Earth and Planetary Science Letters xxx (2008) xxx–xxx

EPSL-09485; No of Pages 6

Contents lists available at ScienceDirect

Earth and Planetary Science Letters

j ourna l homepage: www.e lsev ie r.com/ locate /eps l

ARTICLE IN PRESS

Distribution of the right-lateral strike–slip motion from the Main Recent Fault to theKazerun Fault System (Zagros, Iran): Evidence from present-day GPS velocities

F. Tavakoli a,b, A. Walpersdorf a,⁎, C. Authemayou c,e, H.R. Nankali b, D. Hatzfeld a,M. Tatar d, Y. Djamour b, F. Nilforoushan b, N. Cotte a

a Laboratoire de Géophysique Interne et Tectonophysique, CNRS-INSU UMR 5559, Université Joseph Fourier, Maison des Géosciences, BP 53, 38041 Grenoble Cedex 9, Franceb National Cartographic Center, Meraj Ave. Azadi Sq., P.O.Box 13185-1684, Tehran, Iranc CEREGE, CNRS, Université Aix-Marseille-3, BP 80, Europôle Méditerranéen de l'Arbois, 13 545 Aix-en-Provence Cedex 4, Franced International Institute for Earthquake Engineering and Seismology, P.O. Box 19395/3913, Tehran, Irane Institut de Paléontologie et Géologie, Université de Lausanne, CH-1015 Lausanne, Switzerland

⁎ Corresponding author. Tel.: +33 4 76 63 52 18; fax:E-mail address: Andrea.Walpersdorf@ujf-grenoble.fr

0012-821X/$ – see front matter © 2008 Elsevier B.V. Aldoi:10.1016/j.epsl.2008.08.030

Please cite this article as: Tavakoli, F., et al., DiLett. (2008), doi:10.1016/j.epsl.2008.08.030

a b s t r a c t

a r t i c l e i n f o

Article history:

GPS measurements across t Received 12 December 2007Received in revised form 14 August 2008Accepted 25 August 2008Available online xxxx

Editor: C.P. Jaupart

Keywords:GPSIranZagrosfault slip rates

he Kazerun Fault System in the Zagros mountain belt provide first instantaneousvelocities on the different segments. These results are closely consistent with the geological fault slip rates(over 150 ka), implying stable velocities over a longer period. The present-day strike–slip motion isdistributed from the Main Recent Fault to the N-trending Kazerun Fault System along a preferential en-echelon fault zone included in a more distributed fan-shape fault pattern. The Hormuz salt decoupling layercannot be the only cause of a sedimentary spreading because seismicity attests these faults are rooted in thebasement. The Dena fault (3.7 mm/yr) transfers the MRF fault slip mainly to the Kazerun (3.6 mm/yr) andslightly to the High Zagros and Sabz Pushan faults (1.5 mm/yr), and the Kazerun fault further to the Kareh Basfault (3.4 mm/yr). Total geological horizontal offsets associated with GPS slip rates help inferring precise faultslip onset ages. The successive onsets deduced by this approach imply that the right-lateral strike-slipactivity of the MRF has propagated in time southeastward to the Dena segment, and then to the Kazerunsegment and to the Kareh Bas fault.

© 2008 Elsevier B.V. All rights reserved.

1. Introduction

The NW-trending Zagros fold-and-thrust belt in southwest Iran isthe result of the Neogene collision between the Arabian and Eurasianplates (e.g. Falcon, 1969). The belt is composed of two distinctstructural domains: The narrower North Zagros oriented obliquely tothe direction of regional shortening is associated with orogen-parallelthrust faults and is affected to the north by the large MRF strike-slipfault suggesting partitioning of the oblique motion. The wider CentralZagros oriented perpendicular to the direction of Arabia–Eurasiashortening is also associated with thrust faults experiencing pureshortening, and with the presence of the Hormuz salt detachmentlayer at depth (Talebian and Jackson, 2004).

The present day kinematics of the Zagros is characterized in NorthZagros by 3–6 mm/yr of orthogonal shortening and 4–6 mm/yr oforogen-parallel right-lateral strike–slip motion (2.5 mm/yr located onthe Main Recent Fault), and in Central Zagros by 10 mm/yr of NNEshortening, mainly concentrated in the southern part of the belt closeto the Persian Gulf (Tatar et al., 2002; Walpersdorf et al., 2006). The

+33 4 76 63 52 52.(A. Walpersdorf).

l rights reserved.

stribution of the right-lateral s

Kazerun Fault System (KFS) separates North and Central Zagros(Fig. 1). It is a series of N- to NNW-trending right-lateral strike–slipfaults (Berberian, 1995). The longest (300 km) is the en-echelonKazerun fault, comprising the Dena, Kazerun and Borazjan segments(e. g. Falcon, 1969). KFS further includes the Kareh Bas, Sabz Pushanand Sarvestan faults located further east. These faults are all basementstructures inherited from a Neo-Proterozoic tectonic phase (Talbotand Alavi, 1996). The three segments Dena, Kazerun and Borazjanof the Kazerun fault have similar length (∼100 km) and distort anddisrupt the Zagros folds (e.g. Berberian, 1995; Authemayou et al.,2005). Each segment is terminated to the south by a SE-trendingthrust (Authemayou et al., 2005).

