Introduction

Lakes are major surface fresh water resources (Shiklomanov 1993; Dessie et al. 2015) and are the main sources of meeting the demand for drinking, irrigation, water transport recreation, fishery and industrial purposes (Meybeck 1995; Leonard and Crouzet 1999; Kalff 2001; Shiklomanov and Rodda 2003; Wu et al. 2017). Globally, there are more than 110,000 natural lakes (>1 km2) covering a total area of 2.3 × 106 km2 (Kebede et al. 2006; Chebud and Melesse 2009) and millions of smaller lakes (Meybeck 1995; Finlayson et al. 1999; Kalff 2001). Despite the importance of the lakes, they are facing major problems, which include lowering of water level, siltation, acidification, toxic contamination and eutrophication (Kira 1997; Ballatore and Muhandiki 2002), posing risks to the environment and human health (Li et al. 2016). To cope with such problems and for the management purposes, accurate determination of the water balance of lakes is required.

Groundwater is a life-sustaining resource (Rosenberry et al. 2015) that supplies water to billions of people (Giordano 2009; Sun et al. 2009; Siebert et al. 2010; Gleeson et al. 2012; Li et al. 2013a), plays a central part in recharging the lake basins of the world (Nakayama and Watanabe 2008). It is hydraulically connected to lakes and an understanding of this interaction is fundamental for effective water resource management (Winter et al. 1998; Hayashi and Rosenberry 2002; Brodie et al. 2007; Cook et al. 2001; Owor et al. 2011). Interaction between them influences the characteristics of lake water, including the stability of water level and water quality (Winter et al. 1998; Barr et al. 2000; Schneider et al. 2005). These interactions are complex both in time and space, and are not only influenced by climate, landform, geology, and biotic factors but also by fabricated human activities (Kenoyer and Anderson 1989; Shaw and Prepas 1990; Sebestyen and Schneider 2001).

Environmental isotopes are naturally occurring tracers (Behrens et al. 1971; Martinec et al. 1974; Rodhe 1981; Obradovic and Sklash 1986; Maule and Stein 1990; Maurya et al. 2011; Cable et al. 2011; Wassenaar et al. 2011; Jeelani et al. 2013; Qian et al. 2013) that provide quantitative information about lake inflow components (groundwater, surface water, and precipitation) (Maloszewski and Zuber 1982, 1983; Mattle et al. 2001; Qian et al. 2014). Since each component of a lake’s water budget acquires a different isotopic signature (Bennett et al. 2008; Gibson et al. 2010a, b), well-mixed lakes act as “mixing bowls” (Gibson and Edwards 2002; Turner et al. 2010; Brooks et al. 2014; Gibson et al. 2015) and reflect the net influence of all the components. The isotopic signatures of lake water which comprises the mixture of recharge of these accordingly attribute the particular isotopic signature (Dawidek and Ferencz 2014; Kummu et al. 2014). Due to the enrichment of δ18O and δ2H in lake water that undergoes evaporation (Gonfiantini 1986; Gibson et al. 2002; Gibson and Edwards 2002), the resulting isotopic signature is different from that of ground water and provides an ideal conservative tracer for evaluating the extent of mixing of surface water and groundwater (Gonfiantini 1986; Shivanna et al. 2004; Jeelani et al. 2010). The light isotopic species of water (δ16O) has a higher vapor pressure and diffusivity than the isotopically heavy species (δ18O and δD) and is preferentially removed from a lake during evaporation (Gibson and Edwards 2002; Dawidek and Ferencz 2014; Kummu et al. 2014). The lake thereby becomes enriched in the heavy isotopic species of water. The degree of enrichment depends on climatic conditions, including average annual temperature, relative humidity, and precipitation and evaporation rates. Using these isotopes (δ18O or δ2H) isotope-balance approaches (Coplen, 2011; Sacks et al. 2014; Jeelani et al. 2015) have been used in numerous hydrological studies to quantify groundwater inflow to lakes (Turner et al. 1984; Krabbenhoft et al. 1990; LaBaugh et al. 1997).

