematic GPS, static GPS offsets, seafloor geodesy and near-field and far-field tsunami data into a massively parallel Monte Carlo simulation, we construct an ensemble of samples of the posterior probability density function describing the evolution of fault rupture. We find that most of the slip is concentrated in a depth range of 10–20 km from the trench, and that slip decreases towards the trench with significant displacements at the toe of wedge occurring in just a small region. Estimates of static stress drop and rupture velocity are ambiguous. Due to the spatial compactness of the fault rupture, the duration of the entire rupture was less than approximately 150 s.

Description: Locating earthquakes from the beginning of the modern instrumental period is complicated by the fact that there are few good-quality seismograms and what traveltimes do exist may be corrupted by both large phase-pick errors and clock errors. Here, we outline a Bayesian approach to simultaneous inference of not only the hypocentre location but also the clock errors at each station and the origin time of the earthquake. This methodology improves the solution for the source location and also provides an uncertainty analysis on all of the parameters included in the inversion. As an example, we applied this Bayesian approach to the well-studied 1909 M w 7 Taipei earthquake. While our epicentre location and origin time for the 1909 Taipei earthquake are consistent with earlier studies, our focal depth is significantly shallower suggesting a higher seismic hazard to the populous Taipei metropolitan area than previously supposed.

Description: The estimation of finite fault earthquake source models is an inherently underdetermined problem: there is no unique solution to the inverse problem of determining the rupture history at depth as a function of time and space when our data are limited to observations at the Earth's surface. Bayesian methods allow us to determine the set of all plausible source model parameters that are consistent with the observations, our a priori assumptions about the physics of the earthquake source and wave propagation, and models for the observation errors and the errors due to the limitations in our forward model. Because our inversion approach does not require inverting any matrices other than covariance matrices, we can restrict our ensemble of solutions to only those models that are physically defensible while avoiding the need to restrict our class of models based on considerations of numerical invertibility. We only use prior information that is consistent with the physics of the problem rather than some artefice (such as smoothing) needed to produce a unique optimal model estimate. Bayesian inference can also be used to estimate model-dependent and internally consistent effective errors due to shortcomings in the forward model or data interpretation, such as poor Green's functions or extraneous signals recorded by our instruments. Until recently, Bayesian techniques have been of limited utility for earthquake source inversions because they are computationally intractable for problems with as many free parameters as typically used in kinematic finite fault models. Our algorithm, called cascading adaptive transitional metropolis in parallel (CATMIP), allows sampling of high-dimensional problems in a parallel computing framework. CATMIP combines the Metropolis algorithm with elements of simulated annealing and genetic algorithms to dynamically optimize the algorithm's efficiency as it runs. The algorithm is a generic Bayesian Markov Chain Monte Carlo sampler; it works independently of the model design, a priori constraints and data under consideration, and so can be used for a wide variety of scientific problems. We compare CATMIP's efficiency relative to several existing sampling algorithms and then present synthetic performance tests of finite fault earthquake rupture models computed using CATMIP.

Description: The basic physics of earthquakes is such that strong ground motion cannot be expected from an earthquake unless the earthquake itself is very close or has grown to be very large. We use simple seismological relationships to calculate the minimum time that must elapse before such ground motion can be expected at a distance from the earthquake, assuming that the earthquake magnitude is not predictable. Earthquake early warning (EEW) systems are in operation or development for many regions around the world, with the goal of providing enough warning of incoming ground shaking to allow people and automated systems to take protective actions to mitigate losses. However, the question of how much warning time is physically possible for specified levels of ground motion has not been addressed. We consider a zero-latency EEW system to determine possible warning times a user could receive in an ideal case. In this case, the only limitation on warning time is the time required for the earthquake to evolve and the time for strong ground motion to arrive at a user’s location. We find that users who wish to be alerted at lower ground motion thresholds will receive more robust warnings with longer average warning times than users who receive warnings for higher ground motion thresholds. EEW systems have the greatest potential benefit for users willing to take action at relatively low ground motion thresholds, whereas users who set relatively high thresholds for taking action are less likely to receive timely and actionable information.