geochemical characteristics of mafic and intermediate...

Geochemical Characteristics of Mafic and Intermediate VolcanicRocks from the Hasandağ and Erciyes Volcanoes

(Central Anatolia, Turkey)

AYKUT GÜÇTEKİN & NEZİHİ KÖPRÜBAŞI

Department of Geological Engineering, Kocaeli University, TR–41040 İzmit, Turkey

(E-mail: nezihi@ kocaeli.edu.tr)

Received 23 June 2007; revised typescript received 24 November 2007; accepted 07 December 2007

Abstract: Hasandağ and Erciyes stratovolcanoes, which are the two important stratovolcanoes in Central Anatolia, erupted volcanicproducts with both calc-alkaline and alkaline compositions, although the calc-alkaline activity is more widespread. There are threestages of geochemical evolution in the history of the Hasandağ stratovolcanic complex: (1) Keçikalesi tholeiitic volcanism, (2)Hasandağ calc-alkaline volcanism, and (3) Hasandağ alkaline volcanism. The geochemical evolution of Erciyes volcanic complex alsoexhibits three distinct evolutionary stages: (1) Koçdağ alkaline volcanism, (2) Koçdağ calc-alkaline volcanism and (3) Erciyes calc-alkaline volcanism. The volcanic rocks from both suites show enrichments in LREE relative to HREE. The rocks as a whole showenrichments in large ion lithophile elements (LILE) relative to high field strength elements (HFSE) in N-MORB normalized multi-element diagrams, although the thoeliitic and alkaline rocks have less pronounced effects of HFSE/LILE fractionation comparing tothe calc-alkaline rocks.

Theoretical fractionation models obtained using the whole-rock trace element data indicate two distinct fractionation trends forthe Hasandağ volcanism: amphibole and plagioclase fractionation for the tholeiitic and calc-alkaline rocks and plagioclase, pyroxeneand amphibole fractionation for the alkaline rocks. The alkaline and calc-alkaline rocks of Erciyes volcanism, on the other hand,indicate similar fractionation trends that can be explained by plagioclase and amphibole fractionation. AFC modelling indicate that theeffect of crustal contamination on parental melt compositions is less pronounced in the Erciyes volcanic rocks compared to theHasandağ volcanic rocks. Theoretical melting trends obtained using the non-modal batch melting model indicate degree of meltingbetween 8–9% for the Keçikalesi tholeiitic rocks, 4–5% for the Hasandağ alkaline rocks and 3–8% for the Koçdağ alkaline rocks.The modeling also shows that the Central Anatolian volcanic rocks were originated by variable degree melting of a mantle sourcewith garnet+spinel lherzolite composition, although the effect of residual garnet in the source is more pronounced for the Hasandağalkaline rocks. Geochemical modeling indicates that Central Anatolian volcanic rocks are likely to have originated by partial meltingof a metasomatised lithospheric mantle. Delamination of TBL (termal boundary layer) may be considered as a potential mechanismthat may cause thermal perturbation melting of continental lithospheric mantle. The distribution of volcanic centres along the region-scale strike-slip fault systems may also indicate additional effects of strike-slip faulting on melt production and eruption.

Key Words: Central Anatolia, collision volcanism, lithospheric delamination, Turkey

Hasandağ ve Erciyes Volkanlarının Mafik ve Ortaç VolkanikKayaçlarının Jeokimyasal (Orta Anadolu, Türkiye)

Özet: Orta Anadolu’da iki önemli stratovolkan olan Hasandağ ve Erciyes stratovolkanları kalk-alkalen ve alkalen karakterlivolkanizmalar olmasına rağmen, kalk-alkalen aktivite daha yaygındır. Hasandağ volkanik kompleksi jeokimyasal farklılıklarıylaKeçikalesi toleyitik, Hasandağ kalk-alkalen, Hasandağ alkalen volkanizması olmak üzere üç evrede gelişimini gerçekleşmiştir. Erciyesvolkanik kompleksinin jeokimyasal gelişimi Koçdağ alkalen, Koçdağ kalk-alkalen, Erciyes kalk-alkalen volkanizması olmak üzere üçevredir. Volkanizmalara ait nadir toprak element (REE) içeriklerinde tüm ürünlerde, hafif nadir toprak elementler (LREE) ağır nadirtoprak elementlere (HREE) göre göreceli bir zenginleşme göstermektedir. N-MORB’a göre normalize edilmiş çoklu elementdiyagramlarında tüm volkanik ürünlerde geniş iyonlu litofil (LIL) elementler ve hafif nadir toprak elementlerde (LREE) belirginzenginleşme, Ta, Nb, Ti, Hf gibi kalıcılığı yüksek elementlerde (HFS) ise göreceli bir tüketilme görülmekle birlikte HFSE/LILEfraksiyonasyonunun etkileri kalk-alakali kayalarda daha yüksek ve toleyitik, alkali bileşimli kayalarda daha az oranlardadır.

Tüm kayaç iz element içerikleri kullanılarak yapılan teorik kristalizasyon modellemeleri ile elde edilen eğilim, Hasandağ toleyitikve kalk-alkalen kayaçlarda amfibol, plajiyoklas; alkalen kayaçlarda plajiyoklas, piroksen, amfibol; Erciyes alkalen ve kalk-alkalenkayaçlarda ise plajiyoklas, piroksen ve amfibolun baskın olduğu fraksiyonel kristalizasyon ile açıklanabilir. AFC modellemelerine göreErciyes volkanında kıta kirlemesinin etkisi Hasandağ’a göre daha azdır. Modal olmayan yığın ergime modeliyle hesaplanan ergimedereceleri Keçikalesi toleyitik kayaçlarında yaklaşık %8–9, Hasandağ alkalen kayaçlarında %4–5, Koçdağ alkalen kayaçlarında ise%3–8 arasındadır. Teorik ergime modelleri Orta Anadolu volkanik kayaçları için granat ve spinel-lerzolit bileşimli mantokaynaklarından türeyen değişen miktarlarda magma karışımını önerirken, Hasandağ alkalen kayaçlarında ergime sürecinde artıkgranat etkisi daha baskındır. Metasomatize olmuş litosferik mantodan türediği düşünülen Orta Anadolu volkanizması, termal sınırdüzeyinin (TSD) deleminasyonu ile oluşan geniş çaplı termal düzensizlik sonucu kıta altı litosferik mantonun ergimesiyle açıklanabilir.Bölgedeki volkanizmanın doğrultu atımlı tektonizmanın bulunduğu kuşaklar boyunca görülmesi bu sistemin magma yükselimine olasıkatkı sağladığını göstermektedir.

Anahtar Sözcükler: Orta Anadolu, çarpışma volkanizması, litosferik delaminasyon, Türkiye

1

Turkish Journal of Earth Sciences (Turkish J. Earth Sci.), Vol. 18, 2009, pp. 1–27. Copyright ©TÜBİTAKdoi:10.3906/yer-0806-2 First published online 07 July 2008

Introduction

Collision-related calc-alkaline and alkaline volcanisms areobserved in the Hasandağ and Erciyes volcanoes, whichare two important stratovolcanoes in the Central Anatolia,but calc-alkaline volcanism is more dominant. Tholeiiticrocks of small volumes comprise the first products of theHasandağ volcanism. The plate boundaries betweenAfrican, Arabian and Eurasian plates, which represent thewestern part of the Alpine-Himalayan belt, are irregular inthe Mediterranean region (Jackson & McKenzie 1984).Aeolian and Aegean volcanic arcs were formed in thesouthern Italy and Aegean regions by subduction of theAfrican plate under the Eurasian plate (Keller 1982). Inaddition, geologic and geomorphologic characteristics ofother areas in the Aegean-Anatolian region are mainlyshaped by the collision and post-collision tectonism (Notsuet al. 1995). In most collision zones (Himalaya andAnatolia), calc-alkaline and alkaline associations are foundtogether that are closely associated chronologically andspatially (Pearce et al. 1990; Turner et al. 1996). Bonin(1990) who studied magmatic rocks in the Alps statedthat magmatism changes from calc-alkaline to alkalinetype due to decreasing water content; both types ofvolcanics are derived from almost the same source. TheCentral Anatolian volcanism also is characterized byproducts of calc-alkaline, alkaline and tholeiitic characterwhich developed at various stages of volcanic activity anddo not show any significant chronological and spatialsystematic.

In this study we aimed to provide evidence fordetermining the petrogenetic processes that wereeffective during the formation and evolution of thevolcanic suites. For this purpose, based on trace elementdata on volcanic rocks, a petrogenetic modeling wasperformed. During these studies, in order to determinecrystallization processes of rocks, fractional crystallizationand assimilation effects were examined and then possiblesource components of volcanoes were determined. Thiswas followed by melting modelling studies to delineatepercents of partial melting types which were effectiveduring the magma generation processes.

Geological Setting

The Anatolian volcanism is one of the best examples ofcontinental collision volcanism. With the opening of theSuez Gulf in the Oligocene, the Arabian plate was started

to be separated from the African plate (McKenzie 1972;Cochran 1981; İzzeldin 1987; Bayer et al. 1988; LePichon & Gaulier 1988). Northerly movement of theArabian plate during the middle–late Miocene resulted incontinental collision which was responsible for shorteningand thickening of the crust in the eastern Anatoliabetween the Eurasian and Arabian plates (Şengör 1980;Şengör & Yılmaz 1981; Fyitkas et al. 1984; Buket &Temel 1998; Temel et al. 1998; Yılmaz et al. 1998;Yürür & Chorowicz 1998). The continental collision gaverise to westward escape of the Anatolian block alongNorth Anatolian Fault (NAF) and East Anatolian Fault(EAF) (McKenzie 1972; McKenzie & Yılmaz 1991). Theshortening and thickening of the crust in the easternAnatolia via Arabia-Eurasia collision and westward escapeof the Anatolian block resulted in deformation of theAnatolian block (McKenzie & Yılmaz 1991; Lyberis et al.1992). The volcanic activity in Central Anatolia is relatedto this deformation (McKenzie & Yılmaz 1991; Aydar etal. 1995). The Central Anatolian region in Turkey hasbeen subjected to deformation and volcanism over thepast 10 Ma (Innocenti et al. 1982; Şengör et al. 1985;Pasquare et al. 1988; Aydar 1992; Temel 1992; Aydar etal. 1993; Le Pennec et al. 1994). There are two majoractive fault systems in the Central Anatolian volcanism(Toprak & Göncüoğlu 1993). The first is the left-lateralEcemiş fault zone and the right-lateral Tuz Gölü faultzone and the second is N60–70° E-trending normal faultsthat are consistent with main eruption centres as beingparallel to the volcanic axis (Toprak & Göncüoğlu 1993).NE–SW-extending faults are central Kızılırmak and Niğdefaults (Toprak 1994). Some of these faults are coveredwith young volcanic rocks.

Magmatic, metamorphic and ophiolithic rockassemblages in Central Anatolia form the crystallinebasement complex. This complex is composed ofophiolithic units overlying Palaeozoic–Mesozoic medium-to high-grade metamorphic rocks and intrusive rockscutting those units (Akıman et al. 1993 & Göncüoğlu etal. 1991b). There is no published geophyscial informationregarding the thickness of crust and lithospheric mantlebeneath Central Anatolia.

The products of the Hasandağ volcano (Figure 1),which is one of the important stratovolcanoes in theCentral Anatolia, are composed of lava flows, lava domesand pyroclastic rocks. According to Aydar & Gourgaud(1998), volcanism was developed at four stages, namely,

GEOCHEMISTRY OF HASANDAĞ AND ERCİYES VOLCANOES

2

A. GÜÇTEKİN & N. KÖPRÜBAŞI

3

AK

SA

RA

Y

Kar

apın

ar

Der

inku

yu

Kız

ılırm

akri

ver

Tuz

gölü

Bas

in

Sul

tans

azlı

ğıB

asin

EcemişFaultZone

Tuz-Lak

eFau

lt

DerinkuyuFault

N

010

020

030

0 Km

AN

KA

RA

Bur

sa

İsta

nbul

İzm

ir

Ada

na

Siv

as

Tra

bzon

Erz

urum

Van

BL

AC

KS

EA

ME

Dİ

TE

RR

AN

EA

NS

EA

N

A E G E A N S E A

Ara

bian

Pla

te

Nor

thA

nato

lian

Faul

t

Mt.

Has

an

Erc

iyes

base

men

troc

ks(p

re-M

iddl

eM

ioce

ne)

cont

inen

talc

last

ics

(Lat

eM

ioce

ne-P

lioc

ene)

volc

anic

last

ics

(Lat

eM

ioce

ne-Q

uate

rnar

y)

volc

anic

com

plex

es(L

ate

Mio

cene

-Q

uter

nary

)

Qua

tern

ary

depo

sits

KA

YS

ER

İ

NİĞ

DE

maj

orfa

ults

38No

3330

Eo

ı35

30E

oı

39No

HD

E

Erc

iyes

stra

tovo

lcan

o

Has

anda

ğst

rato

volc

ano

HD E

Cap

pado

cia E

cem

işF

ault

Tuz

gölü

Fau

lt

East

Ana

tolia

nFa

ult

Bitlis

-Zagro

ssutu

rezone

Figu

re 1

.M

ap o

f Ce

ntra

l Ana

tolia

sho

win

g th

e lo

catio

n of

‘the

Cap

pado

cian

vol

cani

c pr

ovin

ce’ a

nd d

istr

ibut

ion

of v

olca

nic

prod

ucts

(m

odifi

ed f

rom

Top

rak

1998

).

Keçikalesi volcano (13 Ma), Palaeovolcano (7 Ma),Mesovolcano and Neovolcano.

The Keçikalesi tholeiitic volcanism is composed of lavaflows of basalt and basaltic andesite and lesser amount ofpyroclastic products while Hasandağ calc-alkalinevolcanism is characterized mainly by andesite lava domesand dacite and rhyolite type lavas and widespreadpyroclastic products. The latest stage of the Hasandağvolcanism is represented by basaltic alkaline volcanics andlinear scoria cones (Aydar & Gourgaud 1998). Productsof the Erciyes volcano are composed of lava flows,pyroclastics and some monogenic structures such asdome, cone and maar. The volcanism was developed intwo stages as Koçdağ and Erciyes (Kürkçüoğlu et al.1998). The products of the Koçdağ volcanism at theeastern part of the Erciyes stratovolcano are alkalibasalts, andesitic lava flows, basaltic andesites andignimbrites (Şen et al. 2003). Dark grey-black colouredbasaltic rocks are monocrystalline and show flowstructure. According to Ayrancı (1969), their age is4.39±0.28 Ma. The Erciyes stratovolcano with an ageranging from 3 Ma to historical times displays a widecompositional spectrum from basalt to rhyodacite(Pasquare et al. 1988; Ayrancı 1991).

