Quaternary International 437 (2017) 4e14

Contents lists available at ScienceDirect

Quaternary International
journal homepage: www.elsevier.com/locate/quaint

Diagenesis of continental carbonate country rocks underlying surficial

travertine spring deposits


Alvaro Rodríguez-Berriguete a, *, Ana María Alonso-Zarza a, b, Rebeca Martín-García a
a
Department of Petrology and Geochemistry, Complutense University, Jose Antonio Novais st 2, 28040, Madrid, Spain
b
Geosciences Institute IGEO-CSIC, Jose Antonio Novais st. 2, 28040, Madrid, Spain

a r t i c l e i n f o a b s t r a c t

Article history: Diagenesis of Miocene-Pliocene continental rocks in extensional Teruel Basin related to deposition of a
Available online 6 October 2016 travertine at surface has been studied. Most of the diagenetic processes described here are in relation to
the travertine deposition, so they are not widespread along the basin. Due to their high induration degree
Keywords: these rocks may be misinterpreted with travertine/tufa facies, so there is a need of clear criteria for their
Diagenesis recognition as they can supply important hydrological and paleoenvironmental information. The
Middle Pleistocene
diagenetic processes, that deeply modified the substrate, include dissolution, recrystallization and
Teruel Basin
cementation and lead to a high induration of the Miocene-Pliocene rocks. These processes were driven
Climate
Terrestrial carbonates
by groundwater flow leading to travertine deposition at surface and appeared to be closely related to the
fluvial incision of Alfambra River and to extensional fracturation during Middle Pleistocene times,
probably under humid conditions. The cementation sequences and the organization of fractures and
vadose micrite point to alternating vadose-phreatic conditions and syndiagenetic movements of the
faults. Isotopic signal show lighter d18O than that of the unaltered Miocene-Pliocene rocks and similar
d13C, suggesting lighter oxygen signal in waters during interglacial Pleistocene periods, and a common
origin for carbon from marine Mesozoic rocks from the substrate of the basin. All of that suggest the
strong control of tectonics, and climate at different timescales in diagenesis of the rocks serving as
substrate for surficial travertine deposition.
© 2016 Elsevier Ltd and INQUA. All rights reserved.

1. Introduction et al., 2012; De Filippis et al., 2013; Gao et al., 2013; Henchiri,
2014) and therefore the related diagenesis of the substrate.
Discontinuities as joints, gravitational fractures, or even In this study we describe the strong modifications of the sub-
inherited volcanic structures can serve as fluid conduits allowing strate of the Villalba Baja travertine (Teruel Basin, Spain) by
the flow of groundwaters to the surface (Pentecost, 2005). As a diagenesis which caused the lithification of the Neogene rocks,
consequence of water flow, new primary calcium carbonates are making it difficult in cases to distinguish between the surface
then formed at surface (travertine ss) and in open fractures and travertine deposits and the diagenetically modified substrate.
other discontinuities (crystalline crusts). Moreover, the flow of Although these modifications can also contain important paleo-
water through the substrate may notably modify it, especially if it environmental signals, they rarely have been discussed in the
consists of carbonates or other permeable rocks. In these cases the literature (Goldstein, 2008; Li et al., 2014). We set out to investigate
travertine deposition at surface is accompanied by diagenetic how changes in hydrology through time plus tectonic processes
modification of the underlying rocks from the substrate. Falls or imprinted characteristic signatures in the rocks that are overlain by
risings of water table due to dry-wet climate cycles of different surface travertine deposition.
temporal scales (annual, decadal, centennial, etc.) can also control
the hydrological system which eventually produces surficial trav- 2. Location and geological setting
ertine deposition, independently of the tectonic activity (Schrøder
Teruel Basin is located in the East of the Iberian Peninsula,
100 km inland from Spanish Mediterranean coast. This basin is a
* Corresponding author. NNE-SSW half-graben of 100 km long, 15e20 km width, filled with

Rodríguez-Berriguete).
E-mail address: [email protected] (A. about 500 m of continental sedimentary rocks over a faulted and

http://dx.doi.org/10.1016/j.quaint.2016.09.013
1040-6182/© 2016 Elsevier Ltd and INQUA. All rights reserved.

