Journal of Volcanology and Geothermal Research 185 (2009) 181–202

Contents lists available at ScienceDirect

Journal of Volcanology and Geothermal Research

j ourna l homepage: www.e lsev ie r.com/ locate / jvo lgeores

Geochemistry and 40Ar/39Ar geochronology of Miocene volcanic rocks from theKaraburun Peninsula: Implications for amphibole-bearing lithospheric mantlesource, Western Anatolia

Cahit Helvacı a, E. Yalçın Ersoy a,⁎, Hasan Sözbilir a, Fuat Erkül b, Ökmen Sümer a, Bora Uzel a

a Dokuz Eylül Üniversitesi, Mühendislik Fakültesi, Jeoloji Mühendisliği Bölümü, TR-35160 İzmir, Türkiyeb Akdeniz Üniversitesi, Teknik Bilimler Meslek Yüksek Okulu TR-07058, Antalya, Türkiye

⁎ Corresponding author. Fax: +90 232 453 87 87.E-mail address: yalcin.ersoy@deu.edu.tr (E.Y. Ersoy)

0377-0273/$ – see front matter © 2009 Elsevier B.V. Adoi:10.1016/j.jvolgeores.2009.05.016

a b s t r a c t

a r t i c l e i n f o

Article history:Received 5 November 2008Accepted 20 May 2009Available online 9 June 2009

Keywords:Karaburun PeninsulaWestern AnatoliaNeogene volcanismlithospheric mantlegeochemistry40Ar/39Ar geochronology

New geochemical and 40Ar/39Ar age data are presented from the Neogene volcanic units of the KaraburunPeninsula, the westernmost part of Western Anatolia. The volcanic rocks in the region are associated withNeogene lacustrine deposition and are characterized by (1) olivine-bearing basaltic-andesites to shoshonites(Karaburun volcanics), high-K calc-alkaline andesites, dacites and latites (Yaylaköy, Armağandağ and Kocadağvolcanics) of ~16–18 Ma, and (2) mildly-alkaline basalts (Ovacık basalt) and rhyolites (Urla volcanics) of ~11–12 Ma. The first group of rocks is enriched in LILE and LREE with respect to the HREE and HFSE on N-MORB-normalised REE andmulti-element spider diagrams. Theyare comparable geochemicallywith volcanic rocks inthe surrounding regions such as Chios Island and other localities in Western Anatolia. The Ovacık basalt isgeochemically similar to thefirst stage early–middleMiocene volcanic rocks but differs fromNWAnatolian lateMiocene alkali basalts.Trace element models indicate that the Kocadağ and Armağandağ volcanics were produced from the Yaylaköyvolcanics by assimilation and fractional crystallization (AFC). Melting models of several trace elements of themost primitive mafic volcanics indicate that their source was metasomatized by subduction-related fluids andwas amphibole-bearing garnet–lherzolitic.

© 2009 Elsevier B.V. All rights reserved.

1. Introduction

The late Cenozoic geodynamic evolution of the Aegean extensionalprovince (Fig. 1) is mainly controlled by: (1) the Aegean subductionzone along which the African plate is sinking underneath the Aegean–Anatolian plate (e.g., Jackson and McKenzie, 1984); (2) collisionalevents between continental blocks; and (3) a post-collisional exten-sional tectonic regime that begun during the latest Oligocene (Seyitoğluand Scott, 1991). The latter has been accommodated by: (a) orogeniccollapse of the Paleogene over-thickened Western Anatolian crustduring which the mid-crustal units of the several metamorphic massifswere exhumed along the low-angle detachment faults throughout thelate Cenozoic (Gessner et al., 2001; Lips et al., 2001; Sözbilir, 2001, Işıket al., 2003; Sözbilir, 2005); (b) thewestward extrusion of the Anatolianblock along theNorthAnatolian Fault (NAF) and the East Anatolian Fault(EAF) (Şengör et al., 1985; Koçyiğit et al., 1999); and (c) southwestwardmigration of the Aegean subduction zone (e.g., Jackson and McKenzie,1984). These geodynamic regimes are also responsible for the volcanicactivity in the region. In the area represented in Fig. 2, wide-spread lateCenozoic volcanic rocks crop out throughout the Western Anatolian

.

ll rights reserved.

coastal region andAegean islands forming the easternpart of theAegeanextensional province.

The Karaburun Peninsula, forming the main scope of this study islocated at centre of the region (Fig. 2) and contains well-preservedNeogene volcanic units of whose geochemical characteristics have notyet been well-documented. The Neogene volcanics in the KaraburunPeninsula are also located between two important areas which havebeen interpreted to have distinct geodynamic features: (1) a volcanicprovince in the south which is the easternmost part of the activeAegean arc located on the front of the Hellenic trench (namely theBodrum Peninsula and surroundings in (Fig. 2); and (2) a large vol-canic province in the north (a region located to the south of the Sea ofMarmara, Fig. 2). The volcanic activity in the latter has producedwide-spread calc-alkaline basalts, andesites, dacites and rhyolites duringEocene to middle Miocene (first stage) and alkaline mafic volcanicsafter the middle Miocene (second stage). Geochemical features of thefirst-stage volcanic products clearly resemble those of other subduc-tion-related volcanic rocks in the world (e.g., Pearce and Parkinson,1993). The second-stage basalts, on the other hand, are represented byOIB-like geochemistry indicating that they originated from a differentmantle source from the first stage volcanic activity (Güleç, 1991;Aldanmaz et al., 2000). These two stages of volcanic activity have beendifferently interpreted by several workers. A general agreement existsamong most workers suggesting that the volcanic units in the region

Fig. 1. Regional map of the Aegean region (from Okay and Tüysüz, 1999; Ring et al., 1999). VIAS: Vardar–İzmir–Ankara suture zone. Distribution of volcanic rocks is from Fytikas et al.(1984).

182 C. Helvacı et al. / Journal of Volcanology and Geothermal Research 185 (2009) 181–202

are the products of post-collisional settings (Yılmaz, 1989, 1990;Güleç, 1991; Aldanmaz et al., 2000; Altunkaynak and Dilek, 2006;Altunkaynak and Genç, 2008; Pe-Piper et al., 2009). On the otherhand, someworkers have proposed that the whole volcanic activity ofthe region was related to the southwestward migrating Aegeansubduction system (Fytikas et al., 1984; Okay and Satır, 2000; Erkül etal., 2005; Innocenti et al., 2005).

The mantle source(s) of the magmas forming the Neogene vol-canic units (except for those of the late Miocene) is accepted to hadbeen metasomatized, either by active or previous subduction in theregion, although its characteristics andmineralogy have not beenwell-constrained. In this study, we present new geochemical and radio-metric agedata from theKaraburun Peninsula, and compare themwiththe other volcanic occurrences in the Aegean region to examine thepetrogenetic evolution of the volcanic rocks in the region throughoutthe late Cenozoic. We also examine their geochemical features toreveal the source characteristics of the volcanic rocks.

2. Distribution of volcanic rocks in the Aegean region

In the eastern part of the Aegean extensional province, wide-spread volcanic activity has taken place from the Eocene to Recent.The volcanic rocks of the Eocene to late Miocene are mainly accepted

Fig. 2. Distributions of the late Cenozoic magmatic rocks in the eastern Aegean (Modified froet al. (1985,1986,1995,1998); Helvacı (1995); Delaloye and Bingöl (2000); Aldanmaz et al. (2et al. (2005); Karacık et al. (2007b); Pe-Piper and Piper (2007b); Altunkaynak and Genç (2

to have been produced in a post-collisional setting. The volcanic rocksin the Karaburun Peninsula lie in the southern part of this largevolcanic province. On the other hand, the younger volcanic units arerelated to active subduction along the Hellenic trench in the southAegean (Aegean subduction system).

In the Biga Peninsula and surroundings, along the southern coastof the Sea of Marmara, the first magmatic activity in the area occurredin the Eocene (Fig. 2). During the Eocene, calc-alkaline volcanic rocks(42–29 Ma, Ercan et al., 1995, 1998; Altunkaynak and Genç, 2008;Altunkaynak and Dilek, 2006) and I-type granitodes (49–45 Ma,Delaloye and Bingöl, 2000; Karacık et al., 2007b) were formed. Themagmatic activity in this region also continued in the Oligocene andearly Miocene (31.3–18 Ma) with similar products (Ercan et al., 1995,1998; Altunkaynak and Genç, 2008), which can also traced in theGökçeada (Ercan et al., 1995) and Samothraki Islands (Seymour et al.,1996) and Thrace Basin (Ercan et al., 1998). Early Miocene volcanicactivity also occurred in Limnos Island with high-K and calc-alkalineand lamproitic products formed at ~22–18 Ma (Pe-Piper and Piper,2007a; Pe-Piper et al., 2009). On the otherhand, there are severalmaficvolcanic centerswhichwere emplaced in the Biga Peninsula during thelate Miocene (11.16–7.65 Ma; Ercan et al., 1995; Kaymakçı et al., 2007).These mafic volcanics are characterized by intra-continental alkalibasaltic lavas with OIB-like isotopic and trace element compositions

mMTA (2002) on the basis of age data of Borsi et al. (1972); Robert et al. (1992); Ercan000); Alther and Siebel (2002); Emre and Sözbilir (2005); Innocenti et al. (2005); Erkül008); Pe-Piper et al. (2009).

183C. Helvacı et al. / Journal of Volcanology and Geothermal Research 185 (2009) 181–202

184 C. Helvacı et al. / Journal of Volcanology and Geothermal Research 185 (2009) 181–202

Fig. 4. Stratigraphic columnar section of the Neogene volcano-sedimentary units in theKaraburun Peninsula.

185C. Helvacı et al. / Journal of Volcanology and Geothermal Research 185 (2009) 181–202

(Aldanmaz et al., 2000, 2005, 2006) and are clearly different from theolder volcanic units with respect to their younger ages and geochem-ical features (Kaymakçı et al., 2007).

Further south, at Ayvalık and surroundings, magmatic activity isrepresented first by the emplacement of the calc-alkaline I-type Kozakpluton, whichyielded 21–17Ma radiometric ages (Boztuğ et al., 2009).In this region, wide-spread early–middle Miocene magmatic activityalso occurs at the southernmost part of the Biga peninsula in the north,at the Lesbos Island in the west and at the Soma–Akhisar region in theeast. These volcanic products are represented by calc-alkaline, high-Kandesites, dacites and rhyolites which were emplaced during ~20–15 Ma (Borsi et al., 1972; Ercan et al., 1995; Altunkaynak and Yılmaz,1998; Aldanmaz et al., 2000; Yılmaz et al., 2001; Innocenti et al., 2005;Pe-Piper and Piper, 2001, 2007a; Altunkaynak andDilek, 2006; Karacıket al., 2007a; Altunkaynak and Genç, 2008). Minor lamproitic lavasat Nebiler were also dated as middle Miocene (15.2 Ma, Aldanmazet al., 2000 and Innocenti et al., 2005). Further east, small high-Kbasalts occur at Akhisar, which has been dated as 16.7 Ma (Innocentiet al., 2005). Similar volcanic activity has also occurred in Lesbos Islandduring ~18–16 Ma producing high-K andesites and lamproites (Fig. 2,Pe-Piper and Piper, 2001, 2007a).

In Chios Island, located ~10 km west of the Karaburun Peninsula,small-voluminous alkaline olivine basalts (nepheline-normativebasalts, 16.2 Ma) and calc-alkaline rhyolites, dacites and andesites(high-Mg adakite-like andesites) (14.3–15.9 Ma) outcrops (Pe-Piperet al., 1995). To the northeast of the Karaburun Peninsula, in the Aliağa–Foça region, Neogene volcanism is represented by early to middleMiocene high-K calc-alkaline andesites and dacites of Yuntdağıvolcanics (Akay and Erdoğan, 2004) and middle Miocene bimodalvolcanism at Foça including mildly alkaline basalts (15.2–14.3 Ma,Aldanmaz et al., 2000; Agostini et al., 2005; Innocenti et al., 2005;) andrhyolites (Akay and Erdoğan, 2004).

Southwestern Anatolia contains younger volcanic rocks located inthe Söke–Kuşadası area (mildly alkaline basalts and high-K andesitesof ~7Ma, Ercan et al., 1986), Bodrum Peninsula (12–8Ma, Robert et al.,1992; Kurt and Arslan, 2001; Karacık, 2006) and Samos (trachytes,rhyolites and basalts of 10.2–7.8 Ma Pe-piper and Piper, 2007b; Robertet al., 1992), Kos, Yali and Nisyros Islands (Pliocene to recent volcanics,Buettner et al., 2005; Pe-Piper and Piper, 2007a; Fig. 2).

