1. Introduction
Despite numerous studies since the 1960s in Turkey, the locations, ages and geological evolution of the branches of the Eastern Mesozoic Tethys Ocean or Eastern Neotethys are disputed (see Sengör & Yılmaz, Reference Sengör and Yılmaz1981; Dercourt et al. Reference Dercourt, Zonenshain, Ricou, Kazminm, Le Pichon, Knipper, Grandjacquet, Sbortshikovm, Geyssant, Lepvrier, Pechersky, Boulin, Sibuet, Savostin, Sorokhtin, Westphal, Bazhenov, Lauer, Biju-Duval, Aubouin, Le Pichon and Monin1986; Robertson et al. Reference Robertson, Dixon, Brown, Collins, Morris, Pickett, Sharp, Ustaömer, Morris and Tarling1996; Göncüoglu, Dirik & Kozlu, Reference Göncüoglu, Dirik and Kozlu1997; Dilek et al. Reference Dilek, Thy, Hacker and Grundvig1999; Göncüoglu et al. Reference Göncüoglu, Turhan, Sentürk, Özcan, Uysal, Bozkurt, Winchester and Piper2000; Stampfli & Borel, Reference Stampfli and Borel2002; Bortolotti & Principi, Reference Bortolotti and Principi2005; Göncüoglu, Yaliniz & Tekin, Reference Göncüoglu, Yaliniz and Tekin2006; Schmid et al. Reference Schmid, Bernoulli, Fügenschuh, Matenco, Schefer, Schuster, Tischler and Ustaszewski2008; Moix et al. Reference Moix, Beccaletto, Kozur, Hochard, Rosselet and Stampfli2008; Göncüoglu, Sayit & Tekin, Reference Göncüoglu, Sayit and Tekin2010). This is mainly owing to two deficiencies. One of them is the fact that next to definite ophiolitic suture belts separating continental blocks with distinct geological histories there are also suture-like alignments or pseudo-belts of allochthonous ophiolitic bodies. The second shortcoming regards the method of ophiolite research yet realized, as it was rarely based upon a multidisciplinary approach. Several detailed petrological studies on different units from different ophiolitic bodies lack age data and vice versa. A considerable number of studies were concentrated on the larger and more or less complete ophiolitic bodies and neglected the remarkable amount of basalt–radiolarian chert associations within the mélange complexes. To overcome these problems we applied a multidisciplinary approach combining petrology and biostratigraphy on crustal remnants of the Neotethys incorporated into the mélange complexes of the Izmir–Ankara–Erzincan Suture Belt (IAESB).
As previously referred to, the location, age and geological evolution of the branch(es) of the Eastern Neotethys are disputed. In fact, most authors favour the hypothesis (presented herein) that implies the existence of two parallel and contemporaneous oceans, namely, the Izmir–Ankara Ocean, to the south, and the Intra-Pontide Ocean, to the north (see Göncüoglu et al. Reference Göncüoglu, Marroni, Sayit, Tekin, Pandolfi and Ellero2012 for an exhaustive review). In contrast, other authors favour the hypothesis that implies the existence of a single oceanic basin represented by the Intra-Pontide Ocean (e.g. Bortolotti & Principi, Reference Bortolotti and Principi2005). In this case the Intra-Pontide Suture (IPS, see below) would be the eastward continuation of the Vardar Ocean suture, whereas the IAESB would represent fragments of the oceanic lithosphere pushed southwards onto the continent by orogenic movements (as happened more westwards, in the Hellenides) (see Saccani et al. Reference Saccani, Photiades, Santato and Zeda2008 a; Bortolotti et al. Reference Bortolotti, Chiari, Marroni, Pandolfi, Principi and Saccani2013b ).
The IAESB separates two completely different units of continental crust with different origins: (a) the Tauride–Anatolide terrane representing the rifted northern margin of NW Gondwana, in the south, and (b) the Sakarya Composite Terrane comprising amalgamated oceanic and continental bodies of Variscan and Cimmerian origin and their alpine platform, in the north. To the north of the Sakarya Composite Terrane another suture, the IPS, represents the boundary towards the Eurasian Istanbul–Zonguldak terrane (Fig. 1).
Figure 1. The main tectonic zones of Turkey (modified after Sengör & Yılmaz, Reference Sengör and Yılmaz1981; Göncüoglu et al. Reference Göncüoglu, Marroni, Sayit, Tekin, Pandolfi and Ellero2012, modified).
The IAESB represents remnants of the Vardar–Izmir–Ankara–Erzincan–Lesser Caucasus Ocean in Turkey. This ocean is considered the main northern branch of the Neotethys, whereas the Intra-Pontide oceanic basin to the north of it is a matter of debate (see above and, for a brief discussion, Göncüoglu et al. Reference Göncüoglu, Marroni, Sayit, Tekin, Pandolfi and Ellero2012 and Tekin et al. Reference Tekin, Göncüoğlu, Pandolfi and Marroni2012). In NW Turkey, the IPS is clearly delineated by the presence of ophiolitic melanges between the Sakarya and the Istanbul–Zonguldak continental plates (e.g. Robertson & Ustaömer, Reference Robertson and Ustaömer2004; Göncüoglu et al. Reference Göncüoglu, Gürsu, Tekin and Köksal2008; Akbayram, Okay & Satir, Reference Akbayram, Okay and Satir2012). By the wedging out of the Sakarya Composite Terrane in northern Central Anatolia, however, the IPS belt juxtaposes with the IAESB along splays of the North Anatolian Shear Zone (e.g. Ellero et al. Reference Ellero, Ottria, Marroni, Pandolfi and Göncüoğlu2015 a) and the neotectonic strike-slip system generated owing to the Tertiary indentation and anticlockwise rotation of the Kırşehir Block (e.g. Cemen et al. Reference Cemen, Göncüoğlu, Erler, Kozlu and Perinçek1993; Kaymakci et al. Reference Kaymakci, Duermeijer, Langereis, White and van Dijk2003). Towards the east and in NE Anatolia, the IAESB is again in its accustomed position between the Anatolides and the units of the Sakarya Composite Terrane (Topuz et al. Reference Topuz, Çelik, Şengör, Altıntaş, Zack, Rolland and Barth2013a ,Reference Topuz, Göçmengil, Rolland, Çelik, Zack and Schmitt b ; Parlak et al. Reference Parlak, Çolakoğlu, Dönmez, Sayak, Yildirim, Türkel, Odabaşi, Robertson, Parlak and Unlugenc2013; Robertson et al. Reference Robertson, Parlak, Ustaömer, Taslı, İnan, Dumitrica and Karaoğlan2013). The vanishing of the IPS belt in the Central Pontides by the North Anatolian Shear Zone during the neotectonic period led some authors to completely ignore the IPS and hence the presence of an oceanic basin (e.g. Elmas & Yiğitbaş, Reference Elmas and Yiğitbaş2001, Reference Elmas and Yiğitbaş2005).
The IAESB is composed of a number S-verging tectonic slices or giant slide blocks representing (1) incomplete sequences of various portions of the Izmir–Ankara–Erzincan oceanic lithosphere (e.g. Floyd et al. Reference Floyd, Göncüoglu, Winchester, Yalınız, Bozkurt, Winchester and Piper2000; Göncüoglu et al. Reference Göncüoglu, Turhan, Sentürk, Özcan, Uysal, Bozkurt, Winchester and Piper2000; Göncüoglu, Reference Göncüoglu2011); (2) an accretionary prism, known as the ‘Ankara Mélange’ (Bailey & McCallien, Reference Bailey and McCallien1953) including blocks derived from different parts of the oceanic basin; (3) imbricated slices of successions formed in a foredeep basin associated with the emplacement of the oceanic material onto the Tauride–Anatolide passive margin.
This study is focused on slide blocks including oceanic lavas associated with pelagic sediments within the accretionary prism and the foredeep basin successions in the eastern part of the Ankara Mélange (for the preliminary data see Bortolotti et al. Reference Bortolotti, Chiari, Göncüoglu, Marcucci, Principi, Tekin, Saccani and Tassinari2013 a). This paper aims to draw a more comprehensive picture of the different basaltic rock types erupted in the Neotethys from Middle Jurassic to Early Cretaceous times. To this purpose, a detailed petrological and geochemical characterization of the volcanic rocks and a detailed biochronological investigation of the associated radiolarian cherts will be carried out. Similar multidisciplinary studies were applied to other parts of the IAESB between the Aegean coast and the western part of the Ankara Mélange (e.g. Bragin & Tekin, Reference Bragin and Tekin1996; Yaliniz, Göncüoglu & Floyd, Reference Yaliniz, Göncüoglu, Floyd, Bozkurt, Winchester and Piper2000; Göncüoglu, Tekin & Turhan, Reference Göncüoglu, Tekin and Turhan2001; Rojay, Yaliniz & Altiner, Reference Rojay, Yaliniz and Altiner2001; Tekin, Göncüoglu & Turhan, Reference Tekin, Göncüoglu and Turhan2002; Gökten & Floyd, Reference Gökten and Floyd2007; Tekin & Göncüoglu, Reference Tekin and Göncüoglu2007, Reference Tekin and Göncüoglu2009; Göncüoglu et al. Reference Göncüoglu, Gürsu, Tekin and Köksal2008; Göncüoglu, Sayit & Tekin, Reference Göncüoglu, Sayit and Tekin2010; Tekin et al. Reference Tekin, Göncüoğlu, Pandolfi and Marroni2012; Moix & Goričan, Reference Moix and Goričan2013; Göncüoglu et al. Reference Göncüoglu, Tekin, Sayit, Bedi and Uzunçimen2015; Soycan, Erdoğan & Konak, Reference Soycan, Erdoğan and Konak2015). These studies reported Middle Triassic to Cretaceous ages for radiolarian cherts stratigraphically associated with basalts showing, in turn, different tectonomagmatic settings of formation, such as mid-ocean ridge, seamount, fore-arc and back-arc. Our new data from the eastern Ankara Mélange aim to interpret the tectonomagmatic processes and their timing, which will be useful for improving the reconstruction of the geodynamic history of the Neotethys Ocean mainly spanning the Jurassic to Early Cretaceous time interval.
2. Geological background
The studied portion of the IAESB was characterized by the pre-Eocene development of a huge accretionary prism (for details see Rojay, Reference Rojay2013), which formed between the Sakarya Composite Terrane to the north and the southerly Kütahya–Bolkardag Belt of the Anatolides, in the western part, and the Central Anatolian Crystalline Complex (CACC), in the eastern part. The first juxtaposition of these oceanic and continental units occurred in latest Cretaceous – Paleocene time. However, compression and related thrusting lasted until Miocene time (Kocyigit et al. Reference Kocyigit, Türkmenoglu, Beyhan, Kaymakçı and Akyol1995).
We sampled some ophiolitic outcrops from the IAESB east of Ankara (Fig. 1), where the concave E–W trend of the belt between Izmir and Ankara makes a sharp turn towards the NNW, caused by the Tertiary indentation (e.g. Cemen et al. Reference Cemen, Göncüoğlu, Erler, Kozlu and Perinçek1993; Kaymakci et al. Reference Kaymakci, Duermeijer, Langereis, White and van Dijk2003) and anticlockwise rotation of the Kırşehir Block.
The Sakarya Unit marks the active margin of the Izmir–Ankara–Erzincan Ocean, and was thrust over the IAESB both to the NNE of Ankara and to the SE of Çorum (Fig. 2). It consists of a composite terrane comprising a Variscan arc complex and its Permian platform, as well as the Palaeotethyan subduction–accretion prism (the Triassic Karakaya Complex; Göncüoglu et al. Reference Göncüoglu, Turhan, Sentürk, Özcan, Uysal, Bozkurt, Winchester and Piper2000; Okay & Göncüoglu, Reference Okay and Göncüoglu2004). It has a Jurassic to ?Upper Cretaceous cover that belonged to a N-facing passive continental margin but was transformed into an active margin by the northward subduction of the Izmir–Ankara–Erzincan oceanic lithosphere.
Figure 2. Sketch geological map of the study area, with the location of the sampled sections (after Bortolotti et al. Reference Bortolotti, Chiari, Göncüoglu, Marcucci, Principi, Tekin, Saccani and Tassinari2013 a, modified and Yolsal-Çevikbilen et al. Reference Yolsal-Çevikbilen, Biryol, Beck, Zandt, Taymaz, Adiyaman and Özacar2012). EFZ – Eldivan Fault Zone; ODFZ – Orta–Devrez Fault Zone; KFZ – Kızılırmak Fault Zone; ESFZ – Ezine Pazarı–Sungurlu Fault Zone; AFZ – Alaca Fault Zone; CF – Çekerek Fault.
