The capital territory of Pakistan constitutes twin cities, Islamabad and Rawalpindi. Islamabad is the capital city, located at the foot of the Margallah Hills, whereas Rawalpindi lies in the Potohar Plateau. Both cities are located in the semi-arid region of Pakistan, where the residents meet their basic need of water through groundwater resources. For the last many years, the quantity and quality of groundwater in these cities has been deteriorating very rapidly. In this study, the foremost recharge source for these cities was identified using tritium and major ion chemistry, and different physiochemical parameters were studied to find out the facts behind the quality deterioration of the groundwater. Tritium values and chemical data suggested that aquifers located in the territory of Islamabad were mainly recharged by upland areas (the Margalla Hills), which accounts for their low electrical conductivity and total dissolved solids contents. A higher sodium adsorption ratio (SAR) suggested an alteration in recharge patterns through soil compaction and cementation (from increased construction activity) that reduced the recharge inputs. The high SAR also disturbs the recharge pattern and had disturbed the natural equilibrium of the groundwater system, deteriorating the quality of the water.

INTRODUCTION

Most of world's arid regions experience increase the dependence on groundwater as a consequence of changes in rainfall patterns and the inconsistency of surface water (Sundary et al. 2006). An aquifer is a geological formation with saturated permeable material that contains groundwater and yields usable quantities of water to boreholes, wells or springs (Lohman 1972). Water in aquifers is held in empty spaces within the rock formations. These geological formations are composed of low and highly permeable rocks that are inter-bedded with each other. Water movement within different layers with a variety of mineral compositions may result in considerable changes in water composition. Groundwater chemistry provides important clues about the processes and dynamics that cause changes in water chemistry (Nelson 2002). There are many factors upon which groundwater chemistry depends, including the aquifer composition through which water flows, recharge source, contamination through land-use activities, ion exchange processes, mixing processes and groundwater age.

In Pakistan, due to the arid climatic conditions, groundwater availability is limited in many parts of the country. The surface and groundwater reserves of Pakistan are about 128,300 million m3 and 50,579 million m3 per year, respectively (Islam et al. 2008). In the current scenario, the residents of Islamabad and Rawalpindi largely depend on groundwater due to the limited, less consistent and polluted surface water reservoir. Increase water demands provoke over exploitation of groundwater, which adversely impacts the quality and quantity of the available groundwater reservoirs. Hence management of this fresh water resource nowadays becomes a pre-requisite to ensure the future water supply (Han et al. 2014). In this context, age assessments provide a detailed database on recharge patterns, contaminant transport and the source of groundwater recharge (Aggarwal et al. 2007). The total time period during which groundwater remains isolated from atmospheric conditions is considered as the age of the groundwater or its residence time (Mozter 2007).

Tracers of modern meteoric water should provide either a characteristic signal that can be preserved in recently recharged waters, or an input that decays once removed from the meteoric source. Tritium (3H or T) has both. It is a short-lived, naturally produced radioactive hydrogen isotope that has a half-life of 12.3 years. It also has a characteristic signal from atmospheric tests of nuclear weapons in the 1960s. Both the decay and position of the thermonuclear peak can provide groundwater age information. The presence of tritium in groundwater, at its most fundamental level of interpretation, is an indication of modern recharge. Tritium (H3) data are widely used in hydrological investigations, particularly in arid and semi-arid areas, to assess the water age/recharge source. Unlike CFCs, 85Kr and other modern tracers, which require very strict sampling protocols, contamination by tritium in atmospheric water vapor during sampling is not a concern (Lucas & Unterweger 2000). Naturally, tritium is produced by cosmic ray neutron and nitrogen (N14) interaction in the upper atmosphere, and is directly incorporated into precipitation (Brown & Grummitt 1956). 
formula
Furthermore, emitted H3 reacts with stratospheric oxygen to form water molecules. Precipitation is the main mode of transport taking this atmospheric water to the ground; later on this water mixes with surface and groundwater components. Tritium decays to helium-3 by emitting low energy (Emax = 18 keV). Large quantities of tritium have been added by human activities such as nuclear bomb testing, which sharply increased the H3 atmospheric concentration to 1,000 TU in the environment. Water with a high tritium level is significant in groundwater development, as it depicts greater water replenishability for that part of the groundwater formation (Moncaster et al. 2000).

In the current investigation, recharge of the aquifer in the study area has been investigated for the first time. Hydrochemical data, including major cations (Ca+2, Mg+2, Na+, K+) and major anions (HCO3, NO3, SO4−2, Cl), are also discussed in order to find the relation of these parameters with tritium and to describe water quality as suggested by other researchers (Bartolino & Cunning 2012; Zhang et al. 2012). Moreover, the outcomes of the study contribute to further investigations and applications as well as management strategy recommendations. This study is expected to provide a better understanding of groundwater resource management in arid and semi-arid zones.

MATERIALS AND METHODS

Study area

The capital territory of Pakistan constitutes the twin cities, Islamabad and Rawalpindi, which are located between the longitude of 72°45′-73°30′N and latitude of 33°30′-33°50′E. Islamabad is a capital territory located at the foot of the Margallah Hills and Rawalpindi lies in the Potohar Plateau. Both cities are located in the semi-arid region of Pakistan, where the residents meet their basic need for water through groundwater resources. This semi-arid sub-tropical climatic region is characterized by dry, cool winters and rainy, hot summers. Monsoon rainfall usually starts in June and ends by September. Annual rainfall ranges from 990 to 1,000 mm. A peak temperature of 45.9 °C was observed in June 1972 and the lowest recorded temperature was −3.9 °C.

Hydrology of the area explains that the ‘Soan’ and ‘Korang’ rivers act as the main surface water reserves emerging from the Margalla and Murree Hills. These two rivers supply water to Simbli and Rawal Lake respectively, from which water is consumed by urban dwellers. These water bodies are drained by many perennial streams that include Gumrah Kas, Tanawala Kas, Bedarwali Kas and the Lai drain. Alluvial and hard rock deposits carry groundwater in these areas (Sajjad et al. 1998; Munir et al. 2006).