3. Neogene stratigraphy of Karaburun Peninsula

The Neogene stratigraphy of the Karaburun Peninsula is repre-sented by a volcano-sedimentary succession including several sedi-mentary and volcanic units (Figs. 3 and 4). These units rest on abasement comprising non-metamorphic and intensely sheared Paleo-zoic to Mesozoic rocks of the Karaburun Belt (e.g., Erdoğan, 1990;Robertson and Pickett, 2000; Çakmakoğlu and Bilgin, 2006; Tatar-Erkül et al., 2008). The Neogene volcano-sedimentary units beginwiththe Bozköy Formation that consists of conglomerates of alluvial fansand fluvial origin. The Bozköy Formation crops out in a limited areathat is generally restricted to the northern parts of the study area. Theunit laterally and vertically passes into the Urla limestone which iscomposed of mainly white-colored fresh-water limestones. The Urlalimestone crops out in a wide area throughout of the peninsula. In theKaraburun Peninsula, several volcanic units interfinger with the Urlalimestone. These are the Karaburun, Armağandağ, Kocadağ andYaylaköy volcanics, which were emplaced contemporaneously withdeposition of the Urla limestone during latest early Miocene. Theseunits are unconformably overlain by the Urla volcanics, and finallythe Ovacık basalt.

The Karaburun volcanics are composed of olivine-bearing basalticlava flows and related pyroclastic rocks. The unit crops out around

Fig. 3. Generalized geological map of the Karaburun Peninsula showing the distribution ofanalyses (modified from Geological Map of Turkey 2002, 1:500 000).

Karaburun, in an area of ~30 km2 in a NW–SE direction (Fig. 3). TheKaraburun volcanics begin with reddish brown oxidized pyroclasticdeposits which are overlain by vesicular and generally altered lavaflows. The lava flows are characterized by olivine phenocrysts found ina fine-grained plagioclase- and clinopyroxene-rich matrix. The lavasinterfinger with the Urla limestone. The Karaburun volcanics havebeen dated by Türkecan et al. (1998) as 18.5 and 16 Ma (K–Ar ages).The Armağandağ volcanics crop out in a large area (~200 km2) be-tween Alaçatı and Nohutalan villages, and are composed of wide-spread pyroclastic rocks of fall and flow origin and massive lava flowswith domes and dykes (Fig. 3). The pyroclastic rocks begin withwhite-colored pumice-and-ash-fall deposits that have regular bed-ding planes. They are overlain by lithic-rich ignimbrites which locallypresents reverse and normal grading of lithic clasts and gas escapepipes. The ignimbrites are overlain by block and ash-flow-deposits.These pyroclastic rocks interfinger with the limestone and marl inter-calations of the Urla limestone. The last products of the Armağandağvolcanics are pink- to grey-colored andesitic lava flows thatpetrographically include euhedral to subhedral plagioclase, brownamphibole and clinopyroxene phenocrysts (~80%, 15% and 5% of totalamount of phenocrysts, respectively) found in a glassy matrix. Thelavas locally exhibit peperitic contacts with the interfingering Urlalimestones. Borsi et al. (1972) obtained 18.2 Ma K–Ar age from theArmağandağ volcanics. The Kocadağ volcanics crop out in a large area(~200 km2) to the west of the Urla. The Kocadağ volcanics have asimilar stratigraphic position to the Armağandağ volcanics, but theirpyroclastic products are dominated by block-and-ash-flow-deposits.The andesitic lavas of the Kocadağ volcanics are characterised byplagioclase and brown amphibole phenocrysts in a glassy matrix.

volcanic rocks and the sample locations collected for geochemical and radiometric age

Table 1Summary of step-heating 40Ar/39Ar data in this study.

Unit Sample(this study)

Material used Preferred age Total gas age Plateau age Isochron age PublishedK–Ar ages⁎(Ma) (Ma) (Ma) (Ma)

Yaylaköy volcanics FK-3 Whole rock 17.0±0.4 21.5±0.2 17.0±0.4 21.3 (1)19.2 (1)

Karaburun basalt Groundmass 18.5 (2)16.0 (2)

Kocadağ volcanics FK-1 Whole rock 17.5±0.1 17.6±0.1 17.5±0.1 16.6 (1)17.3 (2)

Armağandağ volcanics HS-294 Groundmass 17.3±0.1 17.3±0.1 18.2 (1)Uzunkuyu intrusives SK-9 Groundmass 16.7±0.1 17.2±0.1 17.0±0.5 16.7±0.1 15.4 (2)Urla volcanics Whole rock 11.9 (1)Ovacık basalt Whole rock 11.3 (1)

Published K–Ar ages (⁎) are: (1) Borsi et al. (1972); (2) Türkecan et al. (1998). The localities of the samples are indicated on Fig. 3.

186 C. Helvacı et al. / Journal of Volcanology and Geothermal Research 185 (2009) 181–202

Amphiboles are the dominant phenocryst phase in the Kocadağ vol-canics with respect to the lavas of the Armağandağ volcanics (~40–45% of total amount of phenocrysts). The unit interfingers with theUrla limestone and previously dated as 16.6 and 17.3 Ma (K–Ar ages)by Borsi et al. (1972). The Yaylaköy volcanics are exposed to the south-southwest part of the Karaburun town in an area of ~15 km2 (Fig. 3).The lavas of the unit are basalt-like in hand-specimen, but they aregeochemically andesitic and basaltic andesitic in compositions (seebelow), with mainly altered olivine phenocrysts found in a flow-oriented plagioclase microlites-rich black glassy matrix. Borsi et al.(1972) obtained 19.2 and 21.3 Ma K–Ar ages from the Yaylaköyvolcanics.

The Urla volcanics are composed of fine grained rhyolitic lavasand porphyritic domes with trachyte-like porphyritic rhyolite domes.These units crop out in an area of ~25 km2 to the north of the Urla(Fig. 3). The unit cuts and unconformably overlies the Urla limestone.The fine-grained rhyolite lavas are characterized by anhedral toeuhedral quartz and euhedral sanidine phenocrysts in a white coloredvolcanic matrix. The porphyritic rhyolitic domes are pink to greyishred colored and are represented by euhedral sanidine phenocrysts upto 5–6 cm in length. Petrographically, some sanidine phenocrysts havegrown around older plagioclase phenocrysts that include inclusions ofhornblende and opaque minerals. The Urla volcanics have been datedby Borsi et al. (1972) as 11.7 and 11.9 Ma (K–Ar ages). The overlyingOvacık basalt crops out in three different localities around the Urlavillage as small basaltic lava flows (Fig. 3). The lavas are characterizedby olivine phenocrysts in a plagioclase microlite-rich black glassyvolcanic matrix. Borsi et al. (1972) have obtained an 11.3 Ma K–Ar agefrom the unit.

4. Analytical techniques

Age determinations were carried out on a total of 4 lava samplesthat were analyzed by the 40Ar/39Ar incremental heating method inthe Nevada Isotope Geochronology Laboratory at the University ofNevada, Las Vegas. Samples were analyzed by the furnace step heatingmethod utilizing a double vacuum resistance furnace similar to theStaudacher et al. (1978) design. Details of the analytical methods anddata treatment are discussed by Justet and Spell (2001) and Spell andMcDougall (2003). Details of four the new age determinations aresummarized in Table 1 with selected age spectra plots illustrated inFig. 5.

Forty-nine lava samples from the late Cenozoic volcanic rocks of theKaraburun Peninsula were analyzed for their major and trace elementconcentrations. Rock powders of the selected fresh rock samples wereprepared by removing the altered surfaces and powdering with atungsten carbide shatter box at Dokuz Eylül University.Major and traceelement analyses were performed by ACME Analytical LaboratoriesLtd in Vancouver. Major oxides were reported on a 0.2 g sample ana-

lyzed by ICP-emission spectrometry following a lithium metaborate/tetraborate fusion and dilute nitric digestion. Loss on ignition (LOI)was determined by weight difference after ignition at 1000 °C. Theresults are given in Appendix 1 with UTM coordinates of the samplelocations. The major elements and water contents of the samples arelisted in the Appendix 1, but the results have been evaluated on ananhydrous basis in the classification and the other diagrams.

5. 40Ar–39Ar results

Abasaltic lava sample from theYaylaköy volcanics (sample FK-3) hasyielded a total gas age of 21.5±0.2 Ma. Steps 7–10 define a “pseudo-plateau” age of 20.3±0.6 Ma, which represents less than 50% of the39Ar released (Fig. 5a). Steps 5–10 (57% of the 39Ar released) yield asignificantly younger isochron age of 17.0±0.4 Ma and indicates thatexcess argon is present in the sample (40Ar/36Ar=323.0±2.7). The agespectrum is not U-shaped and apparently caused by the presence ofexcess argon in the sample. The isochron age should be consideredthe most reliable date for this sample (Fig. 5b). A lava block from thepyroclastic rocks of the Armağandağ volcanics (sample HS-294 fromthe block and ash flows) gave a total gas age of 17.3±0.1 Ma. There isno plateau or isochron age defined for this sample (Fig. 5c). Discordantage spectra can be caused by excess argon or alteration. Similarly, anandesitic lava sample from the Kocadağ volcanics (sample FK-1) gave atotal gas age of 17.6±0.1Ma(Fig. 5d). There is noplateau for this sample,but steps 6–12 have yielded an isochron age of 17.5±0.1 Ma. Althoughthe isochron is defined by b50% of the gas released it should beconsidered reliable as it is defined by a large number of data points(n=7) with a reasonable spread in radiogenic yield. A mafic dyke fromtheUzunkuyu (Sample SK-9) yielded a total gas age of 17.2±0.1Ma anda plateau age of 17.0±0.5 Ma, as defined by steps 4–7 (55% of the 39Arreleased) (Fig. 5e). On the other hand, steps 4–8 (62% of the 39Arreleased) define a younger isochron age of 16.7±0.1 Ma and indicateexcess argon is present in this sample (40Ar/36Ar=330.0±5.5) (Fig. 5f).The isochronage shouldbe considered themost accurate and reliable forthis sample because of the presence of excess argon, i.e., both the totalgas and plateau ages are anomalously old.

The results of 40Ar–39Aragedating fromthis studyand thepreviouslypublished age data indicate that two-stage volcanism occurred in theKaraburunPeninsula during (1) the earlyMiocene (Burdigalian) and (2)the latest middle Miocene to late Miocene (Serravalian to Tortonian).

6. Major and trace element characteristics

Miocene volcanic rocks of the Karaburun Peninsula have beenplotted on a total alkalis (K2O+Na2O) vs. silica (SiO2) (TAS-IUGS)classification diagram of Le Maitre (2002) on which the alkaline–sub-alkaline discrimination line of Irvine and Baragar (1971) is also shown(Fig. 6a). For comparison several volcanic rocks from the surroundings

Fig. 5. Apparent age spectrum for the dated volcanic rocks in this study. See text for detailed discussion.

187C. Helvacı et al. / Journal of Volcanology and Geothermal Research 185 (2009) 181–202

of the Karaburun Peninsula have also been plotted on these diagrams.These are Pirgi basalt which is associated with rhyolites and andesitesfrom the Chios, rhyolites and mafic rocks from Foça and andesites and

dacites from the Yuntdağı volcanics. Ezine alkaline basalts (~11–7Ma)are also evaluated in order to compare them with the Ovacık basalts(~11 Ma).

Fig. 6. Total alkali-silica (TAS) (a) and K2O versus SiO2 (b) plots with IUGS fields after LeMaitre (2002) for the Karaburun Peninsula. Alkaline–subalkaline line is according to Irvineand Baragar (1971). High-K-shoshonitic line in (b) is from Pecerillo and Taylor (1976).Chios volcanics (Pe-Piper et al.,1995), Foça and Yuntdağı volcanics (Aldanmaz et al., 2000,Akay and Erdoğan, 2004 and Innocenti et al., 2005), and Ezine basalts (Aldanmaz et al.,2000) are also shown for comparison. Data are plotted onwater-free basis.

188 C. Helvacı et al. / Journal of Volcanology and Geothermal Research 185 (2009) 181–202

All volcanic units are sub–alkaline except for the Ezine basalts andsome samples of the Urla volcanics, Ovacık basalts and Foçamafic rocks.The fine-grained lavas of the Urla volcanics are generally alkaline incharacter, while porphyritic rhyolitic domes are generally sub-alkaline.The sub-alkaline volcanic rocks also show calc-alkaline trend. TheOvacık basalt plots very close to the dividing line between alkaline andsub-alkaline fields, and hence, will be classified as mildly alkaline. Themafic rocks from Foça showa very scattered distribution. Some samplesof the Foça mafic rocks are strongly alkaline. On the other hand, somesamples of Foça mafic rocks plot also close to the dividing line or evenon the sub-alkaline field (calc-alkaline). This may be resulted fromrandomly selected samples without detailed petrographic and fieldstudies.