Analytical Methods

A total of 38 samples from lava and dome flows werecollected for chemical analyses, 17 from the Hasandağvolcanism and 21 from the Erciyes volcanism. The rocksamples were powdered at the Mineralogy-Petrographylaboratories of the Geology Department in the KocaeliUniversity. The major and trace element analyses wereconducted at the ALS Chemex (Toronto-Canada) analyticallaboratory using ICP-MS and ICP-OES, respectively.Results of chemical analyses are shown in Table 1.

Geochemical Characteristics

Classification of Volcanic Rocks

The volcanic rocks in the region were classified using theclassification scheme of Le Bas et al. (1986) based on thetotal alkali (Na2O + K2O) vs SiO2 (TAS) diagram (Figure2a). In this diagram, the dashed line separating the alkaliand subalkali magma series was taken from Irvine &Baragar (1971).

Based on the Le Bas et al. (1986) diagram, Hasandağalkaline rocks are basalt and trachybasalt while theKoçdağ alkaline rocks plot in the basalt field. Except forthese two groups, all other rocks are below thesubalkaline line. The Hasandağ calc-alkaline rocks arerepresented by basaltic trachyandesite, basaltic andesiteand andesite while Keçikalesi tholeiitic rocks arerepresented by basaltic andesite. The Erciyes calc-alkalinerocks show a wide compositional spectrum fromtrachybasalt to dacite and calc-alkaline rocks of Koçdağvolcanics mostly plot in the andesite field.

Subalkali series are divided into several sub-groupsbased on K2O and SiO2 contents. This discrimination isshown in the Peccerillo & Taylor (1976) diagram (Figure2b). It is shown in this diagram that samples plot in highK-calcalkaline series and calc-alkaline series. Moreover, intotal alkali (Na2O + K2O)-total FeO-MgO triangulardiagram (Figure 3) proposed by Irvine & Baragar (1971),subalkali series can be discriminated as calc-alkaline andtholeiitic types. In Figure 3 where subalkali samples areused, the Keçikalesi volcanic rock samples plot in tholeiiticfield.

In the major oxide covariation diagrams, for all ofsamples, there is an increase in K2O and a decrease inFe2O3, MnO, CaO with increasing SiO2 (Figure 4). There isalso minor decrease in Al2O3 with SiO2 although thedistributions are rather scattered for most of the samples.Especially, the similar major oxide behaviour in calc-alkaline porphyric rocks is probably due to fractionationof plagioclase dominant mineral assemblages in theshallow depth of long-lived magma chambers.Consumption of CaO can be explained by crystallization ofsignificant amount of plagioclase and clinopyroxene.

The high Fe2O3 and TiO2 concentrations of Hasandağand Erciyes alkaline rocks indicate that they are rich iniron. Especially, TiO2 contents of Hasandağ and Erciyesalkaline rocks increase with SiO2, while there is a negativetrend in other rock groups. MnO, MgO and CaOconcentrations of alkaline Hasandağ and Erciyes rocksamples have high concentrations. There is an increase inMgO with increasing SiO2 in the Hasandağ alkalinevolcanism and this can be explained by the accumulationof forsteritic olivine which exists in the phenocrystsassemblages of these rocks.


4


5

Tabl

e 1.

Who

le-r

ock

maj

or a

nd t

race

ele

men

t da

ta f

or t

he r

epre

sent

ativ

e sa

mpl

es f

rom

Has

anda

ğ* a

nd E

rciy

es v

olca

noes

. D

L– D

etec

tion

Lim

it, R

SD–

Rel

ativ

e St

anda

rt D

evia

tion.

Sam

ple

No

H-2

6H

-80

H-3

6H

-20

H-7

H-4

H-4

2H

-75

H-1

5H

-14

H-2

H-7

2H

-70

H-4

3H

-49

H-6

2H

-64

Loca

lity

Has

anda

ğH

asan

dağ

Has

anda

ğH

asan

dağ

Has

anda

ğH

asan

dağ

Has

anda

ğH

asan

dağ

Has

anda

ğH

asan

dağ

Has

anda

ğH

asan

dağ

Has

anda

ğH

asan

dağ

Has

anda

ğH

asan

dağ

Has

anda

ğ

Roc

k ty

peBa

salt

B. A

ndes

iteAn

desi

teAn

desi

teAn

desi

teAn

desi

teBa

salt

Ande

site

Ande

site

Ande

site

Ande

site

Basa

ltBa

salt

Basa

ltBa

salt

Basa

ltBa

salt

Flow

typ

eLa

va f

low

Lava

flo

wD

ome

flow

Dom

e flo

wLa

va f

low

Lava

flo

wLa

va f

low

Dom

e flo

wD

ome

flow

Dom

e flo

wD

ome

flow

Lava

flo

wLa

va f

low

Lava

flo

wLa

va f

low

Lava

flo

wLa

va f

low

Latit

ude

38o 03

ı 32ıı

38o 10

ı 23ıı

38o 01

ı 06ıı

38o 04

ı 59ıı

38o 11

ı 26ıı

38o 10

ı 47ıı

38o 13

ı 52ıı

38o 10

ı 37ıı

38o 05

ı 54ıı

38o 06

ı 09ıı

38o 08

ı 26ıı

38o 07

ı 45ıı

38o 07

ı 35ıı

38o 13

ı 46ıı

38o 14

ı 21ıı

38o 13

ı 01ıı

38o 14

ı 38ıı

Long

itude

34

o 06ı 03

ıı34

o 05ı 35

ıı34

o 15ı 38

ıı34

o 09ı 51

ıı34

o 11ı 49

ıı34

o 05ı 40

ıı34

o 08ı 38

ıı34

o 03ı 44

ıı34

o 10ı 25

ıı34

o 10ı 24

ıı34

o 06ı 30

ıı34

o 04ı 03

ıı34

o 02ı 38

ıı34

o 08ı 32

ıı34

o 08ı 13

ıı34

o 04ı 25

ıı34

o 03ı 26

ııD

L R

SD (

%)

SiO

255

.12

53.3

360

.10

60.5

057

.71

61.0

954

.23

56.0

260

.70

61.2

960

.30

47.9

652

.34

48.3

648

.66

48.2

649

.05

0.01

0.10

TiO

20.

911.

010.

760.

580.

770.

721.

100.

800.

700.

700.

691.

260.

881.

201.

161.

191.

360.

010.

80

Al2O

317

.10

16.6

515

.50

16.0

016

.40

15.6

515

.60

15.6

016

.30

16.0

516

.15

17.4

017

.25

16.5

515

.95

16.5

516

.10

0.01

0.90

Fe2O

39.

5411

.30

6.52

6.59

7.75

6.78

8.11

6.83

6.58

6.24

6.30

10.5

57.

389.

8510

.05

9.94

10.2

50.

010.

30

MnO

0.15

0.16

0.12

0.11

0.12

0.11

0.13

0.11

0.11

0.10

0.11

0.16

0.13

0.15

0.16

0.15

0.16

0.01

1.50

MgO

3.87

3.88

2.65

2.88

3.56

2.36

7.03

6.02

3.19

2.95

2.79

6.19

4.73

7.56

8.20

7.88

6.58

0.01

1.20

CaO

7.88

8.70

4.65

5.39

6.05

4.95

7.42

7.34

5.73

5.39

5.36

11.0

07.

409.

979.

4311

.30

9.21

0.01

0.70

Na 2

O2.

972.

703.

953.

723.

583.

853.

733.

673.

783.

803.

823.

574.

173.

473.

543.

343.

860.

010.

70

K2O

1.12

1.04

2.25

1.98

1.99

2.33

1.74

1.68

1.93

1.98

2.01

0.96

1.27

1.18

1.18

0.95

1.48

0.01

0.50

P 2O

50.

150.

200.

310.

190.

160.

200.

370.

260.

190.

220.

220.

360.

320.

570.

420.

350.

480.

010.

10

L.O

.I1.

350.

792.

361.

441.

051.

610.

601.

210.

040.

471.

370.

360.

710.

320.

310.

270.

850.

01

Tota

l10

0.16

10

0.06

99.1

799

.48

99.1

499

.65

100.

0699

.54

99.2

599

.10

99.1

299

.77

96.5

899

.18

99.0

610

0.18

99.3

8

Ppm

V34

6.6

299.

338

.128

.831

.95.

114

6.1

102.

977

.148

.356

.623

4.5

156.

320

1.6

139.

923

8.6

52.5

0.1

2.1

Co34

.850

.731

.038

.449

.034

.242

.737

.932

.732

.130

.754

.433

.152

.660

.452

.363

.40.

10.

8

Cu10

4.9

150.

263

.872

.011

5.2

76.1

65.8

52.5

61.7

56.6

49.4

99.8

60.7

85.4

100.

885

.411

4.2

0.1

1.5

Zn99

.310

0.8

65.8

61.7

65.1

69.9

81.3

67.9

65.8

59.2

55.5

92.6

75.1

89.5

85.4

90.5

83.3

0.1

1.5

Ga

22.6

21.6

20.8

20.5

22.4

21.7

20.8

20.0

21.6

20.6

20.6

21.8

19.9

21.0

19.9

20.8

20.2

0.1

1.6

Rb

38.9

39.5

70.8

62.5

59.5

67.8

44.9

50.3

63.9

66.1

64.2

19.6

23.5

25.9

23.3

19.0

30.4

0.1

0.8

Sr39

6.1

403.

237

4.7

392.

049

7.9

411.

360

9.9

566.

140

8.3

398.

139

7.1

671.

061

9.1

676.

166

5.9

656.

770

5.6

0.1

1.4

Y26

.428

.021

.418

.918

.419

.422

.518

.020

.619

.817

.625

.318

.025

.624

.924

.824

.90.

12.

2

Zr10

3.4

105.

620

1.8

138.

612

1.0

149.

015

2.4

127.

914

2.5

151.

213

7.1

130.

511

6.5

147.

314

7.5

136.

916

4.5

0.1

0.2

Nb

5.2

6.3

17.6

11.9

9.5

12.6

17.4

9.4

12.5

13.4

11.5

14.2

11.3

18.3

18.4

15.8

20.4

0.1

2.0

Cs0.

92.

02.

42.

22.

22.

51.

41.

82.

31.

21.

20.

50.

90.

60.

50.

50.

50.

12.

0

Ba30

.633

8.4

497.

845

1.8

468.

254

0.8

430.

413

7.7

459.

048

9.6

460.

035

6.8

240.

238

7.4

384.

436

1.9

387.

20.

12.

7

La15

.82

16.9

230

.69

24.4

720

.17

24.9

929

.11

22.3

725

.77

26.2

022

.76

27.7

420

.99

30.8

231

.12

28.4

529

.38

0.01

0.10

Ce32

.95

35.3

859

.73

44.9

136

.98

45.4

755

.56

42.7

246

.52

48.5

144

.21

52.9

441

.55

60.9

559

.76

58.9

756

.41

0.01

2.10

Pr4.

124.

356.

185.

024.

135.

136.

474.

985.

135.

364.

716.

144.

657.

217.

036.

376.

590.

012.

20

Nd

17.3

518

.62

22.4

618

.38

16.0

118

.82

25.0

119

.21

19.2

120

.24

16.8

624

.53

17.7

327

.17

27.3

225

.16

26.7

30.

010.

50

Sm3.

944.

114.

643.

613.

513.

854.

703.

633.

843.

893.

514.

833.

725.

345.

235.

155.

170.

012.

50

Eu1.

141.

121.

120.

911.

031.

121.

311.

051.

041.

010.

951.

451.

161.

491.

451.

451.

470.

011.

10

Gd

4.15

4.17

4.42

3.36

3.35

3.48

4.47

3.17

3.42

3.89

3.53

4.34

3.73

4.82

4.96

4.35

4.71

0.01

1.20

Tb0.

710.

710.

690.

520.

520.

530.

670.

490.

540.

520.

520.

730.

610.

730.

780.

690.

720.

010.

10

Dy

4.39

4.25

3.75

3.10

3.02

3.20

3.59

2.89

3.16

3.26

3.11

4.34

3.53

4.12

4.42

4.02

4.22

0.01

1.40

Ho

0.92

0.93

0.76

0.63

0.64

0.67

0.74

0.61

0.69

0.72

0.65

0.90

0.71

0.84

0.94

0.83

0.85

0.01

2.40

Er2.

482.

492.

031.

691.

791.

911.

921.

681.

982.

031.

792.

321.

922.

252.

562.

162.

150.

010.

10

Tm0.

350.

380.

310.

260.

260.

280.

270.

240.

290.

290.

270.

320.

270.

310.

370.

300.

310.

012.

70

Yb2.

342.

492.

141.

711.

711.

821.

901.

621.

781.

831.

792.

111.

782.

072.

512.

012.

010.

010.

80

Lu0.

370.

380.

390.

270.

270.

280.

290.

250.

290.

270.

280.

290.

270.

310.

360.

300.

300.

010.

10

Hf

3.94

3.34

6.90

4.35

4.34

4.34

4.23

4.99

4.97

4.35

4.88

4.12

4.92

4.56

4.41

4.12

4.78

0.01

5.20

Ta0.

530.

511.

040.

860.

740.

861.

080.

670.

840.

850.

830.

920.

761.

021.

141.

091.

240.

011

Pb11

.17

7.23

16.5

310

.33

13.4

711

.37

7.28

10.2

511

.37

12.4

027

.90

10.6

310

.33

10.3

37.

487.

217.

230.

012.

80

Th6.

945.

349.

909.

387.

349.

347.

237.

999.

979.

359.

885.

123.

925.

565.

415.

125.

780.

010.

70

U2.

431.

833.

302.

742.

432.

931.

952.

663.

042.

933.

181.

181.

711.

471.

481.

081.

720.

010.

20


6

Tabl

e 1.

