Growth of fissure ridge travertines from geothermal springs of Denizli Basin, western Turkey

Geological Society of America Bulletin

doi: 10.1130/B30606.1 2012;124, no. 9-10;1629-1645Geological Society of America Bulletin

Michele Soligo, Paola Tuccimei and Igor M. VillaLuigi De Filippis, Claudio Faccenna, Andrea Billi, Erlisiana Anzalone, Mauro Brilli, Mehmet Özkul, western TurkeyGrowth of fissure ridge travertines from geothermal springs of Denizli Basin,

Email alerting servicesarticles cite this article

to receive free e-mail alerts when newwww.gsapubs.org/cgi/alertsclick

SubscribeAmerica Bulletin

to subscribe to Geological Society ofwww.gsapubs.org/subscriptions/click

Permission request to contact GSAhttp://www.geosociety.org/pubs/copyrt.htm#gsaclick

official positions of the Society.citizenship, gender, religion, or political viewpoint. Opinions presented in this publication do not reflectpresentation of diverse opinions and positions by scientists worldwide, regardless of their race, includes a reference to the article's full citation. GSA provides this and other forums for thethe abstracts only of their articles on their own or their organization's Web site providing the posting to further education and science. This file may not be posted to any Web site, but authors may postworks and to make unlimited copies of items in GSA's journals for noncommercial use in classrooms requests to GSA, to use a single figure, a single table, and/or a brief paragraph of text in subsequenttheir employment. Individual scientists are hereby granted permission, without fees or further Copyright not claimed on content prepared wholly by U.S. government employees within scope of

Notes

© 2012 Geological Society of America

on August 29, 2012gsabulletin.gsapubs.orgDownloaded from

http://gsabulletin.gsapubs.org/cgi/alerts

http://gsabulletin.gsapubs.org/subscriptions/index.ac.dtl

http://www.geosociety.org/pubs/copyrt.htm#gsa

http://gsabulletin.gsapubs.org/

Growth of fi ssure ridge travertines from geothermal springs of Denizli Basin, western Turkey

Luigi De Filippis1,†, Claudio Faccenna1, Andrea Billi2, Erlisiana Anzalone3, Mauro Brilli2, Mehmet Özkul4, Michele Soligo1, Paola Tuccimei1, and Igor M. Villa5,6

1Dipartimento di Scienze Geologiche, Università Roma Tre, Largo S.L. Murialdo 1, Rome I-00146, Italy2Consiglio Nazionale delle Ricerche, Istituto di Geologia Ambientale e Geoingegneria, c.o. Dipartimento di Scienze della Terra, Sapienza Università di Roma, P.le A. Moro 5, Rome I-00185, Italy3Consiglio Nazionale delle Ricerche, Istituto per l’Ambiente Marino Costiero, Calata Porta di Massa, Naples I-80133, Italy4Department of Geological Engineering, Pamukkale University, Denizli TR-20070, Turkey5Institut für Geologie, Universität Bern, Baltzerstrasse 1 + 3, Bern CH-3012, Switzerland6Dipartimento di Scienze Geologiche Geotecnologie, Università di Milano Bicocca, Piazza della Scienza 4, Milan I-20126, Italy

ABSTRACT

Fissure ridge travertines grown from geothermal springs of Denizli Basin, south-western Turkey, are investigated through stratigraphic, structural, geochemical, and geochronological methods, with the aim of understanding the growth of these elongate mound-shaped structures. Two main types of travertine deposits are recognized: (1) bed-ded travertines, which grew as fl owstone on sloping surfaces and form the bulk of fi ssure ridges, and (2) banded travertines, which grew as veins within the bedded travertine chiefly along its central feeding conduit. Stratigraphic and structural observations shed light on the bedded-banded travertine relationships, where the banded features grew through successive accretion phases, crosscutting the bedded travertine or form-ing sill-like structures. The bedded and banded travertines alternated their growth, as demonstrated by complicated crosscutting relationships and by the upward suture, in places, of banded travertine by bedded trav-ertine that was, in turn, crosscut by younger banded travertine. The bedded travertine is often tilted away from the central axis of the fi ssure ridge, thus leaving more room for the central banded travertine to form. U-series ages confi rm the bedded-banded travertine temporal relationships and show that the growth of the studied fi ssure ridges lasted up to several tens of thousands of years during Quaternary time. The banded travertine was deposited mainly during cold events, possibly in coincidence with seismic

events that might have triggered the outfl ow of deep geothermal fl uids. C and O stable iso-tope and rare earth element data indicate a shallow feeding circuit for the studied struc-tures with a fl uid component deriving from a deeper geothermal circuit. A crack-and-seal mechanism of fi ssure ridge growth is pro-posed, modulated by the interplay of local and regional infl uencing factors and mecha-nisms such as geothermal fl uid discharge, paleoclimate, tectonics, and the progressive tilting of bedded travertine limbs over a soft substratum creating the necessary space for the central veins to grow.

INTRODUCTION

Endogenic travertine deposits are the result of CaCO3 precipitation from hydrothermal waters rising along fractures and faults in Earth’s crust (Barnes et al., 1978; Pentecost, 1995, 2005; Ford and Pedley, 1996; Crossey et al., 2006; Pedley, 2009). As such, endogenic travertines are the result of natural leakage from CO2-charged sys-tems and may serve as useful analogs to assess long-term tectonically related leaks from arti-fi cially charged geologic sites (Shipton et al., 2004, 2005; Haszeldine et al., 2005; Le Guen et al., 2007; Nelson et al., 2009; Uysal et al., 2009; Dockrill and Shipton, 2010). The pre-requi site for the use of endogenic travertines as analogs of anthropogenic activities is, however, the complete understanding of their deposi-tion history and infl uencing factors. Moreover, unraveling the growth of endogenic travertines provides important insights into the feeding geo-thermal circuits and, therefore, into the assess-ment of the geothermal potential (Newell et al., 2005; Crossey et al., 2009; Banerjee et al., 2011).

The main aim of this paper is to understand the growth of fi ssure ridge travertines (Fig. 1), which usually occur in active or recently active extensional settings (Table DR11). We studied fi ssure ridge travertines from the Denizli Basin, western Turkey, where travertines are wide-spread and quarries provide excellent exposure. Previous studies provided important insights into the nature of these structures (Altunel and Hancock, 1993a, 1993b; Altunel, 1994; Altunel and Hancock, 1996; Çakır, 1999; Hancock et al., 1999; Özkul et al., 2002; Altunel and Karabacak, 2005; Uysal et al., 2007, 2009). We take ad-vantage of these previous studies and data, and provide new stratigraphic, structural, geochemi-cal, and geochronological evidence to constrain the nucleation and growth of four fi ssure ridge travertines.

FISSURE RIDGE TRAVERTINES

Endogenic travertine deposits are charac-terized by different shapes and sizes, so that large sedimentary structures such as waterfalls, terrace-mounds, self-built channels, and fi ssure ridges have been identifi ed mainly on a morpho-logical basis (Chafetz and Folk, 1984; Altunel and Hancock, 1993a, 1993b; Özkul et al., 2002). With the term fi ssure ridge, Hayden (1883) de-scribed, for the fi rst time, a travertine ridge at Mammoth Hot Springs (Wyoming, USA). Since then, the term has been applied to several other

For permission to copy, contact [email protected]© 2012 Geological Society of America

1629

GSA Bulletin; September/October 2012; v. 124; no. 9/10; p. 1629–1645; doi: 10.1130/B30606.1; 11 fi gures; 1 table; Data Repository item 2012245.

†E-mail: ldefi [email protected]

1GSA Data Repository item 2012245, additional information concerning U-series dating methods and data, fi ssure ridge travertines known on the Earth’s surface, C- and O-stable isotopes, and correlations with paleoclimate at the regional and global scales, is available at http://www.geosociety.org/pubs/ft2012.htm or by request to [email protected].



De Filippis et al.

1630 Geological Society of America Bulletin, September/October 2012

similar structures in Turkey, Italy, and many other countries (Table DR1 [see footnote 1]) to indicate whale-back-shaped or elongate mound-shaped deposits of travertines (Bargar, 1978; Chafetz and Folk, 1984; Chesterman and Klein-hampl, 1991; Altunel and Hancock, 1993a, 1993b; Altunel, 1994; Altunel and Hancock, 1996; Çakır, 1999; Hancock et al., 1999; Atabey , 2002; Özkul et al., 2002; Temiz, 2004; Altunel and Karabacak, 2005; Uysal et al., 2007; Brogi and Capezzuoli, 2009; Haluk Selim and Yanik, 2009; Uysal et al., 2009; De Filippis and Billi, 2012). These travertine structures (Fig. 1) are between a few tens of meters and ~2700 m in length, 5 and 400 m in width, and 1 and 25–30 m in height. A basic characteristic of a fi ssure ridge is the presence of two types of travertine: one porous and stratifi ed (bedded travertine) that constitutes the bulk of the fi ssure ridge, and one sparitic nonporous travertine (banded trav-ertine) that usually fi lls the interior walls of the fi ssure ridges, forming injection veins and sill-like structures (Fig. 1). The banded travertine is not always observable due to the lack of proper exposures. The bedded travertine may, in fact, entirely encompass and coat the banded one (e.g., Fig. 1B).

In the fi ssure ridges hitherto studied, several main common features can be identifi ed: (1) A fi ssure ridge is a linear elongated travertine mound deposited from underground circulated hot waters; (2) fi ssure ridges can be straight, curved, or even bifurcated in plan view; (3) they are characterized by a central extensional fi s-sure or fracture extending along the crest of the ridge for nearly its entire length, often with as-sociated minor subparallel fractures also named parasitic fi ssures or fractures (Altunel and Han-cock, 1996); (4) ascending hot waters in active fi ssure ridges cause carbonate precipitation both within the fi ssure space and on the ridge fl anks, thus generating, respectively, banded and bedded travertine deposits (Bargar, 1978; Altunel and Hancock, 1993a, 1993b; Altunel and Hancock, 1996; Uysal et al., 2007, 2009); (5) most fi ssure ridges studied are located on the hanging wall of normal faults (Altunel and Hancock, 1993a, 1993b); and (6) fi ssure ridge growth is progres-sive in time and space, involving a spatial growth in all three dimensions (i.e., length, width, and height) and a temporal duration that can be as long as ~400 ka, as determined by U-series dat-ing methods (Altunel and Karabacak, 2005).

So far, fi ssure ridges, either active or ex-tinct, have been studied in the Denizli Basin, Turkey (Altunel and Hancock, 1993a, 1993b, 1996; Hancock et al., 1999; Özkul et al., 2002; Uysal et al., 2007, 2009), in the Kirşehir region, Central Anatolia, Turkey (Atabey, 2002; Temiz, 2004; Temiz et al., 2009; Uysal et al., 2009), in the Sivas area, Turkey (Piper et al., 2007; Mesci et al., 2008), at Rapolano Terme, Italy (Guo and Riding, 1999; Brogi and Capezzuoli, 2009), Mammoth Hot Springs, Wyoming (Bar-gar, 1978), Bridgeport, California (Chesterman and Kleinhampl, 1991; Hancock et al., 1999), Soda Dam, New Mexico (Goff and Shevenell, 1987; Chafetz and Folk, 1984), Paradox Basin, Utah (Shipton et al., 2004, 2005; Dockrill and Shipton, 2010), Zerka Ma’in, Jordan (Khoury et al., 1984), Hammam Meskoutine, Algeria (Pentecost and Viles, 1994; Pentecost, 2005), Band-e-Hajar, Afghanistan (de L’Apparent, 1966), San Antonio Texcala, Mexico (Michal-zik et al., 2001), and elsewhere (Table DR1 [see footnote 1]).

GEOLOGICAL SETTING

The Denizli Basin is located in western Tur-key, which is one of the world’s most rapidly extending regions (e.g., Jackson and McKenzie , 1988; Westaway, 1990). This region, which is part of the Aegean extensional province (Boz-kurt, 2001), is characterized by an extensional strain rate of at least 20 mm/a (Kahle et al., 1998; Reilinger et al., 2006). In particular, the

rate of motion appears to increase from the northern Arabian plate to the Hellenic Trench through western Turkey, where recent mea-surements provide an extensional strain rate of 24.6 ± 1 mm/a (Reilinger et al., 2006). The extensional tectonic regime of western Turkey is regarded by Şengör et al. (1985) as having started in Tortonian time (ca. 7 Ma), when the Anatolian block began to escape toward the west and a N-S stretching regime commenced in western Turkey.

The principal active normal faults of western Turkey generally strike E-W, but grabens lo-cally trending NW-SE and NE-SW also occur. The Denizli Basin occurs at the confl uence of the E-W–trending Menderes and NW-trending Gediz grabens, which are located close to the eastern margin of the Neogene–Quaternary Aegean extensional province (Westaway, 1990; Koçyiğit, 2005) (Fig. 2). The Denizli Basin trends NW-SE, with a length of ~50 km and a width of 20 km, and rests at an altitude of ~200 m above sea level (a.s.l.). The metamor-phic rocks exposed in the Menderes Massif constitute the basement of the Denizli Basin, the formation of which started in early Miocene time (Alçiçek et al., 2007). Marbles are an impor-tant constituent of the metamorphic basement (Pamir and Erentöz, 1974; Okay, 1989) and are, together with Pliocene continental limestone, at the origin of the studied travertines (Uysal et al., 2007). Above the downfaulted meta-morphic basement, continental sedimentary rocks of Neogene age constitute the infi lling of the Denizli Basin, where extensional tectonics are still active at different rates in different parts of the basin (Altunel and Karabacak, 2005). An average strain rate was calculated (derived from heave values of normal faults) as ~0.3 mm/a (Westaway, 1993). The same extensional rate was inferred by Altunel and Karabacak (2005) studying the fi ssure ridge travertines from the Denizli Basin. Recent earthquakes with magni-tudes up to 5.5 (e.g., the 1965 M 5.3, 1976 M 5.0, and 1986 M 5.5 earthquakes) are symptomatic of the active extensional tectonics in the Denizli Basin (Ates and Bayülke, 1982; Taymaz, 1993; Westaway, 1993; Bozkurt, 2001).

An important locality in the Denizli Basin is Pamukkale (Turkish term for Cotton Castle), where actively accumulating travertines are one of Turkey’s most famous tourist sights. Trav-ertine deposition at Pamukkale has been ac-tive since at least 400 ka, partially covering the Roman city and necropolis of Hierapolis (Altunel and Hancock, 1993b). At Pamukkale, travertine originates from geothermal springs that emerge from open fi ssures and at least one fault zone at a temperature of 36 °C (Kele et al., 2011), but, in the Denizli extensional basin, temperatures

A

B

WSW ESE

bedded travertine

sill-likestructure

injectionvein

banded travertine

crestal fissure

minor or parasitic fissure

longitudinal closure

longitudinal closure

5 m

Figure 1. (A) Schematic three-dimensional representation of a typical fissure ridge traver tine deposit. (B) Western end of the San Giovanni Terme fi ssure ridge traver-tine, Rapolano Terme, Tuscany, Italy (for further details, see Brogi and Capezzuoli, 2009).



Growth of fi ssure ridge travertines, Turkey

Geological Society of America Bulletin, September/October 2012 1631

range between 28 °C and 59 °C (Gökgöz, 1998; Özler, 2000; Dilsiz et al., 2004). The studied fi s-sure ridges are all located in the Denizli Basin. The Çukurbağ fi ssure ridge, in particular, lies below the famous Pamukkale travertine.

FISSURE RIDGES OF THE DENIZLI BASIN

We present fi eld observations on fi ssure ridge travertines and related geochronological data. The latter were obtained using the U-Th disequi-librium method, which is the most widely used dating technique applied to speleothems and travertines. The analytical procedure followed Tuccimei et al. (2010); details are reported in the GSA Data Repository item (see footnote 1). All errors are indicated as 2σ (Table 1). We choose

travertine samples where the texture was the best for high precision.

To understand the temporal relationship be-tween the banded and bedded travertines, in the studied fi ssure ridges, we collected samples from fi ssures fi lled by banded travertine and from the adjacent bedded travertine. Samples with mini-mum detrital fraction (i.e., the most sparitic bed-ded travertines with no evidence of secondary recrystallization) were chosen to reduce errors in the age determination. Sampling sites are shown both in plan and cross-sectional views in Fig-ures 3–7, and related coordinates are reported in Table 1 and Table DR2 (see footnote 1) together with the complete results from dating analyses.

