Introduction

The collision between Arabia and Eurasia is considered to be the beginning of the neotectonic regime in and around Turkey by some researchers1,2. Şengör2 stated that the northward movement of the Arabian plate and westward movement of the Anatolian plate formed four neotectonic regions: The East Anatolian Contractional Province, the North Anatolian Province, the Central Anatolian Ova Province and the West Anatolian Extensional Province. In the literature, there are different views regarding the age of this collision, including the late Cretaceous3,4,5, the late Eocene–Oligocene6,7,8, and the Miocene9,10,11,12,13,14,15,16.

The collision between Arabian and Eurasian plates led to a large plateau formation reaching about 2 km in elevation in Eastern Anatolia1. During Miocene, the Anatolian plate, between the North Anatolian Fault Zone (NAFZ) and the East Anatolian Fault Zone (EAFZ), started to move westward (Fig. 1a)10,17,18,19. Due to the compressional tectonic regime in the Eastern Anatolia, east–west trending folds, thrusts, and strike-slip fault systems are formed, ongoing activity of which still can be observed from focal mechanism solutions of the earthquakes in the area (Fig. 1b)1,10,12,18,20,21,22,23,24,25.

Figure 1
figure 1

(a) Tectonic map of the Anatolian, Eurasian and Arabian plates, NAFZ: North Anatolian Fault Zone; EAFZ: East Anatolian Fault Zone; DSFZ: Dead Sea Fault Zone. (b) Seismicity of the Lake Van and its surroundings (M > 4.0 Earthquakes between 1900 and 2019 (earthquake epicenter data from Kandilli Observatory and Earthquake Research Institute (KOERI) and focal mechanism solutions of M > 5.8 earthquakes between 1976 and 2019 (retrieved from “Global Centroid Moment Tensor Catalog”,82,83). Maps were created by using the Generic Mapping Tools software, version 5.1.1. (https://www.soest.hawaii.edu/gmt/)84.

Full size image

East Anatolian Contractional Province consists ophiolitic melange and flysch15,26 that are overlain by Early and Late Miocene sediments. Most of these sediments are then covered by Lower-Mid Miocene lavas13,27,28,29. Many researchers argue that the volcanism that developed in Eastern Anatolia has started in the Late Miocene and continued until today27,28,30,31,32,33,34,35,36. Volcanic activity began soon after the region's uplift and produced volcanic material in many countries, including Turkey, Russia, Georgia, Azerbaijan, Armenia, and Iran. Collision-related volcanism in the Eastern Anatolia covered almost two-thirds of the region and formed volcanic products reaching up to 1 km in thickness29. Keskin28 related this volcanism and the rapid uplift of the region with the delamination of the subducted lithosphere under the Eastern Anatolia. The products of the post-Miocene volcanism are encountered throughout the entire region and among the volcanic centers: Erzurum Kars plateau (EKP)13,27, Nemrut Volcano36, Süphan Volcano37,38, Tendürek Volcano39,40, Karacadağ Volcano41 and Etrüsk Volcano42,43,44 have been studied in detail by many researchers.

In this paper, the results of new paleomagnetic and rock magnetic studies conducted in the north of the Lake Van, Eastern Anatolia are presented. There are only a few previous paleomagnetic studies available close to the research area. The study that was carried out on the Quaternary volcanics throughout Turkey by Sanver45 includes only one site close to our study area indicating a clockwise rotation. Another study Hisarlı et al.16 claimed that there exist at least five blocks bounded by left and right lateral strike-slip faults in the Eastern Anatolia, and noted that the Van Block (VB), rotated clockwise as a single rigid block. Gülyüz et al.46 studied Neogene sedimentary rocks in the southeast of Lake Van, and reported that these sedimentary rocks have a significant 25° clockwise rotations.

The objective of our study is to investigate the tectonic development and related tectonic deformations observed in the region using paleomagnetic samples collected at 86 sites in the north of the Lake Van. In this context, we sampled Late Miocene, Pliocene and Pleistocene volcanic rocks whose ages and rock types are known from the literature40,43,44,47.

