Stephen C. Myers

Improving Sparse Network Seismic Location with Bayesian Kriging and Teleseismically Constrained Calibration Events

Monitoring the Comprehensive Nuclear-Test-Ban Treaty will require improved seismic location capability for small-magnitude events. The International Monitoring System (IMS) is well suited to locate events that are large enough to be recorded at teleseismic distances. However, small events are likely to be recorded on a sparse subset of IMS stations at regional- to upper-mantle distances (less than 30°), and sparse-network locations can be strongly effected by travel-time errors that result from path-specific velocity model inaccuracies. In an effort to improve sparse network location capability, we outline a procedure that applies empirical corrections to travel times determined with an appropriate velocity model. More specifically, Bayesian kriging and calibration events (constrained with a global network) are used to estimate epicenter-specific travel-time corrections. For a test (sparse) network of stations, we calculate travel-time residuals for the calibration events relative to the ak135 velocity model. Travel-time residuals are assigned to the respective calibration epicenter, forming a set of spatially varying travel-time correction points. The spatial set of correction points is declustered to reduce the dimension of the observations with minimal reduction in accuracy of the travel-time corrections. We then use the declustered set of calibration points and Bayesian kriging to form continuous travel-time correction surfaces for each station of the test network. The effectiveness of travel-time correction surfaces is evaluated by locating, with and without corrections, a subset of the 1991 Racha earthquake sequence (Caucasus Mountains), for which we have accurate locations that were independently determined with a dense local network. When no travel-time correction is applied, the mean horizontal distance between the local and test network locations is 42 km, and there is a distinct bias in sparse-network locations toward the north-northwest. The mean difference between local and sparse network locations is cut to 13 km when corrections are applied, and the bias in location is significantly reduced. When calibration events in the Racha vicinity are not used to make the correction surfaces, there is still a significant improvement in location, with mean mislocations of 15 km. When corrections are not applied, only one of the locally determined locations lies within the associated 90% coverage ellipse determined with the test (sparse) network. However, by using travel-time corrections and estimates of model uncertainty determined using kriging, representative error ellipses are obtained. This study demonstrates that kriging correction surfaces based on global-network-constrained calibration events can improve the ability to accurately locate lower magnitude events while providing representative coverage ellipses.

Regional Travel-Time Uncertainty and Seismic Location Improvement Using a Three-Dimensional a priori Velocity Model

We demonstrate our ability to improve regional travel-time prediction and seismic event location accuracy using an a priori 3D velocity model of Western Eurasia and North Africa (WENA1.0). Travel-time residuals are assessed relative to the iasp91 model for approximately 6000 Pg, Pn , and P arrivals, from seismic events having 2 σ epicenter accuracy between 1 km and 25 km (ground truth 1 [GT1] and GT25, respectively), recorded at 39 stations throughout the model region. Ray paths range in length between 0° and 40° (local, regional, and near teleseismic) providing depth sounding that spans the crust and upper mantle. The dataset also provides representative geographic sampling across Eurasia and North Africa including aseismic areas. The WENA1.0 model markedly improves travel-time predictions for most stations with an average variance reduction of 29% for all ray paths from the GT25 events; when we consider GT5 and better events alone, the variance reduction is 49%. For location tests we use 196 geographically distributed GT5 and better events. In 134 cases (68% of the events), locations are improved, and average mislocation is reduced from 24.9 km to 17.7 km. We develop a travel-time uncertainty model that is used to calculate location coverage ellipses. The coverage ellipses for WENA1.0 are validated to be representative of epicenter error and are smaller than those for iasp91 by 37%. We conclude that a priori models are directly applicable where data coverage limits tomographic and empirical approaches, and the development of the uncertainty model enables merging of a priori and data-driven approaches using Bayesian techniques. Online material: Correction surfaces and histograms of travel-time residuals for 40 stations.

Evidence for long-lived subduction of an ancient tectonic plate beneath the southern Indian Ocean

In this study, ancient subducted tectonic plates have been observed in past seismic images of the mantle beneath North America and Eurasia, and it is likely that other ancient slab structures have remained largely hidden, particularly in the seismic-data-limited regions beneath the vast oceans in the Southern Hemisphere. Here we present a new global tomographic image, which shows a slab-like structure beneath the southern Indian Ocean with coherency from the upper mantle to the core-mantle boundary region—a feature that has never been identified. We postulate that the structure is an ancient tectonic plate that sank into the mantle along an extensive intraoceanic subduction zone that migrated southwestward across the ancient Tethys Ocean in the Mesozoic Era. Slab material still trapped in the transition zone is positioned near the edge of East Gondwana at 140 Ma suggesting that subduction terminated near the margin of the ancient continent prior to breakup and subsequent dispersal of its subcontinents.

