Geochemical Constraints on the Provenance, MineralSorting and Subaerial Weathering of Lower Jurassic andUpper Cretaceous Clastic Rocks of the Eastern Pontides,
Yusufeli (Artvin), NE Turkey
ABDURRAHMAN DOKUZ1 & ERKAN TANYOLU2
1Karadeniz Technical University, Gümüflhane Engineering Faculty, Department of Geological Engineering,TR–29000 Gümüflhane, Turkey (E-mail: [email protected])
2F›rat University, Department of Geological Engineering, TR–23119 Elaz›¤, Turkey
Abstract: An integrated petrographic and geochemical study of the shales and sandstones of Early Jurassic andLate Cretaceous age in the Yusufeli area (Turkey) was carried out to obtain more information on their provenance,sedimentological history and tectonic setting, as well as to evaluate the influence of weathering, hydraulic sortingand recycling processes upon source-rock signature. Depending on their matrix and mineralogical content, theLower Jurassic sandstones are identified as arkosic arenite and wacke, while the Upper Cretaceous sandstones aredefined as lithic arenite and wacke.
Elemental ratios such as La/Sc, Cr/Ni, Co/Th, Th/Sc and Cr/Th indicate a mafic source for the majority of eachof the sandstone sequences. However, in several samples, high Th/Sc and low Cr/Th ratios suggest a contributionfrom a felsic source. Using element-ratio diagrams, all of the samples plot along a curve consistent with a two-component mixing model, consisting of a dominant mafic and a subordinate felsic source. The relatively low Cr/Niratios provide no support for significant amounts of ultramafic lithologies in the source area. The lower trace-element contents of the studied sandstones in relation to post-Archean Australian shales (PAAS), which arerepresentative of the composition of the upper continental crust, indicate that recycling processes in their sourceareas were probably less intense than those of the PAAS. The more LREE-depleted patterns of the shales relativeto sandstones likely do not reflect changes in source-area composition but, rather, variations in mineral sorting,chemical weathering and/or sediment recycling.
Chemical index of alteration (CIA) indices observed in the Lower Jurassic and Upper Cretaceous sediments(average of 53 & 46, respectively) suggest that their source area underwent moderate degrees of chemicalweathering processes. However, an upward increase in the values of the CIA indices in the Lower Jurassic profilesmay indicate that the source area gradually underwent more intense chemical weathering, possibly due to climaticand/or tectonic variations. On the other hand, an upward decrease in the CIA indices of the Upper Cretaceousprofiles generally demonstrate that the sediments were derived from a relatively less-weathered source terrain,reflecting an increased erosion rate likely due to increasing tectonic activity.
The compositional immaturity of the analysed sandstone samples is typical of subduction-relatedenvironments, and their SiO2/Al2O3 and K2O/Na2O ratios and Co, Sc, Th and Zr contents reflect their oceanic andcontinental-arc settings.
Key Words: shale and sandstone, geochemistry, provenance, hydraulic sorting, alteration, Yusufeli-Artvin,northeastern Turkey
Do¤u Pontidler’deki Erken Jura ve Geç Kretase Yafll› K›r›nt›l› Kayaçlar›n KaynakAlan›, Mineral Boylanmas› ve Yüzeysel Ayr›flmas› Üzerine Jeokimyasal
S›n›rlamalar, Yusufeli (Artvin), KD Türkiye
Özet: Yusufeli (Artvin) yöresindeki Erken Jura ve Geç Kretase yafll› fleyl ve kumtafllar›; kaynak alanlar›,sedimantolojik geliflimleri ve tektonik ortamlar› hakk›nda daha fazla bilgi edinmek, ayr›flma, hidrolik boylanma vetortul-döngü olaylar›n›n kaynak alan karakteristikleri üzerindeki etkisini de¤erlendirmek amac› ile petrografik vejeokimyasal aç›dan incelenmifltir. Hamur (matriks) ve mineralojik içeriklerine göre, Erken Jura yafll› kumtafllar›arkozik arenit ve vake, Geç Kretase kumtafllar› ise litik arenit ve vake olarak tan›mlanm›fllard›r.
Turkish Journal of Earth Sciences (Turkish J. Earth Sci.), Vol. 15, 2006, pp. 181-209. Copyright ©TÜB‹TAK
181
Introduction
Clastic sedimentary rocks are indicators of pastenvironments, giving clues to their compositions and evento their geodynamic settings by means of theircompositions. The provenance and geodynamicdevelopment of sandstone successions can be classified bya variety of methods, including petrographic analysis,whole-rock and mineral chemistry. Above all, clasticsediments can supply information on continental andoceanic source regions that have been eroded ormetamorphosed through subsequent tectonic processes(Nesbitt & Young 1982; McLennan 1989; McLennan et al.1993; Cullers 1994; Condie et al. 1995). In terrigenousrocks, some detrial minerals are characteristic of certainsource rocks and source regions. Thus, often only thedetrial components of clastic rocks reveal clues to sourceareas that have disappeared. The present study usespetrographic and geochemical methods on Lower Jurassicand Upper Cretaceous sandstones and shales from theYusufeli (Artvin) area to decipher the influence of source-rock characteristics, chemical weathering duringtransport and sedimentation, and post-depositionaldiagenetic reactions, all affecting the chemical record oftheir compositions and, consequently, evidenceconcerning their parental affinities (Nesbitt 1979; Cullerset al. 1979, 1987; Banfield & Eggelton 1989; McLennan1989; McLennan et al. 1993; Condie et al. 1995).
In this regard, it seems obvious that the LowerJurassic clastic succession has been derived from pre-
Jurassic basement that is currently represented by mainlymedium- to high-grade metamorphic rocks ofCarboniferous age (e.g., the Artvin-Yusufeli areametamorphic rocks, Pulur Massif: Dokuz 2000; Topuz etal. 2004; Topuz & Altherr 2004) and cross-cutting high-K granitoids of Permian age (e.g., the Gümüflhane andKöse granitoids: Y›lmaz 1972; Bergougnan 1987).However, while the rock fragments derived metamorphicrocks and granitoids are nearly absent, plagioclase andvolcanic lithic fragments form the main components inthese rocks. The main question here is whence thesevolcanic fragments have been derived. Were they derivedfrom the Late Palaeozoic basement where their exposuresdo not include such volcanic rocks (Y›lmaz 1972;Bergougnan 1987; Okay 1996; Dokuz 2000; Topuz et al.2004; Y›lmaz & Y›lmaz 2004; Erdo¤an & Güngör 2004;Koralay et al. 2004; Gerdjikov 2005), or syn-sedimentarily from Lower Jurassic volcanic activity? Dothey correspond to volcanic clastics derived from theassociated extrusive equivalents (Bozkufl 1990) ofPermian granitoid rocks (which are presently unexposedor nonexistent)? Similar questions are valid for thesecond, or Upper Cretaceous regressive, clasticsuccession. For instance, we have an imperfectunderstanding concerning whether the source area wasthe same or different from those materials deposited inthe Early Jurassic, and whether certain parts of the pre-Jurassic basement were maintained or were not in their‘source position’ until the end of Mesozoic. In the light of
GEOCHEMISTRY OF JURASSIC–CRETACEOUS ROCKS, NE TURKEY
182
La/Sc, Cr/Ni, Co/Th, Th/Sc ve Cr/Th gibi element oranlar› her iki istif için de mafik bir kaynak alana iflaretetmektedir. Bununla birlikte, bir çok örnekteki yüksek Th/Sc ve düflük Cr/Th oranlar› da felsik bir kayna¤›n katk›s›n›öngörmektedir. Örnekler, kullan›lan oransal diyagramlarda, mafik ve felsik kaynaklar›n katk›s› ile oluflan iki bileflenlibir kar›fl›m e¤risi boyunca da¤›lmaktad›rlar. Göreli olarak düflük Cr/Ni oranlar›, kaynak alanda ultramafiklitolojilerin olmad›¤›na iflaret etmektedir. Kumtafllar›n›n, üst k›tasal kabuk bileflimini temsil etti¤i düflünülen post-Arkeen Avusturalya fleyllerine (PAAS) göre daha az bollukta iz element içerikleri, kaynak alanlar›ndaki tortul-döngüolaylar›n›n olas›l›kla daha az yo¤unlukta oldu¤una iflaret etmektedir. fieylerin kumtafllar›na göre daha fazlatüketilmifl hafif nadir toprak element da¤›l›mlar› da, kaynak alandaki kayaçlar›n bileflimlerindeki de¤iflimden ziyademineral boylanmas›, kimyasal ayr›flma ve/veya tortul-döngüdeki de¤iflimleri yans›tmaktad›r.
Alt Jura ve Üst Kretase tortullar›nda gözlenen kimyasal ayr›flma indeksi (chemical index of alteration = CIA)(s›ras› ile ortalama 53 ve 46) kaynak alanlar›n›n orta derecede kimyasal ayr›flma olaylar›na u¤rad›¤›n›öngörmektedir. Bununla birlikte, CIA indeksinin Alt Jura istiflerinde yukar›ya do¤ru artmas›; kaynak alan›n,olas›l›kla iklim ve tektonik de¤iflimler nedeniyle, giderek daha fazla kimyasal ayr›flmaya u¤rad›¤›n› gösterebilir.Di¤er taraftan, Üst Kretase istiflerinde CIA indeksinin yukar› do¤ru genel bir azal›m›; tortullar›n göreli olarak dahaaz ayr›flan bir kaynaktan kaynakland›¤›n› göstermekte ve olas›l›kla da artan tektonik etkinlik nedeni ile artanerozyon oran›n› yans›tmaktad›r.
Analiz edilen kumtafllar›n›n bileflimsel olarak olgunlaflmam›fl olmas› yitimle iliflkili ortamlar›n tipik özelli¤idir.SiO2/Al2O3, K2O/Na2O oranlar› ve Co, Sc, Th ve Zr içerikleri okyanussal ada yay› ve k›tasal yay ortamlar›n›göstermektedir.
Anahtar Sözcükler: fleyl ve kumtafl›, modal analiz, jeokimya, kaynak alan, hidrolik boylanma, kimyasal ayr›flmaindeksi, Yusufeli-Artvin, Kuzeydo¤u Türkiye
A. DOKUZ & E. TANYOLU
183
the lack of geochemical data from the Mesozoic clasticrocks, the elemental abundance data and source-areainterpretations that we present will help to elucidate thepaleogeographical conditions of this area. In a moregeneral context, this study offers a chance to understandin detail the effects of various sedimentary processes onthe geochemical signature of sedimentary rocks.Although analyses of fine-clastic sediments, such asshales, are usually emphasized in geochemical provenancestudies, this sample set also allows direct comparisonbetween sandstones and shales. Assuming thatsandstones and shales of the same age are cogenetic,comparison of the chemical compositions of the twodifferent lithologies allows an evaluation of the control ofthe hydraulic sorting on elemental distributions.
Regional Geological Framework
The Pontides, one of the major tectonic units of Turkey,is a paleo-island arc ranging from Mid-Eocene to LatePalaeozoic in age (Ketin 1966; Ak›n 1978; fiengör &Y›lmaz 1981; Ak›nc› 1984; Bozkurt & Mittwede 2001).Its eastern part is conventionally divided into northernand southern zones (Ak›n 1978; Gediko¤lu et al. 1979;Okay & fiahintürk 1997; Bektafl et al. 1999) that differfrom each other in terms of rock associations. In thenorthern zone, comprising the axial part of the arc,Cretaceous to Eocene basic and acidic volcanic rocks withKuroko-type ore deposits and coeval granitoidspredominate (Arslan et al. 1997; fien et al. 1998; Karsl›et al. 2004; Aslan 2005) (Figure 1). In the southernzone, the Mesozoic section is represented predominantlyby sedimentary units that unconformably overlie theHercynian metamorphic and granitic basesement (Y›lmaz1972; Ço¤ulu 1975; Gediko¤lu et al. 1979; Keskin et al.1989; Okay 1996; Dokuz 2000; Topuz et al. 2004;Topuz & Altherr 2004; Y›lmaz & Y›lmaz 2004). It isgenerally agreed that the Mesozoic basin opened from thePermian until the Dogger as a back arc-basin south of thePontian arc and is a consequence of a southward-dippingoceanic lithosphere (fiengör et al. 1980; fiengör & Y›lmaz1981; Bektafl et al. 1984; Tüysüz 1990; Chatalov 1991;Ustaömer & Robertson 1993; Y›lmaz et al. 1997; Okay& fiahintürk 1997; Floyd et al. 2003; Okay & Göncüo¤lu2004; Pickett & Robertson 2004; Dean 2005). TheHamurkesen formation constitutes the lowermost part ofthe Mesozoic units and comprises pillow basalts, basalticto andesitic lithic tuffs, sandstones and shales, which
locally contain ‘ammonitico rosso’ horizons. Thissuccession has been interpreted as a product of amagmatic arc undergoing rifting (fiengör et al. 1980;fiengör & Y›lmaz 1981; Y›lmaz et al. 1997; Okay &fiahintürk 1997; Floyd et al. 2003). Upper Jurassic toLower Cretaceous platform-type carbonates concordantlyoverlie Middle Jurassic fine-grained clastic rocks and areascribed to the quiescent period that followed rapidtectonic subsidence. The major change in tectonic style –from extensional to compressional – began in theAlbian–Cenomanian (Tüysüz et al. 1995; Okay &fiahintürk 1997; Y›lmaz et al. 1997); this records aregressive turbiditic succession in the southern zone, andvolcanic to volcaniclastic rocks – associated with theirintrusive equivalents – in the northern zone throughoutthe eastern Pontides. It is suggested that the absence ofPalaeocene to Early Eocene lithologies, in general, may beexplained by the Mesozoic basin being raised up to sealevel at the end of the Upper Cretaceous. From thePalaeocene to Early Eocene, the continental margin wastelescoped into a series of stacked north-vergent thrustslices, as a consequence of major shortening along thesouthern zone. Timing the initiation of subduction andcontinental collision is contentious. Some authors suggestthat subduction began during the Jurassic or even earlier(Adamia et al. 1977), while others postulate that its onsetwas initiated during the Cenomanian–Turonian (fiengör &Y›lmaz 1981; Tüysüz et al 1995; Okay & fiahintürk1997). Most workers relate the Eocene volcanism to thenorthward subduction of the Tethyan ocean floor, andhave considered that the collision occurred during theOligocene (Tokel 1977; Ak›n 1978; Robinson et al.1995), while Okay & fiahintürk (1997) maintained a LatePalaeocene–Early Eocene age for continental collisionbetween the Pontides and the Anatolide-Tauride Platformon the basis of the absence of deformation in the Eocenerocks.
