- © 2009 by the Seismological Society of America
Reliable single-array location is important for nuclear monitoring and other Earth science applications (e.g., Lindquist et al. 2005; Guilbert et al. 2005; Rost and Garnero 2006). Effective use of small-to-medium-aperture (20–25 km or less) seismic arrays requires calibration to account for local and distant propagation effects. Array calibration reduces bias in bearing estimation and event location at regional and teleseismic distances. There is a need for calibration at the International Monitoring System array CMAR (Chiang Mai, Thailand). First mentioned by Koch and Kraundofer (1997) and Bondar et al. (1999), then by Guilbert et al. (2005) and Koper and Fatehi (2008), significant effort was devoted to study-specific, empirical corrections of bias in back-azimuth observed at CMAR. However, even though dipping Mohorovičić discontinuity beneath CMAR was mentioned by Bondar et al., 1999 as a possible explanation of the observed parameter residuals, no further research was devoted to solving this hypothesis. We observe cosine-shape back-azimuth and horizontal-velocity residuals at CMAR and propose a fast and effective formula for first-order corrections for a dipping Mohorovičić (Moho) discontinuity beneath the array, valid for near-regional to teleseismic distances.
Empirical slowness (as defined by Aki and Richards 1980) and back-azimuth corrections are currently used for event location using International Monitoring System arrays and single stations (Bondar et al. 1999). The corrections by Bondar et al. (1999) are estimated on automatically analyzed datasets using frequency-wavenumber analysis (Kvaerna and Dornbos 1986) for stations at teleseismic (>20°) epicentral distance. These corrections are also smoothed over a polar grid. A different approach is to tie the observed systematic residuals to a crustal structure beneath the array (e.g., dipping Moho discontinuity beneath the array, differences in elevation of the instruments, or complicated geology at the site) and to find a physical model that explains these residuals. …