Turkish Journal of Earth Sciences (Turkish J. Earth Sci.), Vol. 15, 2006, pp. 123-154. Copyright ©TÜB‹TAK
3D-Architecture and Neogene Evolution of the Malatya Basin: Inferences for the Kinematics of the Malatya and Ovac›k Fault Zones
NURETD‹N KAYMAKCI1, MURAT ‹NCEÖZ2 & PINAR ERTEPINAR1
1RS-GIS Laboratory, Department of Geological Engineering, Middle East Technical University, TR–06531 Ankara, Turkey (E-mail: [email protected]) 2F›rat University, Department of Geological Engineering, TR–23119 Elaz›¤, Turkey
Abstract: The 3D-architecture of the Malatya Basin was studied using remote sensing, seismic interpretation, and palaeostress analysis in the context of the Malatya-Ovac›k fault zone. The results indicate that the Ovac›k and Malatya fault zones are not different segments of a single ‘so called’ Malatya-Ovac›k fault zone; rather, they are two different fault zones that have operated independently. In addition, the Ovac›k fault zone is delimited in the west by the Malatya fault zone, which extends farther north from the point of supposed junction. Maximum individual deflection of streams along the Ovac›k fault zone is about 9.3 km, and summation of all stream deflections along different segments of the Ovac›k fault zone indicates that sinistral displacement of the Ovac›k fault zone has been not more than 20 km following development of the drainage system in the region. Evidence for three different deformation phases were recognized in the Malatya Basin. Deformation phase 1 was characterized by NW–SE-directed extension and operated in the Early to Middle Miocene interval. Deformation σ phase 2 was characterized by WNW–ESE-directed compression and a vertical 2 which indicates transcurrent tectonics. It operated in the Late Miocene to Middle Pliocene. Deformation phase 3 was characterized by σ σ NNE–SSW-directed compression, and vertical stress is interchanged with 2 and 3; this is interpreted as due to near equal magnitudes of these stresses, resulting in stress permutation and interchange of intermediate and minor stress. Deformation phase 3 commenced in the Late Pliocene and has been active since then. The infill of the Malatya Basin has wedge-like geometry in E–W and N–S directions and the basin fed detritus mainly from its eastern margin. During field studies in the basin, a number of inverted normal faults were encountered; these apparently developed as growth faults in the Early to Middle Miocene time interval and then were reactivated or inverted during post-Middle Miocene compressional phases.
Key Words: Malatya Basin, Malatya fault zone, Ovac›k fault zone, palaeostress, reactivation, inversion
Malatya Havzas›n›n 3 Boyutlu Mimarisi ve Neojen Evrimi: Malatya ve Ovac›k Fay Kuflaklar›yla ‹lgili Ç›kar›mlar
Özet: Malatya Havzas›’n›n 3 boyutlu mimarisi Malatya-Ovac›k fay kufla¤› ba¤lam›nda uzaktan alg›lama, sismik yorumlama ve paleostres analizleri kullan›larak ortaya konulmufltur. Var›lan sonuçlar göstermektedir ki Malatya- Ovac›k fay kufla¤› olarak adland›r›lan hat, tek bir fay kufla¤›n›n iki ayr› segmenti olmay›p, aksine bir birinden ba¤›ms›z hareket eden iki ayr› fay kufla¤›d›r. Ayr›ca, Ovac›k fay kufla¤› bat› ucunda Malatya fay kufla¤› ile birleflti¤i öne sürülen noktadan daha kuzeye do¤ru devam eden, Malatya fay kufla¤› taraf›ndan sonland›r›lmaktad›r. Ovac›k fay kufla¤› içerisindeki derelerin üzerinde görülen en büyük ötelenme 9.3 km olup, derelerin hepsi üzerindeki at›m miktarlar›n›n toplam› Ovac›k fay kufla¤›’n›n, bölgedeki akaçlama sisteminin oluflumundan itibaren, toplam ötelenmesinin 20 kilometreden fazla olamayaca¤›n› göstermektedir. Malatya Havzas›’nda üç farkl› deformasyon evresi tespit edilmifltir. ‹lk evre KB–GD yönlü bir geniflleme dönemi σ olup Erken ile Orta Miyosen döneminde hüküm sürmüfltür. ‹kinci evre ise 2‘nin düfley oldu¤u ve bölgesel do¤rultu at›ml› tektonizmaya iflaret eden DGD–BKB yönlü bir s›k›flma ile temsil edilir. Bu dönem Geç Miyosen ile Orta σ Pliyosen evresinde hüküm sürmüfltür. Üçüncü deformasyon evresi ise KKD–GGB yönlü bir s›k›flma alt›nda 2 ile σ 3’ün düfley eksende zaman zaman yer de¤ifltirdi¤i bir deformasyonla temsil edilir. Bu durum iki stres büyüklü¤ünün eflit oldu¤u durumlarda görülen stres de¤ifl-tokuflu (permütasyon) olarak yorumlanm›fl ve ortaç stresle en küçük stresin zaman zaman düfley eksende yer de¤ifltirmesinin nedeni oldu¤u fleklinde yorumlanm›flt›r.
123 MALATYA AND OVACIK FAULT ZONES
Üçüncü deformasyon evresi, Geç Pliyosende bafllay›p günümüze de¤in etkisini sürdürmektedir. Malatya Havzas›’n›n dolgusu D–B ve K–G yönünde kama fleklinde bir geometriye sahip olup genelde do¤u kenar›ndan beslenmifltir. Arazi çal›flmalar› s›ras›nda, havzada bir çok terselmifl faylara rastlanm›flt›r. Bu faylar›n Erken–Orta Miyosen döneminde büyüme faylar› olarak geliflip Orta Miyosen sonras› s›k›flma dönemlerinde yeniden hareket kazand›klar› veya terseldikleri oldukça belirgindir.
Anahtar Sözcükler: Malatya Havzas›, Malatya fay kufla¤›, Ovac›k fay kufla¤›, paleostres, tekrar hareketlenme, terselme
Introduction that might contain complete records of the subduction, The Late Cretaceous to early Tertiary evolution of Turkey collision and post-collisional processes. was dominated mainly by northward subduction along The Malatya Basin is one of the largest basins in the northern and southern branches of the Neotethys eastern Turkey (Figure 2). It is located in the hinterland Ocean and collision processes after their terminal closure. of the Bitlis-Zagros Suture Zone (Figure 1b) which The northern branch is located between the Rhodope- demarcates the former position of the southern branch of Pontide fragments in the north and the Taurides in the the Neotethys, and along which the Arabian and Eurasian south (Figure 1a). The southern branch was located plates collided (Figure 1c). It has infill more than 2 km between the Eurasian and Arabian plates in SE Turkey thick spanning the Early Miocene to Recent, and which (fiengör & Y›lmaz 1981; Görür et al. 1984; Y›lmaz et al. rests on Lower Tertiary volcano-sedimentary units. 1993; Koçyi¤it & Alt›ner 2002). Complete subduction Therefore, it offers a unique opportunity for and obliteration of the former took place in the Late understanding the Neogene development of the region as Cretaceous to early Tertiary (Yal›n›z & Göncüo¤lu 1999; well as the effects of collision and post-collisional Gençalio¤lu-Kuflçu et al. 2001; Rojay et al. 2001) which convergence to which this paper is addressed. Therefore, resulted in collision of the Taurides and Pontides along the primary aim of this paper is to describe and model the the ‹zmir Ankara-Erzincan Suture Zone (fiengör &Y›lmaz 3D-architecture of the Malatya Basin and to unravel its 1981; Koçyi¤it 1991; Kaymakc› 2000; Kaymakc› et al. Neogene evolutionary history using remote sensing, 2000, 2003a, b) during the early Tertiary. Terminal seismic interpretation and palaeostress inversion based on subduction and obliteration of the southern branch took fault-slip data. The information obtained from the place along the Bitlis-Zagros Suture Zone in the Late Malatya Basin will be used to understand the development Miocene (fiengör & Y›lmaz 1981; fiengör et al. 1985; and evolution of the region, especially the kinematics and Dewey et al. 1986; Y›lmaz et al. 1993) and resulted in activity of the two major fault zones of the region, the collision of the Arabian plate with the Eurasian plate. namely the Malatya and Ovac›k fault zones (MFZ and The collision and further northward convergence of the OFZ, respectively). Field studies included ‘ground- Arabian plate gave way to the westward escape of the truthing’ of remotely sensed information and seismic Anatolian Block along the dextral North Anatolian and interpretation, and evaluation of large-scale structures sinistral East Anatolian fault zones (NAFZ and EAFZ, and fault kinematics based on palaeostress analysis from respectively) from the zone of high convergent strain fault-slip data sets collected during field studies. In towards the Aegean trench. Although there is debate addition, the information obtained from the Malatya about the timing of collision and the inception of the Basin will be extrapolated to constrain and discuss the NAFZ and the EAFZ, the neotectonic scheme of Turkey is possible active tectonics and deformation mechanisms of relatively better known than the geological constraints the region. that existed before the Arabian collision took place. This A brief stratigraphic summary of the region with is partly due to difficulties in the dating of continental special emphasis on the Neogene infill is given first, and deposits which prevailed in most of the early Tertiary and followed by remote sensing, seismic interpretation and the Neogene, and partly due to the lack of regional palaeostress analysis, and finally by a discussion of the studies addressing this problem, in addition to incomplete data and our conclusions. kinematic and tectono-stratigraphic studies in the basins
