Petrology of Quaternary volcanic rocks and related plutonic xenoliths from Gölcük volcano, Isparta Angle, Turkey: Origin and evolution of the high-K alkaline series

Page 1

Petrology of Quaternary volcanic rocks and related plutonic xenolithsfrom Gölcük volcano, Isparta Angle, Turkey: Origin and evolutionof the high-K alkaline series

Bernard Platevoet a, Ömer Elitok b, Hervé Guillou c, Jacques-Marie Bardintzeff a,d,⇑, Fuzuli Yagmurlu b,Sébastien Nomade c, André Poisson a, Catherine Deniel e,f,g, Nevzat Özgür b

aUniv Paris-Sud, Laboratoire GEOPS, UMR CNRS 8148, Bât. 504, F-91405 Orsay, Franceb Süleyman Demirel University, Department of Geothermal Energy et Mineral Ressources, Merkez Campus 32260, Isparta, Turkeyc Laboratoire des Sciences du Climat et de l’Environnement (UMR 8212 CEA-CNRS-UVSQ), L’Orme des Merisiers, F-91191 Gif-sur-Yvette, FrancedUniv Cergy-Pontoise, ESPE, F-95000 Cergy-Pontoise, Francee Laboratoire Magmas et Volcans, Clermont Université, Univ Blaise Pascal, BP 10448, F-63000 Clermont-Ferrand, FrancefCNRS, UMR 6524, LMV, F-63038 Clermont-Ferrand, Franceg IRD, R 163, LMV, F-63038 Clermont-Ferrand, France

a r t i c l e i n f o

Article history:

Received 14 April 2014Received in revised form 4 June 2014Accepted 9 June 2014Available online 19 June 2014

Keywords:

Alkaline magmatismXenolithMantle sourceCarbonatiteMetasomatized lithosphereTurkey

a b s t r a c t

The Quaternary volcanism of Isparta, south-western Anatolia, belongs to the post-collisional alkali-potassic to ultrapotassic magmatism, active since Miocene, from Afyon to Isparta. In the so-called IspartaAngle, the magmatism is contemporaneous with the Aegean extensional regime initiated during the LateMiocene and active throughout the Pliocene and Quaternary. The Gölcük volcano-forming stages consistof three main eruptive cycles: Cycle I comprising 200 m-thick pyroclastic flow deposits; Cycle II consist-ing of tephriphonolitic lava dome-flows extruded throughout the caldera; and Cycle III characterized bytuff-ring deposits related to the last phreatoplinian events. These late explosive events sampled plutonicxenoliths that allow to better constrain magma fractionation processes that operated at depth in themagma chamber. Magma evolution was first controlled by accumulation of clinopyroxene, phlogopiteand apatite, then by phlogopite, amphibole and feldspars, with apatite, magnetite, titanite and zirconas accessories. Crystallization of clinopyroxene, phlogopite and amphibole probably controlled the sil-ica-saturation trend of the whole series and faithfully reflect intensive H2O variations in the magma thatwere responsible of explosive cyclic events. The parental magma may have had a lamprophyric-tephriticcomposition. Trace element and isotope ratios indicate a prevalent asthenospheric source versus litho-spheric one. Geochemical features, such as strong enrichment of LILE, REE, HFSE in the Gölcük magmapoint to the involvement of a asthenospheric OIB-type melt with a possible carbonatitic component, thatinteracted with remnants of the delaminated lithosphere during upwelling.

Ó 2014 Elsevier Ltd. All rights reserved.

1. Introduction

The Quaternary Gölcük volcano is located in the Isparta districtin southwestern Anatolia (Lefèvre et al., 1983; Platevoet et al.,2008; Fig. 1). It has a long history beginning about 2 Ma ago withwhat appears to have been a cyclic explosive crisis (Alici et al.,1998; Nemec et al., 1998; Platevoet et al., 2008; Elitok et al.,2010). Located near a large city, it remains a potential major

volcanic hazard for the Isparta district. Its activity was characterizedby mainly explosive eruptions. Its last phreatomagmatic activitydelivered many plutonic xenoliths in the tephra deposits. Thesexenoliths include sedimentary basement xenoliths and a largevariety of plutonic rocks ranging from ultramafic to felsic compo-sitions. Since Lacroix (1893), geologists have paid special attentionto xenoliths found in volcanic rocks, because xenolith studies maygive important evidences about sources of magmas and magmaticprocesses occurring in the chamber such as mineral segregation,fractional crystallization and subsequent geochemical evolutionof magma. The main topic of this paper consists of petrologicaland geochemical studies of plutonic xenoliths, coupled with thoseof their volcanic hosts. This may give us a better understanding of

http://dx.doi.org/10.1016/j.jseaes.2014.06.0121367-9120/Ó 2014 Elsevier Ltd. All rights reserved.

⇑ Corresponding author at: Univ Paris-Sud, Laboratoire GEOPS, UMR CNRS 8148,Bât. 504, F-91405 Orsay, France. Tel.: +33 1 69 15 67 44, mobile: +33 6 26 39 23 17;fax: +33 1 69 15 49 05.

E-mail address: [email protected] (J.-M. Bardintzeff).

Journal of Asian Earth Sciences 92 (2014) 53–76

Contents lists available at ScienceDirect

Journal of Asian Earth Sciences

journal homepage: www.elsevier .com/locate / jseaes

Page 2

the processes occurring at depth in the underneath chamber of theGölcük volcano and the manner in which intermittent explosivecycles are triggered.

2. Geological setting

The Isparta volcanism, SW Turkey, belongs to the post-colli-sional alkali potassic-ultrapotassic magmatism occurring along anorth–south axis from Afyon to the Isparta-Bucak area (Keller,1983; Yagmurlu et al., 1997; Savasçin and Oyman, 1998). TheIsparta volcanic district is located at the apex of the so-called‘‘Isparta Angle’’ (see the tectonic scheme in Fig. 1). This area corre-sponds to the junction of the Hellenic and Cyprus arcs, by completeclosure of both the Northern Neotethys, located North of the Ana-tolian micro-plate, and the Southern Pamphylian Basin of the

Southern Neotethys (see Poisson et al., 2003; Prelevic et al.,2010b and references therein). Associated with an extensionalregime of the Aegean district (Alçiçek et al., 2013), the post-collisional potassic-ultrapotassic magmatism started during theMiocene in the northern district of Afyon and continued duringthe Pliocene till the end of the Pleistocene in Isparta and Burdurareas (Keller, 1983; Lefèvre et al., 1983; Guillou, 1987; Özgür et al.,1990; Seyitoglu et al., 1997; Yagmurlu et al., 1997; Savasçin andOyman, 1998; Nemec et al., 1998; Alici et al., 1998; Hildenbrandet al., 1999; Francalanci et al., 2000; Akal, 2003, 2008; Coban,2005; Prelevic et al., 2008, 2010a, 2012; Ersoy et al., 2010; Dilekand Altunkaynak, 2010; Akal et al., 2013; Ersoy and Palmer, 2013).

Numerous Pliocene dykes and protrusions outcropping in theIsparta and Bucak areas (Lefèvre et al., 1983; Coban and Flower,2006) are the result of the Pliocene magmatic activity whose sur-face manifestations and volcanic products have been now eroded

Fig. 1. Geological map of the central and south-eastern parts of the Gölcük volcano (modified from Platevoet et al., 2008). Tectonic scheme modified after Kumral et al.(2006). See legend and text for more details.

54 B. Platevoet et al. / Journal of Asian Earth Sciences 92 (2014) 53–76

Page 3

away (Cengiz et al., 2005). 40K/40Ar ages on lavas by Lefèvre et al.(1983), Nemec et al. (1998), Platevoet et al. (2008), Prelevic et al.(2012) showed that the first alkaline-potassic volcanism occurredbetween 4.7 and 2.8 Ma. The Gölcük volcano is the only morpho-logically well-preserved volcano in this area. It has a long history,a largely explosive activity and very few lava flows (Alici et al.,1998; Nemec et al., 1998; Platevoet et al., 2008; Elitok et al.,2010). The explosive activity starts at the end of Pliocene. It isthe youngest manifestation of alkaline magmatic activity that con-tinues in this area during the Pleistocene until present (Platevoetet al., 2008; Mouillard, 2011). Built close to the large city of Isparta,and presently at rest, this volcano remains a potential major volca-nic hazard for the entire region.

3. Analytical methods

Analyses of mineral phases were performed with a CamecaSX100 electron microprobe at ‘‘Camparis’’, University ‘‘Pierre etMarie Curie’’, Paris, France (tension: 15KV, current intensity:10nA, counting time: 5s for both peaks and backgrounds; the beamwas defocused during analysis of glass. Ka lines were used. Stan-dards were diopside for Si, Ca and Mg, Fe2O3 for Fe, MnTiO3 forTi and Mn, Cr2O3 for Cr, albite for Na, orthoclase for K and Al.

Some trace and rare earth elements (REE) were analyzed inpyroxene using the LA-ICP-MS technique at Toulouse, France.Whole rock analyses were performed at CRPG Nancy for 54 ele-ments comprising major elements (by the ICP-OES method), traceelements and REE (by the ICP-MS method). Additional technicaldetails can be found in the CRPG website (http://www.crpg.cnrs-nancy.fr). Isotope data were taken from Elitok et al. (2010).Isotope analyses were determined at the Institute of Geosciencesat Tübingen University (Germany) by thermal ionization massspectrometry using a Finnigan MAT 262 mass spectrometer. The87Sr/86Sr isotope ratios were normalized to 86Sr/88Sr = 0.1194 andthe 143Nd/144Nd isotope ratios to 146Nd/144Nd = 0.7219. Over theperiod of analyses, the La Jolla Nd-standard yielded a valueof 0.511828 ± 0.000009 (n = 16, reference value 0.511850) andthe NBS 987 Sr standard yielded a 87Sr/86Sr ratio of0.710245 ± 0.000008 (n = 12, reference value 0.710248). Two addi-tional samples were analyzed for Sr and Nd isotopes at UMR 6524CNRS (Clermont-Ferrand, France), using the procedure described inVlastélic et al. (2009). These data were normalized to the abovestandard values. Owing to the very young ages of the volcanic sam-ples (mainly less than 2 Ma for Gölcük volcano), the measured Ndand Pb isotopic ratios closely approximate the initial ratios, so agecorrections are negligible for Nd and Pb.

4. Eruptive chronology of the Gölcük volcano

Closely to South of Isparta City, the Gölcük volcanic edifice(Fig. 1) partially covers the Lycian nappes, older lavas and volcan-oclastic deposits, as well as trachyandesitic protrusions belongingto the westernmost part of the Pliocene volcanic cycle (Aliciet al., 1998; Platevoet et al., 2008; Elitok et al., 2010). The volcanicedifice corresponds to a maar crater (2.5 km wide across) partlyfilled by a lake and surrounded by a well-preserved tuff cone.The volcanic edifice is composed mainly of tephra deposits(>90 vol.%) including pyroclastic outflows, airfalls and base surgeswith minor lahars and subsidiary lava flows, dykes and domes(Lefèvre et al., 1983; Özgür et al., 1990; Nemec et al., 1998; Aliciet al., 1998; Platevoet et al., 2006, 2008). These products wereemplaced as part of a complex eruptive sequence consisting of atleast three successive cycles herein designated eruptive Cycles I,II, and III (Platevoet et al., 2008).

4.1. Cycle I: Main pyroclastic flow deposits (MPFD)

The explosive activity started about 2,180 ± 44 ka ago (multi-modal age by single feldspar 40Ar/39Ar method, Mouillard, 2011)by huge pyroclastic flows emplacing within the lacustrines of Bur-dur lake. The main pyroclastic flow deposits (MPFD) correspond toa 200 m-thick pile of pyroclastic flow deposits previouslydescribed by Alici et al. (1998) and Platevoet et al. (2008). Multi-modal (same method) ages obtained on MPFD units are scatteredduring Pleistocene, between 440 ± 12 ka and 148 ± 21 ka. Theseages are younger than the regional Pliocene activity (between 4.0and 2.8 Ma) recognized before (Lefèvre et al., 1983; Nemec et al.,1998; Platevoet et al., 2008; Mouillard, 2011).

4.2. Cycle II: Tephriphonolitic lava flow-domes and dykes

Cycle II lava flows outcrop along the north-eastern, south-eastern and western walls of the maar crater (Fig. 1). The lavaflow-domes and dykes consist of massive tephriphonolite lavaswhich have given ages between 115 ± 3 ka and 62 ± 2 ka(40K/40Ar method on lavas, Platevoet et al., 2008), they are slightlyyounger than Cycle I MPFD tephras.

4.3. Cycle III: Tuff-ring deposits and trachyte domes

A second pyroclastic (with the last trachyte domes) episodeproduced a large tuff cone centred on the 2.5 km-wide maar craterand extending beyond the depression rim over the flanks of thevolcano (Fig. 1). The tuff cone has been built between72.7 ± 4.7 ka (multi-modal age on single feldspar 40Ar/39Armethod), and 24 ± 2 ka (by 40K/40Ar method), age of the last tra-chyte dome (Platevoet et al., 2008). A related cryptic diatrem isinferred underneath the Gölcük Lake (Alici et al., 1998). The 100–150 m thick tuff ring consists of a sequence of three distinct pum-ice airfall members separated by paleosol horizons, base-surgedeposits and/or erosion features. Pumice lapilli are always mingledwith very abundant lithoclasts of sedimentary and volcanic origin,including numerous ultramafic, mafic, and felsic plutonic blocks(5–30 cm in size).

5. Petrology and mineralogy of lavas and juvenile pyroclasts

5.1. Mafic dykes

Mafic dykes of lamprophyric affinity, intruding the basementand the Pliocene eruptive formations around the volcano, are theonly visible product of mafic magma generated just before theupper Pliocene–Pleistocene felsic explosive activity of the Gölcükvolcano. The absolute age of these dykes remains unknown. Theirtextures are porphyric with altered olivine phenocrysts includingchromite crystals, abundant euhedral clinopyroxene with skeletalcore and oscillatory zoned rim (colourless to green), yellowish tobrown euhedral mica (phlogopite) and apatite. The groundmassis composed of tiny green pyroxenes and magnetite in a glassymatrix, while feldspar is absent. The predominance of clinopyrox-ene and phlogopite phenocrysts indicates that these rocks are akinto the alkali minette group (Rock, 1991; Mitchell, 1994; Woolleyet al., 1996). However, alkali feldspars are not been observed inthe groundmass of these rocks.

5.2. Tephriphonolitic and trachyandesitic lava flow-domes (Cycle II)

Lava flow-domes, some dykes, and blocks resulting from thepulverization of former lava flow-domes are of tephriphonolitic

B. Platevoet et al. / Journal of Asian Earth Sciences 92 (2014) 53–76 55

Page 4

or trachyandesitic compositions. They display microlitic to latitictextures in which feldspars are predominant in the groundmass.Phenocrysts are green clinopyroxene with complex oscillatory zon-ing, brown mica (frequently totally oxidized), apatite and magne-tite. The groundmass contains sodic plagioclase, anorthoclase tosanidine crystals, clinopyroxene, Fe-oxide, and variable amountsof sodalite-haüyne crystals and glass. Tephriphonolite and trachy-andesite differ by the relative abundances of clinopyroxene andmica phenocrysts. Sodalite is absent in trachyandesite.

5.3. Trachytic domes (Cycle III)

Trachytic lavas of Cycle III exclusively occur in the three domesextruded inside the maar crater. They have a porphyritic microlitictexture with phenocrysts of clinopyroxene, brown mica, rareamphibole, sodic plagioclase and alkali feldspars (anorthoclase tosanidine). Accessories are magnetite, titanite, apatite, and zircon.The groundmass contains alkali feldspar, Fe-oxide and clinopyrox-ene as microlites, with very minor interstitial glass.

5.4. Trachyandesitic and trachytic pyroclasts (Cycle I and III)

Pyroclastic deposits of Cycle I are composed of trachyandesiticpumices. All the pyroclasts are porphyritic, glassy and highly vesic-ulated. Phenocrysts consist of zoned green clinopyroxene, brownamphibole, magnetite, titanite, apatite, mica, rare zircon, plagio-clase and alkali feldspar (anorthoclase to sanidine) frequentlyincluding all other phenocrysts. Some glomeroporphyritic aggre-gates are occasionally present and are mainly formed of alkali feld-spar together with pyroxene and amphibole phenocrysts. Thephenocryst abundance (for each phase) has been determinedunder binocular lens after separation (Table 1). Feldspar, pyroxeneand amphibole are the most frequent phenocrysts than mica, apa-tite, titanite, magnetite and zircon.

Trachytic pyroclasts of the last Cycle III display the same glassyporphyritic texture with similar phenocrysts. However, they con-tain abundant zoned sanidine crystals with anorthoclase coresand rims of oscillatory zoned sanidine. Mica is as common as itis in the late trachytic domes (Table 1) that differ from the trachy-andesitic pumices. Small glomeroporphyritic aggregates (up to5 mm) are common and composed mainly of early mafic minerals(zoned clinopyroxene, amphibole, mica, magnetite and apatite) ina glassy or feldspar-rich matrix.

6. Petrography of plutonic xenoliths

The plutonic xenoliths are found chiefly in the main pumice fallunit of the last explosive Cycle III. Rare plutonic xenoliths are alsoincluded in the youngest trachytic lava domes. Xenoliths belong tothree main petrographic groups: ultramafic pyroxenites, maficmonzonites, and leucocratic monzosyenites. They differ by theirtextures and crystal habits (Fig. 2).

6.1. Ultramafic pyroxenite group

The pyroxenites exhibit a mesocumulate texture (Irvine, 1982)with 70–80 vol.% of euhedral slightly zoned clinopyroxene, euhe-dral to subhedral yellowish-brown phlogopite (10–25 vol.%), anhe-dral to euhedral apatite (1–5%) and interstitial feldspar (5–10%) asubiquitous intercumulus phase. Some of them are incompletelycrystallized and consist of vesiculated glass including small skele-tal crystals of green clinopyroxene, mica and magnetite. Dendriticfeather-like crystals of feldspars (alkali feldspar dominant) oftendevelop and partially replace the glassy matrix (Fig. 2, photos A–C).

6.2. Mafic monzonite group

This is the most variable group in terms of texture and petrog-raphy (Fig. 2, photos D–G). They are more or less mafic according tomafic phase abundance (clinopyroxene and amphibole) relative tofeldspars. Some of them exhibit poikilitic texture with early euhe-dral crystals of zoned clinopyroxene, mica and apatite. Large euhe-dral brown amphibole crystals (5–10 mm in size) commonlyincluding earlier phases (but no feldspars) are very abundant insome xenoliths. The groundmass is composed of poikilitic sodiccm-size plagioclase and alkali feldspar. Other xenoliths display afine-grained matrix composed of radiate feldspar crystals associ-ated with apatite (Fig. 2, photo D), brown mica and green clinopy-roxene with subordinate magnetite and titanite. Some monzonitesalso have crossed elongate skeletal clinopyroxene, mica and apatitein a matrix of coarse-grained plagioclase and alkali feldspar (Fig. 2,photo G).