Geological setting and paleomagnetic sampling

The geological evolution of the region is defined in four different periods20. The first period is designated by the generation of the Paleozoic–lower Mesozoic metamorphic rocks48,49,50,51,52,53,54. The second period is represented by the Upper Cretaceous ophiolitic accretionary complex50,55,56,57,58,59. The third period is depicted as sedimentary rocks deposited between Late Cretaceous and Miocene1,60,61. Volcanic rocks spread throughout the region and continental sediments ranging in age from Late Miocene to the present form the units of last geological period1,26,28,29,42,44,47.

The Eastern Anatolian plateau, one of the youngest and widest plateaus of the world, represents a suture zone at which the northern and southern branches of Neotethys come together9,10. The basement of the Eastern Anatolian plateau consists of micro-continents stacked together between Late Cretaceous–Early Tertiary and separated from each other by ophiolite belts and accretion complexes1,62. Five different tectonic blocks are recognized in Eastern Anatolia: the Eastern Rhodope-Pontide fragment, the Northwest Iranian fragment, the Bitlis-Pötürge unit, Autochthonous units of the Arabian continent and the Eastern Anatolian Accretionary Complex. Except for the Eastern Anatolian Accumulation Complex, all other tectonic blocks correspond to the microcontinents mentioned above29. The Eastern Anatolian Accretionary Complex (EAAC) represents the remnant of a huge subduction accretion complex located between the Rhodop–Pontide and the Bitlis–Pötürge micro-continent, formed within Late Cretaceous–Oligocene26.

The collision between the Eurasian and Arabian continents occurred in Serravallian (~ 13–11 Ma)9,10,29. Volcanic activity started right after the rapid block uplift of Eastern Anatolia and produced different volcanic products that spread throughout the region (Fig. 2)27,29,47,63,64. Volcanic activity first started around Erzurum-Kars plateau with calc-alkaline lavas in the north, then migrated to the south-southeast and became more alkaline28,29. This volcanic activity has produced a large volume of volcanic material that covers almost two-thirds of the area and in some places exceeds 1 km in thickness. Besides fissure eruptions in volcanic activity, there are also many volcanic centers (e.g. Ağrı, Nemrut and Tendürek Mountains).

Figure 2
figure 2

Geological map of the study area including paleomagnetic sample sites of our study. Map were re-arranged from85, 1/500.000 geological map, by using the CorelDRAW Graphics Suite (Education License), Version 2021, (https://www.coreldraw.com/en/).

Full size image

The first stage of volcanism in Middle Miocene produced Aladağ Volcanics, which are widespread to the north-northeast of Lake Van42,65,66. The next stage of volcanism started at the beginning of Late Miocene (~ 10 Ma). Located in the area between the Ilıca and Deliçay rivers, these rocks have a widely varying chemical composition from basalts and sparsely trachybasalts to dacites42. Pliocene volcanism started 5.8–3.7 Ma as trachytic ignimbrites and tuffs that cover a vast area northwest of the town of Erciş. The activity of the Etrüsk Volcano up to trachydacites, trachytes and trachyandesites is the last phase of Pliocene magmatic activity. The activity of this volcano covers a time period between 4.3 and 3.7 Ma42,43,44. Quaternary volcanism in the north of Lake Van, including Girekol Volcano and Karnıyarık cinder cone, started ~ 1 Ma and this young volcanism ended ~ 400 ka.

We collected paleomagnetic oriented samples between 2015 and 2017 from the volcanic rocks whose ages and geochemical compositions have already been known from radiometric aging and petrography methods42,43 to determine the tectonic evolution of the North of Lake Van. The orientations of the samples were determined using magnetic/sun compasses and samples were drilled using a portable oil-powered motorized drill with water-cooling, diamond-coated, non-magnetic drill bits. The paleomagnetic samples were collected around the northeast of Lake Van from Pleistocene volcanics (38 sites), Pliocene volcanics (82 sites), and Late Miocene volcanics (14 sites). To increase the number of samples at paleomagnetic sites and make them statistically more reliable, we have combined two or more very close sites from the same age and rock type to create sites with a larger sample count. Thus, we have 32 Pleistocene, 58 Pliocene and 10 Late Miocene sites (Fig. 2).