124 N. KAYMAKCI ET AL.
BLACK SEA NORTH ANATOLIAN F.Z. EURASIAN PLATE a b Rhodo pe Ponti deFra Z gments IAES IAESZ ES IA Z KIRÞEHÝR TA U BLOCK R I DE IT S BZSZ S OFZ ARABIAN PLATE MEDITERRANEAN SEA n ý .Z. s F Bitlis-Pötürge-Malatya-Alanya metamorphics N a A ophiolites and accretionary wedge materials LI MFZ B O BSZ: Bitlis Zagros Suture Zone AT a N y E Z IAESZ: Izmir-Ankara-Erzincan Suture Zone t A R O a T U N ITS: Intra-Tauride Suture Zone l S UT E SFZ a EA S S M RO AG IS Z ITL B strike-slip faults Kýrþehir Block ophiolites from Northern Tethys Tauride platform units arc- and arc-related rocks Malatya-Keban metamorphic rocks N ophiolites from Southern Tethys Arabian Platform 0 50 100 OFZ: Ovacýk Fault Zone km MFZ: Malatya Fault Zone SFZ: Sürgü Fault Zone c 30°Black Sea 40° 42° NAFZ Eurasian Plate
Aegean Sea Anatolian Block
EAFZ TL B I IS-ZA Ionian Trench GR O S Mediterranean Sea S U T DFZ Arabian Plate U R 34° African Plate E
active convergence zones major strike-slip faults subordinate faults DFZ: Dead Sea Fault Zone, EAFZ: East Anatolian Fautl Zone, NAFZ: North Anatolian Fault Zone
Figure 1. (a) Palaeotectonic outline of Turkey (modified from fiengör et al. 1984 and Görür et al. 1984); (b) isopic zones of the eastern Taurides (modified from Perinçek & Kozlu 1984); (c) neotectonic outline of Turkey and surrounding regions (reproduced from Kaymakc› 2000).
125 MALATYA AND OVACIK FAULT ZONES
38° 39° SIVAS BASIN ERZÝNCAN NA FZ ÝLÝÇ 37
36 NE ZO OVACIK LT DÝVRÝÐÝ AU 38 F KANGAL BASIN KEMALÝYE IK AC OV
1 2 3
4 5 MFZ ARAPKÝR
39° 6 7 1817 20 19 39° strike-slip faults 8 16 thrust faults strike-slip faults 15 with reverse 45 ARGUVAN 46 component HEKÝMHAN 47 plung. anticline 44 9 13 plung. syncline 10 12 43 14 11 probable faults 40 42 35 AYDINLAR Keban Dam 44 sample sites A 41 33 34 TF 32 water bodies Plio-Quaternary NE Sultansuyu Fm. Karakaya Dam ZO (Pliocene) LT MALATYABASINMALATYA AU U. Miocene MFZ 31 F 25 N e clastics 29 IA on 24 26 48 L re Z M. Miocene 30 TO Sutu volcanics YEÞÝLYURT 28 A agros YEÞÝLYURT-BASKILN FAULT ZONEf Bitlis-Z L. Miocene 39 A t o 27 T s clastics 23 S ru 21 A Th L. Miocene 22 E ng marine units adi DOÐANÞEHÝR Le ARABIAN basement in Ma PLATE 20 km 38° 39° N 38° 38°
Figure 2. Geological map of the Malatya Basin area for the Neogene.
Stratigraphy Miocene. In fact, the actual development of the Malatya The lithologies exposed in and around the Malatya Basin Basin succeeded Early Miocene regional extension which are grouped as basement rocks – upon which the basin gave way to widespread deposition in E and SE Anatolia has developed – and as basin infill. All of the pre-Neogene far beyond the present boundaries of the Malatya Basin units are designated basement since their deposition and was accompanied by marine incursions. Therefore, predates the development of the Malatya Basin and are the Early Miocene units are also included in the basin infill not confined to the area within the boundaries of the (Figure 3). Although they are not confined to the present basin (Figure 2). Field studies and seismic interpretations limits of the Malatya Basin, they are related to the indicate that basin development has been accompanied by extensional deformation which stimulated basin extensional deformation which commenced in the Early development.
126 N. KAYMAKCI ET AL. UNCONFORMITY LOCAL
alluvium alluvial fan braided river clastics (Sultansuyu fm.) coal seams lacustrine facies Red clastics very thick basaltic and andesites lava flows and pyroclastics (Yamadað interlayered with Volcanics) fluvio lacustrine units
LATE MIOCENE LATE MIOCENE LATE TO MIDDLE PLIO- QUATERNARY
EARLY TO EARLY MIDDLE PLIOCENE 50 m 50 turbiditic, siltstone, shale, and Nummulite, echinidea, Miliolidea and pelecypoda bearing sandy-, silty-, clayey- limestone alternation tuff/tuffite coal seams fluvio-lacutrine sandstone, siltstone, mudstone and conglomerate alternation, intercalation of plant remains-bearing lacustrine shales and marls reefal, algal limestone marine shale/marl, siltstone, sandstone ANGULAR UNCONFORMITY neritic carbonates rich in pelecypoda and gastropoda UNCONFORMITY LOCAL
continued
EARLY TO MIDDLE MIOCENE MIDDLE TO EARLY LATE EOCENE LATE
BASEMENT INFILL OLIGOCENE Generalized columnar sections for the basement and the infill of the Malatya Basin (modified from Sümengen & Uysal 1990). turbiditic shale, siltstone, sandstone with levels of conglomerate limestone rich in Nummulites marine turbidic sequence including proximal, medial and distal facies. hard Nummulitic limestone yellow, Alveolina Nummulites and yellow to buff bearing sandstone, siltstone,shale and conglomerate lenses pleagic limestones, shale, siltstone and sandstone alternation and conglomerate intercalations sandstone, conglomerate and clayey limetone. Pebbles are derived mainly from Jurassic-Cretaceous carbonates PROGRESSIVE UNCONFORMITY recrystallized limestone/marble various metamorphics belonging to Malatya and Pötürge massifs ANGULAR UNCONFORMITY
Figure 3.
CRETAC.
MIDDLE EOCENE MIDDLE
MESOZOIC PALAEOCENE
LATE PALAEOZOIC-
127 MALATYA AND OVACIK FAULT ZONES
Basement Units We refer the reader to Bingöl (1984), Perinçek & The present study focuses on the Neogene development Kozlu (1984), Özgül & Turflucu (1984), Yazgan (1984), of the Malatya Basin; detailed description of the basement Aktafl & Robertson (1984), Kozlu (1987), Y›lmaz et al. units is beyond the scope of this work. However, a brief (1993), Robertson (2002), Jaffey & Robertson (2004), summary of the basement units will be given below. and Robertson et al. (2004, 2005) for a full description of the basement units and, mainly, interpretations of the The geology of the basement units is not yet fully early Tertiary evolution of the region. understood. In the literature (references herein), there are different models and scenarios that have been set forth to explain the evolution of the region. Among these, Infill the most comprehensive is the one of Perinçek & Kozlu The infill of the Malatya Basin rests unconformably on the (1984) upon which our summary is based. The basement Oligocene and older units and begins, at its base, with in the region comprises various metamorphic rocks, buff to yellow and white clayey marine sandstones, Mesozoic platform carbonates, Late Cretaceous ophiolitic siltstones and marl, and continues upward with an algal slivers and mélanges, various Late Cretaceous to Early fossilifereous limestone, turbiditic, fining-upward Palaeocene volcanic rocks and volcano-sedimentary sequences of sandstone, siltstone and shale, and assemblages, Late Palaeocene to Oligocene volcaniclastic, intercalations of thickly bedded marls (Olukkaya marine clastic, and carbonate rocks, and widespread Formation of Sümengen & Uysal 1990). This unit marks evaporite occurrences (Figure 3). These units were the onset of marine inundation in the region after early intruded by various intrusive igneous bodies in the early Tertiary compression generated thickening and uplift of Tertiary. Regionally, the basement units are included in the region (Perinçek & Kozlu 1984; Y›lmaz et al. 1993). the eastern Tauride belt. Perinçek & Kozlu (1984) This unit has been dated by various researchers (e.g., subdivided the basement units according to the isopic Perinçek & Kozlu 1984; Yalç›n & Görür 1984) who have zones of Özgül (1976) and Özgül & Turflucu (1984) in reported that it is of Early Miocene to Middle Miocene the central Taurides (Figure 1b). age. These marine units are conformably overlain by The tectonic development of the region was continental fluvio-lacustrine assemblages at the centre of dominated mainly by the southward thrusting of ophiolite the basin, while at the basin margins they are locally masses originating from the northern Neotethys along overlain unconformably by Middle Miocene continental the ‹zmir-Ankara-Erzincan Suture zone (Figure 1a) in the units. The Middle Miocene fluvio-lacustrine assemblage Late Cretaceous, and northward thrusting and ophiolite (Zeynepo¤lu Formation of Sümengen & Uysal 1990) is obduction originating from the Intra-Tauride Ocean from exposed mainly in the central and northern parts of the which arc and arc-related intrusive and volcanic rocks Malatya Basin. The thickest and relatively most complete originated in the early Tertiary. In addition, some of the sections of these units are exposed in the central part of ophiolitic units obducted along the south-verging thrust the Malatya Basin – south of the Ayd›nlar thrust fault – in sheets over the Arabian plate originated from southern the core of an anticline (location 35 in Figure 2) and are Tethys in the late Tertiary to Late Miocene (Perinçek & composed of (in decreasing abundance) lacustrine, marl, Kozlu 1984; Aktafl & Robertson 1984; Yazgan 1984; shale, siltstone and sandstone, and includes various Y›lmaz et al. 1993). currently economic coal horizons intensely sheared by The early Tertiary units in and around the basin are thrusting with a top-to-north vergence. In the northern characterized by various microfossil-, pelecypod- and part of the Malatya Basin, these lacustrine units are gastropod-bearing limestones intercalated with interlayered with various lava flows and pyroclastic alternations of turbiditic sandstone, siltstone, shale and materials of the Yamada¤ volcanics. The radiometric ages marl (Figure 3). These units are capped along a local of interlayered volcanic rocks (Leo et al. 1974; Arger et unconformity by Oligocene neritic carbonates rich in al. 2000) are late-Early to Middle Miocene (ca. 18 to 14 pelecypods and gastropods. All of the early Tertiary units Ma). In the NW part of the Malatya Basin, N of Arguvan are unconformably overlain by the Lower Miocene infill of (Figure 2), the Yamada¤ volcanic rocks separate the the Malatya Basin. Malatya Basin from the Kangal Basin and are intercalated with and overlain by a few tens of meters of thick red