6.3. Leucocratic monzosyenite group

These rocks (Fig. 2, photo H) have granular textures, in somecases with a magmatic lamination formed by euhedral to subhe-dral feldspars. The feldspar is zoned from core of oligoclase to an

Table 1

Phenocryst abundance in the volcanic rocks from Gölcük volcano. For lavas, the abundance is determined on thin sections. For pumice samples, phenocrysts have been separatedfrom glass after moderate crushing, and the mineral abundance was determined under binocular lens: 0: absent; 0–1: rare; 2: frequent; 3–4: very frequent; 4: abundant; 5: veryabundant; 10 dominant.

Sample Rock-type Location Lava Tephra Glass Feldspar Cpx Mica Amphibole Titanite Magnetite Apatite

06-013 Lamprophyre Gölcük east X ++ 0 10 5 0 0 0 1–205-012 Trachyte Gölcük crater X + 10 5 5 0-1 1 1 105-014 Trachyte Gölcük crater X + 10 4 5 1–2 1 2 105-008 Tephriphon. Gölcük X +/ÿ 5 506-005 Tephriphon. Gölcük X +/ÿ 5 506-007 Trachyte Gölcük crater X + 10 4 4 0–1 0–1 1–2 1–205-030 Pumice Isparta X +++ 10 4 0–1 4 1–2 1–2 0–105-028 Pumice Hospital X +++ 10 5 0 3–4 1 1 0–105-001 Pumice Cycle I X +++ 10 5 0 0–1 1 2 0–105-024 Pumice Isparta X +++ 10 5 3–4 1–2 1 1 105-034 Pumice Cycle III X +++ 10 5 2 4 1 1 105-036 Pumice Cycle I X +++ 10 5 0 4 1 2 105-033 Pumice Cycle I X +++ 10 5 0–1 2 1 1 105-035 Pumice Cycle I X +++ 10 5 1 3 1 1 105-16b Pumice Cycle I X +++ 10 5 1–2 4 1 105-032 Pumice Cycle I X +++ 10 4 1–2 5 105-005 Pumice Cycle III X +++ 10 5 4 4 0–1 1 0–1

56 B. Platevoet et al. / Journal of Asian Earth Sciences 92 (2014) 53–76

Page 5

extended rim of alkali feldspar with very rare exsolution features.Variable amounts of early-crystallized green pyroxene and magne-tite, titanite and zircon are also observed. Brown mica and

amphibole partly replace the early pyroxene-magnetite phases.Apatite is less common and quartz remains interstitial in the moreevolved syenite type.

A B

C D

E F

G H

1 mm1 mm

2 mm 3 mm

3 mm 3 mm

mica

cpx

F

mica

F amph F

cpx

mica

F

Tit cpx

Mica + cpx

F

cpx

glass

mica

1 mm1 mm

cpx

mica

Fig. 2. Main textures of plutonic xenoliths. (A) Pyroxenite with cumulus clinopyroxene (cpx), intercumulus apatite and mica, and patches of glass. (B) Micaceous pyroxenitewith cumulus cpx and mica, apatite. (C) Pyroxenite with green euhedral elongate cpx and dentritic feldspar growing in the interstitial glass. (D) Micro-crystallized maficmonzonite with euhedral clinopyroxene, mica and apatite in a groundmass of feldspars. (E) Mafic monzonite with intersertal texture made of cpx and mica, oikocrysts ofcoarse-grained plagioclase and alkali feldspar. (F) Monzosyenite containing tiny cpx, mica, apatite and euhedral brown amphibole, set in large oikocrysts of plagioclase andalkali-feldspar. (G) Same texture as E, but very fine-grained with elongate skeletal crystals of cpx and mica, set in a two feldspars matrix. (H) Coarse grained syenite withdominant alkali feldspar, cpx, brown amphibole and mica, minor interstitial quartz. Pictures A, B, C and G under polarized non-analysed light; pictures D, E, F and H undercrossed-nicols. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

B. Platevoet et al. / Journal of Asian Earth Sciences 92 (2014) 53–76 57

Page 6

7. Mineral chemistry

7.1. Olivine and spinel

Olivine (Table 2, see supplementary material) is observed onlyas phenocrysts in the mafic lamprophyric lavas. It is associatedwith spinel inclusions and phlogopite phenocrysts. Olivine crystalsare consistently altered and replaced by oxides and carbonate min-erals. They are Mg-rich (Fo88-89), with relatively low Ni value(0.07–0.24 Ni pfu.). Spinel (Table 2f, see supplementary material)form tiny euhedral crystals within olivine. Its main component ischromite with significant amounts of spinel sensu stricto andmagnesioferrite. Cr/(Cr + Al) ratios are between 0.75 and 0.73, closeto those analyzed in spinels of Macedonian and Serbian lamproites,but less than those analyzed (0.95) in Italian and Spanishlamproites (Prelevic and Foley, 2007; Prelevic et al., 2010a). Olivineand spinel are totally absent from plutonic xenoliths, even inpyroxenites.

7.2. Pyroxene

Clinopyroxene (Cpx) is ubiquitous and is the main early crystal-lized phase in both the lavas and plutonic xenoliths. Except forolivine in lamprophyre, it is always the first mineral to crystallize(with mica) in tephriphonolite until trachyte, and also in pyroxe-nite until quartz-syenite plutonic xenoliths. It commonly has aclear core surrounded by a green rim with complex oscillatory zon-ing in mafic-intermediate lavas and mafic monzonitic xenoliths. Intrachytic lavas, trachytic pumices and leucocratic syenite xeno-liths, the Cpx is pale-green and slightly zoned with a fine greenrim. Orthopyroxene (Opx) has been found only in the Cpx coresin a syenite. Pyroxene formula have been calculated on the basisof 6 oxygens for 4 cations pfu., following the method of Cameron

and Papike (1981) and the main components are calculated follow-ing Cawthorn and Collerson (1974) and Morimoto (1988). Data arereported in Table 2b, supplementary material. The overall evolu-tion of Cpx is from almost pure diopside (Fs2) to augite (Fs25). Feenrichment is restricted in all rocks (Fig. 3). A similar evolutionoccurs from mafic to trachytic lavas and also through the maingroups of plutonic xenoliths. Opx in syenite is magnesium-rich(En92-90).

Complex oscillatory zoning in Cpx with optical gradations fromcolourless to green is correlated with a decreased silica and correl-ative Al and Ti enrichment according to the main substitution: (Mg,Fe)VI + 2SiIV = 2AlIV + TiIV (with Ti never exceeding 0.07 pfu. inmafic to intermediate lavas). Al occurs predominantly as AlIV withdominant ferri-Tchermak component. Calculated AlVI is low, onlyup to 0.10 pfu., with no jadeitic component, a feature which seemsto exclude high pressure crystallisation. Ti and AlIV remain lowfrom trachyandesite to trachyte, probably owing to a silica activityincrease in the most evolved magmas. Cpx from pyroxenite toquartz-syenite has similar chemical characteristics and followsthe same evolution with restricted enrichment in Fe and low Crcontent, even in pyroxenite (Table 2, see supplementary material).However, Cpx rim and skeletal crystals included in the 06–005pyroxenite glassy matrix are Si-poor and Al-Ti-enriched (Table 2,see supplementary material), showing that the trapped residualliquid was probably strongly silica-undersaturated.

Restricted increase of Na is apparent in Cpx rims of both lavasand plutonic xenoliths, following the substitution (Fe2+, Mg)VI +CaM2 = (Fe3+)VI + NaM2. In the diagram Na–Mg–(Fe2+ + Mn) (Fig. 3),the Cpx evolution trend in lavas and xenoliths belongs to a serieswhere crystallization is mainly controlled by the high fO2 coupledwith varied silica activity. A restricted Na–Fe3+ enrichment indi-cates that the liquid was never peralkaline, as Fe3+ first enters inmagnetite and titanite compositions (Gomes et al., 1970; Larsen,

Fig. 3. Clinopyroxene compositions plotted in the enstatite-aegirine-ferrosilite triangle. (a) Clinopyroxenes of trachyandesites and trachytes (pumices and domes). (b)Clinopyroxenes of lamprophyric mafic dykes and tephriphonolites. (c) Clinopyroxenes of plutonic enclaves. Ens: enstatite; AE: aegirine; Fs: ferrosilite.

58 B. Platevoet et al. / Journal of Asian Earth Sciences 92 (2014) 53–76

Page 7

1976; Nielsen, 1979; Bonin and Giret, 1985; Platt and Woolley,1986; Flohr and Ross, 1990; Mitchell and Vladykin, 1996;Dyulgerov and Platevoet, 2006, 2009).

7.3. Mica

Yellowish to brown mica is the second phase to crystallize afterCpx in both lavas and plutonic xenoliths, before amphibole andfeldspars. It crystallized as a cumulus phase in pyroxenite xeno-liths. Its rapid crystallisation with Cpx (crossed skeletal sections)results in a subophitic texture in some monzonites. The micas(Table 2c, see supplementary material) are phlogopite. There areparticularly Mg-rich (Mg# = 0.86–0.66) in lamprophyre rocks butless so (Mg# = 0.68–0.60) in tephriphonolite, trachyandesite andtrachyte lavas. Ba and F contents are maximal in tephriphonolite05–008 (up to 1.81% BaO, 1.5% F). The mica is also phlogopite inthe plutonic xenoliths,. The maximum Mg, Ti, Ba and F contentsare found in the ultramafic pyroxenite group, while Fe enrichmentis restricted in the micas of the monzonite and leucocratic mon-zosyenite groups.

7.4. Amphibole

Magmatic brown euhedral amphibole (Amph) is common onlyin trachyandesitic to trachytic pyroclastic deposits (Table 2d, seesupplementary material). Following the calculations and nomen-clature of Leake et al. (1997) and Hawthorne and Oberti (2007),they belong to the calcic group. Ti content is always less than0.50 Ti pfu.; amphiboles are magnesio-hastingsite, rarely edenite.In plutonic xenoliths, brown amphibole is restricted to the maficmonzonite and the monzosyenite groups. Some mafic monzonitexenoliths can be very enriched in coarse-grained (up to 1–2 cm)euhedral amphibole crystals (magnesio-hastingsite, Table 2d, seesupplementary material) together with numerous euhedral tita-nites, suggesting an accumulation process.

7.5. Feldspars and feldspathoid

Feldspars (Feld) crystallize after Cpx, mica and apatite in thecrystallizing sequence of both lavas and plutonic xenoliths. Feld-spars are absent in lamprophyre. They develop as small euhedralcrystals and microlites in tephriphonolite and trachyandesite lavas.Plagioclase and alkali feldspars are generally associated through amonzonitic texture. Strong zoning is very frequent from core (pla-gioclase or anorthoclase) to rim (anorthoclase or sanidine, respec-tively). In trachyte, 3–5 mm-size phenocrysts of alkali feldspar areobserved with a complex zoning from plagioclase core to a thickzoned sanidine rim. Alkali feldspar is also present as microlites.

Feldspar compositions (Table 2e, see supplementary material)are plotted in the albite-anorthite-orthoclase diagram (Fig. 4a).From tephriphonolitic to trachytic lava, the plagioclase composi-tion has little calcium (between An42 and An19). It commonly formsthe cores of alkali feldspar. The alkali-feldspar is anorthoclase(Or19–Or30) or sanidine (Or35-59 in tephriphonolite, Or31-56 in tra-chyte, Or36-62 in trachyandesitic and trachytic pumices). A trendfrom plagioclase to K-feldspar is observed for the trachyte feldsparphenocrysts. This trend is consistent with crystallization around900 °C for 1 kbar PH2O (Tuttle and Bowen, 1958; Nekvasil, 1994).Plutonic xenoliths have a similar large range of feldsparcompositions.

An extreme compositional variation is also seen in very unusualBa and Sr enrichment of feldspars, especially those of tephriphon-olite, trachyandesitic pyroclastes, pyroxenite and mafic monzonitexenoliths. There is up to 3.49 wt% SrO and 2.05 wt% BaO in feld-spars of tephriphonolite, 4.87 wt% and 3.32 wt% respectively infeldspars of trachyandesite, 4.34 wt% and 3.62 wt% in those of

pyroxenite, and up to 6.86 wt% and 2.72 wt% in those of maficmonzonite. Feldspars in other rock types are less Sr and Baenriched. Ba enters via the classical celsian component BaAl2Si2O8

that reach more than 10% in feldspars from trachyandesitic pyroc-lastics (Fig 4b) and more than 12% in feldspars of mafic monzonite.These values are far lower than those measured in Ba-feldsparsfrom the west-Anatolian Balçikhisar leucitites, Afyon (Akal,2003), in which the celsian component (50%) is equivalent to theorthoclase one. Strontium enters as Sr-rich components that canreach up to 26%, especially in some mafic monzonite feldspars(Fig. 4c). These high contents shift the data in the ‘‘forbidden zone’’of feldspar compositions towards the joint Or–An, but it can be cor-rected by subtracting the Sr components (Fig. 4a). Comparable Srcontents have been found in a meta-eclogite of western Norway(Brastad, 1985) and in the alkaline potassic-ultrapotassic Robitaillesuite, Québec (Hogarth and Robin, 2007). The Sr-bearing phaseequivalent to feldspar is slawsonite (Sr, Ca)Al2Si2O8 (Griffen et al.,1977), and is probably the main Sr feldspar equivalent componentthat enters in Gölcük feldspars. Another Sr-component is stronal-site SrNa2Al4Si4O16 (Hori et al., 1987) which may also be presentespecially in silica undersaturated rocks (Essene et al., 2005). Inthe mafic monzonite feldspars (sample 06-07-E, analyses nb. 31and 34, Table 2e, see supplementary material), the formula unitis non-stoickiometric with more than 5 cations pfu. This is symp-tomatic of the occurrence of the stronalsite component becausethis component has 5.5 cations pfu. when calculated as feldsparequivalent.

Fig. 4. Compositions of feldspars (a) Feldspars plotted in the albite-orthoclase-anorthite diagram. The arrow shows the shift of the unusual compositions whensubstracting the Sr components. A dashed line shows the compositional evolutionof the main feldspars in the lavas. See legend for rock-types. (b) Feldspars in thealbite-celsiane (Cels.)-orthoclase diagram. (c) Feldspars in the albite-Sr-phases(Sr.F)-anorthite diagram.

B. Platevoet et al. / Journal of Asian Earth Sciences 92 (2014) 53–76 59

Page 8

Table

3

(a)W

hole-rock

chem

ical

analyses

formajor

(wt%),traces

andrare

earthelem

ents

(ppm)an

dcalculatedCIPW

norm;DI:differentiationindex

ofTh

orntonan

dTu

ttle

(1960).(b)Traceelem

ents

analyz

edin

clinop

yrox

eneof

thepyrox

enite

06-07-E.

Sample

Lamprophyre(dyk

es)

Cyc

leIItéphriphonolite

(lav

as)

Cyc

leIIItrachytedome(lav

as)

98,423

98,422

98,007

06-013

97-35

97-16

97-25

05-008

05-009

GI-88

GI103

GP9

GP10

97-21

05-012

05-014

06-007

GI-93

GP11

GP12

GP13

SiO2

48.33

50.22

47.56

48.00

56.17

54.03

51.12

54.19

51.37

54.10

51.94

53.52

53.11

61.76

64.03

62.30

61.10

61.06

63.27

63.03

60.53

TiO

21.52

0.97

1.51

1.53

0.57

0.80

0.76

0.70

0.72

0.68

0.73

0.67

0.67

0.52

0.42

0.49

0.50

0.50

0.38

0.37

0.47

Al 2O3

14.16

14.36

14.14

13.99

17.91

17.18

16.67

17.05

15.92

17.09

16.21

16.97

16.98

17.19

16.48

16.91

16.75

16.94

16.48

16.40

16.67

Fe2O3

0.81

0.72

0.80

0.90

0.63

0.75

0.76

0.77

0.87

0.75

0.85

0.76

0.74

0.39

0.38

0.45

0.49

0.46

0.37

0.36

0.45

FeO

5.42

4.83

5.34

5.12

4.19

4.99

5.03

4.38

4.92

4.25

4.80

4.31

4.17

2.61

2.52

3.03

3.27

3.10

2.43

2.40

2.99

MnO

0.12

0.11

0.12

0.11

0.11

0.12

0.12

0.11

0.13

0.10

0.11

0.10

0.10

0.09

0.06

0.08

0.09

0.09

0.18

0.06

0.09

MgO

6.21

6.18

6.08

6.40

1.90

2.95

4.02

3.13

3.73

3.06

3.75

3.12

3.15

1.63

1.90

1.54

1.65

1.55

1.84

1.72

1.57

CaO

10.10

8.98

10.22

9.55

5.56

6.82

8.05

6.70

8.04

6.62

8.10

6.49

6.77

3.91

3.16

3.65

4.23

3.89

3.03

3.16

3.89

Na 2O

1.73

3.86

1.79

2.00

3.81

4.60

4.60

4.10

4.21

3.67

4.58

3.57

4.95

5.24

4.98

5.11

5.25

5.34

5.12

5.17

5.41

K2O

5.44

4.43

5.74

5.14

6.49

4.35

5.66

5.76

5.89

5.75

5.39

6.06

5.80

4.64

5.43

5.16

5.05

5.25

5.80

5.80

5.48

P2O5

0.97

0.70

1.00

1.12

0.43

0.66

0.70

0.63

0.82

0.63

0.81

0.67

0.55

0.30

0.23

0.25

0.29

0.28

0.18

0.23

0.28

H2O

4.87

3.94

4.74

4.94

1.63

0.91

1.28

1.01

1.23

2.00

1.20

2.10

1.30

0.35

0.26

0.43

0.51

0.60

0.10

0.20

0.90

Total

99.69

99.30

99.04

98.81

99.39

98.16

98.77

98.53

97.85

98.70

98.46

98.35

98.28

98.63

99.85

99.40

99.18

99.06

99.18

98.90

98.73

CIPW

Quartz

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

3.52

5.28

3.20

0.85

0.25

2.75

2.54

0.00

Coru

ndum

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

Orthose

32.15

26.18

33.92

30.37

38.35

25.70

33.45

34.04

34.80

33.98

31.85

35.81

34.27

27.42

32.09

30.49

29.84

31.02

34.27

34.27

32.38

Albite

4.71

12.47

0.55

8.18

25.49

29.59

7.61

21.30

9.28

23.02

12.83

20.26

15.50

44.34

42.14

43.24

44.42

45.18

43.32

43.74

45.11

Anorthite

14.80

8.77

13.59

14.02

12.60

13.38

8.12

11.11

7.15

13.18

7.75

12.38

6.98

9.68

6.58

7.96

7.22

6.75

4.86

4.41

5.02

Nep

heline

5.38

10.94

7.91

4.74

3.65

5.05

16.97

7.25

14.27

4.35

14.04

5.39

14.29

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.36

Leucite

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

Wollast./D

12.10

13.03

12.77

10.88

5.08

6.74

11.38

7.52

11.43

6.49

11.33

6.45

9.61

3.24

3.17

3.55

4.96

4.48

3.76

4.08

5.20

Enstatite/D

7.78

8.41

8.20

7.22

2.16

3.29

6.24

3.99

6.16

3.46

6.20

3.43

5.21

1.64

1.71

1.62

2.24

2.02

1.98

2.15

2.39

Ferrosilite/D

3.52

3.75

3.72

2.87

2.94

3.33

4.73

3.30

4.89

2.83

4.72

2.81

4.07

1.53

1.35

1.91

2.69

2.43

1.67

1.81

2.76

Wollast.

