Proceedings World Geothermal Congress 2000

Kyushu - Tohoku, Japan, May 28 - June 10, 2000

GEOLOGY AND GEOCHEMISTRY

OF THE SALIHLI GEOTHERMAL FIELDS, TURKEY

Gultekin Tarcan1, Sevki Filiz1 and Unsal Gemici1

1
Dokuz Eylul University, Geological Engineering Department, 35100-Bornova-Izmir, Turkey

Key Words: Salihli geothermal fields, geochemistry, geothermal activities along the graben indicate. Owing to the
hydrology, geothermometer, ion exchange, saturation index intense tectonic activity of the study area, many thermal
waters, cold springs and running waters were sampled and
ABSTRACT chemically analysed monthly during the period from 1991
December to 1992 July to define the chemical characteristics
The Salihli geothermal fields are located in the southern part of of the waters. For samples collected, temperature, pH, and
the Gediz Graben, which is one of the Western Turkey electrical conductivity were measured in the field. The
grabens. Salihli geothermal systems have been physically remaining chemical constituents of the waters were analysed
divided into four main groups; the Sazdere, Kursunlu, later in the laboratory with techniques described in APHA-
Caferbeyli and Sart-Camur geothermal fields. The reservoir AWWA-WPCF (1980). This paper discusses the
rocks of the geothermal systems are karstic marbles and thermomineral waters located in the geothermal fields of
fractured metamorphic rocks belonging to the Mesozoic Salihli on the basis of hydrological and geochemical
Menderes Massif, which is basement in the study area. Since assessments.
the clayey levels of the overlying Neogene Gobekli and
Acidere units have very low permeabilities, they are the cap 2. GEOLOGICAL AND HYDROLOGICAL SETTING
rock of the systems. Heat sources may be magmatic intrusions
closed to surface, rising along the young faults caused by Geologic maps and models produced by Emre (1996) were
graben tectonism. As with many other geothermal systems in used to explain the relations between geothermal systems and
Western Turkey, all the thermomineral waters in the study area geologic structure. Also were used the same stratigraphic rock
are of meteoric origin and circulation in these systems is units described in Emre (1996). The basement of the study area
closely related to tectonic activity. consists of Menderes Massif rocks, and is made up of high to
low grade metamorphics (gneiss, mica schists, phyllites, quartz
Most of the thermomineral waters in the study area are of the schists, marbles) and granodiorite. The proposed age of the
sodium bicarbonate type. Cold waters in the area are mainly Menderes Massif rocks ranges from Early Triassic to
2+ 2+ – Uppermost Cretaceous (Erdogan and Gungor, 1992). Neogene
dominated by Ca , Mg , and HCO3 ions, and often have no
dominant cation or anion. The major hydrogeochemical sedimentary rocks and Holocene Kula volcanics, which are
processes for thermomineral waters in the geothermal fields are mainly made up of basaltic lava, unconformably cover the
+ 2+ 2+ basement rocks (Fig. 1). Neogene-Quaternary sedimentary
ion exchange between Na and Ca and/or Mg cations,
which are also called natural water softening reactions. The units occur in different facies in the northern and southern parts
thermomineral waters of the study area fall mostly into the of the Gediz Graben, which extends in WNW-ESE direction.
Ca
2+
and Mg
2+
montmorillonite fields and partially the The Acidere formation is mainly made up of pebbles, pebbly
kaolinite and K-feldspar fields on the activity diagrams. The sandstones, and claystone-mudstones. The Gobekli formation
waters are mainly undersaturated with respect to carbonate and consists mainly of intercalated conglomerate, pebbly
sulfate minerals. One important environmental problem in the sandstone, and sandstone. The Asartepe formation consists of
study area is boron contamination in aquifers and soils. The conglomerates comprising sandstone intercalations. The
presence of boron is attributed to the geothermal systems of the Filiztepe formation, which is exposed in the northern part of
study area. To prevent boron contamination of cold waters graben, consists mainly of limestones. The Mevlutlu formation
used for irrigation in the study area, re-injection of produced consists of alternating conglomerate, pebbly sandstone,
thermomineral waters into the geothermal reservoir is sandstone and mudstone.
necessary. Assessments of empirical chemical
geothermometers and mixing models applied to the Permeability within the Menderes Massif rocks is highly
thermomineral waters, which have measured discharge variable and is related to rock and fracture types. Mesozoic
o o
temperatures ranging from 37 C to 155 C, suggest that carbonates (marbles and dolomitic marbles) of the Menderes
o o Massif are highly fractured and karstified, and act as a karstic
reservoir temperatures vary between 150 C and 230 C. Due to
the fact that these waters are not in equilibrium with reservoir aquifer for both cold and thermomineral waters depending
rocks, and are probably dominated by a combination of mixing upon location. Granodiorite, gneiss and quartz-schist units of
phenomena, rock dissolution and ion exchange reactions, the the Menderes Massif rocks form fractured rock aquifers.
geothermometer results should be considered tentative. Where the massif rocks are intersected in topographic lows,
numerous cold springs supply soft and very low salinity
1. INTRODUCTION waters, particularly in the southwestern part of the study area.
Where poorly permeable and impermeable rocks such as schist
The Salihli geothermal fields are located in the southern parts and phyllite units underlie the karstic and fractured aquifers at
of the Gediz graben, which is one of the Western Turkey depth, natural springs are confined to fault and fracture zones
grabens (Fig. 1). The semi-arid climate of the area is and discharge thermomineral waters. Neogene Gobekli and
characterised by hot dry summers and warm wet winters. The Acidere units, which are made up of granular alluvial fan
mean annual temperature and the total annual rainfall at Salihli deposits including poorly cemented clayey levels, have very
o
are about 16.5 C and 492 mm, respectively. The region is still low permeabilities and form the cap rocks of the geothermal
active, as the occurrences of earthquakes and the presence of systems. The shallow regional aquifer consist of Holocene