The most mafic lavas of the early Miocene Karaburun volca-nics (Mg#=54–72) are classified as basaltic–andesite and basaltic–trachyandesite (shoshonite). All samples of theYaylaköy (Mg#=53–64),Armağandağ (Mg#=45–53) andKocadağ (Mg#=24–56) volcanics andthe Uzunkuyu intrusives (Mg#=45–50) are classified as andesites,dacites, trachyandesite (latite) and trachydacites, similarly to Yuntdağvolcanics (Fig. 6a). These volcanic units share also common geochemicalfeatures, and hence, referred as YAKU volcanics in following. The Urlavolcanics differ from the rhyolites of Chios and Foça in respect to their

higherK2O contents. The samples fromtheOvacıkbasalt plot in thebasaltand basaltic–trachyandesite (shoshonite) fields. The Foça mafic alkalinerocks showa scattered distribution ranging from trachybasalts (potassic–trachybasalts and hawaiites) and basaltic trachyandesites (shoshonites)to basaltic andesites and trachyandesites. Some samples also have veryhigh K2O contents and are classified as phonolite and tephriphonolite. Allthe early Miocene volcanics of the Karaburun Peninsula are high-K calc-alkaline except for some samples from theKaraburunvolcanicswhich areshoshonitic according to their K2O and silica contents (Fig. 6b).

Major element contents of the volcanic rocks of the region areplotted versus increasing silica contents as a differentiation index(Fig. 7). All volcanic units show good negative correlations withrespect to Fe2O3, MgO, CaO and TiO2. K2O and Na2O contents of theunits increase as their silica contents increase. Some samples from theFoça mafic rocks which have very high K2O contents are not shown.Al2O3 contents of the Karaburun and Yaylaköy volcanics, the Foçamafic rocks and the Ovacık basalt increase with increasing silicacontents. Al2O3 contents of the YAKU volcanics remain nearly constantas the SiO2 contents increase, while those of the rhyolitic volcanicsdecrease. P2O5 contents of the Karaburun and Ovacık basalts rapidlydecrease in a narrow SiO2 range.

In Fig. 8, trace element concentrations of the volcanic units havebeen plotted versus their silica contents. Rb (and Th) contents of theandesitic and dacitic rocks increase slightly with silica. Sr contents ofthe Karaburun volcanics and the Ovacık basalt also increase, whilethose of the other volcanics decrease with respect to increasing SiO2

contents. Ce, Zr, Y and Yb contents of the Urla volcanics rapidlyincrease distinctly from the rhyolites from Foça and Chios.

The SiO2-dependentmajor and trace element variation diagrams inFigs. 7 and 8 and the classification diagrams in Fig. 6 indicate that: (1)the YAKU volcanics have nearly the same chemical composition andshare the same petrological characteristics, and these are also com-parable with the Yuntdağı volcanics; (2) the Urla volcanics havedifferent chemical composition from the Foça and Chios rhyolites; (3)the Ovacık basalt has a similar composition to the middle MiocenePirgi basalt; (4) some samples from the Foça mafic rocks are alsosimilar to Karaburun volcanics.

On the chondrite-normalized REE diagram and primitive mantle-normalized multi-element diagrams all the volcanic units show en-richment in light rare earth elements (LREE) and large ion lithophileelements (LILE) with respect to high field strength elements (HFSE)(Fig. 9). The Karaburun volcanics and the Pirgi basalt show verysimilar patterns on these diagrams. The Pirgi basalt differs from theKaraburun volcanics only by the absence of a Nb anomaly. The calc-alkaline andesitic–dacitic rocks and the Chios andesites also have verysimilar patterns, with a narrow compositional interval, pronouncedNb and Ti negative anomalies and Eu/Eu⁎ values in a range of 0.68–0.96 (Fig. 9b). The Urla volcanics have different patterns with re-spect to those of the Foça and Chios rhyolites; they have higher LREEcontents and strong Eu negative anomalies (Eu/Eu⁎=0.05–0.41)(Fig. 9c). All rhyolitic rocks also show depletion in Ti. In Fig. 9dnormalized patterns of samples from the Ovacık basalt are showntogether with the Ezine basalts. The samples from the Ovacık basaltaremore enriched in LILE and also show a distinct depletion in Nb andTi with respect to the Ezine basalts. These features of the Ovacık basaltare comparable with those of other samples from the KaraburunPeninsula and surroundings.

7. Petrologic evolution

On the basis of major and trace element characteristics, it is con-cluded that the latest early Miocene volcanism in the KaraburunPeninsula is characterized by (1) basaltic-andesites and shoshonites(Karaburun volcanics) and (2) high-K calc-alkaline andesites anddacites (YAKU volcanics). The Neogene volcanism also continuedduring the latest middle Miocene (~11 Ma) with bimodal products

Fig. 7. (a–f) Major element variation diagrams versus increasing SiO2 contents of the volcanic rocks from the Karaburun Peninsula. Chios volcanics (Pe-Piper et al., 1995) with the Foçaand Yuntdağı volcanics (Aldanmaz et al., 2000, Akay and Erdoğan, 2004 and Innocenti et al., 2005), are also shown for comparison. Data are plotted on water-free basis.Discrimination lines for adakites are from Castillo (2006).

189C. Helvacı et al. / Journal of Volcanology and Geothermal Research 185 (2009) 181–202

comprising the Urla volcanics and Ovacık basalt. The latest earlyMiocene volcanic rocks are characterized by negative Eu anomalies onN-MORB-normalized spider diagrams, and Nb and Ti (HFSE) depletionwith respect to the LILE and LREE on N-MORB-normalized multi-element diagrams. These data may indicate they are co-genetic, andtheir trace and major element variations on Harker diagrams couldhave resulted from: (a) primary geochemical characteristics of sourceregion of the magmas (e.g., mineralogical variations and/or variableenrichment degrees in the mantle source); (b) different degrees ofpartial melting; (c) shallow-level magmatic differentiation processes[e.g., fractional crystallization (FC); assimilation and fractional crys-tallization (AFC)]. In this section we discus differentiation processes

within the magmas that produced the volcanic units, using the maingeochemical differences between them.

7.1. Fractional crystallization and crustal assimilation

In the SiO2-dependent Harker variation diagrams (Figs. 7 and 8),Fe2O3, MgO, Ni, Sc and Cr contents of the Karaburun and YAKUvolcanics and Ovacık basalt decrease with respect to their increasingSiO2 contents, which may be indicative of fractionation assemblagesof mafic phases, such as olivine and pyroxenes. Depletion in Ni isprominent in the Karaburun and Yaylaköy volcanics and the Chiosandesites (Fig. 8). The combined decreases in CaO and Sc may reflect

Fig. 8. (a–h) Trace element variation diagrams versus increasing SiO2 contents of the volcanic rocks from the Karaburun Peninsula. Chios volcanics (Pe-Piper et al.,1995)with the Foça andYuntdağı volcanics (Aldanmaz et al., 2000, Akay and Erdoğan, 2004 and Innocenti et al., 2005), are also shown for comparison. Discrimination lines for adakites are from Castillo (2006).

Fig. 9. N-MORB-normalized (a) chondrite-normalised REE, and (b)multi-element diagrams for the volcanic rocks from the Karaburun Peninsula and surroundings. Normalizing values arefrom Sun andMcDonough (1989). Chios volcanics (Pe-Piper et al., 1995), Foça volcanics (Akay and Erdoğan, 2004) and Ezine basalts (Aldanmaz et al., 2000) are also shown for comparison.

190 C. Helvacı et al. / Journal of Volcanology and Geothermal Research 185 (2009) 181–202

191C. Helvacı et al. / Journal of Volcanology and Geothermal Research 185 (2009) 181–202

Fig.10. (a–b) FC and AFCmodels for the volcanic rocks in the Karaburun Peninsula. Dataare normalized to N-MORB compositions (Sun and McDonough, 1989). Bulk partitioncoefficients used in modeling are indicated on x-axis. “F” values represent the meltremaining after crystallization. “r” values indicate the relative ratio of assimilatedmaterial to crystallized material. Upper continental crust values are from Taylor andMcLennan (1995). Partition coefficients are from Rollinson (1993; and the referencestherein).

192 C. Helvacı et al. / Journal of Volcanology and Geothermal Research 185 (2009) 181–202

clinopyroxene fractionation for all the rock groups, except for theArmağandağ volcanics. Increasing Al2O3 and Sr contents also rule outplagioclase fractionation for the Karaburun volcanics and the Ovacıkbasalt. Instead, combined decreases in CaO, Al2O3 and Sr contents ofYAKU volcanics, especially for the rocks with SiO2N~57, indicative ofplagioclase dominated fractionation. Negative Eu anomalies (meanEu/Eu⁎=0.75, Fig. 9) also support plagioclase dominated fractiona-tion for the YAKU volcanics. On the other hand, small Eu anomaliesin the Karaburun and Ovacık basalt can be explained by crustal as-similation or primary features of the mantle source. Amphibole frac-tionation played a role in petrogenesis of the Kocadağ volcanics andUzunkuyu intrusives, as only their Gd, Tb, Dy, Zr and Y contentsdecrease. Strong depletions in TiO2 strongly suggest the fractionationof Fe–Ti oxides for the Karaburun, Yaylaköy, Uzunkuyu volcanics andthe Ovacık basalt. Rhyolitic rocks in the region show evidence forfeldspar-dominated (plagioclase and K-feldspar) fractionations astheir Al2O3, K2O, Sr and CaO contents decreases and they show strongEu anomalies (Eu/Eu⁎=0.05–0.41, Fig. 9). Systematic increases in theK2O contents of the rocks of the YAKU volcanics also indicate that theyhave not undergone K-feldspar fractionation. These results aresummarized in Table 2.

The Harker variation diagrams illustrated in Figs. 7 and 8 indicatethat the volcanic units in the Karaburun Peninsula may be cogeneticand related to one another. For example, it appears that the Yaylaköyvolcanics and the Uzunkuyu intrusives were both derived via FC–AFCof a primitivemagma that is represented by themoremafic Karaburunvolcanics. To test this hypothesis, we construct an N-MORB-normal-ized REE diagram on which theoretical FC and AFC model results areshown by using equations of De Paolo (1981) (Fig. 10). In Fig. 10a,sample L-30 is chosen from the Karaburun volcanics as C0, repre-senting the most primitive magma that also has the lowest SiO2

and REE contents. The crystallizing phases are chosen as olivine0.60+clinopyroxene0.30+Fe–Ti-oxides0.10 according to both the phenocrystassemblages in the lavas and the results of Harker variation diagrams.Mineral/liquid partition coefficients are compiled from Rollinson(1993). Both FC and AFC (r=0.3; relative ratio of assimilated materialto crystallized material) modeling results indicate that overall REEpatterns of the lavaswould have increased during these processes. Themodeling results agree well with the REE pattern of the sample whichhas highest REE contents in the Karaburun volcanics, and hence, FC orAFC processes with F=0.85 (15% crystallization of olivine andpyroxenes) could account for their REE budgets. On the other hand,the modeling results also show that the Yaylaköy volcanics could notbe produced neither by FC or AFC processes from a magmarepresented by samples of the Karaburun volcanics, as the resultingmagma from FC or AFC of a parental liquidwould have higher contentsof REE.

Table 2Summary of fractionating mineral phases in the genesis of the volcanic units in theKaraburun Peninsula and surroundings.

Volcanicunits

Olivine Pyroxenes Amphibole Plagioclase K-feldspar Fe–Ti-oxides

Karaburunvolc.

+ + − − − +

Yaylaköyvolc.

+ + − +(?) − +

Kocadağ volc. +(?) + + + − +Armağandağvolc.

+(?) − − + − +

Chiosandesites

+ + (?) − − −

Uzunkuyuint.

+(?) + + + − +

Ovacık basalt + + − − − +Urla volc. +(?) +(?) + + +

In Fig. 10b, sample L-39 (from the Yaylaköy volcanics) has also beenmodeled to produce themore felsic Armağandağ and Kocadağ volcanicsand the Uzunkuyu intrusives via FC and AFC processes. The best-fittingpattern of FC modeling suggests 90% crystallization of selected mineralphases (olivine0.50 + pyrocenes0.25 + plagioclase0.20 + amphibole0.05),but this is not realistic for producing other magmas with higher REEcontents. On the other hand, samples of the Armağandağ and Kocadağvolcanics and the Uzunkuyu intrusives have higher REE contents thatcould be produced by AFC processes with higher “r” values. This sug-gests that the crustal assimilation processes played an important rolein genesis of the high-K calc-alkaline andesites and dacites of theKaraburun Peninsula. In Fig.10b, sample L-39 has beenmodeled for AFCprocesses with parameters of r=0.5 and F=0.75 which are morerealistic. The resulting REE-pattern of AFC modeling matches well withthe patterns of samples that have the highest REE, and suggest thatthe Kocadağ, Armağandağ and Uzunkuyu intrusives may be derivedvia AFC processes from a magma that also produced the Yaylaköy vol-canics. It is also important to note that the Kocadağ, Armağandağ andUzunkuyu intrusives have different phenocryst phases. For example, thepetrographic and geochemical features of the Kocadağ volcanics and theUzunkuyu intrusives indicate that amphibole was a fractionating phase

Fig. 11. Th/Yb against Ta/Yb log–log diagram (after Pearce,1983) for theNeogene volcanicrocks in the Karaburun Peninsula and surroundings. Ezine basalts (Aldanmaz et al., 2000),Samos basalts (Pe-Piper and Piper, 2007b), Chios volcanics (Pe-Piper et al., 1995), activeAegean arc lavas (Buettner et al., 2005), Foça mafic alkaline lavas (Aldanmaz et al., 2000and Innocenti et al., 2005) are also shown for comparison.

