Journal of Geochemical Exploration 152 (2015) 54–66

Contents lists available at ScienceDirect

Journal of Geochemical Exploration

j ourna l homepage: www.e lsev ie r .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'aéroport, B.P. 486, Djibouti — ville, Djiboutib Laboratoire Régional, Newalta Châteauguay, 125 Rue Bélanger, Châteauguay, J6J 4Z2 Québec, Canadac Département de biologie, chimie et géographie, Université du Québec à Rimouski, 300, Allée des Ursulines, Rimouski, QC G5L 3A1, Canadad Department of Physics and Earth Sciences “Macedonio Melloni”, University of Parma, Parco Area delle Scienze 157/a, 43124 Parma, Italy

⁎ Corresponding author. Tel.: +39 253 77 84 68 55; faxE-mail address: awaleh@gmail.com (M.O. Awaleh).

http://dx.doi.org/10.1016/j.gexplo.2015.02.0010375-6742/© 2015 Elsevier B.V. All rights reserved.

a b s t r a c t

a r t i c l e i n f o

Article history:Received 8 May 2014Accepted 1 February 2015Available online 7 February 2015

Keywords:Hot springsGeothermometersChemical compositionIsotopic compositionObockDjibouti

This paper examines the hydrochemical features of the waters from the Obock coastal geothermal field. As thesubmarinewaters at the ridge, their chemical and isotope composition shows affinities with sea water–basalt in-teractions at hydrothermal temperature. Moreover, good linear correlations were obtained between hot springsand 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 endmembers exist (seawater andreservoir fluid). After extrapolation of the Mg and SO4 concentrations to zero, the obtained value of SiO2 corre-sponding to the reservoir fluid end member was used to estimate the reservoir temperature by quartzgeothermometers. The obtained temperature of 187 °C is in good agreement with that obtained from multiplemineral equilibrium approach (180–200 °C), cationic geothermometers (172–191 °C) and by the evaluation ofisotopic 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 δ34S(SO4) signature of the dissolved sulfates in Obock thermal waters confirms that these watersresult from the mixture of a hot seawater-derived fluid (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 locatedon the East African Rift System where geology is also resulting fromtwo 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 andZlotnicki, 2001).

In Djibouti, most of the widespread geothermal activity, manifestedin the formof numerous hot springs, fumaroles and hydrothermal alter-ation, is located mainly in thewestern part of the country and along theGulf 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 LakeAsal in order to understand the phenomena related to sea floor

: +39 253 21 35 45 68.

spreading (Mlynarski and Zlotnicki, 2001; Pinzuti et al., 2010). More-over, geothermal studies undertaken in the late 1960s and early 1970sallowed selecting the Asal prospect as themost favorable area for carry-ing out deep drilling exploration (Demange et al., 1971; Lavigne andLopoukine, 1970). Therefore, six geothermal wells (Asal 1 to 6) withvarious depths, between 1137 m and 2105 m, were drilled in theAsal prospect during the last decades (Aquater, 1989; BRGM, 1975).However, the high enthalpy Asal geothermal fluids (about 350 °C)have had high salinity (116 g/kg) (D'Amore et al., 1998).

On the other hand, since the main national electricity productionsources depend on petroleum products, Djibouti is relatively dependenton diesel fuel and fuel oil imports to meet its energy needs. As a conse-quence, the electricity cost is exorbitant in the Republic of Djibouti,where in average 1 kWh costs about 0.23 U.S. $ (GBAD, 2011). To miti-gate this energy burden, which put a brake on the rate of economicgrowth in this country, a national program for development of geother-mal resources (NPDGR) was implemented in 2010. This program aimedto conduct multidisciplinary studies (geochemical, hydrogeological,geophysical, geological and reservoir engineering) on all areaswith geo-thermal activities in the Republic of Djibouti (CERD, 2011, 2012; Awalehet al., submitted for publication).

Fig. 1. A: Simplified 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 thecoast 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).

55M.O. Awaleh et al. / Journal of Geochemical Exploration 152 (2015) 54–66

On the basis of new pre-feasibility studies related to the NPDGR, theWorld Bank granted, in 2012, a loan about 31MU.S. $ to the Republic ofDjibouti to perform four new geothermal wells in the area of Lake Asal(World Bank, 2013). As part of this NPDGR, the hot springs of theObock beach were also studied in 2013 (CERD, 2013). Previously, 3samples of Obock hot springs in the intertidal zone were investigatedto develop a spa center (Aquater, 1982). Nowadays, the thermal watersof Obock beach are used mainly for hydrotherapy. In the early 1990s,Houssein et al. (1993) studied the geochemistry of four Obock beachthermal springs.

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

56 M.O. Awaleh et al. / Journal of Geochemical Exploration 152 (2015) 54–66

Furthermore, the geochemical and isotopic study of the thermal watersof Obock beach, combined with geological and tectonic information aswell as regional hydrogeology, have been used to propose a conceptualmodel 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 ofthe Afar Rift which is at the triple junction between two nascentmid-oceanic ridges (Red Sea and Gulf of Aden) and a continental rift(East African Rift) (Fig. 1B, Doubre, 2004). The main expression ofcrustal rifting process in Afar is, in addition to extensional faulting, abimodal magmatismwhich began as early as 30Ma by the emplacementof trap volcanism in Yemen and Ethiopia (Barberi and Varet, 1977).

The Obock area located in the northern part of the country consistsof a Quaternary formation of coralline limestone (Faure et al., 1980;Gasse et al., 1983; Manighetti, 1993).

The Debergade paleo-valley is characterized by conglomerates withlarge block which have a thickness up to 10 m probably coming fromtheneighboring reliefs (Mablas rhyolites andDahla basalts). TheMablasrhyolitics correspond to a wide range of domes and lava flows, eruptedbetween 18 and 11m.y (Manighetti, 1993) and observed in two sectorson both sides of Tadjourah Gulf (Manighetti et al., 2004). This series un-conformably overlies the Dahla basaltic series (8.6–3.8m.y)which formthe (i) outer rim of the Ali Sabieh antiform and (ii) the western flank ofthe Danakil depression (Barberi et al., 1975). The Dahla basaltic seriesform the outer belt of the Danakil depression, where their thicknessreaches 1000 m in the Randa area (south of the Danakil depression).

Along the coast, alluvial deposits are associated to coral reef materialsand shape small parallel hill which are reminiscent an ancient littoral(Gasse et al., 1983). The North West of the study area is characterizedby red silt plated to coral reef where the clay fraction mainly consists of

Fig. 2. Piezometric map o

illite, smectite and traces of kaolinite and quartz, feldspar, goethite andchlorite minerals (Gasse et al., 1983). Rhyolite pebbles from the Mablasseries form a reg on the surface of the coral reef, in response to climaticchange. The Gulf basalts constitute the basis of these formations (Faureet al., 1980).

