Mineralogy and Petrology

, Volume 96, Issue 3–4, pp 177–196 | Cite as

Lutetian arc-type magmatism along the southern Eurasian margin: New U-Pb LA-ICPMS and whole-rock geochemical data from Marmara Island, NW Turkey

  • P. Ayda Ustaömer
  • Timur Ustaömer
  • Alan S. Collins
  • Jörg Reischpeitsch
Original Paper

Abstract

The rocks of Turkey, Greece and Syria preserve evidence for the destruction of Tethys, the construction of much of the continental crust of the region and the formation of the Tauride orogenic belt. These events occurred between the Late Cretaceous and Miocene, but the detailed evolution of the southern Eurasian margin during this period of progressive continental accretion is largely unknown. Marmara Island is a basement high lying at a key location in the Cenozoic Turkish tectonic collage, with a Palaeogene suture zone to the south and a deep Eocene sedimentary basin to the north. North-dipping metamorphic thrust sheets make up the island and are interlayered with a major metagranitoid intrusion. We have dated the intrusion by Laser Ablation ICP-MS analysis of U and Pb isotopes on zircon separates to 47.6 ± 2 Ma. We also performed major- and trace-elemental geochemical analysis of 16 samples of the intrusion that revealed that the intrusion is a calc-alkaline, metaluminous granitoid, marked by Nb depletion relative to LREE and LIL-element enrichment when compared to ocean ridge granite (ORG). We interpret the metagranitoid sill as a member of a mid-Eocene magmatic arc, forming a 30 km wide and more than 200 km long arcuate belt in NW Turkey that post-dates suturing along the İzmir-Ankara-Erzincan Suture zone. The arc magmatism was emplaced at the early stages of mountain building, related to collision of Eurasia with the Menderes-Taurus Platform in early Eocene times. Orogenesis and magmatism loaded the crust to the north creating coeval upward-deepening marine basins partially filled by volcanoclastic sediments.

Introduction

Understanding of the growth of mountain belts requires robust knowledge about the timing of tectono-thermal events, kinematic analysis and tectonic setting of formation.

Marmara Island, the largest island in the Marmara Sea, is a metamorphic basement high at the edge of S Marmara Shelf, bordered by a deep marine basin (i.e. the Tekirdağ trough) in the north, which is dissected by the dextral North Anatolian Transform Fault (Okay et al. 1999; Seeber et al. 2004). Marmara Island is located at a critical position in northwest Turkey, where several tectono-stratigraphic terranes are juxtaposed (Fig. 1). Several undated meta-sedimentary and meta-volcanic units are exposed in the island, forming N-dipping thrust sheets. These units have been assigned to the Triassic Karakaya Complex (Okay 2000) or have been interpreted as representing a passive margin trench collision zone that was subsequently transformed into an Andean-type arc setting before Permian times (Aksoy 1999).
Fig. 1

Tectonic map of the Aegean region showing the location of the study area and the suture zones discussed in the paper. Data sources are: Şengör and Yılmaz 1981; Brown and Robertson 2000; Okay 1989; Okay and Satır 2006; Okay et al. 2001; Ustaömer and Robertson 1997; Marchev et al. 2005

A WSW-ENE trending, 4 km wide and 20 km long, sill-like intrusion forms a central spine of hills crossing the central part of the island (Fig. 2). This intrusion is highly sheared and contains both a N-dipping foliation and a stretching lineation, trending parallel to the border, between the intrusion and adjacent units. Owing to its deformation and metamorphic state, various ages were proposed for the igneous intrusion. The suggested ages range from Pre-Permian (Caledonian?; Aksoy 1996) to Late Palaeozoic (MTA 2004, 1/500.000 İstanbul sheet) and to Oligo-Miocene (Okay and Satır 2000). If correct, each of these ages has important tectonic implications concerning the tectonic evolution of NW Turkey during the Cadomian, Caledonian, Hercynian or Alpine deformation periods.
Fig. 2

Geological map of the Marmara Island (modified after Aksoy 1995)

In this paper, we present new radiometric age data and geochemical evidence for the timing and petrogenesis of the Marmara Island igneous intrusion. We show that the magmatism is of arc-type and formed in early mid-Eocene, shortly after the collision of the Sakarya Continent with the Menderes-Tauride Platform. A number of plutons were emplaced into the crust in early to mid-Eocene and fed surface eruptions, which generated a narrow and arcuate igneous belt along the suture zones in NW Turkey.

Regional geology

Marmara Island is surrounded by several tectono-stratigraphic terranes of continental and oceanic origin (Fig. 1). The most northerly of these terranes is the Istranca Massif, which is a Palaeozoic-Mid Jurassic continental unit that records earlier Late Palaeozoic (Hercynian) deformation and amphibolite-facies metamorphism and subsequent (Triassic) crustal extension and basin formation (Ustaömer and Robertson 1993; Okay et al. 2001). This basin was closed by Late Jurassic times when the whole terrane underwent a second metamorphism in greenschist facies (Okay et al. 2001). The Armutlu Peninsula, along strike further to the east, comprises two tectonically assembled continental metamorphic units (the Armutlu and İznik Metamorphics) and an ophiolitic suture zone, termed the Intra-Pontide Suture (Şengör and Yılmaz 1981; Göncüoğlu and Erendil 1990; Yılmaz et al. 1995; Robertson and Ustaömer 2004). The Armutlu Metamorphics comprise an amphibolite facies high-grade basement, transgressively overlain by mixed clastic-carbonate and volcanogenic units, metamorphosed to greenschist facies. The volcanogenic rocks in the Armutlu Metamorphics are of MORB-type with little subduction influence (Robertson and Ustaömer 2004; Ustaömer and Robertson 2005; Yiğitbaş et al. 2004). The age of the Armutlu Metamorphics is unknown; with both Precambrian and Palaeozoic ages previously proposed (Göncüoğlu and Erendil 1990; Göncüoğlu et al. 1992; Yılmaz et al. 1995; Robertson and Ustaömer 2004; Ustaömer and Robertson 2005; Yiğitbaş et al. 2004). Recent age data have shown that late Proterozoic (ca 570 Ma) and Ordovician granitoids (ca 460 Ma) occur within the basement (Okay et al. 2008a). The İznik Metamorphics are in tectonic contact with the Armutlu Metamorphics and comprise a basement metavolcanic succession with a rift-like geochemistry (Robertson and Ustaömer 2004). This is unconformably overlain by a Triassic-Late Cretaceous passive margin succession that is overridden by an ophiolitic melange.

A third tectonic terrane occupies a large area to the S of the Marmara Sea and is termed the Sakarya Continent (Şengör and Yılmaz 1981) or Sakarya Zone (Okay 1989; Fig. 1). The pre-Jurassic basement of the Sakarya Zone is made up of a chaotic rock assemblage, called the Karakaya Complex, interpreted as a subduction-accretion complex of Triassic age related to northward subduction of Palaeotethys (Ustaömer and Robertson 1993; Pickett and Robertson 1996, 2004; Okay 2000). Several tectonic units are distinguished within the Karakaya Complex, the most extensive of which is the Nilüfer Unit, consisting of greenschist facies metabasic and metasedimentary rocks. Geochemistry of the metabasic rocks indicates a seamount or LIP (Large Igneous Province) origin for the Nilüfer Unit (Pickett and Robertson 1996; Genç 1998; Okay 2000). The Karakaya Complex is transgressed by Lower Jurassic to Late Cretaceous sedimentary rocks that are interpreted as an unmetamorphosed passive margin succession facing south to Neotethys (Yılmaz et al. 1997). The Sakarya Zone is bound by the İzmir-Ankara-Erzincan Suture Zone to the south and by the Intra-Pontide Suture to the north (Fig. 1). The İzmir-Ankara-Erzincan Suture was closed in Palaeocene-Eocene times, leading to collision of the Menderes-Taurides continent (i.e. Menderes Massif) and the Sakarya Zone (Şengör and Yılmaz 1981; Collins and Robertson 1998; Okay et al. 1996, 1998; Harris et al. 1994). The Intra-Pontide Suture Zone, on the other hand, had already closed in the Late Cretaceous, leading to the amalgamation of the Sakarya Continent with Eurasia (Şengör and Yılmaz 1981; Yılmaz et al. 1995; Göncüoğlu et al. 1992; Robertson and Ustaömer 2004; Ustaömer and Robertson 2005).

Marmara Island is isolated in the Marmara Sea, which is surrounded by a complex tectonic mosaic of several tectono-stratigraphic terranes and fault zones. Because of this, there is no direct evidence as to which terrane the metamorphic units of Marmara Island belong. A direct tie with the İstanbul Fragment to the NE can be ruled out as Palaeozoic to younger units of the İstanbul Fragment do not show any metamorphism (Abdüsselamoğlu 1977; Şengör and Yılmaz 1981; Şengör et al. 1984; Ustaömer and Robertson 1993). A connection with the other units described above is possible when the metamorphic history of the rock units is considered; however, there are important stratigraphic differences between those terranes. Okay and Satır (2000) and Beccaletto and Jenny (2004) have suggested that a suture between the Sakarya Zone and the Rhodope Massif passes to the north of the Marmara Island, leaving Marmara Island at the northern edge of the Sakarya Zone. It is yet not possible to constrain the provenance of the metamorphic units of the Marmara Island.

Tectono-stratigraphy

Three N-dipping thrust sheets are present in Marmara Island (Fig. 2). The lower thrust sheet is made up of a high-grade metamorphic unit (the Gündoğdu Complex of Aksoy 1995), composed of pelitic and psammitic schists and marbles. The pelitic schists have a mineral assemblage of staurolite+garnet+kyanite+quartz+albite+biotite and suggest metamorphism at amphibolite facies conditions (Tanyolu 1979; Aksoy 1995). This unit is over thrust by a metabasite unit (Erdek Complex; Aksoy 1995). Both meta-basic lavas and tuffaceous sedimentary rocks are present in this unit, together with marble lenses and tectonic blocks of serpentinite. Pelitic parts of this unit have a similar mineral assemblage to the meta-pelites of the underlying Gündoğdu Complex, which suggests no metamorphic break is present between the Erdek and Gündoğdu Complexes. The Erdek Complex is overlain by a thick (ca. 1.5 km) marble unit (the Marmara Marble). The contact between the Erdek Complex and the overlying Marmara Marble has been interpreted either as conformable (Yüzer 1971; Tanyolu 1979) or as an unconformity (Aksoy 1995). The contact is cut by the metagranitoid intrusion in the westerly areas, whereas in the eastern part of the island a transition occurs between the two units.