128 N. KAYMAKCI ET AL.
clastics which are also exposed in the Kangal Basin and The Sultansuyu formation is unconformably overlain are overlain by an approximately 100-m-thick lacustrine by various alluvial-fan deposits exposed mainly along the sequence with economic coal horizons. Continental Malatya fault zone and the eastern margin of the Malatya mammalian fossils collected from these units have been Basin. Although no age data have been reported from dated by Hans de Bruijn (Faculty of Geosciences, Utrecht these units, based on their stratigraphic position over the University, The Netherlands). The following faunal Sultansuyu formation, a post-Middle Pliocene age is elements were identified from Kangal Basin: Talpidae assigned to these alluvial-fan deposits. Desmaninae, Archaeodesmana, Insectivora Soricidae, The youngest deposits of the Malatya Basin are Ssciuridae Tamias, Gerbillinae Pseudomeriones, actively accumulating alluvium and alluvial-fan deposits Cricetinae s.l. Blancomys, Murinae Micromys, Murinae along active faults, mainly in the southern part of the Occitanomys, Spalacinae Pliospalax, Castoridae Dipoides, Malatya Basin around Do¤anflehir. Eomyidae Keramidomys, Cricetinae Allocricetus, Murinae Apodemus, Ochotonidae Prolagus, Hyaenidae, Perissodactyla Equidae Hipparion, and Proboscidea Remote Sensing Gompotheriidae Anancus . These fossils characterize the Remote-sensing studies have included processing and European MN-13 Mammal Zone. In addition, interpretation of Landsat TM, Space Shuttle Progonomys sp. was identified in samples collected from Topographical Mission (SRTM) data and Advanced the central part of the Malatya Basin (site 35). It was the Spaceborne Thermal Emission Radiometer (ASTER) first real mouse (Murinea) to arrive in Europe and the images. SRTM was used to produce a digital Middle East and characterizes MN 9 and younger zones. terrain/elevation model (DTM/DEM) of the topography. Similar fauna have also been observed in different parts of the Kangal and Sivas basins (Saraç et al. 2000); These images have different levels of spatial (ground) therefore, a Late Miocene age can be assigned to the resolution and spectral resolution (i.e., bands). For upper levels of the infill of the Malatya Basin. example, Landsat images have a 30-m ground resolution while the SRTM data is at 3 arc seconds (approximately In the southern part of the Malatya Basin, the Late 69.18 m in the E–W direction and 92.46 m in the N–S Miocene units are overlain by very thick fluvial (mainly direction) resolution; these were used to produce a digital braided river) conglomerates and sandstones (the terrain (elevation) model of the region. These images are Sultansuyu formation, first named in the present study). quite useful for detecting large-scale structures, but they However, no fossils could be identified in these units. A fail to detect structures less then their spatial resolution. faunal assemblage comprising Arvicolinae Out of 14 bands, the first three VNIR (visible and Promimomys/Mimomys, Rodentia Eomyidae, Castoridae infrared) bands of the ASTER imagery have a 15-m Castor praefiber and Gerbillinae was identified by ground resolution; these were used in areas where higher Sickenberg (1975) in a similar facies of the Elbistan Basin ground resolution is needed. The SRTM is used to (located approximately 50 km W of the Malatya Basin). In visualize the topography of the region (Figure 4a). This addition, Ünay & Bruijn (1998) reported Rodentia method enhances the detection of lineaments and Castoridae, Murinae Apodemus cf. Dominans, Arvicolinae structures, including active faults of the region. After the Mimomys Moldavicus and Insectivora Soricidae , images were enhanced and lineaments were delineated Cricetinae Mesocricetus aff. Primitivus, Spalacinae, visually on the images, delineated lineaments were Murinae Occitanomys cf. Brailloni, Murinae Apodemus checked during field studies. The lineaments lacking the Dominans, Arvicolinae Mimomys occitanus , and field expression were removed, and existing lineaments Lagomorpha Ochotonidae Ochotonoides from two were labelled as faults; also, fault-slip data were collected different localities at the western margins of the Elaz›¤ from most of these faults. The enhanced image and Basin which has facies characteristics similar to those of resultant map are given in Figure 4. the Sultansuyu formation. These fossils characterize the MN14-15 European Mammal zones. Based on this On the images and in the field, it was observed that information, an Early to Middle Pliocene age is assigned the Malatya fault zone is composed of a number of sub- to the Sultansuyu formation. parallel strands (Figure 4b, c). In general, the MFZ displays an anastomosing pattern characterized by small
129 MALATYA AND OVACIK FAULT ZONES
38° 38°30 39° NORTH ANATOLIAN40°
ERZINCAN BASINFAULT ZONE 39°30 39°30
OVACIK FAULTOVACIK ZONE BASIN (OFZ) 39° 39°
AYDINLAR T.F.
ELAZIÐ BASIN 38°30 38°30 MALATYA
MALATYA FAULT ZONE (MFZ) BASIN
NE ZO LT AU Sultansuyu River F N IA OL AT AN EAST 38° 38° 38° 38°30 38° 38°30
Figure 4. (a) Colour-coded shaded relief image of Malatya Basin area prepared from SRTM data and major lineaments in the region. Note that the Ovac›k fault zone is delimited by the Malatya fault zone. Note also the very straight course of the Sultansuyu River; (b) structural map of the Malatya basin, produced from field and remote-sensing studies. Note displaced steam channels (indicated by blue arrows). (c) structural elements and offset streams are overlaid on ASTER imagery. local highs and depressions, linear valleys and deflected reaches an elevation of 600 m above the Plio–Quaternary streams; these features are characteristic of strike-slip floor of the Malatya Basin (Figure 4a). In the northern fault zones. The MFZ delimits the western margin of the part of the Malatya Basin, the western continuation of the Malatya Basin with a pronounced escarpment that Ovac›k fault zone is has been traced.
130 N. KAYMAKCI ET AL.
ERZÝNCAN BASIN major faults strike-slip faults displaced stream channel inverted or reactivated faults (ticks are on hanging wall) strike-slip faults with reverse component 04 Seismic line probable or concealed faults folds thrust faults subordinate faults
ÝLÝÇ OVACIK
M Figure 15a A LA T Y A F KEMALÝYE A
U
L
T
Z
O
N OFZ
E E E ZONE T
DUTLUCA AUL
ACIK F OV ARAPKÝR
KEBAN
ARGUVAN Figure 14b
04 Figure 13
A YDIN BASKÝL LAR T YAZIHAN .F. ZONE 02
T 06 05 AUL F E A 03 N O TY 10 Z 01 T L 11 U MALA A 07 MALATYA F N IA AKÇADAÐ L EAFZ TO A N A T S A E
ÇELÝKHAN DOÐANÞEHÝR
F.Z SSÜRGÜÜRGÜ F.Z ULT FA N A M A IY N AD ADIYAMAN 50 km Figure 4b
Figure 4b.
131 MALATYA AND OVACIK FAULT ZONES
major faults strike-slip faults displaced stream channel inverted or reactivated faults (ticks are on hanging wall) strike-slip faults with reverse component probable or concealed faults folds thrust faults subordinate faults
N
50 km
Figure 4c.
