Palaeomagnetic study of the Kepezdağ and Yamadağ volcanic complexes, central Turkey: Neogene tectonic escape and block definition in the central-east Anatolides
H. Gürsoy,O. Tatar,J.D.A. Piper,F. Koçbulut,Zafer Akpınar,Baochun Huang,A.P. Roberts,B.L. Mesci
DOI: https://doi.org/10.1016/j.jog.2010.07.004
IF: 2.673
2011-01-01
Journal of Geodynamics
Abstract:Research highlights ▶ The Yamadağ and Kepezdağ volcanic complexes were emplaced in mid Miocene times. ▶ Normal and reversed successions define most eruption between 15 and 13 Ma. ▶ The Yamadağ complex identifies ∼30° anticlockwise rotation relative to Eurasia. ▶ Anatolian rotations are differential and involve blocks on a ∼100 km scale. ▶ They result from continuing northward movement of the Arabian indenter. Abstract The Anatolian accretionary collage between Afro-Arabia and Eurasia is currently subject to two tectonic regimes. Ongoing slip of Arabia relative to Africa along the Dead Sea Fault Zone in the east is extruding crustal blocks away from the indenter by a combination of strike-slip and rotation. This zone of compression gives way to an extensional province in western Turkey, which also includes the eastern sector of Aegean Province. Although it is now well established that rotational deformation throughout Anatolia is distributed and differential, the sizes of the blocks involved are poorly understood. As a contribution towards evaluating this issue in central-east Turkey, we report palaeomagnetic study of the mid-Miocene Kepezdağ and Yamadağ volcanic complexes in central-south Anatolia (38–39.5°N, 37.5–39°E). A distributed sample through the Yamadağ complex identifies eruption during an interval of multiple geomagnetic field reversals (40 normal, 36 reversed, 8 intermediate sites) with a selected mean defined by 63 sites of D / I = 335.4/51.1° ( α 95 = 4.4°). The smaller Kepezdağ complex (8 reversed, 4 normal and 1 intermediate site) yields a comparable mean direction from 12 sites of 338.7/49.8° ( α 95 = 14.1°). In the context of a range of radiometric age evidence, two thick normal polarity zones within the Yamadağ succession probably correlate with zones C5ACn and C5ADn of the Geomagnetic Polarity Time Scale and imply that the bulk of the volcanic activity took place between ∼15 and 13.5 Ma. Comparison of the palaeomagnetic results with the adjoining major plate indenters shows that the Yamadağ complex has rotated CCW by 29.3 ± 5.2° relative to Eurasia; the much smaller dataset from the Kepezdağ complex indicates a comparable CCW rotation of 26.0 ± 11.8° with respect to Eurasia. The Arabian Indenter has also been rotating CCW since mid Miocene times, and the block incorporating these two volcanic complexes north of the East Anatolian Fault Zone (EAFZ) is determined to have rotated 18.2 ± 6.0° CCW relative to the northern perimeter of Arabia. Comparison with data to the north identifies quasi-uniform rotation across a ∼200 km wide block extending from the Central Anatolian Fault Zone in the northwest to close to the East Anatolian transform fault zone in the south east. Although absence of suitable younger rocks does not permit the timing of this rotation to be determined in the study area, analogies with results from the Sivas Basin suggest that it is young, and followed establishment of the major transform faults. Rotation has evidently taken place around bounding arcuate faults and accompanied westward expulsion as the accretionary collage north of Arabia has been subject to ongoing post-collisional indentation. Keywords Palaeomagnetism Volcanic rocks Miocene Anatolides Arabian plate Tectonic rotation Yamadağ Kepezdağ 1 Introduction The African, Arabian and Eurasian plates are presently interacting in the eastern Mediterranean region in a distributed tectonic domain where extension behind the southward-retreating Hellenic Arc, the most rapidly extending region on the globe in the absence of a constructive plate margin, gives way eastwards to a broad zone of extruded crust subject to compressive and strike slip deformation. The complexity of this latter zone results primarily from the inherent weakness of the wide belt of accreted terranes emplaced during the subduction of the former Neotethyan Ocean. Ongoing differential northward motion of the Arabian indenter into this Anatolian collage relative to the African plate is taken up by sinistral motion along the Dead Sea Fault Zone (DSFZ). This northward motion has been accompanied by anticlockwise rotation of Arabia and opening of the Red Sea ( Barka and Reilinger, 1997; McClusky et al., 2000; Gürsoy et al., 2009 ) with the cumulative motion responsible for extruding the dissected Anatolian crust towards the west. This tectonic escape of crustal blocks during post-collisional deformation has been accommodated by differential slip and rotation of crustal blocks; a net consequence has been to contract the crust north of the indenter and progressively expanded the curvature of the Tauride Arc ( Tatar et al., 2002, 2004; Piper et al., 2010 ). Thus the Anatolides are not a plate sensu stricto but are instead undergoing comprehensive internal deformation with strike slip movements between fault blocks on a ∼100 km scale. The differential character of this deformation is now well established by regional palaeomagnetic studies ( Gürsoy et al., 1997, 1998, 2003; Tatar et al., 2002; Piper et al., 2006, 2010 ). These show that block rotations produced by high strains north of the indenter are strongly counter clockwise (CCW) but become relatively less CCW as the strain diminishes to the west of the indenter, so that rotations diminish to near-zero in central Anatolia; still further to the west they become progressively more clockwise (CW) near the western extremity of the extruded terranes. The motions resulting from this deformation are presumably accommodated along major block-bounding faults extending through the upper crust, although these boundaries are rarely well defined due to a combination of superficial cover and presence of over-thrusting in the bordering regions. As a result the sizes of the blocks involved in the tectonic escape are poorly understood. Hence motivations for continuing to refine the post-collisional palaeomagnetic record include the need to (i) define the size of the blocks involved in rotation and delimit zones where contrasting rotations take place, and (ii) extend understanding of tectonic rotation across the Anatolides to compare with the Eurasian margin to the north and the Arabian indenter to the south. As a contribution to these objectives we report a palaeomagnetic study of the Kepezdağ and Yamadağ volcanic complexes sited at 38–39.5°N, 37.5–39°E north of the East Anatolian Fault Zone (see boxed zone in Fig. 1 (A) ); this investigation extends palaeomagnetic sampling of post-collisional volcanism in central-east Anatolia ( Gürsoy et al., 1997, 1998; Platzman et al., 1998; Kissel et al., 2003 ) southwards to a location close to the East Anatolian Transform and near to the north west extremity of the Arabian indenter. 