The Karakaya Complex NNW of Ankara comprises low-grade metamorphic greywackes associated with Permian and Carboniferous limestone blocks and ocean-island-type volcanic rocks with Carnian radiolarian cherts (e.g. Sayit, Tekin & Göncüoglu, Reference Sayit, Tekin and Göncüoglu2011). Unconformably overlying them there is a Mesozoic cover that comprises, from bottom to top, upper Lower Jurassic – Middle Jurassic neritic limestones, Upper Jurassic – Lower Cretaceous (e.g. Altiner et al. Reference Altiner, Koçyiğit, Farinacci, Nicosia and Conti1991) pelagic limestones and Upper Cretaceous turbidites. The oldest common overstep sequence on the IAESB (and the Sakarya Composite Terrane) comprises Upper Paleocene lagoonal sediments, which occur as discontinuous outcrops within the thrust zone between these two units (Göncüoglu et al. Reference Göncüoglu, Turhan, Sentürk, Özcan, Uysal, Bozkurt, Winchester and Piper2000).
The northern-central part of the sampled area has been recently evaluated as the North Anatolian Shear Zone (Ellero et al. Reference Ellero, Ottria, Marroni, Pandolfi and Göncüoğlu2015 a,Reference Bortolotti, Chiari, Marroni, Pandolfi, Principi and Saccani b ). The North Anatolian Shear Zone corresponds to a complex deformation zone where the strain is partitioned in a system of faults, folds and thrusts leading to high-angle faults bounding E–W-elongated blocks and pull-apart basins. The zone is more than 100km wide and comprises several kilometres long, lens-shaped tectonic inlayers belonging to the mélanges of the IPS belt (e.g. Göncüoglu et al. Reference Göncüoglu, Marroni, Pandolfi, Ellero, Ottria, Catanzariti, Tekin and Sayit2014), the Sakarya Composite Terrane (mainly the Upper Jurassic – Lower Cretaceous Sogukcam Limestone, e.g. Catanzariti et al. Reference Catanzariti, Ellero, Göncüoglu, Marroni, Ottria and Pandolfi2013), a Late Cretaceous island arc (Berber, Göncüoglu & Sayit, Reference Berber, Göncüoglu, Sayit, Begiraj, Ionescu, Christofides, Uta, Goga and Marku2014) belonging to the IPS belt, a Late Cretaceous continental arc (Ellero et al. Reference Bortolotti, Chiari, Marroni, Pandolfi, Principi and Saccani2015b ) and the ophiolitic mélanges of the IAESB. The active main strand of the right-lateral North Anatolian Transform Fault is located in the centre of this mega-shear zone, where lateral displacements ranging from 30 to 120km have been proposed in the literature (e.g. Hubert-Ferrari et al. Reference Hubert-Ferrari, Armijo, King, Meyer and Barka2002 and references therein).
The metamorphic rocks of the Sakarya Composite Terrane (the ‘Tokat Massif’, Yilmaz et al. Reference Yilmaz, Serdar, Genc, Yigitbas, Gurer, Elmas, Yildirim, Bozcu and Gurpınar1997), around Çorum, east of the sampled area (Fig. 2), are tectonically overlying the ophiolitic mélange of the IAESB. In this area, both units are in turn unconformably overlain by Middle Eocene rocks. Unfortunately, the primary relations between the Sakarya and IAESB rocks are obscured by intensive Oligocene strike-slip faulting (Fig. 2).
Structurally underlying the IAESB units, the continental crust that was palinspastically located to the south, and west of Ankara, represents the northern rim of the Tauride–Anatolide Platform. It mainly includes high-pressure–low-temperature metamorphosed tectonic slices (e.g. Okay & Tüysüz, Reference Okay, Tüysüz, Durand, Jolivet, Horváth and Séranne1999) whose lithostratigraphic sequences are similar to the Palaeozoic–Mesozoic slope-type successions of the Anatolides (Göncüoglu, Reference Göncüoglu2011). In particular, in the sampled area (Fig. 2), the IAESB units were thrust onto the CACC (Yaliniz, Floyd & Göncüoglu, Reference Yaliniz, Floyd and Göncüoglu1996; Yaliniz, Göncüoglu & Özkan-Altiner, Reference Yaliniz, Göncüoglu and Özkan-Altiner2000), which comprises high-temperature–medium-pressure metamorphic successions similar to the Palaeozoic–Mesozoic Tauride–Anatolide Platform. In turn, they were overthrust by Turonian supra-subduction ophiolites: the Central Anatolian Ophiolites (Yaliniz, Göncüoglu & Floyd, Reference Yaliniz, Göncüoglu, Floyd, Bozkurt, Winchester and Piper2000), which represent remnants of the Izmir–Ankara–Erzincan Ocean. Late Campanian granitoids intrude both the basement rocks and the overlying ophiolite units (Köksal & Göncüoglu, Reference Köksal and Göncüoglu2008), indicating a Late Cretaceous age for the inception of obduction. The oldest overstep sequences in this area are again post-Maastrichtian pre-Lower Eocene (Gülyüz et al. Reference Gülyüz, Kaymakci, Meijers, van Hinsbergen, Lefebvre, Vissers, Bart, Hendriks and Peynircioglu2013) red conglomerates, testifying to a Paleocene age for the main juxtaposition of the CACC and IAESB mélanges.
In the central part of the sampled area, the IAESB rocks are covered by the Paleocene–Eocene post-orogenic marine clastic-volcaniclastic rocks, carbonates and volcanic rocks of the Çankırı Basin, which, in turn, are partly covered by Neogene sediments (e.g. Göncüoglu, Reference Göncüoglu, Sirel and Yazgan1992). Within the basin the anticlockwise rotation of the CACC induced the formation of several NE–SW-trending right-lateral faults and, at its western margin, important left-lateral faults (Fig. 2). When restored, the IAESB palinspastically follows also a roughly E–W trend and the present distortion of the main tectonic units is re-established.
To attempt a reconstruction of the thickness of the IAESB rocks that compose the subducted and accreted remnants of the Izmir–Ankara–Erzincan oceanic lithosphere, together with island arc and sedimentary rocks of several Late Cretaceous – Paleocene piggy-back basins (e.g. Cater et al. Reference Cater, Hanna, Ries and Turner1991; Erdogan, Akay & Ugur, Reference Erdogan, Akay and Ugur1996), geophysical data would be necessary, but at present these date are not available.
3. Description of the sampled sections
All the sampled sequences pertain to the radiolarite–basalt blocks included in the Ankara Mélange. For the numbering of the sections we referred to Bortolotti et al. (Reference Bortolotti, Chiari, Göncüoglu, Marcucci, Principi, Tekin, Saccani and Tassinari2013 a). It is worth noting that in the present paper we do not report the samples from Section 6 of Late Triassic age of Bortolotti et al. (Reference Bortolotti, Chiari, Göncüoglu, Marcucci, Principi, Tekin, Saccani and Tassinari2013 a).
3.a. Section 1
In a massif of pillow basalts along the Elmadag–Kırıkkale road, several metric intercalations of siliceous shales with scattered radiolarian cherts crop out.
We sampled one of these intercalations and the associated basalts (Fig. 3b, 39°55.023′N, 33°21.989′E).
Figure 3. (a) Serpentinite mélange near Beynam. (b) Pillow basalts along the road from Elmadag to Kırıkkale (Section 1). Length of hammer for scale is 32 cm. (c) Overturned sequence of basalts and radiolarian cherts along the road from Çorum to Alaca (Section 5). (d) Outcrop of basalts and radiolarian cherts along the road from Iskilip to Tosya (Section 7). Modified from Bortolotti et al. (Reference Bortolotti, Chiari, Göncüoglu, Marcucci, Principi, Tekin, Saccani and Tassinari2013 a).
Samples: TU10.4, radiolarian chert; TU10.6, TU10.10, basalts below the radiolarian chert intercalation; TU10.9, basalt ~20 m above the radiolarian chert intercalation.
3.b. Section 2
In a massif of mainly pillow basalts along the Sorgun–Çekerek road, thin layers of radiolarian cherts crop out.
We sampled one of these layers and the associated basalts (39°54.889′N, 35°18.063′E); one radiolarian chert sample was collected along the road in an uncertain stratigraphic position.
Samples: TU10.11, radiolarian chert with uncertain stratigraphic position; TU10.12, radiolarian chert; TU10.14 basalt at the contact with the radiolarites; TU10.15, TU10.16, TU10.17, basalts collected several metres from the radiolarites.
3.c. Section 3
In a basalt block immediately north of Gökdere village, along the road, a sequence with radiolarian cherts crops out.
Samples: TU10.19, basalt at the contact with the radiolarian cherts (39°59.924′N, 35°24.272′E); TU10.22, TU10.23, basalts collected some metres southwards (39°59.921′N, 35°24.274′E), respectively 3.5 and 2 m from the contact with the radiolarites; TU10.28, radiolarian chert collected ~22 m from the basalts (this sequence is intensely folded).
3.d. Section 4
An overturned sequence of basalts and radiolarian cherts with a sheared contact, along the road ~3km southeast of Bogazkale (40°00.377′N, 34° 38.762′E).
Samples: TU10.29, TU10.30, TU10.31, radiolarian cherts collected ~20cm above the sheared level; 60cm of sheared argillites and cherts separates the basalts from the above samples; TU10.32, basalt at the contact with the sheared level; TU10.33, basalt collected some metres above.
3.e. Section 5
Big quarry cut by small faults that pull the contact between the basalts–cherts up and down. The sequence could be overturned: the cherts lie under the basalts; the contact is very clear with pockets of argillites and cherts in the mainly massif basalts. The sampled quarry is along the Çorum to Alaca road, south of Küre village (Fig. 3c, 40°15.861′N, 34°48.187′E).
Samples: TU10.34, basalt, sample collected ~6 m from the cherts; TU10.35, TU10.36, TU10.37, samples collected in a radiolarian chert–argillite sequence from the base to 40cm up. A small normal fault shifts the sequence by some metres. TU10.38, radiolarian chert, collected beyond the fault, ~30cm from the basalt sample TU10.40; TU10.40, basalt at the contact with the radiolarian cherts; TU10.39, basalt nearby sample TU10.40.
3.f. Section 7
Small outcrop of basalts and cherts along the Iskilip–Tosya road, 1.2km before the junction to Hacihalil village (Fig. 3d, 40°53.378′N, 34°20.756′E).
Samples: TU10.45, radiolarian chert nodule in the basalts; TU10.46, basalt.
3.g. Section 8
Large body of breccia(?) with basalts, microgabbros, serpentinites and crystallized limestones in a cherty matrix. We sampled the breccia in two close localities northeast of Yukariöz village.
Samples: First locality (40°51.097′N, 33°50.723′E), TU10.47, TU10.48, radiolarian cherts; TU10.49a, TU10.49b, basalts. Second locality (40°51.527′N, 33°49.577′E), TU10.51, M10 radiolarian cherts.
3.h. Section 9
Pillow basalts enveloped in reddish limestones, Eldivan–Sabanozu road (40°31.488′N, 33°28.106′E).
Samples: TU10.52, TU10.53, basalts; TU10.54, TU10.55, limestones.
4. Biostratigraphy
The radiolarian samples were etched with hydrochloric and hydrofluoric acid following the method proposed by Dumitrica (Reference Dumitrica1970), Pessagno & Newport (Reference Pessagno and Newport1972), Baumgartner et al. (Reference Baumgartner, Bjørklund, Caulet, De Wever, Kellogg, Labracherie, Nakaseko, Nishimura, Schaaf, Schimdt-Effing and Yao1981) and De Wever (Reference De Wever1982). The examined samples yielded radiolarians with moderate preservation and the principal marker taxa are illustrated in Figures 4 and 5.