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

Enstatite/H

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

2.42

3.02

2.22

1.87

1.84

2.60

2.14

0.00

Ferrosilite/H

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

2.25

2.38

2.61

2.25

2.22

2.20

1.80

0.00

Forsterite/O

lv5.39

4.90

4.86

6.11

1.80

2.85

2.65

2.67

2.19

2.92

2.20

3.04

1.85

0.00

0.00

0.00

0.00

0.00

0.00

0.00

1.06

Faya

lite/O

lv2.69

2.41

2.43

2.68

2.70

3.18

2.21

2.43

1.91

2.64

1.84

2.75

1.59

0.00

0.00

0.00

0.00

0.00

0.00

0.00

1.35

Mag

netite

1.18

1.05

1.16

1.31

0.91

1.09

1.09

1.12

1.26

1.09

1.23

1.10

1.07

0.57

0.55

0.65

0.71

0.67

0.54

0.52

0.65

Ilmen

ite

2.89

1.84

2.87

2.91

1.08

1.52

1.44

1.33

1.37

1.29

1.39

1.27

1.27

0.99

0.80

0.93

0.95

0.95

0.72

0.70

0.89

Apatite

2.23

1.61

2.30

2.58

0.99

1.52

1.61

1.45

1.89

1.45

1.87

1.54

1.27

0.69

0.53

0.58

0.67

0.65

0.41

0.53

0.65

Total

94.81

95.35

94.28

93.86

97.76

97.25

97.48

97.51

96.61

96.70

97.25

96.24

96.98

98.28

99.58

98.96

98.67

98.46

99.08

98.70

97.83

D.I.

42.23

49.59

42.38

43.29

67.50

60.35

58.02

62.59

58.36

61.35

58.72

61.46

64.07

75.27

79.50

76.93

75.11

76.46

80.35

80.56

77.85

Ba

2244

2446

2506

2930

3204

3004

3303

2750

3587

2808

3181

3023

2990

2344

1823

2407

2417

2489

2154

2277

2592

Ce

196.0

212.0

206.0

226.9

460.5

598.3

407.0

380.0

480.0

420.6

495.7

430.0

419.8

291.3

250.0

283.0

304.8

348.3

265.4

274.9

350.4

Co

26.7

23.5

26.3

27.0

12.0

15.3

20.3

15.2

17.8

19.1

26.5

17.9

16.9

8.2

8.2

7.5

8.6

11.4

10.1

9.8

11.4

Cr

285

370

260

282

516

24

14

30

nd

nd

nd

nd

11

45

15

18

nd

nd

nd

nd

Cs

3.47

3.34

313.61

2.61

2.73

3.48

2.54

3.57

4.1

3.9

9.5

4.5

3.37

2.17

6.14

3.48

11.4

10.1

9.8

11.4

Cu

61

25

60

57

36

40

66

64

60

nd

nd

nd

nd

19

89

21

nd

nd

nd

nd

Dy

5.42

5.37

5.59

5.46

7.84

9.69

7.61

5.93

6.94

6.31

7.03

6.04

6.03

4.90

3.26

3.95

4.29

4.46

3.2

3.35

4.2

Er

2.34

2.25

2.48

2.34

3.11

3.96

3.16

2.42

2.69

2.52

2.76

2.63

2.49

2.09

1.47

1.80

1.93

21.32

1.38

2.03

Eu

3.27

3.25

3.52

3.50

6.83

8.37

6.70

4.89

6.69

5.3

6.69

5.13

4.94

3.94

2.63

3.10

3.44

3.39

2.53

2.81

3.14

Ga

19.2

18.70

19.3

19.92

25.90

26.40

23.90

22.40

21.90

22.7

21.7

22

20.6

27.00

24.10

25.00

26.12

25.5

22.2

24.1

24.7

Gd

8.92

8.98

9.21

8.94

14.54

19.40

14.50

10.60

13.50

10.66

12.72

12.07

10.86

8.65

5.48

6.61

7.86

7.67

5.29

5.75

7.52

Hf

14.7

13.70

14.7

14.49

10.00

12.10

9.80

8.73

9.62

10.1

10.3

11.1

11

10.10

8.32

8.66

9.17

11.3

9.1

9.6

12.5

Ho

0.878

0.76

0.858

0.88

1.06

1.30

1.10

0.93

1.07

0.96

1.1

0.95

0.93

0.71

0.52

0.65

0.70

0.7

0.51

0.53

0.67

La106.0

121.0

113.0

114.4

249.8

341.2

223.8

213.0

268.0

215.5

252.6

184.8

177.7

159.9

132.0

169.0

167.4

165.9

121.5

126.8

152.0

Lu0.295

0.281

0.276

0.295

0.350

0.450

0.380

0.329

0.342

0.380

0.380

0.350

0.350

0.290

0.246

0.295

0.297

0.340

0.250

0.220

0.320

Nb

62.1

56.0

65.2

64.8

44.8

61.6

42.3

38.2

41.7

45.0

42.6

46.6

45.0

45.4

36.5

42.0

43.0

47.3

39.5

37.8

47.2

Nd

79.0

77.6

82.7

82.0

187.2

243.4

169.5

135.0

176.0

152.0

181.7

156.8

149.3

102.6

78.6

90.1

96.6

102.5

83.3

84.8

109.3

Ni

93.4

125.0

79.7

85.0

4.4

12.3

13.5

10.5

15.7

9.0

11.0

4.5

3.8

10.6

40.9

12.7

15.4

12.0

15.1

14.9

3.9

60 B. Platevoet et al. / Journal of Asian Earth Sciences 92 (2014) 53–76

Page 9

Table

3(continued

)

Sample

Lamprophyre(dyk

es)

Cyc

leIItéphriphonolite

(lav

as)

Cyc

leIIItrachytedome(lav

as)

98,423

98,422

98,007

06-013

97-35

97-16

97-25

05-008

05-009

GI-88

GI103

GP9

GP10

97-21

05-012

05-014

06-007

GI-93

GP11

GP12

GP13

Pb

21.8

37.70

21.1

20.20

49.70

65.30

43.00

37.30

38.70

1.7

2.8

2.2

2.6

58.70

35.50

49.30

28.26

4.5

19.3

5.6

3.6

Pr

22.0

22.3

22.5

23.0

53.6

68.8

47.3

37.7

47.8

43.2

51.7

40.7

39.7

29.5

23.9

27.0

28.8

31.0

23.7

24.6

30.7

Rb

201

157

197

182

135

107

152

125

108

123

132

131

126

162

138

149

150

159

144

147

155

Sm12.50

12.90

13.20

13.08

26.19

34.06

24.72

19.50

25.40

22.30

26.10

19.20

18.30

14.12

10.60

12.40

13.27

14.40

9.70

10.20

13.20

Sr1306

1976

1394

1433

5682

5722

5227

3837

4700

4618

5433

5329

4973

3590

2379

2794

3213

3616

3302

3471

3814

Ta

3.98

3.20

4.02

3.99

3.05

4.12

2.84

2.74

2.78

2.90

2.70

2.80

2.70

2.73

2.31

2.61

2.60

2.60

2.10

2.10

2.50

Tb

1.12

1.12

1.15

1.14

1.80

2.22

1.78

1.26

1.57

1.44

1.67

1.41

1.36

1.06

0.68

0.82

0.93

1.01

0.7

0.75

0.94

Th

24

34

22

22

68

103

51

50

49

56

56

57

57

75

59

68

71

79

61

63

77

Tm

0.296

0.289

0.320

0.317

0.420

0.490

0.390

0.344

0.359

0.380

0.440

0.380

0.350

0.300

0.218

0.265

0.278

0.320

0.220

0.250

0.310

U6.30

9.47

6.13

6.24

10.31

22.20

12.27

12.70

12.50

14.30

13.10

14.30

15.00

19.93

16.70

21.80

18.03

18.20

17.70

19.70

18.90

V167

126

161

162

112

122

147

120

135

143

154

148

124

85

53

79

81

89

58

61

90

Y24.9

24.5

26.2

26.1

31.7

37.6

33.4

27.6

31.0

32.3

36.2

31.9

30.3

22.8

16.7

20.7

21.6

24.0

17.0

17.2

23.2

Yb

1.91

1.99

2.02

1.99

2.45

2.96

2.64

2.21

2.30

2.43

2.73

2.37

2.31

2.05

1.52

1.86

1.91

2.07

1.55

1.65

2.08

Zn

83

73

80

80

86

103

88

77

88

nd

nd

nd

nd

71

50

63

66

nd

nd

nd

nd

Zr

555

539

593

714

445

521

424

396

455

422

435

412

395

414

358

370

442

460

320

295

398

Sample

Cyc

leIpyroclastite

(pumices)

Cyc

leIIIpyroclastite

(pumices)

Plutonic

xenoliths

05-

001

06-

015

06-

011c

05-

036

05-

035

05-

016b

05-

032

05-

033

97-

01

05-

005

05-

028

05-

024

05-

030

05-

034

E-3

E-6-

PX

S10

E10B

06-

07-E

97-

66

97-

60

05-

014b

E 10A

97-12

97-

77

E10C

97-

20

SiO2

58.18

60.82

58.21

54.53

55.26

55.53

57.56

60.43

61.57

57.22

58.01

58.52

60.48

57.79

47.67

45.46

41.87

46.49

47.01

48.93

51.54

54.64

56.17

57.88

58.03

61.41

63.88

TiO

20.23

0.28

0.40

0.62

0.59

0.55

0.57

0.46

0.33

0.56

0.45

0.51

0.43

0.56

1.04

1.17

1.45

1.12

1.07

1.23

0.93

0.96

0.71

0.66

0.66

0.31

0.32

Al2O3

16.12

16.87

18.60

15.96

16.35

16.92

16.67

16.77

16.23

16.89

16.66

16.76

16.63

16.67

6.99

8.29

11.92

7.72

11.11

13.34

14.69

16.49

16.80

18.71

16.42

20.62

17.07

Fe2O3

0.26

0.27

0.39

0.66

0.64

0.61

0.50

0.51

0.31

0.70

0.50

0.52

0.48

0.59

0.77

0.85

1.12

1.01

0.97

0.88

0.78

0.90

0.84

0.66

0.71

0.44

0.36

FeO

1.70

1.78

2.59

4.40

4.24

4.08

3.36

3.42

2.04

3.58

3.31

3.46

3.17

3.92

5.15

5.65

7.44

5.16

6.47

5.86

5.19

4.57

4.26

4.39

4.76

2.25

2.40

MnO

0.11

0.06

0.08

0.11

0.11

0.11

0.09

0.10

0.06

0.11

0.10

0.11

0.10

0.11

0.09

0.10

0.14

0.09

0.14

0.13

0.09

0.13

0.11

0.09

0.14

0.10

0.05

MgO

0.42

0.55

0.72

2.22

2.04

2.17

2.13

1.25

1.18

1.70

1.17

1.45

1.17

1.77

17.06

15.98

11.28

17.15

10.23

11.19

6.48

4.64

2.93

1.69

1.84

0.25

1.48

CaO

4.66

3.42

3.35

5.83

5.79

5.53

4.69

3.57

2.21

4.52

3.92

4.03

3.57

4.72

16.31

14.79

13.44

15.00

14.24

7.40

9.59

8.21

6.00

5.30

5.22

3.36

1.97

Na 2O

4.34

4.97

5.05

3.61

3.84

4.65

5.19

5.27

5.66

5.06

5.10

5.05

5.27

5.11

0.48

0.68

1.91

0.58

1.62

2.02

3.19

4.43

4.40

5.29

6.16

6.53

5.84

K2O

6.07

5.08

5.69

4.09

4.41

5.12

4.99

4.99

5.46

5.18

5.06

5.29

5.06

5.16

2.91

3.71

4.37

3.37

3.56

7.01

4.35

3.21

5.16

3.73

4.38

3.63

5.13

P2O5

0.06

0.09

0.14

0.52

0.53

0.49

0.33

0.26

0.22

0.34

0.27

0.29

0.26

0.41

0.10

0.97

2.61

0.08

1.83

0.41

1.00

0.61

0.57

0.31

0.21

0.07

0.23

H2O

7.23

4.47

3.23

6.21

4.41

2.31

3.18

1.29

2.91

3.23

3.84

2.43

2.08

1.90

1.22

1.36

0.70

1.36

0.56

0.89

0.46

1.27

0.32

0.08

0.02

0.29

0.33

Total

99.38

98.66

98.45

98.76

98.21

98.07

99.26

98.32

98.17

99.09

98.39

98.42

98.70

98.71

99.79

99.01

98.25

99.12

98.81

99.29

98.29

100.05

98.27

98.79

98.55

99.26

99.06

CIPW

Quartz

1.51

5.01

0.00

2.71

1.50

0.00

0.00

1.36

1.95

0.00

0.00

0.00

1.57

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.58

3.49

Coru

ndum

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

Orthose

35.87

30.02

33.62

24.17

26.06

30.25

29.49

29.49

32.26

30.61

29.90

31.26

29.90

30.49

0.00

0.00

0.00

0.00

21.05

23.96

25.70

18.97

30.50

22.04

25.88

21.43

30.31

Albite

36.72

42.05

41.27

30.54

32.49

32.57

39.36

44.59

47.89

38.63

43.15

42.63

44.59

39.16

0.00

0.00

0.00

0.00

0.15

0.00

16.31

33.40

32.04

44.76

39.03

55.28

49.41

Anorthite

6.58

8.72

11.28

15.26

14.35

10.17

7.45

7.37

2.75

8.07

7.62

7.44

6.78

7.31

8.32

8.60

11.04

8.53

12.53

6.63

12.92

15.63

10.82

16.29

4.22

16.22

5.21

Nep

heline

0.00

0.00

0.79

0.00

0.00

3.67

2.47

0.00

0.00

2.27

0.00

0.05

0.00

2.21

2.21

3.13

8.75

2.64

7.32

9.26

5.78

2.20

2.83

0.00

7.09

0.00

0.00

Leucite

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

13.49

17.19

20.25

15.61

0.00

13.70

0.00

0.00

0.00

0.00

0.00

0.00

0.00

Wollast./D

3.62

3.20

1.85

4.29

4.56

5.87

5.70

3.61

2.83

5.06

4.20

4.45

3.86

5.61

30.05

24.40

16.11

27.30

19.29

11.44

11.75

8.80

6.36

3.33

8.48

0.01

1.28

Enstatite/

D1.05

1.11

0.60

1.92

2.00

2.69

2.87

1.36

1.36

2.25

1.56

1.82

1.46

2.38

22.98

18.32

10.82

20.99

12.96

8.07

7.49

5.38

3.35

1.31

3.31

0.00

0.63

Ferrosilite/

D2.73

2.18

1.31

2.34

2.55

3.13

2.70

2.31

1.42

2.80

2.73

2.66

2.46

3.24

3.93

3.64

4.08

3.41

4.86

2.39

3.49

2.93

2.82

2.06

5.28

0.01

0.63

Wollast.

3.13

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

Enstatite/

H0.00

0.26

0.00

3.60

3.08

0.00

0.00

1.75

1.57

0.00

0.57

0.00

1.45

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.70

0.00

0.63

3.06

Ferrosilite/

H0.00

0.52

0.00

4.38

3.94

0.00

0.00

2.97

1.64

0.00

1.00

0.00

2.44

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

1.09

0.00

3.45

3.05

Forsterite/

Olv

0.00

0.00

0.84

0.00

0.00

1.90

1.70

0.00

0.00

1.39

0.55

1.26

0.00

1.42

13.67

15.05

12.11

15.23

8.77

13.87

6.06

4.33

2.77

1.54

0.89

0.00

0.00

(continued

onnextpage)

B. Platevoet et al. / Journal of Asian Earth Sciences 92 (2014) 53–76 61

Page 10

Faya

lite/

Olv

0.00

0.00

2.02

0.00

0.00

2.44

1.76

0.00

0.00

1.91

1.07

2.03

0.00

2.13

2.58

3.30

5.03

2.73

3.63

4.52

3.11

2.60

2.58

2.66

1.58

0.00

0.00

Mag

netite

0.38

0.39

0.57

0.96

0.93

0.88

0.72

0.74

0.44

1.01

0.72

0.75

0.70

0.85

1.12

1.23

1.62

1.47

1.41

1.27

1.13

1.30

1.21

0.96

1.03

0.64

0.52

Ilmen

ite

0.44

0.53

0.76

1.18

1.12

1.04

1.08

0.87

0.63

1.06

0.85

0.97

0.82

1.06

1.98

2.23

2.75

2.13

2.04

2.34

1.77

1.83

1.34

1.25

1.25

0.58

0.61

Apatite

0.14

0.21

0.32

1.20

1.22

1.13

0.76

0.60

0.51

0.78

0.62

0.67

0.60

0.94

0.22

2.23

6.01

0.18

4.21

0.94

2.30

1.41

1.30

0.71

0.48

0.15

0.53

Total

92.15

94.19

95.22

92.54

93.79

95.75

96.08

97.03

95.26

95.86

94.55

95.99

96.62

96.81

100.54

99.32

98.60

100.21

98.22

98.40

97.82

98.78

97.94

98.71

98.53

98.97

98.73

D.I.

74.10

77.08

75.68

57.42

60.05

66.49

71.31

75.44

82.11

71.51

73.05

73.94

76.06

71.86

15.70

20.32

29.00

18.24

28.52

46.91

47.80

54.57

65.37

66.80

72.01

77.28

83.22

Ba

822

2113

3401

3179

3268

2920

2362

3495

1469

3585

3072

3262

3157

3637

2905

3809

4070

3901

4554

4231

3926

2299

3028

6050

2115

2820

1832

Ce

323

240

413

479

537

445

285

398

221

375

423

363

341

394

52

128

394

49

392

175

236

304

391

282

290

646

250

Co

23

411

11

12

97

69

88

69

43

46

49

46

36

36

26

21

15

910

26

Cr

96

415

911

22

424

14

411

>414

561

262

68

619

189

1411

152

38

20

511

<L.D.

31

Cs

8.69

5.72

5.42

3.55

4.53

4.68

4.46

6.55

6.91

6.09

8.30

6.65

6.94

5.64

1.13

1.156

2.14

1.36

0.914

2.15

1.50

1.89

0.33

0.77

0.39

0.35

1.08

Cu

5.6

8.3

7.5

34.1

43.1

31.7

24.0

15.5

13.0

16.7

14.4

14.7

11.6

29.5

9.8

10.1

123.0

6.8

32.8

5.7

53.2

40.5

<L.D.