132 N. KAYMAKCI ET AL.
Seismic Interpretation normal fault plane. Similar inversion patterns are also Out of 11 sections, nine seismic lines that were shot by observed all along the MFZ (Figures 5–12). Mobil Exploration Mediterranean Inc. in 1990 were obtained from the Turkish General Directorate of Palaeostress Analysis Petroleum Affairs. The available seismic lines are in time From 49 sites, 696 fault-slip data were collected. The domain and were stacked by the company. The sampled horizons range from Late Cretaceous to interpretation of the sections was performed manually, Plio–Quaternary in age. Therefore, we have a complete directly onto the hard copies. During the interpretation, record of deformation since the Late Cretaceous. the boundaries of stratigraphic horizons and exposed faults were extrapolated from outcrops. Palaeostress configurations were constructed using Angelier’s software. During the analysis of the fault-slip In the seismic sections, Eocene to Oligocene data, a methodology developed by Kaymakc› (2000) and sequences, pre-Eocene units, Early Miocene marine Kaymakc› et al. (2000, 2003a) was followed. The main limestones, Early to Late Miocene units and steps in the palaeostress analysis were: (1) separation of Plio–Quaternary sequences were identified. The lines 01, deformation phases in the field; (2) an automated 02, 06, 07, 10 and 11 (see Figure 4b for their locations) separation process was applied using the grid search traverse the western margin of the Malatya Basin, and method (Hardcastle & Hills 1991) because of the the Malatya fault zone has been interpreted along these possibility of mixed data sets; (3) misfit faults were lines (Figures 5–10). The other lines, including 03, 04 separated from the bulk data and were analyzed and 05 (Figure 11), were shot directly in the Malatya separately [This process was applied until reliable results Basin and do not cross the Malatya fault zone. It has been were obtained. ‘Reliable results’ means that all the faults observed that the MFZ displays a positive flower in one set can have 15º of maximum deviation (ANG of structure (cf. Sylvester 1988), a feature typical of strike- Angelier 1988, and that the maximum value of the quality slip fault zones (Figure 12). Different branches of the estimator (RUP) is 45º (see Angelier 1994)]; (4) the flower structure have both reverse and normal faults which did not fulfil these conditions were regarded components of slip, while the strike slip is the dominant as spurious, possibly due to measurement errors [After slip sense. In all of the seismic sections, some subordinate the analysis of the data, it was found that less than 1% faults were also interpreted. In the combined sections of of the data were spurious, or about 12 faults; the lines 04 and 05 (Figure 11), the Ayd›nlar thrust fault remaining 684 faults yielded reliable stress (ATF) has been recognized, which was also observed configurations. At some of the sites, more than one during field studies (Figure 13a). It is noted that most of deformation phase was recognized in addition to the secondary faults in the Malatya Basin, including the separation in the field; they were then separated and re- ATF, are reactivated normal faults. Among these, the reprocessed]; (5) 56 different stress configurations were most spectacular inverted fault is observed in relation to then constructed; among the 56 stress configurations, 41 the Ayd›nlar thrust fault. In the southern part of the ATF of them belong to Miocene to Recent deformation phases (on the hanging-wall block), the Early to Late Miocene since their data were collected from Neogene units. The infill of the Malatya Basin is quite thick, while it is resultant stress configurations are depicted in Figure 14 extremely thin on the northern block; this indicates a and in Table 1. [The remaining 15 stress configurations typical normal growth-fault pattern. Along the inverted were collected from Late Palaeocene to Oligocene syn- fault near the ATF, the basal limestone facies (Early sedimentary faults, thought to belong to pre-Miocene Miocene reefal limestones) and overlying units are up- deformation phases. Therefore, those data are presented thrown. Collectively, this information indicates that the elsewhere and are not included in this paper]; and (6) the first normal faulting occurred, then the normal fault was constructed stress configurations were grouped inverted. During inversion of the ATF, a roof thrust according to their ages. Dating of events is based on the developed which resulted in formation of a triangular ages of the hosting formations and the relative order of zone (Figure 13b). Moreover, an anticline developed due occurrences as discussed in the next section. to bending of the inverted fault along the pre-existing
133 MALATYA AND OVACIK FAULT ZONES
PLIO-QUATERNARY NW EARLY TO MIDDLE PLIOCENE SE
EARLY TO LATE MIOCENE
MALATYA FAULT ZONE
EOCENE-OLIGOCENE UNDIFFERENTIATED
LINE 01
Figure 5. Uninterpreted and interpreted seismic section along line 01. Arrows mark fore-sets and/or shingles that indicate sediment transport direction. MFZ– Malatya fault zone. See Figure 4b for location.
134 N. KAYMAKCI ET AL.
LINE 02
NW SE
PLIOCENE
EARLY TO LATE MIOCENE
OLIGOCENE - EOCENE
MALATYA FAULT ZONE EARLY MIOCENE (MARINE)
UNDIFFERENTIATED
Figure 6. Uninterpreted and interpreted seismic section along line 02. Arrows mark fore-sets and/or shingles that indicate sediment transport direction. MFZ– Malatya fault zone. See Figure 4b for location.
135 MALATYA AND OVACIK FAULT ZONES
LINE 06
MALATYA FAULT PLIOCENE TO PLIO-QUATERNARY (undifferentiated)
WNW ZONE ESE
E
N O
Z EARLY TO LATE MIOCENE
T L
U EARLY MIOCENE (MARINE) A F OLIGOCENE-
A EOCENE Y T
A
L
A
M
UNDIFFERENTIATED
Figure 7. Uninterpreted and interpreted seismic section along line 06. Arrows mark fore-sets and/or shingles that indicate sediment transport direction. MFZ– Malatya fault zone. See Figure 4b for location.
Field Observations and Deformation Phases overprinting slickensides and block tilting, were used to During field studies, various compressional and differentiate the deformation phases. The most reliable extensional structures were observed at different results for ordering the deformation phases were based stratigraphic horizons. This information, together with on mesoscopic faults sealed by strata that are unaffected by them.
136 N. KAYMAKCI ET AL.
LINE 07 LINE 07
PLIO- QUATERNARY EARLY TO MIDDLE MALATYA FAULT ZONE PLIOCENE
EARLY TO LATE MIOCENE
EOCENE - O L EARLY MIOCENE IG O (MARINE) C E NE
UNDIFFERENTIATED UNDIFFERENTIATED
Figure 8. Uninterpreted and interpreted seismic section along line 07. MFZ– Malatya fault zone. See Figure 4b for location.
137 MALATYA AND OVACIK FAULT ZONES
LINE 10
EARLY TO MIDDLE WNW PLIOCENE PLIO-QUATERNARY ESE
E EARLY TO LATE MIOCENE
N O
Z EARLY MIOCENE T (MARINE) - OLIGOC L E EN EN E U C
A O E F
A
Y
T
A
L
A
M
UNDIFFERENTIATED
Figure 9. Uninterpreted and interpreted seismic section along line 10. MFZ– Malatya fault zone. See Figure 4b for location.
138 N. KAYMAKCI ET AL.
LINE 11
MALATYA FAULT ZONE WNW EARLY TO MIDDLE ESE PLIO-QUATERNARY PLIOCENE
EARLY TO LATE MIOCENE
EARLY MIOCENE MARINE
EOCENE - OLIGOCENE
UNDIFFERENTIATED
Figure 10. Uninterpreted and interpreted seismic section along line 11. MFZ– Malatya fault zone. See Figure 4b for location.
139 MALATYA AND OVACIK FAULT ZONES
LINE 04 LINE 05
AYDINLAR THRUST FAULT (ATF) PLIOCENE TO PLIO-QUATERNARY NNW Figure13 (undifferentiated) SSE
EARLY TO LATE MIOCENE IN V E R T E D EARLY MIOCENE EOCENE - OLIGOCENE N O (MARINE) R M A L FA U LT (IN F)
UNDIFFERENTIATED
Figure 11. Uninterpreted and interpreted combined seismic section of lines 04 and 05. See Figure 4b for location.
140 N. KAYMAKCI ET AL.
MFZ
MFZ
ATF
02
I
N F
01 MFZ 04,05
MFZ 06
MFZ 10
MFZ 11
ATF: Aydýnlar thrust fault INF: inverted normal fault MFZ: Malatya Fault Zone
07
Figure 12. Simplified interpreted sections depicting 3D geometry of structures in the Malatya basin. INF: inverted normal fault. Note flower structure (terminology after Sylvester 1988) of the Malatya fault zone (MFZ).
141 MALATYA AND OVACIK FAULT ZONES c a (b) F AT corresponding
(c)
)
F
T
A
(
M35
T
L
E
U N
titudes of the faults. Location is site 35;
°
A
7 rmal fault (INF) and the ATF;
F
2
T
,
S
W
° U
0
R
2
H
N
T
R
A
L
N
I
D A b
INF ATF 1 2 View of exposed upper part of the Ayd›nlar thrust fault (ATF) developed within the Middle to Late Miocene units, showing the at a simplified model explaining the development of fault-bend fold, roof thrust (ATF) and triangular zone between the inverted no palaeostress configuration constructed using slickensides collected from the mesoscopic faults at the site.
