Vol.:(0123456789)1 3

Mediterranean Geoscience Reviews (2019) 1:179–202 https://doi.org/10.1007/s42990-019-00010-8

ORIGINAL PAPER

Tephrostratigraphy and chronology of the Quaternary Gölludağ and Acıgöl volcanic complexes (Central Anatolia, Turkey)

D. Mouralis1  · J.‑F. Pastre2 · C. Kuzucuoğlu2 · A. Türkecan3 · H. Guillou4

Received: 3 July 2019 / Revised: 13 November 2019 / Accepted: 16 November 2019 / Published online: 9 December 2019 © Springer Nature Switzerland AG 2019

AbstractThe two Göllüdağ and Acıgöl volcanic complexes are located in Cappadocia and belong to the Central Anatolia Volcanic Province. Their activity covers a wide time range from Middle to Upper Pleistocene and the Holocene. The large amount of tephra emitted explains the regional importance of this activity; whereas, the study of obsidian associated with their activity relates to several archaeological topics and research questions. Intensive field work in and around both complexes allows to present here a detailed tephrostratigraphy. In addition, results of ~ 460 single grain geochemical analyses (about 5–10 analyses for each sample out of a total of 48 samples) as well as intensive mineral counting provide an accurate reference set of signatures of Pleistocene rhyolitic tephras in Cappadocia. Correlations are based on stratigraphic observation and characterization of the tephra layers, and are also supported by multivariate statistical analyses. Chronology, constrained by 18 radiometric data (mainly K/Ar), demonstrates the Middle Pleistocene age of the main Göllüdağ tuff (ca 1.3 Ma) and con-firms the Middle to Upper Pleistocene age of the main Acıgöl tuff. Our studies, thus, produce a regional tephrostratigraphic reference, which may be used for long-distance correlation. Besides, the tephra studied and analysed also form key layers used here for geomorphologic reconstruction and dating of stratigraphic and archaeological sequences.

Keywords Tephrostratigraphy · Rhyolitic complexes · Central anatolian volcanic province (CAVP)

Introduction

Establishing a confident stratigraphy of pyroclastic rocks on a regional scale requires that the formations erupted by volcanic complexes be characterised by a specific signature

which can mainly be defined in terms of major element composition of glass shards and mineral content. Additional techniques, such as analyses of trace element composition and statistical treatments of all chemical elements, are also available for discriminating distinct eruptions emitted by a given volcanic complex. Additional indicators also contrib-ute to discrimination: shapes of glass shards, morphology of pumice grains, mineral paragenesis and heavy mineral content.

Tephrochronology also requires sets of dates (whatever the dating method applied) associated with the mineral and geochemical characterization of the tephra layers. Tephras are then potential marker horizons for long-distance corre-lation. As an example, when applied to distal tephra layers, the results allow the chronological control of marine cores. In this domain and according to recent reviews of tephra layers in the Mediterranean region (Hamann et al. 2010; Zanchetta et al. 2011; Satow et al. 2015), the lack of knowl-edge about the pyroclastites of the CAVP is crucially obvi-ous as it generates many unsupported hypotheses about the origin of several Late Pleistocene tephras from the Eastern Mediterranean region.

Electronic supplementary material The online version of this article (https ://doi.org/10.1007/s4299 0-019-00010 -8) contains supplementary material, which is available to authorized users.

* D. Mouralis damase.mouralis@univ-rouen.fr

1 Laboratoire IDEES (CNRS, UMR 6266), Université de Rouen-Normandie, 7 Rue Thomas Becket, 76821 Mont-Saint-Aignan cedex, France

2 Laboratoire de Géographie Physique, Environnements Quaternaires et Actuels (UMR 8591, CNRS & Univ. Paris 1), 1 place Aristide Briand, 92195 Meudon, France

3 Maden Tetkik ve Arama (MTA) Genel Müdürlüğü, Ankara, Turkey

4 Laboratoire des Sciences du Climat et de l’Environnement/IPSL, CEA-CNRS-UVSQ, Gif sur Yvette, France

180 D. Mouralis et al.

1 3

Indeed, Pleistocene and Holocene tephrochronological investigations are still scarce in Anatolia. Apart from the Minoan eruption (Thera-Santorini) which is now confi-dently identified in several lake sequences of western Ana-tolia (Bottema and Woldring 1984; Eastwood et al. 1999; Roberts et al. 1997; Sullivan 1990, 1988), only a few other Upper Pleistocene and Holocene tephras have been reported in sequences of the Western Anatolia (Platevoet et al. 2007; Kazancı et al. 2012), Tuz Gölü (Kashima 2002), Konya plain and Cappadocia (Inoue and Saito 1997; Kashima 2002; Kuzucuoğlu et al. 1998, 1997). Data on mineral and glass composition from these Anatolian tephras and pyroclastites are even rarer with some data available in (Kuzucuoğlu et al. 1997, 1998; Kürkçüoğlu et al. 1998; Pastre et al. 1998; Karabıyıoğlu et al. 1999).

The importance of Central Anatolian pyroclastic depos-its related to the Göllüdağ and Acıgöl rhyolitic complexes also arises from increasingly challenging archaeological topics. In the Göllüdağ massif, recent surveys, excavations, and geoarchaeological investigations focused on prehistoric workshops associated with obsidian outcrops (Balkan-Atlı and Binder 2000; Balkan-Atlı et al. 2011; Kuhn et al. 2015). A significant amount of geochemical analyses for both arte-facts and raw material in and around these ryholitic complex is now available (Binder et al. 2011; Poidevin 1998). So far, the pyroclastites associated with obsidian and those present in excavations have not been studied, with the exception of the Palaeolithic “Kaletepe Deresi 3” excavation (Mouralis et al. 2002; Mouralis 2003; Slimak et al. 2004, 2008; Tryon et al. 2009).

Thus, tephrostratigraphy contributes to the dating of human occupation of the region, and to the reconstruction of the morphological and palaeoenvironmental evolution of Cappadocia during the Middle and Late Pleistocene. Con-sequently, this paper aims at providing a detailed regional tephrostratigraphy based on the identification and char-acterization of the tephra layers related to the Quaternary rhyolithic complexes in Cappadocia, i.e., the Göllüdağ mas-sif and the Acıgöl (Nevşehir) area. Apart from challenging the Eastern Mediterranean tephrochronology, the detailed analyses of the pyroclastites erupted by these rhyolitic vol-canoes address three main palaeoenvironmental issues. (1) The large volumes of tephra emitted buried palaeotopog-raphy and palaeomorphology. (2) The obsidian erupted by these volcanoes was quarried and chopped by prehistoric people, while some tephra layers also bury human artefacts. (3) The volcanic activity covers a wide time range: from the Lower to the Middle Pleistocene in the Göllüdağ com-plex, and from Middle to Upper Pleistocene in the Acıgöl area where Holocene activity is also reported (Kuzucuoğlu et al. 1998). With regard to the necessity of completing our knowledge about these three related topics, our paper intends to produce a detailed tephrostratigraphy and chronology of

both the Göllüdağ and the Acıgöl rhyolitic complexes. It is partially based on a project associating CNRS and Tübitak whose results have been reported in Türkecan et al. (2004), also including new stratigraphic observations as well as new chemical analyses and interpretations.

Regional geological setting

The Göllüdağ and Acıgöl rhyolitic complexes are located south of Cappadocia (Fig. 1). They belong to the Central Anatolian Volcanic Province (CAVP) where volcanic activ-ity began during the Miocene and remained active until the Holocene. Recent papers by Aydar et al. (2012, 2014), Di Giuseppe et al. (2018), Raybarol et al. (2019) or Türkecan (2015) review the dating of the CAVP.

Because of the considerable volume of magma emitted during this long period, the pre-volcanic substratum outcrops very rarely and plays a minor role in the regional morphol-ogy (Fig. 2). The Neogene volcanoes comprise two main types: (1) Andesitic composite volcanoes at the north (Erdaş Dağ), south (Melendiz mountains), and west (Şahinkalesi Tepe) of today’s Göllüdağ; (2) Thick Cappadocian ignim-brites, studied by several geologists after the pioneer work of Pasquarè (1966, 1968) and Innocenti et al. (1976), Besang et al. (1977) and followed by Temel et al. (1998), Le Pen-nec et al. (1994, 2005) and Aydar et al. (2012). The lack of morphological evidence for volcano-tectonic depressions associated with the eruption of these ignimbrites has led to the development of various hypotheses regarding hidden calderas in the region. Sedimentologic features (Le Pennec et al. 1994, 2005) as well as gravimetric data (Froger et al. 1998) indicate two “hidden” calderas: south of Nevşehir (in the area of today’s Acıgöl rhyolitic complex) and west of Derinkuyu village (in the area of today’s Göllüdağ rhyolitic complex).

During the Quaternary, volcanic activity produced two composite volcanoes (Hasandağ and Erciyes), numerous monogenic vents (Doğan-Kulahcı et al. 2018), and two rhy-olitic complexes (Fig. 2): Acıgöl to the north and Göllüdağ to the south (Mouralis et al. 2019, 2002) which are separated by the Erdaş Dağ andesitic massif respectively. Both com-plexes comprise voluminous pyroclastic deposits and domes extruded inside the main structures.

While the Göllüdağ complex has been poorly studied except for obsidian geochemistry, the Acıgöl complex was first described and identified as a caldera by (Yıldırım and Özgür 1981). Later, Druitt et al. (1995) studied its pyroclas-tic products, proposing a first stratigraphy. Kazancı et al. (1995) described the surges of the Eski Acıgöl maar located inside the complex. Kuzucuoğlu et al. (1998) identified seven tephra layers interstratified within the lake sediments of this maar; the 14C ages of these tephra layers range from

181Tephrostratigraphy and chronology of the Quaternary Gölludağ and Acıgöl volcanic complexes…

1 3

14 to 7 ka cal. BP (Kuzucuoğlu et al. 1998; Roberts et al. 2001).

Ages of both rhyolitic complexes have been obtained on obsidian by (mainly) Bigazzi et al. (1993, 1997) and Chataigner et al. (1998), dating the eruptive activity of Göllüdağ volcanoes Middle Pleistocene, and that of Acıgöl Middle to Upper Pleistocene. The stratigraphically subdi-vided Lower and Upper Acıgöl Tuffs (Druitt et al., 1995) were subsequently dated using U-Th and (U-Th)/He geo-chronology to ca. 190 and 160 ka, respectively (Schmitt et al. 2011, partially revised in Atici et al. 2019, this vol-ume). Schmitt et al. (2011) also confirmed the 26–20 ka age of the rhyolitic domes which formed after the main eruption. Radiometric data compiled here not only confirm but also refine this chronology.

Methods

The tephrostratigraphy and tephrochronology proposed in this paper are based on: (1) the systematic study of the stra-tigraphy within and around the Göllüdağ and Acıgöl Quater-nary complexes, combined with (2) establishing a regional

tephra reference set based on the characterization of tephra layers (using petrography, glass chemistry and minerals association) and multivariate statistical analyses; (3) radio-metric datingto constrain the chronology.

Stratigraphical correlations and characterization of tephra layers

The correlations between tephra layers are first based on field evidence: depositional facies and long-distance facies change; pumice morphologies, mineral paragenesis and texture, as well as petrographic compositions of lithics. Additional analyses performed in the laboratory completed the identification of tephra(s) through determination of the mineral associations, and glass geochemistry characteriza-tion (electron microprobe).

Mineral assemblages of Göllüdağ and Hasandağ sam-ples were determined after a first separation using heavy liquid (bromoform; density ≈ 2.89). Extracted heavy miner-als (density > 2.89) were mounted in Canada balsam and counted under a mineralogy microscope. For each sample, counting reached a total of 300 grains, the uncertainty being then lower than 5% (Parfenoff et al. 1970). Figure 3 shows

Fig. 1 Location map showing the studied area and the other Quaternary volcanoes of the CAVP

182 D. Mouralis et al.

1 3

Fig. 2 Geological sketch map of Acıgöl and Gölludağ volcanic com-plexes. 1. Non-volcanic basement. 2. Neogene volcanic rocks (ignim-brites and lava flows). 3. Quaternary mafic to intermediary lava flows (basalt and andesite). 4. Göllüdağ main tuff (GMT). 5. Acıgöl main tuff (AMT). 6. Main obsidian outcrop. 7. Alluvium deposit. 8. Sco-

ria cone. 9. Maar crater. 10. Dome. 11. Sections. 12. Samples Dated. 13. Villages and settlements. 14. Hypothetic limits of the volcano-tectonic structures. Ages refers to the data published in this article (Table 2), except CKUZ-97/8-01, CKUZ-97/8-02 and CKUZ-97/8-04 presented in Kuzucuoğlu et al. (this volume)

183Tephrostratigraphy and chronology of the Quaternary Gölludağ and Acıgöl volcanic complexes…

1 3

that the tephra layers erupted by the Hasandağ volcano are characterised by a high amphibole amount (40–60% of the counted minerals), whereas the tephra layers emitted by the Göllüdağ are ortho- and clinopyroxene rich. On the other hand, pyroclastites erupted by the Acıgöl complex are gen-erally aphyric.

