Application of Autoregressive processing to the analysis of seismograms

Abstract

The application of Autoregressive processing to the analysis of seismograms has been of value to seismologists for several decades. Work in the 1960s and 1970s focused primarily on using Autoregressive (AR) filtering and other error predictive filters for improving the signal to noise ratio of small seismic signals. In the eighties and nineties most interest has been in the application of AR filtering to onset time estimation. Onset time pickers which utilise AR filtering have proved to be very effective for a wide range of seismic signals including, local and teleseismic events. A key advantage of them over other methods is that they are robust even for phases with a small signal to noise ratio. The accuracy of automatic AR picks compare well with those of experienced seismic analysts. *** DIRECT SUPPORT *** A04BD016 00003