1 Introduction

The continental collision between the Arabian and Eurasian plates results in a complex deformation within Iran, which is controlled by the continuing convergence of the Arabian plate toward the Eurasian plate. The present-day Iranian plate indicates different tectonic processes including orogeny (Zagros, Alborz, and Kopeh-Dagh) and subduction of the oceanic lithosphere (Makran) (Fig. 1). The collision between the Arabian and Eurasian plates started in the early Miocene, after the Neotethys Ocean was subducted beneath Eurasia (Jackson and McKenzie, 1984; Dewey et al., 1986; Beghoul and Barazangi, 1989; Boulin, 1991). The closure of the Neotethys Ocean resulted in the emplacement of ophiolites along the Zagros suture zone and the onset of deformation in the Zagros fold and thrust belt (Stoneley, 1981; Richards et al., 2006). The collision process trapped the central Iranian block between the Arabian plate in the south and the Turan shield in the north and led to intra-continental shortening, formation of the Iranian plateau, widespread deformation, and mountain building (Bird, 1978). It is assumed that most deformation is accommodated not only in the major mountain belts (Zagros and Alborz) with large reverse faults, but also along large strike-slip faults that surround the blocks (the Central Iranian block, the Lut block, and the southern Caspian Sea, see Fig. 1) (Jackson and McKenzie 1984; Berberian and Yeats 1999). Distribution of seismicity and the local topography occur at the edges of the deformation zones, which are well defined by previous studies (e.g. Jackson and McKenzie, 1984).

Fig. 1
figure 1

Different tectonic units of the Iranian plate. UL Urumieh Lake, SSZ Sanandaj-Sirjan Zone, ZFTB Zagros Fold and Thrust Belt, MZTF Main Zagros Thrust Fault, HZF High Zagros Fault, MFF Main Front Fault, UDMA Urumieh–Dokhtar Magmatic Arc, Lur. A. Lurestan Arc; Fars A. Fars Arc. Faults are from Jiménez-Munt et al. (2012)

Full size image

The depth of Moho is an important parameter to characterize the structure of the crust. Furthermore, it provides significant constraints on tectonic evolution of the region. The complex tectonic structure of Iran provides an ideal study area for investigation of crustal thickness. A large number of studies have focused on the crustal structure and have shown the topography of the Moho discontinuity beneath different parts of Iran (e.g. Asudeh, 1982; Javan Doloei and Roberts, 2003; Hatzfeld et al., 2003; Sodoudi et al., 2009; Paul et al., 2006, 2010; Taghizadeh-Farahmand et al., 2010, 2013; Radjaee et al., 2010; Abbassi et al., 2010; Afsari et al., 2011, Mohammadi et al. 2013a). However, they were mostly limited to the narrow profiles and could not cover the whole Iran.

Dehgani and Makris (1984) constructed the first Moho depth map of Iran from Bouguer anomaly modeling and seismic data. This map has been often used as the only reference Moho depth map for Iran. Currently, due to the growing number of national Iranian seismological networks and large amounts of available data, a more accurate Moho depth map is required.

The main goal of this paper is to resolve the Moho discontinuity and its lateral depth variations beneath different tectonic zones of Iran using all Iranian broadband and short-period stations for the first time. We calculate the P receiver functions beneath each station and apply the Zhu and Kanamori method (Zhu and Kanamori, 2000) (Z&K) as well as 1-D forward modeling to map the topography of the Moho boundary with a higher resolution (error of  ±2 km) than that previously presented.

2 Data and Analysis

The data used for this study were recorded by the Iranian Telemetry Seismic Networks (ITSN), which consists of 11 seismic networks with 53 permanent short-period seismic stations. In addition, we used the data of 24 broadband stations (Fig. 2). The short-period networks are operated by the Iranian Seismological Center (ISC). They are equipped with SS-1 seismometers with a natural frequency of 1 Hz made by Nanometrics and are connected to the central recording station via a telemetric system. The broadband stations operated by the International Institute of Earthquake Engineering and Seismology (IIEES) are equipped with three component Güralp (CMG-3TD) sensors. Names of the networks and stations and their geographical coordinates are listed in Table 1. Teleseismic data, which were recorded between 1995 and 2011, have been used in this study. More than 1,400 teleseismic events (Fig. 3) with magnitudes greater than 5.5 (Mb) at epicentral distances between 30° and 95° have been used for the P Receiver Function (PRF) analysis. The methodology for PRF analysis used in this paper is the same as described by Yuan et al. (1997).