From a tectonic point of view, the coral reefs are raised by verticalmovements associated with the opening of Tadjourah Gulf (Faure,1976). The Gulf basalts, detrital materials, and coral reefs are affected bynormal faults trending N 20 to N 50° (Faure et al., 1980; Gasse et al.,1983; Richard, 1979). These normal faults take gradually the E–Wdirec-tion of the Tadjourah ridge axis (Gasse et al., 1983). Two major tectonicsstructures with N 110 orientation connecting the two bank of theDebergade paleo-valley with the Obock pit (1600 m of depth) character-ize the Obock area (Fig. 1C). These structures have also been observed inthe magnetic map produced by Courtillot et al. (1980) (Fig. 1D).

The Obock area is known to be the onshore prolongation of theTadjourah spreading ridge (Manighetti, 1993). These prolongationsallow explanation of the large elevation of the coral reefs (Gasse et al.,1983; Manighetti, 1993). The Gulf basalts have been defined as thechronological markers of the dislocation of Tadjourah ridge (Faure,1976; Gasse et al., 1983; Manighetti, 1993). Geophysical data gatheredduring the IV Orgon cruise (1981) on the axial valley of the Tadjourahridge reveals (i) the presence of a huge volcano and (ii) two types ofsediment clearly differentiated by their reflector. The Obock pit wasdrilled at 1070 m depth and the petrography based on cores consistsof clay, carbonate enriched on pelagic organisms, dated from LatePleistocene to Holocene (Moyes et al., 1981).

2.2. Hydrogeology of the area

The piezometric map of Obock region is based upon few points ofmeasurements (Fig. 2), it therefore remains indicative and the valuesof groundwater hydraulic heads and hydraulic gradients could bemodified according to newmeasurements. About ten points irregularly

f the Obock region.

Table1

Chem

ical

analyses

ofthermal

andco

ldwatersfrom

Obo

ckarea.

Samples

Latitude

Long

itud

eT °C

pHEC (μS/cm

)TD

S(m

g/L)

Hyd

roch

emical

type

sCa (m

g/L)

Mg

(mg/L)

Na

(mg/L)

K (mg/L)

HCO

3

(mg/L)

Cl (mg/L)

SO4

(mg/L)

NO3

(mg/L)

IB (%)

Li (mg/L)

F (mg/L)

Br (mg/L)

SiO2

(mg/L)

Thermal

waters

Obo

ck-H

S143

°17′21

.7″

11°57′33

.9″

58.6

6.81

47,450

35,600

Na–

Cl75

394

310

,900

440

118.6

18,800

1992

1.1

21.30

1.03

5682

.59

Obo

ck-H

S243

°17′18

.7″

11°57′33

.5″

71.4

7.01

45,890

33,700

Na–

Cl10

0172

795

6555

013

1.4

16,700

1503

1.6

31.91

1.16

5491

.98

Obo

ck-H

S343

°17′17

.2″

11°57′33

.4″

63.0

6.95

45,130

33,400

Na–

Cl88

184

898

6454

514

2.6

17,800

1699

2.3

11.86

1.19

5310

1.65

Obo

ck-H

S443

°17′17

.3″

11°57′33

.8″

60.0

6.93

41,240

30,000

Na–

Cl86

370

689

5554

415

1.2

16,400

1484

0.6

12.05

0.96

4811

8.6

Obo

ck-H

S543

°17′18

.1″

11°57′33

.3″

63.6

6.36

43,400

32,070

Na–

Cl88

179

796

0451

514

0.4

16,700

1562

2.9

31.85

1.60

5210

1.65

Obo

ck-H

S643

°17′16

.7″

11°57′33

.6″

63.4

6.74

45,170

33,100

Na–

Cl89

280

099

0350

313

3.9

17,600

1622

0.52

22.09

1.14

5410

7.55

Obo

ck-H

S743

°17′16

.6″

11°57′33

.5″

63.5

6.96

45,330

33,650

Na–

Cl80

976

910

,100

548

147.8

17,600

1512

0.88

22.03

1.19

5285

.32

Boreho

leswaters

Illysola

43°05′54

.7″

12°07′00

.6″

29.5

7.58

1948

1374

Na–

Cl–HCO

3–Mg

69.8

76.3

273

3.47

294.7

448

151

301

0.00

71.18

2.02

93.89

Soub

lali

43°12′19

,.4″

12°03′07

.1″

32.1

7.63

2391

1234

Na–

Cl–HCO

3–Mg

62.2

78.7

369

4.84

309.2

569

134

303

0.08

0.84

2.52

96.46

Assassane

143

°09′25

.1″

12°06′47

.1″

34.7

7.34

2030

1346

Na–

Cl–HCO

3–Mg

70.5

77.3

288

3.86

353.6

449

137

331

0.01

61.28

2.00

118.5

Assassane

243

°09′35

.9″

12°06′45

.0″

34.8

7.42

2063

1260

Na–

Cl–HCO

3–Mg

72.7

73.3

300

4.21

350.7

463

142

301

0.00

251.04

2.13

106.37

Wellw

aters

Obo

ck1

43°17′28

.2″

11°58′17

.8″

33.6

7.35

4251

3023

Na–

Cl–Ca

–Mg

206

155

549

27.9

39.99

1607

156

182

0.08

70.77

4.7

42.14

Obo

ck2

43°17′21

.3″

11°58′24

.3″

32.4

7.49

2851

1939

Na–

Cl–Ca

–Mg

140

81.8

320

1681

.93

855

87.7

171

0.00

050.36

2.7

24.24

Kho

maAdo

u43

°10′16

.2″

11°56′50

.7″

27.8

7.72

10,530

7407

Na–

Cl–Mg

405

310

1517

73.6

82.05

3740

387.4

411

0.00

080.65

1215

.53

Alla

-ella

43°13′39

.3″

11°57′47

.7″

31.4

7.94

1322

922

Na–

Cl–HCO

332

.19.17

279

6.7

359.5

209

73.82

422

0.00

020.76

0.84

61.39

Tero-ela

43°08′33

.7″

12°03′17

.5″

28.9

7.01

2590

1933

Na–

Cl–HCO

311

473

.37

378

8.8

406.8

665

157

41-2

0.21

0.7

2.6

n.a.

Seaw

ater

42°31′26

.0″

11°31′36

.5″

27.6

8.06

52,190

41,100

Na–

Cl40

113

8912

,200

350

148.1

21,400

3031

8.2

00.21

51.70

673.63

n.a.=

notan

alyzed

;IB:

ionicba

lanc

e.

57M.O. Awaleh et al. / Journal of Geochemical Exploration 152 (2015) 54–66

spread over an area of 1000 km2 allows drawing up the general featureof the underground flows. The groundwater flow, oriented from theWest to East and the South-East across the Obock plain, seems to bemainly controlled by the mountains of Mabla rhyolites in the Westand the mountains in the North where the main recharge occurs fromthe wadi Sadai and its tributaries.

The piezometric levels evolve rapidly fromObock coastal zone to theupstream area in the mountains. Close to the Obock town large diame-ter wells taping inferoflux underground flow indicates a piezometriclevel of 1 m above sea level. Within the wide sedimentary plain thewells of Oulma and Soublali give respectively 5 m and 20 m. In themountainous zone, the piezometric levels reach 142 and 198 m on thewells Assassan and Illisola respectively. More Westward the site ofWaddi shows a piezometric level of 324m. Consequently, the hydraulicgradients are generally high varying from 4.3‰ to 13‰.