At Margalla Hills, the water table is at a depth of 600 m, which decreases near Soan River to 450 m. The saturated zone lies between 2 and 20 m below the surface. The water table of the study area is on average 7–15 m deep. Wells located on the bank of the Lai drain show a water table 7–10 m deep (Munir et al. 2011). The lowest depth of water table near the Lai drain has led construction of most of the wells being in the vicinity of that area. There is a general groundwater flow towards the center of the basin from the Margalla and Murree Hills. The groundwater discharges its contents in the Korang River. The dominant aquifer system of the area is built up from a very heterogeneous alluvial fill of alternating layers of clay, silt, sand and gravel which reach a thickness of more than 200 m in the central part of the plain. The Lai drain is the longest drain in the study area, and drains both cities. After draining Islamabad, the Lai drain flows to the flatter areas of Rawalpindi. It drains an area of about 211 km2, of which 55 per cent lies in Islamabad and the remaining 45 per cent in Rawalpindi city. The Lai Basin is connected with the neighboring basin of the Kurang River underground. The drainage divide on the northern and north-eastern is functioning partially, compared to its western divide, which is dividing the surface and groundwater (Ali et al. 2012).

Sampling strategy

Groundwater sampling of shallow and deep aquifers were collected from house bore with depths of 80–160 m, and deep water samples were from tube-wells with depths of more than 200 m. A total of 27 samples were collected from different zones of Rawalpindi and Islamabad. Geographic locations of collected samples are shown in Figures 1 and 2.
Figure 1

Geographic locations of collected samples from Rawalpindi and Islamabad.

Figure 1

Geographic locations of collected samples from Rawalpindi and Islamabad.

Figure 2

Depth of sampling sites (in feet).

Figure 2

Depth of sampling sites (in feet).

Hydrochemical parameters

One liter plastic bottles were used to collect the groundwater samples, and rinsed thoroughly with the water to be collected. All the bottles were closed firmly after sampling and ensured that there was no eventual leakage. Before sampling, groundwater was pumped out for 15 minutes in order to evacuate the stagnant water in the pumping system. Samples were carefully handled and stored at room temperature, away from exposure to direct sunlight and heat. Physicochemical parameters such as the pH, specific electrical conductivity (EC) and total dissolved solids (TDS) of samples were measured in situ. Dilute HNO3 was added to each sample until the pH was <2 for major cations, then the sample bottles were stored at about 4 °C. Measurement of pH was done using a digital pH meter (Adwa, Model AD1030). The EC and TDS of collected samples were measured with a portable conductivity meter (WTW-Model LF 95) calibrated with standard solutions from Hanna instruments (Italy).

Determination of major anions included HCO3, Cl, SO4−2 and NO3. The concentration of HCO3 was determined by the titrimetric method, and the ppm concentration was calculated (Clesceriat et al. 2005). Chloride ions were detected using the Argentometric method (Clarke 1950).

For HCO3 determination: 
formula
For Cl determination: 
formula
Sulphate ions in samples were detected using the Turbidimetric method (Wilde & Radke 2001). Nitrate ion detection was carried out through the ion selective electrode method using an ion meter (Ion5/6, Acron series: Akton Instrument) (Nitrate 2007).

Major cations include Na+, K+, Ca+2, and Mg+2 which were analyzed using atomic absorption (Z-2000) (Skoog & West 1980). High purity reagents, high purity de-ionized, MilliQ water filtered through a 0.22 μm filter, were used throughout the analytical work.

Saturation index and sodium adsorption ratio

Saturation index (SI) is used to predict the reactivity and mineral saturation in the groundwater along its flow path. The Langelier Saturation index (LSI) is a useful parameter, as it indicates the chemical stability of the groundwater. It is an equilibrium model derived from the theoretical concept of saturation. Basically, it is an indicator of water saturation with respect to CaCO3 (Langelier et al. 1950). It can be shown that the LSI approximates the base 10 log of the calcite saturation level. Calcium and bicarbonate concentration (mg/L), TDS (mg/L), the actual pH and temperature of the water (°C) are the required parameters for LSI calculation. Values of LSI can be interpreted as follows (http://www.lenntech.com/calculators/langelier/index/langelier.htm):

  • (i) Negative LSI value: No potential to scale, the water will dissolve CaCO3

  • (ii) Positive LSI value: This shows precipitation of CaCO3 may occur

  • (iii) Zero LSI value: Borderline scale potential.

Water quality, changes in temperature, or evaporation could change the SI. Water in wells may be undergoing deposition of carbonate precipitates on well screens, gravel packs and water-yielding rock faces. LSI was used to calculate SI through online application (http://www.lenntech.com/calculators/langelier/index/langelier.htm).

The sodium adsorption ratio (SAR) is useful to assess the potential sodium hazard. Excessive sodium in water cause a sodium hazard to soil. The groundwater SAR was calculated through the following equation (Cannon et al. 2007): 
formula

Tritium analysis

Sample collection and treatment

Tritium, being the radioisotope of hydrogen, emits weak beta particles with low energy and is measured by using a liquid scintillation counter (LSC). Since all natural water samples contain a low tritium content, water has to be enriched prior to measurement to improve the sensitivity of measurement and this is usually done by concentrating the sample to about 20–30 times by an electrolysis process called electrolytic enrichment. Tritium samples are electrolytically enriched in the batch. Environmental tritium samples were collected in one liter polyethylene bottles and assured to be airtight to avoid evaporation and headspace. The samples must be collected while the bore is still pumping (Rosen & Kropf 2009). Bottles were rinsed with sample water prior to sample collection. Groundwater samples were filtered through 0.45 μm filter paper to get rid of fine particles, and then distillation was done to reduce EC (≤50 μS/cm).

Electrolytic enrichment

The electrolytic enrichment unit consists of 20 electrolytic cells (the capacity of each cell is 500 mL), a refrigeration unit and an electric power supply with 0–15A current potential. Electric cells are placed in the special fabricated holes in the refrigerator. The holes are designed in such a way that only the neck of an enrichment cell remains outside, whereas the whole body remains inside the refrigeration unit. An enrichment cell is composed of a cathode and an anode. The cathode is made up of mild steel and the anode is composed of stainless steel. Both electrodes are separated by a Teflon stopper to reduce the risk of a short circuit. A gas outlet, through a silicon oil trap, is attached to each cell to allow the gases that are produced during the electrolysis reaction to escape.

