A comparison of moment magnitude estimates for the European-Mediterranean and Italian regions

作者：Gasperini Paolo^*; Lolli Barbara; Vannucci Gianfranco; Boschi Enzo

来源：Geophysical Journal International, 2012, 190(3): 1733-1745.

DOI：10.1111/j.1365-246X.2012.05575.x

• Earthquake source observations
• Statistical seismology

With the goal of constructing a homogeneous data set of moment magnitudes (Mw) to be used for seismic hazard assessment, we compared Mw estimates from moment tensor catalogues available online. We found an apparent scaling disagreement between Mw estimates from the National Earthquake Information Center (NEIC) of the US Geological Survey and from the Global Centroid Moment Tensor (GCMT) project. We suspect that this is the effect of an underestimation of Mw %26gt; 7.0 (M0 %26gt; 4.0 x 1019 Nm) computed by NEIC owing to the limitations of their computational approach. We also found an apparent scaling disagreement between GCMT and two regional moment tensor catalogues provided by the Eidgenossische Technische Hochschule Zurich (ETHZ) and by the EuropeanMediterranean Regional Centroid Moment Tensor (RCMT) project of the Italian Istituto Nazionale di Geofisica e Vulcanologia (INGV). This is probably the effect of the overestimation of Mw %26lt; 5.5 (M0 %26lt; 2.2 x 1017 Nm), up to year 2002, and of Mw %26lt; 5.0 (M0 %26lt; 4.0 x 1016 Nm), since year 2003, owing to the physical limitations of the standard CMT inversion method used by GCMT for the earthquakes of relatively low magnitude. If the discrepant data are excluded from the comparisons, the scaling disagreements become insignificant in all cases. We observed instead small absolute offsets (=0.1 units) for NEIC and ETHZ catalogues with respect to GCMT whereas there is an almost perfect correspondence between RCMT and GCMT. Finally, we found a clear underestimation of about 0.2 units of Mw magnitudes computed at the INGV using the time-domain moment tensor (TDMT) method with respect to those reported by GCMT and RCMT. According to our results, we suggest appropriate offset corrections to be applied to Mw estimates from NEIC, ETHZ and TDMT catalogues before merging their data with GCMT and RCMT catalogues. We suggest as well to discard the probably discrepant data from NEIC and GCMT if other Mw estimates from different sources are available for the same earthquakes. We also estimate approximately the average uncertainty of individual Mw estimates to be about 0.07 magnitude units for the GCMT, NEIC, RCMT and ETHZ catalogues and about 0.13 for the TDMT catalogue.

• 出版日期2012-9