BASEMENT N60°W,47°NE N60°W,47°NE N60°W,47°NE Figure 13. (a)
142 N. KAYMAKCI ET AL.
In most areas of the Malatya Basin and all along the and the very low Φ values, it is concluded that these sets Malatya fault zone – from Do¤anflehir in the south to ‹liç indicate a new phase of deformation (phase 3). Very low in the north (Figure 4b) – overprinting slickensides were Φ values are interpreted to be the result of equal or near σ σ observed. The primary slickensides, wherever observed, equal magnitudes of 2 and 3; this results in stress indicate normal faulting with a slight strike-slip permutations (Homberg et al. 1997) and coeval component (pitches are not less than 60º), while at other development of strike-slip and reverse faults, which is the localities, the dip-slip and strike-slip components vary case for phase 3. The plane of stress permutation has an among the sites. Based on this information, it is attitude of N72ºW, 82ºSW and is indicated in Figure 14c. concluded that normal faulting pre-dated strike-slip For the timing of deformation phases 2 and 3, refer to deformation and that the oldest Neogene deformation in the next section. the region is extensional (phase 1). The mean orientation of the principal stresses, for deformation phase 1, is as σ σ σ Discussion follows: 1= 065ºN/88º, 2= 220ºN/02º, 3= 316ºN/02º. σ Having 1 vertical and other stresses horizontal indicates Evolution of the Malatya Basin that deformation phase 1 was characterized by NW–SE- The evolution of the Malatya Basin was controlled mainly directed extension (Figure 14a). by three different phases of deformation during which The slickensides that marked deformation phase 1 the regional stress tensor changed; that in turn altered were overprinted by another set of slickensides. Along the style of deformation in the region. The first the Malatya fault zone, some of the mesoscopic normal deformation phase occurred in the Early to Middle faults were reactivated as reverse faults. In addition, Miocene and caused the opening of the Malatya Basin. along the planes of these faults, overprinting slickensides This deformation phase was accompanied by marine are widespread. Based on these information, the inundation in the region (Perinçek & Kozlu 1984; overprinting slickensides are interpreted as a Sümengen & Uysal 1990). Although marine inundation manifestation of a new phase of deformation (phase 2). may have been due to global sea-level rise, the presence The mean orientations of the principal stress for of growth faults that controlled the deposition of the σ σ deformation phase 2 are as follows: 1= 354ºN/03º, 2= Early to Middle Miocene units is evidence for the opening σ σ 244ºN/86º, 3= 080ºN/03º (Figure 14b). Having 1 and of the Malatya Basin in the Early Miocene. The marine σ σ 3 horizontal and 2 vertical indicates that strike-slip inundation suggests that the elevation of the Malatya deformation prevailed during deformation phase 2. Basin was at about sea level. The presence of shallow- In addition to overprinting slickensides along the marine algal limestones and the absence of deep marine normal faults mentioned above, overprinting slickensides deposits and fauna indicate that the Malatya Basin was were also observed within the Late Miocene to Pliocene not very deep during this time interval. The stratigraphic units; at some of these sites, overprinting slickensides relationship of the Lower Miocene units with the Middle with relatively very low pitches (<15º), indicative of two and Upper Miocene units is regressive. Marine conditions different sets of strike-slip deformations, were observed. were replaced by continental ones in which lacustrine The newer slickensides generally indicate reverse conditions dominated. It has been reported that there components. In addition, similar slickensides were also was a global sea-level drop in the Middle Miocene which observed in the Late Pliocene to Quaternary units (phase resulted from a glacial event (cf. Hilgen et al. 1999); this 3). The mean orientations of the principal stress are as implies that in the Middle Miocene, the region was not σ σ σ necessarily uplifted, although continental deposits follows; 1= 015ºN/15º, 2= 131ºN/53º, 3= 285ºN/34º (Figure 14c). The orientations of the constructed dominated in this period. The Yamada¤ volcanic rocks σ were extruded in the Early to Middle Miocene, which may palaeostress configurations indicate that in these sets 1 σ σ also indirectly indicate that extensional conditions is horizontal while the orientations of 2 and 3 are variable (Fig 14c). At some of the sites, none of the prevailed in the Early to Middle Miocene. Based on these principal stresses are vertical. In addition, the stress ratio information and assumptions, we conclude that (Φ) at these sites is quite low (Table 1). In consideration extensional conditions commenced in the Early Miocene of the overprinting slickensides, the age of host lithology and lasted until the Late Miocene.
143 MALATYA AND OVACIK FAULT ZONES
PHASE 1: EARLY TO MIDDLE MIOCENE EXTENSION
M1B M11C M18 M19
M21 M22B M24B M25A
M26B M29 M30A M41
M45B
s1: 065°N/88° s2: 220°N/02° s3: 316°N/02°
direction of extension slickensides direction of compression dextral sinistral s1 thrusting (if point inwards) s2 normal (if point outwards) s3 a
Figure 14. Palaeostress configurations and mean stress orientations determined from contour diagrams for each of the principal stress sets: (a) for phase 1; (b) for phase 2; (c) for phase 3. Note the interchanging (swapping) plane σ σ of 2–and 3–in phase 3.
144 N. KAYMAKCI ET AL.
PHASE 2: LATE MIOCENE - MIDDLE PLIOCENE COMPRESSION
M1A M7A M8A M9
M11B M12 M17 M24A
M27B M32A M38 M45
M46 M47
s1: 354°N/03° s2: 244°N/86°
direction of extension slickensides direction of compression dextral sinistral s1 thrusting (if point inwards) s2 normal (if point outwards) s3
s3: 080°N/03° b
Figure 14b.
145 MALATYA AND OVACIK FAULT ZONES
PHASE 3: PLIO-QUATERNARY COMPRESSION
M2 M4 M5 M10
M13 M15 M22A M23
M27A M35 M36A M38
a
M44 48
s1: 015°N/15° s2: 131°N/53°
direction of extension Slickensides N direction of compression dextral 72 °W, sinistral 82°S s1 W thrusting (if point inwards) s2 normal (if point outwards) s3
s3: 285°N/34° orientations of both s2 and s3 and plane of stress permutation c
Figure 14c.
146 N. KAYMAKCI ET AL.
Table 1. Coordinates of the sample sites and corresponding palaeostress orientations.
σ σ σ Φ Site Code N X Y 1 (D°/P°) 2 (D°/P°) 3 (D°/P°)
M1A 13 355021 4338743 169/16 359/74 259/3 0.56 M1B 8 355021 4338743 220/66 47/23 316/02 0.47 M2 11 368106 4336460 11/8 117/62 277/27 0.31 M4 10 384789 4326857 25/6 121/49 290/41 0.004 M5 6 387803 4324114 41/25 184/62 305/15 0.155 M7A 25 397206 4316636 164/9 294/76 73/11 0.629 M7B 9 397206 4316636 120/17 301/73 210/00 0.367 M8A 10 400894 4314465 191/19 312/57 91/26 0.643 M8B 8 400894 4314465 291/16 152/70 25/12 0.923 M9 11 411432 4288833 210/4 309/66 119/23 0.637 M10 7 417158 4286515 17/26 168/60 280/12 0.341 M11A 14 513988 4280832 309/69 130/21 40/00 0.511 M11B 10 513988 4280832 338/34 147/55 245/05 0.614 M11C 11 513988 4280832 206/66 68/18 333/15 0.761 M12 11 496422 4285258 352/14 207/73 84/09 0.473 M13 8 484149 4287686 34/5 301/26 134/64 0.163 M14 11 483017 4288173 231/77 330/02 61/13 0.836 M15 7 473295 4297850 199/14 58/72 292/11 0.375 M16 14 457498 4314619 107/71 02/05 270/19 0.158 M17 15 455629 4318427 008/43 197/46 102/05 0.615 M18 7 449344 4320217 265/67 37/16 132/16 0.775 M19 8 442246 4318017 028/69 223/20 131/05 0.618 M21 8 411749 4230108 321/66 060/04 152/23 0.008 M22A 8 399178 4222007 330/14 062/07 178/74 0.124 M22B 7 399178 4222007 094/47 220/29 328/29 0.564 M23 6 400610 4227599 164/13 269/50 064/37 0.318 M24A 13 408921 4246093 349/30 167/60 259/01 0.534 M24B 6 408921 4246093 261/69 052/18 145/09 0.55 M25A 13 405623 4249031 096/79 212/05 303/10 0.868 M25B 7 405623 4249031 153/83 313/07 044/02 0.647 M26A 11 426133 4241528 121/05 029/21 223/68 0.122 M26B 8 426133 4241528 090/73 213/09 305/14 0.427 M27A 30 435776 4236304 004/59 178/31 269/03 0.501 M27B 10 435776 4236304 337/01 238/82 067/07 0.741 M27C 11 435776 4236304 052/31 203/55 314/14 0.724 M28A 10 435925 4237093 138/67 263/14 357/19 0.344 M28B 7 435925 4237093 004/69 271/01 181/21 0.426 M29 7 443850 4244746 029/71 216/19 125/02 0.662 M30A 10 445973 4244680 052/67 236/23 145/02 0.595 M30B 8 445973 4244680 320/61 141/29 051/00 0.556 M31 8 483165 4253362 146/66 312/24 44/04 0.536 M32A 32 474115 4272863 343/05 102/80 252/09 0.629 M32B 10 474115 4272863 103/28 305/60 198/10 0.883 M35 7 441583 4277639 008/02 098/15 272/75 0.402 M36A 18 422317 4364543 018/02 110/43 286/47 0.307 M36B 10 422317 4364543 096/22 243/65 001/13 0.919 M38 11 455930 4348504 188/12 359/78 097/02 0.517 M41 14 421725 4277512 127/86 014/02 284/04 0.578 M42 9 420049 4278058 029/65 162/18 258/17 0.679 M44 9 410990 4289155 226/00 136/01 338/89 0.157 M45 23 413901 4298441 157/82 038/04 248/07 0.625 M45B 6 413901 4298441 255/86 45/ 4 135/2 0.685 M46 10 466840 4299107 161/17 341/73 251/00 0.347 M47 16 452331 4291609 356/01 091/81 266/09 0.402 M48 5 430992 4242799 205/10 352/781 114/06 0.795 Φ σ σ σ σ N: Number of faults, X-Y: UTM Zone 37 coordinates of sites, D/P: Direction/Plunge,– = ( 2- 3)/( 1- 3)
147 MALATYA AND OVACIK FAULT ZONES
