Three-dimensional elastic earthquake modelling based on integrated seismological and InSAR data: theM_w= 7.2 Nuweiba earthquake, gulf of Elat/Aqaba 1995 November

Abstract

SUMMARY The Nuweiba earthquake (1995 November 22; Mw= 7.2), the largest seismic event along the Dead Sea Transform (DST) in at least 160 yr, ruptured 45–50 km along the Aragonese segment of the left‐stepping strike‐slip fault system occupying the gulf of Elat/Aqaba (southern segment of the DST). The rupture initiated in a partly normal, low‐slip first subevent near the southern end of the fault and propagated unilaterally north‐northeastward as a high‐slip, nearly pure sinistral second subevent, which was responsible for over 90 per cent of the total seismic moment. The source mechanism and slip distribution, derived from inversion of teleseismic broad‐band waveforms, are used to construct a 3‐D elastic model of the earthquake based on the boundary elements method, resulting in the full 3‐D displacement and stress fields induced by the earthquake. In the absence of sufficient Global Positioning System data, the only other constraints on the geometry and slip distribution of the rupture are provided by interferometric synthetic aperture radar (InSAR) measurements spanning the coseismic and early post‐seismic period. We calculate simulated interferograms by transforming the calculated surface displacement field into the satellite coordinate system and comparing them with the observed interferograms. The model parameters are then iteratively modified until a best‐fitting model is obtained, providing a refined set of static source parameters for the mainshock. This model is then used to calculate the static Coulomb stress changes induced by the mainshock on the step‐over faults, suggesting that the major (Mw≥ 5) aftershocks in the first eight post‐seismic months were triggered by small changes (<1 bar) in the left‐lateral Coulomb stress, with effective friction coefficient not higher than 0.2. Aftershock distribution and mechanisms indicate that the available Coulomb stress dropped below the frictional strength of the fault but was not complete.

GPS measurements of current crustal movements along the Dead Sea Fault

“…The DSF is subjected to a moderate level of seismicity (Figure 1). The highest recorded seismic activity was in the Gulf of Elat (Aqaba), the location of the 1995 Nuweiba earthquake [ Shamir , 1996; Pinar and Turkelli , 1997; Baer et al , 1999; Klinger et al , 1999; Hofstetter , 2003; Hofstetter et al , 2003; Shamir et al , 2003]. Another area of concentrated activity is the area of the Dead Sea, where the M = 6.5, 1927 Jericho earthquake occurred [ Avni et al , 2002].…”

Variable behavior of the Dead Sea Fault along the southern Arava segment from GPS measurements

“…More recently, in 1927, the Jordan Valley and the northern Dead Sea experienced a moderate earthquake (M L 6.2) (Avni et al, 2002;Shapira et al, 1993). The M W 7.3 Nuweiba earthquake in the Gulf of Aqaba was the largest instrumental earthquake in the region (Klinger et al, 1999;Shamir et al, 2003). Recent moderate earthquakes, such as the 2004 (M $ 5.3) earthquake near the northern end of the Dead Sea basin (Al-Tarazi et al, 2006;Lazar et al, 2006) attest to the present-day activity of the fault.…”

The 1995 November 22,Mw7.2 Gulf of Elat earthquake cycle revisited

“…In the absence of direct slip observations and GPS measurements along and in the vicinity of the rupture area, surface deformation associated with the Nuweiba earthquake could be estimated only by interferometric synthetic aperture radar (InSAR) observations (Baer et al 1999, 2001; Klinger et al 2000). The earthquake slip distribution was previously derived from inversion of teleseismic body‐wave seismograms (Hoffstetter et al 2003) and refined by integration of the seismological data with a 3‐D mechanical model through comparison with the InSAR results (Shamir et al 2003). Comparison between several previously published seismological solutions for the source parameters show significant differences (Shamir 1996; Pinar & Turkelli 1997; Klinger et al 1999; Hoffstetter et al 2003; Shamir et al 2003).…”

“…The earthquake slip distribution was previously derived from inversion of teleseismic body‐wave seismograms (Hoffstetter et al 2003) and refined by integration of the seismological data with a 3‐D mechanical model through comparison with the InSAR results (Shamir et al 2003). Comparison between several previously published seismological solutions for the source parameters show significant differences (Shamir 1996; Pinar & Turkelli 1997; Klinger et al 1999; Hoffstetter et al 2003; Shamir et al 2003). Previous InSAR studies of this earthquake also yielded different fault models (Baer et al 1999; Klinger et al 2000; Shamir et al 2003).…”