Figure 4. Scale bar = 50 µm. (a) Archaeodictyomitra sp. cf. A. lacrimula (Foreman), Section 1, TU10.4; (b) Aurisaturnalis variabilis variabilis (Squinabol), Section 1, TU10.4; (c) Hemicryptocapsa sp. cf. H. capita Tan, Section 1, TU10.4; (d) Thanarla brouweri (Tan), Section 1, TU10.4; (e) Eoxitus (?) sp., Section 2, TU10.11; (f) Stichomitra (?) takanoensis Aita, Section 2, TU10.11; (g) Hiscocapsa sp. Section 2, TU10.12; (h) Mirifusus sp. cf. M. guadalupensis Pessagno, Section 2, TU10.12; (i) Stichomitra (?) takanoensis Aita, Section 2, TU10.12; (j) Emiluvia sp. cf. E. ordinaria Ozvoldova, Section 3, TU10.28; (k) Fultacapsa sphaerica (Ozvoldova), Section 3, TU10.28; (l) Podocapsa amphitreptera Foreman, Section 3, TU10.28; (m) Spinosicapsa (?) sp., Section 3, TU10.28; (n) Archaeodictyomitra lacrimula (Foreman), Section 4, TU10.29; (o) Archaeodictyomitra mitra Dumitrica, Section 4, TU10.29; (p) Dicerosaturnalis trizonalis (Rüst), Section 4, TU10.29; (q) Pantanellium sp. cf. P. squinaboli (Tan), Section 4, TU10.29; (r) Thanarla sp. cf. T. gutta Jud, Section 4, TU10.29; (s) Archaeodictyomitra sp., Section 4, TU10.30; (t) Holocryptocanium barbui Dumitrica, Section 4, TU10.30; (u) Pseudodictyomitra lanceloti Schaaf, Section 4, TU10.30; (v) Thanarla sp. cf. T. pacifica Nakaseko & Nishimura, Section 4, TU10.30.
Figure 5. Scale bar = 50 µm. (a) Archaeodictyomitra lacrimula (Foreman), Section 4, TU10.31; (b) Aurisaturnalis carinatus perforatus Dumitrica & Dumitrica-Jud, Section 4, TU10.31; (c) Thanarla brouweri (Tan), Section 4, TU10.31; (d) Eucyrtidiellum pyramis (Aita), Section 5, TU10.35; (e) Podocapsa amphitreptera Foreman, Section 5, TU10.35; (f) Cinguloturris cylindra Kemkin & Rudenko, Section 5, TU10.36; (g) Emiluvia sp. cf. E. ordinaria Ozvoldova, Section 5, TU10.36; (h) Eucyrtidiellum pyramis (Aita), Section 5, TU10.36; (i) Loopus primitivus (Matsuoka & Yao), Section 5, TU10.36; (j) Ristola cretacea (Baumgartner), Section 5, TU10.36; (k) Podocapsa amphitreptera Foreman, Section 5, TU10.37; (l) Praeconosphaera (?) sphaeroconus (Rüst), Section 5, TU10.37; (m) Archaeodictyomitra sp. cf. A. excellens (Tan), Section 5, TU10.38; (n) Podocapsa amphitreptera Foreman, Section 5, TU10.38; (o) Saitoum sp. cf. S. elegans De Wever, Section 5, TU10.38; (p) Zhamoidellum ovum Dumitrica, Section 5, TU10.38; (q) Archaeodictyomitra sp. cf. A. coniforma Dumitrica, Section 7, TU10.45; (r) Cryptamphorella clivosa (Aliev), Section 7, TU10.45; (s) Praeconosphaera (?) sphaeroconus (Rüst), Section 7, TU10.45; (t) Archaeodictyomitra excellens (Tan), Section 8, TU10.47; (u) Cana septemporatus (Parona), Section 8, TU10.47; (v) Halesium sp. cf. H. palmatum Dumitrica, Section 8, TU10.47; (w) Archaeodictyomitra lacrimula (Foreman), Section 8, TU10.51; (x) Pseudoeucyrtis sp. cf. P. hanni (Tan) sensu O'Dogherty (Reference O'Dogherty1994), Section 8, TU10.51.
For the taxonomy and ranges of the Late Jurassic – Early Cretaceous principal marker taxa we refer to Aliev (Reference Aliev1967), Kozur (Reference Kozur1985), Kawabata (Reference Kawabata1988), Aita & Okada (Reference Aita and Okada1986), Goričan (Reference Goričan1994), O'Dogherty (Reference O'Dogherty1994), Baumgartner et al. (Reference Baumgartner, Bartolini, Carter, Conti, Cortese, Danelian, De Wever, Dumitrica, Dumitrica-Jud, Goričan, Guex, Hull, Kito, Marcucci, Matsuoka, Murchey, O'Dogherty, Savary, Vishnevskaya, Widz, Yao, Baumgartner, O'Dogherty, Goričan, Urquhart, Pillevuit and De Wever1995 a,Reference Baumgartner, O'Dogherty, Goričan, Dumitrica-Jud, Dumitrica, Pillevuit, Urquhart, Matsuoka, Danelian, Bartolini, Carter, De Wever, Kito, Marcucci, Steiger, Baumgartner, O'Dogherty, Goričan, Urquhart, Pillevuit and De Wever b ), Dumitrica & Dumitrica-Jud (Reference Dumitrica and Dumitrica-Jud1995), Bak (Reference Bak1996, Reference Bak1999), Dumitrica, Immenhauser & Dumitrica-Jud (Reference Dumitrica, Immenhauser and Dumitrica-Jud1997), Chiari, Marcucci & Prela (Reference Chiari, Marcucci and Prela2004), Danelian et al. (Reference Danelian, Tsikos, Gardin, Baudin, Bellier and Emmanuel2004), Smuc & Goričan (Reference Smuc and Goričan2005), Filippov & Kemkin (Reference Filippov and Kemkin2005), Chiari, Cobianchi & Picotti (Reference Chiari, Cobianchi and Picotti2007), Danelian (Reference Danelian2008), Dumitrica & Zügel (Reference Dumitrica and Zügel2008), O'Dogherty et al. (Reference O'Dogherty, Carter, Dumitrica, Goričan, De Wever, Bandini, Baumgartner and Matsuoka2009), Robin et al. (Reference Robin, Goričan, Guillocheau, Razin, Dromart, Mosaffa, Leturmy and Robin2010), Bandini et al. (Reference Bandini, Baumgartner, Flores, Dumitrica and Jackett2011), Goričan, Pavšič & Rožič (Reference Goričan, Pavšič and Rožič2012) and Chiari et al. (Reference Chiari, Bortolotti, Marcucci, Photiades, Principi and Saccani2012).
From the analysed cherts the following radiolarian assemblages and ages were obtained:
4.a. Section 1
TU10.4. Angulobracchia portmanni Baumgartner, Archaeodictyomitra mitra Dumitrica, Archaeodictyomitra sp. cf. A. lacrimula (Foreman) (Fig. 4a), Archaeodictyomitra sp. cf. A. mitra Dumitrica, Archaeodictyomitra sp., Archaeodictyomitra (?) sp., Aurisaturnalis variabilis variabilis (Squinabol) (Fig. 4b), Hemicryptocapsa sp. cf. H. capita Tan (Fig. 4c), Neorelumbra (?) sp., Pseudodictyomitra sp., Pseudodictyomitra (?) sp., Pyramispongia (?) sp., Tethysetta (?) sp., Thanarla brouweri (Tan) (Fig. 4d), Thanarla sp., Torculum (?) sp., Xitus sp. cf. X. robustus Wu, Xitus sp.
Age. Late Valanginian to late Hauterivian (UAZ. 17–20; UAZones after Baumgartner et al. Reference Baumgartner, Bartolini, Carter, Conti, Cortese, Danelian, De Wever, Dumitrica, Dumitrica-Jud, Goričan, Guex, Hull, Kito, Marcucci, Matsuoka, Murchey, O'Dogherty, Savary, Vishnevskaya, Widz, Yao, Baumgartner, O'Dogherty, Goričan, Urquhart, Pillevuit and De Wever1995 a) based on the presence of Aurisaturnalis variabilis variabilis (Squinabol). Ranges after Baumgartner et al. (Reference Baumgartner, O'Dogherty, Goričan, Dumitrica-Jud, Dumitrica, Pillevuit, Urquhart, Matsuoka, Danelian, Bartolini, Carter, De Wever, Kito, Marcucci, Steiger, Baumgartner, O'Dogherty, Goričan, Urquhart, Pillevuit and De Wever1995b ).
4.b. Section 2
TU10.11. Eoxitus (?) sp. (Fig. 4e), Praewilliriedellum sp. cf. P. japonicum (Yao), Stichomitra (?) takanoensis Aita (Fig. 4f), Striatojaponocapsa (?) sp.
Age. Early–middle Bajocian to late Bathonian–early Callovian (UAZ. 3–7; UAZones after Baumgartner et al. Reference Baumgartner, Bartolini, Carter, Conti, Cortese, Danelian, De Wever, Dumitrica, Dumitrica-Jud, Goričan, Guex, Hull, Kito, Marcucci, Matsuoka, Murchey, O'Dogherty, Savary, Vishnevskaya, Widz, Yao, Baumgartner, O'Dogherty, Goričan, Urquhart, Pillevuit and De Wever1995 a) based on the presence of Stichomitra (?) takanoensis Aita. Range after Baumgartner et al. (Reference Baumgartner, O'Dogherty, Goričan, Dumitrica-Jud, Dumitrica, Pillevuit, Urquhart, Matsuoka, Danelian, Bartolini, Carter, De Wever, Kito, Marcucci, Steiger, Baumgartner, O'Dogherty, Goričan, Urquhart, Pillevuit and De Wever1995b ).
TU10.12. Hiscocaspsa sp. (Fig. 4g), Mirifusus sp. cf. M. guadalupensis Pessagno (Fig. 4h), Parahsuum sp., Parahsuum (?) sp., Praewilliriedellum sp. cf. P. japonicum (Yao), Praewilliriedellum (?) sp., Pseudodictyomitra (?) sp., Stichomitra (?) takanoensis Aita (Fig. 4i), Transhsuum sp., Transhsuum (?) sp., Tritrabs (?) sp.
Age. Early–middle Bajocian to late Bathonian–early Callovian (UAZ. 3–7; UAZones after Baumgartner et al. Reference Baumgartner, Bartolini, Carter, Conti, Cortese, Danelian, De Wever, Dumitrica, Dumitrica-Jud, Goričan, Guex, Hull, Kito, Marcucci, Matsuoka, Murchey, O'Dogherty, Savary, Vishnevskaya, Widz, Yao, Baumgartner, O'Dogherty, Goričan, Urquhart, Pillevuit and De Wever1995 a) based on the presence of Stichomitra (?) takanoensis Aita. Range after Baumgartner et al. (Reference Baumgartner, O'Dogherty, Goričan, Dumitrica-Jud, Dumitrica, Pillevuit, Urquhart, Matsuoka, Danelian, Bartolini, Carter, De Wever, Kito, Marcucci, Steiger, Baumgartner, O'Dogherty, Goričan, Urquhart, Pillevuit and De Wever1995b ).
4.c. Section 3
TU.10.28. Acaeniotylopsis sp., Crolanium (?) sp., Emiluvia sp. cf. E. ordinaria Ozvoldova (Fig. 4j), Eoxitus (?) sp., Fultacapsa sphaerica (Ozvoldova) (Fig. 4k), Mirifusus sp. cf. M. guadalupensis Pessagno, Mirifusus sp., Podocapsa amphitreptera Foreman (Fig. 4l), Spinosicapsa sp. cf. S. triacantha (Fischli), Spinosicapsa (?) sp. (Fig. 4m), Spongocapsula sp., Svinitzium sp., Svinitzium (?) sp., Triactoma (?) sp.
Age. Middle–late Oxfordian to late Kimmeridgian–early Tithonian (UAZ. 9–11; UAZones after Baumgartner et al. Reference Baumgartner, Bartolini, Carter, Conti, Cortese, Danelian, De Wever, Dumitrica, Dumitrica-Jud, Goričan, Guex, Hull, Kito, Marcucci, Matsuoka, Murchey, O'Dogherty, Savary, Vishnevskaya, Widz, Yao, Baumgartner, O'Dogherty, Goričan, Urquhart, Pillevuit and De Wever1995 a) based on the occurrence of Podocapsa amphitreptera Foreman with Fultacapsa sphaerica (Ozvoldova). Ranges after Baumgartner et al. (Reference Baumgartner, O'Dogherty, Goričan, Dumitrica-Jud, Dumitrica, Pillevuit, Urquhart, Matsuoka, Danelian, Bartolini, Carter, De Wever, Kito, Marcucci, Steiger, Baumgartner, O'Dogherty, Goričan, Urquhart, Pillevuit and De Wever1995b ).
4.d. Section 4
TU10.29. Archaeodictyomitra lacrimula (Foreman) (Fig. 4n), Archaeodictyomitra mitra Dumitrica (Fig. 4o), Archaeodictyomitra sp., Dicerosaturnalis trizonalis (Rüst) (Fig. 4p), Holocryptocanium sp. cf. H. barbui Dumitrica, Orbiculiformella (?) sp., Pantanellium sp. cf. P. squinaboli (Tan) (Fig. 4q), Praeconosphaera (?) sp. cf. P. (?) multiconus Yang, Praeconosphaera (?) sp., Praewilliriedellum sp. cf. P. japonicum (Yao), Pseudoeucyrtis sp., Thanarla brouweri (Tan) sensu O'Dogherty (Reference O'Dogherty1994), Thanarla sp. cf. T. gutta Jud (Fig. 4r), Thanarla sp.
