Geochemical Features of Thermal Waters at Benetutti (Sardinia)

GEOCHEMICALRendiconti Seminario FEATURES Facoltà OF ScienzeTHERMAL Università WATERS Cagliari AT BENETUTTIVol. 73 Fasc. (SARDINIA)1 (2003) 39

Geochemical features of thermal waters at Benetutti (Sardinia)

ROSA CIDU(*), ANTONIO DAMIANO MULAS(*)

Riassunto. Questo lavoro riporta le caratteristiche chimiche delle acque termali e fredde nella zona di Benetutti (Sardegna Centrale). In 24 campioni d’acqua sono stati determinati numerosi componenti maggiori ed in tracce utilizzando diverse tecniche analitiche. Le acque termali hanno una composizione cloruro sodica dominante e salinità di 0.5 g/L; sono caratterizzate da pH alcalino (9.5-9.7), concentrazioni molto basse in Mg (0.01-0.1 mg/L) e concentrazioni di F, B, Li, Rb, Cs e Mo molto più elevate rispetto a quelle osservate nelle acque fredde dell’area. Per i campioni di cui si dispongono dati bibliografici, non sono state osservate variazioni significative nei valori di temperatura, portata e composizione chimica negli ultimi 20 anni. Gli acquiferi profondi associati alle acque termali sono stimati avere bassa entalpia (la temperatura calcolata in profondità è di circa 60°C). Oltre l’uso terapeutico, le acque termali possono essere utilizzate per il riscaldamento di edifici e serre.

Parole chiave: acque termali, elementi in tracce, Sardegna.

Abstract. This paper reports the geochemistry of the thermal and cold waters at Benetutti in central Sardinia. Both major and trace components in 24 water samples were analysed using various techniques. The thermal waters show a sodium chloride composition and a salinity of 0.5 g/L; they are characterised by a pH of 9.5-9.7, very low magnesium (0.01-0.1 mg/L), and much higher F, B, Li, Rb, Cs, and Mo concentrations than in the cold waters of the area. No variations in temperature, flow, or chemical composition were observed at the thermal springs for which records dating back 20 years are available. Subsurface reservoirs associated with the thermal springs are of a low enthalpy (calculated temperature at depth is about 60°C), and are therefore candidates for direct uses, such as balneology, space heating, and horticulture.

Key words: thermal waters, trace elements, Sardinia.

(*) Dipartimento Scienze della Terra Università di Cagliari, via Trentino 51, 09127 Cagliari, Italy. Presentato il 27/03/03 40 R. CIDU, A.D. MULAS

INTRODUCTION In the Tirso valley of Central Sardinia, thermal waters occur from north to south at Benetutti, Oddini, and Fordongianus. They show variable emergent temperatures (33-55°C) but all are characterised by high pH (> 9), low salinity (< 1 g/L), sodium-chloride composition, very low (< 20 µg/L) magnesium, and high (up to 9 mg/L) fluoride concentrations [1] [2]. The thermal waters in the Tirso valley are related to important tectonic structures trending NE-SW and E-W [3]. Isotopic studies based on deuterium and oxygen- 18 indicate that the thermal waters from the Tirso valley are meteoric waters that infiltrate at depth and warm up [4]. At Benetutti, a village of about 2,000 inhabitants, thermal waters have long been known, as testified by the Roman bridge on the river Tirso nearby and by the ancient church of San Saturnino. They represent an important resource in an area in which the main economic activities rely on poorly developed farming. Thermal waters have been used therapeutically, especially at the Terme Aurora spa. The local authority is now planning to build a resort area for the development of tourist activities near the thermal springs. This research is aimed at studying the hydrogeochemical features at Benetutti and at highlighting the peculiar characteristics of the thermal waters compared to the cold waters of the area.

STUDY AREA The study area of Benetutti in Central Sardinia extends for about 60 km2, with a maximum elevation of 417 m a.s.l (fig. 1). The climate is characterised by rainy seasons, usually extending from October to April, and long periods of dry weather. At the Bultei meteorological station (the closest to the study area) at 510 m a.s.l., the long-term (1930- 1980 period) mean rainfall is 800 mm/y with a mean of 70 rainy days; the highest mean monthly precipitations occur in December and January, while the lowest are in July and August; the mean annual temperature is 15°C [5]. The geology of the investigated area is shown schematically in Figure 1. Late-tectonic and post-tectonic Hercynian granite are the dominant rocks, and are mainly represented by medium-grained, equigranular granodiorite [6] [7] [8]. In Tertiary times, the Palaeozoic basement of Sardinia was affected by important tectonic activity, and the island was characterised by massive volcanic activity [9] [10]. In the Benetutti area, volcanism developed between the Upper Oligocene and the Lower Miocene, with deposition of lava and ignimbrite tuffaceous sequences [11]. Recent deposits consist of alluvial sediments and outcrop close to the river Tirso (Fiume Tirso in fig. 1). The hydrology is characterised by the river Tirso and its main tributary Rio Mannu; low- flow streamlets occur temporarily over the rainy season. Both the surface and ground waters of the area mainly drain granite rocks; sample no. 16 only partially drains volcanic rocks. The thermal springs are located few kilometres south of Benetutti in a limited area close to the river Tirso (fig. 1). GEOCHEMICAL FEATURES OF THERMAL WATERS AT BENETUTTI (SARDINIA) 41

