Spatio-temporal variations of stress drop in and around the asperity of the Mw 6.1, 6 April 2009 L'Aquila earthquake.

Abstract:

We investigate the variability of Brune stress drop in the normal fault system activated by the Mw 6.1 L’Aquila earthquake in the complex tectonic setting of the central Apennine. We re-analyze the dataset used by Calderoni et al. [2013], augmented by additional earthquakes and additional records at closer distance stations. We refine the EGF method used by Calderoni et al. [2013] applying more restrictive criteria in the selection of the EGF events and removing outliers based on statistical criteria. We focus on spatio-temporal variations in the Paganica fault before the mainshock. Using 51 earthquakes (9 foreshocks, the mainshock, and 42 aftershocks), we show that, after the Mw 4.1 largest foreshock of 30 March 2009, the Brune stress drop goes down to the lowest values (0.4 MPa). This largest foreshock was indicated as a marker for the onset of the temporal variations in efficiency of fault-zone guided waves (Calderoni et al., 2015) and other independent seismic parameters such as the b value [Papadopoulos et al., 2010; Sugan et al., 2014], and the P-to-S wave velocity ratio [Di Luccio et al., 2010; Lucente et al., 2010]. The low values of stress drop after the Mw 4.1 foreshock are consistent with the increase of pore pressure invoked by other authors to explain the increase of the Vp/Vs ratio and the decrease of Vs in the damage fault zone. In contrast, immediate foreshocks occurring a few hours before the mainshock very close to its nucleation are characterized by the highest values observed for foreshocks (≈5 MPa). These high stress drop foreshocks are located in the fault patch where a low b value anomaly indicates highly stressed rock before the main shock rupture [Sugan et al., 2014].

These results provide further evidence to previous observations before major earthquakes suggesting that stress drop variations can provide insight into the preparatory phase of impending earthquakes.