The stratigraphic features of the Taraklı Flysch have been fully reconstructed in the sections cropping out along the northern side of the Akçay Valley, between the Bahçecik and Boyalı Villages and along the Boyalıçay Valley (Fig. 2). The succession, whose thickness has been estimated as at least 700 m, shows a clear thickening and coarsening-upward evolution that can be divided into five different lithofacies (see log in Fig. 3A for more details), which, from the bottom to the top, are: “thin-bedded turbidites”, “medium-grained arenites”, “conglomerates”, “calcareous coarse-grained turbidites”, and “slide-block in shaly-matrix” lithofacies.

The lowermost part of the Taraklı Flysch is characterized by 400-m-thick thin-bedded turbidites consisting of thin-to-medium beds (5–50 cm) of medium- to fine-grained arenites and coarse-grained siltites (Fig. 3B1). The medium-to-fine-grained arenites are often characterized by thin traction carpets. These strata are generally well graded only in their uppermost part, where current ripples and sinusoidal lamina can be also present. In the uppermost part of this lithofacies, decimetric lenticular beds of coarse-grained arenites can be recognized (Fig. 3B2).

The medium-grained arenites lithofacies is characterized by an up to 50-m-thick sequence of turbidites represented by 0.5–2.5-m-thick beds of amalgamated medium-to-fine-grained arenites alternating with subordinate thin beds of shales (Fig. 3A). These strata are characterized by a massive structure without sedimentary features as graded bedding and lamina. The bottom surface of these strata is marked by sole marks and by the widespread presence of organic matter (leaves and tree cortex fragments).

A decimeter-thick level of well-rounded clast- to matrix-supported conglomerates (Fig. 3B3) characterize the medium part of the Taraklı Flysch and represent, in this area, a key level to understand the geometry of the main structures. The most striking feature of this lithofacies is the granite-dominated composition of the pebbles. These beds, derived from high-density erosive flows probably connected to a coarse-grained river-delta system, are characterized by frequent basal erosional features.

The calcareous coarse-grained turbidites lithofacies consists of a sequence of layers (not thicker than 25–30 m) ranging from clast-supported orthoconglomerates to coarse arenites mainly derived from debris flows and high-density turbidity currents. The most common facies is represented by prevalent monomict clast-supported conglomerates characterized by poor sorting. The internal organization of these deposits, characterized by unsorted coarse clasts, is scarce. These beds, derived from high-density erosive flows, are characterized by frequent basal erosional features. The erosional ability is suggested by frequent bottom bedset scours, diffuse amalgamated surfaces, and common rip-up mud clasts. These strata are associated with coarse-grained high-density turbidity current deposits. Thick-to-medium beds without internal structures and with poor sorting are the most common facies. A subtle normal grading and water escape features can be recognized in a few beds.

The most prominent feature of these members is the quasi-monomict composition of the debris characterized by extrabasinal carbonatic clasts.

The upper part of the succession (up to 400 m thick) is dominated by huge slide-blocks embedded in a fine-grained matrix (Fig. 3B4). The matrix of this lithofacies is characterized by varicolored mainly shaly to silty deposits. Subordinate decimetric lenticular beds of coarse-grained arenites have been also recognized. The slide-blocks, usually with lenticular shapes, show different sizes (ranging from boulder up to 100-m-thick blocks) and compositions. Even if the primary relationships between the slide blocks and the surrounding matrix are always tectonized, their emplacement due to submarine landslides for these blocks is suggested by synsedimentary deformation structures recognized in the sediments around the blocks and by slide block-derived monomict pebbly-mudstones and pebbly-sandstones that are present around several slide-blocks. The slide-blocks are mainly made up of granitoids, orthogneisses, metagabbros/amphibolites, Jurassic carbonatic turbidites next to Ordovician quartz-arenites, black shales, crinoidal (Fig. 3B5) and brachiopod-bearing Devonian–Carboniferous limestones and probably Triassic red quartz-arenites (Fig. 3B4) as typical representatives of the IZ Terrane. Few blocks of serpentinites, basalts and cherts have been also recognized in the uppermost part of the sequence.