Hydrogeochemical interaction in various settings is a well-recognized method used into trace the source, for assessing the sensitivity of lakes, seepage estimation and destination of surface water, groundwater and lake water. Physical parameters (pH, EC, and TDS) and major elements (Ca2+, Mg2+, Cl− and HCO3 −) (Zimmerman and Ehhalt 1970; Bentley et al. 1986; LaBaugh et al. 1995, 1997; Bocanegra et al. 2012; Goldhaber et al. 2014; Li et al. 2013b) can be used to trace the flow of groundwater (Allison and Hughes 1978). Groundwater discharge zones can be located by searching for anomalies in pH, electrical conductivity (EC) and total dissolved solids (TDS) values. Most conservative element chloride has been used to check water budgets by the mass balance method (e.g., Balleau 1973; Allison and Hughes 1978; Allison 1988; Allison et al. 1994; Bajjali 2006; Jeelani 2008). Eriksson and Khunakasem 1969 first proposed that chloride concentration in ground water to estimate a recharge flux in aquifers. Understanding the geochemical mass balance of lake (pH, EC, TDS, Ca2+, Mg2+, Cl− and HCO3 −) and their interactions with groundwater is important (1) for water resource management (Showstack 2004) (2) for understanding nutrient balances (Ala-aho et al. 2013; Kidmose et al. 2013) and (3) for quantifying the vulnerability of lakes to pollution (Klove et al. 2011). Lake–groundwater interactions are nevertheless complex, and transport of these solutes is highly variable, both spatially and temporally (Winter et al. 1998). It depends on contribution of subsurface water (groundwater), aquifer geology, topography, and meteorological conditions (Rosenberry and LaBaugh 2008).

Dal Lake has tremendous societal and economic significance in Kashmir Valley. However, it is continuously degraded with the increasing population and anthropogenic pressure. Tourism has also developed rapidly leading to the increasing number of hotels, houseboats in and around the lake. It is important to have both a qualitative and quantitative understanding of lake–groundwater interaction to support the development of effective management plans for sustainable use of water resources. Despite the limited work on water quality of the lake, no study has yet identified the sources of water and the contribution of these sources into the lake. The objectives of the present work are (1) to characterize the geochemical and isotopic signatures in lake water, streams and groundwater, (2) to constrain the different sources of the lake water (3) to assess the connection between groundwater and lake water and (4) to gain a comprehensive understanding of hydrological processes in the Dal Lake.

Study area

Dal, the second largest lake in the Jammu and Kashmir state (Fig. 1), lies between latitudes 34°04′ and 34°11′N and longitudes 74°48′ and 74°53′E in the northeast township of Srinagar city. The lake has a total open water spread area (water surface area) of 12 km2 (excluding areas covered by floating gardens). Some of the morphometric characteristics/hydrological parameters of the Dal Lake are summarized in Table 1. Total area of catchment is about 314 km2, and Teilbal–Dachigam sub-catchment contributes 70% areal distribution of drainage catchment (Jeelani and Shah 2006; Saleem et al. 2015). Panjal Traps (basalts, andesites), Triassic Limestone (limestone, sandstone, dolomite, shale), Agglomeratic Slates (pyroclastic slates, conglomerates, quartzite), Karewas (sand, clay, silt and loess) and Recent Alluvium are the dominant geological formations (Fig. 1) in the lake catchment, as shown in geological map (Thakur and Rawat 1992 and Jeelani and Shah 2006). The lake basin is surrounded mostly by alluvium and Karewas, which form good aquifers with moderate yields (CGWB 2013). Water table depth in the alluvial aquifers in summer ranges from 1.5 to 4 m bgl (below ground level), while in winter it increases and varies from 1.7 to 6 m bgl (CGWB 2013). Toward the eastern side of the lake, the alluvium is in contact with the Zabarwan Range, which consists mostly of Panjal Traps and Triassic Limestone. It is a multi-basin lake, which is divided into four basins; Hazratbal, Boddal, Gagribal and Nigeen basins. The average volume of water estimated in Hazratbal was 35 × 106 m3, Boddal was 18.5 × 106 m3, Gagribal was 13.7 × 106 m3, and Nigeen was 3.3 × 106 m3. Perennial streams are found only in the north and eastern side of the lake including Teilbal main stream, Habbak, Shalimar, Nishat, and Brain streams (Saleem 2015). There are two outlets from the lake, namely the Dalgate and Amir Khan nallah and former is controlled by a weir and lock system. Total annual average outflow from these two outlets has been estimated as 346 × 106 m3 a−1. The drainage pattern in the study area is sub-dendritic patterns with the general flow direction being from east to south west (Fig. 1). The total annual precipitation and average temperature recorded by meteorological station at Shalimar meteorological station (34°8′58″N and 74°52′54″E) in 2015 was 1753 mm and 12.7 °C, respectively. The climate of the area is Sub-Mediterranean type (Bagnolus and Meher-Homji 1959) dominated by the Western disturbances in winter and ISM (Indian summer monsoon) in summer (Jeelani et al. 2017).

Fig. 1
figure 1

Modified from Thakur and Rawat (1992) and Jeelani and Shah (2006)

Location map of the study area with geology and drainage of the catchment.