3. Material and methods

A total of seven geothermal springs, the seawater and ninewells andboreholes cold waters in the Obock area were sampled in January 2013.Coastal thermal waters were collected at very low tide, in which periodthese waters upraised without any mixing with surface seawater.Unstable hydrochemical parameters, including temperature (±0.1 °C),pH (±0.01 unit) and electrical conductivity (±1 μS/cm), weremeasuredon site with, respectively, hand-held meters Hanna CheckTemp, EutechInstruments pH 610, and Eutech Instruments COND 610 that werecalibrated in the field prior to sampling.

Water samples were collected in polyethylene containers afterfiltration through 0.45 μmmembrane filters. All samples used for deter-mination of cations were acidified after collection through addition ofSuprapure® HNO3 (Merck) to bring the pH below 2.

Analyses of anions andmajor cationswere carried out by ionic chro-matography with a Dionex ICS 3000 Ion Chromatograph using analyti-cal procedures and quality assurance for geothermal water chemistryreported by Pang and Armannsson (2006). AS4A SC-4 mm analyticalcolumn (250 mm × 4 mm ID) coupled with AG4A SC 4 mm guard andCS12A analytical column (250 mm × 4 mm ID) coupled with CG12Aguard were used respectively for anions and cations analyses. The ionchromatograph was calibrated through repeated analysis of five work-ing anion and cation standards (with concentrations within the rangeof analyses). Peaks were identified using Chromeleon software (Dionex,Sunnyvale, California). The analytical precision was estimated at ±5%.

For the analysis of aqueous SiO2, the water samples were dilutedtenfold using deionized water to prevent the SiO2 precipitation. SiO2

contents were determined by colorimetry and analyzed using a Jenway6300 spectrophotometer, while HCO3 was analyzed by titration with0.1 M HCl. The ionic balance was below 3%.

Additional samples of untreated waters were collected in 50 mLglass bottles (Quorpak) for stable isotopes analyses of the watermolecule, δ2H(H2O) and δ18O(H2O), and 1000 mL plastic bottles fortritium (3H) analysis, all analyzed at the Bureau de RechercheGéologique etMinéralogique (BRGM) in Orleans, France. The deuteriumand oxygen isotopic ratios were analyzed using a Finnigan MAT 252mass spectrometer and reported in per mil notation (‰) versus theVienna Standard Mean Ocean Water (V-SMOW) standard followingδ = [(Rsample / Rstandard) − 1] × 1000, where R is the 2H/1H and18O/16O absolute isotope amount ratios. The average precision, basedon multiple analyses of various samples and laboratory standards, was±0.1‰ for δ18O(H2O) and ±0.8‰ for δ2H(H2O). Tritium activity wasmeasured by direct liquid scintillation counting. The detection limitwas 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-acidwashed plastic perplex bottles. Cd-acetatewas already added in thebot-tles (5% v/v) prior to sample collection, to fix sulfur as CdS, and then the

58 M.O. Awaleh et al. / Journal of Geochemical Exploration 152 (2015) 54–66

aliquot was filtered through a 0.2 μm nitrocellulose filter before chemi-cal determination of residual sulfate. Dissolved sulfate was precipitatedas BaSO4 at pH b 4 (in order to remove HCO3

− and CO32− species) by

adding a BaCl2 solution. The isotopic analyses on BaSO4 were carriedout using a Delta + XP mass spectrometer coupled in continuous-flowmode to a Thermo Elemental Analyzer in BRGM laboratories. Theδ34S(SO4) and δ18O(SO4) isotope compositions were reported in theusual δ-scale in ‰ with reference to V-CDT (Canyon Diablo Troilite)and V-SMOW (Vienna Standard Mean Ocean Water), respectively. Theanalytical precision was of ±0.3‰ versus V-CDT for δ34S(SO4) and±0.5‰ versus V-SMOW for δ18O(SO4).

4. Results

4.1. Hydrochemistry

Hydrothermal activity in the Obock beach is characterized by low-flow hot springs (1.8 L s−1) (Aquater, 1982). The temperature, pH, elec-trical conductivity (EC), total dissolved solids (TDS), sampling locations,hydrochemical types and major and minor elements of the sampledthermal and cold waters are listed in Table 1. Thermal waters fromObock beach are moderately acid to neutral (pH = 6.36–7.01) with atemperature range of 58.6 to 71.4 °C (Table 1). The chemical composi-tions of thermal waters are similar to that of seawater, with EC valuesin the range of 41 to 47 mS/cm, while in the cold well and boreholewaters from Obock area ranged from 1.3 to 4.2 mS/cm (Table 1). Therelatively high EC (10 mS/cm) of Khoma Adou well, which is locatedinland 200 m from shoreline, likely indicates seawater intrusion.

Chemical compositions of the waters are plotted in the diagram ofPiper (1944), as shown in Fig. 3. Classification of the water samples inTable 1 was made according to principles of IAH (1979). Total equiva-lents of cations and anions separately were accepted as 100% and ionswith more than 20% (meq/L) were taken into consideration in thisclassification.

Fig. 3. A: Piper diagram of seawater and hot springs waters from Ob

The chemical composition of the waters studied described interms of relative concentrations of main anion and cation allows us todistinguish the following two groups of waters (Table 1, Fig. 3):

(1) Thermal waters from the Obock coastal hot springs are of theNa–Cl type;

(2) Cold well and borehole waters fromObock area aremostly of theCl–HCO3–Na–Ca–Mg type.

According to Giggenbach (1991), the chemical composition of thethermal water samples from Obock beach can be classified aschloride-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 chloride–bicarbonateborder (Fig. 4A). When comparing results of this study with previousdata (Aquater, 1982; Houssein et al., 1993), no significant variations influid geochemistry were observed between 1980 and 2013 (Fig. 4B).Such consistent chemical character suggests uniform reservoir for theObock beach geothermal system (OBGS).

4.2. Geothermometry

The Na/1000–K/100–Mg0.5 ternary plot of Giggenbach (1988) canbe used to discriminatematurewaters,whichhave attained equilibriumwith relevant hydrothermal minerals, from immature waters andwaters affected bymixing and/or re-equilibration along their circulationpath (Fig. 5). This provides an indication of the suitability of thewaters for the application of solute geothermometers. Since allhot springs from OBGS were partially equilibrated (Fig. 5), chemicalgeothermometers can be used to estimate geothermal reservoir tem-perature (Arnórsson et al., 1983; Fournier and Potter, 1979; Fournierand Truesdell, 1973; Giggenbach, 1988). Therefore, classical and

ock beach; B: cold borehole and well waters from Obock area.

Fig. 4. Cl–SO4–HCO3 classification diagram of Giggenbach (1991). A: Thermal and coldwaters from theOBGS (this study); B: hot springswaters fromObock beach (1981: Aquater,1982, 1989: Houssein et al., 1993; 2013: this study).

