Journal of Geochemical Exploration 152 (2015) 5466

Contents lists available at ScienceDirect

Journal of Geochemical Exploration

journal homepage: www.elsevier.com/locate/jgeoexp

The geothermal resources of the Republic of Djibouti I:

Hydrogeochemistry of the Obock coastal hot springs
Mohamed Osman Awaleh a,, Farhan Bouraleh Hoch a, Ibrahim Houssein Kadieh b, Youssouf Djbril Soubaneh c,
Nima Moussa Egueh a, Mohamed Jalludin a, Tiziano Boschetti d
a

Centre d'Etudes et de Recherches de Djibouti (CERD), Route de l'aroport, B.P. 486, Djibouti ville, Djibouti
Laboratoire Rgional, Newalta Chteauguay, 125 Rue Blanger, Chteauguay, J6J 4Z2 Qubec, Canada
Dpartement de biologie, chimie et gographie, Universit du Qubec Rimouski, 300, Alle des Ursulines, Rimouski, QC G5L 3A1, Canada
d
Department of Physics and Earth Sciences Macedonio Melloni, University of Parma, Parco Area delle Scienze 157/a, 43124 Parma, Italy
b
c

a r t i c l e

i n f o

Article history:
Received 8 May 2014
Accepted 1 February 2015
Available online 7 February 2015
Keywords:
Hot springs
Geothermometers
Chemical composition
Isotopic composition
Obock
Djibouti

a b s t r a c t
This paper examines the hydrochemical features of the waters from the Obock coastal geothermal eld. As the
submarine waters at the ridge, their chemical and isotope composition shows afnities with sea waterbasalt interactions at hydrothermal temperature. Moreover, good linear correlations were obtained between hot springs
and of sea water sample points when plotting normalized concentrations of elements to chloride (Mg/Cl, SO4/Cl,
K/Cl, Ca/Cl, SiO2/Cl) versus concentration of Li/Cl. This would indicate that two end members exist (seawater and
reservoir uid). After extrapolation of the Mg and SO4 concentrations to zero, the obtained value of SiO2 corresponding to the reservoir uid end member was used to estimate the reservoir temperature by quartz
geothermometers. The obtained temperature of 187 C is in good agreement with that obtained from multiple
mineral equilibrium approach (180200 C), cationic geothermometers (172191 C) and by the evaluation of
isotopic equilibrium between water and sulfate molecule (207 C). Summarizing all the employed approaches,
a mean temperature of 197 10 C has been estimated.
The isotopic 34S(SO4) signature of the dissolved sulfates in Obock thermal waters conrms that these waters
result from the mixture of a hot seawater-derived uid (absence of sulfates) with cold seawater. However,
water isotope data did not exclude the presence of a small contribution from fresh groundwater.
2015 Elsevier B.V. All rights reserved.

1. Introduction
The Republic of Djibouti is one of several African countries located
on the East African Rift System where geology is also resulting from
two other ridges of Red Sea and Gulf of Aden. As in other rifting zones,
the activity of the East African Rift System corresponds to large seismic,
tectonic and volcanic activities (Barberi et al., 1975; Mlynarski and
Zlotnicki, 2001).
In Djibouti, most of the widespread geothermal activity, manifested
in the form of numerous hot springs, fumaroles and hydrothermal alteration, is located mainly in the western part of the country and along the
Gulf of Tadjourah ridge (Fig. 1A).
The most geologically active area in Djibouti is the Lake Asal area,
and the Asal rift is one of two emergent oceanic ridges in the world,
the other being Iceland (Mlynarski and Zlotnicki, 2001). Accordingly,
numerous geological and geophysical studies were completed in Lake
Asal in order to understand the phenomena related to sea oor

Corresponding author. Tel.: +39 253 77 84 68 55; fax: +39 253 21 35 45 68.
E-mail address: awaleh@gmail.com (M.O. Awaleh).

http://dx.doi.org/10.1016/j.gexplo.2015.02.001
0375-6742/ 2015 Elsevier B.V. All rights reserved.

spreading (Mlynarski and Zlotnicki, 2001; Pinzuti et al., 2010). Moreover, geothermal studies undertaken in the late 1960s and early 1970s
allowed selecting the Asal prospect as the most favorable area for carrying out deep drilling exploration (Demange et al., 1971; Lavigne and
Lopoukine, 1970). Therefore, six geothermal wells (Asal 1 to 6) with
various depths, between 1137 m and 2105 m, were drilled in the
Asal prospect during the last decades (Aquater, 1989; BRGM, 1975).
However, the high enthalpy Asal geothermal uids (about 350 C)
have had high salinity (116 g/kg) (D'Amore et al., 1998).
On the other hand, since the main national electricity production
sources depend on petroleum products, Djibouti is relatively dependent
on diesel fuel and fuel oil imports to meet its energy needs. As a consequence, the electricity cost is exorbitant in the Republic of Djibouti,
where in average 1 kWh costs about 0.23 U.S. $ (GBAD, 2011). To mitigate this energy burden, which put a brake on the rate of economic
growth in this country, a national program for development of geothermal resources (NPDGR) was implemented in 2010. This program aimed
to conduct multidisciplinary studies (geochemical, hydrogeological,
geophysical, geological and reservoir engineering) on all areas with geothermal activities in the Republic of Djibouti (CERD, 2011, 2012; Awaleh
et al., submitted for publication).

M.O. Awaleh et al. / Journal of Geochemical Exploration 152 (2015) 5466

55

Fig. 1. A: Simplied geological map and hydrothermal activity in the Republic of Djibouti. B: Digital elevation model (DEM) of the Afar Rift system with the location of Obock (Doubre,
2004) (vertical color scale = altitude in meter; the black area = Obock region). C: Tectonic map of the Gulf of Tadjourah. It is noteworthy the extension of the fault lines, from the
coast to open ocean (after Manighetti, 1993). D: Magnetic anomaly map of the Gulf of Tadjourah (after Courtillot et al., 1980). In the key, numbers are milliTeslas (mT).

On the basis of new pre-feasibility studies related to the NPDGR, the

World Bank granted, in 2012, a loan about 31 M U.S. $ to the Republic of
Djibouti to perform four new geothermal wells in the area of Lake Asal
(World Bank, 2013). As part of this NPDGR, the hot springs of the
Obock beach were also studied in 2013 (CERD, 2013). Previously, 3
samples of Obock hot springs in the intertidal zone were investigated
to develop a spa center (Aquater, 1982). Nowadays, the thermal waters
of Obock beach are used mainly for hydrotherapy. In the early 1990s,
Houssein et al. (1993) studied the geochemistry of four Obock beach
thermal springs.

The main purpose of the present study is to characterize the hydrothermal activity from known and newly investigated hydrothermal
springs and thus to provide a framework for future studies of the
Obock geothermal system, one of the northern active provinces of the
Republic of Djibouti. Towards this aim, detailed geochemical investigations have been carried out on the cold groundwaters (well waters and
boreholes) and the most representative thermal waters (7 hot springs)
from Obock area in order to understand their geochemical evolution
and also estimate the reservoir temperature through chemical and
isotopic geothermometry as well as mineral equilibrium approach.