193C. Helvacı et al. / Journal of Volcanology and Geothermal Research 185 (2009) 181–202

in their genesis on contrast to the Armağandağ volcanics. This would beresulted in MREE depletion on the N-MORB-normalized REE diagrams,and indeed, the samples of the Kocadağ volcanics and the Uzunkuyuintrusives do have depleted MREE patterns.

7.2. Source characteristics

Geochemical characteristics of the latestearlyMiocenevolcanic unitsin theKaraburunPeninsula and surroundings, suchashigh incompatibleelement concentrations, enriched LREE and LILE patterns on N-MORBnormalized elementdiagrams (Figs. 7 and8) suggest that themagmasofthese volcanic rocks were derived from enriched lithospheric mantlesources, as indicated by previous studies in Western Anatolia (e.g.,Aldanmaz et al., 2000). Isotopic compositions of these volcanic units(Table 3) also suggest an enriched lithospheric mantle source withrelatively high 87Sr/86Sr and low 143Nd/144Nd ratios, even in the mostmafic units. Latest middle Miocene alkali basalts in the region (theOvacık basalt and the Foça mafic rocks and also the Samos basalts) alsohave similar isotopic ratios to the older volcanic units, although theyhave slightly lower 87Sr/86Sr compositions.

Negative Nb and Ti anomalies on N-MORB multi-elementdiagrams for the Miocene volcanic units in the region may alsoindicate that the enrichment process is related to subduction zoneprocesses. This interpretation is also supported by Ta/Yb vs. Th/Ybratios of the volcanic units (Fig. 11). Ta/Yb and Th/Yb ratios of allvolcanic units in the Karaburun Peninsula and surroundings areshifted away from mantle array to higher values. This compositionalchange is attributed to subduction-related processes. Hence, it isconcluded that the mantle source region of the volcanic rocks waslithospheric mantle that is enriched by subduction related processes.It is also worth noting that the late Miocene basalts of Samos Islandand active Aegean subduction zone volcanics (Buettner et al., 2005)also have similar Ta/Yb and Th/Yb ratios to those of the volcanicunits from the Karaburun Peninsula and surroundings. On the otherhand, only the late Miocene alkali basalts from the Ezine, located tothe further north of the study area, have distinct Ta/Yb and Th/Ybratios that lie on mantle array. These data also indicate that the lateMiocene Ezine basalts have a different source and they should beevaluated separately from the other Miocene rocks in the region(e.g., Aldanmaz et al., 2000, 2005, 2006).

In order to evaluate the mantle source mineralogy fromwhich thevolcanic rocks were derived, Zr/Nd–Lu/Hf, La/Rb–Yb–Rb, Ce/Yb–Gd/Yb and Ni/Sr–Sc/Sr element ratios have been modeled using non-modal batch melting equations of Shaw (1970). Hypothetical melt-ing curves have been calculated using enriched-depleted MORB mantle

Table 3Available Sr and Nd isotopic data for the volcanic rocks in Karaburun Peninsula andsurroundings.

Volcanic units: 87Sr/86Sr 143Nd/144Nd Data source

Pirgi basalt 0.512560 Pe-Piper et al., 1995Chios andesites 0.709600 0.512330 Pe-Piper et al., 1995Karaburun volcanics 0.706400 Borsi et al., 1972Yaylaköy volcanics 0.708000 Borsi et al., 1972

0.708100Armağandağ volcanics 0.706400 Borsi et al., 1972Kocadağ volcanics 0.708200 Borsi et al., 1972

0.7073000.706700

Uzunkuyu intrusives 0.708000 Borsi et al., 1972Ovacık basalt 0.706232 0.512564 Borsi et al., 1972

0.706262 0.512573 Innocenti et al., 20050.706277 0.512571 Agostini et al., 20050.707540 0.512492

Samos basalts 0.705902 0.512572 Robert et al., 19920.706547 0.512527

(E-DMM) as the source material that proposed by Workman and Hart(2005) (Fig. 12). Small ranges of melting degree curves are shown inthesediagrams (0.01–1.00%). TheEzinebasalts, Samosbasalts and activeAegean arc lavas are also shown for comparisonwith the Karaburun andYaylaköy volcanics and the Ovacık basalt. Melting curves have beendrawn for several mantle facies including garnet lherzolite (Walter,1998, curve A), garnet–amphibole lherzolite (Barry et al., 2003, curve B),garnet–amphibole–phlogopite lherzolite (Barry et al., 2003, curve C),phlogopite–garnet harzburgite (Wang et al., 2004, curve D) and spinel–lherzolite (Kinzler,1997, curve E). In thesediagrams, the effects of FC andAFC have also been shown, where they are effective. In AFC models,upper crust values suggested by Taylor andMcLennan (1995)were usedas the assimilant. The melting parameters and the partition coefficientsused in modeling are given in Table 4. Zr/Nd–Lu/Hf ratios are notdependent on either fractionation of olivine, plagioclase and pyroxenesor on AFC processes. Except for Nd, which is very slightly enrichedin arc supra-subduction regions, these elements are also independentfrom source enrichment by subduction fluxes (e.g., Pearce andParkinson, 1993), so these ratios reflect the melting ratios and sourcemineralogy. Zr/Nd–Lu/Hf ratios are not affected by melting of spinellherzolite, but the presence of the garnet in the sourcewill decrease Lu/Hf ratio rapidly, as Lu is a compatible element in garnet (Fig. 12a). Thesevariations in the element ratios suggest that melting of the mantlesource rocks of the Karaburun basalts, Yaylaköy volcanics and Ovacıkbasalt occurred in the garnet stability field. On the other hand, it is notclear from this diagram whether the source mineralogy includes ahydrous phase such as amphibole/phlogopite or not.

La/Rb–Yb–Rb ratios are not affected by fractionation of olivine,plagioclase, pyroxenes and amphibole, and or AFC processes includingthese fractionating mineral phases. However, these ratios are verysensitive to biotite fractionation; but the phenocryst assemblages ofthe volcanic units do not include biotite, and hence, been concludedthat biotite was not a fractionating phase that would have resulted ingeochemical differentiation in these volcanic rocks. These ratios arealso affected by fluxes released from the subducted lithosphere, as Rb

Fig. 12. (a) Lu/Hf–Zr/Nd, (b) Yb/Rb–La/Rb, (c) Gd/Yb–Ce/Yb and (d) Ni/Sr–Sc/Zr diagrams for the Karaburun volcanics and Yaylaköy volcanics, Ezine, Pirgi and Ovacık basalts, onwhich hypothetical non-modal batch meltingmodels of enriched-depletedMORBmantle (E-DMM,Workman and Hart, 2005) are shown. UCC: Upper Continental Crust of Taylor andMcLennan (1995). Melting curves are drawn for the range of 0.01%–1% melting in a, b and c. A: garnet lherzolite, B: garnet–amphibole lherzolite, C: garnet–amphibole–phlogopitelherzolite, D: phlogopite–garnet harzburgite, E: spinel lherzolite. Vector 1 in b represent biotite fractionation and AFC for F=0.65. Vectors 2 and 3 in c represents amphibolefractionation and AFC, respectively, for F=0.65. F values represent the melt remaining after crystallization. Partition coefficients and the their parameters used in modeling aresummarized in Table 4. See Fig. 11 for the symbols.

194 C. Helvacı et al. / Journal of Volcanology and Geothermal Research 185 (2009) 181–202

concentration increases in this way (Pearce and Parkinson, 1993). Inthis case, La/Rb–Yb–Rb ratios can also be used for evaluate the effectsofmelting processes and/or subduction-related contributions. Fig.12bapparently shows that melting of the mantle source did not occur inthe presence of phlogopite, or in the pure garnet facies, as the La/Rbratios of the volcanic units should be decreased either by melting ofspinel facies (curve E) or by garnet–amphibole facies (curve B)mantlesource.

Ce/Yb–Gd/Yb ratios are affected by amphibole fractionation andrelated crustal assimilation processes, but these effects do not matchthe evolutionary trend of the volcanic units, and hence these ratioscan be used formeltingmodels (Fig.12c). Fig.12c clearly indicates thatthe most favorable melting curve for the volcanic units is representedby garnet–amphibole lherzolite facies mantle. On the other hand,according to this model, the source rocks should be more enriched inCe to fit the compositional trend of the volcanic units on the diagram.Themineralogical composition of the garnet–amphibole facies used inthe modeling includes 5% garnet. The garnet amount in the source,however, should be lower than 5% to produce lower ratios of Gd/Ybthat are similar to those of volcanic units.

In Fig. 12d, Ni/Sr ratios have been modeled against Sc/Sr ratios. Niand Sc values are taken from the depleted mantle compositionsuggested by Salters and Stracke (2004), assuming that the compa-tible elements (Ni and Sc) do not vary considerably in depletedmantleand E-DMM sources. Melting models show that the Karaburun basaltscould be produced by small-degree melting (1–2%) of an enrichedmantle source with garnet–amphibole lherzolite facies. Olivine frac-tionation (~5–10%), as indicated in Fig. 9b, can best account for thecomposition of the Karaburun basalts.

As a conclusion, themeltingmodels indicate that the volcanic unitsin the region were derived from an enriched mantle source. Thisenrichment event cannot be explained by a low-degree of melting.The calc-alkaline and LILE-enriched, and HFSE-depleted nature of thevolcanics (Fig. 9) also implies that the source region had beenmetasomatized by subduction-related processes; i.e., a metasoma-tized lithospheric mantle (e.g., Pearce et al., 1995). It is also importantto note that the trace element compositions of the rocks of lateMiocene Samos basalts and active Aegean arc lavas are similar to thoseof the Miocene volcanic units in the Karaburun Peninsula and sur-roundings. The suggestion of amphibole-bearing source rock (i.e.,

Table 4Partition coefficients and the other parameters used in petrogenetic models in Fig. 12.

Mantle mineralogies⁎

Source modes (X0) Melt modes (P0)

A B C D E A B C D E

Garnet 0.10 0.05 0.05 0.10 – 0.09 0.20 0.20 0.40 –

Spinel – – – – 0.03 – – – – 0.11Olivine 0.60 0.55 0.55 0.60 0.53 0.03 0.05 0.05 0.05 −0.06Clinopyroxene 0.10 0.15 0.15 0.20 0.17 0.88 0.30 0.20 0.32 0.67Ortopyroxene 0.20 0.22 0.20 – 0.27 −0.16 0.05 0.05 – 0.28Biotite/phlogopite – – 0.01 0.10 – – – 0.10 0.23 –

Amphibole – 0.01 0.04 – – – 0.40 0.40 – –

Partition coefficients†

Ba Rb Sr Hf Zr La Ce Nd Gd Yb Lu Ni Sc

Garnet 0.023 0.042 0.012 0.300 0.300 0.010 0.021 0.087 0.498 4.300 5.500 5.100 2.66Spinel – – – – – 0.010 0.010 0.010 0.010 0.010 0.010 – –

Olivine 0.002 0.002 0.002 0.004 0.005 0.001 0.001 0.001 0.002 0.002 0.002 29.000 0.680Clinopyroxene 0.001 0.005 0.096 0.263 0.121 0.044 0.084 0.173 0.336 0.313 0.265 14.000 3.200Ortopyroxene 0.002 0.003 0.007 0.055 0.030 0.002 0.003 0.005 0.016 0.049 0.06 5.000 1.200Biotite/phlogopite 3.480 5.180 0.183 0.190 0.023 0.035 0.030 0.040 0.030 0.040 0.030 1.300 8.300Amphibole 0.450 0.100 0.450 0.500 0.400 0.200 0.350 0.650 0.950 0.800 0.800 6.800 4.200Plagioclase 0.300 0.100 2.000 0.051 0.048 0.270 0.200 0.140 0.066 0.031 0.025 0.040 0.040

Rock compositions‡ (ppm)

E-DMM values 1.219 0.108 9.718 0.186 6.087 0.253 0.726 0.703 0.397 0.382 0.060 1960.0 16.30UCC values 550.0 112.0 350.0 5.800 190.0 30.00 64.00 26.00 3.800 2.200 0.320 20.00 11.00

⁎A: Garnet lherzolite (Walter, 1998), B: Garnet–amphibole lherzolite (Barry et al., 2003), C: Garnet–amphibole–phlogopite lherzolite (Barry et al., 2003), D: Phlogopite harzburgite(Wang et al., 2004), E: spinel lherzolite (Kinzler, 1997).†Partition coefficients are complicated from Rollinson (1993).‡E-DMM values are from Workman and Hart (2005), except for Ni and Sc which are from Salters and Stracke (2004).

195C. Helvacı et al. / Journal of Volcanology and Geothermal Research 185 (2009) 181–202

involvement of a hydrous phase in the mantle) by the melting modelsis also indicative of a metsaomatised source.