Rodríguez-Berriguete et al. / Quaternary International 437 (2017) 4e14
A. 5

folded Mesozoic substrate (Alonso-Zarza and Calvo, 2000; Lafuente Tectonic evolution of the area seems to be controlled mainly by the
et al., 2008; Lafuente and Arlegui, 2009; Gutierrez et al., 2012; activity of Concud fault (Simo
n et al., 2005; Lafuente and Arlegui,
Ezquerro et al., 2014). Sedimentation started during Upper 2009; Lafuente et al., 2011a).
Eocene to Oligocene, but the most continuous sedimentation Travertines and tufas developed during the erosive stage since
occurred through Miocene to Pliocene times. During those times the Pleistocene (Lafuente et al., 2011b; Camuera et al., 2015). This is
clastic alluvial fan deposits formed near the basin margins and the case of the Villalba Baja travertine, developed on top and
lacustrine-palustrine carbonates/evaporites occupied the center of downslope of the Cabezo Agudo Hill which is placed at North of
the basin. Fluvial incision started in the beginning of the Pleisto- Villalba Baja village close to Alfambra River (Figs. 1 and 2A). Within
cene giving place to the formation of four fluvial terrace levels: the hill, Miocene-Pliocene lithologies have been incised and frac-
Upper (probably Lower Pleistocene, 85e90 m above present day tured so at present the hill has a crater like morphology outlined by
Alfambra River level), Middle (Middle Pleistocene, 45e65 m above an arcuate ridge on top and open to the Alfambra River. Both the
actual river level), Lower (Upper Pleistocene, 15e35 m) and Recent Miocene-Pliocene rocks and the travertine are quarried for building
(Holocene, 3.5e5 m) (Gutie
rrez and Pen ~ a, 1976; Pen
~ a et al., 1984; stone.
Lafuente et al., 2011b; Ezquerro et al., 2014).
Teruel Basin may be subdivided into two sectors, Alfambra and 3. Methods
Turia, which are respectively placed at North and South of Teruel
city. The study area is placed at the Alfambra sector close to the 70 thin sections were studied using an optical microscope. XRD
small village of Villalba Baja (Fig. 1). The structure of the Miocene- analyses from 35 samples were performed using a Philips PW-1710
Pliocene rocks from Alfambra Area consists of a wide, asymmetric spectrometer with CuKa at 40 kV and 30 mA. 35 analyses of stable
syncline with gently sloped flanks, the eastern being shorter, and isotopes d18O and d13C were carried out to Scientific-technic ser-
axis coinciding with the basin axis (Corte
s Gracia and Casas Sainz, vices from Barcelona University (Spain). Description of travertine
2000). However, extensional stress dominated the tectonics of bodies was done using historic aerial photographs from the
this area, and such a syncline may be due to higher subsidence in 1973e1986 Spanish Interministerial aerial images series and the
the center than in the margins of the basin (Gutierrez et al., 2012). rejected quarry blocks.

4. Results

4.1. Morphology of travertine bodies

The travertine occurs as different distinct bodies. The main

travertine body which is actually quarried is located at the eastern
end of the arc defined by the hill (Fig. 2A and B). At present neither
the morphology nor the “in situ” facies of the travertine can be
observed due to quarrying activities. So the morphology of the
original build-up has been obtained through the study of the
1973e1986 Spanish Interministerial aerial images series. The main
body consisted of a terraced mound-shaped carbonate body with
some pool-barrier pairs. There were another four travertine bodies
at the eastern side of the incised channel found at the inner part of
the hill. These bodies emerged at the top of the hill and displayed
fan shaped morphology (Fig. 2B). Another travertine body devel-
oped at the western side of the channel, but its morphology re-
mains unclear from aerial photographs. A tongue of indurated rocks
appeared connected to this body downslope (Fig. 2B). This tongue
possesses a convex transversal profile (Fig. 3). The lowest altitude
for the travertines is about 50e60 m above present day Alfambra
River level. The facies of the original travertine are observed on
quarry blocks left in the quarry and consist on Charophyte mounds
and patches (Fig. 4A), bands made of micritic laminae which appear
as gently sloping sets displaying micro-terracettes (Fig. 4B) or
covering Charophyte mounds (Fig. 4C), and eventually travertine
breccia (Fig. 4D).

4.2. Large scale features of the Miocene-Pliocene substrate

Miocene to Pliocene stratigraphic series developed at Cabezo

Agudo Hill show a well stratified pattern, with tabular beds <0.5 m
thick. The Miocene-Pliocene rocks appear tilted, dipping to
Alfambra River (i.e. to the East) in the inner part of the arcuate-
shaped hill, and to the west in the outer part of the hill, the in-
flection point being at the top of the hill.
The series is cross-cut by several fractures of ENE-WSW, NW-SE
and NE-SW directions. Relationships among them are unclear.
Fig. 1. Location of the study area and geological map of the Alfambra-Teruel Area However, all these fracture directions match with those of the
(modified from MAGNA).
s
alpine, tertiary extensional and even late variscan tectonics (Corte
6
Rodríguez-Berriguete et al. / Quaternary International 437 (2017) 4e14
A.

Fig. 2. (A) Digital Elevation Model at 5 m resolution (LiDAR-PNOA © Instituto Geogra
fico Nacional) and contours (10 m; darker contours every 50 m). Stars mark position of studied
carbonate deposits. (B) Aerial views from the Cabezo Agudo Hill in 2012, illustrating the quarry area (PNOA © Instituto Geogra
fico Nacional de Espan
~ a-(Gobierno de Arago
n)) with

rea 1973e1986) with dotted lines delineating the spring
the same stars marking carbonate deposit positions, and (C) historical aerial image (Plan Interministerial de Fotografía Ae
carbonate deposits before quarry activity.

Fig. 3. Image of the western part of the hill showing the convex indurated tongue (arrows) sloping down from the abandoned quarry front.