Fig. 2
figure 2

Location map of the seismological stations used in this study. Stations are shown with red (short period) and blue (broadband) triangles. Active faults are shown with brown lines (Hessami et al., 2003)

Full size image
Table 1 Specification of the seismic stations, Ps conversion times (s), Moho depths (km) and number of PRFs
Full size table
Fig. 3
figure 3

Distribution of teleseismic events recorded by the national permanent stations of Iran between 1995 and 2011 and used to calculate P receiver functions. The green star represents the approximate position of Iran. The black solid circles mark the 30° and 95° epicentral distances, respectively

Full size image

Calculation of PRFs is performed in three different steps including removal of the instrument response, coordinate rotation into the local LQT ray-based coordinate system (as described by Vinnik, 1977), and deconvolution in time domain (as described by Kind et al., 1995), which results in having the converted P-to-S phases on the Q component. A reference slowness of 6.4 s/° is considered for the moveout correction. PRFs are then stacked and filtered with a low-pass filter of 2 s (Butterworth, three poles).

Another step often employed in receiver function analysis is inversion, which can find the most suitable average shear wave velocity and crustal thickness beneath each seismic station. However, unreliable results will occur if no clear converted phases or multiples exist in the time domain receiver functions. In such cases, seismic noise may be transformed into a velocity-depth model. Therefore, we prefer forward modeling (inversion with additional parameters) of the receiver functions (e.g. Kumar et al., 2007). This procedure is more realistic than a blind automatic inversion without phase identification. For this reason, we first identified the Moho conversion in the data, which is often the largest phase on the Q component. Other phases, which are frequently detected in time domain receiver functions are conversions from the bottom of sedimentary layers and crustal multiples. We picked the arrival times of all these phases. A grid search was then performed to find a crustal model, which fits the waveforms reasonably well. For each tectonic zone, we used appropriate velocity models inferred by previous studies (Table 2).

Table 2 Different velocity model that are used as reference in different tectonic zone in this study
Full size table

3 P Receiver Function Observations

Teleseismic events with a relatively high signal-to-noise ratio (larger than 4) have been selected for most of the stations. This criterion significantly reduced the number of PRFs beneath each station. For stations with a relatively high noise level, a large number of data must be deleted (e.g. BJRD). Figure 4 shows individual and stacked PRFs for some short-period (MHD and VIS) and broadband stations located in different tectonic zones. PRFs are sorted by increasing back azimuth. The stacked PRFs (at the top of the Fig. 4) reliably show a clear Ps conversion from the Moho ranging between 4.8 and 6.4 s. This conversion can be clearly followed in the individual traces. Other phases detected in the receiver functions are related to the conversions from the bottom of sedimentary layers and crustal multiples. The minimum arrival time of the Moho converted phase (3.7–3.8 s) is observed beneath the stations CHBR and RMKL located in the southeastern and southwestern part of Iran, respectively. However, the largest arrival time (8.0 s) is seen beneath the station KHMZ located in the SSZ. The stacked PRFs at all stations in different tectonic zones are presented in Fig. 5a–e and arranged after the Moho phase arrival time in seconds. The arrival time of the Moho converted phase can be clearly seen in the PRF data. Small differences in the arrival time of the Moho-converted phase can be observed in some tectonic zones (e.g. Kopeh-Dagh, Alborz). Beneath the Zagros zone and Central Iran these differences seem to be relatively larger. This led us to divide these zones into subregions. Based on our observations, the Moho-converted phase is seen at delay times ranging between 4.8 and 6.5 s beneath the Kopeh-Dagh. It is observed at 5.5–7.8 s delay times beneath the Alborz zone. We found relatively smaller delay times for the Moho phase beneath Central Iran (4.4–6.6 s) and Makran zone (3.8 s). The delay time of the Moho-converted phase ranges between 3.7 and 6.8 beneath Zagros. The largest delay time of the Moho phase is seen beneath the SSZ (station KHMZ, 8 s).