Laboratory studies

The paleomagnetic laboratory studies were performed at the “KANTEK Paleomagnetism Laboratory” which is a collaborative laboratory of Boğaziçi University and Istanbul Technical University (ITU). Cylindrical core samples were cut into standard paleomagnetic specimens (2246 in total) and were subjected to stepwise thermal demagnetization using “Magnetic Measurements MMTD-60” thermal demagnetizer. The thermal demagnetization is applied in increments of 25–50 °C up to a maximum temperature of 650 °C. Molspin spinner magnetometer was used to measure magnetization directions and intensities of the Natural Remanent Magnetization (NRM) after each step of thermal demagnetization. Principal component analysis67 and orthogonal vector diagrams68 were used to describe the Characteristic Remanent Magnetization (ChRM). The ChRM directions and their statistical parameters for each site were determined using standard Fisher statistical analysis with 45° cut-off69. Errors in declination (ΔDx) and inclination (ΔIx) were calculated from the A95 of the Virtual Geomagnetic Pole (VGP) distribution for all sites. The criteria defined by Deenen et al.70,71 indicate that the determined A95 value of the VGP distribution needs to be between the N-dependent values of A95min and A95max to represent the paleosecular variations (PSV) in the geomagnetic field. Remasoft 3.0 Paleomagnetic Data Browser and Analyzer software have been used to interpret the demagnetization diagrams. The reversal test and its classification developed by McFadden and McElhinny72 were used to determine whether the two distributions with the positive and negative polarity means have a common mean direction.

Rock magnetic studies (Isothermal remanent magnetization-IRM and high-temperature susceptibility-HTS) have been carried out to determine the magnetic minerals responsible for permanent magnetization and to determine the change with the coercive force in addition to paleomagnetism studies. All measurements of rock magnetic studies were done in Yılmaz İspir Paleomagnetism Laboratory at Istanbul University-Cerrahpaşa, Department of Geophysics.

HTS measurements have been carried out on 16 representative samples by heating in room conditions. The heating and cooling phases of the grinded sample between room temperature (24 °C) and 650 °C were done using the Bartington MS2 Susceptibility/Temperature System with Bartington MS2 susceptibility meter. IRM studies were performed on pilot samples from each rock type to detect the minerals responsible for the magnetization in the rock.

Analysis results

Rock magnetism studies

We made at least one HTS measurement for each rock type to determine the rocks’ magnetic properties and to identify its magnetic behavior at different temperatures. The analysis of HTS measurements performed in 16 samples of different age and rock types show three different types of behavior. The red curves given in Fig. 3 illustrate the heating phase and the blue curves indicate the cooling phase.

Figure 3
figure 3

High-temperature susceptibility curves (a,c,e,g) and normalized IRM acquisition curves (b,d,f,h) for representative samples. (normalized IRM acquisition curves were obtained from one sample of the site for which HTS was measured. e.g. haci3–haci-3a).

Full size image

Low Curie temperatures are observed between 150 and 250 °C in one of the sample groups (cay1, colp, san2, koz5, inc1), which is an indication of presence of titanomagnetite component in these samples.

Some of the samples show reversible behavior (atok, blk, cay1, inc1, koz5, ykoz5, ykr) with no remarkable difference between heating and cooling curves. On the other hand, the heating and cooling curves of some samples differ considerably, in relation to remarkable degree of alteration (ack, kad2, cay4, colp, yayla2). Heating curves depict a reduction of around 400 °C in two samples (ack, haci), which indicates the transformation to maghemite or the existence of Ti-rich titanomagnetite.

IRM measurements were applied to the samples taken from 33 sites representing each of the volcanic rocks located to the north of Lake Van. IRM curves in Fig. 3 indicate that magnetite is the responsible mineral for magnetization. Two different types of behavior are observed in the IRM acquisition curves. The first type of IRM curve is characterized by low to moderate coercivity phases, reaching saturation between 0.1 and 0.3 T. There are 21 sites (Arg, Bend, Colp, Inc1, Orn, etc.) where the saturation is observed in IRM curves and magnetite mineral is dominantly responsible for their magnetization. The second type of IRM curve is characterized by a rapid increase of magnetization in low fields at first (up to 1 T) and then a small increase without complete saturation at 1 T (Fig. 4). There are 12 sites (Ykoz, Ykr, Haci, Skr, Tprk4, etc.) whose IRM curves are classified as the second type, indicating that both magnetite and hematite are responsible for the magnetization. Hematite was not dominant in any of the samples.