56

M.O. Awaleh et al. / Journal of Geochemical Exploration 152 (2015) 5466

Furthermore, the geochemical and isotopic study of the thermal waters

of Obock beach, combined with geological and tectonic information as
well as regional hydrogeology, have been used to propose a conceptual
model for the Obock beach geothermal system.
2. Geological and hydrogeological settings
2.1. Geology of the area
The Republic of Djibouti is located in the Southeastern part of
the Afar Rift which is at the triple junction between two nascent
mid-oceanic ridges (Red Sea and Gulf of Aden) and a continental rift
(East African Rift) (Fig. 1B, Doubre, 2004). The main expression of
crustal rifting process in Afar is, in addition to extensional faulting, a
bimodal magmatism which began as early as 30 Ma by the emplacement
of trap volcanism in Yemen and Ethiopia (Barberi and Varet, 1977).
The Obock area located in the northern part of the country consists
of a Quaternary formation of coralline limestone (Faure et al., 1980;
Gasse et al., 1983; Manighetti, 1993).
The Debergade paleo-valley is characterized by conglomerates with
large block which have a thickness up to 10 m probably coming from
the neighboring reliefs (Mablas rhyolites and Dahla basalts). The Mablas
rhyolitics correspond to a wide range of domes and lava ows, erupted
between 18 and 11 m.y (Manighetti, 1993) and observed in two sectors
on both sides of Tadjourah Gulf (Manighetti et al., 2004). This series unconformably overlies the Dahla basaltic series (8.63.8 m.y) which form
the (i) outer rim of the Ali Sabieh antiform and (ii) the western ank of
the Danakil depression (Barberi et al., 1975). The Dahla basaltic series
form the outer belt of the Danakil depression, where their thickness
reaches 1000 m in the Randa area (south of the Danakil depression).
Along the coast, alluvial deposits are associated to coral reef materials
and shape small parallel hill which are reminiscent an ancient littoral
(Gasse et al., 1983). The North West of the study area is characterized
by red silt plated to coral reef where the clay fraction mainly consists of

illite, smectite and traces of kaolinite and quartz, feldspar, goethite and
chlorite minerals (Gasse et al., 1983). Rhyolite pebbles from the Mablas
series form a reg on the surface of the coral reef, in response to climatic
change. The Gulf basalts constitute the basis of these formations (Faure
et al., 1980).
From a tectonic point of view, the coral reefs are raised by vertical
movements associated with the opening of Tadjourah Gulf (Faure,
1976). The Gulf basalts, detrital materials, and coral reefs are affected by
normal faults trending N 20 to N 50 (Faure et al., 1980; Gasse et al.,
1983; Richard, 1979). These normal faults take gradually the EW direction of the Tadjourah ridge axis (Gasse et al., 1983). Two major tectonics
structures with N 110 orientation connecting the two bank of the
Debergade paleo-valley with the Obock pit (1600 m of depth) characterize the Obock area (Fig. 1C). These structures have also been observed in
the magnetic map produced by Courtillot et al. (1980) (Fig. 1D).
The Obock area is known to be the onshore prolongation of the
Tadjourah spreading ridge (Manighetti, 1993). These prolongations
allow explanation of the large elevation of the coral reefs (Gasse et al.,
1983; Manighetti, 1993). The Gulf basalts have been dened as the
chronological markers of the dislocation of Tadjourah ridge (Faure,
1976; Gasse et al., 1983; Manighetti, 1993). Geophysical data gathered
during the IV Orgon cruise (1981) on the axial valley of the Tadjourah
ridge reveals (i) the presence of a huge volcano and (ii) two types of
sediment clearly differentiated by their reector. The Obock pit was
drilled at 1070 m depth and the petrography based on cores consists
of clay, carbonate enriched on pelagic organisms, dated from Late
Pleistocene to Holocene (Moyes et al., 1981).
2.2. Hydrogeology of the area
The piezometric map of Obock region is based upon few points of
measurements (Fig. 2), it therefore remains indicative and the values
of groundwater hydraulic heads and hydraulic gradients could be
modied according to new measurements. About ten points irregularly

Fig. 2. Piezometric map of the Obock region.

42.14
24.24
15.53
61.39
n.a.
3.63
4.7
2.7
12
0.84
2.6
67
0.77
0.36
0.65
0.76
0.7
1.70
2
1
1
2
-2
0
18
17
41
42
41
8.2
7.35
7.49
7.72
7.94
7.01
8.06

4251
2851
10,530
1322
2590
52,190

3023
1939
7407
922
1933
41,100

NaClCaMg
NaClCaMg
NaClMg
NaClHCO3
NaClHCO3
NaCl

206
140
405
32.1
114
401

155
81.8
310
9.17
73.37
1389

549
320
1517
279
378
12,200

27.9
16
73.6
6.7
8.8
350

39.99
81.93
82.05
359.5
406.8
148.1

1607
855
3740
209
665
21,400

156
87.7
387.4
73.82
157
3031

3. Material and methods

n.a. = not analyzed; IB: ionic balance.

33.6
32.4
27.8
31.4
28.9
27.6
115817.8
115824.3
115650.7
115747.7
120317.5
113136.5
431728.2
431721.3
431016.2
431339.3
430833.7
423126.0
Well waters
Obock 1
Obock 2
Khoma Adou
Alla-ella
Tero-ela
Seawater

57

spread over an area of 1000 km2 allows drawing up the general feature
of the underground ows. The groundwater ow, oriented from the
West to East and the South-East across the Obock plain, seems to be
mainly controlled by the mountains of Mabla rhyolites in the West
and the mountains in the North where the main recharge occurs from
the wadi Sadai and its tributaries.
The piezometric levels evolve rapidly from Obock coastal zone to the
upstream area in the mountains. Close to the Obock town large diameter wells taping inferoux underground ow indicates a piezometric
level of 1 m above sea level. Within the wide sedimentary plain the
wells of Oulma and Soublali give respectively 5 m and 20 m. In the
mountainous zone, the piezometric levels reach 142 and 198 m on the
wells Assassan and Illisola respectively. More Westward the site of
Waddi shows a piezometric level of 324 m. Consequently, the hydraulic
gradients are generally high varying from 4.3 to 13.

0.087
0.0005
0.0008
0.0002
0.21
0.215

93.89
96.46
118.5
106.37
2.02
2.52
2.00
2.13
1.18
0.84
1.28
1.04
0.007
0.08
0.016
0.0025
1
3
1
1
30
30
33
30
151
134
137
142
448
569
449
463
294.7
309.2
353.6
350.7
3.47
4.84
3.86
4.21
273
369
288
300
76.3
78.7
77.3
73.3
69.8
62.2
70.5
72.7
NaClHCO3Mg
NaClHCO3Mg
NaClHCO3Mg
NaClHCO3Mg
1374
1234
1346
1260
1948
2391
2030
2063
7.58
7.63
7.34
7.42
120700.6
120307.1
120647.1
120645.0
Boreholes waters
Illysola
430554.7
Soublali
431219,.4
Assassane1
430925.1
Assassane2
430935.9

29.5
32.1
34.7
34.8

82.59
91.98
101.65
118.6
101.65
107.55
85.32
56
54
53
48
52
54
52
1.03
1.16
1.19
0.96
1.60
1.14
1.19
1.30
1.91
1.86
2.05
1.85
2.09
2.03
2
3
1
1
3
2
2
1.1
1.6
2.3
0.6
2.9
0.52
0.88
115733.9
115733.5
115733.4
115733.8
115733.3
115733.6
115733.5
431721.7
431718.7
431717.2
431717.3
431718.1
431716.7
431716.6
Thermal waters
Obock-HS1
Obock-HS2
Obock-HS3
Obock-HS4
Obock-HS5
Obock-HS6
Obock-HS7

58.6
71.4
63.0
60.0
63.6
63.4
63.5

6.81
7.01
6.95
6.93
6.36
6.74
6.96

47,450
45,890
45,130
41,240
43,400
45,170
45,330

35,600
33,700
33,400
30,000
32,070
33,100
33,650

NaCl
NaCl
NaCl
NaCl
NaCl
NaCl
NaCl

753
1001
881
863
881
892
809

943
727
848
706
797
800
769

10,900
9565
9864
8955
9604
9903
10,100

440
550
545
544
515
503
548

118.6
131.4
142.6
151.2
140.4
133.9
147.8

18,800
16,700
17,800
16,400
16,700
17,600
17,600

1992
1503
1699
1484
1562
1622
1512

IB
(%)
pH
T
C
Longitude
Latitude
Samples

Table 1
Chemical analyses of thermal and cold waters from Obock area.