8. Discussion and conclusions

The latest earlyMiocenevolcanics fromtheKaraburun Peninsula, thelate Oligocene–middle Miocene lavas in the Western and northwesternAnatolia (Aldanmaz et al., 2000, Altunkaynak and Dilek, 2006) andcentral Western Anatolia (Güleç, 1991; Innocenti et al., 2005, Ersoyet al., 2008) are characterized by (1) basaltic andesites and shoshonites(Karaburun volcanics), and (2) high-K calc-alkaline andesites, dacites,latites and trachydacites (YAKUvolcanics). The latestmiddleMiocene tolateMiocene volcanism is represented bymildly alkaline basalts (Ovacıkbasalt) and rhyolites (Urla volcanics). Similar occurrences in WesternAnatolia share common characteristics: they show LILE and LREE en-richment relative to HFS elements and have high 87Sr/86Sr and low143Nd/144Nd isotopic ratios. These geochemical characteristics are com-parable with volcanic rocks derived from subduction-related metaso-matized lithospheric mantle source (see also Yılmaz, 1989, 1990; Güleç,1991, Aldanmaz et al., 2000; Innocenti et al., 2005; Pe-Piper and Piper2007a,b; Pe-Piper et al. 2009).

Innocenti et al. (2005) have proposed that the source region of theMiocene volcanics in Western Anatolia includes hydrous phases suchas amphibole or phlogopite. Moreover, Ersoy et al. (2008) have pro-posed that the Miocene high-K to shoshonitic and even ultrapotassiclamproitic volcanics from the Selendi Basin (to the further east) werederived from a garnet and amphibole-bearing lherzolitic source.On the other hand, Pe-Piper et al. (2009) have showed that high-K(shoshonitic) calc-alkaline volcanic rocks in the Lesbos, Limnos andsurroundings can be derived by dehydration melting of an under-plated, trace element-enriched, metabasaltic amphibolite source withamphibole, plagioclase, clinopyroxene and minor garnet. The meltingof such underplated metabasaltic source has been interpreted tohave been triggered by asthenospheric upwelling as a result of slabdetachment (Pe-Piper et al., 2009). The distribution of any volcanicactivity resulting from slab detachment should be arranged in a

narrow and linear volcanic zone along which uplift occurred, asdiscussed by Turner et al. (1999) for the Betic–Alboran Domain(Spain). On the other hand, the slab detachment model may notexplain the Neogene volcanism in Western Anatolia, as the surfacedistribution of the volcanism developed in (a) a NE–SW trend in thevicinity of the study area and (b) a nearly E–W trend further north(See Fig. 1). These trends covers large areas rather than the narrowlinearity that would be expected in slab-detachment model.

Conceição and Green (2004) showed that a metasomatic lherzoliticsource rock can produce shoshonitic rocks, such as silica-saturatedtrachyandesites and silica-oversaturated basaltic andesites. Meltingmodels presented in this study also indicate that the early Miocenemafic lavas in the Karaburun Peninsula were derived from a garnetand amphibole-bearing lherzolitic source. The mantle source of thevolcanic rocks was metasomatized by subduction-related fluids asindicated by Ba/Rb–Nb/La trace element ratios of the volcanics. Thetiming of the enrichment event, on the other hand, is highly debated;some authors suggest that the mantle metasomatism occurred duringactive subduction (Fytikas et al., 1984; Okay and Satır, 2000; Innocentiet al., 2005), while others suggest that themetasomatismwas inheritedfrom an ancient subduction event (e.g., Yılmaz, 1989, 1990; Güleç, 1991;Aldanmaz et al., 2000; Altunkaynak and Dilek, 2006). In both cases, themantle source of the volcanics would carry a subduction-relatedgeochemical signature. In the case of metasomatism provided byprevious subduction, the heat source could be explained by astheno-spheric upwelling that triggeredbypartially delaminationof lithosphericroots as suggested by Aldanmaz et al. (2000). This model may alsoexplain the regional uplift andorogenic collapse resulting in exhumationof metamorphic massifs. However, the cause of the southwestwarddecreasing age of volcanism remains unclear in this model.

The amount of mafic volcanic products in the Western Anatolia hasincreased from the Oligocene to middle Miocene times. The composi-tional variations with time have been attributed to crustal thinning inthe region (e.g., Seyitoğlu and Scott,1992; Altunkaynak andDilek, 2006;Altunkaynak and Genç, 2008). Altunkaynak andDilek (2006) have usedSr and Nd isotopic data from the Eocene–late Miocene volcanic rocks in

196 C. Helvacı et al. / Journal of Volcanology and Geothermal Research 185 (2009) 181–202

the region. They have concluded that the 87Sr/86Sr ratios of the volcanicsincreased from Eocene to early Miocene, and interpreted this change asprimarily due to crustal thickening and subsequent thinning. The 87Sr/86Sr ratios of the early Miocene to late Miocene volcanic rocks, on theother hand, decreasedwith time, which is interpreted by them as due todecreasing affects of crustal contamination related to extensionalcollapse of the thickened crust. This interpretation is also valid for theSelendi basin to the east (Ersoy et al., 2008).

The late middle to early late Miocene (b11 Ma) volcanic activity inthe northwestern Anatolia is represented by alkaline basaltic volcanicsand some local rhyolites [the Urla volcanics in the Karaburun PeninsulaandCumaovası volcanics [12.5–9.0Ma, Borsi et al.,1972] to the south). Inthe Foça–Aliağa region, there are calc-alkaline rhyolitic and mildlyalkaline basaltic occurrences describing a bimodal volcanic association.The basaltic volcanic rocks of latest middle Miocene (~11 Ma, Ovacıkbasalt) in the Karaburun Peninsula have similar trace elementcomposition and isotopic ratios to the basaltic outcrops at Söke (Ercanet al.,1986) and Samos (Robert et al.,1992; Pe-Piper and Piper, 2007b). Itis clear, therefore, that thevolcanismmigrated fromnorth to the south inWestern Anatolia during the Eocene to late Miocene times (Fig. 1).

Appendix A

Table A1Representative major and trace element analysis of Karaburun Peninsula volcanics.

Sample L-1 L-2 L-26 L-27 L-28

East 0459500 0459500 0459285 0458360 045836

North 4273500 4273500 4273775 4275425 427542

Unit: (KaV) (KaV) (KaV) (KaV) (KaV)

SiO2 51.40 51.53 51.59 51.07 50.96Al2O3 14.42 14.67 14.49 15.08 14.23Fe2O3 7.34 7.21 7.18 8.00 7.30MgO 7.91 7.51 9.07 5.25 9.16CaO 9.42 9.72 8.47 8.81 8.73Na2O 2.47 2.48 2.50 2.27 2.42K2O 2.68 2.49 2.95 3.41 3.13TiO2 0.83 0.83 0.82 0.93 0.87P2O5 0.43 0.43 0.50 0.69 0.61MnO 0.09 0.10 0.12 0.08 0.12LOI 2.60 2.80 2.00 3.90 2.00Cs 8.50 6.40 3.80 2.60 5.90Rb 89.70 78.90 112.90 138.00 128.30Ba 1162.50 1094.80 1205.10 1195.80 1133.60Sr 650.90 661.70 678.90 677.50 687.20Pb 3.10 4.10 4.00 2.30 2.60Th 12.20 12.50 13.30 14.70 13.50U 2.50 2.50 2.70 2.70 2.60Zr 172.50 172.70 181.60 193.00 185.30Hf 4.70 5.20 5.50 5.70 5.40Ta 0.70 0.70 0.60 0.60 0.50Y 19.70 19.40 19.20 20.20 17.70Nb 9.60 9.80 10.50 10.90 10.00Sc 23.00 23.00 23.00 26.00 24.00Ni 157.10 154.00 157.00 213.30 237.80Co 87.00 49.20 64.50 44.30 109.60V 171.00 166.00 168.00 190.00 174.00La 33.50 32.60 32.60 32.50 30.00Ce 64.10 63.10 64.70 63.80 60.50Pr 7.62 7.29 7.82 8.01 7.48Nd 26.50 26.70 31.20 32.30 29.70Sm 4.80 5.00 5.39 5.43 5.24Eu 1.41 1.37 1.22 1.26 1.19Gd 3.98 4.15 3.86 3.97 3.81Tb 0.60 0.65 0.65 0.66 0.62Dy 3.73 3.35 3.37 3.25 3.21Ho 0.66 0.59 0.64 0.63 0.59Er 2.09 1.87 1.85 1.84 1.68Tm 0.24 0.32 0.26 0.29 0.26Yb 1.89 2.09 1.68 1.70 1.59Lu 0.27 0.27 0.25 0.26 0.27

Fe2O3 is presumably Fe2O3 (total). Major oxides are wt.% and trace elements are in ppm. KaVvolcanics, UV: Urla volcanics, OB: Ovacık basalt, UI: Uzunkuyu intrusive, (⁎) from Innocent

In conclusion, the Miocene volcanism in the Karaburun Peninsulashares common geochemical features with other volcanic rocks insurroundings. The geochemical features of these rocks suggest thatthey were derived from a garnet- and amphibole-bearing lithosphericmantle source that had been metasomatizsed by subduction-relatedfluids. The timing of the enrichment event remains an unsolvedproblem. This problem may be solved by dating of metasomaticminerals from mantle xenolites embedded in the volcanic units.

Acknowledgement

This study was supported by a scientific research project ofDokuz Eylül University (Project No. 03.KB.FEN.058). Special thanksto Martin R. Palmer who helped with the English of final text. TerrySpell is also thanked for his critical comments on the 40Ar/39Argeochronology. Sibel Tatar-Erkül, Özgür Karaoğlu, Orçun Leblebi-cioğlu, Utku Aktaş, Alev Ergene and Çilem Karagöz are alsoacknowledged for their help during field studies and samplepreparation. Georgia Pe-Piper and an anonymous referee are alsothanked for their valuable suggestions.

L-29 L-30 L-32 L-33 L-40

0 0458360 0458200 0455300 0456325 0446355

5 4275425 4275780 4275860 4274260 4273569

(KaV) (KaV) (KaV) (KaV) (KaV)

50.79 49.34 54.05 51.32 52.7315.36 13.81 16.03 15.83 14.627.96 7.33 6.00 7.26 7.375.03 10.19 4.10 8.34 9.368.61 8.59 8.58 8.25 7.822.15 2.17 3.00 2.89 2.533.31 2.93 2.94 2.16 2.510.94 0.85 0.85 0.76 0.800.67 0.60 0.39 0.30 0.440.08 0.12 0.10 0.12 0.124.60 3.60 3.50 2.30 1.301.90 15.10 4.20 2.80 4.80

141.00 118.00 95.40 71.60 96.501221.00 1097.10 1337.50 1423.00 1745.00680.90 608.20 802.50 778.70 864.40

2.90 3.50 4.00 7.30 3.8015.10 13.80 16.80 13.00 18.102.80 2.50 3.40 2.60 3.40

209.50 182.10 191.40 153.70 169.906.00 5.50 5.60 4.10 5.000.60 0.60 0.90 0.60 0.70

20.90 17.50 20.60 19.10 20.1011.80 10.10 13.20 8.70 11.0027.00 24.00 19.00 21.00 27.00

248.70 210.70 90.40 126.10 139.3051.70 73.40 66.70 63.40 55.80176.00 166.00 151.00 161.00 189.0034.10 30.20 36.90 36.10 36.9067.40 61.00 71.50 67.70 72.908.56 7.46 8.51 7.74 8.88

34.50 31.10 33.90 30.70 36.905.69 4.72 5.72 4.70 6.141.36 1.16 1.34 1.18 1.324.26 3.71 4.50 3.69 4.550.68 0.63 0.72 0.62 0.693.51 3.16 3.75 3.16 3.690.67 0.62 0.72 0.67 0.681.91 1.60 2.02 1.95 1.870.30 0.25 0.31 0.29 0.291.81 1.58 1.93 1.87 1.740.27 0.24 0.29 0.29 0.28

: Karaburun volcanics, YV: Yaylaköy volcanics, KoV: Kocadağ volcanics, AV: Armağandaği et al. (2005).