1829
Tarcan et al.

alluvial deposits in the middle of the study area, and of thermomineral waters to the stream. Thermomineral waters from
Pleistocene Mevlutlu formations composed of alluvial deposits Salihli geothermal fields exhibit similar chemical characteristics,
+ 2+ 2+ + – – 2–
in the north. having Na >(Ca +Mg +K ) and HCO3 > (Cl +SO4 ). The
2+
Sazdere thermomineral spring is enriched in Mg . The Sart-
The Salihli geothermal fields are divided into four main groups 2+
Camur spring and MTA2 well are also enriched in Ca as well as
from west to east in the active southern rims of the Gediz 2+
Mg . The variations in groundwater geochemistry for waters in
Graben as the Horzum-Sazdere thermal springs, the Kursunlu
the study area along the anticipated flow direction can also be
thermal springs, the Sart-Camur thermal springs, and the observed in a Piper diagram (Fig. 3). The chemistry of all the
Caferbeyli Geothermal Field (Fig. 1). The Horzum-Sazdere thermomineral waters of the study area is mainly dominated by Na
thermal springs (two springs) are located along the easternmost –
part of the east-west directional tectonic line in the southern and HCO3 ions (>50 %). Sart-Camur thermal spring and MTA2
+ 2+ – 2+
rims of the graben. These thermomineral springs discharge thermal well waters are a Na -Ca -HCO3 (Ca >20 %) water
o + 2+ –
waters at 3 l/s and at a temperature of 35 C. They are not used. type, and Sazdere thermal spring water is a Na -Mg -HCO3
2+
Kursunlu geothermal field has a shallow reservoir about 200 (Mg >20 %) water type. The other thermomineral waters are Na-
meters deep and contains hot springs and wells (Fig. 2). The –
HCO3 rich (all unmentioned ions are < 20 %). Cold waters in the
depth from surface of the reservoir in this geothermal field 2+
study area are different. Dominant ion species are generally Ca
varies between 10 – 200 meters (Yilmazer and Karamanderesi, 2+
and/or Mg . Some waters have no dominant cation species;
1994). The total discharge rate for the MTA1, MTA2 and – 2–
HCO3 and/or SO4 are generally the dominant anions. The
MTA3 wells is 145 l/s and average production temperatures are differences reflect different lithologic and mineralogic controls on
o
90 C. Sart-Camur thermal springs are situated along the water chemistry due to differing flow patterns, and recharge
westernmost part of the southern rim of the Gediz Graben. 2+ 2+ –
sources. Qualitatively, the Ca -Mg -HCO3 water probably
Thermomineral waters in Kursunlu and Sart-Camur geothermal reflects the dissolution of calcite and dolomite in the aquifer
fields are now used for bathing and medicinal purposes. matrix. The chemical reactions involved in the dissolution of the
Caferbeyli Geothermal Field is situated between Sart-Camur carbonate minerals are a combination of hydrolysis and
and Kursunlu fields. When well drilling, was done here in 2+ 2+ – +
dissociation. Other sources of Ca -Mg -HCO3 and Na ions in
o cold waters are silicate weathering and alteration reactions.
1990, the downhole temperature was found to be 155 C at
1189 meters (Karamanderesi, 1997). Owing to the low system +
permeability and resulting low discharge rate (about 2 l/s) However, the high concentrations of Na and HCO3 ions in the
economic fluid production was not possible from this well. thermomineral waters of Salihli geothermal fields cannot be
explained by only dissolution of carbonate and silicate
minerals. These thermomineral waters are probably dominated
As with many other geothermal systems in Western Turkey, by a combination of mixing phenomena, rock dissolution and
the circulation of the thermomineral waters in Salihli ion exchange reactions. The chemical composition of the