EC
(S/cm)

TDS
(mg/L)

Hydrochemical types

Ca
(mg/L)

Mg
(mg/L)

Na
(mg/L)

K
(mg/L)

HCO3
(mg/L)

Cl
(mg/L)

SO4
(mg/L)

NO3
(mg/L)

Li
(mg/L)

F
(mg/L)

Br
(mg/L)

SiO2
(mg/L)

M.O. Awaleh et al. / Journal of Geochemical Exploration 152 (2015) 5466

A total of seven geothermal springs, the seawater and nine wells and
boreholes cold waters in the Obock area were sampled in January 2013.
Coastal thermal waters were collected at very low tide, in which period
these waters upraised without any mixing with surface seawater.
Unstable hydrochemical parameters, including temperature (0.1 C),
pH (0.01 unit) and electrical conductivity (1 S/cm), were measured
on site with, respectively, hand-held meters Hanna CheckTemp, Eutech
Instruments pH 610, and Eutech Instruments COND 610 that were
calibrated in the eld prior to sampling.
Water samples were collected in polyethylene containers after
ltration through 0.45 m membrane lters. All samples used for determination of cations were acidied after collection through addition of
Suprapure HNO3 (Merck) to bring the pH below 2.
Analyses of anions and major cations were carried out by ionic chromatography with a Dionex ICS 3000 Ion Chromatograph using analytical procedures and quality assurance for geothermal water chemistry
reported by Pang and Armannsson (2006). AS4A SC-4 mm analytical
column (250 mm 4 mm ID) coupled with AG4A SC 4 mm guard and
CS12A analytical column (250 mm 4 mm ID) coupled with CG12A
guard were used respectively for anions and cations analyses. The ion
chromatograph was calibrated through repeated analysis of ve working anion and cation standards (with concentrations within the range
of analyses). Peaks were identied using Chromeleon software (Dionex,
Sunnyvale, California). The analytical precision was estimated at 5%.
For the analysis of aqueous SiO2, the water samples were diluted
tenfold using deionized water to prevent the SiO2 precipitation. SiO2
contents were determined by colorimetry and analyzed using a Jenway
6300 spectrophotometer, while HCO3 was analyzed by titration with
0.1 M HCl. The ionic balance was below 3%.
Additional samples of untreated waters were collected in 50 mL
glass bottles (Quorpak) for stable isotopes analyses of the water
molecule, 2H(H2O) and 18O(H2O), and 1000 mL plastic bottles for
tritium (3H) analysis, all analyzed at the Bureau de Recherche
Gologique et Minralogique (BRGM) in Orleans, France. The deuterium
and oxygen isotopic ratios were analyzed using a Finnigan MAT 252
mass spectrometer and reported in per mil notation () versus the
Vienna Standard Mean Ocean Water (V-SMOW) standard following
= [(Rsample / Rstandard) 1] 1000, where R is the 2H/1H and
18
O/16O absolute isotope amount ratios. The average precision, based
on multiple analyses of various samples and laboratory standards, was
0.1 for 18O(H2O) and 0.8 for 2H(H2O). Tritium activity was
measured by direct liquid scintillation counting. The detection limit
was 0.6 TU (Tritium Unit with 1 TU equal to 1 tritium atom in 1018
hydrogen atoms).
The samples for sulfur and oxygen isotope analysis of dissolved
sulfate were collected at the outlet of columns using 250 mL pre-acid
washed plastic perplex bottles. Cd-acetate was already added in the bottles (5% v/v) prior to sample collection, to x sulfur as CdS, and then the

58

M.O. Awaleh et al. / Journal of Geochemical Exploration 152 (2015) 5466

aliquot was ltered through a 0.2 m nitrocellulose lter before chemical determination of residual sulfate. Dissolved sulfate was precipitated
2
species) by
as BaSO4 at pH b 4 (in order to remove HCO
3 and CO3
adding a BaCl2 solution. The isotopic analyses on BaSO4 were carried
out using a Delta + XP mass spectrometer coupled in continuous-ow
mode to a Thermo Elemental Analyzer in BRGM laboratories. The
34S(SO4) and 18O(SO4) isotope compositions were reported in the
usual -scale in with reference to V-CDT (Canyon Diablo Troilite)
and V-SMOW (Vienna Standard Mean Ocean Water), respectively. The
analytical precision was of 0.3 versus V-CDT for 34S(SO4) and
0.5 versus V-SMOW for 18O(SO4).
4. Results
4.1. Hydrochemistry
Hydrothermal activity in the Obock beach is characterized by lowow hot springs (1.8 L s1) (Aquater, 1982). The temperature, pH, electrical conductivity (EC), total dissolved solids (TDS), sampling locations,
hydrochemical types and major and minor elements of the sampled
thermal and cold waters are listed in Table 1. Thermal waters from
Obock beach are moderately acid to neutral (pH = 6.367.01) with a
temperature range of 58.6 to 71.4 C (Table 1). The chemical compositions of thermal waters are similar to that of seawater, with EC values
in the range of 41 to 47 mS/cm, while in the cold well and borehole
waters from Obock area ranged from 1.3 to 4.2 mS/cm (Table 1). The
relatively high EC (10 mS/cm) of Khoma Adou well, which is located
inland 200 m from shoreline, likely indicates seawater intrusion.
Chemical compositions of the waters are plotted in the diagram of
Piper (1944), as shown in Fig. 3. Classication of the water samples in
Table 1 was made according to principles of IAH (1979). Total equivalents of cations and anions separately were accepted as 100% and ions
with more than 20% (meq/L) were taken into consideration in this
classication.

The chemical composition of the waters studied described in

terms of relative concentrations of main anion and cation allows us to
distinguish the following two groups of waters (Table 1, Fig. 3):
(1) Thermal waters from the Obock coastal hot springs are of the
NaCl type;
(2) Cold well and borehole waters from Obock area are mostly of the
ClHCO3NaCaMg type.

According to Giggenbach (1991), the chemical composition of the

thermal water samples from Obock beach can be classied as
chloride-rich mature waters (Fig. 4A).
The cold well and borehole waters also show dominant chloride,
except the Alla Ella well water that plots at the chloridebicarbonate
border (Fig. 4A). When comparing results of this study with previous
data (Aquater, 1982; Houssein et al., 1993), no signicant variations in
uid geochemistry were observed between 1980 and 2013 (Fig. 4B).
Such consistent chemical character suggests uniform reservoir for the
Obock beach geothermal system (OBGS).

4.2. Geothermometry
The Na/1000K/100Mg0.5 ternary plot of Giggenbach (1988) can
be used to discriminate mature waters, which have attained equilibrium
with relevant hydrothermal minerals, from immature waters and
waters affected by mixing and/or re-equilibration along their circulation
path (Fig. 5). This provides an indication of the suitability of the
waters for the application of solute geothermometers. Since all
hot springs from OBGS were partially equilibrated (Fig. 5), chemical
geothermometers can be used to estimate geothermal reservoir temperature (Arnrsson et al., 1983; Fournier and Potter, 1979; Fournier
and Truesdell, 1973; Giggenbach, 1988). Therefore, classical and

Fig. 3. A: Piper diagram of seawater and hot springs waters from Obock beach; B: cold borehole and well waters from Obock area.