Age. Early–early late Berriasian to middle Aptian–early Albian (UAZ. 14 – Costata Subzone of Turbocapsula Zone; zone after O'Dogherty, Reference O'Dogherty1994; UAZone after Baumgartner et al. Reference Baumgartner, Bartolini, Carter, Conti, Cortese, Danelian, De Wever, Dumitrica, Dumitrica-Jud, Goričan, Guex, Hull, Kito, Marcucci, Matsuoka, Murchey, O'Dogherty, Savary, Vishnevskaya, Widz, Yao, Baumgartner, O'Dogherty, Goričan, Urquhart, Pillevuit and De Wever1995 a) based on the occurrence of Archaeodictyomitra lacrimula (Foreman) and Archaeodictyomitra mitra Dumitrica. Ranges after Baumgartner et al. (Reference Baumgartner, O'Dogherty, Goričan, Dumitrica-Jud, Dumitrica, Pillevuit, Urquhart, Matsuoka, Danelian, Bartolini, Carter, De Wever, Kito, Marcucci, Steiger, Baumgartner, O'Dogherty, Goričan, Urquhart, Pillevuit and De Wever1995b ), O'Dogherty (Reference O'Dogherty1994), Dumitrica, Immenhauser & Dumitrica-Jud (Reference Dumitrica, Immenhauser and Dumitrica-Jud1997) and Bandini et al. (Reference Bandini, Baumgartner, Flores, Dumitrica and Jackett2011).
TU10.30. Archaeodictyomitra sp. cf. A. communis (Squinabol), Archaeodictyomitra sp. cf. A. lacrimula (Foreman), Archaeodictyomitra sp. (Fig. 4s), Archaeodictyomitra (?) sp., Emiluvia (?) sp., Eoxitus sp., Hiscocapsa (?) sp., Holocryptocanium barbui Dumitrica (Fig. 4t), Holocryptocanium sp. cf. H. barbui Dumitrica, Pantanellium sp., Praeconosphaera (?) sp., Pseudodictyomitra lanceloti Schaaf (Fig. 4u), Pseudodictyomitra sp. cf. P. lanceloti Schaaf, Thanarla sp. aff. T. brouweri (Tan), Thanarla sp. cf. T. pacifica Nakaseko & Nishimura (Fig. 4v), Thanarla sp. cf. T. praeveneta Pessagno, Thanarla sp., Thanarla (?) sp.
Age. Late Hauterivian to middle Aptian–early Albian (UAZ. 20 – Costata Subzone of Turbocapsula Zone; zone after O'Dogherty, Reference O'Dogherty1994; UAZone after Baumgartner et al. Reference Baumgartner, Bartolini, Carter, Conti, Cortese, Danelian, De Wever, Dumitrica, Dumitrica-Jud, Goričan, Guex, Hull, Kito, Marcucci, Matsuoka, Murchey, O'Dogherty, Savary, Vishnevskaya, Widz, Yao, Baumgartner, O'Dogherty, Goričan, Urquhart, Pillevuit and De Wever1995 a) based on the presence of Pseudodictyomitra lanceloti Schaaf. Range after Baumgartner et al. (Reference Baumgartner, O'Dogherty, Goričan, Dumitrica-Jud, Dumitrica, Pillevuit, Urquhart, Matsuoka, Danelian, Bartolini, Carter, De Wever, Kito, Marcucci, Steiger, Baumgartner, O'Dogherty, Goričan, Urquhart, Pillevuit and De Wever1995b ) and Bandini et al. (Reference Bandini, Baumgartner, Flores, Dumitrica and Jackett2011).
TU10.31. Angulobracchia (?) sp., Archaeodictyomitra lacrimula (Foreman) (Fig. 5a), Archaeodictyomitra mitra Dumitrica, Archaeodictyomitra sp. cf. A. lacrimula (Foreman), Archaeodictyomitra sp., Aurisaturnalis carinatus perforatus Dumitrica & Dumitrica-Jud (Fig. 5b), Dicerosaturnalis trizonalis (Rüst), Halesium (?) sp., Pantanellium sp., Praeconosphaera (?) sp. cf. P. (?) multiconus Yang, Praeconosphaera (?) sp., Praexitus sp., Pseudodictyomitra lanceloti Schaaf, Pseudodictyomitra sp. cf. P. lanceloti Schaaf, Pseudodictyomitra sp., Pseudodictyomitra (?) sp., Spinosicapsa (?) sp., Thanarla brouweri (Tan) (Fig. 5c), Thanarla sp. cf. T. brouweri (Tan), Thanarla sp. cf. T. pacifica Nakaseko & Nishimura, Thanarla sp. cf. T. pulchra (Squinabol), Thanarla sp., Xitus sp., Xitus (?) sp.
Age. Middle late Barremian–early early Aptian based on the occurrence of Aurisaturnalis carinatus perforatus Dumitrica & Dumitrica-Jud. Range after Dumitrica & Dumitrica-Jud (Reference Dumitrica and Dumitrica-Jud1995).
4.e. Section 5
TU10.35. Archaeodictyomitra (?) sp., Dicerosaturnalis trizonalis (Rüst), Dicerosaturnalis sp. cf. D. trizonalis (Rüst), Emiluvia sp. cf. E. ultima Baumgartner, Emiluvia sp., Eucyrtidiellum pyramis (Aita) (Fig. 5d), Mirifusus sp. cf. M. dianae (Karrer) s.l., Mirifusus sp., Napora sp., Podocapsa amphitreptera Foreman (Fig. 5e), Praeconosphaera (?) sp., Pseudodictyomitra (?) sp., Spinosicapsa (?) sp., Svinitzium (?) sp., Syringocapsa (?) sp., Triactoma sp. cf. T. jonesi (Pessagno), Triactoma (?) sp., Zhamoidellum (?) sp.
Age. Early–early late Tithonian to latest Tithonian–earliest Berriasian (UAZ. 12–13; UAZones after Baumgartner et al. Reference Baumgartner, Bartolini, Carter, Conti, Cortese, Danelian, De Wever, Dumitrica, Dumitrica-Jud, Goričan, Guex, Hull, Kito, Marcucci, Matsuoka, Murchey, O'Dogherty, Savary, Vishnevskaya, Widz, Yao, Baumgartner, O'Dogherty, Goričan, Urquhart, Pillevuit and De Wever1995 a) based on the presence of Eucyrtidiellum pyramis (Aita). Range after Baumgartner et al. (Reference Baumgartner, O'Dogherty, Goričan, Dumitrica-Jud, Dumitrica, Pillevuit, Urquhart, Matsuoka, Danelian, Bartolini, Carter, De Wever, Kito, Marcucci, Steiger, Baumgartner, O'Dogherty, Goričan, Urquhart, Pillevuit and De Wever1995b ).
TU10.36. Archaeodictyomitra apiarium (Rüst), Archaeodictyomitra sp. aff. A. excellens (Tan), Archaeodictyomitra sp. aff. A. rigida Pessagno, Archaeodictyomitra sp. cf. A. apiarium (Rüst), Archaeodictyomitra sp. 1, Archaeodictyomitra sp., Cinguloturris cylindra Kemkin & Rudenko (Fig. 5f), Emiluvia sp. cf. E. ordinaria Ozvoldova (Fig. 5g), Emiluvia sp. cf. E. orea Baumgartner, Emiluvia (?) sp., Eucyrtidiellum pyramis (Aita) (Fig. 5h), Eucyrtidiellum (?) sp. cf. E. (?) quinatum Takemura, Loopus primitivus (Matsuoka & Yao) (Fig. 5i), Loopus sp. cf. L. doliolum Dumitrica, Mirifusus dianae s.l. (Karrer), Mirifusus sp., Olanda sp., Podocapsa amphitreptera Foreman, Praeconosphaera (?) sphaeroconus (Rüst), Praeconosphaera (?) sp. cf. P. (?) sphaeroconus (Rüst), Praeconosphaera (?) sp., Pseudodictyomitra sp. cf. P. carpatica (Lozyniak), Pseudodictyomitra (?) sp., Ristola cretacea (Baumgartner) (Fig. 5j), Ristola (?) sp., Sethocapsa horokanaiensis Kawabata, Syringocapsa (?) sp., Zhamoidellum sp. cf. Z. sp. A sensu Goričan (Reference Goričan1994), Zhamoidellum sp.
Age. Early–early late Tithonian (UAZ. 12; UAZone after Baumgartner et al. Reference Baumgartner, Bartolini, Carter, Conti, Cortese, Danelian, De Wever, Dumitrica, Dumitrica-Jud, Goričan, Guex, Hull, Kito, Marcucci, Matsuoka, Murchey, O'Dogherty, Savary, Vishnevskaya, Widz, Yao, Baumgartner, O'Dogherty, Goričan, Urquhart, Pillevuit and De Wever1995 a) based on the occurrence of Cinguloturris cylindra Kemkin & Rudenko, Eucyrtidiellum pyramis (Aita) and Ristola cretacea (Baumgartner) with Loopus primitivus (Matsuoka & Yao). Ranges after Baumgartner et al. (Reference Baumgartner, O'Dogherty, Goričan, Dumitrica-Jud, Dumitrica, Pillevuit, Urquhart, Matsuoka, Danelian, Bartolini, Carter, De Wever, Kito, Marcucci, Steiger, Baumgartner, O'Dogherty, Goričan, Urquhart, Pillevuit and De Wever1995b ).
TU10.37. Archaeodictyomitra (?) sp., Podocapsa amphitreptera Foreman (Fig. 5k), Praeconosphaera (?) sphaeroconus (Rüst) (Fig. 5l), Pseudodictyomitra (?) sp., Spinosicapsa (?) sp.
Age. Late Oxfordian–early Kimmeridgian to latest Valanginian–earliest Hauterivian (UAZ. 10–18; UAZones after Baumgartner et al. Reference Baumgartner, Bartolini, Carter, Conti, Cortese, Danelian, De Wever, Dumitrica, Dumitrica-Jud, Goričan, Guex, Hull, Kito, Marcucci, Matsuoka, Murchey, O'Dogherty, Savary, Vishnevskaya, Widz, Yao, Baumgartner, O'Dogherty, Goričan, Urquhart, Pillevuit and De Wever1995 a) based on the occurrence of Praeconosphaera (?) sphaeroconus (Rüst) with Podocapsa amphitreptera Foreman. Ranges after Baumgartner et al. (Reference Baumgartner, O'Dogherty, Goričan, Dumitrica-Jud, Dumitrica, Pillevuit, Urquhart, Matsuoka, Danelian, Bartolini, Carter, De Wever, Kito, Marcucci, Steiger, Baumgartner, O'Dogherty, Goričan, Urquhart, Pillevuit and De Wever1995b ), Chiari, Cobianchi & Picotti (Reference Chiari, Cobianchi and Picotti2007) and Bandini et al. (Reference Bandini, Baumgartner, Flores, Dumitrica and Jackett2011).
TU10.38. Archaeodictyomitra sp. aff. A. chalilovi (Aliev), Archaeodictyomitra sp. aff. A. excellens (Tan), Archaeodictyomitra sp. aff. A. ioniana Danelian, Archaeodictyomitra sp. cf. A. apiarium (Rüst), Archaeodictyomitra sp. cf. A. excellens (Tan) (Fig. 5m), Archaeodictyomitra sp. cf. A. minoensis (Mizutani), Archaeodictyomitra sp., Archaeodictyomitra (?) sp., Becus (?) sp., Cinguloturris sp. cf. C. carpatica Dumitrica, Dicerosaturnalis trizonalis (Rüst), Dicerosaturnalis sp. cf. D. trizonalis (Rüst), Emiluvia sp., Pantanellium sp., Podocapsa amphitreptera Foreman (Fig. 5n), Praeconosphaera (?) sp., Pseudodictyomitra sp. cf. P. thurowi Dumitrica, Pseudodictyomitra sp., Saitoum sp. cf. S. elegans De Wever (Fig. 5o), Saitoum (?) sp., Spinosicapsa (?) sp., Syringocapsa (?) sp., Zhamoidellum ovum Dumitrica (Fig. 5p), Zhamoidellum sp.