The cathode involves a reduction reaction: 
formula
Oxidation takes place on the anode: 
formula

During electrolysis, cell temperature increase is controlled by a refrigeration unit fitted with a thermo switch to regulate the temperature.

Sodium peroxide (Na2O2) was added in initially distilled samples as an electrolyte. An alkaline solution was formed on addition of Na2O2 in sample water. 
formula
Electrolytic cells were connected in series order. The value of the current was set according to Faraday's law, as to break a water molecule 2.97 amperes is required. At the start of the enrichment process, the current was set in such a manner that only 1Ah was moved from the circuit and the value increased gradually as the enrichment proceeded, to maintain the temperature values at between 2 and 4 °C during the electrolysis process in order to minimize the evaporation. Each run comprised 15 water samples, 3 spike samples and 2 dead water (DW) samples. An electric current of 1,400 Ah was passed through each cell and 500 mL of sample was reduced to about 15–20 mL in each cell. The current supply was controlled through a DC power supply. Precautions were taken not to provide a high current at start of process; the current supply steadily and slowly increased to avoid overheating of the cells that might cause evaporation or bubbling of the sample. The electrolytic enrichment process takes about 20 days under smooth conditions. The cell temperature was measured by thermocouple and it should not be more than 3 °C (0–3 °C) to achieve a similar enrichment in all the cells. The quantity of the sample is determined by weighing the cells. The electrolytic enrichment process provides a detection limit of 0.5TU, which is sufficient for almost all hydrological samples. The enrichment factor depends upon weight reduction, and it is calculated through the formula: 
formula
On completion of enrichment, the electrolytic cells were again weighed.

Neutralization and distillation process

An aliquot of the enriched sample was mixed with 4 g of PbCl2 in order to neutralize the strong alkaline media due to Na+ ions that were present in the water sample. The chemical equation of neutralization using lead chloride is as follows: 
formula
 
formula
After neutralization, the sample was subjected to distillation. The distillation process eliminates the salts formed during the neutralization process. During all of the processes, contact between the sample and air was avoided to eliminate contamination error.

Tritium analysis

The tritium contents of the samples were measured using an LSC (Tri-Carb 3170). The system is provided with an IBM PC-compatible computer along with Windows driven software ‘QuantaSmart’, which is used to access and control all the system features and capabilities via different functional components. Samples for the LSC were prepared by taking 8 mL of distilled samples in 20 mL Teflon vials with 12 mL of Scintillation cocktail (Ultima Gold LLT Packard). Ultima Gold has an aqueous holding capacity of 25%. The scintillation counter requires efficiency calibration, which was done by running spike samples and DW samples along with the groundwater samples whose tritium content has to be determined. Spike samples were placed in LSC to determine the electrolysis yield, whereas tritium free water (DW) served as a background sample to check for possible contamination during the enrichment.

The enrichment factor depends upon weight reduction and the fraction of weight reduced by electrolysis. This factor was calculated using the following equation: 
formula
where: 
formula
The factor for spike cells was calculated using the equation: 
formula
Samples were then placed in a programmed LS counter for spectrum analysis. Counting was done for 10 cycles of 500 minutes, and by averaging the counting values the tritium activity for each sample was calculated. The LSC estimates tritium activity by counting the photons emitted from a liquid sample. Analysis through the liquid scintillation technique has proven to be a method of high-performance measurement for natural H3 concentrations (Morgenstern & Taylor 2009). The quench curve (also known as the efficiency curve) is a plot of the counting efficiency as a function of the quench index parameter, and can be used to determine counting efficiencies. The tritium content in each sample on the counting date was calculated using the following formula: 
formula
where:
  • AT = Measured tritium content in the given sample on counting date

  • NSA = Net count rate of the sample (cpm)

  • NST = Net count rate of the standard (cpm)

  • AST = Tritium activity of the standard on the counting date [TU]

  • ESA = Enrichment factor for the given sample

The calculated tritium activity for the sampling date was corrected, applying decay correction with the following formula: 
formula
where:
  • ATC = Tritium activity of the given sample, corrected for the sampling date

  • D = Decay correction for the period of sampling to counting

  • = Decay constant of tritium = ln(2)/T1/2 = 0.0562662 [y−1]

  • T1/2 = Half life of tritium = 12.32 years = 4,500 days

  • t = Time elapsed between the sampling and measurement [y]

Groundwater characterization analysis

Piper tri-linear diagram

Groundwater chart was used to make a Piper tri-linear diagram for identification of the water type and characterization of groundwater. In a Piper plot, both cations and anions are expressed in meqL−1. Anions are plotted on the left triangle while cations appear on the right triangle as a single plot. The central diamond-shaped area in the tri-linear diagram is a projection of these two plots. Similarities and differences of groundwater samples can be expressed more effectively through groundwater charts as similar qualities of samples plotted together in the form of groups (Hem 1985).

Contour analysis

The results are shown in the form of colored contour maps that were drawn using the kriging method on Golden Surfur software v. 11 (Golden Software, Inc.). A contour plot is comprised of three values, X, Y (which are the coordinates of the representative sample) and Z (which represents the concentration of the given element/parameter).

RESULTS AND DISCUSSION

Chemistry of groundwater

The pH values (7.2–7.9) of groundwater samples were within the satisfactory limit of the WHO, i.e. 6.5–8.5 (WHO 2004), as shown in Figure 3. Groundwater samples collected from the boreholes of Rawalpindi showed higher EC values that exceeded the WHO permissible limits, i.e. 1,400 μS/cm, ranged from 611 μS/cm to 1,743 μS/cm. The surface and groundwater EC values might vary by geographical conditions (Hem 1985), or they might be recharged from the Lai drain, the main drain passing through this area. Naturally through dissolution processes minerals, soil particles and organic material ions enter into groundwater systems. Salinity levels may also fluctuate by anthropogenic activities such as application of natural and synthetic fertilizers, and industrial wastewater (Appelo & Postma 2005; Iqbal et al. 2014). In contrast, the boreholes of Islamabad have comparatively low EC values (593–951 μS/cm). Measured values of TDS exhibited a similar pattern to EC for all groundwater samples (Figure 3). According to WHO (2004) guidelines, the measured TDS content of the Islamabad samples (288–463 mg/L) lay in safe limit (i.e. 0–600 mg/L), whereas the TDS range of Rawalpindi groundwater (297–849 mg/L) was placed in the marginal limit (i.e. 450–1,500 mg/L).
Figure 3

pH, EC and TDS of groundwater of Rawalpindi and Islamabad.

