Hydrological Sciences Journal
ISSN: 0262-6667 (Print) 2150-3435 (Online) Journal homepage: https://www.tandfonline.com/loi/thsj20
Application of geochemical and stable isotopic
tracers to investigate groundwater salinity in the
Ochi-Narkwa Basin, Ghana
Samuel Y. Ganyaglo, Shiloh Osae, Tetteh Akiti, Thomas Armah, Laurence
Gourcy, Tomas Vitvar, Mari Ito & Isaac A. Otoo
To cite this article: Samuel Y. Ganyaglo, Shiloh Osae, Tetteh Akiti, Thomas Armah, Laurence
Gourcy, Tomas Vitvar, Mari Ito & Isaac A. Otoo (2017) Application of geochemical and stable
isotopic tracers to investigate groundwater salinity in the Ochi-Narkwa Basin, Ghana, Hydrological
Sciences Journal, 62:8, 1301-1316, DOI: 10.1080/02626667.2017.1322207
To link to this article:  https://doi.org/10.1080/02626667.2017.1322207
View supplementary material 
Published online: 22 May 2017.
Submit your article to this journal 
Article views: 172
View Crossmark data
Citing articles: 3 View citing articles 
Full Terms & Conditions of access and use can be found at
https://www.tandfonline.com/action/journalInformation?journalCode=thsj20
HYDROLOGICAL SCIENCES JOURNAL – JOURNAL DES SCIENCES HYDROLOGIQUES, 2017
VOL. 62, NO. 8, 1301–1316
https://doi.org/10.1080/02626667.2017.1322207
Application of geochemical and stable isotopic tracers to investigate
groundwater salinity in the Ochi-Narkwa Basin, Ghana
Samuel Y. Ganyagloa,b, Shiloh Osaea, Tetteh Akitib, Thomas Armahc, Laurence Gourcyd, Tomas Vitvare,f, Mari Itog
and Isaac A. Otoob
aNuclear Chemistry and Environmental Research Centre, Ghana Atomic Energy Commission/National Nuclear Research Institute, Legon-
Accra, Ghana; bSchool of Nuclear and Allied Sciences (SNAS), University Of Ghana, Kwabenya, Ghana; cDepartment of Earth Science,
University of Ghana, Legon, Accra, Ghana; dBRGM, Orleans, France; eFaculty of Environmental Science, Department of Applied Ecology,
Czech University of Life Sciences, Prague, Czech Republic; fFaculty of Civil Engineering, Czech Technical University in Prague, Prague, Czech
Republic; gWater Resources Programme, International Atomic Energy Agency, Vienna International Centre, Vienna, Austria
ABSTRACT ARTICLE HISTORY
Rainwater, groundwater and soil-water samples were analysed to assess groundwater geochem- Received 23 June 2015
istry and the origin of salinity in the Ochi-Narkwa basin of the Central Region of Ghana. The Accepted 16 January 2017
samples were measured for major ions and stable isotopes (δ18O, δ2H and δ13C). The Cl− content EDITOR
in rainwater decreased with distance from the coast. The major hydrochemical facies were Na-Cl D. Koutsoyiannis
for the shallow groundwaters and Ca-Mg-HCO3, Na-Cl and Ca-Mg-Cl-SO4 for the deep ground- ASSOCIATE EDITOR
waters. Groundwater salinization is caused largely by halite dissolution and to a minor extent by K. Heal
silicate weathering and seawater intrusion. Stable isotope composition of the groundwaters
followed a slope of 3.44, suggesting a mixing line. Chloride profiles in the soil zone revealed KEYWORDS
the existence of salt crusts, which support halite dissolution in the study area. A conceptual flow geochemistry; stable
model developed to explain the mechanism of salinization showed principal groundwater flow in isotopes; salinity sources;
the NW–SE direction. Ghana
1 Introduction
throughout the country. They include over 15 000
The growing interest in groundwater use in sub- boreholes, 60 000 hand-dug wells and some dugouts.
Saharan Africa is prompted by the fact that surface The aquifer yields are generally low, scarcely exceeding
water resources have been unable to satisfy the water 6 m3/h (Kortatsi 1994).
demand for socio-economic development. In the past Groundwater quality in Ghana is generally consid-
two decades, groundwater sources have become the ered to be good for domestic and agricultural purposes.
preferred drinking water supply means to meet the However, reported low pH values (3.5–6.0), high iron
growing demand of the largely rural and dispersed concentrations (1–64 mg/L) and high salinity content
communities and small urban towns in Ghana. About (5000–14 584 mg/L) in some coastal aquifers (Kortatsi
95% of groundwater use in Ghana is for household and 1994) justify the need to investigate the hydro-geo-
drinking purposes. About 50% of the total number of chemical and isotopic composition of groundwaters
hand-dug wells are used for drinking, whereas about to understand the processes that contribute to elevated
66% are used for both drinking and domestic purposes. ionic concentrations in groundwater.
Less than 5% of groundwater in Ghana is used for In the Ochi-Narkwa basin of the Central Region of
vegetable farming irrigation and livestock watering, Ghana, groundwater is an important source of potable
mostly in the Volta, Upper East, Upper West and water due to limited surface water supplies. It is used
Greater Accra Regions (Obuobie and Barry 2010). mainly for domestic purposes and, to a lesser extent,
Groundwater is abstracted from all geological for- for industrial and agricultural purposes. One of the
mations in the country. Kortatsi (1994) reported the barriers to the exploitation of groundwater in the
existence of 56 000 groundwater abstraction systems Central Region is poor water quality in the majority
consisting of boreholes, hand-dug wells and dugouts. of the boreholes due to high salinity. The origin of the
Currently there are over 75 000 abstraction systems high salinity groundwaters in this area is still poorly
CONTACT Samuel Y. Ganyaglo sganyaglo@yahoo.co.uk; sganyaglo@gmail.com
Supplementary data for this article can be accessed here.
© 2017 IAHS
1302 S. Y. GANYAGLO ET AL.
understood. Armah (2002) attributed the high salinity Meteorological Agency, 2002–2011). The coldest
to multiple sources including seawater intrusion and month is August and the hottest months are March
mixing of freshwater with subsurface saline formation. and April. The mean monthly relative humidity is 79–
The use of chemical and isotopic tracers such as Br/ 89% (Ghana Meteorological Agency, 2002–2011).