We compare our dating results with previous ones from the Akköy, Çukurbağ, and Kocabaş fi ssure ridges (Altunel and Karabacak, 2005;

Uysal et al., 2007, 2009). On the contrary, data from the Kamara fi ssure ridge offer the fi rst ages from this site. To understand the possible infl u-ence of paleoclimate on the fi ssure ridge growth, in Figures 8, DR1, and DR2 (see footnote 1), we compare our radiometric age data with major climate events recorded in speleothems from the Hulu cave (China), Sofular cave (Turkey), and other sites (Wang et al., 2001; Fleitmann et al., 2009). We also consider climate fl uctua-tions during the past 2500 years in Europe (Fig. 8B; Büntgen et al., 2011). From this comparison, we infer that most banded travertine samples fall in cold events during Quaternary time. In par-ticular, Figures 8 and DR1 (see footnote 1) show that our U-series dates for the banded travertine fall within the 16–30 ka time interval ( Table 1), with 6 out of 8 samples coincid ing with cold

Tripolis fault

Akköy fault

Hierapolis fault

B. Menderes River

Çürüksu stream

Kamara f-r

Akköy f-r

Kocabaş f-r

Çukurbağ f-r

0 5 10km

N

Güney

The GüneyWaterfall tufa

Yenice

Gölemezli

Akköy

Yeniköy

Tekkeköy

DereköyKaklik

Belevi

Kocabaş

Honaz

Gürlek

Pinarkent

Karateke

Sarayköy

Irliganli

Karahayit

Pamukkale

DENIZLI

DE

N

N

I

I

I

B

AS

ZL

ANKARA

Çökelez Mt.

Honaz Mt.

24°42°

40°

38°

36°

27° 30° 33° 36° 39° 42° 45°

study area

Legend of the Denizli basinAlluvium

TravertineQuaternary

Neogene unitsOther fissure ridgesStudied fissure ridges (f-r)

Normal faultSettlement

WaterfallHot springPre-Neogene basement

NAF

AE FKTG

B

NAF - North Anatolian FaultAbbreviations:

Tectonic map of Turkey

EAF - East Anatolian Fault KT - Karliova Triple junction

G - Gediz graben B - Büyük Menderes graben

Main fault Graben

20 data

master normal fault

Figure 2. Geological map of the Denizli Basin (Turkey) and distribution of main fissure ridges (the rose diagram shows the distribution of fi ssure ridge long axes in the region) and other travertine deposits in the basin (after Özkul et al., 2002). Rose diagram is done with Daisy software (Salvini et al., 1999). The simplifi ed tectonic map of Turkey shown in the rectangular inset is modified after Reilinger et al. (2006).



De Filippis et al.


climate events such as Heinrich events 1 (H1) and 2 (H2) and the Last Glacial Maximum (LGM) (Bar-Matthews et al., 1999; Wang et al., 2001). This temporal-paleoclimate correlation is also confi rmed at the regional scale, consider-ing results from the Sofular cave in Turkey (Fig. DR2 [see footnote 1]; Fleitmann et al., 2009). There are, moreover, two samples of bedded travertine (having ages of 27.5 ± 3.2 ka and 29.3 ± 1.6 ka; Fig. DR1 [see footnote 1]; Table 1) for which the analytical uncertainty prevents an un-ambiguous assignment to cold or warm periods. The age of the most recent sample (Kamara, 1.7 ± 0.1 ka) corresponds with a well-known dry period (Orland et al., 2009; Büntgen et al.,

2011; Fig. 8B). The general temporal correla-tion between banded travertine deposition in the Denizli Basin and cold climate events has been already ascertained by Uysal et al. (2007, 2009), whose geochronological data are also plotted in Figures 8, DR1, and DR2 (see footnote 1). Concerning the bedded travertine, one sample falls within the H6 cold period (ca. 60 ka). The ages of the other two samples have high ana-lytical uncertainties (36.4 ± 3.0 ka and 53.4 ± 1.6 ka; Fig. DR1 [see footnote 1]), so that their assignment to cold or warm periods is ambigu-ous; it is interesting to note, however, that 36 and 54 ka correspond with warm peaks during Quaternary time (Bar-Matthews et al., 1999).

As mentioned already, due to their sparitic versus detrital quality, travertine samples could not be chosen to systematically characterize the overall timing of fi ssure ridges. The available age data (Figs. 8, DR1, and DR2 [see footnote 1]) do not allow us to understand, for instance, what happened to the travertine deposition in the Denizli Basin during the cold Younger Dryas stadial (ca. 12.8–11.5 ka) or during the rela-tively warm Holocene time (e.g., the Holocene climatic optimum, ca. 9–5 ka). Our interpreta-tions (see the Discussion section) are therefore to be further constrained in the future when a new dating technique will allow for precise dat-ing of detrital travertines as well.

A

A′

20 cm

20 cm

20 cm

35 cm

Ka2

Ka4Ka5

Ka6

Ka7

12

9

7

29

12

933

31

29

10 m0

N

samplehot springfracture (35 cm = aperture)bedding attitude12

2m

2 m

?A (SW) (NE) A′

Ka2

Ka6

Ka4Ka7

waterfallsill-likestructure

banded travertinebedded travertinemeasured beddinginferred beddingsample

A

C D

E

B

NE SW SW SWNE NE5 m 30 cm30 cm

hot spring

central fissure

en echelonfractures

waterfall

Kamara fissure ridge

bandedtravertine

sill-like structure

tbanded ravertine b

ndd

ta

.a

er

v

ri

bandedtrave

t nebanded trave

ti er

ndetritus

bedded travertine bedded travertine bd

ee

dd

trv

re

ae

tin

bedding polesfracture planes

Figure 3. Sketches and photographs from the Kamara fi ssure ridge (see Fig. 2 for its location in the Denizli Basin). (A) Sketch (map view) of the Kamara fi ssure ridge. (B) Cross section (drawn in the fi eld from cross-sectional exposure observation and measurement) through the Kamara fi ssure ridge. Note the asymmetric profi le of the ridge. Stereoplot (Schmidt diagram, lower hemisphere) shows bedding poles and fracture planes. (C) Picture showing a panoramic view of the Kamara fi ssure ridge. Note the central fi ssure exposed along the ridge crest. (D) Banded travertine within the Kamara fi ssure ridge. Note that the vein curves upward to become a sill injected along the strata of the adjacent bedded travertine. (E) Detailed view of the vein of Figure 3D showing the crosscutting relationship with other veins and the adjacent bedded travertine.





Kamara

The Kamara fi ssure ridge is part of the Yenice travertine complex (1.5 km2 in areal extent), which is located in the Büyük Menderes River valley, ~3 km to the northeast of Yenice village, not far from the tourist site of the Güney Waterfall

tufa (Fig. 2; Özkul et al., 2010). The fi ssure ridge is located on Neogene fl uviolacustrine deposits (Fig. 2). Tectonically, the ridge is located on the footwall of the Tripolis fault (Çakir, 1999), which is, in turn, the hanging wall of other normal faults located northeast of Kamara (Yalçınlar, 1983; Kaymakçı, 2006) (Fig. 2). The Tripolis fault is

the northwestern segment of the Pamukkale fault system (Turgay et al., 1985; Çakir, 1999).

Kamara was an active fi ssure ridge, where travertine was being deposited from emerging carbonate-rich hot waters, until 1997; however, drilling to obtain hot water for a thermal spa depleted the spring. The fi ssure ridge (Fig. 3A)

banded travertine

bedded travertine

sample

fracture

inferred beddingmeasured bedding

Roman quarryA′

A34

822121228

31

3787

4453

13915

19

2821

272126 24

26 255

7676

20 33

15

22

173218

27

20 20 24

25

5

Pa5

Cuk1/Cuk3

Cuk5 hot springs (T~ 56°C)sample

bedding attitudehorizontal bedding

21

2 m2

m

A (SSE) (NNW) A′

Pa5


50 m0

N

D

E F

N SS N

S N

1 m

20 cm

2 m

t rt nbanded rave i e

u ureds tdaboun ry

ngbeddi

bedded ravt ertine

bedded rav it ert ne

lens

b n da detrav iert ne

bedded travertine

C

S N1.5 m

bedded travertine

n e tr e tiba d d av r ne

fissure

fissure

bedded travertine

V-growth

A

BFigure 4. Sketches and photographs from the Çukurbağ fi ssure ridge (see Fig. 2 for its location in the Denizli Basin). (A) Map view of the Çukurbağ fi ssure ridge. Stereo-plots (Schmidt diagram, lower hemisphere) show bedding poles and fracture planes. (B) The A-A′ cross-section (drawn from exposure observation and measurement), perpendicular to the long axis of the fi ssure ridge, shows the asymmetric profi le of the ridge. Three subvertical fi ssures (partially quarried) in the central part of the cross section are fi lled by banded travertine (see also Fig. 4C). The contact between bedded and banded travertine is nearly orthogonal. (C) Westward view of the Roman quarry, located in the central portion of the fi ssure ridge, where most of the banded travertine occurred. (D) Enlarged section of Figure C, where a block of lens-shaped bedded trav-ertine (~2 m in length and 0.5 m in width) is surrounded by subvertical banded trav-ertine. (E) Subvertical banded traver-tine partly topped by bedded travertine. (F) Eastward view of the internal exposure of the Çukurbağ fi ssure. A subvertical large V-shaped (i.e., upward widening) banded travertine vein cuts through adjacent bed-ded travertine, which constitutes the fl anks of the fi ssure ridge.



De Filippis et al.


trends N125° (the long axis) and is ~63 m in length, 15 m in width, and 6 m in maximum height over the surrounding plain (Table DR1 [see footnote 1]). Figure 3B displays a cross section (A-A′) through the fi ssure ridge, show-ing an asymmetric profi le, with the northeastern fl ank slightly steeper than the southwestern one. In particular, the northeastern fl ank dips toward

the northeast by ~33°, and the southwestern fl ank dips toward the southwest by ~29°. The top of the ridge shows a system of en-echelon open fractures with an average strike of about N120° (Fig. 3A). Fracture aperture varies be-tween about a few millimeters and a maximum of ~20 cm, and it is usually larger in the central section of the fi ssure ridge than near its lateral

closures (Fig. 3C). Both fl anks of the fi ssure ridge consist of bedded travertines dipping away from the axial fi ssure, with the northeastern fl ank characterized by the presence of a fossil waterfall, determining a marked asymmetry of the Kamara fi ssure ridge.

The bedded travertine, which is very similar to the one observed in the three fi ssure ridges

17 89 3

8915 18 1919

15

1520

26

25

168

8

9 14

16

22

50

408022

70

485241

838

6540

1280

19 1212

3414Ak15 Ak16

Ak17

Ak18 Ak19

Ak27

Ak21Ak22

Ak24Ak25

Ak23 Ak2

Ak26 Ak3

BCD E′ F′ G′

G

F

E

D′ C′ B′

A A’

0 500 m

250 m

25m

150 m 0

N

N

N

N

B-B′C-C′D-D′

E-E′ F-F′ G-G′

A (NW) (SE) A′450 m

170 1158520 60

TravertineFissureBedding attitude

Sample

29

2 m

2m

not t

osc

ale

5m

5 m

not

sa

tocl

e

fracturesample

bedded travertine

banded travertinedeformed travertine

injection veinloose debris/carved travertine

Ak23Ak24

Ak3

Ak2

Ak25

Ak26 Ak27

Ak22

Ak21

Ak17

Ak18Ak19Ak16

Ak15

B

C

D

G

F

E E′

F′

G′

D′

C′

B′

A B

C

E

D

carved banded travertine

carved banded travertine

paraconformity

unconformity

unconformity

unconformity


Figure 5. Sketches from the Akköy fi ssure ridge (see Fig. 2 for its location in the Denizli Ba-sin). (A) Map view of the Akköy fi ssure ridge. The rectangular box shows the study area of the fi ssure ridge, with two main en-echelon fi ssures. (B) Longitudi-nal cross section (A-A′) of the northern branch of the Akköy fi ssure ridge (drawn from ex-posure observation and mea-surement). The quarry sites are here indicated for a better comprehension of the three-dimensional views of parts D and E. (C) Detailed plan view of the study area illustrating the main fi ssures, bedding atti-tude, and sampling sites (black circles). Stereoplot (Schmidt diagram, lower hemisphere) shows bedding poles and frac-ture planes. (D) Three cross sections (drawn from exposure observation and measurement) form a three-dimen sional car-toon of the northern part of the study area (see B and C for cross-section location). Note two main fi ssures fi lled by par-tially quarried banded traver-tine. The fi ssures tend to widen upward with a general incli-nation toward the northeast. In places, veins are injected within (both cutting through and paralleling) the travertine beds. (E) Three cross sections (drawn from exposure observa-tion) form a three-dimensional cartoon of the southern part of the study area (see B and C for cross-section location). Note the unconformable contact between two adjacent fi ssure ridges in the eastern part of the cross sections.





depicted in the following sections, consists of a porous layered travertine with microstructures such as microterracettes, shrubs, and crystalline crusts, which are typical of high-energy slope or cascade environments (Guo and Riding, 1998), where travertine grew as a fl owstone with a clinostratifi cation (Fig. 3).

The Kamara banded travertine is exposed along the southwestern side of the axial fi ssure. The calcite crystals in the banded structure are perpendicular to the vertical fi ssure walls and arranged in parallel, subvertical layers with al-ternating white, red, and brown colors (Figs. 3D and 3E) induced by iron (Özkul et al., 2002). The contact between bedded and banded trav-ertines, when visible, is nearly orthogonal, with the banded travertine cutting through the bedded one (Fig. 3). At least in one case, we observed a

set of banded travertine layers curving to form a sill along the strata of the bedded travertine (Fig. 3D). Another example is a vein cutting out from the axial zone and continuing toward the north-eastern fl ank at a low angle (Fig. 3B).

Travertine from the Kamara fi ssure ridge has never previously been dated. We collected and dated two samples from the banded travertine of Kamara (Ka2 and Ka4) (Figs. 3A and 3B). We ob-tained ages of 1.7 ± 0.1 ka and 2.5 ± 0.1 ka for the two samples (Table 1). The adjacent bedded trav-ertine is too rich in detrital Th to be reliably dated.

Çukurbağ

Çukurbağ is an extinct fi ssure ridge located within the Pamukkale fi ssure zone of Altunel (1994), ~1 km northwest of the active Pamuk-

kale travertine terraces, a World Heritage site of the U.N. Educational, Scientifi c and Cultural Organization (UNESCO) (Fig. 2). Çukurbağ is part of the Pamukkale travertine deposit, which occupies an area of ~7.6 km2 overlying Neogene sediments. The Pamukkale fault system, named as “Pamukkale range front fault” by Altunel and Hancock (1993a), which strikes NW-SE and dips toward the southwest (Çakir, 1999), is the main tectonic feature in the area. It consists of two fault branches, namely, the Hierapolis and Akköy normal faults. The Çukurbağ fi ssure ridge, in particular, grew on the hanging wall of the Hierapolis fault (Fig. 2).

In the Çukurbağ area, there are several fi s-sure ridges (Altunel and Hancock, 1993a), of which the largest one, named Çukurbağ fi ssure ridge, trends E-W and has a length of ~350 m, a width of 36 m, and a maximum height of 11 m (Table DR1 [see footnote 1]). Approximately 70 m toward the south, there is another small fi ssure ridge (Fig. 4A; Table DR1 [see footnote 1]), which we name the Çukurbağ 2 fi ssure ridge. A thermal spring is located at the eastern end of the Çukurbağ fi ssure ridge (Fig. 4A), and it has a temperature of 56 °C (for more information, see Gökgöz, 1998; Kele et al., 2011). An axial fi ssure is exposed on the ridge crest with a strike ranging between N80° and N110° (Fig. 4A). In places, the axial fi ssure is accompanied by minor frac-tures occurring on both fl anks (Fig. 4A). Both the main and minor fi ssures and fractures are nearly vertical and are fl anked by bedded travertines dipping away from the axial fi ssure. The fi ssure aperture, as well as the height and width of the fi ssure ridge, decreases from the central portion toward the lateral closures, where the travertine becomes nearly horizontal (Fig. 4A).

Figure 4B shows a cross section (A-A′) per-pendicular to the long axis of the E-W–trending Çukurbağ ridge. In the cross section (Fig. 4B), the fi ssure ridge profi le is markedly asymmetric, with the northern fl ank steeper than the south-ern one. In this cross section, the fi ssure ridge reaches its maximum height (11 m) and width (36 m). Both fl anks are tilted and, in places, are characterized by fractures that isolate slabs of tilted bedded travertine. In particular, the south-ern fl ank is clearly tilted with beds dipping be-tween 21° and 34° (Fig. 4A). The northern fl ank is also tilted with a maximum dip of 44°. The central part of the cross section is characterized by three subvertical fi ssures fi lled with banded travertine. At present, the banded travertine is visible only in the northernmost fi ssure; the other two central fi ssures have been quarried (Fig. 4B). The fi ssure ridge fl anks, in contrast, consist of bedded travertine. The contact be-tween bedded and banded travertine is nearly orthogonal. In a cross-sectional view (Fig. 4B),

1 m

10 cm

20 cm20 cm

0.5 m

10 m

x

y

z

A B

C

DE

FNE

NE

SW

SW

NENE SW SW

imbricatedtravertine

injection vein

abortedsill-like structures

fissure unconformity

fissure

fissure

injection vein

sill-likestructure

sill-likestructure

injectionveins

CRESTAREASW FLANK NE FLANK

Figure 6. Field photographs of some structural features from the Akköy quarry area (see Fig. 2 for its location in the Denizli Basin). (A) A vertical vein system fi lled by banded trav-ertine cutting through older beds of travertine (bedded travertine). The subvertical vein on the right appears to be injected from the vertical vein on the left. (B) Enlargement of A, showing the injection relationship between banded and bedded travertines. The banded travertine consists of reddish and brownish parallel bands of sparitic calcite usually encom-passing imbricate slabs of bedded travertine (i.e., paper-thin raft). This banded travertine is injected, in its inner portion, by younger generations of white calcitic veins. (C) Lateral section of a vein showing a vertical repetition of subhorizontal sill-like structures. Note the three-dimensional view. (D) Photograph showing an injection vein characterized by fes-toon-like walls, where the acute apexes may be interpreted as incipient, but aborted sill-like structures (see further details in E). (E) Enlarged image from D. Note the festoon-like walls. (F) Often, the quarry activity allows geologists to observe nice exposures of the internal fi s-sure ridge structure and recognize fi ssures, unconformities, and vein networks. We drew the E-E′ cross section of Figure 5E after the analysis of this exposure.