The present day activity of the KFS is emphasized by historical andinstrumental earthquakes that have been located on different partsof the fault system (especially on the Kazerun and Kareh Bas faults),and are associated with right-lateral focal mechanisms (Berberian,1995; Talebian and Jackson, 2004) (Fig. 1). Estimated total offsets ofgeological markers across the Kazerun fault range between 8–27 km(Authemayou, 2006; Authemayou et al., 2006) and 140 km (Berberian,1995). The minimum offset estimate of 8 km total geological offsetis derived from the restoration of the topography that seems to becorrelated between both sides of the Kazerun fault segment. Thelarger offset of 27 km is determined by the restoration of a fold bent

trike–slip motion from theMain Recent Fault to the..., Earth Planet. Sci.

Fig. 1. Map indicating the major tectonic features (Berberian, 1995; Authemayou et al.,2006). Grey circles represent the seismicity from1923 to 2004 (Engdahl et al., 1998, lightgrey circlesMwN4.5, dark grey circlesMwN5.5). Also represented are focal mechanismsof the Global CMT Data Base from 1976 to 2008. Fault names from Berberian (1995):MZRF: Main Zagros Reverse Fault, MRF: Main Recent Fault, DF: Dena Fault, KF: KazerunFault, BF: Borazjan Fault, KB: Kareh Bas Fault, SBZF: Sabz Pushan Fault, SF: SarvestanFault, HZF: High Zagros Fault, MFF: Main Front Fault, ZFF: Zagros Front Fault. The insetshows the location of the study area (black frame) in the global tectonic context of theArabia–Eurasia collision. The white vectors indicate GPS velocities according to theArabia–Eurasia Euler pole of Vernant et al. (2004). The grey vectors show the measuredvelocities (with respect to Eurasia) of the 6 Central Iranian Block stations used toestablish the reference frame for the GPS velocities (Fig. 2 and Table 1).

2 F. Tavakoli et al. / Earth and Planetary Science Letters xxx (2008) xxx–xxx

ARTICLE IN PRESS

against the Kazerun fault by rotation. The maximum offset estimate of140 km is inferred from the apparent offset of the Zagros frontal fault,but this fault might not have been originally a continuous rectilinearstructure. In our opinion, the total offset estimate of 8 km is the mostreliable value.We suggest not taking into account the estimate of 27 kmdue to the complex evolution of folding in the vicinity of the Kazerunfault. Contrasted fold evolution on either side of the Kazerun fault iscontrolled by significant sedimentary facies changes (Sherkaty andLetouzey, 2004), out-of-sequence reactivation, and the synchronicity offolding and strike-slip movements suggesting an independent evolu-tionof folds on either sideof the fault. Theminimumoffset of 8 kmdoesnot suffer from these complications as it is based on topography offsetson either side of a dominantly transcurrent fault. This smaller offsetwill be used later to determine an onset age of the strike-slip motionin this work. This onset age could represent a lower limit, the faultbeing possibly older if the larger total offsets are significant.

The total fault offsets between 8 and 140 km result in a poorlyconstrained average fault slip rate between 1 and 15 mm/yr assumingthat the onset of slip on the KFS is contemporaneous with the mainregional shortening phase in the lateMiocene (e.g. Hessami et al., 2001).

Please cite this article as: Tavakoli, F., et al., Distribution of the right-lateral sLett. (2008), doi:10.1016/j.epsl.2008.08.030

This timing suggesting that the KFS plays the role of a lateral ramp ofthe Central Zagros fold and thrust belt is however poorly constrained.

In this paper we analyze the present day kinematics of theKFS, based on the measurements of a dedicated GPS network. Wequantify the slip rates along each fault and segment and examine theconsistency with geological slip rates (Authemayou et al., 2006;Authemayou et al., submitted for publication). Consistency reinforcesthe results of both methods, while disagreements could highlighteither the occurrence of transient motions affecting the GPS mea-surements, or insufficiencies in the simple rigid block model used tointerpret the GPS velocities, or an overestimation of the precision ofone of the two methods.