Stratigraphy
In the study area, the Lower Jurassic Hamurkesenformation unconformably overlies the metamorphicbasement (Figure 2) and is of variable thickness. Themaximun thickness of the formation, about 964 m, wasmeasured near the village of ‹flhan. Although anEarly–Middle Jurassic age has been assigned to theformation based on its stratigraphic location in the studyarea, it is not necessarily reliable. However, in some early
papers (Alp 1972; Y›lmaz 1972; Özer 1984), an earlyPliensbachian to Toarcian age was reported for the lowerboundary, largely based on ammonites from the‘ammonitico rosso’ horizons. As for the upper boundary,the presence of a Kimmeridgian microfossil assemblage(Protopenerolis striata Weynschenk, Trocholina alpinaLeupold, Koskinobullina socialis Cherci & Schroeder,Nantilaculina oolitica Mohler) in the micritic limestone atthe base of the overlying carbonates (Berdiga formation)indicate a pre-Kimmeridgian age for the top of theformation (Kemal Erdo¤an, personal communication2000). Dokuz (2000) described the formation asconsisting of two stratigraphic units: (i) a lower unit ofpillowed and massive basalt and basaltic andesiteintercalated with some thin-bedded silts and shales, and
(ii) an upper unit of shallow-water sedimentary rockswith some interbedded basic volcanic rocks. The lavaflows of the lower unit are massive or pillowed, with thepillows typically 0.5–1 m in diameter; the pillows possesschilled margins and calcite-, chlorite-, zeolite- and quartz-filled vesicles. Locally, massive basalt grades into pillowbasalt, possibly indicating – in part – that the massivevariety represents extrusive lava flows. Intrusive rocks(dolerite stocks) are scarce in the Hamurkesen formationand include narrow (< 2 m), cross-cutting doleritic dykes.Coal-bearing, thin-bedded green shale is the firstcomponent of the upper unit that overlies the basalts.Upward, the succession grades into a sandstone-shalealternation about 500 m thick and includes intercalatedbasic tuffs, agglomerates and breccias, from several
GEOCHEMISTRY OF JURASSIC–CRETACEOUS ROCKS, NE TURKEY
184
N
40
41
4241403938
0 85 km
BLACK SEA
Trabzon
TRABZON RÝZE
GÝRESUNORDU
NAFS
Kaçkar
Daglarý
Yusufeli
GÜMÜÞHANE
Tortum
Pulur DaðlarýAðvanis
Þebinkarahisar
Dereli
Çambaþý
Fatsa
Kösedað
ARTVÝN
Çoruh River
Ýspir
SinopÝstanbul
BAYBURT
Tokat
Köse
T
AP
BFANKARA
STUDYAREA
undifferantiated metamorphics(Palaeozoic)
Gümüþhane granite(Palaeozoic)
Jurassic-Cretaceous sediments
Lower Cretaceous sediments
Upper Cretaceous sediments
Upper Cretaceous volcano-sedimentary unit
Upper Cretaceous ophiolitic mélange
granitoids (Upper Cretaceous-Lower Tertiary)
gabbro/diorite (Upper Cretaceous-Lower Tertiary)
Middle Eocene volcano-sedimentary unit
Neogene volcanics
Neogene sediments
North Anatolian Fault System (NAFS)
Figure 1. Simplified geological map of the east of the eastern Pontides (modified after the Geological Map of Turkey, scale 1/500,000,MTA 1961) and the location of the mapped area. Inset shows the tectonic units of Turkey with the location of eastern Pontides.P– Pontides, A– Anatolides, T– Taurides and BF– Border Folds (from Ketin 1966).
metres up to 66 m in thickness. The top of the unit isrepresented by about 450 m of shales. The Hamurkesenformation and its equivalents have been interpreted by
many workers as the basin-fill deposits of an extensionalrift (fiengör et al. 1980; fiengör & Y›lmaz 1981; Bektaflet al. 1984; Chatalov 1991; Ustaömer & Robertson
A. DOKUZ & E. TANYOLU
185
KEMERLÝDAÐ
Ýþhan
K ý n a l ý ç a m2399
Tortum Lake
15
25
17
2730
48
36
3018
20
13
1445
28
35
30
14
30 27
30
27
6570
27
3529
75
55
30
3538
18
1530
67
56
23
48
32
35
Çaðlayan
Ormandibi
Çoruh R.
Çoruh
R.
Tortum R.
Tort
um R
.
Oltu R.
Çoruh R
.
22
26
25
30
KARADAÐ
2770
NYUSUFELÝ
1
2
3
4
metamorphics (Carboniferous)
gabbro (pre-Lower Jurassic)
tonalite-trondhjemite (pre-Lower Jurassic)
Lower-Middle Jurassic shales andsandstones (Hamurkesen formation)
Hamurkesen volcanics (Lower Jurassic)
0 2 km
Upper Juarassic-Lower Cretaceouscarbonates (Berdiga formation)
Upper Cretaceous sandstones and shales(Yusufeli formation)
Mid-Eocene sedimentary unit
Quaternary talus
the profile ofanalysed samples thrust fault
anticlineaxis
synclineaxis settlementstrike and dip15
strike-slipfault
Figure 2. Detailed geological map of the Yusufeli (Artvin) area, including profiles of the analysed samples (modified from Dokuz 2000).
1993; Y›lmaz et al. 1997; Okay & fiahintürk 1997; Floydet al. 2003). Crustal stretching has been attributed to theascent of a mantle plume that gave rise to the extrusionof the underlying Hamurkesen volcanic rocks.
On the basis of the determined important taxa andgenera (i.e., Hedbergella trocoidea Gandolfi, Hedbergelladelrioensis Carsey, Globigerina hoterivica Subbotina,Meandrospira favrei Charrolias, Brönniman & Zaninetti,Tintinopsella carpathica Murgeanu & Filipescu, Alveoseptajaccardi Schrodt, Protopeneroplis cf. trochangulataSeptfontaine, Tubiphytes morronensis Crescenti,Conicospirillina basiliensis Mohler, Pseudosyclamminalituus Yokoyama) (Kemal Erdo¤an, personalcommunication 2000), the age of the platform-typecarbonates (Berdiga formation) can be constrained asKimmeridgian–Barremian. The Yusufeli formation isrepresented by arc-derived turbidites of regressivecharacter and was deposited over the carbonate rocksduring the Aptian–Campanian time interval(Globotruncana bulloides Vogler, Globotruncana linneianad' Orbigny, Rosita fornicata Plummer, Pithonella ovalisKaufmann, Stomiosphaera sphaerica Kaufmann,Marssonella oxcona Reus, Lagenidae marginotruncanacoronata Bolli, Marginotruncana pseudolinneianaPessagno, Rotalipora ticinensis Gaundolfi, Rotaliporareichelli Mornod, Globigerinelloides ferreolensisMoullade, Hedbergela trocoidea Gaundolfi) (KemalErdo¤an, personal communication 2000). The unitconsists entirely of sedimentary rocks to the south of thearea, whereas intermediate and acidic volcanic rocks areincorporated into the unit with increasing order to thenorth. It comprises red-clastic lithologies (sandstone,siltstone, shale, micritic limestone, and chert) at fourstratigraphic horizons: one at the bottom, two in themiddle and another one at the top of the sequence. Therest of the unit consists of grey to green sandstones andsiltstones, and subordinate shales. Some basaltic andacidic volcanic levels (up to 4 m thick) are also present.Bed thickness and grain size gradually increase upwardwithin the unit. North of the study area, the turbiditespass vertically and laterally into massive volcanic rocksintermediate to felsic in nature. The bulk of the volcanicrocks are of island-arc calc-alkaline nature (Dokuz 2000).Palaeocene sediments are absent in the area. The Eoceneoverlies the Late Cretaceous Yusufeli formation along anangular unconformity and commences with a basalconglomerate bounded by a red matrix. The nummulite-
bearing sandstones and limestones, which overlie theconglomerate, have yielded a Lutetian age for theinitiation of sediment deposition.
Sampling and Analytical Methods
The present study is based on modal mineralogical andgeochemical analyses of shales and sandstones from fourcomplete profiles (‹flhan, Kemerlida¤, Ormandibi andÇa¤layan) through the Lower Jurassic and UpperCretaceous sequence (Figure 3). A total of 36 medium- tofine-grained samples (11 from the profiles of the LowerJurassic Hamurkesen formation and 25 from the profilesof the Upper Cretaceous Yusufeli formation) were point-counted for detrital modal analysis with a semi-automaticSwift Point Counter (Table 1; geochemically non-analysedsamples are not shown). Depending on groundmasscontent, at least 350–500 grains per sample werecounted. To minimize the effects of grain-size variations,rock fragments were counted using the Gazzi-Dickinsonmethod (Dickinson & Suczek 1979). All componentsbelow 0.063 mm apparent diameter (silt and clayfraction) were counted as matrix, as were cement crystalsregardless of size. Each thin section was stained for K-feldspar. After a thin section was etched in concentratedhydrofluoric acid for 5 s, it was plunged into a solutionprepared by mixing 100 ml of water with 60 ml Na-hegzanitrocobaltat to obtain canary yellow for K-feldspar. According to the Gazzi-Dickinson method, wewere able to distinguish the following frameworkconstituents: quartz (Q), including monocrystalline andpolycrystalline varieties; feldspar (F), comprisingplagioclase (P) and K-feldspar (AF); lithic fragments (L),comprising basaltic-dacitic volcanic and sedimentaryclasts; matrix (M); cement (C); hornblende (hb); biotite(bi); and heavy minerals (HM). Heavy minerals includeopaque grains (magnetite, ilmenite) and chlorite. For thenomenclature of shale samples, the chemical classificationdiagram of Creaser et al. (1997) was used.
Bulk-rock chemical compositions of 36 samples weredetermined partly by inductively coupled plasma-massspectrometry (ICP-MC) at the commercial AcmeAnalytical Laboratories Ltd. in Vancouver, Canada. In theICP-MS analyses, a 0.2 g aliquot is weighed into agraphite crucible and mixed with 1.5 g of LiBO2 flux. Thecrucibles are placed in an oven and heated to 1050 ºC for15 minutes. The molten sample is dissolved in 5% HNO3
GEOCHEMISTRY OF JURASSIC–CRETACEOUS ROCKS, NE TURKEY
186
(ACS-grade nitric acid diluted in demineralized water).Calibration standards and reagent blanks are added to thesample sequence. Sample solutions are aspirated into an
ICP emission spectrograph (Jarrel Ash Atom Comb 975)for determining major oxides and certain trace elements(Ba, Nb, Ni, Sr, Sc, Y & Zr), while the sample solutions
A. DOKUZ & E. TANYOLU
187
0 m
500
0 m
500
964
335*
338339*
348*
349
343*
351
352
431433
451*
452
458 870
0 m
390
0 m
500
896
565566569
570
573
582
590
535536537
542
538*
543*548
pillow basalt
basic sil
shale
sandstone
limestone
basicagglomerate
red limestone
andesiticvolcanics
chert
metamorphicsU. PalaeozoicU. Palaeozoic
Berdiga F.(Kimm.-Barrem.)
Orm
andi
bi P
rofi
le (4
)K
emer
lidað
Pro
file
(3)
Çað
laya
n Pr
ofile
(2)
Ýþha
n Pr
ofile
(1)
Plie
nsba
chia
n - p
re K
imm
erid
gian
Yus
ufel
i For
mat
ion
unconformity
Mid-Eocene
sedimentaryrocks
unconformity
Lithology
Sam
ple
Num
ber
Thi
ckne
ss
Age
Prof
ile (N
um.)
Form
atio
nY
usuf
eli f
orm
atio
nH
amur
kese
n fo
rmat
ion
Lithology
Sam
ple
Num
ber
Thi
ckne
ss
Age
Prof
ile (N
um.)
Form
atio
n
Apt
ian
- Cam
pani
an
Ham
urke
sen
form
atio
n
Plie
nsba
c. -
pre
Kim
m.
Berdiga F.(Kimm.-Barrem.)
425 Hedbergella trocoidae Gandolfi,Hedbergella delioensis Carsey,Hedbergella sp., Ticinella sp. andLagenidae (Aptian)
Globigerinelloides ferreolensisMolludae, Globigerinelloides sp.,Hedbergalla sp., Anomalinidae,Lagenidae
Rotalipora ticinensis Gaundolfi,Rotalipora reichelli Mornod,Stoimosphaera sphaerica Kaufmann,Globigerinidae, Globigerinelloides sp.,Hedbergella sp., Lithothamnium sp.,Ticinella sp.,
Marssonella oxcona Reus,Stomiosphaera Sphaerica Kaufman,Marginotruncana coronata Bolli,Marginotruncana pseudolinneianaPassagno, Pithonella ovalisKaufmann, Globotruncana linneianad� Orbigny
Globotruncana bulloides Vogler,Globotruncana linneiana d�Orbigny,Rosita fornicata Plummer,Pithonella ovalis Kaufmann,Stomiosphaera sphaerica Kaufmann,Pseudoguembelina sp.,Globotruncana sp.,Heterohelicidae, Rugoglobigerina
454*
436
456
453*
Pseudocyclammina cf.massiliensis Myne,Dicarinella concovata Brotzeu,Marginotruncana coronata Bolli,Marginotruncana sp., Dcarinella sp.,Miliolidae
Lithothamnium sp.,Lithophillum sp.,Anomalinidae,Globigerinidae,Lagenidae
574
580*Apt
.- A
lb.
U. A
lb.-S
en.T
ur.-C
am.U
. San
.-Cam
.
430
Fossil Content
Fossil Content
576
581*
Figure 3. Simplified stratigraphic logs of the Lower Jurassic and Upper Cretaceous units with sample localities of modal-mineralogically andgeochemically analysed shales (marked with asteriks) and sandstones. Localities of some sandstones, which were merely analysedin terms of modal mineralogical content, are not shown.
GEOCHEMISTRY OF JURASSIC–CRETACEOUS ROCKS, NE TURKEY
188
Tabl
e 1.
M
iner
alog
ical
and
che
mic
al c
ompo
sitio
ns o
f an
alys
ed s
ampl
es a
nd t
he P
AAS
(Tay
lor
& M
cLen
nan
1985
). M
ajor
ele
men
ts a
re r
epor
ted
in w
eigh
t pe
rcen
t an
d tr
ace
elem
ents
in p
pm.
Wei
ght
perc
ent
oxid
es a
re r
ecal
cula
ted
to 1
00%
on
a vo
latil
e-fr
ee b
asis
.
Low
er J
uras
sic
sam
ples
Upp
er C
reta
ceou
s s.
‹flha
n pr
ofile
Kem
erlid
a¤ p
rofil
eÇa
¤lay
an p
rofil
e
shal
essa
ndst
ones
sand
ston
essh
ales
shal
es
349
335
339
343
348
338
349
351
352
535
536
537
542
548
538
543
PAAS
451
453
454
Q10
.43.
02.
22.
010
.25.
41.
816
.62.
0Pl
2930
.633
.232
.263
.253
.452
.266
.852
.2A
F19
.22.
84.
42.
20.
00.
60.
60.
00.
0L
3529
.827
.434
10.2
7.6
18.6
7.2
8.8
HM
3.4
1.4
0.4
1.6
2.6
0.8
5.2
1.0
1.8
M2.
60.
00.
00.
013
.230
.216
.27.
40.
0C
0.0
32.4
32.4
26.2
0.0
0.0
4.4
0.0
35.2
hb,b
i0.
40.
00.
01.
80.
61.
61.
01.
00.
0
% SiO
280
.63
78.1
380
.44
67.2
575
.20
55.1
252
.43
56.0
863
.78
72.4
760
.08
68.1
460
.59
60.2
060
.96
62.4
48.6
362
.63
64.1
3
TiO
20.
390.
550.
210.
670.
470.
730.
930.
650.
900.
550.
630.
510.
610.
830.
620.
990.
520.
370.
32
Al2O
37.
159.
359.
0514
.92
12.0
318
.11
17.1
317
.24
15.6
413
.46
17.7
017
.31
17.5
718
.73
18.4
818
.78
18.3
017
.19
17.8
0
Fe2O
35.
403.
764.
336.
314.
326.
755.
365.
648.
855.
056.
583.
534.
787.
828.
287.
184.
814.
174.
49
MnO
0.
030.
060.
090.
080.
090.
320.
290.
300.
090.
040.
110.
020.
080.
110.
09-
0.24
0.08
0.06
MgO
1.50
1.40
1.22
3.06
1.34
1.50
1.55
1.92
3.43
2.04
2.26
1.29
0.82
3.19
2.47
2.19
2.11
1.99
2.37
CaO
0.63
1.37
1.35
3.97
2.80
13.4
018
.10
12.9
20.
921.
845.
185.
199.
060.
285.
121.
2919
.41
7.68
1.97
Na 2
O
0.46
0.38
2.93
3.02
1.86
2.58
2.60
3.22
5.95
3.88
6.08
3.63
4.74
6.78
2.97
1.19
4.70
4.74
8.29
K2O
3.
694.
960.
330.
671.
851.
451.
601.
910.
210.
611.
280.
271.
621.
940.
823.
681.
251.
090.
40
P 2O
50.
110.
040.
040.
060.
030.
040.
010.
120.
220.
060.
110.
110.
140.
120.
200.
160.
010.
050.
16
ppm
Rb
152
894.
016
2819
.920
27-
--
--
22-
160
12.8
17.7
-Sr
2336
110
940
465
238
937
236
436
640
729
314
424
513
219
820
011
1354
655
4Cs
3.3
3.5
0.3
0.7
-0.
60.
72.
3-
--
--
0.3
2.4
157.
04.