Gracia and Casas Sainz, 2000). Crystalline crusts developed on similar to micrite of the surrounding matrix where gastropods are
fracture walls (Fig. 5A), and they are not the only feature of water included (Fig. 6B). Micrite may infill partially or completely the
circulation at macroscopic scale. Pipe-like structures consist of gastropod cavities. The former case provides geopetal structures
widening upwards structures made of limestone more hardened with micrite on the base and an empty or partially to totally
and with larger porosity than the surrounding rock (Fig. 5B and C). cemented cavity on top. Few cases indicate that the boundary be-
Therefore these are not infillings but previous rock modification. tween micrite infillings and the cemented cavity is inclined. Os-
Occasionally, structures disrupting stratification have been tracods are in general disarticulated, and valves are usually
observed (Fig. 5D). dispersed in the matrix. Ostracods also show geopetal infillings
similar to those of gastropods, when remain articulated. Charophyte
4.3. Miocene-Pliocene facies gyrogonites are generally unbroken also showing geopetal infillings
(Fig. 6C), but may appear also as accumulated fragments. Oncoids
All analyzed samples are almost 100% low magnesian calcite, range between 1 and 8 mm and exceptionally reach several cm.
with minor amounts of quartz and feldspar. The main facies are: Their nuclei generally consists of fragments of charophyte remains
bioclastic limestone, micritic limestone, and limestone breccia. (Fig. 6D), but also of peloids or intraclasts. The coatings are made of
radially arranged fibers which are non-porous and usually are well
4.3.1. Bioclastic facies preserved.
Bioclastic facies are wackestones-packstones and sometimes
rudstones of different types of bioclasts. They are either banded or 4.3.2. Micritic facies
massive. Banded facies consist of less than 1 mm to more than This is the most abundant facies, and has the wider variety of
5 mm, graded bands of ostracods and other bioclast fragments in a microfacies. It generally consists on mudstones-wackestones and
dense micrite matrix (Fig. 6A). Massive facies commonly contain also peloidal grainstones. Micritic facies are very porous. Four
bioclasts as gastropods, ostracods, oncoids, and charophytes. Ma- microfacies have been defined. Microstromatolitic facies consist of
trix is micritic but often contains few detrital quartz and feldspar thin micritic lamination, generally planar, parallel and regular, but
grains. Gastropods appear mostly unbroken. They commonly show sometimes irregular, and wavy (Fig. 7A). Peloidal is the most com-
micritic infillings of their cavities. Such infillings are in general mon microfacies (Fig. 7B). It appears frequently associated to

Rodríguez-Berriguete et al. / Quaternary International 437 (2017) 4e14
A. 7

Fig. 4. (A) Charophyte patch; (B) Laminated micritic travertine; (C) Charophyte mound covered by laminated micritic travertine (scale is 1.40 m); (D) Travertine breccia covered by
laminated micritic travertine. (Full hammer length is 33 cm).

Fig. 5. (A) Abandoned quarry front with a fractured zone (man height used for scale is about 1.85 m); (B) Pipe-like zone consisting of more indurated than the surrounding
carbonate; (C) Vertical pipe-like structures from a rejected block of the quarry (arrow pointing in water flow direction); (D) Structure disrupting the stratification and producing
brecciation (dotted line delimiting the fracture walls). (Hammer length is 33 cm).

intraclasts and sparse small bioclasts. Sometimes peloids are between adjacent clasts. Fragments range from several millimeters
grouped on loosely shrubby-shaped aggregates. to several centimeters. Voids are of the same magnitude and are
Intraclastic microfacies are made of angular micritic intraclasts totally cemented.
less than 1 mm in size (Fig. 7C). Both, peloidal and intraclastic facies
show high interparticle porosity if micritic matrix is absent. Massive 4.4. Diagenetic features/processes of the Miocene-Pliocene facies
microfacies are sometimes porous but generally dense micrite with
sparse bioclasts. This facies is darker than the other micrites. 4.4.1. Pre-travertine diagenesis
Micritization mainly affected bioclasts (mainly gastropod shells
4.3.3. Breccia facies and charophyte remains, occasionally ostracods). In the case of
This facies consists of fragments of the previous facies, strongly charophyte gyrogonites micritization may be destructive (i.e.
cemented by coarse crystalline calcite (Fig. 7D and E). Brecciation restricted to the original gyrogonite outline) and also constructive
might occur in situ, as is suggested by coincident boundaries (Fig. 8A), producing thickening of their outlines.
8
Rodríguez-Berriguete et al. / Quaternary International 437 (2017) 4e14
A.

Fig. 6. Bioclastic facies: (A) Banded bioclastic facies; (B) Detail of banded bioclastic facies showing dissolved clasts mainly of ostracods; (C) Gastropods with micritized shell,
geopetal micritic infillings, prismatic cements and recrystallized matrix; (D) Charophyte gyrogonites with micritic geopetal infillings.

