Applied Radiation and Isotopes 55 (2001) 731–736

Origin and subsurface history of geothermal water ofMurtazabad area, PakistanFan isotopic evidence

M. Ahmada,*, W. Akrama, S.D. Hussaina, M.I. Sajjada, M.S. Zafarb

aRadiation and Isotope Application Division (RIAD), Pakistan Institute of Nuclear Science and Technology (PINSTECH), P.O. Nilore,

Islamabad, PakistanbDepartment of Physics, University of the Punjab, Lahore, Pakistan

Received 4 October 2000; received in revised form 15 May 2001; accepted 7 June 2001

Abstract

The Murtazabad area represents one of the major geothermal fields in Pakistan, with seven hot springs lying along

the Main Karakoram Thrust. Discharge of the springs is 50–1200 l per minute with the surface temperature from 40 to941C. Environmental isotopes and chemical concentrations have been used to investigate the origin and subsurfacehistory of thermal water. Four sets of water samples were collected and analyzed for various isotopes including 18O, 2Hand 3H of water; 34S and 18O of dissolved sulphates and chemical contents. Isotopic and chemical data show that the

origin of thermal water is meteoric water. On the d-diagram, d18O and d2H data plotting below the local meteoric waterline with a slope around 12.3 show that the original thermal water receives recharge from precipitation at higher altitude(3000m) and undergoes d18O shift of about 1m due to exchange with rocks. Different correlations between isotopes,

temperature and Cl indicate that the observed isotopic compositions have evolved due to mixing of differentproportions of shallow water at different spring paths during movement of thermal water towards the surface. It is alsoinferred from the tritium data along with d18O and d2H that the circulation time is long and is estimated to be more

than 50 years. r 2001 Elsevier Science Ltd. All rights reserved.

Keywords: Origin; Subsurface history; Geothermal waters; Murtazabad; Pakistan

1. Introduction

Research related to geothermal waters in Pakistan isstill at an early stage. Geothermal areas have beenstudied on a regional scale in terms of geological setting,

lithology and general chemical characteristics of ground-water compositions. Studies focusing on the nature andevolution of geothermal water such as the age, origin,

water/rock interaction or mixing between geothermaland shallow groundwater have been very limited.Various applications of isotope geochemical techniquesfor geothermal investigations have been discussed by

Giggenbach et al. (1983); Giggenbach (1992); Truesdelland Hulston (1980); Krouse (1980) and many others.

For a more systematic study on geothermal water,Murtazabad geothermal field located in Northern Areas

of Pakistan at latitudes 361 170 N and longitudes 741 350

E was selected. It is one of the major geothermal fieldshaving an altitude of about 2100m above mean sea

level. The area is characterized by steep topography andU-shaped glaciated valleys and is drained by the riverHunza which joins the great River Indus after joining

Gilgit River. The climate is characterized by coldwinters, and warm and dry summers. June–August arehot months, during which time the mean maximumtemperature is about 301C. Snowfall occurs during the

cold months of December–February when the minimumtemperature goes several degrees below the freezingpoint. Rainfall is scanty. Seven geothermal springs

located along the Gilgit–Hunza road just below theterrace of Murtazabad Town were studied. These

*Corresponding author. Fax: +92-51-9290-275.

E-mail address: manzoor@pinstech.org.pk (M. Ahmad).

0969-8043/01/$ - see front matter r 2001 Elsevier Science Ltd. All rights reserved.

PII: S 0 9 6 9 - 8 0 4 3 ( 0 1 ) 0 0 1 1 9 - 1

springs are scattered over 1.5 km on the right bank ofthe Hunza river, near the bed. The springs emanate at an

elevation of 1945 to 2020m, with surface temperaturesranging from 40 to 941C. The discharge varies from 50to 1200 l per minute. Precipitation samples and four sets

of water samples were collected from the springs and theriver. Knowledge of the origin and physico-chemicalbehaviour of geothermal water is an essential pre-requisite for further detailed exploration.