7.6

7.0

<L.D.

2.1

Dy

3.68

3.07

5.50

5.90

6.92

6.42

4.18

5.11

3.48

5.63

5.50

5.47

4.78

5.56

2.02

3.26

8.15

1.87

6.94

3.52

5.13

5.21

6.54

6.66

7.22

5.26

3.43

Er

2.11

1.47

2.44

2.37

2.69

2.58

1.86

2.24

1.52

2.38

2.36

2.33

2.11

2.39

0.65

1.04

2.93

0.61

2.32

1.45

1.94

2.17

2.68

2.64

3.26

2.40

1.58

Eu

1.96

2.32

4.73

5.38

6.55

5.67

3.19

4.14

2.29

4.97

4.55

4.62

3.78

4.79

2.09

3.59

7.73

1.90

7.12

3.14

5.14

4.45

5.70

6.59

5.11

4.53

2.85

Ga

29.80

26.39

27.97

23.90

26.06

26.01

22.60

24.60

24.50

25.61

27.06

25.04

23.70

24.30

8.77

11.45

16.80

9.52

17.61

20.40

22.60

24.51

24.75

22.80

26.40

30.44

24.00

Gd

5.64

5.18

9.87

11.20

13.61

12.35

6.98

9.00

5.22

10.28

10.60

9.85

8.10

9.83

4.28

7.28

16.55

3.89

15.46

7.29

9.43

9.52

12.43

10.98

11.56

9.10

5.87

Hf

13.01

9.10

10.94

9.20

10.40

9.85

7.51

7.91

8.88

8.67

9.24

9.02

7.96

8.70

1.27

1.66

4.16

1.32

4.80

3.67

8.39

4.35

10.18

9.16

14.20

11.33

8.96

Ho

0.683

0.500

0.850

0.900

1.019

0.952

0.676

0.814

0.528

0.863

0.864

0.846

0.782

0.878

0.291

0.457

1.090

0.262

0.975

0.476

0.730

0.814

1.021

0.960

1.110

0.832

0.495

La204.0

142.7

234.4

264.0

300.7

250.7

169.0

245.0

131.3

212.1

252.1

206.6

202.0

227.0

20.4

58.8

209.9

20.3

188.3

88.6

121.7

152.3

183.9

136.1

121.6

457.1

156.4

Lu0.425

0.256

0.366

0.301

0.353

0.359

0.281

0.330

0.290

0.355

0.355

0.354

0.333

0.351

0.068

0.104

0.280

0.066

0.239

0.156

0.236

0.280

0.363

0.330

0.450

0.352

0.310

Nb

72.8

44.4

63.5

47.2

56.7

44.6

41.6

42.5

42.5

52.1

45.7

51.5

41.3

51.1

4.9

7.5

21.9

5.6

16.2

27.6

35.2

28.3

46.4

41.6

60.4

94.9

33.2

Nd

79.1

69.5

137.6

163.0

183.6

155.5

92.4

123.0

76.0

128.4

134.4

121.3

110.0

132.0

33.6

68.3

192.7

31.7

169.0

75.1

108.6

121.6

161.5

137.0

140.9

159.1

81.5

Ni

7.8

12.6

5.2

11.5

8.9

9.3

9.1

3.8

24.6

7.5

6.7

6.1

3.4

6.2

114.7

105.1

46.5

125.7

47.4

116.0

32.3

28.0

22.1

2.1

2.9

<L.D.

24.0

Pb

73.4

54.3

62.0

66.2

85.5

59.5

40.3

54.4

49.4

42.1

78.7

71.4

54.0

39.4

2.1

3.9

9.8

2.4

12.7

14.2

24.4

24.3

28.0

27.0

32.2

61.2

23.8

Pr

27.42

22.13

41.22

46.20

53.53

44.54

27.20

36.6

22.61

37.31

40.38

35.59

32.10

37.70

7.409

16.3

48.13

7.06

46.93

19.43

28.63

34.10

44.77

35.86

38.03

59.02

24.66

Rb

211

142

133

90

107

116

120

136

178

136

162

144

142

130

78

90

109

91

74

196

93

89

72

41

77

69

137

Sm9.54

8.98

18.17

22.10

24.96

21.71

12.80

17.00

9.86

17.94

18.69

16.93

15.30

18.20

6.94

11.79

28.82

6.32

25.94

11.23

16.46

17.45

23.31

20.09

21.50

18.28

9.81

Sr1164

2353

4501

4635

5363

4780

2439

3448

2012

4746

3299

4257

3163

3960

767

936

2236

1005

3418

2483

4369

4549

5437

10870

4155

6283

2453

Ta

3.45

2.38

4.57

3.12

3.53

2.85

2.50

2.79

2.47

3.10

2.80

3.11

2.63

3.01

0.29

0.44

1.45

0.34

1.05

0.95

2.03

2.00

3.78

3.53

4.29

4.66

1.97

Tb

0.74

0.64

1.19

1.33

1.60

1.41

0.88

1.12

0.69

1.23

1.22

1.17

1.02

1.20

0.47

0.77

1.95

0.43

1.64

0.85

1.14

1.15

1.47

1.50

1.55

1.17

0.74

Th

138.2

77.5

87.1

65.4

77.9

67.6

56.7

75.2

80.6

65.3

94.8

69.6

70.7

62.6

1.9

5.4

23.7

2.3

27.7

19.2

72.3

26.4

24.6

30.0

35.3

145.4

70.6

Tm

0.35

0.22

0.35

0.31

0.35

0.36

0.26

0.32

0.23

0.33

0.33

0.33

0.31

0.34

0.08

0.12

0.30

0.08

0.28

0.18

0.23

0.28

0.36

0.31

0.42

0.34

0.22

U37.11

21.67

22.35

14.30

15.94

18.04

15.40

23.20

25.03

21.55

29.21

23.86

23.70

20.80

0.37

0.99

8.50

0.45

4.31

3.36

13.11

7.82

3.66

4.02

8.71

29.25

18.59

V28

41

63

107

112

112

92

90

44

112

93

100

82

115

131

164

250

138

215

145

174

130

98

142

127

68

53

Y23.5

16.6

26.7

26.6

30.4

29.2

21.0

25.6

16.5

27.0

26.7

26.3

24.2

27.6

7.5

12.5

30.8

7.3

27.1

15.1

21.1

24.2

29.9

27.6

32.7

25.2

15.6

Yb

2.62

1.59

2.42

2.05

2.29

2.32

1.77

2.19

1.61

2.27

2.28

2.31

2.07

2.27

0.48

0.739

1.82

0.45

1.658

0.97

1.51

1.86

2.36

1.98

2.75

2.32

1.67

Zn

88

64

76

87

86

87

70

70

57

80

78

76

66

82

42

52

98

53

82

130

74

129

99

67

108

100

36

Zr

582

471

591

425

523

514

352

363

346

448

456

448

362

399

41

60

172

42

191

193

361

157

506

426

671

660

368

Rock

06-07-E

Cpx1

cCpx1

bCpx2

cCpx3

cCpx3

bElemen

tco

rerim

core

core

rim

Rb

0.0066

0.0033

0.0443

0.0225

<0.0036

Ba

0.21

0.22

1.14

0.80

0.49

Th

0.07

0.07

0.43

0.30

0.26

U0.01

0.01

0.06

0.03

0.02

Nb

0.03

0.03

0.29

0.15

0.14

Ta

0.01

0.00

0.05

0.03

0.03

La9.10

9.08

25.10

17.49

16.12

Ce

31.73

31.43

87.34

61.67

55.58

Pb

0.24

0.24

0.48

0.31

0.34

Pr

5.35

5.28

14.64

10.45

9.28

Sr534.9

529.2

610.3

473.5

602.6

Nd

28.65

28.77

77.21

56.49

49.40

Zr

15.75

14.95

96.31

62.85

46.88

Hf

0.72

0.65

4.35

2.92

2.07

62 B. Platevoet et al. / Journal of Asian Earth Sciences 92 (2014) 53–76

Page 11

Experimental work on the stability of solid solutions betweenBa and Sr components and feldspars has been investigated byLagache (1999) and Essene et al. (2005). They did not find any sol-vus at relatively high temperature; the immiscibility field isencountered only during low hydrothermal temperature or meta-morphic transformation. The preserved high Ba and Sr-bearingfeldspar compositions show that parent magmas must haveunusually high contents of Sr and Ba, especially magmas ofmafic-intermediate compositions. This has already been observedin the K-rich Miocene volcanism of Western Anatolia, Turkey(Akal, 2003; Ersoy et al., 2010).

Feldspathoid commonly occurs as small globular clear crystalsin the groundmass of strongly silica-undersaturated tephriphono-lite. It belongs to the sodalite-haüyne series (Table 2e, see supple-mentary material). The core of crystal is sodalite (up to 86% pfu.),the intermediate zone is nosean, and the rim is haüyne rim (upto 89% pfu.) with greater Ca and K contents, and a sulphate ionreplacing chlorine.

7.6. Magnetite and titanite

Magnetite (Mgt) is present in almost all rock types and titanite(Tit) is the only Ti-rich phase. Titanite is present with Mgt in thegroundmass of tephriphonolite. Titanites form early euhedral yel-lowish crystals in all trachyandesitic and trachytic lavas and intrachytic pyroclasts. Euhedral Tit crystals are also very frequentin plutonic xenoliths except in pyroxenites. Mgt crystals haverestricted amount of the ulvospinel component (Table 2f, see sup-plementary material) probably owing to the partitioning of Tibetween Mgt and Tit. The systematic magnetite-titanite associa-tion is symptomatic of highly oxidizing conditions during the crys-tallisation of intermediate magma and felsic silica-saturatedmagma (Wones, 1989). This is in good agreement with therestricted Cpx Fe-enrichment during magmatic differentiationand the relatively low Ti content of Cpx and amphibole, even in sil-ica undersaturated rock-types (Helz, 1973). The titanite structuralformula has been recalculated on the basis of 4 Si cations p.f.u.(Table 2f, see supplementary material). Relatively Ca-poor titanitewith a sum of cations less than twelve is indicative of significantamount of REE in titanite. The calculated REE total reaches up to0.32 cations p.f.u. in a titanite of mafic monzonite (Table 2f, seesupplementary material).

7.7. Apatite

Abundant apatite (Ap) crystallized early throughout the entirevolcanic series. Apatite has been previously analyzed by Özgüret al. (1992) showing more than 5 wt% F, reflecting a substantialF enrichment in shallow aqueous systems by apatite dissolution(Özgür et al., 1992). Apatites (Table 2f, see supplementary mate-rial) of tephriphonolite and pyroxenite contain high amounts of F(2.5 wt%), until 4.8 wt% in pyroclastic pumices. Sr contents are alsohigh, especially in the apatites of pyroxenite, with more than 1.8wt% SrO. As apatite crystallized before feldspar, it incorporate sig-nificant amount of Sr from the Sr-rich parent magma.

8. Geochemistry

8.1. Whole-rock major and trace elements

Major and trace element analyses of the main volcanic rock-types and plutonic xenoliths are given in Table 3a. All data plotin the alkali-rich field of the TAS diagram (Fig. 5a, according toLe Bas et al., 1986 and Le Maitre, 2002). Volcanic rocks evolvedfrom strongly silica-undersaturated compositions toward

Sm6.54

6.51

17.47

13.38

11.19

Eu

1.78

1.77

4.49

3.72

3.00

Ti

2085

2040

3948

2814

3075

Gd

4.35

4.23

11.38

9.59

7.34

Tb

0.44

0.43

1.24

1.04

0.76

Dy

1.99

2.00

5.82

4.87

3.54

Y7.10

6.90

21.38

17.96

12.20

Ho

0.30

0.28

0.87

0.75

0.53

Er

0.60

0.64

2.02

1.68

1.16

Tm

0.07

0.07

0.25

0.21

0.13

Yb

0.43

0.41

1.46

1.20

0.82

Lu0.06

0.06

0.21

0.17

0.12

B. Platevoet et al. / Journal of Asian Earth Sciences 92 (2014) 53–76 63

Page 12

silica-oversaturated for trachytes. The K2O/Na2O ratio decreasesfrom lamprophyres to trachytes (Fig. 5b). Most lavas plot in theshoshonite field (after Peccerillo and Taylor, 1976), except lampro-phyres in the ultrapotassic field. This is in good agreement with theresults of Elitok et al. (2010), Dilek and Altunkaynak (2010) for theNeogene-Quaternary volcanism of Western Anatolia. According totheir K2O/Na2O values, pyroxenite xenoliths also plot in the ultra-potassic field. The more evolved monzosyenites lie outside of theultrapotassic-shoshonitic field of Western Anatolia defined byDilek and Altunkaynak (2010) (Fig. 5b). As evidenced in Harker dia-grams (Fig. 6), the amounts of CaO, FeOt, MgO, TiO2 and P2O5

decrease with SiO2, while Al2O3 and the Na2O/K2O ratio increasewith SiO2. FeOt/FeOt + MgO ratio first increases and finallydecreases in the last trachytic lavas.

Compared to the tephriphonolite group, pyroxenite xenolithsare highly enriched in CaO, MgO and Ba. Conversely, they areAl2O3, Sr and Zr poor. The Na2O/Na2O + K2O ratios are low.

In the TAS diagram (Fig. 5a), the pyroxenite group representa-tive points plot along a mixing line between their two majorrock-forming minerals (phlogopite and Cpx) chemical composi-tions. This apparent trend of the pyroxenite group results from var-iable mixtures of cumulate of Mg-rich Cpx, phlogopite and apatite.Their apparent ultrapotassic character (Fig. 5b) can be explained byphlogopite accumulation with Na2O-poor phases such as Cpx andapatite. The magma from which pyroxenites are derived may havefractionated Cpx and phlogopite first, then apatite. In intermediatemagmas, feldspar, magnetite and titanite fractionation is also sug-gested as Al2O3 is stabilized while Sr, Ba, TiO2, FeOt/MgO decrease.The very high Al2O3, Sr and Ba contents of some monzosyenitexenoliths show that they may be interpreted as feldspar cumu-lates. Zr is also maximal in trachyandesite and in monzosyenitexenoliths, suggesting possible significant zircon (zirc) cumulation.

Calculated CIPW norms (Table 3) evidence that silica-undersat-uration increase from lamprophyre (5–11 wt% Ne norm) to tephri-phonolite (up to 17 wt%). The rocks become silica-saturated fromtrachyandesite pumices to trachyte lavas with up to 5 wt% of nor-mative quartz. Most of ultramafic–mafic plutonic xenoliths arestrongly silica-undersaturated with up to 20 wt% of calculated nor-mative leucite that reflects the occurrence of phlogopite in therocks. The silica-saturation increases in the more evolved plutonicxenoliths that contain up to 3.5 wt% of normative quartz.

Ba and Sr contents are remarkably high both in volcanic rocksand plutonic xenoliths, respectively up to 6000 ppm and11,000 ppm in syenitic xenoliths (Table 3). Cr, Ni, Co, V transi-tion element contents are notably low, even in lamprophyres(260–370 ppm of Cr), and decrease from mafic to trachytic rocks.Maximum contents are found in pyroxenite and mafic monzonitexenoliths where Cpx is abundant. The Ni and Cr contents ofpyroxenite never exceed respectively 125 ppm and 650 ppm, farbelow most mantle pyroxenites of western Europe and northernAfrica, except some Beni Boussera and Ronda pyroxenites (seeDownes, 2007). Plutonic xenoliths and lavas seem to follow verysimilar trends. The fractionating assemblage may have been becontrolled first by pyroxene, apatite, phlogopite and magnetite,with addition of titanite, amphibole, feldspars and zircon in moreevolved magma compositions.

8.2. Residual glass composition

Microprobe analyses of glass from the most representative lavatypes and from two incompletely crystallized pyroxenites, areplotted in the total alkali versus silica diagram (Table 4, Fig. 7a).Because of variable degree of hydration and alteration processes,all analyses were recalculated to 100% before plotting. From lamp-rophyre to trachyte lavas, residual glass compositions evolve fromphonolitic to trachytic compositions, respectively. Glass found inpyroxenites are highly evolved, between phonolite and trachyte.Residual glasses are alkali-rich and contain substantial amountsof Fe, Ti, Sr and Ba. F contents are low, probably because of earlycrystallization of apatite. Plotted in the K2O versus Na2O diagram(Fig. 7b), lamprophyre glasses are the most potassic. K2O/Na2Oratios decrease below 1 in most of trachyandesitic and trachyticlavas. This evolution may be governed by phlogopite removal.

8.3. Trace element patterns

From tephriphonolite to trachyandesite and trachyte, primitivemantle-normalized (McDonough and Sun, 1995) trace elementpatterns are very similar in shape, showing a strong enrichmentin large ion lithophile elements (LILE) and LREE compared to HREE(Fig. 8a). This suggests a common and relatively homogeneous gar-net-bearing mantle source. Patterns of lamprophyric mafic dykeare smooth compared to other lavas, with a slight positive Zranomaly. Patterns of more evolved lavas (tephriphonolite and tra-chyte) always have more or less Nb, Ta, Hf, Zr, Ti (HFSE) negativeanomalies and K depletion relatively to LILE, Sr and LREE. Ti, Nband Ta negative anomalies are well marked in relation with frac-tionation of magnetite, titanite and amphibole. The global decreaseof REE from tephriphonolite to trachyte is a consequence of bothapatite and titanite removal. The variability of trace elements pat-terns is well expressed in the trachyandesitic Cycle I pyroclastics:the more evolved compositions (with highest Rb, U and Th con-tents) are also the more Ba and Sr depleted, suggesting feldsparremoval. Absolute content of HFSE in lavas and pyroclastics areas high as in mid-Miocene high-K and ultra-K volcanic rocks fromWestern Anatolia (Ersoy et al., 2010); it is also as high as in within-plate continental potassic lavas, like those of the Virunga Province,Rwanda (Rogers et al., 1998).

Fig. 5. Whole rock alkali contents. (a) Total alkali – silica (TAS) diagram (wt%).Dashed line: mixing line betweenmica and clinopyroxene mean compositions (greystars). (b) K2O versus Na2O diagram according to Peccerillo and Taylor (1976). Solidlines: limits of the ultrapotassic field (U), the shoshonitic field (S) and the calc-alkaline field (C). The dotted area is the ultrapotassic-shoshonitic volcanic Provinceof Western Anatolia (from Afyon to Bucak) after Dilek and Altunkaynak (2010).

64 B. Platevoet et al. / Journal of Asian Earth Sciences 92 (2014) 53–76

Page 13

Fig. 6. Major (wt%) and selected trace (ppm) elements versus silica diagrams. Full circle dot: lamprophyric mafic dyke; open diamond: plutonic enclave; full diamond:tephriphonolite cycle II lava; square: trachyandesite pumice of cycle I; triangle: trachytic pumice of cycle III; open circle: trachytic dome. A dashed line shows the correlationwithin the pyroxenite group. Values from Table 3.