(Con

tinue

d)

Sam

ple

noE

-146

E-1

48E

-149

E-1

50E

-152

E-2

33E

-141

E-1

02-b

E-1

08E

-103

E-1

33E

-194

E-1

69E

-196

E-1

85E

-88

E-8

6E

-175

E-1

88Lo

calit

yEr

ciye

sEr

ciye

sEr

ciye

sEr

ciye

sEr

ciye

sEr

ciye

sEr

ciye

sEr

ciye

sEr

ciye

sEr

ciye

sEr

ciye

sEr

ciye

sEr

ciye

sEr

ciye

sEr

ciye

sEr

ciye

sEr

ciye

sEr

ciye

sEr

ciye

sR

ock

type

Basa

ltBa

salt

Basa

ltBa

salt

Basa

ltBa

salt

Ande

site

Ande

site

Ande

site

Ande

site

Ande

site

B.An

desi

teD

asite

Basa

ltAn

desi

teAn

desi

teD

asite

B.An

desi

teAn

desi

teFl

ow T

ype

Lava

flo

wLa

va f

low

Lava

flo

wLa

va f

low

Lava

flo

wLa

va f

low

Lava

flo

w

Lava

flo

w

Lava

flo

w

Lava

flo

w

Lava

flo

w

Lava

flo

w

Dom

e flo

w

Lava

flo

w

Lava

flo

w

Dom

e flo

w

Dom

e flo

w

Lava

flo

w

Lava

flo

w

Latit

ude

38o 19

ı 24ıı

38o 19

ı 46ıı

38o 18

ı 08ıı

38o 17

ı 36ıı

38o 19

ı 41ıı

38o 17

ı 36ıı

38o 20

ı 09ıı

38o 34

ı 20ıı

38o 33

ı 18ıı

38o 32

ı 55ıı

38o 25

ı 49ıı

38o 36

ı 00ıı

38o 19

ı 24ıı

38o 27

ı 17ıı

38o 29

ı 26ıı

38o 39

ı 22ıı

38o 40

ı 22ıı

38o 24

ı 20ıı

38o 27

ı 38ıı

Long

itude

35

o 38ı 15

ıı35

o 38ı 03

ıı35

o 37ı 41

ıı35

o 38ı 10

ıı35

o 37ı 09

ıı35

o 38ı 10

ıı35

o 38ı 57

ıı35

o 39ı 04

ıı35

o 33ı 08

ıı35

o 32ı 17

ıı35

o 34ı 49

ıı35

o 16ı 26

ıı35

o 38ı 15

ıı35

o 21ı 41

ıı35

o 15ı 23

ıı35

o 33ı 43

ıı35

o 32ı 54

ıı35

o 27ı 51

ıı35

o 13ı 35

ıı

Wt

%

SiO

247

.96

47.7

647

.96

47.7

647

.26

47.5

6 61

.39

58.6

158

.71

62.1

957

.31

53.8

364

.08

51.2

460

.00

62.1

964

.08

53.0

360

.40

TiO

21.

771.

671.

731.

741.

751.

700.

680.

980.

760.

750.

721.

230.

521.

420.

650.

490.

490.

960.

70

Al2O

317

.05

17.1

516

.75

16.8

516

.85

16.8

516

.35

15.7

516

.05

15.6

017

.15

17.0

015

.05

17.4

016

.35

15.5

015

.90

17.1

016

.65

Fe2O

312

.45

11.6

511

.90

12.2

512

.15

11.7

55.

357.

176.

315.

316.

797.

924.

738.

785.

934.

404.

737.

675.

96

MnO

0.19

0.17

0.18

0.19

0.18

0.18

0.08

0.10

0.15

0.09

0.10

0.12

0.08

0.13

0.10

0.07

0.08

0.12

0.09

MgO

6.28

6.79

6.78

6.52

6.59

6.75

2.71

3.12

3.46

2.04

4.34

4.29

2.17

5.59

3.15

2.01

1.91

6.14

3.08

CaO

8.56

8.90

8.75

8.55

8.49

8.74

4.85

5.19

5.72

4.04

7.33

6.49

4.32

7.41

5.81

4.70

4.63

8.17

5.84

Na 2

O3.

933.

803.

793.

923.

863.

814.

064.

013.

433.

773.

363.

963.

763.

923.

613.

743.

853.

523.

95

K2O

0.77

0.67

0.73

0.77

0.73

0.73

2.07

2.16

1.96

2.94

1.41

1.62

2.47

1.34

1.69

1.98

2.03

1.14

1.83

P 2O

50.

400.

360.

390.

420.

390.

350.

210.

240.

190.

200.

180.

390.

220.

390.

180.

140.

140.

360.

16

L.O

.I0.

430.

070.

400.

361.

451.

471.

322.

652.

672.

790.

502.

782.

292.

281.

752.

901.

291.

131.

07

Tota

l99

.79

98.9

999

.36

99.3

399

.70

99.8

999

.07

99.9

899

.41

99.7

299

.19

99.6

399

.69

99.9

099

.22

98.1

299

.13

99.3

499

.73

Ppm

V17

1.8

237.

627

1.5

174.

921

0.9

228.

311

5.2

60.7

65.8

45.3

93.6

126.

523

.712

9.6

46.3

80.2

13.4

148.

166

.9

Co61

.154

.551

.258

.257

.654

.322

.435

.634

.624

.137

.335

.525

.639

.329

.718

.422

.939

.631

.8

Cu96

.772

.066

.993

.690

.571

.139

.185

.467

.948

.374

.154

.551

.450

.466

.935

.044

.257

.660

.7

Zn10

1.8

101.

810

9.0

99.8

100.

897

.756

.668

.959

.760

.764

.884

.346

.382

.362

.755

.544

.372

.055

.5

Ga

20.7

21.0

21.7

20.6

21.1

21.1

2.81

22.5

21.4

22.9

20.8

21.7

21.8

21.0

20.8

20.8

20.6

19.1

21.5

Rb

17.8

26.2

17.1

15.2

27.3

26.5

70.5

87.4

60.7

97.9

43.8

42.9

95.8

24.9

45.5

69.4

65.1

26.3

62.7

Sr49

1.8

515.

255

6.9

487.

745

1.1

471.

431

7.7

329.

937

5.7

300.

439

6.1

500.

927

9.0

558.

035

2.3

295.

331

5.6

530.

537

7.7

Y29

.527

.631

.229

.028

.528

.019

.228

.320

.732

.218

.622

.921

.322

.518

.220

.219

.618

.218

.4

Zr20

4.1

185.

519

9.2

199.

719

4.7

194.

520

2.8

229.

617

9.0

260.

613

1.0

167.

514

8.2

188.

515

2.0

165.

515

3.0

133.

515

3.5

Nb

14.2

11.2

12.4

13.1

12.2

12.2

11.5

14.2

12.5

14.7

10.4

18.4

12.6

18.3

11.4

10.5

10.1

16.3

9.4

Cs0.

41.

10.

50.

30.

50.

62.

52.

90.

93.

31.

41.

33.

40.

61.

72.

51.

01.

02.

3

Ba19

3.4

222.

814

8.7

299.

527

1.9

303.

640

0.7

380.

337

8.2

432.

432

1.0

376.

251

8.3

333.

230

3.6

353.

737

3.1

323.

034

2.4

La19

.73

17.6

320

.92

19.6

418

.63

19.1

220

.97

33.9

922

.93

29.4

119

.86

24.8

526

.72

22.5

720

.64

22.6

323

.36

25.5

419

.46

Ce42

.87

37.7

245

.58

42.6

440

.88

40.1

241

.19

55.9

245

.55

57.2

439

.14

49.3

148

.65

47.8

941

.55

45.2

846

.82

48.6

539

.43

Pr5.

114.

655.

495.

245.

044.

794.

596.

135.

036.

754.

515.

435.

155.

564.

595.

035.

145.

134.

35

Nd

21.1

620

.11

22.6

721

.88

20.5

919

.16

16.9

723

.37

18.9

725

.81

16.4

820

.26

18.2

821

.37

16.8

718

.88

18.9

317

.98

16.4

9

Sm5.

124.

835.

615.

314.

914.

893.

615.

523.

945.

033.

594.

263.

814.

713.

614.

073.

813.

683.

54

Eu1.

721.

521.

801.

681.

571.

470.

991.

131.

121.

241.

021.

240.

961.

450.

990.

910.

891.

080.

98

Gd

5.34

4.91

5.43

5.15

5.12

4.71

3.44

5.49

3.89

5.31

3.24

4.26

3.83

4.58

3.64

3.62

3.72

3.38

3.35

Tb0.

890.

800.

910.

840.

870.

780.

570.

920.

620.

860.

530.

670.

590.

720.

590.

620.

610.

560.

53

Dy

5.56

4.83

5.37

5.03

5.24

4.71

3.42

4.95

3.64

4.77

3.20

3.78

3.51

4.31

3.41

3.58

3.34

3.29

3.22

Ho

1.19

1.04

1.15

1.06

1.08

1.02

0.74

1.04

0.75

1.02

0.68

0.84

0.75

0.89

0.72

0.76

0.72

0.72

0.69

Er3.

212.

833.

112.

922.

872.

812.

012.

792.

022.

741.

932.

212.

042.

361.

952.

012.

022.

011.

87

Tm0.

470.

430.

450.

430.

410.

410.

290.

390.

290.

390.

280.

310.

310.

340.

280.

290.

290.

290.

28

Yb3.

182.

762.

982.

872.

682.

651.

932.

511.

912.

701.

792.

312.

012.

321.

881.

991.

981.

871.

80

Lu0.

490.

430.

440.

440.

410.

390.

290.

390.

290.

460.

280.

300.

300.

380.

290.

310.

300.

290.

28

Hf

5.25

5.32

5.49

5.33

5.85

5.94

6.18

7.76

6.64

7.89

4.98

5.83

5.81

5.71

5.60

5.82

5.71

4.62

5.20

Ta0.

930.

750.

830.

820.

820.

840.

860.

950.

880.

930.

731.

020.

821.

120.

830.

730.

721.

010.

75

Pb5.

175.

115.

177.

235.

1710

.33

11.3

713

.45

9.63

13.4

39.

309.

3212

.40

11.3

710

.33

13.5

713

.43

7.23

9.30

Th2.

252.

882.

492.

332.

852.

9411

.18

15.7

69.

6414

.89

7.98

4.83

14.8

14.

716.

6010

.82

10.7

16.

6210

.20

U0.

670.

710.

820.

821.

071.

123.

325.

003.

134.

892.

452.

055.

041.

472.

403.

253.

372.

013.

23

Trace Element Characteristics

In trace element diagrams an increase is observed in Rband Ba concentrations with increasing SiO2 for all of thesamples but there is not a prominent trend defined forthe rest of the elements (Figure 5). However,concentrations of Rb, Zr and Nb indicate positivecorrelations with silica contents of the rocks when theHasandağ and Erciyes alkaline samples are examinedseparately. Also, for the Hasandağ calc-alkaline rocks

there is an increasing trend in Nb, Sr, Zr and Y withincreasing SiO2 contents for the samples with over 60%SiO2. Similar trends are also observed for Rb, Ba, Th inthis rock suite.

Rare Earth Element Characteristics

Chondrite normalized rare earth element diagrams wereprepared separately for the Hasandağ and Erciyes volcanicrocks (Figures 6a, b). These diagrams were used to


7

0

2

4

6

8

10

12

14

16

40 45 50 55 60 65 70 75 80

SiO (wt%)2

Na

O+

KO

(wt%

)2

2

Hasandağ alkaline volcanism

Hasandağ calc-alkaline volcanism

Keçikalesi tholeiitic volcanism

PC B BA A D

R

TB

BTA

TA

TD

Erciyes calc-alkaline volcanism

Koçdağ calc-alkaline volcanism

Koçdağ alkaline volcanismPH

TP

PT

TPP

A l k a l i n e

S u b a l k a l i n e

40 45 50 55 60 65 70 75 800

2

4

6

3

5

1

KO

(wt%

)2

Low-Kandesite

Rhyolite

Dasite

Sho

shon

ite

High-Kandesite

Andesite

Low-K basalt

Basalt

Basalticandesite

Ban

akit

e

High-Kbasaltic andesite

Abs

arok

ite

Low-Kbasaltic andesite






Koçdağ alkaline volcanism

a

b

Kürkçüoğlu (1998)et al.

Deniel (1998)et al.


Deniel (1998)et al.

SiO (wt%)2

Figure 2. Classification of the volcanic rocks from Central Anatolia. (a) TAS diagram of Le Bas etal. (1986). Key to abbreviations: PC– picrobasalt; B– basalt; BA– basaltic andesite; A–andesite; D– dacite; R– rhyolite; TB– trachybasalt; BTA– basaltic trachyandesite; TA–trachyandesite; T– trachyte; TD– trachydacite; BS– basanite; TP– tephrite; TPPH–tephriphonolite; PHTP– phonotephrite; PH– phonolite; F– foidite. (b) K2O vs SiO2

diagram of Peccerillo & Taylor (1976). Data from Deniel et al. (1998) and Kürkçüoğluet al. (1998) are also plotted for comparison.

compare the composition of rock samples of presentstudy with average OIB and MORB compositions of Sun &McDonough (1989). In addition to rock samples, OIB andMORB compositions were also normalized to chondritevalues of Boynton (1984). In these diagrams, one or twosamples were used for each rock group. Symbols and SiO2

contents of these patterns are shown in the same figures.In samples from the Hasandağ stratovolcano, light rareearth element patterns are flat and show sub-paralleltrends and they are more enriched with respect to heavyrare earth elements (Figure 6a). There is a negative Euanomaly in a sample with SiO2 content of 60.1% from theHasandağ calc-alkaline volcanism (Figure 6a). Rare earthelement patterns of the Erciyes stratovolcano closelyresemble those of Hasandağ samples (Figure 6b). In thisdiagram, negative Eu anomalies are noticeable in samplesfrom the Koçdağ calc-alkaline volcanics. The mostnegative Eu anomalies are the product of plagioclasefractionation.

LREE enrichment in REE patterns of samples from theHasandağ and Erciyes stratovolcanoes could be originatedfrom a low-degree partial melting or due to contributionsfrom subducting slab or crustal components.

Multi Element Characteristics

N-type MORB normalized multi-element diagrams for theHasandağ and Erciyes volcanisms are shown in Figure 7a

and b, respectively. In multi-element diagram of theHasandağ strato volcano, there is significant enrichmentof large ion lithophile elements (LIL) (e.g., Rb, Ba, Th, Uand K) and light rare earth elements and a relativedepletion of high field strength elements (HFS) (e.g., Ta,Nb, Ti and Hf) and heavy rare earth elements (HREE).

In comparison to other rock suites, concentrations oflarge ion lithophile elements (LIL) such as Rb, Ba, Th, Uand K in samples from the Hasandağ alkaline volcanismdisplay variable depletion trends. In addition, they displayTa and Nb negative anomalies smaller than the other rockgroups, suggesting that the source of the Hasandağalkaline lavas is different from those of the other rockgroups. This may indicate that the source of the Hasandağalkaline rocks contained less significant subductioncomponent when compared to the other rock groups orthat their derivative melts have not been significantlyaffected by subsequent crustal contamination. Theserocks are significantly enriched in large ion lithophileelements (LIL) and high field strength elements (HFS; Ta,Nb, Zr) in comparison to mid-ocean ridge basalts (N-MORB) and this can be explained by differentmechanisms: (1) enrichment of lithospheric mantle sourcewith small melts fractions derived from asthenosphere;(2) melting of lithospheric source which is depleted inlarge ion lithophile elements (LIL) and light rare earthelements (LREE) by previous melting events andformation of tholeiitic and calc-alkaline magmas; (3) low-degree partial melting of the asthenospheric mantlesource. In some samples of the Keçikalesi tholeiiticvolcanism and Hasandağ calc-alkaline volcanism, depletionin Ta, Nb and P is much more than that of other rocksuites. These negative Ta and Nb anomalies resemblesubduction-related active continental margins andenrichment in large ion lithophile elements (LIL) can beexplained by metasomatism in the mantle source witheffect of subduction component (Pearce 1983).