2. Tectonic setting

The geotectonic development of northern areas ofPakistan during the late Cretaceous to Cenozoic period

involves three tectonic elements viz. the Indo-Pakistanshield and its northern sedimentary cover (the IndianMass), the rocks deposited on the southern part of the

Eurasian Mass, and Kohistan Island Arc Sequence(Powell, 1979). From Archaean times, the Indian Sub-continent was a part of Gondwanaland, which consistedof the continents of South America, Africa, Antarctica,

Australia and India. A vast stretch of Tethys Sea existedbetween the Indo-Australian part of Gondwanaland andthe Eurasian Mass. About 130 million years ago, the

Indian Ocean plate departed from Gondwanaland andstarted drifting towards Eurasia with the simultaneousconsumption of the Tethys Sea plate in between (Khan

et al., 1987). As a result of intra-oceanic subduction infront of the Indo-Pakistan plate, Kohistan Island ArcSequence was produced on the north of the subduction

zone. The first contact of this Island Arc was with theIndo-Pakistan plate, which finally collided with theEurasian Mass. The Kohistan Island Arc Sequence isjuxtaposed between the Indo-Pakistan plate and the

Eurasian plate. A major thrust fault called the mainmantle thrust (MMT) separates the Indian Mass fromthe Kohistan Island Arc Sequence while another thrust

fault called the main Karakoram thrust (MKT) marksthe boundary between the Kohistan Island Arc Se-quence and the Eurasian Mass (Tahirkheli, 1982). The

geothermal manifestations under investigation lie alongthe MKT thrust fault, which is still active. The heatwhich is generated is mainly due to friction along thesefaults (Todaka et al., 1988). The major tectonic features

of the northern areas of Pakistan and the Murtazabadgeothermal site are shown in Fig. 1.

3. Geology

The geothermal springs at Murtazabad lie north ofthe MKT between the Kohistan Island Arc Sequenceand the Eurasian Mass. The rocks constituting the

Kohistan Island Arc Sequence consist of thick calc-alkaline plutonic, volcanic and volcano-sedimentary

rocks, Jurassic–Cretaceous in age (Todaka et al.,1988). The rocks of the Eurasian Mass north of MKT

are late-Paleozoic metasedimentary rocks, mainly flysch(limestone and shale dominant) which are considereddeep sea sediments deposited by turbidity currents along

the Eurasian plate margin in the northern TethysGeosyncline (Khan et al., 1987). The springs emanatefrom a steep cliff, which is made up of fluvial depositslargely comprising gravel. The hard rocks exposed

around the manifestations are garnet staurolite schistand limesilicate marble of the Baltit Group, which isassigned lower Paleozoic to Precambrian age.

4. Materials and methods

Four sets of water samples for measurement ofisotopes (2H, 3H, 18O, 34S) and chloride (Cl) were

collected from hot springs, rivers/streams, etc. Filtrationand acidification of samples was carried out in the fieldwhen required. For analysis of 34S and 18O of dissolvedsulphates, sulphates were precipitated as BaSO4 in the

field (Giggenbach and Goguel, 1989). Physico-chemicalparameters like temperature, pH and electrical conduc-tivity were determined in the field. Analysis of Cl was

carried out using an ion selective electrode (APHA,1985). The stable isotopes 18O and 2H of the water weredetermined relative to Vienna standard mean Ocean

water (VSMOW) on mass spectrometers. The d18O wasmeasured by the CO2 equilibration method (Sajjad,1989). Water samples were reduced to hydrogen gas by

zinc shots for dD measurement (Coleman et al., 1982).Measurement uncertainties for d18O and d2H were 0.1mand 1.0m, respectively. For determination of isotopicvalues of dissolved sulphates, BaSO4 precipitated in the

field, following removal of carbonates, was converted toCO2 for d

18O (Nehring et al., 1977) and to SO2 for d34S

(Yanagisawa and Sakai, 1983). Tritium content of the

samples was determined by liquid scintillation countingafter electrolytic enrichment. The standard error ofmeasurement was of the order of 71 tritium unit (TU).

5. Results and discussion

5.1. Origin of geothermal waters

Isotopic data can adequately differentiate between the

three possible types of origin of thermal water i.e.magmatic, oceanic and meteoric. The ranges of d18O andd2H of all the sampled geothermal manifestations were

�12.8 to �11.6m and �92 to �76m, respectively. Thesedata do not show the presence of any significant amountof magmatic water which generally has d18O: +6 to

+9m and d2H: �40 to �80m (Pearson and Rightmire,1980; Giggenbach, 1992). The electrical conductivity

M. Ahmad et al. / Applied Radiation and Isotopes 55 (2001) 731–736732

(EC) varies from 1603 to 3058 mS/cm while the Cl

content was found to be 13 to 51 ppm. The possibility ofoceanic origin of the encountered thermal waters is ruledout by the absence of highly enriched d18O, d2H and Cl

values. So the origin of thermal waters appears to bemeteoric. d18O vs. d2H has been plotted in Fig. 2. Theline AC in this figure represents the local meteoric water

line (LMWL) determined by Hussain et al. (1995) for thesame area using samples of snow-melt water instead ofrain water which is usually present in trace concentra-

tions only. This line is given by the equation:

d2H ¼ ð870:4Þ d18Oþ ð16:576Þ ðr2 ¼ 0:933; n ¼ 24Þ:

ð1Þ

All the points form a trend below the LMWL.Departure of these points from LMWL may be due to

rock interaction, evaporation or mixing processes(Giggenbach et al., 1983). This aspect will be discussedin the next section.