2. GPS measurements and analysis

A geodetic network of 11 benchmarks (bedrock sites with forcedantenna centering) was established in 2002 around the KFS. In twomeasurement campaigns in 2002 and 2004, each GPS site wasmeasured during at least 48 h. Data from the network surveys havebeen combined with several measurements of adjacent temporarynetworks (Nilforoushan et al., 2003; Vernant et al., 2004; Tatar et al.,2002; Walpersdorf et al., 2006), up to 6 Iranian permanent GPSstations, and 32 GPS stations of the International GNSS Service (IGS)network. We analyzed the data using the GAMIT/GLOBK 10.1 software(King and Bock, 2002). The precision of the campaign measurementscan be characterized by the mean repeatabilities of daily positioningsolutions. These are evaluated for north, east and vertical baselinecomponents in 2002 to 2 mm, 3 mm, 6 mm, and in 2004 to 2 mm,2 mm, 6 mm respectively. These repeatabilities represent only theshort-term errors of the GPS measurement, but they are still aconservative error estimate, as no network adjustments were donebefore comparing daily solutions. Combined over the 2-year measure-ment interval they result in an estimate of the horizontal velocityuncertainties of 2 mm/yr. Details about the processing of the data aregiven by Vernant et al. (2004) and Walpersdorf et al. (2006).

3. Velocity field

We present in Fig. 2 and Table 1 the velocity field around theKazerun Fault System in the Central Iran Block (CIB) frame (Vernantet al., 2004). The average residual velocity for the 6 stations of theCIB (see inset of Fig. 1 for location) is 0.5 mm/yr supporting that thisblock does not deform as suggested by Nilforoushan et al. (2003) andVernant et al. (2004), and attested by the associated low seismicactivity (Jackson and McKenzie, 1984). We ensure that co-seismicdisplacements in the time interval do not affect the surface mea-surements. According to the CMT catalogue, 9 earthquakes ofmagnitudes 5.0 to 5.7 occurred in the study region between the twomeasurement epochs. We calculated the co-seismic displacements forevery GPS site in an elastic half space (Okada 1985), using empiricalrelations (Wells and Coppersmith, 1994) and the CMT focal mechan-isms to define the size, the geometry and the depth of the faults.The highest cumulated displacements were found for stations ALIS,BIG2, LAR2 and TMN2with 0.8, 0.7, 0.6 and 0.6 mm, respectively. Post-seismic displacements affecting our stations can also be excluded. Thelast earthquake of magnitude higher than 6 (Mw=6.2) took placemore than 3 years before the measurements at more than 20 kmdistance of any of the GPS stations.

In a first approximation, all significant (greater than 2 mm/yr)differential velocities are located southwest of the Dena–Kazerun–Kareh Bas–MFF faults. All measurements north of these faults show nomotion relative to CIB suggesting that the MZRF, Sarvestan and SabzPushan faults are not very active.

If we assume that all faults are known and delimit rigid blocks, andthat the GPS benchmarks are located far from the elastic deformationzone at the edges of the blocks, we can increase the precision of the

trike–slip motion from theMain Recent Fault to the..., Earth Planet. Sci.

Fig. 2. Velocity field in the Kazerun region with respect to the Central Iranian Block.Formal uncertainties (Table 1) are indicated at a 1−σ level. Transects T1–T3 of Fig. 3 arehighlighted. CMT focal mechanisms indicate earthquakes during the measurementinterval from 2002 to 2004.

Table 1Site velocities of the Kazerun network, the 6 stations on the CIB and a selection of closeby IGS sites, in mm/yr with respect to Central Iranian Block (CIB) and ITRF2000(Altamimi et al., 2002)