geothermal fields is closely related to major fault and fracture thermomineral waters clearly differs from that of the cold
zones. Fractured rocks of the Menderes Massif, such as quartz waters. Compared to the local cold groundwaters and stream
schists, gneiss and granodiorite and karstic marbles, are the + –
waters, all the thermal waters are enriched in Na and HCO3 .
reservoir rocks in the Salihli geothermal fields. The discharge +
This high concentration of Na in thermal waters of Salihli and
rates are high where the reservoir is predominantly formed by the crystalline basement host rock suggests silicate weathering
marbles. Since the clayey levels of the Neogene Gobekli and and alteration. However, if silicate weathering was the
Acidere units are relatively impermeable, they act as the dominant reaction in the geothermal reservoirs, there should be
system cap rock. Heat sources may be magmatic intrusions indications of the presence of SiO2 in the geothermal fields.
closed to surface intruded via young faults. Meteoric waters However, there is no SiO2 cementation near the thermal
circulate in the Salihli geothermal systems. The meteoric 2+ 2+ –
springs. If dissolved Ca , Mg and HCO3 in both cold and
waters penetrate through the faults and fractures, are heated in thermal water results mainly from the dissolution of carbonate
reservoir rocks, and move up to the surface along the faults. 2+ 2+
18 2 minerals, then the proportions of Ca and/or Mg versus
Isotopic data ( O and H ) suggests that thermal waters in –
18 HCO3 should be linear. The correlation coefficients (r) of
Salihli geothermal fields are of meteoric origin although δ O
rich (Filiz et al., 1992). these ions mentioned above in the waters of the study area vary
from 0.08 to 0.10, which indicate a very poor fit (Table 2). On
+
3. GEOCHEMISTRY the contrary, the correlation coefficient between Na and
–
HCO3 ions is 0.86. This is a good fit. This shows a positive
+ –
The results of the chemical analyses of the waters sampled linear relationship between Na and HCO3 ions. In addition,
from the study area are listed in Table 1. Major constituents, there are negative correlations and a very poor fit between Na
+ 2+ 2+ – – 2– 2+
EC, pH, Na , Ca , Mg , Cl , HCO3 , SO4 and SiO2 and and Ca cations. In accordance with the above, the increase in
+ 3+ sodium and the decrease in calcium and magnesium in the
some secondary constituents, K and B values were obtained
by calculating mean annual values during the period from 1991 thermomineral waters can be explained by ion exchange. It
December to 1992 July. The remaining constituents are values would follow the generalised reaction,
2+ + 2+ 2+
from 1992 July. The locations of the water samples are shown M + Na2–Clay=2Na +M–Clay (1) Where, M is Ca or
in Fig. 1 and Fig. 2. Monthly periodic analytical results show 2+
Mg or other alkaline earth metals. Generally, the chemical
no significant changes in temperatures and chemical reactions involved in the dissolution of carbonate minerals (for
constituent concentrations with time for thermomineral waters calcite) can be summarised as;
in the Salihli geothermal fields. However cold waters, 2+ -
CaCO3+CO2+H2O=Ca +2HCO3 (2)
particularly stream waters, showed significant variation in ion In the above reaction, it is considered that roughly, half of the
+ – 3+ –
concentrations. Na , HCO3 and B concentrations for the dissolved HCO3 comes from carbonates, and the rest is from
Kursunlu Stream increase in summer months due to reduced 2+ 2+
the reaction of removing Ca (and/or Mg ) from the water.
flow and a corresponding increase in the contribution of 2+
For Ca the most likely possibility appears to be the natural