In a geothermal system, d34S of sulphates with amagmatic origin ranges between 0 and +2m canyon

diablo troilite (CDT). Sulphates resulting from the

dissolution of evaporites can have d34S from +10 to+35m whereas in modern oceanic sulphates its value isabout +20m (Krouse, 1980). The d34S values ofsulphates of hot springs are in the range of +4.6–

+10.7m, showing that the sulphates are neither of

Fig. 1. Map showing major tectonic features of the Northern Areas and study site.

Fig. 2. d18O vs. d2H of geothermal springs.

M. Ahmad et al. / Applied Radiation and Isotopes 55 (2001) 731–736 733

magmatic origin nor of modern oceanic origin. Thepossibility of oceanic origin (d18O of SO4=+9.6m) is

further ruled out as the d18O of sulphates of these hotsprings is in the range of �2.6–+1.2m (Fritz et al., 1988;Clark and Fritz, 1998). Relatively low values of d34Sindicate that the major contribution of sulphates isderived from reduced sulphur compounds such assulphide minerals and/or organic sulphides (Pearsonand Rightmire, 1980). Sulphide minerals are exposed at

some places along the cliff of the Murtazabad terrace.The source of sulphates also confirms the origin ofthermal water as meteoric, which dissolves sulphide

minerals during deep circulation.

5.2. Subsurface history

There are two important processes affecting theisotopic and chemical concentrations in geothermal

waters; firstly, steam separation due to adiabaticexpansion of thermal fluids with decreasing pressureand, secondly, dilution/mixing with waters derived from

shallow sources (Giggenbach et al., 1983). The isotopicand chemical signatures of the geothermal fluids havebeen used to identify the prevailing process. The tritiumconcentration of thermal springs ranges from 5 to 66

TU. The variation of tritium over a wide range may bedue to mixing of varying proportions of fresh water withthe thermal waters during movement towards the earth

surface or due to large variation in their residence time.Such a large variation in residence time cannot beexpected for hot springs grouped in a small area.

Negative correlation between tritium and chloride(Fig. 3) favours the mixing of fresh water with thermalwater. With the decrease of surface temperature ofthermal waters, their tritium concentration increases and

chloride content decreases (Fig. 4). This evidence showsthat the cooling of thermal water is mainly due to mixingwith fresh water having high tritium and low Cl contents

(Navada et al., 1995). Fig. 5 shows that d18O isnegatively correlated with temperature and chloridecontent. Lowering of temperature accompanied with

18O enrichment also happens during the evaporation/steam separation process. If this process had been

dominant, the Cl content would have increased withthe enrichment of 18O. However, the observed trend isthe reverse of this. It also indicates that the thermal

water is mixing with the shallow cold water enriched in18O but is depleted in Cl (Mazor et al., 1980;Giggenbach et al., 1983).A plot of d18O vs. d2H (Fig. 2) not only confirms the

mixing process but also gives insight into the history ofthermal water. All the thermal waters plot below theLMWL forming the trend line AB of slope 12.3 and

intersecting the meteoric line at a composition ofd18O=�11.6m and d2H=�77m. This composition isexactly similar to the average d18O and d2H values of

snow (i.e. �11.6 and �76m) collected from theMurtazabad terrace. According to the negative correla-tion of d18O with temperature shown above, the hottest

spring samples lie at the lower end ‘B’ of the line, whilethe samples with minimum temperature plot at theupper end ‘A’. However, reduction in temperature ofthermal waters also happens due to a process of

subsurface steam loss accompanied by enrichmentof d18O and d2H. In this case the thermal waters shouldplot along a line of slope of less than eight because in

such a kinetic process the fractionation factor of 18O is

Fig. 3. Chloride vs. tritium of geothermal springs.

Fig. 4. Temperature vs. tritium and chloride of thermal springs.

Fig. 5. d18O vs. temperature and chloride of thermal springs.