0-
Ba22
044
244
719
047
024
624
420
171
174
168
8821
248
925
065
024
0.1
275.
812
4Th
3.5
12.
84.
28.
422.
51.
61.
3-
--
--
4.7
9.4
14.6
0.8
2.6
-U
0.4
0.3
0.7
1.8
0.5
0.7
0.4
0.4
--
--
-1.
01.
53.
10.
61.
2-
Zr46
.533
.795
.968
.775
57.2
41.6
48.1
210
110
4580
5596
.213
1.2
210
3155
58H
f1.
21
2.8
2.3
3.82
1.8
1.3
1.6
--
--
-2.
94.
05
0.8
2.1
-Y
15.7
13.5
32.1
16.3
1422
.631
.332
.512
116.
35.
54.
716
.215
.727
10.3
7.3
10N
b3
1.8
3.9
5.3
3.4
4.6
2.8
2.3
1511
1010
107.
113
.71.
91.
41.
710
Cr4.
789
27.3
727
.37
6.84
227
.37
20.5
47.8
941
.05
20.5
320
.53
34.2
16.
842
54.7
434
.21
27.3
711
027
.37
34.2
161
.58
Co6.
55.
825
2.5
-13
.510
.412
.9-
--
--
17.6
6.8
2311
.111
.5-
Ni
11.1
6.7
1.9
7.1
2110
6.4
8.2
--
--
-29
7.4
5512
.420
.6-
Cu65
.671
.237
.933
.8-
25.8
28.7
21.5
--
--
--
33.2
5011
.875
.4-
Sc13
2411
1911
1311
1414
1012
1010
1911
1610
1010
V12
318
412
133
-12
914
514
4-
--
--
201
116
150
9885
-La
9.1
6.1
1015
7.6
17.2
26.4
20.1
29.1
--
20.9
-13
.417
.638
.28.
48.
7-
Ce
1710
.521
.933
.512
.831
.647
.435
.751
.3-
-32
.7-
23.4
31.7
79.6
14.1
16.1
-Pr
2.17
1.36
2.59
3.67
1.55
3.76
5.19
4.34
6.69
--
3.85
-2.
973.
648.
831.
782.
01-
Nd
10.2
6.4
11.4
136.
215
.221
.117
.220
.84
--
11.9
8-
1313
.533
.97.
77.
2-
Sm2.
41.
73.
13.
21.
693.
204.
14.
13.
66-
-2.
13-
2.7
2.6
5.55
1.5
1.4
-Eu
0.6
0.57
0.98
0.71
0.81
0.86
0.78
1.01
0.97
--
0.97
-0.
760.
721.
080.
610.
51-
Gd
2.33
1.97
3.63
2.9
1.75
3.52
4.16
4.36
3.16
--
1.82
-3.
082.
384.
661.
681.
33-
Tb0.
460.
350.
740.
50.
50.
560.
740.
720.
66-
-0.
42-
0.48
0.41
0.77
0.29
0.2
-D
y2.
822.
564.
572.
892.
443.
623.
844.
682.
61-
-1.
54-
32.
594.
681.
621.
08-
Ho
0.55
0.54
0.99
0.59
0.54
0.82
1.03
1.1
0.46
--
0.3
-0.
610.
530.
990.
370.
26-
Er1.
441.
433.
221.
772.
022.
302.
653.
131.
57-
-1.
0-
1.74
1.45
2.85
0.96
0.67
-Tm
0.27
0.2
0.47
0.26
0.33
0.34
0.42
0.49
0.23
--
0.14
-0.
30.
240.
400.
150.
1-
Yb1.
61.
382.
851.
751.
762.
342.
673.
361.
26-
-0.
81-
1.93
1.65
2.82
0.94
0.75
-Lu
0.26
0.19
0.51
0.26
0.29
0.32
0.47
0.51
0.21
--
0.13
-0.
330.
240.
430.
160.
12-
A. DOKUZ & E. TANYOLU
189
Tabl
e 1.
con
tinu
ed
Upp
er C
reta
ceou
s sa
mpl
es
Ça¤l
ayan
pro
file
Orm
andi
bi p
rofil
e
sand
ston
essa
ndst
ones
shal
es
430
431
433
436
452
456
458
565
566
569
570
573
574
576
582
590
580
581
Q3.
42.
83.
21.
03.
42.
63.
80.
00.
00.
00.
40.
410
19.2
5.4
9.0
Pl33
.827
.820
29.4
8.8
3428
.424
.232
30.2
17.8
145.
220
.828
5.6
AF
0.0
0.0
0.0
0.4
0.9
0.0
0.0
0.0
0.0
0.0
0.0
0.0
4.4
1.6
0.2
0.0
L32
.439
.236
.623
.282
.931
.841
.556
.465
.263
.266
54.6
60.6
35.2
47.4
79.8
HM
6.6
3.4
5.8
13.6
0.7
1.2
4.4
5.8
1.2
0.6
9.6
1.4
0.2
0.0
4.6
5.6
M0.
00.
00.
00.
00.
00.
020
.213
.61.
25.
45.
80.
00.
00.
00.
00.
0C
19.2
25.2
32.6
323.
412
.80.
00.
00.
00.
00.
029
.219
.623
10.6
0.0
hb,
bi4.
61.
61.
80.
40.
917
.61.
70.
00.
40.
60.
40.
40.
00.
23.
80.
0
% SiO
252
.63
51.3
647
.98
50.9
857
.31
54.7
268
.33
63.1
663
.29
65.9
060
.34
55.0
267
.40
71.8
852
.28
76.2
962
.75
68.9
1
TiO
20.
620.
670.
770.
870.
610.
450.
650.
590.
750.
500.
740.
830.
560.
200.
760.
450.
570.
06
Al2O
316
.89
16.2
015
.08
16.9
816
.64
17.3
915
.93
19.1
517
.77
17.6
619
.19
14.7
313
.63
8.21
15.6
315
.55
14.7
311
.95
Fe2O
37.
017.
098.
665.
497.
845.
324.
015.
305.
464.
125.
753.
153.
621.
438.
863.
986.
394.
19
MnO
0.16
0.17
0.18
0.09
0.13
0.11
0.04
0.05
0.05
0.04
0.06
0.14
0.14
0.07
0.16
0.03
0.16
0.10
MgO
1.88
1.54
2.19
1.03
3.85
2.96
0.95
2.75
2.20
0.96
1.21
4.45
2.79
0.71
3.87
0.88
1.96
2.13
CaO
15.3
117
.58
19.5
420
.07
8.50
13.9
23.
393.
111.
391.
605.
8217
.15
6.15
13.8
813
.75
1.63
8.99
8.24
Na 2
O
3.90
3.76
3.86
2.97
3.17
4.15
4.44
4.36
7.91
8.68
5.08
3.76
4.98
1.48
3.73
0.26
3.18
2.84
K2O
1.
541.
511.
621.
461.
860.
842.
121.
290.
990.
421.
680.
630.
522.
130.
870.
881.
111.
17
P 2O
50.
060.
120.
120.
060.
090.
120.
150.
230.
190.
120.
120.
150.
210.
010.
080.
050.
160.
42
ppm
Rb
--
--
--
-21
.9-
6.3
-13
.311
36.5
-27
.921
.918
.2Sr
634
550
692
826
268
935
968
629.
362
386
759
931
122
115
730
934
323
423
5Cs
1.5
0.3
1.4
0.4
0.7
20.
60.
4Ba
431
403
331
1658
272
191
458
473
234
507
262
130
548
270
146
230
260
404
Th-
--
--
--
4.7
-6.
7-
3.8
7.4
2.2
-6.
42.
14.
5U
--
--
--
-1.
1-
1.2
-3
1.1
0.8
-1.
50.
60.
9Zr
4745
5154
5037
9183
8310
372
7098
3830
138
5873
Hf
--
--
--
-2.
4-
32.
22.
81.
23.
41.
52.
8Y
1210
1511
1010
1013
.411
14.6
1013
.615
.96.
311
12.4
18.7
29.4
Nb
3.8
2.7
5.6
4.3
3.5
5.1
2.2
6.2
1211
.85.
510
.910
.93.
33.
18.
33.
43.
9Cr
27.3
720
.53
20.5
347
.89
6.84
261
.58
20.5
354
.74
41.0
520
.53
54.7
475
.26
41.0
541
.05
102.
641
.05
47.8
934
.21
Co-
--
--
--
12-
8-
17.6
11.2
2.2
-4.
515
.17.
3N
i-
--
--
--
16.7
-9.
7-
28.3
26.7
4.4
-8.
36.
24.
4Cu
--
--
--
-19
.8-
56.1
-75
.617
.34
-8.
522
.125
.6Sc
67
1010
1314
1010
1010
1011
1010
2010
1611
V-
--
--
--
70-
78-
222
7626
-61
163
82La
12.7
--
--
11.7
-22
.623
.927
.421
.323
.935
.48.
6-
29.9
9.9
18.8
Ce
21.3
--
--
24.4
-40
.140
.249
.236
.631
.555
.514
.1-
45.7
17.5
31.7
Pr2.
83-
--
-2.
93-
4.48
4.93
5.22
4.4
4.67
6.47
1.74
-5.
072.
193.
87N
d9.
72-
--
-10
.72
-15
.515
.79
1714
.15
17.6
24.5
6.1
-17
.87.
715
.6Sm
2.16
--
--
2.52
-3.
12.
973.
22.
692.
83.
61.
2-
3.1
2.2
3.4
Eu0.
76-
--
-0.
78-
0.82
0.79
0.83
0.96
0.95
1.01
0.41
-0.
720.
730.
8G
d2.
00-
--
-2.
04-
2.6
2.57
2.64
2.23
2.55
3.17
1.22
-1.
92.
453.
97Tb
0.48
--
--
0.47
-0.
370.
550.
430.
480.
390.
470.
16-
0.31
0.4
0.64
Dy
2.12
--
--
1.89
-2.
172.
212.
281.
91.
952.
420.
91-
1.73
2.81
3.88
Ho
0.38
--
--
0.34
-0.
450.
40.
490.
330.
460.
510.
21-
0.39
0.61
0.95
Er1.
31-
--
-1.
12-
1.22
1.36
1.33
1.13
1.24
1.37
0.66
-1.
171.
732.
53Tm
0.19
--
--
0.16
-0.
180.
20.
220.
160.
180.
20.
08-
0.2
0.27
0.42
Yb1.
09-
--
-0.
9-
1.22
1.15
1.28
0.87
1.33
1.42
0.62
-1.
381.
852.
56Lu
0.16
--
--
0.14
-0.
190.
20.
220.
140.
180.
220.
07-
0.21
0.29
0.44
are aspirated into an ICP mass spectrometer (Perkins-Elmer Elan 6000) for determination of the traceelements, including rare-earth elements (La, Ce, Pr, Nd,Sm, Eu, Gd, Tb, Dy, Ho, Er, Tm, Yb, Lu).
Petrography
The results of the point counts are shown in Table 1. Onthe basis of the roundness of clasts and quartz contents,the sandstones are generally immature and moderatelysorted. The matrix constitutes between 2 and 38 modal% and mainly comprises quartz, plagioclase, chlorite, andclay minerals in the sandstones. Fe-Ti minerals are alsosignificant components as matrix in most of the UpperCretaceous sandstone. Only one thin section (548) fromthe Lower Jurassic sandstones has a limonite and/orhematite matrix, and that of 35.2 modal %. Carbonategrains are absent, and they are only a significantcomponent in most of the Lower Jurassic samples ascement. However, mudstone fragments with subordinatefossil shells are also present up to 23 modal % of someof the Upper Cretaceous sandstones (geochemically non-analysed due to their unsuitable modal compositions).Ternary diagrams of detrital modal analyses are usuallyapplied in the classification of sandstones, using the threedetrital framework components quartz, feldspar and rockfragments (e.g., McBride 1963; Dott 1964; Dickinson &Suczek 1979; Pettijohn et al. 1987). In the presentstudy, all grains having a size between 30–60 µm (silt)were counted as matrix. Depending on the matrixcontent, two ternary graphs are in use: one for matrix-rich wackes (>15 % matrix), and the other for matrix-poor arenites ( <15 % matrix; Dott 1964; Pettijohn et al.1987). In those graphs, chert grains are assigned to thelithic corner. Because in this study the Gazzi-Dickinsonmethod was applied, the QFL plot was selected forsandstone classification.
Seven samples from the Lower Jurassic and thirteensamples from the Upper Cretaceous sandstones can bedisplayed in the ternary diagram for matrix-rich wackesdue to their matrix contents of more than 15 modal %.Feldspar and lithic fragments dominate, whereas quartzfragments are fairly scarce. Thus, samples from theLower Jurassic sandstones are defined as feldspaticwacke, while samples from the Upper Cretaceoussandstones are defined as lithic wacke (Figure 4a). Theremaining samples must be plotted in the ternary
diagram for arenites because of their matrix contents lessthan 15 modal %. Nearly all of the samples from theUpper Cretaceous sandstones plot as lithic arenite,whereas samples from the Lower Jurassic sandstonesplot as arkosic arenite (Figure 4b), except for one thatfalls into the field of lithic arenite.
In the QFL ternary provenance diagram (e.g.,Dickinson & Suczek 1979; Dickinson et al. 1983), fourfeldspar-rich sandstones of the Lower Jurassic profile fallinto the continental-block field (basement uplift as asubfield for feldspar-rich compositions), while others fallinto the transitional-arc field (Figure 4c). The greatmajority of the Upper Cretaceous sandstones plot in theundissected-arc field except for five that plot in thetransitional-arc field, exhibiting a distinct discriminationfrom the Lower Jurassic samples.
Lithic fragments are the first-order component interms of abundance of the Upper Cretaceous and thesecond-order component after feldspar grains in theLower Jurassic sandstones. Lithic fragments constitute anaverage of about 30% of the Lower Jurassic sandstonesand 47% of the Upper Cretaceous sandstones. In spite ofthe fact that they are dominated mainly by basaltic andandesitic types in nearly all of the sandstone samples,variable amounts of dacitic and sedimentary fragmentsare also present. The majority of the basaltic fragmentshave a brownish or blackish appearence due todecomposition. In large dacitic fragments, anhedralquartz, subhedral to euhedral plagioclase with oscillatoryzoning, and amphibole grains within the microgranularmatrix are present. Sedimentary fragments are absent inmost of the Lower Jurassic sandstones and are moreabundant in some of the Upper Cretaceous sandstones,mainly as carbonate fragments and fossil shells.
Feldspar grains are dominated by plagioclase, mostlyalbitized, and typically show normal to oscillatory zoning.Some feldspar grains have been partially sericitized, andothers partially or wholly replaced by calcite. Unalteredsmall plagioclase grains may be confused with quartzgrains. All thin sections were stained for K-feldspar, andthe K-feldspar present is entirely orthoclase. Although K-feldspar contents reach significant levels (up to 19%) insample 338 from the base of the ‹flhan profile, the LowerJurassic sandstones generally contain K-feldspar up to4.4%. On the other hand, K-feldspar generally is absent,although it reaches 4% in some thin sections of the UpperCretaceous sandstones.
GEOCHEMISTRY OF JURASSIC–CRETACEOUS ROCKS, NE TURKEY
190
Quartz is generally quite scarce in both lithologies;only one sample from each profiles has a significantquartz content (up to 19%). Quartz grains typically havea moderate degree of sphericity and roundness, and bothmonocrystalline and polycrystalline varieties occur thoughmost grains are monocrystalline. Some monocrystallinegrains are strained and were most probably derived fromplutonic and volcanic rocks. Polycrystalline grains aregenerally made up of inequant individual crystals. Largepolycrystalline grains consist of smaller, elongatedindividual crystals with smooth or sutured boundaries;such grains are most probably of metamorphic origin.
The bulk heavy-mineral assemblage is in appreciableand is higher in the Upper Cretaceous samples (Table 1);it consists of opaque oxides (i.e., magnetite, ilmenite) and
chlorite in decreasing order of abundance. Chlorite occursin appreciable amounts in some samples. High amphibolecontents, up to 18%, were found in the UpperCretaceous sandstones, an important fact to beconsidered when interpreting detrital modes andgeochemistry in so far as amphibole is sensitive to MREE,HREE and Y. Biotite and muscovite are quite scarce (upto 1.8%) in some of the Lower Jurassic samples.