CIB ITRF2000

Site Longitude °E Latitude °N Ve Vn Ve Vn σ Ve σ Vn

Zagros networks (Kazerun network and adjacent sites)ABAD 52.568 31.228 1.58 −1.32 27.39 20.54 1.41 1.40ABAR 53.308 31.123 1.42 −2.03 27.41 19.80 1.43 1.41ALIS 51.082 28.919 .45 5.13 26.40 26.72 0.95 0.92BAMO 50.980 30.109 1.44 3.38 27.01 25.35 1.43 1.41BIG2 53.637 27.852 .04 1.19 26.85 22.45 0.94 0.92CHEL 50.098 32.482 −1.10 .09 23.98 24.07 1.53 1.51DASH 51.814 29.945 .26 .39 26.11 22.23 1.45 1.43DAYY 51.836 27.850 6.18 4.93 32.30 26.67 1.47 1.43DEDA 50.578 0.990 2.10 1.86 27.50 23.73 1.19 1.18FAR2 52.106 28.851 −1.13 4.90 25.15 26.32 0.90 0.87GOT2 53.631 28.624 − .14 1.96 26.59 23.29 1.01 0.97ISL2 53.066 28.347 − .66 2.24 25.96 23.57 0.88 0.86KAN2 52.056 27.834 1.70 7.79 28.10 29.17 0.90 0.88KHO2 54.126 29.923 − .86 − .16 .78 21.25 0.89 0.87KRD2 50.531 31.808 −1.72 .15 23.69 22.00 1.51 1.50LAR2 54.320 27.644 1.25 4.03 28.24 25.21 0.91 0.88MARV 52.752 29.798 2.11 − .92 28.17 20.92 1.45 1.42OSL2 52.607 27.474 1.11 9.49 27.72 30.63 1.65 1.54QIR2 53.029 28.477 −1.30 1.32 25.22 22.58 0.86 0.85QOMS 51.799 32.250 − .43 − .74 25.22 21.10 1.52 1.50SAA2 53.146 30.087 .77 −1.33 27.11 20.13 0.97 0.90SARD 50.026 30.325 1.73 4.49 7.05 26.41 1.20 1.18SEDE 52.179 30.726 1.97 −3.09 27.74 18.78 1.41 1.39SEMI 51.430 31.225 .28 −2.41 25.83 19.49 1.20 1.18SEPI 51.358 30.610 .02 1.05 25.61 23.06 1.43 1.42SHAN 51.775 28.400 1.47 3.51 27.51 25.31 1.44 1.42SVR2 53.244 29.281 − .74 − .23 25.72 21.17 0.90 0.87TMN2 54.316 9.239 −1.24 −1.06 25.54 20.28 0.88 0.86YAGH 52.235 29.617 .56 .36 26.54 22.16 1.42 1.40

Central Iranian Block stationsARDA 53.822 32.313 .58 −1.45 26.71 20.15 1.04 1.02BIJA 47.930 36.232 − .61 .16 23.15 22.35 1.68 1.52HARA 54.608 30.079 .75 1.45 27.50 22.74 1.65 1.49KERM 57.119 30.277 − .02 1.19 27.29 22.31 2.46 1.68MIAN 46.162 36.908 − .24 1.14 22.96 23.48 1.61 1.49SHAH 50.748 32.367 −1.04 − .39 24.38 21.33 1.58 1.47

Close by IGS sitesANKR 32.758 39.887 −19.44 −10.72 −0.53 12.29 0.79 0.79BAHR 50.608 26.209 5.54 6.97 31.62 28.26 0.73 0.73IRKT 104.316 52.219 −7.98 −18.93 23.94 −8.44 0.74 0.73KIT3 66.885 39.135 .79 −16.68 28.50 3.51 0.74 0.74LHAS 1.104 29.657 11.55 −2.25 44.81 12.92 0.75 0.75POL2 74.694 42.680 −86 −15.30 28.03 3.46 0.74 0.74RAMO 34.763 30.598 −4.63 −4.44 17.35 17.95 1.06 1.03ZECK 41.565 43.788 6.50 −12.92 26.57 10.09 0.80 0.80ZWEN 36.759 55.699 8.39 −14.92 23.31 8.29 0.84 0.84

3F. Tavakoli et al. / Earth and Planetary Science Letters xxx (2008) xxx–xxx

ARTICLE IN PRESS

estimated fault velocities by averaging several measurements locatedon the same rigid block. The GPS site emplacements are effectivelyselected according to this strategy, far enough from an active fault tobe representative for the constant inter-seismic velocity between thetwo tectonic blocks separated by the fault. Relocated seismicity (Tataret al., 2002) indicate a brittle layer in the Zagros of 10–15 km deep,which means that in 10 km distance from the fault, a major part ofthe long term fault slip can be observed (according to the arctangentmodel of Savage and Burford, 1973). Only the station YAGH is closerthan 10 km to the Kareh Bas fault trace, because no good rock outcropin a safe place could be found further from the fault in this area. Thismeans that the displacement rate of YAGH, for example with respectto DASH, could underestimate the slip rate across the northernsegment of the Kareh Bas fault. All other sites are at least in 10 kmdistance from the next fault.