De Filippis et al.


the banded travertine occupies ~15% of the fi s-sure ridge area.

At Çukurbağ, we collected and dated two sam-ples, one from the bedded (Pa5) and one from the banded travertine (Cuk1) (Figs. 4A and 4B). We obtained ages of 24.1 ± 0.3 ka and 60.0 ± 1 ka for the banded and bedded travertines, respectively (Table 1). Ages between 24.9 ka and 25.2 ka from the Çukurbağ banded travertine published by Uysal et al. (2009) are consistent with our re-

sults (Figs. 8, DR1, and DR2 [see footnote 1]) as well as ages (with error bars around 4%–5%, however) determined by Altunel and Karabacak (2005) (i.e., 21.5 ± 1.0 ka and 29.5 ± 1.4 ka).

Akköy

The Akköy fi ssure ridge is part of the Kara-kaya Hill, a system of NW-SE–elongated reliefs between the villages of Akköy and Pamukkale.

The Karakaya Hill is situated along the south-west part of the Pamukkale travertine plateau (Fig. 2). This part of the Denizli Basin is charac-terized by a set of fi ssure ridges, either active or extinct, most of which are found in the stepover zone between the Akköy and Hierapolis faults, where an intense fracture network likely en-hanced the leakage of mineralizing fl uids. The Akköy fi ssure ridge is inactive and located on the hanging wall of the Hierapolis fault, which is 2 km away from the fi ssure ridge. Next, we present observations made on the northern seg-ment of the Akköy fi ssure ridge (Fig. 5A), where excellent exposures of travertine are available in active quarries.

The Akköy fi ssure ridge trends NW-SE and has a length of ~1900 m, a width of 200 m, and an average height of 25 m (Table DR1 [see footnote 1]). In the studied part of the Akköy fi ssure ridge, two main en-echelon fi ssures run along the crestal region of the ridge with a NW-SE trend (N145°) (Fig. 5A). These axial fi ssures are accompanied by minor fractures. Both main and minor fractures are nearly verti-cal in the central portion of the fi ssure ridge and become inclined toward the northeast, forming a fan-shaped pattern (Figs. 5D and 5E). These fractures are fi lled by banded travertine (at present, almost entirely quarried; Figs. 5 and 6) and are fl anked by bedded travertine, which dips away from the axial zone.

Figure 5B shows a longitudinal cross sec-tion of the Akköy ridge, whereas Figure 5C is a plan view of the study area with dip domains and sampling sites. Figures 5D and 5E show two three-dimensional views, each one based on three cross sections elaborated perpendicularly to the Akköy ridge.

Cross sections of Figure 5D show two main fi ssures dipping toward the southwest and other associated fractures and veins. These fi ssures are fi lled by banded travertine and can be followed upward from cross-section D-D′ to cross-sec-tions C-C′ and B-B′, where most banded trav-ertine has been quarried. In general, the fi ssures fi lled by banded travertine tend to widen upward and to become inclined toward the northeast from a subvertical attitude in the central portion of the ridge. Blocks of bedded travertine entirely surrounded by banded travertines are visible on the cross sections (see cross-section D-D′ in Fig. 5). In particular, in cross-section D-D′, the block of bedded travertine lined by banded trav-ertine is pervaded by veins (i.e., fi lled by banded travertine) both cutting through and paralleling (i.e., travertine sills) the travertine beds.

The three-dimensional view including cross-sections E-E′, F-F′, and G-G′ (Fig. 5E) shows features similar to the ones observed in Figure 5D, with fi ssures fi lled by banded travertine

Ko2 Ko4Ko1

A (SW) (NE) A′quarried area

sill-like structurebookshelfbookshelfbookshelfstructurestructurestructure

(NE) B′B (SW)

fissure

A79

42

17

60

4365

6837

8135

3029

A′

B′

BKo2

Ko4Ko1 0 100 m

0 1000 m N

N

N 5m

5 m

fractureshear fracture

sample

bedded travertine

banded travertineimbricated trav./breccia zone

loose debris/carved trav.injection vein

travertinefissurebedding attitudesample

29

NENE SWSW 1 m1 m

fracture

brecciazone

bandedtravertine

shear fracture

sill-likestructure

loose debrissi l-l ke truc u e

l i s t r in

i

jectionve

n

banded travertine

breccia zone

A

B

C

FD

E


Figure 7. Sketches and photographs from the Kocabaş fi ssure ridge (see Fig. 2 for its loca-tion in the Denizli Basin). (A) Map view of the Kocabaş fi ssure ridge. The rectangular box (in black) shows the study area of the fi ssure ridge, with an inferred fi ssure path (dotted line). (B) Detailed plan view of the study area illustrating the inferred fi ssure (the dotted line with diagonal direction), bedding attitude, and sampling sites (gray circles). Stereoplot (Schmidt diagram, lower hemisphere) shows bedding poles and fracture planes. (C) Two cross sections (drawn from exposure observation and measurement) form a pseudo–three-dimensional cartoon of the southern part of the Kocabaş fi ssure ridge (see part B for cross-section location). Note a series of three fi ssures fi lled by partially quarried banded travertine covered, in places, by loose debris from quarry activity. The fi ssures tend to widen upward with a general inclination toward the southwest. The fi ssure in the middle of the A-A′ cross-section (close to Ko2 sample) is characterized by a thick vein (~1.5 m in thickness) injected within the travertine beds, both cutting through and paralleling them. The main fi ssure zone (the inferred path of which is shown in A and B) is visible in the northeastern end of the A-A′ cross section and in the southwestern end of B-B′ cross section (see the thick black dotted line linking the two cross sections, representing the inferred fi ssure from parts A and B). (D) Part of a large injection vein (~1.5 m in thickness) cutting through preexisting strata of bedded travertine. Upward, the injection vein is curved to form a sill-like structure within the bedded travertine. (E) Photograph showing an enlargement of D. Note the travertine breccia within the vein. (F) A joint system with associated injection veins forming sill-like structures (see the southwestern end of the A-A′ cross section for its location).





(now mostly quarried) slightly widening up-ward and becoming inclined toward the north-east, and with the bedded travertine dipping away from the axial fi ssures. One important feature (cross-section E-E′) is a bedded traver-tine unconformity marking the contact between two adjacent fi ssure ridges coalesced to form this branch of the Akköy fi ssure ridge (Fig. 5A). Other unconformities and paraconformities are visible in cross-sections D-D′, F-F′, and G-G′, marking the contact between travertine deposits formed in temporal successions.

At Akköy, we collected and dated nine sam-ples, seven from the banded travertine (Ak15, Ak16, Ak17, Ak21, Ak22, Ak23, Ak26) and two from the bedded travertine (Ak24 and Ak27) (Figs. 5C–5E). Our analyses provided ages be-tween 16.3 ± 0.3 ka and 29.3 ± 1.6 ka for the banded travertine, and ages of 36.4 ± 3 ka and 53.4 ± 1.6 ka for the two bedded travertine sam-ples (Table 1). Uysal et al. (2007, 2009) dated the banded travertine within the southeastern branch of the Akköy fi ssure ridge (south of our study area) and obtained ages between ca. 49 ka and 74 ka, whereas ages obtained by the same authors on the banded travertine from our study area are between ca. 21 ka and 26 ka. Moreover, Altunel and Karabacak (2005) obtained an age of 34.9 ka for a single sample from the banded travertine in our study area.

Kocabaş

Several inactive fi ssure-ridges occur close to the village of Kocabaş along the Denizli-Afyon Road in the eastern portion of Denizli Basin (Fig. 2) (Altunel and Karabacak, 2005). We studied one of these fi ssure ridges, hereafter named the Kocabaş fi ssure ridge, where excel-lent exposures are present in two active quarries (see also Altunel and Karabacak, 2005). In par-ticular, these exposures allowed us to analyze the

eastern portion of the ridge (Fig. 7). Concerning the Kocabaş fi ssure ridge, Hancock et al. (1999) indicated that this fi ssure ridge was buried be-neath at least 200 m of continental sedi ments and then exhumed to the present setting.

The Kocabaş fi ssure ridge trends WNW-ESE for ~2700 m in length, 300 m in maximum width, and 20 m in height. Figure 7A shows a plan view of the fi ssure ridge, whereas in Figure 7B, the study area is magnifi ed to show bedding attitudes and sample location. In Figure 7C, two artifi cial exposures, ~120 m (A-A′) and 30 m (B-B′) long, across the eastern portion of the fi ssure ridge are shown. Four main fi ssures fi lled by banded travertine (at present almost entirely quarried) are visible in these cross sec-tions (A-A′ and B-B′ in Fig. 7C). These fi ssures are rather thick (up to ~10–12 m) compared to the previously described fi ssure ridges (Figs. 3–6). In the north-northeastern part of the cross sections (Fig. 7C), a block of bedded travertine is entirely surrounded by banded travertine. In the south-southwestern part, a highly inclined fi ssure fi lled with banded travertine is curved to form a gently dipping thick sill (~1.5 m in thickness) within the SSW-dipping bedded trav ertine (Fig. 7D). Other evidence of banded traver tine within the surrounding bedded trav-ertine is present on the southwestern termina-tion of the A-A′ cross section (Figs. 7C and 7D). On the northeastern termination of the A-A′ cross section (Fig. 7C), a bookshelf struc-ture, including some shear fractures, affects the bedded travertine.

At Kocabaş, we collected and dated three samples from the banded travertine (Ko1, Ko2, and Ko4) (Figs. 7B and 7C). Our analyses pro-vided two ages exceeding the range of the U-Th disequilibrium dating method (>350 ka) and one age of 160.0 ± 5 ka (Table 1). In the same study area, Altunel and Karabacak (2005), for two samples of banded travertines, obtained one

age of ca. 105 ka and one >350 ka. Some addi-tional age data were obtained ranging from 300 to 350 ka using the thermoluminescence (TL) method (Özkul, 2011, personal commun.).

MICROSTRUCTURAL ANALYSIS OF THE BANDED TRAVERTINE

Microstructural observations from 25 thin sections were made from veins filled with banded travertine (Fig. 9). The thin sections, in particular, are from a total of 11 veins affecting three of the four previously described fi ssure ridges (Kamara, Çukurbağ, and Akköy). The representativeness is limited by the mentioned extensive quarries. The thin-sectioned samples are, in fact, from the boundaries of the largest quarried veins or from minor veins accompany-ing the largest ones.

The fabric observed in the analyzed veins is rather complex and heterogeneous with several textural zones. In synthesis, main common fea-tures are as follows.

(1) All the veins are composed of visible bands both under the optical microscope and with the naked eye (1–25 mm thick) with brownish-yellowish to white colors. Such banded fabric is produced by the multiphase growth history of the veins (Altunel and Karabacak, 2005; Uysal et al., 2007; compare with Nuriel et al., 2011, 2012).

(2) When elongated or fi brous, the calcite crystals are perpendicular to the vein walls (Figs. 9A and 9B), attesting that the veins were formed by opening and crystals grew synkinematically.

(3) Under the microscope, veins appear to be composed of several textural zones (Fig. 9), which are depicted in more detail in the follow-ing points. Each vein has its particular sequence and pattern of textural zones, but the various types of zones (except the exotic zones that are rarer; see following) are present in all the veins.

(4) From the size and width of calcite crystals, which increase in the direction of crystal growth, we infer that the crystals grew toward the center of the vein (symmetric syntaxial veins, Fig. 9A) or from one wall toward the other wall (asym-metric syntaxial veins, Fig. 9C). The general syntaxial growth of veins is also confi rmed by radiometric dating of vein bands (Altunel and Karabacak, 2005; Uysal et al., 2007, 2009).

(5) Three main textural zones are recognized in all veins at the microscopic scale: hiatus and crystal termination zones, zones of dynamic crystallization, and exotic zones (Fig. 9). Hiatus zones may be thick and complex with various layers including micro- and macrocrystalline euhedral calcite and impure calcite (Figs. 9D and 9E). This type of hiatus zone markedly sep-arates adjacent textural zones usually consisting of elongate to fi brous crystals. Otherwise, hiatus

TABLE 1. AGE OF TRAVERTINES SAMPLED IN THE DENIZLI BASIN*†

Fissure ridge Sample Travertine type Age (ka)1.0±7.1dednaB2aKaramaK1.0±5.2dednaB4aKaramaK

Çukurbağ 3.0±1.42dednaB1kuCÇukurbağ 1±06deddeB5aP

6.1±3.92dednaB51kAyökkA5.0±2.02dednaB61kAyökkA3.0±3.61dednaB71kAyökkA2.3±5.72dednaB12kAyökkA1.0±4.12dednaB22kAyökkA6.0±5.32dednaB32kAyökkA6.1±4.35deddeB42kAyökkA4.0±1.42dednaB62kAyökkA

3±4.63deddeB72kAyökkAKocabaş dednaB1oK ≥350Kocabaş 5±061dednaB2oKKocabaş dednaB4oK ≥350

*See Table DR2 (see text footnote 1) for a complete list of data.†Errors are always quoted as 2σ.



De Filippis et al.


zones are very thin with a serrate (rarely linear) profi le and are assumed to mark short pauses of crystal growth (Fig. 9D). These hiatus zones are usually accompanied by pockets of micro-crystalline blocky calcite and separate adjacent zones of dynamic crystallization with very simi-lar textures.

(6) Zones of dynamic crystallization include elongate-blocky to fi brous crystals with length to width ratio between ~5 and more than 100 (Fig. 9D). Calcite crystals show growth com-

petition textures and more rarely elongate to fi brous columnar textures (denoting no or limited competition) with rather sharp crystal boundaries. The analyzed zones of dynamic crystallization often show a marked homoge-neous extinction under polarized microscopic light. Moreover, some veins are characterized by thin-twinned crystal morphology as well as stylolitic hiatus zones (Fig. 9D), suggesting postformation defor mation, possibly due to vein sealing (see also Nuriel et al., 2012).

(7) Exotic zones (Fig. 9F) are the rarest textural zones occurring only in a few of the studied thin sections. These zones include ex-otic grains (either rounded or angular) from the bedded travertine either in contact with one another or separated and surrounded by a cement of microcrystalline or coarse cal-cite. Voids in the exotic zones indicate that dilation was accompanied by cementation. A sec ondary vein cuts through the exotic zone shown in Figure 9F.

Figure 8. (A) Figure modifi ed after Uysal et al. (2009) and references therein. Compari-son of U/Th ages (diamonds with horizontal error bars = 2σ) of the bedded and banded travertines from the Denizli Basin with high-resolution paleo climate records. U/Th age data are from this work (Table 1) and from Uysal et al. (2007, 2009). Gray bars show main correlations with high-resolution climate records (Figs. DR1 and DR2 [see text footnote 1]). Bands with dashed lines indicate Younger Dryas (YD) (Wang et al., 2001), Hein-rich events (H) (Wang et al., 2001), and Last Glacial Maxi-mum (LGM) (Mix et al., 2001). Also shown are low summer insolation (40°N, June) periods in the Northern Hemisphere (Berger, 1978), low δ18O val-ues of Greenland ice (GISP2, 1997), and high δ18O values of Hulu/Sanbao cave stalagmites (Wang et al., 2001, 2008). Num-bers indicate Greenland ice sheet (GISs) and their correla-tion with Hulu data. The δ18O values of the numbered peaks are reversed for Hulu (increas-ing down) in comparison with GISP2 (increasing up). San-bao δ18O record (broken lines between 2.14 and 11.6 ka and 76 and 90 ka) is plotted 1.6‰ more positive to match with the higher Hulu values. (B) Figure modifi ed after Büntgen et al. (2011). Comparison of U/Th ages (diamonds with horizontal error bars = 2σ) of the banded travertine from the Kamara fi ssure ridge with high-resolution paleoclimate records of Europe during the past 2500 years. U/Th age data are from this work (Table 1). Reconstructed AMJ (April-May-June) precipitation totals (top) and JJA (June-July-August) temperature anomalies (bottom) with respect to the 1901–2000 period are also shown. The curves are 60 years low-pass fi lters of the original curves shown in Büntgen et al. (2011). Periods of demographic expansion, economic prosperity, and societal stability are reported, as are periods of political turmoil, cul-tural change, and population instability. VPDB—Vienna Peedee belemnite; VSMOW—Vienna standard mean ocean water.