Dissolution affected micritic particles (peloids, intraclasts) but Spar Cementation 2 consists of equant mosaics, with crystals
also bioclast shells (mainly those of gastropods and ostracods). It being both, larger than and without size transition from the pre-
produced an increase in porosity amount and size, enhancing both vious cementing phases. Crystals are hundreds of microns in size,
primary porosity in micritic facies and moldic porosity in bioclastic equidimensional and subhedral, producing equant mosaics
facies. Secondary voids from micritic facies are 1 mm in size, with (Fig. 8C).
irregular and often elongated shapes (Fig. 8B).
Recrystallization to Microspar occurred on micritic matrix, 4.4.2. Syn-travertine diagenesis
components such as peloids (Fig. 8C), and infillings of bioclast Dissolution during this period produced the largest voids,
voids. There is no evidence of recrystallization of previously ranging from several millimeters to centimeters mainly in micritic
micritized bioclast shells. facies. Voids formed in this stage usually have irregular morphol-
Spar Cementation 1 is seen as small (tens of micrometers) ogies consisting on planar top walls and concave basal parts.
equidimensional spar crystals surrounding the dissolution voids, Prismatic Cementation Crusts and Micrites: Crystalline crusts are
peloids and intraclasts (Fig. 8B and C). made of prismatic crystals ranging from several hundreds of

Fig. 7. (A) Microstromatolitic microfacies; (B) Peloidal facies with voids cemented by spar crystals; (C) Intraclastic facies cemented by spar crystals; (D) and (E) Scanned thin sections
of breccia facies totally cemented.

Rodríguez-Berriguete et al. / Quaternary International 437 (2017) 4e14
A. 9

Fig. 8. (A) Charophyte gyrogonite showing constructive micritized outline (arrow showing the thickened micrite envelope); (B) Peloidal facies with spar cement phase 1 (arrows
pointing to; (C) Peloids and intraclasts with spar cement phases 1 and 2, and large voids due to dissolution (arrow is over the void); (D) Detail of peloidal facies with spar cement
phases 1 and 2, and recrystallized micrite to microsparite.

micrometers to more than 2 mm. They grew normal-to-subnormal dissolution around micrite containing gyrogonites, cementation
to walls of open fractures (Fig. 8D), or large voids (Fig. 9A and B), by crystalline crusts and subsequent micritization producing
and frequently display corrosive boundaries with the substrate on bands (Fig. 9E).
which they grew. Such corrosive boundaries are apparently unre-
lated to mechanical erosion, being therefore produced due to
4.4.3. Late (post-travertine times?) diagenesis
chemical disequilibrium between micrite and the water precipi-
Micritization of large prismatic crystals and spar mosaics
tating these prismatic crystals. Usually they alternate with thin
occurred related to organic filaments (Fig. 9F) in small fractures and
micritic bands which drape the apexes of the previous crystalline
pore walls. The boundary between micritized and sparite crystal is
crust. In other cases the thicknesses are opposite, that is prismatic
irregular. This process is not localized as bands within the crystals
crystals are smaller and the micritic infillings may reach more than
as in the previously described micritization, but occurred near the
2 mm. Sometimes, this kind of crystals can grow within large voids,
pore walls or small fractures, and results in massive micritized
surrounding their walls and without any micritic infilling (and
areas. Such massive micritized areas can contain relicts of the
eventually are followed by subsequent spar cementation 3).
previous sparitic crystal and areas which are not totally micritized
Brecciation affected the micritic facies but also the crystalline
consisting on microsparite crystals.
crusts growing over them. Brecciation resulted in the formation of
mm to cm size angular fragments. It seems to have occurred in situ
as is suggested by the coincident boundaries between adjacent 4.5. Stable isotopes
fragments (Fig. 9C).
Spar Cementation 3 occurred in breccias, probably immediately 34 selected samples provided d18O values between
6.35
after brecciation but also in large voids (Fig. 9B and C). It usually and
8.43‰ (mean value:
7.63 ± 0.41‰) (VPDB), and d13C values
overlies the crystalline crusts, and also appear cementing the cav- from
2.08 to
6.11‰ (mean value:
3.89 ± 1.17‰) (VPDB)
ities partially filled with prismatic cementation crusts. This (Fig. 10A). There is one sample displaying anomalous isotopic
cementing phase consists of drusy calcite mosaics. Crystals range values of
1.02‰ (VPDB) in d13C and
5.76‰ (VPDB) in d18O.
from tens of micrometers close to the pore wall to hundreds of Positive correlation (R2 ¼ 0.45) has been obtained between both
micrometers to the center. Crystals are equidimensional, anhedral isotopes using all samples, but R2 decreases to 0.37 if the anoma-
to subhedral, and occasionally are interlocked. lous sample is discarded. d18O and d13C were also plotted by facies
Recrystallization accompanied the syn-travertine cementation (Fig. 10B) and compared to reported stable isotope data for the
phases and produced changes from micrite to microsparite. basin (Fig. 10C). Coarse crystalline facies have the lower d13C values
Micritization of Large Crystals. Large prismatic crystals from whereas bioclastic facies have the higher d13C values. Micritic
crusts may appear partially micritized. This micritization often did facies appear across the whole d13C range, but are mainly associ-
not affect the whole crystal, but it occurred in bands (Fig. 9D and E) ated with the higher values. The other facies do not seem to display
and typically produces degraded crystal apexes. Therefore, these any trend. Differences between crystalline and micritic facies are
large crystals appear alternatively micritized or unmicritized in general high in both isotopes, mainly in d13C. However, differ-
along their length. Coated grain-like morphologies in micritic ences in d13C and d18O are very low between crystalline crusts and
facies containing charophytes have been produced as a result of banded facies.
10
Rodríguez-Berriguete et al. / Quaternary International 437 (2017) 4e14
A.