Fig. 4
figure 4figure 4figure 4

Individual PRFs with summation traces for seven broadband and two short-period stations (MHD, VIS) in different tectonic zones of Iran. Individual seismograms are plotted equally spaced and sorted by increasing back azimuth (red rectangles). Black dots indicate the epicentral distances (shown on the right). They are filtered with a low-pass filter of 2 s. The P onset is fixed at zero time. The Ps conversion phases from the Moho are marked with red dashes lines (labeled Moho Ps)

Full size image
Fig. 5
figure 5

Stacked PRFs obtained for each tectonic zone. A low-pass filter of 2 s is applied. PRFs are sorted by the increasing arrival time of the Moho converted phase. Ps conversions from the Moho discontinuity are shown by red bar lines (labeled Moho Ps). a Stacked PRFs for stations located in the Zagros orogenic system (ZFTB, SSZ and UDMA). b Same as (a) for the Alborz tectonic zone. c Same as a for the Kopeh-Dagh mountain range. d Same as a for Central Iran. e Same as a for the Makran subduction zone

Full size image

4 Crustal Thickness

We estimated Moho depths based on Ps conversion times using available velocity models obtained from the previous geophysical studies in Iran for each tectonic zone (see Table 2). The Moho depths are listed in Table 1. The Moho depth varies between 32 ± 2 km at CHBR station in the Makran zone and 61 ± 2 km at KHMZ station in the SSZ. In the next step, we have used the arrival times of crustal multiples to determine the crustal thickness using the Z&K stacking approach. We applied this method only for stations which show clear multiple phases. We chose weight factors of 0.5, 0.25, and 0.25 for Moho conversion and crustal multiples, and performed a grid search for estimating the Moho depth and crustal Vp/Vs ratio. The maximum amplitude of stacked traces occurs where the three phases add constructively. The results of the Z&K method for some stations (shown in Fig. 4) are presented in Fig. 6. We also show the stacked moveout corrected receiver functions for the Ps and the multiple phases (PpPs and PpSs), respectively. As Fig. 6 shows the multiple phases are amplified after the correct moveout correction and fit the arrival times predicted by the final model. The Moho depths obtained from the Z&K method are also listed in Table 1. Based on our finding, the Moho depth varies between 31.5 ± 1.5 km at station CHBR in the Makran zone and 64.5 ± 2 km at station KHMZ in the SSZ. At some stations with very weak multiples, we found relatively large differences between the estimated Moho depths obtained from the Ps arrival time and Z&K method (see Table 1).

Fig. 6
figure 6figure 6figure 6

Examples of Z&K stacks of receiver functions for six stations indicated also in Fig. 4. A grid search is performed to estimate the Moho depth and crustal Vp/Vs ratio. The amplitude is shown in the lower part and ranges from 0.8 to 1. The optimal combination of the crustal thickness and Vp/Vs ratio is defined where the largest amplitude (1) occurs as marked by a red solid circle. The three traces on the left side of each panel show receiver functions after moveout correction for the Ps, PpPs, and PpSs phases, respectively. Black arrows mark the predicted times of the three phases obtained from the grid search

Full size image

5 P Receiver Function Modeling

We used forward modeling of the receiver functions to find the most suitable crustal thickness beneath each station (see also Taghizadeh-Farahmand et al. 2010 and Afsari et al., 2011). P wave velocity models shown in Table 2 were used as starting models for each tectonic zone. Figures 7 and 8 illustrate the results of forward modeling for one short-period (MHD) and one broadband station (SNGE) (see also Fig. 4). We first determined the Moho depth and tried to find the simplest model, which fits all other converted phases (e.g. sedimentary layers, Moho multiples). As an example, we present in Figs. 7 and 8 three selected models among many other models calculated for stations MHD and SNGE. A simple model containing a pronounced sedimentary layer and a Moho boundary can be well matched with the observed seismograms. Moho depths obtained from forward modeling are listed in Table 1. We also showed the differences between the results of forward modeling and those obtained from Ps arrival time in Fig. 9. In general, the differences are not larger than 4 km (except nine stations).

Fig. 7
figure 7

Forward modeling of the stacked PRF for station MHD located in Central Iran. The dashed line in the right panel is the observed P receiver function. The solid line represents the synthetic P receiver function corresponding to the model shown in the left. a Synthetic PRF calculated for a model with a Moho boundary at 47 km depth. b Same as a for a model with a 2 km thick sedimentary layer and a Moho boundary at 48 km. c The model with the best fit contains a 5 km thick sedimentary layer and a Moho boundary at 49 km. Ps_sed: conversion from the bottom of a sedimentary layer, Ps_Moho: conversion from the Moho boundary; PpPs_Moho: the first Moho multiple with positive amplitude

Full size image
Fig. 8
figure 8

Same as Fig. 7 for station SNGE located in the SSZ

Full size image
Fig. 9
figure 9

Differences between the estimated Moho depths calculated by Ps arrival time (according to velocity models presented in Table 2) and those obtained from Zhu and Kanamori approach (2000)