Single grain geochemical analyses (WDS). 48 samples were analysed through 460 single glass shards electron probe microanalyses (EPMA) using a Cameca SX100 microprobe at the “Magmas et Volcans” Laboratory (UMR 6524, Cler-mont-Ferrand), and in the “Camparis” Laboratory (Sor-bonne University, Paris VI University, Paris). Nine major element contents were measured, according to a conven-tional protocol (Frogatt 1992; Lowe 2011): after mounting shards in epoxy and careful polishing, microprobe analyses were performed with a beam intensity of ~ 6–8 nA, with 15 kV accelerating potential. When analysing glass shards,

the beam was defocused to 10 μm to limit the volatilization of alkaline elements. At least 5–10 different glass shards have been analysed for each sample. Results are expressed in mass percent of each oxide analysed.

We already published a small set of the data obtained in Tryon et al. 2009 (347 analyses). New sampling and analyses presented here improve this initial data set by including more tephra layers from Acıgöl and Göllüdağ as well as some geo-chemical data from Karadağ volcano (a composite volcano located in the Konya plain, near the well-known Çatalhöyük Neolithic site).

Regional tephra reference data set and multivariate analyses

To establish a reference data set, we sampled systemati-cally the rhyolitic tephra layers from both the Göllüdağ and

Fig. 3 Paragenesis of selected tephras layers from the Göllüdağ complex

184 D. Mouralis et al.

1 3

Acıgöl complexes. For comparisons, we also sampled tephra layers from the nearby composite Quaternary volcanoes of the CAVP (Hasandağ, Erciyes Dağ and Karadağ).

Our reference data set is structured according to the scale of observation and sampling. We distinguish four different levels; the structure of the reference data set used in this study is presented in complementary data 1. The first (larger) level (1) corresponds to six “groups of tephra layers”. Each group presents similar characteristics (geochemical and mineralogical) and is formed by tephra layers from a single volcano. Resolution increases at level (2) where 19 tephra layers were distinguished on the basis of inter-comparison of field sections (stratigraphic logs in Figs. 8 and 9) and stratigraphic position. Level (3) corresponds to the whole set of 48 samples collected in the field, and level (4) groups the chemical content results of 460 single grain analyses. Table 1 shows the mean and standard deviation of the geo-chemical analyses for level (1) (“groups of tephra layer”) whereas complementary data 2 gives the mean and standard deviation for each of the 48 samples (level 3).

To reduce and graphically interpret the variations observed within this dataset, we use multivariate analyses based on discriminant function analyses (DFA). It is one of the most powerful methods, allowing to test statistically the strength of our classification based on the stratigraphic evidences and to assign any distal tephra of unknown origin to one of the regional Quaternary volcanoes.

DFA are computed using Systat 12.02 for Windows. Data from all nine major elements analysed are treated through a classical discriminant analysis. Prior probabilities are weighted proportionally to group size differences within our samples. Data used in the model are not normalised because neither normalisation nor log ratio transformation showed improvement of the results discrimination.

Radiometric ages

Twenty-two K/Ar dates have been obtained at the LSCE (UMR 8212). All samples were dated using the unspiked K–Ar technique described by Charbit et al. (1998). Both K and Ar measurements were performed on the ground-mass of lava samples, since the groundmass is considered to be representative of the phase that crystallised during the solidification of the lava. After crushing, sieving to the 250–125 mm fraction size, samples were washed in ultra-sonically bath of acetic acid (1 N) during 45 min at a tem-perature of 60 °C, to remove any secondary mineral phases. Phenocrysts and xenocrysts, which may carry excess 40Ar, were then removed using heavy liquids of appropriate den-sities, and magnetic separations to produce a clean ground-mass separate.

Potassium was analysed by atomic absorption and flame emission spectrophotometry. Argon was extracted from 1 to 2.6 g of groundmass sample by radio frequency heating

Table 1 Summary data of major elements characterization using electron microprobe

Mean (first line, in bold) and standard deviation for each group of tephra layer. Data are normalised to 100%. In the last column, the analytical total (not normalised) indicates the quality of the analyses. The complete data for each tephra layer are indicated in Complementary data 2Analytical conditions: Electron probe micro analyses (EPMA) using a Cameca SX100 microprobe at the “Magmas et Volcans” Laboratory (UMR 6524, Clermont-Ferrand), and in the “Camparis” Laboratory (Paris VII University, Paris). Beam intensity of 6–8 nA, with 15-kV accelerating potential. Beam was defo-cused to 10 μm for glass shardsa Number of individual analysesb Total Fe expressed as FeO

Groups of tephra layers

nbra K2O CaO TiO2 MnO FeOb Na2O MgO Al2O3 SiO2 Anlytical total

GDT 128 4.833 0.439 0.087 0.083 0.735 3.299 0.041 12.631 77.851 94.5690.517 0.089 0.048 0.073 0.110 0.453 0.027 0.200 0.371 1.398

AMT 168 4.574 0.815 0.093 0.060 1.180 3.898 0.093 13.519 75.768 95.8920.255 0.072 0.052 0.049 0.152 0.376 0.033 0.295 0.516 2.063

ATT 63 4.579 0.403 0.036 0.085 0.737 4.187 0.022 12.634 77.317 97.1400.221 0.068 0.040 0.053 0.125 0.205 0.032 0.212 0.395 2.030

HSDG 64 3.858 0.849 0.185 0.045 0.908 2.410 0.177 13.277 78.292 96.7700.458 0.154 0.087 0.045 0.203 0.763 0.061 0.665 0.673 1.489

ERCI 26 3.503 1.503 0.198 0.055 1.278 2.827 0.248 13.947 76.441 95.8330.109 0.124 0.131 0.056 0.118 0.417 0.054 0.239 0.436 1.570

KARA 11 3.653 1.447 0.131 0.068 1.230 2.740 0.297 14.639 75.794 95.1020.168 0.180 0.084 0.046 0.172 0.254 0.046 0.334 0.538 1.976

185Tephrostratigraphy and chronology of the Quaternary Gölludağ and Acıgöl volcanic complexes…

1 3

induction in a high vacuum glass line, and purified with tita-nium sponge and Zr–Al getters. Argon was analysed using a 180°, 6-cm radius, 620-V accelerating potential mass spec-trometer working in a semi-static mode. Beam sizes were measured simultaneously on a double Faraday collector in sets of 100 online acquisitions with a 1-s integration time. A manometrically calibrated dose of atmospheric Ar is used to convert beam intensities into atomic abundances. A sepa-rate measurement of atmospheric argon is used to monitor the atmospheric correction. The manometric calibration is based on periodic, replicate determinations of the interna-tional dating standard HD-B1 (24.21 ± 0.32 Ma), (Fuhrmann et al. 1987; Hautmann and Lippolt 2000; Hess and Lippolt 1994) using the same procedure as described in Charbit et al. (1998).

If the unspiked K–Ar method allows to measure accu-rately small amounts of radiogenic 40Ar (Guillou et  al. 2011), it cannot check two important assumptions ruling the K–Ar clock, (1) that the initial 40Ar/36Ar ratio of the sample is of atmospheric composition and (2) that the iso-topic system remained closed since the age of crystallisation. Therefore, K–Ar ages may be affected by unresolved excess argon or argon loss, fractionated 40Ar, implying real errors higher than the analytical ones.

Morgan et al. (2009), Flude et al. (2018) and Clay et al. (2015) demonstrate that mobility of 40Ar in obsidian glass may produce erroneous results. Flude et  al. (2018) and Clay et al. (2015) observed this on hydrated and subglacial obsidians where interaction of water with the magma is an important disruptive vector of the K–Ar clock. Given the fact that the obsidians studied here emplaced subaerially, we consider that our obsidian samples did not suffer from this perturbation. Following Morgan et al. (2009) recommenda-tions, to reduce at its maximum erroneous results due to atmospheric alteration effects on obsidian and lava flows all our samples were collected as far as possible from the rims of the flows and domes. This approach has produced accu-rate and reliable unspiked K–Ar ages (Le Bourdonnec et al. 2012) based on agreement with 40Ar/39Ar ages. Age bias due to Ar-isotope fractionation, however, remains a potential source of bias.

K–Ar ages may significantly overestimate eruption ages in the case of samples containing extraneous argon that includes both excess and inherited argon (Sasco et al. 2017 and references therein). This is mostly the case with conti-nental mafic rocks containing significant amounts of xeno-liths, xenocrysts and phenocrysts. Because all our samples are free of such artefacts and because direct comparisons between the unspiked method and the 40Ar/39Ar (Guillou et al. 2004, 2011; Laj et al. 2014; Singer et al. 2008, 2009), which verifies via the age spectra and isochron formalism all the basic assumption ruling the K–Ar clock, successfully qualified the unspiked method when applied to unaltered

and free of excess 40Ar samples, we trust on the geological significance of our K–Ar results. So even if we consider our ages reliable, in the absence of direct comparison with the 40Ar/39Ar method, the analytical errors given in Table 2 may in principle be less than the real error.

Moreover, a single 14C dating was performed by accel-erated β counting at the UMR 1572 (M. Fontugne; LSCE, Gif-sur-Yvette) on a sample from a palaeosol interstratified within pyroclastites of the Acıgöl complex (complementary data 3).

Results

Main characteristics of the tephra layers

Petrographic and mineralogical definition of pumice

Pumice erupted from the Acıgöl complex is aphyric, whereas pumice from the Göllüdağ and the Hasandağ erup-tions is microphyric to porphyric with minerals < 0.5 mm and proportions of heavy minerals ranging between 0.01 and 4%. We did not perform exhaustive chemical analyses of minerals associated with each tephra, but focused on their relative content. The counting of heavy minerals within the samples collected in the Göllüdağ area reveals three main groups (Fig. 3): (1) Hasandağ pyroclastites contain a high amount of amphiboles; (2) pyroclastic deposits related to Göllüdağ associated with the paroxysmal eruption, noted TG-1 to TG-10, present a relative abundance of minerals in the following order: orthopyroxene, pyroxene, opaque min-erals, clinopyroxene and amphibole; (3) pyroclastites related to extrusion of domes in and around the Göllüdağ complex differ from other groups by the dominance of amphibole and opaque minerals-rich products.

Geochemical results

The Quaternary pyroclastic deposits from Cappadocia belong to the calc-alkaline medium potassic series repre-sented by the complete range from basalt to rhyolite. Pumice glass shards of both complexes belong to the most differenti-ated products in the CAVP, with SiO2 content above 72.5%. In the TAS diagram (Le Maitre 2002), pumice composition is located in the rhyolite field (Fig. 4). Mantle sources and magmatic evolution have already been studied (Aydar et al. 1995, 2012; Ercan et al. 1987, 1990, 1992; Notsu et al. 1995; Olanca 1994; Siebel et al. 2011); we focus here on the cor-relation of these tephra layers by using their geochemical properties.

The geochemical analyses of 48 samples of Quaternary rhyolitic tephras of the CAVP allow distinguishing six

186 D. Mouralis et al.

1 3

Tabl

e 2

K/A

r dat

ing

of la

va fl

ows i

n an

d ar

ound

Göl

lüda

ğ, A

cıgö

l vol

cano

es

Latit

ude

Long

itude

Mat

eria

lLo

calis

atio

nSa

mpl

eK

*(w

t.%)

Wei

ght m

olte

n (g

)40

Ar*

(%)

40A

r*

(10−

12

mol

es/g

)

Age

(±2s

)ka

Age

mea

n va

lue

A—

In a

nd a

roun

d G

ollu

dağ

volc

anic

com

plex

M y

ears

 38°

14′3

1.04″N

34°2

8′54

.07″

EPo

rphy

ric

rhyo

-da

cite

Wes

tern

par

t of t

he B

ozkö

y-K

ayırl

ı ro

adD

MO

/TF-

514.