The Marmara Marble is overlain by the Saraylar Complex (Aksoy 1995). This is a chaotic unit, metamorphosed to greenschist facies conditions. The matrix of the unit is represented by meta-pelitic-psammitic sediments with occasional interbedded layers of meta-carbonates and metacherts. The blocks in this unit are dominantly marbles. Gabbro, metabasite and ultramafic blocks are also present (Aksoy 1995).

İlyasdağ metagranitoid

The İlyasdağ metagranitoid is a WSW-ENE trending sill-like intrusion that cuts the Marmara Marble and the Erdek Complex. It can be traced across the island (Fig. 2). At the present level of exposure its N-S width is nearly 4 km but thins down to 1.5 km in the central part of the island. The intrusion is dark grey coloured on fresh surfaces. It is a medium to coarse grained intrusion and is cut by aplitic and pegmatitic veins. The grain size is reduced towards the margins. Feldspar, quartz, amphibole and biotite are the main minerals present in the intrusion. Screens of host rock (amphibolite) form narrow and semi-continuous zones parallel to the general strike of the intrusion (Fig. 2).

A number of aplitic, pegmatitic and granitoid veins and dykes, related to intrusion of the İlyasdağ metagranitoid are found in the nearby Marmara Marble and Erdek Complex. Evidence for contact metamorphism was determined locally in the neighbouring metacarbonates (Tanyolu 1979; Aksoy 1995). The contact metamorphic mineral assemblage indicates a metamorphism at albite-epidote hornfels (quartz+actinolite+epidote+sphene) and hornblende-hornfels (hornblende+epidote+plagioclase+quartz+garnet) facies (Tanyolu 1979).

The İlyasdağ metagranitoid was previously termed gneiss (Ketin 1946), orthogneiss (Yüzer 1971), granodioritic orthogneiss (Tanyolu 1979) and İlyasdağ Metagranodiorite (Aksoy 1995). The metagranitoid exhibits a weakly developed mylonitic foliation and a stretching lineation. The mylonitic foliation appears to have formed by the parallel alignment of micas and amphiboles on the foliation planes. The stretching lineation is marked by elongation of quartz and needle-shaped amphibole crystals. The intrusion comprises occasional deformed mafic mineral enclaves (MME) and xenoliths of the country rocks, which were described and studied by Aksoy (1995). The MMEs are parallel to foliation planes and are elliptic in shape. The long axes of the MME reach up to 1 m. The deformation amount was calculated from the shape analysis of MME as 62% elongation (parallel to the mean strike; i.e. WSW-ENE) and 38% shortening (perpendicular to the foliation plane, i.e. N-S) (Aksoy 1995).

According to Aksoy (1995), the foliation and lineation of the metagranodiorite is at the same attitude as those in the adjacent metamorphic units in the Marmara Island. He therefore thinks that all the units were deformed and metamorphosed together at the same stage.

Petrography of the metagranitoid

The main mineral phases in the metagranitoid are quartz+plagioclase (andesine)+alkali feldspar (orthoclase and microcline)+amphibole (hornblende)+ biotite, accompanied by titanite, zircon, apatite and some opaque minerals in the accessory phases. A secondary mineral assemblage consisting of epidote ± chlorite ± actinolite ± muscovite is developed as a result of hydrothermal alteration.

The igneous fabric of the medium to coarse grained İlyasdağ metagranitoid has largely been replaced by a brittle-ductile deformation resulting in grain-size reduction, recrystallization and development of a pervasive mylonitic foliation. The most prominent feature of the cataclastic structure is the grain size reduction, recrystallisation and development of a mylonitic foliation which was observed in all thin sections examined. In general, the mafic phases (biotite and amphiboles) envelope the felsic mineral rich zones and creates an anastomosing foliation. A secondary shear band foliation is also observed locally, forming an SC-structure.

Both plagioclase and alkali feldspars suffered brittle deformation and their grain sizes were reduced considerably. Such anhedral-subhedral feldspar porphyroclasts or augens exhibit deformation lamellae and pericline twins. Feldspar shows core- and -mantle structures. Strongly zoned plagioclase minerals are partly replaced by secondary epidote group minerals and chlorites. Some feldspar crystals have sericitized cores.

Quartz underwent dislocation creep that resulted newly formed small recrystallized grains, the latter showing occasional ribbon structure. Larger quartz grains in such aggregates show undulose extinction and are composed of elongate subgrains. Grain boundary migration is the principle deformation mechanism of quartz in the metagranitoid.

Biotites are brown-greenish brown and exhibit plastic deformation into elongate lenses. Such coarse grained biotites are seen to have been replaced with smaller neocrystallised grains of biotite, muscovite, titanite, chlorite, epidote and opaques. Kinking along, and tight folding outside the foliation planes are common microstructures exhibited by biotite. In several sections, biotite shows a fish-eye fabric, preserved along shear planes.

Amphiboles are also affected by grain size reduction as a result of brittle deformation. Both euhedral to subhedral amphibole grains occur and show pale to dark green pleochroism. Poikiloblastic amphiboles contain biotite inclusions. Fine grained aggregates of biotite also mantles the euhedral grains locally. The amphiboles break down to an assemblage of actinolite, chlorite and epidote.

Despite these deformation structures dominating the rock fabric, the relict igneous fabrics encountered at individual minerals or mineral groups record evidence for magma replenishment or mixing processes. These include oscillatory zoning in plagioclase, poikilitic feldspars containing amphibole and plagioclase inclusions, and presence of apatite needles in feldspars. Perthitic fabrics are encountered occasionally in alkali feldspar porphyroclasts. Titanite forms euhedral and subhedral grains. Euhedral zircons are commonly present as small inclusions in plagioclase.

Analytical methods

Geochronology

Zircons were separated by conventional methods, then were hand-picked and mounted in epoxy resin. Zircon mounts were imaged by cathodoluminescence (CL), using standard imaging techniques on a Phillips XL20 SEM with attached Gatan CL.

U-Pb analysis of individual zircons was conducted using the laser inductively coupled plasma mass spectrometer (LA-ICPMS) at the University of Adelaide. Equipment and operating conditions for zircon analysis are identical to those reported by Payne et al. (2006). A spot size of 30 μm and repetition rate of 5 Hz was used for U-Pb data acquisition, producing a laser power density of approximately 8 J/cm−2. Zircon ages were calculated using the GEMOC GJ-1 zircon standard to correct for U-Pb fractionation (TIMS normalisation data 207Pb/206Pb = 608.3 Ma, 206Pb/238U = 600.7 Ma and 207Pb/235U = 602.2 Ma—Jackson et al. (2004)), and the GLITTER software for data reduction (Jackson et al. 2004). Over the duration of this study the reported average normalised ages for GJ-1 are 607.9 ± 3.6, 600.7 ± 1.0 and 602.4 ± 0.8 Ma for the 207Pb/206Pb, 206Pb/238U and 207Pb/235U ratios, respectively (n = 209). Accuracy was monitored by repeat analyses of the Sri Lankan in-house internal standard (BJWP-1, 207Pb/206Pb = 720.9 ± 1.8 Ma, 206Pb/238U = 720.4 ± 0.5 Ma and 207Pb/235U = 720.5 ± 0.6 Ma, unpublished Massachusetts Institute of Technology TIMS data). Over the duration of this study the reported average ages for BJWP-1 are 744 ± 22, 721.4 ± 6.6 and 727.1 ± 6.1 Ma for the 207Pb/206Pb, 206Pb/238U and 207Pb/235U ratios, respectively (n = 7).

U-Pb LA ICP-MS Zircon ages

The analysed zircon grains ranged in 206Pb/238U age from 56.7 ± 0.8 Ma to 46.1 ± 0.7 Ma (Table 1, Fig. 3). Six analyses are older than 50 Ma and are considered to be mixed analyses that sampled different age domains within the zircon grain. These analyses are thought to consist partially of xenocrystic zircon incorporated into the magma during its ascent through the crust. A tight age-group of 39 analyses yield a precise 206Pb/238U age of 47.6 ± 0.2 Ma (95% confidence, MSWD = 0.99). This concordant population, that correlates with euhedral, oscillatory-zoned zircon, is interpreted as dating crystallisation of the İlyasdağ metagranitoid (Fig. 3).
Table 1

Quadrupole laser ablation ICP-MS U/Pb analysis of zircons

 

Ages (Ma)

Analysis

Pb206/U238

±(1)

Pb207/U235

±(1)

Pb207/Pb206

±(1)

Pb207/Pb206

±(1)

Pb206/U238

±(1)