0 10

–10

–8

–6

–4

–36

–40

–44

20 30 40 50 60 70 80 90

460

480

500

520

LGM

12

3 4 5 6 7 891011

1213

1415 1617 1819 20

21

12

3

4 567

8910 11

12 13 1415

16 17

18 1920 21

50

150

250

350

–3

–2

–1

0

1

2

3

400 200 1/1 200 400 600 800 1000 1200 1400 1600 1800 2000

banded travertine fromKamara fissure ridge

(this work)

banded travertine

banded travertine (Uysal et al., 2007, 2009)bedded travertine

this work

U/Th-dates (this work)


Uysal et al., 2007, 2009

YD H1 H2Hulu cave

GISP2

H5 H6

Age (ka)

Year (B.C.E. / C.E.)

2δ

δPSIG

18)

WOMSV,(

O‰

evaculuH18

)BDPV,(

O‰

mW(noitalosn I

–2)

ColdDry

Late IronAge

RomanEmpire Medieval Period

Tnoit atipicerP

)m

m(sl ato

)C°(

sei lamon

Aerutare p

meT

noisnapxE

citleC

enimaFtaer

Ghtae

Dkca l

B

raW’sr aeY

0 3

noita rgiM

nredoM

d oirePnoitargi

M

A

ts euqn oC

namo

R

B

insolation

Sanbaocave

Sanbaocave





BANDED-BEDDED TRAVERTINE GEOMETRIC RELATIONSHIPS

Characterization of the geometric and cross-cutting relationships between banded and bedded travertines is fundamental to under-standing the growth of fi ssure ridges. Here, we summarize our fi eld observations of the bed-ded-banded travertine geometric relationships.

(1) Banded travertine forms usually high-angle veins fi lled by growth bands (often syntaxial as shown in the previous section) of sparitic calcite (Altunel and Hancock, 1993a). Bedded travertine forms low-angle thick car-bonate strata (Figs. 3D and 6F).

(2) The banded travertine either crosscuts preexisting strata or forms sill-like structures paralleling the strata (Altunel and Hancock, 1993b) (Figs. 3D, 4D, 6, and 7). In places (in cross-sectional view), bedded travertine blocks are entirely surrounded by banded travertine veins (Altunel and Hancock, 1996) (Fig. 4D).

(3) Veins of banded travertine usually form a fan-shaped pattern with a subvertical large V-shaped vein in the core and thinner inclined veins in the fl anks (Figs. 4B and 4F).

(4) Bedded travertine is often crosscut by a pervasive vein network of banded travertine usually departing from the main veins (Figs. 5 and 6A–6E).

(5) The banded travertine veins belong to dif-ferent generations, with younger veins cutting preexisting ones (Altunel and Karabacak, 2005; Uysal et al., 2011).

(6) Frequently, the boundary walls of the in-jection veins are undulated rather than linear. Such geometry is hardly explicable solely by a brittle process of fracturing, suggesting that veins formed by brittle cracking and were then chemically weathered (i.e., their walls; Figs. 6A–6E; e.g., Billi et al., 2007). In some other cases, in contrast, walls of injection veins are festoon-like, where the festoon acute apexes may be interpreted as incipient but aborted sill-like structures (Fig. 6D).

(7) Some veins show fl uid-like structures (formed during fl uidization). This can be inter-preted as imbricate clasts of bedded travertines occurring within some veins, in places associ-ated with pockets of exotic material (Uysal et al., 2009) (Figs. 6B and 7E).

(8) Locally, the banded travertine is sutured by bedded travertine (Altunel and Hancock,

1993b). This younger travertine, in turn, can host even younger veins of banded travertine, demonstrating a temporal succession of bed-ded and banded travertine depositional events (Fig. 4E).

CARBON AND OXYGEN ISOTOPE DATA

Figure 10 shows results from the oxygen and carbon stable isotope analyses. Delta values fall between +2.9‰ and +7.8‰ versus Peedee belemnite (PDB) for the carbon isotopes and between –14.1‰ and –7.1‰ versus PDB for the oxygen isotopes. Figure 10 shows a gen-eral positive correlation between δ13C and δ18O data. Travertines from Kamara, Çukurbağ, and Kocabaş display minimal carbon isotope vari-ability, whereas a larger variability character-izes the Akköy travertines. The variability of the oxygen isotope data is somewhat marked for all the fi ssure ridges, there being a difference be-tween the banded and bedded travertines in the Kamara, Çukurbağ, and Kocabaş deposits. By contrast, the difference between the two types of Akköy travertine, which has the largest data set (Table DR3 [see footnote 1]), is minimal.

500 µm

500 µm 250 µmA B C

D

E

1 mm

500 µm

HR

GD

GCHR HRGC

H

FC

GD

GDBC

CC GC

H

GDGC

H

BCFC

GDFC

BCBC

H

FC

FC

H

F

500 µmVo

Ce

CB

BC = blocky crystals; CB = clast of bedded travertine; CC = columnar crystals; Ce = cement; EC = elongate crystals;FC = fibrous crystals; GC = growth competition; GD = growth direction; H = hiatus and hiatus zone;HR = host rock; FC = fibrous crystals; Vo = void

EC

Figure 9. Microphotographs in cross-polarized light of veins filled by banded travertine. (A) Calcite growth bands from the host rock toward the vein center (Çukurbağ, sample Cuk1). (B) Calcite growth bands from the host rock toward the vein center. Note the alternation of elongated-fi brous and blocky crys tals (Çukurbağ, sample Cuk3). (C) Calcite growth bands from one wall toward the op-posite one (Kamara, sample Ka4). (D) Photomosaic showing growth phases with long fi brous crystals in competition with one another (Akköy, sample Ak26). (E) Photomosaic show-ing growth phases with differ-ent textures (blocky and fi brous crystals) separated by marked hiatus zones (Çukurbağ, sample Cuk3). (F) Exotic zone includ-ing clasts of bedded travertine surrounded by microcrystalline cement. Note that the clasts are not in contact with one another. Note also the presence of voids (Akköy, sample Ak3).



De Filippis et al.


The difference between the banded and bedded travertines for Kamara, Çukurbağ, and Kocabaş (1‰–5‰ in δ18O versus PDB) may be ascribed to a difference in the precipitation conditions (e.g., temperature, isotope composition of pre-cipitation water). The isotope data of Akköy indicate that the origin and temperature range of the water from which the bedded and banded travertines precipitated were the same.

The travertines we studied were also sampled and isotopically analyzed by Uysal et al. (2007). Their results are plotted in Figure 10 together with our data. The data provided by Uysal et al. (2007) are from banded travertines sampled in the Akköy and Çukurbağ fi ssure ridges. The range of the carbon isotope data does not ex-pand when the data by Uysal et al. (2007) are added, whereas that of the oxygen isotopes widens toward negative values up to –16.3‰ (Fig. 10). The positive correlation of the isotope data in the δ13C versus δ18O plot is less evident when the data by Uysal et al. (2007) are added. Indeed, if the data from the banded and bed-ded travertines are taken separately, and if we consider that data by Uysal et al. (2007) derive exclusively from the banded travertine, the cor-relation (in the δ13C vs. δ18O plot) is no longer evident for banded travertines, but it is still pres-ent for bedded travertines (Fig. 10). This last correlation may be the expression of the differ-

ent depositional facies present in the geothermal depositional system, as already proposed by Kele et al. (2008, 2011). In this view, different δ13C values represent a signifi cant downstream increase due to continuous CO2 degassing (es-timated to be ~6‰ at Pamukkale; Kele et al., 2011). The downstream increase in δ18O values results, instead, from the superimposed isotope effects caused by carbonate precipitation, evap-oration, and decrease in temperature.

The range of carbon isotope data (Table DR3 [see footnote 1]) falls within the fi eld of thermo-gene travertines according to the distribution of δ13C (PDB, ‰) of travertines and allied carbon-ates based on site/sample means of the Pente-cost (2005) compilation. Generally, the carbon source of travertines with such an isotope composition is considered to derive from CO2 originating from limestone decarbonation (Turi, 1986), often with a signifi cant magmatic component (Pentecost, 2005) and water-rock inter action (carbonate dissolution). Rapid CO2 evasion contributes to the carbon isotope en-richments encountered in travertine formation. In our case, we may argue that dissolution of the limestones and marbles in the substratum of the Denizli Basin and re-precipitation along the ascent of deeply circulated hot fl uids cannot have dominantly contributed to the travertine formation. Indeed, such a process would have

produced δ13C values for the travertines close to those of the limestones and marbles, the δ13C values of which range, on the contrary, between about –1‰ and +2‰ (Attanasio et al., 2006; Uysal et al., 2007). Thus, the carbon isotope values for the travertines collected in our study sites must have been affected by a major contri-bution from another carbon source, likely in the form of dissolved CO2, which is very abundant in the modern active travertine precipitation sys-tem in the Pamukkale area (Dilsiz et al., 2004).

The origin of CO2 can be determined by using the carbon isotope composition of the travertine. If we assume that there is isotopic equilibrium during precipitation of the traver-tine between CO2 and calcite, it is possible to calculate the δ13CCO2

from our data at the range of temperatures that are representative of the Denizli hydro thermal system (25–58 °C; Dilsiz , 2006) by using the CO2/CaCO3 temperature equations of Bottinga (1968). Unfortunately, the CO2 degassing most certainly enriched 13C in the residual CO2 in the precipitation solution, thus modifying the pristine carbon-CO2 isotope composition (Gonfi antini et al., 1968). Panichi and Tongiorgi (1976) proposed an empirical equation (δ13C[CO2] = 1.2 × δ13C[trav] –10.5) that correlates the carbon isotopes of the travertine with those of the parental CO2 after analyses of 11 active thermal springs of central Italy. Us-ing this equation, we obtain a range of carbon isotope compositions between ~–7‰ and –2‰. This range supports the hypothesis that the CO2 originated from varying proportions of both magmatic mantle degassing and thermal decar-bonation of limestones and marbles, bearing in mind that the magmatic end member has a range of –6.5‰ ± 2.5‰ (Sano and Marty, 1995) and that carbonates in the Denizli Basin substratum are in the –1‰ to +2‰ range (Attanasio et al., 2006; Uysal et al., 2007). Measurements of helium isotopes in the modern gas emission in the area by Güleç et al. (2002) indicate that the source of He is the mantle, thereby providing further evidence of this type of contribution in the geothermal circulation of the Denizli Basin.

The relatively wide-ranging δ18O values for the banded and bedded travertines may be in-dicative of precipitation from isotopically dif-ferent waters at varying temperatures. Using the carbonate-water temperature equation of O’Neil et al. (1969) and assuming that the trav-ertine precipitation occurred near the isotopic equilibrium at a temperature within the present range of temperatures for the Denizli hydro-thermal system (25–58 °C; Dilsiz, 2006), we can calculate that the oxygen isotopic compo-sitions for the precipitation waters ranged be-tween –12‰ and +1‰ (versus standard mean ocean water [SMOW]).

3

4

5

6

7

8

–15 –13 –11 –9 –7

Kamara bedded Kamara bandedÇukurbağ bedded Çukurbağ banded

Akköy bedded Akköy bandedKocabaş bedded Kocabaş banded

Uysal et al. (2007)(banded)

δ18O (‰, PDB)

δ13C

(‰, P

DB

)

Figure 10. Results from oxygen and carbon stable isotope analyses conducted in this study and in Uysal et al. (2007). See Table DR3 for the complete list of data (see text footnote 1). The approximate correlation between oxygen and carbon stable isotope (except data from banded travertines previously published by Uysal et al., 2007) may represent the effect of downstream distance of samples from the main feeding conduits as suggested by Kele et al. (2011). Mean values for δ18O are –10.24‰ and –11.10‰ for bedded and banded travertines, respectively (excluding data by Uysal et al., 2007). Mean values for δ13C are 5.51‰ and 5.42‰ for bedded and banded traver-tines, respectively (excluding data by Uysal et al., 2007).





Studies on thermogene travertines showed that the equilibrium condition is rare for oxy-gen in carbonate precipitation, especially at the spring orifi ces. For instance, at Egerszalok (Hungary), which is a natural laboratory for the study of active travertine deposition, Kele et al. (2008) showed that carbonate precipita-tion generally does not occur at equilibrium and that the oxygen isotope composition of the carbonate deposit is enriched slightly over 1‰ if compared with the equilibrium condition, though this value decreases in proportion to the distance from the spring orifi ce. This tendency is very common in thermogene travertines. Kele et al. (2011) reported that Pamukkale travertines often precipitated outside the isotopic equilib-rium, as at the Egerszalok site, but were close to the equilibrium in a few other cases. These observations indicate that the range of the pre-cipitation waters of the travertine reported in this study is quite reasonable. Furthermore, if we also take into account the disequilibrium ef-fect, the range of possible δ18O values of the pre-cipitation water would decrease. It is therefore evident that the precipitation waters have a con-sistent meteoric water component. The variabil-ity of the δ18O of travertines could be therefore the expression of the mixing, to a varying extent, of water of meteoric origin (with a low isotope composition and temperature) and deep thermal waters from the Denizli hydrothermal system (with a high isotope composition and tempera-ture). The waters of the modern thermal springs of Pamukkale (Kele et al., 2011) and adjacent areas (Dilsiz, 2006) have an isotope composi-tion range of –9.5‰/–8.4‰ and –57‰/–61‰ for oxygen and hydrogen isotopes, respectively, thus displaying an entirely meteoric origin. If we assume that the water cycle in the area studied did not vary in the past (during the deposition of the travertines studied), the δ18O variability may be ascribed solely to the temperature variations, and it is, therefore, likely to be the result of the variability in the mixing of (1) deep geothermal waters deriving from the infi ltration of meteoric waters and (2) shallow cold waters related to rel-atively recent precipitation (Dilsiz et al., 2004; Dilsiz, 2006).

DISCUSSION

Our fi eld observations and laboratory results help to constrain the growth of the studied fi s-sure ridges and, therefore, the long-term hy-draulic circuit feeding the travertine deposits. Observations and results are synthesized in the growth model of Figure 11 through the following conceptual four main growth phases: (1) infant phase, (2) Kamara phase, (3) Çukurbağ phase, and (4) Akköy phase.

In the model, we propose a fi ssure ridge his-tory starting with an infant phase (Fig. 11A), where a linear elongated travertine mound (bedded travertine) starts to form on top of hot springs. The interplay between fi ssure ridges and fractures was previously suggested by Bar-gar (1978), observing that, at Mammoth Hot Springs (Wyoming, USA), many fi ssure ridge deposits developed from hot-spring waters fl ow-ing from preexisting linear vertical planes of weakness. As the fi ssure ridge grows (Kamara phase, Fig. 11B), the weight of the accumulated

bedded travertine increasingly hinders direct outfl ow from the central feeding conduit. The hydrothermal system thus develops an overpres-sure, which may even allow hydraulic fractur-ing. At the same time, in the feeder conduit, conditions differ from the open-air conditions, under which the early bedded travertine formed. In an (almost) sealed environment, fi eld and laboratory evidence (e.g., Boegli, 1980) sug-gests that chemical corrosion of the preexisting bedded travertine (e.g., Fig. 6D) can occur. De-pressurization by cracking and exchange with

bandedtravertine

sill-likestructure

seal phase crack phase

newfracture

Infant phase

Kamara phase

Çukurbağ phase

Akköy phase

unconformitybrecciaV-growth

subsidence

subsidingflank

tiltingtilting

beddedtravertine

banded travertine (?)

soft substratum

unconformity

unconformity

paraconformity

young bandedtravertine

travertine lateral growth

A

B

C

D

b1,2

b1

c1

d1

d1

c1

b2

Figure 11. Fissure ridge growth model. (A) Infant phase: Along a fracture system, an elon-gated travertine mound is built by hot-spring endogenic water. Cyclically, chemical corro-sion events within the fi ssures can occur. (B) Kamara phase: Tilting of the bedded travertine fl anks, due to the soil subsidence (caused by the general weighting of the fi ssure ridge), trig-gers a continuous divergent expansion of the fractures with deposition of the banded traver-tine. During a seal phase, the system is “closed,” and fl uids are injected within the host rock to form sill-like structures (see box b1), whereas in the crack phase, the system is “opened,” and new fractures can occur on a fl ank of the fi ssure ridge (see box b2). (C) Çukurbağ phase: The fi ssure ridge shows a marked asymmetric profi le, and within the fi ssures, deposition of the banded travertine occurs with a typical V-shaped structure. Often, breccias are as-sociated to the banded travertine (see box c1). (D) Akköy phase: This is the mature stage of a fi ssure ridge. The travertine deposition proceeds on a fl ank, characterized by maximum tilting and fracturing. New fi ssure ridges form on these fl anks, where other conduits permit leakage of hydrothermal fl uids (see box d1).



