Earthquake parameters of historical earthquakes in Europe (Invited)
Abstract
The assessment of earthquake parameters of historical earthquakes is a key issue for understanding the seismic potential and evaluation the seismic hazard of a region. The processing of historical data is a complicated affair, still performed according to subjective, non repeatable procedures. In the last ten years varied methods using macroseismic datapoints have been developed: Bakun & Wentworth, or BW (Bakun and Wentworth, 1997), Boxer (Gasperini et al., 1999), MEEP (Musson and Jimenez, 2008). These methods allow the assessment to be performed by means of computer codes and, therefore, results are independent from the operator to some extent. With the aim of re-assessing earthquake parameters for the main historical events (M>5) of the entire European region, the NA4 module (Distributed Archive of Historical Earthquake Data) of the European project NERIES started to calibrate the three methods, in a homogeneous way, in five areas: Aegean, Iberian, Italian, Great Britain and Switzerland. A dataset of about fifteen earthquakes of the 20th century was compiled according to homogeneous procedures for each area. Earthquakes were selected among those with both macroseismic datapoints and reliable instrumental moment magnitude (Mw) and epicentre available, in such a way to cover the largest possible magnitude range and geographical distribution of earthquakes. Boxer and MEEP come with codes which can provide calibrated coefficients from the input dataset. The BW method requires an intensity attenuation relation as function of the Mw and hypocentral distance. Such relations were obtained for each area calibrating the coefficients of a log-linear function with regression procedures from the same dataset used for calibrating the other methods. Recently published relations were also considered. The obtained, calibrated coefficients were then validated determining the parameters of another ten events. For Boxer and MEEP the validation exercise was devoted to survey whether the obtained coefficients would lead to realistic results. For BW at least two alternatives were available in all areas; the scope was, therefore, to select the “best performing” one. The three methods and the relevant calibrations have been then used for assessing the parameters of the historical earthquakes (1000-1750) in the five regions. The results are encouraging although some problems are still to be explored. The epicenter determinations appear rather successful and not so much regionally dependent, although to date no method is completely successful in distinguishing coastal and offshore events. The magnitude determinations, which come with uncertainty estimates, strongly depend on the regional calibration; in some cases they show large divergences between the methods, and also with respect to the values reported by current national catalogues, often obtained through unreported procedures. We believe that such differences can be reduced, but they do represent an estimate of the epistemic uncertainty of the process.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2009
- Bibcode:
- 2009AGUFM.S34B..07S
- Keywords:
-
- 7212 SEISMOLOGY / Earthquake ground motions and engineering seismology;
- 7215 SEISMOLOGY / Earthquake source observations;
- 7230 SEISMOLOGY / Seismicity and tectonics