1830
Tarcan et al.

+
softening reaction of cation exchange for Na on clay levels of geothermometers seen in Table 4 are impossible since they are
schists or Neogene sediments as discussed below; lower than the measured surface temperatures. Discarding
2+ + these data, the rest of the data can be used to estimate the
Ca +2Naex=Caex+2Na (3)
If this equation (3) is combined with the former reaction (eqn. reservoir temperature. The geothermometry results suggest the
2), We obtain reservoir temperatures of Salihli geothermal fields are between
+ o o
CaCO3+H2CO3+2Naex=Caex+2Na +2HCO3 (4) or 150 C - 230 C. Solute geothermometers give reliable results
2+ – + + – 2+ when the water-rock interactions reach equilibrium. Mixing of
Ca +2(HCO3 )+solid-2Na =2Na +2(HCO3 )+solid-Ca (5)
Dissolution of host rocks as a function of fluid temperatures thermal waters with cold waters while moving up to the surface
occurs at all sites. The hot waters are mixed with different will not lead to equilibration and thus the results are doubtful.
0.5
proportions of cold groundwater and the ion exchange occurs A ternary plot of Na/1000-K/100-Mg was proposed by
through the circulation path of the thermomineral waters. Giggenbach (1988) as a method to determine reservoir
temperature and to recognize waters, which have attained
Major hydrogeochemical processes for cold waters in the study equilibrium with the host lithology (Fig. 4). The graphical
area seem to be carbonate solution and silicate weathering representation of the combined geothermometer for this study
reactions. The major hydrogeochemical processes in the is illustrated in Fig. 4. All the thermomineral waters in the
geothermal fields is ion exchange, which is also referred to as Salihli geothermal fields fall into the immature fields
+ 2+
natural water softening and occurs between Na and Ca indicating none of these waters have attained equilibrium with
2+
and/or Mg cations. Solution of calcareous materials in their associated host rocks. Because these waters are not in
2+ equilibrium with reservoir rocks, and are probably dominated
geothermal systems leads to an increase in Ca which is then
+
exchanged for Na from clay minerals (a number of clayey by rock dissolution, mixing with cold groundwater, and ion
levels are associated with the cap rocks of the Salihli exchange, geothermometer results can only be considered
tentative. Hence, the applications of cation geothermometers
geothermal systems). Such a process yields a NaHCO3 type
groundwater, which also has a thermal and mineral water must be considered in doubt and correlated with the results
character (Tarcan and Filiz, 1997). obtained by mixing models to predict the reservoir
temperature. The thermal waters in the study area have a fast
One of the major environmental problems for cold waters in hydrological circulation and/or mix with the cold groundwaters
the study area is boron contamination of aquifers and soils. along the flow path to the surface. Geothermometry techniques
Boron contents of the thermomineral waters are quite high appear to be unreliable for these waters because none of the
(Table 1). Boron is well correlated with chloride, sodium, fluids are at equilibrium with reservoir rocks.
bicarbonate and silica in the all waters of the study area (Table
2). This suggests that the cause of the high boron contents in Enthalpy-Chloride and Enthalpy-Silica mixing models were
cold groundwaters and surface waters is thermomineral fluids proposed by Fournier (1977b) to estimate reservoir
issuing from the Salihli geothermal fields. All the temperatures and the mixing rates of the cold and hot water
– 3+ components of mixing waters (Fig. 5 and Fig. 6). The
thermomineral waters are mainly HCO3 dominated and B
– estimated reservoir temperatures obtained by the enthalpy
and HCO3 are well correlated with each other. Therefore, chloride mixing model for the thermomineral waters in Salihli
– o o
HCO3 is probably the main anion responsible for the high geothermal fields are from 194 C to 288 C (Fig. 5). The hot
boron content in the thermomineral waters. To prevent boron water components of the thermomineral waters obtained by the
contamination of cold waters used for irrigational purposes, re- enthalpy chloride mixing model are 33% in Sart Camur, 55%
injection of produced thermomineral waters to the geothermal in Celikli, 56% in Kursunlu, 62% in MTA2, and 73% in
reservoir may be necessary. MTA3. The reservoir temperatures obtained by the enthalpy
o
silica mixing models were found to be between 187 C and
It is important to know saturation indices for minerals to o
predict which ones may precipitate during the extraction and 227 C (Fig. 6). The hot water components of the
use of the waters. A saturation index of zero indicates that thermomineral waters (mixing ratios) in the study area
thermodynamic equilibrium exists with the solid phase. A obtained by the enthalpy silica mixing model vary from 27 %
to 90 % (Sazdere=27%, Celikli=34%, Sart Camur=36%,
negative or positive index indicates undersaturation and
oversaturation, respectively. Mineral saturation indices for 13 MTA3=45%, Caferbeyli=75%, MTA2=81%, Kursunlu=90%).
mean annual water analyses were calculated on the basis of
measured surface temperatures using the Solmineq.88 5. SUMMARY AND CONCLUSIONS
(Kharaka et al., 1988) computer program (Table 3). All of the
waters are undersaturated with respect to gypsum and Fractured rocks such as gneiss, quartz schist and granodiorite,
anhydrite, and oversaturated with respect to quartz and and karstic rocks such as marbles of the Menderes Massif form
chalcedony, except for Kursunlu Ufuruk cold spring. All the basement reservoir rocks of Salihli geothermal fields.
saturation indices for calcite, aragonite, dolomite, siderite and Neogene aged Acidere and Gobekli units, including clayey
magnesite reflect undersaturation except for the Caferbeyli levels, are the cap rocks of the Salihli geothermal systems.
deep well and the Sazdere thermomineral spring waters. These Heat sources may be magmatic intrusions closed to the surface
results coincide with the field observations. Gypsum and via the young faults. All the thermomineral waters are of
anhydrite minerals have not been observed in the aquifer meteoric origin. Meteoric waters which percolate through the
systems and there is a paucity of travertine near the springs. faults, cracks, fissures, and karstic voids are heated at depth
and then are convected to the surface. As with many other
geothermal systems in Western Turkey, the circulation of the
4. GEOTHERMOMETER APPLICATIONS
thermomineral waters in the study area is closely related to
tectonic activity. Thermomineral waters of the Salihli
A number of solute geothermometers were used to estimate the
geothermal fields exhibit similar chemical characteristics, having
reservoir temperature of some of the thermomineral waters in + 2+ 2+ + – – 2–
Salihli geothermal fields (Table 4). Some results of the Na >(Ca +Mg +K ) and HCO3 > (Cl +SO4 ). Cold

