Introduction

Fractured rocks pose a great challenge to the earth scientists (hydrogeologists, geophysicists, geochemists) as well as to engineers and environmentalists. They play an important role in the groundwater flow, sustainability of groundwater, transport of pollutants, siting of successful bore holes and wells, geothermal resources, etc. This challenge emerges from the heterogeneities and discontinuities in fractured rocks, as the flow and transport processes within discontinuities are complex in comparison to those in porous media. There exists a great contrast between the hydraulic conductivity of the fractures and that of the matrix, and the permeability of some fracture zones may be higher by orders of magnitude than that of matrix. Flow through fractures and transport of contaminants could be very significant, especially when the fractures serve as preferential flow paths.

During the last three decades, a great effort employing multi-disciplinary scientific tools has helped in addressing the problems associated with the fractured rock aquifers (Faybishenko et al. 2000). One of the important approaches has been to use isotopic and geochemical methods (Scanlon et al. 2002; de Vries and Simmers 2002). Isotopic methods have proved valuable in elucidating the recharge processes and flow mechanisms in fractured hard rocks. Within the framework of the unsaturated zone, quantification of recharge through the matrix and that through the fractures is vital for making a realistic assessment of groundwater recharge through precipitation and secondary sources (Allison and Hughes 1978; Verhagen et al. 1979; Sukhija et al. 1996; Wood et al. 1997; Rangarajan and Athavale 2000). Of late, it has been observed that groundwater flow through macro pores in the unsaturated zone is significant (Nativ et al. 1995; Davidson et al. 1998; Sukhija et al. 2000, 2003). The preferential flow through the macro-pores is even observed in alluvial tracts (Dahan et al. 2000), notwithstanding the bulk of transport taking place through the micro-pores, thus showing the piston flow model to be valid (Jury and Wang 2000).

The purpose of this paper is to study the recharge processes and flow dynamics involving weathered and fractured zones using environmental isotopes and geochemical tracer (chloride) in the hard rock region of Hyderabad (India). The research work was carried out mainly around Hyderabad (India) urban conglomerate and its suburbs (Ranga Reddy District). The role of recharge processes and also the combined effect on the quality of groundwater and the flow mechanism involved is inferred. The isotopic and chloride data demonstrate that the groundwater flow can be very complex in the hard rock type systems. Using radiocarbon ages, δ18O and chloride concentration of groundwater samples, a conceptual hydrogeologic model has evolved to address complex flow dynamics and recharge processes of groundwater in the studied fractured rock aquifers in granites. Such a model is compared with other known models (Cook et al. 1996).

The study area

The study area mainly comprises Hyderabad and its suburbs, the Ranga Reddy district, Andhra Pradesh state, India (Fig. 1). The region is underlain by a Precambrian granite-gneissic complex with dolerite intrusives and many outcrops; undulating topography characterizes the area. The granite gneissic complex occurs in two colors, gray and pink. Pink granites are coarse grained and younger than gray granites. The contact zone between these two rock types is favorable for weathering, creating potential zones for groundwater development.

Fig. 1
figure 1

Location map of deep groundwater sampled wells in and around Hyderabad City, Andhra Pradesh (India)

Full size image

As there is no primary porosity in these rocks, water-holding capacity and water-yielding capacity of the rocks depend on secondary porosity developed due to weathering, fracturing, joints, etc. The thickness of the weathered zone is minimum (0.5–2.0 m) near the outcrops and hill slopes, and maximum (10–15 m) in topographical lows and valleys. The area has a large number of shallow lakes (surface reservoirs), though urbanization has encroached upon many of them, replacing them with huge masonry structures. Apart from a few reservoirs being used for domestic water supply, like Himayat Sagar and Osman Sagar (not shown in Fig. 1) in the west of Hyderabad, many other big and small reservoirs such as Hussain Sagar, Miralam Cheruvu, etc. (Fig. 1) are evaporating water bodies. The average annual rainfall in the studied area is 750 mm and the climate is semi-arid with recorded maximum and minimum temperatures of 44 and 10°C respectively.