Cl, δ18O, δ2H, 3H, 87Sr/86Sr, and δ11B to determine the The largest part of the study area is drained by
origin of salinity in groundwater has been demon- the River Ochi-Narkwa and its tributaries. It rises
strated by various authors (Jørgensen and Banoeng- from the northern section of the study area, flows
Yakubo 2001, Kim et al. 2003, Kortatsi 2006, into the Narkwa Lagoon at the coast and discharges
Bouchaou et al. 2008). In this paper, chemical tracers into the sea east of Narkwa village (Fig. 1). A small
(Br−/Cl−, Na+/Cl−, SO42−/Cl−), stable oxygen and section of the study area is drained by the River
hydrogen isotopes (δ18O, δ2H) and carbon isotopes Ayensu. The Ayensu basin drains a total area of
(δ13C) were employed to determine the origin of sali- 1709 km2 and forms part of the Coastal River Basin
nity in the coastal aquifers of the Central Region of System of Ghana. The source of this river is located in
Ghana, with a focus on the Ochi-Narkwa basin. the Atewa hills in the East Akim District at an altitude
of 610 m a.m.s.l.
The geology of the study area contains
2 Study area Paleoproterozoic Supracrustals and intrusive rocks
The study area lies between latitudes 5°11 45 5°28.66 constituting the Birimian Supergroup and rocks of′ ″– ′
30 N and longitudes 0°33 41.6ʹ 1°11 17 W and has a the Eburnean Plutonic Suite (Fig. 2). The Birimian″ ′ ’– ′ ″
total area of 700 km2, bounded to the south by the Gulf Supergroup consists of volcanic belts and sedimentary
of Guinea (Fig. 1). basins. The volcanic belts are composed of low-grade
The prevailing climatic conditions in the area are metamorphic tholeiitic basalts with intercalated volca-
dry equatorial and moist semi-equatorial. Annual rain- niclastics as well as minor andesitic and felsic flow
fall ranges from 1000 mm along the coast to 2000 mm rocks and local chemical sediments. The volcanic
in the hinterland (Ghana Meteorological Agency, belts are intruded by coeval, comagmatic and synvol-
2002 2011). The wettest months are May June and canic granitoid plutons such as tonalite and granodior-– –
September October, whereas drier periods occur in ite. The sedimentary basins comprise wacke and–
December February. The range of mean monthly volcaniclastic sediments, mica schist, amphibolite of–
minimum temperature is 22.8 24.9°C, whereas mean contact-metamorphic origin, biotite gneiss, locally mig-–
monthly maximum temperature is 27.2 31.9°C (Ghana matitic and minor biotite schist and undifferentiated–
Figure 1. Location map of the study area with sampling points.
HYDROLOGICAL SCIENCES JOURNAL – JOURNAL DES SCIENCES HYDROLOGIQUES 1303
Figure 2. Geological map of the study area (modified from the New Geological Map of Ghana, Geological Survey Department of
Ghana) showing spatial distribution of EC and inferred flow direction (→).
hornblende-biotite granitoid. The Eburnean Plutonic the highest (179.7 L/min) occurred in the biotite
Suite consists of undifferentiated biotite granitoid, bio- gneiss/biotite granite contact in the Katakyiase area in
tite gneiss and hornblende–biotite granitoid. Typical the hinterland (Fig. 2). Of the boreholes, 52% were
rock types in these groups have been reported (Leube classified as confined and 48% as semi-confined. The
et al. 1990, Hirdes et al. 1992, Taylor et al. 1992) as transmissivity of the boreholes in the study area
quartz diorites, tonalites and trondhjemites, granodior- obtained from discharge and drawdown data ranged
ites, adamellites and granites which intrude into the from 0.3 to 26.9 m2/d, with a mean of 4.1 m2/d for
sedimentary basins. Part of the study area closer to the unconfined aquifers. In the case of confined aquifers
sea (Ekumfi Asafa area) falls within the Amisian group the transmissivity values were 0.3–35.7 m2/d. A
of the Mesozoic and includes conglomerates, mud- hydraulic head map of the area showed that ground-
stones, micaceous sandstones and arkoses (Fig. 2). water flows in the NW–SE direction, towards the coast-
Groundwater is present in all principal lithologies in line (Fig. 3).
the area: sediment/volcaniclastic sediment, wacke sedi-
ment, biotite gneiss and biotite granitoid. Data held on
41 wells in the study area by the Community Water 3 Methods
and Sanitation Agency (CWSA), Central Region, indi- A total of 78 water samples from 70 boreholes (deep
cated that the thickness of the weathered zone varied wells, BH) and eight hand-dug wells (shallow wells,
from 0 to 30 m below the surface with a mean of HD) were obtained. In total, 137 rainwater samples
14.1 m. The borehole depth was 18–94.5 m with a were collected at the Saltpond and Twifo Praso meteor-
mean of 34.5 m. The static water level varied from 0 ological stations from 2010 to 2011 on an event basis.
to 20 m below the land surface. The mean static water Thirty-five soil samples were obtained from four sites
level was 5.8 m. The borehole yield was 7.5–179.7 L/ at various depths to assess the contribution of Cl– in
min and the mean yield was 32.1 L/min. The lowest the soil to groundwater salinity. All water samples were
yield of 7.5 L/min occurred in the schist aquifer; and collected in high-density polyethylene (HDPE) bottles,
1304 S. Y. GANYAGLO ET AL.
Figure 3. Hydraulic head (meters) map of the study area showing groundwater flow direction.
which were conditioned by washing initially with measurement of sodium (Na+) and potassium (K+). A
detergent, then with 10% nitric acid, and finally rinsing Dionex 90 ion chromatograph was used for the deter-
several times with distilled water. This was carried out mination of Cl–, SO 2–, Br–, NO –, PO 3–4 3 4 and F
–. A
to ensure that the sample bottles were free from con- charge balance error of ±10% was accepted.
taminants. Representative in situ groundwater samples The Los Gatos Research (LGR) instrument, LGR
were collected after purging the aquifers to evacuate DLT-100 (model 908–0008), was used to determine
stagnant water in the borehole casing. Temperature, the δ18O and δ2H values of the water samples at the
pH and redox potential (Eh) were measured in water Isotope Hydrology Laboratory of the International
samples in the field using a WTW 3110 field probe, Atomic Energy Agency (IAEA) in Vienna and the
and electrical conductivity (EC), total dissolved solids NNRI of GAEC. Isotopic results were reported on the
(TDS) and salinity were measured using a WTW 3210 delta-scale (δ) with respect to Vienna Standard Mean
field probe. Bicarbonates (HCO –3 ) were measured in Ocean Water (V-SMOW) (Coplen 1996). The instru-
the field by a titrimetric method as total alkalinity. ment has a precision of approximately 1‰ for δ2H and
Samples were filtered through 0.45-µm cellulose filters 0.2‰ for δ18O. Internal standards were calibrated
with the aid of a hand-operated vacuum pump. Part of against the international standards Vienna Standard
the filtered water was used to rinse the sample bottles Mean Ocean Water (VSMOW2) and Standard Light
three times before sampling, as suggested by Boghici Antarctic Precipitation (SLAP2) for δ2H and δ18O.
