To study the shallowest part of the crust spanning the Dead Sea transform fault (DSF), in the Dead Sea rift valley, tomographic inversion techniques were applied to first-arrival travel times of direct P-waves generated by vibrating trucks. The line was 100 km long, was oriented approximately perpendicular to the DSF (WNW-ESE), and was centered on the DSF. Vibroseis spread length was 18 km. Travel times (250,000) were picked manually for 1700 shots and 180 receivers. The 2-D version of FAST by Zelt and Bartov (1998) was used for the tomographic inversion. Since the inversion is non-unique, extensive testing of model and inversion parameters was carried out. By varying model parameters (block size and shape) and inversion parameters (number of iterations, vertical and horizontal smoothing parameters, choice of reasonable starting models), a stable final velocity model was found. Checkerboard tests indicate that P-wave velocity structure is well resolved down to a depth of several kilometers. The final velocity model shows features that correlate with surface geology and also some buried features. The Dead Sea rift valley is approximately 27 km wide, bounded by inward-facing normal faults at kilometers 40 and 67 on our line. The DSF is located in the center of the rift valley, at kilometer 54, and the east-dipping, normal Zofar fault (ZF) is located on the west side of the valley at kilometer 42. Iso-velocity lines are seen to do the following from west to east across the rift valley. At the western boundary fault, lines $<$2.5 km/s, corresponding to young sediments, step downward $\sim$200-300 m. At the ZF, lines $<$3.5 km/s step downward 1 km along a gently east-dipping ramp. Given this iso-velocity line configuration, the block between the western bounding fault and the ZF may be a slump block. At the DSF, iso-velocity lines $<$2.5 km/s step downward $\sim$100 meters on the edge of the rift valley, but deeper lines, $<$3.5 km/s, step upward $\sim$100 m, suggesting that there are different formation thicknesses on either side of this fault. Deeper lines yet, $>$5 km/s, probably corresponding to ``basement'' (Precambrian rocks), step upward more than 1 km across the DSF. A few km east of the DSF, all iso-velocity lines ($>$5 km/s) step down $\sim$1 km into a buried basin not expressed at the surface. Near the eastern bounding fault, all iso-velocity lines step upward 2-3 km in 2 steps, with one step buried a few km east of the fault and one step at the fault. The buried step looks like a buried slump block along the eastern bounding fault. In summary, a preliminary interpretation of the shallow structure of the rift valley includes slump blocks along the eastern and western bounding normal faults and differing geological structure across the DSF, with a buried basement high on the east side.

Eos Trans. AGU, 82(47), Fall Meet. Suppl., 2001