Patterns in multi-element variation diagram for theErciyes stratovolcano are similar to those of the Hasandağstratovolcano. In this diagram, concentrations of large ionlithophile elements (LIL) (Th, U and K) in samples of theKoçdağ and Erciyes calc-alkaline volcanism aresignificantly enriched in comparison to other samples.Depletion of Hf and Ti with respect to other high fieldstrength elements (HFS) is attributed to mantleenrichment due to intra-continental processes orderivation from a source with low-degree partial melts.


8

MgO

FeOt

Calc-alkaline

Tholeiitic


Deniel (1998)et al.

Has

anda

ğca

lc-a

lkal

ine

volc

anis

m

Keç

ikal

esi t

hole

iitic

volc

anis

m

Erciyescalc-alkaline

volcanism

Koçdağ

calc-alkalinevolcanism

Na2O + K2O

Figure 3. Distrubution of the samples from the Central Anatolianvolcanic rocks on Irvine & Baragar (1971) AFM diagram.Data from Deniel et al. (1998) and Kürkçüoğlu et al. (1998)are also plotted for comparison.

Post-collisional within plate origin is also reported inprevious works (Kürkçüoğlu et al. 1998).

Concentrations of large ion lithophile elements (LIL)(Rb, Ba, Th, U and K) in samples of the Koçdağ alkalinevolcanism are relatively less enriched compared to thosein other rock suites. However, Ta and Nb display negativeanomalies. These patterns are similar to those of theHasandağ alkaline volcanism and can be evaluated in thesame way.

Trace Element Ratios

Diagrams where Th/Yb and Ta/Yb trace element ratios areused can determine fractional crystallization and/or partial

melting. This diagram proposed by Pearce (1983) makesa distinction between rocks derived from regular mantlesuch as ocean island basalts (OIB) or mid-ocean ridgebasalts (MORB) which generally lie on a diagonal line ofmantle array, and rocks derived from mantle which isenriched via subduction process or rocks affected by thecrustal contamination.

The deviation observed along the diagonal mantlearray for all the samples from the Hasandağ and Erciyesstratovolcanoes could be explained with subduction-related metasomatism or mantle-derived melts that aresignificantly modified by the crustal melts (Figure 8). Thetrend of samples from the Hasandağ alkaline volcanism isalmost parallel to that diagonal mantle array. Although


9

0.0

0.5

1.0

1.5

2.0

45 50 55 60 65 700.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

45 50 55 60 65 70

2

3

4

5

6

45 50 55 60 65 70

CaO

0

2

4

6

8

10

12

45 50 55 60 65 70

MgO

0

2

4

6

8

10

45 50 55 60 65 70

MnO

0.00

0.05

0.10

0.15

0.20

45 50 55 60 65 70

0

2

4

6

8

10

12

14

45 50 55 60 65 7013

14

15

16

17

18

45 50 55 60 65 70


Koçdağ calc-alkaline volcsnism

Koçdağ alkaline volcsnism




Al O2 3

Na O2

K O2

Fe O2 3

TiO2

SiO (wt%)2SiO (wt%)2

Figure 4. Major element variation of the Central Anatolian volcanic rocks.

this does not rule out variable effect of crustalcontamination, the relative enrichment of LILEs observedeven in the least modified mafic compositions suggests asubduction influence in the source of Hasandağ alkalinevolcanics. The effect of subduction modified mantlesource is also observed in the alkaline series of Erciyesvolcanism but the effect of crustal contamination is lesspronounced.

Petrogenetic Modeling

Interpretation of Crystallization Process in VolcanicRocks

Changes in trace element concentrations can be modelledusing binary vector diagrams which show direction andamount of variation occurring as a result of certainprocesses. The mineral vectors used in modeling


10

Ba

0

100

200

300

400

500

600

45 50 55 60 65 70







V

0

100

200

300

400

45 50 55 60 65 70

Y

0

10

20

30

40

45 50 55 60 65 70

Zr

0

50

100

150

200

250

300

45 50 55 60 65 70

Sr

0

200

400

600

800

45 50 55 60 65 70

Rb

0

20

40

60

80

100

120

45 50 55 60 65 70

2

6

10

14

18Th

45 50 55 60 65 70

Nb

0

10

20

30

45 50 55 60 65 70

(ppm)(ppm)

(ppm) (ppm)

(ppm)

(ppm)

(ppm)

(ppm)

SiO (wt%)2SiO (wt%)2

Figure 5. Trace element variation of the Central Anatolian volcanic rocks.

represent the compositions arising from fractionation ofany mineral phase or assemblage. Theoretically, thefractionation trends are computed using the ‘Rayleigh’equation.

The mineral vectors used in modeling were formedwith computer software by Keskin (2002). In fractionalcrystallization modeling, Rb and Th were used as thefractionation index since they are incompatible elements.Other elements used in diagrams include Y and Sm.Yttrium was used to examine the fractionation effectduring the formation of the volcanic rocks. As known,yttrium has a high distribution coefficient in garnet andamphibole. Samarium was used for discrimination ofanhydrous and hydrous crystallization assemblages.

In general, two different trends are found in rocksuites of the Hasandağ stratovolcanism (Figure 9).

Fractionation effects of amphibole and plagioclase arevery distinct in the Hasandağ calc-alkaline rocks (Figure9a–c).

The Hasandağ alkaline rocks, which are the youngestproducts of volcanism, are conformable with trend no. 2in Figure 9 and also mineral assemblages (plag[%50] +amp[%30] + cpx[%10] + opx[%10]) determined in petrographicstudies (Figure 9a–c). The phenocrystal assemblage inthese rocks are composed of olivine, plagioclase andclinopyroxene and do not contain amphibole. This can beattributed to earlier amphibole fractionation anddepletion of Y(amp-liquidKdY≈1.1) and Sm(amp-liquidKdSm≈2.3)elements in magmas. Previous studies report the presenceof little garnet in these rocks (Aydar et al. 1995; Aydar &Gourgaud 2002). However, in trace element modeling,no garnet fractionation was observed that could affect the


11

1

10

100

1000Erciyes volcanism

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

average N-MORB

average OIBCho

ndri

te(C

1)no

rmal

ised

10

100

1000


Hasandağ volcanism

Cho

ndri

te(C

1)no

rmal

ised

1average N-MORB

average OIB

1: H-26 (55.12 % SiO ), Keçikalesi tholeiitic volcanism2

2: H-7 (55.71 % SiO ), Hasandağ calc-alkaline volcanism4: H-36 (60.10 % ), Hasandağ calc-alkaline volcanism

2

SiO2

5: H-49 (48.66 % SiO ), Hasandağ alkaline volcanism3: H-62 (48.26 % ), Hasandağ alkaline volcanism

2

SiO2

1: E-148 (47.76 % SiO ), Koçdağ alkaline volcanism2

6: E-102-b (58.61 % SiO ), Koçdağ calc-alkaline volcanism5: E-103 (62.19 % SiO2), Koçdağ calc-alkaline volcanism

2

3: E-169 (64.08 % SiO ), Erciyes calc-alkaline volcanism2: E-196 (51.24 % SiO2), Erciyes calc-alkaline volcanism

2

a

b

Figure 6. Chondrite-normalised REE element patterns for the Central Anatolian volcanic rocks.Chondrite normalising values are from Boynton (1984).

trace element concentrations. The presence of pyroxeneand plagioclase is supported by decreasing Y (cpx-

liquidKdY≈1.2, plg-liquidKdY≈0.15) contents vs Th.Fractionation trends of rock suites from the Erciyes stratovolcano are quite similar and consistent withpetrographically proposed fractionation trends(plg[%50]+amp[%50]) (Figure 10a–c).

Assimilation-fractional Crystallization Effect

As a result of heat transfer from hot magma to the coldcrust, magmas, which are undergone fractionalcrystallization and assimilated around the crust, are verycommon (De Paolo 1981; Spera & Bohrson 2001;Thompson et al. 2002; Kuritani et al. 2005). Most oflarge-volume continental silicic magmas are combinationof continental rocks and mantle-derived basaltic melts (De


12

b

0.1

1

10

100

1000

CsRb

BaTh

UK

TaNb

LaCe

PrP

NdZr

HfSm

TiGd

TbDy

YHo

ErTm

YbLu

12

3

45

3: H-26 (55.12 % SiO ), Keçikalesi tholeiitic volcanism2

4: H-7 (55.71 % SiO ), Hasandağ calc-alkaline volcanism5: H-36 (60.10 % SiO2), Hasandağ calc-alkaline volcanism

2

1: H-49 (48.66 % SiO ), Hasandağ alkaline volcanism2: H-62 (48.26 % ), Hasandağ alkaline volcanism

2

SiO2

Hasandağ volcanismN

-Mor

bno

rmal

ised

0.1

1

10

100

1000

Erciyes volcanism

3: E-148 (47.76 % SiO ), Koçdağ alkaline volcanism2

4: E-102-b (58.61 % SiO ), Koçdağ calc-alkaline volcanism5: E-103 (62.19 % ), Koçdağ calc-alkaline volcanism

2

SiO2

6: E-169 (64.08 % SiO ), Erciyes calc-alkaline volcanism1: E-196 (51.24 % ), Erciyes calc-alkaline volcanism

2

SiO2

6

12

3

45

CsRb

BaTh

UK

TaNb

LaCe

PrP

NdZr

HfSm

TiGd

TbDy

YHo

ErTm

YbLu

N-M

orb

norm

alis

ed

a

Figure 7. N-MORB normalised multi-element patterns for the Central Anatolian volcanic rocks. N-MORB normalising values are from Sun & McDonough (1989).

Paolo et al. 1992). Although physical mixing formed byhybridization is preserved in xenoliths and xenocrysts,most workers state that chemical evidence (particularlyradiogenic isotopes) are important for mixing betweenmagmas with contrasting geochemical compositions(Beard et al. 2005). By the early 1970’s it was recognizedthat isotopic variations in the Sierra Nevada batholith, forexample, correlated with the character of the basemenetand, implicitly, required mixing of crustal- and mantle-derived melts (Kistler 1974). Recent geochemical worksindicated that fractional crystallization and assimilationare the most important processes for differentiation ofvolcanic rocks (De Paolo 1981; Spera & Bohrson 2001).Mantle-derived magmas are subjected to contaminationwhen they rise through the thick crust and therefore, areaffected by the contamination with certain degree.

In order to examine the effects of magma minglingand assimilation processes in rocks of the Hasandağ andErciyes stratovolcanoes, theoretical AFC models (De Paolo1981) were used that are based on trace elementchemistry. In the modeling, ‘r’ value (the ratio of

assimilation to fractional crystallization) was used as theAFC numeric index.

Since mantle xenoliths are absent in rocks of theHasandağ and Erciyes stratovolcanoes, alkaline rockscomprising the most primitive rocks in the region wereused in studies regarding end-member of the mantlesource. The average crust values provided by Taylor &McLennan (1985) were used as the crust composition.

Distribution coefficients (D) used in AFC modelingwere computed from mineral/melt distributioncoefficients (Kd) which were used in previous fractionalcrystallization modeling (Keskin 1994; Keskin et al.1998). AFC vectors for the Hasandağ and Erciyesstratovolcanoes were separately computed. Alldistribution coefficients were calculated consideringfractional assemblages of (plg[50%]+amp[50%]) for theHasandağ stratovolcano and (plg[505%]+amp[50%]) for theErciyes stratovolcano.

Assimilation and fractional crystallization effects of allthe rock samples from the Hasandağ and Erciyesstratovolcanoes were studied on the Rb vs Rb/Thdiagrams. In the AFC modeling, Rb/Th ratio was usedsince Rb and Th are not affected significantly bycrystallizing phases. Rb/Th ratio of the crust is higherthan basaltic lavas and increases in this ratio may bepossible reflections of the crustal assimilation. Rubidiumwas used as the fractionation index. Rb is stronglyretained in biotite which is not involved significantly in thecrystallization assemblages of volcanic rock suites exceptfor some high silica rocks. Calculated theoretical curveswere included to the Rb vs Rb/Th diagrams. In thismodeling, due to potential variations in distributioncoefficient values during the course of fractionalcrystallization and uncertainties in end-membercompositions, ‘r’ values represent rather broadapproximations. However, r values significantly reflectthe most primitive composition of rocks and AFC effectsbetween the rocks. Similar trends and assimilation ratiosare also reported by Keskin et al. (1998) for the EastAnatolian post-collisional volcanic rocks.

As shown from the diagram, the Hasandağ alkalinerocks is the rock group which is the least contaminated bythe crust. This rock group is found between r= 0.01 and0.2 theoretical curves. Most of the Keçikalesi tholeiiticand Hasandağ calc-alkaline rock samples are observed atr= 0.2–0.3 while a few samples are at r>0.4 (Figure11a).


13

subd

ucti

onen

rich

men

t

decr

easin

g degr

eeof

parti

alm

eltin

g

MANTLE

ARRAY

Th

/Yb

Ta/Yb0.01 0.1 1.0 10

0.01

0.1

1.0

10

100

averagecrust



Keçikalesi tholeiitic volcanismErciyes calc-alkaline volcanism



N-MORB

Figure 8. Th/Yb against Ta/Yb log±log diagram (after Pearce 1983)for the Central Anatolian volcanic rocks. Average OIB andMORB values used for comparasion are from Sun &McDonught (1989).

AFC effects on samples from the Erciyes stratovolcanoare examined in Figure 11b. In this diagram, the Koçdağalkaline rocks that comprise the most primitive rocks ofvolcanism are found between theoretical vectors of r=0.01 and 0.1. All the calc-alkaline rocks are foundbetween similar vectors. In diagram, Rb/Th ratios span arather narrow range particularly for the Koçdağ alkalineand Erciyes calc-alkaline rocks, although there is asignificant increase in Rb concentrations indicating thatfractional crystallization was an effective process duringthe magmatic evolution.

Mantle-melting Modeling

In mantle melting modeling, non-modal batch meltingequations of Shaw (1970) were used. REE abundancesand their ratios were used for limiting the sourcecharacteristics of magmas. Distribution coefficients ofREE were taken from McKenzie and O’Nions (1991,1995). Values proposed by Kostopoulos & James (1992)were utilized for the modal mineralogies and meltingratios.