Full size image
Table 1 Morphometric characteristics/hydrological parameters
Full size table

Materials and methods

Water sampling

For major ion chemistry, lake water (n = 444), groundwater (n = 480) and stream samples (n = 36) were collected in high density polyethylene (HDPE) bottles during 2015 from representative sites (Fig. 2) on monthly basis. Groundwater samples were collected from open dug wells (n = 36), closed tube wells (n = 372) and springs (n = 108) located nearby the lake, and its catchment. Stream water samples were also collected from inflow streams (Teilbal, Habbak, Shalimar, Nishat, and Brain) before the confluence with lake water. Water samples were collected in 125 ml HDPE bottles from precipitation (n = 27), lake water (n = 18) and groundwater (n = 32) for δ18O and δ2H analysis. Precipitation samples were collected in homemade precipitation collectors, fitted at many selected sites across the study area. Each collector consisted of a 10-L plastic container fitted with a funnel and a long narrow tube to avoid evaporation.

Fig. 2
figure 2

Data collection and sampling location sites in and around Dal Lake (from springs, tube wells, monitoring wells)

Full size image

Analytical techniques

Physical parameters like, pH, total dissolved solids (TDS) and electrical conductivity (EC) were measured in field (in situ) using portable HACH+ (HQD Sension+). Standard methods were adopted for determination of major ions in water samples (APHA 2005). Ca2+, Mg2+, Na+ and K+ were measured by atomic absorption spectrophotometer. HCO3 − and Cl− were analyzed by acid–base and AgNO3 titration, respectively. SO4 2− and NO3 − were determined by colorimetric method (HACH DR/890). Stable water isotope ratios hydrogen (δ2H) and oxygen (δ18O) were analyzed at Physical Research Laboratory (PRL), Ahmedabad, by Dual Inlet Isotope Ratio Mass Spectrometer through gas equilibration method (Revesz and Coplen 2006). Isotope results are reported in the conventional delta notation (δ) (Coplen 1996), as given by the following equation:

$$\updelta {\text{D }}\left( {{\text{or}}\,\,\updelta^{18} {\text{O}}} \right) = \left\{ {\frac{{R_{\text{sample}} - R_{\text{standard}} }}{{R_{\text{standard}} }}} \right\} \times 10^{3}$$

where R sample represents the isotope ratio of the sample (2H/1H or 18O/16O) and R standard represents the corresponding ratio of the standard, VSMOW). The δ values are expressed in parts per thousand (permil, ‰). The precision of measurement for δ2H is ±1‰ and that of δ18O is ±0.1‰.

Water level measurement

Water table (WT) is hidden beneath the land surface; it can be directly observed through monitoring wells. It was monitored/observed on seasonal basis (2015) at forty (40) observation sites including; dug wells, tube wells, and springs (Fig. 2). It was recorded with a hand-held sensor capable of accuracy to about 0.1 inch, whereas elevation of ground surface was measured with the help of GPS (Trimble Juno). Besides, the water level of the lake was also measured simultaneously nearer to these observation locations along the periphery of the lake. The recorded data was later interpolated in Arc-GIS 10.2 software using geostatistical analyst tool for contour maps.

Isotopic mass balance

Isotopic mass balance (IMB) equation provided by Gonfiantini (1965), Dincer (1968), Krabbenhoft et al. (1990), and subsequently by Gibson et al. (2002), Gibson and Reid (2010) and Sacks et al. (2014) were simplified and used. The Dal Lake, complex dynamic system is connected to different components of hydrological cycle, through evaporation, precipitation, groundwater, surface inflows and outflow. The generalized mass balance equation for unsteady lakes is:

$$\frac{{{\text{d}}v}}{{{\text{d}}t}} = \left( {\text{Inflow}} \right) - ({\text{Outflow}} + E) = 0$$
(1)
$$\frac{{{\text{d}}v}}{{{\text{d}}t}} = (G_{\text{i}} + P + S_{\text{i}} ) - (G_{\text{o}} + S_{\text{o}} + E) = 0$$
(2)

where G i and S i are the groundwater and surface water inflow rates, respectively, G o and S o are the groundwater and surface water outflow rates, respectively, P and E are the precipitation and evaporation rates, respectively, v is the volume of water in the lake, and t is time. The annual isotope mass balance equation for a well-mixed lake may be written as:

$${\text{d}}\frac{{\left( {v\delta L} \right)}}{{{\text{d}}t}} = S_{\text{i}} \delta_{\text{Si}} + G_{\text{i}} \delta_{\text{Gi}} + P\delta_{\text{P}} - E\delta_{\text{E}} - G_{\text{o}} \delta_{\text{L}} - S_{\text{o}} \delta_{\text{L}} = 0$$
(3)

where each water budget elements (term) in (Eq. 2) has been multiplied by its respective isotopic composition expressed in delta notation (δ) to obtain the isotopic mass balance equation in order to separate and quantify groundwater inflow.