2 Tectonic framework Present day Anatolia is part of the Alpine-Himalayan collisional belt and lies within the zone of convergence between the Eurasian and Afro-Arabian plates. The accretionary complex comprises a collage of continental fragments which moved across the Palaeotethys Ocean from former sites at the northern margin of Gondwana with a new ocean (Neotethys) opening in their wake ( Şengör and Yılmaz, 1981; Robertson et al., 1996 ). Closure of the Neotethys commenced during the Late Cretaceous and subduction accompanied accretion of continental fragments along the Eurasian margin, mainly during Eocene times. By mid Miocene times the major tectonic elements of present-day Anatolia were in place following terminal subduction and collision of Arabia with this terrane collage to form the Bitlis-Zagros Suture Zone (BZSZ). Subsequent post-collisional deformation has been driven by northward movement and differential slippage between Africa and Arabia along the Dead Sea Fault Zone (DSFZ, Fig. 1 ) which initially produced a compressive regime of thrusting and crustal thickening. Since the establishment of the intracontinental transforms was substantially later than suturing, this regime has been progressively succeeded by strike slip faulting and lateral escape to the west. As a consequence present day central Anatolia is now dominated by tectonic escape. Strike slip has been primarily focussed along major intracontinental transforms comprising the right-lateral North Anatolian Fault Zone (NAFZ) and the left-lateral EAFZ ( Fig. 1 ) with an anatomising network of subsidiary faults also accommodating differential rotations. The successive tectonic phases comprising Cenozoic deformation in the Anatolides are loosely divided into ‘Palaeotectonic’ and ‘Neotectonic’ phases with the former dominated by collisional tectonics and the latter comprising the post-collisional deformation. These are not rigid temporal divisions because the transition from collision to intracrustal deformation occurred much earlier in the north than in the south. In most kinematic models slip along the DSFZ is considered to meet the EAFZ at an unstable FFF triple junction sited close to city of Kahramanmaraş in southern Turkey ( Fig. 1 (A)). North of the EAFZ the second-order faults fall broadly into two sets: the first comprises NE-SW fractures that mostly splay off from the NAFZ into the Anatolian collage; the longest examples comprise the Kırıkkale-Erbaa (KEF) and Almus fault (AFZ) zones shown in Fig. 1 . Members of this system can display both right and left-lateral strike-slip movements, plus normal or reverse-slip components, where there are variations from this general trend ( Koçyiğit and Beyhan, 1998 ); dextral strike-slip is dominant in the east of Anatolia and includes the faults noted below which traverse the study region ( Kaymakçı et al., 2006 ). A second set of faults has NW-SE trends and oblique-slip motions, usually with a dominant right-lateral motion. Members of this latter set divide the western half of central Anatolia into horsts and grabens. They are concentrated within a transitional tectonic zone between the strike-slip regime of eastern Anatolia and the extensional regime of the Aegean and western Anatolia ( Yılmaz et al., 1987 and Fig. 1 ). 3 Geological framework 3.1 Regional setting The study area is located in Fig. 1 (A) and the volcanic outcrop amplified in Fig. 2 ; more detailed tectonic maps of the region are given by Kaymakçı et al. (2006) . The basement in this part of SE Anatolia comprises rock assemblages associated with closure of the Neotethyan Ocean and incorporates Mesozoic carbonates of the Anatolide-Tauride platform and magmatic, metamorphic and ophiolitic rocks ( Yılmaz and Özel, 2007 ) together with a range of intrusive rocks emplaced during Late Cretaceous times ( Boztuğ et al., 2007 ). The collisional history was dominated by southward-thrusting of ophiolites derived from the northern arm of Neotethys in Late Cretaceous times; collisional remnants are now sited along the Izmir-Ankara-Erzincan Suture Zone ( Fig. 1 (A)). From Early Tertiary times onwards deep to shallow water marine and continental sediments intercalated with volcano-sedimentary units, and including evaporates and lignite, infilled depressions developed on this basement. These post-collisional successions are sited within E–W to NE–SW trending sedimentary basins which from southeast to northwest comprise the Malatya basin bordering the BZSZ, the Darende-Balaban-Gürün basin and the Sivas Basin ( Akkuş, 1971; Poisson et al., 1996; Gürbüz and Gül, 2005; Kaymakçı et al., 2006 ). The infill into the Malatya Basin is mostly Eocene to Miocene in age, although at the southern margin a thick succession of fluviatile deposits comprises the Sultansuyu Formation and is of Early to Middle Pliocene age ( Kaymakçı et al., 2006 ). The ongoing evolution of the area is also recorded by continental sedimentation in the Elbistan and Kangal Neogene basins in the northwest sector of this region. Extensive volcanism associated with N–S convergence between the Arabian and Eurasian plates has been a prominent feature of the Neogene to Quaternary history throughout East Anatolia ( Yılmaz, 1990 ). This magmatism has a complex relationship to the tectonic history since it embraces the last phases of Neotethyan subduction, ongoing post-collisional compression, lateral extrusion by tectonic escape and intermittent regional extension ( Leo et al., 1974; Yılmaz, 1990; Pearce et al., 1990; Keskin et al., 1998; Yalçın et al., 1998; Arger et al., 2000; Keskin, 2003; Alpaslan et al., 2004; Kürüm et al., 2004; Ekici et al., 2005; Alpaslan, 2007 ). As a consequence of the close temporal relationship of these multiple tectonic regimes, the Anatolian region is not strictly analogous to the Tibetan or any other volcanic environment within the Alpine-Himalayan Orogen and appears to have a unique plate tectonic and crustal-lithospheric signature ( Keskin, 2003; Şengör et al., 2003 ). The Yamadağ and Kepezdağ volcanic complexes comprising this study are the record of Miocene activity within the western extension of the Eastern Anatolian Volcanic Province ( Yılmaz, 1990 ). Volcanic activity here has been focused along the north western rim of the Malatya Basin and near to the juncture between the NE–SW trending Ovacık Fault Zone (OFZ) and the linked NNE–SSW trending Malatya Fault Zones (MFZ, Fig. 1 ). The Malatya zone is >30 km wide and shows multiple phases of deformation ( Kaymakçı et al., 2006 ); the first comprised extensional deformation in Early to Middle Miocene times, which resulted in the opening of the basin and extensive volcanic activity along the northern perimeter. Successive reactivation of the two fault zones has included reverse and sinistral strike slip motions. These faults have continued to be active to the present day although post-mid-Miocene stress regimes have not permitted access to asthenosphere melts. 3.2 The Kepezdağ volcanic complex (KVC) The Kepezdağ volcanic complex (KVC) forms a morphological feature with a general N–S trend dissected by deep NE–SW trending valleys. It is composed of a pile of basaltic lavas, individually 3–20 m thick, of jointed and vesicular columnar units intercalated with pyroclastic deposits. Tuffaceous units are extensively developed at the base of the succession and observed mostly in the north and eastern parts of the KVC where they lie on marine sediments ( Turkish Geological Maps (TGM), Sivas sheet, 2002; Kaymakçı et al., 2006 ). At the northern margin of the volcanic complex the sediments include Upper Eocene-Lower Oligocene evaporites ( TGM Sivas sheet, 2002 ) and Lower Miocene marine sediments and carbonates of the Tahtalıtepe Formation ( Akkuş, 1971 ) which fix an older age of ∼16 Ma for the volcanic activity. Samples taken from palaeomagnetic field sites 11 and 12 ( Table 1 and Fig. 2 ) have yielded 39 Ar/ 40 Ar ages of 15.39 ± 0.23 and 15.9 ± 0.17 Ma with the latter coming from the oldest exposed unit in this part of the complex. These results accord with K–Ar ages between 14.1 and 15.3 Ma for 5 samples from the northern and central parts of the volcanic complex ( Ekici et al., 2007 , personal communication); one further result is marginally younger at 13.6 Ma ( Table 1 ). The geochronological evidence thus indicates extensive lava effusion over an interval of about 2 Myr between late Lower Miocene and Middle Miocene times. No significant eruption center is recognised within the present outcrop of the KVC. 3.3 The Yamadağ volcanic complex (YVC) The Yamadağ volcanic complex was emplaced onto the Munzur carbonate platform at the junction between the NE extension of the Tauride belt and the Sivas, Darende-Balaban-Gürün and Malatya sedimentary basins. It comprises a thick pile of