B. Platevoet et al. / Journal of Asian Earth Sciences 92 (2014) 53–76 65

Page 14

Table

4

Glass

microprobean

alyses.Datahav

ebee

nrecalculatedto

100%beforeplottingin

diagram

s.Se

ealso

Table

3fortherock

-typ

es.Datanam

ed‘‘E

girdir’’arefrom

apyroc

lastic

dep

osit

studiedbyNem

ecet

al.(1998).

98,423

Lamprophyre

06-013

Lamprophyre

05-008

Lava

05-036

Pyroclastite

I

Glass

910

415

58

59

60

61

62

63

64

711

712

56

79

10

11

12

13

14

SiO2

59.20

58.23

54.95

54.55

54.24

55.13

55.72

56.38

56.09

56.34

56.02

61.91

62.85

59.03

59.31

58.69

59.53

59.68

59.91

59.24

60.50

59.17

TiO

20.42

0.39

0.44

0.51

0.45

0.48

0.57

0.34

0.43

0.42

0.26

0.25

0.54

0.43

0.40

0.51

0.32

0.51

0.40

0.44

0.37

0.33

Al 2O3

21.51

21.27

20.12

20.19

20.48

21.14

21.57

21.92

21.53

21.88

21.48

17.45

18.76

19.90

19.46

19.11

21.01

19.32

19.37

19.51

19.71

18.67

Cr 2O3

0.01

0.03

0.00

0.00

nd

nd

0.00

0.00

0.00

0.04

0.00

0.00

0.05

0.01

0.00

0.00

0.00

0.00

0.00

0.10

0.00

0.00

FeOt

1.06

1.03

2.05

1.88

1.61

1.28

1.63

1.19

1.19

1.07

1.28

1.40

1.47

2.96

3.10

3.33

2.99

2.88

2.94

3.09

3.01

2.88

MnO

0.04

0.00

0.11

0.20

0.03

0.60

0.10

0.00

0.00

0.12

0.09

0.12

0.00

0.11

0.09

0.01

0.15

0.11

0.13

0.17

0.14

0.00

MgO

0.01

0.01

0.07

0.06

0.01

0.05

0.03

0.02

0.02

0.00

0.00

1.97

0.99

0.55

0.49

0.49

0.48

0.55

0.51

0.51

0.45

0.48

CaO

0.73

0.79

1.62

1.24

0.81

0.74

0.98

0.80

0.76

0.76

0.83

3.35

2.36

3.32

2.99

2.98

2.64

2.99

3.04

3.04

3.30

3.00

Na 2O

2.78

3.01

5.57

4.20

3.90

3.96

4.19

4.55

3.92

4.12

4.08

3.60

5.04

6.12

5.91

5.72

5.76

5.81

5.81

6.23

6.06

5.72

K2O

7.71

8.12

8.80

9.25

9.16

9.23

7.33

7.32

7.80

7.55

7.27

7.27

6.41

5.62

5.67

5.90

5.63

5.56

5.78

5.69

5.35

5.73

SrO

nd

nd

nd

nd

nd

nd

0.05

0.23

0.05

0.00

0.14

0.23

0.18

0.09

0.50

0.37

0.32

0.00

0.37

0.18

0.59

0.00

BaO

nd

nd

nd

nd

nd

nd

0.13

0.00

0.28

0.10

0.20

0.03

0.33

0.13

0.35

0.08

0.35

0.15

0.13

0.15

0.10

0.33

P2O5

nd

nd

nd

nd

0.09

0.09

0.36

0.04

0.18

0.00

0.04

0.00

0.00

0.08

0.16

0.27

0.06

0.07

0.05

0.02

0.07

0.03

F0.00

0.17

nd

nd

0.30

0.25

0.30

0.20

0.29

0.20

0.28

0.00

0.00

0.06

0.03

0.00

0.06

0.44

0.32

0.09

0.00

0.16

Cl

0.19

0.19

nd

nd

0.06

0.10

0.21

0.21

0.13

0.15

0.08

0.03

0.01

0.19

0.17

0.17

0.25

0.22

0.29

0.26

0.24

0.20

Total

93.66

93.24

93.73

92.08

91.14

93.05

93.17

93.20

92.67

92.75

92.05

97.61

98.99

98.6

98.64

97.62

99.55

98.28

99.06

98.71

99.88

96.69

05-032

Pyroclastite

I05-030

Pyroclastite

III

05-034

Pyroclastite

III

Glass

81

82

23

24

25

26

27

28

15

18

19

20

21

22

22b

32

52

12

34

5

SiO2

58.61

58.35

63.40

64.37

63.77

62.26

60.69

62.31

66.51

62.91

63.21

63.98

64.40

65.27

63.45

64.31

63.12

64.06

62.66

61.36

59.86

64.68

TiO

20.44

0.47

0.16

0.15

0.19

0.24

0.40

0.35

0.05

0.16

0.24

0.12

0.20

0.09

0.41

0.18

0.15

0.19

0.24

0.29

0.13

0.26

Al2O3

20.46

19.54

18.86

18.80

20.61

18.67

17.68

18.88

18.01

16.18

16.32

16.87

16.93

17.28

17.41

17.57

17.13

18.68

18.66

18.67

17.96

18.75

Cr 2O3

0.00

0.00

0.02

0.00

0.00

0.00

0.38

0.03

0.07

0.26

0.25

0.00

0.04

0.04

nd

nd

nd

0.02

0.08

0.05

0.00

0.01

FeOt

3.26

2.87

2.33

2.16

1.70

2.27

2.19

2.29

1.46

1.69

1.47

1.65

1.58

1.96

1.89

1.62

1.64

1.94

2.04

1.93

1.66

1.56

MnO

0.16

0.06

0.00

0.00

0.03

0.00

0.12

0.04

0.08

0.01

0.00

0.17

0.09

0.06

0.11

0.10

0.11

0.03

0.04

0.04

0.03

0.07

MgO

0.55

0.50

0.31

0.27

0.25

0.29

0.25

0.32

0.15

0.12

0.10

0.13

0.14

0.44

0.13

0.14

0.15

0.16

0.37

0.19

0.21

0.09

CaO

2.78

3.00

2.21

2.20

2.36

2.08

2.00

1.91

1.25

1.24

1.22

1.27

1.39

2.06

1.30

1.19

1.39

1.59

1.95

1.59

2.71

1.13

Na 2O

6.74

6.17

7.16

6.52

6.35

6.17

5.56

6.28

5.76

5.54

5.46

6.04

5.91

6.04

5.90

5.53

5.78

6.18

6.11

5.84

6.73

5.20

K2O

6.17

6.28

3.90

4.37

3.96

5.29

5.31

5.25

5.45

5.19

4.83

5.09

5.24

5.03

5.65

6.02

5.82

6.06

5.50

5.54

5.35

6.88

SrO

0.16

0.50

0.14

0.36

0.50

0.14

0.09

0.05

0.00

0.00

0.00

0.00

0.00

0.00

0.16

0.05

0.04

0.14

0.27

0.00

0.32

0.32

BaO

0.05

0.16

0.03

0.60

0.65

0.03

0.20

0.05

0.18

0.00

0.25

0.00

0.23

0.00

0.04

0.04

0.00

0.28

0.30

0.15

0.20

0.38

P2O5

0.10

0.07

0.01

0.13

0.06

0.03

0.05

0.12

0.05

0.00

0.00

0.03

0.00

0.00

0.04

0.00

0.00

0.00

0.00

0.07

0.11

0.07

F0.07

0.04

0.00

0.00

0.09

0.00

0.06

0.16

0.13

0.13

0.00

0.19

0.10

0.00

0.28

0.21

0.04

0.16

0.17

0.14

0.32

0.06

Cl

0.22

0.22

0.26

0.27

0.18

0.31

0.27

0.25

0.42

0.39

0.49

0.37

0.50

0.23

0.40

0.42

0.58

0.15

0.12

0.07

1.61

0.02

Total

99.77

98.23

98.75

100.18

100.71

97.77

95.25

98.27

99.58

93.83

93.85

95.91

96.73

98.5

97.17

97.38

95.95

99.62

98.51

95.93

97.2

99.48

Egirdir

Pyroclastite

E3

Xen

olith

E06ÿ005

Xen

olith

similar

toE3

Glass

61

23

45

621

22

23

24

25

14

17

18

23

38

39

SiO2

64.04

66.08

62.79

62.41

64.47

62.44

61.32

61.97

53.75

53.01

52.41

56.46

55.80

54.14

56.04

56.47

55.92

55.83

55.04

TiO

20.36

0.18

0.36

0.48

0.00

0.39

0.47

0.29

0.35

0.20

0.24

0.21

0.33

0.27

0.24

0.21

0.24

0.25

0.34

Al2O3

18.37

19.20

17.34

17.41

19.77

18.12

18.22

19.10

18.66

18.72

19.47

19.72

19.40

21.98

21.58

21.77

21.79

21.88

21.78

Cr 2O3

0.00

nd

nd

nd

nd

nd

nd

nd

0.01

0.00

0.05

0.04

0.00

nd

nd

nd

nd

nd

nd

FeOt

2.14

1.21

2.15

3.37

0.20

2.89

3.27

2.46

2.45

2.42

2.25

2.52

2.74

1.37

1.43

1.53

1.44

1.43

1.43

MnO

0.17

0.00

0.11

0.17

0.00

0.16

0.17

0.10

0.06

0.05

0.10

0.05

0.08

0.04

0.08

0.12

0.11

0.09

0.16

MgO

0.23

0.08

0.12

0.35

0.02

0.31

0.12

0.09

0.76

0.77

0.50

0.65

0.78

0.09

0.08

0.11

0.06

0.09

0.07

CaO

1.76

1.15

1.42

1.65

0.60

1.68

1.49

0.98

3.47

3.49

3.17

3.37

3.96

1.70

1.76

1.74

1.50

2.18

2.07

Na 2O

6.15

5.69

4.69

4.91

5.25

4.51

4.50

4.68

5.26

5.33

6.02

5.44

5.55

4.71

5.38

5.22

4.90

4.75

5.14

K2O

5.73

6.69

7.49

7.11

7.40

7.01

6.59

7.12

8.22

8.21

8.71

9.07

8.61

8.01

7.07

7.01

7.58

6.47

6.27

SrO

0.00

nd

nd

nd

nd

nd

nd

nd

0.56

0.25

0.56

0.34

0.80

nd

nd

nd

nd

nd

nd

BaO

0.25

nd

nd

nd

nd

nd

nd

nd

0.52

0.34

0.49

0.32

0.41

nd

nd

nd

nd

nd

nd

P2O5

0.02

nd

nd

nd

nd

nd

nd

nd

0.13

0.11

0.11

0.15

0.10

0.00

0.03

0.09

0.04

0.00

0.02

F0.22

nd

nd

nd

nd

nd

nd

nd

0.20

0.00

0.00

0.00

0.00

0.44

0.13

0.14

0.15

0.15

0.20

Cl

0.24

nd

nd

nd

nd

nd

nd

nd

0.23

0.16

0.26

0.21

0.20

0.28

0.24

0.24

0.23

0.24

0.25

Total

99.67

100.28

96.47

97.86

97.71

97.51

96.15

96.79

94.65

93.05

94.34

98.55

98.76

93.03

94.06

94.65

93.96

93.36

92.77

66 B. Platevoet et al. / Journal of Asian Earth Sciences 92 (2014) 53–76

Page 15

In the ultramafic–mafic plutonic xenoliths (Fig. 8a), the less U,Th, REE-enriched patterns are those where Cpx and apatite are veryabundant. This is in accordance with trace element patterns ofindividual Cpx obtained with LA–ICP–MS analyses (Fig. 8b,Table 3b). The great variation of the ultramafic–mafic xenolith pat-terns is explained by variable ratios between cumulus phases (Cpx,mica, apatite) and intercumulus ones (feldspars and/or residualglass). Monzonite and leucocrate monzosyenite enclave patternsare very similar to those of tephriphonolites and Cycle I pumices,trachytes and Cycle III pumices, suggesting a coeval evolution forboth lavas and plutonic xenoliths. However, they also evidencelarge variations: the most Ba and Sr-enriched xenoliths are thosewith large proportions of cumulus feldspar, and the most Rb, Th,U-enriched xenoliths are those with high residual liquid contentsafter feldspar removal. They also display pronounced REE, Ta, Nbdepletions correlated with sorting of amphibole and REE-bearingaccessories (titanite and apatite).

Whole-rock REE patterns normalized to CI chondrite (Table 3,Fig. 9a) are very similar, from lamprophyre to trachyte, with a pro-nounced LREE enrichment. All patterns are more or less MREE-depleted relative to La, Ce and Yb, Lu. This MREE depletion is morepronounced in trachytic lavas and pumices of Cycle III, comparedto trachyandesitic to trachytic pumices of Cycle I. The global REEcontents display some variations linked to Ap + Tit + Amph frac-tionation effects (Flohr and Ross, 1990; Wilson et al., 1995;Hughes et al., 1997; Tiepolo et al., 2007). The complete lack of anEu anomaly, even in the most evolved lava types, suggests that oxi-dizing conditions were prevalent during feldspar crystallizationand sorting (Drake and Weill, 1975).

Ultramafic–mafic xenoliths have a wide range of patterns withenrichment factors from 100 to 1000, correlated with Cpx and apa-tite modal fractions. REE patterns of Cpx (Fig. 9b, Table 3b) showsimilarities with those of the ultramafic xenoliths: note the relativelower values of La, Ce, Pr, Nd for both ultramafic–mafic xenolithsand Cpx. Monzonitic to syenitic xenoliths also show MREE deple-tion and no Eu anomaly. REE decrease in the most evolved syenitesand their patterns are very similar to Cycle III trachytes and relatedpumices.

8.4. Sr–Nd–Pb isotopes

Isotopic compositions of Gölcük rocks were first investigated byAlici et al. (1998) and were revisited by Dilek and Altunkaynak(2010) and Elitok et al. (2010). We add two new analyses of lam-prophyres. Sr, Nd and Pb isotopic data (Table 5) concern both thesub-actual Gölcük volcanism and the older (2–5 Ma) extra calderavolcanism (see Elitok et al., 2010). The less evolved lavas of Gölcükgenerally display relatively lowmeasured 87Sr/86Sr ratios (between0.70374 and 0.704536 for lamprophyres and between 0.70350 and0.70363 for tephriphonolites, Table 5, Fig. 10b). For lamprophyres98,007 and 06–013, initial 87Sr/86Sr ratios are between 0.704492and 0.704519 (for an age of 3 Ma).

The positive correlation between Sr ratios and SiO2 reported forthe ‘‘extra-caldera’’ volcanism (Fig. 10b, triangle) is less apparentfor the more recent Gölcük series (filled diamond and opensquare). All rocks display high Sr contents, higher than 1000 ppmand up to 8000 ppm (Fig. 10c). A crustal (high 87Sr/86Sr, and highSr concentration) contamination is evidenced for the most evolvedrocks (Fig. 10c). Moreover, the most depleted Sr isotope signaturesare clearly associated with the most radiogenic Pb isotope signa-tures suggesting that the mantle sources must exhibit such charac-teristics. However, the contribution of a lithospheric source,metasomatized during previous subduction processes, is also sug-gested by the strong LILE and REE enrichments relative to HFSEevidenced in the trace element patterns. The relatively unradio-genic Sr character of this metasomatised mantle might eitherreflect (i) a young age for the subducted slab and/or the subductionprocess, (ii) a strongly depleted original lithospheric mantle (byprevious melting events), or (iii) a prevalent asthenospheric originof Sr.

9. Discussion

9.1. Relationship between plutonic xenoliths and volcanic series

Gölcük pyroxenites are never associated with peridotite xeno-liths. They never bear olivine, nor high-pressure minerals such asspinel or garnet. Their feldspar-bearing matrix evidences crystalli-zation under ‘‘crustal’’ conditions at relatively low pressure. More-over, pyroxenites and mafic related xenoliths highly differ totypical mantle pyroxenites (Downes, 2007) according to theirREE compositions and trace element patterns. Ultramafic–maficplutonic xenoliths from Gölcük may not be assumed to be mantlexenoliths from a hypothetic metasomatized mantle source (Sßenet al., 2008).

Three main types of plutonic xenoliths have been observed thatdiffer by their texture: (i) Pyroxenites show dominant cumulatetextures with few intercumulus phases crystallizing from atrapped residual liquid. (ii) Mafic monzonitic xenoliths also exhibitcumulate-like features with a set of early euhedral phases in anundercooled felsic-dominant matrix. (iii) Sometimes, they alsoshow a ‘‘sub-doleritic’’ texture with skeletic elongate crystalssymptomatic of rapid crystallization under strong undercooledconditions (Lofgren, 1980, 1983). These ‘‘sub-doleritic’’ xenoliths

Fig. 7. Glass compositions. (a) TAS (total alkali-silica) diagram. Solid line anddashed line separate the strongly silica-depleted rock field (SU), the slightly silica-undersaturated rocks field (U–S) and the silica-oversaturated rock field (OV) (LeMaitre, 2002). See legend for the origin of glass. Grey crosses are for thecorresponding chemical whole rock analyses, MD: mafic dyke; E3: pyroxeniticxenolith; TP: tephriphonolite; TA1: cycle I trachyandesite; T3: cycle III trachyte. Allmicroprobe glass analyses have been recalculated to 100% before plotting (Table 4).(b) Na2O versus K2O (wt%) for glass compositions. Same limits as Fig. 5b.

B. Platevoet et al. / Journal of Asian Earth Sciences 92 (2014) 53–76 67

Page 16

are likely to be close to liquid compositions. Coarse felsic texturesoccur in monzosyenite xenoliths with dominant euhedral feld-spars. A similar sequence of crystallization is observed both in plu-

tonic samples and volcanic rocks (Francalanci et al., 2000), with asimilar mineral succession: Cpx + Phl + Ap + Mgt + Tit + Amph +Feld + Zirc.

Cycle III pumices

1

10

100

1000

10000

Cs Rb Ba Th U K Ta Nb La Ce Sr Nd Hf Zr Sm Eu Ti Tb Y Yb

Trachytic domes

1

10

100

1000

10000

Cs Rb Ba Th U K Ta Nb La Ce Sr Nd Hf Zr Sm Eu Ti Tb Y Yb

Syenitic xenoliths

1

10

100

1000

10000

Cs Rb Ba Th U K Ta Nb La Ce Sr Nd Hf Zr Sm Eu Ti Tb Y Yb

Mafic dykes

1

10

100

1000

10000

Cs Rb Ba Th U K Ta Nb La Ce Sr Nd Hf Zr Sm Eu Ti Tb Y Yb

Ultramafic-mafic xenoliths1

10

100

1000

10000

Cs Rb Ba Th U K Ta Nb La Ce Sr Nd Hf Zr Sm Eu Ti Tb Y Yb

Monzonitic xenoliths

1

10

100

1000

10000

Cs Rb Ba Th U K Ta Nb La Ce Sr Nd Hf Zr Sm Eu Ti Tb Y Yb

Rock / Primitive mantle

Cpx / Primitive mantle

0.01

0.10

1.00

10.00

100.00

Rb Ba Th U Nb Ta La Ce Pb Pr Sr Nd Zr Hf Sm Eu Ti Gd Tb Dy Y Ho Er Tm Yb Lu

Cpx1 Core

Cpx2 Core

Cpx3 Core

a

b

Tephriphonolites II1

10

100

1000

10000

Cs Rb Ba Th U K Ta Nb La Ce Sr Nd Hf Zr Sm Eu Ti Tb Y Yb

Cycle I pumices

1

10

100

1000

10000

Cs Rb Ba Th U K Ta Nb La Ce Sr Nd Hf Zr Sm Eu Ti Tb Y Yb

Fig. 8. Primitive mantle normalized diagrams for trace elements. (a) Whole rock patterns; normalizing value from McDonough and Sun (1995). (b) Normalized diagrams forselected trace elements in clinopyroxene from the pyroxenite xenolith 06-07-E (Table 3b).