In the modeling, we first attempted to test whetherthe trace element composition of depleted MORB mantle


14

1

10

100

1 10 100

10

9

7

6

5

432

1

8

Rb

Y

10 1001

10

1

1

2

3

4

5

6

7

8

9

10

Rb

Sm

1

10

100

1 10

1

234

5

7,8

Th

Y

9

6

100

1) %65 pl+%25 cpx+%10 opx (I)2) %50 pl+%30 am+%10 cpx+%10 opx (I)3) %50 pl+%50 cpx (A)4) %50 pl+%50 am (I)5) %50 pl+%50 am (A)6) %20 pl+%60 cpx+%10 opx+%10 ol (I)7) gt (I)8) am (I)9) cpx (A)10) %50 pl+%15 cpx+%35 ol (B)11) %50 pl+%15 cpx+%20 ol+%15 gt (B)

11

11 11

Hasandağ alkaline volcanismHasandağ calc-alkaline volcanismKeçikalesi tholeiitic volcanism

alkaline rocks alkaline rocks

tholeiitic and calc-alkaline rocks

a b

c

Figure 9. Trace-element diagrams used to study the petrogenesis of the Hasandağ strato volcanicrocks. Diagram showing theoretical Rayleigh fractionation vectors for 50% crystallisationof the phase combinations indicated in the inset (a– logRb–logY; b– logTh–logY; c– logRb–logSm). Thick marks on each vector correspond to 5% crystallization intervals am–amphibole; pl– plagioclase; ol– olivine; cpx– clinopyroxene; opx– orthopyroxene; gt–garnet; B– basic; I– intermediate; A– acidic; numbers refer to phase proportions.

(DMM) can be a suitable source for the Central Anatolianvolcanic rocks. DMM was assumed to represent theasthenospheric mantle circulation which is characterizedby a theoretical depleted MORB source proposed byMcKenzie and O’Nions (1991, 1995). Since the mantlesource mineralogy is not accurately known, the modellingwas based on a mixture of three different sourcecompositions; garnet lherzolite, spinel lherzolite andgarnet-spinel lherzolite (50%: 50%). In the modelings,priority was given to incompatible element concentrations(LRE) which are not significantly affected by changes inthe source mineralogy.

Theoretic melting curves in the La-Ce, La-La/Nd andLa/Ce-La/Yb diagrams are computed based on the batch

melting modeling (Figure 12). Melting degrees and sourcecompositions used in calculations are shown on themelting curves. As shown from diagrams in Figure 12,melting degrees of the Keçikalesi tholeiitic rocks, theHasandağ alkaline rocks and Koçdağ alkaline rocks are 8–9%, 4–5% and 3–8%, respectively.

In order to determine the source mineralogy, La-La/Sm and La/Sm-Sm/Yb diagrams (MREE/HREE andLREE/MREE ratios are considered) were used (Figure13). Since La and Sm are not significantly affected bychanges in the source mineralogy, they may be very usefulfor determination of bulk chemical composition of thesource (Aldanmaz et al. 2000). These diagrams were usedto distinguish the melting of garnet-lherzolite and spinel-


15

Th

123

4

5

7,8Y

96

1

10

100

1 10 100100

9

Rb

Y123

5

6, 10

7

8

4

1

10

100

1 10

1

10

1 10 100

Sm

10

4

1

6

8

3

95

7

Rb

2

1) %45 pl+%15 cpx+%10 opx+%30 am (I)2) %50 pl+%25 am+%25 cpx+ (I)3) %50 pl+%50 cpx (A)4) %50 pl+%50 am (I)5) %50 pl+%50 am (A)6) %50 pl+%30 cpx+%20 ol (I)7) gt (I)8) am (I)9) cpx (A)10) %50 pl+%15 cpx+%35 ol (B)

Erciyes calc-alkaline volcanismKoçdağ calc-alkaline volcanismKoçdağ alkaline volcanism

a b

c

Figure 10. Trace-element diagrams used to study the petrogenesis of the Erciyes strato volcanicrocks. Diagram showing theoretical Rayleigh fractionation vectors for 50% crystallisationof the phase combinations indicated in the inset (a– logRb–logY; b– logTh–logY; c– logRb–logSm). Thick marks on each vector correspond to 5% crystallization intervals am–amphibole; pl– plagioclase; ol– olivine; cpx– clinopyroxene; opx– orthopyroxene; gt–garnet; B– basic; I– intermediate; A– acidic; numbers refer to phase proportions.

lherzolite sources. Melting of spinel-lherzolite sourcecauses only a small change in La/Sm ratio of the melt,because none of these elements is fractionated by spinelfacies melting. Likewise, only a small change occurs inSm/Ym ratio of the melt. Therefore spinel-lherzolitemelting is observed as a horizontal trend. Melting ofgarnet-lherzolite source, on the other hand, producesmelt with La/Sm ratio higher than the mantle array andthe melt fraction produced from such a source will bedisplaced from the mantle array towards higher La/Smand Sm/Yb ratios on a La/Sm-Sm/Yb diagram. The resultsare given in Table 2, together with estimated mantlesource composition for the Central Anatolian volcanicrocks.

C1 chondrite-normalized REE diagram includingdepleted MORB mantle (DMM), primordial mantle (PM)and estimated using the above approach Central Anatolianmantle is shown in Figure 14. As shown from this figure,the computed Central Anatolian mantle composition ismore enriched in LREE’s with respect to DMM and PM. Inaddition, LREE concentrations are not affected by garnetor spinel lherzolite source while M-HREE concentrationsare significantly affected by the source mineralogy. Figure14 indicates that the mantle sourcing the CentralAnatolian volcanic rocks is represented by a source withgarnet + spinel lherzolite composition and the abundanceof garnet component is more pronounced for theHasandağ alkaline rocks.


16

1

10

10 100Rb

Rb

/Th

A

B r =0.011

r =0.12

r =0.23

r =0.34

r =0.45r =0.5

6

r =0.77r =0.85

8C /C =FL 0

(D-1)

Hasandağ alkaline volcanismHasandağ calc-alkaline volcanismKeçikalesi tholeiitic volcanism

composition component A:Taylor & McClennan (1985)Rb= 91; Th= 1.4D(Rb)= 0.1; D(Th)= 0.08

composition component B:C94-14 (Şen 2004)et al.Rb= 14.13; Th= 4

a

1

10

100

1 10 100

Rb

/Th

Rb

A

B

r =0.45

r =0.011

r =0.12

r =0.23

r =0.34

r =0.56

r =0.858

r =0.77

composition component A:Taylor & McClennan (1985)Rb= 91; Th= 1.4D(Rb)= 0.1; D(Th)= 0.08

composition component B:ER-24 (Notsu 1995)et al.Rb= 6; Th=1.5

C /C =FL 0

(D-1)

Erciyes calc-alkaline volcanismKoçdağ calc-alkaline volcanismKoçdağ alkaline volcanism

b

Figure 11. Rb/Th vs Rb diagram for the Central Anatolian volcanc rocks showing theresults of AFC (assimilation-fractional crystallization modelling) as discussedin the text.

Source Characteristics and Melting Mechanisms

The Central Anatolian volcanism is characterized by calc-alkaline, alkaline and tholeiitic type volcanic rocks whichwere formed at various stages of volcanic activity and donot show any systematic relationships. Based ongeochemical characteristics, the Central Anatolian calc-alkaline and tholeiitic volcanism is reported to be similarto Andean basalts (Hickey et al. 1986), basaltic rocks

from the Dikili-Ayvalık-Bergama area (Aldanmaz et al.2000) and Big Pine volcanic rocks (Ormerod et al. 1991)which are generally derived from lithospheric mantleenriched in subduction components (Şen et al. 2004). Thealkaline rocks resemble within plate oceanic island basalt(OIB) and to some extent they also indicate mantle sourcewith a subduction component.

The Central Anatolian Quaternary volcanism is mainlyrelated to an extensional tectonism without an activesubduction process (Deniel et al. 1998; Şen et al. 2004).As mentioned previously, considering the trace elementcontent of volcanic rocks in MORB normalized multi-element diagrams, LIL element concentrations aresignificantly enriched with respect to HFS elements whichcan be attributed to either the addition of subductioncomponent to the source or crustal contamination onmantle-derived primary magmas.

As shown from the Th/Yb-Ta/Yb diagram, most ofsamples have compositions indicative of derivation fromsubduction modified source and resulting melts possiblymodified further by crustal contamination. Alkaline rocksseem to be less affected from these processes. Inaddition, low Zr/Nb (Hasandağ alkaline rocks: 8–10,Erciyes alkaline rocks: 14–17) and high La/Yb (Hasandağalkaline rocks: 13–15, Erciyes alkaline rocks: 6–14)ratios observed in alkaline rocks of the Hasandağ andErciyes stratovolcanoes may indicate derivation from amantle source similar to those of alkaline OIB magmas. Information of calc-alkaline series, a combination ofassimilation and fractional crystallization processes areeffective.

Alkaline rocks of the Hasandağ stratovolcano arecharacterized by low Sr isotopic ratios and the crustcontribution is decreased in time (Deniel et al. 1998).Alkaline rocks of the Erciyes stratovolcano have beenreported to have 87Sr/86Sr ratios higher than the typicalMORB but significantly lower than the chondritic silicateearth compositions and this may indicate the presence ofan OIB source in the development of this magma(Kürkçüoğlu et al. 1998).

Hofmann & White (1982) state that enriched isotopecompositions of oceanic island basalts is due to old crustmaterial sent back to the mantle. On the other hand,McKenzie and O’Nions (1983) suggest that lower parts ofthe lithosphere thickening in the orogenic belts aredetached in time and separated from the lithosphere and


17

1

La/Yb

1%

1%

2%

2%

2%

4%

6%

8%10%

4%

10%

4%

15%

25%

20%25%

10

100

1 10 100

La

Ce

4%

6%

8%

10%

15%

20%

25%

4%

6%

8%

10%

15%

20%

25%

garnet lherzolitepartial melting curve

spinel lherzolitepartial melting curve

2%

0.1

1

10

La

La/Nd

4%6%

1%2%

6% 4%2%

1%

garnet-spinel lerzolitepartial melting curve

10 100

8%10%

10% 8%

La/Ce1

10

100

0.1

0.01%

0.5%

0.01%

0.5%

8%

8%





source compositionLa: 1.44Ce: 3.14




source compositionLa: 1.44Nd: 1.86

source compositionLa: 1.44Ce: 3.14Yb: 0.41

a

b

c

garnet-spinel lerzolitepartial melting curve

Figure 12. Plots of (a) Ce and (b) La/Nd against La shows meltingcurves obtained using the non-modal batch meltingequations (Shaw 1970) and data from Central Anatolianrocks. The absolute abundances of plotted incompatibleelements are unaffected by the mantle mineralogy,therefore the sole control is the bulk chemical compositionof the source. (c) Variation of La/Yb vs La/Ce showingmelting curves obtained using non-modal batch meltingequations for varying proportions of source mineralogy.The points on the curves represent degree of partialmelting. Mineral/matrix partition coefficients are from thecompilation of McKenzie & O’Nions (1991, 1995).

added to convectional currents in the asthenosphere and,they rise as plumes in certain places thus forming thesource of oceanic island basalts.

McKenzie (1989) defined the continental basalticmagmas as those derived probably from themetasomatized subcontinental lithospheric mantle(SCLM). He also described melts that are significantly

enriched in incompatible elements and volatiles and alsoupward migration of such melts which may result inmetasomatism at the base of SCLM and the combinationof low-degree partial melt of the asthenospheric mantlematerial under the continents. Anderson (1994) proposedconvective mantle that is represented by trace elementand isotope characteristics (like DMM) depleted prior to


18

0.1

1

10

0.1 1 10 100

15%

La/Sm

Sm/Yb

PMDMM N-MORB

OIB

E-MORB

depletion enrichment

1%2%

4%

10%

20%

6%8%

15%

1%2%

4%10%

6%

0.5%

1%2%4%8%0.01%0.5%

1%2%4%10%0.5%

15%

1%2%

4%

10%

20%

6%8%

0.01%

0.5%

WAM

gt+sp-lherzolite

sp-lherzolite

gt-lherzolite

sp-lherzolite

0.1

1

10

0.1 1 10 100

La

La/Sm

DMM

PMWAM

N-MORB

E-MORBdepletion

enrichment

4%

6%

10%

20%30%

4%6%10%

20%25%

2%

1%

1%2%

10%15%

20%30%

50%70%

50%70%

OIB

2%4%

25% 20%15% 10%

0.1%

2%4%

1%gt-lherzolite

sp-lherzolite

gt-lherzolite

sp-lherzolite

gt+sp-lerzolite

source compositionLa: 1.44Sm: 0.42




source compositionLa: 1.44Sm: 0.42Yb: 0.41

CAM

CAM

a

b

Figure 13. Plots of La/Sm vs La and Sm/Yb vs La/Sm showing melt curves (orlines) obtained using the non-modal batch melting equations of Shaw(1970). Melt curves are drawn for spinel-lherzolite (with mode andmelt mode of ol0.578 + opx0.270 + cpx0.119 + sp0.033 and ol0.10 + opx0.270

+ cpx0.50 + sp0.13) and for garnet-lherzolite (with mode and meltmode of ol0.598 + opx0.211 + cpx0.076 + gt0.115 and ol0.05 + opx0.20 +cpx0.30 + gt0.45; respectively; Kostopoulos & James 1992).Mineral/matrix partition coefficients and DMM are from thecompilation of McKenzie & O’Nions (1991, 1995); PM, N-MORB andE-MORB compositions are from Sun & McDonough (1989). WesternAnatolian Mantle (WAM) composition is from Aldanmaz et al.(2000). The heavy line represents the mantle array defined usingDMM and PM compositions. Solid curves (or lines) are the meltingtrends from DMM and Central Anatolian Mantle (CAM), respectively.Tick marks on each curve (or line) correspond to degrees of partialmelting for a given mantle source.


19

Tabl

e 2.

Man

tle s

ourc

e co

mpo

sitio

ns f

or t

he c

entr

al A

nato

lian

volc

anic

roc

ks.

Trac

e el

emen

t co

ncen

trat

ions

wer

e es

timat

ed u

sing

thr

ee d

iffer

ent

min

eral

mod

es a

s ga

rnet

-lher

zolit

e, s

pine

l–lh

erzo

lite

and

garn

et+

spin

el-lh

erzo

lite

(50%

– 50

%).

LaCe

PrN

dSm

Eu

Gd

TbD

yH

oE

rTm

YbLu

(F)

Est

imat

ion

of m

antl

e so

urce

com

posi

tion

of

the

Cent

ral A

nato

lia s

uite

Deg

ree

of

part

ial

1.44

3.14

0.40

1.86

0.42

0.14

0.50

0.09

0.61

0.13

0.40

0.06

0.41

0.06

mel

ting

Conc

entr

atio

n of

tra

ce e

lem

ent

in t

le li

quid

(ga

rnet

-lher

zolit

e)

0.25

5.76

12.5

21.

597.

411.

680.

581.

960.

352.

400.

531.

540.

231.

540.

230.

207.

1515

.48

1.95

9.01

2.01

0.68

2.27

0.40

2.62

0.55

1.55

0.22

1.32

0.18

0.15

9.45

20.2

92.

5211

.50

2.49

0.82

2.68

0.45

2.87

0.58

1.55

0.20

1.15

0.14

0.10

13.9

029

.40

3.57

15.8

83.

281.

043.

280.

533.

180.

611.

550.

191.

020.

120.

0817

.13

35.8

54.