dacitic-rhyolitic, trachyandesite-andesitic, basaltic lavas and associated pyroclastics. The underlying basement is formed of ophiolitic rock assemblages (Divriği, Hekimhan and Hasançelebi regions) intruded by acid plutons during Late Cretaceous times and variably covered by Tertiary basin fill ( Boztuğ et al., 2007 ), the Munzur Carbonate Platform and Lower Miocene marine clastics, neritic carbonates (Arapkir town, TGM Sivas sheet, 2002 ). The youngest rocks predating the YVC appear to be the Uğurlu Formation comprising claystone-siltstone alternations, and conglomerate and sandstone with gypsum lenses ( Uçurum, 1992; Yalçın et al., 1998 ); this succession is dated Late Oligocene-Early Miocene. Eroded landscape features indicate a sub-volcanic environment where columnar jointed, pink to grey rhyolitic-dacitic and trachyandesitic rocks, variably associated with altered zones, are observed within deep eroded valleys near the central parts of the YVC. Some minor eruption centres are observed in the central part of Yamadağ volcanic complex and these are highlighted by the dotted lines encircled in Fig. 2 . Palaeomagnetic site 64 at the summit of the Yamadağ is a basaltic lava flow dated 9.7 ± 0.2 Ma ( Table 1 ); this is also the youngest dated unit identified within this complex. The results of previous geochronological studies are summarised in Table 1 . As part of the present study 13 samples were taken for 39 Ar/ 40 Ar analysis and 11 produced acceptable ages. One result (site 77) yielding an age of 29.5 ± 0.9 Ma is anomalously high and much older than the youngest substrate. The remaining 10 sites however, yield 39 Ar/ 40 Ar ages in the range 9.9–17.6 Ma ( Table 1 ). Further ages of 14.1, 16.6 and 18.7 Ma have been determined from basaltic and andesitic flows in west-southwest parts of the complex ( Leo et al., 1974 ) whilst Platzman et al. (1998) obtained an age of 14.1 Ma from the northern part and 18.3 Ma in the southwest. Wider expression of post-collisional volcanism in this region has been dated at 5.9 and 4.0 Ma from isolated lava flows around Kangal and Gürün respectively to the north west, and 15.2–15.9 Ma ( Arger et al., 2000 ) and 12.2–15.8 Ma ( Kürüm et al., 2004 ) from flows near Arapkir; Ekici (2003) also obtained closely comparable K–Ar ages of 13.9, 14.3, and 14.8 Ma from the southern edge of the volcanic complex. The collective age and stratigraphic data are linked to the locations of the palaeomagnetic sites in Table 1 . The age data combined with the geochemical and stratigraphic evidence identify the YVC as the focus of major calc-alkaline activity concentrated within several million years of mid-Miocene time and perhaps continuing episodically until about 10 Ma. This magmatism thus correlates more widely with the last phase of the Eurasia-Arabia collision and regionally with extensional opening of the Malatya Basin ( Yalçın et al., 1998; Ekici et al., 2009 ). 4 Analysis of neotectonic deformation Methods for quantifying neotectonic crustal deformation apply to three contrasting time scales: in the short term (10–10 2 years) surface kinematics are resolvable from GPS and ground surveying, with earthquake evidence yielding an holistic analysis of contemporary crustal deformation. On archaeological to late geological time scales (∼10 3 –10 5 years) surface deformation can be evaluated by geomorphic study of features such as stream offsets and changing elevations of erosional and depositional surfaces. In the longer term (>10 5 years) recovery of palaeomagnetic directions is usually required to quantify crustal deformation. This latter approach has now been effectively applied over much of Anatolia due to the widespread preservation of young volcanic rocks that tend to be good recorders of the palaeofield direction. In deforming terranes palaeomagnetism aims to resolve cumulative kinematics by comparing palaeomagnetic directions with reference directions of the same age from the adjoining (and presumed rigid) indenting continental blocks. Contrasts in magnetic declination are the signature of cumulative tectonic rotation since the age of magnetisation and, in principle, contrasts in magnetic inclination (following appropriate adjustments for any tectonic tilt) are the signature of relative latitudinal movement. In practise these latter differences are more difficult to apply on time scales covered by the post-collisional history because geomagnetic inclination has, in the majority of cases, been found to be too shallow along the Alpine-Himalayan Orogen to be accommodated by actual northward latitudinal movements ( Cogné et al., 1999; Beck et al., 2001; Piper et al., 2006 ); as a result the tectonic significance of magnetic inclination differences is more difficult to quantify but may potentially be assessed from statistical analysis of directional distributions in large sample populations ( Tauxe and Kent, 2004 ). By combining geological observation with distributed palaeomagnetic sampling, the aim of palaeomagnetic investigation in deformed terranes is then to resolve the size and integrity of the fault blocks involved in the rotation. Palaeomagnetic sites for this investigation were sampled, typically by 7 distributed cores drilled over several metres of outcrop, using a portable drilling motor with cores oriented in situ by both magnetic and sun compasses. Specific site data and GPS-determined locations are summarised in Table 1 and the generalised spatial distribution is shown on the outline geological map of Fig. 2 . Whilst young lavas are typically high quality recorders of the palaeomagnetic field direction during rapid cooling at the time of eruption with weak fabrics providing a reliable record of the geomagnetic field direction, this is instantaneous in geological terms and therefore a component of the secular variation; sampling of a number of lavas is always required to yield a mean direction of palaeomagnetic, rather than virtual geomagnetic, significance. The attitude of flow structures and pseudo-bedding in lavas and layering in the pyroclastic units was routinely measured although indications of bedding were not determinable at massive outcrops; elsewhere dips contrasting significantly with the palaeohorizontal (>10°) were evident at only a small number of sites ( Table 1 ). In these cases the relevance to the palaeomagnetic data is normally unclear because the volcanic units may have flowed down primary slopes meaning that most or all of the measured tilt is primary. In this study tilt-adjusted data are used in the overall rotational analysis; however, since the number of sites to which the adjustment is applicable is relatively small (17 of 63 used in the mean calculation) and corrections are mostly small, the overall tectonic conclusions are not appreciably affected. 5 Methodology Magnetic susceptibilities were determined by Bartington Bridge and the majority of cores from each site treated by progressive alternating field (a.f.) demagnetisation using incremented steps of 5 milliTesla (mT) to 40 or 50 mT and then in 10 or 20 mT steps to 140 mT or until behaviour ceased to be systematic. A minority of cores from each site was treated by thermal demagnetisation in steps of 50 or 100 °C to the Curie points of the magnetic carriers using Magnetic Measurements MM2D demagnetisers; all measuring and demagnetisation was conducted within large sets of Rubens coils with the ambient geomagnetic field continuously nullified to minimise secondary remanence acquisition. Results of progressive treatment were projected onto horizontal and vertical planes to produce standard orthogonal projections; components comprising the total Natural Remanent Magnetisation (NRM) were resolved by eye for calculation of equivalent directions by principal component analysis (PCA). Since each volcanic unit is expected to record a near-instantaneous record of the geomagnetic field, within-site component directions recognised as belonging to the same population are combined to produce site means recording virtual geomagnetic field (VGP) directions. To complement the palaeomagnetic analysis small pieces of cores from selected sites were used for rock magnetic analysis using a Variable Frequency Translation Balance (VFTB). Four experiments were performed comprising Isothermal Remanent Magnetisation (IRM) acquisition, Back Field subtraction of IRM, hysteresis analysis and thermomagnetic (Curie point) determinations. 