We compared site velocities on three west-east transects (T1, T2,T3 from north to south) crossing the Kazerun, Kareh Bas and SabzPushan faults (Fig. 2) to estimate the differential motion of the rigidblocks on each side of the faults. As the faults are approximatelyNS oriented, the north component of the velocities is assumed torepresent the strike-slip component of the fault motion (Fig. 3). Theeast component of the velocities would represent the shorteningmotion, but in all cases they are smaller than 2 mm/yr (and thereforewithin the uncertainties), indicating the absence of significant short-ening perpendicular to the KFS. For each transect, we compute thetotal slip velocity on a fault as the difference between the averagevelocities of the stations on each site of the fault (Fig. 3). The relative

Please cite this article as: Tavakoli, F., et al., Distribution of the right-lateral sLett. (2008), doi:10.1016/j.epsl.2008.08.030

distances between GPS sites and faults can be presented correctly onEW transects across NS oriented faults. However, in Fig. 3, the EWtransects cut also faults oriented NNW–SSE. In particular, the SBZF isoblique to transect 3. The intersections of the faults are indicated withrespect to the emplacement of the transect as shown in Fig. 2. Therefore,the relative distances to the GPS sites can only be interpreted in aqualitative way in Fig. 3.

For the northernmost transect T1 across the Dena segment wemeasure awell resolved right-lateral strike–slip rate of 3.7±0.8mm/yr.The middle transect T2 across the Kazerun segment yields a similarsignificant rate of 3.6±0.6 mm/yr, and the 1.5 mm/yr velocitycumulated over the Sabz Pushan (SBZF) and High Zagros (HZF) faultshas an uncertainty of 0.2 mm/yr.

The southernmost transect T3 crosses the Borazjan segment,the Kareh Bas and the Sabz Pushan faults. There is no significanthorizontal motion neither on the Borazjan fault (constrained only byALIS) nor on the Sabz–Pushan fault. Only the Kareh Bas fault shows asignificant dextral strike–slip motion of 3.4±0.3 mm/yr.

trike–slip motion from theMain Recent Fault to the..., Earth Planet. Sci.

Table 2Total horizontal fault slips and age of strike–slip onsets from Authemayou (2006) andAuthemayou et al. (2006, submitted for publication)

Fault Totalhorizontal slip

Age offault onset

Averagegeologicalslip rate

GPS derivedslip rate

GPS inferredfault onset

[km] [Ma] [mm/yr] [mm/yr] [Ma]

Dena 13 2.6–4.3 3–5 2.9–4.5 2.8–4.5Kazerun 8.2 2.7–8.2 1–3 3.0–4.2 2.0–2.7Borazjan 0 Not applicable Not applicable 0–1 Not

applicableKareh Bas 7.2 Still

undeterminedStillundetermined

3.1–3.7 1.9–2.3

MRF 50 5–8.3 6–10 0.5–4.5 11–100

GPS velocities from this study except for MRF from Walpersdorf et al. (2006).

4 F. Tavakoli et al. / Earth and Planetary Science Letters xxx (2008) xxx–xxx

ARTICLE IN PRESS

4. Comparison with geological displacement rates

Horizontal fault slip rates have been estimated from lateral offsetsof streams, fans and terraces and in situ cosmogenic 36Cl exposuredating of cobbles sampled on the surface of these geomorphic features(Authemayou, 2006; Authemayou et al., submitted for publication).The oldest alluvial fan is dated between 121 and 169 ka. Thus theseestimates are valid for the Quaternary period. These slip rates inferredby geochronology are more precise than the previously availableestimate of Berberian (1995) based on the poorly constrained faultonset age in the Late Miocene. Therefore, we can now compare thefirst well constrained long term average slip rates over severalthousands of years produced by Authemayou (2006) and Authemayouet al. (submitted for publication), with our GPS inferred instantaneousvelocities (Table 2).

For the Dena and the Kazerun fault segments, geological displace-ment rates are available (Authemayou et al., submitted for publica-tion) and evaluated to 3–5 mm/yr and 1–3 mm/yr, respectively. Thecomparison of these long term slip rates with the instantaneousslip rate of 2.9–4.5 and 3.0–4.2 mm/yr, respectively, shows that thegeological and the instantaneous displacement rates are consistent(within the uncertainties) and less than the 15 mm/yr suggested byBerberian (1995). The coherent observations also suggest that the

Fig. 3. Modeling of the Dena (T1), Kazerun (T2) and Kareh Bas (T3) fault kinematics asfaults separating rigid blocks. Total strike–slip velocities are indicated for each fault bydifferences between average NS velocities per block, corresponding to the strike–slipmotion on the roughly NS oriented faults. T2 also evaluates 1.5 mm/yr of cumulatedstrike–slip rate over HZF and SBZF. Individual NS velocity components per site areindicated with their 1−σ uncertainty as in Fig. 2.