Thirty-five thin sections from the Taraklı Flysch (30 arenites and 5 rudites) were analyzed by means of a polarizing microscope. A modal analysis was performed on 19 selected medium- to coarse-grained arenites. Point counting (500 points) of arenites was performed using the Gazzi–Dickinson technique (Zuffa, 1987 and quoted references) to minimize the dependence of arenite composition on grain size. The point counting results are plotted on the triangular diagrams of Fig. 3C.

No large differences can be recognized in the framework composition of arenites from the Taraklı Flysch. They range from quartz-poor mixed arenites up to calclithites. The total framework is characterized by a mixed siliciclastic–carbonate framework composition (Fig. 3C1) where the important carbonatic extrabasinal contribution (CE, up to the 50%) led us to classify these rocks as mixed arenites (Zuffa, 1980).

The extrabasinal siliciclastic framework is characterized by a common presence of mono- and polycrystalline quartz (12÷41% of the total framework) and feldspar (6÷31%). Felsic intrusive coarse-grained rock fragments, such as granitoids, are common (0÷3% of the total framework), while low grade metamorphic rock fragments are not common and include coarse-grained gneisses, metaquartzites, fine-grained schists, and mica-schists (1÷8% of the total framework).

A striking feature of the Taraklı Flysch arenites is the widespread presence of carbonate rock fragments (Fig. 3C1). In all the studied samples, both intra- (1÷30% of the total framework) and extrabasinal (22÷48%) carbonate fragments have been recognized. Carbonate extrabasinal fragments are represented by carbonate platform derived rocks, mainly mudstones, wackestones, and grainstones of Jurassic–Early Cretaceous age (Fig. 3B6). The allochems in the grainstone fragments are peloids, ooids, minor benthic foraminifera, and undeterminable macrofossil fragments. The intrabasinal carbonate fragments are instead represented by mudstone, showing deformed and squeezed soft margins and by isolated bioclasts, mainly benthic foraminifera and macrofossils (Fig. 3B6). The presence of intrabasinal carbonate fragments became relevant in the calcareous coarse-grained turbidites. The composition in arenites of this lithofacies indicates a strong supply from a coeval carbonate platform (up to 65% of the total framework, Fig. 3C1).

The lacking of ophiolite-derived rock fragments represents one of the more striking features of Taraklı Flysch arenites. Therefore, the source areas of these sediments were mainly characterized by a continental basement made up of granitoids, metamorphic rocks, felsic volcanic rocks and the relative sedimentary covers, represented by extrabasinal non-coeval carbonate rock successions.

This source area can be related to a typical continental margin. The lithic fragments in the arenites debris and the composition of the main slide-blocks in the uppermost part of the succession seem to indicate the IZ Terrane as the most probable source area of the Taraklı Flysch.

Fifty-five samples were analyzed for the calcareous nannofossil study of the Taraklı Flysch. The analyses were performed on smear slides, prepared from unprocessed material, using a light microscope at ×1250 magnification. The taxa have been recognized following the taxonomy proposed by Perch-Nielsen (1985a, b) and Bown (1999). Several samples (19) are barren, while the investigated assemblages suffered overgrowth and dissolution. However, some information about the age of the identified lithofacies has been obtained.