Age. Middle–late Oxfordian to late Kimmeridgian–early Tithonian (UAZ. 9–11; UAZones after Baumgartner et al. Reference Baumgartner, Bartolini, Carter, Conti, Cortese, Danelian, De Wever, Dumitrica, Dumitrica-Jud, Goričan, Guex, Hull, Kito, Marcucci, Matsuoka, Murchey, O'Dogherty, Savary, Vishnevskaya, Widz, Yao, Baumgartner, O'Dogherty, Goričan, Urquhart, Pillevuit and De Wever1995 a) based on the presence of Podocapsa amphitreptera Foreman and Zhamoidellum ovum Dumitrica. Ranges after Baumgartner et al. (Reference Baumgartner, O'Dogherty, Goričan, Dumitrica-Jud, Dumitrica, Pillevuit, Urquhart, Matsuoka, Danelian, Bartolini, Carter, De Wever, Kito, Marcucci, Steiger, Baumgartner, O'Dogherty, Goričan, Urquhart, Pillevuit and De Wever1995b ) and Smuc & Goričan (Reference Smuc and Goričan2005).
4.f. Section 7
TU10.45. Archaeodictyomitra sp. cf. A. communis (Squinabol), Archaeodictyomitra sp. cf. A. coniforma Dumitrica (Fig. 5q), Archaeodictyomitra sp., Archaeodictyomitra (?) sp., Cryptamphorella clivosa (Aliev) (Fig. 5r), Godia sp. cf. G. decora (Li & Wu), Godia (?) sp., Holocryptocanium sp. cf. H. barbui Dumitrica, Praeconosphaera (?) sphaeroconus (Rüst) (Fig. 5s), Praeconosphaera (?) sp., Pseudodictyomitra (?) sp., Tethysetta sp. cf. T. usotanensis (Tumanda), Thanarla brouweri (Tan), Thanarla sp. cf. T. pulchra (Squinabol), Thanarla sp., Thanarla (?) sp.
Age. Valanginian to middle Aptian–early Albian (Valanginian to Costata Subzone of Turbocapsula Zone; zone after O'Dogherty, Reference O'Dogherty1994) based on the presence of Cryptamphorella clivosa (Aliev) with Praeconosphaera (?) sphaeroconus (Rüst). Ranges after Aliev (Reference Aliev1967), O'Dogherty (Reference O'Dogherty1994), Filippov & Kemkin (Reference Filippov and Kemkin2005), Chiari, Cobianchi & Picotti (Reference Chiari, Cobianchi and Picotti2007) and Bandini et al. (Reference Bandini, Baumgartner, Flores, Dumitrica and Jackett2011).
4.g. Section 8
TU10.47. Archaeodictyomitra excellens (Tan) (Fig. 5t), Archaeodictyomitra lacrimula (Foreman), Archaeodictyomitra sp. cf. A. communis (Squinabol), Archaeodictyomitra sp. cf. A. ioniana Danelian, Archaeodictyomitra sp. cf. A. vulgaris Pessagno, Archaeodictyomitra sp., Archaeodictyomitra (?) sp., Cana septemporatus (Parona) (Fig. 5u), Halesium sp. cf. H. palmatum Dumitrica (Fig. 5v), Hiscocapsa sp. cf. H. grutterinki (Tan), Hiscocapsa sp., Holocryptocanium barbui Dumitrica, Holocryptocanium sp. cf. H. barbui Dumitrica, Pantanellium sp., Praeconosphaera (?) sp. cf. P. (?) sphaeroconus (Rüst), Praeconosphaera (?) sp., Pseudodictyomitra sp. cf. P. sp. 5 in Dumitrica et al. (Reference Dumitrica, Immenhauser and Dumitrica-Jud1997), Pseudodictyomitra sp., Pseudodictyomitra (?) sp., Pseudoxitus (?) sp., Syringocapsa (?) sp., Thanarla sp. cf. T. brouweri (Tan), Thanarla sp., Thanarla (?) sp., Williriedellum (?) sp., Xitus sp. cf. X. normalis (Wu & Li), Xitus sp., Xitus (?) sp.
Age. Late Valanginian–early Barremian (UAZ. 17–21; UAZones after Baumgartner et al. Reference Baumgartner, Bartolini, Carter, Conti, Cortese, Danelian, De Wever, Dumitrica, Dumitrica-Jud, Goričan, Guex, Hull, Kito, Marcucci, Matsuoka, Murchey, O'Dogherty, Savary, Vishnevskaya, Widz, Yao, Baumgartner, O'Dogherty, Goričan, Urquhart, Pillevuit and De Wever1995 a) based on the presence of Cana septemporatus (Parona). Ranges after Baumgartner et al. (Reference Baumgartner, O'Dogherty, Goričan, Dumitrica-Jud, Dumitrica, Pillevuit, Urquhart, Matsuoka, Danelian, Bartolini, Carter, De Wever, Kito, Marcucci, Steiger, Baumgartner, O'Dogherty, Goričan, Urquhart, Pillevuit and De Wever1995b ).
TU10.48. Age not determinable because of the very low preservation of radiolarians.
TU10.51. Acaeniotyle sp. cf. A. umbilicata (Rüst), Archaeodictyomitra lacrimula (Foreman) (Fig. 5w), Archaeodictyomitra sp. cf. A. communis (Squinabol), Archaeodictyomitra sp., Archaeodictyomitra (?) sp., Crucella (?) sp., Praeconosphaera (?) sp., Pseudodictyomitra (?) sp., Pseudoeucyrtis sp. cf. P. hanni (Tan) sensu O'Dogherty (Reference O'Dogherty1994) (Fig. 5x), Spinosicapsa (?) sp., Thanarla sp. cf. T. pulchra (Squinabol), Thanarla sp.
Age. Early–early late Berriasian to middle Aptian–early Albian (UAZ. 14 – Costata Subzone of Turbocapsula Zone; zone after O'Dogherty, Reference O'Dogherty1994; UAZone after Baumgartner et al. Reference Baumgartner, Bartolini, Carter, Conti, Cortese, Danelian, De Wever, Dumitrica, Dumitrica-Jud, Goričan, Guex, Hull, Kito, Marcucci, Matsuoka, Murchey, O'Dogherty, Savary, Vishnevskaya, Widz, Yao, Baumgartner, O'Dogherty, Goričan, Urquhart, Pillevuit and De Wever1995 a) based on the presence of Archaeodictyomitra lacrimula (Foreman). Range after Baumgartner et al. (Reference Baumgartner, O'Dogherty, Goričan, Dumitrica-Jud, Dumitrica, Pillevuit, Urquhart, Matsuoka, Danelian, Bartolini, Carter, De Wever, Kito, Marcucci, Steiger, Baumgartner, O'Dogherty, Goričan, Urquhart, Pillevuit and De Wever1995b ) and O'Dogherty (Reference O'Dogherty1994).
TU10.M10. Archaeodictyomitra sp., Archaeodictyomitra (?) sp., Kilinora (?) sp., Triversus (?) sp.
Age. Age not determinable because of the low preservation of radiolarians.
5. Petrography and geochemistry of the basaltic rocks
5.a. Petrography
All the studied rocks are affected by low-temperature, ocean-floor alteration, which resulted in the replacement of primary minerals, though primary igneous textures are well preserved. Plagioclase is usually replaced by albite or calcite and rarely by clay mineral assemblages. Clinopyroxene is normally pseudomorphosed either by chlorite or actinolitic amphibole. In samples TU10.22 and TU10.23 (Section 3) clinopyroxene is replaced by brown hornblende, though fresh clinopyroxene relics are locally observed. The groundmass secondary phases mainly consist of chlorite and clay minerals. Regardless of the secondary mineralogical transformation, the following petrographic description of the various rock types has been made on the basis of the primary igneous phases. Moreover, for a better understanding, it has been made according to the geochemical groups described below.
Group 1. Pillow and massive lavas have aphyric, micro-crystalline sub-ophitic textures in which only small laths of plagioclase can be recognized. Pillow breccias are generally monogenetic and show a coarse-grained, intergranular texture with euhedral plagioclase and interstitial clinopyroxene.
Group 2. Massive lavas show both aphyric and porphyritic (PI = ~40) textures. In both varieties, the groundmass texture is hyalopilitic. Phenocrysts are represented by large crystals of plagioclase.
Groups 3 and 4. Pillow and massive lavas most commonly display aphyric, ophitic or sub-ophitic textures with crystal sizes ranging from micro-crystalline to coarse grained. Nonetheless, a few samples display slightly porphyritic textures with plagioclase microphenocrysts. In addition, a hyalopilitic texture is locally observed. The groundmass mineral assemblage includes plagioclase, clinopyroxene and variable amounts of opaque phases. Pillow breccias are generally monogenetic and the individual fragments are texturally and compositionally similar to the pillow lavas. Most of these rocks are characterized by variable abundances of varioles filled by calcite. In all the studied rock groups, the crystallization order is: plagioclase + clinopyroxene ± Fe–Ti-oxides.
5.b. Analytical methods
Whole-rock major- and some trace-element analyses were obtained by X-ray fluorescence (XRF) on pressed-powder pellets, using an ARL Advant-XP automated X-ray spectrometer. The matrix correction methods proposed by Lachance & Trail (Reference Lachance and Trail1966) were applied. Volatile contents were determined as loss on ignition (LOI) at 1000°C. In addition, Rb, Sr, Nb, Hf, Ta, Th, U and the rare earth elements (REEs) were determined on 12 representative samples by inductively coupled plasma-mass spectrometry (ICP-MS) using a Thermo Series X-I spectrometer.
The CO2 content was determined by the simple volumetric technique (Jackson, Reference Jackson1958) only on the samples affected by calcite veins and amygdales. This technique was calibrated using standard amounts of reagent grade CaCO3. In addition, for the discussion of the geochemical characteristics and for a better comparison of chemical data, the major-element compositions of these samples were recalculated on a calcite-free basis. In detail, CaO content in secondary calcite has been calculated according to stoikiometric proportions with CO2 contents, given that the secondary carbonates are exclusively composed of calcite. Major-element composition has then been recalculated to 100 wt% without considering LOI and CaO in calcite.
The accuracy of the data for XRF and ICP-MS analyses were evaluated using results for international standard rocks run as unknowns. The detection limits for XRF and ICP-MS analyses were evaluated using results from several runs of ~29 international standards. Accuracy and detection limits for the CO2, analyses were determined using different amounts of reagent grade CaCO3 run as unknowns. Results are given in the online Supplementary Material available at http://journals.cambridge.org/geo. All whole-rock analyses were performed at the Dipartimento di Science della Terra, Università di Ferrara. The results are shown in Table 1.
Table 1. Representative major- and trace-element analyses of Middle Jurassic – Early Cretaceous volcanic rocks from the Ankara Mélange
Abbreviations: bas – basalt; tra – trachyte; Fe-bas – ferrobasalt; trans – transitional-type; sub-alk – sub-alkaline-type; OIB – ocean-island basalt; P-MORB – plume-type mid-ocean ridge basalt; E-MORB – enriched-type mid-ocean ridge basalt; G-MORB – garnet-influenced mid-ocean ridge basalt; E – Early; M – Middle; L – Late; Jr – Jurassic; Cr – Cretaceous; mlf – massive lava flow; n.d. – not detected. Mg no. = 100×Mg/(Mg+Fe2+). Fe2O3 = 0.15×FeO. Normalizing values for REE ratios are from Sun & McDonough (Reference Sun, McDonough, Saunders and Norry1989).
5.c. Geochemistry
The following geochemical description is made mainly using those major and trace elements that are virtually immobile during low-temperature alteration and metamorphism (e.g. Pearce & Norry, Reference Pearce and Norry1979). These elements include many incompatible elements, such as Ti, P, Zr, Y, Sc, Nb, Ta, Hf, Th, middle (M-) and heavy (H-) REEs, as well as some transition metals (e.g. Ni, Co, Cr, V). Light REEs (LREEs) may be affected by some mobilization during alteration. However, the good correlations between these elements and many immobile elements (not shown) indicate that LREEs have not been mobilized by the alteration. For example, the correlation coefficients (r2) for the linear correlation Zr–La and Zr–Ce are 0.96 and 0.90, respectively. Rb, Ba, K and Sr are commonly mobilized during alteration. In fact, these elements plotted against Zr generally show low or very low r2. However, the fairly good correlation with Zr and Rb in Group 4 samples (r2 = 0.91), as well as Ba in Group 4 (r2 = 0.87) and in Group 2 samples (r2 = 0.85), suggests that these elements were only slightly mobilized during alteration in these rock types.