De Filippis et al.


the atmosphere can then trigger the precipita-tion of banded crack-fi lling travertine. The oc-currence of sill-like structures (Figs. 3D, 3E, and 11B) demonstrates that the banded traver-tine may have grown by exerting pressure on the host rock (e.g., Rossetti et al., 2007) constituted by the bedded travertine. The large extension of the bedded travertine (e.g., Fig. 6F), however, implies that, at the core of fi ssure ridges, a large volume is progressively made available for the growing banded travertine (Çukurbağ and Akköy phases; Figs. 11C and 11D). This space was probably obtained through the interplay of four main mechanisms: (1) extensional tecton-ics, (2) chemical corrosion, (3) fl uid and crystal-lization pressure induced by the vein formation, and (4) lateral collapse of fi ssure ridge fl anks.

The role of the tectonics is addressed later in this section together with other regional infl u-encing factors such as paleoclimate. It is inter-esting to note, however, that the growth rate of subvertical banded travertine as determined by data from Uysal et al. (2007) is only ~0.015 mm/a. Consequently, at the scale of the single fi ssure ridge, the rate of extensional tectonics is very low, only equal to or smaller than 0.015 mm/a, which is probably the result of the inter-play between tectonics and other factors that are addressed as follows.

Chemical corrosion along the central feeding conduit of the studied fi ssure ridges is evident from the irregular boundaries of some veins (Fig. 6D). We have no data to quantify this process (chemical corrosion); however, considering the elongate shape of the studied veins (Fig. 6D), we infer that these structures must have chiefl y formed by brittle cracking, and, subsequently, their boundaries were slightly reworked by chemical corrosion. We estimate, therefore, that the process of chemical corrosion had to be limited and that it can only account for a small portion of the volume presently occupied by the banded travertine (e.g., Fig. 6F).

Lateral collapse of fi ssure ridge fl anks must have been an important mechanism driving the deposition of banded travertine at the core of the studied fi ssure ridges (Figs. 11C and 11D). The occurrence of this mechanism (lateral col-lapse) is supported by a series of fi eld observa-tions such as the lateral tilting of fi ssure ridge fl anks (Figs. 4B and 7C), the V-shape of several veins at the core of fi ssure ridges (Figs. 4B, 4F, 5, and 6F), and the occurrence of travertine un-conformities on the fl anks of fi ssure ridges (Figs. 5E, 6F, 10C, and 11D). The lateral collapse of fi ssure ridge fl anks may have been induced by both the force exerted by the precipitation of banded travertine (i.e., crystallization pressure; Winkler and Singer, 1972; Bargar, 1978; Noiriel et al., 2010) and by the progressive subsidence

of soft sediments lying below the fl anks of fi s-sure ridges. In this view, the rotation (about a subhorizontal axis) of the ridge fl anks away from the central fi ssure may have been favored by the fact that the central portion of the fi ssure ridge, where the travertine load is maximum and subsidence should therefore be maximum as well, may be sustained by the banded travertine rooted in the fi ssure ridge substratum, whereas the distal portions of the ridge fl anks are not rooted, but simply rest on the substratum.

The lateral collapse of fi ssure ridge fl anks most likely implies, in turn, the opening of the hydraulic system in the axial region of the fi s-sure ridge itself, thus leading to the outfl ow of water rich in bicarbonate and associated lateral deposition of bedded travertine. The upward suture of some banded travertine by the bedded one (Fig. 4E) as well as the presence of brec-cias and, more in general, exotic material within some veins (Figs. 7E and 9F) all support the hypothesis of cyclic opening of the hydraulic system within the fi ssure ridges and alternate phases of deposition of banded and bedded trav-ertines. In summary, we propose, for the studied fi ssure ridges, a crack-and-seal model of growth (e.g., Gratier et al., 2003; Renard et al., 2005), where the deposition of bedded travertine mostly corresponds with the crack phases and the banded travertine with the seal phases. The microscopic observations, in particular, support the synkinematic progressive growth of banded travertine in a sealed environment (Fig. 9). On the other hand, the geochronological results (Table 1) give a broad picture, indicating that the growth of each edifi ce lasted several tens of thousands of years; while the age data do not resolve whether the alternate deposition of banded and bedded travertines was punctuated or continuous, they are certainly compatible with a repeated cycle of massive bed deposition followed by pressure-cracking vein fi lling and renewed bed deposition. Moreover, the differ-ence between equilibrium and disequilibrium precipitation in a crack-and-seal regime may be one cause for the oxygen isotopic heterogeneity that we observe.

In addition to local mechanisms such as chemical corrosion (Fig. 6D), fl uid and crystal-lization pressure, and lateral collapse of fi ssure ridge fl anks (Fig. 4B), it is now important to understand the external factors infl uencing the growth of fi ssure ridges, such as paleoclimate, tectonics, and geothermal circuit and outfl ow. Uysal et al. (2007, 2009) observed that the banded travertine from Denizli Basin mostly grew during cold periods of Quaternary time. Our data are substantially consistent with this observation (Figs. 8, DR1, and DR2 [see foot-note 1]). Unlike other types of travertine de-

posits, which mostly formed during warm-wet periods thanks to the abundant fl uid discharge (Rihs et al., 2000; Faccenna et al., 2008), Uysal et al. (2009) rightly ascribed the growth of banded travertine during cold periods to the CO2 oversaturation of deep reservoirs in connection with the general reduction in surface discharge of CO2 by spring or geothermal waters during these periods. Host-rock fracturing in response to seismic shaking and fl uid overpressure may have resulted in rapid exsolution and expansion of the dissolved gas, leading to hydrothermal outfl ow (Uysal et al., 2009). This mechanism is very similar to that proposed by Tuccimei et al. (2006) to explain growth and erosion phases of speleothems from the Colli Albani volcano area. Also REE (rare earth element) data from the banded travertine of Denizli Basin indicate that the banded travertine formed as thermo-gene deposits from rapidly ascending (from deep reservoirs) CO2-rich fl uids (Uysal et al., 2009). Mixing of CO2-rich geothermal fl uids with shallow groundwaters led to the formation of banded travertines with the isotopic signa-ture observed in Figure 10 and discussed in a previous section. We can ascribe the process of chemical weathering along some fi ssure walls (Fig. 6D) to these ascending CO2-rich geother-mal fl uids. This model is also consistent with the general crack-and-seal mechanisms proposed in this paper (Fig. 11).

From a morphotectonic point of view, the ori-entation of the studied fi ssure ridges and other ones in the region compared with the main nor-mal fault (Fig. 2) suggests that the growth of fi s-sure ridges in the Denizli Basin is only in part controlled by the regional extensional tectonics; otherwise all the fi ssure ridges would be paral-lel and would parallel the master normal fault. Rather, local deformation mechanisms such as those ones occurring in stepover zones between normal fault segments (Hancock et al., 1999; Brogi and Capezzuoli, 2009) may have been de-termining factors in the location and orientation of fi ssure ridges.

CONCLUSIONS

The growth of fi ssure ridges in the Denizli Basin has occurred by competing deposition of bedded and banded travertines formed dur-ing Quaternary time, similar to a volcano edifi ce that results from the long time interplay between intrusive (dikes, sills, etc.) and effusive (lava) products. Deposition of banded and bedded travertines alternated through a crack-and-seal mechanism modulated by several local and ex-ternal factors. Although the exact role and im-portance of each factor are still undefi ned and may vary from case to case, we propose that





these factors are: at the local scale, lateral col-lapse of fi ssure ridge fl anks, chemical corro-sion, and fl uid and crystallization pressure; at the regional scale, paleoclimate and geothermal degassing modulated by tectonics.

The studied fi ssure ridge travertines may represent a natural analog of the long-term (up to tens of thousands of years) evolution of an overpressured H2O + CO2 aquifer, such as could arise in the neighborhood of an artifi cial under-ground CO2 repository.

ACKNOWLEDGMENTS

We warmly thank A. Braathen, L. Crossey, J. Peder-son, N. Riggs, Z. Shipton, and T. Uysal for handling our manuscript and for very constructive comments. We also thank F. Rossetti for insightful help, F. De Angelis for useful suggestions, and D. Hawk for im-proving our English writing.

REFERENCES CITED

Alçiçek, H., Varol, B., and Özkul, M., 2007, Sedimentary fa-cies, depositional environments and palaeogeographic evolution of the Neogene Denizli Basin, SW Anatolia. Turkey: Sedimentary Geology, v. 202, p. 596–637, doi:10.1016/j.sedgeo.2007.06.002.

Altunel, E., 1994, Active Tectonics and the Evolution of Quaternary Travertines at Pamukkale, Western Tur-key [Ph.D. thesis]: Bristol, UK, University of Bristol, 234 p.

Altunel, E., and Hancock, P.L., 1993a, Active fi ssuring and faulting in Quaternary travertines at Pamukkale, west-ern Turkey, in Stewart, I.S., Vita-Finzi, C., and Owen, L.A., Neotectonics and Active Faulting: Zeitschrift für Geomorphologie, v. 94, supplement, p. 285–302.

Altunel, E., and Hancock, P.L., 1993b, Morphology and structural setting of Quaternary travertines at Pamuk-kale, Turkey: Geological Journal, v. 28, p. 335–346, doi:10.1002/gj.3350280312.

Altunel, E., and Hancock, P.L., 1996, Structural attributes of travertine-fi lled extensional fi ssures in the Pamukkale plateau, western Turkey: International Geology Review, v. 38, p. 768–777, doi:10.1080/00206819709465360.

Altunel, E., and Karabacak, V., 2005, Determination of horizontal extension from fi ssure-ridge travertines: A case study from the Denizli Basin, southwestern Tur-key: Geodinamica Acta, v. 18, p. 333–342, doi:10.3166/ga.18.333-342.

Atabey, E., 2002, The formation of fi ssure ridge type lami-nated travertine-tufa deposits; microscopical charac-teristics and diagenesis, Kirsehir Central Anatolia: Mineral Research Exploration Bulletin, v. 123–124, p. 59–65.

Ates, R.C., and Bayülke, N., 1982, The 19 August 1976 Denizli, Turkey, earthquake: Evaluation of the strong accelograph record: Bulletin of the Seismological So-ciety of America, v. 72, p. 1635–1649.

Attanasio, D., Brilli, M., and Ogle, N., 2006, The Isotopic Signature of Classical Marble: Rome, Ed. L’Erma di Bretschneider, 297 p.

Banerjee, A., Person, M., Hofstra, A., Sweetkind, D., Cohen, D., Sabin, A., Unruh, J., Zyvoloski, G., Gable, C.W., Crossey, L., and Karlstrom, K., 2011, Deep permeable fault-controlled helium transport and limited mantle fl ux in two extensional geothermal systems in the Great Basin, USA: Geology, v. 39, p. 195–198, doi:10.1130/G31557.1.

Bargar, K.E., 1978, Geology and thermal history of Mam-moth Hot Springs, Yellowstone National Park, Wyo-ming: U.S. Geological Survey Bulletin 1444, p. 1–55.

Bar-Matthews, M., Ayalon, A., Kaufman, A., and Wasser-burg, G.J., 1999, The Eastern Mediterranean paleo-climate as a refl ection of regional events: Soreq cave, Israel: Earth and Planetary Science Letters, v. 166, p. 85–95, doi:10.1016/S0012-821X(98)00275-1.

Barnes, I., Irwin, W.P., and White, D.E., 1978, Global Distri-bution of Carbon-Dioxide Discharges and Major Zones of Seismicity: U.S. Geological Survey Water Re-sources Investigations Open-File Report 78–39, 12 p.

Berger, A., 1978, Long-term variations of daily insolation and Quaternary climatic changes: Journal of the At-mospheric Sciences, v. 35, p. 2362–2367, doi:10.1175/1520-0469(1978)035<2362:LTVODI>2.0.CO;2.

Billi, A., Valle, A., Brilli, M., Faccenna, C., and Funiciello, R., 2007, Fracture-controlled fl uid circulation and dissolutional weathering in sinkhole-prone carbonate rocks from central Italy: Journal of Structural Geology, v. 29, p. 385–395, doi:10.1016/j.jsg.2006.09.008.

Boegli, A., 1980, Karst Hydrology and Physical Speleol-ogy (translated by June C. Schmid): Berlin, Springer-Verlag, 300 p.

Bottinga, Y., 1968, Calculation of fractionation factors for car-bon and oxygen in the system calcite–carbon dioxide–water: Journal of Physical Chemistry, v. 72, p. 800–808, doi:10.1021/j100849a008.

Bozkurt, E., 2001, Neotectonics of Turkey—A synthe-sis: Geodinamica Acta, v. 14, p. 3–30, doi:10.1016/S0985-3111(01)01066-X.

Brogi, A., and Capezzuoli, E., 2009, Travertine deposi-tion and faulting: The fault-related travertine fi ssure-ridge at Terme S. Giovanni, Rapolano Terme (Italy): International Journal of Earth Sciences/Geologische Rundschau, v. 98, p. 931–947, doi:10.1007/s00531-007-0290-z.

Büntgen, U., Willy, T., Nicolussi, K., McCormick, M., Frank, D., Trouet, V., Kaplan, J.O., Herzig, F., Heuss-ner, K.-U., Wanner, H., Luterbacher, J., and Esper, J., 2011, 2500 years of European climate variability and human susceptibility: Science, v. 331, p. 578–582, doi:10.1126/science.1197175.

Çakır, Z., 1999, Along-strike discontinuity of active normal faults and its infl uence on Quaternary travertine depo-sition; examples from western Turkey: Turkish Journal of Earth Sciences, v. 8, p. 67–80.

Chafetz, H.S., and Folk, R.L., 1984, Travertines: Depositional morphology and the bacterially constructed constituents: Journal of Sedimentary Petrology, v. 54, p. 289–316.

Chesterman, C.W., and Kleinhampl, F.J., 1991, Travertine Hot Springs, Mono County, California: California De-partment of Conservation, Division of Mines and Geol-ogy, August 1991, California Geology, v. 44, no. 8, p. 171–182.

Crossey, L.J., Fischer, T.P., Patchett, P.J., Karlstrom, K.E., Hilton, D.R., Newell, D.L., Huntoon, P., Reynolds, A.C., and de Leeuw, G.A.M., 2006, Dissected hydro-logic system at the Grand Canyon: Interaction between deeply derived fl uids and plateau aquifer waters in modern springs and travertine: Geology, v. 34, p. 25–28, doi:10.1130/G22057.1.

Crossey, L.J., Karlstrom, K.E., Springer, A.E., Newell, D., Hilton, D.R., and Fischer, T., 2009, De gassing of mantle-derived CO2 and He from springs in the south-ern Colorado Plateau region—Neotectonic connections and implications for groundwater systems: Geological Society of America Bulletin, v. 121, no. 7/8, p. 1034–1053, doi:10.1130/B26394.1.

De Filippis, L., and Billi, A., 2012, Morphotectonics of fissure ridge travertines from geothermal areas of Mammoth Hot Springs (Wyoming) and Bridgeport (California): Tectonophysics, v. 548–549, p. 34–38, doi.org/10.1016/j.tecto.2012.04.017.

de L’Apparent, A.F., 1966, Les depôts de travertins des montagnes Afghanes à l’ouest de Kabul: Revue de Géographie Physique et de Géologie Dynamique, v. 8, p. 351–357.

Dilsiz, C., 2006, Conceptual hydrodynamic model of the Pamukkale hydrothermal fi eld, southwestern Turkey, based on hydrochemical and isotopic data: Hydrogeol-ogy Journal, v. 14, issue 4, p. 562–572, doi:10.1007/s10040-005-0001-4.

Dilsiz, C., Marques, J.M., and Carreira, P.M.M., 2004, The impact of hydrological changes on travertine deposits related to thermal springs in the Pamukkale area (SW Turkey): Environmental Geology, v. 45, no. 6, p. 808–817, doi:10.1007/s00254-003-0941-8.

Dockrill, B., and Shipton, Z.K., 2010, Structural controls on leakage from a natural CO2 geologic storage site: Cen-

tral Utah, U.S.A.: Journal of Structural Geology, v. 32, p. 1768–1782, doi:10.1016/j.jsg.2010.01.007.

Faccenna, C., Soligo, M., Billi, A., De Filippis, L., Funi-ciello, R., Rossetti, C., and Tuccimei, P., 2008, Late Pleistocene depositional cycles of the Lapis Tibur-tinus travertine (Tivoli, central Italy): Possible in-fluence of climate and fault activity: Global and Planetary Change, v. 63, p. 299–308, doi:10.1016/j.gloplacha.2008.06.006.

Fleitmann, D., Cheng, H., Badertscher, S., Edwards, R.L., Mudelsee, M., Göktürk, O.M., Fankhauser, A., Pick-ering, R., Raible, C.C., Matter, A., Kramers, J., and Tüysüz, O., 2009, Timing and climatic impact of Greenland interstadials recorded in stalagmites from northern Turkey: Geophysical Research Letters, v. 36, L19707, doi:10.1029/2009GL040050.

Ford, T.D., and Pedley, H.M., 1996, A review of tufa and traver-tine deposits of the world: Earth-Science Reviews, v. 41, p. 117–175, doi:10.1016/S0012-8252(96)00030-X.

GISP2, 1997, The Greenland Summit Ice Cores: Boulder, Colorado, National Snow and Ice Data Center, Uni-versity of Colorado at Boulder, and the World Data Center-A for Paleoclimatology, National Geophysical Data Center, http://www.ncdc.noaa.gov/paleo/icecore/greenland/summit/document/gispingfo.htm (accessed 18 May 2012).