Geochemistry
Analysed shales and sandstones have been grouped inTable 1 on the bases of age and stratigraphic sections.Likewise, average data for post-Archean shales (PAAS)(Taylor & McLennan 1985), which are considered
A. DOKUZ & E. TANYOLU
191
Q
F L
Q
F L
Q
F L
>15 % matrix >15 % matrix
quartzwacke
quartzarenite
Upper Cretaceous samples
Lower Jurassic samples
arkose
dissectedarc
undissectedarc
arkosicarenite
lithic arenitelithicwacke
feldspathicwacke
basementuplift
recycledorogen
cratoninterior
50
50transitional arc
(c)
(a) (b)
Figure 4. QFL diagrams of Lower Jurassic and Upper Cretaceous sandstones. Q– total quartz, F– feldspar, L–lithic fragments; (a) wackes, (b) arenites, Koç QFL diagram, with provenance fields, of Dickinson &Suczek (1979).
representative of the upper continental crust, areincluded for reference. In addition, correlationcoefficients obtained from matrix correlation arereported in Table 2.
Major Elements
The Lower Jurassic and Upper Cretaceous clastic rocksshow significant lithological variability, which is reflectedin their variation in chemical composition (Table 1); theyhave commonly overlapping abundances with regard toSiO2 contents, from 52.4 to 80.6% and 47.9 to 76.2%,respectively, and fall within the greywacke, litharenite,and arkose fields in the chemical classification diagram forthe sedimentary rocks (Pettijohn et al. 1987; Creaser etal. 1997) (Figure 5a). These rocks have broadly similarSiO2/Al2O3 ratios, but variable K2O/Na2O ratios (Table 3).However, the low K2O/Na2O ratios (<1) in most samplesare remarkable with an average ratio of about 1:6; thissituation is attributed to a paucity of K-feldspar andbiotite. The two points – with an average ratio of 10:1 –represent the shales, reflecting the higher content in K-bearing clay phases and other clay-sized phases comparedto sandstones of the same age. Unexpectedly, the shaleshave higher SiO2 and lower Al2O3, CaO and Na2O than theassociated sandstones (Table 1), except for those in theKemerlida¤ profile which have average values, suggestingthe enrichment of fine quartz particles as compared toplagioclase. This behaviour is also supported by the highnegative correlation coefficient (r = -0.91) between SiO2
and Al2O3 (Table 2). Generally, the Lower Jurassic rockstend to have lower (CaO+MgO)/SiO2+K2O+Na2O) ratiosthan the Upper Cretaceous rocks (Figure 5b). The rocksshow variable degrees of negative correlation of SiO2
versus TiO2, MnO, MgO and CaO. There is a positivecorrelation between Al2O3 and TiO2, indicating chemicalweathering in the source area and resulting in a relativeconcentration of these residual elements. Although therocks do not exhibit any obvious discrimination in termsof major elements – compared to PAAS (Taylor &McLennan 1985) (Figure 6a) – the Upper Cretaceoussandstones have slightly more depleted patterns for Ti,Fe, Ca and K than their Lower Jurassic equivalents. LowAl values (0.6–0.8 x PAAS) suggest that the source areawas not subjected to intense weathering. Na and Caenrichments accord with their high plagioclase andhornblende contents. However, excessive enrichments inCa may also be due to the occurrence of some diagenetic
calcite cements and increase in the abundance oflimestone clasts.
Large-Ion Lithophile Elements (LILEs: Rb, Cs, Ba, Sr)
In comparison with PAAS (Figure 6b), our Lower Jurassicsamples generally tend to display more depleted LILE andHFSE patterns. The mean LILE and HFSE concentrationsof the fine-grained and coarse-grained fractions from theeach units are slightly to moderately depleted, whereasthe mean Sr and Nb concentrations are moderatelyenriched, particularly in the sandstones. Theseenrichments are in accordance with modally highplagioclase, heavy-mineral (i.e., magnetite, ilmenite) andopaque- cement (i.e., limonite, hematite) contents ofcertain sandstones. While obvious depletions in the LILEand HFSE contents of the Upper Cretaceous shalesrelative to associated sandstones are present, their LowerJurassic counterparts show similarly depleted LILE andHFSE contents. In the Lower Jurassic rocks, Rb, Cs andBa have moderate to high negative correlationcoefficients with Al2O3 (-0.64, -0.45 & -0.36,respectively) (Table 2). These correlations suggest thattheir distributions are not controlled by kaolinitic andmontmorillonitic phyllosilicate phases. In contrast, thehigher positive linear correlations with K2O (0.81, 0.75and 0.44) suggest a strong link to illite and K-feldspar.No correlations have been found between Sr and otherelements/oxides, probably due to its occurrence inmultiple mineral phases. On the other hand, in the UpperCretaceous rocks, the LILE have no significantcorrelations with Al2O3, except for Sr and Cs; these haverelatively high positive correlation coefficients (0.63 and0.48), probably indicating that their distributions aremainly controlled by plagioclase.
High-Field-Strength Elements (HFSEs: Zr, Hf, Y, Nb,Ta, Th, U)
High-field-strength element (HFSE) concentrations in theLower Jurassic and Upper Cretaceous rocks show slightlyto moderate positive correlations with TiO2 and P2O5 andnegative correlations with Cr, Ni and Sc (Table 2), anexpected trend due to their incompatible behaviours. Theaverage Th/U ratio of 5.7 in the Lower Jurassic rocks ishigher than in the Upper Cretaceous rocks (3.6), in goodagreement with the average ratio of 3.8 in the uppercontinental crust (Taylor & McLennan 1985). The Th/Sc
GEOCHEMISTRY OF JURASSIC–CRETACEOUS ROCKS, NE TURKEY
192
A. DOKUZ & E. TANYOLU
193
Tabl
e 2.
Corr
elat
ion
coef
fcie
nts
(r)
from
the
cor
rela
tion
mat
rix
obta
ined
usi
ng g
eoch
emic
al d
ata
from
the
sha
les
and
sand
ston
es (
n= 3
5).
CIA
(che
mic
al i
ndex
of
alte
ratio
n) =
[Al
2O3
/(A
l 2O
3+Ca
O+
Na 2
O+
K2O
)]10
0.
Low
er J
uras
sic
Upp
er C
reta
ceou
s
rA
l 2O
3K
2O
TiO
2Zr
ThY
P 2O
5CI
Ar
Al 2
O3
K2O
Ti
O2
ZrTh
YP 2
O5
CIA
SiO
2-0
.91
0.29
-0.7
90.
020.
09-0
.22
-0.2
2-0
.14
SiO
2-0
.38
-0.1
0-0
.59
0.70
0.56
0.18
0.29
0.61
Al 2
O3
1-0
.47
0.69
0.14
0.17
0.04
0.35
0.18
Al 2
O3
1-0
.22
0.46
0.07
0.06
-0.2
3-0
.12
0.43
Rb
-0.6
40.
81-0
.31
-0.5
2-0
.12
-0.4
70.
330.
01R
b-0
.55
0.81
-0.4
3-0
.13
-0.3
0-0
.27
-0.2
5-0
.02
Cs-0
.45
0.75
-0.2
9-0
.31
-0.0
1-0
.41
0.36
-0.0
6Cs
0.49
0.15
0.11
-0.4
3-0
.58
-0.4
5-0
.52
-0.1
1
Th0.
17-0
.36
-0.1
20.
761.
00-0
.53
0.47
0.66
Th0.
06-0
.65
0.10
0.90
1.00
0.29
0.37
0.61
Ba
-0.3
60.
44-0
.29
-0.2
40.
160.
25-0
.45
-0.1
0B
a0.
060.
170.
280.
060.
630.
06-0
.02
-0.1
6
U0.
40-0
.60
0.16
0.65
0.51
-0.3
10.
420.
58U
0.06
-0.4
60.
560.
290.
26-0
.09
0.02
0.02
Sr0.
10-0
.04
0.28
-0.0
10.
13-0
.03
-0.3
6-0
.10
Sr0.
630.
070.
26-0
.09
-0.1
9-0
.30
-0.2
50.
02
Sc-0
.20
0.58
0.20
-0.1
5-0
.36
-0.0
6-0
.10
-0.2
3Sc
-0.0
9-0
.21
0.08
-0.2
5-0
.32
0.12
0.04
-0.0
4
V0.
370.
550.
71-0
.33
-0.1
0-0
.49
0.13
-0.2
0V
0.18
-0.4
30.
67-0
.14
-0.2
20.
210.
13-0
.24
Ni
0.26
0.12
0.29
0.19
0.38
-0.4
70.
170.
32N
i0.
31-0
.52
0.67
0.08
0.23
-0.2
4-0
.01
-0.0
5
Co0.
02-0
.38
-0.2
70.
29-0
.18
0.60
-0.1
70.
06Co
0.46
-0.4
50.
75-0
.20
-0.1
90.
120.
17-0
.14
Cr0.
390.
080.
32-0
.21
-0.1
80.
31-0
.01
0.52
Cr0.
02-0
.44
0.17
0.08
-0.0
5-0
.05
0.05
-0.0
3
Cu-0
.83
0.82
-0.4
7-0
.31
-0.6
3-0
.63
-0.0
40.
30Cu
0.31
-0.4
90.
34-0
.03
0.02
-0.0
80.
02-0
.08
Hf
0.26
-0.5
9-0
.17
0.89
0.90
-0.2
70.
360.
69H
f0.
12-0
.63
-0.0
10.
920.
910.
390.
440.
73
Yb0.
09-0
.05
0.00
-0.3
3-0
.46
0.96
-0.3
4-0
.61
Yb-0
.26
-0.3
3-0
.28
0.34
0.28
0.98
0.82
0.15
ΣLR
EE
0.40
-0.4
80.
600.
46-0
.22
0.35
0.19
-0.5
8ΣL
RE
E0.
01-0
.48
0.14
0.68
0.96
0.34
0.35
0.46
ΣMR
EE
0.25
-0.2
70.
27-0
.05
-0.5
40.
90-0
.12
-0.5
7ΣM
RE
E0.
02-0
.49
0.03
0.48
0.59
0.86
0.89
0.28
ΣHR
EE
0.03
-0.1
1-0
.05
-0.2
8-0
.44
0.96
-0.3
8-0
.56
ΣHR
EE
-0.2
1-0
.30
-0.2
60.
290.
280.
990.
850.
13
(La/
Yb) C
0.42
-0.5
30.
340.
570.
25-0
.46
0.53
0.28
(La/
Yb) C
0.24
-0.2
90.
470.
660.
82-0
.28
-0.0
40.
51
(Gd/
Yb) C
0.33
-0.4
20.
440.
60-0
.32
-0.4
40.
540.
29(G
d/Yb
) C0.
340.
040.
41-0
.10
0.31
-0.4
0-0
.01
-0.0
6
La/S
c0.
60-0
.60
0.50
0.36
-0.0
50.
180.
30-0
.27
La/S
c0.
13-0
.47
0.34
0.82
0.95
0.18
0.26
0.55
CIA
(ch
emic
al in
dex
of a
lter
atio
n) =
[A
ll 2O
l 3/(
All 2
Ol 3
+Ca
O+
Nal
2O+
Kl 2
O)]
100
GEOCHEMISTRY OF JURASSIC–CRETACEOUS ROCKS, NE TURKEY
194
Tabl
e 3.
Elem
enta
l rat
ios
in t
he a
naly
sed
grou
ps o
f sh
ales
, sa
ndst
ones
and
the
PAA
S. O
nly
the
sam
ples
of
Tabl
e 1,
whi
ch h
ave
full
trac
e-el
emen
t co
vera
ge,
are
show
n in
thi
s ta
ble.
Low
er J
uras
sic
sam
ples
Upp
er C
reta
ceou
s sa
mpl
es
shal
essa
ndst
ones
shal
essa
ndst
ones
335
339
343
348
538
543
338
349
351
352
PAAS
451
453
580
581
565
569
573
574
576
590
SiO
2/Al
2O3
11.3
8.4
8.9
4.5
3.2
3.3
6.3
3.0
3.1
3.3
3.3
2.7
3.6
4.3
5.8
3.3
3.7
3.7
4.9
8.8
4.9
K2O
/Na 2
O
8.0
13.0
0.1
0.2
0.3
0.3
1.0
0.6
0.6
0.6
3.1
0.3
0.2
0.3
0.4
0.3
0.0
0.2
0.1
1.4
3.3
Zr/H
f38
.833
.734
.329
.933
.232
.819
.631
.832
.030
.142
39.3
26.0
38.8
26.0
34.5
34.4
31.8
34.9
31.3
40.7
Zr/Y
b29
.124
.433
.639
.349
.879
.542
.624
.415
.614
.374
.533
.472
.831
.528
.468
.080
.752
.668
.760
.510
0.3
Zr/T
h13
.333
.734
.316
.420
.514
.08.
922
.926
.037
.014
.339
.321
.027
.716
.217
.615
.418
.413
.217
.021
.6
Th/U
8.8
3.3
4.0
2.3
4.7
6.3
16.8
3.6
4.0
3.3
4.7
1.3
2.2
3.5
5.0
4.3
5.6
1.3
6.7
2.8
4.3
Th/S
c0.
30.
00.
30.
20.
20.
90.
80.
20.
10.
10.
90.
10.
30.
10.
40.
50.
70.
30.
70.
20.
6
Co/T
h1.
95.
88.
90.
63.
70.
70.
05.
46.
59.
91.
613
.94.
47.
21.
62.
61.
24.
61.
51.
00.
7
Cr/T
h1.
427
.49.
81.
67.
32.
93.
38.
229
.931
.67.
534
.213
.222
.87.
611
.63.
119
.85.
518
.76.
4
Cr/N
i0.
44.
114
.41.
01.
23.
71.
32.
17.
55.
02.
02.
21.
77.
77.
83.
32.
12.
71.
59.
34.
9
Cr/V
0.0
0.1
2.3
0.1
0.2
0.2
-0.
20.
30.
30.
70.
30.
40.
30.
40.
80.
30.
30.
51.
60.
7
V/N
i11
.127
.56.
318
.76.
915
.70.
012
.922
.717
.62.
77.
94.
126
.318
.64.
28.
07.
82.
85.
97.
3
La/S
c0.
70.
30.
90.
80.
71.
60.
71.
32.
41.
42.
40.
80.
90.
61.
72.
32.
72.
23.
50.
93.
0
La/T
h2.
66.
13.
63.
62.
91.
90.
96.
916
.515
.52.
610
.53.
34.
74.
24.
84.
16.
34.
83.
94.
7
ΣREE
51.2
35.3
67.0
80.0
67.7
79.3
40.3
85.6
121.
010
0.8
185
40.3
40.4
50.6
89.6
95.0
111.
789
.713
6.3
36.1
109.
6
ΣLR
EE38
.524
.445
.965
.252
.866
.428
.267
.810
0.1
77.3
166
32.0
34.0
37.3
70.0
82.7
98.8
77.7
121.
930
.598
.5
ΣMR
EE9.
27.
714
.010
.810
.69.
27.
712
.614
.716
.017
.66.
14.
89.
213
.69.
59.
99.
111
.24.
18.
2
ΣHR
EE3.
63.
27.
14.
04.
33.
64.
45.
36.
27.
511
.32.
21.
64.
16.
02.
83.
12.
93.
21.
43.
0
(La/
Sm) C
2.4
2.3
2.0
3.0
3.1
4.3
2.8
3.4
4.1
3.1
4.3
3.5
3.9
2.8
3.5
4.6
5.4
5.4
6.2
4.5
6.1
(Gd/
Yb) C
1.2
1.2
1.0
1.3
1.3
1.2
0.8
1.2
1.3
1.1
1.4
1.4
1.4
1.1
1.3
1.7
1.7
1.6
1.8
1.6
1.1
(La/
Yb) C
3.8
3.0
2.4
5.8
4.7
7.2
2.9
5.0
6.7
4.0
9.2
6.0
7.8
3.6
5.0
12.5
14.5
12.1
16.8
9.4
14.6
Eu/E
u*0.