M. Ahmad et al. / Applied Radiation and Isotopes 55 (2001) 731–736734

greater than that of 2H (Gonfiantini, 1986). In thepresent situation, the slope of line AB in Fig. 2 is 12.3

and so the process of steam separation is ruled out. Dueto location at relatively low altitude the snow of the localarea is isotopically enriched compared to general

meteoric water which originates from precipitation athigh altitudes drained by rivers (d18O=�13.8m,d2H=�94m). The local snow being located very nearto point ‘A’ representing the thermal water sample with

lowest temperature, can be considered as an end-mixingcomponent. It shows that the thermal water of meteoricorigin depleted in 18O and 2H gets mixed with the

shallow groundwater resulting from local snow beforedischarging to ground surface. Variation in isotopic andchemical concentration of thermal waters flowing

through various paths is due to mixing with varyingproportions of fresh water.The d18O/d2H plot (Fig. 2) also explains the process

after recharge of meteoric water in the hot zone beforemixing with shallow cold water. Tritium values decreasefrom 66 to 5 TU along the line AB. On extrapolationthis line meets the possible 18O shift line (line CD),

originating from mean values of d18O and d2H formeteoric water (Point C), at the Point ‘D’. According tothe proportionality of extrapolated length, the point D

approximately represents tritium concentration as zeroTU. It can be inferred that before mixing, the thermalwater had values of d18O and d2H similar to those of the

point D (i.e. �13 and�94m) and a tritium concentrationof about zero TU. This suggests pre-1953 recharge ofthese waters (i.e. recharge water was free of bombgenerated tritium). Considering the meteoric water as

the origin of geothermal water, the departure fromLMWL, which is the d18O shift of about 1m, isattributed to rock water interaction at higher tempera-

ture. In this process the 18O of fluid exchanges withreservoir rocks having very high d18O (Clark and Fritz,1998). Depletion of about 2.4m in 18O of the original

meteoric water compared to local precipitation of theMurtazabad Area gives an approximate difference inaltitude of the recharge area of 1200m using the

gradient of d18O as 0.2m/100m in cold regions(Eriksson, 1983). This indicates that the meteoric watergets recharged in areas having an altitude of about3000m above mean sea level.

6. Conclusions

The thermal waters evolve from meteoric waterrecharged at higher altitude that subsequently undergoes

d18O enrichment of about 1m due to exchange of 18Owith rocks at higher temperature. Mixing with varyingproportions of shallow fresh groundwater is mainly

responsible for cooling and isotopic enrichment of thedischarge compositions of different manifestations. No

tritium in the original thermal water, as has beenestimated, indicates a residence time of recharging water

in the geothermal system of more than 50 years.

7. Potential applications

This study has proved that by the combined use of

isotopic and geochemical techniques, origin of water anddissolved salts, subsurface history, and circulation timecan be investigated. Therefore, these techniques can be

potentially applied in the fields of volcanology, geother-mics and heat extraction from deep hot rocks. Source ofthe volcanic or geothermal water can be identified,

whether it is purely from magmatic source or meteoricwater recharged in the ground or mixture of both thesources. In case of circulation of fresh water, itsresidence time can be determined. In heat extraction

from deep rocks, fresh water is circulated. The changesin isotopic signatures of circulation water would giveinformation about the mixing with deep formation

water, which is normally very old and has quite differentisotopic values. Surface exploration using isotopic andchemical techniques provides information to assess the

maturity, replenishment rate and temperature of athermal reservoir. Hence, these techniques are veryuseful for assessing the geothermal water reservoir

potential for exploitation of thermal energy to generateelectric power or space heating.

References

American Public Health Association, APHA, 1985. Standard

Methods for the Examination of Water and Wastewater,

sixth edition, APHA, Washington DC.

Clark, I.D., Fritz, P., 1998. Environmental Isotopes in

Hydrogeology. Lewis Publishers, Boca Raton, New York.

Coleman, M.L., Shepherd, T.J., Durham, J.J., Rouse, J.E.,

Moor, G.R., 1982. Reduction of water with zinc for

hydrogen isotope analysis. Anal. Chem. 54, 993–995.

Eriksson, E., 1983. Stable isotopes and tritium in precipitation.

In: International Atomic Energy Agency, Guide Book on

Nuclear Techniques in Hydrology, Technical Report Series

No. 91, Vienna, Austria.

Fritz, P., Lapcevic, A., Miles, M., Frape, S.K., Lawson, D.E.,

O’Shea, K.J., 1988. Stable isotopes in sulphate minerals

from the salina formation in southwestern Ontario. Can.

J. Earth Sci. 25, 195–205.

Giggenbach, W.F., 1992. Isotopic shift in waters from

geothermal and volcanic systems along convergent plate

boundaries ad their origin. Earth Planet. Sci. Lett. 113,

495–510.

Giggenbach, W.F., Goguel, R.L., 1989. Collection and Analysis

of Geothermal and Volcanic Water and Gas Discharges.