Full size image

6 Results and Discussion

We presented new Moho depth maps for Iran derived from our results of Ps arrival time, Z&K approach, and forward modeling (Fig. 10a-c). To construct the Moho depth map we interpolate all the depth values directly obtained from our analysis beneath stations. Our interpolation is reliable for the areas, which are well covered by stations (e.g. western Iran). This is in contrast to some other areas (e.g. Central Iran), where our estimations are limited to the results of few stations. For these areas, the depth values beneath each station are also shown with colors and are more accurate than those obtained from the linear interpolation. All our three Moho depth maps reveal the same trend beneath Iran, implying that the thickest crust is beneath the Alborz zone and along the SSZ zone. While the crustal thickening beneath the Alborz is related to the shortening process associated with the orogenic belt, the crustal thickening beneath the SSZ may reveal the underthrusting of the Arabian plate beneath the Iranaian plate (Paul et al., 2010; Mohammadi et al. 2013a). The average crustal thickness elsewhere in Iran is about 40-45 km except in southeast Iran which shows a normal continental crust of about 33 ± 2 km, which has not been significantly influenced by collisional processes.

Fig. 10
figure 10figure 10

The Moho depth map (in km) obtained from P receiver function analysis. a using Ps arrival time, b using Z&K approach, and c using forward modeling

Full size image

Our presented Moho depth map (from PRF modeling, Fig. 10c) is the first Moho depth map obtained from high frequency receiver functions and appears to be more accurate (error of  ±2 km) than the global Moho depth maps (e.g. Mooney et al., 1998; Bassin et al., 2000), which provide rough estimates of the Moho depth within Iran and those previously obtained by gravity data (Dehgani and Makris 1984), partitioned waveform inversion (Shad Manaman et al., 2011) and regional/residual Bouguer anomalies (Jiménez-Munt et al., 2012). In the following subsections, we summarize and compare our results obtained from modeling for each tectonic zone of Iran with those shown by previous geological and geophysical studies.

6.1 The Caspian Basin and Surrounding Mountain Ranges (Alborz, Binalud, and Kopeh-Dagh)

The Alborz mountains form a seismically active fold-and-thrust belt along the southern Caspian Sea coast extending from the southern end of the Talesh Mountains in the west to their junction with the Kopeh-Dagh Mountains in the east and central Iran in the south (Fig. 1). A number of geophysical studies have focused on the crustal structure of the Alborz region (e.g. Sodoudi et al., 2009; Abbassi et al., 2010; Radjaee et al., 2010; Motavalli-Anbaran et al., 2011; Nasrabadi et al., 2011) and provided different estimates of the Moho depths due to various assumed body wave velocities and model resolutions. Our results showed a thickening of the crust from 47 ± 2 km beneath the western part of the Alborz mountains (station ZNJK) to ~54 ± 2 km below the central part of this region. Beneath the southeastern part of the Alborz mountains in the Binalud zone, the crust thins out and is 45 ± 2 km thick (station SHRD). We also observed a decrease in the Moho depth towards the north and south of central Alborz. A crustal thickness of ~51–54 km was shown by the joint analysis of P and S receiver functions beneath the central Alborz by Sodoudi et al. (2009). They found an unusual crustal thickness of about 67 km beneath the Damavand volcano, which is located on southern flank of the range. However, their analysis was based on the data obtained from one short period station (DMV). Moreover, they used the IASP91 reference model (Kennett and Engdahl, 1991) for the depth estimation, which may have higher crustal velocities than the local model used in this work. Our results revealed a relatively large crustal thickness beneath the Alborz region, which can be related to the shortening process. A local crustal thickening to about 57 ± 2 km beneath the central part of Alborz (obtained from three stations, DMV and DAMV and THKV) is consistent with the result of Sodoudi et al. (2009) beneath the Damavand volcano if we take the errors of depth estimation produced by using a reference model into account (~5 %). This local thick crust may be attributed to the magmatic addition at the base of the crust beneath the volcanic region (Sodoudi et al., 2009). If it is valid, we may confirm the earlier suggestions (e.g. Dehgani and Makris 1984; Jackson et al., 2002) showing no deep root beneath the high-elevated central Alborz. Our findings are also consistent with the results shown by joint inversion of receiver functions and Rayleigh wave group velocity (Abbassi et al., 2010; Nasrabadi et al., 2011), and those obtained from fundamental mode Rayleigh wave group velocities (Radjaee et al., 2010).