Goff, F., and Shevenell, L., 1987, Travertine deposits of Soda Dam, New Mexico, and their implications for the age and evolution of the Valles caldera hydrothermal system: Geological Society of America Bulletin, v. 99, p. 292–302, doi:10.1130/0016-7606(1987)99<292:TDOSDN>2.0.CO;2.

Gökgöz, A., 1998, Geochemistry of the Kızıldere-Tekkehamam-Buldan-Pamukkale Geothermal Fields: Reykjavik, Iceland, United Nations University, Geothermal Train-ing Programme, Reports, p. 115–156.

Gonfi antini, R., Panichi, C., and Tongiorgi, E., 1968, Iso-topic disequilibrium in travertine deposition: Earth and Planetary Science Letters, v. 5, p. 55–58, doi:10.1016/S0012-821X(68)80012-3.

Gratier, J.-P., Favreau, P., and Renard, F., 2003, Modeling fl uid transfer along California faults when integrat-ing pressure solution crack sealing and compaction processes: Journal of Geophysical Research, v. 108, p. 2104, doi:10.1029/2001JB000380.

Güleç, N., Hilton, D.R., and Mutlu, H., 2002, Helium isotope variations in Turkey: Relationship to tectonics, volcanism and recent seismic activities: Chemical Geology, v. 187, p. 129–142, doi:10.1016/S0009-2541(02)00015-3.

Guo, L., and Riding, R., 1998, Hot-spring travertine facies and sequences, late Pleistocene, Rapolano Terme, Italy: Sedimentology, v. 45, p. 163–180, doi:10.1046/j.1365-3091.1998.00141.x.

Guo, L., and Riding, R., 1999, Rapid facies changes in Holo-cene fi ssure ridge hot spring travertines, Rapolano Terme, Italy: Sedimentology, v. 46, p. 1145–1158, doi:10.1046/j.1365-3091.1999.00269.x.

Haluk Selim, H., and Yanik, G., 2009, Development of the Cambazlı (Turgutlu/MANISA) fi ssure-ridge–type trav-ertine and relationship with active tectonics, Gediz gra-ben, Turkey: Quaternary International, v. 199, no. 1–2, p. 157–163, doi:10.1016/j.quaint.2008.04.009.

Hancock, P.L., Chalmers, R.M.L., Altunel, E., and Çakir, Z., 1999, Travitonics: Using travertine in active fault stud-ies: Journal of Structural Geology, v. 21, p. 903–916, doi:10.1016/S0191-8141(99)00061-9.

Haszeldine, R.S., Quinn, O., England, G., Wilkinson, M., Shipton, Z.K., Evans, J.P., Heath, J., Crossey, L., Ballentine, C.J., and Graham, C.M., 2005, Natural geochemical analogues for carbon dioxide storage in deep geological porous reservoirs, a United Kingdom perspective: Oil & Gas Science and Technology, v. 60, p. 33–49, doi:10.2516/ogst:2005004.

Hayden, F.V., 1883, Twelfth Annual Report of the U.S. Geo-logical Survey of the Territories: Part II. Yellowstone National Park Geology–Thermal Springs–Topography: Washington, D.C., U.S. Government Printing Offi ce, 503 p.

Jackson, J.A., and McKenzie, D.P., 1988, Rates of active deformation in the Aegean Sea and surrounding re-gions: Basin Research, v. 1, p. 121–128, doi:10.1111/j.1365-2117.1988.tb00009.x.



De Filippis et al.


Kahle, H.G., Straub, C., Reilinger, R., McClusky, S., King, R., Hurst, K., Veis, G., Kastens, K., and Cross, P., 1998, The strain rate fi eld in the Eastern Mediterranean region, estimated by repeated GPS measurements: Tectonophysics, v. 294, p. 237–252, doi:10.1016/S0040-1951(98)00102-4.

Kaymakçı, N., 2006, Kinematic development and paleo-stress analysis of the Denizli Basin (western Turkey): Implications of spatial variation of relative paleostress magnitudes and orientations: Journal of Asian Earth Sciences, v. 27, p. 207–222, doi:10.1016/j.jseaes.2005.03.003.

Kele, S., Demény, A., Siklósy, Z., Németh, T., Tóth, M., and Kovács, M.B., 2008, Chemical and stable isotope com-position of recent hot-water travertines and associated thermal waters, from Egerszalók, Hungary: Deposi-tional facies and non-equilibrium fractionation: Sedi-mentary Geology, v. 211, no. 3–4, p. 53–72, doi:10.1016/j.sedgeo.2008.08.004.

Kele, S., Özkul, M., Fórizs, I., Gökgöz, A., Baykara, M.O., Alçiçek, M.C., and Németh, T., 2011, Stable isotope geochemical study of Pamukkale travertines: New evidences of low-temperature non-equilibrium calcite-water fractionation: Sedimentary Geology, v. 238, p. 191–212, doi:10.1016/j.sedgeo.2011.04.015.

Khoury, H., Salameh, E., and Udluft, P., 1984, On the Zerka Ma’in (Therma Kallirhoes) travertine/Dead Sea (hydro chemistry, geochemistry and isotopic composi-tion): Neues Jahrbuch für Geologie und Paläontologie, Monatshefte, v. 8, p. 472–484.

Koçyiğit, A., 2005, The Denizli graben-horst system and the eastern limit of western Anatolian continental exten-sion: Basin-fi ll, structure, deformational mode, throw amount and episodic evolutionary history, SW Turkey: Geodinamica Acta, v. 18, p. 167–208, doi:10.3166/ga.18.167-208.

Le Guen, Y., Renard, F., Hellmann, R., Brosse, E., Col-lombet, M., Tisserand, D., and Gratier, J.P., 2007, Enhanced deformation of limestone and sandstone in the presence of high PCO2 fl uids: Journal of Geo-physical Research, v. 112, B05421, 21 p., doi:10.1029/2006JB004637.

Mesci, L.B., Gürsoy, H., and Tatar, O., 2008, The evolution of travertine masses in the Sivas area (central Turkey) and their relationships to active tectonics: Turkish Jour-nal of Earth Sciences, v. 17, p. 219–240.

Michalzik, D., Fischer, R., Hernandez, D., and Oezen, D., 2001, Age and origin of the “Mexican Onyx” at San An-tonio Texcala (Puebla, Mexico): Geologische Beiträge Hannover, v. 2, p. 79–89.

Mix, A.C., Bard, E., and Schneider, R., 2001, Environmen-tal processes of the ice age: Land, oceans, glaciers (EPILOG ): Quaternary Science Reviews, v. 20, p. 627–657, doi:10.1016/S0277-3791(00)00145-1.

Nelson, S.T., Mayo, A.L., Gilfi llan, S.M., Dutson, S.J., Harris , R.A., Shipton, Z.K., and Tingey, D.G., 2009, Enhanced fracture permeability and accompanying fl uid fl ow in the footwall of a normal fault: The Hur-ricane fault at Pah Tempe hot springs, Washington County, Utah: Geological Society of America Bulletin, v. 121, p. 236–246.

Newell, D.L., Crossey, L.J., Karlstrom, K.E., and Fischer, T.P., 2005, Continental-scale links between the mantle and groundwater systems of the western United States: Evidence from travertine springs and regional He iso-tope data: GSA Today, v. 15, no. 12, p. 4–10, doi:10.1130/1052-5173(2005)015[4:CSLBTM]2.0.CO;2.

Noiriel, C., Renard, F., Doan, M.L., and Gratier, J.P., 2010, Intense fracturing and fracture sealing induced by min-eral growth in porous rocks: Chemical Geology, v. 269, p. 197–209, doi:10.1016/j.chemgeo.2009.09.018.

Nuriel, P., Rosenbaum, G., Uysal, I.T., Zhao, J.X., Gold-ing, S.D., Weinberger, R., Karabacak, V., and Avni, Y., 2011, Formation of fault-related calcite precipitates and their implications for dating fault activity in the East Anatolian and Dead Sea fault zones, in Fagereng, A., Toy, V.G., and Rowland, J.V., eds., Geology of the Earthquake Source: A Volume in Honour of Rick Sib-son: Geological Society of London Special Publication 359, p. 229–248.

Nuriel, P., Weinberger, R., Rosenbaum, G., Golding, S.D., Zhao, J.-X., Uysal, I.T., Bar-Matthews, M., and Gross,

M.R., 2012, Timing and mechanism of late-Pleistocene calcite vein formation across the Dead Sea fault zone, northern Israel: Journal of Structural Geology, v. 36, p. 43–54, doi:10.1016/j.jsg.2011.12.010.

Okay, A.İ., 1989, Denizli’nin güneyinde Menderes masifi ve Likya naplarının jeolojisi: Mineral Research Explora-tion Bulletin, v. 109, p. 45–58.

O’Neil, J.R., Clayton, R.N., and Mayeda, T.K., 1969, Oxy-gen isotope fractionation in divalent metal carbonates: The Journal of Chemical Physics, v. 51, p. 5547–5558, doi:10.1063/1.1671982.

Orland, I., Bar-Matthews, M., Kita, N.T., Ayalon, A., Mat-thews, A., and Valley, J., 2009, Climate deterioration in the Eastern Mediterranean as revealed by ion micro-probe microanalysis of speleothems that grew from 2.2 to 0.9 ka in Soreq Cave Israel: Quaternary Research, v. 71, p. 27–35, doi:10.1016/j.yqres.2008.08.005.

Özkul, M., Varol, B., and Alcicek, M.C., 2002, Depositional environments and petrography of the Denizli traver-tines: Mineral Research Exploration Bulletin, v. 125, p. 13–29.

Özkul, M., Gökgöz, A., and Horvatincic, N., 2010, Depo-sitional properties and geochemistry of Holocene perched springline tufa deposits and associated spring waters: A case study from the Denizli Province, west-ern Turkey, in Pedley, H.M., and Rogerson, M., eds., Tufas and Speleothems: Unravelling the Microbial and Physical Controls: Geological Society of London Special Publication 336, p. 245–262, doi:10.1144/SP336.13.

Özler, H.M., 2000, Hydrogeology and hydrochemistry in the Curuksu (Denizli) hydrothermal fi eld, western Turkey: Environmental Geology, v. 39, no. 10, p. 1169–1180, doi:10.1007/s002540000139.

Pamir, H.N., and Erentöz, C., 1974, Geological Maps of Turkey: Denizli Sheet: Ankara, Turkey, Mineral Re-search Exploration Directorate Turkey (MTA), scale 1:500,000.

Panichi, C., and Tongiorgi, E., 1976, Carbon isotopic com-position of CO2 from springs, fumaroles, mofettes, and travertines of central and southern Italy, in Proceed-ings of the 2nd U.N. Symposium on Development and Use of Geothermal Resources, 1975: San Francisco, p. 815–825.

Pedley, H.M., 2009, Tufas and travertines of the Mediter-ranean region: A testing ground for freshwater carbon-ate concepts and developments: Sedimentology, v. 56, p. 221–246, doi:10.1111/j.1365-3091.2008.01012.x.

Pentecost, A., 1995, The Quaternary travertine deposits of Europe and Asia Minor: Quaternary Science Reviews, v. 14, p. 1005–1028, doi:10.1016/0277-3791(95)00101-8.

Pentecost, A., 2005, Travertine: Berlin, Springer-Verlag, 445 p.

Pentecost, A., and Viles, H.A., 1994, A review and reassess-ment of travertine classifi cation: Géographie Physique Quaternaire, v. 48, p. 305–314.

Piper, J.D.A., Mesci, L.B., Gürsoy, H., Tatar, O., and Davies , C.J., 2007, Palaeomagnetic and rock magnetic properties of travertine: Its potential as a recorder of geomagnetic palaeosecular variation, environmental change and earthquake activity in the Sıcak Çermik geothermal fi eld, Turkey: Physics of the Earth and Planetary Interiors, v. 161, p. 50–73, doi:10.1016/j.pepi.2007.01.006.

Reilinger, R., McClusky, S., Vernant, P., Lawrence, S., Ergin-tav, S., Cakmak, R., Ozener, H., Kadirov, F., Guliev, I., Stepanyan, R., Nadariya, M., Hahubia, G., Mahmoud, S., Sakr, K., ArRajehi, A., Paradissis, D., Al-Aydrus, A., Prilepin, M., Guseva, T., Evren, E., Dmitrotsa, A., Filikov, S.V., Gomez, F., Al-Ghazzi, R., and Karam, G., 2006, GPS constraints on continental deformation in the Africa–Arabia–Eurasia continental collision zone and implications for the dynamics of plate interactions: Journal of Geophysical Research, v. 111, B05411, 26 p., doi:10.1029/2005JB004051.

Renard, F., Andreani, M., Boullier, A.M., and Labaume, P., 2005, Crack-seal patterns: Records of uncorrelated stress release variations in crustal rocks, in Gapais, D., Brun, J.P., and Cobbold, P.R., eds., Deformation Mechanisms, Rheology and Tectonics: From Minerals to the Lithosphere: Geological Society of London Spe-cial Publication 243, p. 67–79.

Rihs, S., Condomines, M., and Poidevin, J.L., 2000, Long-term behaviour of continental hydrothermal systems: U-series study of hydrothermal carbonates from the French Massif Central (Allier Valley): Geochimica et Cosmochimica Acta, v. 64, no. 18, p. 3189–3199, doi:10.1016/S0016-7037(00)00412-9.

Rossetti, F., Tecce, F., Billi, A., and Brilli, M., 2007, Patterns of fl uid fl ow in the contact aureole of the late Miocene Monte Capanne pluton (Elba Island, Italy): The role of structures and rheology: Contributions to Mineral-ogy and Petrology, v. 153, p. 743–760, doi:10.1007/s00410-006-0175-3.

Salvini, F., Billi, A., and Wise, D.U., 1999, Strike-slip fault-propagation cleavage in carbonate rocks: The Mattinata fault zone, Southern Apennines, Italy: Journal of Struc-tural Geology, v. 21, p. 1731–1749, doi:10.1016/S0191-8141(99)00120-0.

Sano, Y., and Marty, B., 1995, Origin of carbon in fumarolic gas from island arcs: Chemical Geology, v. 119, p. 265–274, doi:10.1016/0009-2541(94)00097-R.

Şengör, A.M.C., Gorur, N., and Saroglu, F., 1985, Strike-slip faulting and related basin formation in zones of tec-tonic escape: Turkey as a case study, in Biddle, K.T., and Christie-Blick, N., eds., Strike-Slip Faulting and Basin Formation: Society for Sedimentary Geology Special Publication 37, p. 227–264.

Shipton, Z.K., Evans, J.P., Kirchner, D., Kolesar, P.T., Wil-liams, A.P., and Heath, J., 2004, Analysis of CO2 leak-age through “low-permeability” faults from natural reservoirs in the Colorado Plateau, southern Utah, in Baines, S.J., and Worden, R.H., eds., Geological Storage of Carbon Dioxide: Geological Society of London Special Publication 233, p. 43–58.

Shipton, Z.K., Evans, J.P., Dockrill, B., Heath, J., Williams, A., Kirchner, D., and Kolesar, P.T., 2005, Natural leak-ing CO2-charged systems as analogs for failed geologic sequestration reservoirs, in Thomas, D., and Benson, S.M., eds., Carbon Dioxide Capture for Storage in Deep Geologic Formations; Results from the CO2 Capture Project: Amsterdam, Elsevier Science, v. 2, p. 699–712.

Taymaz, T., 1993, The source parameters of the Çubukdag (W. Turkey) earthquake of October 11: Geophysical Journal International, v. 113, p. 260–267, doi:10.1111/j.1365-246X.1993.tb02545.x.

Temiz, U., 2004, Kirsehir Dolayinin Neotektoniği ve Depremselliği [Neotectonics and Seismicity of the Kirsehir Region] [Ph.D. thesis]: Ankara, Department of Geological Engineering, Ankara University, 113 p.

Temiz, U., Gökten, E., and Eikenberg, J., 2009, U/Th dat-ing of fi ssure ridge travertines from the Kirsehir re-gion (Central Anatolia Turkey): Structural relations and implications for the Neotectonic development of the Anatolian block: Geodinamica Acta, v. 22, no. 4, p. 201–213, doi:10.3166/ga.22.201-213.

Tuccimei, P., Giordano, G., and Tedeschi, M., 2006, CO2 release variations during the last 2000 years at the Colli Albani volcano (Roma, Italy) from speleothems studies : Earth and Planetary Science Letters, v. 243, p. 449–462, doi:10.1016/j.epsl.2006.01.009.

Tuccimei, P., Soligo, M., Ginés, J., Ginés, A., Fornós, J., Kramers, J., and Villa, I.M., 2010, U-Th ages of phre-atic overgrowths on speleothems from coastal caves in Mallorca (Western Mediterranean): Earth Surface Proc esses and Landforms, v. 35, p. 782–790.