According to Bortolotti et al. (Reference Bortolotti, Chiari, Göncüoglu, Marcucci, Principi, Tekin, Saccani and Tassinari2013 a), four groups of volcanic rocks can be recognized in the studied sections. Group 1 is represented by basalts cropping out in Sections 5 and 8. These rocks have a clear sub-alkaline nature with low Nb/Y ratios (Fig. 6) and show relatively high TiO2, P2O5, Zr and Y contents. Ni is generally low, whereas Cr is relatively high, with the only exception of sample TU10.49b. V content is also generally high (Table 1). The values of these elements are similar to those of Group 2 basalts (see below). By contrast, Hf, Ta, Th and U contents are low. These rocks show a sharp increase in FOt and Y and a sharp decrease in Mg no. and Ni with increasing Zr, as well as very low Th/Tb ratios (Fig. 7). Group 1 basalts are characterized by flat normalized incompatible element patterns (Fig. 8a). The REE abundance (Fig. 8b) varies from ~10 to ~20 times that of chondrite and displays LREE/MREE smoothly depleted patterns with (La/Sm)N ratios ranging from 0.78 to 0.82. The overall geochemical features of these rocks, as well as both the incompatible elements and REE patterns, resemble those of typical N-MORB (normal mid-ocean ridge basalt) (Sun & McDonough, Reference Sun, McDonough, Saunders and Norry1989). Nonetheless, HREEs are slightly depleted with respect to both LREEs and MREEs (Fig. 8b) with (La/Yb)N ratios = 1.06–1.30 and (Sm/Yb)N ratios = 1.31–1.58. In particular, the (Sm/Yb)N ratios are higher than that of typical N-MORB (SmN/YbN = 0.96, Sun & McDonough, Reference Sun, McDonough, Saunders and Norry1989). These values are comparable to those of garnet-influenced MORBs (G-MORB) from the External Ligurides of the Northern Apennines (SmN/YbN = 1.25–1.50, Montanini, Tribuzio & Vernia, Reference Montanini, Tribuzio and Vernia2008), Corsica (SmN/YbN = 1.30–1.81, Saccani et al. Reference Saccani, Principi, Garfagnoli and Menna2008b ) and Elba Island (SmN/YbN = 1.49–2.10, Saccani & Principi, Reference Saccani and Principi2016), as well as to those of the Palaeozoic Misho mafic complex in north Iran (Saccani et al. Reference Saccani, Allahyari, Beccaluva and Bianchini2013 a) and the Mesozoic Kermanshah ophiolites in south Iran (Saccani et al. Reference Saccani, Azimzadeh, Dilek and Jahangiri2013b ). The G-MORB affinity of Group 1 rocks is also suggested by Th–Nb co-variation and by their (Ce/Yb)N and (Dy/Yb)N ratios (Fig. 9), although in the most common tectonic discrimination diagrams these basalts plot in the field for N-MORBs (e.g. Fig. 10).
Figure 6. Ti/Y v. Nb/Y discrimination diagram (Pearce, Reference Pearce and Thorpe1982) for Middle Jurassic – Early Cretaceous volcanic rocks from the Ankara Mélange. Modified from Bortolotti et al. (Reference Bortolotti, Chiari, Göncüoglu, Marcucci, Principi, Tekin, Saccani and Tassinari2013 a).
Figure 7. Variation diagrams for some representative major and trace elements versus Zr for Middle Jurassic – Early Cretaceous volcanic rocks from the Ankara Mélange. Major elements are recalculated on a volatile-free and calcite-free basis. Abbreviations: pl – plagioclase; ol – olivine; cpx – clinopyroxene; opx – orthopyroxene; mt – magnetite. Mg no. = 100×Mg/(Mg+Fe2+). Lines represent the inferred fractionation trends for the different rock groups.
Figure 8. (a, c, e, g) N-MORB normalized incompatible element patterns and (b, d, f, h) chondrite-normalized REE patterns for Middle Jurassic – Early Cretaceous volcanic rocks from the Ankara Mélange. Normalizing values and the compositions of normal mid-ocean ridge basalt (N-MORB), enriched mid-ocean ridge basalt (E-MORB) and ocean-island basalt (OIB) are from Sun & McDonough (Reference Sun, McDonough, Saunders and Norry1989).
Figure 9. (a) ThN v. NbN and (b) (Dy/Yb)N v. (Ce/Yb)N discrimination diagrams for Middle Jurassic – Early Cretaceous volcanic rocks from the Ankara Mélange. Modified after Saccani (Reference Saccani2015). N-MORB and chondrite normalization values for panels (a) and (b), respectively, are from Sun & McDonough (Reference Sun, McDonough, Saunders and Norry1989).
Figure 10. Th–Ta–Hf/3 (Wood, Reference Wood1980) discrimination diagrams for Middle Jurassic – Early Cretaceous volcanic rocks from the Ankara Mélange.
Group 2 is represented by basalts cropping out in Sections 4 and 7. These rocks have a sub-alkaline nature with low Nb/Y ratios ranging from 0.27 to 0.40 (Fig. 6) and show relatively high TiO2, P2O5, Zr and Y contents (Table 1). Ni, though variable, is generally low, whereas Cr and V are rather high. As exemplified in the variation diagrams in Figure 7, these rocks have major-element contents and compositions of many trace elements largely overlapping those of Group 1 basalts. Nonetheless, as a distinctive feature, Group 2 basalts are characterized by large ion lithophile element (LILE)/high-field-strength element (HFSE) smoothly enriched patterns (Figs 8c). Likewise, REE patterns show a slight enrichment in LREEs compared to HREEs (Fig. 8d), with (La/Sm)N and (La/Yb)N ratios ranging from 1.12 to 1.51 and from 1.27 to 2.19, respectively. These patterns are similar to that of the typical enriched-type MORB (E-MORB) of Sun & McDonough (Reference Sun, McDonough, Saunders and Norry1989). The Th–Nb values are also similar to those of the typical E-MORB (Fig. 9a). In the most common tectonic discrimination diagrams these basalts plot across the boundary between the fields for N-MORB and E-MORB (e.g. Fig. 10).
Group 3 is represented by pillow basalts and ferrobasalts from Section 2, as well as by massive lava basalts from Section 9. These rocks have a transitional nature, as testified to by their high Nb/Y ratios (Fig. 6). Basalt TU10.15 (Section 2) displays a rather primitive composition with relatively high Mg no. and low TiO2 and P2O5 contents (Table 1). In contrast, ferrobasalts from Section 2 and basalts from Section 9 have rather evolved compositions with relatively low Mg no. and very high TiO2 and P2O5 contents (Table 1). Accordingly, Ni and Cr contents are generally high in basalts, whereas they are relatively low in the differentiated rocks. Except for Mg no., the variation of many elements with respect to Zr display roughly common evolutionary trends towards high contents of the incompatible elements and FeOt and low contents of the compatible elements for all Group 3 samples (Fig. 7). In contrast, Mg no. for samples from Sections 2 and 9 show distinct evolutionary trends with respect to Zr. All these trends are compatible with a magmatic evolution by fractional crystallization. The Th/Tb ratios of these rocks are significantly higher than those of Groups 1 and 2 rocks and significantly lower than those of Group 4 rocks (Fig. 7). Group 3 rocks show high abundance in low-field-strength elements (LFSEs) with respect to N-MORB and display regularly decreasing N-MORB normalized patterns from Rb to Y (Fig. 8e). The REE abundance (Fig. 8f) varies from ~30 to ~100 and from ~10 to ~12 times that of chondrite for LREEs and HREEs, respectively. These rocks displays LREE/MREE and LREE/HREE enriched patterns, with (La/Sm)N ratios = 2.61–4.15 and (La/Yb)N ratios = 4.65–9.17. These incompatible elements and REE patterns are comparable to those of P-MORB (plume-type MORB). This conclusion is also supported by the co-variation of the Th and Nb concentrations (Fig. 9a). In the most common tectonic discrimination diagrams (e.g. Fig. 10) these rocks generally plot in the fields for E-MORBs and alkaline ocean-island basalts (OIB).
Group 4 is represented mainly by alkaline basalts and subordinate trachytes from Sections 1 and 3. These rocks have a clear alkaline nature, as testified to by their very high Nb/Y ratios (Fig. 6). Alkaline basalts from Section 3 mainly display relatively primitive compositions, whereas alkaline basalts from Section 1 represent rather evolved compositions. Generally, all samples have high contents of TiO2, P2O5, Zr, Nb, Hf and Th and low contents of compatible elements (Table 1; Fig. 7). In the variation diagrams in Figure 7, rocks from the different sections show different evolutionary trends, most likely reflecting the distinct evolutions of magmas of different initial compositions. This is particularly evident in the FeOt, Mg no., Ni and Y v. Zr diagrams. Nonetheless, the well-defined trends observed for samples from each single section suggest that each section consists of rocks belonging to a comagmatic suite. Group 4 rocks show high abundance in LFSEs with respect to N-MORB and display regularly decreasing N-MORB normalized patterns from Rb to Y (Fig. 8g). The REE abundance (Fig. 8h) varies from ~60 to ~400 and from ~9 to ~10 times that of chondrite for LREEs and HREEs, respectively. These rocks displays LREE/MREE and LREE/HREE strongly enriched patterns, with (La/Sm)N ratios = 4.15–5.25 and (La/Yb)N ratios = 18.75–23.13. Group 4 rocks can readily be distinguished from Group 3 rocks, as they show higher Zr, incompatible element and LREE concentrations, as well as Th/Tb ratios at comparable Mg no. (Figs 7, 8g, h). The overall geochemistry of these basalts resembles that of alkaline basalts generated at within-plate ocean-island settings (OIB), as also exemplified by the generally high Ti/V ratios (Table 1). This conclusion is also supported by the co-variation of the Th and Nb values (Fig. 9), as well as by the most common tectonic discrimination diagrams (e.g. Fig. 10).
6. Discussion
6.a. Mantle melting processes and magma generation
One of the main goals of this study is to assess the nature and tectonic significance of the magmatic events that occurred in the Turkish sector of the Neotethys during the Middle Jurassic – Early Cretaceous time span. According to many authors (e.g. Pearce, Reference Pearce and Thorpe1982), the compositional differences between magma types are related to different source characteristics that are associated, in turn, with distinct tectonomagmatic settings of formation. We will therefore focus our petrogenetic discussion on the identification of the possible mantle sources and related tectonic setting of formation of the four distinct lava groups identified in the previous section, which are: (1) G-MORBs (Group 1); (2) E-MORBs (Group 2); (2) P-MORBs (Group 3); (4) OIB-type alkaline volcanic rocks (Group 4). Unfortunately, the chemical variation due to fractional crystallization cannot be defined in detail, as the mélange nature of the sampled rocks prevents us from establishing definite genetic relationships between rocks within each single chemical group. Nonetheless, some trace-element ratios (e.g. Zr/Nb, Ce/Y, Th/Ta, Th/Tb) are little affected by fractional crystallization of predominantly olivine + clinopyroxene + plagioclase. Therefore, even in the presence of significant amounts of fractionation, they are believed to represent the elemental ratios in the source (e.g. Allègre & Minster, Reference Allègre and Minster1978; Beker et al. Reference Beker, Menzies, Thirlwall and Macpherson1997). Ratios of incompatible elements (Ce/Y, Nb/Yb), ratios of hygromagmatophile element ratios ((Th/Ta)/(Th/Tb)) (Table 1), as well as distinct normalized multi-element and REE patterns (Fig. 8) suggest that the different magmatic rocks from the Ankara Mélange units have been most likely originated from chemically distinct mantle sources. In order to constrain the possible mantle sources of the different rock series of the Ankara Mélange, non-modal, batch partial melting models are presented in Figure 11. A rigorous quantification of the melting processes is not possible as the composition of the mantle sources are difficult to constrain. However, a semi-quantitative modelling of the REEs can place some solid constraints. The melt modelling uses plots of LREE/HREE (i.e. La/Yb) v. MREE/HREE (i.e. Dy/Yb) ratios, which are particularly useful for distinguishing between melting in the spinel and garnet stability fields (Thirlwall, Upton & Jenkins, Reference Thirlwall, Upton and Jenkins1994). Partial melting of a mantle source in the spinel-facies produces little change in Dy/Yb ratios in melts with respect to melt fraction. In contrast, mantle partial melting in the garnet-facies produces large changes in Dy/Yb ratios with melt fraction. In both cases, La/Yb ratios are particularly responsive to melt fraction change (Fig. 11). Another important feature of these plots is that mixing between different melt fractions will generate linear mixing arrays (e.g. Beker et al. Reference Beker, Menzies, Thirlwall and Macpherson1997).