Figure 3

pH, EC and TDS of groundwater of Rawalpindi and Islamabad.

The measured values of Ca+2 and Na+ ions ranged from 8–220 mg/L and 13–170 mg/L, respectively, for all groundwater samples. The values of both Ca+2 and Na+ ions were within the WHO permissible limits (i.e. 300 and 200 mg/L, respectively). The measured Mg+2 ion concentrations varied from 10 to 38 mg/L in the groundwater of the twin cities (Figure 4). According to the results of the groundwater samples, both cities have Mg+2 values (50–150 mg/L) within the permissible limits of the WHO. The potassium ion concentration within the sampling profile varied between 0.65 and 7.25 mg/L with an average value of 1.79 mg/L. However, all samples lay within the permissible limits recommended by WHO (2004), which is 200 mg/L.
Figure 4

Major cations in the groundwater of Rawalpindi and Islamabad.

Figure 4

Major cations in the groundwater of Rawalpindi and Islamabad.

Of the anions, bicarbonate, chloride, sulphates and nitrate ions were measured. Results showed that the groundwater of Rawalpindi has higher bicarbonate values (153–565 mg/L) compared to Islamabad groundwater samples, with a minimum value of 136 mg/L and a maximum value of 240 mg/L. The chloride ions were ranged between 11­ and 145 mg/L for the groundwater samples of the twin cities. Overall, Rawalpindi samples had high Cl values (Figure 5) compared to Islamabad groundwater samples, but were within permissible limits (250 mg/L by WHO 2004). High Cl values are mainly contributed to by wastewater seepage. In Rawalpindi groundwater samples, the sulphate and nitrate concentrations ranged between 6–62 mg/L and 16–125 mg/L, respectively. Contour analysis showed the boreholes of Rawalpindi contained significant concentrations, within the range of 17–125 mg/L, which exceeds the WHO permissible limit of 50 mg/L. Higher nitrate concentrations were observed in GWR-15 (Muslim Town), GWR-16 (Khanna Pul), GWR-19 (Adiala Road), GWR-20 (Morgah) and GWR-23 (Dhamial). This is attributed to the presence of these boreholes near the Lai drain, as shown in Figure 5. Data suggested that the Lai drain served as an active recharge source for these wells. Chemical data of the groundwater samples are shown in Table 1. High concentrations of chlorides, nitrates, sulfates, carbonates and bicarbonates in the groundwater samples of Rawalpindi are attributed to high concentrations of these anions in the wastewater of the Lai drain passing through the Rawalpindi area. It is worth mentioning that the part of the Lai drain passing through the Rawalpindi area has a higher concentration of cations and anions compared to the part passing through Islamabad, as reported by Munir (Munir et al. 2011).
Table 1

Physiochemical characteristics of groundwater samples (Rawalpindi and Islamabad)

    Major cations
Major anions
 
ECTDSCa+2Na+Mg+2K+HCO3ClSO4−2NO3
Sample codespHμS/cmmg/Lmg/Lmg/LTritium (TU)
GWR-1 Westridge-RWP 7.37 1,072 522 55 43 8.13 2.09 404 61.34 48.575 79.4 8.13 
GWR-4 Harley Street-RWP 7.51 979 477 51 45 25.42 1.5 474 47.71 36.09 60.0 25.42 
GWR-5, Dheri-RWP 7.48 1,019 502 23 47 10.29 1.66 500 54.32 28.96 46.7 10.29 
GWR-6 Gawalmandi-RWP 7.7 1,163 567 45 62 8.26 2.31 488 67.02 40.85 70.9 8.26 
GWR-7 LiaquatBagh-RWP 7.34 611 297 39 25 10.86 7.25 294 23.28 18.28 17.6 10.86 
GWR-8 Khyaban-e-Sir Syed-RWP 7.41 908 441 19 27 11.35 1.59 468 36.35 19.46 64.9 11.35 
GWR-10 Raja Bazzar-RWP 7.38 1,070 523 65 38 8.13 0.7 565 48.28 20.05 40.6 8.13 
GWR-11 Satellite Town-RWP 7.4 984 481 32 35 12.69 1.59 431 47.14 30.75 62.4 12.69 
GWR-12 Dhoke Kala Khan -RWP 7.9 682 332 24 18 11.17 1.37 329 24.42 9.05 56.4 11.17 
GWR-15 Muslim Town -RWP 7.3 1,572 769 55 23 10.22 2.96 498 74.4 62.23 115.7 10.22 
GWR-16 Khanna Road -RWP 7.53 1,400 683 87 17 8.1 2.81 524 55.09 43.22 107.2 8.1 
GWR-17 Pakistan Town -RWP 7.46 1,163 566 90 6.02 1.67 450 85.2 14.5 63.6 6.02 
GWR-19 Adiala Road -RWP 7.28 1,168 566 74 56 9.14 0.79 388 67.02 20.06 113.3 9.14 
GWR-20 Morgah -RWP 7.58 1,507 731 220 75 12.94 0.94 450 71 45.01 125.4 12.94 
GWR-23 Dhamial -RWP 7.65 1,743 849 61 170 6.83 1.4 466 145.97 25.99 121.8 6.83 
GWI-1 (H-9 (A))- ISB 7.53 716 348 69 20 9.12 0.78 163 19.88 10.72 52.7 9.12 
GWI-2 (H-9(B))- ISB 7.79 759 370 85.2 25 5.1 1.98 157 21.02 6.63 56.4 5.1 
GWI-3 (H-10)- ISB 7.2 757 369 94.9 34 9.43 0.84 240 21.58 6.78 49.1 9.43 
GWI-4 (I-8/4)- ISB 7.68 735 358 115.8 18 6.28 0.95 230 19.88 19.06 46.7 6.28 
GWI-5(I-9)- ISB 7.47 748 364 81.6 22 5.67 0.86 178 19.31 11.93 54.0 5.67 
GWI-6 (I-11)- ISB 7.5 951 463 66.8 35 17.65 0.77 155 52.26 18.60 70.9 17.65 
GWI-7 (I-10)- ISB 7.48 691 337 65.8 19 7.44 1.2 192 29.54 18.15 62.4 7.44 
GWI-8 (G-10/3)- ISB 7.78 653 319 61.58 15 18.52 0.65 197 11.36 37.56 37.0 18.52 
GWI-9 (G-9)- ISB 7.6 759 372 80.86 26 13.65 1.29 234 22.72 29.52 68.5 13.65 
GWI-10 (F-9)- ISB 7.34 593 288 28.4 27 36.08 0.9 136 14.2 14.81 16.4 36.08 
GWI-11 (F-8/4)- ISB 7.44 705 344 45.7 13 5.181 2.2 213 22.72 20.42 62.4 5.181 
GWI-12 (F-8/1)- ISB 7.33 682 332 21.8 14 17.53 1.1 143 21.3 14.66 52.7 17.53 
WHO Guidelines (2004)  6.5–8.5 _ 600 300 200 150 200.0 _ 250 250 50   
    Major cations
Major anions
 