Turgay, I., Özgüler, M.E., and Sahin, H., 1985, Resistivity Report on Geothermal Energy Research in Denizli, Buldan and Pamukkale Areas: General Directorate of Mineral Research & Exploration (M.T.A.) Report 6958, 19 p. (In Turkish.)

Turi, B., 1986, Stable isotope geochemistry of travertines, 207–235, in Fritz, B.P., and Fontes, J.C., eds., Hand-book of Environmental Isotope Geochemistry: Amster-dam, Netherlands, Elsevier, v. 2, 557 p.

Uysal, I.T., Feng, Y., Zhao, J.X., Altunel, E., Weatherley, D., Karabacak, V., Cengiz, O., Golding, S.D., Lawrence, M.G., and Collerson, K.D., 2007, U-series dating and geochemical tracing of late Quaternary travertine in co-seismic fi ssures: Earth and Planetary Science Let-ters, v. 257, no. 3–4, p. 450–462, doi:10.1016/j.epsl.2007.03.004.

Uysal, I.T., Feng, Y., Zhao, J.X., Isik, V., Nuriel, P., and Golding, S.D., 2009, Hydrothermal CO2 degassing in





seismically active zones during the late Quaternary: Chemical Geology, v. 265, p. 442–454, doi:10.1016/j.chemgeo.2009.05.011.

Uysal, I.T., Feng, Y., Zhao, J.X., Bolhar, R., Isik, V., Baublys, K.A., Yago, A., and Golding, S.D., 2011, Seismic cycles recorded in late Quaternary calcite veins: Geochronological, geochemical and microstruc-tural evidence: Earth and Planetary Science Letters, v. 303, p. 84–96, doi:10.1016/j.epsl.2010.12.039.

Wang, Y.J., Cheng, H., Edwards, R.L., An, Z.S., Wu, J.Y., Shen, C.C., and Dorale, J.A., 2001, A high-resolution absolute-dated late Pleistocene monsoon record from Hulu Cave, China: Science, v. 294, p. 2345–2348, doi:10.1126/science.1064618.

Wang, Y.J., Cheng, H., Edwards, R.L., Kong, X.G., Shao, X.H., Chen, S.T., Wu, J.Y., Jiang, X.Y., Wang, X.F., and

An, Z.S., 2008, Millennial- and orbital-scale changes in the East Asian monsoon over the past 224,000 years: Nature, v. 451, p. 1090–1093, doi:10.1038/nature06692.

Westaway, R., 1990, Block rotation in western Turkey: 1. Observational evidence: Journal of Geophysi-cal Research, v. 95, p. 19,857–19,884, doi:10.1029/JB095iB12p19857.

Westaway, R., 1993, Neogene evolution of the Denizli re-gion of western Turkey: Journal of Structural Geology, v. 15, p. 37–53, doi:10.1016/0191-8141(93)90077-N.

Winkler, E.M., and Singer, P.C., 1972, Crystallization pres-sure of salts in stone and concrete: Geological Society of America Bulletin, v. 83, p. 3509–3514, doi:10.1130/0016-7606(1972)83[3509:CPOSIS]2.0.CO;2.

Yalçınlar, Î., 1983, Türkiye’de Neojen ve Kuvaterner Omurgalı Arazileri ve Jeomorfolojik Karakterleri [Neogene and

Quaternary Vertebrate Fields of Turkey and their Geo-morphologic Characteristics]: Publication of Faculty of Literature, Istanbul Technical University 2741, p. 1–240 (in Turkish).

SCIENCE EDITOR: NANCY RIGGS

ASSOCIATE EDITOR: JOEL PEDERSON

MANUSCRIPT RECEIVED 14 SEPTEMBER 2011REVISED MANUSCRIPT RECEIVED 29 MARCH 2012MANUSCRIPT ACCEPTED 6 APRIL 2012

Printed in the USA



List of Supplementary Material and Related Captions

Supplementary Material 1. 230Th/234U dating method.

Table DR1. Main known fissure ridge travertines in the world and related attributes.

Table DR2. Uranium content, uranium and thorium activity ratios, and age of travertines sampled in

the Denizli basin for this work.

Table DR3. Oxygen and carbon stable isotope data of travertines sampled in the Denizli basin for this

work.

Figure DR1. Ages (with error bars) of Denizli travertines (from this work and from Uysal et al.,

2007, 2009) plotted over figure 1 from Wang et al. (2001). Travertines from Denizli

basin are distinguished in banded (red diamonds for data from this work and green

diamonds for data from Uysal et al., 2007, 2009) and bedded travertine (blue diamonds,

only from this work). Vertical yellow bars indicate Younger Dryas (YD) and Heinrich

paleoclimate events (H) (Wang et al., 2001). The band encompassed by the red dashed

line indicates Last Glacial Maximum (LGM) (Mix et al., 2001). Also shown are δ18O of

Hulu Cave stalagmites (purple, green, and red), Greenland Ice (GISP2, 1997) (dark blue),

and insolation (33°N June, July, and August) (Berger, 1978; Paillard et al., 1996) (black)

versus time. Numbers indicate GISs and their correlation with Hulu data. The δ18O values

of the numbered peaks are reversed for Hulu (increasing down) in comparison with

GISP2 (increasing up) (see Wang et al., 2001 for further details).

References Berger, A., 1978, Long-term variations of daily insolation and quaternary climatic changes: Journal of the Atmospheric

Science 35, 2362–2367.

DR2012245 De Filippis et al.

GISP2, 1997, The Greenland Summit Ice Cores, National Snow and Ice Data Center, University of Colorado at

Boulder, and the World Data Center-A for Paleoclimatology, National Geophysical Data Center, Boulder

Colorado.

Mix, A.C., Bard, E., Schneider, R., 2001, Environmental processes of the ice age: land, oceans, glaciers (EPILOG).

Quaternary Science Reviews 20 (4), p. 627–657.

Paillard, D., Labeyrie, L., and Yiou, P. (1996), Macintosh Program performs time-series analysis, Eos Transactions

AGU, 77(39), 379, doi:10.1029/96EO00259.

Uysal, I.T., Feng, Y., Zhao, J.X., Altunel, E., Weatherley, D., Karabacak, V., Cengiz, O., Golding, S.D., Lawrence,

M.G., and Collerson, K.D., 2007, U-series dating and geochemical tracing of late Quaternary travertine in co-

seismic fissures: Earth and Planetary Science Letters, v. 257 (3–4), p. 450–462.

Uysal, I.T., Feng, Y., Zhao, J.X., Isik, V., Nuriel, P., and Golding, S.D., 2009, Hydrotermal CO2 degassing in

seismically active zones during the Late Quaternary: Chemical Geology, v. 265, p. 442-454.

Wang, Y.J., Cheng, H., Edwards, R.L., An, Z.S., Wu, J.Y., Shen, C.C., and Dorale, J.A., 2001, A high-resolution

absolute-dated late Pleistocene monsoon record from Hulu Cave, China: Science 294 (5550), 2345–2348.

Figure DR2. Ages (with error bars) of Denizli travertines (from this work and from Uysal et al.,

2007, 2009) plotted over figure 1 from Fleitmann et al. (2009). Travertines from Denizli

basin are distinguished in banded (red diamonds for data from this work and green

diamonds for data from Uysal et al., 2007, 2009), and bedded travertine (blue diamonds,

only from this work). (a-i) Vertical grey bars indicate Heinrich events (H) (Wang et al.,

2001), YD is for Younger Dryas (Wang et al., 2001); the band encompassed by the red

dashed line indicates Last Glacial Maximum (LGM) (Mix et al., 2001). See Fleitmann et

al. (2009) for all the details about the δ18O and δ13C records, and U-Th dates worldwide.

References Fleitmann, D., Cheng, H., Badertscher, S., Edwards, R.L., Mudelsee, M., Göktürk, O.M., Fankhauser, A., Pickering, R.,

Raible, C.C., Matter, A., Kramers, J., and Tüysüz, O., 2009, Timing and climatic impact of Greenland

interstadials recorded in stalagmites from northern Turkey: Geophysical Research Letters, vol. 36, L19707,

doi:10.1029/2009GL040050, 2009.

Mix, A.C., Bard, E., Schneider, R., 2001, Environmental processes of the ice age: land, oceans, glaciers (EPILOG).

Quaternary Science Reviews 20 (4), p. 627–657.




Uysal, I.T., Feng, Y., Zhao, J.X., Isik, V., Nuriel, P., and Golding, S.D., 2009, Hydrotermal CO2 degassing in

seismically active zones during the Late Quaternary: Chemical Geology, v. 265, p. 442-454.

Wang, Y.J., Cheng, H., Edwards, R.L., An, Z.S., Wu, J.Y., Shen, C.C., and Dorale, J.A., 2001, A high-resolution

absolute-dated late Pleistocene monsoon record from Hulu Cave, China: Science 294 (5550), 2345–2348.


Supplementary Material 1

230Th/234U dating method

The method is based on the extreme fractionation of the parent isotopes 238U and 234U from their

long-lived daughter 230Th in the hydrosphere. Uranium, markedly more soluble than Th in the

surface and near-surface environments, is readily mobilized as the highly soluble uranyl ion (UO2

2+) and its complexes, whereas Th is easily hydrolyzed and precipitated or adsorbed on detrital

particles. Uranium is co-precipitated with CaCO3 on exsolution of CO2, while Th is generally

negligible. In the absence of detrital Th, 230Th only forms in situ by radioactive decay of co-

precipitated U. In a closed system the extent to which the 230Th/234U activity ratio has returned

towards unity is a function of time, taking into account also the state of disequilibrium between 234U

and 238U (Kaufman and Broecker, 1965). About 500 mg of travertine samples were spiked with a

mixed 236U–229Th tracer and dissolved in concentrated distilled HCl. The dried chloride was then

converted to nitrate. U and Th were separated on miniaturized PTFE columns containing U-

TEVA™ resin using HNO3 and HCl. As the Th eluate is strongly bound to rather volatile organic

matter washed from the resin, the Th fraction was evaporated at <50°C to avoid loss by

evaporation. Organic molecules from the resin were oxidized according to the protocol described in

Tuccimei et al. (2010). After oxidation, U and Th samples were dissolved in 3 M HCl, and

measured on a Nu Instruments™ multicollector plasma source mass spectrometer equipped with an

APEX™ desolvating nebulizer. The WARP™ filter limiting access to one of the ion counters of the

mass spectrometer is able to reduce the background on masses 230 and 229 to less than 0.5 ppm and

less than 0.2 ppm of the peak height at mass 232, respectively. Measurement protocols are as

described by Fleitmann et al. (2007). Activity ratios for MC-ICPMS data were calculated with the

decay constants described in Cheng et al. (2000). The ages of all samples were calculated by means


of Isoplot/Ex (version 3.0), a plotting and regression program designed by Ludwig (2003) for

radiogenic-isotope data.

Uranium content, uranium and thorium activity ratios, and the age of travertine are reported in

Table DR2. Uranium abundances range from 6 to 392 ppb. Banded travertines from the Akkoy fissure

ridge (the site with most dated samples) display uranium concentration higher than the values

observed for bedded deposits. (234U/238U) activity ratios do not show large fluctuations and equal to

1.196 – 1.347, with the exception of sample Cuk1 (from Çukurbag fissure ridge) with a value of

1.435. The detrital content is negligible as evidenced by (230Th/232Th) activity ratios, always higher

than 100, with the exception of sample Ko1 (from Kocabaş fissure ridge) where this ratio

approaches 50. Therefore, no correction scheme is required to obtain a reliable age.

References

Cheng, H., Edwards, R.L., Hoff, J., Gallup, C.D., Richards, D.A., and Asmero, Y., 2000, The half-lives of uranium-234

and thorium-230: Chemical Geology 169, p. 17-33.

Fleitmann, D., Burns, S.J., Mangini, A., Mudelsee, M., Kramers, J., Villa, I.M., Neff, U., Al-Subbary, A.A., Buettner,

A., Hippler, D., and Matter, A., 2007, Holocene ITCZ and Indian monsoon dynamics recorded in stalagmites

from Oman and Yemen (Socotra): Quaternary Science Reviews 26, p. 170-188.

Kaufman, A., and Broecker, W.S., 1965, Comparison of 230Th and 14C ages of carbonate materials from Lakes

Lahantan and Bonneville: Journal of Geophysical Research 70, p. 4039-4054.

Ludwig, K.R., 2003, Using Isoplot/Ex, Version 3. A GeochronologicalToolkit for Microsoft Excel: Berkeley

Geochronology Ctr. Spec. Pub. 4.

Tuccimei, P., Soligo, M., Ginés, J., Ginés, A., Fornós, J., Kramers, J., and Villa, I.M., 2010, U-Th ages of phreatic

overgrowths onspeleothems from coastal caves in Mallorca (Western Mediterranean): Earth Surface Processes

and Landforms 35, p. 782-790.


Table DR1. Main known fissure ridge travertines in the world and related attributes. (1) The term “isolated” refers to the occurrence of fissure ridges as single structures or in swarms. (2) Note that the banded travertine occurs in the inner portion of fissure ridges and only in some cases it is visible. (3) Not interpreted as fissure ridge but as “line of tufa cones” or “fossil spring deposits” deposited from hot thermal waters.

fissure ridge locality country lat long tectonic setting position to main

fault long-axis trend length

(m) width (m)

height (m)

isolated (1)

observed banded

travertine (2)

references

Kamara Yenice (Denizli basin) Turkey N 38°03'24" E 28°58'16" extensional basin hanging wall N125° 63 15 6 no yes Cakir 1999

Çukurbağ Pamukkale (Denizli basin) Turkey N 37°55'53" E 29°06'58" extensional basin hanging wall N85° 350 36 11 no yes Altunel & Hancock 1993a, Altunel &

Karabacak 2005, Uysal et al. 2007, 2009

Çukurbağ 2 Pamukkale (Denizli basin) Turkey N 37°55'51" E 29°07'04" extensional basin hanging wall N77° 130 14 1 no no Altunel & Hancock 1993a

Akköy (Karakaya Hill)

Akköy (Denizli basin) Turkey N 37°56'56" E 29°05'28" extensional basin hanging wall N116° - N146° 1900 200 25 no yes Altunel & Hancock 1993a, Altunel &

Karabacak 2005, Uysal et al. 2007, 2009

Karahayit Karahayit (Denizli basin) Turkey N 37°57'26" E 29°06'12" extensional basin hanging wall N145° 600 200 7 no yes Altunel & Hancock 1993a, 1993b, 1996

Kizilseki Hill Kizilseki Hill (Denizli basin) Turkey N 37°57'16" E 29°05'55" extensional basin hanging wall N147° 1200 220 35 no yes Altunel & Hancock 1993a, 1993b, 1996

Hanife Hill Hanife Hill (Denizli basin) Turkey N 37°56'43" E 29°06'14" extensional basin hanging wall N113° - N161° 750 160 10 no yes Altunel & Hancock 1993a, 1993b, 1996

Kocabaş Kocabaş (Denizli basin) Turkey N 37°48'39" E 29°18'59" extensional basin hanging wall N115° - N126° 2700 300 20 no yes Altunel & Karabacak 2005

Bal Balkayasi (Gediz graben) Turkey N 38°22'07" E 28°42'43" extensional basin hanging wall N135° 350 25 no Cakir 1999

Cambazli Cambazli (Gediz graben) Turkey N 38°34'14" E 27°50'21" extensional basin hanging wall 200 5 15 no yes Haluk Selim & Yanik 2009

Kirsehir Kirsehir (Central Anatolia) Turkey N 39°07'50" E 34°07'52" extensional basin hanging wall N170° 800 30 4 no yes Atabey 2002, Temiz et al. 2009, Uysal et

al. 2009

Ihlara Ihlara Valley (Central Anatolia) Turkey N 38°17'15" E 34°14'23" volcanic field no yes Karabacak & Altunel 2005

Sicak Çermik Sivas area (Central Turkey) Turkey N 39°44’55” E 36°42’58” pull apart N30° 1600 no yes Piper et al. 2007, Mesci et al. 2008

Delikkaya Sivas area (Central Turkey) Turkey N 39°44’55” E 36°42’58” pull apart N150° 500 no yes Piper et al. 2007, Mesci et al. 2008

Sarikaya Sivas area (Central Turkey) Turkey N 39°44’55” E 36°42’58” pull apart N115° 750 no yes Piper et al. 2007, Mesci et al. 2008

Terme S. Giovanni

Rapolano Terme (Siena) Italy N 43°16'46" E 11°35'31" extensional basin hanging wall N117° 250 30 10 yes no Guo & Riding 1999, Brogi 2004, Brogi &

Capezzuoli 2009

Abano Terme Abano Terme (Padova) Italy N 45°21'09" E 11°46'38" volcanic field hanging wall N120° yes no Zampieri et al. 2010

Nymphopetres (3)

Lakes Volvi and Langada

(Mygdonia basin) Greece N 40°41’52” E 23°19’30” extensional basin hanging wall NNW-SSE 3 5 no Pavlides & Kilias 1987, Traganos et al.