In the study area, groundwater occurs under phreatic conditions in shallow weathered zones and under semi-confined and confined conditions in the fractured and sheared zones at deeper levels. However, it is very difficult to separate the different aquifers on a regional scale because the system is very heterogeneous. Natural recharge measurements carried out in the hard rock areas of India using tracers estimated the groundwater recharge to be 7–8% of precipitation considering only the piston flow component of recharge (Sukhija et al. 1996). However, it is estimated that a substantial quantity of recharge takes place through preferred pathways (Sukhija et al. 2000, 2003) in the unsaturated zone. At present, except for some low-lying areas, groundwater occurs at a depth of 20–30 m below ground level, while the top weathered zone is practically dry. Therefore, the present groundwater exploitation is from fracture zones occurring between 50–150 m depths. This scenario has emerged because of over-exploitation of groundwater necessitated by extensive and intensive domestic and industrial usage.

Methods

The environmental tracers methods were utilized to characterize the movement of deep groundwater in fractured rock aquifer through determination of radiocarbon ages, δ18O and chloride concentration in groundwater. Isotopes and geochemical tracers have been extensively utilized to identify the source and movement of groundwater (Vogel 1970; Geyh and Backhaus 1979; Sukhija et al. 1996; Dahan et al. 2000).

14C Isotope

Radioactive isotope of carbon (14C6), produced in the upper atmosphere due to cosmic ray interaction with atmospheric nitrogen, gets oxidized to CO2 and mixes with the atmospheric CO2 reservoir, then assimilated by plants for photosynthesis, released in the subsoil through root respiration, and reacts with soil moisture and soil calcium forming CaCO3 and Ca(HCO3)2. For groundwater age determinations 14C is selected due to availability of its measuring facility with us. The 14C dating of groundwater can extend up to 45,000 years before present with respect to 1950 (45 Ky B.P.). Dissolved inorganic carbon species from the collected groundwater samples were obtained in the form of precipitates, using Ba(OH)2 and BaCl2. Carbon dioxide evolved through acidification of the precipitates was converted to methane, which was used for radiocarbon measurements.

The groundwater ages were determined using the radiocarbon activity of groundwater samples. The age calculations are based on radiocarbon half-life 5,730 yr and using 100 PMC (percent modern carbon) as initial 14C activity. No attempt was made to correct 14C age using δ13C, because there is no dead carbon source in the form of CO3 or HCO3 in the weathered or solid granite. This argument is further supported from the radiocarbon measurements of the samples from a number of shallow wells where the 14C ages were found to be modern. Since absolute dates of chronological events of paleo-waters were not used, relative ages can be used for elucidating recharge processes. In addition, the δ13C content of most of the measured groundwaters did not vary much (range: −11 to −12‰). The precision for 14C measurements is ±1.8 PMC (percent modern carbon). The other isotopes viz. 36Cl and 41Ca require accelerator mass spectrometry (AMS) for dating but the age limit starts from 10 Ky onwards. Groundwater ages in association with other isotopic and tracer data enables the investigators to infer its source and flow dynamics.

18O Isotope

The stable isotope (deuterium and oxygen-18) composition of precipitation undergoes variations due to altitude, latitude, amount of rainfall, temperature, etc. The heavy isotope composition of groundwater can be helpful as an indication of the genesis and mixing of groundwater in hydrogeological studies (Gat and Gonfiantini 1981). The water molecules with different masses exhibit differing vapour pressure so that during a change of phase viz. evaporation, condensation and sublimation, enrichment of heavier isotopes in the more volatile phase takes place. This process is known as isotope fractionation. The higher temperature and low precipitation during the early rainfalls of the summer monsoon result in the enriched heavy isotope content, and depleted heavy isotope content of heavy rainfall during subsequent months of the summer monsoon.

Soil and subsurface waters inherit the isotopic characteristics of the meteoric and surface water inputs, and change in isotopic composition occurs as a result of recharge process and mixing with waters of different compositions. In semi-arid zones, the water loss by evaporation assumes importance in the sense that heavy isotope content of groundwater gets enriched relative to that of precipitation.

Oxygen-18 (18O) stable isotope is utilized to identify the recharge process and mixing of groundwater from different sources. Surface waters contain a distinct composition of stable isotopes due to enrichment caused by evaporation. Therefore, 18O isotope makes it possible to distinguish between evaporated waters and waters directly recharged from precipitation. Artificial recharge, large-scale irrigation and over exploitation of aquifers alter the isotopic composition of groundwater through admixture of different flow regimes.