By assuming that the Dal Lake is isotopically well mixed (i.e., δS o = δL = δG o) can be simplified to the following equation:

$$G_{\text{i}} = \left\{ {\frac{{P\delta_{\text{P}} - E\delta_{\text{E}} + {\text{Si}}\delta_{\text{Si}} - \left( {P - E - V + S_{\text{i}} } \right)\delta_{\text{L}} - \Delta \left( {V\delta_{\text{L}} } \right)}}{{\left( {\delta_{\text{L}} - \delta_{\text{Gi}} } \right)}}} \right\}$$
(4)

where δ values in ‰ represent the composition of δ18O and δ2H for all components of the lake system; ∆ V is change in lake volume with the isotopic composition of the lake δ L; S i and G i are the inflow of surface water and groundwater, with their isotopic compositions δS i and δG i, respectively; P is precipitation, with its isotopic signature δ P; S o and G o are the outflow of surface water and groundwater with their isotopic compositions δS o and δG o, respectively; and E is evaporation, with the isotopic signature of evaporative flux δ E. In the present study, seasonal weighted values of δP and δE are used. Seasonal δP and δE values are multiplied by seasonal percentage of total precipitation and evaporation, respectively.

The linear resistance model given by Craig and Gordon (1965) was used to determine the isotopic composition of evaporated water from lake. The equation from Horita and Wesolowski (1994) and Clark and Fritz (1997) was used to solve for the isotopic value for evaporation.

$$\delta_{\text{E}} = \delta_{\text{L}} - \varepsilon_{{{\text{l}} - {\text{v}}}} - \Delta \varepsilon_{{{\text{bl}} - {\text{v}}}}$$
(5)

where δ L is the isotopic value for the lake, δ E Isotopic value for evaporated water vapor, ε l–v is fractionation factor for equilibrium, liquid–water vapor exchange, Δε bl–v is the kinetic fractionation factor for evaporation. Following equation (Eq. 6) was used to solve the Δε bl–v for oxygen and hydrogen isotopes (Skrzypek et al. 2015)

$$\Delta \varepsilon ^{18} Obl_{ - v} = - 14.2 \times \left( {1 - h} \right)\permil\,\,\hbox{and}\,\,\Delta \varepsilon ^{2} Hbl_{ - v} = - 12.5 \times \left( {1 - h} \right)\permil$$
(6)

where h is the relative humidity.

Meteorological data

Daily data of air temperature, precipitation, vapor pressure, humidity and pan evaporation were procured from the meteorological station in Shalimar (latitude: 34°8′58″N, longitude: 74°52′54″ and elevation: 1608 m amsl) for the year 2015, which is just about 2 km in east side of the lake. Monthly values for atmospheric precipitation were multiplied by the lake surface area on the monthly precipitation occurred to determine the monthly volume of precipitation input to the lake (Lancaster 2013).

$${\text{PPT}}_{\text{inflow}} \,\left( {{\text{m}}^{3} } \right) = 12\,{\text{ km}}^{2} \times {\text{ppt}}\,\left( {\text{mm}} \right)$$
(7)

Evaporation from the lake surface was estimated from pan evaporation using conversion coefficient (Dunne and Leopold 1978; Qu 1994; Skrzypek et al. 2015):

$$L_{\text{E}} = k \times E_{\text{pan}}$$
(8)

where L E is evaporation from Dal lake, E pan is evaporation from pan (data from Shalimar meteorological station), k is the conversion coefficient (0.59).

Chloride mass balance

Chloride mass balance (CMB) is well-recognized and widely used method to trace recharge sources and to estimate their contributing fractions to a given system. In this study, the groundwater mass balance technique given by Eriksson and Khunakasem (1969) was revisited and applied.

$$Q_{\text{GW}} = \frac{{{\text{PCl}}_{\text{Wap}} }}{{{\text{Cl}}_{\text{GW}} }}$$
(9)

where Q GW = groundwater recharge flux (mm season−1); ClGW = chloride concentration in ground water (mg L−1); P = precipitation (mm season−1); and Clwp = weighted average chloride concentration in precipitation (mg L−1). The weighted average chloride in precipitation was calculated using the following equation:

$${\text{Cl}}_{\text{wp}} = \frac{{\mathop \sum \nolimits_{i = 1}^{n} (P_{i} {\text{Cl}}_{\text{Pi}} )}}{{\mathop \sum \nolimits_{i = 1}^{n} (P_{i} )}}$$
(10)

where ClPi is chloride in precipitation (mg L−1) of the ith sample; P is precipitation (mm) of the ith sample; and n is the number of samples.