M.O. Awaleh et al. / Journal of Geochemical Exploration 152 (2015) 5466

59

deep parent uids of about 169195 C (Table 2). Furthermore, the

NaK geothermometer of Fournier (1979) and Arnrsson et al. (1983)
estimated the temperature of the geothermal reservoir respectively at
about 156185 C and 153179 C (Table 2). Those temperatures are
in agreement with the values obtained with NaKCa geothermometer,
173192 C (Table 2). It is of interest to note that Houssein et al. (1993)
obtained almost the same range of estimate temperature for OBGS
(Table 2).
The NaLi geothermometer was initially developed by Fouillac and
Michard (1981) from a statistic study about groundwaters in granitic
and volcanic domains. Later on, NaLi equation was recalibrated by
Kharaka et al. (1982) and Verma and Santoyo (1997) obtaining different
equations. Moreover, the Na/Li geothermometer appears to be sensitive
to the total dissolved solids of the water at depth, locally controlled by
contributions of NaCl, and the rock type (Sonney and Vuataz, 2010).
Based on these results, these Na/Li geothermometers cannot be considered reliable for the Obock beach geothermal system.
However, a Na/Li geothermometer adapted for high-temperature
uids derived from seawater interacting with basalt in emerged
rifts was recently proposed (Sanjuan, 2010). Although it might be
affected by mixing, adsorption and precipitation of secondary phases,
the temperature of the OBGS was estimated at 210 C (Sanjuan, 2010).
In our calculation, temperatures obtained by the Na/Li and other
geothermometers are in agreement, except for the Obock-HS1
(Table 2). This exception might be explained considering that the
Obock-HS1 water showed low silica, high SO4 and high Mg concentrations, which indicate a higher contribution of cold seawater.

Fig. 4. ClSO4HCO3 classication diagram of Giggenbach (1991). A: Thermal and cold

waters from the OBGS (this study); B: hot springs waters from Obock beach (1981: Aquater,
1982, 1989: Houssein et al., 1993; 2013: this study).

isotopic geothermometers as well as geothermometrical modeling

approach were used to estimate the reservoir temperature of OBGS.

4.2.1.2. Silica geothermometers. The silica geothermometers, based

on the content aqueous silica and the solubility of different silica
phases as a function of temperature, can be used to determine the
temperature of waterrock interaction processes at depth in a rising
uid before its discharge. However, among the several existing silica
geothermometers, in this study only equations for quartz were used because this is the main silica phase found in the local rift (D'Amore et al.,
1998; Zan et al., 1990). In Table 3, the results obtained from the application of the quartz equations proposed by Fournier (1977) are presented.
Estimated temperatures in the OBGS reservoir range between 125 and
147 C, in agreement with values (111137 C) obtained by using previous data (Houssein et al., 1993). Quartz temperatures were generally
lower than those obtained with cationic geothermometers (Table 2).
Such differences might be due to disequilibrium between quartz and
the hot water, probably caused by mixing between waters of different
temperature and silica content (Fig. 6). In fact, the cold terrestrial waters
of meteoric origin, probably representing an end-member, appear
equilibrated with silica phases (amorphous silica, opal, chalcedony)
more soluble than quartz (Fig. 6). Precipitation of silica phases during
mixing between the hot uid with cold seawater, as typically occurs in
hydrothermal vents (e.g. Bethke, 2008), could also explain the lack of
equilibrium with respect to quartz.

4.2.1. Chemical geothermometers

4.2.1.1. Cationic geothermometers. The NaK geothermometer was initially developed to locate the major upow in high enthalpy geothermal
systems, because a general decrease in Na/K ratios of geothermal uids
with increasing temperatures was observed (D'Amore and Arnrsson,
2000). This geothermometer is related to the variation of sodium and
potassium in thermal waters due to ion exchange of these elements
between coexisting alkali feldspars (Nicholson, 1993). NaK equations
are adapted for reservoir temperatures in the range 180350 C
(Ellis, 1979), but are limited at lower temperatures, notably less than
120 C (Nicholson, 1993).
Several NaK geothermometers have been proposed and applied to
calculate reservoir temperature in the past three decades. The NaK
geothermometer of Giggenbach (1988), based on thermodynamic
data of albite and K-feldspar, gives an equilibrium temperature of

4.2.2. Geothermometrical modeling

The state of equilibrium between water and mineral is a
function of temperature; therefore saturation indices can be used as
geothermometers (Pang and Reed, 1998; Reed and Spycher, 1984). In
other words, estimation of reservoir temperature can be achieved by
simultaneous consideration of the equilibrium state between specic
water and many hydrothermal minerals as a function of temperature.
Geothermometrical modeling calculations have been performed
for the OBGS hot springs using the SOLVEQ computer program (Reed,
1982).
Fig. 7 depicts the mineral saturation indices versus increasing
temperature for Obock coastal hot springs. For those hot springs, saturation indices with respect to quartz, chlorite, brucite, epidote, prehnite,
microcline, forsterite and monticellite minerals tend to get closer to
the zero (SI = 0) around the temperature of 180200 C at which

60

M.O. Awaleh et al. / Journal of Geochemical Exploration 152 (2015) 5466

Fig. 5. Relative Na/1000, K/100 and Mg1/2 contents in mg/L basis of Obock beach hot spring and cold waters from the area (data from this study and Houssein et al., 1993). Dashed and
dotted lines depict the K/Mg and Na/K isotherms at 100, 200 and 300 C, respectively. Arrow represents the isochemical dissolution of 100 g of basalt in 1 kg of chloride water
(Giggenbach, 1988).

Table 2
Cationic geothermometry results for the Obock hot spring water samples (all values displayed in C).
Reference

This study
This study
This study
This study
This study
This study
This study
Mean (except HS 1)
Standard deviation (except HS 1)
Houssein et al. (1993)
Houssein et al. (1993)
Houssein et al. (1993)
Houssein et al. (1993)

Hot spring code

Obock-HS1
Obock-HS2
Obock-HS3
Obock-HS4
Obock-HS5
Obock-HS6
Obock-HS7
187
3
Obock-HS1
Obock-HS2
Obock-HS3
Obock-HS4
Mean
Standard deviation

NaKCa

NaK

NaK

NaK

NaK

NaLi

Fournier and Truesdell

(1973)a

Fournier
(1979)b

Arnorsson et al.
(1983)c

Giggenback
(1988)d

Verma and Santoyo

(1997)e

Sanjuan
(2010)f

173
188
187
191
185
183
188
178
4
190
186
193
187
189
3

156
181
178
185
176
172
177
173
4
185
175
185
178
181
5

153
176
173
179
171
168
172
189
4
179
170
179
172
175
5

169
192
189
195
187
183
188
176
4
195
186
195
188
191
5

155
178
176
182
174
170
175
172
7
182
173
182
175
178
5

127
170
165
183
167
176
171

In the equation a and f, concentration of Ca, Na, K and Li is in mol/Kg, whereas Na and K are expressed as ppm from b to e.
: data not used in the evaluation of the reservoir temperature.
a
T = 1647/[2.24 + Log(Na/K) + Log(Ca0.5/Na)] 273.15; = 4/3 for T b 100 C and b = 1/3 for T N 100 C and Log(Ca0.5/Na) b 0.
b
T = 1217/(1.438 + Log(Na/K)) 273.15.
c
T = 1319/(1.699 + Log(Na/K)) 273.15.
d
T = 1390/(1.75 + Log(Na/K)) 273.15.
e
T = 1390/(1.75 + Log(Na/K)) 273.15.
f
T = 1390/(1.75 + Log(Na/K)) 273.15.

143
115
148
120

M.O. Awaleh et al. / Journal of Geochemical Exploration 152 (2015) 5466

61

Table 3
Silica geothermometry results for the Obock hot spring water samples (all values displayed in C).
Reference

This study
This study
This study
This study
This study
This study
This study
Houssein et
Houssein et
Houssein et
Houssein et
a
b
c
d

Hot spring code

al. (1993)
al. (1993)
al. (1993)
al. (1993)

Obock-HS1
Obock-HS2
Obock-HS3
Obock-HS4
Obock-HS5
Obock-HS6
Obock-HS7
Obock-HS1
Obock-HS2
Obock-HS3
Obock-HS4

Quartz

Quartz

Quartz

Quartz

Fournier (1977)a

Fournier (1977)b

Fournier and Potter (1982)c

Verma (2001)d

127
133
138
147
138
141
129
127
111
133
113

124
129
133
141
133
136
125
132
118
137
120

127
133
138
147
138
141
129
128
111
133
113

124
130
136
146
136
140
126
124
106
130
109

T = {1309/[5.19 Log(S)]} 273.15.