1831
Tarcan et al.

waters in the study area reflect different water types. Dominant Emre, T. (1996). Gediz Grabeni’nin jeolojisi ve tektonigi.
2+ 2+ Turkish Journal of Earth Sciences. Tubitak, no. 5, pp.171-185,
ion species are generally Ca and/or Mg . Some waters have
– 2– (in Turkish, special issue).
no dominant cation; HCO3 and/or SO4 are generally the
dominant anions. Monthly analytical results showed no
significant changes in temperatures and major and secondary Erdogan, B., and Gungor, T. (1992). Menderes Masifi'nin kuzey
constituents with time for the thermomineral waters in the kanadinin stratigrafisi ve tektonik evrimi. TPJD Bult. Vol.. 4/1
Salihli geothermal fields. However, sodium, bicarbonate, and pp.9-34 (In Turkish).
boron concentrations of cold waters increase during the
summer season. Filiz, S., Gokgoz, A., and Tarcan, G. (1992). Hydrogeologic
comparisons of geothermal fields in the Gediz and Buyuk
Geochemical evolution of thermal waters progresses from Menderes Grabens. XI. Congress of World Hydrothermal
alteration reactions of silicates and carbonates to ion exchange Organisation. 13-18/V/1990, Istanbul-Pamukkale-Turkey,
reactions. All the thermal waters in the study area ultimately Proceedings, pp.129-153.
become a sodium bicarbonate water due to ion exchange
+ 2+ 2+ Fouillac, C., and Michard, G., 1981, Sodium/Lithium ratio in
reactions between Na and Ca and/or Mg cations. Major
hydrogeochemical processes in the study area are a water applied to the geothermometry of geothermal waters.
combination of mixing phenomena, rock dissolution, and ion Geothermics, Vol.10, pp.55-70.
exchange reactions. The waters are mainly undersaturated with
respect to gypsum and anhydrite and oversaturated with respect Fournier, R.O. (1977a). A Review of chemical and isotopic
to the quartz and chalcedony. All the saturation indices for geothermometers for geothermal systems. In Proceedings of
calcite, aragonite, dolomite, and magnesite minerals reflect the Symp. on Geot. En. Cento Sci. Prog., Ankara, pp.133-143.
undersaturation except for the Caferbeyli deep well and
Sazdere thermomineral spring waters. One of the most Fournier, R.O. (1977b). Chemical geothermometers and
important environmental problems is boron contamination of mixing models for geothermal systems. In Proceedings of the
cold groundwaters and surface waters used for irrigational Symp. on Geoth. Ene. Cento Sci. Prog., Ankara, pp.199-210.
purposes by mixing with rising thermal waters. Thermomineral
waters should be reinjected to their own reservoir to prevent Fournier, R.O., and Potter, R.W. (1979). Magnesium
boron contamination of soil and water. This process would also Correction to the Na-K-Ca Chemical Geothermometer.
recharge the geothermal systems. The reservoir temperatures Geochimica et Cosmoschimica Acta, Vol. 43, pp.1543-1550.
calculated from the cation geothermometers, except for the Mg
corrected Na-K-Ca geothermometer, are generally higher than Fournier, R.O., and Truesdell, A. H. (1973). An Empirical Na-
those of silica geothermometers. Chemical geothermometers K-Ca Geothermometer for Natural Waters. Geochimica et
and mixing models indicated that the reservoir temperatures of Cosmochimica Acta, Vol.37, pp.1255-1275.
o o
Salihli geothermal field vary between 150 C and 230 C.
Enthalpy-Chloride and Enthalpy-Silica mixing models Giggenbach, W. F. (1988). Geothermal Solute Equilibria.
indicated that the hot water components of the thermal waters Derivation of Na-K-Mg-Ca Geoindicators. Geochimica et
range from 27% to 90%. All the thermomineral waters in Cosmochimica Acta, Vol.52, pp.2749-2765.
Salihli geothermal fields fall into the immature field of these
plots indicating none of these waters have attained equilibrium Giggenbach, W. F., Gonfiantini, R., Jangi, B.L., and Truesdell,
with their associated host rocks. Since thermomineral waters of A.H. (1983). Isotopic and Chemical Composition of Parbati
the fields are not equilibrated in water-rock interactions, Valley Geothermal Discharges, NW Himalaya, Indiana.
chemical geothermometers were not successful in estimating Geothermics, Vol. 5, pp.51-62.
reservoir temperature.
Karamanderesi, I.H. (1997). Salihli-Caferbeyli (Manisa Ili)
ACKNOWLEDGEMENTS jeotermal sahasi potansiyeli ve gelecegi. DEK TMK, Turk 7.
En. Kong. Tekn. Ot. bil. met., pp.247-261, (In Turkish).
The authors thank Mike Whitworth and Halim Mutlu for their
helpful suggestions and reviews of the manuscript. Kharaka, Y.K., Gunter, W. D., Aggarwal, P. K., Perkins, E. H.,
and De Braal, J. D. (1988). Solmineq.88: A computer Program
REFERENCES Code for Geochemical Modelling of Water-Rock Interactions.
In U.S.Geological Survey Water Investigations Report 88-05,
APHA-AWWA-WPCF (1980). Standard Methods for
Examination of Water and Waste Water. Fifteenth Edition, Tarcan, G., and Filiz, S. (1997). Hydrogeology of the Turgutlu
Copyright by Am. Pub. Health As. Washington D.C. 1193 pp. Geothermal Field. Turkish Journal of Earth Sciences, vol.6,
no.2, pp.43-64.
Arnnorsson, S., Gunnlaugsson, E., and Svavarsson, H. (1983).
The chemistry of geothermal waters in Iceland. III. Chemical Yilmazer, S., and Karamanderesi, İ.H., 1994, Kursunlu
geothermometry in geothermal investigations. Geoch. et Cosm. jeotermal alaninin (Salihli-Manisa) jeolojisi ve jeotermal
Acta, vol. 47, pp. 567-577. Pergamon Press. USA. potansiyeli. Dun. En. Kon..Turkiye 6. En. Kong., TMK, 17-22
Ekim 1994, Izmir, Tek. Ot. Tebl. Vol.I, pp.68-181, (In Turkish).