80.
90.
90.
70.
80.
91.
40.
80.
60.
70.
651.
21.
11.
00.
70.
90.
81.
10.
91.
00.
8
CIA
5048
5858
5960
5643
3641
7036
4845
4161
5433
4625
80
ratio, a good indicator for the bulk composition of theprovenance, has average values of 0.30 and 0.39 for theLower Jurassic and Upper Cretaceous rocks, respectively.For example, the Lower Jurassic and Upper Cretaceousrocks yield a broad relationship between Th/Sc and Zr/Scthat is consistent with provenance-dependentcompositional variation (trend 1 in Figure 7). An additionof zircon by sorting and recycling to samples would resultin an increase in Zr/Sc ratios, as exemplified by trend 2 inFigure 7. Furthermore, strong positive correlationbetween Zr and Hf, as attested by their high correlationcoefficients (0.89 and 0.92; Table 2), indicate that Thand Zr behave similarly during magmatic differentiation.Also, the choice of coarse-grained rocks in this studyincreases the possibility that heavy-mineralconcentrations might significantly affect the compositionof the rocks. Y has strong positive correlation with Yband heavy-rare-earth elements (HREEs), which seems toreflect that the HREEs are controlled mainly byhornblende abundance. The mean Y contents are lower inthe Upper Cretaceous rocks (12.5 ppm) than in theLower Jurassic rocks (17.6 ppm), having contents similarto the PAAS (27 ppm). In addition, the negativecorrelation between Y and SiO2 in the Lower Jurassicrocks seems to reflect its association with mineralspreferentially concentrated in quartz-poor rocks.
A. DOKUZ & E. TANYOLU
195
CaO+MgO
SiO /10
Na O+K O22
2
Granites
Ultramafites
Basalts
0.1 1.0 10.01
10
100
K O/N a O22
LA A
G
(a) (b)
SiO
2/Al
2O3
RbSr
CsBa
ThU
ZrHf
YNb
CrCo
NiCu
ScV
Shale
SandstoneSandstoneL ower Jurassic samples Upper Cretaceous samples
Shale
A l Fe Mg Ca K PSi Ti Na
Shale
Lower Jurassic samples
Upper C retaceous samples
Sandstone
Sandstone
Shale
10.0
1.0
0.1
10.0
1.0
0.1
Sam
ple/
PAAS
Sam
ple/
PAAS
(a)
(b)
Figure 5. (a) Chemical classification diagram showing the ranges in composition ofthe Lower Jurassic and Upper Cretaceous shales and sandstones (afterCreaser et al. 1997; Pettijohn et al. 1987). G– greywacke, LA– lithicarenite, A– arkose. (b) (CaO+MgO)–SiO2–(Na2O+K2O) diagram showingcomparison between major-element compositions of the analysed samplesand those of ultramafites, basalts and granites (after Taylor & McLennan1985).
Figure 6. (a) and (b) elemental variations for shales and sandstonesof the Lower Jurassic and Upper Cretaceous sedimentaryunits, respectively from the Yusufeli (Artvin) area. Dataare average values normalised to PAAS (Talor & McLennan1985).
Transition Trace Elements (TTEs: Cr, Co, Ni, Cu, Sc, V)
As observed for the great majority of the other trace andmajor elements, the transition trace-element (TTE)contents of the studied shales and sandstones are alsodepleted relative to those of the PAAS (Figure 6b), withthe exception of the Cu and Sc values for the LowerJurassic shales, which have patterns quite similar to thePAAS. These rocks have low Cr and Ni contents (andCr/Ni ratios in general), demonstrating the absence ofultramafic rocks in the source area. However, the higherconcentrations of Cr (42 ppm on average) in the UpperCretaceous sandstones with respect to those in the LowerJurassic rocks (26 ppm on average) indicate a higherproportion of mafic rocks in their source area (Taylor &McLennan 1985; Wronkijewicz & Condie 1987). TheCr/Th ratio, a good indicator for provenance (Condie &Wronkiewicz 1990), has average values of 12 and 14 inthe Lower Jurassic and Upper Cretaceous rocks,respectively (Table 3). In general, all of the TTEs in theUpper Cretaceous sandstones are positively correlatedwith TiO2 and Al2O3 (Table 2), indicating that they aremainly concentrated in the heavy minerals andphyllosilicates. In contrast, there are no systematiccorrelations between the TTEs and TiO2 and Al2O3 in theLower Jurassic shales and sandstones, suggesting thatadditional factors also control their distribution.
Rare-Earth Elements (REEs)
Chondrite- and PAAS-normalized REE patterns are givenin Figures 8 and 9. Although the total REE contents andthe magnitude of LREE-enrichment and Eu anomaliesvary, the patterns for the Lower Jurassic and UpperCretaceous rocks are broadly similar to one another, andto the PAAS (Figure 8). The Lower Jurassic shales seemto have lower REE contents in comparison to associatedsandstones. In general, they are characterized by low(La/Yb)C ratios (<7 except for sample 543 with a value of7.2), and slight to moderate negative Eu anomalies(Euc/Euc* = 0.57–0.94), except for a high positive Euanomaly (EuC/EuC* = 1.42 in sample; Table 3). Althoughall the mean REEs are depleted relative to the PAAS(Figure 9a), depletions in the LREE concentrations aregreater than the mean HREE concentrations, with theexception of the mean REE concentration of sandstonesfrom the Kemerlida¤ profile, which show roughly a flatREE pattern. If LREE, MREE and HREE are separatelyconsidered, all show strong positive correlation with Y (r= 0.92; Table 2), although the LREE and MREE show
GEOCHEMISTRY OF JURASSIC–CRETACEOUS ROCKS, NE TURKEY
196
1
2
1.0
0.1
1 10 100Zr/Sc
Th/S
c zircon addition
compo
sition
al
varia
tion
Figure 7. Th/Sc vs Zr/Sc variation plot for sedimentary rocks of theYusufeli area. The compositional variation trend line (1)was defined by McLennan et al. (1993) for sedimentaryrocks from active margins that are least affected bysedimentary sorting and recycling. Zircon accumulation bysorting and recycling would result in Zr enrichmentrelative to Th, defined by trend line (2) (McLennan et al.1993).
1
10
100
1
10
100
La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu
La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu
sandstoneshalePAAS
Upper Cretaceous samples
sandstoneshalePAAS
Lower Jurassic samples
Sam
ple/
Chon
drite
Sam
ple/
Chon
drite
(a)
(b)
Figure 8. Chondrite-normalised REE patterns of shales andsandstones of Lower Jurassic and Upper Cretaceous agefrom the Yusufeli area. The PAAS pattern is given forreference.
slightly to moderately positive correlations with Al2O3,TiO2 and Zr. All of these correlations seem to indicate that– in addition to hornblende – various mineral phases, suchas phyllosilicates, opaque phases and zircon, have playedsome role in controlling the REE contents.
The Upper Cetaceous rocks have highly enriched LREEpatterns (LaC/SmC = 2.83–6.18), and flat to moderatelyfractionated HREE patterns (GdC/YbC = 1.07–1.83)(Figure 8b; Table 3). The samples are characterized byslightly negative to moderately positive Eu anomalies(EuC/EuC* = 0. 7–1.17). The total REE contents are lowerin the Upper Cretaceous rocks (average ΣREE = 87 ppm)than in the Lower Jurassic rocks (average ΣREE = 93ppm). Compared to the PAAS (Figure 9b), the mean REEcontents exhibit flat but depleted patterns, except for themean REE contents of shales from the Ça¤layan profile,which are similar in appearance to the PAAS. Y and P2O5
show moderate to high positive correlations with MREEand HREE (Table 2). In addition, Zr and Th show strongpositive correlations with LREE and MREE. All of thesecorrelations indicate that REE distribution is likelycontrolled by the collective influence of hornblende,zircon, phosphates and opaque phases.
Discussion and Interpretation
Provenance
The major- and trace-element compositions of the twounits overlap, but nevertheless are distinct enough tosuggest that the bulk compositions of their sourceregions were slightly different. The relative abundance ofmodal plagioclase in the Lower Jurassic rocks is theprincipal difference between them and the UpperCretaceous rocks, represented by lithic arenite and lithicwacke petrofacies. The Lower Jurassic samples arescattered in the transitional-arc and basement-upliftfields, while the Upper Cretaceous samples mainly fall inthe undissected-arc field of Dickinson & Suczek (1979)(Figure 4c). In all samples from these units, quartz isdominantly monocrystalline, suggesting an igneoussource for the majority of the material. The paucity ofpolycrystalline quartz and sedimentary lithic fragments(fossil shells excluded), and the abundance of basaltic anddacitic fragments, suggests that these feldspathic andlithic arenites and wackes were derived from a volcanicsource where basic and acidic rocks were exposed andaltered through chemical and physical weathering duringa single cycle of sedimentation. Thus, the most likelysource terrain for these rocks would be volcanic rocksolder than Early Jurassic, possibly Late Palaeozoic, whichare no longer exposed/present (Y›lmaz 1972;Bergougnan 1987).
Plots of relatively insoluble elements in waters asratios, such as Th/Sc, La/Sc, Co/Th, Cr/Th, Cr/V, V/Ni andCr/Ni, are good indicators of provenance (McLennan et al.1983, 1990; Taylor & McLennan 1985; Wronckiewicz &Condie 1989, 1990; Cullers 1994), and help to elucidateconcentration or dilution effects of sorting duringsediment transport. The trace-element data from theseunits were plotted on a Sc-Th crossplot in two fields(Figure 10a). One field is characterized by a Th/Sc rationear 1, typical of continental crust that is enriched inincompatible elements. The other field is characterized byan enrichment in Sc (Th/Sc <1), usually found in moremafic sources that are enriched in compatible elements.The trace-element composition of each unit results fromvariable mixing of sediments deposited at each location.The contributions from each source are similar for thetwo units in so far as they are controlled by geographicproximity to each source. Figure 10b is an element-ratioplot of Th/Sc vs. Cr/Th. This plot varies in contrast withdifferent proportions from continental and mafic sources;also drawn in Figure 10b is a curve that best fits the data.
A. DOKUZ & E. TANYOLU
197
0.1
1.0
La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu
LowerJurassic samples
0.1
1.0
La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu
Upper Cretaceous samples
Sam
ple/
PAAS
Sam
ple/
PAAS
average compositions from ‹flhan profileshale sandstone
average compositions from Kemerlida¤ profileshale sandstone
shale sandstone
shale sandstone
average compositions from Ça¤layan profile
(a)
(b)
Figure 9. PAAS-normalized distribution of selected trace elementsin average Lower Jurassic and Upper Cretaceous shalesand sandstones. PAAS-normalising values are from Taylor& McLennan (1985).
This curve is consistent with a two-component mixingmodel between felsic and mafic end-members. It seemsapparent that at least two source areas were involved inproviding detritus to the basin during deposition of theunits. The dominant source has a mafic signature, withsome of the cratonic material less felsic than averagecontinental crust. The signatures of the felsic enrichmentover upper crust – represented by highly evolved alkalineigneous rocks – and mafic enrichment close to MORB arenot indicated. The relatively low Cr/Ni ratios from bothshales and sandstones of each unit provide no evidence ofsignificant amounts of ultramafic lithologies in the sourcearea. The relative abundance of Cr and Ni contents in theUpper Cretaceous rocks compared to the Lower Jurassic
rocks has been interpreted to reflect unroofing of uppercrust. In general, values of Zr/Hf, Th/Sc, La/Sc, Th/Sc,La/Yb ratios and CIA are lower and values of Co/Th,Cr/Th, V/Ni and Cr/Ni ratios are higher in the shales andsandstones of each unit compared to those of the PAAS(Table 3). All of these ratios indicate a more basic sourcethan the upper crust; therefore, they are close to those ofmid-continental crust. Table 4 gives elemental ratios ofdifferent igneous and sedimentary materials; theseindicate that the shales and sandstones of the Yusufeliarea have values similar to those the andesites of Condie(1993), and lie between sands derived from basic andacidic rocks (Cullers et al. 1988; Cullers 1994). As aresult, it is suggested that the contribution of
GEOCHEMISTRY OF JURASSIC–CRETACEOUS ROCKS, NE TURKEY
198
0 5 10 15 20 25 300
5
10
15
20
25
Sc (ppm)
UC
0 10 20 30 40 500.0
0.4
0.8
1.2
Cr/Th
granitesTh/Sc=10Cr/Th=0.02
MORBTh/Sc=0.02Cr/Th>500
Upper Cretaceous samples
Lower Jurassic samples
continental signature
mafic signature
Th/Sc = 0.6
Th/Sc =
1
(a) (b)
Th (
ppm
)
Th (
ppm
)
UC
Figure 10. (a) Th vs Sc for the Lower Jurassic and Upper Cretaceous samples. Th is an incompatible element that isenriched in felsic rocks, and Sc is a compatible element that is enriched in mafic rocks. Th/Sc ratios nearunity are typical of upper continental-crustal (UC) derivation, and Th/Sc ratios near 0.6 suggest a moremafic component. (b) Th/Sc vs Cr/Th for the samples. The samples lie on the lower side of a curveconsistent with mixing of a continental source enriched in incompatible elements (Th) and a more maficsource enriched in compatible elements (Cr and Sc). Values for average upper continental crust (UC),average granites, and average mid-ocean-ridge basalts are given for comparison.
Table 4. Elemental ratios typical of granites, andesites, ophiolites, and sands from basic and silicic rocks, the upper continental crust (UCC), lowercontinental crust (LCC) and oceanic crust (OC), and from terrigenous rocks of the Yusufeli area reported in this study.
Elemental sand from sand from Ratio granitesa andesitesa ophiolitesb basic rocksc felsic rocksc UCCd LCCd OCd shalese sandstonese
La/Sc 8.0 0.9 0.25 0.4–1.1 2.5–16 2.7 0.30 0.10 0.66 1.46Sc/Th 0.28 4.65 56 20–25 0.05–1.2 1.0 34 1.73 5.00 3.34Cr/Th 0.44 9.77 410 22–100 0.5–7.7 3.3 222 1227 10.04 18.24Co/Th 0.17 4.65 70 7.1–8.3 0.22–1.5 0.9 33 214 4.30 5.46Eu/Eu* 0.34 0.66 1.0 _ _ 0.61 1.07 1.02 0.83 0.88
a Condie (1993)b Spadea et al. (1980)c Cullers (1994); Cullers et al. (1988)d Taylor & McLennan (1985)e This study
intermediate and basic rocks rather than acidic rocks tothe the sedimentation seems to have predominated.
A more distinct discrimination between the sourceareas of the units may be inferred by using REEcharacteristics. The Lower Jurassic shales and associatedsandstones have moderately fractionated REE patterns,with average LaC/YbC values of 3.7 and 7.9, and negativeEu anomalies (EuC/EuC* = 0.8 and 0.9 on average),respectively. Sample 338 is an exception with itsmoderately positive Eu anomaly (EuC/EuC* = 1.4), whichis due to its high K-feldspar and plagioclase contents(Table 1). The average LREEs are moderatelyfractionated (LaC/SmC = 2.4 and 4.0), and the HREEpatterns are almost flat (GdC/YbC = 1.2 & 1.3,respectively) (Figure 8a). These characteristics indicatethat the original source area for the Lower Jurassic rockswas felsic, and the negative Eu anomalies are regarded asevidence of a differentiated source, possibly similar todacite or rhyolite (Taylor & McLennan 1985; Slack &Stevens 1994). On the other hand, depletions in K2O, Rband Cs, and enrichments in Na2O, CaO and Sr incomparison with PAAS (Figure 6a & b) suggest theoccurrence of plagioclase-rich rocks rather than K-feldspar-rich rocks in the source area. In addition,depleted REE patterns in comparison with PAAS (Figure9a) reflect a greater contribution from intermediaterather than felsic rocks. Therefore, they plot near shalesand sands originated from mid-continental crust.