Nonstationary Bayesian kriging: a predictive technique to generate spatial corrections for seismic detection, location and identification

Abstract Seismic characterization works to improve the detection, location, and identification of seismic events by correcting for inaccuracies in geophysical models. These inaccuracies are caused by inherent averaging in the model, and, as a result, exact data values cannot be directly recovered at a point in the model. Seismic characterization involves cataloging reference events so that inaccuracies in the model can be mapped at these points and true data values can be retained through a correction. Application of these corrections to a new event requires the accurate prediction of the correction value at a point that is near but not necessarily coincident with the reference events. Given that these reference events can be sparsely distributed geographically, both interpolation and extrapolation of corrections to the new point are required. In this study, we develop a closed-form representation of Bayesian kriging (linear prediction) that incorporates variable spatial damping. The result is a robust nonstationary algorithm for spatially interpolating geophysical corrections. This algorithm extends local trends when data coverage is good and allows for damping (blending) to an a priori background mean when data coverage is poor. Benchmark tests show that the technique gives reliable predictions of the correction value along with an appropriate uncertainty estimate. Tests with travel-time residual data demonstrate that combining variable damping with an azimuthal coverage criterion reduces the large errors that occur with more classical linear prediction techniques, especially when values are extrapolated in poor coverage regions. In the travel-time correction case, this technique generates both seismic corrections along with uncertainties and can properly incorporate model error in the final location estimate. Results favor the applicability of this nonstationary algorithm to other types of seismic corrections such as amplitude and attenuation measures.

Reply to “Comment on ‘Improving Sparse Network Seismic Location with Bayesian Kriging and Teleseismically Constrained Calibration Events,’ by Stephen C. Myers and Craig A. Schultz,” by A. Douglas

We thank Alan Douglas (Douglas, 2004) for his thoughtful comments on Myers and Schultz (2000). In Myers and Schultz (2000) we improve seismic location accuracy and uncertainty estimates for a sparse, regional network by improving travel-time predictions and better characterizing residual uncertainties. Station-specific, travel-time corrections and associated uncertainty estimates are calculated using the Bayesian kriging methodology of Schultz et al. (1998). The Bayesian kriging methodology is an empirical approach that exploits the spatial correlation of travel-time residuals to interpolate an empirical (calibration) set of travel-time residuals into a minimum-variance, geographically continuous surface of travel-time corrections and an associated uncertainty surface. In Myers and Schultz (2000) we test the Bayesian kriging approach by relocating events in the 1991 Racha, Georgia earthquake sequence (∼ 42°N, 43°E). We compare uncalibrated locations (based on P -wave travel times for the ak135 model; Kennett et al. , 1995) and calibrated locations (calibrated travel-times and uncertainties) to benchmark locations that are constrained with a dense, local-distance network. The calibration data set consists of event locations that are constrained by a teleseismic network. One goal of the study was to assess the utility of teleseismically constrained events for calibration. We find that better travel-time prediction for regional paths can be achieved by using the teleseismically constrained calibration events, and the improved travel-time prediction improves regional-network location accuracy. A major advantage of Bayesian kriging over many other interpolation techniques is the propagation of uncertainty from the calibration data set into a travel-time-prediction uncertainty surface. Sources of data set uncertainty include calibration-event locations, phase observations for calibration events, and calibration-event spacing. We find that coverage ellipses (Evernden, 1969) are representative of observed epicenter accuracy (i.e., confidence level of the coverage ellipse is indicative of the number of occurrences of the benchmark location within the ellipse) when the …

Assessment of Regional-Distance Location Calibration Using a Multiple-Event Location Algorithm

Abstract We test the use of a multiple-event seismic location method to improve epicenter accuracy estimates. Regional arrival-time observations of 74 Nevada Test Site explosions with known locations comprise the test data set. We investigate epicenter accuracy as a function of the number of events in the multiple-event system that are constrained at the known hypocenter (calibration), the effect of distance between calibration and unconstrained events, and the use of velocity models with varying travel-time prediction accuracy. Further, we test the utility of using a posteriori travel-time residuals to assess location and travel-time prediction accuracy. We find that constraining one event at the known hypocenter reduces epicenter error for all other events by 58% on average compared to locations produced without constraining events. The incremental improvement in epicenter accuracy rapidly diminishes as more hypocenters are constrained, and incremental location improvement is minimal when the number of constrained hypocenters exceeds 10. Events closest to a constrained event exhibit small location bias. Distinct epicenter bias occurs when the distance between the calibration event and the relocated event is greater than a few tens of kilometers. Last, we confirm that metrics based on a posteriori travel-time residuals are poor indicators of both epicenter accuracy and velocity model-based travel-time prediction accuracy.

Erratum to Deep Earthquakes beneath the Northern Caucasus: Evidence of Active or Recent Subduction in Western AsiaErratum

An intermediate-depth earthquake is confirmed at a depth of 158 4 km under the northern foothills of the Greater Caucasus. Separate methods were used to confirm the depth: data from local and regional networks, teleseismic depth phases, and examination of waveforms. Additional examination of global catalogs suggests the presence of a (perhaps remnant) northeast-dipping subduction zone under the Greater Caucasus. The most likely explanation appears to be subduction of oceanic crust with the interface at the northern edge of the Kura Basin. Events at depths of 30 – 50 km in the Kura Basin may be related to underthrusting by the South Caspian basin rather than subduction in the Greater Caucasus.