Fig. 9. (A) Void filled with large calcite crystals; (B) Void cemented by prismatic crystals surrounding the void and spar mosaic cement 3 closing the porosity; (C) Breccia cemented
by spar mosaic of cement phase 3; (D) Crystalline crusts with crystals only partially micritized in their upper part; (E) The dissolution of micrite has formed large voids filled by
crystalline crusts which nucleate on charophyte debris. Note that crystal apexes are degraded (arrow) suggesting micritization instead recrystallization of micrite to sparite; (F)
Large crystals covered by filaments and partially micritized. Arrow points to small unmicritized relict of sparite crystal.

5. Discussion (surficial) part made of several travertine bodies. The diagenetic

modification described has only been recognized in this area of the
5.1. Villalba Baja spring system and the relative age of the basin, and is related to the travertine occurrence. Groundwater flow
diagenetic processes in this small area occurred through both, open fractures and
permeable carbonates, as is suggested by the presence of open
Travertine deposition at surface has its lower altitude at about fractures with crystalline crusts and pipe-shaped zones of modified
50e60 m above the present day Alfambra River level, which fits substrate.
very well with the highest altitude of the Middle Terrace (Middle
Pleistocene) (Gutie
rrez and Pen~ a, 1976; Pen
~ a et al., 1984; Lafuente 5.2. Modification of stable isotope signals due to diagenesis
et al., 2011a, b; Ezquerro et al., 2014). This suggests that the pro-
cesses that triggered travertine deposition and also diagenesis of Stable isotopes from Villalba Baja samples were compared to the
the substrate may have started during and/or immediately after reported isotope signals from Miocene rocks of Teruel Basin
that stage of fluvial incision of Alfambra River. The system (Fig. 10C) and fluvial tufas, also probably Pleistocene, deposited
described in the present work consists of an inner part formed by over the Mesozoic rocks from the eastern basin boundary in an area
fractures with crystalline crusts at their walls and permeable dia- close to Villalba Baja (Tortajada system; Camuera et al., 2015). Vil-
genetized zones of Miocene-Pliocene substrate, and an outer lalba Baja samples seem to have lighter d18O and similar to slightly

Rodríguez-Berriguete et al. / Quaternary International 437 (2017) 4e14
A. 11

Fig. 10. Stable isotopes from Villalba Baja area: (A) General plot; (B) By facies; and (C) Compared to other reported rocks from Teruel Basin.

higher values of d13C compared to that reported from Miocene suggests that diagenesis and travertine deposition at Villalba Baja
rocks (Alonso-Zarza and Calvo, 2000; Alonso-Zarza et al., 2012). should probably occur during a humid period.
This lighter oxygen signal may account for different processes than The lowest d13C values of Villalba Baja samples are similar to the
those affected the Miocene-Pliocene sedimentation, as for example highest d13C values of Tortajada fluvial tufa (Camuera et al., 2015)
lower evaporation in Pleistocene than in Miocene, as rainwater and similar to slightly higher than those of Miocene-Pliocene rocks
recharged the same aquifer during those times. Mean calcite d18O in (Alonso-Zarza and Calvo, 2000; Ezquerro et al., 2014). Miocene to
Pleistocene tufas from eastern Spain (Sancho et al., 2016) and mean Pliocene carbonates from Teruel Basin should display carbon
d18O from Villalba Baja diagenetic calcite are very similar. This isotope values close to that of Tortajada because both had the same
12
Rodríguez-Berriguete et al. / Quaternary International 437 (2017) 4e14
A.

carbon source: Mesozoic rocks from the basement of the basin.

Differences in carbon isotope signal between lacustrine-palustrine
carbonates and tufas may account for differences in precipitation
processes and physico-chemical parameters of CaCO3-precipitating
waters. However, the signal of Villalba Baja samples may be more
complex, as primary (e.g. crystalline crusts in open fractures) but
also diagenetic signals are involved. In fact, values from Villalba
Baja samples are between those of Tortajada tufa (Camuera et al.,
2015) and unaltered Miocene to Pliocene rocks. This suggests that
diagenetic modifications locally lead to oxygen isotope depletion
(relative to whole rock signal) and possibly different carbon isotope
variations (sometimes enrichment, other times depletion)
depending on the intensity and type of diagenetic processes
affecting these rocks.