6 Rock magnetism Rock magnetic properties are summarised in Table 2 and typical sample results illustrated in Fig. 3 . Thermomagnetic determinations (saturation magnetisation, M s , versus temperature) mostly yield discrete determinations with heating paths and Curie points identifying low-Ti titanomagnetite as the ferromagnet ( Fig. 4 ); cooling paths lie variably below heating paths expressing incipient conversion of titanomagnetite to hematite with low M s in the furnace. In some samples Curie points as low as ∼510 °C indicate that these young volcanics have not experienced prolonged subsolidus exsolution to produce the Curie points near 580 °C as is typically observed in older volcanics. Hysteresis properties are a function of ferromagnet mineralogy and domain state: titanomagnetites and titanomaghemites saturate in fields of 300 mT or less, whilst titanohematites require much larger fields to achieve saturation. Parameters used to express these properties are saturation magnetisation, M s , saturation remanence, M rs , and coercive force, H c ( Table 2 ). Parameters M s and M rs are dependent on the concentration and type of magnetic minerals present: M s is independent of grain size but M rs is smaller for multidomain (MD) than for single domain (SD) grain properties. Hence the ratio of M rs / M s is a useful indicator of domain states: if M rs / M s is <0.1 the sample is dominated by MD grains; if M rs / M s >0.1 small but significant fractions of SD grains are present within a dominant MD assemblage. If mixed domain sizes are present within magnetite only (as indicated by M rs / M s values of 0.2–0.5), the equation X MD = (0.5 − ( M rs / M s ))/0.48 yields a semi-quantitative estimate of the fraction of MD grains present. IRM acquisition is also characteristic of dominant titanomagnetite with saturation largely achieved by 30 mT and little or no indication of further climb into the higher coercivities characterised by titanohematites ( Fig. 3 ). Hysteresis characteristics ( Table 2 ) indicate that domain sizes are predominantly MD but with significant fractions (10–40%) of SD grains; the latter are presumably responsible for the high magnetic stabilities observed during demagnetisation in the bulk of this collection. Observed shapes of the hysteresis loops are all pot-bellied; mixed high and low coercivity components do not appear to be present here, this is in general conformity with the absence of high IRM characteristics that might otherwise suggest that deuteric or hydrothermal hematite is pervasive in these volcanics. 7 Palaeomagnetic results Total Natural Remanent Magnetisations (NRMs) were measured using ‘Minispin’ magnetometers. Intensities of NRM vary by more than two orders of magnitude but are strong in all units, as is typical of young, mostly basic, volcanic rocks. A total of 645 cores were subjected to laboratory demagnetisation with 433 (67%) demagnetised by a.f. fields and the remaining 33% thermally demagnetised. With few exceptions sample magnetisations are highly stable to progressive demagnetisation and, often following removal of a viscous component, illustrate long linear trajectories converging to the origin of orthogonal projections ( Figs. 4 and 5 ); sometimes samples with reversed magnetisations illustrate the removal of a normal component in the present field (for example, site 71 in Fig. 4 ) or more rarely normally magnetised samples have a reversed component removed first. Within-site groupings of Characteristic Remanent Magnetisations (ChRMs) were evaluated and means calculated using standard Fisher (1953) analysis. Results are summarised in Table 3 and tight within-site dispersions and alpha95 angles of no more than a few degrees achieved from grouping data from dispersed samples within sites are an indication of rapid local chilling and magnetisation (e.g. Biggin et al., 2008; Johnson et al., 2008 ); they also imply that strong local fields over strongly magnetised young lava (which may have the potential for deflecting the field from the local direction during cooling, see Valet and Soler, 1999 ) do not seem to be a significant problem here. The normal and reversed directions of the present day geomagnetic field in this region are D / I = 0/57.5° and D / I = 180/−57.5° ( Table 4 ) and site mean data summarised in Table 3 show a mixture of normal and reversed polarities with the former somewhat more numerous (40 normal, 36 reversed, Fig. 6 ) and broadly compatible with the slight bias of the geomagnetic field towards normal polarity (54%) during the ∼10–20 Ma interval embraced by the bulk of the volcanism represented here ( Cande and Kent, 1995 ). Although postdating the collisional deformation, these volcanics are locally influenced by neotectonic faulting and we are unable to rule out effects of fault-related tilting on some directions although the magnitudes of these effects are likely to be small. Some site mean directions are clearly anomalous and probably record transitions of the palaeofield between polarities. Whilst the iterative method of Vandamme (1994) provides a means of excluding transitional field directions, the possible contributions of unaccomodated tilts and local rotations about fault planes means that these collections are unlikely to be perfect samples of palaeosecular variation ( Calderone and Butler, 1991 ). Accordingly we regard directions as anomalous, and possibly transitional, if the equivalent virtual geomagnetic poles (VGPs) are >45° from the mean pole. On this basis one site in the Kepezdağ complex and 8 of 76 sites in the Yamadağ complex are anomalous and presumed to reflect the high number of reversals evident in the lava successions here: the anomalous sites tend to cluster spatially and occur between lavas of normal and reversed polarity ( Table 3 and Fig. 7 ) suggesting that they record transitional fields. Twelve of 14 sites in the Kepezdağ complex define N or R polarities with 10 rotated counterclockwise (CCW) with respect to the present field and two (6, 12) slightly CW. With CCW rotations ranging up to 60°, the mean ( D / I = 158.7/−49.8°) is fairly dispersed ( α 95 = 14.1°) but identifies a mean CCW rotation of about 20° with respect to the present north-south axis ( Table 4 ). The bulk of the sites from the Yamadağ Complex, with similar numbers of normal and reversed polarities, are also characterised by CCW rotation from ambient north-south ( Fig. 6 ) typically by ∼30° but ranging from near zero to ∼60°. Acknowledging this to be the combined signature of secular variation and CCW tectonic rotation in the region as a whole, we isolate a small number of anomalous, sometimes large, CW rotations; these include site 19 (where the magnetic record is complex) and sites 45, 52, 55 and 62. Excluding these and 8 intermediate, possibly transitional, site mean directions we derive a mean of D / I = 335.5/51.1° ( α 95 = 4.4°) from the remaining 83% of the collection from this complex. Because these two volcanic complexes lie within a tectonically active domain the standard tests for dispersive properties of the site mean directions are likely to be imperfect. Also, with just four sites of normal polarity in the Kepezdağ collection there are too few to yield a meaningful reversal test ( McFadden and McElhinny, 1990 ). Although the angle between normal and reverse axes is only 3.9° the test is indeterminate ( p = 0.511, critical angle = 38.4°). The much larger collection from the Yamadağ complex