(2003). The remaining filtered water was then trans- Values obtained were normalized using VSMOW2
ferred into two 250-mL HDPE bottles for cation and and SLAP2 on the δ-scale. The internal standards
anion analyses. The samples for cation analysis were were used as unknowns against certified reference sec-
acidified with 0.2M HNO3 to preserve the ions in ondary standards from the IAEA to measure the
solution. Samples for stable isotope analysis were not unknown samples. All δ2H and δ18O values for water
filtered. The sample bottles (50-mL HDPE) were filled samples were reported based on the VSMOW/SLAP
directly with the borehole water and then capped. The scale.
rainwater samples were collected from the rain gauges
at the meteorological stations of Saltpond and Twifo
Praso. Thirty-five soil samples from four profiles were 4 Results
collected by means of a hand auger. The samples were
4.1 Hydrochemistry
collected at 20-cm intervals to maximum depths of
160–200 cm and transferred into clean polyethylene 4.1.1 Rainfall
zip-locked bags with a plastic trowel (IAEA 2008). Proportions of anions in rainfall from the Saltpond
The bags were immediately sealed, labelled and kept station showed the sequence Cl− > SO 2− −4 > NO3 >
in ice chests at a relatively low temperature (4°C). PO 3−4 > F
− > Br− > NO −2 , whilst the analogous
Chemical analyses of the water samples were carried sequence for the Twifo Praso station was Cl− > NO −3
out at the National Nuclear Research Institute (NNRI) > SO 2−4 > PO
3−
4 > F
− > Br− > NO −2 . The Cl
− concen-
of the Ghana Atomic Energy Commission (GAEC) trations from the two meteorological stations, Saltpond
using an atomic absorption spectrometer for calcium and Twifo Praso, are shown in Figure 4. The Cl− anion
(Ca2+), magnesium (Mg2+) and trace metal determina- was the most abundant anion at both stations; Cl−
tions. Flame emission spectrometry was used for the concentrations from Saltpond station varied between
HYDROLOGICAL SCIENCES JOURNAL – JOURNAL DES SCIENCES HYDROLOGIQUES 1305
Figure 4. Concentrations of Cl− in rainfall events between May and August 2010 at the Twifo Praso and Saltpond stations.
1.07 and 22.32 mg/L, with a mean of 9.71 mg/L, while borehole CR2-54 at Anokye and lower Cl− was found in
those from Twifo Praso station varied between 0.48 borehole CR4-FZ-14 at Abora Dunkwa. Concentrations
and 8.28 mg/L, with a mean of 3.12 mg/L. Higher of NO −3 varied from 0.15 to 0.18 mg/L, with a mean of
values were observed at the Saltpond station closer to 0.15 mg/L. Generally, low NO −3 concentrations occurred
the coast, suggesting that the ocean is the major con- in the shallow groundwaters of the entire study area.
tributor of Cl− in rainfall.
4.1.3 Deep groundwaters
4.1.2 Shallow groundwaters Temperatures of the deep groundwaters were in the
A statistical overview of the hydrochemical results of 23.2–30.9°C range, with a mean of 28°C. A statistical
the shallow groundwaters is presented in Table 1(a) overview of measured parameters for the deep ground-
(detailed results are presented in the Supplementary waters is presented in Table 1(b) (detailed results are in
material, Table S1). The minimum pH of the shallow given in the Supplementary material, Table S2). The
groundwaters was 5.2 and occurred in hand-dug well pH varied from 5.32 to 7.84, indicating slightly acidic
CR4-FZ-03 in the northwestern section of the study to slightly alkaline conditions. A wide variation of EC
area. The maximum pH value of 9.3 occurred in hand- from 120.6 to 29 000 µS/cm was observed, with a mean
dug well CR4-01 at Ekumfi Asafa in the southern of 2347 µS/cm. Higher EC occurred in the deep
section of the study area close to the Gulf of Guinea groundwaters as compared to the shallow ground-
(Fig. 2). waters (Fig. 2). The highest value of 29 000 µS/cm
Electrical conductivity (EC) of the shallow ground- occurred in borehole CR2-45 at Gomoa Abora, which
waters was in the range 540–6050 μS/cm, with a mean is located 5.9 km from the coast within the amphibolite
of 2102 μS/cm. Elevated EC values in wells CR2-54 (6050 aquifers (Fig. 2).
μS/cm) and CR2-53 (5050 μS/cm) occurred at 4 and The cation Na+ was the dominant cation in the deep
16 km from the coast (Fig. 5). The concentration of groundwaters. The abundance of cations was in the
TDS ranged between 270 and 3030 mg/L, with a mean order Na+ > Ca2+ > Mg2+ > K+. The Na+ concentration
of 1098 mg/L. The highest TDS of 3030 mg/L occurred in varied from 22.2 mg/L in borehole CR4-FZ-08 at
the hand-dug well CR2-54 at Anokye. The dominant Ayeldu to 2770 mg/L in borehole CR2-45 at Gomoa
cation in the shallow groundwaters was Na+, followed Abora (Fig. 6(a)), with a mean of 335.9 mg/L. The
by Ca2+, Mg2+ and K+. The amount of Na+ varied from dominant anion in the deep groundwater was Cl−.
26.4 to 1430 mg/L. The mean Na+ content of the shallow The abundance of the anions was in the order Cl− >
groundwaters was 366.9 mg/L. The highest Na+ concen- HCO − > SO 2−3 4 . Concentrations of Cl
− varied from
tration (1430 mg/L) occurred in borehole CR2-53 at 30.6 mg/L in borehole CR4-FZ-08 at Ayeldu to
Abonko and the lowest (26.40 mg/L) in borehole CR4- 4799 mg/L in borehole CR2-49 at Ekumfi Akwakrom
FZ-14 at Abora Dunkwa. The dominant anion in the (Fig. 6(b)), with a mean of 595.4 mg/L. High values of
shallow groundwaters was Cl−, followed by HCO −3 and 1210–4799 mg/L were found in boreholes CR2-23 at
SO 2−4 . The Cl
− in the shallow groundwaters ranged Ekumfi Engow, CR2-45 at Gomoa Abora and CR2-49
between 39.4 and 2750 mg/L, with a mean of 788 mg/L. at Akwakrom (Fig. 6(b)). These values also corre-
Higher concentrations of Cl− (2750 mg/L) occurred in sponded to high concentrations of Na+. The least
1306 S. Y. GANYAGLO ET AL.
dominant anion in the deep groundwater was SO 2−4 ,
ranging between 7.5 and 751.6 mg/L, with a mean of
129.1 mg/L. The highest SO 2−4 concentration of
751.6 mg/L occurred in borehole CR4-04 at Otuam in
the aquifer underlain by biotite granite. Extremely low
HCO −3 values of 9.2 mg/L occurred further north in
the biotite granite and biotite gneiss aquifers further
from the coast. The generally low NO −3 concentrations
in the study area varied from 0.02 to 2.3 mg/L, with a
mean of 0.5 mg/L. Similarly, the concentrations of
PO 3−4 and F
− were low in the entire study area.