Beneath Kopeh-Dagh we estimated an average crustal thickness of about 45 ± 2 km. Furthermore, we showed that the Moho depth varies from ~43 ± 2 km beneath the southern Kopeh-Dagh foreland basin to ~49 ± 2 km below the northern part of the basin. Our results can be confirmed by those obtained by Mangino and Priestley (1998) and Jiménez-Munt et al. (2012) for the NE Iran. They are also in good agreement with those shown by Motavalli-Anbaran et al. (2011), who estimated the Moho depths using gravity, geoid, topography, and surface heat flow data.

6.2 The Zagros Orogenic System

The Zagros Mountain belt in southwestern Iran results from the collision of Arabia and Eurasia plates most likely in the early Miocene. Previous studies in the Zagros (e.g. gravity and seismic studies) indicated a relatively thick crust (~40–45 km) beneath this region (Snyder and Barazangi 1986; Hatzfeld et al. 2003; Paul et al. 2006; Shad Manaman and Shomali 2010; Afsari et al. 2011; Mohammadi et al. 2013). We that found various Moho depths related to the different structure units exist in this area. According to our results (Fig.10c), the crust has an average thickness of about 43 ± 2 km beneath the NW Zagros Fold and Thrust Belt (ZFTB, Fig. 1) and shows a significant thickening (~51 ± 2 km) beneath the central part of this region, which most likely represents the overthrusting system beneath this area (Berberian 1995). This result is consistent with that shown by Afsari et al. (2011). The average Moho depth increases to about 48 ± 2 km below the central part of the ZFTB. At the end of the SE Zagros between the Zagros continental collision and Makran subduction zone, our Moho depth map indicates a crustal thickness of about 52 ± 2 km confirming the results shown by Yamini-Fard and Hatzfeld (2008) and Tatar and Nasrabadi (2013). In general, our Moho depths are consistent with those obtained from other studies in the ZFTB (e.g. Hatzfeld et al., 2003; Paul et al., 2006; 2010). Our results indicate that crustal thickening and shortening in the collision zone of Zagros is not constant. These results are consistent with those shown by Vernant et al. (2004) and Vernant and Chéry (2006), who indicated that the convergent rate varies from 4.5 ± 2 mm year−1 in the northwestern part, to 9 ± 2 mm year−1 in the southeastern part of the Zagros.

Beneath the SSZ we observe an average crustal thickness of about 54 ± 2 km with a strong increase to about 66 ± 2 km (beneath KHMZ). Paul et al. (2010) compared the Bouguer anomaly data and the Moho depths obtained from two profiles crossing the ZFTB and the SSZ. Their comparison significantly showed that the location of the maximum Moho depth beneath the SSZ does not coincide with the minimum Bouguer anomaly. To reconcile the gravity data with Moho depths they proposed that the localized thickening beneath the SSZ reveals the overthrusting of the crust of Central Iran onto the Zagros crust along the MZT. A more recent study based on S receiver functions (Mohammadi. et al., 2013) clearly imaged a significant crustal thickening (~70 km) beneath the SSZ and resolved the presence of two different lithospheric blocks beneath Iran separated in the northeast of the UDMA (along a profile crossing northwest Zagros). Shad Manaman and Shomali (2010) showed similar results beneath the ZFTB (~45 km) and SSZ using a partitioned waveform inversion method. Based on residual Bouguer anomalies, Jiménez-Munt et al. (2012) indicated a maximum crustal thickness of about 60 km beneath the Zagros collision zone between Fars and Lurestan arcs. Furthermore, they argued that the crust thins towards the Central Iran block and Persian Gulf and reaches about 42 km. We found an average crustal thickness of about 46 ± 2 km beneath UDMA, which is also supported by the results shown by Paul et al. (2006, 2010), Shad Manaman and Shomali (2010), Afsari et al. (2011), and Tatar and Nasrabadi (2013).