M1Q102

0.00761

0.00013

0.05554

0.00199

0.05297

0.00194

327.3

81.1

48.8

0.8

M1Q106

0.00748

0.00012

0.0504

0.00148

0.04888

0.00144

142.3

67.7

48.0

0.7

M1Q107

0.00741

0.00012

0.05327

0.00182

0.05215

0.0018

291.8

77.0

47.6

0.8

M1Q109

0.00718

0.00011

0.04649

0.00133

0.04693

0.00133

45.2

66.6

46.2

0.7

M1Q110

0.00747

0.00011

0.05264

0.00141

0.05115

0.00136

247.4

60.0

47.9

0.7

M1Q111

0.00741

0.00013

0.0507

0.00195

0.04962

0.00194

177.3

88.8

47.6

0.8

M1Q113

0.00743

0.00012

0.05196

0.00151

0.05073

0.00147

228.8

65.7

47.7

0.7

M1Q114

0.00855

0.00012

0.06126

0.00135

0.05197

0.0011

284.1

47.8

54.9

0.8

M1Q208

0.00737

0.00011

0.0529

0.0015

0.05204

0.00148

287.2

63.8

47.4

0.7

M1Q210

0.00745

0.00012

0.05086

0.00157

0.04953

0.00153

173.0

70.6

47.8

0.8

M1Q305

0.00728

0.00012

0.05421

0.00175

0.05401

0.00176

371.5

71.6

46.8

0.8

M1Q312

0.00752

0.00012

0.05527

0.00155

0.0533

0.00149

341.5

62.0

48.3

0.7

M1Q316b

0.00733

0.00012

0.06748

0.00203

0.06681

0.00204

831.7

62.4

47.1

0.8

M1Q318

0.00756

0.00014

0.05075

0.00231

0.04869

0.00227

132.7

106.3

48.6

0.9

M1Q403b

0.00742

0.00012

0.04895

0.00168

0.04787

0.00166

91.7

81.1

47.6

0.8

M1Q407

0.0074

0.00011

0.05247

0.00146

0.05142

0.00143

259.7

62.5

47.5

0.7

M1Q409

0.00732

0.00011

0.04884

0.00121

0.04838

0.00118

118.0

56.5

47.0

0.7

M1Q410

0.00744

0.00012

0.05284

0.00161

0.05147

0.00158

262.1

68.8

47.8

0.8

M1Q411

0.00764

0.00011

0.05465

0.00145

0.05188

0.00136

279.9

58.9

49.1

0.7

M1Q412

0.00751

0.00013

0.07896

0.00263

0.07627

0.00262

1,102.2

67.3

48.2

0.8

M1Q414a

0.00879

0.00013

0.06263

0.00154

0.05171

0.00125

272.7

54.3

56.4

0.8

M1Q415a

0.00735

0.00011

0.05776

0.00161

0.05697

0.00159

489.9

61.0

47.2

0.7

M1Q418

0.00747

0.00011

0.05057

0.00133

0.04909

0.00128

152.3

60.0

48.0

0.7

M1Q508

0.00755

0.00011

0.05085

0.00125

0.04885

0.00119

140.8

56.0

48.5

0.7

M1Q510

0.00768

0.00012

0.05842

0.00178

0.05514

0.0017

417.8

66.8

49.3

0.8

M1Q512

0.00736

0.00011

0.05108

0.00137

0.05038

0.00133

212.4

60.2

47.2

0.7

M1Q516a

0.00745

0.00011

0.05162

0.00138

0.05027

0.00133

207.6

60.4

47.8

0.7

M1Q518

0.00747

0.00011

0.04924

0.0014

0.04784

0.00135

90.2

66.7

48.0

0.7

M1Q606

0.00724

0.0001

0.04785

0.00121

0.04796

0.0012

96.2

59.2

46.5

0.7

M1Q607

0.00745

0.00012

0.05197

0.00164

0.05064

0.0016

224.6

71.4

47.8

0.8

M1Q608

0.00724

0.0001

0.04775

0.0011

0.04787

0.00107

91.6

53.0

46.5

0.7

M1Q609b

0.00736

0.00011

0.0502

0.00136

0.04949

0.00132

171.1

61.2

47.3

0.7

M1Q610

0.00733

0.00011

0.04844

0.0014

0.04795

0.00137

95.7

67.6

47.1

0.7

M1Q611

0.00779

0.00011

0.05407

0.00106

0.05034

0.00093

210.6

42.4

50.0

0.7

M1Q612

0.0074

0.00012

0.05363

0.00165

0.0526

0.00162

311.4

68.6

47.5

0.8

M1Q613a

0.00746

0.00013

0.05511

0.00201

0.05362

0.00199

354.9

81.5

47.9

0.8

M1Q613b

0.00756

0.00012

0.05395

0.0017

0.05177

0.00164

275.1

70.8

48.6

0.8

M1Q615a

0.00743

0.00011

0.05029

0.00112

0.04912

0.00105

153.6

49.4

47.7

0.7

M1Q615b

0.00735

0.00012

0.04998

0.0015

0.04933

0.00147

163.6

68.3

47.2

0.8

M1Q616

0.00738

0.0001

0.05114

0.0013

0.05021

0.00129

204.9

58.3

47.4

0.7

M1Q617

0.00883

0.00013

0.0636

0.00135

0.05227

0.00105

297.4

45.1

56.7

0.8

M1Q618

0.00718

0.00011

0.04857

0.00151

0.04905

0.00155

150.1

72.3

46.1

0.7

M1Q619a

0.00788

0.0001

0.05841

0.00095

0.05377

0.00079

361.1

32.8

50.6

0.7

M1Q619b

0.00731

0.00011

0.05008

0.00144

0.04971

0.00143

181.3

65.5

46.9

0.7

M1Q620

0.00797

0.00013

0.05201

0.00194

0.04736

0.0018

66.7

88.6

51.1

0.8

M1Q621

0.00795

0.00012

0.05438

0.00161

0.04962

0.00149

177.2

68.6

51.0

0.8

(1) Absolute errors, 1 sigma

Fig. 3

U-Pb concordia diagram of zircon analyses from the İlyasdağ tonalite. Insert is a typical cathodoluminescence image of an oscillatory zoned zircon from the rock and also a plot of the analyses used to create the weighted mean

As noted above, the age of the İlyasdağ metagranitoid was previously thought to be Late Palaeozoic or older based on its deformation state. However, our new age data has shown that the İlyasdağ metagranitoid intrusion is in fact an early mid-Eocene pluton. The age of the brittle-ductile deformation that the pluton exhibits, therefore, also occurred in the syn- to post mid-Eocene period.

Magmatism of Eocene age is known in NW Anatolia and NE Greece (Serbo-Macedonian and Rhodope Massifs). The distribution of granitoids, volcanic rocks and sedimentary units of Eocene age in NW Turkey, together with the ophiolitic thrust sheets of the Tethyan suture zones are shown on Fig. 4. The İzmir-Ankara-Erzincan Suture to the south was formed in Late Palaeocene-early Eocene by complete closure of N Neotethys (Şengör and Yılmaz 1981; Robertson and Dixon 1984; Collins and Robertson 1998; Okay 2000). The Intra-Pontide Suture to the north, on the other hand, is a Late Cretaceous feature (Yılmaz et al. 1995; Göncüoğlu et al. 1992; Robertson and Ustaömer 2004; Ustaömer and Robertson 2005). The only ocean basin that known to have survived Eocene times was South Neotethys, which was open until the Miocene and partially closed along the Bitlis Suture Zone by collision of the Arabian promontory with the E Taurides (Robertson et al. 2006). The present Eastern Mediterranean basin is a remnant of South Neotethys in this area, which is actively closing by northward subduction along the Hellenic trench (Fig. 1).
Fig. 4

Simplified geological map of NW Turkey showing distribution of Eocene magmatic rocks and sedimentary basins. Known suture zones are also shown. Data sources are: Harris et al. 1994; Delaloye and Bingöl 2000; Okay and Satır 2006; Beccaletto et al. 2007

Eocene granitoids are found in two separate, E-W trending, regions in NW Anatolia. The N-S widths of the northern and southern plutons are 25 and 30 km, respectively, and the N-S distance between the two groups of plutons is ~100 km. The southern group (also termed as the suture zone granitoids; Altunkaynak 2007; Karacık et al. 2008) includes several plutons that intrude the ophiolites of the İzmir-Ankara-Erzincan Suture zone and the underlying Late Cretaceous blueschist facies metamorphic rocks (Tavşanlı Zone; Okay 1984, 1989) of the Menderes-Taurus platform. These southerly plutons were dated by 40Ar/39Ar (Harris et al. 1994), K/Ar (Delaloye and Bingöl 2000) and U/Pb methods (Okay and Satır 2006) to early-middle Eocene (53 to 44 Ma), one being nearly identical in age to the granitoid of the Marmara Island at 47.8 ± 0.4 Ma (the Topuk granodiorite; Harris et al. 1994). Eocene magmatism further to the south of the blueschist belt is not known. All the dated post-Triassic granitoids in the Menderes Massif yield Early Miocene emplacement ages (e.g. Ring and Collins 2005).

The northern group of plutons occurs in Armutlu and Kapıdağ Peninsulas and in Karabiga, to the north of the Intra-Pontide Suture Zone (termed the Marmara granitoids; Altunkaynak 2007; Karacık et al. 2008). There are no known Eocene plutons in the Sakarya Continent, between the southern and northern plutonic belts (Fig. 4). This area is occupied by Oligocene to Miocene intrusions and volcanics, constructed upon Cretaceous to older sedimentary-metamorphic units (Okay et al. 2008b). The northern plutons are also associated with Eocene volcanics that form a continuous belt from Çanakkale-Biga to the Armutlu Peninsula and further east (Fig. 4). In contrast to the southerly plutons, the northern plutons have mostly been dated using the K/Ar method on mineral separates and different minerals yielded different ages for individual plutons (Delaloye and Bingöl 2000). Uncertainty, therefore, exists for the exact age of crystallisation of these plutons. For instance, two biotite analyses yielded 48.2 ± 1 and 35.4 ± 0.8 Ma and one muscovite analysis yielded 34.3 ± 0.9 Ma cooling ages for the Fıstıklı pluton in Armutlu Peninsula (Fig. 4; Delaloye and Bingöl 2000). A similarly wide age range was gathered for the north and south Kapıdağ plutons. Two biotite analyses yielded 39.9 ± 0.8 and 38.3 ± 0.8 Ma, and one hornblende analysis yielded 42.2 ± 1 Ma for the N Kapıdağ pluton whereas 38.2 ± 0.8 and 36.1 ± 0.8 Ma ages were determined from two biotite analyses from the S Kapıdağ pluton (Delaloye and Bingöl 2000). Only one biotite analysis was made from the Karabiga pluton that yielded an age of 45.3 ± 0.9 Ma (Delaloye and Bingöl 2000). Recently, Beccaletto et al. (2007) dated xenotime in the Karabiga pluton and obtained a crystallisation age of 52.7 ± 1.9 Ma. Thus, the U-Pb age dating of the İlyasdağ Pluton provided one of the most robust age data gathered from the northern plutons.

While the deeper part of the crust was invaded by these plutons in mid- to late Eocene times, stratigraphic data suggest that the surface was covered with a shallow to deep sea, as carbonates and clastics of this age are found throughout north west Turkey (Saner 1985; Okay and Satır 2006; Siyako and Huvaz 2007). Before this, the early Eocene was probably the time of uplift and erosion due to closure of Neotethys and subsequent collision of the Menderes and Sakarya continents, since marine sediments of this age are scarce in NW Anatolia.

