The Effect of Crustal Structure on Station Magnitude Anomalies (Magnitude Bias).

Some analyses of P- and Rayleigh wave amplitudes observed at North American stations

“…When station effects are assumed to be zero and dropped from the analysis (analysis method 2 , Table 1) then the variance of a single observation for the SP data is significantly larger (at 0 100 II. The predicted crustal effects used are the SP50 and LP50 effects given by Der et al (1977); that is the effect the crust is predicted to have on aP-wave from a 50 kt explosion in granite.…”

“…Note that stations HY, J P and SI have to be dropped from these analyses because Der et al (1977) d o not predict the crustal effects for these stations.…”

“…Stations with material of low P-wave speed and low density at the free surface, for example, would be expected to record larger amplitudes than stations with material of high wave speed and density. Der, McElfresh & Mrazek (1977) have made a detaded study of the effects of crustal structure on the amplitude of P-waves at the LRSM stations and conclude that, although crustal structure does have an effect on the observed amplitude, by far the largest contribution t o the SP station effects observed by Booth et al (1974) is due to variations in Q,. Here we describe a similar analysis made using the data of Booth et al (1974) to try and assess the contribution to station effects of variations in Q, predicted from a, with that due to crustal structure.…”

“…Booth et al (1974) suggest that the Lp effects depend on crustal structure and demonstrate that large positive station effects can be accounted for by assuming that the stations showing these effects are underlain by large thicknesses of sediment. Der et al (1977) have now published estimates of the LP Station effects computed from the information available on crustal structure at each station, So we examine how far the LP station effects obtained by Booth et al (1974) can in fact be @counted for by variations in crustal structure between stations.…”

Determination of Event Magnitudes with Correlated Data and Censoring: A Maximum Likelihood Approach

“…Both of these effects should produce regional correlation between stations such as those regional patterns exhibited across western and eastern North America associated with the tectonically active and tectonically stable regions of the continent. Also, crustal effects such as thick sedimentary basins are known to introduce station effects that amplify-deamplify short-period records (Der et al 1977;1981;Blandford & Shumway 1982). In order to determine these two effects it is necessary to average the station magnitude residuals over a range of azimuths and incidence angles.…”