5.3. Diagenesis, tectonics and climate

5.3.1. Order of the diagenetic processes

The fact that strong modifications of Miocene-Pliocene facies
have only been observed in rocks from Villalba Baja suggests that
such modifications are related to the hydrological system which
also produced travertine deposition at surface. However, micrites
seem to be indurated before the strong diagenetic modifications
described here. The less modified micrites only show micritization
of some bioclasts, slight dissolution, aggrading recrystallization
from micrite to microsparite, and two phases of calcite spar ce-
ments (Fig. 11). All of that suggests early lithification of Miocene-
Pliocene rocks with low burial. Such lithification has been re-
Fig. 11. Sequence of diagenetic processes affecting the Mio-Pliocene carbonates prior
ported from other lacustrine-palustrine carbonates from Madrid and during travertine deposition.
Basin (Spain) of similar age (Wright et al., 1997). These authors
argued that recrystallization from micrite to microsparite needs a
The common dynamic of the systems show alternating vadose-
driving force such as stabilization of mineralogy (i.e. from aragonite
phreatic conditions indicated by the micrite bands alternating with
or HMC to LMC). XRD data suggests all samples studied here are
spar cements (Alonso-Zarza et al., 2012), but also changes in water
LMC, and no aragonite relicts have been found. However, provided
saturation which explain the micritization of some bands of the
the source rocks are Mesozoic marine carbonates from the basin
prismatic crystals (Martín-García et al., 2009, 2014). In addition the
bedrock (containing limestones and dolostones) and therefore
change in crystal morphologies from prismatic to equidimensional
there is a source for Mg, it is possible that the original micrite
mosaics is probably related to water becoming progressively poorer
contained some HMC. This recrystallization probably occurred after
in Mg (Folk, 1974).
some dissolution and related at least to the first described
cementation phase. On the other hand there was at least one strong
dissolution process prior to strong cementation. Such dissolution 5.3.2. Tectonic and climatic controls
generated very large voids which do not seem to have generated The first diagenetic process affecting the Miocene-Pliocene
collapses of the micrite facies, but generated local brecciation, with substrate was dissolution. Dissolution was produced at a basin
breccias supporting the early lithification of Miocene-Pliocene scale coeval with the entrenchment of the fluvial network. Disso-
rocks prior to the subsequent strong diagenetic modifications. lution, due to the percolation of meteoric, calcite-undersaturated
Therefore, diagenetic processes driven by hydrological system waters still goes on at present, but started prior to travertine
which generated travertine at the surface started after the strong deposition, probably during the Middle Villafranquian, producing
dissolution process (Fig. 11). Crystalline crusts alternating (or not) karstic features widely described in the literature (Webb et al.,
with micrite bands occupied the newly generated voids and sur- 1992; Jaillet et al., 2004; Alonso-Zarza et al., 2012). In addition to
faces. Breccias did not form immediately after this dissolution, as is basin scale dissolution, the study area shows larger (now cemen-
suggested by broken crystalline crusts found in them. Therefore ted) voids which are present only in the strongly modified facies.
there was another process different to dissolution leading to brec- This local dissolution could be driven by the arrival of groundwa-
ciation. This process was probably not a seismic event as brecciation ters with higher temperature than that of surrounding rocks which
is not a general feature of the country rocks (i.e. there are not very cooled in their contact with the rocks and then became undersat-
many breccia facies). Rather, the fact that the rock had been urated with respect to calcite (Rossi et al., 2002; Goldstein, 2008).
cemented by crystalline crusts prior to breakage, and that the rock The entrance of these waters, driven probably by the extensional
was totally cemented immediately after the breakage by calcite spar tectonic pulses and related seismic activity in Teruel Basin during
mosaics (under phreatic conditions), may suggest hydromechanical Middle-Upper Pleistocene (Lafuente et al., 2008), also contributed
fracturation as the brecciation cause. This process has been to the hydromechanical fracturation of the substrate. The diage-
described for burial diagenetic and hydrothermal environments netically modified substrate also records signs of coetaneous
(Wangen, 2001; Katz et al., 2006). The local and restricted character movements of the structures which may be related to this
of this brecciation together with pipe-shaped structures also points tectonically-influenced activity, as for example the originally hori-
to hydromechanical fracturing. In any case, brecciation seems to zontal vadose micrite infillings with different inclinations in the
have been an isolated process eventually disrupting the usual dy- same sample, or the progressive change of inclinations through
namics of the hydrological/diagenetic system. laminae from the crystalline crusts related to fractures (Fig. 8F). The