ECTDSCa+2Na+Mg+2K+HCO3ClSO4−2NO3
Sample codespHμS/cmmg/Lmg/Lmg/LTritium (TU)
GWR-1 Westridge-RWP 7.37 1,072 522 55 43 8.13 2.09 404 61.34 48.575 79.4 8.13 
GWR-4 Harley Street-RWP 7.51 979 477 51 45 25.42 1.5 474 47.71 36.09 60.0 25.42 
GWR-5, Dheri-RWP 7.48 1,019 502 23 47 10.29 1.66 500 54.32 28.96 46.7 10.29 
GWR-6 Gawalmandi-RWP 7.7 1,163 567 45 62 8.26 2.31 488 67.02 40.85 70.9 8.26 
GWR-7 LiaquatBagh-RWP 7.34 611 297 39 25 10.86 7.25 294 23.28 18.28 17.6 10.86 
GWR-8 Khyaban-e-Sir Syed-RWP 7.41 908 441 19 27 11.35 1.59 468 36.35 19.46 64.9 11.35 
GWR-10 Raja Bazzar-RWP 7.38 1,070 523 65 38 8.13 0.7 565 48.28 20.05 40.6 8.13 
GWR-11 Satellite Town-RWP 7.4 984 481 32 35 12.69 1.59 431 47.14 30.75 62.4 12.69 
GWR-12 Dhoke Kala Khan -RWP 7.9 682 332 24 18 11.17 1.37 329 24.42 9.05 56.4 11.17 
GWR-15 Muslim Town -RWP 7.3 1,572 769 55 23 10.22 2.96 498 74.4 62.23 115.7 10.22 
GWR-16 Khanna Road -RWP 7.53 1,400 683 87 17 8.1 2.81 524 55.09 43.22 107.2 8.1 
GWR-17 Pakistan Town -RWP 7.46 1,163 566 90 6.02 1.67 450 85.2 14.5 63.6 6.02 
GWR-19 Adiala Road -RWP 7.28 1,168 566 74 56 9.14 0.79 388 67.02 20.06 113.3 9.14 
GWR-20 Morgah -RWP 7.58 1,507 731 220 75 12.94 0.94 450 71 45.01 125.4 12.94 
GWR-23 Dhamial -RWP 7.65 1,743 849 61 170 6.83 1.4 466 145.97 25.99 121.8 6.83 
GWI-1 (H-9 (A))- ISB 7.53 716 348 69 20 9.12 0.78 163 19.88 10.72 52.7 9.12 
GWI-2 (H-9(B))- ISB 7.79 759 370 85.2 25 5.1 1.98 157 21.02 6.63 56.4 5.1 
GWI-3 (H-10)- ISB 7.2 757 369 94.9 34 9.43 0.84 240 21.58 6.78 49.1 9.43 
GWI-4 (I-8/4)- ISB 7.68 735 358 115.8 18 6.28 0.95 230 19.88 19.06 46.7 6.28 
GWI-5(I-9)- ISB 7.47 748 364 81.6 22 5.67 0.86 178 19.31 11.93 54.0 5.67 
GWI-6 (I-11)- ISB 7.5 951 463 66.8 35 17.65 0.77 155 52.26 18.60 70.9 17.65 
GWI-7 (I-10)- ISB 7.48 691 337 65.8 19 7.44 1.2 192 29.54 18.15 62.4 7.44 
GWI-8 (G-10/3)- ISB 7.78 653 319 61.58 15 18.52 0.65 197 11.36 37.56 37.0 18.52 
GWI-9 (G-9)- ISB 7.6 759 372 80.86 26 13.65 1.29 234 22.72 29.52 68.5 13.65 
GWI-10 (F-9)- ISB 7.34 593 288 28.4 27 36.08 0.9 136 14.2 14.81 16.4 36.08 
GWI-11 (F-8/4)- ISB 7.44 705 344 45.7 13 5.181 2.2 213 22.72 20.42 62.4 5.181 
GWI-12 (F-8/1)- ISB 7.33 682 332 21.8 14 17.53 1.1 143 21.3 14.66 52.7 17.53 
WHO Guidelines (2004)  6.5–8.5 _ 600 300 200 150 200.0 _ 250 250 50   

RWP, Rawalpindi city.; ISB, Islamabad city.

Figure 5

Major anions in the groundwater of Rawalpindi and Islamabad.

Figure 5

Major anions in the groundwater of Rawalpindi and Islamabad.

Sodium adsorption ratio

SAR predicts the degree to which irrigation water enters into cation-exchange reactions in soil media. Sodium ions can replace the adsorbed Mg+ and Ca+, which would lead to the high SARs. High SAR values correspond to a sodium hazard that badly impacts the soil structure, as the adsorbed sodium causes dryness and compaction of the soil. Compacted soils are more impervious to water penetration. Fine textured soils with high clay content are more susceptible to this action. Islamabad samples showed SAR within satisfactory limits (0.5–1.4), whereas the SAR values for Rawalpindi groundwater were within the limits of 0–5.9, which is an alarming situation related to sodium hazard. Five out of 15 samples showed higher SAR, viz., GWR-17, GWR23, GWR-5, GWR-6 and GWR-19. These samples also showed high TDS and EC values. Natural factors that affect TDS content may also influence the relative concentration of sodium, calcium and magnesium. Therefore, SAR values of water samples are positively correlated with the dissolved solids concentrations. These groundwater samples also have high sodium concentrations relative to calcium and magnesium that purely relate to the soil nature and composition (Cannon et al. 2007).

