Validation of Regional and Teleseismic Travel-Time Models by Relocating Ground-Truth Events

Abstract

SUMMARYRecently developed three-dimensional global seismic-velocity models have demonstrated location improvements through independent regional and teleseismic travel-time calibration. Concurrently, a large set of high quality ground truth (GT) events with location accuracies 10 km or better (GT0-GT10) has been collected for Europe, the Medite rranean, North Africa, the Middle East, and Western Eurasia. In this study, we validate event location improvements using this new data set by applying the regional and teleseismic modelbased travel-time calibrations (independently and jointly) to demons trate that significant improvements can be achieved using 3D global models for locating small events with sparse network data. Besides relocating events using all station arrivals, a subset of the GT events was also relocated using controlled station geom etries generated from a "constrained bootstrapping" technique. The advantages of this approach include: (1) generating simulated sparse networks (Simulated Sparse Network Bulletin or SSNB), (2) increasing the statistical power of the tests, (3) reducing the effect of correlated errors to ensure valid 90% error ellipse coverage statistics, and (4) measuring location bias due to un-modeled three-dim ensional (3D) Earth structures.With respect to the GT events, we compared event relocations, with and without t raveltime calibrations, considering statistics of mislocation, error ellipse area, 90% coverage, origin time bias, origin time errors, and misfit. Relocations of more than 1000 GT0-GT10 reference events show significant reductions in location bias and uncertainty. Pn and/or P calibration reduces mislocation for between 60% and 70% of the events. Joint regional Pn and teleseismic P travel-time calibration provided the largest location improvements, and approximately achieved GT5 accuracy levels. Due to correlated 3 errors, calibrated event locations using large numbers of stations have deficient 90% error ellipse coverage. However, the coverages derived from the model errors are appropriate for sparse regional and teleseismic networks. This validation effort demonstrates that the global model-based travel-time calibrations of Pn and teleseismic P travel-time reduce both location bias and uncertainty over wide areas.

Multiscale Modeling and Process Optimization for Engineered Microstructural Complexity

“…These path effects are also being referred to as source-specific station corrections or SSSCs (e.g. Morozov et al, 2005;Murphy et al, 2005;Ritzwoller et al, 2003;Yang et al, 2001Yang et al, , 2004.…”

“…Both models have been shown to improve seismic locations and achieve significant variance reduction in travel-time predictions (e.g. Bondár et al, 2004b;Ritzwoller et al, 2003;Yang et al, 2004). Figure 4 shows the GT event clusters we use throughout this study.…”

“…Recognizing the problem of location bias due to inadequate velocity models, most recent location calibration efforts focused on developing better travel-time predictions, either model based or empirical (e.g. Flanagan et al, 2007;Morozov et al, 2005;Murphy et al, 2005;Myers and Schultz, 2000;Reiter et al, 2005;Ritzwoller et al, 2003;Yang et al, 2004).…”

“…This assumption is violated when the separation between stations is less than the scale length of local velocity heterogeneities. To illustrate the effect of the correlated error structure on seismic event location, we performed a constrained bootstrapping (Yang et al, 2004) experiment on the 7 October 1994 Lop Nor, China nuclear explosion. The explosion is considered GT1 (Fisk, 2002) and recorded by some 400 stations at teleseismic distances.…”

The Prospect of Using Three-Dimensional Earth Models to Improve Nuclear Explosion Monitoring and Ground-motion Hazard Assessment

“…Although a number of studies have demonstrated the utility of using 3-dimensional models for determining seismic locations (Antolik et al, 2003;Yang et al, 2004;Flanagan et al, 2007), the broader application of 3-dimensional models in operational monitoring systems has yet to occur. Changing operational practice is a serious matter for any monitoring agency, because operational changes are costly, and there are risks to operational robustness.…”