Figure 4
figure 4

(a) Mean remanent magnetization directions of Pleistocene volcanic rocks. (b) Mean remanent magnetization directions of Pliocene volcanic rocks. (c) Mean remanent magnetization directions of Late Miocene volcanic rocks. Maps were created by using the Generic Mapping Tools software, version 5.1.1. (https://www.soest.hawaii.edu/gmt/)84.

Full size image

Paleomagnetic studies

In the study area, there are volcanics of different rock types in the Late Miocene–Pleistocene range. During the interpretation of the paleomagnetic data, we divided these volcanics into 3 different age groups as Late Miocene, Pliocene and Pleistocene. Late Miocene volcanism in the region is represented by products from Aladağlar and Meydan Mountains. These volcanics are located to the north of the Zilan Valley and Etrüsk Volcano within the vicinity of the Tendürek Mountain (N-NW of Erciş Village). Late Miocene-aged sites are generally located in the north-northeast of Erciş district. The Pliocene-aged volcanics, which are the most common rocks in the study area, are around Van, Erciş, Etrüsk Mountain and Muradiye. Pleistocene volcanics are located in the north of the Erciş district and the west of Etrüsk Mountain as Girekol, Yüksektepe and Karnıyarık volcanics. Pleistocene aged sites are also located in these regions.

Group mean directions

Pleistocene aged rocks in the north of Lake Van have been sampled from 32 paleomagnetic sites. Except for 11 statistically unreliable paleomagnetic sites which are excluded from evaluation, all sites have normal polarities (Table 1). Seven sites show suspicious directions that are diverging from the average distribution and sites with A95 values outside the envelope of A95min and A95max were not taken into evaluation. Eight of the 14 remaining sites with reliable magnetization were in paleo-horizon and no tilt correction was applied. The remaining six sites are tilt corrected and the results of all 14 sites are given in Table 1.

Table 1 Paleomagnetic results from Late miocene to pleistocene volcanic rocks.
Full size table

These 14 normal polarity sites in Pleistocene volcanics show a mean direction in geographic coordinates of D = 350.0° and I = 48.9° with statistical parameters k = 113.76 and α95 = 4.3° and in the stratigraphic coordinates D = 348.0° and I = 50.4° with statistical parameters N = 14, k = 166.85 and α95 = 3.5° (Table 1; Fig. 4a).

A total of 58 paleomagnetic sites have been sampled from Pliocene aged rocks in the north of Lake Van. Five sites are excluded from evaluation because of statistical unreliability (Table 1). A total of 11 sites showing suspicious directions deviating from the average distribution and sites with A95 values outside the envelope of A95min and A95max were not taken into evaluation.

Nine of the Pliocene-aged sites have normal polarity and the rest have reverse polarity. A positive reversal test is obtained with classification of “B”72 and the k ratio is 1.3. The normal polarity sites (N = 9) give a clockwise rotation (D = 11.7°, I = 55.8°, R = 8.7 and k = 30.4). Reverse polarity sites (N = 33) give a counterclockwise rotation (D = 178.4°, I = − 58.2°, R = 32.1 and k = 36.4). The observed angular difference (γ = 7.6°) is smaller than the critical angular difference (γc = 9.2°) indicating that the secondary overprints have been removed in the Pliocene paleomagnetic data.

Pliocene volcanic rocks are geographically close to each other and split into four different groups during evaluation: The south of the Erciş Fault (Pliocene_SW), the north of the Erciş Fault (Pliocene_NE), around Muradiye (Pliocene_EM), and the north of the Van (Pliocene_EV) (Table 1; Fig. 4b).

For the Pliocene sites in the East of Van (Pliocene_EV), mean directions of the group are found to be D = 24.4° and I = 63.7° with statistical parameters N = 3, k = 73.19 and α95 = 14.5°, which are updated as D = 25.8° and I = 65.2° with statistical parameters N = 3, k = 133.13 and α95 = 10.7° after tilt correction (Table 1; Fig. 4b). All of three sites have normal polarity. For the Pliocene sites in the South of the Erciş Fault (Pliocene_SW), group mean direction is calculated as D = 349.3° and I = 61.2° with statistical parameters N = 17, k = 70.13 and α95 = 4.3° before tilt correction, D = 347° and I = 61.0° with statistical parameters N = 17, k = 65.30 and α95 = 4.4° after tilt correction. Two sites have normal and 15 sites have reverse polarity (Table 1; Fig. 4b).

