Bulletin of the Seismological Society of America · 2008 · 14 citations · 22 references
Natural SciencesEarth SciencesGeophysicsApplied GeophysicsEarthquake PhysicsSeismologyEarthquake MagnitudeGeophysical Data ProcessingGeophysical InversionGeophysical Signal ProcessingGround Motion CharacterizationSeismic Data ProcessingDistance CorrectionEarthquake Magnitude EstimationEarthquakesEvent Magnitude EstimationInversionsLocal MagnitudeNonparametric EstimationSeismic Data Inversion
Local-magnitude scales are derived for northeastern Iran (Khorasan province) from waveform data recorded at six stations from 205 local earthquakes, ranging in distance from 10 to 600 km. By averaging the horizontal components in a single measure, we used 1506 zero-to-peak amplitudes from synthetic Wood- Anderson seismograms to determine, in a least-squares sense, the appropriate logA0 attenuation functions, the event local magnitude, and the station corrections. Both a parametric and a nonparametric description of logA0 is considered while performing the inversions. In both cases, the constraint of 1-mm motion recorded at 100 km for M 3:0 earthquakes was used. To evaluate the distance correction curves in determining the local magnitude, ML, in northeastern Iran we applied both linear and trilinear inversions to our datasets. The result of the linear inversion for dis- tance correction is given by: logA0 1:370 0:050logR=100 0:0020 0:0001R 100 3. For trilinear inversion we have applied the Monte Carlo technique. The resulting coefficients evaluated for the area are R1 106 5 km; R2 347 49 km; n1 1:380 0:045; n2 0:597 0:132; n3 0:415 0:236; k 0:0033 0:0003, where n1, n2, and n3 are the coefficients of geometrical spreading for distances from the source to R1, R1 to R2, and beyond R2. k is the coef- ficient of inelastic attenuation. The remarkable agreement between the parametric and nonparametric results confirms that both linear and trilinear attenuation functions that we made for deriving the parametric distance correction are equally reasonable. More- over, inversion of bootstrap replications of our dataset furnished stable solutions. Sta- tion magnitude corrections range between 0:17 and 0.27, suggesting a variable and noticeable effect of station-site properties on recorded amplitudes.
22
Bootstrap Methods: Another Look at the Jackknife
B. Efron · The Annals of Statistics · 1979 · 17.4K citations · Full text