1832
Tarcan et al.
N. N.
MANISA Gediz River Marmara KULA
Ahmetli Lake
TURGUTLU
Caferbeyli Deep SALIHLI
Well Kursunlu
Sart-Camur
Hot Spring Celikli Hot Spring
Spring
ALASEHIR Aktepe
429

KIRAZ EXPLANATIONS KIZILHAVLU GEOLOGICAL MAP OF THE

Study Area
ANKARA
ADALA SALIHLI AREA
Road DOMBAYLI
MANISA
Hot Spring
EXPLANATION
0 7,5 15km South of North of
Settlement Area
Gediz Graben Gediz Graben

Quater

Hol
oce
ne
nary
Gediz River
Pleis
Old Alluvium basalts
M evlutlu

Young Alluvium

N
O

N
C

C
E

o
g

n
e

e
I
Kula
Acidere Unit.
S AL MEVLUTLU tocene Asartepe Unit Unit.
IHL Alase Y’ Pliocene Gobekli Unit. Filiztepe
I-ALA hir Stream
mCaferbeyli K
Deep Well
r
SALIHLI
SE H IR Unit.
Late
a

s
tr e
u

S
n
l
PLA NE Miocene
a k u
b S
a t
T

r
e
a
YENIPAZAR Menderes Granodiorite
Sart-Camur m Massif
Thermal Spring Rocks Metamorphic rocks
Kursunlu GOBEKLI
CO2. Discharge Thermal Spring
Contact Geothermal Field
0 2km DEREKOY
Detachment Fault Natural CO2
Normal Fault discharge
Y. Deep Geothermal
Y’ Cross section line Well
SW
1250m
750 Y. NE
250 Sazdere
Thermal
Y. Springs
SSW Y’

Figure 1. Location, geological map and cross section of the study area and locations of geothermal fields (modified from Emre, 1996)

1-MTA2 deep well.

Salihl

2-MTA3 deep well 7-Kursunlu hot spring

N 3-Caferbeyli well
i

Ilica 8-Tabakstream
stream 4-Sart camur hot spring 9-Karaoluk cold water
0 5-Kursunlu celikli hot 11 9
10-Kursunlu stream
40 spring
0 0 0 6-Sazdere hot spring 12-Salihli-stream water
3
5 3