Pliocene sites in the North of the Erciş Fault (Pliocene_NE) show mean direction of the group as D = 3.8° and I = 51.8° with statistical parameters N = 16, k = 29.88 and α95 = 6.9°, and D = 0.9° and I = 56.2° with statistical parameters N = 16, k = 34.21 and α95 = 6.4° after tilt correction (Table 1; Fig. 4b). Four sites have normal and 12 sites have reverse polarity.

The group near the Muradiye is named Pliocene_EM and has six sites. The mean direction of this group is calculated as D = 5.9° and I = 43.8° with statistical parameters N = 6, k = 38.84 and α95 = 10.9° before tilt correction, D = 8.3° and I = 46.2° with statistical parameters N = 6, k = 38.84 and α95 = 10.9° after tilt correction. One site has normal and five sites have reverse polarity (Table 1; Fig. 4b).

Late Miocene aged rocks in the north of Lake Van were sampled from 10 paleomagnetic sites. There is not any statistically unreliable paleomagnetic sites (Table 1). A total of four sites showing suspicious directions diverging from the average distribution and sites with A95 values outside the envelope of A95min and A95max were not taken into evaluation.

Late Miocene rocks (N = 6 sites) yield D = 2.6°, I = 46.5° with statistical parameters k = 19.49, α95 = 15.6° in in-situ coordinates, and D = 359.7°, I = 53.9° with statistical parameters k = 27.06 and α95 = 13.1° after tilt correction from five reverse and one normal polarity sites (Table 1; Fig. 4c). Figure 5 shows stereographic projections and statistical parameters of different aged paleomagnetic groups after tilt corrections.

Figure 5
figure 5

Paleomagnetic mean directions of the north of Lake Van in different time intervals (Pleistocene, Pliocene and Late Miocene). Red stars represent the current magnetic field direction and inclination angle (D/I = 0°/58°) in the latitude where the study area is located.

Full size image

Paleosecular variation

Paleosecular variation (PSV) should be averaged in paleomagnetic studies, so that the paleomagnetic directions show only tectonic movement70. PSV should be considered unreliable if A95 values are above or below the A95min and A95max limits70. In this study VGPs were calculated from paleomagnetic rotations for each site and group. A95 values for all of the groups and the most of the individual samples fall within the A95min/A95max confidence envelope; samples otherwise were not included within the evaluation. Therefore, it is plausible to consider that PSV is adequately averaged in our paleomagnetic data set.

A95 values of the whole group mean directions of Late Miocene–Pleistocene aged volcanic rocks are within the required A95min–A95max envelope. Equal area projections of the ChRM and VGP directions for all groups are shown in Fig. 6.

Figure 6
figure 6

Equal area projections of the ChRM and VGP directions following Deenen’s criteria70, respectively. Red circles indicate α95 cone of confidence and solid red dots indicate rejected results. All directions are converted to normal polarity.

Full size image

Discussion

The Pleistocene pole position is calculated as 77.4° N, 278.3° E (dp = 4.7, dm = 3.2, α95 = 3.5°) showing counterclockwise rotation of R ± ΔR = 13.4° ± 3.8° when compared with the expected stable Eurasia reference pole (λref, /Φref = − 88.5°/353.9°, α95 = 1.9°) of73 using PMGSC (version 4.2) software74 (Fig. 7a). Examining the rotations in the southern part of the Erciş Fault (group Pliocene_SW) implies a counterclockwise rotation: R ± ΔR = 14.5° ± 6.1° (Fig. 7b). On the other hand, no significant rotation (R ± ΔR = 0.6° ± 7.4°) is observed on the northern side of Erciş Fault (group Pliocene_NE) (Fig. 7b). Also, the remarkable clockwise rotation R ± ΔR = 24.4° ± 17.0° to the north of Van (group Pliocene_EV) and clockwise rotation R ± ΔR = 6.9° ± 9.4° of the sites near Muradiye indicate that the prevailing tectonic movement of the region is clockwise (Fig. 7b). In summary, Pliocene paleomagnetic results reveal that there are a lot of remarkable rotation differences around the Erciş Fault, most of which are caused by the Erciş Fault itself. The results from Late Miocene volcanic rocks indicate that there is almost no significant rotation (R ± ΔR = 1.8° ± 13.7°) in the region (Fig. 7c). In Table 2, net rotation amounts of other ages and location groups are given respectively. These values were used in the tectonic interpretation of paleomagnetic data obtained in this study. The difference between the observed poles (λobs, ϕobs) and the reference poles and expected tectonic rotations (R) were computed using the pole-space method of Beck75 and the 95% confidence limits (ΔR) were determined after Demarest76.