Below, we will present the geochemical properties of the İlyasdağ pluton (Table 2). We incorporate published geochemical data from the northern and southern group of plutons (Harris et al. 1994; Delaloye and Bingöl 2000; Köprübaşı and Aldanmaz 2004) with an aim to show the similarities and differences of the İlyasdağ pluton from the rest of the Eocene plutons. As we will show, the İlyasdağ pluton forms the most basic Eocene pluton in NW Anatolia.
Table 2

Geochemical analysis of the İlyasdağ tonalite

Sample

MR1

MR4

MR5

MR6

MR7

MR8

MR10

MR12

MR14a

MR15

MR16

MR18

MR20

MR25

MR27

M1

SiO2

62,59

65,49

63,03

64,96

62,65

57,2

63,32

60,01

62,71

60,61

66,17

65,09

74,7

75,6

62,13

68,53

TiO2

0,48

0,41

0,49

0,4

0,43

0,76

0,43

0,63

0,49

0,51

0,38

0,4

0,08

0,04

0,49

0,32

Al2O3

17,65

17,18

17,62

16,94

17,75

19,53

17,49

18,32

17,54

17,96

16,7

17,39

13,46

13,9

17,99

16,27

Fe2O3

4,63

4,04

4,56

4,11

4,56

6,97

4,58

5,99

4,66

5,05

3,67

3,99

0,83

0,39

4,91

3,21

MnO

0,15

0,13

0,14

0,13

0,13

0,16

0,12

0,16

0,13

0,14

0,12

0,13

0,04

0,02

0,13

0,12

MgO

1,73

1,53

1,68

1,58

1,99

2,25

1,93

2,26

1,85

2,02

1,32

1,46

0,21

0,06

2,06

1,13

CaO

6,16

5,35

5,93

5,32

6,17

7,82

5,73

7,15

6,4

6,11

4,88

5,23

1,4

0,97

6,44

4,25

Na2O

4,03

4,06

4,01

3,95

3,82

3,78

3,68

3,87

3,91

3,79

3,96

4,03

3,71

5,28

3,88

4,15

K2O

1,2

1,11

1,2

1,41

1,24

0,27

1,47

0,85

0,85

1,08

1,2

1,4

3,81

2,86

0,94

2

P2O5

0,13

0,11

0,13

0,12

0,14

0,21

0,15

0,18

0,14

0,17

0,11

0,12

0,02

0,01

0,16

0,09

SO3

0,02

0,01

0,01

0,02

0,01

0,01

0,01

0,01

0,01

0,01

0,01

0,01

0,01

0,01

0,01

0,02

LOI

0,88

0,56

0,74

0,81

0,83

1,04

0,9

0,61

0,86

1,07

0,92

0,62

0,34

0,2

0,83

0,44

Total

99,65

99,97

99,55

99,74

99,71

100

99,81

100,04

99,54

98,53

99,44

99,88

98,6

99,32

99,98

100,52

Zr

99

90

109

91

79

97

82

101

82

93

97

110

44

68

90

85

Nb

2

3

2

2

2

3

3

2

3

3

2

3

2

2

1

1

Y

21

18

23

18

16

35

16

23

19

20

15

19

9

18

21

15

Sr

337

281

336

309

413

362

392

396

368

378

318

347

143

42

386

292

Rb

34

32

35

36

37

5

41

24

27

36

39

42

76

60

27

50

U

2

1

1

2

3

1

2

0

1

2

1

1

2

3

2

1

Th

6

4

3

5

6

4

6

1

4

3

4

6

10

13

4

6

Pb

10

10

12

12

15

7

16

11

11

11

13

16

28

28

11

27

Ga

18

17

16

15

16

20

16

17

15

16

16

17

10

14

18

15

Cu

18

14

17

12

7

44

7

19

13

15

9

27

7

16

15

9

Zn

62

57

62

57

58

77

58

70

53

64

56

55

15

3

59

45

Ni

3

2

3

2

4

4

4

4

7

6

6

2

1

1

6

3

Ba

280

295

295

360

346

198

359

223

224

289

361

340

929

290

218

325

Sc

8

6

7

6

8

10

7

11

8

9

5

6

2

2

10

5

Co

66

47

53

54

38

35

45

48

52

28

29

36

40

63

43

6

V

79

62

81

64

92

154

88

116

83

86

60

62

14

6

88

52

Ce

43

28

27

30

24

28

30

18

17

18

20

26

21

16

22

25

Nd

14

10

10

7

11

13

9

4

9

7

6

11

4

4

9

5

La

16

8

7

6

7

4

8

0

3

4

4

8

10

3

3

8

Cr

1

1

3

0

3

0

2

5

17

8

16

2

2

5

8

7

Geochemistry

A total of 16 samples were selected for geochemical analysis. Fourteen of these are from the granitoid and two from the cross-cutting aplitic veins. The samples were analyzed at the University of Adelaide, South Australia.

Whole rock (XRF) chemistry of the granitoid

The analytical method has been described in Ustaömer et al. (2009) and the results are given in Table 2. The graphics in this section were produced by using the software GCDKit (Janousek et al. 2006).

Major elements

SiO2 values of the İlyasdağ metagranitoid samples vary between 57 to 69%. The cross-cutting aplitic veins, however, have higher silica values (> 71%). The Mg number of the granitoid samples varies between 39 and 46 but that of the aplitic veins is less than 35.

The İlyasdağ metagranitoid samples plot in the tonalite field, whereas the aplitic samples plot in the monzogranite field on the Q’ versus ANOR nomenclature diagram (Fig. 5A). Several other nomenclature diagrams (P versus Q diagram of Debon and Le Fort 1983; R1 versus R2 diagram of De la Roche et al. 1980; TAS diagram, Middlemost 1985) yielded a similar result (not shown). The K2O content of the İlyasdağ metatonalite varies from 0.27 % to 2 %. This is well illustrated on a SiO2 versus K2O diagram (Peccerillo and Taylor 1976) in which the metatonalite samples plot in the low- to medium-K fields (Fig. 5B), indicating a tholeiitic to calc-alkaline nature of the melt. The aplitic samples, on the other hand, plot in the medium to high-K fields on the same diagram. The metatonalite is classified as a calcic pluton in a SiO2 versus MALI (Modified alkali-lime index: Na2O + K2O-CaO; Frost et al. 2001) diagram (Fig. 5C). Two of the aplite samples plot in calc-alkaline and one in alkali-calcic field on the same diagram. The SiO2 versus FeOt/FeOt + MgO diagram of Frost et al. (2001) is used to measure the iron enrichment of the intrusive rocks. The metatonalite and one aplite sample plot in the magnesian field and two aplite samples plot in ferroan field (Fig. 5D). The boundaries of Caledonian (post-collisional) and Cordilleran plutons (magmatic arc) are also shown on these diagrams. The İlyasdağ metatonalite samples exhibit a similar composition to the Cordilleran plutons (Fig. 5C, D). The Shand index is used to distinguish aluminium saturation level of the samples studied (Fig. 6). The metatonalite samples plot in the metaluminous and the aplitic samples plot in the peraluminous field. Normative corundum in the aplitic set is, however, less than 0.8, suggesting that the aplites are weakly peraluminous.
Fig. 5

a Q vs ANOR nomenclature diagram; b SiO2 versus K2O diagram (Peccerillo and Taylor 1976); c SiO2 vs MALI (Modified Alkali Lime Index) diagram (Frost et al. 2001); D) SiO2 vs FeOt/FeOt + MgO diagram (Frost et al. 2001)

Fig. 6

A/CNK (molar Al2O3/CaO + Na2O + K2O) versus A/NK (molar Al2O3/Na2O + K2O) diagram (Shand 1927). See text for explanation

There is a good correlation of the major and minor oxides with SiO2. The correlation coefficient for most of these oxides is above 0.97, except Na2O which has a correlation coefficient of 0.68. Scatter of Na2O of the aplitic and only a few intrusive samples is interpreted as an indication of mobility of this element during post-crystallisation alteration.

Published geochemical data on Eocene plutons of NW Anatolia are plotted along with the İlyasdağ data on Harker variation diagrams (Fig. 7). Note that the ages of these plutons are variable, recording at least 15 Ma of magmatic processes that operated in NW Anatolia between 50 to 35 Ma. Field and petrographic features of the northern plutons suggest that magma mixing is one of the principle mechanisms for differentiation of these plutons, together with assimilation and fractional crystallisation (Köprübaşı and Aldanmaz 2004). The trends observed on Harker variation diagrams of Eocene plutons are the result of these magmatic processes. Overall, Al2O3, CaO, MgO, FeO*, TiO2 and P2O5 show strong negative correlation with increasing SiO2, while K2O and Na2O show a positive correlation with increasing SiO2 (Fig. 7). The İlyasdağ pluton is the most basic pluton in the data set with its low SiO2 and high MgO content. Some of the Kapıdağ samples plot off trend, otherwise the Eocene plutons lay on the same differentiation trend on major oxide Harker diagrams. A great scatter of Na2O of the Eocene plutons (except the İlyasdağ metatonalite) reflects the high mobility of this oxide during weathering and alteration. Therefore, the nomenclature and variation diagrams involving Na2O content of such plutons should be interpreted with caution.
Fig. 7

Major element Harker variation diagrams of the Marmara Island intrusives. See text for explanation (Lapseki-Armutlu-Kapıdağ plutons data are compiled from Köprübaşı and Aldanmaz 2004; Orhaneli-Topuk plutons data are compiled from Harris et al. 1994)

Trace elements

The İlyasdağ metatonalite is characterised by low Nb, Rb, Th and high V, Sc and Pb contents. Unlike major oxides, the Eocene plutons display different variation trends based on trace element compositions, indicating differences in their source compositions and subsequent crystallisation history. For instance, the Kapıdağ and Orhaneli plutons show first enrichment and depletion whereas the İlyasdağ, Topuk, Lapseki and Armutlu plutons show depletion of Sr with increasing SiO2 (not shown). Different variations in the Ba content of the plutons with SiO2 are also evident. The İlyasdağ pluton shows a moderate Ba enrichment, whereas the Kapıdağ pluton shows a strong enrichment in Ba, and the Armutlu and Lapseki plutons exhibit strong Ba depletion (not shown). Rb enrichment with increasing SiO2 is also strong in the Kapıdağ pluton, moderate in the İlyasdağ pluton, but strongly depleted in the Armutlu pluton, although some variations in the data may relate to mobility of this element. Nb is highly depleted in the İlyasdağ pluton in comparison to rest of the Eocene plutons. Nb decreases weakly with increasing SiO2 in the İlyasdağ pluton but it decreases strongly in all the Armutlu, Kapıdağ and Lapseki plutons (not shown).