333 ±

0.04

31.

0132

746

.302

47.0

446.

25 ±

0.12

6.21

± 0.

090.

4520

642

.638

46.4

716.

17 ±

0.12

 38°

18′3

.42″

N34

°30′

29.6

2″E

Bas

alt

Top

of K

ayırl

ı vill

age.

Und

er th

e M

ain

Gol

luda

ğ Tu

f.D

MO

/TF-

540.

963 ±

0.01

01.

0895

96.

399

2.77

91.

66 ±

0.04

1.71

± 0.

030.

9870

210

.320

2.92

31.

75 ±

0.04

 38°

19′7

.80″

N34

°32′

50.8

0″E

Perli

teW

este

rn p

art o

f Boz

tepe

DM

O/T

F-73

3.71

1 ± 0.

041.

1249

014

.726

6.26

00.

972 ±

0.02

0.96

6 ± 0.

014

0.85

309

15.5

516.

172

0.95

9 ± 0.

02 3

8°15′5

7.11″N

34°3

3′34

.32″

ER

hyol

iteN

orth

easte

rn sl

ope

of th

e G

öllü

dağ

DM

O/T

F-75

3.84

9 ± 0.

038

1.15

505

6.58

35.

794

0.86

8 ± 0.

018

0.85

5 ± 0.

013

0.40

880

4.95

75.

623

0.84

2 ± 0.

018

 38°

16′5

9.12″N

34°3

4′17

.30″

EO

bsid

ian

Kal

etep

eD

AM

-129

3.76

9 ± 0.

038

1.57

885

31.6

677.

143

1.09

± 0.

021.

10 ±

0.02

1.18

947

31.4

447.

292

1.12

± 0.

02 3

8°15′1

.91″

N34

°30′

39.3

6″E

Obs

idia

nW

este

rn p

art o

f Boz

tepe

DA

M-1

313.

752 ±

0.03

81.

0939

412

.917

6.94

91.

07 ±

0.02

1.08

± 0.

020.

7076

823

.067

7.13

01.

10 ±

0.02

 38°

13′4

8.03″N

34°3

0′28

.13″

ER

hyol

iteK

üçük

Göl

lüda

ğD

MO

/TF-

713.

692 ±

0.03

71.

0681

11.

487

2.75

30.

430 ±

0.00

90.

444 ±

0.00

70.

4368

21.

588

2.93

70.

459 ±

0.01

0B

—In

and

aro

und

Acı

gol v

olca

nic

com

plex

k ye

ars

 38°

35′4

8.51″N

34°3

0′12

.03″

EB

asal

tIn

alli-

Asa

ğıkı

zıl T

epe 

DM

-05

0.96

0 ± 0.

010

1.00

578

5.07

21.

051

631 ±

1561

8 ± 14

1.57

229

6.31

11.

007

605 ±

13 3

8°34′2

4.73″N

34°3

0′23

.39″

EB

asal

tY

ukar

ı Kız

ıl Te

pe (S

outh

of I

nalli

)D

AM

-120

0.73

9 ± 0.

007

1.14

992

3.03

30.

687

536 ±

1653

8 ± 12

1.08

652

3.27

60.

692

540 ±

18 3

8°32′5

3.35″N

34°3

6′7.

20″E

Obs

idia

nA

smad

ağı T

epe

DA

M- 1

243.

683

±0.

037

1.61

181

5.53

71.

034

162 ±

416

0 ± 3

1.53

257

6.25

31.

011

158 ±

4 3

8°31′4

.48″

N34

°38′

35.0

0″E

Bas

alt

Kız

ıl Te

pe (T

uluc

e). S

outh

of

Boğ

azkö

yD

M-0

21.

468 ±

0.01

51.

5990

63.

672

0.39

215

4 ± 5

154 ±

42.

5056

94.

005

0.39

015

3 ± 4

 38°

30′5

4.90″N

34°3

8′46

.76″

EB

asal

tK

ızıl 

Tep

e (N

orth

). So

uth

of

Boğ

azkö

y,D

M-0

31.

656 ±

0.01

71.

5990

64.

035

0.37

313

0 ± 4

134 ±

32.

6744

63.

617

0.39

813

8 ± 4

 38°

28′5

0.79″N

34°3

8′13

.81″

EB

asal

tÖ

zyay

la T

epe.

Top

of t

he se

ctio

nD

M-0

61.

573 ±

0.01

61.

6001

00.

978

0.28

610

5 ± 4

110 ±

51.

1484

20.

940

0.31

711

6 ± 6

 38°

28′5

0.79″N

34°3

8′13

.81″

EB

asal

tO

zyay

la T

epe.

Bot

tom

of t

he se

c-tio

nD

M-0

71.

517 ±

0.01

51.

4620

11.

103

0.28

010

6 ± 4

109 ±

41.

5243

01.

229

0.29

611

2 ± 4

 38°

32′7

.98″

N34

°38′

45.4

9″E

Obs

idia

nEa

st of

Koc

adağ

DA

M-1

263.

711 ±

0.03

71.

5368

23.

525

0.61

495

± 3

93 ±

21.

5165

73.

652

0.57

890

± 2

 38°

28′1

6.77″N

34°3

8′15

.76″

EB

asal

tK

ızıl 

Tep

e (K

ızılc

ın)

DM

O/T

F-68

1.60

2 ± 0.

016

1.30

442

0.37

90.

227

82 ±

481

± 4

0.92

328

0.40

40.

2198

79 ±

8

187Tephrostratigraphy and chronology of the Quaternary Gölludağ and Acıgöl volcanic complexes…

1 3

groups of tephra layers (Figs. 4, 5, and 6, Tables 1 and 3) which are:

1. Göllüdağ Tephra Layers (GDT), emitted by the Göllüdağ volcanic complex either during paroxysmal eruptions or during later dome extrusion (dated ca 1 Ma in Batum (1978) and Bigazzi et al. (1993, 1998).

2. Acıgöl main tuff group (AMT), represented by the Lower and Upper Acıgöl tuffs described by Druitt et al. (1995), and dated ca 206–163 ka in Schmitt et al. (2011).

3. Pyroclastic deposits associated with rhyolitic domes and maars extruded after the Acıgöl main tuff deposi-tion during the Acıgöl terminal activity phase (ATT) and which are dated ca 30–20 ka according to Bigazzi et al. (1993) and Schmitt et al. (2011).

In addition, samples collected in and around the other Quaternary composite volcanoes of the CAVP, form three additional groups of tephra layers in our data set (Figs. 1, and 4, complementary data 1, complementary data 2):

4. Hasandağ (HSDG) where recent ages of rhyolitic events range from Middle to Upper Pleistocene (younger than 35 ka according to Kuzucuoğlu et al. (1998) and Pastre et al. (1998);

5. Erciyes (ERCI);6. Karadağ in the Konya plain (KARA).

Discriminant function analyses (DFA)

The canonical variates (CV) plots reveal a clear separation between the different groups within the multivariate space defined by CV1 and CV2 axes which account for 93.2% of the total variance (Fig. 5, Tables 1 and 3). The most discri-minant oxides are, in decreasing order: CaO, Al2O3, SiO2 and Na2O (Table 3). The discriminant analysis correctly classifies the individual analyses of tephra layers within the correct group of tephra layers defined on the basis of the stratigraphic correlation in 97% cases (Table 4). Misclassi-fications within our reference set concern primarily Karadağ because of similarities with the geochemical composition of Erciyes and Hasandağ pyroclastic deposits. Some mis-classification also occurs between Göllüdağ tephras (GDT) and Acıgöl terminal tuff (ATT) groups. However, the good separation along the canonical axes and the high proportion of correctly classified samples emphasise the strength of our model allowing us to classify distal tephra layers from pri-marily unknown origin to one of the Quaternary volcanoes.

Stratigraphic correlations

In both rhyolithic complexes, we distinguish two types of deposits related to two main eruption phases: (1) a “major C

orre

latio

ns b

etw

een

elem

ent o

xide

abu

ndan

ce (i

n w

t. %

) for

the

first

two

cano

nica

l axe

s of t

he re

fere

nce

set.

Can

onic

al d

iscr

imin

ant f

unct

ions

stan

dard

ised

by

with

in v

aria

nces

Age

cal

cula

tion

is b

ased

on

Stei

ger a

nd Jä

ger (

1977

) con

stan

ts

Tabl

e 2

(con

tinue

d)

Latit

ude

Long

itude

Mat

eria

lLo

calis

atio

nSa

mpl

eK

*(w

t.%)

Wei

ght m

olte

n (g

)40

Ar*

(%)

40A

r*

(10−

12

mol

es/g

)

Age

(±2s

)ka

Age

mea

n va

lue

 38°

30′7

.36″

N34

°36′

47.0

7″E

Bas

alt

Obr

uk T

epe

DM

O/T

F-67

1.54

3 ± 0.

015

1.52

772

0.71

40.

854

32 ±

432

± 3

1.61

170.

744

0.83

931

± 4

188 D. Mouralis et al.

1 3

tuff” mainly composed of ash and pumice flows emitted during the main activity phase; and (2) other pyroclastites linked to terminal activity (mainly maar formation and dome extrusions). These major tuffs are probably linked to col-lapse of part of each complexes, although the present mor-phology in both complexes lacks an expression of a caldera with the exception of a few fault-controlled cliffs.

In the case of the Göllüdağ complex, we group all Göllüdağ tephra layers into one geochemical group (GDT): the geochemical composition of major elements does not allow differentiating both eruptive phases. In the case of the Acıgöl complex, however, it is possible to distinguish the main tuff (AMT) from the terminal activity linked to later dome extrusions (ATT).

In Figs. 7 and 9, letters “TG” label the tephra layers emitted during both Göllüdağ volcanic activity phases. These « TG » letters are followed by a number (1–13) rep-resenting the stratigraphic and chronological position of

each layer distinguished (1 = the basal one, i.e., the oldest tephra of the series). The same labelling system is used for the Acıgöl (TA-1 to TA-12) and Hasandağ (TH) pyroclastic deposits. In some cases, the pyroclastic deposits found in and around the Göllüdağ area were produced by the volcanic activity of the Acıgöl or Hasandağ volcanoes. They are, thus, named TGx-A and TGx-H, respectively. Finally, the letter β indicates the basaltic nature of some analysed tephra layers. In addition, all tephra units within the sedimentary sections studied in the field are identified by the following notation: section number/unit number (e.g., 17/2 refers to unit two within Sect. 17).

Tephrostratigraphy of the paroxysmal phase within Göllüdağ volcanic complex

In the western part of the Göllüdağ complex, old reworked pyroclastic deposits are interstratified with thick

Fig. 4 Total Alkali-Silica diagram (TAS). After Le Maitre (2002). Data are 100% normalised

189Tephrostratigraphy and chronology of the Quaternary Gölludağ and Acıgöl volcanic complexes…

1 3

fluvio-lacustrine formations composed of a succession of cross-stratified pebbles, clay material and diatomite. These clastic and lacustrine formations overlap Neogene ignim-brites (Fig. 7: sites 14, 15, 16). At site 14, the thickness of these formations is more than 28 m (Fig. 8a). Two different reworked pyroclastic deposits cover these basal terrestrial formations. According to chemical and mineralogical evi-dences, tephra layer TG-1 is attributed to Göllüdağ volcano, and TG-2H to Hasandağ.

The main Göllüdağ volcanic activity comprises post TG-1 pyroclastic flows and fall deposits (Fig. 7). The tephra layers TG-3, TG-6 and TG-7 are the thickest ones and are preserved in the western part of the complex (logs 14, 15, 16). TG-3 (15/2 and 16/7; Fig. 8c, d) is a 15-m thick, white and ashy flow, lithics poor and rich in accretionary lapilli. Its lower part is composed of surges identified by wavy features. The heavy mineral assemblage contains orthopyroxene and amphibole (25% each) associated with clinopyroxene, rare biotite and zircon.

TG-6 is a 10-m-thick, local, pyroclastic flow deposit only observed near sites 15 (unit 15/5) and 17 (unit 17/2). It is a pumice and lapilli flow containing pseudo-fibrous pumice.

Its lithic fraction is composed of perlitic rhyolite and obsid-ian, with rare weathered basaltic and andesitic blocks. The heavy mineral assemblage is composed of clinopyroxene and orthopyroxene (30% each) associated with opaque minerals and amphibole (10%).

