Abstract: Earthquake events have discrete properties in space and time during their occurrences and migration. When the seismogenic system is in tranquil(or normal) state, the probability distribution for time sequence of variance X ( i.e. seismicity in this paper) will show single-peak distribution such as Poisson distribution; the time intervals for two events(or waiting time) Δt can be described by negative exponent function. While the system enters into non-tranquil (or abnormal) state, the probability distribution for time sequence of X shows two-peaks or multiple-peak distribution; their waiting time can be described by power function. The distribution properties of X can be discriminated by calculating 1 to 4 order of moment estimators, which will further help to identify the current state of the system. According to this approach, we try to investigate the evolutions of time and space structure in seismogenic system and identify its state. We also apply these methods to the prediction practice of strong aftershocks of the 2008 Wenchuan M8.0 earthquake sequence.

Key words: Seismic Activity, Statistical Group Properties, Earthquake prediction, Wenchuan earthquake