Figure 1. Schematic geological map of the Benetutti area and location of the water samples.

METHODS Water sampling was carried out in May 2001 after two months of drought and a low- rain winter season. The sampling period therefore represents low-flow conditions. At the sampling site, temperature, pH, redox potential, conductivity, hydrogen sulphide and 42 R. CIDU, A.D. MULAS

nr nr nr nr nr nr

0.01

0.37

0.71 1.19

0.63

<0.1

<0.1 <0.1

<0.1 <0.1 <0.1 <0.1 <0.1 <0.1

<0.1

<0.1 <0.1

mg/L

nr nr

nr nr nr nr

2.9

249

210

0.62

0.22

4.35

0.22 0.23

<0.1

<0.1 <0.1

<0.1

<0.1 <0.1

<0.1 <0.1 <0.1

<0.1 <0.1 <0.1 <0.1

<0.1

mg/L

nr nr nr nr nr nr

0.2

0.53

16.7

<0.1

<0.1 <0.1

<0.1 <0.1 <0.1 <0.1 <0.1 <0.1

<0.1 <0.1

mg/L

nr nr nr

nd nd nd nd nd nd nd

5.0

0.6

0.1 0.2

0.9

0.4

0.7

HS-

0.05

mg/L

<0.05

<0.05 <0.05

40 40 42

40 40

45 42

41 39

40 39

40 37 17

46 49 42

58 42 49

44 26 25

25 51 15

mg/L

213

203

219 221 219

215 208

211

233

250 231

235

228

229

224 225 220

219 217

235 244 242 204

229 245 204

150 121 157 142

mg/L

- -

- - -

- - - -

nr nr nr

nr nr

6.0

4.5

6.0 7.2

6.6 8.4

9.1 9.4

9.6

7.0

5.9 5.5

8.0

10.8

10.5

mg/L

31 34

34 35 38

29 31

25 24

25 27 30

32 24 26

82 25 28 40

218

287 163 207

175 173 102

mg/L

HCO

3.3

3.0

3.1 3.1 3.1

3.0 2.8

3.0

2.8

2.8 3.0

3.2

3.1

2.7 2.6 2.6 2.2

2.3 2.0 2.9

2.8 2.9

2.7 2.6 2.6 6.1

1.7 5.7

3.3 1.0 1.2

18.4

mg/L

78 98

154

165 177 169

162 167 156

146

161

167 163

161 155

158 153

152 152 151

160 165

170 167 184

161 164

147

101

mg/L

0.1

Benetutti are reported in italics for comparison.