The REE patterns of the Upper Cretaceous samplesare highly fractionated (LaC/YbC = 6.0–16.8), withslightly negative to positive Eu anomalies (EuC/EuC* =0.8–1.2), except for the two shale fractions which arecharacterised by moderate REE fractionation (LaC/YbC =3.6 and 5.0) and negative to neutral Eu anomalies(EuC/EuC* = 0.7–1.0) (Table 3; Figure 8b). The LREEsand MREEs are also moderately fractionated (LaC/SmC =2.8–6.9, GdC/YbC = 1.1–1.8) and HREE enrichment islower than that of the Lower Jurassic rocks. Thesefeatures suggest that the source area was geochemicallysimilar to that of the Lower Jurassic rocks in general, butcontributions from less or moderately evolved rocksseems to have been more widespread. The markedpositive CaO, Sr and Eu anomalies in the PAAS-normalised diagrams (Figure 6a & b) indicate apronounced contribution of moderately evolved and evenmafic materials to sedimentation. Compared to the PAAS,the relatively flat and depleted REE patterns (Figure 9b)
reflect the abundance of low to moderately evolved calc-alkaline materials in the modal mineralogical content.
Mineralogical Control and the Effect of SedimentaryProcesses on REE Distribution
In the Lower Jurassic rocks, depletions in LREE againstHREE (Figure 9) imply that the concentrations of theLREEs in these rocks were provided by an abundance oflow total-REE-bearing minerals, such as quartz andplagioclase. These features of the shales and associatedsandstones are consistent with their relatively high silicaand alumina contents, respectively (Table 1). Strongpositive correlations between Y and HREE (r = 0.96) andMREE (r = 0.90) (Table 2) indicate that control on theseelements was supplied mainly by hornblende. But LREE-enriched and HREE-flat patterns in the chondrite-normalised plot (Figure 8) rule out hornblende control onREE distribution as the sole factor. Strong positivecorrelation between Zr and Th (r = 0.76) implies that thecontents of both elements may have been derived fromzircon. However, the mean Zr contents of the shales aresimilar to those of associated sandstones, indicating thatthis mineral was not preferentially concentrated and/ordid not exert a strong influence on Zr and Thdistributions. Moreover, although zircon accumulationusually causes marked enrichment in LREEs, poornegative correlation of Zr with LREEs (r = -0.33)indicates that zircon has little or no influence on theabundance of these elements. Slight to moderate positivecorrelations between LREE and Al2O3 and TiO2 (r = 0.40and 0.60) suggest the compound influence of plagioclase,Ti-bearing minerals (such as biotite, ilmenite, titaniferousmagnetite) and clay minerals (in addition to hornblende)present in the sediments rather than a selective effect ofone of these minerals.
In the Upper Cretaceous rocks, the moderate positivecorrelations between LREEs and Zr (r = 0.68) and Th (r= 0.96) (Table 2) suggest that LREE enrichment wasproduced by the occurrence of zircon and opaqueminerals. In addition, positive correlations of Y with Yband HREEs (r = 0.98 & 0.99, respectively) and positivecorrelations of P2O5 with Yb, MREEs and HREEs (0.82,0.89 & 0.85, respectively) indicate that hornblende andphosphate played important roles in the distribution ofMREEs and HREEs; this observation is supported by themodal compositions of certain sandstones, which include
A. DOKUZ & E. TANYOLU
199
varying amounts of hornblende. Slight to moderatecorrelation of CIA with LREEs and MREEs (r = 0.46 &0.28, respectively) also indicate the influence of opaquecement present in some of the Upper Cratecoussandstones. These sandstones generally are characterisedby positive Eu anomalies, indicating the abundance ofbasaltic fragments which themselves contain abundantplagioclase laths.
Transportation, hydraulic sorting and deposition maysignificantly affect the chemical compositions ofterrigenous sediments, controlling the distribution ofmajor elements (P2O5 and TiO2) and some trace elements(e.g., REE, Th, U, Zr, Hf & Y); therefore, these elementsmay not be representative of provenance if heavy-mineralconcentrations affect the elemental distributions (cf.Reimer 1985; Cullers et al. 1987; McLennan 1989).
In the Lower Jurassic sediments, there is a strongtextural dependency upon chemical composition. SiO2,K2O, Zr and Rb contents are enriched, whereas Al2O3,MgO, CaO, Cr, Ni and particularly REE are depleted inshales relative to associated sandstones (Table 1). Theseelemental distributions suggest the existence of arelatively high energy level, which impeded theaccumulation of clay-sized sediments during deposition ofthe shales. Moreover, the shales yield a more depletedLREE pattern on average relative to associatedsandstones in the PAAS-normalized plot (Figure 9), thussupporting the argument for enrichment of silt-sizedquartz during sedimentation; this is because quartz haslow total REE contents even though it gives an LREE-enriched pattern against HREE on REE-normaliseddiagrams (Götze 1998). Shales have depleted CaO andNa2O values relative to associated sandstones (Figure 6a)thus demonstrating the influence of hydraulic sorting anddegree of recycling on the chemical composition ofterrigenous sediments. All of these differences betweenthe shales and sandstones indicate that sedimentarysorting was an important process that controlled thechemical composition of these clastic rocks.
The Upper Cretaceous sandstones yield highlyfractionated REE patterns with (La/Yb)C values rangingfrom 6.0 to 16.8, and slightly positive to negative Euanomalies (Eu/Eu* = 0.8–1.1) (Table 3). When theabundance and angular shape of lithic fragments aretaken into consideration, these features can beinterpreted as reflecting the influence of the REEcharacteristics of the source area on sandstone chemistry,
which may have resulted from a single cycle ofsedimentation. In general, shale fractions that have low(La/Yb)C ratios (3.6–7.8) and negative Eu anomalies maybe regarded as reflecting the homogenized effects on theREEs of hydraulic sorting, chemical weathering and/ordegree of recycling.
Subaerial Weathering
Variable degrees of weathering in source areas may havean important influence on the abundances of alkalis andalkaline-earth elements in siliciclastic sediments. Cationssuch as Rb and Ba are often fixed in weathering profiles,whereas cations with smaller ionic radii – such as Na, Caand Sr – are more rapidly removed from weatheringprofiles as dissolved species (Nesbitt et al. 1980). Acommon approach to quantifying the degree of source-area weathering is to use the chemical index of alteration(CIA; Nesbitt & Young 1982). This index can be calculatedusing molecular proportions
CIA = [Al2O3/(Al2O3+CaO*+Na2O+K2O)]
where CaO* is the amount of CaO incorporated in thesilicate fraction of the rock.
CIA values for the Lower Jurassic shales andassociated sandstones range from 36 to 60 (Table 3),with an average of 53, a value typical of moderatelyweathered rocks. The Upper Cretaceous rocks have CIAvalues of 25–61, with an average of 46 (excluding sample590, which has an unusually high CIA value of 80),indicating a low degree of chemical weathering in thesource area. Compared with the PAAS, these shales andsandstones are enriched in Ca, Na and Sr, whereas K, CsRb, Ba are depleted (Figure 6a & b). This result is at oddswith studies by Nesbitt et al. (1980) and Wronkiewicz &Condie (1987), in which they conclude that smallercations are selectively leached, whereas cations withrelatively large ion radii have been fixed by preferentialexchange and adsorption on clays. Therefore, in general,weathering conditions were not more intensive in theprimitive source areas(s) of the shales and sandstonesthan that/those of the PAAS. A distinct increase in theclay/sand ratio upwards in the Lower Jurassic sequencesmay suggest erosion of an intensely weathered sourcearea and a lessening of tectonic activity, and/or agradually decreasing subsidence rate. There is also anincrease in CIA values stratigraphically upwards (CIA = ~60), suggesting an increase in the amount of weathering
GEOCHEMISTRY OF JURASSIC–CRETACEOUS ROCKS, NE TURKEY
200
or a decrease in the erosion rate of the source terrain.
The Upper Cretaceous rocks have CIA values thatgenerally decrease stratigraphically upwards. Initially highvalues from the Ormandibi profile (CIA = 61) may reflecterosion of a deeply weathered source terrain duringdeposition of the basal sediments. With increasingtectonic activity, relatively unweathered source materialmay have commenced to enter the basin, resulting in adecrease in CIA values in the Ormandini sequence (CIA =25). However, the existence of some red carbonate andsilty to clayey horizons (up to 170 m in thickness) withinthe succession suggest that tectonic and/or volcanicactivity waned, resulting in decreased erosion rates andincreased input of chemically weathered material into thebasin. This may explain the increase in the CIA value ofsandstone sample 590 (CIA = 80) in the Ormandibiprofile.
On the other hand, the red colour of these sedimentsmay have originated from volcanic activity of the palaeo-Pontian arc, just north of the basin, during the LateCretaceous. However, this probability seems to be ruledout in so far as this kind of colouration derives fromsolutions rich in Fe, Mg, Mn and Ti and has no effect onthe CIA indices of the sediments since the CIA formuladoes not use any of these elements. The effects ofhydraulic sorting and transportation on CIA indices is alsoruled out in so far as these processes mainly influencephysical weathering. However, it seems possible thatselective enrichment of a mineral (using one or twocomponents of the CIA formula – such as plagioclase, K-feldspar, corundum and calcium phosphates) occursduring transportation. Such an obvious enrichment isreflected only in the SiO2 contents and SiO2/Al2O3 ratios(Table 1 & 3) of the Lower Jurassic shales relative to theassociated sandstones, indicating that enrichmentoccurred more in the quartz content of the shales than intheir plagioclase and K-feldspar contents. Additionally,the abundance of angular components, particularly in theUpper Cretaceous sandstones, demonstrates that noselective enrichment in the modal percentages of anymineral has occurred along the transport path.Consequently, the tendency for increase in the CIA indicesof the red Upper Cretaceous sediments may beattributed, at least during some time periods, to increasesin the rate of chemical weathering in the source area(s)prior to intense erosion.
On the Al2O3–(CaO*+Na2O)–K2O (A-CN-K) diagram(Figure 11) of Nesbitt & Young (1984, 1989), the LowerJurassic and Upper Cretaceous rocks, as expected, definea linear array along the A-CN join. Two shale samplesfrom the Lower Jurassic rocks plot away from theexpected weathering trend of mafic rocks along the A-CNjoin. Similar trends have been observed in ancientweathering profiles (Grandstaff et al. 1986; Nesbitt &Young 1989) and in Archean shales, such as those of theBuhwa Greenstone Belt (Fedo et al. 1996), and areinterpreted to be a result of K-metasomatism, reflectingeither peculiar soil-forming processes in the Precambrianor post-depositional alteration (Kimberley & Holland1992). Data from the Upper Cretaceous profile plot onthe A-CN join at ~ 80, which may be attributed torelatively intense chemical weathering of the source areaduring some periods. Degree of chemical weathering ismainly a function of climate and rate of erosion, the latterof which is controlled by relief and vegetative cover in thesource area. Assuming high erosion rates (as indicated bythe abundance of coarse detritus in the Upper Cretaceoussediments) – due to the tectonically active setting of thebasin and the degree of weathering – a lack of vegetationat least during some periods of the Late Cretaceous and astrong influence of climate on chemical weathering are
A. DOKUZ & E. TANYOLU
201
cpxhbl
pl
sm
kao
ilms
kf
0
20
40
60
80
100
Upper Cretaceous samples
Lower Jurassic samples
weathering trend
metasomatic trend
rock composition
mineral composition
Al2O3
CaO* + Na2O K2O
average gabbro
averagegranite
Figure 11. Al2O3–(CaO*+Na2O)–K2O (A-CN-K) diagram of the LowerJurassic and Upper Cretaceous shales and sandstones(compositions as molar proportions, CaO* represents CaOof the silicate fraction only). Selected rock and mineralcompositions and weathering trends (after Nesbitt &Young 1984) are given. Kao– kaolinite, il– illite, ms–muscovite, kf– alkali-feldspar, sm– smectite, pl–plagioclase, hbl– hornblende, cpx– clinopyroxene; averagegranite and gabbro from Nesbitt & Young (1984).
suggested. Severe weathering may have been related to aCO2-rich atmosphere and/or elevated surfacetemperatures; an arid climate may thus be postulated forthe Upper Cretaceous in the source area (e.g., Young1991; Kasting 1993; des Marais 1994).
Tectonic Setting
Plate-tectonic setting discrimination of ancientsedimentary basins are usually accomplished using major-and trace-element bivariate and multivariate plots.Although such graphs are not really significant for specificlocal plate-tectonic settings, some correlations betweentectonics and geochemical processes in sandstones, aswell as the relationships among and temporal and spatialvariations within various lithostratigraphic units, can beevaluated.
The ratio of SiO2 vs K2O/Na2O provides a tectonic-setting discriminator for sandstone-mudstone suites(Roser & Korsch 1986). In Figure 12a, the investigatedsamples plot mainly in the fields of arc and activecontinental-margin settings. The samples plotting in thepassive-margin setting represent quartz- andphyllosilicate-rich shales of Early Jurassic age that showsimilar patterns in most of the following tectonic-discrimination diagrams. A similar plot using theSiO2/Al2O3 ratio instead of SiO2 discriminates thesandstones and mudstones of different settings (Maynardet al. 1982) (Figure 12b). In this plot, the investigatedsamples mainly group in the arc field due to their lowSiO2/Al2O3 and K2O/Na2O ratios, and can be separated intotwo categories: A1 (<10 % Q, basaltic and andesiticdetritus) and A2 (higher Q, acidic volcanic detritus). Shalesthat are mineralogically mature (quartz-rich) scatter inthe active continental-margin and passive-margin fields,as seen in the graph of SiO2 vs K2O/Na2O. Two ternaryplots, in which the highly incompatible trace elements Thand Zr are plotted against the highly compatible elementsCo and Sc, have been used (Bhatia & Crook 1986) (Figure12c & d). The analysed samples plot in the fields foroceanic island arc and continental island arc, withgenerally high Zr/Th ratios and moderate abundances ofCo and Sc. Two shale samples from the Lower Jurassicrocks plot outside the fields on the Th-Co-Zr/10 plot dueto their low Co contents; this situation is probably due todepletion of ferromagnesian minerals and basalticfragments and increased abundance of quartz during
sedimentary processes. In general, there is a distinctoverlap, albeit with slight difference, of the compositionsof the Lower Jurassic and Upper Cretaceous units.Another important point to note is that with decreasedgrain size (e.g., the shales), the samples show a tendencyto fall into the passive-margin field. The reason for this isthe ‘grain-size effect’ on chemical composition. Roser &Korsch (1985) found that the modal and chemicalcompositions of greywackes and argillites are stronglycontrolled by grain size. As mean grain size decreases intransition from sandstone to argillite, modal quartz,feldspar and lithic fragments decline in abundance andmodal matrix and phyllosilicate correspondingly increase;this strongly affects bulk chemistry. SiO2 and Na2Odecrease regularly from sandstone to argillite, and K2Oincreases. These changes produce a progressive increasein K2O/Na2O ratio with decreasing SiO2. As for the LowerJurassic shales, they also display higher K2O/Na2O ratiosrelative to associated sandstones, except for higher SiO2
contents which are undoubtedly related to theirunexpectedly high quartz contents, as mentioned above.
Evolution of the Basin
Data from the sandstones and shales of the Yusufeli area,when considered in conjunction with sedimentological,structural and geodynamic data given in earlier papers,allow a reconstruction of the sedimentological andtectonic events in the Yusufeli area from the earliestJurassic to present. In light of all of these data, it isproposed that the basin evolved from a rifted margin toa passive margin (fiengör &Y›lmaz 1981; Bektafl et al.1984; Koçyi¤it et al. 1992; Ustaömer & Robertson1993; Y›lmaz et al. 1997; Okay & fiahintürk 1977;Adamia et al. 1997; Dean 2005), and then to a forelandbasin (Koçyi¤it 1991; Tüysüz et al. 1995; Y›lmaz et al.1997; Okay & fiahintürk 1997) prior to completeclosure. Sedimentary detritus that contributed to eachunit was not only derived from the crystalline basementbut, also, and more significantly, from the volcanic rocksof a Late Palaeozoic arc. Figure 13 shows a summary andchronology of events for the Mesozoic basin in theYusufeli area. Each stage of this basin’s evolution isrepresented by a lithological sequence comprising anumber of rock units. On the basis of the sedimentaryarchitecture, the depositional history of this basin can bedivided into four stages.