δ18O was measured on water samples using a mass spectrometer at the Physical Research Laboratory, in Ahmedabad, India, and δ18O concentrations (‰) were obtained in comparison to SMOW (Standard Mean Ocean Water). The precision for δ18O was ±0.1‰. During the study period, rainfall data and δ18O content was available only for the year 1998. Nevertheless, the weighted mean values of δ18O would represent the typical value for Hyderabad (India) rainfall.

Environmental chloride

Environmental chloride has potential utility in hydrology. It is deposited on land by atmospheric deposition processes (rainfall and dry fallout) and chloride fallout is concentrated near coastlines. The other important sources of chloride in addition to precipitation could be from applied fertilizers, human waste, etc. If the chloride present in the unsaturated zone and groundwaters has surface environment (i.e. sea) as its source, and there exists no other source or sink underground, it is possible to obtain chloride balance. The ratio of chloride concentration in rainwater to that in percolated water is utilized to estimate recharge. In semi-arid regions, combined geochemical and isotopic studies of both unsaturated zones and shallow groundwaters can be effectively undertaken to determine not only the present day direct recharge component, but also a recharge chronology of immediate historic times, which may be important in the estimation of long-term water resources (Edmunds and Walton 1980).

Many researchers have effectively utilized the environmental chloride present in the unsaturated zone and groundwater regimes for natural and artificial recharge estimates, salinity problems, soil water movement studies, etc. Continuous monitoring of the chloride concentration and hydraulic head in space and time can reveal important information about change in groundwater recharge characteristics as a result of deforestation (Williamson et al. 1987).

The chloride concentration of sampled groundwaters and surface waters were measured using an ion-selective electrode. The precision for chloride measurements was ±5%.

In all, 34 deep groundwater samples were collected from Hyderabad and its suburbs. Sampling was done covering the maximum possible geomorphic units including elevated grounds, low-lying areas and close to surface reservoirs to observe their influence. Depth of the sampled wells ranged between 20–200 m. However, water-yielding zones are restricted to only limited thickness i.e., sheet joints or fracture zones. Most of the sampled wells draw water from sheet joints/deep fractures, as there is no water in the phreatic zone. The information about fractures was obtained from the driller’s records. Water samples were collected during pumping, and in case of more than one fracture zone it is likely that the water from different fractures was mixed, as no attempt was made to collect water from individual fracture zones.

Results

In order to elucidate recharge processes and groundwater flow mechanism in fractured granitic aquifers, wells were sampled from Hyderabad and the adjoining District of Ranga Reddy.

The highly heterogeneous character of the fractured hard rock aquifers in the study area is evident from the exploratory drilling carried out by the Central Ground Water Board (CGWB) in Hyderabad. The drilling indicated the existence of fracture zones at various depths between 20 and 173 m, and the discharge rate (measured using 90° V-notch) during drilling varied from 0.02 to 6.91 l/s (l/s) (Fig. 2) (Reddy et al. 2002). The fracture depths from the ground level are recorded by the driller using the length of drilling pipe.

Fig. 2
figure 2

Litho-logs of some typical exploratory wells drilled by Central Ground Water Board in Hyderabad City, indicating different sheet joints and its discharge during drilling

Full size image

As can be seen from Fig. 2, wells from different parts of Hyderabad indicate that the thickness of the weathered zone is meager (0.5–2.0 m) in elevated areas (Borabanda and Film Nagar, Fig. 1), and quite substantial (about 15 m thick) in the low-lying areas along the river/stream (Gowliguda and Lower Tank Bund).

Though Borabanda well passed deeply into basement rock, it encountered two fractures at a depth of 93–95 m and 135–137 m with a discharge of only 0.02 and 0.08 l/s, respectively, during drilling. However, another sheet joint at a depth of 172–173 m provided a discharge of about 7.0 l/s. The Film Nagar well, also in basement rock, encountered a sheet joint at a depth of 127–128 m (Fig. 2) with a discharge rate of 5.5 l/s. Though a thick weathered zone was encountered in the Lower Tank Bund and Gowliguda areas (Fig. 2), the discharge during drilling of wells in these areas was nil. This shows that the weathered zone in most of the study area is presently non-yielding. The above drilling data demonstrate the control of fractures on well yields in the hard rock aquifers of the studied area.

