Geology Reference
In-Depth Information
typical for carbonates precipitated from the fresh
water system where
14
C activity of dissolved
inorganic carbon (DIC) is result of mixture of
biogenic/atmospheric CO
2
and dissolved mineral
carbonates, for example, limestone. The d
13
C and
d
18
O values of active and passive tufa samples are
in the range from 29.13 to 26.0‰, and from
28.44 to 27.40‰, respectively (Table 2). More
negative d
13
C values of the G¨ney site indicate
that most of the
mounds in a downslope direction. This is not
always the case, hovever, as new tufa lobes can
develop above older sites of deposition (Pedley
et al. 2003). Furthermore, Mart
´
n-Algarra et al.
(2003) pointed out that well-defined tufa steps in
Granada basin (south Spain) were deposited by
springs that migrated downslope in time.
Tufa-depositing spring waters have distinct
chemical compositions. Most waters emerging
from karstic massifs commonly are of Ca - HCO
3
and Ca - HCO
3
-SO
4
types and supersaturated with
respect to calcite (Andreo et al. 1999; Horvatin
ˇ
i
´
et al. 2005). In some areas, however, the chemical
compositions of karstic waters may be slightly
different as in the Honaz site, Denizli, Turkey
(Horvatin
ˇ
i
´
et al. 2005). In addition, Pentecost
(2000) assumed that the Matlock Bath deposit and
associated spring waters, Derbyshire, UK were
'thermogene travertine' and 'thermal water', respe-
ctively, depending on chemical compositions. The
Matlock Bath waters are really deep-cycled meteo-
ric waters, which are only a few degrees above
ambient temperature (pers. comm., Pedley 2009).
The Matlock Bath site is a good example of a
perched springline tufa model in aspect of its for-
mation and morphology and fits his classification
(Pedley 1990) very well.
At the G¨ney site, the hydrochemical parameters
of the tufa-precipitating waters were monitored sea-
sonally for more than one year (Table 1). Downflow,
the pH, SIC, Mg/Ca values increased whereas the
CO
2
, EC, Ca, Mg, HCO
3,
and DIC decreased. The
Na, K, Cl and SO
4
values did not change during
the study period. The DIC concentrations were
higher in January (winter season) and lower in
November such as in the recent freshwater carbon-
ates of River Krka, Croatia (Lojen et al. 2004).
Apart from the G¨ney site, there are two other
tufa sites in the Denizli basin, one at Sakızcılar
and the other at Honaz. The tufa deposits and
spring waters of G¨ney and Sakızcılar sites show
similar characteristics in water chemistry. In con-
trast, the water and tufa chemistry at the Honaz
site, which is located along a normal fault zone
(Bozku¸ et al. 2000), is different. The water temp-
erature, electrical conductivity, dissolved anion
and cation values of the Honaz site are higher than
the other two tufa sites because of longer residence
time. Waterfall tufa area is one of the most impor-
tant components of perched springline and fluvial
models as investigated in some European sites pre-
viously. Also this area takes place in front of tufa
barrage downstream in a fluvial model and passes
into lake deposits upstream. The most rapid tufa pre-
cipitation occurs in waterfall front of a perched
springline tufa site (Pedley et al. 1996; Pentecost
2000; Pedley et al. 2003; Carthew et al. 2003;
Andrews 2006; Pedley 2009).
14
C is of atmospheric CO
2
and/or
biogenic origin.
Discussion
The prominent features of the studied site were com-
pared with some well-known tufa sites throughout
Europe in Table 3.
The G¨ney waterfall site is composed of coalesc-
ent tufa bodies and includes an active waterfall
area. The two distinct lobe-top terrace areas took
place above the active waterfall. Horizontal -
subhorizontal lobe-top area is an indication of
mature stage of perched springline tufa deposits
(Pedley 1990; Ford & Pedley 1996; Pedley et al.
2003). Consequently, some of the tufa lobes have
reached a mature stage. In the study area, the flat ter-
races possibly developed in consequence of prograd-
ing lobe fronts and issue of antecedent topography.
A self-built channel has been developed at one of
the lobe-top terraces. This kind of channels was
observed at thermal areas in Pamukkale and sur-
roundings in the Denizli basin, western Turkey
(Bean 1971; Altunel & Hancock 1993;
¨
zkul et al.
2002). Similar channels have been mentioned from
another perched springline tufa site (i.e. Rochetta a
Volturno, Italy; Violante et al. 1994), but not seen
in the Spanish examples (Pedley et al. 2003).
The inactive Holocene tufa samples are from
2000 - 5800 yr BP old, based on the
14
C dating
method (Horvatinˇi´ et al. 2005). Age relationships
between coalescent tufa bodies in perched spring-
line sites are quite complex (Mart´n-Algarra et al.
2003; Pedley et al. 2003). In the study area, the
present multi-spring resurgences and individual
tufa lobes coalesced occur at various elevations.
Thus, a regular downslope trend in ages is not
expected. One of the reasons of this situation may
be heterogenity within the aquifer rocks. Earthquake
activity in the region (Kumsar et al. 2008, p. 95 - 98)
may have been changed the settings of the spring
resurgences through time. In some Spanish sites,
regular trends have been recorded in downslope
directions. Gonz´les Mart´n et al. (1989), for
example, indicated that the higest deposits at
Taju˜a sites, NE of Madrid, are no longer active,
whereas those in the valley bottom are still develop-
ing. This suggests a decrease in age of individual
Search WWH ::
Custom Search