HR: 1340h
AN: S33A-0305    [Abstracts]
TI: Earthquake Depth Determination Using the F Trace and Associated Probability
AU: * Heyburn, R
EM: rossh@earth.ox.ac.uk
AF: Oxford University, Department of Earth Sciences, Parks Road, Oxford, OX1 3PR United Kingdom
AU: * Heyburn, R
EM: rossh@earth.ox.ac.uk
AF: AWE Blacknest, Brimpton Common, Reading, RG7 4RS United Kingdom
AU: Bowers, D
EM: bowers@blacknest.gov.uk
AF: AWE Blacknest, Brimpton Common, Reading, RG7 4RS United Kingdom
AU: Woodhouse, J
EM: john.woodhouse@earth.ox.ac.uk
AF: Oxford University, Department of Earth Sciences, Parks Road, Oxford, OX1 3PR United Kingdom
AB: Determining the depths of seismic disturbances is of importance when monitoring compliance with the Comprehensive Nuclear-Test-Ban Treaty, in plate tectonic studies, and in seismic hazard assessments. We have developed a semi-automated method that uses the output from the F-detector (a processing technique that under certain conditions optimally detects correlated signals crossing an array) as a tool for identifying candidate depth phases (pP and sP), and hence estimating source depth. Analysis of F-traces (recorded from earthquakes located in the Hindu-Kush and Lop Nor regions) shows that the distribution of an F-trace which highlights depth phases has a significant positive skewness. The skewness of the calculated F distribution can therefore be used by analysts to determine whether or not the seismogram and F-detector output contain interpretable arrivals after direct P, and to determine the optimal passband for filtering. For medium aperture seismic arrays (10-20 km), the F trace can be interpreted approximately in terms of a probability. For a given signal-to-noise ratio and bandwidth this probability can be used to help interpret signals and identify candidate depth phases. To develop a semi-automated method for depth determination, the values of the probability trace at the predicted times for pP and sP for a given depth are weighted and summed over several high quality medium-aperture array stations. A clear maximum in the trace sum function with depth is interpreted as the probable hypocenter depth. To determine confidence in a depth estimate, we estimate whether it is possible that the phases interpreted as pP and sP are in fact S-to-P conversions from some structure in the below source region. This is done by comparison of observed pP/P and sP/P amplitude ratios with S-to-P/P amplitude ratios calculated from theory. Results for a set of shallow earthquakes in the Lop Nor region indicate that this method of depth estimation is effective for most earthquakes with magnitude greater than about 4.0{m}_b (National Earthquake Information Center).
DE: 7219 Seismic monitoring and test-ban treaty verification
SC: Seismology [S]
MN: Fall Meeting 2005