however, yields a class B reversal test; the angle between the normal and reversed axes is 4.3° compared with a critical angle of 9.2° ( p = 0.511). The individual sites represent rapidly chilled volcanic units and are expected to record palaeofield components of the secular variation of geomagnetic rather than to have longer term palaeomagnetic significance; in this circumstance a population of site directions is always required to yield a time-averaged mean for tectonic analysis. The dispersion of VGPs can be used to assess whether the secular variation has been effectively averaged and therefore whether the overall means can yield time-averaged palaeomagnetic poles. The circular standard deviation ( S ≈ δ 63 ) for the distribution of Kepezdağ VGPs is 21.1° from a fairly small sample ( N = 12 sites, Cox, 1969 ). This compares with a value of S λ of 20° and lower and upper values of 17.7° and 22.6° in the latitude band 20–40° between 22 and 45 Ma for palaeosecular variation model G of McFadden et al. (1991) indicating that the group mean result from this volcanic complex is a realistic time-averaged result, as supported by the presence of at least one reversal. The standard deviation from the much larger Yamadağ complex is 20.8° with a 95% confidence interval between 18.1° and 23.9° and this sample (in a succession which includes multiple reversals) is evidently an efficient sample of mid-Miocene secular variation. In each case these conclusions are qualified by the observation that differential vertical axis rotations between fault blocks ( MacDonald, 1980 ) are likely to have modified the directional dispersions. The group mean palaeomagnetic inclinations derived from the Kepezdağ and Yamadağ volcanic complexes (49.8° and 51.1°, respectively) are shallower than the mid-Miocene inclination of ∼55° predicted from the apparent polar wander paths of Eurasia and Africa ( Besse and Courtillot, 2002 , Table 5 ). Hence the inclination-shallowing effect, which is extensively present in Tertiary rocks from the Alpine-Himalayan region (e.g. Cogné et al., 1999 ), is evidently also present here, as is observed further to the west in Anatolia (c.f. Beck et al., 2001; Piper et al., 2006 ). Factors potentially contributing to a shallowing of the mean inclination include a variety of effects related to primary deposition (especially important in clastic sediments, e.g. Gilder et al., 2001 ), the geomagnetic field source ( Wilson, 1971 ) and possibly unaccomodated tilt corrections ( Calderone and Butler, 1991 ). Tauxe and Kent (2004) provide a test for isolating the contribution of inclination-shallowing from departures from the geocentric axial dipole (GAD) by comparing an elongation parameter E defined by the ratio of eigenvectors of the directional distribution with the elongations predicted from palaeosecular variation models at different latitudes. Unfortunately with a recommended sample size of N ≥ 100, application of this method is not merited here. 8 Discussion 8.1 Magnetostratigraphy The palaeomagnetic sample includes a number of sections through lava successions which, together with the GPS record of site altitudes ( Table 3 ), provides an approximate control over the lateral extrapolation of polarity intervals from section to section. The collective evidence is summarised in Fig. 7 together with radiometric data for the age of specific levels. The latter are somewhat dispersed but a realistic comparison is constrained by the late Lower Miocene age of the substrate and an average radiometric age of ∼14.8 Ma. The contemporaneous Geomagnetic Polarity Time Scale (GPTS) is shown in Fig. 7 after Cande and Kent (1995) where we observe two protracted normal polarity chrons between 13.7 and 14.6 Ma that would appear to correlate with the two thick normal polarity successions observed in the central-southern part of the volcanic complex ( Fig. 7 ). There are clearly uncertainties in lateral correlation of polarity changes across this volcanic complex and contributory effects of faulting, and hence the precise number of reversals recorded here are unknown; nevertheless the record of polarity intervals shown in Fig. 7 is substantially more limited than the number of polarity chrons between maximum and minimum ages recorded from the KVC ( Table 1 ). The recognition of the two thick normal polarity divisions indicates that the bulk of the preserved volcanic succession is likely to have been emplaced over an interval of less than 2 Myr between ∼15 and 13.3 Ma ( Fig. 7 ). Since the volcanism is a record of access to mantle melts to the palaeosurface, it is reasonable to link this interval to the extensional regime responsible for initiating the adjoining Malatya Basin (Section 3.1 ). 8.2 Tectonic rotation Tectonic rotation, R ′, in deformed terranes is determined by comparison of the mean directions with the predicted palaeofield axis from the adjoining stable plates. Both observed and reference directions have confidence limits (Δ D and Δ D ref , respectively). Beck (1980) proposed a confidence limit on R ′ defined by Δ R ′ = ( Δ D 2 + Δ D ref 2 ) ; this is qualified by Demarest (1983) who showed that, provided α 95 is small and preferably less than 10°, a standard correction factor is applicable. For n ≥ 6 this correction factor lies between 0.78 and 0.80; Δ R ′ can then be derived from the equation Δ R ′ = 0.8 ( Δ D 2 + Δ D ref 2 ) . Only if R ′ > Δ R ′ can a rotation be regarded as significant. Group mean directions are compared with reference palaeofield directions ( Table 5 ) at the location of the study area calculated from the apparent polar wander path of plates bordering the Alpine Himalayan orogen ( Besse and Courtillot, 2002 ). Using 10–20 Ma reference palaeofields and applying this approach the YVC defines a CCW rotation of 28.6 ± 4.9° with respect to the African Plate and 29.3 ± 5.2° with respect to Eurasia. The smaller collection from the KVC defines a comparable CCW rotation of 25.3 ± 11.8° with respect to the African Plate and 26.0 ± 11.8° with respect to Eurasia. Thus there has been clear CCW rotation of the block incorporating the Yamadağ and Kepezdağ volcanoes. Collectively these two volcanic complexes occupy a block at least 90 km east-west and 120 km north-south ( Fig. 2 ). The comparison with the bordering Arabian Indenter however, is qualified because this latter block has also been subject to CCW rotation relative to Africa since Miocene times as it has moved northwards and the Red Sea has opened ( Gürsoy et al., 2009 ). Volcanic successions on the Arabian Plate south of the EAFZ and Bitlis Suture Zone of similar mid Miocene age to the YVC and KVC volcanoes yield an overall mean direction of D / I = 353.6/51.6° ( α 95 = 5.4°) rotated CCW with respect to Africa and Eurasia but significantly less than the well-constrained Yamadağ result of this study. Hence we recognise a substantial CCW rotation of ∼29° with respect to Africa and Eurasia. A smaller CCW rotation of 18.2 ± 6.0° is identified with respect to contemporaneous results from the northern margin of the Arabian Indentor from which KVC and YVC are separated by the EAFZ and the Bitlis Suture. Comparable CCW rotations to those recognised from these palaeomagnetic studies are approximately defined by the orientation of the remanent dipolar signatures resolved from the aeromagnetic anomalies over this region ( Büyüksaraç, 2007 ). 