Some features of aftershock patterns in Greece

A quantitative study of 12 aftershock sequences, which occurred in Greece and the surrounding regions from 1971 to 1980, has been made. The aim of the study was to investigate the temporal pattern of each aftershock sequence and to examine the possibility of 'predicting' the largest aftershock. The magnitudes of the main shocks were M(L) greater than or equal to 5.0. The minimum and maximum numbers of aftershocks were 25 and 191, respectively. The temporal analysis was performed using the modified Omori formula and Akaike information criterion. First, we examine the Omori law to see if holds for the whole aftershock sequence. Second, we examine the same data set in order to find out if there is any possibility of 'distinguishing' two aftershock sequences as follows: (1) the first must include all the shocks since the occurrence of the main event until the occurrence of the largest aftershock and (2) the second must include all events following the occurrence of the largest aftershock until the end of the aftershock activity. In 10 cases out of 12, the model of two different aftershock sequences provides the best fit to the data. The aftershock activity prior to the large aftershocks was then examined. Although the main shocks occurred in different seismogenic regions, an almost identical behaviour of aftershock activity was observed. The activity shows an appreciable decrease from the level expected from the modified Omori formula. Also, in some cases, the aftershock activity recovers to a normal level or increases beyond it prior to the occurrence of the main aftershock. These results seem to be independent of the magnitude of the 'large' aftershock. If the aftershock activity is monitored in real time, this observed temporal pattern may be useful in predicting a large aftershock.