The accumulation of sodium in soils results most often from the use of irrigation water with high SAR or water with high levels of HCO3−2 and CO3−2 ions. Cation exchange processes (among them sodium and calcium) take place when soil contains Ca+2 and Mg+2 as the predominant cations adsorbed on soil particles. On the other hand, the excessive sodium tends to deflocculate soil, disperses the clay and may cause a reduction of permeability when the amount of adsorbed sodium increases to 10% of the total cations. The higher SAR values of Rawalpindi city showed that rain recharge in the area is adversely affected and the groundwater quality of this city is deteriorating rapidly.

Langelier saturation index

The SI for the groundwater of Rawalpindi and Islamabad is +0.7 and −0.8 to +0.39 on the scale, respectively. The GWI-6 showed maximum SI with a value of 59, which shows highly supersaturated water with respect to calcium carbonate (CaCO3). The higher SI values of the Rawalpindi groundwater samples suggested a recharge of groundwater from the Lai drain. Furthermore, data suggested that the quality of the groundwater of Rawalpindi city will not be suitable for drinking purposes in future.

Piper plot

The Rawalpindi Piper plot (Figure 6) interpretation shows that most of the water samples are of the Na-Mg-Ca type, with Na+ and Ca+2 as the predominant cations, while of the anions HCO3 is the most dominant followed by Cl and SO4−2. The Piper plot (Figure 7) of Islamabad also showed a similar output, with HCO3 dominance. Here, the sites showed the Ca-Na-Mg-HCO3 type in most groundwater samples except for two (i.e. GWI-10 and GWI-12) that have Na+ as the dominant anion rather than Ca+2. Both of these samples have high tritium values too.
Figure 6

Piper plot of Rawalpindi.

Figure 6

Piper plot of Rawalpindi.

Figure 7

Piper plot of Islamabad.

Figure 7

Piper plot of Islamabad.

Tritium, nitrate, chloride, pH and EC

A groundwater recharge source can be identified by comparison of the tritium concentration in the surface and groundwater with the local rainfall, streams and drains. The detected level of H3 for Islamabad precipitation was 32TU in 1993 (Sajjad 1998), but presently this concentration falls in the range of 35–40TU.

The spatial variation of tritium has been shown in Figure 8. The groundwater from these locations exhibited dissimilar values of tritium concentration due to the mixing of a modern recharge (most probably rain) component with the recharge from the 1960s and 1970s (Ravikumar & Somashekar 2011). According to the frequency histogram (Figure 9), 3H concentrations in most of the groundwater samples lie between 5 and 14 TU and indicate modern water according to the mentioned criteria for Pakistan. Four sites, GWI-6, GWI-8, GWI-12 and GWR-4, showed an H3 content of 17TU, 18TU, 17TU and 25TU, respectively, which shows mixing of modern and 1960s to 1970s recharge. According to Cox (Cox 2003) the samples with a tritium value of 15–30TU might have a mixture of older and young water. Only one sampling point, GWI-10 (F-9), has a tritium concentration of 36TU, which points towards recharge during the 1960s and 1970s and is evident in lacking mixing. However, according to Moran (Moran et al. 2004), water with significant tritium is regarded as younger water. Normally, tritium concentrations in precipitation fluctuate between 1TU and 10TU, from oceanic high precipitation to dry arid inland regions. The tritium quantity of all groundwater samples lies above 1TU, which is another indication of modern recharge (Ravikumar & Somashekar 2011). Tritium data for both cities (Table 1) suggested that rainfall contributed more to groundwater recharge in Islamabad and more or less in Rawalpindi.
Figure 8

Tritium (TU) concentrations in groundwater of Rawalpindi and Islamabad.

Figure 8

Tritium (TU) concentrations in groundwater of Rawalpindi and Islamabad.

Figure 9

Frequency histogram of Tritium in the groundwater samples of Rawalpindi and Islamabad.

Figure 9

Frequency histogram of Tritium in the groundwater samples of Rawalpindi and Islamabad.

The groundwater samples of Rawalpindi have higher nitrate content compared to the Islamabad area. This suggests that groundwater has anthropogenic recharge too, most probably through the Lai drain that passes throughout the city. Among the aquifers, shallow boreholes are more susceptible to nitrate contamination where the soil is more porous. This behavior is more explicit in the boreholes of Rawalpindi, as they have an average well depth of 38 m. A significant relationship between tritium and chloride/nitrates corresponds to some anthropogenic inputs in groundwater (Figure 10).
Figure 10

Relationship of Tritium with nitrates and chlorides in groundwater of Rawalpindi areas.

Figure 10

Relationship of Tritium with nitrates and chlorides in groundwater of Rawalpindi areas.

The groundwater samples of Islamabad, with low nitrate and chloride contents, suggest that the areas of Islamabad have not been influenced by anthropogenic contamination yet. The relationship between the chlorides and tritium and nitrates with tritium showed a good pattern, suggesting that precipitation in this area is the main recharge source and the input of chlorides and nitrates in groundwater is, mostly, through precipitation (Figure 11).
Figure 11

Relationship of Tritium with nitrates and chlorides in the groundwater of the Islamabad area.

Figure 11

Relationship of Tritium with nitrates and chlorides in the groundwater of the Islamabad area.

The relation between the tritium and pH values of the groundwater samples of both cities are shown in Figures 12 and 13. The pH values of the groundwater samples of both cities ranged between 7.2 and 7.9. The presence of limestone minerals in the study area has an important role in carbonate buffering that gives neutral pH values for given sites.
Figure 12

Relationship between Tritium and pH values of groundwater of Rawalpindi areas.

Figure 12

Relationship between Tritium and pH values of groundwater of Rawalpindi areas.