L-3 L-4 L-34 L-35 L-36 L-37 L-38 L-39 L-11 L-12 L-13

0453070 0452710 0456325 0454630 0453515 0452600 0452280 0452090 0460711 0464682 0463768

4270500 4270435 4274260 4271790 4270500 4270430 4269910 4269475 4244641 4245683 4240921

(YV) (YV) (YV) (YV) (YV) (YV) (YV) (YV) (KoV) (KoV) (KoV)

56.79 56.18 58.58 56.27 57.48 56.84 56.10 55.80 62.50 62.70 59.7617.10 16.80 17.83 17.03 16.88 17.34 16.87 16.84 16.44 17.21 16.764.85 5.38 4.36 5.30 5.17 5.47 5.50 5.52 4.91 4.47 5.233.04 3.24 2.81 4.99 4.92 4.91 5.25 4.45 2.32 1.48 3.487.69 7.92 6.15 6.79 6.06 6.77 6.78 7.28 4.78 4.43 6.003.22 3.19 3.35 3.40 3.63 3.52 3.34 3.33 3.62 3.84 3.362.11 2.23 2.44 2.59 2.78 2.43 2.36 2.35 3.56 3.04 2.950.64 0.66 0.78 0.70 0.76 0.75 0.71 0.71 0.65 0.62 0.630.17 0.18 0.24 0.23 0.24 0.21 0.20 0.21 0.23 0.18 0.220.07 0.09 0.06 0.10 0.07 0.08 0.09 0.10 0.06 0.05 0.084.20 4.00 3.20 2.30 1.70 1.40 2.50 3.10 0.80 1.90 1.405.80 6.00 5.00 4.70 2.20 2.50 4.40 4.70 4.40 3.30 4.90

92.00 94.00 99.80 94.30 97.10 86.00 81.00 86.30 126.00 96.70 118.10739.90 744.30 749.00 682.00 693.20 613.40 614.90 584.40 1208.50 948.80 1220.10521.60 543.40 538.20 542.70 511.30 489.00 468.60 518.70 498.30 410.00 529.10

4.50 2.50 3.80 1.70 3.50 7.00 3.60 3.00 1.80 3.10 2.0011.70 11.60 12.70 14.00 11.80 12.00 10.30 10.40 17.00 12.70 16.703.40 3.80 3.90 3.80 3.70 3.40 3.30 3.50 4.70 4.10 4.20

166.60 162.20 186.80 173.40 183.90 165.50 158.20 165.80 196.10 151.10 174.604.80 4.70 5.20 4.60 5.20 4.40 4.30 4.60 5.70 4.70 5.600.90 1.00 1.20 1.00 1.10 1.00 0.90 1.00 1.00 0.90 0.90

19.40 19.10 19.80 18.80 19.40 19.80 17.70 18.90 22.40 21.80 23.1010.20 10.40 15.80 13.00 16.20 12.40 12.40 12.30 10.80 9.30 9.7015.00 16.00 15.00 14.00 14.00 16.00 16.00 16.00 13.00 10.00 15.0048.10 65.50 19.00 55.70 33.20 43.70 56.70 48.30 4.90 3.20 7.1043.40 49.90 165.90 48.00 29.30 32.30 68.80 52.80 54.70 30.00 51.9098.00 103.00 98.00 93.00 100.00 101.00 98.00 101.00 102.00 87.00 116.0030.80 28.30 32.00 29.70 30.70 27.00 25.90 25.80 41.20 34.00 39.7055.10 53.90 59.00 57.60 59.30 51.60 52.20 51.20 79.50 66.90 74.206.37 5.96 7.07 6.68 7.01 6.29 5.99 6.09 8.89 7.47 8.57

23.10 23.60 27.20 26.20 27.40 23.40 22.80 23.90 31.50 26.70 31.603.70 4.20 4.60 4.42 4.57 4.19 4.01 4.07 6.10 5.10 4.901.16 0.99 1.09 1.03 1.11 1.05 1.00 1.05 1.14 1.16 1.303.69 3.31 3.74 3.39 3.70 3.55 3.22 3.39 4.33 3.80 3.720.67 0.62 0.62 0.62 0.64 0.64 0.58 0.61 0.77 0.64 0.803.40 3.56 3.48 3.22 3.31 3.51 3.04 3.39 3.59 3.80 3.810.64 0.63 0.65 0.63 0.65 0.65 0.61 0.62 0.73 0.67 0.751.98 1.91 1.93 1.77 1.82 1.82 1.66 1.90 2.33 2.17 2.180.31 0.27 0.28 0.28 0.29 0.30 0.31 0.30 0.34 0.36 0.351.68 1.88 1.68 1.69 1.70 1.83 1.74 1.61 2.09 1.86 2.100.30 0.23 0.26 0.27 0.28 0.29 0.25 0.27 0.32 0.34 0.35

197C. Helvacı et al. / Journal of Volcanology and Geothermal Research 185 (2009) 181–202

Table A1 (continued)

Sample L-19 L-20 L-21 L-22 L-23 L-24 L-25 L-44 L-45 L-46

East 0467250 0467250 0467250 0467250 0467250 0465446 0465446 0451794 0451794 0451794

North 4249861 4249861 4249861 4249861 4249861 4251829 4251829 4236840 4236840 4236840

Unit: (KoV) (KoV) (KoV) (KoV) (KoV) (KoV) (KoV) (AV) (AV) (AV)

SiO2 61.67 62.54 62.91 62.12 62.36 59.81 60.16 61.17 60.16 60.06Al2O3 17.36 17.54 17.33 17.26 17.54 17.62 17.06 16.86 17.53 17.54Fe2O3 4.74 4.57 4.60 4.68 4.56 5.01 5.22 4.77 5.41 5.47MgO 2.03 1.17 1.67 1.89 1.12 2.54 2.58 2.89 2.78 2.45CaO 4.68 4.65 4.55 4.61 4.47 5.25 5.09 5.22 5.71 5.27Na2O 3.55 3.77 3.70 3.57 3.77 3.39 3.08 3.43 3.49 3.53K2O 3.14 3.27 3.26 3.18 3.30 2.71 2.73 3.34 2.82 2.80TiO2 0.62 0.63 0.62 0.62 0.63 0.66 0.63 0.60 0.66 0.66P2O5 0.18 0.17 0.17 0.17 0.18 0.16 0.17 0.24 0.17 0.17MnO 0.08 0.04 0.06 0.07 0.04 0.07 0.09 0.08 0.09 0.08LOI 1.70 1.40 1.00 1.70 1.90 2.50 2.90 1.20 0.90 1.70Cs 5.40 4.60 4.50 5.70 4.60 4.00 4.00 3.80 4.00 2.90Rb 121.90 121.20 114.50 122.40 125.10 104.40 106.60 118.20 96.60 98.10Ba 775.90 792.70 769.70 791.80 780.50 803.80 1043.60 1172.80 791.30 802.40Sr 413.80 427.10 409.30 425.50 423.50 462.20 470.30 573.40 496.50 464.90Pb 2.10 3.20 2.30 1.50 2.30 1.70 1.60 1.60 1.70 2.90Th 18.10 18.40 16.60 18.30 16.30 16.60 16.00 19.10 16.30 16.10U 4.50 4.80 4.80 4.80 5.00 4.10 4.20 4.20 3.90 4.10Zr 184.80 191.00 191.60 186.90 192.80 151.40 156.20 187.70 151.70 155.60Hf 5.30 5.40 5.80 5.20 5.80 4.40 4.10 5.40 4.40 4.70Ta 0.90 1.00 1.00 1.00 0.90 0.70 0.80 0.80 0.80 0.70Y 21.50 19.70 20.60 21.60 23.40 22.60 23.10 24.40 22.60 22.90Nb 10.60 10.80 10.90 10.90 11.30 8.40 8.50 11.00 8.50 8.70Sc 10.00 10.00 9.00 9.00 9.00 13.00 12.00 14.00 13.00 13.00Ni 2.00 3.00 2.60 1.50 2.80 3.00 2.80 3.50 2.70 3.90Co 46.40 40.10 39.40 42.20 31.20 79.20 46.40 58.90 61.40 39.80V 75.00 73.00 75.00 75.00 74.00 117.00 108.00 108.00 119.00 121.00La 33.40 32.20 33.50 33.50 35.20 35.00 36.10 41.70 32.30 33.50Ce 65.00 60.30 62.80 65.90 66.00 63.10 67.60 80.40 63.50 63.00Pr 7.52 7.24 7.48 7.66 7.97 7.65 7.99 9.69 7.42 7.53Nd 30.10 27.00 29.10 29.10 29.50 29.60 32.10 38.50 28.30 31.00Sm 4.88 4.60 4.94 4.74 5.21 4.99 5.07 6.14 4.77 5.13Eu 1.04 1.06 1.04 1.05 1.11 1.11 1.10 1.25 1.05 1.13Gd 3.73 3.54 3.70 3.69 3.96 4.14 4.07 4.76 3.96 3.99Tb 0.65 0.60 0.65 0.67 0.69 0.70 0.70 0.80 0.71 0.70Dy 3.43 3.31 3.32 3.43 3.62 3.94 3.91 4.28 3.75 3.81Ho 0.68 0.63 0.67 0.68 0.71 0.74 0.72 0.82 0.77 0.75Er 1.97 1.95 2.00 2.07 2.21 2.23 2.13 2.34 2.14 2.18Tm 0.31 0.30 0.32 0.32 0.36 0.33 0.34 0.35 0.34 0.33Yb 1.95 1.84 1.89 2.01 2.12 2.10 2.16 2.12 2.10 2.03Lu 0.31 0.27 0.29 0.31 0.35 0.32 0.32 0.35 0.33 0.33

198 C. Helvacı et al. / Journal of Volcanology and Geothermal Research 185 (2009) 181–202

L-47 KB-11 KB-12 KB-13 L-15 L-16 L-48 L-49 L-50a L-50b

0451794 0452077 0452746 0453733 0477299 0477299 0477299 0477299 0479500 0479500

4236840 4236600 4236004 4236324 4246739 4246739 4246739 4246739 4245340 4245340

(AV) (AV) (AV) (AV) (UV) (UV) (UV) (UV) (UV) (UV)

59.54 61.01 60.73 61.61 70.40 71.15 69.65 69.25 69.74 68.7317.40 17.06 16.48 16.63 16.39 15.86 16.94 17.11 16.47 16.775.47 5.33 5.10 5.06 0.89 0.72 0.62 0.72 1.15 1.242.82 2.31 2.79 2.62 0.01 0.01 0.01 0.01 0.23 0.255.60 5.31 5.13 5.10 0.12 0.10 0.21 0.21 0.79 0.963.50 3.68 3.20 3.47 5.86 4.63 6.02 6.00 4.62 4.872.82 2.50 3.78 3.37 5.45 6.83 5.98 5.93 5.41 5.420.67 0.69 0.62 0.60 0.14 0.15 0.18 0.18 0.37 0.390.17 0.17 0.24 0.22 0.04 0.04 0.03 0.03 0.09 0.090.09 0.07 0.08 0.07 0.01 0.01 0.01 0.01 0.01 0.011.80 1.70 1.50 0.90 0.80 0.60 0.30 0.50 0.90 1.203.90 2.50 3.90 3.60 7.30 6.70 11.20 11.40 8.80 8.80

94.30 95.00 132.10 120.50 383.80 475.50 327.70 333.80 207.10 200.10748.10 823.00 1282.00 1223.70 27.50 46.50 109.40 97.90 223.30 261.30480.10 421.90 507.00 498.80 8.30 12.70 26.90 31.70 108.80 124.10

1.20 3.60 1.00 4.80 21.50 27.40 24.10 21.50 2.50 2.1014.60 11.10 17.20 16.00 71.80 84.90 75.40 70.90 47.20 42.803.60 3.70 4.20 4.10 11.10 13.70 13.20 13.30 16.10 13.40

150.70 141.10 182.00 170.20 480.50 782.10 646.30 642.30 467.20 421.904.70 4.20 6.10 5.50 14.60 18.80 17.40 16.80 11.30 10.000.70 0.70 0.80 0.70 7.60 7.50 5.50 5.60 2.90 2.60

21.60 24.90 23.10 23.30 60.60 84.50 37.40 39.40 50.00 43.208.60 8.30 10.50 10.20 75.30 81.20 62.40 63.10 37.10 36.40

13.00 14.00 14.00 14.00 1.00 1.00 1.00 1.00 2.00 2.002.80 1.20 0.80 0.70 0.90 1.20 1.20

54.70 42.50 48.40 55.30 52.00 71.00 30.80 21.00 49.00 11.80122.00 19.00 119.00 107.00 17.00 13.0032.20 32.40 37.00 40.00 84.00 101.30 71.90 73.00 127.70 95.4062.90 68.50 83.50 80.00 191.80 212.10 151.80 144.80 232.40 176.907.39 7.70 9.00 9.40 19.95 22.32 16.08 16.01 25.26 19.46

29.90 30.30 34.90 35.30 63.60 70.90 53.40 54.60 86.70 64.504.98 5.10 5.90 6.10 12.10 14.40 8.45 9.27 13.53 9.991.06 1.00 1.20 1.30 0.19 0.22 0.37 0.37 1.09 0.964.02 4.00 4.30 4.40 8.80 11.54 5.77 6.61 9.01 7.070.66 0.70 0.80 0.80 1.79 2.32 1.17 1.27 1.67 1.253.81 3.80 3.90 3.80 10.11 13.30 6.21 7.03 8.46 6.970.72 0.80 0.70 0.80 2.15 2.80 1.25 1.34 1.56 1.332.13 2.50 2.20 2.10 6.75 8.87 4.04 4.29 4.74 4.010.34 0.40 0.40 0.40 1.07 1.37 0.70 0.72 0.80 0.622.09 2.10 2.20 1.90 6.83 9.01 4.50 4.89 5.07 4.330.31 0.30 0.30 0.30 1.13 1.42 0.71 0.77 0.80 0.70