mg/L

0.02

0.01

0.12 0.12 0.14

<0.1

11.4

11.2 0.10

28.6 38.2 11.9

19.7 26.9 29.6

0.045

0.042

0.029

0.019 0.063

0.023 0.012 0.012

0.011 0.022 0.043

0.045 0.067

0.064

11 12

11 11 14

14 12

11 11 11

14 14

13 18 12 45 77 16 32

39 37 14 9.3

9.4

9.5 8.6 8.7 8.7

mg/L

502

477 528 509

502 508 489

409

520

545 525

519

507

512 498 503 492

492 486

274 525 545

531 709 512 836

897 615 476

574 686 197

TDS

mg/L

nr nr nr

nr nr

0.76

0.91

1.27

1.10 1.08

0.98 1.03

1.05 0.99

1.00 0.95 1.00 1.01

0.59 1.07 1.10

1.07 1.01 0.85 1.14 1.11 1.01

0.62 0.89 0.75

0.30

Cond

mS/cm

nr nr

73 56

-20

110 100

120

290 170

290

100

185

410 150 360 420 330 220

450 450 430 480 450 490 460

9.4

9.3

9.4 9.2 9.2

9.3 9.0

9.0

9.6

9.5

9.6

9.5 9.5

9.6 9.6

9.6 9.7 9.6 9.6

7.5 9.5 9.5

6.6 8.0 9.5

7.0 6.8 7.4

6.6 7.4 6.2

7.7

41 41 42 42 42 42

40 36 42

27 41

42 33 33 39 42 19 26 21 24 18 18 35 18 22 24 15 18 17 27

T°C

water

nr nr nr nr nr nr

22 22 24

24 29

29 29 28 27 26 22 25 20 21 22 26 33 29 29 33 33 30 32 32

air

T°C

L/s

0.2

0.3 0.1 0.1 0.3 0.3

0.1

100

0.03

0.05

0.02 0.05 0.05

0.03 0.05 0.05 0.05 0.03

Flow

2000

1985 1988

1990

1982

July 1989

Apr. 1989 Oct. 1989

sampling

May 2001 May 2001 May 2001

May 2001 May 2001

May 2001 May 2001 May 2001 May 2001 May 2001 May 2001 May 2001 May 2001 May 2001 May 2001 May 2001 May 2001 May 2001 May 2001 May 2001 May 2001 May 2001 May 2001 May 2001

Name Date of

Terme Aurora A Terme Aurora [17] Terme Aurora B Banzu sos beccos

Banzu sos beccos [15] Banzu sos beccos [1] Banzu sos beccos [16] Banzu sos beccos [16] Banzu sos beccos [16] Banzu sos beccos [2] Sorgente Sos Beccos Banzu Mazzore

Banzu Mazzore [3] Sorgente Casa Tanda San Saturnino chiesa San Saturnino Su Giudice

Banzu sos nervios Abba Putida Fiume Tirso Pozzo Agriturismo Lai Fontana Lai Maurizio Sorgente degli occhi Pozzo N.ghe Luzzanas Sorgente Ziu Paulu Sorgente Urchi Sorgente F.lli Lai Sorgente Linia Sorgente Su Cantaru Sorgente Boloe Sorgente Pauleddu Rio Mannu

Table 1. Chemical composition of the waters at Benetutti. Previous records of the thermal waters at

No.

1 1 2 3 3 3 3 3 3 3 4 5 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

TDS: total dissolved solids; [17]: reference; nr: not reported. GEOCHEMICAL FEATURES OF THERMAL WATERS AT BENETUTTI (SARDINIA) 43

