Crustal structure beneath the south and southeast Iran using receiver function and Rayleigh waves group velocity dispersion

Iran is one of the seismically active areas of the world because it is located in the Alpine-Himalayan orogenic belt, at a 1000-km-wide zone of the compression between the colliding Eurasian and Arabian continents. Studying the crust velocity structure and Moho discontinuity in Iranian plateau is conducive to an understanding of its evolution and the tectonic history of its seismotectonic zones. Nowadays, it is indispensable to acquire sufficient and accurate data from the crust and upper mantle velocity structure or its specification.
To specify the receiver functions with an iterative approach, we made use of a two-year teleseismic data (with epicentral distance 25o-90o) recorded by six seismic stations located in the southeast Zagros (BNDS, NIAN and GENO), Makran (CHBR) and eastern Iran (SZD1 and ZHSF) . In order to delete high frequencies, Gaussian parameter 1.0 was used. So as to augment the signal to noise ratio, RFs were clustered in 10˚ azimuthal and less than 15˚ epicentral distance ranges. Finally, the RFs were stacked.
Receiver functions (RFs) show Earth’s local structure response to P-wave vertical arrival approximately beneath a three-component seismometer; these functions are sensitive to shear-wave velocity impedance. Depth-velocity trade-off in RFs information poses inversion non-uniqueness issues, but a combined inversion of receiver functions and surface wave dispersion increases the uniqueness of the solution over separate inversions, further facilitating the explicit parameterization of layer thickness in the model space, providing more exact constraints as to the crustal structure. Surface wave velocity dispersion depends more on S wave velocity than on P wave velocity, and its dependence on density is slight. In previous studies, it has been shown that it improves the inversions of receiver functions for crustal structures (Julia et al. 2000). Surface wave velocity dispersion provides information as to the absolute seismic shear velocity, yet is relatively insensitive to sharp velocity changes. The group velocities were incorporated into our joint inversion scheme from an independent surface wave tomography study by Rham (2009). Group velocities from regional events, recorded at permanent and broadband stations, were measured for fundamental mode Rayleigh waves within 10–100s period range. The region was parameterized using a uniform, 1×1°, grid of constant slowness cells. The dispersion curve is the result of separate tomographic imaging for each period. Fundamental mode Rayleigh wave group velocities are taken from the corresponding tomographic cell containing the stations. The joint inversion of the two independent data sets was performed considering a proper combination of weighting parameters done by Herrmann and Ammon’s program (2003). Minimizing the standard error between the real and predicted data is the criterion for the desired final model which is close to Earth’s real model.
Models resulting from joint inversion in the south-east Zagros (Hormozgan province) suggest that Moho discontinuity depths beneath BNDS, GENO and NIAN stations are about 54, 54 and 48 kilometers, respectively, while the average depth of Moho discontinuity in the region is about 52±2 kilometers. In the Makran’s seismotectonic state, the resulted models pertaining to single station in the region (CHBR, near the city of Chabahar) show that the average depth of Moho discontinuity in this region is about 28 kilometers and thickness of the sediments is about 10 km, consistent with the shallow subduction of a high-velocity oceanic crust of Arabian plate beneath the southern side of Makran. In the Flysch zone (eastern Iran), the models of the two stations (SZD1, ZHSF) show that the average depth of Moho is about 40±2 kilometers.