Results

Water table (WT) and lake water level (LwL) followed a natural cyclic pattern of seasonal fluctuations. The WT elevation (hydraulic head) in alluvium (on-shore) varied from 1587 to 1590 m amsl, while the LwL varied from 1583 to 1585 m amsl. Higher LwL was observed in summer, followed by declining LwL in autumn and low LwL in winter. Lowest hydraulic head was found at Dalgate, Sadribal, Saidakadal, Rainawari, Khanyar and Nigeen sites and highest at Harwan, Shalimar and Nishat sites. WT varied from 1 to 2 m amsl during the summer season and 0.5–1 m amsl during the winter season.

Water samples from lake, groundwater and streams in the Dal catchment showed distinct chemical variations as summarized in Table 2. pH of groundwater was found to be lower than that of the lake water. It ranged from 7.2 to 7.7 with an average of 7.4 in groundwater and 7.2–8.7 with an average of 8.3 in lake water. Electrical conductivity (EC) of the groundwater was 1–3 times higher than the lake water. It ranged from 177 to 1344 µs cm−1 with an average of 601 µs cm−1 in groundwater and 175–287 µs cm−1 with an average of 218 µs cm−1 in the lake water. The overall trend of concentration of major ions was found to be similar in lake water, groundwater and streams with dominant cation Ca2+ followed by Mg2+ and Na+, and dominant anion HCO3 − followed by SO4 2− and Cl−. The concentration of Ca2+ and Mg2+ varied from 24.4 to 43 and 1.6 to 3.5 mg L−1 in lake water, 9.6 to 120.2 and 6.7 to 81.6 mg L−1 in stream water, 9.6–192.3 and 8.6–86.3 mg L−1 in groundwater, respectively. The concentration of Cl− was lower in streams (2.5–4.9 mg L−1) than the lake water (2.8–17 mg L−1) and groundwater (2.8–38.3 mg L−1). The precipitation weighted average chloride (Clwp) concentration of precipitation samples ranged from 2 to 6 mg L−1.

Table 2 Summary of physio-chemical parameters of Dal Lake in different sides (NE, NW, W, E, SE and SW)
Full size table

The summary of stable isotopic composition of streams, precipitation, groundwater and lake water is presented in Table 3. The isotopic composition of lake water ranged from −8.9 to −4‰ with an average of −6.7 for δ18O and from −52 to −30‰ with an average of −41 for δ2H (Table 3). Groundwater δ18O ranged from −8.6 to −4.7‰, with an average of −7.9‰ and δ2H ranged from −29.9 to −53.7‰ with an average of −45.5‰. The δ18O and δ2H values in streams ranged from −8.9 to −6 with an average of −7.8 and −52.4 to −25.4 with an average of −44.7, respectively. The isotopic composition of precipitation ranged from −11.9 to −1.6 with an average of −6.8 for δ18O and δ2H ranged from −81.7 to 6 with an average of −37.8. The groundwater was found to be highly depleted in heavier isotopes, while the lake water was enriched in heavier isotopes.

Table 3 Summary table of isotopic values in precipitation, streams, lake water and groundwater
Full size table

Discussion

Variation in groundwater and lake water level

Throughout the year, WT was found to be higher than the adjacent LwL across the shoreline, irrespective of the season. When the hydraulic head of the WT is higher than elevation (head) of LwL, the groundwater flows into the lake (higher to lower head). From the hydrological data (Fig. 3), it clearly excluded that the groundwater is fed by the lake water. Water level contour map also corroborates and supports that the lake basin is effluent in nature (Fig. 3). The higher hydraulic gradient of WT was found toward northeast (NE) and northwest (NW) side, which indicates maximum flux of groundwater into the lake. Lowest flux in east (E) and southeast (SE) sides is attributed to the occurrence of volcanic rocks (Panjal Traps). It is pertinent to mention here that the Panjal Traps are massive volcanic rocks with lesser fracture density. The lesser hydraulic gradient in western side of the lake is found due to the clay-rich alluvium with lower hydraulic conductivity and permeability.

Fig. 3
figure 3

Water level contour map with groundwater flow direction

Full size image

The difference between the WT and adjacent LwL of water level is higher (1.5 m) toward NE, NW and E than SE, SW and W also indicated the higher flux of groundwater toward these sides. Seasonally the difference is higher (2 m) in summer and lower (1 m) in winter reflecting the higher groundwater flux into the lake during summer. The observed WT fluctuations have been mainly caused by seasonal variations in weather with differential recharge from rainfall and snow/ice melt.