TG-7 is ubiquitous in most sections studied in the western part of the complex (14/11, 15/7, 16/10, 18A/4 and 18B/6). This flow is characterised by abundant lithics with obsidian reaching 15 cm in size, and with rhyolite and basalt clasts. The length of the fibrous and pseudo-fibrous pumice in this flow sometimes reaches 10 cm; pumice clasts are sub-aphy-ric, with a heavy mineral content < 0.1%.

The thickness of each of these pyroclastic flow deposits decreases from the centre to the periphery of the complex, suggesting radial centrifuge flow directions. These pyroclas-tic units bury an ancient river network underlined by alluvial deposits (15/1, 15/2) under TG-3 (the oldest) pyroclastic formation. All flows seem to have been channelized toward the present Çiftlik and Derinkuyu depressions, suggesting that a lowland palaeotopography with a river network existed at the start of volcanic events.

In addition to these pyroclastic flow deposits, four pum-ice fallout deposits have been identified in the field. Pumice fallout deposit TG-4 has only been observed in the eastern part of the complex (18B/2; 20/1; 22/3; 23/4 and 30/3). Its pumice present oriented vesicles and are aphyric. The thick-ness of the deposit is always < 1 m. TG-5 (14/8; 15/3; 16/9; 18A/2 and 18B/2) shows quartz and feldspar-rich pumice presenting unorganised vesicles. The heavy mineral compo-sition associates opaque minerals, orthopyroxene, clinopy-roxene and amphibole, with rare black mica and zircon. The lithic fraction contains weathered rhyolitic lava and obsid-ian lapilli. This deposit is composed of two units: a basal 2-m-thick pumice fall deposit is overlain by a ca. 3-m-thick surge deposit within which a thin ash fall is intercalated (site 21, units 3, 4 and 5).

On the southern slopes dominating the Çiftlik plain (sites 4–7), two additional pumice fall deposits, TG-8 and TG-9, cover both TG-3 and TG-5. TG-8 (5/3, 6/2 and 7/4) comprises two pumice fall deposits containing pumice with orthopyroxene, clinopyroxene and opaque minerals. TG-9 (5/5, 6/4, 7/5) shows a different paragenesis with black micas and opaque minerals.

Pyroclastic activity related to dome extrusion during the Göllüdağ terminal phase

After the eruption of the main tuff (TG-3 to TG-9), the extru-sion of ten distinct domes ended in forming a cumulo-dome (Fig. 2) dominated by the highest one, the Büyük Göllüdağ (2172 m). The site 21 (Fig. 8e) shows fall and surge deposits associated with Büyük Göllüdağ dome. The fall deposits are bedded with inverse grading, while pumice forms ca. 80%

Fig. 5 Result of the discriminant function analyses. Individual analy-ses and mean of each group of tephra layer classified according to the statistical model showing good separation between each group

190 D. Mouralis et al.

1 3

of the clasts with size reaching 15 cm. The upper part of the section shows surge deposits with finer clasts (ash and lapilli), undulating layering and cross-bedding.

Morphostratigraphic evidence allows reconstructing the relative chronology of these extrusions, with younger

domes overlying older ones (e.g., Kabak Tepe and Büyük Göllüdağ, Fig. 2). Most of the extrusions are associated with local pyroclastic deposits.

Fig. 6 Selected major (wt %) Harker variation diagrams

Table 3 Analysis of the discriminant functions for the geochemical data set: correlations between element oxide abundance (in wt %) for the first two canonical axes of the reference set. Canonical discriminant functions standardised by within variances

K2O CaO TiO2 MnO FeO* Na2O MgO Al2O3 SiO2

Canonical variate 1 0.286 − 0.836 0.124 0.054 0.101 0.311 0.332 0.127 0.054Canonical variate 2 0.372 0.022 0.126 0.042 0.524 0.668 − 0.282 0.856 0.778

191Tephrostratigraphy and chronology of the Quaternary Gölludağ and Acıgöl volcanic complexes…

1 3

Tephra layers deposited after the end of the Göllüdağ volcanic activity

In and around the Göllüdağ complex, reworked deposits (e.g., alluvium or colluvium) are intercalated between these pyroclastic formations suggesting a hiatus between two eruption phases. For example, south of Büyük Göllüdağ, three pumice-rich and lithic-poor (rhyolitic lapilli) fall deposits are located at site 24 (unit 24/2, 24/3a

and 24/3b; Fig. 7). Units 24/2 and 24/3a are ~60-cm thick; unit 24/4b is 150-cm thick, and its summit is reworked by runoff. These three units comprise only ash and lapilli, excluding blocks; they are all heavy mineral-rich domi-nated by amphibole, a mineral paragenesis different from the typical (i.e. older) Göllüdağ pyroclastic deposits (Fig. 3). Similar tephra layers are also observed north of Büyük Göllüdağ at site 27 (samples 27/3 and 27/5), and east of Kabak Tepe at site 12. Mean and standard deviation

Table 4 Jackknifed classification matrix of volcanic sources and eruptive phases used in the linear discriminant analysis of the reference set (this paper), based on untransformed abundances (in wt. %) of all elements

The end right column lists the correct classification (expressed in %) with regard to a volcanic source defined a priori. This percentage is, thus, an estimate of the strength of the classificatory model

GDT AMT ATT HSDG ERCI KARA % Correctly clas-sified to known source

GDT 125 0 3 0 0 0 98AMT 0 167 0 1 0 0 99ATT 3 1 59 0 0 0 94HSDG 4 0 0 61 0 0 94ERCI 0 0 0 0 25 1 96KARA 0 0 0 1 2 8 73TOTAL 132 168 62 63 27 9 97

Fig. 7 Stratigraphy of the tephra layers in and around the Göllüdağ complex. “TG” labels the tephra layers emitted during Göllüdağ vol-canic activity phases. The following numbers (1–13) represent the stratigraphic order and chronological position of each layer (1 = the basal one, i.e., the oldest tephra of the series). The same labelling sys-tem is used for the Acıgöl (TA-1 to TA-12) and Hasandağ (TH) pyro-clastic units. TGx-A and TGx-H refer to pyroclastic deposits found in

and around the Göllüdağ area and erupted by the Acıgöl or Hasandağ volcanoes, respectively. The letter β indicates the basaltic nature of some analysed tephra layers. In addition, all tephra units within the sedimentary sections studied in the field are identified by the follow-ing notation: section number/unit number (e.g., 17/2 refers to unit 2 within Sect. 17)

192 D. Mouralis et al.

1 3

of the chemical composition of these distal tephra layers are given in Table 5.

To test our data set and to classify these distal tephra lay-ers of unknown origin, we have used our DFA model. The classification of the six single grain analyses from sample 24-4 and of seven analyses from sample 27-3, shows that these distal tephras pertain to the Hasandağ “Group of tephra layers” with a posterior probability of 0.9 and 1.0, respec-tively (complementary data 4).

Other tephra layers have been observed at site 35, which is a section opened during the “Kaletepe Dere 3” archaeo-logical excavation. Here, five rhyolitic tephra layers (TG-13-A: see Fig. 7, log 35, unit 2a–2e) and a younger trachytic tephra layer are interstratified with the most recent occupa-tion levels of this Palaeolithic site (Mouralis 2003; Mouralis et al. 2002; Slimak et al. 2004, 2008; Tryon et al. 2009). In previous publications, we demonstrated that the five rhy-olitic tephra layers are associated with the emission of the Acıgöl Main Tuff. The classification associated with our new

dataset confirms this initial identification (complementary data 4).

The main pyroclastites emitted by the Acıgöl volcanic complex (AMT)

As already pointed out by Druitt et al. (1995), the Acıgöl pyroclastic deposits are mainly observable in the eastern part of the Acıgöl complex (the Kumtepe hills: Fig. 2). In this area, thick pyroclastic deposits emitted during the main activity phase are overlain by deposition of alluvium and colluvium and then by younger pyroclastic deposits that were emitted during the extrusion of domes in and around the complex.

The initial activity is poorly documented. Site 51 (Figs. 9 and 10c) shows a 1.5-m-thick plinian unit (TA-1). Pumice clasts are dense with few small vesicles. A thin basaltic scoria layer (TA-2β) is intercalated between the

Fig. 8 Photography of some units in and around Göllüdağ complex. Location of the sites is given in Fig. 2. a Site 14 (1713 m), stratified alluvial and lake sediments. The upper part of the section shows TG-3 and TG-7 (Göllüdağ main tuff). Black line is ca. two metres high. b Site 15: Tilted alluvium containing well-rounded andesitic and basaltic pebbles reaching 15 cm in diameter, with no pumice. This alluvial deposit is deformed by a ca 33° eastward dip, pointing to a collapse of the centre of the volcanic complex. The col-league indicates the scale. c Site 16, 1798 m, western part of the Göllüdağ complex, ancient valley has been filled in with TG-3 and TG-7 (Göllüdağ main tuff) and then eroded. Look-ing to the east. d Site 26: TG-3 tuff, locally eroded by a gully. The hammer (low right of the picture) is shown for scale. e Site 21: fall and surge associ-ated with Büyük Göllüdağ dome. Fall deposits are bedded with inverse grading. Pumices are 80% of the clasts with size reaching 15 cm. Upper surges show finer clasts (ash and lapilli) with undulating and cross-bedding features. Black line represents one metre

193Tephrostratigraphy and chronology of the Quaternary Gölludağ and Acıgöl volcanic complexes…

1 3

lower and upper part of TA-1. At site 64, TA-1 overlays the Plio-Quaternary andesite substratum.

Site 52 (Kumtepe hills, see Fig. 10a) allows a descrip-tion of all main Acıgöl pyroclastites (TA-3 to TA-12). A synthetic log of these formations shows (1) two basal units (TA-3 and TA-4) covered by (2) a middle group formed by TA-4 to TA-7 fall, followed by (3) the youngest group exhibiting interstratified basaltic products (TA-β8 to TA-12). In site 52 located, in a proximal situation, these layers constitute a continuous record from TA-3 to TA-12. However, the inter-bedding of basaltic tephra layers (TA-β8 and TA-β10) indicates the interruption of rhyolitic accumulation before the renewing of rhyolitic eruption and the deposition of TA-11. Moreover, the major ele-ments content is identical in the rhyolitic units from TA-03 to TA-12 with, for example, a SiO2 content ranging from 75.11 to 76.27% (see complementary data 2 and Fig. 4).

The basal group is composed of a fall deposit (TA-3) covered by a pyroclastic flow deposit (TA-4). At site 52, TA-3 fallout thickness exceeds 3 m. Pumice is aphyric with distinctive tubular and fibrous facies. Rhyolite domi-nates the lithic fraction; there are no obsidian clasts. The

impressive size (20 cm) reached by the pumice points to the proximity of site 52 to the vent.

Overlying TA-3, TA-4 is composed of three pyroclastic flow units (52/2, 52/6 and 52/7), associated with a surge (52/3), an ash fallout (52/4) and lag-breccias (52/5). The flow units outcrop all around the northern and eastern parts of the complex, whereas the ash fall and surge deposits have a more local extension. At site 52, TA-4 total thickness reaches 17 m, of which 10 m are represented by flow depos-its. Pumice is aphyric, with two main facies: (1) elongated and oriented vesicles, and (2) flexuous. The lithic fraction contains obsidian, rhyolite and diabase lapilli. This deposit is matrix-supported. The ashy unit (52/3) is interpreted as a co-ignimbritic ash fall. North-eastwards at site 69 (units 2–4), the lag breccia layer disappears because of a higher distance to the vent and the total thickness of TA-4 decreases to < 10 m. At site 80 (Fig. 2), TA-4 consists only of flow units and presents a total thickness < 6 m. Furthermore, TA-4 pyroclastites seem to have filled-in river channels directed northwards and eastwards.