59M.O. Awaleh et al. / Journal of Geochemical Exploration 152 (2015) 54–66

isotopic geothermometers as well as geothermometrical modelingapproach were used to estimate the reservoir temperature of OBGS.

4.2.1. Chemical geothermometers

4.2.1.1. Cationic geothermometers. The Na–K geothermometer was ini-tially developed to locate themajor upflow in high enthalpy geothermalsystems, because a general decrease in Na/K ratios of geothermal fluidswith increasing temperatures was observed (D'Amore and Arnórsson,2000). This geothermometer is related to the variation of sodium andpotassium in thermal waters due to ion exchange of these elementsbetween coexisting alkali feldspars (Nicholson, 1993). Na–K equationsare adapted for reservoir temperatures in the range 180–350 °C(Ellis, 1979), but are limited at lower temperatures, notably less than120 °C (Nicholson, 1993).

Several Na–K geothermometers have been proposed and applied tocalculate reservoir temperature in the past three decades. The Na–Kgeothermometer of Giggenbach (1988), based on thermodynamicdata of albite and K-feldspar, gives an equilibrium temperature of

deep parent fluids of about 169–195 °C (Table 2). Furthermore, theNa–K geothermometer of Fournier (1979) and Arnórsson et al. (1983)estimated the temperature of the geothermal reservoir respectively atabout 156–185 °C and 153–179 °C (Table 2). Those temperatures arein agreement with the values obtainedwith Na–K–Ca geothermometer,173–192 °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 Na–Li geothermometer was initially developed by Fouillac andMichard (1981) from a statistic study about groundwaters in graniticand volcanic domains. Later on, Na–Li equation was recalibrated byKharaka et al. (1982) andVermaand Santoyo (1997) obtaining differentequations. Moreover, the Na/Li geothermometer appears to be sensitiveto the total dissolved solids of the water at depth, locally controlled bycontributions of Na–Cl, and the rock type (Sonney and Vuataz, 2010).Based on these results, these Na/Li geothermometers cannot be consid-ered reliable for the Obock beach geothermal system.

However, a Na/Li geothermometer adapted for high-temperaturefluids derived from seawater interacting with basalt in emergedrifts was recently proposed (Sanjuan, 2010). Although it might beaffected 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 othergeothermometers are in agreement, except for the Obock-HS1(Table 2). This exception might be explained considering that theObock-HS1 water showed low silica, high SO4 and high Mg concentra-tions, which indicate a higher contribution of cold seawater.

4.2.1.2. Silica geothermometers. The silica geothermometers, basedon the content aqueous silica and the solubility of different silicaphases as a function of temperature, can be used to determine thetemperature of water–rock interaction processes at depth in a risingfluid before its discharge. However, among the several existing silicageothermometers, in this study only equations for quartz were used be-cause 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 applica-tion of the quartz equations proposed by Fournier (1977) are presented.Estimated temperatures in the OBGS reservoir range between 125 and147 °C, in agreementwith values (111–137 °C) obtained by using previ-ous data (Houssein et al., 1993). Quartz temperatures were generallylower than those obtained with cationic geothermometers (Table 2).Such differences might be due to disequilibrium between quartz andthe hot water, probably caused by mixing between waters of differenttemperature and silica content (Fig. 6). In fact, the cold terrestrialwatersof meteoric origin, probably representing an end-member, appearequilibrated with silica phases (amorphous silica, opal, chalcedony)more soluble than quartz (Fig. 6). Precipitation of silica phases duringmixing between the hot fluid with cold seawater, as typically occurs inhydrothermal vents (e.g. Bethke, 2008), could also explain the lack ofequilibrium with respect to quartz.

4.2.2. Geothermometrical modelingThe state of equilibrium between water and mineral is a

function of temperature; therefore saturation indices can be used asgeothermometers (Pang and Reed, 1998; Reed and Spycher, 1984). Inother words, estimation of reservoir temperature can be achieved bysimultaneous consideration of the equilibrium state between specificwater and many hydrothermal minerals as a function of temperature.

Geothermometrical modeling calculations have been performedfor the OBGS hot springs using the SOLVEQ computer program (Reed,1982).

Fig. 7 depicts the mineral saturation indices versus increasingtemperature for Obock coastal hot springs. For those hot springs, satura-tion indices with respect to quartz, chlorite, brucite, epidote, prehnite,microcline, forsterite and monticellite minerals tend to get closer tothe zero (SI = 0) around the temperature of 180–200 °C at which

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

Reference Hot spring code Na–K–Ca Na–K Na–K Na–K Na–K Na–Li

Fournier and Truesdell(1973)a

Fournier(1979)b

Arnorsson et al.(1983)c

Giggenback(1988)d

Verma and Santoyo(1997)e

Sanjuan(2010)f

This study Obock-HS1 173 156 153 169 155 127This study Obock-HS2 188 181 176 192 178 170This study Obock-HS3 187 178 173 189 176 165This study Obock-HS4 191 185 179 195 182 183This study Obock-HS5 185 176 171 187 174 167This study Obock-HS6 183 172 168 183 170 176This study Obock-HS7 188 177 172 188 175 171Mean (except HS 1) 187 178 173 189 176 172Standard deviation (except HS 1) 3 4 4 4 4 7Houssein et al. (1993) Obock-HS1 190 185 179 195 182 143Houssein et al. (1993) Obock-HS2 186 175 170 186 173 115Houssein et al. (1993) Obock-HS3 193 185 179 195 182 148Houssein et al. (1993) Obock-HS4 187 178 172 188 175 120

Mean 189 181 175 191 178 –

Standard deviation 3 5 5 5 5 –

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) + β ∗ Log(Ca0.5/Na)] − 273.15; β = 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.

60 M.O. Awaleh et al. / Journal of Geochemical Exploration 152 (2015) 54–66

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

Reference Hot spring code Quartz Quartz Quartz Quartz

Fournier (1977)a Fournier (1977)b Fournier and Potter (1982)c Verma (2001)d

This study Obock-HS1 127 124 127 124This study Obock-HS2 133 129 133 130This study Obock-HS3 138 133 138 136This study Obock-HS4 147 141 147 146This study Obock-HS5 138 133 138 136This study Obock-HS6 141 136 141 140This study Obock-HS7 129 125 129 126Houssein et al. (1993) Obock-HS1 127 132 128 124Houssein et al. (1993) Obock-HS2 111 118 111 106Houssein et al. (1993) Obock-HS3 133 137 133 130Houssein et al. (1993) Obock-HS4 113 120 113 109

a T = {1309/[5.19 − Log(S)]}− 273.15.b T = {1522/[5.75 − Log(S)]}− 273.15.c T = −42.2 + 0.28831 ∗ (S) − 3.6686 ∗ 10−4 ∗ (S)2 + 3.1665.10−7 ∗ (S)3 + 77.034 ∗ Log(S).d T = {−1175.7/[Log(S)− 4.75]}− 273.15; where (S) = silica concentration as SiO2 (ppm).

61M.O. Awaleh et al. / Journal of Geochemical Exploration 152 (2015) 54–66

temperature these minerals are assumed to be in equilibrium withwater giving rise the estimated reservoir temperature (Fig. 7).