Please cite this article as: Tavakoli, F., et al., Distribution of the right-lateral sLett. (2008), doi:10.1016/j.epsl.2008.08.030

faults might have slipped at constant rates for a long time period.Assuming a constant slip rate since the onset of the faults andconsidering the total fault offsets of 13 and at least 8 km on the Denaand the Kazerun fault segments, respectively (Authemayou, 2006), theshort and long term slip rates yield comparable onset ages for Dena(2.8–4.5 Ma by GPS with respect to 2.6–4.3 Ma by geology) andKazerun (2.0–2.7 Ma by GPS with respect to 2.7–8.2 Ma by geology,minimum values, Table 2).

We do not have a geological displacement rate for the Kareh Basfault. Supposing a constant slip rate of 3.1–3.7 mm/yr as measured byGPS, the 7.2 km fault offset (Authemayou, 2006) observed on theKareh Bas fault could be achieved in 1.9–2.3 Ma. Fig. 4 presents a mapwith the comparison between GPS present day slip rates and

Fig. 4. Compilation of GPS and geological slip rates for the major faults in the Kazerunregion. The inset presents the GPS inferred strike–slip onset ages.

trike–slip motion from theMain Recent Fault to the..., Earth Planet. Sci.

5F. Tavakoli et al. / Earth and Planetary Science Letters xxx (2008) xxx–xxx

ARTICLE IN PRESS

geological slip rates. The inset of Fig. 4 shows the successive fault agesinferred from these instantaneous slip rates and total geologicaloffsets, and the direction of fault propagation.

5. Discussion and conclusions

OurGPSmeasurementsprovide aprecise estimateof thepresent-daydextral strike–slip motion on the Dena (3.7±0.8 mm/yr) and Kazerun(3.6±0.6mm/yr) segments, and on the Kareh Bas fault (3.4±0.3mm/yr).These velocities indicate that the Dena, Kazerun and Kareh Bas faultsaccommodate most of the differential motion between North andCentral Zagros, with 3–6 mm/yr and 10 mm/yr shortening across Northand Central Zagros, respectively. Effectively, the 3–6 mm/yr NS short-ening across North Zagros are increasing from 3 mm/yr in the west to6 mm/yr in the east, close to the Kazerun Fault System. Therefore weexpect the KFS to absorb rather 4 than 4–7 mm/yr of differential NSshorteningbetweenNorth andCentral Zagros. Ourmeasurements of theslip rates across the Dena, Kazerun and Kareh Bas faults reach the sameorder of magnitude, in particular taking into account that part of thestrike–slip is also found on the High Zagros fault and the Sabz Pushanfault further east.

Themajor result of our study is that the comparisonof the geologicalslip rates in the Kazerun Fault System (Dena and Kazerun faults,Authemayou et al., submitted for publication) with instantaneous sliprates indicates generally stable rates over ∼150 ka (Table 2). This resultis particularly remarkable in view of the complex fault pattern of theKazerun Fault System. The agreement enhances the significance of theslip rates obtained by completely independent methods.

Exception to the general agreement between instantaneous andlong term slip rates is the MRF. Its GPS velocity of 2.5±2 mm/yr(Walpersdorf et al., 2006) is slightly slower than the Kazerun fault sliprate and than the geological slip rates (Table 2).

However, the GPS orogen-parallel strike-slip motion across the totalNorth Zagros mountain belt is 4–6mm/yr (Walpersdorf et al., 2006) andis coherentwith theMRFgeological slip rateof6–10mm/yr (Authemayouet al., 2006). Therefore, theMRFGPS velocity could be underestimated bythe use of the rigid block model to explain the GPS velocities, suggestingNorth Zagros strike–slip deformation is rather continuous thanconcentrated on individual faults, or the errors of the geochronologicalor geodetical measurement technique are underestimated.

If we consider the GPS velocity uncertainties as significant, we caninfer slip onset times with lower uncertainties than the geologicalmethods and try to obtain as a secondary result new informationabout the chronology of slip onset in the Kazerun Fault System. Totalgeological horizontal offsets and GPS slip rates for the Dena andKazerun segments (Authemayou, 2006) permit to determine faultages of 2.8–4.5 and 2.0–2.7 Ma, respectively (Table 2). The Kareh Basfault age was determined supposing constant GPS slip rates to 1.9–2.3 Ma, comparable to the Kazerun segment. The comparison of GPSinferred strike–slip onset times with respect to the MRF (6–10 Ma)(Authemayou, 2006) could imply that the right lateral strike–slipactivity of the MRF has propagated in time southeastward to the Denasegment, and then to the Kazerun segment and the Kareh Bas fault.