Another very important field observation was the existence of laterally extensive sheet joints. These were encountered in drilling at different depths in granites as revealed in a quarry shown in the photograph (Fig. 3). This figure represents the geologic section obtained from an open rock quarry of about 10 m deep and about 200–300 m long. However, the sheet joints extend laterally much beyond the exposed quarry. The geologic section shows three sets of sheet joints dipping southwards with a gradient of about 5–10°. The thickness of these sheet joints varies from 30–50 cm with 2–3 m thick solid granite below and above. It is expected that once the recharge water enters the sheet joint, it moves horizontally for a considerable distance. In fact, such sheet joints are explored in fractured hard rock areas, which serve commonly as a source for groundwater that occurs under sub-artesian conditions due to the presence of a confining medium. A general hydrogeological perception is that such sheet joints may extend laterally from a few hundred meters (as shown in Fig. 3) to perhaps up to kilometers, though no detailed investigations are available to support this. Some of the drilled wells shown in Fig. 2 may be tapping sheet joints similar to those shown in Fig. 3, yet because of lack of sufficient borehole data, it is difficult to prove their lateral extent. Lateral extent of fractures shown in Fig. 2 cannot be determined from the available bore well data.

Fig. 3
figure 3

Photograph shows the sheet joints in the granite basement rock in the exposed quarry near Hyderabad, which act as potential groundwater zones

Full size image

Table 1 presents the results of measurements of chloride concentration and δ18O of Hyderabad precipitation. The radiocarbon ages, chloride concentration and δ18O of groundwater samples with their location and fracture depths are shown in Table 2.

Table 1: δ18O and chloride data of Hyderabad rainfall for the period 1997–2001
Full size table
Table 2: Radiocarbon dates (uncorrected), chloride and δ18O concentrations of groundwaters collected from fractured hard rock aquifers
Full size table

Monthly average chloride concentration of precipitation at Hyderabad and its environs varies from 1–10 mg/l. Generally, a low chloride concentration is associated with heavy precipitation and vice-versa (amount effect). Chloride concentration of collected groundwater samples varies between 20 and 350 mg/l. However, the majority of samples have chloride concentration between 50 and 200 mg/l, several samples have concentrations <50 mg/l, and very few samples have concentration >200 mg/l. The measured average chloride concentrations in four major surface water bodies are 243 mg/l (Mir Alam Tank), 210 mg/l (Hussain Sagar), 11 mg/l (Himayat Sagar) and 11 mg/l (Osman Sagar). Out of these, Himayat Sagar and Osman Sagar, are located away from the urban area and are being used for drinking water supply to Hyderabad, while the Hussain Sagar and Mir Alam Tank are surrounded by thickly populated areas and are not protected, hence, enormously polluted water finds its way into them.

The measured δ18O concentration in the Hyderabad rainwater varied from –10.2 to +4.2‰ SMOW. The δ18O concentration was positive for initial showers of the summer monsoon, and in subsequent rainfall it was relatively depleted (Table 1). Kumar and Kingi (1981) measured δ18O concentration in a surface reservoir which is being used for domestic water supply as ∼4.0‰ (SMOW). The groundwater samples collected for this study show the δ18O content between –3.2‰ and +1.7‰ (SMOW). However, about 50% of the samples show around –2‰ and about 25% of the samples show enrichment of δ18O (positive values).

As can be seen from Table 2, there is a wide range of uncorrected radiocarbon ages from “Modern (≤50 yr)” to 5570 yr B.P. More than half of the samples show “Modern” age and quite a few samples show higher ages (>1000 yrs) particularly from the northwestern part of the study area.

Discussion

For fractured hard rock aquifers, some conceptual models have been developed to address groundwater flow and solute transport, using short-term tracer methods as well as environmental isotopes (especially chlorofluorocarbon, tritium, tritium-helium-3 ratio) methods (McKay et al. 1993; Cook et al. 1996; Cook and Simmons 2000; Flint et al. 2000; Shapiro 2001). The advantage of environmental tracers is quite understandable, as experiments using dye tracers require field injections to be pursued for a considerable time in order to study equilibrium between matrix and fracture flows. Further, the transport models evolved from small-scale measurements of aquifer properties may not truly represent the actual large-scale solute migration rates as the aquifer parameters in fractured hard rocks are prone to large spatial variability (Aeschbach-Hertig et al. 1998). A conceptual hydrogeologic model, evolved from the present study, is compared with some of the conceptual models developed by others in various geologic settings. The environmental tracer data discussed in this paper include the development of a conceptual model of the recharge processes in a granitic terrain.