4.2.3. Water isotope, sulfate isotope and their equilibriumThe results concerning the isotope composition of the water mole-

cule and the dissolved sulfate from four selectedObock hot springs sam-ples and the local seawater are in Table 4. In the δ2H(H2O) versusδ18O(H2O) plot of Fig. 8A, all the samples fall on or near the RedSea–Gulf of Aden line of Craig (1969): δ2H(H2O) = 6 × δ18O(H2O). Inparticular, the local seawater sample, with δ2H(H2O) = +5.2‰ andδ18O(H2O)=+1.1‰, is enriched in heavierwater isotopes, accordinglyto a coastal seawater from an arid climate and in comparison with the

0

20

40

60

80

100

120

140

160

180

200

220

0 100 200 300 400

seawater

m

mixi

ng

Ental

SiO

(pp

m)

2

amor

phou

s si

lica

(Gun

nars

son

& A

rnór

sson

, 199

8)

opal

(Fou

rnie

r, 19

91)

chalce

dony

(Arnorsson et a

l., 1983)

Fig. 6. Dissolved silica-enthalpy diagram of the Obock hot springs from this study and literatucontent of the local geothermal reservoir is SiO2 = 212 mg/L. Quart steam and no-steam lossArnórsson, 2000) and opal (Fournier, 1991) curves are also represented for comparison.

mean values of the Obock's springs, δ2H(H2O) = +2.6 ± 0.7‰ andδ18O(H2O) = +0.55 ± 0.17‰, which are slightly depleted in compari-son to the typical Gulf of Aden values (Fig. 8A). This is probably due to thefact that: i) the main recharge of the hot springs didn't occur near theiroutpouring sites but in a more open (and deep) ocean environment;ii) a possible mixing with freshwater of meteoric origin. According tothis latter hypothesis, the analyzed sample with lower salinity falls onthe local meteoric water line (LMWL in Fig. 8A; Fontes et al., 1980), prob-ably due to the flow of local groundwater towards the coast (Awalehet al., submitted for publication). However, the δ34S(SO4) and δ18O(SO4)values of the springs confirms a signature quite similar to the local(Fig. 8B) and mean ocean water, i.e. δ34S(SO4) = +21.1‰ (Coplen

500 600 700 800

ixing

py (kJ/kg)

quar

tz(V

erm

a, 2

001)

quar

tz no s

team

loss

stea

m lo

ss

reservoir

this study

boreholes

wells

1993

hotsprings

coldwaters

re (Houssein et al., 1993), local cold waters and seawater. The calculated dissolved silicacurve are from Anórsson (2000). Amorphous silica-water equilibrium (Gunnarsson and

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 theconvergence 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 otherObock beach hot springs.

62 M.O. Awaleh et al. / Journal of Geochemical Exploration 152 (2015) 54–66

et al., 2001) and δ18O(SO4)=+8.6‰ (Boschetti and Iacumin, 2005). Fur-thermore, a comparisonwith the deep Red Sea waters and hydrothermalbrines of Atlantis II (Craig, 1969; Zierenberg and Shanks, 1986) reveals anisotopic affinity with the former, and at present a trend towards moreenriched δ34S(SO4) should be excluded (Fig. 8B).

The application of bisulfate–water or sulfate–water isotopicgeothermometer, which are based on the equilibrium exchange ofoxygen isotopes between aqueousHSO4

− or SO42− andH2O, respectively,

depends by various factors, mainly i) the dominant dissolved sulfur spe-cies in solution; ii) the kinetic of equilibration; and iii) the presence ofconductive or mixing cooling (Boschetti, 2013; Boschetti et al., 2003).Considering the evident mixing between Obock thermal waters and sea-water and to check if the mother thermal water is equilibrated withHSO4

− or SO42−, the δ18O(SO4) and δ18O(H2O) values obtained in this

study was plotted on a 103lnα(SO4–H2O) versus 106/T2 diagram andusing the temperatures obtained by the chemical geothermometers(Fig. 8C). Similarly to theAtlantis II brine from theRed Sea, Obock thermalwaters plot between the HSO4

−–H2O and SO42−–H2O equilibrium lines:

103lnαðHSO−4 –H2OÞ ¼ 3:26106

=T2−5:81 ð1Þ

103lnαðSO2−4 –H2OÞ ¼ 2:545106

=T2−6:61 ð2Þ

where T is the temperature in Kelvin, Eq. (1) from Seal et al. (2000) andEq. (2) the best fit of the data from Halas and Pluta (2000) and Zeebe(2010) (see Boschetti, 2013 for details). SO4

2− is the main sulfate species

Table 4Isotopic data of the hot springs and seawater from Obock.

Sample δ18O(H2O)(‰ vsV-SMOW)

δ2H(H2O)(‰ vsV-SMOW)

3H(T.U.)

δ34S(SO4)(‰ vsV-CDT)

δ18O(SO4)(‰ vsV-SMOW)

Obock-HS1 0.7 3.0 b0.6 20.9 9.1Obock-HS2 0.6 1.6 b0.6 21.2 8.7Obock-HS3 0.6 2.9 b0.6 21.1 8.9Obock-HS6 0.3 2.9 b0.6 21.3 8.9Seawater 1.1 5.2 1.3 20.8 9.1

at the outpourings of the coastal hot springs, howeverHSO4−would be the

prevailing oxidized sulfur species in the deep hydrothermal vent, proba-bly originated by the disproportionation of magmatic SO2 (after Mülleret al., 2013):

103lnαðHSO−4 –H2OÞ ¼ 1:852106

=T2−0:288 ð3Þ

The approaching of the Obock spring with the highest estimatedtemperature to the intersection of the HSO4

−–H2O equilibrium linessustains this hypothesis (Fig. 8C). In fact, the obtained temperaturefrom 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 thegeothermometrical modeling.

5. Discussion

It has been reported that the fluids resulting from seawater–basaltinteraction at temperature higher than 100 °C have been completelydepleted of magnesium and sulfate (Bowers et al., 1988; Mottl andHolland, 1978; Tomasson and Kristmannsdottir, 1972). Moreover,straight lines were obtained between points of hot springs and pointof sea water when plotting normalized concentrations of elements tochloride (Mg/Cl, SO4/Cl, K/Cl, Ca/Cl, SiO2/Cl) versus concentration ofLi/Cl (Fig. 9). Therefore, one may extrapolate the Mg and SO4 concentra-tions to zero, which corresponds to the maximum interaction betweenseawater and basalt. The Mg/Cl and SO4/Cl straight lines get zero for thesame Li/Cl value of 3.228 × 10−4 (Fig. 9A). The silicate concentration ofthe hot end-member (e.g. Li/Cl = 3.228 × 10−4) was estimated about212 mg/L (Fig. 9B). When the value obtained for dissolved silica is usedas quartz geothermometer (Verma, 2001), it gives a geothermal reservoirtemperature of about 187 °C, which is in good agreement with that ob-tained from the chemical geothermometer (172–191 °C, Table 2) aswell as from multiple mineral equilibrium approach (180–200 °C, Fig. 7).