According to seismic, GPS and geomorphologic data, the MZRF ispresently inactive (Berberian and King, 1981; Jackson and McKenzie,1984; Tatar et al., 2002; Yamini-Fard et al., 2006). According totectonostratigraphic relations between the MRF, the MZRF andPliocene deposits, cut by the MRF, the MZRF ceased to be activewith the strike–slip onset along the MRF (Authemayou et al., 2006).TheMZRF activity is not disturbed by the connection between theMRFand the KF, but by the onset of the MRF. We suggest that this is due tothe fact that the two faults use the same lithospheric interface: Thesuture zone between Arabia and Central Iran. The slip along thissuture zone changes from reverse to strike–slip in the Late Miocene.

The SBZF andHZF showa lowcumulative slip rate (1.5±0.2mm/yr).This slip rate inferred only from NS velocity components could be

Please cite this article as: Tavakoli, F., et al., Distribution of the right-lateral sLett. (2008), doi:10.1016/j.epsl.2008.08.030

underestimated due to the departure of the HZF from the general NSfault trend in the KFS. However, no geomorphic analyses on the SBZFand HZF have been realized so far to evidence their recent activity. Nosignificant strike–slip motion has been revealed on the southernmostBorazjan segment of the Kazerun fault, as well as on the easternmostfault of the KFS, the Sarvestan fault.

The general pattern of slip rate suggests that the present-daydeformation is distributed from the MRF to the Dena fault, from theDena fault mainly to the Kazerun and slightly to the HZF and the SabzPushan fault, and from the Kazerun to the Kareh Bas fault. Thecoherent slip in this complex en-echelon system shows that thelocation of deformation is different from the more distributed fan-shape fault pattern evidenced by tectonics. This implies that the slippattern evolves in time.

The evidence for slip in an en-echelon system rather than on allfault segments in the fan-shape pattern also implies that thedistribution of strike–slip motion from the MRF to the KFS is probablynot related to the presence of the very ductile Hormuz salt layerrestricted to the Central Zagros, east of the KFS. The Hormuz salt caneffectively permit the southward migration of the Zagros front actingas a low-resistance boundary. This would be in agreement with thesouth-eastwardmigration of the strike–slip along the KFS. However, ifthe slip motion along the KFS is only due to the salt tectonics in thecover, the maximum present-day slip rate must be localised along thesouthern part of the KFS as the active shortening location in theCentral Zagros. Nevertheless, this is not the case, themaximum strike–slip rate along the KFS is localised along the northern and centralsegments. Moreover, these fast slip rates along the northern andcentral Kazerun segments are coherent with the regional basementdeformation and seem to be due to the MRF slip transfer to the KFS.This transfer implies that the motion along the strike–slip faults isnot confined to the cover but also present in the basement as attestedby seismic data (Yamini-Fard et al., 2006).

Acknowledgements

We are grateful to all the French and Iranian observers withoutwhom these results would not exist. We had fruitful discussions withJ. Jackson, O. Bellier and M. Vergnolle. We thank Roland Bürgmann andan anonymous reviewer for the constructive comments that helpedclarifying the discussion of our results. This work has been supported byNCC, IIEES, and CNRS/INSU via the program Dyeti. F. Tavakoli benefitedof a scholarship from Total.

References

Altamimi, Z., Sillard, P., Boucher, C., 2002. ITRF2000: a new release of the internationalterrestrial reference frame for earth science applications. J. Geophys. Res. 107 (B10),2214.

Authemayou, C., 2006. Partitionnement de la convergence oblique en zone de collision:Exemple de la chaîne du Zagros (Iran), Ph.D. thesis, 334p, Univ. Paul Cézanne, Aix-Marseille, France.

Authemayou, C., Bellier, O., Chardon, D., Malekzade, Z., Abassi, M., 2005. Role of KazerunFault System in active deformation of the Zagros fold-and-thrust belt (Iran). C. R.Geoscience 337, 539–545.

Authemayou, C., Chardon, D., Bellier, O., Malekzadeh, Z., Shabanian, E., Abbassi, M.R.,2006. Late Cenozoic partitioning of oblique plate convergence in the Zagros fold-and-thrust belt (Iran). Tectonics 25, TC3002. doi:10.1029/2005TC001860.

Authemayou, C., Bellier, O., Chardon, D., Benedetti, L., Malekzadeh, Z., Claude, C.,Angeletti, B., Shabanian, E., and Abbassi, M.R. Quaternary Kazerun and Main RecentFaults' slip-rates: Evidence for strike-slip partitioning in the Zagros fold-and-thrustbelt, Geophys. J. Int., submitted for publication.

Berberian, M., 1995. Master blind thrust faults hidden under the Zagros folds: activebasement tectonics and surface morphotectonics. Tectonophysics 241, 193–224.

Berberian, M., King, G.C.P., 1981. Towards a paleogeography and tectonic evolution ofIran. Can. J. Earth Sci. 18, 210–265.