Depth variation of environmental tracer data

In order to elucidate the recharge process and flow mechanism in fractured aquifers of granite- gneisses, measurements of radiocarbon, δ18O and chloride were made. These parameters were plotted as a function of depth (Fig. 4). Fig. 4a shows the depth variation of drilling discharge for individual fractures at that depth; few of the sampled wells depicted in Fig. 4a are shown in Fig. 2. There is no systematic trend of either increase or decrease of drilling discharge with depth. They show much higher discharge than average. For example, high drilling discharge is encountered at different depths viz. 16–25, 34–72, 127–128, 172–173 m in different wells (Fig. 2).

Fig. 4
figure 4

Depth variation of (a) exploratory well discharge during drilling, (b) chloride concentration, (c) δ18O content and (d) uncorrected 14C ages of sampled groundwaters (exploratory and sampled wells for groundwaters are different)

Full size image

Figure 4b and c show chloride and δ18O plots, with respect to depth of fracture respectively, which indicate a random variation. Such variation can be attributed to multiple sources of recharge as well as multiple flow paths at various stages of the recharge process. Figure 4d is the plot of depth variation of 14C ages. Unlike the chloride and δ18O variations, which are primarily controlled by environmental conditions during the recharge process, the variation in 14C ages is because of residence time of groundwater in the aquifer. The measured 14C ages do not show any increase with depth, which is in contrast to the observation of Cook et al. (1996) and Aeschbach-Hertig et al. (1998), wherein they found increasing groundwater ages with depth. Our observations indicate that because of highly undulating topography, deeper fractures also may have recharge sources close by.

In the present study area (granite basement), the hydrogeologic setting varies on a regional scale because of different sets of fractures/joints as well as thickness of the weathered zone. There are thick weathered zones at places, and at other places, outcrops with a thin weathered zone accompanied by vertical fracture zones and a combination of thick weathered and fracture zones is observed elsewhere. In such a hydrogeologic frame work, since there is substantial flow through vertical fractures and inter-granular pores in thick weathered zones, and there exists a little lateral flow, particularly in low-lying valley areas, the residence time of water in the aquifer is small as water does not travel to any great depth/distances before it is withdrawn from the wells. Thus, quite young ages (Modern) in this scenario are expected. On the other hand, if the recharging water from the vertical fractures in weathered zones enters the sheet joints, which are confined by the compact hard rock on top and bottom, it subsequently maintains its stable isotope and chemical composition (chloride, assuming non-reactive system), but radiocarbon concentration is expected to decrease with time. Thus it can be inferred that high radiocarbon ages in groundwater could result from the movement of groundwater in extensive sheet joints (spanning tens of meters to few kilometers) connected either to vertical joints in the vadose zone or thin weathered zone which are found to be prevalent in the fractured granites (as evidenced in Fig. 3). Thus considering the overall hydrogeological situation, no systematic depth variation of different tracers would occur.

Relationship between 14C ages, chloride and δ18O

The studied groundwater ages, chloride concentration and δ18O data are plotted in Fig. 5. This figure shows broadly two types of groundwaters, one possessing age range from 1300–5600 yr B.P. and the other group having Modern to 800 yr B.P. The high ages could result from movement of the groundwater to large distances through the sheet joints (Fig. 3), which are confined. Once the recharging waters from the weathered zone enters these sheet joints, the hydrochemical signatures would not alter except for a decrease in the 14C content due to radioactive decay. If the transit time of the recharging water were large, before its entry into the abstraction point, the groundwater age would be proportionately high and vice-versa. The wells, which possessed the high ages, are: 4, 12, 15, 17, 18, 27, 39 and 47. The chloride content of these wells is ≤100 mg/l except well no. 39, which has 325 mg/l of chloride content. The δ18O content of these wells ranged from −3.2 to −1.7‰. The enrichment of chloride and δ18O content, in relation to the rainwater values, suggests normal evaporation process during recharge through unsaturated zone. After passage of these waters through the vadose zone, the waters enter the extensive sheet joints, thereby, the chloride and δ18O contents remain intact. The exceptionally high chloride content of well no. 39 suggests a contamination from a local subsurface chloride rich water source (e.g. septic tank, leaking drainage etc.).