Figure 7
figure 7

Rotations of blocks in the study area. (a) Rotation of the Pleistocene sites, (b) Rotations of Pliocene sites and, (c) Rotation of Late Miocene sites. Maps were created by using the Generic Mapping Tools software, version 5.1.1. (https://www.soest.hawaii.edu/gmt/)84.

Full size image
Table 2 Group mean ChRM directions (D, I, α95) after tectonic correction and expected tectonic rotations (R) with respect to the stable Eurasia reference poles73.
Full size table

Hisarlı et al.16 named the VB as a region that includes the Lake Van and its surroundings, from the Karlıova Triple Junction to the east. According to the paleomagnetic data obtained from this study, VB must have been rotating clockwise starting from Late Miocene.

Our study area does not coincide with the area of the other relevant studies16,45,46 and covers a relatively narrower area compared to them, allowing the investigation of smaller “micro-block” movements instead of mono-block movements. Our study also includes more paleomagnetic samples and distribution in the north of Lake Van and results show both clockwise and counterclockwise rotations in the area which can be interpreted that the Van Block is divided into micro-blocks that have different rotation directions.

If the lineament (i.e. fault) between the blocks of same aged rocks determined by the paleomagnetic rotations coincide with the lineament observed in the active fault map, it can be said that the slips along the faults by the current earthquakes are the continuations of the past tectonic movements.

According to the paleomagnetic rotations of the Pleistocene volcanic rocks, it can be concluded that the whole region rotated counterclockwise (~ 13.4° ± 3.8°) and moved as a mono-block. Selçuk et al.77 claimed that the slip rate of the Çaldıran Fault was ~ 3 mm/year (for 290,000 years) and that the maximum slip was ~ 900 m. The rotation resulting from such a small slip cannot be measured with paleomagnetic data. For this reason, it is not relevant to expect a clockwise rotation in the area between the Çaldıran Fault and the Erciş Fault, two of which are dextral strike-slip faults. Therefore, it is clear from the obtained data that the region has rotated counterclockwise (~ 13.4° ± 3.8°) since Pleistocene.

Rotations from Pliocene rocks are given in Fig. 7.a. To determine the rotations in the Pliocene–Pleistocene time interval, counterclockwise rotation of the Pleistocene rocks (~ 13.4° ± 3.8°) is needed to be rotated clockwise. After this period, Pliocene_NE rocks were rotated ~ 13° clockwise and Late Miocene rocks were rotated ~ 11° clockwise in the Late Miocene–Pleistocene period. It is observed that the Çakırbey Fault is formed before the Erciş Fault (Pliocene) and causes a rotation of ~ 37.8° towards the east of the Erciş Fault. This suggests that the region was subjected to an active rotation in the counterclockwise direction after Late Miocene until the Quaternary.

Emre et al.78,79 does not mark an active fault in the northwest of Lake Erçek. However, according to our results, we suggest that the Çakırbey Fault, which is roughly oriented northeast-southwest towards Erciş Fault, must be active (Fig. 7) and thus different rotations may have developed on both sides of this fault.

Copley and Jackson80, investigating the active tectonics of the Turkish–Iranian plateau, examined the right lateral strike-slip faults and claimed that the parallel striking Erciş Fault and Çaldıran Fault are 11 km and 1.3 km offset, respectively. Furthermore, they stated that the combined slip rate for these faults was about 8 mm/year and that a time of about 1.5 Ma would be necessary to generate the combined 12.3 km shift at this speed. Authors stated that the removal of 11 km of dextral movement along the Erciş Fault restores the mountain-fronts (Fig. 8a) and the edge of volcanic rocks (Fig. 8b).

Figure 8
figure 8

Restoration of 11 km movement on Erciş Fault. (a) Present state of the area after 11 kms of tectonic shift. (b) The state of the area before the tectonic shift, restored by reversing the process visually. Red dotted line represents Erciş Fault and the dashed black lines represent mountain fronts marked in80. Maps were created by using the Generic Mapping Tools software, version 5.1.1. (https://www.soest.hawaii.edu/gmt/)84.