GEOCHEMISTRY OF JURASSIC–CRETACEOUS ROCKS, NE TURKEY
202
The earliest stage was characterised by the break-upof an arc during Early Jurassic time in which syn-riftsequences, including the Hamurkesen formation andequivalent basic volcanic rocks along the southern marginof the arc, accumulated. During this break-up, rifting isenvisaged to have taken place not only along the southernmargin of the arc but also in the interior of Neotethyanbasin to the south, forming a series of intra-arc rifts
parallel to the continental margin, which later evolved toform grabens or half grabens (fiengör et al. 1980;fiengör & Y›lmaz 1981; Bektafl et al. 1984; Tüysüz1990; Y›lmaz et al. 1997; Okay & fiahintürk 1997).
The transition from a rifted to a passive margin isrepresented by the deposition of first marine sediments(after coal-bearing non-marine sediments) above anunconcormity, upon which the lowermost part
A. DOKUZ & E. TANYOLU
203
Co
ACM
PM
OIA
C IA
Th
ACM
PM
OI A
CIA
Zr/10Zr/10 Sc
Upper Cretaceous samples
Lower Jurassic samples
Th
PM
ACMArc
SiO2 (%)
100.0
10.0
1.0
0.1
0.040 50 60 70 80 90 100
K2O
/Na 2O
(a)
(c) (d)
0.01 0.10 1.00 10.00 100.000
2
4
6
8
10
PMA1
A2
K2O/Na2O
ACM (b)
Figure 12. Plot of samples from the Yusufeli area on tectonic discrimination diagrams in order to distinguish sources forthe sedimentary rocks. (a) SiO2–Log K2O/Na2O diagram of Roser & Korsch (1986), and (b) SiO2/Al2O3 vs.K2O/Na2O diagram of Maynard et al. (1982). On both diagrams, sandstone samples fall in the fields for arc andactive continental-margin settings, whereas some of the shale samples fall in the field for a passive-marginsetting which is caused by grain-size effects on the chemical compositions of the rocks. As the grain size reduces,the phyllosilicate abundance correspondingly increases in the transition from sandstones to shales (Roser &Korsch 1985). (c) Th-Co-Zr/10 and (d) Th-Sc-Zr/10 diagrams reported by Bhatia & Crook (1986). PM– passivemargin, ACM– active continental margin, CIA– continental island arc, OIA– oceanic island arc, A1– arc setting,basaltic and andesitic detritus, A2– evolved arc setting, acidic volcanic detritus.
(Hamurkesen formation) of the Mesozoic sequence wasdeposited. During this stage the basin is interpreted tohave been in an intra-arc or a back-arc setting withrespect to the Early Jurassic Pontian arc to the north(fiengör &Y›lmaz 1981; Koçyi¤it et al. 1992; Ustaömer &Robertson 1993; Y›lmaz et al. 1997; Okay & fiahintürk1977). Back-arc extension in response to generation ofthe magmatic arc may lead to rapid subsidence andaccumulation of coarse-clastic sediments in the deeperparts of the basin. In so far as the basin was tectonicallyunstable during this period, the horst sites – especiallythose near the shore of the basin – could not have
received clastic sediments and may have remained assource areas to grabens even though they were below sealevel. The absence of carbonates in the Lower Jurassicprofiles probably demonstrates that sedimentation wasrelatively rapid and that the region was near the shore tothe north of the Neotethyan basin. Throughout the timein which the horst-graben topography of the basin wassmoothed out, sedimentation seems to have dominated ingraben sites. The variable thickness of the Lower Jurassicsuccessions and the occurrence or absence of coarseclastics at the bottoms of the successions may beexplained by this horst-graben topography of seafloor.
GEOCHEMISTRY OF JURASSIC–CRETACEOUS ROCKS, NE TURKEY
204
J U
R A
S S
I C
C R
E T
A C
E O
U S
LIA
SD
OG
GE
RM
ALM
HettangianSinemurianPliensbachianToarcianAalenianBajocianBathonianCallovianOxfordian
Kimmeridgian
TithonianBerrriasianValanginianHauterivian
Barremian
Aptian
Albian
CenomanianTuronianConiacianSantonianCampanianMaastrichtian
PALEOCENE
EOCENE
OLIGOCENE
MIOCENE
LOW
ER
UP
PE
R
170
155
140
125
110
95
80
65
50
35
20 Ma
200
185
Depositional History
Burial and Uplift History
InterpretedTectonic Events
Hamurkesenvolcanics
Hamurkesenshales andsandstones
platform typecarbonates(Berdiga formation)
Yusufelisandstonesandshales
clasticsedimentary
continuoussubsidence
rapid subsidence
Initiation ofsubsidence
maximum burial
renewedsubsidence
tilting, uplift anderosion
renewed thrusting
extension, reversal of thrust,Tertiary basin development
compression, reversal fromextension to compression anddevelopment of an active marginalong the north of Neo-Tethyanbasin
tilting and uplift ofMesozoicformations
compression, uplift and erosionof paleo - continental arc rocks,exposure of Pontian basementrocks
extension, development ofblock faulted continental riftbasin
growth of rift basin to a shallowmarginal basin with the initiationof f i rs t mar ine sed imentdeposition
become slowerand cessationof subsidence
erosion of Yusufeliformation from the
TE
RT
IAR
Y
initiation of tilting
closure of Neotethyan
continuation of marginalbasin growth
Figure 13. Summary and chronology of events from Early Jurassic to present in the Yusufeli area, Turkey.
Then, as the basin gradually deepened with its smoothbase topography and with the possible decrease intectonic activity, it became a suitable environment for thedeposition of shales, as abundantly represented in theupper parts of the Lower Jurassic successions. Theexistence of some basaltic sills within the formationsuggests that back-arc extension in the south advancedand led to break-up in the arc to the north. Therefore,during this period, detritus may have been derived fromthe Late Palaeozoic arc where mainly volcanic rocks ofcalc-alkaline character were exposed, along withsubordinate metamorphic rocks and granites. Thereseems to have been little or no contribution of detritusoriginating from syn-rift volcanic rocks due to theirposition at the base of succession. Conformably overlyingplatform-type carbonates (the Berdiga formation) are theend products of this extensional regime, and weredeposited from the Kimmeridgian to the Cenomanian;these are characterized by thin-bedded micritic limestoneat the base, but acquire a thick or massive beddedstructure upwards. Thus, the basin may be interpreted tohave formed during a period in which subsidencerelatively slowed or ceased, even though upper parts of itmay have been compressed or uplifted in the Aptian orAlbian.
The third stage of the basin coincided with depositionof Yusufeli formation, which generally comprisesturbidites. Upward coarsening in clast size and bedthickness in the unit indicates that the basin was filledgradually. A likely tectonic scenario may have involved areverse in tectonic style from extension to compressionand development of an active margin along the northernside of the Neotethyan basin (fiengör & Y›maz 1981;Koçyi¤it 1991; Tüysüz et al. 1995; Y›lmaz et al. 1997;Okay & fiahintürk 1997). Red clastics, up to 23 m thickat the base of Ormandibi profile, were the first depositsof the Yusufeli formation over the carbonate rocks andtend to show higher CIA indices in compared to the greyto green clastics of the unit. All of these data suggest thata long, tectonically stable period and oxidizing climaticconditions in the source area(s) prevailed during thedeposition of carbonates from the Kimmeridgian to theBarremian. Additionally, a sudden transition fromdeposition of carbonates to red clastics in the Aptian maybe interpreted as a transition from a tectonically stablesituation to an unstable one, and as initiation of uplift inthe source area. Consequently, there was a change in theweathering conditions of the source area(s), such as from
intensely oxidizing weathering to predominantly physicalweathering. The other red horizons in the upper levels ofthe unit may also be interpreted as products of crustalprovenance areas, where arid climatic conditions andstatic periods (in terms of crustal movements) prevailedduring particular periods prior to intense physicalweathering and transportation. Until the end of thePalaeocene, the present-day Yusufeli area was above sealevel (Robinson et al. 1995; Okay et al. 2001). Thus,nearly all of the Upper Cretaceous formations wereeroded from atop the eastern Pontides.
The other short-lived change in tectonic regime –from compression to extension – occurred in the mid-Eocene. After a short period of extension, which is knownas the Lutetian transgression throughout the Pontides(fiengör & Y›lmaz 1981; Okay et al. 1997, 2001; Y›lmazet al. 1997), this area and the whole Pontian belt wereaffected by renewed compression and, hence, a finalepisode of tilting, uplift and erosion occurred.
Conclusions
In the present study, standard petrographic techniqueswere combined with major- and trace element analyses inorder to elucidate the provenance, sedimentary historyand geochemistry of the Lower Jurassic and UpperCretaceous terrigenous rocks of the Yusufeli area (Artvin,Turkey). Petrographic and geochemical data indicate thatthe sediment source for Lower Jurassic and UpperCretaceous rocks can be directly related to moderatelyevolved volcanic-arc rocks. Thus, the likely source terrainseems to have comprised pre-Jurassic volcanic rocks, andvery possibly the volcanic rocks of a Late PalaeozoicPontian arc which are presently not exposed (or no longerexist). La/Sc, Th/Sc, Cr/Th, Co/Th and Cr/Ni ratios, whichare good indicators of provenance, demonstrate that atleast two types of rocks were exposed in the source areaand provided detritus to the basin, consisting of adominant mafic volcanic source and a subordinate felsicvolcanic source.
Mineral fractionation is not observed in the rocks,indicating compound influences on REE distributionrather than a single, selective one. However, positivecorrelation between Zr and LREEs and between HREEsand Y and P2O5 in the Upper Cretaceous samples indicatesthat zircon, hornblende and phosphate minerals seem tohave at least partially controlled REE distribution.
A. DOKUZ & E. TANYOLU
205
High SiO2 contents and the LREE-depleted patterns ofthe shales relative to associated sandstones in the PAAS-normalised plot indicate enrichment in silt-sized quartzparticles, which have a low total REE content even thoughthey yield an LREE-enriched pattern. This situation maybe regarded as an effect of hydraulic sorting on REEchemistry.
The average CIA indices for subaerial weatheringindicate that the source terrain for the Lower Jurassic andUpper Cretaceous sediments was not affected by intensechemical weathering. However, a distinct upward(stratigraphically) increase in shale/sand ratio and CIAindices in the Lower Jurassic profile suggest that thesource area underwent gradually intensifying chemicalweathering, possibly due to a decrease in tectonic activity.In contrast, a general upward decrease in CIA indices in
the Upper Cretaceous profiles demonstrates that thesediments were derived from a gradually less-weatheredsource terrain, reflecting increased erosion rates likelydue to increasing tectonic activity. However, red horizonsin the unit may correspond to periods in which tectonicactivity and erosion slowed and weathering increased.
Acknowledgements
The authors thank Kemal Erdo¤an (General Directorateof Mineral Research and Exploration [MTA], Ankara) forhis determination of fossil assemblages from our thinsections. The manuscript was constructively reviewed byA. Sami Derman, N. Terzio¤lu and an anonymous referee;their insightful comments are gratefully acknowledged.
GEOCHEMISTRY OF JURASSIC–CRETACEOUS ROCKS, NE TURKEY
206
ADAMIA, S.A., LORDKIPANIDZE, M.B. & ZAKARIADZE, G.S. 1977. Evolution ofan active continental margin as exemplified by the Alpine historyof the Caucasus. Tectonophysics 40, 183–199
AKIN, H. 1978. Geologie, Magmatismus und Lagerstaettenbildung imostpontischen Gebirge-Turkei aus der Sicht der plattentektonik.Geologische Rundschau 68, 253–283.
AKINCI, Ö.T. 1984. The eastern Pontide volcano-sedimentary belt andassociated massive sulphide deposits. In: DIXON, J.E. & ROBERTSON
A.H.F. (eds), The Geological Evolution of the EasternMediterranean. Geological Society, London Special Publications17, 415–428.
ALP, D. 1972. Amasya Yöresinin Jeolojisi [Geology of Amasya Region].Ph.D. ‹stanbul Üniversitesi Fen Fakültesi Monografileri Tabii‹limler K›sm› 22, 101 p.
ASLAN, Z. 2005. Petrography and petrology of the calc-alkaline Sar›hangranitoid (NE Turkey): an axample of magma mingling andmixing. Turkish Journal of Earth Sciences 14, 183–207.
ARSLAN, M., TÜYSÜZ, N., KORKMAZ, S. & KURT, H. 1997. Geochemistry andpetrogenesis of the eastern Pontide volcanic rocks, NortheastTurkey. Chemie der Erde 57, 157–187.
BANFIELD, J.F. & EGGELTON, R.A. 1989. Apatite replacement and rareearth mobilization, fractionation and fixation during weathering.Clays and Clay Minerals 37, 113–127.
BEKTAfl, O., PEL‹N, S. & KORKMAZ, S. 1984. Do¤u Pontid yay gerisihavzas›nda manto yükselimi ve polijenik ofiyolit olgusu [The riseof mantle in the eastern Pontide back-arc basin and polygenicophiolite fact]. Türkiye Jeoloji Kurumu Ketin Sempozyumu,175–188 [in Turkish with English abstract].
BEKTAfl, O., fiEN, C., ATICI, Y. & KÖPRÜBAflI, N. 1999. Migration of theUpper Cretaceous subduction-related volcanism towards the backarc basin of the eastern Pontide magmatic arc (NE Turkey).Geological Journal 34, 95–106.
BERGOUGNAN, H. 1987. Etudes geologiques dans I’Est Anatolien. PhDThesis, University Pierre et Marie Curie, France, 606 p.
BHATIA, M.R. & CROOK, K.A.W. 1986. Trace element characteristics ofgraywackes and tectonic setting Discrimination Of SedimentaryBasin. Contributions To Mineralogy And Petrology 92, 181–193.
BOZKURT, E. & MITTWEDE, S.K. 2001. Introduction to the geology ofTurkey–a synthesis. International Geology Review 43, 578–594.
BOZKUfl, C. 1990. Olur-Kömürlü-Akflar (Erzurum) Aras›n›n Stratigrafikve Tektonik Özellikleri [Stratigraphic and Tectonic Characteristicsof Olur-Kömürlü-Akflar (Erzurum) Area]. PhD Thesis, CumhuriyetÜniversitesi, Sivas, Türkiye, 162 p [in Turkish with Englishabstract, unpublished].
CHATALOV, A.G. 1991. Triassic in Bulgaria. In: DEWEY, J.F. (ed), SpecialIssue on Tectonics. Bulletin of the Technical University of ‹stanbul44, 103–135.
CONDIE, K.C. 1993. Geochemical composition and evolution of the uppercontinental crust: contrasting results from surface samples andshales. Chemical Geology 104, 1–37.
CONDIE, K.C., DENGATE, J. & CULLERS, R.L. 1995. Behavior of rare earthelements in a paleoweathering profile on granodiorite in the frontrange, Colorado, USA. Geochimica et Cosmochimica Acta 59,279–294.
CONDIE, K.C. & WRONKIJEWICZ, D.J. 1990. The Cr/Th ratio inPrecambrian pelites from the Kaapvaal craton as an index ofcraton evolution. Earth and Planetary Science Letters 97,256–267.
CREASER, R.A., ERDMER, P., STEVENS, R.A. & GRANT, S.L. 1997. Tectonicaffinity of Nisutlin and Avil assemblages strata from the Teslintectonic zone, northern Canadian Cordillera: constraints fromneodymium isotope and geochemical evidence. Tectonics 16,107–121.
References
CULLERS, R.L. 1994. The controls on the major and trace elementvariation of shales, siltstones, and sandstones of Pennsylvanian-Permian age from uplifted continental blocks in Colorado toplatform sediment in Kansas, USA. Geochimica et CosmochimicaActa 58, 4955–4972.
CULLERS, R.L., BARRET, T., CARLSON, R. & ROBINSON, B. 1987. Rare earthelement and mineralogical changes in Holocene soil and streamsediment: a case study in the Wet Mountains, Colorado, USA.Chemical Geology 63, 275–295.
Cullers, R.L., Basu, A. & Suttner, L. 1988. Geochemical signature ofprovenance in sand-size material in soils and stream sedimentsnear the Tobacco Root batholith, Montana, USA. ChemicalGeology 70, 335–348.