Fig. 5
figure 5

Relationship between 14C ages, δ18O and chloride concentration of studied groundwaters in the fractured hard rocks, reveal broadly two types of groundwaters. One type groundwaters is characterized by high 14C ages (1300 to 5600 yr B.P.), δ18O −3.2 to −1.5‰ and chloride concentration of <100 mg/l, which imply recharge processes involving weathered zone followed by large transit time through extensive sheet joints. The second type is comprised of 14C age (Modern to ∼ 800 yr B.P.), a wide range of δ18O (−3.2 to +1.7‰) and chloride concentrations (up to 350 mg/l). The three sub-sets are resultant of recharge involving weathered zone and vertical fractures (δ18O:−3.2 to −1.5‰, Cl: ≤100 mg/l, 14C ages: Modern to 800 yr B.P.), local contamination from chloride rich sub-surface sources (δ18O ∼ −1.5‰, Cl: 100–150 mg/l, 14C ages: Modern), and contribution from polluted and evaporative surface water bodies ((δ18O ∼ +0.03 to +1.7‰, Cl: >150 mg/l, 14C ages: Modern)

Full size image

The second broad group having 14C ages Modern to 800 yr B.P., contained chloride in the range 20 to 350 mg/l and δ18O in the range −3.2 to +1.7‰. Such a wide range in chloride and δ18O contents of groundwater demonstrates multiple processes and sources of recharge. This group is further sub divided into three subgroups. The groundwaters that contained chloride <100 mg/l and δ18O < -1.5 ‰, indicate normal recharge through weathered zone and fractures. Some of these waters have residence time to the extent of 800 years, implying moderate transit time through fractures. The above subgroup with <100 mg/l of chloride, shows enrichment of δ18O during percolation of recharging waters through the unsaturated zone, before entering the fracture network. The well nos. 16, 21, 22, 35, and 42 exemplify this type of groundwaters.

The other subgroup of groundwaters with Modern ages, chloride content > 150 mg/l and having positive δ18O values (0.3 to +1.7‰) clearly show the recharge from polluting and evaporative surface water bodies. The δ18O content of surface water bodies undergo enrichment due to continued evaporation. The typical δ18O content of one of the surface water bodies in the present study area was 4.0‰ (Kumar and Kingi 1981). Thus, the observed positive δ18O values seem to have resulted from mixing of normal groundwater and feed from surface water bodies. The wells 29 and 38 represent this category of groundwaters.

The contribution of evaporative surface reservoirs to the groundwaters, having δ18O in the range of −0.9 to +1.7‰, is computed taking the δ18O content of recharging waters in the unsaturated zone and fractures as one end member while the second end member is the δ18O value of tank water, which was determined to be +4.0‰ (Kumar and Kingi 1981) for one of the surface water bodies in the present study area. The first δ18O end member is computed using the rainfall δ18O value (−6.5‰) and assuming 5% annual average recharge (Sukhija et al. 2003) through matrix flow and about 50% of bypass flow to the total recharge (Sukhija et al. 2003). The resultant δ18O input from 5% matrix flow would be −0.32‰, and from bypass flow, it will be −3.25‰. Thus the total input δ18O content would be −3.57‰ (as one end member). Then the contribution from surface reservoir, computed taking −3.57‰ (recharging waters) and +4.0‰ (tank waters) as two end members, ranged from 43 to 70% corresponding to the resultant δ18O content −0.3 to +1.7‰ of groundwaters.

Quantification of the mixing of recharging groundwaters and waters from chloride-rich subsurface sources is not attempted owing to various non-point diffusive sources like septic tanks and leaking drainage pipes in the subsurface. Alternatively, enriched δ18O groundwaters are considered to address the mixing phenomenon from sample reservoirs.

The groundwaters with intermediate δ18O values around −1.0‰ and chloride contents in the range of 100 to 150 mg/l and having Modern 14C ages, demonstrate contribution from subsurface contaminant sources such as septic tanks and leaking drainage pipes. The groundwaters from well nos. 19, 26, 33, 34 etc. belong to this category. The various recharge processes and their representative wells are listed in Table 3.