T = {1522/[5.75 Log(S)]} 273.15.
T = 42.2 + 0.28831 (S) 3.6686 104 (S)2 + 3.1665.107 (S)3 + 77.034 Log(S).
T = {1175.7/[Log(S) 4.75]} 273.15; where (S) = silica concentration as SiO2 (ppm).

mean values of the Obock's springs, 2H(H2O) = + 2.6 0.7 and

18O(H2O) = +0.55 0.17, which are slightly depleted in comparison to the typical Gulf of Aden values (Fig. 8A). This is probably due to the
fact that: i) the main recharge of the hot springs didn't occur near their
outpouring sites but in a more open (and deep) ocean environment;
ii) a possible mixing with freshwater of meteoric origin. According to
this latter hypothesis, the analyzed sample with lower salinity falls on
the local meteoric water line (LMWL in Fig. 8A; Fontes et al., 1980), probably due to the ow of local groundwater towards the coast (Awaleh
et al., submitted for publication). However, the 34S(SO4) and 18O(SO4)
values of the springs conrms a signature quite similar to the local
(Fig. 8B) and mean ocean water, i.e. 34S(SO4) = + 21.1 (Coplen

temperature these minerals are assumed to be in equilibrium with

water giving rise the estimated reservoir temperature (Fig. 7).
4.2.3. Water isotope, sulfate isotope and their equilibrium
The results concerning the isotope composition of the water molecule and the dissolved sulfate from four selected Obock hot springs samples and the local seawater are in Table 4. In the 2H(H2O) versus
18O(H2O) plot of Fig. 8A, all the samples fall on or near the Red
SeaGulf of Aden line of Craig (1969): 2H(H2O) = 6 18O(H2O). In
particular, the local seawater sample, with 2H(H2O) = + 5.2 and
18O(H2O) = +1.1, is enriched in heavier water isotopes, accordingly
to a coastal seawater from an arid climate and in comparison with the

220

reservoir

qu
00
1)

am

qu

(V artz
erm
a,
2

100

ste

ixi
ng

los

120

SiO2 (ppm)

art

140

no
s

tea

ml

oss

xin

mi

am

160

o
(Gu rphou
nna
s
rsso silica
n&
Arn
r

sso

180

pa
(Fo l
urn
ie

n, 1

r, 1
9

998

91)

200

80

60
)

83

40

ny al.,
do et

19

e n
alc so
ch rnors
A
(

20

hot
springs

this study

cold
waters

boreholes

1993

wells

seawater

0
0

100

200

300

400

500

600

700

800

Entalpy (kJ/kg)
Fig. 6. Dissolved silica-enthalpy diagram of the Obock hot springs from this study and literature (Houssein et al., 1993), local cold waters and seawater. The calculated dissolved silica
content of the local geothermal reservoir is SiO2 = 212 mg/L. Quart steam and no-steam loss curve are from Anrsson (2000). Amorphous silica-water equilibrium (Gunnarsson and
Arnrsson, 2000) and opal (Fournier, 1991) curves are also represented for comparison.

62

M.O. Awaleh et al. / Journal of Geochemical Exploration 152 (2015) 5466

Fig. 7. Diagrams showing the change in calculated saturation indices (log Q/K) of various minerals as a function of temperature (Obock-HS2 sample). The temperature range with the
convergence of saturation indices to zero is assumed to indicate the temperature of deep thermal reservoir for OBGS. It should be noted that similar diagrams were obtained for other
Obock beach hot springs.

et al., 2001) and 18O(SO4) = +8.6 (Boschetti and Iacumin, 2005). Furthermore, a comparison with the deep Red Sea waters and hydrothermal
brines of Atlantis II (Craig, 1969; Zierenberg and Shanks, 1986) reveals an
isotopic afnity with the former, and at present a trend towards more
enriched 34S(SO4) should be excluded (Fig. 8B).
The application of bisulfatewater or sulfatewater isotopic
geothermometer, which are based on the equilibrium exchange of
2
oxygen isotopes between aqueous HSO
4 or SO4 and H2O, respectively,
depends by various factors, mainly i) the dominant dissolved sulfur species in solution; ii) the kinetic of equilibration; and iii) the presence of
conductive or mixing cooling (Boschetti, 2013; Boschetti et al., 2003).
Considering the evident mixing between Obock thermal waters and seawater and to check if the mother thermal water is equilibrated with
18
18
2
HSO
4 or SO4 , the O(SO4) and O(H2O) values obtained in this
study was plotted on a 103ln(SO4H2O) versus 106/T2 diagram and
using the temperatures obtained by the chemical geothermometers
(Fig. 8C). Similarly to the Atlantis II brine from the Red Sea, Obock thermal
2
waters plot between the HSO
4 H2O and SO4 H2O equilibrium lines:
3

10 ln HSO4 H2 O 3:2610 =T 5:81

10 ln SO4 H2 O 2:54510 =T 6:61

where T is the temperature in Kelvin, Eq. (1) from Seal et al. (2000) and
Eq. (2) the best t of the data from Halas and Pluta (2000) and Zeebe
is the main sulfate species
(2010) (see Boschetti, 2013 for details). SO2
4
Table 4
Isotopic data of the hot springs and seawater from Obock.
Sample

18O(H2O)
( vs
V-SMOW)

2H(H2O)
( vs
V-SMOW)

3
H
(T.U.)

34S(SO4)
( vs
V-CDT)

18O(SO4)
( vs
V-SMOW)

Obock-HS1
Obock-HS2
Obock-HS3
Obock-HS6
Seawater

0.7
0.6
0.6
0.3
1.1

3.0
1.6
2.9
2.9
5.2

b0.6
b0.6
b0.6
b0.6
1.3

20.9
21.2
21.1
21.3
20.8

9.1
8.7
8.9
8.9
9.1

at the outpourings of the coastal hot springs, however HSO

4 would be the
prevailing oxidized sulfur species in the deep hydrothermal vent, probably originated by the disproportionation of magmatic SO2 (after Mller
et al., 2013):
3

10 ln HSO4 H2 O 1:85210 =T 0:288

The approaching of the Obock spring with the highest estimated

temperature to the intersection of the HSO
4 H2O equilibrium lines
sustains this hypothesis (Fig. 8C). In fact, the obtained temperature
from the intersection of Eqs. (1) and (3) occurs at T = 207 C,
which is in line with the temperature of 210 C estimated by Sanjuan
(2010) and our maximum temperature of 200 C estimated by the
geothermometrical modeling.

5. Discussion
It has been reported that the uids resulting from seawaterbasalt
interaction at temperature higher than 100 C have been completely
depleted of magnesium and sulfate (Bowers et al., 1988; Mottl and
Holland, 1978; Tomasson and Kristmannsdottir, 1972). Moreover,
straight lines were obtained between points of hot springs and point
of sea water when plotting normalized concentrations of elements to
chloride (Mg/Cl, SO4/Cl, K/Cl, Ca/Cl, SiO2/Cl) versus concentration of
Li/Cl (Fig. 9). Therefore, one may extrapolate the Mg and SO4 concentrations to zero, which corresponds to the maximum interaction between
seawater and basalt. The Mg/Cl and SO4/Cl straight lines get zero for the
same Li/Cl value of 3.228 104 (Fig. 9A). The silicate concentration of
the hot end-member (e.g. Li/Cl = 3.228 104) was estimated about
212 mg/L (Fig. 9B). When the value obtained for dissolved silica is used
as quartz geothermometer (Verma, 2001), it gives a geothermal reservoir
temperature of about 187 C, which is in good agreement with that obtained from the chemical geothermometer (172191 C, Table 2) as
well as from multiple mineral equilibrium approach (180200 C, Fig. 7).
As stated above, the 34S(SO4) compositions of Obock thermal
waters are similar to that of the present-day seawater. These results
rule out any evolution of the seawaterbasalt system that would not

M.O. Awaleh et al. / Journal of Geochemical Exploration 152 (2015) 5466

At
lan
tis

II

8.0

A
7.0

Obock
seawater

LM

18

WL

6.0

5.0

4.0

Gulf of Aden
seawater

3.0

H(H2O) ( vs V-SMOW)

63

20

60

0
10

0
14

0
18

22

26

30

Temperature (C)

2.0

-H O
2

12
1.0

0.8

1.0

1.2

1.4

10.0

9.5

Obock
seawater

9.0

Mean Ocean

Gulf of Aden

Obock

seawater

7
?