10 13-Gobekli village well

0
2
5
13
MTAo-3. 8 12
(95 C)
4
SA-o10 MTAo-2 1
(62 C) (95 C) 6 2
Kursunlu 5
MTAo-1 7
(88.9 C) 3
SA-o 9
(35 C) SA-12
Celikli Kursunlu Hot Springs
SA-5 250
o
(89 C) Anticipated Groundwater Motion
SA-o6
SA-o8 SA- o7
(43 C) (40 C)
(85 C) Figure 3. Piper trilineer diagrams of the waters from the study area
Kukurtlu Mineral SA-4o 300
water (63,5 C) cNa/1000+cK/100+c(Mg)1/2=S Na/1000
Allahdiyen Village %Na = cNa/10S 90
SA-3 %Mg = 100c(Mg)1/2/S
SA-o2 EXPLANATIONS c = mg/l 80
(55 C)
MTA 1
Deep Well 70 Fully Equilibrated
I
SA-o1 Waters
0 0
60 160 140
(20 C) SA 1 Spring %-Na 200
180
0
0
120
0

0
Boundary of 50 220
0 100

SA-11 the Shallow 240

0
0
Geothermal 40 260 Partially Equilibrated Waters
Field II
350
Road 30

0 200m
20
400 3. 1.
7.
10
Immature Waters 5. 2. 10
III
K/100 20 30 50 60 70 80 6. 4. 9.
Figure 2. Spring and well locations in the Kursunlu
90 (M g)

10 40
% Mg
8. 1/2
geothermal field (modified from Yilmazer and
Karamanderesi, 1994). Figure 4. Distribution of the thermal waters from the study
area in Na-K-Mg tri-linear diagram(Giggenbach, 1988).

1833
Tarcan et al.

o
Table 1. Chemical Analyses of Waters from the Salihli Geothermal Fields (Concentrations are in ppm, T ( C)=measured temperature )
Sampling No and Name Date T (OC) EC S/cm PH Na K Ca Mg Fe Li Mn Ni Cu Pb Cl HCO3 SO4 SiO2 B
1-MTA2 Tw 25.4.1992 85 3000 6.3 462 55 130 13 0.100 Nd nd 0.147 0.008 0.076 69 1378 125 267 65
2-MTA3 Tw 24.6.1992 95 2750 6.03 431 50 31 13 90 1220 100 135 31
3-Caferbeyli Tw 20.9.1990 155 2700 7.8 680 70 42 6 0.100 Nd nd 0.147 0.008 0.076 115 1983 34 214 67
4-Sart-Camur Ts An. Av. 51 1431 6.03 199 24 134 23 0.133 1.8 0.04 0.082 0.006 0.057 37 1076 81 101 13
5-Celikli Ts An. Av 42 2310 5.98 500 62 45 15 0.063 1.788 0.005 0.120 0.008 0.087 68 1513 119 182 23
6-Sazdere Ts 21.5.1993 37 3070 7.55 417 50 29 65 200 2477 43 145
7-Kursunlu Ts An. Av. 90 1850 4.85 426 51 10 9 0.110 1.842 0.103 0.005 0.072 64 1080 107 186 38
8-Tabak Stream Sw An. Av. 20 586 7.10 45 6 34 12 16 305 73 29 2.3
9-Karaoluk Cs An. Av. 14 210 7.25 5 1 13 5 0.054 0.014 nd 0.022 0.005 0.033 14 66 71 11 nd
10-Kursunlu Stream Sw An. Av. 19 410 6.76 29 4 26 16 0.072 0.082 nd 0.033 0.004 0.038 16 204 92 19 3
11-Kursunlu Ufuruk Cs 31.5.1992 15 1420 5.75 12 7 180 77 21 647 541 2
12-Salihli Stream Sw 21.5.1993 18 1299 7.19 140 15 17 92 41 535 174 19
13-Gobekli Village Cw 21.5.1993 18 556 4.9 30 4 14 15 12 293 120 26
Tw=Thermal well, Ts=Thermal spring, Sw= Stream water, Cs= Cold spring, Cw= Cold well, An. Av. = Annual averages (from 1991 Dec. to 1992 July), Nd not determined.

(mg/l
Steam point (639 cal/g)

2)
600 Maximum
steam loss

Si
O
600

500 500 Quartz solubility

400

400
300 1.
3.
200 5. 7.
C4 (344 C) O
100
6. 4. 2.
O (187O.C) (227O.C)
C1 (288 C) 0
300 C7 (274OC) 0 50 100 150 200 250 300 350
C5 (267OC) Reservoir Enthalpy (cal/g)