However, many factors influence LwL, including rainfall, snow/ice melt, evaporation and water consumption. Drawdown of LwL was also influenced by the amount and timing of groundwater and surface water inputs to the streams feeding the lake, which are in turn controlled by different hydrological processes in the catchment.

Hydrogeochemical approach

pH values of groundwater varies in a narrow range, while the pH of lake water varies widely. The wide range in pH of lake water is due to the diel and seasonal changes in photosynthetic rate in aquatic plants (Williams 1998; Saleem et al. 2015). Lowest EC in groundwater was recoded at high altitude springs and tube wells (Harwan, Dara, Brain, and Chesmashahi), and higher was found in Habbak, Batpora, Shalimar, Nehru Park and Nishat areas. From Fig. 4, it is clear that the EC showed a general decreasing trend from all sides of the lake toward the lake interior. From large discrepancies in EC between lake and groundwater and decreasing trend with distance from the shoreline, it can be excluded that the groundwater is fed by lake water. pH and EC values also corroborate and support the results from the LwL–WT contour maps that the Dal Lake is effluent in nature.

Fig. 4
figure 4

EC map of Dal Lake with respect to surrounding

Full size image

Major ions of lake water were compared with groundwater samples, to see the contribution of groundwater to lake using geochemical mass balance (GMB) approach. pH, EC, TDS, Ca2+, Mg2+, Cl−, and HCO3 − were used for GMB respect to different zones (sides) of the lake [i.e., northeast (NE), northwest (NW), east (E), west (W), southeast (SE) and southwest (SW) sides] (Fig. 4). NO3 − and SO4 2− did not exhibit substantial differences, and their sources in the lake are from anthropogenic activities so that they were not used for the GMB method. Groundwater showed higher values of Ca2+ (>185%), HCO3 − (>158%), Mg2+ (>1300%), Cl− (>181%), Na+ (>400%) and K+ (>400%) than lake and stream water. Higher concentrations of chemical constituents in groundwater suggest the carbonate and silicate weathering (Jeelani and Shah 2006; Saleem et al. 2015). These results show that there appears to be good relationship in groundwater with the water quality of the lake. However, the groundwater samples showed no evidence of elevated SO4 2− and NO3 − concentrations than lake water. This suggests that groundwater is not the primary source of SO4 2− and NO3 − to the lake; rather, they may primarily come from anthropogenic sources. From Piper trilinear (Fig. 5) and Stiff diagrams (Fig. 6), two hydrochemical water types were identified in lake water: Ca2+–Mg2+–HCO3 − type and mixed type, whereas only one water type Ca2+–Mg2+–HCO3 − was found in the groundwater and streams. This clearly indicates that the contribution of lake water to groundwater is negligible and there is a strong hydraulic connection between the streams and groundwater.

Fig. 5
figure 5

Piper trilinear diagram (hydrochemical facies) for the data obtained from chemical analysis of lake, streams and groundwater

Full size image
Fig. 6
figure 6

Stiff or radar diagram showing the relative proportions of major ions (Ca2+, Mg2+, Cl−, HCO3 −, Na+ and K+)

Full size image

Based on GMB, the zone (side) wise relative proportion of major ions Fig.7 revealed that groundwater contributes 19–58% to the lake (Figs. 7, 8 and  9). The higher contribution of groundwater was found from NW (58%) and E (56%) sides. Lowest contribution was found from SE (20%) and W (37%) sides of the lake (Fig. 8). EC and WT contour maps also corroborate and support the higher fraction of groundwater from NW and E sides of the lake.

Fig. 7
figure 7

Mean major ion concentrations (lake, groundwater and streams) for different sides of the lake (NE, NW, E,W, SE and SW) (I) pH, (II) EC, (III) TDS, (IV) Ca2+, (V) Mg2+, (VI) HCO3 −, (VII) Cl− and (VIII)

Full size image
Fig. 8
figure 8

Contribution of groundwater inflow/flux from different sides (NE, NW, E, W, SW and SE) of the lake

Full size image
Fig. 9
figure 9

Annual contribution of groundwater in different sides of the lake (NE, NW, E, W, SE and SW) using geochemical approach (Ca2+, Mg2+, Cl−, pH, EC, TDS and HCO3 −)

Full size image

In other words, the overall annual contribution of groundwater in eastern and western side of the lake was 18 and 21%, respectively (Fig. 10). The contribution of streams in eastern and western side of the lake was estimated to be 19 and 17%, respectively. Higher contribution of groundwater in western side is because of lesser streams along that side. Rest of the contribution in eastern (13%) and western side (11%) of the lake was found from other sources including: (1) outflows from sewage treatment plants, (2) small drains, (3) sullage from hotels/houses, and (4) waste water from the houseboats present within the lake basin. Teilbal stream was found to have a distinct effect on the chemical quality of Hazratbal basin on northern side and contributes 100% of water in northern side. However, it does not contribute much to the water quality in other sides (southern) of the lake basin. The contribution of other streams (Shalimar, Nishat, Brain) is very less compared to the main stream.