Figure 11. Melt curve models based on Dy/Yb v. La/Yb. Melt curves are calculated using non-modal, batch melts of garnet and spinel lherzolites. (a) Melt curves for DMM mantle (Workman & Hart, Reference Workman and Hart2005); (b) melt curves for a theoretical DMM mantle (Workman & Hart, Reference Workman and Hart2005) enriched in LREEs by OIB-type components; (c) melt curves for a theoretical enriched (OIB-type) mantle. Garnet lherzolite mode is: 0.598 ol, 0.211 opx; 0.076 cpx, 0.115 gt that melts in the proportions 0.05 ol, 0.20 opx, 0.30 cpx, 0.45 gt. Spinel lherzolite mode is: 0.578 ol, 0.270 opx, 0.119 cpx, 0.033 spl that melts in the proportions 0.10 ol, 0.27 opx, 0.50 cpx, 0.13 spl. Mantle mode and melting proportions are from Thirlwall, Upton & Jenkins (Reference Thirlwall, Upton and Jenkins1994). Arrays representing the mixing between various proportions of melt fractions from the garnet-facies mantle and melt fractions from the spinel-facies mantle are also shown. Distribution coefficients are from Irving & Frey (Reference Irving and Frey1984) with the exception of those for spinel, which are from McKenzie & O'Nions (Reference McKenzie and O'Nions1991). Normalizing values are from Sun & McDonough (Reference Sun, McDonough, Saunders and Norry1989).
Group 1 basalts. As observed in the previous section, the Group 1 rocks (G-MORBs) show unusually high MREE/HREE ratios (DyN/YbN = 1.20–1.34) compared to the typical N-MORB (DyN/YbN = 1, Sun & McDonough, Reference Sun, McDonough, Saunders and Norry1989). REE modelling for Group 1 basalts (Fig. 11a) shows that these rocks cannot have been simply derived from partial melting of a typical depleted MORB mantle (DMM) source (Workman & Hart, Reference Workman and Hart2005) in the spinel-facies. Rather, their significant HREE/MREE fractionation can be interpreted as a garnet signature, which can be related either to a deep initiation of melting in the garnet peridotite stability field, or to the melting of a heterogeneous mantle source characterized by garnet-bearing mafic/ultramafic layers (e.g. Montanini, Tribuzio & Vernia, Reference Montanini, Tribuzio and Vernia2008; Saccani et al. Reference Saccani, Principi, Garfagnoli and Menna2008b ; Saccani, Reference Saccani2015). However, melting of a DMM source bearing garnet mafic/ultramafic layers would generate primary melts characterized by high MREE/HREE ratios coupled with low (<0.8) LREE/HREE ratios (not shown). By consequence, the high (1.06–1.30) LREE/HREE ratios observed in the Group 1 basalts are inconsistent with this hypothesis. Therefore, in Figure 11a the partial melting model of a DMM source that starts in the garnet-facies and continues to larger degrees in the spinel-facies (with various combinations of melting fractions in the garnet- and spinel-facies) is shown. It can be observed that the REE composition of the Group 1 basalts is compatible with the calculated compositions for 2.5% melting in the garnet-facies and 10% melting in the spinel-facies, assuming mixing of ~70–80% of melt derived from spinel-facies mantle with ~30–20% melt from garnet-facies mantle.
Group 2 basalts. Group 2 basalts (E-MORBs) show variable LREE/HREE enrichments (Fig. 8d), which can be considered a result of variable mixing between depleted and enriched asthenospheric sources or, alternatively, from lower degrees of partial melting of a DMM source, compared to N-MORBs. In fact, basalt TU10.32 is generally compatible with a low degree (~2.5%) of partial melting of a DMM source in the spinel-facies (Fig. 11b). However, basalt TU10.46 shows HREE values lower than those of the N-MORBs of Group 1, which cannot be generated by lower degrees of partial melting of a common mantle source (Fig. 8b, d). In addition, lower degrees of partial melting of a DMM source in the spinel-facies cannot generate the (La/Yb)N and (Dy/Yb)N ratios of this sample. The most appropriate solution for the genesis of this E-MORB sample is not straightforward. The LREE/HREE and MREE/HREE ratios are compatible with mixing of melt derived from a very low degree (~1%) of partial melting of a DMM source in the garnet-facies with melt derived from a much larger degree (~10%) of partial melting in the spinel-facies mantle (Fig. 11a). The REE composition of these basalts is compatible with mixing of <10% of melts generated in the garnet-facies with > 90% of melts generated in the spinel-facies. Nonetheless, a possible alternative solution is to invoke a more enriched source than DMM. Figure 11b shows that the melting curve of a hypothetical DMM source slightly enriched in LREEs with (La/Yb)N = 1.09 and (Dy/Yb)N = 1.49, and has a Yb concentration equal to that of DMM (YbN = 2.08). The model shows that the REE composition of E-MORB TU10.46 is compatible with the calculated composition for ~8% partial melting of this theoretical source in the spinel-facies.
Group 3 and Group 4 rocks. The high LREE/HREE ratios displayed by the transitional basalts of Group 3 rocks (P-MORBs) and by the alkaline basalts of Group 4 (Fig. 8f, h) suggest an involvement of a garnet peridotite source. Moreover, the high La/Yb ratio observed in these basalts implies a source more enriched in LREEs than DMM. In fact, variable degrees of partial melting of a DMM source in the spinel-facies cannot generate the observed La/Yb ratios, and variable degrees of partial melting of a DMM source in the garnet-facies cannot generate the observed La/Yb and Dy/Yb ratios (Fig. 11a). Therefore, the most appropriate solution is to invoke a more enriched source than DMM, although it is impossible to assess the exact composition of this source. The model in Figure 11c illustrates melting curves for a hypothetical LREE-enriched source (Beker et al. Reference Beker, Menzies, Thirlwall and Macpherson1997) with (La/Yb)N = 1.32 and (Dy/Yb)N = 1.49, and has a Yb concentration equal to that of DMM (YbN = 2.08). Higher Yb concentrations (e.g. PM values) in the source would generate concentrations of HREEs in the melts that are too high compared with the Group 3 and Group 4 volcanic rocks.
The co-variation in La/Yb–Dy/Yb systematics of Group 3 and Group 4 samples (Fig. 11c) cannot, however, be explained by variable degrees of partial melting of this enriched source in either the spinel- or garnet-facies. Melting in the garnet-facies produces melts with much higher Dy/Yb ratios than those of both Group 3 and Group 4 samples at reasonable degrees of melting (i.e. <20%). Alternatively, the mantle source would require an unusually low Dy/Yb ratio if the samples were to be simply the product of garnet-facies mantle melting. By contrast, melting in the spinel-facies produces melts with both La/Yb and Dy/Yb ratios lower than those of both Group 3 and Group 4 samples. Therefore, the simplest model to account for the REE systematics of these rocks involves mixing of small melt fractions from garnet-facies enriched mantle with relatively larger melt fractions from spinel-facies (Fig. 11c). This figure shows that the La/Yb–Dy/Yb systematics of Group 3 basalts can be explained by mixing of small-degree melts (~0.5–1%) and larger degree melts (~5%) from garnet- and spinel-facies mantle, respectively. Likewise, the observed REE data for Group 4 basalts can be accounted for by mixing of melts from ~0.5% and ~5% partial melting from garnet- and spinel-facies mantle, respectively. Alternatively, the La/Yb–Dy/Yb systematics of Group 4 basalts can be explained by a comparatively higher degree of melting (~1%) in the garnet-facies mantle and comparatively lower degree of melting (~2.5%) in the spinel-facies mantle. In any case, the different La/Yb–Dy/Yb ratios shown by Group 3 and Group 4 basalts can be accounted for by mixing of different proportions of melts generated in the garnet- and spinel-facies mantle. In detail, Group 4 basalts may have resulted from the mixing of ~60% of melt derived from spinel-facies mantle with ~40% melt from garnet-facies mantle, whereas Group 3 basalts may have resulted from the mixing of ~90% of melt derived from spinel-facies mantle with ~10% melt from garnet-facies mantle (Fig. 11c).
Figure 11b shows that the La/Yb and Dy/Yb ratios of Group 3 basalts can also be compatible with a very low degree (<0.8%) of partial melting in the spinel-facies of the slightly metasomatized mantle source hypothesized for the genesis of Group 2 rocks. Such a very low degree of partial melting is, however, unreasonable. In fact, experimental studies on melt mobility in peridotites showed that very small melt fractions are not readily mobile and therefore they remain within the host peridotite rather than migrating to form volcanic rock melt (see Warren, Reference Warren2016 for a more detailed discussion). Finally, it should be noted that variations in the degree of partial melting in the spinel field are difficult to constrain owing to the small range in La/Yb ratios generated by spinel-facies melting. Some scatter in the La/Yb–Dy/Yb systematics (Fig. 11c) might be accounted for by small fluctuations in the degree of melting of garnet-facies mantle. In any case, from a semi-quantitative point of view, melt fractions in the garnet field are restricted to <1% whereas those in the spinel field are probably several per cent (~5%).
The possible influence of crustal contamination can be excluded as the Th–Nb compositions of all rock types plot within the MORB–OIB array (Fig. 9). Other geochemical indicators further support this conclusion. For example, high Th/Ta and low Nb/U ratios are effective indicators of crustal contamination. All groups of Middle Jurassic – Early Cretaceous basalts from the Ankara Mélange show very low Th/Ta ratios (<1.74), as well as Nb/U ratios averaging 44 in N-MORBs, 50 in E-MORBs, 40 in P-MORBs and 43 in OIBs. These Nb/U ratios are comparable to those of the typical N-MORB (49.6), E-MORB (46.1) and OIB (47.1) of Sun & McDonough (Reference Sun, McDonough, Saunders and Norry1989).
Bortolotti et al. (Reference Bortolotti, Chiari, Göncüoglu, Marcucci, Principi, Tekin, Saccani and Tassinari2013 a) have noted that the association of depleted basalts (N-MORBs), moderately enriched basalts (E-MORBs) and variably enriched rocks (P-MORBs and OIBs) occurring in the Ankara Mélange is also observed in many peri-Mediterranean ophiolitic complexes (e.g. Saccani & Photiades, Reference Saccani and Photiades2005; Saccani et al. Reference Saccani, Beccaluva, Photiades and Zeda2011 and references therein), as well as from several Middle East ophiolites (Allahyari et al. Reference Allahyari, Saccani, Pourmoafi, Beccaluva and Masoudi2010; Saccani et al. Reference Saccani, Delavari, Beccaluva and Amini2010, Reference Saccani, Allahyari, Beccaluva and Bianchini2013a ,Reference Saccani, Azimzadeh, Dilek and Jahangiri b ), where it is interpreted as the result of partial melting of a MORB-type asthenospheric source enriched in HFSEs and LREEs by an OIB-type chemical component (plume-type component). Bortolotti et al. (Reference Bortolotti, Chiari, Göncüoglu, Marcucci, Principi, Tekin, Saccani and Tassinari2013 a) have used the co-variation of Zr/Y and Zr/Nb to qualitatively depict the influence of a plume-type component on MORB compositions in the Middle Jurassic – Early Cretaceous basalts from the Ankara Mélange (Fig. 12). From Figure 12 it is evident that the data conform extremely well to the mixing curve calculated using the OIB and N-MORB end-members. Such mixing relationships are consistent with either magma mixing or source region mixing (or eventually, a combination of these).
Figure 12. Zr/Y v. Zr/Nb diagram for volcanic rocks from the Middle Jurassic – Early Cretaceous volcanic rocks from the Izmir–Ankara Mélange (modified from Bortolotti et al. Reference Bortolotti, Chiari, Göncüoglu, Marcucci, Principi, Tekin, Saccani and Tassinari2013 a). The compositions of modern normal mid-ocean ridge basalt (N-MORB) and ocean-island basalt (OIB) are from Sun & McDonough (Reference Sun, McDonough, Saunders and Norry1989). The compositional variation for ocean-floor basalts erupted in the North Atlantic Ocean is shown for comparison (data from Hanan et al. Reference Hanan, Blichert-Toft, Kingsley and Schilling2000). The dashed line represents the mixing curve calculated using the OIB and N-MORB end-members.