Table 3 The characteristics of studied deep groundwaters
Full size table

Conceptual hydrogeologic model

The source and recharge processes of different types of groundwaters with their isotope and chloride data can be interpreted using Fig. 6, which represents a conceptual hydrogeologic model of granite-gneissic area. The model takes into account the following observations: (i) lithology as revealed through a number of boreholes in and around Hyderabad and quarries, (ii) geomorphic variations, (iii) qualitative yields of fracture zones, (iv) variation of δ18O and chloride concentration in groundwater, (v) high and low radiocarbon ages etc.

Fig. 6
figure 6

A conceptual hydrogeologic model evolved to elucidate various recharge processes and flow mechanisms in the studied fractured hard rock aquifers, based on varying 14C ages, chloride and δ18O contents of groundwater. Hypothetical wells like BH-1 and BH-2 are recharged from distant sources and water moves through the weathered zone and sheet joints resulting in quite large residence times of groundwaters. Wells like BH-3 are characterized by young 14C ages due to low turnover time, and relatively high chloride concentrations in comparison to BH-1 and BH-2 result from natural recharge through a thick weathered zone. Groundwaters being recharged from surface water bodies connected through an intense fracture network are characterized by young 14C ages and positive δ18O values and are represented by wells like BH-4. The groundwaters in the valley (low lying) areas are characterized by a thick weathered zone, intensive fracturing in the subsurface, and minimal lateral flow of groundwaters are represented by wells like BH-5 which possess young 14C ages and high levels of chloride due to mixing of waters from subsurface contaminant sources (soak pits, drainage lines etc.)

Full size image

The model shows varying thickness of weathered zone overlying hard and fractured basement. Field observations indicate that a thick weathered zone is generally associated with an intense fracture network as well as with the courses of past and present streams. From the observation of a number of quarries (Fig. 3) and boreholes (Fig. 2) the general perception is that sheet joints which are a few cm to tens of cm thick, are quite extensive and are connected to high yielding wells. Such sheet joints can be found at various depths in the study area, and are connected to an upper horizon, which may have a thin weathered zone for recharging the aquifer or vertical joints in the unsaturated zone. Another important aspect of the hydrogeologic system is the presence of surface water bodies, which could serve as one of the recharge sources.

Thus the type of groundwater samples with 14C ages >1300 yr B.P., chloride concentration ≤100 mg/l, and δ18O <−1.7‰ (Table 3) are indicative of having relatively large transit time and recharged through upper weathered zones and sheet joints. This situation can be described from the conceptual hydrogeologic model (Fig. 6) in which hypothetical wells like BH-1 and BH-2 are located where the source of recharge is some distance away. In an actual field situation, the sampled groundwater wells numbered 4, 12, 15, 17, 18, 27, 39 and 47, mainly from the Tolichowki area (Fig. 1), represent the above hydrogeological situation. From the radiocarbon age data and considering extensive fracture zones (Fig. 2) with good yield (5–7 l/s) prevalent in the area, it is envisaged that groundwater moves laterally through such fracture zones for sufficiently long periods and over distant areas rather than consider that groundwater is locked in a deeper horizon in low permeability fractures as we have not observed any depth variation. Flint et al. (2000) in their attempt to estimate regional recharge and travel time through the unsaturated zone of arid fractured rocks in Yucca Mountains, Nevada, conceptualised a model of groundwater recharge. According to them, recharge is nearly vertical and the travel time to the saturated zone is dependent on the thickness of the unsaturated zone and effective travel path porosity. The flow path may be in a rock matrix or in fractures. Further, a significant flux of recharge occurs through fractures under saturated or nearly saturated conditions. Infiltrated water is expected to travel laterally through sheet joints until intercepted by deep wells. A higher infiltration rate is assumed to take place through fractures while a lower value of infiltration moves through the matrix. However, Cook and Simmons (2000), in their study using environmental tracers of 14C, CFC-12, and 3H, and in combination with hydraulic properties in Clare valley, Australia, have identified two separate groundwater systems with contrasting vertical recharge to be 100 mm/yr for the upper fractured system and less than 0.1 mm/yr for the deeper flow system. Since the present study area is in the part of the Peninsular granitic shield, extensive sheet joints are common, through which groundwater moves laterally for a long distance, whereas groundwater movement in Yucca Mountains and Clare valley is almost vertical.