Differences in source characteristics of these plutons are clearer in compatible-incompatible element ratios diagram. In Th versus Th/Nb diagram (Fig. 8A), the İlyasdağ and Kapıdağ-Orhaneli-Topuk-Lapseki plutons form two parallel positive trends and the Armutlu pluton occupies the space between these two trends. A similar feature is observed in the Rb versus Rb/Nb, Y/Nb versus Rb/Nb and Rb/Zr versus Nb diagrams (Fig. 8B, C, D), implying that the source characteristics of the İlyasdağ Pluton is different from the rest of the plutons. Low Rb (4–49 ppm), Th (1–6.4 ppm), Nb (0.8–2.8 ppm), Rb/Sr (0.01–0.12) and high Y/Nb (6–13) values of the İlyasdağ pluton distinguish it from the other Eocene plutons of NW Anatolia.
Fig. 8

a- Th vs Th/Nb diagram of the NW Anatolian Eocene granitoids (İlyasdağ and Armutlu, Kapıdağ-Orhaneli-Topuk-Lapseki plutons); b Rb versus Rb/Nb diagram; c Y/Nb versus Rb/Nb diagram; d Rb/Zr vs Nb diagram. See Fig. 7 for the key and text for explanation

Further information can be gleaned from multi-element spidergrams. On an ORG- (Ocean Ridge Granites) normalized spidergram, both aplitic and tonalitic samples give similar overall patterns, but the aplitic samples are richer in LIL- (Large Ion Lithophile) elements and are more depleted in HFS- (High Field Strength) elements compared to tonalitic samples (not shown). There is a marked LIL-element enrichment and a Nb depletion relative to LREE (i.e. Ce) in all samples. Nb is depleted 5–10 times more relative to ORG.

The samples are also plotted on primitive mantle normalized spidergrams (Fig. 9). The distinction between the aplitic and metatonalitic samples is much clearer in this diagram. The aplitic set is enriched in Th, U, Pb, and K and depleted in Ti, Zr and P relative to the metatonalitic set.
Fig. 9

Primitive Mantle normalised spidergrams of İlyasdağ tonalite and cross-cutting aplitic veins. See text for explanation. Normalising values are from Sun and McDonough (1989)

Tectonic discrimination

In the Y versus Nb diagram (Pearce et al. 1984), all the samples plot in the VAG (volcanic arc granites) and syn-COLG (syn-collisional granites) fields (not shown). The Y + Nb versus Rb diagram is used to separate between these tectonic settings. All samples plot in the VAG field in this diagram (not shown). In R1 versus R2 diagram of Batchelor and Bowden (1985), the aplitic samples plot in the syn-collison to post-orogenic setting while the tonalitic samples plot in pre-plate collision setting, near the mantle fractionates dividing line (Fig. 10a). The Nb versus Rb/Zr diagram is used to compare the arc maturity of Brown et al. (1984). The Marmara Island’s plutonic rocks are grouped in primitive island arcs or continental arcs fields rather than normal or mature continental arcs (Fig. 10b).
Fig. 10

a R1 versus R2 discrimination diagram (Batchelor and Bowden 1985) for İlyasdağ Pluton; b Distribution of Eocene and Oligocene plutons on the Nb versus Rb/Zr arc maturity diagram (Brown et al. 1984). Note that the Ilyasdağ Pluton is characterised by low Nb values and plots in the primitive arc field. See Fig. 7 for the key of the symbols

Interpretation of geochemical data

Our study has shown that the İlyasdağ metatonalite is a metaluminous, calc-magnesian granitoid, emplaced at 47.6 ± 2 Ma in the Lutetian. It is a unique pluton among the Eocene plutons of NW Anatolia with its low SiO2 and K2O, high MgO, and the lowest Nb, Th, Rb/Sr and highest Y/Nb ratios. It is also characterised by having a high Pb content.

Harker variation diagrams indicate that compositional variation of the İlyasdağ tonalite was formed either from fractional crystallisation of an andesitic melt (57 wt% SiO2) or by differentiation and fractional crystallisation of a basic melt (more primitive than 57 wt% SiO2) at a magma chamber in the continental crust. Enrichment of Ba in the İlyasdağ pluton is controlled mainly by the crystallisation of clinopyroxene. Enrichment of Sr, up to SiO2 value of 64%, and subsequent depletion in the more evolved samples is thought to have been controlled by clinopyroxene and plagioclase, respectively. Rapid depletion of Sc can also be assigned to crystallisation of clinopyroxenes. Enrichment of Rb is much weaker in the İlyasdağ pluton than in the Kapıdağ and Karabiga plutons. Crystallisation and removal of clinopyroxene and plagioclase should have caused the slight enrichment of Rb in the remaining melt. The low K2O content of the İlyasdağ metatonalite rules out the enrichment of Rb with the assimilation of continental crustal material. Minor phases such as titanite, apatite and zircon control the depletion of Zr, Nb, TiO2 and P2O5. Amphibole, on the other hand, had an effect on the depletion of Y in the İlyasdağ pluton.

The cross-cutting aplitic set is different from the metatonalite in its geochemical properties. The aplitic set generally follows the same fractional crystallisation trend with the metatonalite in major- and trace-element Harker variation diagrams but they are more felsic (> 71 wt% SiO2) than the metatonalite and are peraluminous in composition.

Aplitic veins are characterised by low concentrations of Al2O3, CaO, MgO, Fe2O3 and TiO2 relative to the metatonalite. The enrichment in Th, K2O, Pb and U and depletion in Zr and Y relative to the metatonalite suggest that crustal contamination may have played an important role in the chemical evolution of the aplitic set in addition to differentiation and fractional crystallisation processes. Therefore we think that the aplites were originated from the same tonalitic melt but differentiated from it by assimilation of crustal material and by fractionation of amphiboles, which is responsible for low Y concentrations.

Previous petrological studies on the Eocene magmatism of NW Anatolia indicated metasomatised mantle sources (Köprübaşı and Aldanmaz 2004; Altunkaynak 2007; Karacık et al. 2008). Major and trace element characteristics of the İlyasdağ pluton are also compatible with a metasomatised mantle source. The metaluminous, rather than peraluminous nature, and high Mg # of the İlyasdağ pluton rule out derivation from a lower crustal amphibolitic/metabasaltic source area.

Both metatonalites and aplitic veins are characterised by LIL-element enrichment, strong Nb depletion, LREE enrichment, negative Ti and P anomalies and a positive Pb anomaly. These features are compatible with supra-subduction zone magmatism. In that regard, the İlyasdağ pluton can be considered to have emplaced at a continental margin arc setting.

Discussion

Mantle involved magmatism is common in all tectonic settings, including rifts, magmatic arcs and post-collisional settings. If the mantle concerned is metasomatised previously with the fluids from a subducting slab, the resulting initial melts will be of arc-type, regardless of the mantle melting mechanism (e.g. Pearce et al. 1984; Seyitoğlu et al. 1997; Reiners et al. 2000; Keskin 2003). This is the reason for a wide range of models proposed in the previous work for the tectonic setting of the Eocene magmatism in NW Anatolia and NE Greece. All the previous studies emphasised that the Eocene magmatic rocks in this region are of arc-type. Some tectonic models, however, inferred contemporaneous subduction (Delaloye and Bingöl 2000; Okay and Satır 2006) whereas others suggested post-collisional setting with the view that Neothethys in this area was sutured by early Eocene (Köprübaşı and Aldanmaz 2004; Altunkaynak 2007; Karacık et al. 2008). Eocene active subduction was also suggested for the north Greek plutons (Pe-Piper 2004).

Below we will discuss these alternative tectonic models, highlighting their positive and negative sides in the light of regional geological data.

a) Mid-Eocene post-collision magmatism

A post-collision setting was first proposed for the southerly plutons by Harris et al. (1994) and Okay et al. (1998). Harris et al. (1994) suggested that the Orhaneli and Topuk granodiorites are post-collisional granites emplaced during the 53–48 Ma interval (Ypresian, early Eocene). They showed that the granodiorites are either derived from a mantle wedge or by anatexis of lower continental crust; in the latter case advective heating of the crustal source by mantle-derived melts is needed to generate crustal melting. They envisaged that cessation of subduction may have heated the hydrated mantle and this resulted in melting. Okay et al. (1998), on the other hand, tied the mantle melting mechanism to asthenospheric upwelling, which occurred following a rapid isostatic rebound of subducted continental crust (the lower plate) by 60 km within a 35 myr period between Late Cretaceous and early Eocene, due to slab breakoff. In that model, the upwelling asthenosphere underwent decompression melting and the ascending melts triggered dehydration melting of lower crust by providing convective heat. Okay et al. (1998) argued that the location of the granodiorites in the isostatically upwelling zone (i.e. in Tavşanlı Zone) supports their model.

Köprübaşı and Aldanmaz (2004) extended the post-collision model to the northern plutons and inferred that either slab breakoff or delamination of a thermal boundary layer of subcontinental mantle produced the Eocene magmatism in this area. According to this model, magmatism occurred by means of conductive heating of lithospheric mantle by upwelling asthenosphere. Altunkaynak (2007) similarly suggested a slab breakoff model for generation of the Eocene magmatism in NW Anatolia. In this model, the southern plutons were assumed to be older (52 to 48 Ma) and the northern plutons younger (48 to 35 Ma). The southern plutons were emplaced during the initial slab breakoff (rift stage) and the northern plutons were emplaced when the slab was detached and the slab window was formed completely. Melting of mantle was triggered by asthenospheric upwelling through the slab window according to this model. The slab in these models is the northward descending Neotethyan oceanic lithosphere attached to the northern passive margin of the Menderes-Taurus platform.

The main discrepancy of the post-collisional models involving the slab breakoff mechanism lies in inconsistency of the timing of slab breakoff. The northern passive margin of the Menderes-Taurus Platform subducted to mantle depths (ca. 70 km) and underwent HP/LT metamorphism in Late Cretaceous times (80 Ma, Okay 2002). This blueschist belt, known as the Tavşanlı Zone, extends over 500 km in NW-SE direction with a width of ca. 80 km in NE-SW direction. The blueschist belt was exhumed in the latest Cretaceous since detritus derived from it is found in the latest Cretaceous sediments of the Sakarya Continent (Okay and Satır 2006). The exhumation of the subducted northern margin of the large Menderes-Taurus Platform in latest Cretaceous has been proposed to be due to the slab breaking off in late Campanian to early Maastrichtian (Okay 2002; Okay and Satır 2006).