It is generally accepted that the collision of the a number of normal faults along which the southern Arabian plate with the Eurasian plate took place by the blocks (hanging-wall blocks) displaced downward, end of the Middle Miocene (fiengör & Y›lmaz 1981; indicating that the basin deepened from north to south Y›lmaz et al. 1993; Westaway & Arger 2001; Westaway and from west to east in the Early to Middle Miocene 2003 and references therein) and it is presumed that the (deformation phase 2). Some of these normal faults were Late Miocene was dominated by compressional inverted during Late Miocene and Pliocene compressional deformation in the region. In the seismic sections and events (deformation phases 2 and 3). during field studies, no important (regional) unconformity was detected between the Middle Miocene and Upper Miocene units; however, the Late Miocene Malatya Fault Zone (MFZ) units in the Malatya Basin are dominated by fluvial red As discussed previously, the Malatya and Ovac›k fault clastic materials. This has two implications: (1) the zones are different segments of a single ‘so called’ evolution of the Malatya Basin continued without break in Malatya-Ovac›k fault zone (MOFZ, Koçyi¤it & Beyhan its sedimentation in the Late Miocene, although a re- 1998; Westaway & Arger 2001). In addition, Westaway organization occurred in the regional stress field and, & Arger (2001) postulated that the MOFZ was active 5–3 hence, in the deformation pattern; (2) extensional Ma and was a plate boundary between the Arabian and deformation lasted until the end of the Miocene and re- Anatolian plates which had taken up approximately 29 km organization of the stress pattern and compressional of relative motion between them. They further speculated deformation did not take place until the Pliocene, during that after the development of the East Anatolian Fault which the Sultansuyu formation was deposited; this Zone (EAFZ) as a single, through-going fault zone, the further implies that the effect of collision along the Bitlis- MOFZ became inactive. In addition to the seismic data, Zagros suture zone was not transferred until the our observations about the Malatya fault zone are as Pliocene, or that the main collision occurred in the follows. All along the escarpment at the western margin Pliocene. These questions are still debated and require of the Malatya Basin (Figures 4a & 12), widespread further study. However, we favour the first assumption mesoscopic faults with sinistral strike-slip and normal since the argument for Late Miocene collision is firmly offset were observed (Figure 15). Such faults are established in the literature. Therefore, it is thought that observed all along the Malatya fault zone, especially in the deformation phase 2 commenced in the Late Miocene and area between Yaz›han and Do¤anflehir (see Figure 4 for resulted from collision of the Arabian plate with the their locations). These faults indicate stepping-down of Eurasian plate. The Late Miocene units are unconformably the escarpment towards the Malatya Basin and the fault’s overlain by the Sultansuyu formation. Based on this reactivation. The Malatya fault zone is at least a 30-km- information, it is proposed that deformation phase 3 wide fault zone characterized by parallel to sub-parallel commenced in the Late Pliocene and recently has still been branches displaying an anastomosing pattern due to in operation. bifurcation and rejoining of different branches (Figures The geometry of the infill of the Malatya Basin is 4); it also has a positive flower-structure geometry in wedge-like in the E–W direction in which the western seismic sections, typical of strike-slip fault zones (cf. margin is thicker and the infill is thinning eastward, Sylvester 1988) (Figure 12). As discussed in the previous although sediment transport took place mainly from the section, most of the observed mesoscopic faults have eastern margin, as indicated by fore-set and shingle overprinting slickensides indicating at least three different patterns in the seismic section (Figures 5–7). In the N–S phases of deformation. The first movement was direction, similar wedge-like geometry is also manifested characterized by a normal dip-slip component; other in the seismic sections (Figure 11). There is a major step movement was dominantly strike-slip with reverse and in the geometry of the Malatya Basin proximal to the normal components. Normal components in the later Ayd›nlar thrust fault (ATF). The stepping is towards the movements are observed mainly in the releasing bends, south and corresponds to an inverted normal fault which and reverse components were observed along the facilitated an accommodation space for deposition of the restraining bends of the major branches. Based on this Lower to Upper Miocene sequences. In addition, there are information, we propose that the Malatya fault zone is a
148 N. KAYMAKCI ET AL. Normal and sinistral mesoscopic faults developed along the Malatya fault zone near sites 24 & 25 (see Figure 2 for locations). Figure 15.
149 MALATYA AND OVACIK FAULT ZONES
reactivated fault zone. It developed in the Early Miocene Malatya-Ovac›k fault zone (MOFZ), but without as a normal fault with a slight sinistral component and presenting any detailed supporting data. Westaway & controlled the development of the Malatya Basin in the Arger (2001) further speculated about the role of MOFZ Early to Middle Miocene. That activity corresponds to in the evolution of the region based only on two deformation phase 1. Then, possibly in the Late Miocene observation points for the whole region, and on and/or in the Pliocene, it that fault was reactivated and topographical and morphological information obtained some of its normal branches were inverted; that activity from topographical maps. They also speculated that corresponds to deformation phase 2. During this activity on the MOFZ ceased about 3 Ma, as mentioned deformation phase, the Ayd›nlar thrust fault also previously developed and the Sultansuyu formation was deposited. Our observations about the OFZ are as follows: (1) In the post-Middle Pliocene, a regional reorganization in the OFZ is characterized by a number of fault segments the stress field took place, corresponding to deformation less than 100 km length, and the zone bifurcates from phase 3. During this period, the Upper Plio–Quaternary the North Anatolian Fault Zone near the SE corner of the units were deposited together with alluvial-fan deposits Erzincan Basin (Figure 4a), as discussed by Westaway & along the Malatya fault zone at the western margin of the Arger (2001); (2) the escarpment that delimits the Malatya Basin. northern margin of the Ovac›k Basin is the most There is no direct evidence, such as medium- to large- prominent feature along the OFZ. Arpat & fiaro¤lu magnitude earthquakes, along the Malatya fault zone. (1975) argued that the segment of the OFZ at the However, displaced stream courses (Figure 4) along the northern margin of Ovac›k Basin dips southward and has MFZ between Yaz›han and Do¤anflehir, the presence of a a normal component of slip which resulted in the number of small-scale earthquakes as high as Ms 3.5, and formation of the Ovac›k Basin; (3) different segments of a very prominent escarpment indicate that Malatya fault the OFZ trend N70ºE and are delimited in many places by zone is presently active. These data imply that the quite prominent lineaments trending N20ºE, parallel to Malatya fault zone is in a period of seismic quiescence. the trend of the MFZ; (4) the Ovac›k fault zone further There is the possibility, therefore, that this fault zone bifurcates in the west starting at the western margin of might produce a large magnitude earthquake in the near the Ovac›k Basin. The northernmost branch follows an future. approximately N80°W trend and is delimited in the west The other striking conclusion yielded by the seismic near Sand›k by the Malatya fault zone. In the Malatya section is the lack of evidence for the Sultansuyu fault Basin, the OFZ is not constrained into a single fault zone; (Figure 4a), the presence of which was first proposed rather, it is distributed over a very large area, all of which along the Sultansuyu River by Perinçek & Kozlu (1984), is delimited in the west by the Malatya fault zone (Figure based on very persistent linearity of the river channel. 16). In this area, almost all of the streams and the Later Westaway & Arger (2001) discussed the same Euphrates River are deflected sinistrally. The maximum lineament. In line with our field observations, neither deflection is about 9.3 km near Dutluca (Figure 16a). Perinçek & Kozlu (1984) nor Westaway & Arger (2001) Based on this deflection and some other morphotectonic have encountered any trace of faulting along the data, Westaway & Arger (2001) argued that the Sultansuyu River, despite its rectilinear course. maximum displacement of the OFZ is about 27 km. However, the sum of the deflections all along the different branches of the OFZ (Figure 4b) indicates a Ovac›k Fault Zone (OFZ) maximum displacement of not more than 20 km; (5) in The Ovac›k fault zone was first recognized by Arpat & addition, along one of the southernmost branches of the fiaro¤lu (1975), and was later studied by Aktimur OFZ about 10 km east of Arguvan, a very pronounced (1979), Perinçek et al. (1987) and Chorowicz et al. anticline developed on the southern block of the fault (Fig (1994) using satellite images and aerial photos. Later, 15b). In this area, it is observed that, this fault is a Koçyi¤it & Beyhan (1998) proposed that the Ovac›k and reactivated fault; it developed as a normal fault dipping Malatya fault zones are two different segments of a single south and later was reactivated as a sinistral strike-slip