199C. Helvacı et al. / Journal of Volcanology and Geothermal Research 185 (2009) 181–202

Table A1 (continued)

Sample L-52 1/220 L-14 IZ-115? IZ-117? SK-9 KB-6 L-6 L-7 L-9

East 0478980 0470980 0483044 0470949 0458590 0458590 0458590 0458590 0457120

North 4245490 4232125 4238199 4232100 4235851 4235851 4235851 4235851 4233554

Unit: (UV) (OB) (OB) (OB) (OB) (UI) (UI) (UI) (UI) (UI)

SiO2 67.70 49.40 47.18 47.90 48.42 63.37 65.47 59.06 51.26 57.46Al2O3 16.77 18.16 15.06 16.57 17.94 16.92 16.01 17.97 13.59 17.93Fe2O3 1.69 8.68 9.62 3.18 3.67 4.05 3.54 6.09 6.58 6.25MgO 0.33 5.35 9.43 5.63 5.58 1.91 1.56 3.16 6.32 3.12CaO 1.22 9.36 10.54 9.92 9.49 4.28 3.43 6.31 11.49 6.86Na2O 4.95 3.22 2.60 2.74 3.13 3.46 3.99 3.68 1.19 3.57K2O 5.60 2.22 1.40 1.64 2.02 3.19 4.00 2.47 6.76 2.52TiO2 0.40 1.35 1.37 1.49 1.43 0.55 0.44 0.73 0.59 0.84P2O5 0.14 0.35 0.30 0.16 0.13 0.18 0.14 0.17 0.09 0.26MnO 0.02 0.13 0.14 0.29 0.32 0.04 0.06 0.09 0.11 0.08LOI 1.10 1.50 2.20 2.90 1.32 1.90 1.20 0.20 2.10 1.00Cs 4.90 3.40 101.00 5.10 6.00 4.30 5.90 4.40Rb 196.70 80.80 45.70 52.00 52.00 125.60 182.30 83.90 371.70 81.30Ba 322.30 585.10 396.70 530.00 530.00 943.90 899.80 747.90 541.10 863.40Sr 157.80 639.80 535.20 727.00 727.00 455.70 364.80 506.30 519.40 565.70Pb 5.40 9.70 6.70 8.90 5.70 3.30 2.50 3.60Th 40.90 10.80 8.10 13.00 3.00 20.10 23.70 12.80 7.50 9.80U 7.90 3.10 1.40 1.00 8.00 4.60 4.00 3.40 2.10 2.80Zr 399.60 163.20 111.70 145.00 178.00 164.30 161.20 139.30 95.40 162.40Hf 9.90 4.90 3.10 5.40 5.80 3.90 3.60 3.90Ta 2.70 1.10 0.70 1.00 1.20 0.70 0.50 0.80Y 37.50 27.40 24.00 30.00 25.00 22.70 18.00 22.80 23.40 25.40Nb 36.50 13.80 10.00 12.00 13.00 10.90 11.00 7.20 4.50 10.50Sc 3.00 20.00 29.00 11.00 9.00 15.00 27.00 17.00Ni 1.50 15.00 122.60 15.00 17.00 6.20 28.90 13.60Co 20.80 43.90 65.70 34.00 27.00 53.60 56.60 58.40 51.60 63.00V 20.00 195.00 222.00 238.00 179.00 97.00 72.00 134.00 163.00 139.00La 81.90 34.90 28.40 43.00 38.00 37.80 36.20 31.70 21.40 31.80Ce 136.00 78.60 60.40 51.00 61.00 83.50 76.30 62.40 44.50 61.20Pr 15.32 8.76 6.92 8.80 7.60 7.06 5.48 7.32Nd 52.50 35.20 27.80 35.80 26.70 26.00 21.80 28.50Sm 7.98 6.80 5.70 5.80 4.00 4.60 4.50 5.20Eu 0.95 1.89 1.78 1.20 0.90 1.21 1.05 1.34Gd 6.25 5.90 5.21 4.60 3.40 3.93 4.11 4.39Tb 1.09 0.87 0.86 0.70 0.60 0.66 0.69 0.75Dy 5.85 4.46 4.19 3.70 2.70 3.80 3.95 3.66Ho 1.14 0.89 0.86 0.70 0.50 0.77 0.77 0.85Er 3.49 2.60 2.29 2.20 1.60 2.25 2.17 2.59Tm 0.56 0.38 0.33 0.30 0.30 0.31 0.38 0.41Yb 3.36 2.32 1.99 2.00 1.60 2.29 1.98 2.41Lu 0.57 0.36 0.30 0.30 0.20 0.35 0.30 0.42

200 C. Helvacı et al. / Journal of Volcanology and Geothermal Research 185 (2009) 181–202

201C. Helvacı et al. / Journal of Volcanology and Geothermal Research 185 (2009) 181–202

References

Agostini, S., Doglioni, C., Innocenti, F., Manetti, P., Savaşçın, M.Y., Tonarini, S., 2005.Tertiary high-Mg volcanic rocks from Western Anatolia and their geodynamicsignificance fort he evolution of the Aegean area. In: Fytikas, M., Vougioukalakis, G.E. (Eds.), The South Aegean Active Volcanic Arc. Elsevier, Amsterdam, pp. 345–361.

Akay, E., Erdoğan, B., 2004. Evolution of Neogene calc-alkaline to alkaline volcanism inthe Aliağa–Foça region (Western Anatolia, Turkey). J. Asian Earth Sci. 24, 367–387.

Aldanmaz, E., Pearce, J.A., Thirwall, M.F., Mitchell, J.G., 2000. Petrogenetic evolution oflate Cenozoic, post-collision volcanism in western Anatolia, Turkey. J. Volcanol.Geotherm. Res. 102, 67–95.

Aldanmaz, E., Gourgaud, A., Kaymakçı, N., 2005. Constraints on the composition andthermal structure of the upper mantle beneath NW Turkey: evidence from mantlexenoliths and alkali primary melts. J. Geodyn. 39, 277–316.

Aldanmaz, E., Köprübaşı, N., Gürer, Ö.F., Kaymakçı, N., Gourgaud, A., 2006. Geochemicalconstraints on the Cenozoic, OIB-type alkaline volcanic rocks of NW Turkey:implications for mantle sources and melting processes. Lithos 86, 50–76.

Alther, R., Siebel, W., 2002. I-type plutonism in a continental back-arc setting: Miocenegranitoids and monzonites from the central Aegean Sea, Greece. Contrib. Miner.Petrol. 143, 397–415.

Altunkaynak, Ş., Yılmaz, Y., 1998. The Mount Kozak magmatic complex, WesternAnatolia. J. Volcanol. Geotherm. Res. 85, 211–233.

Altunkaynak, Ş., Dilek, Y., 2006. Timing and nature of post collisional volcanism inWestern Anatolia and geodynamic implications. In: Dilek, Y., Pavlides, S. (Eds.),Post-Collisional Tectonics and Magmatism in the Mediterranean Region and Asia.Geol. Soc. Am. Special Papers, vol. 409, pp. 321–351.

Altunkaynak, Ş., Genç, Ş.C., 2008. Petrogenesis and time-progressive evolution of theCenozoic continental volcanism in the Biga Peninsula, NWAnatolia (Turkey). Lithos102, 316–340.

Barry, T.L., Saunders, A.D., Kempton, P.D., Windley, B.F., Pringle, M.S., Dorjnamjaa, D.,Saandar, S., 2003. Petrogenesis of Cenozoic basalts fromMongolia: evidence for therole of asthenospheric versus metasomatized lithospheric mantle sources. J. Petrol.44, 5–91.

Borsi, J., Ferrara, G., Innocenti, F., Mazzuoli, R., 1972. Geochronology and petrology ofrecent volcanics in the eastern Aegean Sea (West Anatolia and Lesbos Island). Bull.Volcanol. 36, 473–496.

Boztuğ, D., Harlavan, Y., Jonckheere, R., Can, İ., Sarı, R., 2009. Geochemistry and K–Arcooling ages of the Ilıca, Çataldağ (Balıkesir) and Kozak (İzmir) granitoids, westAnatolia, Turkey. Geol. J. 44, 79–103.

Buettner, A., Kleinhanns, I.C., Rufer, D., Hunziker, J.C., Villa, I.M., 2005. Magmageneration at the easternmost section of the Hellenic arc: Hf, Nd, Pb and Sr isotopegeochemistry of Nisyros and Yali volcanoes (Greece). Lithos 83, 29–46.

Çakmakoğlu, A., Bilgin, Z.R., 2006. Pre-Neogene stratigraphy of the Karaburun Peninsula(W of İzmir Turkey). Bull. Miner. Res. Explor. Inst. Turk. 132, 33–61.

Castillo, P.R., 2006. An overview of adakite petrogenesis, Chinese Sci. Bull. 51, 257–268.Delaloye, M., Bingöl, E., 2000. Granitoids from western and northwestern Anatolia:

geochemistry and modelling of geodynamic evolution. Int. Geol. Rev. 42,241–268.

Conceição, R.V., Green, D.H., 2004. Derivation of potassic (shoshonitic) magmas bydecompression melting of phlogopite+pargasite lherzolite. Lithos 72, 209–229.

De Paolo, D.J., 1981. Trace element and isotopic effects of combined wall-rockassimilation and fractional crystallisation. Earth Planet. Sci. Lett. 53, 189–202.

Emre, T., Sözbilir, H., 2005. Küçük menderes grabeni doğu ucundaki andezitlerin(Başova–Kiraz/İzmir) jeolojisi, petrografisi ve jeokimyası [Geology, petrographyand geochemistry of andesites in the eastern margin of the Küçük MenderesGraben (Başova–Kiraz/İzmir)]. Bull. Miner. Res. Explor. Inst. Turk. 131, 1–19.

Ercan, T., Satir, M., Kreuzer, H., Türkecan, A., Günay, E., Çevikbaş, A., Ateş, M., Can, B., 1985.Batı Anadolu Senozoyik volkanitlerine ait yeni kimyasal, izotopik ve radyometrikverilerinyorumu [Interpretation of thenewgeochemical, isotopic and radiometric agedata from thewestern Anatolia Cenozoic volcanics]. Bull. Geol. Soc. Turk. 28, 121–136.

Ercan, T., Akat, U., Günay, E., Savaşçın, Y., 1986. Söke–Selçuk–Kuşadası dolaylarınınjeolojisi ve volkanik kayaçların petrokimyasal özellikleri [Geology of the areaaround Söke–Selçuk–Kuşadası and petrochemical characteristics of volcanic rocks].Bull. Miner. Res. Explor. Inst. Turk. 105/106, 15–38.

Ercan, T., Satır, T., Steinitz, G., Dora, A., Sarıfakıoğlu, E., Adis, C., Walter, H.-J., Yıldırım, T.,1995. Biga Yarimadası ile Gökçeada, Bozcaada ve Tavşan Adalarındaki (KB Anadolu)Tersiyer Volkanizmasının Özellikleri [Features of the Tertiary volcanism in the BigaPeninsula and Gökçeada, Bozcaada and Tavşan islands (NWAnatolia)]. Bull. Miner.Res. Explor. Inst. Turk. 177, 55–86.

Ercan, T., Türkecan, A., Guillou, H., Satır, M., Sevin, D., Şaroğlu, F., 1998. Marmara Deniziçevresindeki Tersiyer volkanizmasının özellikleri [Features of Tertiary volcanismaround the Sea of Marmara]. Bull. Miner. Res. Explor. Inst. Turk. 120, 199–221.

Erdoğan, B., 1990. Tectonic relations between İzmir–Ankara Zone and Karaburun Belt.Bull. Miner. Res. Explor. Inst. Turk. 110, 1–15.

Erkül, F.,Helvacı, C., Sözbilir, H., 2005. Evidence for twoepisodesof volcanism in theBigadicborate basin and tectonic implications for western Turkey. Geol. J. 40, 545–570.

Ersoy, Y., Helvacı, C., Sözbilir, H., Erkül, F., Bozkurt, E., 2008. A geochemical approach toNeogene–Quaternary volcanic activity of western Anatolia: an example of episodicbimodal volcanism within the Selendi Basin, Turkey. Chem. Geol. 255, 265–282.

Fytikas, M., Innocenti, F., Manetti, P., Mazzuoli, R., Peccerillo, A., Villari, L., 1984. TertiarytoQuaternary evolution of volcanism in the Aegean region. In: Dixon, J.E., Robertson,A.H.F. (Eds.), TheGeological Evolution of the EasternMediterranean: Geol. Soc. Spec.Publ., vol. 17, pp. 687–699.

Geological Map of Turkey-İzmir (1:500000), 2002. Publication of Mineral Research andExploration Institute of Turkey.

Gessner, K., Ring, U., Johnson, C., Hetzel, R., Passchier, C.W., Güngör, T., 2001. An activebivergent rolling-hinge detachment system; central Menderes metamorphic corecomplex in western Turkey. Geology (Boulder). 29, 611–614.