8.5
Red Sea

6
Atlantis II

8.0
5
3

7.5

8
6

10 /T(kelvin)

18

O(SO4) ( vs V-SMOW)

CaSO

10 ln

10

S O 24 -H
2O

0.6

18

0.4

O(H2O) ( vs V-SMOW)

-H2 O

0.2

2 -dis

0.0

SO

11
0.0

HSO -

SMOW

10

11

12

13

7.0

Atlantis II

6.5

6.0
20.0

20.5

21.0

21.5

22.0

34

S(SO4) ( vs V-CDT)

Fig. 8. Isotope diagrams of the hot springs (squares) and seawater (diamonds) from Obock. A: Water isotope data are compared with the Local Meteoric Water Line (LMWL; Fontes et al.,
1980), the seawater from the Gulf of Aden (Craig, 1969; Ganssen and Kroon, 1991), the Craig's line 2H(H2O) = 6 18O(H2O) relating surface seawaters of the area (Craig, 1969) and the
Atlantis II hydrothermal brine from the Red-Sea Rift (Pierret et al., 2001). B: The sulfate isotope values are compared with Red-Sea seawater sampled at different depth (triangles) and
Atlantis II hydrothermal brine (Zierenberg and Shanks, 1986). C: Sulfatewater oxygen isotope fractionation vs. temperature (after Boschetti, 2013). Gray squares and area depict the calculated fractionation using the obtained 18O(SO4) and 18O(H2O) isotope data and the temperature at the sampling site of the Obock's hot springs (Table 4) and Atlantis II (Zierenberg and
Shanks, 1986), respectively. Red squares and yellow area are the same samples, but using the temperature calculated by geothermometers. Horizontal arrows represent conductive and/or
mixing cooling. Please note that the 18O(SO4) data of Zierenberg and Shanks (1986) were recalculated considering the value of the mean present-day seawater 18O(SO4) = +8.6
(Boschetti, 2013; Boschetti and Iacumin, 2005).

yet attain the equilibrium (e.g. no end-member depleted of sulfate and

magnesium). Indeed, when the diminution of SO4 and Mg is due to a degree of advancement of the reaction without reaching the nal state, it
should remain SO4 in the solution that would present therefore
34 32
S/ S ratio higher than that of seawater because of isotopic fractioning
(Camo et al., 1991), which is not the case in Obock coastal geothermal
eld.
In the other hand, when the diminution of SO4 is the result of a blend
between an end-member, depleted of Mg and SO4, and the surrounding
seawater which mix each other during ascent, the 34S/32S ratio of the
end-member is similar to that of the seawater, which is the case of the
present study (Table 4).

The Obock region was dened as the nal termination of a rift

segment (Manighetti, 1993). In addition, Manighetti (1993) revealed
the occurrence of hydrothermalism in the Obock ridge. Therefore, the
following model was proposed for the Obock coastal geothermal eld
(Fig. 10): the magma chamber of the submarine volcano located beneath the Obock ridge could constitute the main heat source of the
Obock geothermal system. Seawater would inltrate through the basalts fractures to reach the geothermal reservoir. In this later, seawater
could react with the basalt bedrock at the temperature of the geothermal reservoir (197 10 C) for a relatively long time so that the equilibrium can be reached. Therefore, the seawaterbasalt interaction in the
geothermal reservoir could produce a geothermal uid which were

64

M.O. Awaleh et al. / Journal of Geochemical Exploration 152 (2015) 5466

Fig. 9. A: Mg/Cl and SO4/Cl versus Li/Cl; B: SiO2/Cl versus Li/Cl; C: Ca/Cl and K/Cl versus Li/Cl.

depleted of magnesium and sulfate. Then, during their ascent by

convection through basalt fractures, the geothermal uid is mixed in
varying degrees with cold seawater which, after an additional possible

mixing with freshwater, give rise to coastal hot springs of Obock city
(Fig. 10). This model could explain the fact that the thermal waters in
the present study are partially equilibrated at the outpourings. Likewise,

M.O. Awaleh et al. / Journal of Geochemical Exploration 152 (2015) 5466

65

Fig. 10. Simplied sketch of the conceptual model of the Obock geothermal system.

the lower amount of chloride in hot springs than in seawater suggests

the dilution by freshwater. This is further supported by the Obock
regional piezometric map which shows that the meteoric water ow
toward the sea (Fig. 2). However, one cannot preclude any phase separation process that could be responsible of the lower concentration of
chloride in Obock hot springs than in seawater as reported for some
submarine hydrothermal sources (Bischoff and Rosenbauer, 1989).
In a recent pre-feasibility study, a geophysical survey of the Obock
geothermal prospect, based on electro telluric, magnetotelluric and
gravimetric measurements, was conducted by the CERD (2013). These
geophysical investigations reveal a geothermal anomaly in an area
near the coast of Obock city. Therefore, these results support the conceptual model proposed for Obock beach geothermal manifestations
(Fig. 10).
6. Conclusions and perspectives
The reservoir temperatures in the Obock beach geothermal
system were assessed using a number of geothermometry techniques.
Chemical geothermometers (i.e. the Na/K, NaKCa, SiO2) and multiple
mineral equilibrium approaches estimate the temperatures of the deep
geothermal reservoir at 197 10 C, which is also conrmed by the
application of the 18O/16O isotope fractionation (HSO
4 H2O).
A conceptual model was proposed for the Obock beach geothermal
system. During their ascent through fractures, the geothermal uid is
mixed in varying degrees with cold seawater which, after an additional
possible mixing with small amounts of fresh water, give rise to coastal
hot springs of Obock city.
Similarly to the Atlantis II hydrothermal water from the Red Sea, the
hot waters from Obock have 34S(SO4) values very near to that of the
present-day seawater. Moreover, the magnesium and sulfate depletion
of the water are consistent with a uid having interacted with hot
axial-rift basalts.
However, future investigations, such as offshore geophysical
surveys, should be undertaken to locate the Obock geothermal reservoir.

