Lithospheric-scale transform faults play an important role in the dynamics of global plate motion. Near-surface deformation fields for such faults are relatively well documented by satellite geodesy, strain measurements and earthquake source studies 1,2 , and deeper crustal structure has been imaged by seismic profiling 3.
Relatively little is known, however, about deformation taking place in the subcrustal lithosphere—that is, the width and depth of the region associated with the deformation, the transition between deformed and undeformed lithosphere and the interaction between lithospheric and asthenospheric mantle flow at the plate boundary. Here we present evidence for a narrow, approximately 20-km-wide, subcrustal anisotropic zone of fault-parallel mineral alignment beneath the Dead Sea transform, obtained from an inversion of shear-wave splitting observations along a dense receiver profile. The geometry of this zone and the contrast between distinct anisotropic domains suggest subhorizontal mantle flow within a vertical boundary layer that extends through the entire lithosphere and accommodates the transform motion between the African and Arabian plates within this relatively narrow zone.  At the southern end of the Dead Sea transform (DST), between the Dead Sea and the Red Sea, the Wadi Arava fault is the main active strike-slip fault 4–6 trending approximately N20E. Near the Figure 1 ,Study area and shear-wave splitting parameters. a, Map with topography and the locations of seismic stations for which the shear-wave splitting analysis was performed. The Arava fault (black line) strikes at approximately N20E. The bars indicate measured SKS splitting parameters for the period range of 2–5 s (blue) and 5–7 s (red). 
The orientation corresponds to the polarization direction of the fast shear wave (fast polarization f) and the length is proportional to the delay time dt–1 s. b , Measured shearwave splitting parameters (circles) along the profile (delay times and fast polarization directions) for the two period bands (2–5 and 5–7 s). The parameters are obtained by application of an inverse splitting operator to minimize the energy of the transverse SKS component 17. A measure of error has been derived from the 95% confidence region as determined by the x 2 distribution. For reasons of representation a factor of 0.4 is applied to the error scales used in this figure. The lines represent a smoothed version of the measurements, calculated by averaging the results within a sliding window.

العنوان

Nature 425, 497-501 (2 October 2003)

الناشر

--

بلد الناشر

فلسطين

نوع المنشور

Both (Printed and Online)

المجلد

--

السنة

2003

الصفحات

--