2818

.74

3.76

1.17

3.60

0.57

3.32

0.62

1.55

0.18

0.98

0.11

0.06

22.3

245

.91

5.34

22.8

54.

401.

334.

000.

613.

480.

631.

550.

180.

940.

110.

0432

.01

63.8

47.

1129

.26

5.30

1.55

4.48

0.66

3.65

0.64

1.55

0.17

0.90

0.10

0.02

56.5

910

4.71

10.6

240

.68

6.65

1.84

5.11

0.72

3.84

0.66

1.55

0.17

0.87

0.10

0.01

91.8

715

4.01

14.0

950

.55

7.64

2.04

5.49

0.75

3.95

0.66

1.55

0.17

0.85

0.10

0.00

513

3.46

201.

4216

.85

57.5

38.

242.

155.

700.

774.

000.

671.

550.

170.

840.

090.

0001

239.

9328

8.47

20.8

566

.53

8.94

2.28

5.93

0.79

4.05

0.67

1.55

0.17

0.84

0.09

Conc

entr

atio

n of

tra

ce e

lem

ent

in t

le li

quid

(sp

inel

-lher

zolit

e)

0.25

5.76

12.5

51.

607.

451.

700.

581.

980.

362.

440.

541.

580.

241.

600.

240.

27.

1515

.49

1.96

9.07

2.05

0.70

2.38

0.43

2.92

0.65

1.89

0.29

1.91

0.28

0.15

9.42

20.2

12.

5311

.57

2.59

0.87

2.98

0.54

3.63

0.81

2.36

0.36

2.37

0.35

0.1

13.7

929

.10

3.57

15.9

83.

511.

163.

980.

714.

791.

073.

120.

483.

130.

460.

0816

.93

35.3

04.

2718

.86

4.10

1.34

4.60

0.82

5.49

1.23

3.59

0.55

3.59

0.53

0.06

21.9

344

.87

5.32

23.0

04.

921.

595.

450.

976.

441.

444.

230.

644.

210.

610.

0431

.11

61.5

57.

0429

.47

6.15

1.94

6.68

1.18

7.77

1.75

5.13

0.77

5.08

0.74

0.02

53.5

397

.97

10.4

341

.00

8.19

2.50

8.62

1.51

9.82

2.23

6.52

0.98

6.41

0.92

0.01

83.6

913

9.14

13.7

250

.98

9.83

2.92

10.0

81.

7611

.30

2.57

7.55

1.13

7.37

1.05

0.00

511

6.51

176.

1516

.30

58.0

410

.92

3.19

11.0

21.

9112

.22

2.79

8.20

1.23

7.97

1.13

0.00

0118

9.22

238.

2719

.98

67.1

612

.25

3.50

12.1

22.

1013

.28

3.04

8.94

1.33

8.66

1.22

Conc

entr

atio

n of

tra

ce e

lem

ent

in t

le li

quid

(sp

inel

+gt

-lher

zolit

e)

0.25

5.76

12.5

41.

597.

431.

690.

581.

970.

362.

420.

541.

560.

241.

570.

230.

27.

1515

.49

1.95

9.04

2.03

0.69

2.32

0.41

2.77

0.60

1.72

0.25

1.61

0.23

0.15

9.43

20.2

52.

5311

.54

2.54

0.85

2.83

0.50

3.25

0.69

1.95

0.28

1.76

0.25

0.1

13.8

429

.25

3.57

15.9

33.

401.

103.

630.

623.

990.

842.

340.

332.

080.

290.

0817

.03

35.5

84.

2818

.80

3.93

1.26

4.10

0.69

4.41

0.92

2.57

0.36

2.28

0.32

0.06

22.1

245

.39

5.33

22.9

24.

661.

464.

720.

794.

961.

042.

890.

412.

570.

360.

0431

.56

62.6

97.

0829

.37

5.72

1.74

5.58

0.92

5.71

1.20

3.34

0.47

2.99

0.42

0.02

55.0

610

1.34

10.5

240

.84

7.42

2.17

6.86

1.12

6.83

1.44

4.04

0.58

3.64

0.51

0.01

87.7

814

6.57

13.9

150

.77

8.73

2.48

7.79

1.26

7.62

1.62

4.55

0.65

4.11

0.57

0.00

512

4.99

188.

7916

.58

57.7

89.

582.

678.

361.

348.

111.

734.

870.

704.

410.

610.

0001

214.

5826

3.37

20.4

166

.84

10.5

92.

899.

031.

448.

671.

865.

250.

754.

750.

66

ascension and interacted at the beginning with shallowand enriched mantle. This portion proposed by Anderson(1994) is below the thermal boundary layer or within theasthenosphere. The same author also states that enrichedpart of the shallow mantle may correspond to chemicalcharacteristics of a source region similar to continentalbasaltic or OIB magmatism. In order to explaincomposition of all oceanic and continental within platebasic volcanics which necessitate an enriched source,McKenzie and O’Nions (1995) proved that SCLM could beenriched and remelted afterwards.

No evidence was found for spatial and chronologicsystematic variation of the source component and themagmatism sourcing alkaline and calc-alkaline volcanics inthe Central Anatolia. Therefore, it is thought that alkalineand calc-alkaline magmatic products were represented bymagmas derived from a heterogeneous mantle sourcewith melts of varying rates and affected from crustalcontamination at varying rates.

In order to examine melting mechanisms which gaverise to generation of Central Anatolian post-collisionalvolcanism, P-T diagram of Pearce et al. (1990) proposedfor the Eastern Anatolia was taken into consideration.Pearce et al. (1990) state that a typical non-cratonic

continental lithosphere has a mechanical boundary layer(MBL) of varying thickness and is separated from theconvective asthenosphere with a thermal boundary layer(TBL). Bailey (1987) suggests that geothermal gradientof MBL cuts the lithospheric solidus which ismetasomatized by small melt pockets ascending fromasthenosphere to the lithosphere and it is metasomatizedin lower parts of the region.

The potential mechanisms of magma generation in the(subduction component bearing) heterogeneouslyenriched mantle source of the Central Anatolian volcanismmay include: (a) mantle-plume, (b) slab break-off, (c)extension and associated strike-slip faulting, and (d)lithospheric delamination and associated orogeniccollapse. These possible options are discussed below.

(a) Melting of lithospheric mantle as a result ofthermal perturbation in association with ascending mantleplume or asthenospheric mantle may give rise toformation of widespread volcanism in the CentralAnatolia. It was proposed by McKenzie & Bickle (1988)that asthenosphere could start to melt at a potentialtemperature of 1280 ºC and at an initial depth of 50 kmunder a stress factor of 2.5. Under hot mantle conditions,mantle plume can also start melting at great depthswithout substantial stress (at 100 km, potential


20

1

10

100

1000



Koçdağ alkaline volcanismKeçikalesi tholeiitic volcanism

2%4%6%8%10%

2%4% 6%

8%10%

2%

4%6%8%10%

0

Chondrite(C1)normalised

C /C =1/[D +F(1-P)]L 0 0

PMCAMDMM

garnet-lherzolitespinel-lherzolitegarnet+spinel lherzolite

Figure 14. Chondrite (C1) normalised REE patterns showing the compositions of modelled batch meltsproduced by melting of a CAM source with mineral proportions of garnet-lherzollite, spinel-lherzolite and garnet (%50) + spinel (%50) lherzolite. The modal mineral proportions anddistribution coefficients used same as figure 13. Normalised values are srom Boynton(1984) and the batch melting equation is from Shaw (1970).

temperature is 1480 ºC) and affects the regionaltopography. The typical mantle plume is described byWhite and McKenzie (1989) as a spreading center (limitedto 150–200 km) forming a mushroom-shaped hot spotwith a diameter of 1000–2000 km that dynamicallyuplifts an area of 1000–2000 km in diameter. There is noregional stress evidence that could be associated with amantle plume responsible for the post-collisional CentralAnatolian volcanism and there are no topographicindicators necessary for such a mantle plume either.

(b) Melting of lithospheric mantle due to thermalperturbation arising from ascending asthenosphericmantle by slab breakoff could be an alternativemechanism for the source of Central Anatolia volcanism.According to Davies and Von Blanckenburg (1995), slabbreakoff is very common for many orogenic belts and itexplains the coexistence of ultra-high pressuremetamorphics and mantle-derived magmatism. Slabshould be separated at the early stages of continentalcollision (Davies & Von Blanckenburg 1995; Wong A Ton& Wortel 1997). This is due to decrease in subductionrate by the positive lifting power of continentallithosphere intruded into the subduction zone (Gerya etal. 2004).

In contrast to previously proposed several models forthe Eastern Anatolia, Şengör et al. (2003) and Keskin(2003, 2007) suggested the model of slab breakoff andsteepening under widely spread subduction accretionarycomplex. In the model proposed by Davies & VonBlanckenburg (1995), if the subduction rate is 1 cm/year,the breakoff was stated to occur at a depth of 50–120km. The volcanism was thought to start as a result ofrelease of this substantial load and upwelling of less denseasthenospheric mantle. Unlike the model of ‘original slabbreakoff’ proposed by Davies & Von Blanckenburg(1995), Keskin (2003), Şengör et al. (2003) suggest anew model where subducting oceanic lithosphere beneaththe accretionary prism gets steepened and eventuallydetached from the continental lithosphere. According tothese workers, the melting occurs in the asthenosphericmantle and eventually volcanic activity spreads in a largearea apart from the suture zone.

Under the light of above information, a similarmechanism can be discussed for Central Anatoliaconsidering its closeness to Eastern Anatolia and similargeochemical characteristics of the two regions. Given thedistance of the volcanic centres to any possible subduction

zone in this region, however, the volcanism in CentralAnatolia is unlikely to be directly affected by thesubduction process that resulted in formation of theEastern Anatolia collision system. One can infer that theclosure of inner-Tauride Ocean could be a possiblesubduction. Geochemical characteristics of Paleocene–Eocene volcanic and intrusive products in the region andtheir radiometric ages may indicate that these volcanicsare cogenetically formed (Gürer & Aldanmaz 2002). Inmost of previous works, these events are evaluated in theconcept of arc magmatism associated with closure of theinner-Tauride Ocean (Şengör &Yılmaz 1981; Görür et al.1984; Gökten & Floyd 1987; Erdoğan et al. 1996). Morerecent studies, however, showed that the regionalmagmatic activity took place in a post-collisionalenvironment and may have an intra-plate character(Gürer & Aldanmaz 2002). This is evident from thetemporal evolution of the activity that changes from asyn-collisional episode (e.g., S-type, peraluminousgranites) through a post-collisional, calc-alkaline episode(e.g., I-type, high-K, metaluminous granites) to a post-collisional, within-plate alkaline episode (e.g. A-type,peralkaline granites) (e.g., Göncüoğlu & Türeli 1993;İlbeyli & Pearce 1997; Boztuğ 2000, 2008; Köksal &Göncüoğlu 2008).

The age of tholeiitic volcanism of the Hasandağ stratovolcano ranges from 12.4±0.6 to 13.7±0.3 Ma (Besanget al. 1977) and alkaline type latest products are dated as34000±7000 years (Aydar 1992). Alkaline rocks as thefirst products of the Erciyes stratovolcano are of4.39±0.28 Ma old (Ayrancı 1969). Since the volcanism inboth regions is not a subduction-type volcanism related tothe closure of inner-Tauride Ocean and the subsequentslab breakoff event is not evidenced by significantchronological and spatial systematic, this model is notsuitable one. In addition, considering the alkaline lavas ofvarying ages, the presence of a heterogeneous mantlesource is reasonable.

(c) Another alternative mechanism is that regionalextension developing in association with strike-slip faultsand the melting of mantle lithosphere due to thermalperturbation could be related to tectonic regime in theCentral Anatolia. Formation of lithospheric melting incontinental extensional environments is studied by severalworkers (Latin et al. 1989; Hawkesworth & Norry 1982;Menzies & Hawkesworth 1987). Lithospheric thermalperturbation arising from upwelling of hot asthenosphere


21

is the reason for lithospheric extension necessary for themelting of lithospheric mantle material (Dixon et al.1981) or the position of solidus above the metasomatizedlevel. Since melting point of these rocks will be lower inmetasomatized mantle conditions, small thermalirregularities could be sufficient for starting of melting.Thus, even small-size extensions can trigger volcanism.

The magma generation in pull-apart basins inassociation with N–S-trending fractures and strike-slipregimes in the Eastern Anatolia is attributed to fastlithospheric extension and small-scale delamination underthe pull-apart basins (Dewey et al. 1986). AlthoughPearce et al. (1990) suggest that delamination process isthe dominant mechanism for widespread magmageneration in the same region they also state that othermechanisms such as formation of pull-apart basinsaccompany these processes. Cooper et al. (2002)proposed that mantle upwelling and melting under thestrike-slip fault systems could be the origin of maficmagmas in northwest Tibet. Likewise, Aldanmaz et al.(2005, 2006) mention partial effect of strike-slip faultsin generation of alkaline volcanics in NW Turkey.

The first group of active fault systems extendingthrough the Central Anatolian volcanism is left-lateralEcemiş fault zone and right-lateral Tuz Gölü fault zoneand the second is N60–70° E-trending normal faults thatare parallel to the volcanic axis and concordant with themain eruption centres (Toprak & Göncüoğlu 1993). TheErciyes basin is a pull-apart basin with an importantstrike-slip component (Koçyiğit & Beyhan 1998). Due tolateral stress arising from transtensional tectonic regimein the Central Anatolia coupled with regional extension,formation of thermal perturbation in the mantlelithosphere could be a possible mechanism. In addition, asmentioned previously, little effect of AFC on the Erciyesstrato volcano could be explained with crustal thinning asa result of extensional tectonism in the region associatedwith pull-apart basin.

(d) Another alternative mechanism could be thermalperturbation and melting triggered by lithosphericdelamination and associated orogenic collapse followingthe post-collisional lithospheric thickening. Thermalperturbation and melting arising from lithosphericdelamination could be the reason of volcanism in theCentral Anatolia. Although thermal perturbation inassociation with strike-slip faults in lithosphere can bethought as the mechanism to start melting, delamination

of thermal boundary layer (TBL) could be proposed as themore efficient mechanism (Pearce et al. 1990). Ingeneral, one of the fundamental consequences of thelithospheric delamination is the collapse event associatedwith rapid uplifting and extensional system which occursas a result of isostatic replacing of relatively dense (cold)lithospheric mantle by the less dense (hot) asthenosphericmantle (Dewey 1988; England & Houseman 1998;Nelson 1992; Platt & England 1993). Thickening TBL isdenser and colder than the surrounding medium and sincethickened levels of metasomatized lithosphere are in closecontact with the asthenosphere, they are convectivelyreplaced by the asthenosphere (Houseman et al. 1981;England & Houseman 1998). Due to thermalperturbation arising from direct contact betweenmetasomatized mantle lithosphere and hotasthenosphere, delamination of TBL could result inmelting of metasomatized mantle lithosphere.