68 B. Platevoet et al. / Journal of Asian Earth Sciences 92 (2014) 53–76

Page 17

Trachytic domes cycle III

1

10

100

1000

La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu

Pumices cycle III

1

10

100

1000

La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu

Cycle I pumices

1

10

100

1000

10000

La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu

Tephriphonolites II

1

10

100

1000

10000

La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu

Mafic dykes

1

10

100

1000

10000

La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu

Syenitic xenoliths

1

10

100

1000

10000

La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu

E 10A

97-12

97-77

E 10C

97-20

Ultramafic-mafic xenoliths1

10

100

1000

La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu

E-3

E-6-PX

S10

E 10B

06-07-E

Monzonitic xenoliths

1

10

100

1000

La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu

97-66

97-60

05-014b

Rock / CI chondrite

1.00

10.00

100.00

1000.00

La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu

Cpx1 c core

Cpx1 b rim

Cpx2 c core

Cpx3 c core

Cpx3 b rim

Cpx / CI chondrite

a

b

Fig. 9. REE chondrite CI-normalized diagrams. (a) Whole rock analyses. (b) clinopyroxenes of the pyroxenitic xenolith 06-07E (Table 3).

B. Platevoet et al. / Journal of Asian Earth Sciences 92 (2014) 53–76 69

Page 18

Main mineral compositions are very similar in plutonic and vol-canic rocks. The persistence of magnetite along with titanite forboth rock-types also evidences high oxidizing conditions duringmagma crystallization. Calcic amphibole crystallization prior tofeldspars in both rock types requires a minimum PH2O of 0.25GPa (Holloway and Burnham, 1972) and phlogopite crystallizationprior to feldspars requires an even more elevated PH2O. Accordingto the amphibole geobarometer (Schmidt, 1992; Anderson andSmith, 1995, for temperature correction), total pressure calcula-tions using the amphibole Al content for trachyandesitic andtrachytic silica-oversaturated pumices (samples 05-036 and 05-030) give a total pressure range between 0.3 and 0.6 GPa, givinga depth of at least 9 km. This is higher than those (Ptotal = 0.1 GPa)assumed by Kumral et al. (2006) from their calculation of liquidusphase succession (clinopyroxene, feldspar, amphibole) using theMELTS algorithm of Ghiorso and Sack (1995). Geochemical dataof lavas and xenoliths also show similar global trace element pat-terns, illustrating processes of mineral fractionation / sorting andsuggesting a common parent magma (Figs. 8 and 9). The majorityof plutonic xenoliths may be interpreted as cumulates or partiallycumulative rocks. Accordingly, these xenoliths are interpreted ascognate plutonic fragments sampled from the crystallizing andcumulative zones close to the magma chamber boundaries, under-neath the Gölcük volcano. Plutonic xenoliths were mostly sampledduring the last phreato-magmatic events of Cycle III, during whichhighly vesiculated trachytic magma quickly rose from a relativelydeep magma chamber, and then met a shallow phreatic reservoirleading to maar-type eruptions.

9.2. Parent magma and fractionation processes

Tephriphonolites and lamprophyres display great mineralogicalaffinities with plutonic xenoliths, but tephriphonolites are too

evolved to represent primary magma of the series. Lamprophyres,which are the least evolved rock-type, plot between the pyroxenitecumulates and tephriphonolites in most diagrams (see Figs. 5 and6), but some geochemical discrepancies concerning TiO2, Zr and Hf,are apparent in these diagrams. The presence of lamprophyricdykes in the same area as the one with evolved trachyandesiticand trachytic rocks, indicates that some mafic melts might havebeen produced from the mantle slightly before the last volcanicepisodes of Gölcük volcano. Moreover, some mafic monzonitexenoliths have been found in the most recent trachytes. Their tex-tures are characteristic of rapid crystallization, suggesting min-gling between trachytic magma and a less evolved magma withlamprophyric affinities. Comingling features between lamprophyreand evolved magmas have been described elsewhere (Pagel andLeterrier, 1980; Sabatier, 1980; Plà Cid et al., 2003). Lamprophyricmagma cannot be totally ruled out as possible mafic component ofthe Gölcük series.

Monzonite xenoliths are also less evolved than the lavas of thetephriphonolite group (Fig. 5). Moreover, some of these xenolithsdisplay an ophitic texture, without any significant accumulatingprocess of early minerals and seem to be close to the magma com-position. To test this hypothesis, we use a monzonite (monzonite97-60, Table 3a) as normalizing value (Fig. 11a and b); the pyroxe-nite xenoliths show depleted incompatible trace element patterns(Th, U, Hf, Zr) and selective Rb, Ba, K and Ti enrichments comparedto the monzonite reference. These geochemical features are symp-tomatic of Cpx and phlogopite accumulation. Strong variation inREE is also linked to variable apatite accumulation. The moreevolved monzosyenite xenoliths display the reverse patterns(Fig. 11a and b) with a global enrichment compared to the hypo-thetical monzonitic parent, strong variations in Ba and Sr due tovariable feldspar content, and relative depletion of MREE due toamphibole-titanite-apatite removal. The monzosyenite xenoliths

Table 5

Sr, Nd and Pb isotope ratios.

Sample Rock-type Rb (ppm) Sr (ppm) 87Sr/86Sr Nd (ppm) Sm (ppm) 143Nd/144Nd Pb (ppm) U (ppm) Th (ppm) 206Pb/204Pb 207Pb/204Pb 208Pb/204Pb

98,007 Lamprophyre c 197 1394 0.704536 82 13 0.512688 21 6 2206-013 Lamprophyre c 182 1433 0.704508 81 13 0.512691 20 6 22GI-40 Lamprophyre a 119 3144 0.70374 114 16 0.51272 6 5 31 19.45 15.65 39.39GI-70 Lamprophyre a 42 3068 0.70377 114 17 0.51274 8 5 34 19.41 15.65 39.33GI-103 Tephriphonolitea 132 5433 0.70361 182 26 0.51274 3 13 56 19.45 15.66 39.26GI-93 Trachyte cycle III a 159 3616 0.70388 103 14 0.51271 5 18 79 19.33 15.68 39.27IG-5 Tephriphonoliteb 107 4498 0.70363 – – – – – – – – –IG-18 Tephriphonoliteb 110 5684 0.70353 – – 0.51278 – – – – – –IG-19 Tephriphonoliteb 109 5395 0.70350 – – – – – – – – –I-92-6 Tephriphonoliteb 117 4075 0.70360 – – 0.51276 – – – – – –BC-1 Bucak lamproitea 239 2757 0.70386 104 14 0.51269 14 6 24 19.54 15.65 39.49GI-2 Trachytea 144 2431 0.70412 81 12 0.51267 53 11 44 19.31 15.65 39.23GI-5 Trachy-andesitea 164 3097 0.70366 112 13 0.51273 11 14 71 19.49 15.64 39.44GI-8 Trachytea 174 1315 0.70504 51 8 0.51264 7 11 45 19.32 15.65 39.2GI-28 Trachytea 183 2237 0.70417 57 8 0.51268 23 15 49 19.28 15.65 39.21GI-45 Trachytea 191 1374 0.70485 62 9 0.51264 35 16 55 19.27 15.66 39.17GI-50 Trachy-andesite a 143 6121 0.70398 141 20 0.51271 5 13 55 19.19 15.66 39.12GI-82 Trachy-andesite a 98 1494 0.70450 63 10 0.51269 3 9 31 19.26 15.67 39.18GI-104 Trachy-andesitea 130 2203 0.70408 83 13 0.51269 1 10 38 19.41 15.66 39.36GI-105 Trachytea 112 2817 0.70407 54 7 0.51269 4 12 31 19.31 15.66 39.24GI-96 Trachy-andesitea 29 6536 0.70365 215 30 0.51276 19 5 66 19.49 15.66 39.44GI-109 Trachy-andesitea 131 5679 0.70365 189 24 0.51273 2 14 67 19.45 15.66 39.46GI-34 Monzosyenitea 128 3920 0.70386 26 4 0.51268 15 12 45 19.42 15.67 39.38IG-23 Trachyteb 144 2488 0.70384 – – – – – – – – –IG-24 Trachy-andesiteb 106 4172 0.70360 – – – – – – – – –IG-17 Trachy-andesiteb 146 4241 0.70373 – – – – – – – – –IG-1 Trachy-andesiteb 133 3072 0.70383 – – – – – – – – –I-93-46 Trachy-andesiteb 124 4942 0.70366 – – 0.51276 – – – – – –IG-3 Trachyte cycle IIIb 151 3314 0.70370 – – – – – – – – –IG-4 Trachyte cycle IIIb 150 3256 0.70384 0.51273 – – – – – –

All 87Sr/86Sr and 143Nd/144Nd in this table are measured ratios.a Data from Elitok et al. (2010).b Data from Alici et al. (1998) and Dilek and Altunkaynak (2010)c New data (this paper).

70 B. Platevoet et al. / Journal of Asian Earth Sciences 92 (2014) 53–76

Page 19

can be interpreted as feldspar cumulates with a variable ratio oftrapped residual liquid. The most evolved monzosyenite xenolithsare very close to trachyte composition and are Ba, Sr, Ti and REE-depleted due to feldspar, magnetite and titanite sorting. During thislast step of the fractionation process, feldspar sorting is already pre-valent upon Cpx, phlogopite, amphibole and accessories.

9.3. Coexistence of silica-undersaturated and silica-oversaturated

magmas

Several hypotheses have been explored to explain the coexis-tence of silica-oversaturated and undersaturated series at the sametime. (i) A crustal contamination of a strongly alkaline basaniticparental magma could explain the occurrence of both series(Fitton and Upton, 1987; Briot et al., 1991). (ii) Two basaltic mag-mas, a weakly alkaline one and a strongly alkaline one, may resultfrom variable mantle partial melting and/or heterogeneous mantlesources (Bonin and Giret, 1985; Wilson et al., 1995). In the Mio-cene-Quaternary Afyon-Isparta magmatic province, strongly alka-line undersaturated rocks coexist with rhyolitic ignimbrites(Keller and Villary, 1972; Savasçin et al., 1994), with at least twodistinct mantle sources and a third crustal component. At Gölcükvolcano, strongly undersaturated tephriphonolite coexists withsilica-oversaturated trachyte, but without any basaltic lava as apotential primary magma. The role of silica-poor hydrous mineralssuch as amphibole and mica has often been discussed to explainthe silica content increase and the production of silica-oversaturat-ed residual liquids from undersaturated basaltic magmas (Leterrierand Maury, 1978; Maury et al., 1980; Bonin and Giret, 1985; Fittonand Upton, 1987; Wilson et al., 1995). A sorting mixture with dom-inant Cpx, equivalent to the Gölcük pyroxenites, can lead to a

residual tephriphonolite liquid from a tephritic or lamprophyricparent magma (Fig. 12). Such cumulates progressively increasethe silica-undersaturation character of the successive magmas, asevidenced by the occurrence of strongly silica-undersaturatedtephriphonolites and residual glasses in the incompletely crystal-lized pyroxenites (Fig. 7). Reversely, a sorting mixture of prevalentphlogopite and/or amphibole and feldspars may lead to silica-saturation and oversaturation of the Gölcük residual magma. Inmagmas of intermediate compositions, amphibole becomesprevalent upon phlogopite. The paragenesis of mafic monzonitexenoliths is in agreement with the successive sorting of phlogopite,amphibole, and feldspar. Trachyandesitic pyroclasts (cycle I) alsohave similar phenocryst assemblage where amphibole is dominantupon phlogopite.

As a whole, a liquid line of descent and a correlative ‘‘cumulateline’’ corresponding to the main cumulative plutonic types mayillustrate the evolution from silica-undersaturated to silica-over-saturated evolved liquids (Fig. 12). At high PH20, this process wasoperative frommafic-intermediate magmas first by Cpx and phlog-opite sorting, then by amphibole and feldspars sorting. This is con-firmed by the abundance of euhedral amphibole phenocrysts intrachyandesitic pyroclastites and in some plutonic xenoliths (seeFig. 2, photo F). This crystal fractionation scheme is different fromthe model assumed by Kumral et al. (2006), who calculated thatphlogopite and amphibole played a minor role compared to Cpxand feldspars.

9.4. Explosive volcanism and deep fractionation processes

High PH2O is the main factor governing the crystallization ofhydrous minerals (Phl and Amph) prior to feldspar. Phlogopite is

Fig. 10. 143Nd/144Nd and 87Sr/86Sr variations in Gölcük magmatism (whole rock analyses). (a) 143Nd/144Nd versus 87Sr/86Sr diagram. Open triangle: extra-caldera volcanism;full dot: lamprophyric mafic dyke; full diamond: tephriphonolite; open square: trachyandesite and trachyte; open diamond: monzosyenitic plutonic xenolith (see Table 5).Full line: Mantle array; Dashed line: correlation within the extra-caldera volcanism. Data after Elitok et al. (2010), Dilek and Altunkaynak (2010), Alici et al. (1998) and thispaper. (b) SiO2 (wt%) versus 87Sr/86Sr isotopic ratio diagram. Same symbols as (a); Square: trachyandesites cycle I. (c) Sr (ppm) versus 87Sr/86Sr isotopic ratio diagram. Samesymbols as (b). (d) 206Pb/204Pb versus 87Sr/86Sr isotopic ratio (whole rock): same symbols; star: Bucak lamproite. See Table 5.

B. Platevoet et al. / Journal of Asian Earth Sciences 92 (2014) 53–76 71

Page 20

always the first phase to crystallize in Gölcük lamprophyre andtephriphonolite. In trachyandesitic flow-domes phlogopites aregenerally destabilized and, in trachyandesitic pyroclastites, amphi-bole becomes the main hydrous mineral in place of Phl. Phlogopiteor amphibole crystallization strongly depends of the K2O/Na2Oratio of magma. Cpx and Phl are the main liquidus phases of tephri-phonolite and pyroxenite xenoliths. During Cpx and Phl sorting,the K2O/Na2O ratio rapidly decreases in the liquid, leaving amphi-bole as the main H2O-bearing phase in place of Phl in the trachy-andesitic residual magma. In Cycle I trachyandesitic pyroclastics,Amph is always the main hydroxyl-bearing phenocryst in placeof Phl. This change may strongly influence the H2O saturationbehaviour because the H2O content of amphibole is less than thoseof phlogopite. Amphibole nucleation and growth in place of phlog-opite crystallization (with Cpx) lead to increase PH2O in the magmachamber leading quickly to H2O saturation and vesiculation in theresidual liquid (Bardintzeff and Bonin, 1987). The main explosiveCycle I may have been triggered at fluid saturation by dominant

Fig. 11. Normalized trace elements patterns of plutonic enclaves. The values of the monzonite 97-60 are used for normalization; (a) pyroxenitic group and twomonzosyeniticxenoliths (samples 97-66 and 05-014b) for comparison; (b) monzosyenitic and syenitic groups; full diamond: pyroxenitic group, grey diamond: monzosyenitic group; opendiamond: syenitic group. See also Table 3.

Fig. 12. Illustration of fractionation processes in the total alkali – silica diagram.Solid line: partition line between strongly silica-undersaturated rock domain (US)and slightly-undersaturated to over-saturated rock domain (OS). Grey dashedcurve: assumed liquid line of descent; dotted curve: assumed successive cumu-lates; stars: main pure cumulate types. Main sorting minerals: clinopyroxene (cpx),mica, amphibole (amph), K-feldspar (KF).

72 B. Platevoet et al. / Journal of Asian Earth Sciences 92 (2014) 53–76

Page 21

amphibole crystallization upon phlogopite in trachyandesiticmagma. This cycle ended with the less evolved tephriphonoliteflow-domes protruding within the caldera. Within a stratifiedmagma chamber, the tephriphonolite magma forms a H2O-under-saturated magma layer beneath the H2O-saturated trachyandesiticmagma whose vesiculation is responsible of the main explosiveevents.

Mixing of mafic and felsic magmas is commonly regarded as amajor factor for triggering explosive eruptions in zoned, oftensilicic magma reservoirs (Anderson, 1976; Sparks et al., 1977;McBirney, 1980; Eichelberger, 1980; Marshall and Sparks, 1984;Sparks and Marshall, 1986; Platevoet and Bonin, 1991;Bardintzeff, 1992; Snyder and Tait, 1996; Wiebe, 1996; Peruginiand Poli, 2000; Viccaro et al., 2006; Kratzmann et al., 2009). Someevidence of mafic-felsic glass mingling has been described for theGölcük volcano by Nemec et al. (1998) in tephra deposits near Egir-dir Lake, as well as in the last trachytic protrusions (this work). AtGölcük volcano, evidence of magma mingling within the chamberexists but remains relatively rare. The major role was played by thefractionation process, such as phlogopite-amphibole succession,that involved H2O saturation and a rapid fluid pressure increaseunder the roof of the chamber, triggering subsequent explosiveeruptions.

9.5. Hypothetical magma sources

Strong LILE and REE enrichment relative to HFSE anomalies forlong have been related to convergent plate-tectonic settings(Pearce and Norry 1979; Pearce, 1982, 1983), and are symptomaticof lithospheric mantle metasomatism due to subduction processes.These geochemical features are observed elsewhere in the Mioceneto Quaternary calc-alkaline, potassic and ultrapotassic volcanismof Western Anatolia. Many authors regard the metasomatizedlithospheric mantle as a major source component of this potassicto ultrapotassic magmatism (Guleç, 1991; Foley, 1992a, 1992b;Savasçin et al., 1994; Seyitoglu et al., 1997; Aldanmaz et al.,2000; Francalanci et al., 2000; Prelevic et al., 2008; Akal, 2008;Ersoy et al., 2010; Prelevic et al., 2010a, 2010b, 2012). Gölcük lavasand associated plutonic xenoliths share similar chemical character-istics. However, the negative HFSE anomalies relative to LILE andREE are not so prominent in the most mafic Gölcük lavas, andare emphasized, due to mineral fractionation, in more evolvedlavas and in the syenite enclave group. Moreover, initial 87Sr/86Srand 143Nd/144Nd ratios remains in the ‘‘depleted quadrant’’ (fig.10a) for the great majority of Gölcük rocks. These rocks are slightlymore radiogenic in Sr than the Quaternary Kula basalts (Alici et al.,2002), but have substantially less radiogenic Sr than the Mioceneultrapotassic, shoshonitic and high-potassic (USH) volcanism ofWestern Anatolia (Agostini et al., 2007; Ersoy et al., 2010).