The following pyroclastites (TA-5 to TA-7) are charac-terised by a plinian regime. TA-5 consists of surges (52/8)

Table 5 Mean and standard deviation of some distal tephra layers

Mean (upper line, in bold) and standard deviation of some distal tephra layers. Data are normalised to 100%. In the last column, the analytical total (not normalised) indicates the quality of the analyses*Total Fe expressed as FeO References of the dataa Primary published in Mouralis, 2002 and then in Tryon et al. 2008b This study

n K2O CaO TiO2 MnO FeO* Na2O MgO Al2O3 SiO2 Analytical total

Tephra layer top of Göllüdag tuffKaletepe Tephra fall deposits—TG13A Site 35a

 35/2a (C2D9-R1) 17 4.706 0.734 0.089 0.062 0.906 4.052 0.048 13.161 76.242 95.7330.161 0.087 0.063 0.058 0.197 0.191 0.028 0.224 0.379 2.205

 35/2b (C3D4-R2) 7 4.771 0.754 0.055 0.054 0.963 3.863 0.067 13.011 76.464 95.1460.144 0.091 0.035 0.058 0.127 0.157 0.037 0.148 0.410 0.423

 35/2c (C2D11-R3) 12 4.913 0.660 0.071 0.028 0.836 3.979 0.059 12.906 76.547 97.2220.428 0.161 0.048 0.038 0.214 0.224 0.033 0.253 0.451 2.630

 35/2d (C2D12-R4) 10 4.746 0.760 0.083 0.044 0.976 3.922 0.089 13.155 76.227 94.6380.199 0.062 0.045 0.029 0.096 0.218 0.048 0.113 0.208 0.494

 35/2e (C3D8-R5) 6 4.821 0.777 0.120 0.049 1.034 4.015 0.075 13.292 75.817 94.1430.282 0.129 0.045 0.045 0.056 0.143 0.021 0.142 0.318 0.945

Other tephra layer top of Göllüdag tuffb

 24/3b (GD2_5) 6 4.061 1.040 0.264 0.041 1.148 3.934 0.226 12.727 76.559 92.5170.241 0.036 0.078 0.038 0.140 0.308 0.046 0.306 1.292 1.167

 27/3 (KOM3_48) 7 3.985 0.971 0.267 0.075 1.026 2.766 0.196 12.719 77.995 94.2340.103 0.073 0.060 0.082 0.180 0.556 0.030 0.210 0.438 1.698

Tephra layer top of Acigöl main tuff (2) 53/11 (KA_1122) 22 4.521 0.418 0.028 0.060 0.747 4.114 0.024 12.769 77.319 98.228

0.135 0.055 0.036 0.041 0.076 0.355 0.020 0.212 0.355 1.689

194 D. Mouralis et al.

1 3

and a plinian fall (52/9). Surges are 3–5-m thick, with fibrous pumice. TA-6 includes an ash and pumice flow (52/10), a pumice fall (52/11) and a thinly bedded ash fall (52/12) interpreted as a co-ignimbritic fall probably linked to unit 10. TA-7 comprises fall deposits (52/13 and 52/14); their pumice clasts display elongated vesicles. Unit 13 is massive; whereas unit 14 presents a bedded framework with 10–20-cm-thick beds.

The uppermost and youngest group of pyroclastites (TA-β8 to TA-12) presents interbedding of scoria layers and rhyolitic pyroclastites. Two scoria fall deposits (TA-β8 and TA-β10) are interstratified within TA-9 rhyolitic pyroclastites. Also, at site 57, in the southern part of the Acıgöl complex, scoria layers are interstratified with rhy-olitic pyroclastites. At the top of Sect. 52, a rhyolitic ash and pumice flow (TA-11) is overlain by lahars (TA-12).

Pyroclastic deposits associated with monogenic vents in and around Acıgöl complex

Most of the domes extruded in and around the Acıgöl com-plex are associated with pyroclastic products, especially with phreatomagmatic tephras. Four domes (Susamsivrisi, Kuzey, Asmadağı and Kocadağ) (Fig. 2) are partially overlain by TA-4 ash and pumice flow. This observation demonstrates that these domes are older than the AMT (i.e., “Acıgöl main tuff”).

On the other hand, many vents located inside the Acıgöl complex were active after the AMT. Three of them are domes extruded in maars (Güneydağ, Kaleci Tepe and Korudağ), one is a maar without a dome (Eski Acıgöl), and the other ones are scoria cones (e.g., Obruk Tepe). Sections in the three rhyolitic domes show an initial

Fig. 9 Stratigraphy of the tephra layers in and around the Acıgöl complex. “TA” labels the tephra layers emitted during Acıgöl vol-canic activity phases. The following number (1–12) represent the stratigraphic order and chronological position of each layer (1 = the basal one, i.e., the oldest tephra of the series). The letter β indicates

the basaltic nature of some analysed tephra layers. In addition, all tephra units within the sedimentary sections studied in the field are identified by the following notation: section number/unit number (e.g., 17/2 refers to unit 2 within Sect. 17)

195Tephrostratigraphy and chronology of the Quaternary Gölludağ and Acıgöl volcanic complexes…

1 3

phreatomagmatic activity preceding the extrusion of the dome (Fig. 10d). Domes and scoria cones have emitted pyroclastic products of local extension (i.e., found near the vents and in a few sections only). These pyroclastic depos-its, thus, do not play any significant role in the regional tephrostratigraphy, although in some cases they are good chronological markers of the local Late Pleistocene/Holo-cene landscape evolution.

For example, at the bottom of section site 53, the Acıgöl Main Tuff is overlain by three scoria falls and colluvium. Unit 53/11, a 20–40-cm-thick ash and lapilli-sized pum-ice fall is overlain by a palaeosol (53/12) the age of which is 29.9 ± 1 ka BP (Complementary data 3). The compari-son between the geochemistry of this tephra layer and our regional referential database indicates that it has been emit-ted during the Acıgöl terminal activity. Complementary data 4 show the results of the classification using the DFA: 20 of the 22 single grain analyses are classified into ATT group of tephra layers, with a posterior probability of 0.9. However, the chemical analyses of major elements do not allow to specify the emitting vent.

Chronology of the volcanic activity of the Göllüdağ and Acıgöl complexes

To constrain the chronology of both complexes 17 K/Ar dates have been obtained from volcanic products (lava flows and extrusions) (Table 2). In addition, a single 14C date (complementary data 3) has been obtained from a palaeosol.

Radiometric ages in and around the Göllüdağ complex

In our dataset, two samples pre-date the volcanic activity of the Göllüdağ complex: DMO/TF51 (porphyric rhyodacite flow) and DMO/TF54 (basalt flow). Both flows were sam-pled in the western part of the complex, west of Bozköy-Kayırlı road (DMO/TF54) and above Kayırlı village (DMO/TF51). Both flows are covered by the Göllüdağ main tuff (TG-7). Their respective K/Ar age is 6.21 ± 0.09 (TF51) and 1.71 ± 0.03 Ma (TF54). In addition, five post-caldera domes have been K/Ar dated. Four of them yielded ages ranging from 1.1 ± 0.02 Ma to 0.855 ± 0.013 Ma, whereas

Fig. 10 Photography of selected units in and around Acıgöl complex. Location of the sites is indicated in Fig. 2. a Site 52 (Kumtepe hills), 1447 m. Quarry showing the different units associated with main Acıgöl tuff. b Site 52, section showing unit TA-4 to TA-12. c Site 51, TA-1. d Site 72 (Güneydağ): surges associ-ated to the maar, before dome extrusion

196 D. Mouralis et al.

1 3

the fifth one (Küçük Göllüdağ) has a younger K/Ar age of 0.444 ± 0.007 Ma.

Radiometric ages in and around Acıgöl complex

In the northern part of the Acıgöl complex, two scoria cones, the Asağı and the Yukarı Kızıl Tepe, have been K/Ar dated at 618 ± 14 ka (DM-05) and 538 ± 12 ka (DM-120), respec-tively. The flanks of these cones being covered by the AMT, the lava from both these cones predate the AMT.

The obsidian related to the setting of the Asmadağı Tepe dome, and covered by TA-4 pyroclastic deposit, yielded an age of 160 ± 3  ka (DAM-124). At site 56, obsidian intruded into the TA-4 deposit has been dated at 93 ± 2 ka (DAM-126).

Inside the Acıgöl complex, two scoria cones (Kızıl Tepe North and South), respectively, 154 ± 4 and 134 ± 3 ka old (DM-02 and DM-03), appear to be older than the AMT pyroclastites that overlie them. A third scoria cone (Obruk Tepe), dated to the late Pleistocene (32 ± 3 ka: DAM-126), produced basaltic ashes which, at site 55, are interstratified with rhyolitic tephra layers erupted from Korudağ (Acıgöl Terminal Activity).

In the southern part of the complex, two other scoria falls include > 1 m large bombs. These falls were emitted by the nearby Özyayla Tepe. They are interbedded with TA-9 tephra layer. The bottom scoria layer was K/Ar dated at 109 ± 4 ka (DM 07), whereas the top scoria layer has been K/Ar dated at 110 ± 5 ka (DM 06), thus dating TA-9 tephra emission. South of Özyayla Tepe, Küzlük Tepe was dated at 81 ± 4 ka (DMO/TF68).

At site 53 (Figs. 2 and 7), a palaeosol, overlain by a thin in situ pumice layer, has been 14C dated 29.950 ± 1.080 uncal. BP yrs (GIF-11,550: M. Fontugne, LSCE; com-plementary data 3). The age of these uppermost rhyolitic tephras (equal to that of the dome which produced them) is, thus, the youngest age in our tephrochronology dataset.

Discussion

Stratigraphy of the tephra layers

Around the Göllüdağ complex, stratigraphic observation supports the assumption of three distinct eruptive stages. During the initial activity (1), a thin tephra layer (TG-1) was deposited in the valleys located in the western part of the complex. At the same time, another tephra layer was emit-ted from the Hasandağ (TG-2H). Both these pyroclastites were reworked by running water. During the paroxysmal stage (2), the main tuff was emitted: it is characterised by the deposition of tephra layers TG-3 to TG9, infilling and preserving paleo-valleys running southwards to the Çiftlik

plain and eastwards to northwards to the Derinkuyu plain. During the final stage (3), domes were extruded within the complex. The tephra layers related to this late volcanic activ-ity had only a local dispersion. A fourth group of pyroclas-tites identified in the Göllüdağ complex (4) corresponds to the tephra layers emitted by the neighbouring Acıgöl com-plex (TG13A, TG14A) and Hasandağ volcano (TG-11H, TG12H). All these late tephra layers are separated from the Göllüdağ pyroclastic units by alluvial or colluvial deposits that indicate a time-gap between the end of the Göllüdağ activity and the deposition of these younger tephra layers derived from external sources.

In the Acıgöl complex, the stratigraphic reconstruction (Fig. 9) shows the following succession:

1. From TA-3 to TA-7, the absence of alluvium, colluvium or palaeosol points to the continuity of deposition. These TA-3 to TA-7 tephra layers form the Lower Acıgöl Tuff (LAT) defined by Druitt et al. (1995).

2. At the top of TA-7, interruption of the volcanic activity and related pyroclastic deposition are evidenced by: (1) interstratified basaltic scoria layers; (2) unconformity of the basaltic scoria layer TA-β10 and of young TA-11 and TA-12 rhyolitic units above the TA-3 to TA-9 units in Sect. 61. These TA-11 to TA-12 pyroclastic deposits form the Upper Acıgöl Tuff (UAT) defined by Druitt et al. (1995).

It is to be noticed that the sections we have observed in proximal situations do not show any palaeosol, in contrast to the description by Druitt et al. (1995) of a soil separating LAT and UAT. In sites 51, 57, 61 and 62, scoria layers are intercalated in-between the initial units TA-3 to TA-7 (cor-responding to LAT) and TA-11 and TA-12 (corresponding to UAT) without any palaeosol. In a proximal situation, the volcanic activity of strombolian cones succeeded in insert-ing scoria layers in-between rhyolitic tephra layers (TA-3 to TA-7 and TA-11 to TA-12).

Chronology of both complexes

The chronology of both complexes is constrained here by 18 radiometric dates.

Chronology of Göllüdağ complex

In the north-western part of the complex, the substratum (Miocene, Pliocene and Early Pleistocene volcanics) has been locally dated at 1.71 Ma by DMO/TF-54. Other K/Ar dates obtained from volcanics in the Göllüdağ are from dome lavas extruded inside the volcanic complex after the eruption of the main Göllüdağ tuff. The oldest of these dated domes is the Kaletepe dome (ca 1.1 Ma). Accordingly, the

197Tephrostratigraphy and chronology of the Quaternary Gölludağ and Acıgöl volcanic complexes…

1 3

main Göllüdağ tuff was emitted between 1.7 and 1.1 Ma. These data are in good agreement with the chronology pub-lished by Aydın et al. (2014): using U–Pb in zircon, these authors indicates a 1.083 Ma age on the southern slope of the Kaletepe dome, while we obtain a 1.1 Ma (K/Ar, DAM-129) on the south-eastern slope of the same dome.