4.2 Stable isotope composition of groundwater
Detailed results of the stable isotope compositions of
the groundwaters are presented in the Supplementary
material, Table S3. Stable isotope compositions of the
shallow groundwaters in the study area for δ18O ranged
between −3.09‰ and −1.60‰, with a mean of −2.44‰
and a standard deviation of 0.50 (Supplementary mate-
rial, Table S4(a)). The stable isotopic composition for
δ2H of shallow groundwater ranged between −12.35‰
and −6.07‰, with a mean of –9.25‰ and a standard
deviation of 2.30 (Supplementary material, Table S4
(a)). In deep groundwater, the isotopic composition
for δ18O ranged between –3.07‰ and –1.43‰, with a
mean of –2.43‰ and a standard deviation of 0.41
(Supplementary material, Table S4(b)). The δ2H for
deep groundwater ranged between –17.61‰ and –
7.04‰, with a mean of –11.99‰ and a standard devia-
tion of 2.73 (Supplementary material, Table S4(b)). The
spatial distribution of δ18O in the study area showed
waters with more positive δ18O inland and more nega-
tive δ18O towards the coast (Fig. 7(a)). The δ18O and
δ2H data are plotted in Figure 7(b). As revealed in
Figure 7(b), the groundwaters are indicative of waters
of meteoric origin, i.e. samples close to the Global
Meteoric Water Line (GMWL) and Local Meteoric
Water Line (LMWL). Shallow and deep groundwaters
show similar isotopic compositions suggesting a
hydraulic connection between shallow groundwater
and deep groundwater as observed from the chemical
data. This implies deep groundwater is recharged ver-
tically from the unsaturated soil zone.
5 Discussion
5.1 Hydrochemistry
Shallow and deep groundwater in the study area exhib-
ited similar sequences of abundance of cations and anions
signifying a hydraulic connection between them. The
highest EC of 29 000 μS/cm in borehole CR2-45 at
Table 1. Statistical summary of hydrochemical data of (a) shallow groundwater and (b) deep groundwater.
Parameter Temp (°C) pH Eh (mV) EC (μS/cm) TDS (mg/L) Salinity Ca2+ (mg/L) Na+ Mg2+ (mg/L) K+ (mg/L) Cl−(mg/L) SO 2−(mg/L) HCO − (mg/L) NO −4 3 3 (mg/L)
(mg/L) (mg/L)
(a) Shallow groundwater
Minimum 5.20 −104.10 540.00 270.00 300.00 9.34 26.40 8.33 4.45 39.38 20.50 12.20 0.15
26.90
Maximum 9.30 104.50 6050.00 3030.00 3300.00 183.20 1429.50 168.00 468.50 2749.50 184.67 277.53 0.81
30.80
Mean 13.92 2192.00 1097.78 1144.44 68.54 366.93 45.14 83.83 788.00 88.18 141.31 0.51
28.54
Parameter Temp (OC) pH Eh (mV) EC (μS/cm) TDS (mg/L) Sal Ca2+ (mg/L) Na+ Mg2+ (mg/L) K+ (mg/L) Cl−(mg/L) SO 2−4 (mg/L) HCO
−
3 (mg/L) NO
−
3 (mg/L)
(mg/L) (mg/L)
(b) Deep groundwater
Minimum 23.20 5.32 −36.60 120.60 60.30 100 10.87 22.15 2.88 3.70 30.57 7.46 9.15 0.02
Maximum 30.90 7.84 107.40 29000.00 14420.00 17700 456.70 2769.50 416.70 101.90 4798.50 751.55 780.80 2.26
Mean 27.91 44.16 2346.92 1183.59 1298.57 109.33 335.96 44.76 18.72 595.36 129.09 2.26 0.55
Parameter F− (mg/L) PO 3−4 (mg/L) Br
−
(mg/L)
Minimum 0.00 0.00 0.00
Maximum 0.79 0.51 219.00
Mean 0.31 0.07 41.77
HYDROLOGICAL SCIENCES JOURNAL – JOURNAL DES SCIENCES HYDROLOGIQUES 1307
Figure 5. Distribution of EC levels of the groundwater with distance from the coast. The classification of the waters into fresh,
brackish and saline is based on the scheme employed by Park et al. (2012).
Figure 6. Spatial distribution of (a) Na+ and (b) Cl− with respect to the geology of the area.
Gomoa Abora can be explained by the proximity of the cm were classified as “brackish”, and those with EC levels
coast, and also by weathering of minerals of the rocks in >3000 μS/cm were described as “saline”. Plotting of EC
the area. Park et al. (2012) evaluated the relationship values in the Ochi-Narkwa basin against distance from
between EC levels and distance from the coast in South the coast (Fig. 5) reveals a weak negative correlation (r = –
Korean aquifers and observed negative correlation in 0.41), implying that salinity does not depend on distance
alluvial coastal aquifers, but no correlation in bedrock from the coast. According to the classification scheme of
coastal aquifers. They developed a classification scheme Park et al. (2012), Figure 5 shows that some boreholes
to describe the quality of water in South Korea. within 7 km of the coast had EC of 100–1000 μS/cm and,
Groundwaters with EC < 1500 μS/cm were classified as therefore, can be classified as fresh. Some boreholes and
“fresh”, those having EC within the range 1500–3000 μS/ hand-dug wells within the same distance from the coast
1308 S. Y. GANYAGLO ET AL.
Figure 7. (a) Spatial distribution of δ18O in the study area. (b) Relationship between δ2H‰ V-SMOW and δ18O‰ V-SMOW for
shallow and deep groundwater.
