Applying coda envelope measurements to local and regional waveforms for stable estimates of magnitude, source spectra and energy

Magnitude estimation forms an integral part in any seismic monitoring endeavor. For monitoring compliance of the Comprehensive Nuclear-Test-Ban Treaty, regional seismic discriminants are often functions of magnitude such as m{sub b}:M{sub 0} high-to-low spectral ratios, and nuclear yield estimation. For small-to-moderate magnitude events that cannot be studied by a large regional or global network of stations, there is a need for stable magnitudes that can be obtained from as few as one station. To date, magnitudes based on coda envelopes are by far the most stable because of the coda's averaging properties. Unlike conventional magnitudes which utilize the direct phases such as P (P{sub n}, P{sub g}) or S (S{sub n}, L{sub g}), or M{sub g}, a coda envelope magnitude is not as sensitive to the undesirable effects of source radiation pattern, 3-D path heterogeneity, and constructive/destructive interference near the recording site. The stability of the coda comes from a time-domain measurement made over a large portion of the seismogram thereby averaging over the scattered wavefield. This approach has been applied to earthquakes in the western United States where it was found that a single-station coda magnitude was approximately equivalent to an average over a 64 station network which used only the direct waves such as L{sub g} (Mayeda and Walter, JGR, 1996). In this paper we describe in detail our calibration procedure starting with a broadband recording, correlation with independent moment estimates, formation of narrowband envelopes, coda envelope fitting with synthetics, and finally the resultant moment-rate spectra. Our procedure accounts for all propagation, site, and S-to-coda transfer function effects. The resultant coda-derived moment-rate spectra are then used to estimate seismic moment (M{sub o}), narrowband magnitudes such as m{sub b} or M{sub L}, and total seismic energy. For the eastern Mediterranean region a preliminary study was completed for earthquakes in the Gulf of Aqaba region using two regional broadband stations, KEG and BGIO. As was found in the western U.S., a significant reduction in magnitude scatter was achieved when using the coda. This procedure provides a means of unbiased, unsaturated magnitude estimation that will be tied to a physical measure of earthquake size (seismic moment), unlike conventional magnitudes such as m{sub b}, M{sub L}, M{sub D}, and M{sub S}. We outline a calibration procedure that can be used in software codes such as SAC on both UNIX and PC platforms. This paper describes the calibration technique and the application to regional stations of the IMS