Time variations of aftershock decay parameters of the 2009 April 6 L’Aquila (central Italy) earthquake: evidence of the emergence of a negative exponential regime superimposed to the power law

Abstract

We analysed the time evolution of the decay parameters of the L’Aquila aftershock sequence, neglecting spatial variability. During the first two months after the main shock, the sequence showed quite unusual properties: a particularly slow decay of the aftershock rate that progressively accelerated and a very scarce sensitivity to the occurrence of strong aftershocks. In the first few days, the decrease of the aftershock rate was compatible with an Omori’s process with power-law exponent p ≈ 0.5. The successive increase of the exponent up to about p = 1.2 in the following months can be interpreted as the emergence of a negative exponential regime that has been found to control the decay of other sequences occurred in Italy and California. In fact, two decay models, even including a negative exponential term, reproduce the aftershock rate in the first 60 days significantly better than the Omori’s law according to the Akaike information criteria. In this time interval, the strongest aftershocks do not seem to have produced significant increases of the aftershock rate while a couple of them seem to be preceded, rather than followed, by a slight increase of the rate. Consequently, epidemic models do not perform significantly better than non-epidemic ones for durations shorter than 60–80 days. A slow change of decay parameters seems to have been preceded a clear increase of the rate occurred 80 days after the main shock in correspondence of a relatively strong aftershock in the main fault area and of the activation of a previously silent fault segment in the NW. As a consequence of such reactivation, epidemic models become preferable with respect to non-epidemic ones for longer durations. The L’Aquila main shock productivity is the highest ever observed in Italy since the installation of a modern seismic network in Italy in mid 1980s, as the number of generated aftershock is from three to 10 times higher than for any previous earthquake of similar magnitude.