Pearce et al. (1990) stated that the close contactbetween thickened metasomatized lithospheric mantleand asthenosphere is an effective mechanism forcollisional magma generation developed in the EasternAnatolia. While delaminated block of the mantlelithosphere sinks into the asthenosphere, increasingmelting may release water (Elkins-Tanton 2004). Thesetwo mechanisms play an important role in formation ofcommon partial melt in the mantle and can producewidespread volcanism in the region.

As a result, delamination of TBL and associatedsubstantial thermal perturbation could be alternativemodels that facilitate melting of continental lithosphereresponsible for initiation of Central Anatolia volcanism. Inaddition, the close association of volcanism with thestrike-slip tectonism in the region indicates that thissystem possibly contributes to the magma rising.

Conclusion

The Hasandağ and Erciyes stratovolcanoes represented bywidespread calc-alkaline volcanism are two importantmembers of collisional volcanism in the Central Anatolia.The calc-alkaline products are accompanied by tholeiiticand alkaline rocks in the Hasandağ volcano and by alkalinerocks in the Erciyes stratovolcano. Rocks of bothvolcanisms show a wide composition spectrum frombasalt to rhyolite.

In the chondrite normalized diagrams the rocks fromthe Hasandağ and Erciyes stratovolcanoes, display


22

uniform and sub-parallel light rare earth element patternsand show relatively enrichment in LREE. In the multi-element diagram normalized to N-MORB, tholeiitic andcalc-alkaline rocks of the Hasandağ strato volcano show asignificant enrichment in LIL elements (Rb, Ba, Th, U andK) and LREE’s and relatively depletion in HFS elements(Ta, Nb, Ti and Hf). In the calc-alkaline rocks of theErciyes stratovolcano that are characterized by similarpatterns, LIL elements such as Th, U and K as well asLREE’s also show a distinct enrichment. In the alkalinerocks of the both stratovolcanoes, LIL elements are inlower concentrations in comparison to other rock groupsand they do not have negative Ta and Nb anomaliesindicating that alkaline rocks are derived from a differentsource. In MORB normalized multi-element diagrams ofthe same rocks, it was observed that LIL elements aresignificantly enriched compared to HFS elements. Thisenrichment may be attributed to the effect of subductioncomponent in the source region or crustal contamination.Depletion of Hf and Ti with respect to other HFS elementscan be explained with mantle enrichment inintracontinental processes or derivation from a sourcewith small degree partial melts.

According to fractionation vectors constructed on thebasis of variations in trace element concentrations, rocksof the Hasandağ volcanism generally display two differenttrends. Tholeiitic and calc-alkaline rocks are modified byfractionation of amphibole and plagioclase while alkalinerocks are represented by fractionation effects ofplagioclase, pyroxene and amphibole minerals. In alkalineand calc-alkaline rocks of the Erciyes volcano,fractionation trends are similar and conformable withthose of plagioclase and amphibole.

Theoretically constructed AFC modelings show thatHasandağ alkaline rocks are less affected from the crustalcontamination in comparison to tholeiitic and calc-alkalinerocks. However, in alkaline and calc-alkaline rocks of theErciyes volcano which display similar AFC trends, theeffect of crustal contamination is less than that ofHasandağ volcanics.

According to theoretical melting curves calculated withnon-modal batch melting model, the melting degrees ofHasandağ and Erciyes volcanic rocks are 8–9% for theKeçikalesi tholeiitic rocks, 4–5% for the Hasandağalkaline rocks and 3–8% for the Koçdağ alkaline rocks.

The Central Anatolian volcanic rocks are generated by amantle source of garnet + spinel lherzolite compositionwhile Hasandağ alkaline rocks are dominated by thegarnet composition. The calculated Central Anatolianmantle composition is more enriched in LREE’s incomparison to DMM and PM.

Accordingly, it is believed that alkaline and calc-alkaline magmatic products were generated from aheterogeneous mantle source under varying meltingconditions and affected by the crustal contamination atvarying degrees.

Considering its geochemical characteristics, theCentral Anatolian volcanism seems to be derived from alithospheric mantle that is enriched in subductioncomponents. Solutions which are strongly enriched byincompatible elements and volatiles result inmetasomatism at the base of continental lithosphere. TBLwhich is thickened by lithospheric thickening inassociation with collisional tectonism is denser and colderthan its environs. When thickened levels of metasomitizedlithosphere are in close contact with asthenosphere, it canbe convectively replaced by the asthenosphere.Delamination of TBL may result in melting due to thermalperturbation arising from direct contingence ofmetasomitized mantle lithosphere to hot asthenosphere.Hence, the Central Anatolian volcanism can be generatedby melting of continental lithosphere as a result of asubstantial thermal perturbation occurring bydelamination of TBL. In addition, the observation thatvolcanism is closely associated with strike-slip tectonismin the region may indicate that this system possiblycontributes to the magma rising. The mechanismsuggested for the melt generation is rather similar to thatproposed by Pearce et al. (1990) and Keskin et al. (1998)for the east Anatolian and by the Aldanmaz et al. (2000)for the west Anatolian volcanisms.

Acknowledgements

This research was financially supported by the ResearchFund of Kocaeli University (Project No: 2003/74). We aregrateful to Dr. Ercan Aldanmaz for his constructivediscussion. We thank the guest editors, Prof. Nilgün Güleçand Prof. Durmuş Boztuğ, for the editorial handling ofthe paper.


23


24

ReferencesAKIMAN, O., ERLER, A., GÖNCÜOĞLU, M.C., GÜLEÇ, N., GEVEN, A., TÜRELİ, K.

& KADIOĞLU, Y. 1993, Geochemical characteristics of granitoidsalong the western margin of the Central Anatolian CrystallineComplex and their tectonic implications. Geological Journal 28,371–382.

ALDANMAZ, E., PEARCE, J.A., THIRWALL, M.F. & MITCHELL, J.G. 2000.Petrogenetic evolution of Late Cenozoic post-collision volcanism inwestern Anatolia, Turkey. Journal of Volcanology and GeothermalResearch 102, 67–95.

ALDANMAZ, E., GOURGAUD, A. & KAYMAKCI, N. 2005. Constraints on thecomposition and thermal structure of the upper mantle beneathNW Turkey: evidence from mantle xenoliths and alkali primarymelts. Journal of Geodynamics 39, 277–316.

ALDANMAZ, E, KÖPRÜBAŞI, N., GÜRER, Ö.F., KAYMAKCI, N. & GOURGAUD, A.2006. Geochemical constraints on the Cenozoic, OIB-type alkalinevolcanic rocks of NW Turkey: implications for mantle sources andmelting processes. Lithos 86, 50–76.

ANDERSON, D.L. 1994. The sublithospheric mantle as the source ofcontinental flood basalts; the case against the continentallithosphere and plume had reservoirs. Earth and Planetary ScienceLetters 123, 269–280.

AYDAR, E. 1992. Etude volcano-structurale et magmatologique dustratovolcan Hasan Dagi, Anatolie Centrale, Turquie. PhD Thesis,Universite Blasie Pascal Clermont-Ferrand France [unpublished].

AYDAR, E. & GOURGAUD, A. 1998. The geology of Mount Hasanstratovolcano, Central Anatolia, Turkey. Journal of Volcanologyand Geothermal Research 85, 129–152.

AYDAR, E. & GOURGAUD, A. 2002. Garnet-bearing basalts: an examplefrom Mt. Hasan, Central Anatolia, Turkey. Mineralogy andPetrology 75, 185–201.

AYDAR, E., GOURGAUD, A., DENIEL, C., LYBERIS, N. & GÜNDOĞDU, M.N. 1995.Le volcanisme quaternaire d’Anatolie central Turquie: associationde magmatisme calco-alcalin et alcalin en domaine de convergence.Canadian Journal of Earth Sciences 327, 1058–1069.

AYDAR, E., GÜNDOĞDU, M.N., BAYHAN, H. & GOURGAUD, A. 1993. Kapadokyabölgesinin Kuvaterner yaşlı volkanizmasının volkanik, yapısal vepetrolojik incelenmesi [Volcanic, stuctural and petrologic study ofthe Quaternary volcanism in Cappadocia region]. Doğa TürkYerbilimleri Dergisi 3, 25–42 [in Turkish with English abstract].

AYRANCI, B. 1969. Zur Petrologie und Geologie des Erciyes-Vulkan-Geibetes bei Kayseri in Zentralanatolien/Türkei. Ph.D. Thesis,Würzburg University, Germany [unpublished].

AYRANCI, B. 1991. The magnificent volcano of central Anatolia: Mt.Erciyes near Kayseri. Bulletin of Technical University of İstanbul44, 375–417.

BAILEY, D.K. 1987. Mantle metasomatism-perspective and prospect. In:FITTON, J.G. & UPTON, B.G.J. (eds), Alkaline Igneous Rocks.Geological Society, London, Special Publications 30, 1–13.

BAYER, H.J., HOTZL, H., JADO, A.R., ROCHER, B. & VOGGENREILER, W. 1988.Sedimentary and structural evolution of the northwest ArabianRed Sea Margin. Tectonophysics 204, 1–15.

BEARD, J.S., RAGLAND, P.C. & CRAWFORD, M.L. 2005 Reactive bulkassimilation: a model for crust-mantle mixing in silicic magmas.Geology 33, 681–684.

BESANG, C., ECKHARDT, F.J., HARRE, W., KREUZER, H. & MULLER, P. 1977.Radiometrische Altersbestimmungen an NeogenenEruptivegesteinen der Turkei. Geologisches Jahrbuch B25, 3–36.

BONIN, B. 1990. From orogenic to anorogenic settings: evolution ofgranitoid suites after a major orogenesis. Geologica Journal 25,261–270.

BOYNTON, W.V. 1984. Geochemistry of the rare earth elements:meteorite studies. In: HENDERSON, P. (ed), Rare Earth ElementGeochemistry, Elsevier Amsterdam, 63–114.

BOZTUĞ, D. 2000. S-I-A-type intrusive associations: geodynamicsignificance of synchronism between metamorphism andmagmatism in Central Anatolia, Turkey. In: BOZKURT, E.WINCHESTER, J.A. & PIPER, J.D.A. (eds), Tectonics and Magmatismin Turkey and the Surrounding Area. Geological Society, London,Special Publications 173, 441–85.

BOZTUĞ, D. 2008. Petrogenesis of the Kösedağ pluton, Suşehri NE Sivas,East-Central Pontides, Turkey. Turkish Journal of Earth Sciences17, 241–262.

BUKET, E. & TEMEL, A. 1998. Major-element, trace element and Sr-Ndisotopic geochemistry, and genesis of Varto (Muş) volcanic rocks,eastern Turkey. Journal of Volcanology and Geothermal Research85, 405–22.

COCHRAN, C.R. 1981. The Gulf of Aden: structure and evolution of ayoung ocean basin and continental margin. Journal of GeophysicalResearch 86, 263–287.

COOPER, K.M., REID, M.R., DUNBAR, N.W. & MCINTOSH, W.C. 2002. Originof mafic magmas beneath nortwestern Tibet: constraints from230Th–238U disequilibria. Geochemistry, Geophysics 3, articleno.1065.

DAVIES, J.H. & BLANCKENBURG, V.F. 1995. Slab breakoff: A model oflithosphere detachment and its test in the magmatism anddeformation of collisional orogens. Earth and Planetary ScienceLetters 129, 85–102.

DENIEL, C., AYDAR, A. & GOURGAUD, A. 1998. The Hasan Dağıstratovolcano Central Anatolia, Turkey: evolution from calc-alkaline to alkaline magmatism in a collision zone. Journal ofVolcanology and Geothermal Research 87, 275–302.

DE PAOLO, D. 1981. Trace element and isotopic effects of combined wallrock assimilation and fractional crystallization. Earth andPlanetary Science Letters 53, 189–202.

DE PAOLO, D.J., PERRY, F.V. & BALDRIDGE, W.S. 1992. Crustal vs mantlesources of granitic magmas: A two parameter model based on Ndisotopic studies. Royal Society of Edinburgh Transactions, EarthSciences 83, 439–446.

DEWEY, J.F. 1988. Extensional collapse of orogens. Tectonics 7, 1123–1139.

DEWEY, J.F., HEMPTON, M.R., KIDD, W.S.F. & ŞENGÖR, A.M.C. 1986.Shortening of continental lithosphere; the neotectonics of easternAnatolia, a young collision zone. In: COWARD, M.P. & RIES, A.C.(eds), Collision Tectonics. Geological Society Special Publications19, 3–36.

DIXON, J.E., FITTON, J.G. & FROST, R.T.C. 1981. The tectonic significanceof post–Carboniferous igneous activity in the North Sea basin. In:ILLING, L. & HOBSON, G.D. (eds), Petroleum Geology of theContinental Shelf of North West Europe. Blackwell London, 29–52.

ELKINS-TANTON, L.T. 2004. Continental magmatism caused bylithospheric delamination. http://www.mantleplume.org/LithDelam.html.

ENGLAND, P.C. & HOUSEMAN, G.A. 1988. The mechanics of the TibetanPlateau. Philosophical Transactions of the Royal Society, LondonA326, 301–320.

ERDOĞAN, B., AKAY, E. & UĞUR, S.M. 1996. Geology of the Yozgat regionand evolution of the collisional Çankırı basin. InternationalGeology Review 38, 788–806.

FYITKAS, M., INNOCENTI, F., MANETTI, P., PECCERILLO, A. & VILLARI, L. 1984.Tertiary to Quaternary evolution of volcanism in the Aegeanregion. In: DIXON, J.E. & ROBERTSON, A.H.F. (eds), The GeologicalEvolution of the Eastern Mediterranean. Geological Society,London, Special Publications 17, 687–699.

GERYA, T.V., YUEN, A.D. & MARECSH, W.V. 2004. Thermomechanicalmodelling of slab detachment. Earth and Planetary Science Letters226, 101–116.

GÖKTEN, E. & FLOYD, P. 1987. Geochemistry and tectonic environment ofthe Şarkışla area volcanic rocks in Central Anatolia: Turkey.Mineralogical Magazine 51, 553–559.

GÖNCÜOĞLU, M.C., TOPRAK, V., KUŞÇU, I., ERLER, A. & OLGUN, E. 1991b.Orta Anadolu Masifi Batı Kesiminin Jeolojisi, Bölüm I: GüneyKesim [Geology of Western Part of the Central Anatolian Massif,Part I: Southern Section]. Türkiye Petrolleri Anonim Ortaklığı(TPAO) Rapor no. 2909 [in Turkish, unpublished].

GÖNCÜOĞLU M.C. & TÜRELİ, T.K. 1993. Petrology and geodynamicinterpretation of plagiogranites from Central Anatolian ophiolites(Aksaray, Turkey). Turkish Journal of Earth Sciences 2, 195–203.

