UTIG RESEARCH PROJECTS ARCHIVE

Principal Investigators: Cliff Frohlich, Steve Grand, and Jay Pulliam

Funded by: Defense Special Weapons Agency

An essential element for monitoring the Comprehensive Test Ban Treaty (CTBT) concerns how to most effectively use seismic network data to identify and locate small magnitude (M ~ 2.0 to 4.5) seismic events. An overall difficulty is that, except in a few geographic areas, such small events are usually recorded by very few seismic stations. This is both an immediate (1-3 years) and a long-term (5-10 year) problem. The immediate problem is that the in-place methods for locating the source of seismic signals rely heavily on travel-time based methods; these methods are notoriously poor at obtaining reliable focal depths. Moreover, in geographic regions where there is regular activity and many stations, there may be so many signals recorded that it can be difficult to sort out which signals belong to which events, prior to attempting location. The longer-term problem is that travel-time based methods rely on having reliable phase information from several stations; thus, to monitor smaller-magnitude events it would first appear that we must cover the entire Earth with a very dense seismic network. A more economical, and attractive alternative is to develop reliable methods which accurately locate events using waveforms recorded at only a single station.

How, then, can we make better use of the recorded signals to address these problems? An obvious path towards solving these problems is to utilize the entire waveform of recorded signals rather than just phase arrival times. Our past research program [Frohlich and Zhao, 1994; Zhao and Frohlich, 1995; 1996; Zhao et al., 1996] has focused on accomplishing this by comparing observed waveforms with synthetic waveforms generated by various now-standard methods. The preliminary results are quite promising; however, to most effectively utilize coda information it is necessary to determine the near-station crustal structure separately for each seismic station of interest.

In our ongoing work, we are focusing on five research elements:

1. Focal depth - The problem is how to utilize full-waveform information to determine accurate focal depths for seismic events when there are already-existing locations determined by other methods. This research element addresses immediate CTBT concerns and augments existing methods

2. Phase association - The problem is to determine tentative preliminary locations using single-station methods for a significant fraction of signals recorded by a network. This would be a part of routine preliminary signal processing which would aid in the phase association problem. After this preprocessing step, well-enough-recorded events could presumably be located using travel-time-based methods. This research element addresses immediate CTBT concerns, and augments existing methods

3. Single-station location - The problem is how to make best use of auxiliary seismic stations, many situated in regions which are remote from other nearby stations, or in regions not currently considered a risk for CTBT violations. This research element addresses longer-term CTBT concerns

4. Future events in 'quiet' areas - The problem is develop a workable strategy for identifying and locating future events which may occur in areas which are politically at risk for CTBT violations, but where there are no naturally-occurring past seismic events. This research element addresses longer-term CTBT concerns

5. Testing and simulation - The problem is to determine the range of structure over which the software we develop is applicable. This will involve various kinds of modeling or simulation; e.g., to determine under what conditions the use of a flat-layered crustal model is adequate, it would be instructive to use finite difference methods to construct synthetic seismograms for various realistic laterally heterogeneous crustal structures. We could then determine focal depth or locations using single-station methods which assume flat-layered crustal models, and evaluate the error incurred by the simplifying assumptions.

Although this research leans heavily on our past CTBT-related research, a new feature is that, where possible, we will developing exportable demonstration software packages that incorporate the results of our work. At the 18th Annual Seismic Research Symposium in Annapolis in September, 1996, it was clear that there is increasing emphasis on applications-based research. As university researchers we do not confront the continuous data streams that form the basis for CTBT monitoring; however, our intent is that the demonstration software will allow CTBT monitoring personnel to test, modify, and utilize our research results.

A second new feature is that we plan to investigate how best to incorporate array data when it is available. Our most recent research [Zhao and Frohlich, 1996] showed that station-event azimuth was the least-well determined parameter for single-station locations; array determinations of phase slowness vectors would clearly be superior to azimuths determined from single-station transverse energy minima. Moreover, stacking of body-wave signals from array stations should improve focal depth determination as well, since relative time differences of P-wave coda are one of the better indicators of focal depth.

Frohlich, C., and L.-S. Zhao, Location of regional seismic events using single-station broadband data, Proc., 16th Annual Seismic Research Symposium, edited by J. J. Cipar, J. F. Lewkowicz, and J. M. McPhetres, 99-105, 1994.

Zhao, L.-S. and C. Frohlich, Crustal and upper mantle structure beneath seismic stations from modeling teleseismic waveforms, Proc. 16th Annual Seismic Research Symposium, edited by J. J. Cipar, J. F. Lewkowicz, and J. M. McPhetres, 400-406, 1994.

Zhao, L.-S. and C. Frohlich, Determination of near-station crustal structure and the regional seismic event location problem, Proc., 17th Annual Seismic Research Symposium, edited by J. F. Lewkowicz, J. M. McPhetres, and D. T. Reiter, 941-950, 1995.

Zhao, L.-S., and C. Frohlich, Teleseismic body waveforms and receiver structures beneath seismic stations, Geophys. J. Int., 124, 525-540, 1996.

Zhao, L.-S., M. K. Sen, P. L. Stoffa, and C. Frohlich, Application of very fast simulated annealing to the determination of the crustal structure beneath Tibet, Geophys. J. Int., 125, 355-370, 1996.