In the lowermost thin-bedded turbidites lithofacies, a sample with a monogeneric assemblage (ABT04), composed of common Micula staurophora, Micula concava, and Micula sp., has been collected. This sample can be attributed to the Maastrichtian on the basis of paleoclimatic consideration. The genus Micula is considered a cold water indicator and the genus Watznaueria is related to warm water conditions. Bojar Melinte et al. (2009) record a Watznaueria/Micula crossover in Early Maastrichtian times as a result of colder water conditions. The monospecific assemblage of Micula spp. found in our sample could be related to this cold event and indirectly dated to the Early Maastrichtian. Reworking is evidenced by the sample ABT02, where the finding of Ephrolithus floralis indicates the Late Aptian–Late Cenomanian time interval. On the contrary, samples TC21 and TC22 from marls of the shaly-matrix lithofacies around the slide-blocks have been dated to the Middle Paleocene (Selandian zone NP5 of Martini, 1971) on the basis of the occurrence of Heliolithus cantabriae, Fasciculithus ulii, Fasciculithus tympaniformis, and Sphenolithus anarrhopus. The characteristics of this lithofacies resulted in a strong reworking of the nannofossil assemblages. For instance, the sample with Polycostella beckmannii and Diazomatolithus lehmanii (ABT22) can be attributed to the Polycostella beckmannii Subzone (NJ20B) of Bralower et al. (1989) attributable to the Middle Tithonian, and samples bearing Nannoconus sp., Hexalithus noeliae, H. chiastia, P. beckmannii and D. lehmanii (TC196, TC197, TC198) have been dated to the NJK Zone (Jurassic–Cretaceous boundary) of Bralower et al. (1989).

In summary, the nannofossil analyses indicate an Early Maastrichtian age for the lowermost level of the studied succession, whereas a Middle Paleocene age can be proposed for its top, i.e. the slide-block in shaly-matrix lithofacies.

The Taraklı Flysch is characterized by a complex deformation pattern, even if the whole succession is non-metamorphic. This deformation pattern is the result of two main deformation phases, referred to as D1 and D2 phases, whose structures are well identifiable in the field.

The structures of the D1 phase are represented by F1 folds that commonly display a geometry ranging from isoclinal to subisoclinal. The F1 fold axes show a NW–SE to NNW–SSE trend with steep plunges (Fig. 4A).

At the outcrop scale, the F1 folds are overprinted by a centimeter- to tens of meter-scale structures (Fig. 4A) represented by a complex association of faults, thrusts and folds, all attributed to the D2 phase (Fig. 4B and C).

The faults occur as high-angle brittle shear zones grouped into three main systems with east-west, NNW–SSE and NNE–SSW strike, respectively attributed to the S1, S2, and S3 systems (Fig. 4D). The faults of S1 and S3 systems show predominantly dextral strike-slip movements, but slip sense indicators of normal movements are also observed. On the contrary, the faults of S2 system are mainly represented by sinistral strike-slip faults. The thrusts are represented by flat shear zones with medium to low dip and NNW–SSE and ENE–WSW strike (Fig. 4C). The thrusts are characterized by both northward and southwestward senses of shear. In the field, examples of structures where the thrusts are rooted into the high angle strike-slip faults are common.

The F2 folds can be instead grouped into two main groups based on their geometry and/or relationships with faults and thrusts. The first type F2a corresponds to upright folds directly associated with steeply dipping faults (Fig. 4B). The strike of axial planes are roughly parallel to that of the related faults. These upright folds show hinge zone that may be partially cut out by the faults, whereas the vertical fold limbs are cut by an array of faults parallel to the bedding planes or cross-cutting at low angles to produce imbricate zones. The second type F2b includes the folds associated with the thrusts. The F2b folds display sub-horizontal axes and axial planes, whose strikes is roughly parallel to the mean directions of the thrusts (Fig. 4C).

The close association among folds, faults and thrusts as well as their cross-cutting relationships clearly indicates their belonging to flower structures, as suggested by the close parallelism between the strike of PA2 axial planes with those of thrusts and faults. All these structures can be regarded as related to the NAFZ system.

The structural setting of the Taraklı Flysch identified in the field can be also recognized in a north-south trending geological cross-section of the Boyalı area reported in Fig. 2. In this cross-section, asymmetric flower structures can be identified, even if the distribution of these structures is not homogeneous. Along the cross-section, the southern contact of the Taraklı Flysch with the Eocene deposits corresponds to a dextral high-angle fault whereas, at the northern edge of the cross-section, the Taraklı Flysch is separated from the IPS units by a low-angle thrust (Fig. 2).