GÖRÜR, N., OKTAY, F.Y., SEYMEN, İ. & ŞENGÖR, A.M.C. 1984. Paleotectonicevolution of the Tuzgölü basin complex; Sedimentary record of aNeotethyan closure. In: DIXON, J.E. & ROBERTSON, A.H.F. (eds), TheGeological Evolution of the Eastern Mediterranean. GeologicalSociety, London, Special Publications 17, 467–82.

GÜRER, Ö.F. & ALDANMAZ, E. 2002. Origin of the Upper Cretaceous–Tertiary sedimentary basins within the Tauride-Anatolide platformin Turkey. Geological Magazine 139, 191–197.

HAWKESWORTH, C.J. & NORRY, M.J. 1982. Continental Basalts and MantleXenoliths. Shiva, Cheshire, U.K. 230–249.

HICKEY, R.L., FREY, F.A., GERLACH, D.C. & ESCOBAR, L.J. 1986. Multiplesource basaltic arc rocks from the southern volcanic zone of theAndes (34o–41oS): trace element and isotopic evidence forcontributions from subducted oceanic crust. Journal ofGeophysical Research 91, 5963–5983.

HOFMANN, A.W. & WHITE, W.M. 1982. Mantle plums from acient oceaniccrust. Earth and Planetary Science Letters 58, 44–62.

HOUSEMAN, G.A., MCKENZIE, D.P. & MOLNAR, P. 1981. Convectiveinstability of a thickened boundary layer and its relevance for thethermal evolution of continental convergent belts. Journal ofGeophysical Research 86, 6115–6132.

INNOCENTI, F., MANETTI, P., MAZZUOLI, R., PASQUARE G. & VILLARI, L. 1982.Anatolia and North-Western Iran. In: THORPE, R.S. (ed), Andesites:Orogenic Andesites and Related Rocks. The Open University,Department of Earth & Environmental Sciences, Milton Keynes,UK, 327–349.

IRVINE, T.N. & BARAGAR, W.R.A. 1971. A guide to the chemicalclassification of the common volcanic rocks. Canadian Journal ofEarth Sciences 8, 448–523.

İLBEYLİ, N. & PEARCE, J. A. 1997. Petrogenesis of the collision-relatedAnatolian granitoids, Turkey. European Union of Geosciences(EUG), Strasbourg, France 9, p. 502.

IZZELDIN, A.Y. 1987. Seismic gravity and magnetic surveys in central partof the Red Sea: their interpretation and implications for thestructure and evolution of the Red Sea. Tectonophysics 143, 269–306.

JACKSON, J.A. & MCKENZIE, D.P. 1984. Active tectonics of the Alpine-Himalayan belt between western Turkey and Pakistan.Geophysical Journal of Royal Astronomical Society 77, 185–264.

KELLER, J. 1982. Mediterranean island arcs. In: THORPE, R.S. (ed),Andesite: Orogenic Andesite and Related Rocks. Wiley, New York,307–325.

KESKİN, M. 1994. Genesis of Collision-related Volcanism on theErzurum–Kars Plateau, Northeastern Turkey. PhD Thesis,University of Durham, UK [unpublished].

KESKİN, M. 2002. FC-Modeler: A Microsoft Excel spreadsheet programfor modeling Rayleigh fractionation vectors in closed magmaticsystems. Computers & Geosciences 28, 919–928.

KESKİN, M. 2003. Magma generation by slab steepening and breakoffbeneath a subduction-accretion complex: an alternative model forcollision-related volcanism in Eastern Anatolia, Turkey.Geophysical Research Letters 30 (24), 8046, doi: 10.1029/2003GL0180019.

KESKİN, M. 2007. Eastern Anatolia: a hotspot in a collision zone withoutha mantle plume. In: FOULGER, G.R. & JURDY, D.M. (eds), Plates,Plumes and Planetary Processes. Geological Society of AmericaSpecial Paper 430, 693–722, doi:10. 1130/2007. 2430 (32).

KESKİN, M., PEARCE, J.A. & MITCHELL, J.G. 1988. Volcano-stratigraphyand geochemistry of collision-related volcanism on the Erzurum-Kars Plateau, northeastern Turkey. Journal of Volcanology andGeothermal Research 85, 355–404.

KISTLER, R.W. 1974. Phanerozoic batholiths in western North America:a summary of some recent work on variations in time, space,chemistry, and isotopic compositions. Annual Review of Earth andPlanetary Sciences 2, 403–417.


25

KURITANI, T., KITAGAWA, H. & NAKAMURA. E. 2005. Assimilation andfractional crystallization controlled by transport process of crustalmelt: implications from an alkali basalt dacite suite from RishiriVolcano, Japan. Journal of Petrology 46, 1421–1442.

KOÇYİĞİT, A. & BEYHAN, A. 1998. A new intracontinental transcurrentstructure: the Central Anatolian Fault Zone, Turkey.Tectonophysics 284, 317–336.

KOSTOPOULOS, D.K. & JAMES, S.D. 1992. Parameterization of the meltingregime of the shallow upper mantle and the effects of variablelithospheric stretching on mantle modal stratification and trace-element concentrations in magmas. Journal of Petrology 33,665–691.

KÖKSAL, S. & GÖNCÜOĞLU, M.C. 2008. Sr and Nd isotopic charaestericticof some S-, I- and A- type granitoids from Central Anatolia.Turkish Journal of Earth Sciences 17, 111–127.

KÜRKÇÜOĞLU, B., ŞEN, E., AYDAR, E., GOURGAUD, A. & GÜNDOĞDU, N. 1998.Geochemical approach to magmatic evolution of Mt. Erciyesstratovolcano, Central Anatolia, Turkey. Journal of Volcanologyand Geothermal Research 85, 473–494.

LATIN, D.M., DIXON, J.E. & FITTON, J.G. 1989. Rift related magmatism inthe North Sea basin. In: BLUNDELL, D.J. & GIBBS, A. (eds), TectonicEvolution of the North Sea Rifts. Oxford University Press, Oxford,101–104.

LE BAS, M.J., LE MAITRE, R.W., STRECKEISEN, A. & ZANETTIN, B. 1986. Achemical classification of volcanic rocks based on the totalalkalisilica diagram. Journal of Petrology 27, 445–450.

LE PENNEC, J.L., BOURDIER, J.L., FROGER, J.L., TEMEL, A., CAMUS, G. &GOURGAUD, A. 1994. Neogene ignimbrites of the Nevşehir plateauCentral Turkey: stratigraphy, distribution and source constraints.Journal of Volcanology and Geothermal Research 63, 59–87.

LE PICHON, X. & GAULIER, J.M. 1988. The rotation of Arabia and theLevant fault system. Tectonophysics 153, 271–294.

LYBERIS, N., YÜRÜR, T., CHOROWICZ, J., KASAPOĞLU, E. & GÜNDOĞDU, N.1992. The East Anatolian Fault: an oblique collisional belt.Tectonophysics 204, 1–15.

MCKENZIE, D. 1989. Some remarks on the movement of small meltfractions in the mantle. Earth and Planetary Science Letters 95,53–72.

MCKENZIE, D. & YILMAZ, Y. 1991. Deformation and volcanism in westernTurkey and the Aegean. Bulletin of Technical University 44, 345–373.

MCKENZIE, D.P. 1972. Active tectonics of the Mediterranean region.Geophysical Journal of Royal Astronomical Society 30, 109–185.

MCKENZIE, D.P. & BICKLE, M.J. 1988. The volume and composition ofmelt generated by extension of the lithosphere. Journal ofPetrology 29, 625–679.

MCKENZIE, D. P. & O’NIONS, R. K. 1983. Mantle reservoirs and oceanisland basalts. Nature 301, 229–231.

MCKENZIE, D.P. & O’NIONS, R.K. 1991. Partial melt distributions frominversion of rare earth element concentrations. Journal ofPetrology 32, 1021–1091.

MCKENZIE, D.P. & O’NIONS, R.K. 1995. The source regions of oceanisland basalts. Journal of Petrology 36, 133–159.

MENZIES, M.A. & HAWKESWORTH, C.J. 1987. Mantle Metasomatism.Academic Press, London.

NELSON, K.D. 1992. Are crustal thickness variations in old mountain beltslike the Appalachians a consequence of lithospheric delamination?Geology 20, 498–502.

NOTSU, K., FUJITANI, UI, T., MATSUDA, J. & ERCAN, T. 1995. Geochemicalfeatures of collision-related volcanic rocks in central and easternAnatolia, Turkey. Journal of Volcanology and GeothermalResearch 64, 171–192.

ORMEROD, D.S., ROGERS, N.W. & HAWKESWORTH, C.J. 1991. Melting of thelithosphere mantle: Inverse modeling of alkali olivine basalts fromthe Big Pine volcanic field, California. Contributions to Mineralogyand Petrology 108, 305–317.

PASQUARE, G., POLI, S., VEZZOLI, L. & ZANCHI, A. 1988. Continental arcvolcanism and tectonic setting in Central Anatolia, Turkey.Tectonophysics 146, 217–230.

PEARCE, J.A. 1983. Role of the sub-continental lithosphere in magmagenesis at active continental margins. In: HAWKESWORTH, C.J. &NORRY, N.J. (eds), Continental Basalts and Mantle Xenoliths. Shiva,Cheshire, UK, 230–249.

PEARCE, J.A., BENDER, J.F., DE LONG, S.E., KIDD, W.S.F., LOW, P.J., GÜNER,Y., ŞAROĞLU, F., YILMAZ, Y., MOORBATH, S. & MITCHELL, J.G. 1990.Genesis of collision volcanism in Eastern Anatolia, Turkey. Journalof Volcanology and Geothermal Research 44, 189–229.

PECCERILLO, A. & TAYLOR, S.R. 1976. Geochemistry of Eocene calcalkalinevolcanic rocks from the Kastamonu area, Northern Turkey.Contributions to Mineralogy and Petrology 58, 63–91.

PLATT, J.P. & ENGLAND, P.C. 1993. Convective removal of lithospherebeneath mountain belts: thermal and mechanical consequences.American Journal of Science 293, 307–336.

SHAW, D.M. 1970. Trace element fractionation during anatexis.Geochimica et Cosmochimica Acta 34, 237–243.

SPERA, F.J. & BOHRSON, W.A. 2001. Energy-constrained open systemmagmatic processes I: General model and energy-constrainedassimilation and fractional crystallization (EC-AFC) formulation.Journal of Petrology 42, 999–1018.

SUN, S.S. & MCDONOUGH, W.F. 1989. Chemical and isotopic systematicsof oceanic basalts: implications for mantle composition andprocesses. In: SAUNDERS, A.D. & NORRY, M.J. (eds), Magmatism inthe Ocean Basins. Geological Society, London, Special Publications42, 313–345.

ŞEN, E., KÜRKÇÜOĞLU, B., AYDAR, E., GOURGAUD, A. & VINCENT, P.M. 2003.Volcanological evolution of Mount Erciyes stratovolcano and originof the Valibaba Tepe ignimbrite (Central Anatolia, Turkey).Journal of Volcanology and Geothermal Research 261, 1–22.

ŞEN, P.A., TEMEL, A. & GOURGAUD, A. 2004. Petrogenetic modeling ofQuaternary post-collisional volcanism: a case study of central andeastern Anatolia. Geological Magazine 141, 81–98.

ŞENGÖR, A.M.C. 1980. Principles of Turkey Neotectonics. GeologicalAssociation Conference Series no. 2, [in Turkish].


26

ŞENGÖR, A.M.C., GÖRÜR, N. & ŞAROĞLU, F. 1985. Strike-slip faulting andrelated basin formation in zones of tectonic escape: Turkey as acase study. In: BIDDLE, K.T. & CHRISTIE-BLICK, N. (eds), Strike-SlipDeformation, Basin Formation and Sedimentation. Society ofEconomic Palaeontologists Mineralogists, Special Publications 37,227–264.

ŞENGÖR, A.M.C., ÖZEREN, S., ZOR, E. & GENÇ, T. 2003. East Anatolianhigh pleteau as a mantle-supported, N–S shortened domalstructure. Geophysical Research Letters 30 (24), doi:10.1029/2003GL017858,

ŞENGÖR, A.M.C. & YILMAZ, Y. 1981. Tethyan evolution of Turkey; a platetectonic approach. Tectonophysics 75, 181–241.

TAYLOR, S.R. & MCLENNAN, S.M. 1985. The Continental Crust: ItsComposition and Evolution. Blackwell Scientific, Oxford.

TEMEL, A. 1992. Kapadokya Eksplozif Volkanizmasının Petrolojik veJeokimyasal Özellikleri [Petrologic and GeochemicalCharacteristics of Cappadocian Explosive Volcanism]. PhD Thesis,Hacettepe Üniversitesi [in Turkish with English abstract,unpublished].

TEMEL, A., GÜNDOĞDU, N., GOURGAUD, A. & LE PENNEC, J.-L. 1998.Ignimbrites of Cappadocia, Central Anatolia, Turkey: petrology,geochemistry and evolution through time. Journal of Volcanologyand Geothermal Research 85, 447–441.

THOMPSON, A.B., MATILE, L. & ULMER, P. 2002. Some thermal constraintson crustal assimilation during fractionation of hydrous, mantle-derived magmas with examples from central Alpine batholiths.Journal of Petrology 43, 403–422.

TOPRAK, V. 1994. Central Kızılırmak fault zone: northern margin of theCentral Anatolian volcanics. Turkish Journal of Earth Sciences 3,29–38.

TOPRAK, V. & GÖNCÜOĞLU, M.C. 1993. Tectonic control on thedevelopment of Neogene-Quaternary Central Anatolian VolcanicProvince, Turkey: Geological Journal 28, 357–369.

TURNER, S., ARNAUD, N., LIU, J., ROGERS, N., HAWKESWORTH, G., HARRIS, N.,KELLEY, S., VAN CALSTEREN, P. & DENG, M. 1996. Post-collision,shoshonitic volcanism on the Tibetan Plateau: implications forconvective thinning of the lithosphere and the source of oceanisland basalts. Journal of Petrology 37, 45–71.

WHITE, R. & MCKENZIE, D.P. 1989. Magmatism at rift zones: thegeneration of volcanic continental margins and flood basalts.Journal of Geophysical Research 94, 7685–7729.

WONG, S.Y.M., TON, A. & WORTEL, M.J.R. 1997. Slab detachment incontinental collision zones: an analysis of controlling parameters.Geophysical Research Letters 24, 2095–2098.

YILMAZ, Y., GÜNER, Y. & ŞAROĞLU, F. 1998. Geology of the Quaternaryvolcanic centers of the east Anatolia. Journal of Volcanology andGeothermal Research 85, 173–210.

YÜRÜR, T. & CHOROWICZ, J. 1998. Recent volcanism, tectonics and platekinematics near the junction of the African, Arabian and Anatolianplates in the eastern Mediterranean. Journal of Volcanology andGeotermal Research 85, 1–15.


27

geochemical characteristics of mafic and intermediate...

Documents