200 B1 C2 (194OC) Figure 6. Enthalpy-Silica diagram of waters in the study area

B7 Table 2. Correlations of the chemical parametres of the waters
3 in the study area (Concentrations are in meq/l)
B4 B5 B2
Na K Ca Mg Cl HCO3 SO4 SiO2 pH B EC
100 Na 1,0 0,9 -0,05 -0,24 0,72 0,86 -0,37 0,92 -0,12 0,86 0,89
7 1 2
K 1,0 -0,01 -0,22 0,72 0,87 -0,32 0,92 -0,18 0,83 0,91
A Ca 1,0 0,26 -0,12 0,10 0,65 0,08 -0,40 0,22 0,20
4 5
Mg 1,0 0,18 0,08 0,57 -0,34 0,44 -0,59 0,07
Cl 1,0 0,93 0,32 0,62 0,18 0,88 0,81
0 HCO3 1,0 -0,24 0,76 0,03 0,81 0,92
0 10 20 30 40 50 60 70 80 90 100 110 120 130 Chloride (mg/l)
SO4 1,0 -0,36 -0,17 -0,06 -0,11
SiO2 1,0 -0,40 0,96 0,85
Figure 5. Enthalpy-Chloride diagram of waters in the PH 1,0 -0,34 -0,18
study area. B 1,0 0,85
EC 1,0
Table 3. Mineral saturation indices of waters calculated in respect to the discharge temperature conditions
Calcite Aragonite Dolomite Siderite Magnesite Quartz Chalcedony Gypsum Anhydrite
MTA 2. Well (85OC) -0.09 -0.21 1.48 -0.28 -0.19 0.96 0.72 -2.04 -1.84
MTA 3. Well (95OC) -2.28 -0.38 1.20 - -0.37 0.56 0.33 -2.08 -1.77
Caferbey well 1.99 1.92 5.83 0.42 1.19 1.20 0.05 -2.39 -1.41
Sart-Camur Ther. Spring (37OC) -0.12 -0.24 0.62 -1.01 -0.79 0.92 0.67 -1.59 -1.67
Kursunlu Thermal Spring (90OC) -2.03 -2.13 -2.02 -1.70 -1.80 0.75 0.52 -2.51 -2.25
O
Celikli Thermal Spring (42 C) -0.65 -0.77 -0.25 -1.43 -1.11 1.29 1.04 -1.98 -2.13
Sazdere Thermal Spring 0.80 0.67 3.44 - 1.13 1.25 1.01 -2.74 -2.93
Tabak Stream -0.36 -0.51 -0.75 - -1.10 0.74 0.30 -1.98 -2.22
Karaoluk spring -1.35 -1.45 -2.59 - -2.13 0.40 -0.06 -2.29 -2.54
Kursunlu Stream -1.03 -1.18 -1.82 - -1.54 0.58 0.13 -1.98 -2.22
Kurs.Ufuruk Spring -0.88 -1.03 -1.57 - -1.57 -0.34 -0.81 -0.77 -1.01
Salihli Stream 1.64 1.50 4.37 - 2.15 0.40 -0.03 -2.12 -2.34
Gobekli Well -2.91 -3.06 -5.45 - -3.10 0.61 0.18 -2.15 -2.37

Table 4. Chemical geothermometry results applied on the hot springs in the study area
GEOTHERMOMETERS K.T.S. K.C.T.S S. C T.S. K MTA 2. Well KMTA 3 well H.S.T.S.
(90 oC)* (42 oC)* ( 51 oC)* (85 oC)* (95 oC )* ( 37 oC)*
a- SiO2 (mg/l) (Amorphous Silica) 52 50 17 76 33 37
a- SiO2 (mg/l) (Alpha cristobalite) 125 124 87 152 104 109
a- SiO2 (mg/l) (Chalcedony) 153 151 111 183 130 135
a- SiO2 (mg/l) (Quartz) 175 173 137 201 154 159
a- SiO2 (mg/l) (Quartz max. steam loss) 164 163 133 185 147 151
b- SiO2 (mg/l) (Chalcedony cond. cooling) 148 146 109 175 127 131
b- SiO2 (mg/l) (Quartz steam loss) 149 147 107 178 126 131
b- SiO2 (mg/l) (Quartz steam loss) 164 162 132 185 146 150
b- Na/K (mg/l) 214 218 215 214 211 165
b- Na/K (mg/l) 230 233 231 230 227 192
c- Na/K (mg/l) 206 210 207 205 202 150
d -Li (mol/l) 177 176 176 - - -
d -Na/Li (mol/l) 176 160 251 - - -
e- K/Mg (mg/l) 109 107 77 106 103 84
c- Na-K-Ca (mol/l) 216 =1/3 181 =4/3 94 =4/3 139 =4/3 180 =4/3 182 =4/3
f- Na-K-Ca – Mg corrected (meq/l) 81 88 not applicable 116 79 7
(KTS=Kursunlu thermal spring, KCTS= Kursunlu-Celikli thermal spring, SCTS= Sart Camur thermal spring, KMTA2=Kursunlu MTA2 well, KMTA3=Kursunlu MTA3 well, HSTS=Horzum Sazdere thermal
spring, a= Fournier 1977b; b= Arnorsson et al. 1983, c= Fournier and Truesdell 1973, d= Fouillac and Michard 1981, e= Giggenbach et al. 1983, f= Fournier and Potter 1979)

1834