Fig. 10
figure 10

Annual total contribution of groundwater and streams to different sides (NE, NW, E, W, SW and SE) of the lake

Full size image

Chloride mass balance

The results indicate a range of recharge values to groundwater that varies between 18.5 and 70 mm/season. Maximum recharge (70 mm) was found in spring season, when the total rainfall was recorded as 230 mm (Fig. 11), while minimum recharge (41 mm) was estimated in winter, when the precipitation was higher. The groundwater recharge is influenced by the melting of accumulated winter snow which is higher during winter. In summer, the catchment of the lake the extent of snow packs decreases significantly. The same is reflected in the recharge estimates. The groundwater recharge in summer (35 mm) was higher than in autumn season (18 mm). Based on the mean annual values recharge to the groundwater is probably 165 mm (10% of the annual rainfall), excluding recharge from additional water.

Fig. 11
figure 11

Recharge flux of groundwater from precipitation

Full size image
Fig. 12
figure 12

a Isotopic signatures in precipitation, lake water, streams and groundwater, b in four basins

Full size image
Fig. 13
figure 13

Monthly fluxes of various water balance components in Dal Lake

Full size image

Isotopic approach

δ18O versus δ2H plot (Fig. 12a) of surface water, lake water and groundwater from the Dal catchment clearly shows that the samples from groundwater and streams have similar isotopic values than the lake water. Groundwater and stream water samples were characterized by lighter isotopes, with higher slope (6.8 and 8.5) and intercept (8 and 21) than the local meteoric water line (LMWL). On the other hand, the isotopic values of lake water were relatively enriched in the heavy isotopes, with lower slope (3.9) and intercept (15) than LMWL. This suggested that the isotopic composition of lake water is modified by evaporation, while the isotopic composition of the stream and groundwater has retained their original signature. The degree of evaporation is also reflected in the isotopic composition of different basins (Fig.12b) of the lake: Hazratbal basin (δ18O = −8.2), Boddal basin (δ18O = −6.8), Nigeen basin (δ18O = −6.9) and Gagribal basin (δ18O = −5.3). The lower isotopic values in Hazratbal basin reflects the dominant influence of with the main inflow stream (Telbal nala), whereas the higher isotopic value in Gagribal basin is attributed to the evaporation (the last basin with outflow). Samples collected to the west and south east side of the lake (Boddal and Gagribal basins) plot further away from the GMWL suggesting higher evaporation rate due to the stagnation of water. Different evaporation rates/trends in four basins reflect inputs contribution from different sources with different isotopic values. It also suggests that these basins have different residence time of water, which is then reflected in the isotopic composition (Gonfiantini 1986; Gibson et al. 2002). Increase in the residence time of water in the lake makes the water more enriched in δ18O. The most depleted values were during spring in Nigeen basin, and the most isotopically enriched values were observed during the autumn months in Gagribal basin. During spring season, the lake water is heavily influenced by snow melt, so lake water samples plot very closely to the meteoric water line. When lake water is exposed to sunlight and heat during the summer and autumn, lighter isotopes (δ16O and δ1H) are evaporated preferentially (Gibson and Edwards 2002). With fewer light isotopes, the remaining lake water is enriched in δ18O and δ2H, causing isotopic composition of that water to show more positive δ18O and δ2H values.

The percentage of groundwater component that mixes with lake has been computed using IMB equation (Eq. 4) on seasonal basis. Throughout the study, IMB results showed direct relationship between groundwater and lake water. It was found that groundwater is entering into the lake at all locations and there is no contribution of lake water into the groundwater recharge. It was found that groundwater contributes 23–40% to the lake. Maximum contribution was found in winter season (40%) and lower contribution in autumn season (23.5%). Spring and summer contribute 24.7 and 34.8% respectively (Table 4).