8.3 Regional assessment—distributed deformation of the Anatolides For comparison of these results with the regional palaeomagnetic record. Table 6 lists palaeomagnetic data from the Anatolides and adjoining regions embracing the time period beginning with closure of NeoTethys in Late Cretaceous times and continuing to the Quaternary. The compilation of published results from Turkey by Cengiz Tapırdamaz ( http://www.mam.gov.tr/enstituler/ydbe/paleo_katalogu.zip ) is included here together with some additional results in the Global Palaeomagnetic Database (GPDB) and recent results from the literature after Piper et al. (2010) . Reversed polarity palaeomagnetic directions are plotted in Fig. 8 to illustrate the pattern of magnetic declination change across Anatolia. The tectonic history preceding neotectonism in central-southern Anatolia is considered by Yılmaz (1993) to fall into five phases comprising (i) ophiolite emplacement (Cretaceous), (ii) initial collision between the metamorphic massifs and the Taurides (Eocene), (iii) initiation of suturing between the northerly-sited tectonic mosaic and the Arabian Plate (Early Miocene), (iv) closure of a remnant basin in south east Anatolia in post-Late Eocene times (e.g. Yiğitbaş and Yılmaz, 1996 ) and (v) final elimination of remnant seas during mid-Miocene times. Following this history of successive collision-related events, the beginning of the post-collisional neotectonic domain is recognised when compressional stresses were transferred laterally for the first time because this is a novel event compared with the earlier events, and specifically defined by development of the NAFZ and EAFZ ( Şengör, 1979 ). Transform initiation was contemporaneous with growth of the Karlıova Basin where the two fault systems intersect and the oldest infill is estimated from stratigraphic data to be latest Miocene-Early Pliocene ( Şaroğlu and Yılmaz, 1986 ). Within the long and complex history of Alpine orogenesis in this region it is difficult to decompose the observed cumulative palaeomagnetic rotations into a temporal history, although significant inferences can be made from comparison of results from units predating and post-dating collision, and from comparison of results from rock units of different ages such as in the Sivas Basin where a protracted history of volcanic activity is recorded. This latter basin lying immediately to the north of the KVC and YVC comprises a thick sequence of sediments and volcanics laid down on ophiolite sheets emplaced during late stages of the palaeotectonic history on the Akdağmadeni metamorphic complex; this is considered to be essentially a post-collisional tectonic element ( Yılmaz, 1994; Temiz, 1996 ) with deformation dominated by N–S to NW–SE compression. The infilling succession is undeformed in the central part of the basin near the town of Sivas but becomes progressively incorporated into southward migrating folds and thrusts towards the north (including the CATB; Fig. 1 ) and in northward migrating thrusts towards the south ( Guezou et al., 1996 ). Extensive palaeomagnetic studies in this region by Gürsoy et al. (1997) , Platzman et al. (1998) and Kissel et al. (2003) identify CCW rotations between 20 and 30° in Miocene and younger units. Comparable rotations are observed in Miocene, Pliocene and Quaternary units ( Gürsoy et al., 1997; Piper et al., 2006 ) implying that rotation has been confined to a late phase in the post-collisional history and accompanied the tectonic escape since establishment of the NAFZ and EAFZ. Results 32 and 36 in Table 6 incorporate the central part of the Sivas Basin between the CATB and the northward-directed Sivas Thrust and they identify CCW rotation in reversed declinations between 148° and 158° whilst results 33–35 embrace neotectonic units in blocks emplaced successively to the southeast between the Sivas Thrust, the Deliler Fault, the Sarız-Kangal Fault, and south of this latter fault system, respectively (see Fig. 3 of Gürsoy et al., 1997 ). The implied rotations recorded in Miocene age units are all similar, and statistically identical, to the results of this study where the YVC has an equivalent reversed polarity declination of 155° and the KVC 159°. Hence we find consistent CCW rotation of ∼25° since mid-Miocene times in a block extending from the CATB southwards for 150–200 km at least to the Malatya Fault Zone (MFZ see Figs. 1 and 8 ) and within 50–80 km of the East Anatolian Transform. Thus a differential rotation of ∼20° relative to the Arabian Plate is evidently focussed between the MFZ-OFZ and the EAFZ. This zone incorporates a number of ENE-WSW faults with left-lateral motions dissecting the Malatya Basin ( Kaymakçı et al., 2006 ), although since these are substantially straight, motions along them cannot apparently explain CCW rotation of the crust to the northwest. The arcuate form of the system incorporating the MFZ and OFZ is also unlikely to be able to accommodate all the rotation because the former fault seems to cut the YVC without major offset ( Fig. 2 ). At its north western border the rotation of this eastern sector of the Anatolides may have been focussed on the arcuate Central Anatolian Fault Zone (CAFZ, Figs. 1 and 8 ) interpreted by Koçyiğit and Beyhan (1998) as a left lateral intracontinental transform. This fault zone currently exhibits lower levels of seismic activity than the EAFZ ( Demirtaş and Yılmaz, 1996 ) but may have been a zone of differential CCW rotation during part of the post-collisional history. This point is reinforced by the observation that rotations on the west side from Cappadocia to Erenlerdağı are small CCW or insignificant ( Fig. 8 ), whereas on the east side declinations resolved from Miocene sedimentary succession by Krijgsman et al. (1996, results 37–45 in Table 6 ) identify strong CCW rotations. The latter rotations (which are clearly subject to an inclination-shallowing signature) are larger than the results from volcanics noted above from regions further to the east from a number of independent studies (32–36) and results of the present investigation ( Fig. 8 ). The larger CCW rotations observed in the sediments by Krijgsman et al. (1996) probably record an episode of intra-Lower-Middle Miocene rotation because they are significantly older (Late Oliocene-Early Miocene) than the volcanics where the bulk of the radiometric evidence indicates Middle Miocene and younger ages. Jaffey and Robertson (2001) found that Plio-Quaternary motion across the CAFZ has been mainly extensional and suggest that progressive CCW rotation in this sector has rotated the fault zone away from an optimal orientation for accommodating strike-slip motions during tectonic escape. This interpretation predicts that left-lateral motion around this arcuate fault system has been responsible for the differential rotation here but has now essentially ceased; it has presumably been transferred to the southeast to become focussed along the EAFZ and perhaps to the north on the side splays from the NAFZ. Thus the results of this study amplify a systematic pattern of CCW rotation beyond the north western tip of the Arabian Indenter to show that a uniform post-Miocene rotation has influenced a large segment of Eastern Anatolia extending from the CATB in the north to close to the EAFZ in the south. Substantial differential rotation is recognised near the NW margin of this block across the CAFZ and across the SE margin near the EAFZ. Although the rotations cannot be decomposed within the post-mid-Miocene history of the directions resolved from the KVC and YVC, the similarity of rotations in pre- and post-collisional units, and comparisons within the Sivas Basin suggest that they have been concentrated within the latter part of the post-collisional history. 