nr nr nr nr nr

0.19

13.3

1.82

11.2 1.26

1.08 0.23

<0.1

<0.1 <0.1

<0.1 <0.1 <0.1

<0.1 <0.1

<0.1

µg/L

nr nr nr nr nr

9.8

8.7 8.1 9.5

0.1

5.7 1.8

0.2 0.3 0.2

15.0

11.9

13.1

12.2

12.0 10.8

10.9

10.0

12.7

13.5 10.9

12.3

<0.1 <0.1 <0.1

µg/L

nr nr nr nr

nr nr

0.3

0.5

0.3

<0.3

<0.3 <0.3

<0.3 <0.3 <0.3

<0.3 <0.3 <0.3 <0.3

<0.3 <0.3 <0.3 <0.3 <0.3 <0.3 <0.3

<0.3

µg/L

nr nr nr nr nr

8.0

0.6

0.3 0.3

0.2

1.2

0.5

0.2 0.4 0.4

1.1

0.5

0.4

0.2

<0.2

<0.2 <0.2

<0.2 <0.2 <0.2

<0.2

<0.2 <0.2 <0.2

<0.2

µg/L

nr nr nr

3.0

1.2 2.3

1.7

0.8 0.5 0.4

0.4 0.6 1.4

1.3 0.6 0.8

0.4 1.6

0.6 0.3 0.3

2.1 0.7 0.3

0.6

17.0

<0.2

µg/L

nr nr

nr nr nr nr

0.2

0.08

0.15

0.24

0.35 0.12

0.34

0.24 0.29

0.07 0.77

0.09

<0.1

µg/L

<0.05

<0.05 <0.05

<0.05

<0.05 <0.05

<0.05

nr nr nr

nr nr

0.05

0.06 0.06

0.06 0.05

0.07

0.06

µg/L

<0.05 <0.05

<0.05

<0.05 <0.05 <0.05 <0.05

<0.05 <0.05

<0.05 <0.05 <0.05

<0.05

nr nr nr nr

0.6

7.5

1.4 3.6

2.3 3.5

9.5 4.3

4.5 4.9

5.2

µg/L

nr nr

nr nr nr

257 17

0.21

0.77 0.79

0.14 1.76

0.66 0.22

0.22 1.17 1.24 5.07

0.51

1.92

0.73

µg/L

nr nr

nr nr nr

µg/L

1150

3212229

nr nr

11 280 30 6.5

32 <4 470 8.5

36 23 4.3 6.2

Ba Fe Mn Zn

0.3 <4 0.4 15

4.9 6 3.3 2.5

9.1 260 16 7.7

110

0.08

0.38

0.37 10 0.46 19

0.18 <4

0.22 13

0.78 5

0.14 12

0.46 8 1.01 18

0.33 <4 1.06 24 0.79 83 8.5 7.3

0.34 12

13.4 214 100 6.2 16.1 16 4.3 11

µg/L

180

167 175

167 161 168

169

204 173

169

146

152 131 136 116

114 116

215 223 190

125 183 315 488 114 190

220 260

µg/L

10 13 17

4.7

4.5

4.4 4.1

4.2 3.8

4.1 3.7 3.9

3.7 2.8 1.5

3.6 4.1

2.8 1.5 3.8

0.32

<0.1

<0.1 <0.1 <0.1 <0.1

<0.1 <0.1

µg/L

20 10 15

24 29

1.8

6.9

1.2 4.4

27.5

27.6 25.3

26.0 24.3

25.1

22.9 23.9 23.0

18.1 15.9

26.7 28.9 25.6

24.8

0.13

0.48 0.87 0.71

<0.1

µg/L

84 93 86 87 94 79

83 86

87 88 89 78 81 74 74 73

89 89 85 19 85 64 19 35

2.9

5.4 9.5 8.9 2.5

µg/L

67 66

61 43 42

43 28 14

290

100 105 111 111 121 109

108

105 108

112 116

117 109 112 112 115

102 101 105

109

209

µg/L

nr nr nr nr nr

3.0

2.2

1.9 2.1

2.2 2.4

2.5 2.5 2.7 2.7 2.2 1.2

1.3 1.6

2.2

1.9

0.16

0.37

0.41

0.56 0.17 0.19 0.11

<0.1

µg/L

nr nr nr nr nr nr

44 12

11 16 13 69

19 25 11 14

33 41

<6 <6

650

µg/L

6.3

6.7

37.7

41.8 42.1 39.1 38.3 42.0 40.7

41.0

38.4

35.5 37.5

37.8 39.1 39.3 37.9 38.9 38.8 38.3 31.4

34.0 34.6 30.3 14.9 35.7 22.2 26.7 18.9 22.2 29.8 23.1

SiO

mg/L

8.9 8.8 8.7 9.1 9.7 9.1

20.6

9.37 9.43 9.83

9.81 8.79

8.88 8.07 8.22 7.06 6.82 7.49 0.16 8.82 8.81 8.84 1.54 9.22 0.42 0.23 0.38 0.19 0.25 0.22 0.13

mg/L

nr nr nr nr nr nr

4.85

0.74 0.79 0.75

0.71 0.76

0.71 0.67 0.76 0.69 0.74 0.68 0.20 0.70 0.79 0.80 0.63 0.71 0.78 0.45 0.44 0.23 0.44 0.48 0.18

mg/L

Table 1. Continued.

No.

1 1 2 3 3 3 3 3 3 3 4 5 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

44 R. CIDU, A.D. MULAS alkalinity were measured, and the waters were filtered (0.4 µm, Nuclepore 111130), acidified, and stored for metal analyses. The redox potential (Eh) measured by a platinum electrode was corrected against the Zobell’s solution [12]. Hydrogen sulphide was estimated by the Visicolor Macherey-Nagel-Duren test kit, anions were determined by ion chromatography, metals by inductively coupled plasma optical emission spectroscopy (ICP-OES, ARL3520¨) and inductively coupled plasma mass spectrometry (ICP-MS, ELAN5000¨), while Hg, As and Sb were determined by ICP-MS after flow injection Hg-vapour or hydride generation [13]. The detection limit for chemical components was calculated at five times the standard deviation of blank solutions. The accuracy was evaluated using the NIST1643d standard reference solution. Both precision and accuracy were estimated at ± 6% and ± 11%, or better, at the mg/L and µg/L concentrations, respectively. The ionic balance calculated by the PHREEQC [14] computer program was always less than ± 5% suggesting that the analyses are of reasonable good accuracy. In this paper, the terms ÇdissolvedÈ and Çin solutionÈ refer to components present in the fraction below 0.4 µm. Speciation and equilibrium calculations were carried out using the PHREEQC program.