As stated above, the δ34S(SO4) compositions of Obock thermalwaters are similar to that of the present-day seawater. These resultsrule out any evolution of the seawater–basalt system that would not

5

6

7

8

9

10

11

12

3 4 5 6 7 8 9 10 11 12 13

Atlantis II

Obock

Temperature (°C)

Mean Ocean

Gulf of Aden

Red Sea

seawater

310

lnα

6 210 /T(kelvin)

SO

-HO

42

2-

HS

O-H

O4

2

-S

O-d

isp

2

CaS

O-H

O4

2

2060300

260

220

180

140

100

6.0

6.5

7.0

7.5

8.0

?8.5

9.0

9.5

10.0

20.0 20.5 21.0 21.5 22.0

Atlantis II

Atla

ntis

II0.0

1.0

2.0

3.0

4.0

5.0

6.0

7.0

8.0

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4

SMOW

2

18

δH =

6 x

δO

Gulf of Adenseawater

18δ O(H O) ( vs V-SMOW)2 ‰

2 δH

(HO

) (

vs

V-S

MO

W)

2‰

34δ S(SO ) ( vs V-CDT)4 ‰

18 δO

(SO

) (

vs

V-S

MO

W)

4‰

LMW

L

Obockseawater

Obockseawater

A

B

C

Fig. 8. Isotope diagrams of the hot springs (squares) and seawater (diamonds) from Obock. A:Water isotope data are compared with the Local MeteoricWater Line (LMWL; Fontes et al.,1980), the seawater from the Gulf of Aden (Craig, 1969; Ganssen and Kroon, 1991), the Craig's line δ2H(H2O)= 6 δ18O(H2O) relating surface seawaters of the area (Craig, 1969) and theAtlantis 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) andAtlantis II hydrothermal brine (Zierenberg and Shanks, 1986). C: Sulfate–water oxygen isotope fractionation vs. temperature (after Boschetti, 2013). Gray squares and area depict the cal-culated fractionation using the obtained δ18O(SO4) and δ18O(H2O) isotopedata and the temperature at the sampling site of theObock's hot springs (Table 4) andAtlantis II (Zierenberg andShanks, 1986), respectively. Red squares and yellow area are the same samples, but using the temperature calculated by geothermometers. Horizontal arrows represent conductive and/ormixing cooling. Please note that the δ18O(SO4) data of Zierenberg and Shanks (1986) were recalculated considering the value of the mean present-day seawater δ18O(SO4) = +8.6‰(Boschetti, 2013; Boschetti and Iacumin, 2005).

63M.O. Awaleh et al. / Journal of Geochemical Exploration 152 (2015) 54–66

yet attain the equilibrium (e.g. no end-member depleted of sulfate andmagnesium). Indeed,when the diminution of SO4 andMg is due to a de-gree of advancement of the reaction without reaching the final state, itshould remain SO4 in the solution that would present therefore34S/32S ratio higher than that of seawater because of isotopic fractioning(Camo et al., 1991), which is not the case in Obock coastal geothermalfield.

In the other hand, when the diminution of SO4 is the result of a blendbetween an end-member, depleted of Mg and SO4, and the surroundingseawater which mix each other during ascent, the 34S/32S ratio of theend-member is similar to that of the seawater, which is the case of thepresent study (Table 4).

The Obock region was defined as the final termination of a riftsegment (Manighetti, 1993). In addition, Manighetti (1993) revealedthe occurrence of hydrothermalism in the Obock ridge. Therefore, thefollowing model was proposed for the Obock coastal geothermal field(Fig. 10): the magma chamber of the submarine volcano located be-neath the Obock ridge could constitute the main heat source of theObock geothermal system. Seawater would infiltrate through the ba-salts fractures to reach the geothermal reservoir. In this later, seawatercould react with the basalt bedrock at the temperature of the geother-mal reservoir (197±10 °C) for a relatively long time so that the equilib-rium can be reached. Therefore, the seawater–basalt interaction in thegeothermal reservoir could produce a geothermal fluid which were

A

B

C

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.

64 M.O. Awaleh et al. / Journal of Geochemical Exploration 152 (2015) 54–66

depleted of magnesium and sulfate. Then, during their ascent byconvection through basalt fractures, the geothermal fluid is mixed invarying 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 inthe present study are partially equilibrated at the outpourings. Likewise,

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

65M.O. Awaleh et al. / Journal of Geochemical Exploration 152 (2015) 54–66

the lower amount of chloride in hot springs than in seawater suggeststhe dilution by freshwater. This is further supported by the Obockregional piezometric map which shows that the meteoric water flowtoward the sea (Fig. 2). However, one cannot preclude any phase sepa-ration process that could be responsible of the lower concentration ofchloride in Obock hot springs than in seawater as reported for somesubmarine hydrothermal sources (Bischoff and Rosenbauer, 1989).

In a recent pre-feasibility study, a geophysical survey of the Obockgeothermal prospect, based on electro telluric, magnetotelluric andgravimetric measurements, was conducted by the CERD (2013). Thesegeophysical investigations reveal a geothermal anomaly in an areanear the coast of Obock city. Therefore, these results support the con-ceptual model proposed for Obock beach geothermal manifestations(Fig. 10).

6. Conclusions and perspectives

The reservoir temperatures in the Obock beach geothermalsystem were assessed using a number of geothermometry techniques.Chemical geothermometers (i.e. the Na/K, Na–K–Ca, SiO2) and multiplemineral equilibrium approaches estimate the temperatures of the deepgeothermal reservoir at 197 ± 10 °C, which is also confirmed by theapplication of the 18O/16O isotope fractionation α(HSO4

−–H2O).A conceptual model was proposed for the Obock beach geothermal

system. During their ascent through fractures, the geothermal fluid ismixed in varying degrees with cold seawater which, after an additionalpossible mixing with small amounts of fresh water, give rise to coastalhot springs of Obock city.

Similarly to the Atlantis II hydrothermal water from the Red Sea, thehot waters from Obock have δ34S(SO4) values very near to that of thepresent-day seawater. Moreover, the magnesium and sulfate depletionof the water are consistent with a fluid having interacted with hotaxial-rift basalts.

However, future investigations, such as offshore geophysicalsurveys, should be undertaken to locate the Obock geothermal reservoir.

The conceptual model proposed here clearly indicates that deep deviateddrilling should be used to intercept the Obock geothermal reservoir.

Acknowledgments

This researchworkwas financially supported by the Centre d'Etudeset de Recherche de Djibouti (CERD). We are grateful to Abdi AbdillahiDjibril and Samaleh Idriss for their assistances in the field works. Wewould like to thank the Editor-in-Chief, Prof. Benedetto De Vivo, andan anonymous reviewer for their very constructive comments thatsubstantially improved the manuscript.

References

Anórsson, S., 2000. Mixing processes in upflow zones and mixing models. In: Arnórsson, S.(Ed.), Isotopic and Chemical Techniques in Geothermal Exploration, Development andUse. Sampling Methods, Data Handling, Interpretation. IAEA, Vienna, pp. 200–211.