Engdahl, E.R., Van der Hilst, R.D., Buland, R.P., 1998. Global teleseismic earthquakerelocationwith improved travel times and procedure for depth determination. Bull.Seism. Soc. Am. 88, 722–743.

Falcon, N.L., 1969. Problem of the relationship between surface structures and deepdisplacement illustrated by the Zagros range. In: Kent, P.E., Satterwaite, G.E.,Spencer, A.M. (Eds.), Time and Place in Orogeny. Geol. Soc. Spec. Publ., 3, pp. 9–22.

trike–slip motion from theMain Recent Fault to the..., Earth Planet. Sci.

6 F. Tavakoli et al. / Earth and Planetary Science Letters xxx (2008) xxx–xxx

ARTICLE IN PRESS

Hessami, K., Koyi, H.A., Talbot, C.J., Tabasi, H., Shabanian, E., 2001. Progressiveunconformities within an evolving foreland fold-thrust belt, Zagros mountains.J. Geol. Soc. London 158, 969–981.

Jackson, J., McKenzie, D., 1984. Active tectonics of the Alpine–Himalayan belt betweenWestern Turkey and Pakistan, Geophys. J. R. Astr. Soc. London 77, 185–264.

King, R.W., Bock, Y., 2002. Documentation for the GAMIT analysis software, Release 10.1.Massachusetts Institute of Technology, Cambridge, MA.

Nilforoushan, F., Masson, F., Vernant, P., Vigny, C., Martinod, J., Abbasi, M., Nankali, H.R.,Hatzfeld, D., Bayer, R., Tavakoli, F., Ashtiany, M., Doerflinger, E., Daignieres, M.,Collard, P., Chéry, J., 2003. GPS network monitors the Arabia–Eurasia collisiondeformation in Iran. J. of Geodesy 77, 411–422.

Okada, Y., 1985. Surface deformation due to shear and tensile faults in a half-space. Bull.Seism. Soc. Am 75 (4), 1135–1154.

Savage, J., Burford, R., 1973. Geodetic determination of relative plate motion in CentralCalifornia. J. Geophys. Res 95, 4873–4879.

Sherkaty, S., Letouzey, J., 2004. Variation of structural style and basin evolution in thecentral Zagros (Izeh zone and Dezful Embayment), Iran. Iran. Mar. Pet. Geol 21 (5),535.

Talbot, C.J., Alavi, M., 1996. The past of a future syntaxis across the Zagros. In: Alsop, G.I.,Blundell, D.J., Davison, I. (Eds.), Salt Tectonics. Geol. Soc. Spec. Publ., London, vol.100, pp. 89–110.

Please cite this article as: Tavakoli, F., et al., Distribution of the right-lateral sLett. (2008), doi:10.1016/j.epsl.2008.08.030

Talebian, M., Jackson, J., 2004. A reappraisal of earthquake focal mechanisms and activeshortening in the Zagros mountains of Iran. Geophys. J. Int. 156, 506–526.

Tatar, M., Hatzfeld, D., Martinod, J., Walpersdorf, A., Ghafory-Ashtiany, M., Chéry, J.,2002. The present-day deformation of the central Zagros from GPS measurements.Geophys. Res. Lett. 29 (19). doi:10.1029/2002GL015159.

Vernant, P., Nilforoushan, F., Hatzfeld, D., Abbassi, M.R., Vigny, C., Masson, F., Nankali, H.R.,Martinod, J., Ashtiany, A., Bayer, R., Tavakoli, F., Chéry, J., 2004. Present-day deformationand plate kinematics in theMiddle East constrained by GPSmeasurements in Iran andnorthern Oman. Geophys. J. Int. 157, 381–398.

Walpersdorf, A., Hatzfeld, D., Nankali, H., Tavakoli, F., Nilforoushan, F., Tatar, M., Vernant, P.,Chéry, J., Masson, F., 2006. Difference in the GPS deformation pattern of North andCentral Zagros (Iran). Geophys. J. Int.167 (3),1077. doi:10.1111/j.1365-246X.2006.03147.x.

Wells, D.L., Coppersmith, K.J., 1994. New empirical relationships among magnitude,rupture length, rupturewidth, rupture area and surface displacement. Bull. Seismol.Soc. Am. 84, 1940–1959.

Yamini-Fard, F., Hatzfeld, D., Tatar,M., Mokhtari, M., 2006.Microearthquake seismicity atthe intersection between the Kazerun fault and theMain Recent Fault (Zagros, Iran).Geophys. J. Int. 166 (1), 186–196. doi:10.1111/j.1365-246X.2006.02891.x.

trike–slip motion from theMain Recent Fault to the..., Earth Planet. Sci.