RESULTS AND DISCUSSION Analytical results of the 24 waters considered in this study are reported in table 1. The following elements are not reported because were found below the detection limit: Ag (< 0.1 µg/L), Be (< 0.5 µg/L), Co (< 0.2 µg/L), Cr (< 0.4 µg/L), Ni (< 0.8 µg/L), Sb (< 0.6 µg/L), Tl (< 0.2 µg/L), and Bi (< 0.2 µg/L). Previous records on thermal waters at Benetutti, and related references, are also shown in table 1. It can be observed that most of trace components were not reported prior to this study. A comparison of data on thermal waters collected at Benetutti over 20 years with data from this study shows a remarkably stable chemical composition, as well as similar temperature, pH and concentrations of minor components. The water samples show temperatures of 15-42°C. The air temperature over the sampling period reached peaks of 33°C affecting the temperature of the surface and low- flow spring waters (e.g., samples No. 12, 24, 20); for this reason, when the water temperature is below 30°C, it does not appear to be a reliable parameter in distinguishing the thermal waters. Total dissolved solids (TDS) range from 0.2 to 0.9 g/L with the lowest TDS values observed in the Tirso and Rio Mannu surface waters (samples No. 12 and 24, respectively). The pH ranges from 6.2 to 9.7. The waters with a high pH (9.5-9.7) generally show higher temperatures and lower Eh values; they are characterised by a strong H2S smell at emergence, and often have a detectable amount of HS- (see table 1). Gas emission is observed in most of the high-pH waters. The gas composition was not analysed in this study, but previous records show dominant N2 (98.6 vol. %), with minor Ar (1.25 vol. %) and CH4 (0.084 vol. %), and H2S <0.005 vol. % [18]. GEOCHEMICAL FEATURES OF THERMAL WATERS AT BENETUTTI (SARDINIA) 45

Nitrogen species (NO3 and NO2) and PO4 are usually below the detection limit of 0.1 mg/L in the high-pH waters. Nitrate concentrations higher than the drinking water limit (50 mg/L) established by Italian regulations [19] occur in two water samples (No. 19 and

23). The occurrence of NO3 and PO4 in sample No. 19 is associated with the highest K, Al, Sr, Ba, Fe, Mn, Zn, and Cu concentrations (see table 1) and may reflect anthropogenic inputs, such as fertilisers and uncontrolled discharge of untreated urban wastes. Sample No. 19 will be omitted when discussing water-rock interaction processes. The concentrations of toxic components, such as heavy metals (Cd, Pb, Hg) and As, are very low, much lower than the drinking water limits, and often below the detection limits of the used methods. The different redox environments affect the concentration of dissolved uranium: the reduced waters do not show any detectable U, while at oxidising conditions U occurs in the range of 0.2 to 13 µg/L. The Piper diagram (fig. 2) shows that all the waters have a dominant Na-Cl composition with a subordinate Ca, Mg, and SO4 contribution. A group of waters (sample

Figure 2. Piper diagram showing the main chemical composition of the waters at Benetutti. 46 R. CIDU, A.D. MULAS

250

200 R2 = 0.81

150

Na/Cl in seawater Na (mg/L)

100

Figure 3. So- low-Mg waters dium versus chloride con- 0 050100150200 250 300 centrations Cl (mg/L) in the Bene- tutti waters.

1,0

Br/Cl in seawater

R2 = 0.93 0,8

0,6 Br (mg/L) 0,4

0,2

low-Mg waters

0,0 050100 150 200 250 300 Cl (mg/L)

Figure 4. Bromide versus chloride concentrations in the Benetutti waters. GEOCHEMICAL FEATURES OF THERMAL WATERS AT BENETUTTI (SARDINIA) 47