In addition to the radiometric chronology, our field results refine the relative chronology of the extrusion of several domes. The palaeo-Kabak Tepe (dated 1.1 Ma), possibly one of the oldest domes, is cut and filled by vol-canics from the younger Kabak Tepe and Kayışkıran Tepe domes. Kayışkıran Tepe is in turn cut-and-filled by volcanics from (1) the Bozdağ dome (0.966 Ma) and (2) the palaeo-Büyük Göllüdağ. Palaeo-Büyük Göllüdağ is in turn also partially destroyed by the younger Büyük Göllüdağ dome (0.855 Ma). This age (DMO/TF-75) is consistent with the one (0.899 Ma) published by Aydın et al. (2014) on pum-ice collected at site 21, associated with the extrusion of the Büyük Göllüdağ dome.

Finally, the youngest K/Ar age (0.4 Ma) was obtained from a rhyolite lava forming the basal cliff of the Küçük Göllüdağ. This cliff and other morphological characteristics point to this dome being the youngest of all domes inside the complex.

Chronology of the Acıgöl complex

Our tephrostratigraphy refines the chronology of the Acıgöl complex volcanic activity as defined by Schmitt et  al. (2011). On the summit of the Asmadağı Tepe, TA-4 covers an obsidian K/Ar dated 160 ± 3 ka (DAM-124). The main Acıgöl pyroclastic units are thus younger than 160 ± 3 ka. The Acıgöl Main Tuff (AMT) is divided into two sub-groups as already noticed by Druitt et al. (1995) who distinguish the Lower Acıgöl Tuff (LAT) from the Upper Acıgöl Tuff (UAT). The earliest one corresponds to our TA-1 to TA-7 tephra layers ending with basaltic deposits (TA-β8 and TA-β10: see Sects. 52, 61, 90, 91, 92 and 93). At site 57, two basaltic bombs emitted by the Özyayla Tepe strombo-lian cone have been dated at 110 ka (DM-07 and DM-06). Our results also show that the AMT is contemporaneous of the activity of two small basaltic cones K/Ar dated 154 ± 4 (DM-02) and 134 ± 5 ka (DM-03). The TA-3 to TA-7 Acıgöl activity, thus, extends between 160 and 110 ka. Immediately after 110 ka, two younger rhyolitic tephra layers (TA-11 and TA-12) were emitted.

Subsequently, the volcanic activity persisted with the extrusion, within the volcanic complex, of the Kocadağ dome which is accompanied by an obsidian-filled dyke dated 93 ± 2 ka (DAM-126). This dyke intruded into the TA-4 pyroclastic deposit which, thus, predates de dyke.

After ~ 90 ka, an activity lag is indicated by a palaeo-sol and some colluvial deposits (site 53, 62 and 55). The

volcanic activity started again with eruptions of (1) a basal-tic scoria cone (Obruk Tepe dated 32 ± 3 ka K/Ar, DMO/TF-67), and (2) four rhyolitic maars and domes (Korudağ, Güneydağ, Eski-Acıgöl maar, Kaleci Tepe) all dated at 20 ka (Bigazzi et al. 1993; Schmitt et al. 2011). Our results provide also a more detailed, although relative, chronology of the activity of these three youngest vents of the complex. In the NW part of the complex, the Güneydağ dome extruded first into an initial maar. This dome was partly destructed during the eruption of the Eski-Acıgöl maar which cuts the older structures. Finally, Kaleci Tepe dome was extruded in an initial maar, the ring of which is today well preserved around the dome. During this final event, Kaleci Tepe tephra layers blanketed all previous volcanoes and their products. This rel-ative chronology is in agreement with Schmitt et al. (2011) who dated these vents (using U/Th and U-Th/He geochronol-ogy) 23.8 ± 0.9 ka (Güneydağ), 23.2 ± 3 ka (Eski Acigöl) and 20.3 ± 0.6 ka (Kaleci Tepe). Noteworthy is a 28.3 ka uncal BP 14C date from a frost-deformed soil associated with a pure obsidian fall in the Göçü quarry NW of the Konya plain (Karabıyıoğlu et al. 1999; Kuzucuoğlu et al. 1998).

Quaternary volcano‑tectonic structures

According to our field observation, the pyroclastic deposits assigned to GDT and AMT groups cover areas of ~ 720 and 1.100 km2 respectively (Fig. 2). In the case of Acıgöl, the value we report is close to that proposed by Druitt et al. (1995). Such a high amount of pyroclastic deposits produced by rhyolitic complexes deserves discussion about possible caldera structures.

In the Göllüdağ complex, several indicators suggest a col-lapsed volcano structure.

1. The Göllüdağ complex is located in a depression posi-tioned at the foot of cliffs cut-into a Mio-Pliocene sub-stratum (the Şahinkalesi Tepe to the west; the Melendiz mountains, to the south-east). (Fig. 2).

2. Elevation values of the substratum in comparison to the pyroclastic layers also suggest a collapse of the central part of the complex. In the western part (Sect. 10), the substratum outcrops at 1760 m; whereas, in Sect. 16 it outcrops at 1660 m. In the southern part of the com-plex in Sects. 22 and 23, the substratum buried by the pyroclastites is situated at 1480 m elevation, while the top of the pyroclastites reaches only 1560 m. Thus, the elevation of the substratum–pyroclastite contact presents a 100–300-m altitude difference between the central part of the complex and in its periphery.

3. Moreover, in Sect. 15 (Fig. 8b), early alluvium contains well-rounded andesitic and basaltic pebbles reaching 15 cm in diameter, with no pumice. This alluvial deposit is tilted with a ~ 33° eastward dip, pointing to a collapse

198 D. Mouralis et al.

1 3

of the centre of the volcanic complex. Most units in the section also dip in the same direction but weaker (15°), suggesting a decrease of the subsidence rate with time.

However, field observation and careful mapping did not allow the precise identification of the limits of the collapsed centre and its surrounding structures.

In the case of the Acıgöl complex, the situation is more complex. Yıldırım and Özgür (1981) first identified and described a caldera. According to these authors, the size of the collapsed structure is 6 × 10 km, its southern limit being the Erdaş Dağ, and its eastern limit being the obsidian “wall” (cliff) east of the Kocadağ. Druitt et al. (1995) disagreed with this interpretation citing that topographic steps in these locations are not clearly identifiable.

We argue for the presence of a caldera based on four con-siderations (e.g. Ferrari et al., 1991): (1) the outcrops of the pre-caldera substratum are located outside the structure and never in its centre; (2) pyroclastic flows are associated with the structure without occurrence of any other alterna-tive volcanic source; (3) domes extruded in the centre and around the collapse structure, leave untouched some mor-phological parts of the collapse fault scarps; (4) the volumes of both the emitted pyroclastic deposits and the collapsed part of the structure present a similar magnitude. Whereas, the volume of the collapse structures remains unconstrained due to their uncertain boundaries, the first three criteria of Ferrari et al. (1991) are fulfilled for both, the Göllüdağ and Acıgöl complexes.

Relationship with Tertiary calderas

It is noticeable that the studied complexes occur near the inferred location of Neogene collapse structures that must have accompanied the emission of the Cappadocian ignim-brites during the Mio-Pliocene. According to Froger et al. (1998), the Acıgöl-Nevşehir and Derinkuyu areas host sig-nificant negative Bouguer anomalies, usually attributed to calderas and acidic eruptions. Using Le Pennec et al. (1994)’s reconstructions of ignimbrite flow directions, Froger et al. (1998) showed that the “Kavak” and “Zelve” Neogene ignimbrites came from an area south of Nevşehir, where gravimetric data point to the presence of a caldera; whereas, the “Sarımaden”, “Cemilköy”, “Gördeles” and “Kızılkaya” Neogene ignimbrites were emitted by a caldera located in the Derinkuyu area.

Our results (Fig. 11) show that the Göllüdağ Quaternary complex is located inside the “Derinkuyu” Neogene cal-dera, whereas part of Acıgöl caldera is located inside the “Acıgöl–Nevşehir” Neogene caldera as defined by Froger et al. (1998). According to these authors, measured negative Bouguer anomalies result from the accumulation of pyro-clastites (Neogene) in the caldera areas, as the density of

pyroclastic units is lower than that of the substratum (Neo-gene andesites and pre-volcanic substratum). Our results stress two important facts supporting this interpretation: (1) the negative Bouguer anomalies in Cappadocia are also due to the presence of younger pyroclastic accumulation related to structures which collapsed during the Quaternary; (ii) the morphologies related to the Neogene calderas are now invis-ible for two main reasons: erosion and volcanic activity dur-ing the Quaternary. Indeed, the Early and Late Pleistocene volcanic activity emplaced two rhyolitic complexes which destroyed and buried older morphologies.

Conclusion

The volcanic complexes analysed in the present paper have erupted voluminous pyroclastic deposits whose stratigraphy, petrographic and chemical characterization and dating are presented here for the first time altogether. Eruptive activ-ity occurred around 1.3 Ma in the Göllüdağ complex, and around 150 ka in the Acıgöl complex. The large amount of tephras erupted during each of these major events buried and destroyed the previous morphology, largely modifying the regional landscape. As this volcanic activity probably re-activated Neogene structures, the morphology of the Neogene calderas became indistinguishable in the present landscape.

Intensive and systematic field observation allowed us to reconstruct the detailed regional tephro- and chronostratig-raphy we present here. The robustness of the analysis of dis-criminant functions differentiating our geochemical data is well underlined by the obvious separation of the data within the multivariate space, as well as by the high proportion of correctly classified data.

It is noticeable that human populations were affected by the Middle to Upper Pleistocene Acıgöl eruption. For exam-ple, in the Palaeolithic “Kaletepe Dere 3” excavation, several tephra layers buried human artefacts (Mouralis et al. 2002, Mouralis 2003; Slimak et al. 2004, 2008; Tryon et al. 2009). In these archaeology-oriented publications, we demonstrated the capacity of our regional tephra reference to provide a chronological framework for archaeological studies.

This regional tephra reference is also useful for under-standing landscape evolution. In Mouralis (2003), Mouralis et al. (2004) and Kuzucuoğlu (2004), we demonstrate that the studied tephra layers are key layers for the reconstruction of past landscapes and of their evolution under the influence of sediment erosion and/or accumulation.

Finally, the tephrostratigraphy and tephrochronology pro-posed here may not only be useful for understanding the volcanic complexes but also for establishing long-distance correlations that are important in different fields of research,

199Tephrostratigraphy and chronology of the Quaternary Gölludağ and Acıgöl volcanic complexes…

1 3

from geomorphology to geoarchaeology, oceanography and climatology.

Acknowledgements A part of the research presented here was con-ducted thanks mainly to the MTA-CNRS-TÜBITAK Project on “Upper

Pleistocene Volcanism and Palaeogeography in Cappadocia, Turkey” (MTA-CNRS-TÜBITAK 2001–2003 RESEARCH PROGRAMME N° 101Y109) and to a Turkish-French TÜBITAK-CNRS agreement, both programmes co-directed by A. Türkecan and C. Kuzucuoğlu. The authors benefited from financial support from MTA, CNRS (LGP) and from French Ministry of Foreign Affairs (Bosphorus Programme). DM

Fig. 11 Göllüdağ and Acıgöl volcanic complexes re-use Neogene structures.  1. Hypothetic limits of collapsed structures associated with both volcanic complexes (see text for discussion). 2. Main faults. 3. Bouger negative anomaly (< 90 mGal), after Froger et al. (1998). 4. Hypothetic limits of Neogene calderas according to Froger et  al.

(1998). 5. Neogene andesitic massifs: Erdaş Dağ, Melendiz and Keçi-boyduran. Quaternary volcano: 6. Scoria cone. 7. Dome. 8. Maar. 9. Lava flows and pyroclastites emitted by Hasandağ. 10. Villages and towns

200 D. Mouralis et al.

1 3

benefited from a PhD scholarship from the University of Paris 12/Créteil (under the direction of Prof. B. Dumas), from the technical and scientific support of Kaletepe excavation team (directed by Prof. N. Balkan-Atlı from Istanbul University & Prof. D. Binder from CEPAM Lab. In Sophia-Antipolis). DM thanks Dr. Lütfi Dokuzoğlu for his help and friendship in Aksaray and Cappadocia. The authors are grateful to A. Schmitt and an anonymous reviewer for numerous suggestions that have greatly contributed improving the initial manuscript.

Compliance with ethical standards

Conflict of interest On behalf of all authors, the corresponding author states that there is no conflict of interest.