The type of groundwater samples having Modern to 800 yr B.P. radiocarbon ages with 20 to 350 mg/l chloride concentration and δ18O values from −3.2 to +1.7‰ (Table 3) involve various recharge processes and sources as detailed below:

Wells like BH-3 (Fig. 6) form subgroup-1 (14C age range: Modern to 800 yr B.P., Chloride: ≤100 mg/l, δ18O: ≤−1.5‰, Table 3) where the source of recharge is at a moderate distance, essentially receives recharge from a thick weathered zone and there may be vertical recharge to a considerable extent. In comparison to the groundwater whose residence time in the sheet joints is >1300 yr B.P., this type of well may have less travel time as well as residence time in the aquifer. This group consists of the samples like 16, 21, 22, 35 and 42, where the weathered zone is thick and water is recharged through this zone.

Wells like BH-4 form subgroup-2 (14C ages: Modern, chloride: >150 mg/l, δ18O: +0.3 to +1.7‰, Table 3), which are connected to the nearby polluting and evaporative surface water bodies through weathered and fractured zone. In the study area, there are prominent polluting surface water bodies like Hussain Sagar, Mir Alam Cheruvu, Nacharam Cheruvu (Fig. 1), which are likely to contribute to groundwater recharge. Groundwater samples like 29 and 38 may have their recharge from such sources, as evidenced by enriched δ18O and higher chloride concentrations.

Wells like BH-5 form subgroup-3 (14C ages: Modern, chloride: 100 to 150 mg/l, δ18O: around − 1.0‰, Table 3), situated in low-lying areas where a thick weathered zone with an intense fracture system, are prone to accumulation of groundwater along with feed from subsurface pollution sources. Groundwater samples like 19, 26, 33, 34 may have their recharge from such sources, as evidenced by higher chloride concentrations.

The hydrogeologic model presented above has also some similarities with a conceptual model of groundwater flow developed by Cook et al. (1996) for Saprolite and fractured rock using environmental tracers. In this model, most of the recharge is through discrete fractures in the vadose zone that becomes saturated during rainy season, though the micropores may remain unsaturated and most recharge to groundwater is transmitted laterally. In the hydrogeologic model developed in this study, it is believed that sheet fractures, which are connected to the vadose zone, carry recharge water laterally a few hundreds of meters to a few kilometers, which result in ageing of the recharged water. However, in contrast to the above situation, younger radiocarbon ages with similar chemical and isotopic characterization are also obtained. This situation may occur due to infiltration and movement of water through the matrix as well as the fractures within the weathered zone, and mix together resulting in apparent young age of groundwater. Such systems are generally unconfined in nature. In the former case involving sheet joints, higher groundwater ages are associated with semi artesian conditions.

However, despite the fact that the studied fractured hard rock aquifer system is a complex geologic environment comprising unsaturated rock matrix and multiple fracture network, the study has identified dominant recharge processes and flow mechanisms operative in the area. The possible sources of errors that would offset the data from its real value could be recharge contributions from unidentifiable subsurface regimes as well as contribution from multiple fractures during pumping of wells.

Conclusion

The isotopic and geochemical analysis of various groundwaters of the studied area with undulating topography, fractured hard rock outcrops and basement, varying thickness of weathered zone with low primary porosity and dominant secondary porosity, identified various recharge processes operative in the study area. In such a complex geological setting, the recharge processes involved unsaturated rock matrix, high permeable paths in the regolith, sheet joints etc., thus demonstrating the groundwater flow pattern to be quite complicated involving various sources. Apart from rainfall, the recharge sources include surface water bodies and the subsurface polluting sources like leaking drainage pipes and septic tanks. The contribution from surface reservoir to groundwater is evaluated and found to be significant (40 to 70%) for certain type of groundwaters. To understand such a heterogeneous hydrogeological scenario, a model has been developed based on the isotopic and geochemical data. The developed model could be applicable to areas having similar hydrogeological environments. The importance of such models further increases from the practical utility of studying the transport of contaminants to the aquifers in the industrial and urban areas.