LAUSR

Abstract The seismicity rate variations prior to the 1993 and 1995 Gulf of Aqaba mainshocks have been investigated using statistical analysis, focusing on the precursory phenomena of the seismic activity progression. A statistical estimation of the seismic quiescence was applied to detect a significant reduction in the rate of seismic activity, using the normal standard deviate Z-test. The catalogue data both from the Egyptian National Seismological Network (ENSN) and the International Seismological Center (ISC) which cover the interval from 1983 to 2003 was used in this analysis. For our computations, only events with completeness magnitude ≥3.5 in the source area of the two Gulf of Aqaba mainshocks were considered. We investigated seismicity rate variations as a function of time and space in terms of the Z value preceding the 1993 and 1995 Gulf of Aqaba mainshocks respectively, using a declustered catalogue. Mapping the spatial changes of the seismicity over the entire period of catalogue manifests clear seismic quiescence anomalies related to the epicenters of the 1993 and 1995 Gulf of Aqaba mainshocks. The long-term average and the corresponding cumulative number of events versus time plot for anomalous areas have been constructed as a way of determining the starting time of the seismic quiescence and its duration for both mainshocks. The plot of the 1993 mainshock shows a decrease in the seismicity rate with a maximum Z value level (Zmax) equal to 5.4 with an observed seismic quiescence starting two years before its occurrence while the 1995 mainshock plot gives a Zmax = 4.5 with 1.3 seismic quiescence duration before it. The estimated periods of quiescence are consistent with the duration of the second type seismic gap derived from the detailed investigations of the spatio-temporal variations of the seismicity pattern which precedes the 1993 and 1995 Gulf of Aqaba main earthquakes. Earthquakes of magnitudes smaller than those which have shown the second type of seismic gap reflect pattern similarities of the activity before both 1993 and 1995 earthquakes. Both mainshocks are preceded by anomalous patterns of activity, which include both swarm type activity and short-term seismic gap, respectively.