Secondly, magmatism associated with slab breakoff (and delamination) mechanism is shown to emplace into thickened crust with elevated topography at the surface (see plate tectonic diagrams in Atherton and Ghani 2002; Altunkaynak 2007). The stratigraphic record in NW Anatolia, however, shows that the crust was at its normal thickness when the mid-Eocene plutons were emplaced (Okay and Satır 2006). For instance, an important mid-Eocene to Oligocene stratigraphic sequence is preserved to the north of the Marmara Island, in the Thrace Basin to the NW of Marmara Sea. The sequence is unconformable on older units and starts with neritic limestones and clastics of middle Eocene age (the Soğucak Limestone) and deepens upward, marked by deposition of overlying turbiditic sandstones (Ceylan Formation). The basin was then filled with volcanogenic conglomerates and sandstones. After volcanism ended, a thick sequence of grey shales was deposited in the basin, together with thin interlayers of turbiditic sandstones. The basin was then filled by the deposition of channelised conglomerates and sandstones of late Eocene to Oligocene age. Similar mid-Eocene shallow marine limestones are exposed in the Biga and Armutlu Peninsulas, to the south of Marmara Island. The ophiolites in the İzmir-Ankara-Erzincan Suture are also unconformably overlain by mid-Eocene shallow marine sediments. Thus, the stratigraphic record reveals that NW Anatolia was below sea level during the emplacement of the İlyasdağ pluton at ~48 Ma and remained as a subsiding area until the end of Eocene in this area. Apparently, the granitoid magmatism aided this subsidence by thermally weakening the crust. The subsiding area became an erosional area in Oligocene times, accompanied with folding and faulting.

Thirdly, the spatiotemporal distribution of the Eocene plutons (53–48 Ma in the south and 48–35 Ma in the north) is not valid since the Tepeldağ pluton in the southern plutons and the Karabiga pluton in the northern plutons have recently been dated as ~44 Ma and ~53 Ma, respectively by U/Pb method (Okay and Satır 2006; Beccaletto et al. 2007), and therefore the magmatism actually took place contemporaneously both in the northern and southern plutonic belts.

Fourthly, alkaline magmatism directly derived from the asthenospheric melts are absent in this area.

b) Mid-Eocene arc magmatism

Assuming that the Neotethyan slab was detached in Late Cretaceous times, a new subducting slab is needed for generation of arc magmatism in NW Anatolia. Several alternatives were proposed for such an oceanic lithosphere.

Northward subduction of the African Plate along the Hellenic trench was proposed for the origin of Eocene to Quaternary magmatism in Anatolia and the Aegean Sea (Delaloye and Bingöl 2000). The present distance between Marmara Island and the Hellenic trench exceeds 800 km. However, the area between the two is known to be the one of the fastest extending area in the world at present. The extension of the upper plate commenced in Early Miocene (Seyitoğlu and Scott 1996), causing retreat of the trench southward. Several lines of evidence suggest that the extension factor is ~1.2 (Eyidoğan 1988). The distance between the two amounts to 600 km when this extension is restored. Major discrepancies of this model are first, the great distance involved between arc magmatism and the trench, which is possible only if the subduction angle was unusually low, and second, the uncertainty in the age of northward subduction initiation along the Hellenic trench.

Eocene subduction along the Vardar suture

Okay and Satır (2006) envisaged that the dual nature of the mid-Eocene plutons in NW Anatolia is due to strike-slip displacement of a once contiguous, E-W trending plutonic belt in post-Eocene times. This plutonic belt is interpreted to be formed by arc magmatism developed in response to northeastward subduction of a remnant ocean basin, located in the north Aegean Sea region along the trace of the Vardar Suture.

Blueschist and eclogite facies metamorphism of 52 to ca. 40 Ma (Eocene) in the Cycladic Massif and equivalents in west Anatolia have been reported (Oberhänsli et al. 1998; Tomaschek et al. 2003; Lagos et al. 2007). These ages are indicative of active subduction in this area. The subduction trench is inferred to be NW-SE trending, oblique to the general E-W trend of the Eocene plutons in NW Anatolia. Palaeomagnetic data collected from the Late Miocene volcanics in NW Anatolia indicate dominantly anticlockwise rotations at ca. 20 degrees (Kaymakçı et al. 2007). When such rotation is restored, the plutonic belt rotates to NW-SE direction and becomes parallel to the inferred active margin. However, anticlockwise rotations were determined from the coastal areas only and it is not certain if the whole area in consideration underwent such rotation.

Discrepancies of this model are first, that no Eocene blueschist has yet been found along the İzmir-Ankara-Erzincan Suture. The Eocene plutons are also found in the Sivrihisar area some 200 km to the east and therefore a tie with the subduction of the Vardar Ocean requires anticlockwise rotation of whole Anatolian Peninsula, which is difficult to visualise. Secondly, there are some differences in the northern and southern plutons. The main difference between the two granitoid zones, apart from their location in the upper (northern group) and lower plates (southern group) of the Sakarya-Menderes convergence system, is the absence of late Eocene (?) plutons and associated volcanics in the southern zone. Altunkaynak (2007) also found out that both the northern and southern plutonic belts were derived from the metasomatised mantle however the northern plutons were modified by larger amount of crustal contamination that resulted in their stronger depletion in Eu, Ba, Sr and P and higher contents of Pb, K, Ni and SiO2. Due to the fact that the exact age of most of the northern plutons are not known, useful comparisons can only be made with the İlyasdağ (~48 Ma) and Topuk (~48 Ma) plutons as they were emplaced at nearly the same time in the upper and lower plates, respectively. The Topuk and İlyasdağ plutons display similar patterns on N-MORB normalised spidergrams. However, the Topuk pluton is enriched in most of the elements in comparison to İlyasdağ tonalite, except Ti and Y (not shown).

c) Mid-Eocene extensional magmatism associated with the collapse of the orogen

Late Palaeocene to early Eocene sedimentary rocks are largely absent in NW Anatolia. This interval possibly corresponds to uplift and erosion due to deformation of older rocks in response to collision of the Menderes-Taurus Platform and the Sakarya Continent. The presence of a mid-Eocene shallow marine sedimentary cover on the older units in Tavşanlı Zone and NW Anatolia suggests that the uplifted area rapidly subsided below sea-level in mid-Eocene times. If the subsidence of uplifted orogen in the middle Eocene is related to lithospheric extension, then the coeval magmatism can easily be interpreted as an extensional magmatism. In that case, asthenospheric melts of decompressional origin could have provided the necessary heat to melt the metasomatised subcontinental mantle.

In NW Anatolia there is no evidence for compressional deformation during the mid to late Eocene, as described above. However, in marked contrast, the mid-Eocene was the time when the Menderes Massif underwent its peak metamorphism under the thrust load of the Lycian and Beyşehir-Hoyran-Hadim Nappes. Therefore, it appears that while the Menderes Massif was telescoping by thin- to thick-skin thrusting, the Tavşanlı Zone and the area further north subsided rapidly.

Delamination of the thermal boundary layer under the Tavşanlı Zone and the areas further north is one of the possible explanations for this regional subsidence. However, orogenic collapse, in most cases, requires renewed subduction at a nearby ocean basin to accommodate the extension of the orogen. A remnant ocean basin located at the present Aegean Sea to the north of Cycladic Massif may have started subduction northward at this stage.

The available evidence does not prove any one or other of the models, although we favour a hybrid model, in which extension in NW Anatolia occurred after the final collision between the Menderes-Tauride Block and the Sakarya continent. We interpret that this extension was due to the rapid retreat of the southern Eurasian subduction system from the (present) central Aegean and the Izmir-Ankara Zone to the site of the active Hellenic trench. The reason for this retreat, we suggest, was due to the Palaeocene-Early Eocene collision of the Menderes-Tauride Block with the Sakarya continent (the southern margin of Eurasia at the time). This collision caused the >200 km translation of the Lycian Thrust Sheets over the Menderes-Tauride Block to the present Lycian front that lies along strike of the termination of the Hellenic Trench (Collins & Robertson 1997; 1998).

Conclusions

The İlyasdağ metatonalite is a N-dipping, sill-like intrusion, emplaced into a metamorphic rock association, made up of several thrust sheets of metaclastics, metabasics and metacarbonates. The intrusion exhibits strong brittle-ductile deformation fabrics, with the development of both a N-dipping foliation and an E-W plunging stretching lineation.

Despite the previous estimates of Palaeozoic or older intrusion ages, LA-ICP MS dating of U and Pb isotopes on zircon separates from the İlyasdağ metatonalite yielded an early Lutetian (47.6 ± 2 Ma) age for igneous crystallisation.

Eocene granitoids form two E-W trending belts in NW Anatolia; the one in the north is parallel to the Late Cretaceous Intra-Pontide Suture and the other in the south is parallel to the Early Cenozoic İzmir-Ankara-Erzincan Suture. Robust radiometric ages were available for the southern belt whereas the published ages of the northerly plutons have mostly been made using the K/Ar method on mineral separates that yielded diverse ages for individual plutons. Therefore, the age data presented in this paper are the one of the most robust age data gathered from the northern plutonic belt.

Geochemical data shows that the İlyasdağ metatonalite is a calcic-magnesian, metaluminous intrusion with arc-type geochemical characteristics. The İlyasdağ metatonalite was derived from a depleted mantle source with some contamination from continental crustal rocks, and is the most basic and the northernmost member in the northern plutonic belt of NW Anatolia.

The İlyasdağ pluton was emplaced into a crust of normal thickness, following the collision of the Sakarya and Menderes platforms along the İzmir-Ankara-Erzincan Suture Zone in Late Palaeocene-early Eocene times. The pluton underwent severe brittle-ductile deformation following its mid-Eocene emplacement.

All previous tectonic models have some sort of problems in explaining the genesis of the Eocene magmatism in NW Anatolia. However, a model of extension in the hanging-wall of a rapidly retreating arc setting is favoured here for the genesis of the İlyasdağ Pluton. The arc retreat being recorded in the rapid southerly stepping of subduction from the central Aegean (Cyclades) in the Eocene to the Hellenic Trench at the present day. This retreat is recorded in the punctuated history of thrusting recorded in the Lycian Thrust Sheets (Collins and Robertson 1998). Younger Eocene and Oligocene granitoids in NW Anatolia may relate to post-collisional processes. Future radiometric dating of the plutons and the blueschists are needed to better understand the tectonic setting of formation of magmatism.