6.b. Tectonomagmatic significance
The melting models carried out for the different groups of volcanic rocks, which are presented in the previous section, allow the following conclusions to be drawn: (1) the geochemically distinct groups of Middle Jurassic – Early Cretaceous volcanic rocks in the Ankara Mélange are related to different mantle source compositions and partial melting degrees; (2) regardless of their geochemical affinities, all the studied volcanic rocks were generated by partial melting starting in the garnet-facies mantle and continuing to larger degrees in the spinel-facies mantle. Perhaps, Group 2 basalts from Section 7 may represent the only exception to this conclusion. In fact, their chemistry is compatible either with partial melting starting in the garnet-facies mantle and continuing to larger degrees in the spinel-facies mantle, or partial melting in the spinel-facies from a slightly enriched source. As shown before, the formation of enriched alkaline and P-MORB-type rocks implies the occurrence of mantle sources strongly metasomatized by OIB-type components. Two alternative hypotheses can account for such OIB-type metasomatism of depleted mantle sources: (1) the existence of plume activity in the region during Middle Jurassic – Early Cretaceous times and (2) the existence of deep mantle heterogeneously modified by previous mantle plume activity that occurred in the same area in association with the opening of the Neotethys. However, the extant geological evidence suggests that the first hypothesis can be disregarded. In fact, the lack of magmatic evolution from more depleted to more enriched rocks that is commonly observed in plume-related magmatism, the absence of basaltic plateaus and a relatively small volume of plume-related volcanic rocks collectively argue against the existence of a well-established, long-lasting mantle plume in the region. Therefore, we favour the hypothesis that the different Middle Jurassic – Early Cretaceous volcanic rock types from the Ankara Mélange were formed from partial melting of a strongly heterogeneous mantle, with OIB-type components inherited from a previous mantle plume activity associated with the opening of the Neotethys. In fact, the Anisian alkaline volcanic rocks intruding the Kütahya–Bolkardag have been interpreted as the early products of the rifting of the Neotethys Ocean with the involvement of a plume (Göncüoglu, Turhan & Tekin, Reference Göncüoglu, Turhan and Tekin2003; Göncüoglu, Reference Göncüoglu2010; Akal et al. Reference Akal, Candan, Koralay, Oberhänsli, Chen and Prelevic2012). Recent studies on modern oceanic basins further support our favoured hypothesis. In fact, these studies have demonstrated that the upper mantle is much more heterogeneous than previously thought (e.g. Brunelli et al. Reference Brunelli, Seyler, Cipriani, Ottolini and Bonatti2006; Warren, Reference Warren2016). The mantle heterogeneities can be either the result of earlier tectonic events or ancient episodes of melting, melt extraction and melt entrapment, and they can occur at a local or at a regional scale (e.g. Liu et al. Reference Liu, Snow, Hellebrand, Brügmann, von der Handt and Hofmann2008; Warren et al. Reference Warren, Shimizu, Sakaguchi, Dick and Nakamura2009).
A possible tectonomagmatic model that can explain the formation of the different volcanic rocks from the Ankara Mélange during Middle Jurassic – Early Cretaceous times is shown in Figure 13. In this model, the OIB-type metasomatized portions are likely to be prominent in the asthenospheric mantle. The uprising asthenospheric mantle underwent polybaric partial melting, which started in the garnet-facies and continued in the spinel-facies. The strongly enriched alkaline and P-MORB rocks were generated from various, but limited degrees of polybaric partial melting of OIB-type metasomatized portions. G-MORB and E-MORB were generated from various degrees of polybaric partial melting of depleted asthenosphere portions. Alternatively, E-MORB (particularly, that of Section 7) may have been generated from partial melting of a slightly enriched mantle source in the spinel stability field. The tectonomagmatic model presented in Figure 13 implies that the different rocks were formed by partial melting of chemically different portions of the sub-oceanic mantle in different times. In consequence, this model does not necessarily imply the existence of a genetic relationship between different rock groups or within a single rock group.
Figure 13. Two-dimensional cartoon showing the tectonomagmatic mechanisms responsible for the formation of garnet-influenced (G-), enriched (E-) and plume-type (P-) mid-ocean ridge basalts (MORB), as well as alkaline ocean-island-type (OIB) basalts from the Ankara Mélange during Late Jurassic – Early Cretaceous times. Other abbreviations: sp – spinel; gt – garnet.
6.c. Geodynamic implications
The ages obtained for the studied radiolarian assemblages coupled with the geochemistry of the associated basalts indicate that a composite oceanic crust including G-MORB, E-MORB, P-MORB and alkaline basalts was forming during Middle Jurassic – Early Cretaceous times. These ages are in agreement with the radiolarians (Bragin & Tekin, Reference Bragin and Tekin1996; Tekin, Reference Tekin1999; S. Celik, unpub. M.Sc. thesis, Hacettepe Univ., 2010; T. Üner, unpub. Ph.D. thesis, Hacettepe Univ., 2010; Tekin, Göncüoglu & Uzuncimen, Reference Tekin, Göncüoglu and Uzuncimen2012; Göncüoglu et al. Reference Göncüoglu, Tekin, Sayit, Bedi and Uzunçimen2015) and foraminifera (Boccaletti, Bortolotti & Sagri, Reference Boccaletti, Bortolotti and Sagri1966; Bortolotti & Sagri, Reference Bortolotti and Sagri1968; Yaliniz, Göncüoglu & Özkan-Altiner, Reference Yaliniz, Göncüoglu and Özkan-Altiner2000; Rojay, Yaliniz & Altiner, Reference Rojay, Yaliniz and Altiner2001) data obtained from different parts of the IAESB. The age ranges from the published biostratigraphical data indicate gaps in Early Jurassic and late Early Cretaceous times (see Göncüoglu, Sayit & Tekin, Reference Göncüoglu, Sayit and Tekin2010). Moreover, Middle Jurassic findings were restricted to a limited number of samples. Our new findings partially fill the gap during the late Early Cretaceous period (see Göncüoglu, Sayit & Tekin, Reference Göncüoglu, Sayit and Tekin2010). In fact, the radiolarian cherts associated with the G-MORB of Section 8 indicated a late Valanginian – early Barremian age.
The Middle Jurassic – Early Cretaceous rock assemblage found in the mélange complexes of the IAESB, as well as the tectonomagmatic model presented in the previous section can be framed within the geodynamic model presented by Göncüoglu (Reference Göncüoglu2010) (Fig. 14). According to this model, the rifting of the Neotethys Ocean started in Late Triassic time with the involvement of a mantle plume (Fig. 14a). Anisian and Norian volcanic rocks showing alkaline affinity were erupted at this stage (Göncüoglu, Turhan & Tekin, Reference Göncüoglu, Turhan and Tekin2003; Göncüoglu Reference Göncüoglu2010; Akal et al. Reference Akal, Candan, Koralay, Oberhänsli, Chen and Prelevic2012; Bortolotti et al. Reference Bortolotti, Chiari, Göncüoglu, Marcucci, Principi, Tekin, Saccani and Tassinari2013 a). The Early Jurassic history of this oceanic basin cannot be straightforwardly constrained owing to the lack of data. In fact, Early Jurassic (Hettanginian–Sinemurian) radiolarian chert blocks were found as single blocks in the Ankara Mélange, but they are not associated with volcanic rocks (S. Celik, unpub. M.Sc. thesis, Hacettepe Univ., 2010; Göncüoglu et al. Reference Göncüoglu, Tekin, Sayit, Bedi and Uzunçimen2015). Nonetheless, it is reasonable to postulate that during this time the Neotethys Ocean experienced an oceanic spreading phase. During the Middle Jurassic – Early Cretaceous time span the formation of a composite oceanic crust including G-MORB, E-MORB, P-MORB and alkaline basalts suggests that the volcanic rocks of the IAESB formed at a mid-ocean ridge setting by tapping different portions of a deep, highly heterogeneous mantle (Fig. 14b). Mantle heterogeneities consisted of portions variably enriched by OIB-type components, which were inherited from the Triassic mantle plume activity associated with the opening of this Neotethys oceanic branch (Fig. 14a). Our findings show that P-MORBs have a Middle Jurassic age, G-MORBs have a Late Jurassic age, E-MORBs were erupted during Early Cretaceous time and alkaline basalts were erupted from Late Jurassic to Early Cretaceous times. Partial melting of different portions of a heterogeneous sub-oceanic mantle (Figs 13, 14b) can explain the formation of different and genetically unrelated magmatic rocks in the same time span.
Figure 14. Two-dimensional cartoon showing the geodynamic evolution of the Izmir–Ankara Neotethys branch from Late Triassic to late Early Cretaceous times (modified from Göncüoglu, Reference Göncüoglu2010). Abbreviations: MORB – mid-ocean ridge basalts; G- – garnet-influenced MORB; E- – enriched-MORB; P- – plume-type MORB; alk – alkaline basalts; SSZ – supra-subduction zone.
The intra-oceanic subduction likely started from late Early Cretaceous time on the northern side of the Neotethys branch, leading to the production of supra-subduction zone (SSZ)-type rocks during late Early Cretaceous time (Fig. 14c). Previous data on the earliest ages related to supra-subduction type volcanism and the formation of a metamorphic sole indicate an early Late Cretaceous age (e.g. T. Üner, unpub. Ph.D. thesis, Hacettepe Univ., 2010); therefore, the intra-oceanic decoupling within the ocean was considered as pre-Late Cretaceous in age. This age could not be verified in this study, since in contrast to the western part of the IAESB, none of the basalt samples collected from the eastern part yielded supra-subduction characteristics. The different rock types formed in the ocean were then incorporated into the accretionary prism during the Late Cretaceous closure.
7. Conclusion
This study is focused on slide blocks including oceanic lavas associated with pelagic sediments within the eastern part of the Ankara Mélange. A detailed petrological characterization of the volcanic rocks and a detailed biochronological investigation of the associated radiolarites was carried out. The main conclusions can be summarized as follows.
(1) The radiolarian cherts associated with volcanic rocks show the following ages: late Valanginian to late Hauterivian (Section 1, OIB); early–middle Bajocian to late Bathonian–early Callovian (Section 2, P-MORB); middle–late Oxfordian to late Kimmeridgian–early Tithonian (Section 3, OIB); middle late Barremian–early early Aptian (Section 4, E-MORB); middle late Oxfordian to late Kimmeridgian–early Tithonian and early–early late Tithonian (Section 5, G-MORB); Valanginian to middle Aptian–early Albian (Section 7, E-MORB); late Valanginian–early Barremian (Section 8, G-MORB).
(2) Volcanic rocks are largely represented by basalts and minor ferrobasalts and trachytes. They show different geochemical affinities and overlapping ages including: (a) Late Jurassic – Early Cretaceous G-MORB; (b) Early Cretaceous E-MORB; (c) Middle Jurassic P-MORB; (d) Late Jurassic – Early Cretaceous alkaline basalts. All rock types show a clear garnet signature, as testified to by their high MREE/HREE ratios.
(3) REE modelling shows that their garnet signature is related to polybaric partial melting starting in the garnet peridotite stability field and continuing to larger degrees in the spinel-facies mantle. However, the different geochemical affinities displayed by the studied rocks are related to different mantle source compositions. G-MORBs were generated from 2.5% melting in the garnet-facies and 10% melting in the spinel-facies of a DMM source. E-MORB composition is compatible with mixing of melts derived from a very low degree (~1%) of partial melting of a DMM source in the garnet-facies with melts derived from a much larger degree (~10%) of partial melting in the spinel-facies mantle. Alternatively, these rocks may have derived from ~8% partial melting of a theoretical slightly enriched source in the spinel-facies. P-MORBs and alkaline basalts have REE compositions that can be accounted for by mixing of different proportions of melts generated in the garnet- and spinel-facies from an enriched mantle source metasomatized by OIB-type (plume-type) components.
(4) The coexistence of chemically different rock types from Middle Jurassic to Early Cretaceous times suggests that they were formed in a mid-ocean ridge setting from partial melting of a highly heterogeneous mantle characterized by the extensive occurrence of OIB-metasomatized portions, which were likely inherited from a Triassic mantle plume activity associated with the continental rift and opening of the Neotethys branch.
Acknowledgements
The Italian Ministry of Education, University and Research (MIUR) is acknowledged for the financial support (Prin 2010–2011). This research has also been funded by the Ferrara University (FIR-2016 Project). Many thanks go to Mirella Bonora (Ferrara University) for her support with analytical techniques. Radiolarian micrographs were taken with a Philips XL20 of the Ivalsa Institute (CNR) by Simona Lazzeri and with a Zeiss EVO MA15 of the MEMA (University of Florence) by Maurizio Ulivi. Special thanks go to Špela Goričan and anonymous reviewer for their constructive reviews of this paper.
Supplementary material
To view supplementary material for this article, please visit https://doi.org/10.1017/S0016756817000401.