Rodríguez-Berriguete et al. / Quaternary International 437 (2017) 4e14
A. 13

inclination of the laminae may suggest that the movements were information to a better understanding of the hydrological system
mostly related to the progressive tilting of the blocks due to frac- giving place to travertine formation and it significance within the
turing. In other words, progressive tilting and water circulation basin infill.
were the dominant processes during travertine deposition, how-
ever the local brecciation events may reflect some seismic activity. Acknowledgements
The intense cementation processes, the lithification of the sub-
strates and the formation of the surface travertine bodies, required We acknowledge Alfonso Mele
ndez and Andrea Martín for their
a relatively continuous water recharge of the aquifer, and subse- discussion and assistance during fieldwork. Special thanks to the
quent incorporation of Ca2þ and CO2 as dissolved inorganic carbon Editors and reviewers (Drs. Brasier and Huerta) for their helpful
(DIC). Therefore, climatic conditions should be humid enough (i.e. contributions to improve the manuscript. This work was funded by
periodic rains at recharge areas) to allow the dissolution of the project CGL2014-54818-P from the Spanish Ministerio de Ciencia e
Mesozoic and Cenozoic substrates. The more favorable periods for Innovacio
n and is a contribution to UCM Research Group 910404.
tufa formation in this area and the closer Ebro Basin are the humid Ma
rmoles Llorens Company is thanked for the facilities given to
interglacial times during Upper Pleistocene, but also during Middle work in the quarry area.
Pleistocene MIS 11, 7, and glacial but humid MIS 6 (Sancho et al.,
2016). So very probably the formation of the travertine and the References
modification of the substrates occurred during the relatively more
humid periods. However the presence of alternating micrite-spar Alonso-Zarza, A.M., Calvo, J.P., 2000. Palustrine sedimentation in an episodically
subsiding basin: the Miocene of the northern Teruel Graben (Spain). Palae-
cements in some voids and in the crystalline crusts suggests ogeography, Palaeoclimatology, Palaeoecology 160, 1e21.
alternating vadose-phreatic conditions in a shorter timescale, Alonso-Zarza, A.M., Mele
ndez, A., Martín-García, R., Herrero, M.J., Martín-Pe
rez, A.,
which may respond to hydrological changes throughout a year or 2012. Discrimination between tectonism and climate signatures in palustrine
deposits: Lessons from the Miocene of the Teruel Graben, NE Spain. Earth-
longer temporal scales (i.e. droughts followed by wetter periods). Science Reviews 113, 141e160.
Camuera, J., Alonso-Zarza, A.M., Rodríguez-Berriguete, A., Mele
ndez, A., 2015. Var-
6. Conclusions iations of fluvial tufa sub-environments in a tectonically active basin, Pleisto-
cene Teruel Basin, NE Spain. Sedimentary Geology 330, 47e58.
Corte
s Gracia, A.L., Casas Sainz, A.M., 2000. ¿Tiene el Sistema de fosas de Teruel
The studied area of Villalba Baja consists of surficial travertine origen extensional? Revista de la Sociedad Geolo
gica de Espan ~ a 13 (3e4),
deposits which are difficult to observe due to quarry activity, and 445e470.
De Filippis, L., Faccena, C., Billi, A., Anzalone, E., Brilli, M., Soligo, M., Tuccimei, P.,
the subsurface diagenetically modified Neogene substrate. The
2013. Plateau versus fissure ridge travertines from Quaternary geothermal
latter may be considered as the roots of the travertine deposit. springs of Italy and Turkey: Interactions and feedbacks between fluid discharge,
Surficial travertine and the modification of the continental Neogene paleoclimate, and tectonics. Earth-Science Reviews 123, 35e52.
Ezquerro, L., Luzo
n, A., Navarro, M., Liesa, C., Simo
n, J.L., 2014. Climatic vs. tectonic
substrate occurred during the Middle Pleistocene in an extensional
signals in a continental extensional basin (Teruel, Ne Spain) from stable isotope
setting within an open sedimentary basin. (d18O) and sequence stratigraphical evolution. Terra Nova 26 (5), 337e346.
The diagenetic modification was the result of the interaction of Folk, R.L., 1974. The natural history if crystalline calcium carbonate: effect of mag-
several processes which include: micritization, dissolution, nesium content and salinity. Journal of Sedimentary Petrology 44, 40e53.
Gao, J., Zhou, X., Fang, B., Li, T., Tang, L., 2013. U-series dating of the travertine
cementation, recrystallization, vadose micrite infills, and breccia- depositing near the Rongma hot springs in northern Tibet, China, and its pa-
tion. Some of these processes repeated through time and occurred leoclimatic implication. Quaternary International 298, 98e106.
prior, during and after travertine deposition. However, the fact that Goldstein, R.H., 2008. Fluid inclusion geothermometry in sedimentary systems:
from paleoclimate to hydrothermal. SEPM Special Publication 103, 45e63.
only in this area of the basin the primary Neogene facies (bioclastic, Gutie
rrez, M., Pen
~ a, J.L., 1976. Glacis y terrazas en el curso medio del río Alfambra
micritic and breccias) are more diagenetically modified permits to (provincia de Teruel). Boletín Geolo
gico y Minero 87 (6), 561e570.
identify the processes which are linked to the supply of water to Gutierrez, F., Carbonel, D., Guerrero, J., McAlpin, J.P., Linares, R., Roque
, C.,
Zarroca, M., 2012. Late Holocene displacement on fault scarps related to
form the surficial travertine. Dissolution and brecciation were
interstratal dissolution of evaporates (Teruel Neogene Graben, NE Spain).
probably driven by hydrodynamic fracturation and circulation of Journal of Structural Geology 34, 2e19.
warm water within the created fractures, which were the conduits Henchiri, M., 2014. Quaternary paludal tufas from the Ben Younes spring system,
Gafsa, southwestern Tunisia: Interactions between tectonics and climate.
for water supplies to construct the surface build-ups. The creation
Quaternary International 338, 71e87.
of the fractures was due to the extensional tectonics within the Jaillet, S., Pons-Branchu, E., Brulhet, J., Hamelin, B., 2004. Karstification as
basin. Water, Ca2þ and CO2 were supplied from groundwater geomorphological evidence of river incision: the karst of Cousance and the
circulating through the Mesozoic and Neogene deposits of the ba- Marne valley (eastern Paris Basin). Terra Nova 16, 167e172.
Katz, D.A., Eberli, G.P., Swart, P.K., Smith Jr., L.B., 2006. Tectonic-hydrothermal
sin. Isotopic data suggest that there was very low or no input of brecciation associated with calcite precipitation and permeability destruction
endogenous CO2 and the obtained values indicate the interaction of in Mississipian carbonate reservoirs, Montana and Wyoming. AAPG Bulletin 90
meteoric fluids mainly with the Mesozoic substrate. Water circu- (11), 1803e1841.
Lafuente, P., Rodríguez-Pascua, M.A., Simo
n, J.L., Arlegui, L.E., Liesa, C.L., 2008. Sis-
lation also caused cementation, micritization and recrystallization mitas en depo
sitos pliocenos y pleistocenos de la fosa de Teruel. Revista de la
of the micrite Neogene muds. Sociedad Geolo
gica de Espan ~ a 21 (3e4), 133e149.
Travertine deposition and the related diagenesis of the substrate Lafuente, P., Arlegui, L., 2009. Where faults meet: palaeostress analysis at the
juncture of the Concud and Teruel faults. Trabajos de Geología 29, 417e423.
probably occurred during a humid (probably interglacial) period. Lafuente, P., Arlegui, L.E., Casado, I., Ezquerro, L., Liesa, C.L., Pueyo, O.,
Simo
n, J.L.,
The common alternating vadose-phreatic conditions during the 2011a. Geometría y cinem
atica de la zona de relevo entre las fallas neo
geno-
activity of this hydrogeological system may reflect short-term cuaternarias de Concud y Teruel (Cordillera Ibe
rica). Revista de la Sociedad
Geolo
gica de Espan ~ a 24 (1e2), 117e133.
climate fluctuations.
n, J.L., 2011b. Paleoseismological analysis of
Lafuente, P., Arlegui, L.E., Liesa, C.L., Simo
Our study shows that the hydrological system that determines an intraplate extensional structure: the Concud fault (Iberian Chain, eastern
the formation of surface travertines also diagenetically modifies the Spain). International Journal of Earth Sciences 100, 1713e1732.
Li, Z., Goldstein, R.H., Franseen, E.K., 2014. Climate, duration, and mineralogy con-
bedrock substrate in which the water circulates. If this occurs
trols on meteoric diagenesis, La Molata, southeast Spain. Interpretation 2 (3),
within continental carbonates (e.g. palustrine or groundwater cal- SF111eSF123.
cretes) it may be difficult to distinguish the surface travertine de- Martín-García, R., Alonso-Zarza, A.M., Martín-Pe
rez, A., 2009. Loss of primary
posits from the diagenetically modified continental substrates. Up texture and geochemical signatures in speleothems due to diagenesis: Evi-
dences from Castan ~ ar Cave, Spain. Sedimentary Geology 221, 141e149.
to now very few papers have described these modifications, despite Martín-García, R., Alonso-Zarza, A.M., Martín-Pe
rez, A., Schro €der-Ritzrau, A.,
their distribution and characteristics containing invaluable Ludwig, T., 2014. Relationships between colour and diagenesis in the aragonite-
14
Rodríguez-Berriguete et al. / Quaternary International 437 (2017) 4e14
A.