1995, Hancock et al. 1999

Hot Tub Ridge Bridgeport (California) U.S.A. N 38°14'45" W 119°12'18" extensional basin hanging wall N204° - N221° 165 7 4.5 no yes Chesterman & Kleinhampl 1991,

Hancock et al. 1999

White Elephant Back Terrace

Mammoth Hot Springs

(Wyoming) U.S.A. N 44°57'49" W 110°42'47" volcanic field hanging wall N45° 235 15 8 no yes Bargar 1978

Pyramid Lake (3)

Pyramid Lake (Nevada) U.S.A. N 40°08’47” W 119°40’44” extensional basin N145° 2000 165 10-65 no Benson 1994, Hancock et al. 1999

Searles Lake (3)

Searles Lake (California) U.S.A. N 35°43'27" W 117°20'37" extensional basin hanging wall N-S no Scholl & Taft 1964, Hancock et al. 1999

Soda Dam Jemez Springs (New Mexico) U.S.A. N 35°47'29" W 106°41'11" volcanic field hanging wall N240° 100 25 15 no Chafetz & Folk 1984, Goff & Shevenell

1987 Crystal Geyser

system Paradox Basin

(Utah) U.S.A. N 38°56’18” W 110°08’08” extensional basin footwall WNW-ESE 80 70 3 no yes Shipton et al. 2004, 2005, Dockrill & Shipton 2010

San Antonio Texcala

San Antonio Texcala (Puebla) Mexico N 18°24'01" W 97°26'43" yes Michalzik et al. 2001

Zerka Mai'in Madaba (Dead Sea region) Jordan N 31°36'11" E 35°36'21" Khoury et al. 1984

Hammam Meskoutine

Hammam Meskoutine Algeria N 36°27'42" E 07°16'10" no Pentecost & Viles 1994, Pentecost 2005


References

Altunel, E., and Hancock P.L., 1996, Structural attributes of travertine-filled extensional fissures in the Pamukkale plateau, western Turkey: International Geology Review, v. 38, p. 768-777.

Altunel, E., and Hancock, P.L., 1993a, Active fissuring and faulting in Quaternary travertines at Pamukkale, western Turkey, in Stewart, I.S., Vita-Finzi, C., and Owen, L.A., Neotectonics and Active Faulting:

Zeitschrift fuer Geomorphologie Supplement, v. 94, p. 285-302.

Altunel, E., and Hancock, P.L., 1993b, Morphology and structural setting of Quaternary travertines at Pamukkale, Turkey: Geological Journal, v. 28, p. 335–346.

Altunel, E., and Karabacak, V., 2005, Determination of horizontal extension from fissure-ridge travertines: a case study from the Denizli Basin, southwestern Turkey: Geodinamica Acta, v. 18, p. 333–342.

Atabey, E., 2002, The formation of fissure ridge type laminated travertine-tufa deposits microscopical characteristics and diagenesis, Kirsehir Central Anatolia: Mineral Research Exploration Bulletin (MTA), v. 123-

124, p. 59–65.

Bargar, K.E., 1978, Geology and thermal history of Mammoth Hot Springs, Yellowstone National Park, Wyoming: U. S. Geological Survey Bulletin, v. 1444, p. 1–55.

Benson, L., 1994, Carbonate deposition, Pyramid Lake Subbasin, Nevada: 1. Sequence of formation and elevation distribution of carbonate deposits (Tufas): Palaeogeography, Palaeoclimatology, Palaeoecology, v.

109, p. 55-87.

Brogi, A., 2004, Faults linkage, damage rocks and hydrothermal fluid circulation: tectonic interpretation of the Rapolano Terme travertines (southern Tuscany, Italy) in the context of the Northern Apennines

Neogene-Quaternary extension: Eclogae Geologicae Helvetiae, v. 97, p. 307–320, doi:10.1007/s00015-004-1134-5.

Brogi, A., and Capezzuoli, E., 2009, Travertine deposition and faulting: the fault-related travertine fissure-ridge at Terme S. Giovanni, Rapolano Terme (Italy): International Journal of Earth Sciences Geol Rundsch,

v. 98, p. 931– 947, doi:10.1007/s00531-007-0290-z.

Çakır, Z., 1999, Along-strike discontinuity of active normal faults and its influence on Quaternary travertine deposition; examples from western Turkey: Turkish Journal of Earth Sciences, v. 8, p. 67–80.

Chafetz, H.S., and Folk, R.L., 1984, Travertines: depositional morphology and the bacterially constructed constituents: Journal Sedimentary Petrology, v. 54, p. 289–316.

Chesterman, C.W., and Kleinhampl, F.J., 1991, Travertine Hot Springs, Mono County, California: California Geology, California Department of Conservation, Division of Mines and Geology, August 1991, v. 44,

no. 8, p. 171-182.

Dockrill, B., and Shipton, Z.K., 2010, Structural controls on leakage from a natural CO2 geologic storage site: central Utah, U.S.A.: Journal of Structural Geology, v. 32, issue 11, p. 1768-1782, doi:

10.1016/j.jsg.2010.01.007.

Goff, F., and Shevenell, L., 1987, Travertine deposits of Soda Dam, New Mexico, and their implications for the age and evolution of the Valles caldera hydrothermal system: Geological Society of America Bulletin,

v. 99, 292-302, August 1987.

Guo, L., and Riding, R., 1999, Rapid facies changes in Holocene fissure ridge hot spring travertines, Rapolano Terme, Italy: Sedimentology, v. 46, p. 1145–1158.

Haluk Selim, H., and Yanik, G., 2009, Development of the Cambazlı (Turgutlu/MANISA) fissure-ridge-type travertine and relationship with active tectonics, Gediz Graben, Turkey: Quaternary International, v. 199,

issues 1-2, p. 157-163, doi:10.1016/j.quaint.2008.04.009.

Hancock, P.L., Chalmers, R.M.L., Altunel, E., and Çakir, Z., 1999, Travitonics: using travertine in active fault studies: Journal of Structural Geology, v. 21, p. 903–916.

Karabacak, V., and Altunel, E., 2005, Relationship between travertine deposition and volcanic activity: in Özkul, M., Yağiz, S., and Jones, B., eds., Proceedings of 1st International Symposium on Travertine,

September 21-25, 2005, Pamukkale University (Denizli, Turkey).

Khoury, H., Salameh, E. and Udluft, P., 1984, On the Zerka Ma’in (Therma Kallirhoes) travertine/Dead Sea (hydrohemistry, geochemistry and isotopic composition): Neues Jahrbuch für Geologie und Paläontologie,

Monatshefte, v. 8, p. 472–484.

Mesci, L.B., Gürsoy, H., and Tatar, 2008, The evolution of travertine masses in the Sivas Area (Central Turkey) and their relationships to active tectonics: Turkish Journal of Earth Sciences, v. 17, p. 219–240.

Michalzik, D., Fischer, R., Hernandez, D., and Oezen, D., 2001, Age and origin of the “Mexican Onyx” at San Antonio Texcala (Puebla, Mexico): Geologische Beiträge Hannover, v. 2, p. 79-89.

Pavlides, S.B., and Kilias, A.A., 1987, Neotectonic and active faults along the Serbomacedonian zone (SE Chalkidiki, northern Greece): Annales Tectonicae, v. 1, p. 97-104.

Pentecost, A., 2005, Travertine: Springer-Verlag, Berlin Heidelberg, 445 pp.

Pentecost, A., and Viles, H.A., 1994, A review and reassessment of travertine classification: Géographie physique Quaternaire, v. 48, p. 305–314.

Piper, J.D.A., Mesci, L.B., Gürsoy, H., Tatar, O., and Davies, CJ., 2007, Palaeomagnetic and rock magnetic properties of travertine: Its potential as a recorder of geomagnetic palaeosecular variation, environmental

change and earthquake activity in the Sıcak Çermik geothermal field, Turkey: Physics of the Earth and Planetary Interiors, v. 161, p. 50–73, doi:10.1016/j.pepi.2007.01.006.

Scholl, D.W., and Taft, W.H., 1964, Algae, contributors to the formation of calcareous tufa, Mono Lake, California: Journal Sedimentary Petrology, v. 34, p. 309–319.

Shipton, Z.K., Evans, J.P., Kirchner, D., Kolesar, P.T., Williams, A.P. and Heath, J., 2004, Analysis of CO2 leakage through "low-permeability" faults from natural reservoirs in the Colorado Plateau, southern Utah:

in Baines, S.J. and Worden, R.H., eds., Geological storage of carbon dioxide: Geological Society, London, Special Publications, no. 233, p. 43-58.


Shipton, Z.K., Evans, J.P., Dockrill, B., Heath, J., Williams, A., Kirchner, D., and Kolesar, P.T., 2005, Natural leaking CO2-charged systems as analogs for failed geologic sequestration reservoirs: in Thomas, D., and

Benson, S.M., eds., Carbon Dioxide Capture for Storage in Deep Geologic Formations; Results from the CO2 Capture Project, v. 2, Elsevier Science, p. 699-712.

Temiz, U., Gökten, E., and Eikenberg, J., 2009, U/Th dating of fissure ridge travertines from the Kirsehir region (Central Anatolia Turkey): structural relations and implications for the Neotectonic development of the

Anatolian block: Geodinamica Acta, v. 22/4, p. 201-213.

Traganos, G., Jux, U., and Steuber, T., 1995, Isotopic characteristics of geothermal waters and fossil spring deposits in Mygdonia Basin, Northern Greece: Geothermics, v. 24, no. 1, p. 61-80.

Uysal, I.T., Feng, Y., Zhao, J.X., Altunel, E., Weatherley, D., Karabacak, V., Cengiz, O., Golding, S.D., Lawrence, M.G., and Collerson, K.D., 2007, U-series dating and geochemical tracing of late Quaternary

travertine in co-seismic fissures: Earth and Planetary Science Letters, v. 257 (3–4), p. 450–462.

Uysal, I.T., Feng, Y., Zhao, J.X., Isik, V., Nuriel, P., and Golding, S.D., 2009, Hydrotermal CO2 degassing in seismically active zones during the Late Quaternary: Chemical Geology, v. 265, p. 442-454.


Table DR2. Uranium content, uranium and thorium activity ratios and the age of travertine. Errors are always quoted as 2σ .

fissure ridge sample travertine

type lat long ppb U (230Th/232Th) (234U/238U) (230Th/234U) age (ka)

Kamara Ka2 banded N 38°03’24” E 28°58’16” 392.49 ± 0.67 150.62 ± 2.13 1.287 ± 0.002 0.015 ± 0.001 1.7 ± 0.1

Kamara Ka4 banded N 38°03’24” E 28°58’16” 383.94 ± 0.68 136.52 ± 2.01 1.347 ± 0.002 0.023 ± 0.001 2.5 ± 0.1

Çukurbağ Cuk1 banded N 37°55’54” E 29°06’58” 93.24 ± 0.19 225.02 ± 4.32 1.435 ± 0.004 0.199 ± 0.002 24.1 ± 0.3

Çukurbağ Pa5 bedded N 37°55’53” E 29°06’56” 27.08 ± 0.07 250 ± 2.87 1.285 ± 0.005 0.432 ± 0.007 60 ± 1

Akköy Ak15 banded N 37°57’02” E 29°05’22” 96.94 ± 0.17 187 ± 2.24 1.196 ± 0.002 0.237 ± 0.011 29.3 ± 1.6

Akköy Ak16 banded N 37°57’02” E 29°05’22” 195.53 ± 0.33 173.65 ± 2.41 1.243 ± 0.002 0.180 ± 0.004 20.2 ± 0.5



Akköy Ak22 banded N 37°56’57” E 29°05’26” 271 ± 0.46 333.39 ± 6.41 1.236 ± 0.002 0.179 ± 0.001 21.4 ± 0.1

Akköy Ak23 banded N 37°56’54” E 29°05’26” 355.71 ± 0.60 102.23 ± 2.02 1.214 ± 0.002 0.195 ± 0.004 23.5 ± 0.6

Akköy Ak24 bedded N 37°56’54” E 29°05’26” 70.13 ± 0.13 198 ± 2.04 1.220 ± 0.003 0.392 ± 0.009 53.4 ± 1.6

Akköy Ak26 banded N 37°56’56” E 29°05’27” 259.93 ± 0.44 125.68 ±1.82 1.226 ± 0.001 0.200 ± 0.003 24.1 ± 0.4

Akköy Ak27 bedded N 37°56’56” E 29°05’27” 31.75 ± 0.15 127.70 ± 1.96 1.237 ± 0.005 0.286 ± 0.023 36.4 ± 3

Kocabaş Ko1 banded N 37°48’39” E 29°19’01” 12.94 ± 0.03 50.26 ± 0.99 1.203 ± 0.004 1.100 ± 0.020 ≥ 350

Kocabaş Ko2 banded N 37°48’39” E 29°18’56” 21.82 ± 0.05 321 ± 4.56 1.122 ± 0.003 0.786 ± 0.012 160 ± 5

Kocabaş Ko4 banded N 37°48’39” E 29°19’00” 6.55 ± 0.02 297 ± 3.87 1.218 ± 0.006 1.007 ± 0.035 ≥ 350


Table DR3. Oxygen and carbon stable isotope data.

fissure ridge or fault sample travertine

type lat long δ18O (PDB) ‰

δ13C (PDB) ‰

references

Kamara Ka2 banded N 38°03’24” E 28°58’16” -13.60 2.92 this work

Kamara Ka4 banded N 38°03’24” E 28°58’16” -13.10 3.34 this work Kamara Ka5 bedded N 38°03’24” E 28°58’16” -8.34 5.01 this work Kamara Ka6 bedded N 38°03’24” E 28°58’16” -12.30 4.38 this work Kamara Ka7 banded N 38°03’24” E 28°58’16” -13.20 3.19 this work

Çukurbağ Cuk1 banded N 37°55’54” E 29°06’58” -14.10 5.79 this work Çukurbağ Cuk5 bedded N 37°55’53” E 29°06’56” -9.74 6.43 this work Çukurbağ Pa5 bedded N 37°55’53” E 29°06’56” -9.82 7.83 this work

Akköy Ak2 banded N 37°56’54” E 29°05’26” -7.14 6.17 this work Akköy Ak3 bedded N 37°56’52” E 29°05’30” -10.30 5.35 this work Akköy Ak3 banded N 37°56’52” E 29°05’30” -11.00 5.27 this work Akköy Ak17 banded N 37°57’02” E 29°05’22” -12.20 4.55 this work Akköy Ak18 bedded N 37°57’02” E 29°05’22” -11.30 4.70 this work Akköy Ak19 banded N 37°57’02” E 29°05’22” -11.85 4.97 this work Akköy Ak21 banded N 37°56’57” E 29°05’26” -8.75 6.46 this work Akköy Ak22 banded N 37°56’57” E 29°05’26” -9.23 7.11 this work Akköy Ak23 banded N 37°56’54” E 29°05’26” -9.12 7.51 this work Akköy Ak24 bedded N 37°56’54” E 29°05’26” -9.86 5.80 this work Akköy Ak25 bedded N 37°56’54” E 29°05’26” -9.59 5.34 this work Akköy Ak25 banded N 37°56’54” E 29°05’26” -10.40 4.71 this work Akköy Ak26 banded N 37°56’55” E 29°05’27” -8.78 7.52 this work Akköy Ak27 bedded N 37°56’55” E 29°05’27” -9.56 6.21 this work

Kocabaş Ko1 banded N 37°48’39” E 29°19’01” -14.00 5.42 this work Kocabaş Ko2 bedded N 37°48’39” E 29°18’56” -9.67 5.04 this work Kocabaş Ko2 banded N 37°48’39” E 29°18’56” -11.10 5.51 this work

Pamukkale PI-1 banded N 37°55’44” E 29°06’56” -16.3 5.8 Uysal et al., 2007

Akköy PII-1 banded N 37°56’52” E 29°05’35” -15.2 4.7 Uysal et al., 2007 Akköy PII-3 banded N 37°56’52” E 29°05’35” -14.4 4.8 Uysal et al., 2007

Hanife Hill PIII-1 banded N 37°56’47” E 29°06’19” -14.5 4.9 Uysal et al., 2007 Hanife Hill PIII-2 banded N 37°56’47” E 29°06’19” -12.4 4.7 Uysal et al., 2007 Pamukkale PIV-1 banded N 37°55’47” E 29°06’35” -14.3 5.7 Uysal et al., 2007 Pamukkale PIV-2 banded N 37°55’47” E 29°06’35” -15.0 5.6 Uysal et al., 2007

Pamukkale Range-front Fault HB1 banded N 37°57’12” E 29°06’57” -14.6 4.7 Uysal et al., 2007

Pamukkale Range-front Fault HCE banded N 37°57’08” E 29°06’59” -11.3 5.1 Uysal et al., 2007


References





LGM

banded travertine

banded travertine (Uysal et al., 2007, 2009)

bedded travertinethis work


Uysal et al., 2007, 2009

Figure DR1


LGM

banded travertine

banded travertine (Uysal et al., 2007, 2009)

bedded travertinethis work

Figure DR2

Age (ka B.P.)

Uysal et al., 2007, 2009 U/Th-dates

(this work)


Documents

Growth of fissure ridge travertines from geothermal springs of Denizli Basin, western Turkey