References

Atici G, Schmitt AK, Friedrichs B, Sparks S, Danišík M, Yurteri E, Atakay Gündoğdu E, Schindlbeck-Belo J, Çobankaya M, Wang K-L, Lee H-Y (2019) Ages and glass compositions for paired large-volume eruptions from the Acigöl volcanic complex, Cappadocia (Turkey). Mediterranean Geosci Rev. https ://doi.org/10.1007/s4299 0-019-00013 -5

Aydar E, Gourgaud A, Deniel C, Lyberis N, Gündoğdu N (1995) Le volcanisme quaternaire d’anatolie centrale (Turquie): association de magmatisme calco-alcalin et alcalin en domaine de conver-gence. Can J Earth Sci 32:1058–1069

Aydar E, Schmitt AK, Çubukçu HE, Akın L, Ersoy O, Şen E, Dun-can RA, Atıcı G (2012) Correlation of ignimbrites in the central Anatolian volcanic province using zircon and plagioclase ages and zircon compositions. J Volcanol Geoth Res 213–214:83–97. https ://doi.org/10.1016/j.jvolg eores .2011.11.005

Aydın F, Schmitt AK, Siebel W, Sönmez M, Ersoy Y, Lermi A, Dirik K, Duncan R (2014) Quaternary bimodal volcanism in the Niğde volcanic complex (Cappadocia, central Anatolia, Turkey): age, petrogenesis and geodynamic implications. Contrib Mineral Petrol 168:1078. https ://doi.org/10.1007/s0041 0-014-1078-3

Balkan-Atlı N, Binder D (2000) L’atelier néolithique de Kömürcü–Kaletepe: fouilles de 1999. Anatolia Antiqua 8:199–214

Balkan-Atlı N, Kuhn SL, Astruc L, Kayacan N, Dinçer B (2011) Göllüdağ survey 2010. Anatol Antiqu XIX:259–278

Batum I (1978) Nevsehir güneybatısındaki Göllüdağ ve Acıgöl volkan-itlerinin jeokimyası ve petrolojisi. Yerbilimleri 4:50–69 (70–88)

Besang C, Eckhardt FJ, Harre W, Kreuzer H, Müller P (1977) Radi-ometrische alterbestimmungen an neogenen eruptivegestinen des Türkei. Geol Jahrb 25:3–36

Bigazzi G, Yegingil Z, Ercan T, Oddone M, Özdoğan M (1993) Fis-sion track dating obsidians in Central and Northern Anatolia. Bull Volcanol 55:588–595. https ://doi.org/10.1007/BF003 01811

Bigazzi G, Yegingil Z, Ercan T, Oddone M, Özdoğan M (1997) Doğu Anadolu’daki obsidiyen içeren volkaniklerin “Fisyon Track” yöntemiyle yaş tayini (Age determination of obsidian bearing vol-canics in eastern Anatolia using the Fission-track dating method). Geol Bull Turk 40:57–72

Binder D, Gratuze B, Mouralis D, Balkan-Atlı N (2011) New inves-tigations of the Göllüdağ obsidian lava flows system: a multi-disciplinary approach. J Archaeol Sci 38:3174–3184. https ://doi.org/10.1016/j.jas.2011.05.014

Bottema S, Woldring H (1984) Late Quaternary vegetation and climate of south-western Turkey. Paleohistoria 26:123–149

Charbit S, Guillou H, Turpin L (1998) Cross calibration of K-Ar stand-ard minerals using an unspiked Ar measurement technique. Chem Geol 150:147–159. https ://doi.org/10.1016/S0009 -2541(98)00049 -7

Chataigner C, Poidevin J-L, Arnaud NO (1998) Turkish occurrences of obsidian and use by prehistoric peoples in the Near East from 14,000 to 6000 BP. J Volcanol Geoth Res 85:517–537. https ://doi.org/10.1016/S0377 -0273(98)00069 -9

Clay PL, Busemann H, Sherlock SC, Barry TL, Kelley SP, McGarvie DW (2015) 40Ar/39Ar ages and residual volatile contents in degassed subaerial and subglacial glassy volcanic rocks from Iceland. Chem Geol 403:99–110. https ://doi.org/10.1016/j.chemg eo.2015.02.041

Di Giuseppe P, Agostini S, Manetti P, Savaşçın MY, Conticelli S (2018) Sub-lithospheric origin of Na-alkaline and calc-alkaline magmas in a post-collisional tectonic regime: Sr–Nd–Pb iso-topes in recent monogenetic volcanism of Cappadocia, Central Turkey. Lithos 316–317:304–322. https ://doi.org/10.1016/j.litho s.2018.07.018

Doğan-Kulahcı GD, Temel A, Gourgaud A, Varol E, Guillou H, Deniel C (2018) Contemporaneous alkaline and calc-alkaline series in Central Anatolia (Turkey): spatio-temporal evolution of a post-collisional Quaternary basaltic volcanism. J Volcanol Geoth Res 356:56–74. https ://doi.org/10.1016/j.jvolg eores .2018.02.012

Druitt TH, Brenchley PJ, Gökten YE, Francaviglia V (1995) Late quaternary rhyolitic eruptions from the Acigöl complex, central Turkey. J Geol Soc 152:655–667. https ://doi.org/10.1144/gsjgs .152.4.0655

Eastwood WJ, Pearce NJG, Perkins WT, Lamb HF, Westgate JA, Rob-erts N (1999) Geochemistry of Santorini tephra in lake sediments from southwest Turkey. Glob Planet Change 21:17–29

Ercan T, Yıldırım T, Akbaşlı A (1987) Gelveri (Niğde)—Kızılcin (Nevşehir) arasindaki volkanizmanin özellikleri. Jeomorfoloji Dergisi 15:27–36

Ercan T, Tokel S, Can B, Fişekçi A, Fujitani T, Notsu K, Selvi Y, Ölmez M, Matsuda JI, Ui T, Yıldırım T, Akbaşlı A (1990) Hasan dağı—karacadağ (Orta Anadolu) dolaylarindaki senozoyik yaşli volkanizmanin kökeni ve evrimi. Jeomorfoloji Dergisi 18:39–54

Ercan T, Tokel S, Matsuda JI, Ui T, Notsu K, Fujitani T (1992) Hasandağ-karacadağ (Orta Anadolu) kuvaterner volkanizmasina ilişkin yeni jeokimyasal, izotopik ve radyometrik veriler. Türkiye Jeoloji Kurultayı Bülteni 7:8–21

Ferrari L, Garduño VH, Pasquarè G, Tibaldi A (1991) Geology of Los Azufres Caldera, Mexico, and its relationships with regional tectonics. J Volcanol Geoth Res 47:129–148. https ://doi.org/10.1016/0377-0273(91)90105 -9

Flude S, Tuffen H, Sherlock SC (2018) Spatially heterogeneous argon-isotope systematics and apparent 40Ar/39Ar ages in per-litised obsidian. s 480:44–57. https ://doi.org/10.1016/j.chemg eo.2017.05.018

Frogatt PC (1992) Standardization of the chemical analysis of tephra deposits. Report of the ICCT working group. Quat Int 13(14):93–96

Froger J-L, Lénat J-F, Chorowicz J, Le Pennec J-L, Bourdier J-L, Köse O, Zimitoğlu O, Gündoğdu N, Gourgaud A (1998) Hidden calderas evidenced by multisource geophysical data; example of Cappadocian Calderas, Central Anatolia. J Volcanol Geoth Res 85:99–128. https ://doi.org/10.1016/S0377 -0273(98)00052 -3

Fuhrmann U, Lippolt H, Hess JC (1987) HD-B1 Biotite reference mate-rial for K-Ar chronometry. Chem Geol 66:41–51

Guillou H, Singer BS, Laj C, Kissel C, Scaillet S, Jicha BR (2004) On the age of the Laschamp geomagnetic excursion. Earth Planet Sci Lett 227:331–343. https ://doi.org/10.1016/j.epsl.2004.09.018

Guillou H, Nomade S, Carracedo JC, Kissel C, Laj C, Perez Torrado FJ, Wandres C (2011) Effectiveness of combined unspiked K–Ar and 40Ar/39Ar dating methods in the 14C age range. Quat Geo-chronol 6:530–538. https ://doi.org/10.1016/j.quage o.2011.03.011

Hamann Y, Wulf S, Ersoy O, Ehrmann W, Aydar E, Schmiedl G (2010) First evidence of a distal early Holocene ash layer in Eastern Mediterranean deep-sea sediments derived from the

201Tephrostratigraphy and chronology of the Quaternary Gölludağ and Acıgöl volcanic complexes…

1 3

Anatolian volcanic province. Quatern Res 73:497–506. https ://doi.org/10.1016/j.yqres .2009.12.004

Hautmann S, Lippolt HJ (2000) 40Ar/39Ar dating of central Euro-pean K–Mn oxides—a chronological framework of supergene alteration processes during the Neogene. Chem Geol 170:37–80. https ://doi.org/10.1016/S0009 -2541(99)00241 -7

Hess JC, Lippolt HJ (1994) Compilation of K–Ar measurements on HD-B1 standard biotite. In: Odin GS (ed) Phanerozoic time scale., Bull. Liais. Inform. I.U.G.S., Subcom. Geochronol., vol 12, pp 19–23

Innocenti F, Mazzuoli R, Pasquare’ G, Radicati Di Brozolo F, Villari L (1976) Evolution of the volcanism in the area of interaction between the arabian, anatolian and iranian plates (Lake van, Eastern Turkey). J Volcanol Geothermal Res 1:103–112. https ://doi.org/10.1016/0377-0273(76)90001 -9

Inoue K, Saito M (1997) Climatic changes in the Konya basin, Tur-key, estimated from physicochemical, mineralogical, and geo-chemical characteristics of its lacustrine sediments. Jpn Rev 8:147–165

Karabıyıoğlu M, Kuzucuoğlu C, Fontugne M, Kaiser B, Mouralis D (1999) Facies and depositional sequences of the late pleistocene Göçü shoreline system, Konya basin, Central Anatolia: implica-tions for reconstructing lake-level changes. Quatern Sci Rev Qua-tern Sci Rev 18:593–609

Kashima K (2002) Environmental and climatic changes during the last 20,000 years at Lake Tuz, central Turkey. CATENA 48:3

Kazancı N, Gevrek A, Varol B (1995) Facies changes and high calorific peat formation in a quaternary maar lake, central Anatolia, Tur-key: the possible role of geothermal processes in a closed lacus-trine basin. Sed Geol 94:255–266. https ://doi.org/10.1016/0037-0738(94)00092 -9

Kazancı N, Boyraz S, Özkul M, Alçiçek MC, Kadıoğlu YK (2012) Late holocene terrestrial tephra record at western Anatolia, Turkey: possible evidence of an explosive eruption outside Santorini in the eastern Mediterranean. Global Planet Change 80–81:36–50. https ://doi.org/10.1016/j.glopl acha.2011.09.007

Kuhn SL, Dinçer B, Balkan-Atlı N, Erturaç MK (2015) Paleolithic occupations of the Göllü Dağ, Central Anatolia, Turkey. J Field Archaeol 40:581–602. https ://doi.org/10.1179/20424 58215 Y.00000 00020

Kürkçüoğlu B, Şen E, Aydar E, Gourgaud A, Gündoğdu N (1998) Geo-chemical approach to magmatic evolution of Mt. Erciyes stratovol-cano Central Anatolia, Turkey. J Volcanol Geoth Res 85:473–494. https ://doi.org/10.1016/S0377 -0273(98)00067 -5

Kuzucuoğlu C (2004) Upper and late pleistocene environmental changes in Anatolia reconstructed from sediment sequences and landform studies. In: Barla B, Mater B, Algan O (eds) Proceed-ings of international symposium on earth system sciences 2004 Istanbul–Turkey: dedicated to the memory of professor sirri erinç. Istanbul University Press, Istanbul, pp 19–27

Kuzucuoğlu C, Karabiyikoğlu M, Fontugne M, Pastre J-F, Ercan T (1997) Environmental changes in holocene lacustrine sequences from Karapinar in the Konya Plain (Turkey). In: Dalfes HN, Kukla G, Weiss H (eds) Third millennium BC climate change and old world collapse, NATO ASI series. Springer, Berlin Heidelberg, pp 451–463

Kuzucuoğlu C, Pastre J-F, Black S, Ercan T, Fontugne M, Guillou H, Hatté C, Karabıyıoğlu M, Orth P, Türkecan A (1998) Iden-tification and dating of tephra layers from quaternary sedimen-tary sequences of Inner Anatolia, Turkey. J Volcanol Geoth Res 85:153–172. https ://doi.org/10.1016/S0377 -0273(98)00054 -7