Notes

Acknowledgement

This study was partly supported by a TÜBİTAK (The Scientific and Technical Research Council of Turkey) grant to P.A.U. (TÜBİTAK post-doctoral overseas research grant). The authors would like to thank John Stanley, David Bruce, Justin Payne and Ben Wade from the University of Adelaide for their help during the laboratory studies. Mineral Research and Exploration Institute in Ankara is greatly appreciated for preparation of thin sections. We are also grateful to Mehmet Keskin of Istanbul University for his suggestions on geochemistry section and to Tolga Görüm of Yıldız Technical University for his assistance in drafting some of the figures. Constructive and helpful reviews, provided by Gernold Zulauf, Aral Okay and A. Möller are greatly acknowledged.

References

  1. Abdüsselamoğlu MŞ (1977) The Palaeozoic and Mesozoic in the Gebze region explanatory text and excursion guidebook. 4th Colloquium on the Aegean Region, Exc. 4, İ.T.Ü. Maden Fakültesi, Istanbul, 16 ppGoogle Scholar
  2. Aksoy R (1995) Stratigraphy of the Marmara Island and the Kapıdağ Peninsula (Marmara Adası ve Kapıdağ Yarımadası’nın stratigrafisi). TAPG (Turkish Association of Petroleum Geologists) Bulletin 7/1:33–49 (in Turkish)Google Scholar
  3. Aksoy R (1996) Mesoscopic tectonic features of the Marmara Island and Kapıdağ Peninsula (Marmara Adası ve Armutlu Yarımadası’nın mezoskopik tektonik özellikleri). Turk J Eart Sci 5:187–195 in TurkishGoogle Scholar
  4. Aksoy R (1999) Relation between progressive regional metamorphism and tectonic history in the Marmara Island (Marmara Adası’nda ilerleyen bölgesel metamorfizma ve tektonik geçmiş arasındaki ilişki). Geol Bull Turk 42(11):1–14 in TurkishGoogle Scholar
  5. Altunkaynak Ş (2007) Collision-driven slab breakoff magmatism in northwestern Anatolia, Turkey. J Geol 115:63–82CrossRefGoogle Scholar
  6. Atherton MP, Ghani AA (2002) Slab breakoff: a model for Caledonian, Late Granite syn-collisional magmatism in orthotectonic (metamorphic) zone of Scotland and Donegal (Ireland). Lithos 62:65–85CrossRefGoogle Scholar
  7. Batchelor RA, Bowden P (1985) Petrogenetic interpretation of granitoid rock series using multicationic parameters. Chem Geol 48:43–55CrossRefGoogle Scholar
  8. Beccaletto L, Jenny C (2004) Geology and correlation of the Ezine Zone: a Rhodope fragment in NW Turkey? Turk J Eart Sci 13:145–176Google Scholar
  9. Beccaletto L, Bonev N, Bosch D, Bruguier O (2007) Record of a Palaeogene syn-collisional extension in the north Aegean region: evidence from the Kemer micaschists (NW Turkey). Geol Mag 144(2):393–400CrossRefGoogle Scholar
  10. Brown SAM, Robertson AHF (2000) Sedimentary geology as a key to the understanding the tectonic evolution of the Mesozoic-Early Tertiary Paikon Massif, Vardar Suture Zone, N Greece. Sed Geol 160:179–212CrossRefGoogle Scholar
  11. Brown GC, Thorpe RS, Webb PC (1984) The geochemical characteristics of granitoids in contrasting arcs and comments on magma sources. J Geol Soc London 141:413–426CrossRefGoogle Scholar
  12. Collins AS, Robertson AHF (1997) Lycian melange, southwestern Turkey: an emplaced Late Cretaceous accretionary complex. Geology 25:255–258CrossRefGoogle Scholar
  13. Collins AS, Robertson AHF (1998) Processes of Late Cretaceous to Late Miocene episodic thrust-sheet translation in the Lycian Taurides, SW Turkey. J Geol Soc London 155:759–772CrossRefGoogle Scholar
  14. De La Roche H, Leterrier J, Grandclaude P, Marchal M (1980) A classification of volcanic and plutonic rocks using R1R2-diagram and major element analyses—its relationships with current nomenclature. Chem Geol 29:183–210CrossRefGoogle Scholar
  15. Debon F, Le Fort P (1983) A chemical-mineralogical classification of common plutonic rocks and associations. T Roy Soc Edin-Earth Sci 73:135–149Google Scholar
  16. Delaloye M, Bingöl E (2000) Granitoids from western and northwestern Anatolia: geochemistry and modelling of geodynamic evolution. Geol Bull Turk 42:241–268Google Scholar
  17. Eyidoğan H (1988) Rates of crustal deformation in Western Turkey as deduced from major earthquakes. Tectonophysics 148:83–92CrossRefGoogle Scholar
  18. Frost BR, Barnes CG, Collins WJ, Arculus RJ, Ellis DJ, Frost CD (2001) A geochemical classification for granitic rocks. J Petrol 42:2033–2048CrossRefGoogle Scholar
  19. Genç ŞC (1998) Evolution of the Bayramiç magmatic complex, Northwestern Anatolia. J Volcanol Geoth Res 85:233–249CrossRefGoogle Scholar
  20. Göncüoğlu MC, Erendil M (1990) Pre-late cretaceous tectonic units of the Armutlu peninsula (Armutlu Yarımadasının Geç Kretase öncesi tektonik birimleri). Proceedings of 8th Petroleum Congress of Turkey, pp 161–168 (in Turkish)Google Scholar
  21. Göncüoğlu MC, Özcan A, Turhan N, Işık A (1992) Stratigraphy of the Kütahya region: In: A Geotraverse across Tethyan suture zones in NW Anatolia. Field guide book for the Symposium on the Geology of the Black Sea region, Ankara, pp 3–11Google Scholar
  22. Harris NBW, Kelley S, Okay Aİ (1994) Post-collisional magmatism and tectonics in northwestern Anatolia. Contrib Mineral Petrol 117:241–252CrossRefGoogle Scholar
  23. Jackson SE, Pearson NJ, Griffin WL, Belousova EA (2004) The application of laser ablation inductively coupled plasma-mass spectrometry to in-situ U/Pb zircon geochronology. Chem Geol 211:47–69CrossRefGoogle Scholar
  24. Janousek V, Farrow CM, Erban V (2006) Interpretation of whole-rock geochemical data in igneous geochemistry: introducing Geochemical Data Toolkit (GCDkit). J Petrol 47:1255–1259CrossRefGoogle Scholar
  25. Karacık Z, Yılmaz Y, Pearce JA, Ece ÖI (2008) Petrochemistry of the South Marmara granitoids, northwest Anatolia, Turkey. Int J Earth Sci (Geol Rundsch) 97:1181–1200CrossRefGoogle Scholar
  26. Kaymakçı N, Aldanmaz E, Langereis C, Spell TL, Gürer OF, Zanetti KA (2007) Late Miocene transcurrent tectonics in NW Turkey: evidence from paleomagnetism and 39Ar-40Ar dating of alkaline volcanic rocks. Geol Mag 144:379–392CrossRefGoogle Scholar
  27. Keskin M (2003) Magma generation by slab steepening and breakoff beneath a subduction-accretion complex: an alternative model for collision-related volcanism in Eastern Anatolia, Turkey. Geophys Res Lett 30:1–9CrossRefGoogle Scholar
  28. Ketin İ (1946) Geological investigations in the Marmara Island and the Kapıdağ Peninsula (Marmara Adası ve Kapıdağ Yarımadası’nda jeolojik incelemeler). İU Sci Fac Bull Serie B XI(2):69–81 in TurkishGoogle Scholar
  29. Köprübaşı N, Aldanmaz E (2004) Geochemical constraints on the petrogenesis of Cenozoic I-type granitoids in northwestern Anatolia, Turkey: evidence for magma generation by lithospheric delamination in a post-collisional setting. Int Geol Rev 46:405–429Google Scholar
  30. Lagos M, Scherer EE, Tomaschek F, Münker C, Keiter M, Berndt J, Ballhaus C (2007) High precision Lu/Hf geochronology of Eocene eclogite-facies rocks from Syros, Cyclades, Greece. Chem Geol 243:16–35CrossRefGoogle Scholar
  31. Marchev P, Kaiser-Rohrmeier M, Heinrich C (2005) Hydrothermal ore deposits related to Bulgaria and Greece. Int Geol Rev 27:53–89Google Scholar
  32. Middlemost EAK (1985) Magmas and magmatic rocks: An introduction to igneous petrology: London. Longman, UKGoogle Scholar
  33. MTA (2004) Geological map of Turkey scale 1/500.000. MTA-Mineral Research and Exploration Institute, AnkaraGoogle Scholar
  34. Oberhänsli R, Monie P, Candan O, Warkus FC, Partzsch JH, Dora O (1998) The age of blueschist metamorphism in the Mesozoic cover series of the Menderes Massif. Schweiz Mineral Petrogr Mitt 78(2):309–316Google Scholar
  35. Okay Aİ (1984) Distribution and characteristics of the northwest Turkish blueschists. In: Dixon JE, Robertson AHF (eds) The geological evolution of the Eastern Mediterranean. Geological Society of London, Special Publication 17: 455–466Google Scholar
  36. Okay Aİ (1989) Tectonic units and sutures in the Pontides, northern Turkey. In: Şengör AMC (ed) Tectonic evolution of the Tethyan Region. Dordrecht, Kluwer, pp 109–115Google Scholar
  37. Okay Aİ (2000) Was the Late Triassic orogeny in Turkey caused by the collision of an oceanic plateau? In: Bozkurt B, Winchester JA, Piper JDA (eds) Tectonics and magmatism in Turkey and the surrounding area. Geological Society of London, Special Publication 173: 25–42Google Scholar
  38. Okay Aİ (2002) Accretional and collisional orogens in Anatolia. Geol Soc Am Abst 34:329–330Google Scholar
  39. Okay Aİ, Satır M (2000) Coeval plutonism and metamorphism in a latest Oligocene metamorphic core complex in northwest Turkey. Geol Mag 137:495–516CrossRefGoogle Scholar
  40. Okay Aİ, Satır M (2006) Geochronology of Eocene plutonism and metamorphism in northwest Turkey: evidence for a possible magmatic arc. Geodin Acta 19:251–266CrossRefGoogle Scholar