6.3 Central Iran

The Central Iranian Micro-Continent (CIMC) consists of separated blocks that drifted from Gondwana in the Permian to early-Triassic, and subsequently accreted onto Eurasia along the Alborz and Kopeh-Dagh sutures during the late Triassic closure of the Paleo-Tethys (Falcon 1974; Stoneley 1981). Based on our findings (Fig. 10c), the average Moho depth beneath Central Iran, from the north to the south, varied between 42 ± 2 to 46 ± 2 km. However, our results were obtained only from few stations, but are in good agreement with the results indicated by Paul et al. (2006, 2010), Shad Manaman and Shomali (2010), Afsari et al. (2011), and Motavalli-Anbaran et al. (2011). Sodoudi et al. (2009) calculated the PRFs beneath the northern part of Central Iran. Using IASP91 reference model, they estimated the Moho at about 51 km depth, which is deeper than our estimate. The reason for this difference is related to the IASP91 model, which is relatively faster than the average velocity model we used for Central Iran (see Table 2). Beneath Azarbaijan (see Fig. 1), which is located in the northwestern part of Central Iran between two thrust belts—the Caucasus to the north, and the Zagros mountain belt to the south—we found an average crustal thickness of about 45 ± 2 km in good agreement with the results shown by Mangino and Priestley (1998). According to Fig. 10c, the Moho depth increases from west to east. It is not completely flat and increases smoothly from 40 ± 2 km under the Urumieh Lake in the west to about 50 ± 2 km in the east of the region. There is also a decrease in the Moho depth towards north. The Moho depth map presents a crustal thickening towards the northeast. Taghizadeh-Farahmand et al. (2010) attributed this variation to the collision between Central Iran and South Caspian plate, which most likely shows the crustal shortening processing in this part of Iran. Our estimations are also consistent with those obtained from a joint inversion of PRFs and Rayleigh waves beneath NW Iran (Nasrabadi et al., 2011) and those shown beneath eastern Turkey (~45 km) (e.g. Zore et al., 2003; Angus et al., 2006). Furthermore, our results beneath the eastern part of Central Iran (~42 ± 2 km) are consistent with those shown by Nasrabadi et al. (2011), Rajab-beiki et al. (2011) and Jiménez-Munt et al. (2012).

6.4 Southeastern Iran (Makran)

The Makran region is the Oceanic-Continental subduction zone which is located in southeastern Iran and southern Pakistan. It is expanded ~1,000 km from west (Iran) to east (Pakistan) and its width is around 300 km. The north border reaches the Jazmoorian depression, and the southern range of this zone is limited to the Oman seacoast (Fig. 1). Few geophysical studies have focused on the Makran subduction zone, which mostly resolved the shallow seismic structure of this zone. Unfortunately, our estimation is bounded to the result of only one station (CHBR) showing the thinnest crust (~33 ± 2 km) within Iran. A crustal thickness of 33 km is very close to the average thickness of the continental crust and may show that the crust beneath this area has not been significantly thickened. This result is in good correlation with the absence of collisional processes beneath this region. Dehgani and Makris (1984) estimated the Bouguer anomaly for the whole Iranian plate and implied a crustal thickness of ~30 km beneath the Makran region. Moreover, our result is consistent with the results shown by Shad Manaman et al. (2011), who found a thin crust (25–30 km) under the Oman seafloor and Makran foreacre setting. A PRF study beneath the western end of the Makran prism (Yamini-Fard and Hatzfeld, 2008) showed also the Moho boundary at ~32 km depth.

7 Conclusions

PRFs were calculated for the teleseismic events recorded between 1995-2011 at 77 national permanent stations (24 broadband and 53 short period) of Iran. We presented the first Moho depth map by forward modeling of PRFs. Our estimated Moho depth values coincided fairly well with those obtained from previous analysis using different geophysical approaches. Because of the different deformation zones existing in the study area, our results showed significant variations of the Moho depth beneath the Iranian plate. The maximum Moho depth (~66 ± 2 km) was seen along the Zagros mountain belt beneath the SSZ, where the crust of Central Iran is assumed to overthrust the Zagros crust along the MZT. The average crustal thickness beneath the ZFTB and the SSZ was estimated to be about 43 ± 2 km and 50 ± 2–55 ± 2 km, respectively. In general, we found average crustal thicknesses of about 40 ± 2 to 45 ± 2 km beneath the Iranian plate increasing northwards to about 50 ± 2 km beneath the Alborz Mountains, and up to 56 ± 2 km near the Damavand volcano due to the shortening process related to the orogenic belt. The crustal thickness ranges between 40 ± 2 and 44 ± 2 km beneath the Central Iran decreasing towards the SE and reaching about 33 ± 2 km beneath the Makran region, due to the lack of significant collisional processes. The Moho depth increases northwards to the Kopeh-Dagh Mountains with values varying between 43 ± 2 and 50 ± 2 km.