had EC above 1000 μS/cm and therefore they are saline. Generally, Cl− is not a significant constituent of
From 7 km to about 15 km from the coast, some bore- silicate rocks. The presence of Cl− in groundwater is
holes had an EC below 1000 μS/cm, indicating freshwater. usually attributed to atmospheric sources, decomposi-
Similar observations can be made in Figure 2, which tion of organic matter and trace impurities in rocks
shows the spatial distribution of EC with respect to the and minerals (Freeze and Cherry 1979). It may also
geology of the study area. The origin of the electrical result from seawater encroachment and intrusion due
conductivities in the area is complex, indicating weath- to proximity to the coast. Like Cl−, SO 2−4 is not a major
ering of rocks, seawater intrusion, sea aerosol spray, halite constituent of silicate rocks and therefore its content in
dissolution and slow movement of groundwater in the the study area is low. The elevated SO 2−4 concentration
area. Of the deep groundwaters, 59% had EC < 1500 μS/ of 751.6 mg/L in the biotite granite aquifer may be
cm and 41% had EC > 1500 μS/cm. Figure 2 indicates the explained by oxidation of pyrite observed in the rocks
possible groundwater flow direction from the inland of the area. Very low NO −3 concentrations may be
areas in the north with low EC values towards the coast attributed to less vigorous anthropogenic activities in
in the south with increasing EC. This agreed with the flow the area. Low F− and PO 3−4 concentrations in deep
direction obtained in Figure 3 from hydraulic heads. groundwater may be attributed to lack of phosphate
The cations Na+, Ca2+, Mg2+ and K+ are significant minerals and fluorites in the rocks of the study area.
constituents of silicate rocks (Freeze and Cherry 1979) Elevated concentrations of F− in groundwater in
and hence Na+ in groundwater in the Ochi-Narkwa basin Northern Ghana, however, are attributable to the dis-
is considered to originate from silicate weathering. A solution of the mineral fluorite in the rocks (Apambire
secondary source of Na+ is halite, concentrated in the 1997).
soil zone by evaporation and leached into the ground-
water by infiltrating rain. The Ca-feldspars (CaAl2Si2O8)
are considered the main source of calcium, releasing Ca2+ 5.2 Hydrochemical facies
in the presence of carbonic acid (H2CO3) generated in the The Piper (1944) trilinear plot of shallow unconfined
soil zone. Other possible sources of Ca2+ in the ground- groundwaters showed two main water types: NaCl and
waters include dissolution of the mineral hornblende non-dominant (mixed) (Fig. 8(a)). In the case of deep
[Ca (MgFeAl) (AlSi) O ] and pyroxenes. The Mg2+2 5 8 22 in confined groundwater, four main water types were
groundwater probably comes from biotite [K(Mg,Fe)3 identified (Fig. 8(b)): Ca-Mg-HCO3, labelled I; NaCl,
(AlSi3)O10(OH,F)2] and hornblende, which are constitu- labelled II; Ca-Mg-Cl-SO4, labelled III; non-dominant,
ents of the aquifers in the study area. Micas (muscovite labelled IV. The groundwater evolved from a Ca-Mg-
and biotite) are major constituents of the sediment/vol- HCO3 water type (freshwater) in a recharge area to a
caniclastic sediment in the area and therefore responsible NaCl water type (saline), indicating a discharge area. A
for K+ in the groundwater. general flow direction is thus shown in Figure 8(b)
HYDROLOGICAL SCIENCES JOURNAL – JOURNAL DES SCIENCES HYDROLOGIQUES 1309
Figure 8. Hydrochemical facies in the study area using the Piper (1944) trilinear diagram for (a) shallow unconfined groundwater,
(b) deep confined groundwater, and (c) unconfined and confined groundwater plotted together; (d) Scholler diagram showing the
relationship between unconfined and confined groundwater.
from inland (freshwater zone) to the coast (saline produced by waves breaking from the ocean burst
zone), as observed in Figure 3. One shallow and two after rising back to the air–sea interface and produce
deep groundwater samples located near the seawater seasalt aerosols (Keene et al. 1986). The seasalt aerosols
zone of the Piper trilinear diagram in Figure 8(c) produced are removed from the atmosphere by preci-
indicate that they have been affected by seawater intru- pitation. The chemistry of precipitation thus becomes
sion. Generally, the most common hydrochemical dominated by marine components (Keene et al. 1986).
facies in both shallow unconfined and deep confined The various ionic species contributing to seasalt in the
groundwaters is NaCl. The Schoeller diagram in Ochi-Narkwa study area were calculated using linear
Figure 8(d) is used to explain the relationship between regression techniques based on three major
unconfined and confined aquifers in the study area. assumptions:
Although both aquifers have similar hydrochemical
patterns, the confined aquifers are more mineralized (1) all reference species are contributed by seasalt;
than the unconfined aquifers. This may be explained by (2) no fractionation occurs during formation of the
lower transmissivities observed in confined aquifers in aerosol; and
the area, which in turn result in slow movement and (3) no fractionation occurs during atmospheric
longer transit times of groundwater. The spatial distri- transport and scavenging.
bution of the various water types in Figure 9 shows that
NaCl waters predominantly occur in the mica schist If the species of interest are contributed solely by seasalt,
and in part of the biotite granite aquifers. a regression of the concentrations of the species of interest
and reference species yields a slope approximately equal to
the seawater ratio and intercept approximately equal to
5.3 Origin of salinity in the groundwaters
zero. A slope that differs from the seawater ratio and an
The ocean is a major source of aerosols and gases for intercept that differs from zero are indications of a non-
the overlying atmospheric boundary layer. Bubbles seasalt contribution. Keene et al. (1986) calculated various
1310 S. Y. GANYAGLO ET AL.
Figure 9. Spatial distribution of water types in the study area.
ionic ratios in seasalt including the Na/Cl ratio and found indicating silicate weathering. This is further confirmed
them to be 0.86. This ratio has been employed widely to by the plot of TDS against Na/Na+Ca (Fig. 10(c)), as
determine the origin of salinity in groundwater (De proposed by (Singh et al. 2011), where most of the
Montety et al. 2008, Al-Charideh 2011, Cheong et al. samples are plotted in the rock dominance region,
2011). Groundwater having a Na/Cl molar ratio equal to accounting for a weathering origin. Three percent of
0.86 is considered to have originated from seasalt or sea- the samples have Na/Cl molar ratio less than 1 and
water and halite dissolution. suggest possible seawater intrusion.