9 Conclusions • The Kepezdağ and Yamadağ volcanic complexes were emplaced during a relatively brief interval of mid-Miocene times and were the products of major volcanic activity, probably during extensional deformation initiating the Malatya Basin. • Palaeomagnetic study recognises multiple reversals within the volcanic successions. Correlation with the GPTS indicates that the bulk of the volcanism occurred between 13 and 15 Ma, a range considerably shorter than the distribution of radiometric ages from these two complexes. • The mean palaeomagnetic directions define CCW rotation of the block incorporating the Yamadağ volcanic complex of 28.6 ± 4.9° with respect to the African Plate and 29.3 ± 5.2° with respect to Eurasia; since Arabia has also been rotating CCW since mid-Miocene times this represents a smaller CCW rotation of 18.2 ± 6.0° with respect to the adjoining Arabian Indenter. The smaller collection from the KVC defines a comparable CCW rotation of 25.3 ± 11.8° with respect to the African Plate and 26.1 ± 11.8° with respect to Eurasia. • Although the absence of suitable younger rocks does not permit temporal control over the rotation determined here, comparison with blocks to the north and west suggests that it is young and concentrated within the latter part of the Neotectonic Era as the transform faults became established and Anatolian terranes began extruding to the west by tectonic escape. • Comparable CCW rotation is observed across a broad crustal unit extending ∼200 km from the Central Anatolian Thrust Belt in the north to within 80 km of the East Anatolian Transform in the south east. The differential rotation between this division and the Arabia therefore appears to have been concentrated within a relatively narrow zone south of the MFZ and OFZ. Acknowledgements We are grateful to TUBITAK (Grant YDABAG 101Y023 and ÇAYDAG 104Y262) for supporting the field and laboratory studies. Erasmus and the British Council are thanked for supporting academic links between the Geomagnetism Laboratory of the Department of Earth and Ocean Sciences at Liverpool and the Department of Geology, Cumhuriyet University, Sivas. We are grateful to Douwe van Hinsbergen and an anonymous reviewer for comments which materially helped to improve the manuscript. References Alpaslan et al., 2004 M. Alpaslan H. Yılmaz A. Temel Geochemistry of postcollision Pliocene-Quaternary Karasar basalt (Divriği-Sivas, Eastern Turkey): evidence for partial melting processes Geol. Carpath. 55 2004 487 500 Alpaslan, 2007 M. Alpaslan Early to Middle Miocene intra-continental basaltic volcanism in the northern part of the Arabian plate, SE Anatolia, Turkey: geochemistry and petrogenesis Geol. Mag. 144 5 2007 867 882 Akkuş, 1971 M.F. Akkuş Darende-Balaban Havzasının (Malatya ESE, Anadolu) jeolojik ve stratigrafik incelenmesi MTA dergisi 76 1971 1 60 Arger et al., 2000 J. Arger J.G. Mitchell R.W.C. Westaway Neogene and quaternary volcanism of southeastern Turkey E. Bozkurt J.A. Winchester J.D.A. Piper Tectonics and Magmatism in Turkey and its Surrounding Area 2000 Geological Society London 459 487 Special Publications 173 Barka and Reilinger, 1997 A. Barka R. Reilinger Active tectonics of the Eastern Mediterranean region: deduced from GPS, neotectonic and seismicity data Annali Di Geofisica 40 1997 587 610 Beck, 1980 M.E. Beck Palaeomagnetic record of plate margin processes along the western edge of North America J. Geophys. Res. 87 1980 7115 7131 Beck et al., 2001 M.E. Beck R.F. Burmester P. Kondopoulou A. Atzemoglou The palaeomagnetism of Lesbos, NE Aegean, and the eastern Mediterranean inclination anomaly Geophys. J. Int. 145 2001 233 245 Besse and Courtillot, 2002 J. Besse V. Courtillot Apparent and true polar wander and the geometry of the geomagnetic field over the last 200 Myr J. Geophys. Res. 107 2002 B11, 2300, doi:10.1029/20000JB000050 Biggin et al., 2008 A. Biggin D.J.J. van Hinsbergen C.G. Langereis G.B. Straathof M.H. Deenen Geomagnetic secular variation in the Cretaceous Normal Superchron and in the Jurassic Phys. Earth Planet. Inter. 169 2008 3 19 Boztuğ et al., 2007 D. Boztuğ Y. Harlavan G.B. Arehart M. Satır N. Avcı K–Ar age, whole-rock and isotope geochemistry of A-type granitoids in the Divriği–Sivas region, eastern-central Anatolia, Turkey Lithos 97 2007 193 218 Büyüksaraç, 2007 A. Büyüksaraç Investigation into the regional wrench tectonics of inner East Anatolia (Turkey) using potential field data Phys. Earth Planet. Inter. 160 2007 86 95 Cande and Kent, 1995 S.C. Cande D.V. Kent Revised calibration of the geomagnetic polarity timescale for the Late Cretaceous and Cenozoic J. Geophys. Res. 100 1995 6093 6095 Calderone and Butler, 1991 G.J. Calderone R.F. Butler The effects of noise due to random undetected tilts and palaeosecular variation on regional palaeomagnetic directions J. Geophys. Res. 96 1991 3973 3977 Cogné et al., 1999 J.P. Cogné N. Halim Y. Chen V. Courtillot Resolving the problem of shallow magnetisations of Tertiary age in Asia: insights from palaeomagnetic data from the Qiangtang, Kunlun and Qaidam blocks (Tibet, China), and a new hypothesis J. Geophys. Res. 104 1999 17715 17734 Cox, 1969 A. Cox Confidence limits for the precision parameter κ Geophys. J. R. Astr. Soc. 17 1969 545 550 Demarest, 1983 H.H. Demarest Error analysis for the determination of tectonic rotation from palaeomagnetic data J. Geophys. Res. 88 1983 4321 4328 Demirtaş and Yılmaz, 1996 R. Demirtaş R. Yılmaz Seismotectonics of Turkey 1996 Ministry of Public Works and Settlement Ankara Ekici, 2003 Ekici, T., 2003. Petrology of the Neogene volcanics along the Malatya fault between Arguvan and Arapkir (Malatya). Ph.D. Thesis. Çukurova University, 121 pp. (unpublished). Ekici et al., 2005 Ekici, T., Alpaslan, M., Parlak, O., Uçurum, A., 2005. Çarpışmayla ilişkili Orta Miyosen yaşlı Yamadağı (Doğu Anadolu) kalkalkalin volkanizmasının jeokimyası, 58.Türkiye Jeoloji Kurultayı Bildiri özleri, s.246 MTA Kültür sitesi, Ankara. Ekici et al., 2007 T. Ekici M. Alpaslan O. Parlak A. Temel Geochemistry of the Pliocene basalts erupted along the Malatya-Ovacık Fault Zone (MOFZ), eastern Anatolia, Turkey: implications for source characteristics and partial melting processes Chemie Der Erde-Geochem. 67 2007 201 212 Ekici et al., 2009 T. Ekici M. Alpaslan O. Parlak A. Uçurum Geochemistry of the Middle Miocene collision-related Yamadağ calc-alkaline volcanics Turk. J. Earth Sci. 18 2009 511 528 Fisher, 1953 Sir R.A. Fisher Dispersion on a sphere Proc. R. Soc. A217 1953 295 305 Gilder et al., 2001 S. Gilder Y. Chen S. Sen Oligo-Miocene magnetostatigraphy and rock magnetism of the Xishuigou section, Subei (Gansu Province, western China) and implications for shallow inclinations in central Asia J. Geophys. Res. 106 30 2001 505–530, 521 Guezou et al., 1996 J.C. Guezou H. Temiz A. Poisson H. Gürsoy Tectonics of the Sivas Basin. The Neogene record of the Anatolian accretion along the Inner Tauric Suture Int. Geol. Rev. 38 1996 901 925 Gürbüz and Gül, 2005 K. Gürbüz M. Gül Evolution of and factors controlling Eocene sedimentation in the Darende-Balaban Basin, Malatya (Eastern Turkey) Turk. J. Earth Sci. 14 2005 311 335 Gürsoy et al., 1997 H. Gürsoy J.D.A. Piper O. Tatar H. Temiz A palaeomagnetic study of the Sivas Basin, Central Turkey: crustal deformation during lateral escape of the Anatolian Block Tectonophysics 271 1997 89 106 Gürsoy et al., 1998 H. Gürsoy J.D.A. Piper O. Tatar L. Mesci Palaeomagnetic study of the Karaman and Karapınar volcanic complexes, central Turkey: neotectonic rotation in the south-central sector of the Anatolian Block Tectonophysics 29 1998 191 211 Gürsoy et al., 2003 H. Gürsoy O. Tatar J.D.A. Piper A. Heimann L. Mesci Neotectonic deformation in the Gulf of Iskenderun, Southern Turkey, deduced from paleomagnetic study of the Ceyhan-Osmaniye Volcanics Tectonics 22 6 2003 10.1029/2003TC001524 Gürsoy et al., 2009 H. Gürsoy O. Tatar J.D.A. Piper A. Heimann F. Koçbulut B.L. Mesci Palaeomagnetic study of Tertiary volcanic domains in Southern Turkey and Neogene anticlockwise rotation of the Arabian Plate Tectonophysics 465 2009 114 127 Jaffey and Robertson, 2001 N. Jaffey A.H.F. Robertson New sedimentological and structural data from the Ecemiş Fault Zone, Southern Turkey: implications for its timing and offset and the Cenozoic tectonic escape of Anatolia J. Geol. Soc. Lond. 158 2001 367 378 Johnson et al., 2008 C.L. Johnson C.G. Constable L. Tauxe R. Barendregt L.L. Brown R.S. Coe P. Layer V. Mejia N.D. Opdyke B.S. Singer H. Staudigel D.B. Stone Recent investigations of the 0–5 Ma geomagnetic field recorded by lava flows Geochem. Geophys. Geosyst. 