Nos. 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11,13, 14, 15, 17) is confined at the Cl and Na corners, and appears well distinguished by the low Mg concentration (0.01-0.1 mg/L, see table 1); hereafter, this group of waters will be indicated as low-Mg waters. The other samples show relatively higher HCO3, Ca, and Mg. In all the waters, the dissolved Na is correlated with Cl (R2 = 0.81, fig. 3). The samples showing a higher Na/Cl ratio than observed in seawater probably reflect an increased interaction of the water with Na-plagioclase. The concentration of Br increases with increasing Cl. The observed high correlation (R2 = 0.93, fig. 4) probably indicates a common source for these elements. The Br/Cl ratio is slightly lower than observed in seawater. Figure 5 shows that Sr concentrations increase with increasing Ca. Both Ca and Sr derive mostly from plagioclase and feldspar dissolution. The low-Mg waters have a Sr/ Ca ratio higher than observed in the other waters. Equilibrium or slight supersaturation with respect to calcite occurs in the low-Mg waters, and the possible precipitation of calcite might control the concentration of Ca, while Sr appears to be uncontrolled indicating that it is only partially incorporated in the calcite lattice. Figure 6 shows the comparison between element concentrations in spring n. 23 representing the shallow groundwater and spring n. 10 representing the low-Mg waters. Figure 6 and table 1 show that the low-Mg waters have significantly higher F, B, Li, Rb, Cs, Mo, and Ga concentrations than observed in the shallow groundwater and the surface water of the area. Since all the waters, except sample no. 16, apparently circulate in granite rocks of similar composition, this behaviour might indicate a higher mobility of these elements when the waters warm up and/or deep circuits with longer residence time. The increased mobility of these elements with increasing temperature has been reported in several studies (eg [2] [20] [21]). The concentration of fluoride seems to be controlled by the fluorite mineral: the low-Mg waters are at approximate equilibrium (SIfluorite ranges from Ð0.09 to +0.15) while the other waters are undersaturated (SIfluorite ranges from Ð1.28 to -3.12). In figures 7 to 10 the concentrations of F, Li, Rb, and Mo are plotted versus the concentration of Cl. Chloride is assumed to behave conservatively, i.e. it remains in solution, since it is unaffected by processes such as precipitation of solid phases, ionic exchange, or sorption. When considering all samples, none of these components appear correlated with the concentration of Cl, while a rough correlation is observed among the low-Mg waters.

Silica concentration in the surface waters is low (6-7 mg/L SiO2) and increases in the low-Mg waters (30-39 mg/L SiO2). Silica concentration can be used to estimate the temperature of waters in the deep reservoir, provided that mixing does not occur during uprise to the surface [22]. At Benetutti, the low Mg concentrations observed in the high- pH waters indicate that mixing between the water at depth and the cold waters is unlikely to occur because the Mg concentrations of the cold waters are about 2 orders of magnitude higher (9 mg/L in the less saline water, sample No. 24). H4SiO4 activity derived from speciation is nearly constant (0.46±0.04 10Ð3 mole/L) in the low-Mg waters. These waters 48 R. CIDU, A.D. MULAS

500

450

Sr/Ca in seawater 400

350

300

250 Sr (µg/L) 200

150

100

50 low-Mg waters

0 051015 20 25 30 35 40 45 50 Ca (mg/L)

Figure 5. Strontium versus calcium concentrations in the Benetutti waters.

1000000

Alk Cl 100000 SO4 Na Mg Ca SiO 10000 2

K 1000

Sr Br F 100

Ba Fe B

10 Zn Li Figure 6. Compa- Mn Al rison of element 1 concentrations in Rb Ga U water sample No. Concentrations in the spring No. 23 (µg/L) 23 representing the 0,1 Mo Cs shallow groundwater, and in water 0,01 sample No. 10 0,01 0,1 1 10 100 1000 10000 100000 1000000 representing the Concentrations in the spring No. 10 (µg/L) low-Mg water. GEOCHEMICAL FEATURES OF THERMAL WATERS AT BENETUTTI (SARDINIA) 49

low-Mg waters

6 F (mg/L)

0 050100 150 200 250 300 Cl (mg/L)

Figure 7. Fluoride versus chloride concentrations in the Benetutti waters.

100

low-Mg waters

60 18 Li (µg/L) 40

0 050100 150 200 250 300 Cl (mg/L)

Figure 8. Lithium versus chloride concentrations in the Benetutti waters. 50 R. CIDU, A.D. MULAS

35 low-Mg waters

Rb (µg/L) 15

20 5

0 050100 150 200 250 300 Cl (mg/L)

Figure 9. Rubidium versus chloride concentrations in the Benetutti waters.

low-Mg waters 14

8 Mo (µg/L) 6

2 18

0 050100 150 200 250 300 Cl (mg/L)