150 N. KAYMAKCI ET AL.
a NE ZO LT AP AU PROX IK F IMAT OVAC Sandýk E BOUNDARY OF 38 f f e e a NE c ZO a LT DUTLUCA AU ZONE b F T IK C b A c V O AUL d F O F d Y M38 A R A D N TY U BO TE IMA ROX MALA PP A a-a:8.1 km d-d: 7.8 km b-b: 3.6 km e-e: 2.5 km
c-c: 9.3 km f-f: 2.3 km a
b M47 B (N)
47 B
LATE MIOCENE
LACUSTRINE FACIES MIDDLE MIOCENE
S S S
C C C I I I
T T T
S S S
A A A
L L L
C C C
D D D
E
R
E
N
E
C O I
M
E T
ALLUVIUM A L A A(S)
Figure 16. (a) Shaded relief image of the junction of the Malatya and Ovac›k fault zones [marking major deflections in the streams (a-a’ to f-f’)] and the palaeostress configuration of site 38. Note that the Ovac›k fault zone is delimited by the Malatya fault zone. Note also that the maximum deflection is 9.3 km; (b) simplified map and cross-section of the anticline on the hanging-wall block of an inverted normal fault overlaid on to ASTER image. See Figure 4b for location. This fault is one of the southernmost branches of the Ovac›k fault zone.
151 MALATYA AND OVACIK FAULT ZONES
fault with a reverse component. This implies that some of 3. Basin development was initiated in deformation the pre-existing normal faults in the Early to Late Miocene phase 1 (in the Early Miocene) and continued until (deformation phase 1) were reactivated in the successive the Late Micoene without any significant hiatuses. deformation phases and were incorporated into different Basin development was accompanied by normal segments of the OFZ. Palaeostress data collected from faulting, during which the Malatya fault zone was this area, where the OFZ dominates, indicate that major also a normal fault zone. compression was roughly N–S, which possibly drove the 4. The Malatya Basin displays a wedge-like geometry sinistral movement of the OFZ with a reverse component. in the E–W and N–S directions, and has a major Briefly, the Ovac›k fault zone is a not a single through- step in the north proximal to the Ayd›nlar thrust going fault zone and is not only bifurcated into two fault. branches around Arapkir, as proposed Westaway & Arger 5. The Malatya and Ovac›k faults are two (2001). Rather, it developed in an area where very independent fault zones contra Koçyi¤it & Bayhan diffuse deformation is occurring. Moreover, the OFZ has (1998) and Westaway & Arger (2001), who a number of branches, some of which are reactivated proposed that they are a single fault zone, the ‘so- normal faults with sinistral strike-slip components. called’ Malatya-Ovac›k fault zone. Recent seismic activity along branches of the OFZ 6. Both the Malatya and Ovac›k fault zones are indicates that it is presently active. presently active, contra Westaway & Arger (2001) All of this information invalidates the assumptions of who proposed that they become inactive about 3 Koçyi¤it & Beyhan (1998) and Westaway & Arger Ma. (2001), and the kinematic model proposed by Westaway 7. Most of the faults, including the Malatya and some (2003). Therefore, current kinematic models for the branches of the Ovac›k fault zones, are reactivated Eurasian, Arabian and Anatolian plates should be revised. pre-existing faults which developed as normal faults in the Early to Late Miocene and reactivated Conclusions as sinistral strike-slip faults, possibly during later This study leads us to the following conclusions: compressional events in the post-Middle Miocene. 1. Three different deformation phases have operated 8. Current kinematic models for the Eurasian, in and around the Malatya Basin: (a) The first Arabian and Anatolian plates should be revised on phase was characterized by NW–SE-directed the basis of the data presented here. extension; (b) The second phase was characterized 9. The seismic potential of the Malatya and Ovac›k by approximately NNW–SSE-directed compression faults should be assessed in detail since they have σ and a sub-vertical 2, indicating the operation of the potential of producing large earthquakes in transcurrent tectonics in the region; (d) The third the near future. phase was characterized by NNE–SSW-directed compression and variable vertical σ and σ 2 3 Acknowledgement orientations at various sites. It is interpreted that variable vertical stress indicates stress This study was supported by the “Middle East Basins σ σ permutation between 2 and 3, implying that the Evolution” Programme, project no. MEBE 16-02 (CNRS- magnitudes of the intermediate and minor France). We are thankful to Eric Barrier who encouraged principal stresses were almost equal. The latest us to take part in the MEBE Program and critically phase deformation (phase 3) involved reviewed the field reports. Hans de Bruijn, Gerçek Saraç, transcurrent tectonics, commenced by the Late Engin Ünay, and Erksin Güleç are thanked for the dating Pliocene, and is still active. of the fossils and for generously sharing their mammal- fossil database of Turkey. Cengiz Demirci provided 2. The Malatya Basin comprises infill ranging from seismic sections. Early Miocene to Recent in age.
152 N. KAYMAKCI ET AL.
References
AKTAfl¸ G. & R OBERTSON A.H.F. 1984. The Maden complex, SE Turkey: HILGEN, F. J., A BDUL-AZIZ, H., K RIJGSMAN, W., L ANGEREIS, C.G., L OURENS, evolution of a Neotethyan active margin. In: D IXON J.E. & L.J., M EULENKAMP, J.E., R AFFI, I., S TEENBRINK, J., T URCO, E., VAN ROBERTSON, A.H.F. (eds), The Geological Evolution of the Eastern VUGT, N., W IJBRANS, J. R. & Z ACHARIASSE, W.J. 1999. Present Mediterranean. Geological Society, London, Special Publications status of the astronomical (polarity) time-scale for the 17, 375–402. Mediterranean Late Neogene. Philosophical Transactions Royal Society London A 357, 1931–1947. AKT‹MUR, S. 1979. Malatya-Sivas Dolay›n›n Uzaktan Alg›lama Yöntemi ile Çizgiselliklerinin ‹ncelenmesi [The Lineament Analysis of Malatya- HOMBERG, C., H U, J.C., A NGELIER, J., B ERGERAT, F. & L ACOMBE, O. 1997. Sivas Region Using Remote Sensing]. MTA Report No. 6651 Characterization of stress perturbation near major fault zones: [unpublished]. insights from 2-D distinct-element numerical modeling and field studies (Jura Mountains). Journal of Structural Geology 19, ANGELIER, J. 1988. Tector: Determination of Stress Tensor Using Fault 703–718. Slip Data Set. Quantitative Tectonics, University Pierre and Marie Curie, Paris. JAFFEY, N. & R OBERTSON, A.H.F. 2004. Non-marine sedimentation associated with Oligocene–Recent exhumation of the Central ANGELIER, J. 1994. Fault slip analysis and palaeostress reconstruction. Taurus Mountains, S. Turkey. Sedimentary Geology 173, 53–89. In: HANCOCK, P.L. (ed), Continental Deformation. Pergamon Press, Oxford, 53–100. KAYMAKCI, N. 2000. Tectono-stratigraphical Evolution of the Çankiri Basin (Central Anatolia, Turkey). PhD Thesis, Geologica ARGER, J., M ITCHELL J. & W ESTAWAY, R. 2000. Neogene and Quaternary Ultraiectina. No. 190, Utrecht University Faculty of Earth volcanism of south-eastern Turkey. In: BOZKURT, E., W INCHESTER, J.A., PIPER, J.D.A. (eds), Tectonics and Magmatism, in Turkey and Sciences, The Netherlands, ISBN 90-5744-047-4. the Surrounding Area. Geological Society, London, Special KAYMAKCI, N., WHITE, S.H. & VAN DIJK P.M. 2000. Palaeostress inversion Publications 173, 459–487. in a multiphase deformed area: kinematic and structural evolution