calcite speleothems in Basajaún Etxea cave, Spain. Sedimentary Geology 312, for interstadials in the Levant: The chronology and climate history of the
63e75. Palaeolithic cultures from Yabroud (Syria). Quaternary International 257,
Pentecost, A., 2005. Travertine. Springer-Verlag, Berlin. 27e33.
~ a, J.L., Gutie
Pen
rrez, M., Ib

an~ ez, M.J., Lozano, M.V., Rodríguez, J., S
anchez, M.,
n, J.L., Lafuente, P., Arlegui, L.E., Liesa, C.L., Soriano, M.A., 2005. Caracterizacio
Simo
n
Simo
n, J.L., Soriano, M.A., Yetano, L.M., 1984. Geomorfología de la provincia de paleosísmica preliminar de la falla de Concud (fosa del Jiloca, Teruel). Geogaceta
Teruel. Instituto de Estudios Turolenses, Teruel. 38, 63e66.
Rossi, C., Goldstein, R.H., Ceriani, A., Marfil, R., 2002. Fluid inclusions record thermal Wangen, M., 2001. A quantitative comparison of some mechanisms generating
and fluid evolution in reservoir sandstones, Khatatba Formation, Western overpressure in sedimentary basins. Tectonophysics 334, 211e234.
Desert, Egypt: a case for fluid injection. American Association of Petroleum Webb, J.A., Fabel, D., Finlayson, B.L., Ellaway, M., Shu, L., Spiertz, H.-P., 1992.
Geologists Bulletin 86, 1773e1799. Denudation chronology from cave and river terrace levels: the case of
Sancho, C., Arenas, C., V
azquez-Urbez, M., Pardo, G., Lozano, M.V., Pen ~ a-Monne
, J.L., the Buchan Karst, southeastern Australia. Geological Magazine 129 (3),
Hellstrom, J., Ortiz, J.E., Osa
car, M.C., Auque
, L., Torres, T., 2016. Climatic impli- 307e317.
cations of the Quaternary fluvial record in the NE Iberian Peninsula over the last Wright, V.P., Alonso-Zarza, A.M., Sanz, M.E., Calvo, J.P., 1997. Diagenesis of Late
500 ka. Quaternary Research 84 (3), 398e414. Miocene micritic lacustrine carbonates, Madrid Basin, Spain. Sedimentary Ge-
Schrøder, N., Jensen, G.M., Limborg, M., 2012. Groundwater spring deposits as proxy ology 114, 81e95.