9 2008 Q04032 doi:10.1029.2007GC001696 Kaymakçı et al., 2006 N. Kaymakçı M. İnceöz P. Ertepınar 3D-Architecture and Neogene evolution of the Malatya Basin: inferences for the kinematics of the Malatya and Ovacık fault zones, Turkish J. Earth Sci. 15 2006 123 154 Keskin, 2003 M. Keskin Magma generation by slab steepening and breakoff beneath a subduction-accretion complex: an alternative model for collision-related volcanism in Eastern Anatolia, Turkey Geophys. Res. Lett. 30 24 2003 8046 10.1029/2003GL018019 Keskin et al., 1998 M. Keskin J. Pearce J.G. Mitchell Volcano-stratigraphy and geochemistry of collision-related volcanism on the Erzurum–Kars Plateau, northeastern Turkey J. Volcanol. Geotherm. Res. 85 1998 355 404 Kissel et al., 2003 C. Kissel C. Laj A. Poisson N. Görür Palaeomagnetic reconstruction of the Cenozoic evolution of the Eastern Mediterranean Tectonophysics 362 2003 199 217 Koçyiğit and Beyhan, 1998 A. Koçyiğit A. Beyhan A new intracontinental transcurrent structure: the Central Anatolian Fault Zone, Turkey Tectonophysics 284 1998 317 336 Krijgsman et al., 1996 W. Krijgsman C.E. Duermeijer C.G. Langeris H. Bruijn G. Sarac P.A.M. Andriessen Magnetic polarity stratigraphy of late Oligocene to middle Miocene mammal-bearing continental deposits in central Anatolia (Turkey) Newsl. Stratigr. 34 1 1996 13 29 Kürüm et al., 2004 S. Kürüm A. Önal D. Boztuğ M. Arslan T. Spell 40Ar/39Ar age determination and geochemistry of widespread Yamadag volcanics, in Arapkir, N. Malatya region, Eastern Anatolia, Turkey A.A. Chatzipetros S.B. Pavlides Proceedings 5th International Symposium on Eastern Mediterranean Geology, Abstracts, vol. 3 Thesssaloniki, Greece 2004 1295 1298 Leo et al., 1974 G.W. Leo R.F. Marvin H.H. Mehnert Geologic framework of the Kuluncak-Sofular area, east central Turkey, and K–Ar ages of igneous rocks Geol. Soc. Am. Bull. 85 1974 1785 1788 MacDonald, 1980 W.D. MacDonald Net tectonic rotation, apparent tectonic rotation and the structural tilt correction in palaeomagnetic studies J. Geophys. Res. 85 1980 3659 3669 McClusky et al., 2000 S. McClusky S. Balassanian A. Barka C. Demir S. Ergintav I. Georgiev O. Gürkan M. Hamburger K. Hurst H. Kahle K. Kastens G. Kekelidze R. King V. Kotzev O. Lenk S. Mahmoud A. Mishin M. Nadariya A. Ouzounis D. Paradissis Y. Peter M. Prilepin R. Reilinger I. Şanlı H. Seeger A. Tealeb M.N. Toksöz G. Veis Global positioning system constraints on plate kinematics and dynamics in the eastern Mediterranean and Caucasus J. Geophys. Res. 105 2000 5695 5719 McFadden and McElhinny, 1990 P.L. McFadden M.W. McElhinny Classification of the reversal test in Palaeomagnetism Geophys. J. Int. 103 1990 725 729 McFadden et al., 1991 P.L. McFadden R.T. Merrill M.W. McElhinny S. Lee Reversals of the Earth's magnetic field and temporal variations of the Dynamo Families J. Geophys. Res. 96 1991 3023 3933 Pearce et al., 1990 J.A. Pearce J.F. Bender S.E. De Long W.S.F. Kidd P.J. Low Y. Güner F. Şaroğlu¸ Y. Yılmaz S. Moorbath J.G. Mitchell Genesis of collision volcanism in eastern Anatolia, Turkey J. Volcanol. Geotherm. Res. 44 1990 189 229 Piper et al., 2006 J.D.A. Piper O. Tatar H. Gürsoy F. Kocbulut B.L. Mesci Palaeomagnetic analysis of Neotectonic deformation in the Anatolian accretionary collage, Turkey Y. Dilek S. Pavlides Post-Collisional Tectonics and Magmatism in the Eastern Mediterranean Region 2006 Geol. Soc. Amer. 417 440 Spec. Paper 409 Piper et al., 2010 J.D.A. Piper H. Gürsoy O. Tatar M.E. Beck A. Rao F. Koçbulut B.L. Mesci Distributed Neotectonic deformation in the Anatolides of Turkey: a palaeomagnetic analysis Tectonophysics 488 2010 31 50 Platzman et al., 1998 E.S. Platzman C. Tapırdamaz M. Sanver Neogene anticlockwise rotation of central Anatolia (Turkey): preliminary palaeomagnetic and geochronological results Tectonophysics 299 1998 175 189 Poisson et al., 1996 A. Poisson J.C. Guezou A. Öztürk S. İnan H. Temiz H. Gürsoy K.Ş. Kavak S. Özden Tectonic setting and evolution of the Sivas Basin, Central Anatolia, Turkey Int. Geol. Rev. 38 1996 838 853 Robertson et al., 1996 A.H.F. Robertson J.E. Dixon S. Brown A. Collins A. Morris E. Pickett I. Sharp T. Ustaömer Alternative tectonic models for the late Palaeozoic—early Tertiary development of Tethys in the Eastern Mediterranean region A. Morris D.H. Tarling Palaeomagnetism and Tectonics of the Mediterranean Region vol. 105 1996 Geol. Soc. London 239 263 Spec. Publ. Şaroğlu and Yılmaz, 1986 F. Şaroğlu Y. Yılmaz Geological evolution and basin modelling in East Anatolia during neotectonic period (in Turkish) MTA Bull. 107 1986 73 94 Şengör, 1979 A.M.C. Şengör The North Anatolian transform fault: its age, offset and tectonic significance J. Geol. Soc. Lond. 136 1979 269 282 Şengör and Yılmaz, 1981 A.M.C. Şengör Y. Yılmaz Tethyan evolution of Turkey: a plate tectonic approach Tectonophysics 75 1981 181 241 Şengör et al., 2003 A.M.C. Şengör S. Özeren E. Zor T. Genç East Anatolian high plateau as a mantle-supported, N-S shortened domal structure Geophys. Res. Lett. 30 24 2003 10.1029/2003GL017858 Tatar et al., 2002 O. Tatar H. Gürsoy J.D.A. Piper Differential neotectonic rotations in Anatolia and the Tauride Arc: palaeomagnetic investigation of the Erenlerdağ Volcanic Complex and Isparta volcanic district, south-central Turkey J. Geol. Soc. Lond. 159 2002 281 294 Tatar et al., 2004 O. Tatar J.D.A. Piper H. Gürsoy A. Heimann F. Koçbulut Neotectonic deformation in the transition zone between the Dead Sea Transform and the east Anatolian fault Zone, Southern Turkey: a palaeomagnetic study of the Karasu Rift Volcanism Tectonophysics 385 2004 17 43 Tauxe and Kent, 2004 L. Tauxe D. Kent A simplified statistical model for the geomagnetic field and the detection of shallow bias in palaeomagnetic inclinations: was the ancient magnetic field dipolar? Am. Geophys. Union 2004 1 15 Temiz, 1996 H. Temiz Tectonostratigraphy and thrust tectonics of the central and eastern parts of the Sivas Tertiary Basin, Turkey Int. Geol. Rev. 38 1996 957 971 TGM, 2002 (Turkish Geological Map) TGM 1/500.000 Scale Sivas Sheet 2002 MTA Publications Ankara Uçurum, 1992 Uçurum, A., 1992. The geology of northern Hasançelebi and mineralogic petrographic and geochemical study of the volcanics, NW Malatya, Turkey. Cumhuriyet University. MSc Thesis. 125 pp. (unpublished, in Turkish, abstract in English). Valet and Soler, 1999 J.-P. Valet V. Soler Magnetic anomalies of lava fields in the Canary Islands. Possible consequences for palaeomagnetic records Phys. Earth Planet. Inter. 115 1999 109 118 Vandamme, 1994 D. Vandamme A new method to determine palaeosecular variation Phys. Earth Planet. Inter. 85 1994 131 142 Wilson, 1971 R.L. Wilson Dipole offset—the time average palaeomagnetic field over the past 25 million years Geophys. J. R. Astr. Soc. 22 1971 481 504 Yalçın et al., 1998 H. Yalçın M.N. Gündoğdu A. Gourgaud P. Vidal A. Uçurum Geochemical characteristics of Yamadağı volcanics in central east Anatolia: an example from collision-zone volcanism J. Volcanol. Geotherm. Res. 85 1998 303 326 Yiğitbaş and Yılmaz, 1996 E. Yiğitbaş Y. Yılmaz Post-Late Cretaceous strike-slip tectonics and its implications for the southeast Anatolian orogen, Turkey Int. Geol. Rev. 38 1996 818 831 Yılmaz, 1994 A. Yılmaz An example of a post-collisional trough: Sivas Basin, Turkey Proceedings of the 10th Petroleum Congress of Turkey Ankara 1994 21 23 Yılmaz and Özel, 2007 H. Yılmaz S. Özel Crustal structure of the eastern part of central Anatolia (Turkey) Turk. J. Earth Sci. 18 2007 169 185 Yılmaz, 1990 Y. Yılmaz Comparison of young volcanic associations of western and eastern Anatolia formed under a compressional regime: a review J. Volcanol. Geotherm. Res. 44 1990 69 87 Yılmaz, 1993 Y. Yılmaz New evidence and model on the evolution of the southeast Anatolian Orogen Geol. Soc. Am. Bull. 105 1993 251 271 Yılmaz et al., 1987 Y. Yılmaz F. Şaroğlu Y. Güner Initiation of the neomagmatism in East Anatolia Tectonophysics 134 1987 177 199