CULLERS, R.L., CHAUDHURI, S., KILBANE, N. & KOCH, R. 1979. Rare earthsin size fractions and sedimentary rocks of Pennsylvanian-Permianage from the mid-continent of the USA. Geochimica etCosmochimica Acta 43, 1285–1301.
ÇO⁄ULU, E. 1975. Gümüflhane ve Rize Granitik Plütonlar›n›n MukayeseliPetrojeolojik ve Jeokronometrik Etüdü [Petrologic andGeochronological Investigation and Comparison of Gümüflhaneand Rize Plutons]. DSc Thesis, ‹stanbul Technical UniversityPublications 1034, ‹stanbul, 112 p [in Turkish with Englishabstract, unpublished].
DES MARAIS D.J. 1994. The Archean atmosphere: its composition andfate. In: CONDIE, K.C. (ed), Archean Crustal Evolution.Developments in Precambrian Geology 11, Elsevier.
DEAN, W.T. 2005. Trilobites from the Çal Tepe formation (Cambrian),near Seydiflehir, central Taurides, southwestern Turkey. TurkishJournal of Earth Sciences 14, 1–71.
DICKINSON, W.R., BEARD, L.S., BRAKENRIDGE, G.R., ERJAVEC, J.L., FERGUSON,R.C., INMAN, K.F., KNEPP, R.A., LINDBERG, F.A. & RYBERG, P.T.1983. Provenance of North American Phanerozoic sandstone inrelation to tectonic setting. Geological Society of America Bulletin93, 222–235.
DICKINSON, W.R. & SUCZEK, C.A. 1979. Plate tectonics and sandstonecompositions. American Association of Petroleum GeologistsBulletin 63, 2164–2182.
DOKUZ, A. 2000. Yusufeli (Artvin-Türkiye) Yöresinin Jeolojisi,Jeotektoni¤i, Magmatik-Metamorfik Kayaçlar›n Jeokimyas› vePetrojenezi [Geology, Geotectonics, Geochemistry andPetrogenesis of Magmatic and Metamorphic Rocks in Yusufeli(Artvin-Turkey)]. PhD Thesis, Karadeniz Teknik Üniversitesi,Trabzon, Türkiye, 311 p [in Turkish with English abstract,unpublished].
DOTT, R.H. 1964. Wacke, graywacke and matrix: what approach toimmature sandstone classification? Journal of SedimentaryPetrology 34, 625–632.
ERDO⁄AN, B. & GÜNGÖR, T. 2004. The problem of the core-coverboundary of the Menderes Massif and an emplacement mechanismfor regionally extensive gneissic granites, western Anatolia(Turkey). Turkish Journal of Earth Sciences 13, 15–36.
FEDO, C.M., ERIKSSON, K.A. & KROGSTAD, E.J. 1996. Geochemistry ofshales from the Archean (~ 3.0 Ga) Buhwa Greenstone Belt,Zimbabwe: implications for provenance and source areaweathering. Geochimica et Cosmochimica Acta 60, 1751–1763.
FLOYD, P.A., ÖZGÜL, M. & GÖNCÜO⁄LU, C. 2003. Metabasite blocks fromthe Koçkaya HP-LT metamorphic rocks, Konya, central Anatolia:geochemical evidence for an arc-back-arc pair? Turkish Journal ofEarth Sciences 12, 157–174.
GED‹KO⁄LU, A., PEL‹N, S. & ÖZSAYAR, T. 1979. The main lines ofgeotectonic development of the east Pontides in the Mesozoic era:Proceedings of the 1st Geological Congress of the Middle East(GEOCOME), 555–580.
GERDJIKOV, I. 2005. Alpine metamorphism and granitoid magmatism inthe Strandja zone: new data from the Sakar unit, SE Bulgaria.Turkish Journal of Earth Sciences 14, 167–183.
GÖTZE, J. 1998. Geochemistry and provenance of the Altendorffeldspathic sandstone in the Middle Bunter of the Thuringianbasin, Germany. Chemical Geology 150, 43–61.
GRANDSTAFF, D.E., EDELMAN, M.J., FOSTER, R.W., ZBINDEN, E. & KIMBERLEY,M.M. 1986. Chemistry and mineralogy of Precambrian paleosolsat the base of the Dominion and Pongola groups (Transvaal,South Africa). Precamrian Research 32, 97–131.
KARSLI, O., AYDIN, F. & SADIKLAR, M.B. 2004. Magma interactionrecorded in plagioclase zoning in granitoid systems, Ziganagranitoid, Eastern Pontides, Turkey. Turkish Journal of EarthSciences 13, 287–305.
KASTING, J.F. 1993. Earth’s early atmosphere. Science 259, 920–926
KESK‹N, ‹., KORKMAZ, S., GED‹K, ‹., ATEfl, M., GÖK, L., KÜÇÜMEN, Ö. & ERKAL,T. 1989. Geology of the Region Around Bayburt. Maden Tetkikve Arama Genel Müdürlü¤ü (MTA), Report No. 8995, 128 p [inTurkish, unpublished].
KIMBERLEY, M.M. & HOLLAND, H.D. 1992. Introduction to Precambrianweathering and paleosols. In: SCHIDLOWSKI, M., GOLUBIC, S.,KIMBERLEY, M.M., MCKIRDY, D.M. & TRUDINGER, P.A. (eds), EarlyOrganic Evolution:. Implications for Mineral and EnergyResources, Springer, 9–15.
KETIN, ‹. 1966. Tectonic units of Anatolia (Asia Minor). MTA Bulletin 66,23–34
KOÇY‹⁄‹T, A. 1991. An example of an accretionary fore-arc basin fromnorthern central Anatolia and its implications for the history ofsubduction of Neotethys in Turkey. Geological Society of AmericaBulletin 103, 22–36.
KOÇY‹⁄‹T, A., ALTINER, D., FARINACCI, A., UMBERTO, N. & CONTI M.A. 1992.Late Triassic-Aptian evolution of the Sakarya divergent margin:implications for the opening history of the northern Neotethys innorthwest Anatolia, Turkey. Geologica Romana 27, 81–99.
KORALAY, E., DORA, O.Ö., CHEN, F., SATIR, M. & CANDAN, O. 2004.Geochemistry and geochronology of orthogneisses in the Derbent(Alaflehir) area, eastern part of the Ödemifl-Kiraz Submassif: Pan-African magmatic activity. Turkish Journal of Earth Sciences 13,37–61.
A. DOKUZ & E. TANYOLU
207
MAYNARD, J.B., VALLONI, R. & YU, H.S. 1982. Composition of moderndeep-sea sands from arc related basins. In: LEGGET, J.K (ed),Trench-Forearc Geology: Sedimentation and Tectonics on Modernand Ancient Active Plate Margins. Geological Society, London,Special Publications 10, 551–561.
MCBRIDE, E.F. 1963. Diagenetic processes that affect provenancedeterminations in sandstone. In: ZUFFA, G.G. (ed), Provenance ofArenites. Reidel Dordrecht, 95–114.
MCLENNAN, S.M. 1989. Rare earth elements in sedimentary rocks:influence of provenance and sedimentary processes. In: LIPIN, B.R.& MCKAY, G.A. (eds), Geochemistry and Mineralogy of REE.Reviews in Mineralogy 21, 169–200.
MCLENNAN, S.M., HEMMING, S.R., MCDANIEL, D.K. & HANSON, G.N. 1993.Geochemical approaches to sedimentation, provenance andtectonics. In: JHONSON, M.J & BASU, A. (eds), Processes Controllingthe Composition of Clastic Sediments. Geological Society ofAmerica Special Paper 284, 21–40.
MCLENNAN, S.M., TAYLOR, S.R. & KRONER, A. 1983. Geochemical evolutionof Archean shales from South Africa: I. the Swaziland and PongolaSupergroups. Precambrian Research 22, 93–124.
MCLENNAN, S.M., TAYLOR, S.R., MCCULLOCH, M.T. & MAYNARD, J.B. 1990.Geochemical and Nd-Sr isotopic composition of deep-seaturbidites: crustal evolution and plate tectonic associations.Geochimica et Cosmochimica Acta 54, 2015–2050.
NESBITT, H.W. 1979. Mobility and fractionation of REE duringweathering of granodiorite. Nature 279, 206–210.
NESBITT, H.W., MARKOVICS, G. & PRICE, R.C. 1980. Chemical processesaffecting alkalis and alkaline earths during continental weathering.Geochimica et Cosmochimica Acta 44, 1659–1666.
NESBITT, H.W. & YOUNG, G.M. 1982. Early Proterozoic climate and platemotions inferred from major element chemistry of lutites. Nature299, 715–717.
NESBITT, H.W. & YOUNG, G.M. 1984. Predictions of some weatheringtrends of plutonic and volcanic rocks based on thermodynamic andkinetic considerations. Geochimica et Cosmochimica Acta 48,1523–1534.
NESBITT, H.W. & YOUNG, G.M. 1989. Formation and diagenesis ofweathering profiles. Journal of Geology 97, 129–147.
OKAY, A.‹. 1996. Granulite facies gneisses from the Pulur region,eastern Pontides. Turkish Journal of Earth Sciences 5, 55–61.
OKAY, A.‹. & GÖNCÜO⁄LU, C. 2004. The Karakaya complex: a review ofdata and concepts. Turkish Journal of Earth Sciences 13, 75–95.
OKAY, A.‹. & fiAH‹NTÜRK, Ö. 1997. Geology of the eastern Pontides. In:ROBINSON A.G. (ed), Regional and Petroleum Geology of the BlackSea and Surrounding Region. AAPG Memoir 68, 291–311.
OKAY, A.‹., TANSEL, ‹. & TÜYSÜZ, O. 2001. Obduction, subduction andcollision as reflected in the Upper Cretaceous–Lower Eocenesedimentary record of western Turkey. Geological Magazine 138,117–142.
ÖZER, E. 1984. The geology of Bayburt (Gümüflhane) region. KaradenizTeknik Üniversitesi Dergisi Jeoloji 3, 77–89.
PETTIJOHN, F.J., POTTER, P.E. & SIEVER, R. 1987. Sand and Sandstone(2nd ed). Springer, Berlin, Heidelberg, NewYork.
PICKETT, E.A. & ROBERTSON, A.H.F. 2004. Significance of thevolcanogenic Nilüfer unit and related components of the TriassicKarakaya Complex for Tethyan subduction/accretion. TurkishJournal of Earth Sciences 13, 97–143.
REIMER, T.O. 1985. Volcanic rocks and weathering in the EarlyProterozoic Witwatersrand Supergroup, South Africa. GeololgicalSurvey of Finland Bulletin 331, 33–49.
ROBINSON, A.G., BANKS, C.J., RUTHERFORD, M.M. & HIRST, J.P.P. 1995.Stratigraphic and structural development of the eastern Pontides,Turkey. Journal of the Geological Society, London 152, 861–872.
ROSER, B.P. & KORSCH, R.J. 1985. Plate tectonics and geochemicalcomposition of sandstones: a discussion. Journal of Geology 93,81–84.
ROSER, B.P. & KORSCH, R.J. 1986. Determination of tectonic setting ofsandstone-mudstone suites using SiO2 content and K2O/Na2Oratio. Journal of Geology 94, 635–694.
SLACK, J.F. & STEVENS, P.J. 1994. Clastic metasediments of the EarlyProterozoic Broken Hill Group, New South Wales, Australia:geochemistry, provenance, and metallogenic significance.Geochimica et Cosmochimica Acta 58, 3633–3652.
fiEN, C., ARSLAN, M. & VAN, A. 1998. Geochemical and petrologicalcharacteristics of eastern Pontide Eocene (?) alkaline volcanicprovince, NE Turkey. Turkish Journal of Earth Sciences 7,231–239.
fiENGÖR, A.M.C. & YILMAZ, Y. 1981. Tethyan evolution of Turkey: a platetectonic approach. Tectonophysics 75, 181–241.
fiENGÖR, A.M.C., YILMAZ, Y. & KET‹N, ‹. 1980. Remnants of Pre-LateJurassic ocean in northern Turkey: fragments of Permian-TriassicPaleo-Tethys. Geological Society of America Bulletin 91,599–609.
SPADEA, P., TORTORICI, L. & LANZAFAME, G. 1980. Ophiolites of Tyrrheniancoastal chain. In: BORTOLOTTI, V. (ed), VIth Ophiolite FieldConference, Field Excursion Guidebook. Gruppo Lavoro OfiolitiMediterranee, Florence, Italy, 19–27.
TAYLOR, S.R. & MCLENNAN, S.M. 1985. The Continental Crust: ItsComposition and Evolution. Oxford, Blackwell ScientificPublishing, 312 p.
TOKEL, S. 1977. Do¤u Karadeniz bölgesinde Eosen yafll› kalk-alkalenandezitler ve jeeotektonizma [Geotectonics and Eocene calc-alkaline andesites in eastern Black Sea region]. Türkiye JeolojiKurumu Bülteni 20, 49–54 [in Turkish with English abstract].
TOPUZ, G. & ALTHERR, R. 2004. Pervasive rehydration of granulitesduring exhumation – an example from the Pulur complex , easternPontides, Turkey. Mineralogy and Petrology 81, 165–185.
TOPUZ, G., ALTHERR, R., KALT, A., SATIR, M., WERNER, O. &. SCHWARZ,W.H. 2004. Aluminous granulites from the Pulur complex, NETurkey: a case of partial melting, efficient melt exctraction andcrystallisation. Lithos 72, 183–207.
TÜYSÜZ, O. 1990. Tectonic evolution of a part of the Tethyside orogeniccollage: the Karg› Massif, northern Turkey. Tectonics 9,141–160.
GEOCHEMISTRY OF JURASSIC–CRETACEOUS ROCKS, NE TURKEY
208
TÜYSÜZ, O., DELLALO⁄LU, A.A. & TERZ‹O⁄LU, N. 1995. A magmatic beltwithin the Neo-Tethyan suture zone and its role in the tectonicevolution of northern Turkey. Tectonophysics 243, 173–191.
USTAÖMER, T. & ROBERTSON, A.H.F. 1993. A Late Palaeozoic-EarlyMesozoic marginal basin along the active southern continentalmargin of Eurasia: evidence from the central Pontides (Turkey)and adjacent regions. Geological Journal 28, 219–238.
WRONKIJEWICZ, D.J & CONDIE, K.C. 1987. Geochemistry of Archean shalesfrom the Witwatersand supergroup, South Africa: source areaweatheing and provenance. Geochimica et Cosmochimica Acta 51,2401–2416.
WRONKIJEWICZ, D.J & CONDIE, K.C. 1989. Geochemistry and provenanceof sediments from the Pongola Supergroup, South Africa:evidence for a 3.0 Ga-old continental craton. Geochimica etCosmochimica Acta 53, 1537–1549.
WRONKIJEWICZ, D.J & CONDIE, K.C. 1990. Geochemistry and mineralogyof sediments from the Ventersdrop and Transvaal supergroups,South Africa: cratonic evolution during the Early Proterozoic.Geochimica et Cosmochimica Acta 54, 343–354.
YILMAZ, Y. 1972. Petrology and Structure of the Gümüflhane Graniteand Surrounding Rocks, North-Eastern Anatolia. PhD Thesis,University of London, 260 p [unpublished].
YILMAZ, Y., TÜYSÜZ, O., Y‹⁄‹TBAfl, E., GENÇ, fi.C. & fiENGÖR, A.M.C. 1997.Geology and tectonic evolution of the Pontides. In: ROBINSON, A.G.(ed), Regional and Petroleum Geology of the Black Sea andSurrounding Region. AAPG Memoir 68,183–226.
YILMAZ, A. & YILMAZ, H. 2004. Geology and structural evolution of theTokat Massif, eastern Pontides, Turkey. Turkish Journal of EarthSciences 13, 231–246.
YOUNG, G.M. 1991. The geologic record of glaciation: relevance to theclimatic history of the earth. Geoscience Canada 18, 100–108.
A. DOKUZ & E. TANYOLU
209
Received 02 February 2005; revised typescript accepted 25 November 2005