In the less evolved Gölcük rocks (lamprophyre and tephriphon-olite), low Zr/Nb ratios relative to Nb/Ta ratios (Table 3a, Fig. 13a)suggest a still more Nb-enriched mantle source than the USH Mio-cene volcanism of Western Anatolia (Ersoy et al., 2010) and theRoman Province, Italy (Peccerillo et al., 1984; Beccaluva et al.,1991). The mantle source(s) of Gölcük lavas thus appear(s) to havea relatively enriched asthenospheric component (Savasçin et al.,1994; Francalanci et al., 2000; Chakrabartia et al., 2011). High206Pb/204Pb ratios relative to low 87Sr/86Sr ratios compared to southEuropean lamproites (Prelevic et al., 2010a) show that the appar-ent asthenospheric component has some affinity with the OIB pole.High 207Pb/204Pb ratios relative to 206Pb/204Pb ratios compared tothe Macedonian lamproites (Fig. 13b; Prelevic et al., 2010a) seemto confirm the dual asthenospheric-lithospheric origin of Gölcükvolcanism with a dominant asthenospheric component; this wasalso suggested for Vulture volcanism compared to the Roman Prov-ince volcanism (Rogers et al., 1985; De Fino et al., 1986).

However, some peculiar geochemical features characterize theGölcük volcanism. These include most notably: (i) very high Ba,Sr, Th and LREE contents, (ii) HFSE as high as in continental or oce-anic within-plate magmatism, (iii) K and HFSE depletion relative toLILE and LREE, (iv) low 87Sr/86Sr. All these characteristics are fea-tures more commonly found in carbonatites and related rocks(Nelson et al., 1988; Woolley and Kempe, 1989; Bailey, 1990;Woolley and Ross, 1990; Gwalani et al., 2010; Bell and Simonetti,2010; Doroshkevich et al., 2010; Tappe et al., 2013), as previouslysuggested by Prelevic et al. (2010b) for the South European lam-proitic magmatism involving asthenospheric carbonatitic meltsinteracting with the partially melted lithospheric mantle. Carbon-atitic fluids of sub-lithospheric origin, or generated by partial melt-ing of an asthenospheric enriched mantle (Wyllie et al., 1990;Dallou et al., 2009; Bell and Simonetti, 2010), might have playeda significant role in the Gölcük volcanism.

High-potassic melt of shoshonitic affinity has been producedexperimentally at relatively low pressure (P < 1.5 GPa) duringdecompression melting of a metasomatized lithosphere(Conceição and Green, 2004). The produced melt has a silica-over-saturated shoshonite (trachyandesitic) composition in whichphlogopite crystallized during the destabilization melting of earlyolivine. However, lamproites and strongly K-rich undersaturatedtypes such as leucitites may not be explained by this process alone,as they coexist with more evolved silica-saturated compositions inthe nearby northern Afyon district (Akal, 2003, 2008; Prelevic et al.,

Fig. 13. Zr/Nb versus Nb/Ta and 207Pb/204Pb versus 206Pb/204Pb diagrams. (a) U–S–H(ultrapotassic, shoshonitic and high-potassic) West Anatolia primitive volcanic fieldafter Ersoy et al. (2010). DMM: depleted MORB mantle; U-DMM, E-DMM: ultradepleted DMM and enriched DMM (from Workman and Hart, 2005); PM1:primitive mantle composition from Palme and O’Neill (2002); PM2: primitivemantle composition from Hofmann (1988), McDonough and Sun (1995). Dashedellipse: least evolved rocks from Gölcük volcano. Black dashed line: limit betweenhypothetical depleted and enriched mantle sources. Full dot: lamprophyric rock;full diamond: tephriphonolite; triangle: extra-caldera volcanism (Elitok et al.,2010). (b) 207Pb/204Pb versus 206Pb/204Pb diagram. Grey: M. L. area = MacedonianLamproïte area after Prelevic et al. (2008); white area: MORB to OIB basaltsdistribution. Same symbols as in (a); square: trachyandesitic pumice; star: Bucaklamproite from Dilek and Altunkaynak (2010).

B. Platevoet et al. / Journal of Asian Earth Sciences 92 (2014) 53–76 73

Page 22

2008) and southern Isparta and Bucak areas (Lefèvre et al., 1983;Guillou, 1987; Coban and Flower, 2006). At least, several distinctmantle sources can be inferred to have collectively contributedto the regional volcanism in this complex post-collisional tectonicsetting. The volcanics of the young Gölcük edifice indicate that thepresent-day source may have a main asthenospheric component(Savasçin and Oyman, 1998) and that the resulting magma is prob-ably of OIB alkaline basaltic type. A possible contribution of car-bonatitic fluid is considered, interacting with remnants of adelaminated lithosphere (Cvetkovic et al., 2004; Prelevic et al.,2010b). The role of this carbonatitic fluid has been decreasing sinceMiocene times and from northern to southern part of the Afyon-Isparta Miocene-Quaternary volcanism.

10. Concluding remarks

(1) Plutonic xenoliths and volcanic rocks found at Gölcük vol-cano show great mineralogical and geochemical affinities.They came from the same magma chamber emplaced atcrustal depth. Great majority of xenoliths are ultramafic–mafic and felsic cumulates or partially cumulative rocksextracted from the chamber boundaries.

(2) Fractional/sorting processes, first governed by clinopyrox-ene, phlogopite, apatite, and then by amphibole, feldsparsand others accessories, can explain the whole set ofxenoliths and the differentiation of the volcanic series. Thesilica-saturation of the magma at the end of the fraction-ation process seems to have been governed by the ratiobetween anhydrous and hydrous phases under high PH2O.

(3) The phlogopite-amphibole succession, governed by thedecrease of the K2O /Na2O ratio in the liquid, seems to beresponsible of the confined fluid pressure rising at the topof the chamber, triggering the main explosive events (CycleI MPFD) of the volcano.

(4) The geochemical features of Gölcük magmatism argue for aprevalent asthenospheric source with possible, but decreas-ing, lithospheric components, as potassic melts generatedfrom remnants of the delaminated metasomatized litho-sphere. The possible role played by deeply-sourced carbon-atitic liquids, and their interaction with alkaline OIB-typebasaltic melt, is questionable but need further evaluation.

Acknowledgements

This work was supported by a grant fromMAE, France and Tübi-tak, Turkey operated through the CNRS-Tübitak project n°18066.We thank particularly the authorities of the Sûleyman Demirel Uni-versity of Isparta for providing logistical support for fieldwork andtransport. The UMR GEOPS-CNRS 8148 (Orsay) and UMS LMC 14(Saclay) are thanked for laboratory facilities. The LSCE (UMR8212, CNRS-CEA-UVSQ) is also thanked for laboratory support.A.R. McBirney is thanked for fruitful remarks. Comments and sug-gestions on an earlier version of this manuscript by D. Prelevic wereof great aid. We are grateful to two anonymous reviewers for usefulremarks.

Appendix A. Supplementary material

Supplementary data associated with this article can be found, inthe online version, at http://dx.doi.org/10.1016/j.jseaes.2014.06.012.

References

Agostini, S., Doglioni, C., Innocenti, F., Manetti, P., Tonarini, S., Savasßçin, M.Y., 2007.The transition from subduction-related to intraplate Neogene magmatism inthe Western Anatolia and Aegean area, In: Beccaluva, L., Bianchini, G., Wilson,M. (Eds.), Cenozoic Volcanism in the Mediterranean Area. Geological Society ofAmerica Special Paper 418, pp. 1–15.

Akal, C., 2003. Mineralogy and geochemistry of melilite leucitites, Balçikhisar,Afyon, Turkey. Turkish J. Earth Sci. 12, 215–239.

Akal, C., 2008. K-richterite-olivine-phlogopite-diopside-sanidine lamproites fromthe Afyon volcanic province, Turkey. Geol. Mag. 145 (4), 570–585.

Akal, C., Helvaci, C., Prelevic, D., van den Bogaard, P., 2013. High-K volcanism in theAfyon region, western Turkey: from Si-oversaturated to Si-undersaturatedvolcanism. Int. J. Earth Sci. 102 (2), 435–453.

Alçiçek, M.C., Brogi, A., Capezzuoli, E., Liotta, D., Marco Meccheri, M., 2013.Superimposed basin formation during Neogene-Quaternary extensionaltectonics in SW-Anatolia (Turkey): insights from the kinematics of the DinarFault Zone. Tectonophysics 608, 713–727, erratum 617, 150, 2014.

Aldanmaz, E., Pearce, J.A., Thirlwall, M.F., Mitchell, J.G., 2000. Petrogenetic evolutionof late Cenozoic, post-collision volcanism in western Anatolia, Turkey. J.Volcanol. Geoth. Res. 102 (1–2), 67–95.

Alici, P., Temel, A., Gourgaud, A., Kieffer, G., Gündogdu, N., 1998. Petrology andgeochemistry of potassic rocks in the Gölcük area (Isparta, SW Turkey): genesisof enriched alkaline magmas. J. Volcanol. Geoth. Res. 65, 1–24.

Alici, P., Temel, A., Gourgaud, A., 2002. Pb–Nd–Sr isotope and trace elementgeochemistry of Quaternary extension-related alkaline volcanism: a case studyof Kula region (western Anatolia, Turkey). J. Volcanol. Geoth. Res. 115 (3–4),487–510.

Anderson, A.T., 1976. Magma mixing: petrological process and volcanological tool. J.Volcanol. Geoth. Res. 1, 3–33.

Anderson, J.L., Smith, D.R., 1995. The effects of fO2 on the Al-in-hornblendebarometer. Am. Mineral. 80, 549–559.

Bailey, D.K., 1990. Mantle carbonatite eruptions: crustal context and implications.Lithos 26 (1–2), 37–42.

Bardintzeff, J.M., 1992. Magma mixing processes in volcanic contexts, athermodynamical approach with the examples of St. Vincent Soufrierevolcano, West Indies and Cerro Chiquito, Guatemala. Terra Nova 4, 553–566.

Bardintzeff, J.M., Bonin, B., 1987. The amphibole effect: a possible mechanism fortriggering explosive eruptions. J. Volcanol. Geoth. Res. 33, 255–262.

Beccaluva, L., Di Girolamo, P., Serri, G., 1991. Petrogenesis and tectonic setting of theRoman volcanic Province, Italy. Lithos 26, 191–221.

Bell, K., Simonetti, A., 2010. Source of parental melts to carbonatites-critical isotopicconstraints. Mineral. Petrol. 98, 77–89.

Bonin, B., Giret, A., 1985. Clinopyroxene compositional trends in oversaturated andundersaturated alkaline ring complexes. J. Afr. Earth Sci. 3, 175–183.

Brastad, K., 1985. Sr metasomatism, and partition of Sr between the mineral phasesof a meta-eclogite from Bjorkedalen, West Norway. Tschermaks Mineralogischeund Petrogrographische Mitteilungen 34, 87–103.

Briot, D., Cantagrel, J.M., Dupuy, C., Harmon, R.S., 1991. Geochemical evolution incrustal reservoirs: trace element and Sr–Nd–O isotopic variations in twocontinental intraplate series at Monts Dore, Massif Central, France. Chem. Geol.89, 281–303.

Cameron, M., Papike, J.J., 1981. Structural and chemical variations in pyroxenes. Am.Mineral. 66, 1–50.

Cawthorn, R.G., Collerson, K.D., 1974. The recalculation of pyroxene end-memberparameters and the estimation of ferrous and ferric iron content from electronmicroprobe analysis. Am. Mineral. 59, 1203–1208.

Cengiz, O., Sener, E., Yagmurlu, F., 2005. A satellite image approach to the study oflineaments, circular structures and regional geology in the Golcuk Craterdistrict and its environs (Isparta, SW Turkey). J. Asian Earth Sci. 27 (2), 155–163.

Chakrabartia, R., Basu, A.R., Ghatak, A., 2011. Chemical geodynamics of WesternAnatolia. Int. Geol. Rev., iFirst article, pp. 1–22.

Coban, H., 2005. New geochronologic, geochemical and isotopic constraints on theevolution of plio-quaternary alkaline volcanism from Isparta district, SWTurkey. International Symposium on the Geodynamics of EasternMediterranean: Active Tectonics of the Aegean Region, 15–18 June 2005,Istanbul, Turkey.

Coban, H., Flower, M.F.J., 2006. Mineral phase compositions in silica-undersaturatedleucitite lamproites from the Bucak area, Isparta, SW Turkey. Lithos 89, 275–299.

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

Cvetkovic, V., Prelevic, D., Downes, H., Jovanovic, M., Vaselli, O., Pecskay, Z., 2004.Origin and geodynamic significance of Tertiary post-collisional basalticmagmatism in Serbia (Central Balkan Peninsula). Lithos 73, 161–186.

Dallou, C., Koga, K.T., Hammouda, T., Poitrasson, F., 2009. Trace element partitioningbetween carbonatitic melts and mantle transition zone minerals: Implicationsfor the source of carbonatites. Geochim. Cosmochim. Acta 73, 239–255.

De Fino, M., La Volpe, L., Peccerillo, A., Piccaretta, G., Poli, G., 1986. Petrogenesis ofMonte Vulture volcano (Italy): inferences from mineral chemistry, major andtrace element data. Contrib. Mineral. Petrol. 92, 135–145.

Dilek, Y., Altunkaynak, S., 2010. Geochemistry of Neogene-Quaternary alkalinevolcanism in western Anatolia, Turkey, and implications for the Aegean mantle.Int. Geol. Rev. 52 (4–6), 631–655.

74 B. Platevoet et al. / Journal of Asian Earth Sciences 92 (2014) 53–76

Page 23

Doroshkevich, A.G., Ripp, G.S., Moore, K.R., 2010. Genesis of the Khaluta alkaline-basic Ba–Sr carbonatite complex (West Transbaikala, Russia). Mineral. Petrol.98, 245–268.

Downes, H., 2007. Origin and significance of spinel and garnet pyroxenites in theshallow lithospheric mantle: ultramafic massifs in orogenic belts in WesternEurope and NW Africa. Lithos 99, 1–24.

Drake, M.J., Weill, D.F., 1975. Partition of Sr, Ba, Ca, Y, Eu2+, Eu3+ and other REEbetween plagioclase feldspar and magmatic liquid: an experimental study.Geochim. Cosmochim. Acta 39, 689–712.

Dyulgerov, M., Platevoet, B., 2006. Unusual Ti and Zr aegirine-augite and potassicmagnesio-arfvedsonite in the peralkaline potassic oversaturated Buhovo-Seslavtzi complex, Bulgaria. Eur. J. Mineral. 18, 127–138.

Dyulgerov, M., Platevoet, B., 2009. Comparative mineralogical study of maficminerals from Variscan potassic-alkaline rocks in Stara Planina Mountains,Bulgaria. Rev. Bulgar. Geol. Soc. 70 (1–3), 47–62.

Eichelberger, J.C., 1980. Vesiculation of mafic magma during replenishment of silicicmagma reservoirs. Nature 288, 446–450.

Elitok, Ö., Özgür, N., Drüppel, K., Dilek, Y., Platevoet, B., Guillou, H., Poisson, A.,Scaillet, S., Satir, M., Siebel, W., Bardintzeff, J.M., Deniel, C., Yilmaz, K., 2010.Origin and geodynamic evolution of late Cenozoic potassium-rich volcanism inthe Isparta area, southwestern Turkey. Int. Geol. Rev. 52 (4–6), 454–504.

Ersoy, E.Y., Helvaci, C., Palmer, M.R., 2010. Mantle source characteristics andmelting models for the early-middle Miocene mafic volcanism in WesternAnatolia: implications for enrichment processes of mantle lithosphere andorigin of K-rich volcanism in post-collisional settings. J. Volcanol. Geoth. Res.198, 112–128.

Ersoy, E.Y., Palmer, M.R., 2013. Eocene-Quaternary magmatic activity in the Aegean:implications for mantle metasomatism and magma genesis in an evolvingorogeny. Lithos 180–181, 5–24.

Essene, E.J., Clafin, C.L., Giorgetti, G., Mata, P.M., Peacor, D.R., Arkai, P., Rathmell,M.A., 2005. Two-, three- and four-feldspar assemblages with hyalophane andcelsian: implications for phase equilibria in BaAl2Si2O8–CaAl2Si2O8–NaAlSi3O8–KAlSi3O8. Eur. J. Mineral. 17, 515–535.

Fitton, J.G., Upton, B.G.J., 1987. Alkaline Igneous Rocks. Geol. Soc. Spec. Publ. 30.Blackwell, Oxford, 568 pp.

Flohr, M.J.K., Ross, M., 1990. Alkaline igneous rocks of Magnet Cove, Arkansas:mineralogy and geochemistry of syenites. Lithos 26, 67–98.

Foley, S.F., 1992a. Vein-plus-wall-rock melting mechanisms in the lithosphere andthe origin of potassic alkaline magmas. Lithos 28, 435–453.

Foley, S.F., 1992b. Petrological characterization of the source components ofpotassic magmas: geochemical and experimental constraints. Lithos 28, 187–204.

Francalanci, L., Innocenti, F., Manetti, P., Savaçin, M.Y., 2000. Neogene alkalinevolcanism of the Afyon-Isparta area, Turkey: petrogenesis and geodynamicimplications. Mineral. Petrol. 70, 285–312.

Ghiorso, M., Sack, R.O., 1995. Chemical mass transfer in magmatic processes: IV. Arevised internally consistent thermodynamic model for the interpolation andextrapolation of liquid-solid equilibria in magmatic systems at elevatedtemperatures and pressures. Contrib. Miner. Petrol. 119, 197–212.

Gomes, C.B., Moro, S.L., Dutra, C.V., 1970. Pyroxenes from the alkaline rocks fromItapirapuã, São Paulo, Brazil. Am. Mineral. 55, 224–230.

Griffen, D.T., Ribbe, P.H., Gibbs, G.V., 1977. The structure of slawsonite, a strontiumanalog of paracelsian. Am. Mineral. 62, 31–35.

Guillou, H., 1987. Pétrologie du volcanisme pliocène de la région d’Isparta, Turquie.Unpublished Research Master Subject (Mémoire de DEA), Université de Paris-Sud, 60 pp.

Guleç, N., 1991. Crust-mantle interaction in western Turkey: implications from Srand Nd isotope geochemistry of tertiary and Quaternary volcanics. Geol. Mag.128 (5), 417–435.

Gwalani, L.G., Moore, K., Simonetti, A., 2010. Carbonatites, alkaline rocks and themantle. Min. Petrol., Special issue, vol. 98, 1–4, 345 pp.