The plot of Na+ against Cl− concentrations in These findings were further examined using the Br−/
groundwater in the study area (Fig. 10(a)) shows a Cl− ratio as a reliable indicator of the origin of salinity
strong correlation of 0.93, indicating most likely a due to its specific composition in various saline sources
common source of saline water. The Na–Cl relation- (Vengosh et al. 2005, De Montety et al. 2008, El-Fiky
ship in Figure 10(a) and (b) shows most of the samples 2010). The Br−/Cl− ratio of seawater observed by
along the 1:1 line, indicating that the sources of Na+ Helstrup et al. (2007) in parts of Ghana and Togo
and Cl− in the groundwater may originate from seasalt close to the Gulf of Guinea was 0.0035 ± 0.0002 (weight
aerosols, direct seawater intrusion and halite dissolu- ratio). The Br−/Cl− ratios of the groundwaters in the
tion. The geology of the area does not indicate halite Ochi-Narkwa basin ranged from 0.0054 to 2.075. Four
deposits. It is therefore supposed that halite, originat- of the boreholes (CR2-01 at Kweikrom, CR2-22 at
ing from high concentrations of Cl− in rainfall (Fig. 4), Ekumfi Techiman, CR2-23 at Ekumfi Engow and
accumulated over time in the soil zone by evaporation CR2-45 at Gomoa Abora) are plotted close to the sea-
and was flushed into the groundwater zone by infil- water line and therefore show characteristics of sea-
trated rainwater. The Na/Cl molar ratios of the water transgression into groundwater, sea aerosol spray
groundwaters ranged from 0.36 to 5.18, and 83% of or halite dissolution (Fig. 10(d)). The remaining sam-
the samples have a Na/Cl molar ratio equal to 1, ples occur above the Br−/Cl− ratio for seawater, sug-
suggesting that halite dissolution is a major hydro- gesting salinity sources other than seawater intrusion.
geochemical process in the area. A ratio equal to 1 However, Figure 10(d) reveals a general decrease in
implies that halite dissolution is responsible for con- Br−/Cl− ratio with increasing Cl− concentration, as
tributing Na+ ions and a ratio greater than 1 means shown by the arrow AB, suggesting a tendency to sea-
that silicate weathering would contribute to Na+ ions in water characteristics.
the groundwater (Singh et al. 2011). In this study, 14% The SO 2− −4 /Cl ratio was also used to investigate a
of the samples have a Na/Cl molar ratio greater than 1, possible seawater intrusion. Marie and Vengosh (2001)
HYDROLOGICAL SCIENCES JOURNAL – JOURNAL DES SCIENCES HYDROLOGIQUES 1311
Figure 10. (a) and (b) Na–Cl relationship of groundwater in the coastal areas of the Central Region. (c) Scatter plot between TDS
and Na/(Na+Ca) showing rock-dominant weathering in the area (after Singh et al. 2011). (d) Br−/Cl− weight ratio vs Cl− (mg/L).
recorded a seawater ratio for SO 2−/Cl−4 of 0.15; a ratio an average of 5.8 m. The borehole depths (bottom of
greater than 0.15 indicates additional SO 2−4 input from borehole) in the area were 18–94.5 m, with a mean of
fertilizer application, leaching of contaminated landfill 34.5 m. The depths to water levels show that ground-
or oxidation of sulphides (Marie and Vengosh 2001). A water is accessed close to the surface. For the shallow
plot of SO 2−4 against Cl
− from the Ochi-Narkwa study groundwaters the depths ranged between 6 and 12 m.
area shows a moderate positive correlation of 0.62 Both shallow and deep groundwaters have similar
(Fig. 11(a)). SO 2− −4 /Cl ratios for samples collected in hydrochemical facies and similar δ
18O and δ2H com-
the study area ranged between 0.02 and 4.09. Of the positions, implying recharge from the shallow ground-
samples, 68% had SO 2−4 /Cl
− ratios above the seawater water to deep groundwater via a preferential path. This
value, indicating additional sources of SO 2−4 into the exhibits mixing of shallow groundwater and deep
groundwater system (Fig. 11(b)); 32% had SO 2−4 /Cl
− groundwater, hence the mixing line.
ratios either equal to or less than the seawater value, The majority of the boreholes have been con-
hence showing characteristics of seawater origin. structed at depths closer to the surface and are there-
In Figure 7(b) both shallow and deep groundwaters fore affected by evaporation. Salt layers in the
deviate from the GMWL and the LMWL. The ground- unsaturated zone are therefore possible. The effect
water regression line is defined by the equation of a possible seawater intrusion was also investigated
δ2H = 3.44δ18O – 3.27. A regression line of slope 3.44 by plotting Cl− against δ18O‰ (Fig. 12), which
suggests either evaporation from the soil zone before showed no significant relationship between the para-
recharge to the aquifers or mixing of shallow ground- meters. Figure 12 depicts that there is no distinct
water and deep groundwater. Deviation from the change in δ18O‰ with increasing Cl− concentration,
GMWL and LMWL with a slope of less than 8 provides implying that salinity originates dominantly from
evidence of evaporation from the soil zone. The depths dissolution of salts from soils and rocks. This
to water levels in the study area support this assertion. explains why the majority of the groundwaters are
The depths to water levels, as indicated in Section 2, located in the dissolution band of the diagram. Two
varied between 0 and 20 m below the land surface with boreholes appear in the mixing band, suggesting
1312 S. Y. GANYAGLO ET AL.
Figure 11. (a) SO 2−4 (meq/L) vs Cl
− (meq/L), and (b) SO 2−/Cl−4 vs Cl (mg/L) .
Figure 12. Relationship between Cl (mg/L) and δ18O‰ V-SMOW.
seawater intrusion. One of the hand-dug wells shows major Cl− peak occurs at 140 cm (249.2 mg/kg). Thus,
elevated Cl− concentration as a result of evaporation. Cl− had probably already accumulated in this zone over
The Cl− concentration data from four soil profiles time. The lower Cl− values below 140 cm depth indicate
(Fig. 13) in the vicinity of Ekumfi Akwakrom near that Cl− is being dissolved and leached to deeper hori-
Ekumfi Asokwa were used to assess the presence of Cl− zons. Minor peaks are observed between 20 and 100 cm.
in the unsaturated soil. The Cl− profiles displayed large The Cl− peaks show periods of maximum accumulation
variations in Cl− concentration: low concentrations of in the unsaturated soil zone. It can be concluded that
50 mg/kg occurring between 20 and 60 cm and higher NaCl zones occur in lenses and support the existence of
concentrations (99.9–449.8 mg/kg) between 80 and salt crusts at different depths, notably between 80 and
120 cm below the surface (Fig. 13). All measured Cl− 120 cm, and therefore that dissolution of these salts in the
profiles are bulge shaped, with decreasing Cl− concentra- soil zone contributes to the high salinity in the ground-
tions below the bulge at depth. The upper parts of pro- waters, as revealed by results from both geochemical and
files 1, 3 and 4 reveal a constant Cl− value of 49.9 mg/kg, stable isotopic considerations.