The conceptual model proposed here clearly indicates that deep deviated
drilling should be used to intercept the Obock geothermal reservoir.
Acknowledgments
This research work was nancially supported by the Centre d'Etudes
et de Recherche de Djibouti (CERD). We are grateful to Abdi Abdillahi
Djibril and Samaleh Idriss for their assistances in the eld works. We
would like to thank the Editor-in-Chief, Prof. Benedetto De Vivo, and
an anonymous reviewer for their very constructive comments that
substantially improved the manuscript.
References
Anrsson, S., 2000. Mixing processes in upow zones and mixing models. In: Arnrsson, S.
(Ed.), Isotopic and Chemical Techniques in Geothermal Exploration, Development and
Use. Sampling Methods, Data Handling, Interpretation. IAEA, Vienna, pp. 200211.
Aquater, 1982. Etude technico-conomique de pr-faisabilit gothermique. Rapport
spcique Nouvelles perspectives de dveloppement intgr avec les ressources
gothermiques (Rpublique de Djibouti, 90 pp.).
Aquater, 1989. Geothermal Exploration Project. Republic of Djibouti. Final Report. CERD,
Rpublique de Djibouti (159 pp.).
Arnrsson, S., Gunnlaugsson, E., Svavarsson, H., 1983. The chemistry of geothermal waters
in Iceland III: chemical geothermometry in geothermal investigations. Geochim.
Cosmochim. Acta 47, 567577.
Awaleh, M.O., Hoch, F.B., Soubaneh, Y.D., Boschetti, T., Egueh, N.M., Elmi, S.A., 2015. The geothermal resources of the Republic of Djibouti II: geochemical study of the Lake Abhe
geothermal eld. J. Geochem. Explor. (submitted for publication).
Barberi, F., Varet, J., 1977. Volcanism of Afar: small-scale plate tectonics implications. Geol.
Soc. Am. Bull. 88, 12511266.
Barberi, F., Ferrara, G., Santacroce, R., Varet, J., 1975. Structural evolution of the Afar triple
junction. In: Pilger, A., Rosler, A. (Eds.), Afar Depression of Ethiopia, Inter-Union Commission on Geodynamics, Scientic Report 14, E. Schweizerbart'sche
Verlagsbuchhandlung (Ngele u Obermiller), Stuttgart, pp. 3854.
Bethke, C.M., 2008. Geochemical and Biogeochemical Reaction Modeling. Cambridge University Press, New York.
Bischoff, J.L., Rosenbaeur, R.J., 1989. Salinity variations in submarine hydrothermal systems by layered double diffusive convection. J. Geol. 97, 613623.
Boschetti, T., 2013. Oxygen isotope equilibrium in sulphatewater systems: a revision of
geothermometric applications in low-enthalpy systems. J. Geochem. Explor. 124,
92100.

66

M.O. Awaleh et al. / Journal of Geochemical Exploration 152 (2015) 5466

Boschetti, T., Iacumin, P., 2005. Continuous ow 18O measurements: new approach to
standardization, high temperature thermodynamics and sulphate analysis. Rapid
Commun. Mass Spectrom. 19, 30073014.
Boschetti, T., Cortecci, G., Bolognesi, L., 2003. Chemical and isotopic study of the
shallow groundwater system of Vulcano Island, Aeolian Archipelago, Italy. GeoActa
2, 134.
Bowers, T.S., Campbell, A.C., Measures, C.I., Spivack, A.J., Khadem, M., Edmond, J.M., 1988.
Chemical controls on the composition of vent uids at 1311N and 21N, East Pacific Rise. J. Geophys. Res. 93 (B5), 45224536.
BRGM, 1975. Territoire Franais des Afars et des Issas: rapport de n de sondage: rsultats
des premiers essais de production (75SGN442.GTH. 18 pp.).
Camo, T., Sakai, H., Ki., E.S., Shitashima, K., Ishibashi, J.I., 1991. High alkalinity due to sulphate reduction in Clam hydrothermal eld, Okinawa through. Earth Planet. Sci.
Lett. 107, 328338.
Centre d'Etudes et de Recherche de Djibouti (CERD), 2011. Etude de prfaisabilit de
Goubet (Rpublique de Djibouti).
Centre d'Etudes et de Recherche de Djibouti (CERD), 2012. Etude de prfaisabilit du Lac
Abh (Rpublique de Djibouti).
Centre d'Etudes et de Recherche de Djibouti (CERD), 2013. Etude de prfaisabilit
d'Obock (Rpublique de Djibouti).
Coplen, T.B., Hopple, J.A., Bhlke, J.K., Peiser, H.S., Rieder, S.E., Krouse, H.R., Rosman, K.J.R.,
Ding, T., Vocke Jr., R.D., Rvsz, K.M., Lamberty, A., Taylor, P., De Bivre, P., 2001. Compilation of minimum and maximum isotope ratios of selected elements in naturally
occurring terrestrial materials and reagents. US Geological Survey WaterResources
Investigations Report 01-4222.
Courtillot, V., Galdeano, A., Le Mouel, J.L., 1980. Propagation of an accreting plate boundary: a discussion of new aeromagnetic data in the Gulf of Tadjourah and southern
Afar. Earth Planet. Sci. Lett. 47, 144160.
Craig, H., 1969. Geochemistry and origin of Red Sea brines. In: Degens, E.T., Ross, D.A.
(Eds.), Hot Brines and Recent Heavy Metal Deposits in the Red Sea. Springer-Verlag,
New York, pp. 208242.
D'Amore, F., Arnrsson, S., 2000. Geothermometry. In: Arnrsson, S. (Ed.), Isotopic and
Chemical Techniques in Geothermal Exploration, Development and Use. Sampling
Methods, Data Handling, Interpretation. IAEA, Vienna, pp. 152199.
D'Amore, F., Giusti, D., Abdallah, A., 1998. Geochemistry of the high-salinity geothermal
eld of Asal, Republic of Djibouti, Africa. Geothermics 27, 197210.
Demange, J., Di Paola, G.M., Lopoukine, M., Stieltjes, L., 1971. Etude gothermique du
T.F.A.I., B.R.G.M., Rapport 7104063.
Doubre, C., 2004. Structure et mcanismes des segments de rift volcano-tectoniques;
tudes de rifts anciens (Ecosse et Islande) et d'un rift actif (AsalGhoubbet). (PhD
Thesis), Universit du Maine (420 pp.).
Ellis, A.J., 1979. Chemical geothermometry in geothermal systems. Geothermics 25,
219226.
Faure, H., 1976. Les dformations des ctes et le modle godynamique de l'Afar. In:
Pilger, A., Rosler, A. (Eds.), Afar between Continental and Oceanic Rifting. International Symposium on the Afar Region and Related Rift Problems, Bad Bergzabern, April
1974, Vol. 2, Inter-Union Commission on Geodynamics, Scientic Report, 16, Stuttgart, pp. 148155.
Faure, H., Hong, C.T., Lalou, C., 1980. Datation Th/U des calcaires coralliens et mouvements
verticaux Djibouti. Bull. Soc. Geol. Fr. 6, 959962.
Fontes, J.C., Pouchon, P., Saliege, J.F., Zuppi, G.M., 1980. Environmental isotope study of
groundwater systems in the Republic of Djibouti. Arid-zone Hydrology: Investigations with Isotope Techniques. IAEA, Vienna, pp. 237262.
Fouillac, R., Michard, S., 1981. Sodium/Lithium ratio in water applied to geothermometry
of geothermal reservoirs. Geothermics 10, 5570.
Fournier, R.O., 1977. Chemical geothermometers and mixing models for geothermal systems. Geothermics 5, 4150.
Fournier, R.O., 1979. A revised equation for the Na/K geothermometer. Geotherm. Resour.
Counc. Trans. 3, 221224.
Fournier, R.O., 1991. Water geothermometers applied to geothermal energy. In: D'Amore,
F. (coordinator), Application of geochemistry in geothermal reservoir development.
UNITAR/UNDP publication, Rome, 3769.
Fournier, R.O., Potter, R.W.I.I., 1979. Magnesium correction to NaKCa geothermometer.
Geochim. Cosmochim. Acta 43, 15431550.
Fournier, R.O., Truesdell, A.H., 1973. An empirical NaKCa geothermometer for natural
waters. Geochim. Cosmochim. Acta 37, 12551275.
Fournier, R.O., Potter, R.W., 1982. II A revised and expanded silica (quartz)
geothermometer. Geotherm. Resourc. Counc. Bull. 11, 312.
Ganssen, G., Kroon, D., 1991. Evidence for Red Sea surface circulation from oxygen isotopes of modern surface waters and planktonic foraminiferal tests. Paleoceanography
6, 7382.
Gasse, F., Fournier, M., Richard, O., 1983. Carte gologique de la Rpublique de Djibouti
1:100 000. Djibouti. Notice explicative. ISERST, Ministre franais de la Coopration,
Paris.
Giggenbach, W.F., 1988. Geothermal solute equilibria. Derivation of NaKMgCa
geoindicators. Geochim. Cosmochim. Acta 52, 27492765.
Giggenbach, W.F., 1991. Chemical techniques in geothermal exploration. In: D'Amore, F.
(coordinator), Application of geochemistry in geothermal reservoir development.
UNITAR/UNDP publication, Rome, 119142.
Groupe de la Banque Africaine de Dveloppement (GBAD), 2011. Document de
Stratgie Pays, 20112015 (Rpublique de Djibouti).
Gunnarsson, I., Arnrsson, S., 2000. Amorphous silica solubility and the thermodynamic
properties of H4SiO4 in the range of 0 to 350 C at Psat. Geochim. Cosmochim. Acta
64, 22952307.