Aquater, 1982. Etude technico-économique de pré-faisabilité géothermique. Rapportspécifique — Nouvelles perspectives de développement intégré avec les ressourcesgéothermiques (République de Djibouti, 90 pp.).

Aquater, 1989. Geothermal Exploration Project. Republic of Djibouti. Final Report. CERD,République de Djibouti (159 pp.).

Arnórsson, S., Gunnlaugsson, E., Svavarsson, H., 1983. The chemistry of geothermal watersin Iceland III: chemical geothermometry in geothermal investigations. Geochim.Cosmochim. Acta 47, 567–577.

Awaleh, M.O., Hoch, F.B., Soubaneh, Y.D., Boschetti, T., Egueh, N.M., Elmi, S.A., 2015. The geo-thermal resources of the Republic of Djibouti — II: geochemical study of the Lake Abhegeothermal field. J. Geochem. Explor. (submitted for publication).

Barberi, F., Varet, J., 1977. Volcanism of Afar: small-scale plate tectonics implications. Geol.Soc. Am. Bull. 88, 1251–1266.

Barberi, F., Ferrara, G., Santacroce, R., Varet, J., 1975. Structural evolution of the Afar triplejunction. In: Pilger, A., Rosler, A. (Eds.), Afar Depression of Ethiopia, Inter-Union Com-mission on Geodynamics, Scientific Report 14, E. Schweizerbart'scheVerlagsbuchhandlung (Nägele u Obermiller), Stuttgart, pp. 38–54.

Bethke, C.M., 2008. Geochemical and Biogeochemical Reaction Modeling. Cambridge Uni-versity Press, New York.

Bischoff, J.L., Rosenbaeur, R.J., 1989. Salinity variations in submarine hydrothermal sys-tems by layered double diffusive convection. J. Geol. 97, 613–623.

Boschetti, T., 2013. Oxygen isotope equilibrium in sulphate–water systems: a revision ofgeothermometric applications in low-enthalpy systems. J. Geochem. Explor. 124,92–100.

66 M.O. Awaleh et al. / Journal of Geochemical Exploration 152 (2015) 54–66

Boschetti, T., Iacumin, P., 2005. Continuous flow δ18O measurements: new approach tostandardization, high temperature thermodynamics and sulphate analysis. RapidCommun. Mass Spectrom. 19, 3007–3014.

Boschetti, T., Cortecci, G., Bolognesi, L., 2003. Chemical and isotopic study of theshallow groundwater system of Vulcano Island, Aeolian Archipelago, Italy. GeoActa2, 1–34.

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 fluids at 13°–11°N and 21°N, East Pacif-ic Rise. J. Geophys. Res. 93 (B5), 4522–4536.

BRGM, 1975. Territoire Français des Afars et des Issas: rapport de fin de sondage: résultatsdes 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 sul-phate reduction in Clam hydrothermal field, Okinawa through. Earth Planet. Sci.Lett. 107, 328–338.

Centre d'Etudes et de Recherche de Djibouti (CERD), 2011. Etude de préfaisabilité deGoubet (République de Djibouti).

Centre d'Etudes et de Recherche de Djibouti (CERD), 2012. Etude de préfaisabilité du LacAbhé (République de Djibouti).

Centre d'Etudes et de Recherche de Djibouti (CERD), 2013. Etude de préfaisabilitéd'Obock (République de Djibouti).

Coplen, T.B., Hopple, J.A., Böhlke, J.K., Peiser, H.S., Rieder, S.E., Krouse, H.R., Rosman, K.J.R.,Ding, T., Vocke Jr., R.D., Révész, K.M., Lamberty, A., Taylor, P., De Bièvre, P., 2001. Com-pilation of minimum and maximum isotope ratios of selected elements in naturallyoccurring terrestrial materials and reagents. US Geological Survey Water—ResourcesInvestigations Report 01-4222.

Courtillot, V., Galdeano, A., Le Mouel, J.L., 1980. Propagation of an accreting plate bound-ary: a discussion of new aeromagnetic data in the Gulf of Tadjourah and southernAfar. Earth Planet. Sci. Lett. 47, 144–160.

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. 208–242.

D'Amore, F., Arnórsson, S., 2000. Geothermometry. In: Arnórsson, S. (Ed.), Isotopic andChemical Techniques in Geothermal Exploration, Development and Use. SamplingMethods, Data Handling, Interpretation. IAEA, Vienna, pp. 152–199.

D'Amore, F., Giusti, D., Abdallah, A., 1998. Geochemistry of the high-salinity geothermalfield of Asal, Republic of Djibouti, Africa. Geothermics 27, 197–210.

Demange, J., Di Paola, G.M., Lopoukine, M., Stieltjes, L., 1971. Etude géothermique duT.F.A.I., B.R.G.M., Rapport 7104063.

Doubre, C., 2004. Structure et mécanismes des segments de rift volcano-tectoniques;études de rifts anciens (Ecosse et Islande) et d'un rift actif (Asal–Ghoubbet). (PhDThesis), Université du Maine (420 pp.).

Ellis, A.J., 1979. Chemical geothermometry in geothermal systems. Geothermics 25,219–226.

Faure, H., 1976. Les déformations des côtes et le modèle géodynamique de l'Afar. In:Pilger, A., Rosler, A. (Eds.), Afar between Continental and Oceanic Rifting. Internation-al Symposium on the Afar Region and Related Rift Problems, Bad Bergzabern, April1974, Vol. 2, Inter-Union Commission on Geodynamics, Scientific Report, 16, Stutt-gart, pp. 148–155.

Faure, H., Hong, C.T., Lalou, C., 1980. Datation Th/U des calcaires coralliens etmouvementsverticaux à Djibouti. Bull. Soc. Geol. Fr. 6, 959–962.

Fontes, J.C., Pouchon, P., Saliege, J.F., Zuppi, G.M., 1980. Environmental isotope study ofgroundwater systems in the Republic of Djibouti. Arid-zone Hydrology: Investiga-tions with Isotope Techniques. IAEA, Vienna, pp. 237–262.

Fouillac, R., Michard, S., 1981. Sodium/Lithium ratio in water applied to geothermometryof geothermal reservoirs. Geothermics 10, 55–70.

Fournier, R.O., 1977. Chemical geothermometers and mixing models for geothermal sys-tems. Geothermics 5, 41–50.

Fournier, R.O., 1979. A revised equation for the Na/K geothermometer. Geotherm. Resour.Counc. Trans. 3, 221–224.

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, 37–69.

Fournier, R.O., Potter, R.W.I.I., 1979. Magnesium correction to Na–K–Ca geothermometer.Geochim. Cosmochim. Acta 43, 1543–1550.

Fournier, R.O., Truesdell, A.H., 1973. An empirical Na–K–Ca geothermometer for naturalwaters. Geochim. Cosmochim. Acta 37, 1255–1275.

Fournier, R.O., Potter, R.W., 1982. II A revised and expanded silica (quartz)geothermometer. Geotherm. Resourc. Counc. Bull. 11, 3–12.

Ganssen, G., Kroon, D., 1991. Evidence for Red Sea surface circulation from oxygen iso-topes of modern surface waters and planktonic foraminiferal tests. Paleoceanography6, 73–82.