indicating a period of non-deposition of Cl−, or that Cl− The δ13C of total dissolved inorganic carbon
may have been leached to deeper zones of the profiles (TDIC) of groundwater was employed to further
and eventually to the saturated zone (Fig. 13(a), (c) and investigate possible seawater intrusion processes. As
(d)). In the case of Profile 2 (Fig. 13(b)), a minor Cl− peak the geology of the area does not contain carbonate
occurs at the upper part of the profile (49.9 mg/kg), and a rocks, carbonate minerals could only occur by
HYDROLOGICAL SCIENCES JOURNAL – JOURNAL DES SCIENCES HYDROLOGIQUES 1313
Figure 13. Variation of Cl− (mg/kg) concentration in the unsaturated soil zone with depth (cm) .
secondary mineralization. It is therefore assumed seawater intrusion is the cause of salinity of ground-
that δ13C of seawater is unlikely to be affected by water, an increase in Cl− concentration of ground-
geochemical reactions and thus be conserved for water with a corresponding enrichment of δ13C
tracing of seawater intrusion in the groundwaters in content of groundwater will occur. A plot of δ13C
the study area. against Cl− (mg/L) for 24 groundwater samples in
Carbon-13 (δ13C) content of seawater usually var- the study area reveals no such correlation (r = 0.07),
ies from –2 to 2‰ V-PDB. In the Mediterranean Sea, indicating that the samples are not of a common
Yechieli et al. (2001) reports a value of 0‰ V-PDB. source (Fig. 14). It can also be deduced from
The δ13C for the ocean is quoted as 0‰ V-PDB, while Figure 14 that δ13C becomes more depleted with
that of groundwater ranges from 0‰ to –20‰ V-PDB increasing Cl− concentration. This indicates that the
(Clark and Fritz 1997). If groundwater was intruded significant component of the dissolved inorganic car-
by seawater, δ13C values would be close to those of the bon (DIC) in the Ochi-Narkwa basin is not derived
ocean or show δ13C enrichment towards the value of from marine sources, but rather from biogenic
the ocean. Carbon-13 of groundwater in the study sources. This indicates that groundwater salinization
area varied from –19.30 to –7.94‰ V-PDB, with a may not be as a result of seawater intrusion.
mean of −14.56‰ V-PDB. The groundwaters in the
study area are more depleted in δ13C compared to the
13 5.4 Conceptual flow model of the study areaocean. Applying δ C‰ V-PDB as an indicator of
seawater intrusion to coastal aquifers, δ13C values of A conceptual flow model showing the mechanism of
saline groundwater and non-saline groundwater are salinization in the study area has been proposed
plotted against chloride concentrations in mg/L. If (Fig. 15). It reveals that groundwater flows
1314 S. Y. GANYAGLO ET AL.
Figure 14. Plot of δ13C (‰) versus V-PDB against Cl− (mg/L) in the Central Region.
dominantly from the northwestern section (inland) 6 Summary and conclusion
of the study area to the southeastern section
Anion chemistry of rainfall showed a higher Cl− concen-
(towards the coast). Evaporation over the sea sur-
tration at the coast than inland. Shallow and deep
face results in the formation of suspended particles
groundwaters showed similar hydrochemical patterns,
in the atmosphere referred to as sea aerosols, mostly
dominated by Na+ and Cl−. Cation concentrations
rich in NaCl. The aerosols are deposited on the
occurred in the order Na+ > Ca2+ > Mg2+ > K+ and the
ground surface. Rain infiltrating into the soil dis-
anions in the order Cl− > HCO − > SO 2−. The major
solves the salts on the surface, which are precipi- 3 4
hydrochemical facies of the shallow groundwater was Na-
tated as lenses of salt crust in the unsaturated soil
Cl and the major facies of the deep groundwater were Ca-
zone. These salts are later leached into the ground-
Mg-HCO , Na-Cl, Ca-Mg-Cl-SO and non-dominant
water, causing salinization of groundwater as 3 4
water types. Of the deep groundwaters, 59% had EC <
demonstrated in Figure 15. Rock weathering also
1500 and 41% had EC > 1500 μS/cm. A conceptual model
accounts for salinization of groundwater in the
of groundwater flow and processes of groundwater
study area.
Figure 15. Conceptual flow model explaining the mechanism of salinization in the study area (modified from Geological Survey
field sheet no. 32).
HYDROLOGICAL SCIENCES JOURNAL – JOURNAL DES SCIENCES HYDROLOGIQUES 1315
salinization was developed. Groundwater flow was Apambire, W.B., 1997, Geochemistry, genesis, and health
hypothesized from the inland areas (northwestern section implications of fluoriferous groundwaters in the upper
of the area) towards the coast (southeastern section). regions of Ghana. Environmental Geology, 33 (1), 13-24.
Armah, T., 2002. Hydrochemical and geophysical studies
Elevated EC in the study area was attributed to the of groundwater salinity, Central Region, Ghana. Thesis
impervious nature of the rocks, which results in longer (PhD). University of Ghana.
contact time with the rocks, slow groundwater movement Boghici, R., 2003. A field manual for groundwater sampling,
and therefore a greater amount of dissolved minerals. user manual 51 [online]. TX: Texas Water Development
The Na/Cl molar ratio within the range 0.36–5.18 Board. Available from: http://www.twdb.texas.gov/publica
revealed that the principal source of high groundwater tions/reports/manuals/UM-51/FieldManual.pdf. [Accessed
18 November 2016].
salinity is halite dissolution, because 83% of the samples Bouchaou, L., et al., 2008. Application of multiple isotopic
had ratios equal to 1. Fourteen percent had ratios greater and geochemical tracers for investigation of recharge, sal-
than 1, suggesting that silicate weathering or rock disso- inization, and residence time of water in the Souss–Massa
lution contributed to the salinity sources. Three percent aquifer, southwest of Morocco. Journal of Hydrology, 352,
of the groundwater samples showed characteristics of 267–287. doi:10.1016/j.jhydrol.2008.01.022
Cheong, J.Y., et al., 2011. Groundwater nitrate contamina-
seawater intrusion. These findings were also supported tion and risk assessment in an agricultural area, South
by the analysis of Br−/Cl− ratio versus Cl, the ratio SO 2−4 / Korea. Environmental Earth Sciences, 66, 1127–1136.
Cl− and the Cl−–δ18O relationship. doi:10.1007/s12665-011-1320-5
The Cl− concentrations in the unsaturated soil zone Clark, I.D., and Fritz, P., 1997. Environmental isotopes in
indicated the occurrence of NaCl salt crusts at different hydrogeology. Boca Raton, FL: CRC Press.
depths, notably between 80 and 120 cm. The 13C Cl− Coplen, T.B., 1996. New guidelines for reporting stableδ –
hydrogen, carbon, and oxygen isotope-ratio data.
relationship indicated no impact of seawater intrusion. Geochimica et Cosmochimica Acta, 60 (17), 3359–3360.