Güleç, N., 1991. Crust–mantle interaction in western Turkey: implications from Sr and Ndisotope geochemistry of Tertiary and Quaternary volcanics. Geol. Mag. 23, 417–435.

Helvacı, C., 1995. Stratiggraphy, mineralogy, and genesis of the Bigadiç borate deposits,Western Turkey. Econ. Geol. 90, 1237–1260.

Innocenti, F., Agostini, S., Di Vincenzo, G., Doglioni, C., Manetti, P., Savaşçın, M.Y.,Tonarini, S., 2005. Neogene and Quaternary volcanism inWestern Anatolia: magmasources and geodynamic evolution. Mar. Geol. 221, 397–421.

Irvine, N., Baragar, W.R.A., 1971. A guide to chemical classification of the commonvolcanic rocks. Can. J. Earth Sci. 8, 523–548.

Işık, V., Seyitoğlu, G., Çemen, İ., 2003. Ductile–brittle transition along the Alaşehirdetachment fault and its structural relationship with the Simav detachment fault,Menderes Massif, western Turkey. Tectonophysics 374, 1–18.

Jackson, J., McKenzie, D., 1984. Active tectonics of the Alpine–Himalayan Belt betweenwestern Turkey and Pakistan. Geophys. J. R. Astron. Soc. 77, 185–264.

Justet, L., Spell, T.L., 2001. Effusive eruptions from a large shallow magma chamber: theBearhead Rhyolite, Jemez Volcanic Field, NewMexico. J. Volcanol. Geotherm. Res. 107,241–264.

Karacık, Z., 2006. Stratigraphy and volcanology of the Türbükü volcanics: products of astratovolcano in the Bodrum Peninsula, SW Anatolia. Geol. Mag. 41, 145–162.

Karacık, Z., Yılmaz, Y., Pearce, J.A., 2007a. The Dikili–Çandarlı volcanics,Western Turkey;magmatic interactions as recorded by petrographic and geochemical features. Turk.J. Eart. Sci. 16, 493–522.

Karacık, Z., Yılmaz, Y., Pearce, J.A., Ece, Ö.İ., 2007b. Petrochemistry of the south Marmaragranitoids, northwest Anatolia, Turkey. Int. J. Earth Sci. 97, 1181–1200.

Kaymakçı, N., Aldanmaz, E., Langereis, C., Spell, T.L., Gürer, Ö.F., Zanetti, K.A., 2007. LateMiocene transcurrent tectonics in NW Turkey: evidence from palaeomagnetismand 40Ar–39Ar dating of alkaline volcanic rocks. Geol. Mag. 144, 379–391.

Kinzler, R.J., 1997. Melting of mantle peridotite at pressures approaching the spinel togarnet transition: application to mid-ocean ridge basalt petrogenesis. J. Geophys.Res. 102, 853–874.

Koçyiğit, A., Yusufoğlu, H., Bozkurt, E., 1999. Evidence from the Gediz graben for episodictwo-stage extension in western Turkey. J. Geol. Soc. 156, 605–616.

Kurt, H., Arslan, M., 2001. Bodrum (GB Anadolu) volkanik kayalarının jeokimyasal vepetrolojik özellikleri: fraksiyonel kristalleşme, magma karışması ve asimilasyonailişkin bulgular [Geochemical and petrological characteristics of the Bodrum (SWAnatolia) volcanic rocks: evidence for fractional crystallization, magma mixing andassimilation. Yerbilimleri 23, 15–32.

Le Maitre, R.W., 2002. Igneous rocks: a classification and glossary of terms:recommendations of the International Union of Geological Sciences. Subcommis-sion on the Systematics of Igneous Rocks. Cambridge University Press. 236 p.

Lips, A.L.W., Cassard, D., Sözbilir, H., Yılmaz, H.,Wijbrans, J.R., 2001.Multistage exhumationof the Menderes Massif, western Anatolia (Turkey). Int. J. Earth Sci. 89, 781–792.

Okay, A.İ., Tüysüz, O., 1999. Tethyan sutures of northern Turkey. In The MediterraneanBasins: Tertiary extension within the Alpine orogen. In: Durand, B., Jolivet, L.,Horváth, F., Séranne, M. (Eds.), Geol. Soc. London Spec. Publ. N.156, pp. 475–515.

Okay, A.İ., Satır, M., 2000. Coeval plutonism and metamorphism in a latest Oligocenemetamorphic core complex in northwest Turkey. Geol. Mag. 5, 495–516.

Pearce, J.A., 1983. Role of the sub-continental lithosphere in magma genesis at activecontinental margins. In: Hawkesworth, C.J., Norry, M.J. (Eds.), Continental Basaltsand Mantle Xenoliths. Shiva Publishing, Nantwich, pp. 230–249.

Pearce, J.A., Parkinson, I.J., 1993. Trace element models for mantle melting: application tovolcanic arc petrogenesis. In: Prichard, H., et al. (Eds.), Magmatism and Plate Tectonics(I.G. Gass Memorial Volume): Geol. Soc. London Spec. Publ., vol. 76, pp. 373–403.

Pearce, J.A., Ernewein, M., Bloomer, S.H., Parson, L.M., Murton, B.J., Johnson, L.E., 1995.Geochemistry of Lau Basin volcanic rocks: influence of ridge segmentation and arcproximity. In: Smellie, J.L. (Ed.), Volcanism Associatedwith Extension at ConsumingPlate Margins. Geol. Soc. Spec. Publ., pp. 53–75.

Pecerillo, A., Taylor, S.R., 1976. Geochemistry of Eocene calc-alkaline volcanic rocks fromthe Kastamonu area, northern Turkey. Contrib. Mineral. Petrol. 58, 63–81.

Pe-Piper, G., Piper, D.J.W., 2001. Late Cenozoic, post-collisional Aegean igneous rocks:Nd, Pb and Sr isotopic constraints on petrogenetic and tectonic models, Geol. Mag.138, 653–668.

Pe-Piper, G., Piper, D.J.W., 2007a. Neogene backarc volcanism of the Aegean: newinsights into the relationship between magmatism and tectonics. In: Beccaluva, L.,Bianchini, G., Wilson, M. (Eds.), Cenozoic Volcanism in the Mediterranean Area:Geol. Soc. Am. Spec. Paper., vol. 418, pp. 17–31.

Pe-Piper, G., Piper, D.J.W., 2007b. Late Miocene igneous rocks of Samos: the role oftectonism in petrogenesis in the southeastern Aegean. In: Taymaz, T., Yılmaz, Y.,Dilek, Y. (Eds.), The Geodynamics of the Aegean and Anatolia: Geol. Soc. London,Spec. Publ., vol. 291, pp. 75–97.

Pe-Piper, G., Piper, D.J.W., Kotopouli, C.N., Panagos, A.G., 1995. Neogene volcanoes ofChios, Greece. The relative importance of subduction and back-arc extension: Geol.Soc. London Spec. Publ., 81, pp. 213–232.

Pe-Piper, G., Piper, D.J.W., Koukouvelas, I., Dolansky, L.M., Kokkalas, S., 2009.Postorogenic shoshonitic rocks and their origin by melting underplated basalts:the Miocene of Limnos, Greece. Bull. Geol. Soc. Am. 121, 39–54.

Ring, U., Susanne, L., Matthias, B., 1999. Structural analysis of a complex nappe sequenceand late orogenic basins from the Agean Island of Samos, Greece. J. Struct. Geol. 21,1575–1601.

Robert, U., Foden, J., Varne, R., 1992. The Dodecanese Province, SE Aegean: a model fortectonic control on potassic magmatism. Lithos. 28, 241–260.

Robertson, A.H.F., Pickett, E.A., 2000. Paleozoic–Early Tertiary Tethyan evolution ofmélanges, rift and passive margin units in the Karaburun Peninsula (western

202 C. Helvacı et al. / Journal of Volcanology and Geothermal Research 185 (2009) 181–202

Turkey) and Chios Island (Greece). In: Bozkurt, E., Winchester, J.A., Piper, J.D.A.(Eds.), Geol. Soc. London, Spec. Publ., vol. 173. Tectonics and Magmatism in Turkeyand Surrounding Area, pp. 43–82.

Rollinson, H.R., 1993. Using Geochemical Data: Evaluation, Presentation, Interpretation,Longman Scientific & Technical. John Wiley & Sons, UK. 352 pp.

Salters, V., Stracke, A., 2004. Composition of the depleted mantle. Geochem. Geophys.Geosyst. (ISSN: 1525-2027) 5 (5). doi:10.1029/2003GC000597.

Şengör, A.M.C., Görür, N., Şaroğlu, F., 1985. Strike–slip faulting and related basinformation in zones of tectonic escape: Turkey as a case study. In: Biddle, K., Christie-Blick, N. (Eds.), Strike–Slip Deformation, Basin Formation and Sedimentation: Soc.Econom. Paleontolo. Min. Spec. Publ., vol. 37, pp. 227–264.

Seyitoğlu, G., Scott, B.C., 1991. Late Cenozoic crustal extension and basin formation inwest Turkey. Geol. Mag. 128, 155–166.

Seyitoğlu, G., Scott, B.C., 1992. Late Cenozoic volcanic evolution of the NE Aegean region.J. Volcanol. Geotherm. Res. 54, 157–176.

Seymour, K.St., Tsikouras, V., Kotopouli, K., Hatzipanayiotou, K., Pe-Piper, G., 1996. Awindow to the operation of microplate tectonics in the Tethys Ocean: thegeochemistry of the Samothrace granite, Aegean Sea. Miner. Petrol. 56, 251–272.

Shaw, D.M., 1970. Trace element fractionation during anatexis. Geochim. Cosmochim.Acta 34, 237–243.

Spell, T.L., McDougall, I., 2003. Characterization and calibration of 40Ar/39Ar datingstandards. Chem. Geol. 198, 189–211.

Staudacher, T., Jessberger, E.K., Dörflinger, D., Kiko, J., 1978. A refined ultrahigh-vacuumfurnace for rare gas analysis. J. Phys. E J. Sci. Instrum. 11, 781–784.

Sun, S.S., McDonough, W.F., 1989. Chemical and isotopic systematics of oceanic basalts:implications formantle compositions andprocesses. In: Saunders, A.D., Norry,M.J. (Eds.),Magmatism in the Oceanic Basin. Geol. Soc. London Spec. Publ, London, pp. 313–345.

Sözbilir, H., 2001. Extensional tectonics and the geometry of related macroscopicstructures: field evidence from the Gediz Detachment, Western Turkey. Turk. J. Eart.Sci. 10, 51–67.

Sözbilir, H., 2005. Oligo–Miocene extension in the Lycian orogen: evidence from theLycian molasse basin, SW Turkey. Geodin. Acta 18, 257–284.

Tatar-Erkül, S., Sözbilir, H., Erkül, F., Helvacı, C., Ersoy, E.Y., Sümer, Ö., 2008. Geochemistryof I-type granitoids in the Karaburun Peninsula, West Turkey: Evidence for Triassiccontinental arc magmatism following closure of the Palaeotethys. Isl. Arc. 17,394–418.

Taylor, S.R., McLennan, S.M., 1995. The geochemical evolution of the continental crust.Rev. Geophys. 33, 241–265.

Türkecan, A., Ercan, T., Sevin, D., 1998. Karaburun Yarımadası'nın Neojen volkanizması[Neogene volcanism in Karaburun Peninsula]. Report no: 10185, General Directo-rate of Mineral research and Exploration, Ankara.

Turner, S.P., Platt, J.P., George, R.M.M., Kelley, S.P., Pearson, D.G., Nowell, D.G., 1999.Magmatism associated with orogenic collapse of the Betic–Alboran domain, SE Spain.J. Petrol. 40, 1011–1036.

Walter, M.J., 1998. Melting of garnet peridotite and the origin of komatiite and depletedlithosphere. J. Petrol. 39, 29–60.

Wang, K.-L., Chung, S.-L., O'Reilly, S.Y., Sun, S.-S., Shinjo, R., Chen, C.-H., 2004. Geochemicalconstraints for the genesis of post-collisional magmatism and the geodynamicevolution of the Northern Taiwan Region. J. Petrol. 45, 975–1011.

Workman, R.K., Hart, S.R., 2005. Major and trace element composition of the depletedMORB mantle (DMM). Earth Planet. Sci. Lett. 231, 53–72.

Yılmaz, Y., 1989. An approach to the origin of young volcanic rocks of Western Turkey.In: Şengör, A.M.C. (Ed.), Tectonic evolution of the Tethyan Region. Kluwer Ac. Publ.,pp. 159–189.

Yılmaz, Y., 1990. Comparison of young volcanic associations of western and easternAnatolia formedunder a compressional regime: a review. J. Volcanol. Geotherm. Res.44, 1–19.

Yılmaz, Y., Genç, Ş.C., Karacık, Z., Altunkaynak, Ş., 2001. Two contrasting magmaticassociations of NWAnatolia and their tectonic significance. J. Geodyn. 31, 243–271.