Figure 10. Molybdenum versus chloride concentrations in the Benetutti waters. GEOCHEMICAL FEATURES OF THERMAL WATERS AT BENETUTTI (SARDINIA) 51 are slightly supersaturated with respect to quartz, and at equilibrium with respect to chalcedony. Assuming that the deep water is at equilibrium with respect to chalcedony, the calculated temperature at depth is estimated at 60±5°C in the low-Mg waters, i.e. all these waters are thermal. If this is the case, and assuming a normal geothermal gradient, it can be inferred that the thermal waters reach maximum depths of about 1,500 m. The calculated temperature at depth is not much higher than the temperature at emergence (maximum measured temperature: 42°C) and suggests a relatively fast rise of the water from the deep aquifer to the surface. Occurrence of low-magnesium, alkaline waters that have a high pH, dominant sodium, and low salinity, is well documented in the granite areas of France, Spain, Bulgaria, Italy, and Sweden [21] [23]. Similar characteristics are also observed in the high-temperature (150-160°C) waters in the Himalayan granite [24]. Alkaline waters can be considered as the equilibrium composition of a water reacting with granite minerals [21] [25]. Primary minerals, such as Na-Ca-feldspars, are transformed into newly formed minerals that are more stable at the interaction temperature [21]. This interpretation appears reliable for the thermal waters at Benetutti. The dissolution of aluminosilicate minerals consumes protons and contributes to an increase in pH. The pH is also controlled by the equilibrium with carbonate minerals. In fact, the thermal waters at Benetutti are slightly supersaturated with respect to calcite. The precipitation of calcite will lead to low concentrations of calcium and carbonate species. Therefore, silicate hydrolysis combined with carbonate geochemistry might play an important role in producing the observed water composition of the alkaline waters at Benetutti: sodium will be the dominant cation, and chloride the dominant anion. The observed low salinity might be due to the lack of deep origin CO2 inputs, which limit the supply at depth of carbonic acid (i.e. the main source of protons necessary for silicate hydrolysis). The source of main and trace components, including that of the rare alkaline elements, F and Mo, is compatible with the weathering of granite minerals. Fluid inclusions in the granite rocks of north Sardinia have been documented [26], and may represent an additional source of Na, Cl, and Br in solution.

CONCLUSION At Benetutti in central Sardinia, both the surface waters and the groundwaters show a dominant sodium-chloride composition and relatively low salinity (0.2-0.9 g/l). The thermal waters are characterised by gas emission, detectable amounts of HS-, a high pH, very low magnesium, and much higher concentrations of SiO2, F, B, Li, Rb, Cs, and Mo than in the cold waters of the area. The overall characteristic of the thermal waters can be considered the result of a water reacting with granite minerals where silicate hydrolysis combined with carbonate geochemistry play an important role. No variation in temperature at emergence, discharge, or chemical composition were observed at the thermal springs for which records dating back 20 years are available. The deep reservoirs 52 R. CIDU, A.D. MULAS associated with thermal springs are of low enthalpy (calculated temperature at depth is about 60°C), and are suitable for direct uses, such as balneology, space heating, and horticulture.

Acknowledgements. This study was supported by funds from the MURST (ex 60%; Scientific Co-ordinator: R. Cidu). The manuscript benefited from the review of L. Fanfani.

REFERENCES

[1] R. CABOI, R. CIDU, A. CRISTINI, L. FANFANI, P. ZUDDAS, The geothermal area of the Tirso Valley (Sardinia, Italy). In: Proc. 6th Int. Symp. on Water-Rock Interaction (D.L. Miles ed.), Balkema, Rotterdam, 125-128 (1989). [2] F. FRAU, Selected trace elements in groundwaters from main hydrothermal areas of Sardinia (Italy) as a tool in reconstructing water-rock interaction. Mineralogica Petrografica Acta, 36: 281-296 (1993). [3] B. DETTORI, A.R. ZANZARI, P. ZUDDAS, Le acque termali della Sardegna. In: Ricerche geotermiche in Sardegna con particolare riferimento al graben del Campidano. CNR-PFE- RF10, Pisa, 56-86 (1982). [4] B. DETTORI, R, FANCELLI, S. NUTI, A.M. PASSINO, Il termalismo della provincia di Sassari. Le acque termo-minerali di S. Saturnino presso Benetutti (SS). Nota II. Atti 3° Convegno Int. sulle acque sotterranee, Palermo, 1-5 novembre (1975). [5] RAS, Regione Autonoma della Sardegna, Assessorato dei Lavori Pubblici, Sezione Autono- ma per il Servizio Idrografico della Sardegna, Cagliari (2000). [6] A. BRALIA, C. GHEZZO, G. GUASPARRI, G. SABATINI, Aspetti genetici del batolite sardo-corso. Rendiconti Soc. It. Mineralogia Petrografia 38: 701-764, (1982). [7] C. GHEZZO, J.B. ORSINI, Lineamenti strutturali e composizionali del batolite ercinico sardo- corso in Sardegna. In: Guida alla geologia del Paleozoico sardo, Guide Geologiche Regionali, Memorie Soc. Geologica It., 88-102 (1982). [8] L. CARMIGNANI, S. BARCA, R. CAROSI, A. DI PISA, M. GATTIGLIO, G. MUSUMECI, G. OGGIANO, P.C. PERTUSATI, Schema dell’evoluzione del basamento sardo. In: L. Carmignani et al. (Eds.), «Struttura della catena ercinica in Sardegna. Guida all’escursione». Gruppo Informale di Geologia Strutturale, pp. 11-38 (1992). [9] C. COULON, Le volcanism calco-alcalin cénozoïque de la Sardaigne (Italie). Petrographie, géochimie et genese des laves andesitiques et des ignimbrites – signification géodinamique. ÇThese Doct. 3° Cycle Univ. Aix-Marseille IIIÈ, (1977). [10] L. BECCALUVA, P. BROTZU, G. MACCIOTTA, L. MORBIDELLI, G. SERRI, G. TRAVERSA, Cenozoic tectono-magmatic evolution and inferred mantle in the sardo-Tyrrenian area. The lithosphere in Italy. Advances in Earth Science Research, Accad. Naz. Lincei, 80: 229-248 (1987). [11] S. VARDABASSO, Questioni paleogeografiche relative al terziario antico in Sardegna. Memo- rie Soc. Geologica It., 3: 665-678 (1962). [12] D.K. NORDSTROM, Thermochemical redox equilibria of ZoBell’s solution. Geochim. Cosmochim. Acta, 41: 1835-1841 (1977). [13] R. CIDU, Inductively coupled plasma Ð mass spectrometry and Ð optical emission spectrometry determination of trace elements in water. Atomic Spectroscopy, 17: 155-162 (1996). GEOCHEMICAL FEATURES OF THERMAL WATERS AT BENETUTTI (SARDINIA) 53

