European Seismological Commission, 32nd General Assembly, Sept. 6-10, 2010, Montpellier, France
A. Orefice and A. Zollo
The objective of this work is the estimation of the attenuation (P and S quality factors) and source parameters (seismic moment, source radius, seismic energy and stress release) of the 2009 L’Aquila earthquake sequence by using a spectral inversion technique. For this aim an accelerometric waveform archive of 605 earthquakes recorded between 30 March 2009 and 30 April 2009 by DPC-RAN (National Accelerometric Network) (35 stations) and by INGV (29 stations) permanent and temporary seismic networks has been formatted and compiled. The total number of analyzed three-component records is 32275 for events with local magnitude ranging between 2.5 and 5.9.
The S-wave displacement spectra have been inverted using the non-linear Levenberg-Marquardt least-square algorithm for curve fitting, assuming the Boatwright (1980)’ source model.
Preliminary analyses have shown that the frequency dependent Q-model does not provide a significant misfit improvement relative to the constant-Q model, which has finally adopted to correct the spectra for the path attenuation effect. The optimal attenuation model is chosen according to the minimum of the Akaike Information Criterion in the frequency range 0.5-40 Hz.
To get more robust estimations of the attenuation parameter, a two step inversion procedure is applied to the S-wave displacement spectra. In the first step the spectra are inverted for estimating the t*=T/Q parameter for each source-receiver couple, as well as the event-average estimations of low-frequency spectral level Ωo and corner frequency ωc. In the second step the spectral inversion is performed by fixing Ωo and ωc at the previous found average values providing with estimation of parameter t* only for each source and receiver couple.
The frequency-dependent site response function are retrieved by an iterative procedure based on the computation of displacement spectra residuals and stack at each receiver site.
Refined estimation of Ωo and ωc are therefore obtained by inverting the observed S-displacement spectra , corrected for estimated site response and the attenuation function.
The data set enables to investigate the scaling of source size, energy and dynamic/apparent stress release as a function of the seismic moment in a wide magnitude range in an dominant extensional tectonic environment. The radiated seismic energy is computed from the integral of the squared velocity spectrum applying the correction for the finite instrument bandwidth (Ide and Beroza, 2001). This allowed us to obtain estimates of the apparent stress release to be compared to the dynamic stress drop inferred from corner frequencies.