The overall assessment of groundwater in the study doi:10.1016/0016-7037(96)00263-3
area revealed a strong geological background to the De Montety, V., et al., 2008. Origin of groundwater salinity
major ion hydrochemistry in the coastal zone of and hydrogeochemical processes in a confined coastal
Central Ghana with no significant seawater intrusion. aquifer: case of the Rhône delta (Southern France).
Applied Geochemistry, 23, 2337–2349. doi:10.1016/j.
apgeochem.2008.03.011
Acknowledgements El-Fiky, A.A., 2010. Hydrogeochemical characteristics and
evolution of groundwater at the Ras Sudr-Abu Zenima
We thank the staff of the isotope hydrology and the Area, Southwest Sinai, Egypt. JKAU: Earth Sciences, 21
inorganic laboratories for analysing the samples, in par- (1), 79–109. doi:10.4197/Ear.21-1.4
ticular Nash Bentle, Eunice Agyeman, Ruby Torto and Freeze, R.A. and Cherry, J.A., 1979. Groundwater prentice.
Godfred Anyanu. The Community Water and Sanitation Englewood Cliffs, NJ: Englewood Cliffs.
Agency (CWSA), Central Region, is acknowledged for Helstrup, T., Jørgensen, N.O., and Banoeng-Yakubo, B.,
providing existing hydrogeological and hydrochemical 2007. Investigation of hydrochemical characteristics of
data on boreholes. groundwater from the Cretaceous-Eocene limestone
aquifer in southern Ghana and southern Togo using
hierarchical cluster analysis. Hydrogeology Journal, 15,
Disclosure statement 977–989. doi:10.1007/s10040-007-0165-1
Hirdes,W.,Davis,D.W., andEisenlohr, B.N., 1992. Reassessment
No potential conflict of interest was reported by the authors. of Proterozoic granitoid ages in Ghana on the bases of U/Pb
zircon and monazite dating. Precambrian Research, 56 (1–2),
89–96. doi:10.1016/0301-9268(92)90085-3
Funding IAEA (International Atomic Energy Agency), 2008. Field
estimation of soil water content––A practical guide to
This research work is based on the technical cooperation methods, instrumentation and sensor technology.
project GHA/8009 supported by the International Atomic Vienna: IAEA-TCS-30. ISSN 1018–5518.
Energy Agency (IAEA). We would also like to thank the Jørgensen, N.O. and Banoeng-Yakubo, B.K., 2001.
National Nuclear Research Institute (NNRI) of the Ghana Environmental isotopes (18O, 2H, 87Sr/86Sr) as a tool in
Atomic Energy Commission (GAEC) for their financial and groundwater investigations in the Keta Basin, Ghana.
logistic support in conducting this study [GHA/8009]. Hydrogeology Journal, 9 (2), 190–201. doi:10.1007/
s100400000122
Keene, W.C., et al., 1986. Seasalt corrections and interpreta-
References tion of constituent ratios in marine precipitation. Journal
of Geophysical Research, 91 (6), 6647–6658. doi:10.1029/
Al-Charideh, A., 2011. Geochemical and isotopic characteriza- JD091iD06p06647
tion of groundwater from shallow and deep limestone aquifers Kim, Y., et al., 2003. Hydrogeochemical and isotopic evidence of
system of Aleppo basin (north Syria). Environmental Earth groundwater salinization in a coastal aquifer: a case study in
Sciences, 65, 1157–1168. doi:10.1007/s12665-011-1364-6
1316 S. Y. GANYAGLO ET AL.
Jeju volcanic island, Korea. Journal of Hydrology, 270 (3–4), Park, Y., et al., 2012. National scale evaluation of ground-
282–294. doi:10.1016/S0022-1694(02)00307-4 water chemistry in Korea coastal aquifers: evidences of
Kortatsi, B.K., 2006. Hydrochemical characterization of seawater intrusion. Environmental Earth Sciences, 66,
groundwater in the Accra plains of Ghana. 707–718. doi:10.1007/s12665-011-1278-3
Environmental Geology, 50 (3), 299–311. doi:10.1007/ Piper, A.M., 1944. A graphic procedures in the geochem-
s00254-006-0206-4 ical interpretation of water analyses. American
Kortatsi, B.K., 1994. Groundwater utilization in Ghana. In: J. Geophysical Union Trans, 25, 914–923. doi:10.1029/
Soveri and T. Suokko, eds. Proceedings of the helsinki con- TR025i006p00914
ference on groundwater resources at risk, Helsinki, Finland. Singh, C.K., Shashtri, S., and Mukherjee, S., 2011. Integrating
Vol. 222. Wallingford, UK: International Association of multivariate statistical analysis with remote sensing and
Hydrological Sciences, IAHS Publ, 149–156. GIS for geochemical assessment of groundwater quality: a
Leube, A., et al., 1990. The early proterozoic birimian super- case study of Rupnagar district in Shiwaliks of Punjab.
group of Ghana and some aspects of its associated gold Environmental Earth Sciences, 62 (7), 1387–1405.
mineralization. Precambrain Research, 46, 139–165. doi:10.1007/s12665-010-0625-0
doi:10.1016/0301-9268(90)90070-7 Taylor, P.N., et al., 1992. Early Proterozoic crustal evolu-
Marie, A. and Vengosh, A., 2001. Sources of salinity in ground- tion in the Birimian of Ghana: constraints from geo-
water from Jericho Area, Jordan Valley.Groundwater, 39 (2), chronology and isotope geology. Precambrian Research,
240–248. doi:10.1111/gwat.2001.39.issue-2 56, 97–111. doi:10.1016/0301-9268(92)90086-4
Obuobie, E. and Barry, B., 2010. Groundwater in sub- Vengosh, A., et al., 2005. Sources of salinity and boron in the
Saharan Africa: implications for food security and liveli- Gaza Strip: National contaminant flow in southern
hoods. Ghana Country Status on Groundwater [online]. Mediterranean coastal aquifer. Water Resources Research,
Final Report, International Water Management Institute, 41, W01013. doi:10.1029/2004WR003344
48. Available from: http://gwafrica.iwmi.org/Data/Sites/24/ Yechieli, Y., et al., 2001. Radiocarbon in seawater intruding
media/pdf/Country_Report-Ghana.pdf [Accessed 17 into the Israeli Mediterranean coastal aquifer.
November 2016]. Radiocarbon, 43, 773–781.