Figure 13

Relationship between Tritium and pH values of the groundwater of the Islamabad area.

Figure 13

Relationship between Tritium and pH values of the groundwater of the Islamabad area.

The distribution of common ions in the groundwater samples and their relations with groundwater residence time are consistent that have a similar recharge source and undergo modest chemical evolution of the major ions' composition as the groundwater moves along a flow path (Savenije 2004). Arid and semi-arid climatic conditions in this zone reveal that most of the recharge comes through rain. The aquifer located in the territory of Islamabad possibly received recharge from rainfall on upland areas (the Margalla Hills), which accounts for their low EC and TDS contents. Dissolved solids concentrations are lower in the upland recharge areas, and increased in samples collected from down gradient locations. Higher Ca+2 contents were explained by Cox (2003). According to him, calcium and magnesium together make up more than 80% of the cations in solution, with calcium generally present in larger concentrations than magnesium, particularly in wells down gradient from the recharge area. Whereas, in the case of the Rawalpindi groundwater samples, severe nitrate, sulphate and chloride contamination have been observed in samples mostly taken from the Lai basin, which ensures recharge inputs from the Lai drain.

Furthermore, the majority of wells in the Rawalpindi area and some in the Islamabad area have poor quality water with limited recharge (as is clearly evident from the age assessment) in the twin cities. The abstraction rate in these areas also rose sharply due to over population. The continuous increase in use of groundwater results in the lowering of the water table. The water table is dropping at the rate of 3 m per year approximately in the area of Rawalpindi. In the same way, a number of studies have reported a decline of groundwater levels in Islamabad due to extensive withdrawal practices. Over abstraction adversely affects the quantity as well as the quality of groundwater. The graph between EC and tritium in both cities (Figures 14 and 15) suggested the groundwater movement within different layers with a variety of mineral compositions may result in considerable changes in water composition and ion content. Data suggested that most of the samples with high tritium content and low EC values are considered to have a shallow circulating water profile. Some samples in the same area have higher EC values and lower tritium values, suggesting deep circulating water. Such data suggest that the region has water resources where shallow and deep water mix together in different proportions resulting in a large residence time for groundwater (Wright & Belitz 2011). Hydrochemistry data reveal that groundwater quality also deteriorated, with an excessive concentration of nitrates, sulphates and chlorides that shows contamination of the aquifer systems. The higher rate of water abstraction and the deterioration of surface water (commonly from waste dumping) poses a serious threat to water quality. The water quality of the surface water resources strongly influences the groundwater quality, as both these water reservoirs are interconnected. Moreover, alteration in recharge patterns through soil compaction and cementation (by increase construction activity) results the low recharge inputs more evident from SAR values. Aquifers in study area are highly dependent on rainfall for the recharge due to the semi-arid climatic conditions. SARs for current investigation also indicated potential sodium hazard. High SAR induces the sodium hazard that leads towards soil compaction and also disturbs the recharge pattern. The rapid increase in population increased the consumption of groundwater than its recharge. This phenomenon has disturbed the natural equilibrium of groundwater system. More use of groundwater due to increase in population will cause further depletion of groundwater sources in the near future. Hence, a groundwater database and regular monitoring are pre-requisites for better use of groundwater in the current scenario.
Figure 14

Relationship between Tritium and EC values of groundwater of Rawalpindi areas.

Figure 14

Relationship between Tritium and EC values of groundwater of Rawalpindi areas.

Figure 15

Relationship between Tritium and EC values of groundwater of Islamabad areas.

Figure 15

Relationship between Tritium and EC values of groundwater of Islamabad areas.

CONCLUSION

Groundwater is a renewable fresh water resource, but an increased rate of abstraction would convert it to a non-renewable resource. The tritium and chemical data suggested that the groundwater in the study areas was freshly recharged. The tritium concentration of these cities suggested mixing of a modern recharge (most probably rain) component with the canals and drains. The groundwater samples of Rawalpindi, with higher chloride and nitrate contents, and the graph of tritium/chloride, suggested that groundwater has anthropogenic recharge most probably through the Lai drain passing throughout the city. Among the aquifers, shallow boreholes are more susceptible to nitrate contamination where the soil is more porous. Furthermore, in the case of the Rawalpindi groundwater samples, severe nitrate, sulphate and chloride contamination has been observed in samples mostly taken from the Lai basin, which ensures recharge inputs from the Lai drain. Moreover, alteration in recharge patterns through soil compaction and cementation (by increased construction activity) results in the low recharge inputs more evident from the SAR values. A high SAR induces the sodium hazard, which leads towards soil compaction and also disturbs the recharge pattern. Hydrochemistry data reveals that groundwater quality also deteriorated, with excessive concentration of nitrates, sulphates and chlorides, which shows contamination of the aquifer systems.

The groundwater samples of Islamabad with low nitrates and tritium content suggested that the areas in Islamabad have not yet been influenced by anthropogenic contamination.

Tritium values in the Islamabad groundwater samples also suggested that aquifers in the study area are highly dependent on rainfall for recharge. The aquifer located in the territory of Islamabad possibly received recharge from the upland areas (the Margalla Hills), which accounts for their low EC and TDS contents. Dissolved solids concentrations are lower in the upland recharge areas and increased in samples collected from down gradient locations.

The higher rate of water abstraction and deterioration of surface water pose a serious threat to water quality. Continuous increases in water withdrawal activities have resulted in the lowering of the water table in both cities. If the withdrawal activities proceed with a similar pattern, the aquifers may yield more poor quality water. Hence, a groundwater database and regular monitoring are pre-requisites for better use of groundwater in the current scenario.

Effective management requires awareness of the status of groundwater corresponding to the quality and the quantity of available groundwater in the region. In this perspective, the results of the current investigation must be regarded as preliminary, and further extensive studies should be done on large scale covering the areas in the vicinity of the twin cities. Furthermore, rules and regulations should be formulated at federal as well as provisional level to stop domestic and industrial wastewater input into the Lai drain and ensure sustainable use of groundwater.