Laj C, Guillou H, Kissel C (2014) Dynamics of the earth magnetic field in the 10–75 kyr period comprising the Laschamp and Mono Lake excursions: new results from the French Chaîne des Puys in a global perspective. Earth Planet Sci Lett 387:184–197. https ://doi.org/10.1016/j.epsl.2013.11.031

Le Bourdonnec F-X, Nomade S, Poupeau G, Guillou H, Tushabram-ishvili N, Moncel M-H, Pleurdeau D, Agapishvili T, Voinchet P, Mgeladze A, Lordkipanidze D (2012) Multiple origins of Bondi Cave and Ortvale Klde (NW Georgia) obsidians and human mobility in Transcaucasia during the Middle and Upper Palaeo-lithic. J Archaeol Sci 39:1317–1330. https ://doi.org/10.1016/j.jas.2011.12.008

Le Maitre RW (ed) (2002) Igneous rocks: a classification and glossary of terms: recommendations of the international union of geologi-cal sciences, subcommission on the systematics of igneous rocks. Cambridge University Press, Cambridge

Le Pennec J-L, Bourdier J-L, Froger J-L, Temel A, Camus G, Gour-gaud A (1994) Neogene ignimbrites of the Nevsehir plateau (Cen-tral Turkey): stratigraphy, distribution and source constraints. J Volcanol Geoth Res 63:59–87. https ://doi.org/10.1016/0377-0273(94)90018 -3

Le Pennec J-L, Temel A, Froger J-L, Şen S, Gourgaud A, Bourdier J-L (2005) Stratigraphy and age of the Cappadocia ignimbrites, Tur-key: reconciling field constraints with paleontologic, radiochro-nologic, geochemical and paleomagnetic data. J Volcanol Geoth Res 141:45–64. https ://doi.org/10.1016/j.jvolg eores .2004.09.004

Lowe DJ (2011) Tephrochronology and its application: a review. Quat Geochronol 6:107–153. https ://doi.org/10.1016/j.quage o.2010.08.003

Mouralis D, Kuzucuoğlu C, Pastre JF, Guillou H, Slimak L, Kunesch S (2004) Göllüdag quaternary volcanic complex (central Anatolia): morphologic evolution and environmental implications. In: Barla B, Mater B, Algan O (eds) Proceedings of international sympo-sium on earth system sciences 2004, Istanbul–Turkey: dedicated to the memory of professor Sırrı Erinç. Istanbul University Press, Istanbul, pp 67–74

Morgan LE, Renne PR, Taylor RE, WoldeGabriel G (2009) Archaeo-logical age constraints from extrusion ages of obsidian: examples from the Middle Awash, Ethiopia. Quat Geochronol 4:193–203. https ://doi.org/10.1016/j.quage o.2009.01.001

Mouralis D (2003) Les complexes volcaniques quaternaires de Cap-padoce (Göllüdag et Acigöl)—Turquie : évolutions morphody-namiques et implications environnementales (Thèse doctorat). Paris 12, Paris

Mouralis D, Pastre J-F, Kuzucuoglu C, Türkecan A, Atıcı Y, Slimak L, Guillou H, Kunesch S (2002) Les complexes volcaniques rhyo-lithiques quaternaires d’anatolie centrale (Göllü Dag et Acigöl, Turquie): genèse, instabilité, contraintes environnementales. Quaternaire Revue de l’Association Française pour l’Étude du Quaternaire 13:219–228. https ://doi.org/10.3406/quate .2002.1714

Mouralis D, Aydar E, Türkecan A, Kuzucuoğlu C (2019) Quaternary volcanic landscapes and prehistoric sites in Southern Cappado-cia: Göllüdağ, Acıgöl and Hasandağ. In: Kuzucuoğlu C, Çiner A, Kazancı N (eds) Landscapes and landforms of Turkey, world geomorphological landscapes. Springer International Publishing, Cham, pp 551–563. https ://doi.org/10.1007/978-3-030-03515 -0_32

Notsu K, Fujitani T, Ui T, Matsuda J, Ercan T (1995) Geochemical features of collision-related volcanic rocks in central and eastern Anatolia, Turkey. J Volcanol Geoth Res 64:171–191. https ://doi.org/10.1016/0377-0273(94)00077 -T

Olanca K (1994) Geochimie des Laves Quaternaires de Cappadoce (Turquie). Les Appareils Monogéniques (Thèse doctorat—PhD). Université Blaise Pascal, Clermont-Ferrand

Parfenoff A, Pomerol C, Tourenq J, Wyart J (1970) Les Minéraux en grains: méthodes d’étude et détermination. Masson, Paris

Pasquarè G (1966) Outlines of the neogene and quaternary volcan-ism of Asia Minor. Atti Della Accad Nazionale Dei Lincei 8:1077–1085

Pasquarè G (1968) Geology of the cenozoic volcanic area of Central Anatolia. Atti della Accademia nazionale dei Lincei 9:53–204

202 D. Mouralis et al.

1 3

Pastre J-F, Kuzucuoğlu C, Fontugne M, Guillou H, Karabiyikoğlu M, Ercan T, Türkecan A (1998) Séquences volcanisées et cor-rélations téphrologiques au N–W du Hasan dağ (Haut bassin de la Melendiz, Anatolie centrale, Turquie). Quaternaire Revue de l’Association Française pour l’Étude du Quaternaire 9:169–183. https ://doi.org/10.3406/quate .1998.1601

Platevoet B, Scaillet S, Guillou H, Blamart D, Nomade S, Massault M, Poisson A, Elitok Ö, Özgür N, Yağmurlu F, Yılmaz K (2007) Pleistocene eruptive chronology of the Gölcük volcano, Isparta Angle, Turkey. Quaternaire Revue de l’Association Française pour l’Étude du Quaternaire 19:147–156. https ://doi.org/10.4000/quate rnair e.3092

Poidevin J-L (1998) Les gisements d’obsidienne de turquie et de trans-caucasie. Géologie, géochimie et chronométrie. In: Cauvin M-C, Gourgaud A, Gratuze B, Arnaud N, Poupeau G, Poidevin J-L, Chataigner C (eds) L’Obsidienne Au Proche et Moyen Orient. Du Volcan à l’outil, BAR international series 738. Archeopress, Oxford, pp 105–156

Rabayrol F, Hart CJR, Thorkelson DJ (2019) Temporal, spatial and geochemical evolution of late Cenozoic post-subduction magma-tism in central and eastern Anatolia, Turkey. Lithos 336–337:67–96. https ://doi.org/10.1016/j.litho s.2019.03.022

Roberts N, Eastwood WJ, Lamb HF, Tibby JC (1997) The Age and causes of Mid-Holocene environmental change in Southwest Tur-key. In: Dalfes HN, Kulkla G, Weiss H (eds) Third millennium BC climate change and old world collapse, NATO ASI series, series I, global environmental change. Springer, Berlin, pp 409–429

Roberts N, Meadows ME, Dodson JR (2001) The history of mediterra-nean-type environments: climate, culture and landscape. Holocene 11:631–634. https ://doi.org/10.1191/09596 83019 5663

Sasco R, Guillou H, Nomade S, Scao V, Maury RC, Kissel C, Wan-dres C (2017) 40Ar/39Ar and unspiked 40 K–40Ar dating of upper Pleistocene volcanic activity in the Bas-Vivarais (Ardèche, France). J Volcanol Geoth Res 341:301–314. https ://doi.org/10.1016/j.jvolg eores .2017.06.003

Satow C, Tomlinson EL, Grant KM, Albert PG, Smith VC, Man-ning CJ, Ottolini L, Wulf S, Rohling EJ, Lowe JJ, Blockley SPE, Menzies MA (2015) A new contribution to the Late quaternary tephrostratigraphy of the Mediterranean: aegean Sea core LC21. Quatern Sci Rev 117:96–112. https ://doi.org/10.1016/j.quasc irev.2015.04.005

Schmitt AK, Danišík M, Evans NJ, Siebel W, Kiemele E, Aydin F, Harvey JC (2011) Acigöl rhyolite field, Central Anatolia (part 1): high-resolution dating of eruption episodes and zircon growth rates. Contrib Mineral Petrol 162:1215–1231. https ://doi.org/10.1007/s0041 0-011-0648-x

Siebel W, Schmitt AK, Kiemele E, Danišík M, Aydın F (2011) Acıgöl rhyolite field, central Anatolia (part II): geochemical and isotopic (Sr–Nd–Pb, δ18O) constraints on volcanism involving two high-silica rhyolite suites. Contrib Mineral Petrol 162:1233–1247. https ://doi.org/10.1007/s0041 0-011-0651-2

Singer BS, Hoffman KA, Schnepp E, Guillou H (2008) Multiple Brunhes Chron excursions recorded in the West Eifel (Germany) volcanics: support for long-held mantle control over the non-axial dipole field. Physics of the Earth and Planetary Interiors. Palaeomagnetism Earth’s Deep Inter 169:28–40. https ://doi.org/10.1016/j.pepi.2008.05.001

Singer BS, Guillou H, Jicha BR, Laj C, Kissel C, Beard BL, John-son CM (2009) 40Ar/39Ar, K-Ar and 230Th–238U dating of the Laschamp excursion: a radioisotopic tie-point for ice core and climate chronologies. Earth Planet Sci Lett 286:80–88. https ://doi.org/10.1016/j.epsl.2009.06.030

Slimak L, Roche H, Mouralis D, Buitenhuis H, Balkan-Atlı N, Binder D, Kuzucuoğlu C, Grenet M (2004) Kaletepe Deresi 3 (Turquie), aspects archéologiques, chronologiques et paléontologiques d’une séquence pléistocène en Anatolie centrale. CR Palevol 3:411–420. https ://doi.org/10.1016/j.crpv.2004.04.005

Slimak L, Kuhn SL, Roche H, Mouralis D, Buitenhuis Hijlke, Balkan-AtlI N, Binder D, Kuzucuoglu C, Guillou H (2008) Kaletepe Deresi 3 (Turkey): archaeological evidence for early human set-tlement in Central Anatolia. J Hum Evol 54:99–111. https ://doi.org/10.1016/j.jhevo l.2007.07.004

Steiger RH, Jäger E (1977) Subcommission on geochronology: conven-tion on the use of decay constants in geo- and cosmochronology. Earth Planet Sci Lett 36:359–362. https ://doi.org/10.1016/0012-821X(77)90060 -7

Sullivan DG (1990) Minoan tephra in lake sediments in Western Tur-key dating the eruption and assessing the atmospheric dispersal of the ash. In: Hardy DA, Renfrew C (eds) Thera and the Aegean World—Proceedings of the Third International Congress, San-torini, Greece The Thera Foundation, London, pp 114–118

Sullivan DG (1988) The discovery of Santorini Minoan tephra in west-ern Turkey. Nature 333:552–554. https ://doi.org/10.1038/33355 2a0

Temel A, Gündoğdu MN, Gourgaud A, Le Pennec J-L (1998) Ign-imbrites of Cappadocia (Central Anatolia, Turkey): petrology and geochemistry. J Volcanol Geoth Res 85:447–471. https ://doi.org/10.1016/S0377 -0273(98)00066 -3

Tryon CA, Logan MAV, Mouralis D, Kuhn S, Slimak L, Balkan-Atlı N (2009) Building a tephrostratigraphic framework for the Paleo-lithic of Central Anatolia, Turkey. J Archaeol Sci 36:637–652. https ://doi.org/10.1016/j.jas.2008.10.006

Türkecan A (2015) Türkiye’nin senozoyik volkanitleri. Maden Tetkik Arama Müdürlüğü, Ankara

Türkecan A, Kuzucuoğlu C, Mouralis D, Pastre J-F, Atıcı Y, Guillou H, Fontugne M (2004) Upper pleistocene volcanism and palaeogeog-raphy in Cappadocia, Turkey. MTA-CNRS-TÜBİTAK 2001–2003 Research Programme. Tübitak Project No. 101Y109 (No. 10652). M.T.A, Ankara

Yıldırım T, Özgür R (1981) Acıgöl kalderası. Jeomorfoloji Dergisi 10:59–70

Zanchetta G, Sulpizio R, Roberts N, Cioni R, Eastwood WJ, Siani G, Caron B, Paterne M, Santacroce R (2011) Tephrostratigraphy, chronology and climatic events of the Mediterranean basin dur-ing the Holocene: an overview. Holocene 21:33–52. https ://doi.org/10.1177/09596 83610 37753 1

Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.