  41. Okay Aİ, Satır M, Maluski H, Siyako M, Metzger R, Akyüz HS (1996) Palaeo and Neo-Tethyan events in northwest Turkey: Geological and geochronological constraints. In: Yin A, Harrison M (eds) Tectonics of Asia. Cambridge Univ Press, Cambridge, UK, pp 420–441Google Scholar
  42. Okay Aİ, Harris NBW, Kelley SP (1998) Exhumation of blueschists along a Tethyan suture in northwest Turkey. Tectonophysics 285:275–299CrossRefGoogle Scholar
  43. Okay Aİ, Demirbağ E, Kurt H, Okay N, Kuşçu İ (1999) An active, deep marine strike-slip basin along the North Anatolian Fault in Turkey. Tectonics 18(1):129–147CrossRefGoogle Scholar
  44. Okay Aİ, Satır M, Tüysüz O, Akyüz HS, Chen F (2001) The tectonics of the Strandja Massif: Variscan and mid-Mesozoic deformation and metamorphism in the northern Aegean. Int J Earth Sci (Geologische Rundschau) 90:217–233CrossRefGoogle Scholar
  45. Okay Aİ, Bozkurt E, Satır M, Yiğitbaş E, Crowley QG, Shang CK (2008a) Defining the southern margin of Avalonia in the Pontides: geochronological data from the Late Proterozoic and Ordovician granitoids from NW Turkey. Tectonophysics 461:252–264CrossRefGoogle Scholar
  46. Okay Aİ, Satır M, Zattin M, Cavazza W, Topuz G (2008b) An Oligocene ductile strike-slip shear zone: the Uludağ Massif, northwest Turkey-implications for the westward translation of Anatolia. Geol Soc Am Bull 120:893–911CrossRefGoogle Scholar
  47. Payne JL, Barovich KM, Hand M (2006) Provenance of metasedimentary rocks in the northern Gawler Craton, Australia: implications for Palaeoproterozoic reconstructions. Precambrian Res 148:275–291CrossRefGoogle Scholar
  48. Pearce JA, Harris NBW, Tindle AG (1984) Trace element discrimination diagrams for for the tectonic interpretation of granitic rocks. J Petrol 25:956–983Google Scholar
  49. Peccerillo A, Taylor SR (1976) Geochemistry of Eocene calc-alkaline volcanic rocks in the Kastamonu area, Northern Turkey. Contrib Mineral Petrol 58:63–81CrossRefGoogle Scholar
  50. Pe-Piper G (2004) Unique features of Greek Cenozoic igneous rocks: tectonics and magma sources. 5th International Symposium on Eastern Mediterranean Geology, Thessaloniki, Greece, 14–20 April 2004, 1208–1211Google Scholar
  51. Pickett E, Robertson AHF (1996) Formation of the Late Paleozoic-Early Mesozoic Karakaya Complex and related ophiolites in NW Turkey by Palaeotethyan subduction-accretion. J Geol Soc London 153:995–1009CrossRefGoogle Scholar
  52. Pickett E, Robertson AHF (2004) Significance of the volcanogenic Nilüfer Unit and related components of the Triassic Karakaya Complex for the Tethyan subduction/accretion processes in NW Turkey. Turk J Eart Sci 13:97–144Google Scholar
  53. Reiners PW, Hammond PE, McKenna JM, Duncan RA (2000) Young basalts of the central Washington Cascades, flux melting of the mantle and trace element signatures of primary arc magmas. Contrib Mineral Petrol 138:249–264CrossRefGoogle Scholar
  54. Ring U, Collins AS (2005) U-Pb SIMS dating of syn-kinematic granites: timing of core-complex formation in the northern Anatolide belt of western Turkey. J Geol Soc London 162:289–298CrossRefGoogle Scholar
  55. Robertson AHF, Dixon JE (1984) Introduction: aspects of the geological evolution of the Eastern Mediterranean. In: Dixon JE, Robertson AHF (eds) The geological evolution of the Eastern Mediterranean. Geological Society of London, Special Publication 17: 1–74Google Scholar
  56. Robertson AHF, Ustaömer T (2004) Tectonic evolution of the Intra-Pontide suture zone in the Armutlu Peninsula, NW Turkey. Tectonophysics 381:175–209CrossRefGoogle Scholar
  57. Robertson AHF, Ustaömer T, Parlak O, Ünlügenç Ü, Taşlı K, İnan N (2006) The Berit transect of the Tauride thrust belt, S Turkey: Late Cretaceous-Early Cenozoic accretionary/collisional processes related to closure of the Southern Neotethys. J Asian Earth Sci 27:108–145CrossRefGoogle Scholar
  58. Saner S (1985) Sedimentary sequences and tectonic setting of Saros Gulf region, NE Aegean Sea, Turkey. Bull Geol Soc Turk 28:1–10Google Scholar
  59. Seeber L, Emre Ö, Cormier MH, Sorlien CC, McHugh CMG, Polonia A, Özer N, Çağatay N (2004) Uplift and subsidence from oblique slip: the Ganos-Marmara bend of the North Anatolian Transform, Western Turkey. Tectonophysics 391:239–258CrossRefGoogle Scholar
  60. Şengör AMC, Yılmaz Y, Sungurlu O (1984) Tectonics of the Mediterranean Cimmerides: nature and evolution of the western termination of Palaeo-Tethys. In: Dixon JE, Robertson AHF (eds) Geological evolution of the Eastern Mediterranean. Geological Society of London, Special Publication 17: 77–112Google Scholar
  61. Şengör AMC, Yılmaz Y (1981) Tethyan evolution of Turkey: a plate tectonic approach. Tectonophysics 75:181–241CrossRefGoogle Scholar
  62. Seyitoğlu G, Scott B (1996) The cause of N-S extensional tectonics in western Turkey: tectonic escape vs back-arc spreading vs orogenic collapse. Journal of Geodynamics 22:145–153CrossRefGoogle Scholar
  63. Seyitoğlu G, Anderson D, Nowell G, Scott BC (1997) The evolution from Miocene potassic to Quaternary sodic magmatism in Western Turkey: implications for enrichment processes in the lithospheric mantle. J Volcanol Geoth Res 76:127–147CrossRefGoogle Scholar
  64. Shand SJ (1927) Eruptive rocks: Their genesis, composition, classification, and their relation to ore-deposits, with a chapter on meteorites. Wiley, New YorkGoogle Scholar
  65. Siyako M, Huvaz O (2007) Eocene stratigraphic evolution of the Thrace Basin, Turkey. Sed Geol 198:75–91CrossRefGoogle Scholar
  66. Sun S, McDonough WS (1989) Chemical and isotopic systematics of oceanic basalts: implications for mantle processes and compositions. In: Saunders AD, Norry MJ (eds), Magmatism in the oceanic basins. Geological Society of London, Special Publication 42: 313–345Google Scholar
  67. Tanyolu E (1979) Petrological studies of the Marmara Island metamorphic series (Marmara Adası metamorfik serilerinin petrolojik çalışması). Zonguldak Engineering and Architecture Academy, Habiliation Thesis 108 pp (in Turkish)Google Scholar
  68. Tomaschek F, Kennedy AK, Villa IM, Lagos M, Balhaus C (2003) Zircons from Syros-Cyclades, Greece-recrystallization and mobilization of zircon during high-pressure metamorphism. J Petrol 44:1977–2002CrossRefGoogle Scholar
  69. Ustaömer T, Robertson AHF (1993) A Late Palaeozoic-Early Mesozoic marginal basin along the active southern continental margin of Eurasia: evidence from the central Pontides (Turkey) and adjacent regions. Geol J 28:219–238CrossRefGoogle Scholar
  70. Ustaömer T, Robertson AHF (1997) Tectonic-sedimentary evolution of the north Tethyan margin in the Central Pontides of northern Turkey. AAPG Memoir 68:255–290Google Scholar
  71. Ustaömer T, Robertson AHF (2005) Discussion: Reply to discussion contribution by A. Elmas and Yiğitbaş, 2005’ on Tectonic evolution of the Intra-Pontide suture zone in the Armutlu Peninsula, NW Turkey by A.H.F. Robertson and T. Ustaömer, Tectonophysics: 381 (2004), 175–209. Tectonophysics 405: 223–231Google Scholar
  72. Ustaömer PA, Ustaömer T, Collins AS, Robertson AHF (2009) Cadomian (Ediacaran-Cambrian) arc magmatism in the Bitlis Massif, SE Turkey: Magmatism along the developing northern margin of Gondwana. Tectonophysics doi:10.1016/j.tecto.2008.06.010
  73. Yiğitbaş E, Kerrich R, Yılmaz Y, Elmas A, Xie Q (2004) Characteristics and geochemistry of Precambrian ophiolites and related volcanics from the Istanbul-Zonguldak Unit, northwestern Anatolia: following the missing chain of the Precambrian South European suture zone to the east. Precambrian Res 132:179–2006CrossRefGoogle Scholar
  74. Yılmaz Y, Genç ŞC, Yiğitbaş E, Bozcu M, Yılmaz K (1995) Geological evolution of the late Mesozoic continental margin of Northwestern Anatolia. Tectonophysics 241:155–171CrossRefGoogle Scholar
  75. Yılmaz,Y, Tüysüz O, Yiğitbaş E, Genç, ŞC, Şengör AMC (1997) Geology and tectonic evolution of the Pontides. In: Robinson AG (ed) Regional and Petroleum Geology of the Black Sea and Surrounding Region. American Association of Petroleum Geologists Memoir 68: 183–226Google Scholar
  76. Yüzer E (1971) Engineering geology and anisotropic futures investigation of the Marmara Island marbles (Marmara Adası mermerlerinin mühendislik jeolojisi ve anizotropik özelliklerinin incelenmesi). Istanbul Technical University, Habilitation thesis 128 pp (in Turkish)Google Scholar

Copyright information

© Springer-Verlag 2009

Authors and Affiliations

  • P. Ayda Ustaömer
    • 1
  • Timur Ustaömer
    • 2
  • Alan S. Collins
    • 3
  • Jörg Reischpeitsch
    • 4
  1. 1.Doğa Bilimleri Araştırma MerkeziYıldız Teknik ÜniversitesiBeşiktaş-IstanbulTurkey
  2. 2.Mühendislik Fakültesi, Jeoloji Mühendisliği Bölümüİstanbul ÜniversitesiAvcılar-IstanbulTurkey
  3. 3.Tectonics and Resources Exploration (TREX), School of Earth and Environmental SciencesUniversity of AdelaideAdelaideAustralia
  4. 4.Department of GeologyUniversity of HamburgHamburgGermany

Personalised recommendations