Radial Anisotropy in the Upper Crust Beneath the Tehran Basin and Surrounding Regions

Abstract

Radial anisotropy is characterized by Rayleigh–Love wave discrepancy indicating crustal past and ongoing deformation. For this study, data from 1075 micro-earthquakes were collected based on recordings by 36 stations between 2004 and 2016, and after using data selection criteria, 375 events were used to calculate radial anisotropy in the Tehran region. More than 1908 and 1705 source-station Rayleigh and Love wave group velocity dispersion curves were, respectively, measured in the period band of 0.6–3.0 s. Furthermore, the tomographic inversion method was carried out to obtain group velocity maps for each period individually. Next, a damped least square iterative process was performed using a 3.5 × 3.5 km geographic grid size to calculate both V_SV and V_SH models. Horizontal and vertical spatial extents of the radial anisotropy beneath the Tehran region are revealed as maps of anisotropy as a percentage. Furthermore, the average value of the radial anisotropy as a function of depth indicates three sharp anomalies including: (1) relatively negative within a depth range from subsurface to 1.5 km, (2) relatively positive anomaly within a depth range from 2.0 to 4.0 km, and (3) an approximately isotropic half-space for a depth greater than 4.0 km. In general, the redeposition of former sediments near fault systems, geological and tectonic setting features are correlated with radial anisotropy anomalies at various depths as shown in the horizontal maps. In the radial anisotropy profiles, sedimentary thickness varies from ~ 500 to ~ 1500 m for southward transects, and it is constant for the eastward transect. These profiles clearly indicate the edges of three different tectonic settings.

Appendices

Appendix 1

To investigate the checkerboard resolution test, an initial velocity model was used by adding ± 0.25 km/s to the observed mean value of the Rayleigh and Love group velocities at periods of 1, 2 and 3 s. The checkerboard cell size was also assigned three times the observed data inversion cell size (Trampert and Sneider 1996) and then synthetic travel times were individually calculated for all available source-station pairs in corresponding periods. As shown in Fig. 7, the checkerboard test resolutions were recovered reasonably well except for the southwestern part of the studied area. Unreliable regions, which are not recovered in the checkerboard test resolution in different periods, were masked. Thus, the region with good resolution for horizontal slices, vertical profiles and further interpretation were used. The corner of the input model is indicated by green addition symbols (+) on all recovered result maps.

Fig. 7

Checkerboard test resolutions of Rayleigh (top maps) and Love (bottom maps) waves with periods of 1, 2 and 3 s

Appendix 2

Sensitivities of dispersion data are the most important issues in surface wave studies, especially in considering the discrepancy between fundamental mode of Rayleigh and Love waves based on a linearized inversion method. Short periods of surface waves carry information on relatively shallow structures than the longer periods. In other words, different periods of fundamental modes of Rayleigh and Love waves sample different depths of the velocity structure. Thus, the sensitivity kernel was calculated to obtain the maximum penetration depth and effective depth range using the average of local Rayleigh and Love wave dispersion curves. Figure 8 (1–6) show the normalized depth sensitivity kernel for the periods of 0.6, 1.0, 1.5, 2.0, 2.5 and 3.0 s, individually. Minima and maxima values of penetration depths are approximately 200 m and 6 km corresponding to 0.6 and 3 s, as shown in Fig. 8 (1, 6) respectively. However, surface wave has very small sensitivity around the maximum penetration depth. Thus, the effective depth sensitivity kernels were used for further interpretations. Moreover, differences in sensitivity kernels of surface wave as a function of depth may cause appear artifact anomalies in radial anisotropy result. Thus, to obtain different sensitivity kernels of Rayleigh and Love waves at period of 0.6 and 3.0 s, a similar formula [2 × [Sen_Love − Sen_Rayleigh]/[Sen_Love + Sen_Rayleigh]] was used. In this formula, Sen_Rayleigh and Sen_Love are the sensitivity kernels of Rayleigh and Love waves as a function of depth respectively. Also, if Sen_Rayleigh or Sen_Love closes to zero, the value of formula limits to ± 2%. As shown in Fig. 8 (7), the value of formula for minimum and maximum periods is between ± 1% in this study.

Fig. 8

Normalized sensitivity kernel of Rayleigh (black) and Love (gray) waves with periods of 0.6 (1), 1.0 (2), 1.5 (3), 2.0 (4), 2.5 (5) and 3.0 (6) s. The effective depth sensitivity ranges of these periods are 0.0–0.7, 0.5–1.2, 0.9–2.5, 1.2–3.2, 1.5–4.0 and 1.8–5.0 km, respectively. The difference between the Rayleigh and Love wave sensitivity kernels, which is calculated using [2 × [Sen_Love − Sen_Rayleigh]/[Sen_Love + Sen_Rayleigh]] formula, are depicted by gray (0.6 s) and black (3 s) traces in (7)