REFERENCES

REFERENCES
Aggarwal
P. K.
Froehlich
K.
Kulkarni
K. M.
2007
Environmental isotopes in groundwater studies
. In:
Groundwater–Encyclopedia of Live Support Systems
(
Silveira
L.
Usunoff
E. J.
, eds).
EOLSS Publishers/UNESCO
,
Paris
.
Ali
M.
Qureshi
A. A.
Waheed
A.
Baloch
M. A.
Qayyum
H.
Tufail
M.
Khan
H. A.
2012
Assessment of radiological hazard of NORM in Margalla Hills limestone, Pakistan
.
Environmental Monitoring and Assessment
184
(
8
),
4623
4634
.
Appelo
C. A. J.
Postma
D.
2005
Geochemistry, Groundwater and Pollution
.
CRC Press
,
Boca Raton, FL, USA
.
Bartolino
J. R.
Cunning
W. L.
2012
Groundwater depletion across the nation
.
Drinking Water & Backflow Prevention, USGS Water Science
29
(
1
),
26
29
.
Brown
R. M.
Grummitt
W. E.
1956
The determination of tritium in natural waters
.
Canadian Journal of Chemistry
34
(
3
),
220
226
.
Cannon
M. R.
Nimick
D. A.
Cleasby
T. E.
Kinsey
S. M.
Lambing
J. H.
2007
Measured and Estimated Sodium-Adsorption Ratios for Tongue River and its Tributaries, Montana and Wyoming, 2004-06
.
US Geological Survey
.
Clesceriat
L. S.
Greenberg
A. E.
Eaton
A. D.
Rice
E. W.
Franson
M. A. H.
2005
Standard Methods for the Examination of Water and Wastewater
. 20th edn.
American Public Health Association
,
Washington, DC
.
Cox
S. E.
2003
Estimates of Residence Time and Related Variations in Quality of Ground Water Beneath Submarine Base Bangor and Vicinity, Kitsap County, Washington
.
US Department of the Interior, US Geological Survey
.
Hem
J. D.
1985
Study and Interpretation of the Chemical Characteristics of Natural Water
(
Vol. 2254
),
Department of the Interior, US Geological Survey
.
Iqbal
N.
Rafiq
M.
Hashmi
Z.
Latif
Z.
Ghaffar
A.
2014
Hydrological investigation of groundwater under salt affected land in Shorkot area with special emphasis on seepage from link canals
.
Nucleus
51
(
1
),
63
74
.
Islam
U. H. T.
Cheema
W. A.
Ahmed
C. N.
2008
Multifaceted groundwater quality and recharge mechanism issues in a mega-city (Rawalpindi, Pakistan), and mitigation strategies
. In:
Securing Groundwater Quality in Urban and Industrial Environments, Proceedings of the 6th International IAHS Groundwater Quality Conference, Fremantle, Western Australia, 2–7 December 2007
.
IAHS Publ. no. XXX, 8-16
.
Langelier
W. F.
Caldwell
D. H.
Lawrence
W. B.
Spaulding
C. H.
1950
Scale control in sea water distillation equipment-contact stabilization
.
Industrial & Engineering Chemistry
42
(
1
),
126
130
.
Lohman
S.W.
1972
Ground-Water Hydraulics. US Geological Survey Professional Paper 708, 7-9
.
Lucas
L. L.
Unterweger
M. P.
2000
Comprehensible review and critical evaluation of the half-life of Tritium
.
Journal of Research of the National Institute of Standards and Technology
105
(
4
),
541
549
.
Moncaster
S. J.
Bottrell
S. H.
Tellam
J. H.
Lloyd
J. W.
Konhauser
K. O.
2000
Migration and attenuation of agrochemical pollutants: insights from isotopic analysis of groundwater sulphate
.
Journal of Contaminant Hydrology
43
(
2
),
147
163
.
Moran
J. E.
Hudson
G. B.
Eaton
G. F.
Leif
R.
2004
A Contamination Vulnerability Assessment for the Santa Clara and San Mateo County Groundwater Basins. Lawrence Livermore National Laboratory Internal Report, UCRL-TR-201929, 49
.
Morgenstern
U.
Taylor
C. B.
2009
Ultra low-level tritium measurement using electrolytic enrichment and LSC
.
Isotopes in Environmental and Health Studies
45
(
2
),
96
117
.
Mozter
W. E.
2007
Tritium age dating of groundwater
. In:
Hydro Vision Groundwater Resource Association of California, Santa Rosa, CA, USA
.
Munir
H. M.
Baig
M. S.
Mirza
K.
2006
Upper cretaceous of Hazara and Paleogene biostratigraphy of Azad Kashmir, North-West Himalayas, Pakistan
.
Geology Bulletin, Punjab University
40–41
.
Munir
S.
Mashiatullah
A.
Mahmood
S.
Javed
T.
Khan
M. S.
Zafar
M.
2011
Assessment of groundwater quality using physiochemical and geochemical analysis in vicinity of Nala Lai, Islamabad
.
The Nucleus
48
,
149
158
.
Nelson
D.
2002
Natural variations in the composition of groundwater
.
Presented at Groundwater Foundation Annual Meeting, Drinking Water Program, Oregon Department of Human Services, Springfield, OR, USA
.
Nitrate
2007
Method 9210a Potentiometric Determination of Nitrate in Aqueous Samples with an Ion-selective Electrode
.
SW-846
.
Rosen
M. R.
Kropf
C.
2009
Nitrates in Southwest groundwater
.
Southwest Hydrology
8
(
4
),
20
28
.
Sajjad
M.
Rahim
S.
Tahir
S. S.
1998
Chemical quality of groundwater in Rawalpindi/Islamabad. In:
24th WEDC Conference, Water, Engineering and Development Centre, pp.
271
274
.
Skoog
D. A.
West
D. M.
1980
Principles of Instrumental Analysis
.
Saunders College
,
Philadelphia
, p.
158
.
WHO
2004
Guidelines for Drinking Water Quality
. 3rd edn,
WHO
,
Geneva, Switzerland
.
Wilde
F. D.
Radke
D. B.
2001
US Geological Survey TWRI Book 9
.
Wright
M. T.
Belitz
K.
2011
Status and Understanding of Groundwater Quality in the San Diego Drainages Hydrologic Province, 2004—California GAMA Priority Basin Project: US Geological Survey Scientific Investigations Report 2011-5154
, p.
102
.