Hawthorne, F.C., Oberti, R., 2007. Classification of the amphiboles. In: Hawthorne,F.C., Oberti, R., Della Ventura, G., Mottana, A. (Eds.), Amphiboles: CrystalChemistry, Occurrence, and Health Issues, Reviews in Mineralogy andGeochemistry, vol. 67, pp. 55–88.

Helz, R.T., 1973. Phase relations of basalts in their melting ranges at PH2O = 5Kb as afunction of oxygen fugacity I: mafic phases. J. Petrol. 14, 249–302.

Hildenbrand, A., Platevoet, B., Poisson, A., Monath, F., Nauret, F., Guillou, H., Lefèvre,Ch., 1999. The Shoshonitic Volcanic Series of Isparta Triangle, Turkey: Evidencefor Fractionation Processes in a Shallow Reservoir and Mantle Source. EUGStrasbourg, Post-collisional Magmatism Symposium, Terra Nova abstract, p.695.

Hogarth, D.D., Robin, M.J.L., 2007. Strontium in feldspars of high-K proterozoicigneous rocks of the Robitaille suite, Buckingham, Québec. Can. Mineral. 45,1293–1306.

Holloway, J.R., Burnham, C.W., 1972. Melting relations of basalt with equilibriumwater pressure less than total pressure. J. Petrol. 13, 1–29.

Hori, H., Nakai, I., Nagashima, K., Matsubara, S., Kato, A., 1987. StronalsiteSrNa2Al4Si4O16, a new mineral from Rendai, Kochi City, Japan. Mineral. J. 13,368–375.

Hughes, J.M., Erich, S., Bloodaxe, J., Hanchar, M., Foord, E., 1997. Incorporation ofrare earth elements in titanite; stabilization of the A2/a dimorph by creation ofantiphase boundaries. Am. Mineral. 82 (5–6), 512–516.

Irvine, T.N., 1982. Terminology of layered intrusions. J. Petrol. 23 (2), 127–162.Keller, J., 1983. Potassic lavas in the orogenic volcanism of the Mediterranean area. J.

Volcanol. Geoth. Res. 18, 321–335.

Keller, J., Villary, L., 1972. Rhyolitic ignimbrites in the region of Afyon (CentralAnatolia). Bull. Volcanol. 36, 342–358.

Kratzmann, D.J., Carey, S., Scasso, R., Naranjo, J.A., 2009. Compositional variationsand magma mixing in the 1991 eruptions of Hudson volcano, Chile. Bull.Volcanol. 71, 419–439.

Kumral, K., Coban, H., Gedikoglu, A., Kilinc, A., 2006. Petrology and geochemistry ofaugite trachytes and porphyritic trachytes from the Gölcük volcanic region,Isparta, SW Turkey: a case study. J. Asian Earth Sci. 27 (5), 707–716.

Lacroix, A., 1893. Les enclaves des roches volcaniques. Protat frères, imprimeurs,710 pp.

Lagache, M., 1999. Equilibre entre phases du système ternaire: albite, anorthite,feldspath strontique et solution chlorurée à 750 °C, 200 MPa. C. R. Acad. Sci.Paris, Géomatériaux/Minéralogie, vol. 328, pp. 303–308.

Larsen, L.M., 1976. Clinopyroxenes and coexisting mafic minerals from the alkalineIlimaussaq intrusion, South Greenland. J. Petrol. 17, 258–290.

Le Bas, M.J., Le Maître, R.W., Streckeisen, A., Zanettin, B., 1986. A chemicalclassification of volcanic rocks based on the total alkali-silica diagram. J. Petrol.27, 745–750.

Le Maitre, R.W., 2002. Igneous Rocks: A Classification and Glossary Terms, seconded. IUGS, Cambridge University Press, 236 pp.

Leake, B.E., Wooley, A.R., Arps, C.E.S., Birch, W.D., Gilbert, M.C., Grice, J.D.,Hawthorne, F.C., Kato, A., Kisch, H.J., Krivovichev, V.G., Linthout, K., Laird, J.,Mandarino, J., Maresch, W.V., Nickel, E.H., Rock, N.M.S., 1997. Nomenclature ofamphiboles. Report of the subcommittee on amphiboles of the InternationalMineralogical Association Commission on new minerals and mineral names.Eur. J. Mineral. 9, 623–651.

Lefèvre, C., Bellon, H., Poisson, A., 1983. Présence de leucitites dans le volcanismePliocène de la région d’Isparta (Taurides occidentales, Turquie). C.R. Acad. Sci.Paris 297, II, pp. 367–372.

Leterrier, J., Maury, R.C., 1978. Représentation graphique de la cristallisationfractionnée des séries volcaniques. Application aux séries alcalines. Geol.Runschau 67 (3), 943–952.

Lofgren, G., 1980. Experimental studies on the dynamic crystallization of silicatemelts. In: Hargraves, R.B. (Ed.), Physics of Magmatic Processes. PrincetonUniversity Press, Princeton, New Jersey, pp. 487–543.

Lofgren, G., 1983. Effect of heterogeneous nucleation on basaltic textures: adynamic crystallisation study. J. Petrol. 24 (3), 229–255.

Marshall, L.A., Sparks, R.S.J., 1984. Origin of some mixed magma and net-veined ringintrusions. J. Geol. Soc. London 141, 171–182.

Maury, R.C., Brousse, R., Villemant, B., Joron, J.L., Jaffrezic, H., Treuil, M., 1980.Cristallisation fractionnée d’un magma basaltique alcalin: la série de la Chaînedes Puys (Massif Central Français). I: Pétrologie. Bull. Minér. 103, 250–266.

McBirney, A.R., 1980. Mixing and unmixing of magmas. J. Volcanol. Geoth. Res. 7,357–371.

McDonough, W.F., Sun, S.S., 1995. The composition of the Earth. Chem. Geol. 120,223–253.

Mitchell, R.H., 1994. The lamprophyre facies. Mineral. Petrol. 51, 137–146.Mitchell, R.H., Vladykin, N.V., 1996. Composition variation of pyroxene and mica

from the little Murun ultrapotassic complex, Aldan Shield, Russia. Mineral. Mag.60, 907–925.

Morimoto, N., 1988. Nomenclature of pyroxenes. Bull. Minér. 111, 535–550.Mouillard, P., 2011. Etude téphrochronologique du bassin de Burdur, Anatolie de

l’Ouest, Turquie. Unpublished Master 1 Memoir, Université Paris-Sud Orsay, 30pp.

Nekvasil, H., 1994. Ternary feldspar/melt equilibria: a review. In: Parsons, I. (Ed.),Feldspars and their Reactions, NATO ASI Series, Series C: Mathematical andPhysical Sciences, 421. Kluwer Academic Publ., Dordrecht, pp. 195–219.

Nelson, D.R., Chivas, A.R., Chappell, B.W., McCulloch, M.T., 1988. Geochemical andisotopic systematics in carbonatites and implications for the evolution of ocean-island sources. Geochim. Cosmochim. Acta 52, 1–27.

Nemec, W., Kazanci, N., Mitchell, J.G., 1998. Pleistocene explosions and pyroclasticcurrents in west-central Anatolia. Boreas 27, 311–332.

Nielsen, T.D.F., 1979. The occurrence and formation of Ti-aegirine in peralkalinesyenite: an example from the Tertiary ultramafic alkaline Gardiner complex,East Greenland. Contrib. Mineral. Petrol. 69, 235–244.

Özgür, N., Pekdeger, A., Schneider, H.J., 1990. Pliocene volcanism in the Gölcük area,Isparta/Western Taurides. In: IESCA, 1990 Proceedings, pp. 411–419.

Özgür, N., Pekdeger, A., Schneider, H.J., 1992. High fluorine contents in the Pliocenevolcanic rocks of the Gölcük area, Isparta, SW Turkey. Bull. Geol. Soc. Greece 28(2), 417–427.

Pagel, M., Leterrier, J., 1980. The subalkaline potassic magmatism of the BallonsMassif (southern Vosges, France): shoshonitic affinity. Lithos 13, 1–10.

Palme, H., O’Neill, H.S.C., 2002. Cosmochemical estimates of mantle composition. In:Holland, D.H., Turrekian, K.K. (Eds.), Treatrise on Geochemistry. V2, Elsevier,Amsterdam, pp. 1–38.

Pearce, J.A., 1982. Trace element characteristics of lavas from destructive plateboundaries. In: Thorpe, R.S. (Ed.), Andesites. Wiley, Chichester, pp. 525–548.

Pearce, J.A., 1983. Role of the sub-continental lithosphere in magma genesis atactive continental margins: Trace element characteristics of lavas fromdestructive plate boundaries. In: Hawkesworth, C.J., Norry, M.J. (Eds.),Continental Basalts and Mantle Xenoliths. Shiva Publications, pp. 230–249.

Pearce, J.A., Norry, M.J., 1979. Petrogenetic implications of Ti, Zr, Y and Nb variationsin volcanic rocks. Contrib. Mineral. Petrol. 69, 33–47.

Peccerillo, A., Poli, G., Tolomeo, L., 1984. Genesis, evolution and tectonic significanceof K-rich volcanics from the Alban Hills (Roman comagmatic region) as inferredfrom trace element geochemistry. Contrib. Mineral. Petrol. 86, 230–240.

B. Platevoet et al. / Journal of Asian Earth Sciences 92 (2014) 53–76 75

Page 24

Peccerillo, A., Taylor, S.R., 1976. Geochemistry of Eocene calc-alkaline volcanic rocksin the Kastamonu area, Northern Turkey. Contrib. Mineral. Petrol. 58, 63–81.

Perugini, D., Poli, G., 2000. Chaotic dynamics and fractals in magmatic interactionprocesses: a different approach to the interpretation of mafic microgranularenclaves. Earth Planet. Sci. Lett. 175, 93–103.

Plà Cid, J., Nardi, L.V.S., Stabel, L.Z., Conceição, R.V., Balzaretti, R.V., 2003. High-pressure minerals in mafic microgranular enclaves: evidence for co-minglingbetween lamprophyric and syenitic magmas at mantle conditions. Contrib.Mineral. Petrol. 145, 444–459.

Platevoet, B., Bonin, B., 1991. Enclaves and mafic-felsic associations in the Permianalkaline province of Corsica, France. Physical and Chemical Interactionsbetween Coeval Magmas. In: Didier, J., Barbarin, B. (Eds.), Enclaves andGranites Petrology. Developments in Petrology, vol. 13. Elsevier, pp. 191–204.

Platevoet, B., Scaillet, S., Guillou, H., Nomade, S., Blamart, D., Poisson, A., Elitok, Ö,Özgür, N., Yagmurlu, F., Yilmaz, K., 2006. Activité récente plinienne etphréatoplinienne du volcan Gölcük, angle d’Isparta, Turquie. SymposiumAFEQ-SGF: Téphras et séquences quaternaires. Société Géologique de France,livret des communications, Paris (06.11.06).

Platevoet, B., Scaillet, S., Guillou, H., Nomade, S., Blamart, D., Poisson, A., Elitok, Ö.,Özgür, N., Yagmurlu, F., Yilmaz, K., 2008. Pleistocene eruptive chronology of theGölcük volcano, Isparta Angle, Turkey. Quaternaire 19 (2), 147–156.

Platt, R.G., Woolley, A.R., 1986. The mafic mineralogy of the peralkaline syenites andgranites of the Mulanje complex, Malawi. Mineral. Mag. 50, 85–99.

Poisson, A., Yagmurlu, F., Bozcu, M., Sentürk, M., 2003. New insights on the tectonicsetting and evolution around the apex of the Isparta Angle (SW Turkey). Geol. J.38, 257–282.

Prelevic, D., Foley, S.F., 2007. Accretion of arc-oceanic lithospheric mantle in theMediterranean: evidence from extremely high-Mg olivines and Cr-rich spinelinclusions from lamproites. Earth Planet. Sci. Lett. 256 (1–2), 120–135.

Prelevic, D., Foley, S.F., Romer, R., Conticelli, S., 2008. Mediterranean Tertiarylamproites derived from multiple source components in postcollisionalgeodynamics. Geochim. Cosmochim. Acta 72, 2125–2156.

Prelevic, D., Akal, C., Romer, R.L., Foley, S.F., 2010a. Lamproites as indicators ofaccretion and/or shallow subduction in the assembly of south-western Anatolia,Turkey. Terra Nova 22 (6), 443–452.

Prelevic, D., Stracke, A., Foley, S.F., Romer, R.L., Conticelli, S., 2010b. Hf isotopecompositions of Mediterranean lamproites: mixing of melts fromasthenosphere and crustally contaminated mantle lithosphere. Lithos 119,297–312.

Prelevic, D., Akal, C., Foley, S.F., Romer, R.L., Stracke, A., Van Den Bogaard, P., 2012.Ultrapotassic mafic rocks as geochemical proxies for post-collisional dynamicsof orogenic lithospheric mantle: the case of southwestern Anatolia, Turkey. J.Petrol. 53 (5), 1019–1055.

Rock, N.M.S., 1991. Lamprophyres: the global occurrence, petrology, origin andeconomic significance of some rocks of deep origin. Blackie, Glasgow, 285 pp.

Rogers, N.W., Hawkesworth, C.J., Parker, R.J., Marsh, J.S., 1985. The geochemistry ofpotassic lavas from Vulsini, Central Italy, and implications for mantleenrichment processes beneath the Roman region. in the upper mantle.Contrib. Mineral. Petrol. 90, 244–257.

Rogers, N.W., James, D., Kelley, S.P., De Mulder, M., 1998. The generation of potassiclavas from the eastern Virunga Province, Rwanda. J. Petrol. 39 (6), 1223–1247.

Sabatier, H., 1980. Vaugnérites et granites: une association particulière de rochesgrenues acides et basiques. Bull. Minéral. 103, 507–522.

Savasçin, M.Y., Birsoy, R., Dag, N., Nohutçu, E., 1994. Kirka-Afyon-Isparta structuraltrend and alkaline rock associations (Anatolia). Bull. Geol. Soc. Greece 30 (3),89–98.

Savasçin, M.Y., Oyman, T., 1998. Tectono-magmatic evolution of alkaline volcanicsat the Kirka-Afyon-Isparta structural trend, SW Turkey. Turkish J. Earth Sci. 7,201–214.

Schmidt, M.W., 1992. Amphibole composition in tonalite as a function of pressure:an experimental calibration of the Al-in-hornblende barometer. Contrib.Mineral. Petrol. 110, 304–310.

Sßen, P., Temel, A., Sßen, E., Gourgaud, A., Kieffer, G., 2008. The Origin of GölcükVolcanism (Isparta-SW Türkey): Evidence from Mantle Xenoliths. 61st TurkishGeological Congress, March 24–28, 2008, abstract program.

Seyitoglu, G., Anderson, D., Nowell, G., Scott, B., 1997. The evolution from Miocenepotassic to Quaternary sodic magmatism in western Turkey: implications forenrichment processes in the lithospheric mantle. J. Volcanol. Geoth. Res. 76,127–147.

Snyder, D., Tait, S., 1996. Magma mixing by convective entrainment. Nature 379,529–531.

Sparks, R.S.J., Marshall, L.A., 1986. Thermal and mechanical constraints on mixingbetween mafic and silicic magmas. J. Volcanol. Geoth. Res. 29, 99–124.

Sparks, R.S.J., Sigurdsson, H., Wilson, L., 1977. Magma mixing: a mechanism fortriggering acid explosive eruptions. Nature 267 (5609), 315–318.

Tappe, S., Graham Pearson, D., Prelevic, D. (Eds.), 2013. Kimberlite, Carbonatite, andPotassic Magmatism as Part of the Geochemical Cycle. Chemical Geology,Special issue, p. 353.

Thornton, C.P., Tuttle, O.F., 1960. Chemistry of igneous rocks. I: The differentiationindex. Am. J. Sci. 258, 664–684.

Tiepolo, M., Oberti, R., Zanetti, A., Vannucci, R., Foley, S.F., 2007. Trace-Elementpartitioning between amphibole and silicate melt. In: Hawthorne, F.C., Oberti,R., Della Ventura, G. Mottana, A. (Eds.), Amphiboles: Crystal Chemistry,Occurrence, and Health Issues, Reviews in Mineralogy and Geochemistry, vol.67, pp. 417–451.

Tischendorf, G., 1997. On Li-bearing micas: estimating Li from electron microprobeanalyses and an improved diagram for graphical representation. Mineral. Mag.61, 809–834.

Tuttle, O.F., Bowen, N.L., 1958. Origin of granite in the light of experimental studiesin the system NaAlSi3O8–KAlSi3O8–SiO2–H2O. Geol. Soc. Am. Memoir 74, 153pp.

Viccaro, M., Ferlito, C., Cortesogno, L., Cristofolini, R., Gaggero, L., 2006. Magmamixing during the 2001 event at Mount Etna (Italy): effects on the eruptivedynamics. J. Volcanol. Geoth. Res. 149, 139–159.

Vlastélic, I., Deniel, C., Bosq, C., Telouk, P., Boivin, P., Bachèlery, P., Famin, V.,Staudacher, T., 2009. Pb isotope geochemistry of Piton de la Fournaise historicallavas. J. Volcanol. Geotherm. Res. 184 (1–2), 63–78.

Wiebe, R.A., 1996. Mafic-felsic layered intrusions: the role of basaltic injections onmagmatic processes and evolution of silicic magma chambers. Trans. Roy. Soc.Edinburgh: Earth Sci. 87, 233–242.

Wilson, M., Downes, H., Cebria, J.M., 1995. Contrasting fractionation trends incoexisting continental alkaline magma series; Cantal, Massif Central, France. J.Petrol. 36 (6), 1729–1753.

Wones, D.R., 1989. Significance of the assemblage titanite + magnetite + quartz ingranitic rocks. Am. Mineral. 74, 744–749.

Woolley, A.R., Bergman, S.C., Edgar, A.D., Le Bas, M.J., Mitchell, R.H., Rock, N.M.S.,Scott Smith, B.H., 1996. Classification of Lamprophyres, lamproites, kimberlitesand the kalsilitic, melilitic and leucitic rocks. Can. Mineral. 34, 175–186.

Woolley, A.R., Kempe, D.R.C., 1989. Carbonatites: nomenclature, average chemicalcompositions, and element distribution. In: Bell, K. (Ed.), Carbonatites: Genesisand Evolution. Unwin Hyman, London, pp. 500–545.

Woolley, A.R., Ross, M., 1990. Alkaline Igneous Rocks and Carbonatites. Lithos,Special issue, 26, 1, 2, 177 pp.

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

Wyllie, P.J., Baker, M.B., White, B.S., 1990. Experimental boundaries for the originand evolution of carbonatites. In: Wooley, A.R., Ross, M. (Eds.), Lithos, Specialissue, Alkaline Igneous Rocks and Carbonatites, vol. 26, 1–2, 3–20.

Yagmurlu, F., Savasçin, Y., Ergün, M., 1997. Relation of alkaline volcanism and activetectonism within the evolution of the Isparta Angle, SW Turkey. J. Geol. 105 (6),717–728.

76 B. Platevoet et al. / Journal of Asian Earth Sciences 92 (2014) 53–76