Gasse, F., Fournier, M., Richard, O., 1983. Carte géologique de la République de Djibouti à1:100 000. Djibouti. Notice explicative. ISERST, Ministère français de la Coopération,Paris.

Giggenbach, W.F., 1988. Geothermal solute equilibria. Derivation of Na–K–Mg–Cageoindicators. Geochim. Cosmochim. Acta 52, 2749–2765.

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, 119–142.

Groupe de la Banque Africaine de Développement (GBAD), 2011. Document deStratégie — Pays, 2011–2015 (République de Djibouti).

Gunnarsson, I., Arnórsson, S., 2000. Amorphous silica solubility and the thermodynamicproperties of H4SiO4 in the range of 0° to 350 °C at Psat. Geochim. Cosmochim. Acta64, 2295–2307.

Halas, S., Pluta, I., 2000. Empirical calibration of isotope thermometer δ18O(SO42−)–δ18-

O(H2O) for low temperature brines. V Isotope Workshop European Society for Iso-tope Research, Kraków, Poland, pp. 68–71.

Houssein, I.K., Sanjuan, B., Michard, G., 1993. Indices géochimiques de l'existence d'unfluide à 210 °C dans la région d'Obock (République de Djibouti). C. R. Acad Sci.Paris II 316, 771–776.

IAH, 1979. Map of mineral and thermal water of Europe. Scale 1:500,000. InternationalAssociation of Hydrogeologists, UK.

Kharaka, Y.K., Lico, M.S., Law, L.M., 1982. Chemical geothermometers applied to formationwaters, Gulf of Mexico and California basins. Am. Assoc. Petrol. Geol. Bull. 66, 588.

Lavigne, J., Lopoukine, M., 1970. Reconnaissance géothermique du T.F.A.I., B.R.G.M., 70SGN, 109 GTM.

Manighetti, I., 1993. Dynamique des systèmes extensifs en Afar. (Thèse de Doctorat),Université de Paris VI (242 pp.).

Manighetti, I., King, G., Sammis, C.G., 2004. The role of off-fault damage in the evolution ofnormal faults. Earth Planet. Sci. Lett. 217, 399–408.

Mlynarski, M., Zlotnicki, J., 2001. Fluid circulation in the active emerged Asal rift (eastAfrica, Djibouti) inferred from self-potential and telluric–telluric prospecting.Tectonophysics 339, 455–472.

Mottl, M.J., Holland, T.J., 1978. Chemical exchange during hydrothermal alteration of ba-salt by seawater. Experimental results for major and minor components of seawater.Geochim. Cosmochim. Acta 42, 1103–1115.

Moyes, J., Daprat, J., Faugeres, J.C., Gonthier, E., Pujol, C., 1981. Etude Stratigraphique etsédimentologique. In: C.E.P.M., C.N.E.X.O. (Eds.), Orgon IV, Golfe d'Aden, MerD'Oman. Novembre 1978. C.N.R.S., Paris, pp. 189–263.

Müller, 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 be-tween laboratory experiments and a hydrothermally active site in the Manus Basin,Papua New Guinea. (PhD thesis), Universität Bremen, pp. 105–140 (Chapter 4).

Nicholson, K., 1993. Geothermal fluids. Chemistry and Exploration Techniques. Springer-Verlag, Berlin Heidelberg (263 pp.).

Pang, Z.H., Armannsson, H., 2006. Analytical Procedures and Quality Assurance for Geo-thermal Water Chemistry. United Nations University Geothermal TrainingProgramme.

Pang, Z., Reed, M.H., 1998. Theoretical chemical geothermometry on geothermal waters:problems and methods. Geochim. Cosmochim. Acta 62, 1083–1091.

Pierret, M.C., Clauer, N., Bosch, D., Blanc, G., France-Lanord, C., 2001. Chemical and isotopic(87Sr/86Sr, δ18O, δD) constraints to the formation processes of Red-Sea brines.Geochim. Cosmochim. Acta 65, 1259–1275.

Pinzuti, P., Mignan, A., Geoffrey, C.P.K., 2010. Surface morphology of active normal faultsin hard rock: implications for the mechanics of the Asal Rift, Djibouti. Earth Planet.Sci. Lett. 299, 169–179.

Piper, A.M., 1944. A graphic procedure in the geochemical interpretation of water-analyses. Trans. Am. Geophys. Union 25, 914–923.

Reed, M.H., 1982. Calculation of multicomponent chemical equilibria and reaction pro-cesses in systems involving minerals, gases and an aqueous phase. Geochim.Cosmochim. Acta 46, 513–528.

Reed,M.H., Spycher,W.H., 1984. Calculation of pH andmineral equilibria in hydrothermalwaters with applications to geothermometry and studies of boiling and dilution.Geochim. Cosmochim. Acta 48, 1479–1492.

Richard, O., 1979. Etude de la transition dorsale océanique-rift émergé: le Golfe deTadjourah (République de Djibouti). (Thesis). Université de Paris-Sud.

Sanjuan, B., 2010. Use of a new Sodium/Lithium (Na/Li) geothermometric relationship forHigh-Temperature (HT) geothermal fluids derived from seawater/basalt interactionprocesses: application to the Djibouti case. Third East African Rift Geothermal Confer-ence ARGEO-C3-DJIBOUTI. Exploring and Harnessing the Renewable and PromisingGeothermal 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: Crystallogra-phy, Geochemistry, and Environmental Significance. Reviews inMineralogy and Geo-chemistry 40. Mineralogical Society of America and Geochemical Society,Washington, D.C., pp. 541–602.

Sonney, R., Vuataz, F.D., 2010. Validation of Chemical and Isotopic Geothermometers fromLow Temperature Deep Fluids of Northern Switzerland. Proceedings World Geother-mal Congress 2010 Bali, Indonesia.

Tomasson, J., Kristmannsdottir, H., 1972. High temperature alteration minerals and ther-mal brines, Reykjanes, Iceland. Contrib. Mineral. Petrol. 36, 123–134.

Verma, M.P., 2001. Silica solubility geothermometers for hydrothermal systems. In: Cidu,R. (Ed.), Proc. of the 10th Int. Symp. on Water–Rock Interaction, Vol. 1, Villasimius,Italy, June 10–15, pp. 349–352.

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. 79,9–24.

World Bank, 2013. Communiqués de Presse, 13 Octobre 2013. La Banque mondiale etDjibouti signent un accord sur la prospection de l'énergie géothermique. http://www.banquemondiale.org/fr/news/press-release/2013/10/13/world-bank-and-djibouti-sign-agreement-to-explore-geothermal-energy.

Zan, L., Gianelli, G., Passerini, P., Troisi, C., Haga, A., 1990. Geothermal exploration in theRepublic of Djibouti: thermal and geological data of the Hanlé and Asal areas.Geothermircs 19, 561–582.

Zeebe, R.E., 2010. A new value for the stable oxygen isotope fractionation between dis-solved sulphate ion and water. Geochim. Cosmochim. Acta 74, 818–828.

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, 2205–2214.