Table 4 Summary of the total monthly inflows, outflows in Dal Lake
Full size table

Water budget of lake/hydrological budget

The monthly estimated discharge measurements (total inflow and total outflow from streams) were combined with the results from the chemical and isotope mass balance approaches. The results were used to determine daily/monthly contribution of groundwater to the water budget of the lake (Table 4, Fig. 13). All the components were taken into consideration including; stream inflows, precipitation, evaporation, groundwater and outflow except runoff. It was observed that total volume of water in Dal Lake varied substantially in time and space (basin wise). The total volume of water stored in the lake varied from 79.11 × 106 m3 to 56.78 × 106 m3, highest was recorded in the summer months and lowest in winter months. Spatially, Hazratbal basin exhibits more volume than Boddal, Gagribal and Nigeen basins. In Hazratbal basin, it ranged from (28.13 × 106 to 39 × 106 m3), whereas in Boddal, Gagribal and Nigeen basins, the total volume of water varied from (15 × 106 to 20.82 × 106 m3), (10.81 × 106 to 15.45 × 106 m3) and (2.7 × 106 to 3.78 × 106 m3), respectively. Maximum lake volume and water was found in summer months, particularly in June and lower in winter months, especially in January. Precipitation over the catchment ranged from 14.8 to 440 mm month−1 with an annual total of 1753 mm. Spring season contributes ~40% of the total annual precipitation. The results revealed that annual contribution of precipitation to the lake is 6.43 × 105 m3 (5.6%). Evaporation was also found to be one of the main components of the water budget, and it ranged from 0 to 0.44 m3 day−1 with an average of 0.24 m3 day−1. Total 88 m3 of water is lost as evaporation from the lake annually. Similarly, the seasonal groundwater contribution to the lake varied from 31.2 × 103 to 674 × 103 m3 day−1 with an average of 276 × 103 m3 day−1. Groundwater contributes maximum during summer and minimum in winter. It was observed that 112 × 103 to 2389 × 103 m3 of water is discharged daily from the lake through two main outflows (Dalgate and Amirkhan), highest being recorded in May and lowest in December.

Uncertainties

Lake–groundwater interactions have their own uncertainties and, until now, they have very rarely been combined (geochemical, isotopic and hydrological approaches) in a single study as we have done in this work. There are uncertainties at the time of measurement and computation due to human, instruments and shortage of data. Water balance estimates of Dal Lake involved different parameters and variables recorded in meteorological station (Shalimar) and sometimes obtained from equations/manually. Human or instrumental errors are found associated with such estimates. Similarly, estimation of precipitation can have a wide range of error, depending on the gage placement in the catchment. Errors in measurement of individual rainfall events directly on lake can be as high as 7%. Errors in estimates of evaporation from lake can also vary widely depending on instrumentation and methodology. If pans are used that are located a distance from the lake of interest, errors can be considerable. Evaporation estimates for this study may also have more uncertainty because of the lack of direct estimates taken from Dal Lake. Isotopic mass balance is based on several factors, and some of these were neglected in this approach, such as runoff component, leading to considerable uncertainties. The cross section of a stream channels where a discharge measurement is made is an important control on the quality of the measurement. Yet, because selection of the measurement sites was good but there can be considerable variation in measured discharge related to the section of the channel chosen for a discharge measurement. Errors in estimates of stream discharge (inflow and outflow) which was measured using area-velocity method are often considered to be within 10 percent. The measurement errors are in 7% range. Sensitivity analysis has been made to all variables considered in the lake water balance. The isotopic mass balance equation is more sensitive to evaporation and precipitation. 10% change in the evaporation can cause up to 14% increase or decrease in groundwater contribution. Similarly by changing rainfall over the lake by 10% from what has been measured a change up to 17% increase or decrease in groundwater contribution. The more uncertainty and sensitivity errors arise either due to random and coarse observations. It can be reduced by additional (sampling count) and systematic observations, ensuring proper instrument configuration (advanced instrumentation) and postprocessing technique.

Conclusion

This study presents a lake–groundwater investigation to quantify or classify the relative role of groundwater inflow to Dal Lake. Stable isotopes coupled with physical flux measurements were crucial for accomplishing the major objectives. The study concludes that LwL is lower than the adjacent WT, suggesting the effluent nature of the lake. The concentration of major ions in lake remained relatively constant throughout the year and was influenced by groundwater inflows, reflecting its quality and the local geology. WT, EC and GMB values suggested that NE and NW sides of the lake have higher contribution from groundwater. Isotopic composition of δ18O and δ2H indicates that water in the different basins of the lake is significantly influenced by evaporation and shows evaporative enrichment. The groundwater generally shows an average stable isotopic signal of the local precipitation indicating active recharge with annual average of 10%. The total volume of water in lake varied from 79.11 × 106 m3 to 56.78 × 106 m3. Both geochemical and isotopic approaches suggest that groundwater (flux) contributes major component (23–40%) to lake water budget. It was found that the average groundwater inflow to Dal Lake was 32 × 103 m3 day−1 in winter months and 508 × 103 m3 day−1 in summer month with an annual average of 276 × 103 m3 day−1. Uncertainty in precipitation (10%) and evaporation (10%) measurements suggests that groundwater inflow rates may increase or decrease 17 and 14%, respectively.