[14] D. PARKHURST, PHREEQC: A computer program for speciation, batch-reaction, one- dimensional transport, and inverse geochemical calculations. USGS. http:// www.ndsu.nodak.edu/webphreeq (2001) [15] R. CABOI, R. CIDU, L. FANFANI, P. ZUDDAS, Geochemistry of thermal waters in Sardinia (Italy). In: Proc. 5th Int. Symp. on Water-Rock Interaction, Reykjavik, August 8-17 (1986). [16] R. CABOI, R. CIDU, L. FANFANI, P. ZUDDAS, Geothermal studies in Sardinia. Department of Earth Science, University of Cagliari, internal report (1990). [17] TERME AURORA, Le cure termali alle Terme Aurora di Benetutti (SS). 20 pages. Stamperia Artistica, Sassari (2000). [18] A. MINISSALE, G. MAGRO, F. TASSI, F. FRAU, O. VASELLI, The origin of natural gas emissions from Sardinia island, Italy. Geochemical J., 33: 1-12 (1999). [19] GAZZETTA UFFICIALE DELLA REPUBBLICA ITALIANA, Decreto Legislativo 2 febbraio 2001, n. 31. In: Supplemento n. 52, 3 marzo 2001. Istituto Poligrafico dello Stato, Roma, Italy (2001). [20] A.J. ELLIS, W.A.J. MAHON, Chemistry and geothermal systems. Academic Press, New York (1977). [21] G. MICHARD, Behaviour of major elements and some trace elements (Li, Rb, Cs, Sr, Fe, Mn, W, F) in deep hot waters from granitic areas. Chem. Geology, 89: 117-134 (1990). [22] S. ARNORSSON, E. GUNNLAUGSSON, H. SVARVARSSON, The chemistry of geothermal waters in Iceland. III. Chemical geothermometry in geothermal investigations. Geochim. Cosmochim. Acta, 47: 567-577 (1983). [23] D.K. NORDSTROM, J.W. BALL, R.J. DONAHOE, D. WHITTEMORE, Groundwater chemistry and water-rock interactions at Stripa. Geochim. Cosmochim. Acta, 53: 1727-1740 (1989). [24] D. JI, Z. PING, Characteristics and genesis of the Yangbajing geothermal field, Tibet. In: Proc. World Geothermal Congress, Kyushu Ð Tohoku, Japan, May 28 Ð June 10, 1083-1088 (2000). [25] D. GRIMAUD, C. BEAUCAIRE, G. MICHARD, Modelling of the evolution of ground waters in a granite system at low temperature: the Stripa ground waters, Sweden. Appl. Geochemistry, 5: 515-525 (1990). [26] G. GIORGETTI, M.L. FREZZOTTI, C. GHEZZO, Structural and microthermometric studies of fluid inclusions in the Gallura intrusive complex (N Sardinia). European J. Mineralogy, 4: 1175- 1185 (1992). 54 R. CIDU, A.D. MULAS

Pagina bianca