Report No. CD 2401. Chemistry Division DSIR, Petone,

New Zealand.

M. Ahmad et al. / Applied Radiation and Isotopes 55 (2001) 731–736 735

Giggenbach, W., Gonfiantini, R. and Panichi, C., 1983.

Geothermal Systems. In: International Atomic Energy

Agency, Guidebook on Nuclear Techniques in Hydrology,

Technical report series No. 91, IAEA, Vienna, pp. 359–379.

Gonfiantini, R., 1986. Environmental isotopes in lake studies.

In: Fritz, P., Fontes, J.-Ch. (Eds.), Handbook of Environ-

mental Isotope Geochemistry, Vol. 2. Elsevier, Amsterdam-

Oxford, New York.

Hussain, S.D., Ahmad, M., Gonfiantini, R., Akram, W.,

Sajjad, M.I., Tasneem, M.A., (1995). Isotopic and chemical

studies of geothermal waters of Northern Areas of Pakistan,

Proceedings of the Final Research Coordination Meeting on

the Application of Isotope and Geochemical Techniques in

Geothermal Exploration in the middle East, Asia, the

Pacific and Africa, Dumaguete City, Philippines, 12–15

October 1993, IAEA-TECDOC-788, Vienna, Austria.

Khan, K.S.A., Khan, I.H., Leghari, A.L., Khan, M.S.Z., 1987.

Geology along the Karakoram Highway from Hassan

Abdal to Khunjerab Pass. Geological Survey of Pakistan,

Quetta.

Krouse, H.R., 1980. Sulphur isotopes in our environment. In:

Fritz, P., Fontes, J.-Ch. (Eds.), Handbook of Environ-

mental Isotope Geochemistry, Vol. 1. Elsevier, Amsterdam,

pp. 435–472.

Mazor, E., Levitte, D., Truesdell, A.H., Healy, J., Nissenbaum,

A., 1980. Mixing models and ionic thermometers applied to

warm (up to 601C) springs: jordan rift valley, Israel.

J. Hydrol. 45, 1–19.

Navada, S.V., Nair, A.R., Sharma, S., Kulkarni, U.P., 1995.

Geochemical and isotopic studies of the Central and

Northern India. Proceedings of the Final Research

Coordination Meeting on the Application of Isotope and

Geochemical Techniques in Geothermal Exploration in the

Middle East, Asia, the Pacific and Africa, Dumaguete City,

Philippines, 12–15 October 1993, IAEA-TECDOC-788,

Vienna, Austria.

Nehring, N.L., Bowen, P.A., Truesdell, A.H., 1977. Techniques

for the conversion to carbon dioxide of oxygen from

dissolved sulphates in thermal waters. Geothermics 5,

63–66.

Pearson Jr., F.J., Rightmire, C.T., 1980. Sulphur and oxygen

isotopes in aqueous sulphur compounds. In: Fritz, P.,

Fontes, J.-Ch. (Eds.), Handbook of Environmental Isotope

Geochemistry, Vol. 1. Elsevier, Amsterdam-Oxford, New

York, pp. 227–257.

Powell, C.McA., 1979. A speculative tectonic history of

Pakistan and surroundings: some constraints from the

Indian Ocean. In: Farah, A., Kees, DeJong. (Eds.),

Geodynamics of Pakistan, Geological Survey of Pakistan,

Quetta.

Sajjad, M.I., 1989. Isotope Hydrology in Pakistan, Instrumen-

tation-Methodology-Applications. PhD thesis, University

of the Punjab, Lahore.

Tahirkheli, R.A., 1982. Geology of the Himalayas, Karakoram

and Hindukush in Pakistan, Geological Bulletin, University

of Peshawar, Special Issue, Vol. III.

Todaka, N., Shuja, T.A., Jamiluddin, S., Khan, N.A., Pasha,

M.A., Iqbal, M., 1988. A Preliminary Study for Geothermal

Development Project in Pakistan. Geological Survey of

Pakistan, Quetta.

Truesdell, A.H., Hulston, J.R., 1980. Isotope evidence on

environments of geothermal systems. In: Fritz, P., Fontes,

J.-Ch. (Eds.), Handbook of Environmental Isotope

Geochemistry., Vol. 1. Elsevier, Amsterdam, pp. 179–226.

Yanagisawa, F., Sakai, H., 1983. Thermal decomposition of

barium sulphate–vanadium pentaoxide–silica glass mixture

for preparation of sulfur dioxide in sulfur isotope ratio

measurements. Anal. Chem. 55 (6), 985–987.

M. Ahmad et al. / Applied Radiation and Isotopes 55 (2001) 731–736736