ARPAT, E. & fi ARO⁄LU, F. 1975. Türkiye’deki baz› önemli genç tektonik of the Çank›r› Basin (central Turkey), Part 1. In: B OZKURT, E., olaylar [Some recent tectonic events in Turkey]. Türkiye Jeoloji WINCHESTER, J.A. & PIPER, J.A.D. (eds), Tectonics and Magmatism Kurumu Bülteni 18, 91–101. in Turkey and the Surrounding area. Geological Society London Special Publication 173, 445–473. B‹NGÖL, A.F. 1984. Geology of the Elaz›¤ area in the eastern Taurus region. In: T EKEL‹, O. & G ÖNCÜO⁄LU, M. (eds), Geology of the KAYMAKCI, N., D UERMEIJER, C.E., L ANGEREIS, C., W HITE, S.H. & VAN DIJK, Taurus Belt. Mineral Research and Exploration Institute (MTA) P.M. 2003b. Oroclinal bending due to indentation: a Publication, Ankara, Turkey, 209–216. paleomagnetic study for the early Tertiary evolution of the Çankiri Basin (central Anatolia, Turkey). Geological Magazine 140, CHOROWICZ, J., LUXEY, P., LYBERIS, N., CARVALHO, J., PARROT, J.-F., YÜRÜR, T. & GÜNDO⁄DU, M.N. 1994. The Maras Triple Junction (southern 343–355. Turkey) based on digital elevation model and satellite imagery KAYMAKCI, N., W HITE, S.H. & VAN DIJK P.M. 2003a. Kinematic and interpretation. Journal of Geophysical Research 99/B10, structural development of the Çank›r› Basin (central Anatolia, 20225–20242. Turkey): a palaeostress inversion study. Tectonophysics 364, DEWEY, J.F., HEMPTON, M.R., KIDD, W.S.F., fiARO⁄LU, F. & fiENGÖR, A.M.C. 85–113. 1986. Shortening of continental lithosphere: the neotectonics of KOÇY‹⁄‹T, A. & A LT›NER, D. 2002. Tectonostratigraphic evolution of the Eastern Anatolia – a young collisional zone. In: C OWARD, M.P. & North Anatolian palaeorift (NAPR): Hettangian-Aptian passive RIES, A.C. (eds), Collision Tectonics. Geological Society, London, continental margin of the northern Neo-Tethys, Turkey. Turkish Special Publications 19, 3–36. Journal of Earth Sciences 11, 169–191. GENÇAL‹O⁄LU-KUflCU, G., G ÖNCÜO⁄LU, M.C. & K UflÇU, ‹. 2001. Post- KOÇY‹⁄‹T, A. 1991. An axample of an accretionary forearc basin from collisional magmatism on the northern margin of the Taurides and north central Anatolia and its implications for the history of its geological implications: geology and petrology of the Yahyal›- subduction of Neotethys in Turkey. Geological Society America Karamadaz› granitoid. Turkish Journal of Earth Sciences 10, Bulletin 103, 22–36. 103–119. KOÇY‹⁄‹T, A. & B EYHAN, A. 1998. A new intracontinental transcurrent GÖRÜR, N., O KTAY, F.Y., S EYMEN, ‹. & fi ENGÖR, A.M.C. 1984. Palaeotectonic evolution of the Tuzgölü basin complex, central structure: the Central Anatolian Fault Zone, Turkey. Tectonophysics 284, 317–336. Turkey: sedimentary record of a Neotethyan closure. In: D IXON, J.E. & R OBERTSON, A.H.F. (eds), The Geological Evolution of the KOZLU, H. 1987. Misis-And›r›n dolay›n›n stratigrafisi ve yap›sal evrimi Eastern Mediterranean. Geological Society, London, Special [Stratigraphy and structural evolution of Misis-And›r›n region]. Publications 17, 77–112. Türkiye 7. Petrol Kongresi, Bildiriler Kitab› , 104–116.
HARDCASTLE, K.C. & H ILLS, L.S. 1991. Brute3 and Select: QuickBasic 4 LEO, G.W., MARVIN, R.F. & MEHNERT H.H. 1974. Geologic framework of programs for determination of stress tensor configurations and the Kuluncak-Sofular area, east-central Turkey, and K-Ar ages of separation of heterogeneous populations of fault-slip data. igneous rocks. Geological Society America Bulletin 85, Computers and Geosciences 17, 23–43. 1785–1788.
153 MALATYA AND OVACIK FAULT ZONES
ÖZGÜL, N. 1976. Toroslar›n baz› temel jeolojik özellikleri [Basic geologic fiENGÖR, A.M.C., fiARO⁄LU, F. & GÖRÜR, N. 1985. Strike-slip deformation characteristics of Taurides]. Türkiye Jeoloji Kurumu Bülteni 19, and related basin formation in zones of tectonic escape: Turkey as 65–78. a case study. In: BIDDLE, K.T. & CHRISTIE-BLICK, N. (eds), Strike-Slip Deformation Basin Formation and Sedimentation . Society of ÖZGÜL, N. & T URflUCU A. 1984. Stratigraphy of the Mesozoic carbonate Economic Paleontologists and Mineralogists, Special Publication sequence of the Munzur Mountains (eastern Taurides). In: TEKEL‹, 37, 227–264. O. & G ÖNCÜO⁄LU, M. (eds), Geology of the Taurus Belt . General Directorate of Mineral Research and Exploration (MTA), Special SICKENBERG, O. 1975. Die Gliederung des höheren Jungtertiärs und Publication, Ankara, Turkey, 173–180. Altquartärs in der Türkei nach Vertebraten und ihre Bedeutung für die internationale Neogen-Stratigraphie. Geologisches PER‹NÇEK, D. & KOZLU, H. 1984. Stratigraphy and structural relations of Jahrbuch 15, 1094–1116. the units in the Afflin-Elbistan-Do¤anflehir region (eastern Taurus). In: T EKEL‹, O. & G ÖNCÜO⁄LU, M. (eds), Geology of the SÜMENGEN, M. & U YSAL, fi. 1990. A¤z›kara Project. Petroleum Potential Taurus Belt. General Directorate of Mineral Research and of Kayseri-Gürün-Kahramanmarafl-Malatya Region. Mobil Exploration (MTA), Special Publication, Ankara, Turkey, Exploration Mediterranean Inc. Report [unpublished]. 181–198. SYLVESTER, A.G. 1988. Strike-slip faults. Geological Society America Bulletin 100, 1666–1703. PER‹NÇEK, D., GÜNAY, Y. & KOZLU, H. 1987. Do¤u ve güneydo¤u Anadolu bölgesindeki yanal at›ml› faylar ile ilgili yeni gözlemler [New ÜNAY, E. & B RUJIN, H. 1998. Plio-Pleistocene rodents and lagomorphs observations on the strike-slip faults of east and southeast from Anatolia. Proceedings of SEQS-EuroMam Symposium 60, Anatolia]. Türkiye 7. Petrol Kongresi Bildiriler Kitab› , 89–103. 432–485.
ROBERTSON, A.H.F. 2002. Overview of the genesis and emplacement of WESTAWAY, R. 2003. Kinematics of the Middle East and eastern Mesozoic ophiolites in the eastern Mediterranean Tethyan region. Mediterranean updated. Turkish Journal of Earth Sciences 12, Lithos 65, 1– 67. 5–46.
ROBERTSON, A.H.F., Ü NLÜGENÇ, U., ‹ NAN, N. & T ASLI, K. 2004. The WESTAWAY, R. & ARGER, J. 2001. Kinematics of the Malatya–Ovac›k fault Misis–Andirin complex: mid-Tertiary subduction/accretion and zone. Geodinamica Acta 14, 103–131.
melange formation related to closure and collision of the South- YALÇ›N, M.N. & GÖRÜR, N. 1984. Sedimentological evolution of the Adana Tethys in S Turkey. Journal of Asian Earth Sciences22, 413–453. Basin. In: T EKEL‹, O. & G ÖNCÜO⁄LU, M.C. (eds), Geology of the
ROJAY, B., Y AL›N›Z, M.K. & A LT›NER, D. 2001. Tectonic implications of Taurus Belt. General Directorate of Mineral Research and some Cretaceous pillow basalts from the North Anatolian Exploration (MTA) Special Publication, 165–172. ophiolitic mélange (central Anatolia-Turkey) to the evolution of YAL›N›Z, M.K. & G ÖNCÜO⁄LU, M.C. 1999. Clinopyroxene compositions of Neotethys. Turkish Journal of Earth Sciences 10, 93–102. the isotropic gabbros from the Sar›karaman Ophiolite: new evidence on supra-subduction zone type magma genesis in central SARAÇ, G., DE BRUIJN, H., GÜLEÇ, E., ÜNAY, E. WHITE, T. & VAN MEULEN, A. Anatolia. Turkish Journal of Earth Sciences 8, 103–112. 2000. Türkiye Omurgal› Fosil Yataklar› [Mammalian Fossils of Turkey]. MTA Report [unpublished] YAZGAN, E. 1984. Geodynamic evolution of the Eastern Taurus regi›on. In: T EKEL‹, O. & G ÖNCÜO⁄LU, M.C. (eds), Geology of the Taurus fiENGÖR, A.M.C. & YILMAZ, Y. 1981. Tethyan evolution of Turkey: a plate Belt. General Directorate of Mineral Research and Exploration tectonic approach. Tectonophysics 75, 181–241. (MTA) Special Publication, 199–208. fiENGÖR, A.M.C., Y› LMAZ, Y. & S UNGURLU, O. 1984. Tectonics of the Y›LMAZ, Y., Y ‹⁄‹TBAfl¸ E. & G ENÇ, C.fi. 1993. Ophiolitic and metamorphic Mediterranean Cimmerides: nature and evolution of the western assemblages of southeast Anatolia and their significance in the termination of palaeo-Tethys. In: DIXON, J.E. & ROBERTSON, A.H.F. geological evolution of the orogenic belt, Tectonics 12, (eds), The Geological Evolution of the Eastern Mediterranean . 1280–1297. Geological Society, London, Special Publications 17, 77–112.
Received 23 February 2005; revised typescript accepted 03 January 2006
154