Full size image

Koçyiğit81 claimed that the Erciş Fault cut across Early-Middle Miocene marine sequence and Quaternary volcanic rocks to sedimentary packages, also Çakırbey Fault cuts across the Early–Middle Miocene marine limestone, Quaternary volcanic rocks. However, as it is seen from our new paleomagnetic rotations, the northern part of the older Çakırbey Fault was cut and shifted by Erciş Fault. This displacement caused the oppositely directed rotations around Muradiye.

In Fig. 9, the focal mechanism distributions of M > 4.5 earthquakes on the possible Çakırbey Fault are given on the northeastern border of Lake Van. Also, Koçyiğit81 shows that the Eastern plateau is under the influence of the N-S direction of the Arabian plate by using GPS data and focal mechanism solutions of different earthquakes. Our results and the distribution and the focal mechanisms of earthquakes are examined, it is clear that a left lateral strike slip Çakırbey fault exists and is active.

Figure 9
figure 9

Seismicity of the Lake Van and the surrounding area. M > 4.0 earthquakes between 01.01.1900–21.02.2021. Earthquake epicenter data are retrieved from Kandilli Observatory and Earthquake Research Institute (KOERI) and focal mechanism solutions of M > 4.5 earthquakes between 23.10.2011 (after the 23 October 2011, M = 7.2)—21.02.2021 from “Disaster and Emergency Management Authority of Turkey (AFAD) Presidential of Earthquake Department” Fault Mechanism system. Only focal mechanism solutions of the earthquakes on the possible Çakırbey Fault and 23 October 2011 Van earthquake were drawn. (dashed blue line represents possible Çakırbey Fault). Map was created by using the Generic Mapping Tools software, version 5.1.1. (https://www.soest.hawaii.edu/gmt/)84.

Full size image

On the northern southern parts of the Erciş Fault, ~ 13° clockwise and ~ 1° counterclockwise rotations are seen, respectively, between Pliocene–Pleistocene time intervals. These rotations indicate a deformation around a right lateral strike-slip fault and the block movements on both sides of the fault. In summary, according to the results obtained from this study, the study area had a rotation of ~ 2° counterclockwise in the Late Miocene–Pliocene time interval. It was rotated clockwise ~ 13° in the Pliocene–Pleistocene period. Since the Pleistocene, it has rotated ~ 14° counterclockwise.

Conclusions

Widespread and intense seismic activity, rapid uplift of the Eastern Anatolian plateau, westward movement of the Anatolian plate along the NAFZ and EAFZ, strike-slip fault zones in the region and Neogene magmatic activity throughout the region are indicators of the region's complex tectonic structure. At the same time, the fact that the paleomagnetic rotations obtained from our study using the same aged rock groups scattered in different regions is another indicator of the high tectonic activity of the region.

In this study, 62 reliable paleomagnetic sites were taken from the Late Miocene to Quaternary volcanic rocks located to the north of Lake Van, Eastern Anatolia, in order to examine the tectonic deformation of the north of the Lake Van from a paleomagnetic aspect. Hisarlı et al.16 stated that the region moved as a single block rotating clockwise by using a limited number of paleomagnetic sites sparsely covering the study area. By increasing the number and distribution of paleomagnetic sites, our study revealed smaller blocks that are moving in different directions, both clockwise and counter clockwise in the area.

When the inclination angles of the group mean values given in Table 2 are considered, the value is very close (with 2–3° difference) to the expected inclination angle value (58°) for the region, which is a sign that the region has not been making a latitudinal movement since the Late Miocene.

In the vicinity of the western part of Erciş Fault, rotations are observed in all age groups in the Late Miocene–Pleistocene time interval depending on the activity of the Erciş fault. Also, paleomagnetic data from the Pliocene volcanics on the north-western part of Erçek Lake, Van-Muradiye highway (around Çolpan village near the Lake Van) defined R ± ΔR = 24.4 ± 17.0 clockwise rotation. However, there is no active fault in this area on the new active fault map of Turkey78,79. These paleomagnetic rotations and focal mechanism solutions of the recent earthquakes in the field indicate that the former Çakırbey Fault might still be active.