18Halas, S., Pluta, I., 2000. Empirical calibration of isotope thermometer 18O(SO2
4 )
O(H2O) for low temperature brines. V Isotope Workshop European Society for Isotope Research, Krakw, Poland, pp. 6871.
Houssein, I.K., Sanjuan, B., Michard, G., 1993. Indices gochimiques de l'existence d'un
uide 210 C dans la rgion d'Obock (Rpublique de Djibouti). C. R. Acad Sci.
Paris II 316, 771776.
IAH, 1979. Map of mineral and thermal water of Europe. Scale 1:500,000. International
Association of Hydrogeologists, UK.
Kharaka, Y.K., Lico, M.S., Law, L.M., 1982. Chemical geothermometers applied to formation
waters, Gulf of Mexico and California basins. Am. Assoc. Petrol. Geol. Bull. 66, 588.
Lavigne, J., Lopoukine, M., 1970. Reconnaissance gothermique du T.F.A.I., B.R.G.M., 70
SGN, 109 GTM.
Manighetti, I., 1993. Dynamique des systmes extensifs en Afar. (Thse de Doctorat),
Universit de Paris VI (242 pp.).
Manighetti, I., King, G., Sammis, C.G., 2004. The role of off-fault damage in the evolution of
normal faults. Earth Planet. Sci. Lett. 217, 399408.
Mlynarski, M., Zlotnicki, J., 2001. Fluid circulation in the active emerged Asal rift (east
Africa, Djibouti) inferred from self-potential and tellurictelluric prospecting.
Tectonophysics 339, 455472.
Mottl, M.J., Holland, T.J., 1978. Chemical exchange during hydrothermal alteration of basalt by seawater. Experimental results for major and minor components of seawater.
Geochim. Cosmochim. Acta 42, 11031115.
Moyes, J., Daprat, J., Faugeres, J.C., Gonthier, E., Pujol, C., 1981. Etude Stratigraphique et
sdimentologique. In: C.E.P.M., C.N.E.X.O. (Eds.), Orgon IV, Golfe d'Aden, Mer
D'Oman. Novembre 1978. C.N.R.S., Paris, pp. 189263.
Mller, I.A., Brunner, B., Max, T., Breuer, C., Reeves, E.P., Thal, J., Bernasconi, S., Bach, W.,
2013. The isotope signature of magmatic SO2 disproportionation: a comparison between laboratory experiments and a hydrothermally active site in the Manus Basin,
Papua New Guinea. (PhD thesis), Universitt Bremen, pp. 105140 (Chapter 4).
Nicholson, K., 1993. Geothermal uids. Chemistry and Exploration Techniques. SpringerVerlag, Berlin Heidelberg (263 pp.).
Pang, Z.H., Armannsson, H., 2006. Analytical Procedures and Quality Assurance for Geothermal Water Chemistry. United Nations University Geothermal Training
Programme.
Pang, Z., Reed, M.H., 1998. Theoretical chemical geothermometry on geothermal waters:
problems and methods. Geochim. Cosmochim. Acta 62, 10831091.
Pierret, M.C., Clauer, N., Bosch, D., Blanc, G., France-Lanord, C., 2001. Chemical and isotopic
(87Sr/86Sr, 18O, D) constraints to the formation processes of Red-Sea brines.
Geochim. Cosmochim. Acta 65, 12591275.
Pinzuti, P., Mignan, A., Geoffrey, C.P.K., 2010. Surface morphology of active normal faults
in hard rock: implications for the mechanics of the Asal Rift, Djibouti. Earth Planet.
Sci. Lett. 299, 169179.
Piper, A.M., 1944. A graphic procedure in the geochemical interpretation of wateranalyses. Trans. Am. Geophys. Union 25, 914923.
Reed, M.H., 1982. Calculation of multicomponent chemical equilibria and reaction processes in systems involving minerals, gases and an aqueous phase. Geochim.
Cosmochim. Acta 46, 513528.
Reed, M.H., Spycher, W.H., 1984. Calculation of pH and mineral equilibria in hydrothermal
waters with applications to geothermometry and studies of boiling and dilution.
Geochim. Cosmochim. Acta 48, 14791492.
Richard, O., 1979. Etude de la transition dorsale ocanique-rift merg: le Golfe de
Tadjourah (Rpublique de Djibouti). (Thesis). Universit de Paris-Sud.
Sanjuan, B., 2010. Use of a new Sodium/Lithium (Na/Li) geothermometric relationship for
High-Temperature (HT) geothermal uids derived from seawater/basalt interaction
processes: application to the Djibouti case. Third East African Rift Geothermal Conference ARGEO-C3-DJIBOUTI. Exploring and Harnessing the Renewable and Promising
Geothermal Energy, Djibouti.
Seal II, R.R., Alpers, C.N., Rye, R.O., 2000. Stable isotope systematics of sulphate minerals.
In: Alpers, C.N., Jambor, J.L., Nordstrom, D.K. (Eds.), Sulphate Minerals: Crystallography, Geochemistry, and Environmental Signicance. Reviews in Mineralogy and Geochemistry 40. Mineralogical Society of America and Geochemical Society,
Washington, D.C., pp. 541602.
Sonney, R., Vuataz, F.D., 2010. Validation of Chemical and Isotopic Geothermometers from
Low Temperature Deep Fluids of Northern Switzerland. Proceedings World Geothermal Congress 2010 Bali, Indonesia.
Tomasson, J., Kristmannsdottir, H., 1972. High temperature alteration minerals and thermal brines, Reykjanes, Iceland. Contrib. Mineral. Petrol. 36, 123134.
Verma, M.P., 2001. Silica solubility geothermometers for hydrothermal systems. In: Cidu,
R. (Ed.), Proc. of the 10th Int. Symp. on WaterRock Interaction, Vol. 1, Villasimius,
Italy, June 1015, pp. 349352.
Verma, M.P., Santoyo, E., 1997. New improved equations for Na/K, Na/Li and SiO2
geothermometers by outlier detection and rejection. J. Volcanol. Geotherm. Res. 79,
924.
World Bank, 2013. Communiqus de Presse, 13 Octobre 2013. La Banque mondiale et
Djibouti signent un accord sur la prospection de l'nergie gothermique. http://
www.banquemondiale.org/fr/news/press-release/2013/10/13/world-bank-anddjibouti-sign-agreement-to-explore-geothermal-energy.
Zan, L., Gianelli, G., Passerini, P., Troisi, C., Haga, A., 1990. Geothermal exploration in the
Republic of Djibouti: thermal and geological data of the Hanl and Asal areas.
Geothermircs 19, 561582.
Zeebe, R.E., 2010. A new value for the stable oxygen isotope fractionation between dissolved sulphate ion and water. Geochim. Cosmochim. Acta 74, 818828.
Zierenberg, R.A., Shanks, W.C., 1986. Isotopic constraints on the origin of the Atlantis II,
Suakin and Valdivia brines, Red Sea. Geochim. Cosmochim. Acta 50, 22052214.