Introduction

Natural uranium is radioactive, having three isotopes 238U (99.27%), 234U (0.72%) and 235U (0.005%) and are present in every sphere of the environment, such as soil, rock, water and biota with low specific activity (25.4 Bq/mg) (IAEA, 2009; UNSCEAR, 2017). Uranium mainly exists in the 4 + and 6 + oxidation states, with the hexavalent state having higher solubility than the tetravalent state. Uranium often combines with oxygen to form a uranyl ion. Uranium has chemical and radiological toxicants, which are a potential health hazard for humans (ATSDR, 1999).

Water is indeed one of the most valuable resources on earth, and its quality and availability are essential for sustaining life on the planet. However, the quality of water, particularly surface and ground water, is degrading due to rapid urbanisation, industrialisation, and modern agricultural practices that use harmful chemicals such as pesticides, herbicides, and fertilisers. Several states in India have reported the presence of high concentrations of toxic metal ions such as uranium, fluoride, nitrate, iron, and arsenic in their water, which have led to a significant deterioration in water quality. This degradation can be attributed to the complex chemical interactions that occur between rock and water systems in various lithological regions (CGWB, 2020). The most common radionuclides found in water are uranium, radium, and radon, which dissolve from specific rocks and soils upon contact with water. Several studies have been conducted to estimate the uranium concentration and ingestion dose in drinking water in India (Sharma et al., 2017; Daulta et al., 2018; Kumar et al., 2018; Sahoo et al., 2021), which indicated that uranium concentrations in water depend on numerous factors of geological and lithological conditions. Ground water plays an imperative role in migrating and redistributing uranium into the environment. Under reducing conditions, uranium gets precipitated from ground water (Babu et al., 2008). Continuous intake of uranium in drinking water, exceeding the acceptable limit, may cause damage to internal organs such as kidneys (Limson et al., 1998) and bones (Arzuaga et al., 2010). Numerous studies concluded no direct link between higher uranium concentration (5.6–3410 µgL−1) (Prat et al., 2009; UNSCEAR, 2017) and undesirable health effects. Many countries like Sweden, India, Iran and Jordan have high uranium concentrations in drinking water and high radiation background area (Godeon et al., 1994; Singh et al., 1995; Kannan et al., 2002; Faghihi et al., 2011); but no significant clinical health effect was observed in these areas.

In India, uranium concentrations in ground water in shallow aquifers ranged from 0.1 to 2876 µgL−1 (CGWB, 2020), indicating that it varies by several orders of magnitude. A recent report showed that many states in India have a higher uranium distribution in ground water like 11.71–113.70 µgL−1 in Bhatinda, Punjab (Singh et al., 1995), 0.3–1442.9 µgL−1 in Kolar, Karnataka (Babu et al., 2008), 0.6–521.15 µgL−1 in Peddagattu, Andhra Pradesh (Raghavendra et al., 2014) and 0.6–260 µgL−1 in western Rajasthan and northern Gujarat (Pandit et al., 2022) that exceeds the acceptable limit of 30 µgL−1 (WHO, 2017), while places like Jaduguda, Jharkhand (Das et al., 2020), Nadia, West Bengal (Sahu et al., 2019), and Nalbari, Assam (Saikia et al., 2021) have observed lower concentration of uranium (< 30 µgL−1) in water sources. The present study evaluated the concentration of uranium and associated water quality parameters in surface water and ground water of the Chittorgarh district of Rajasthan, India.

Materials and methods

Study area

The Chittorgarh district, which lies in the south-eastern part of Rajasthan (Fig. 1), is located between 23°32′ and 25°13′ north latitudes and between 74°12′ and 75°53′ east longitude, which has a 10,856 sq. km area (3.17% of the Rajasthan State). The population of this district is distributed in rural areas (89.39%) and urban areas (15.61%), which uses water for the domestic purpose from surface water (river, dam, lake, stream, and pond) and ground water (well, handpump, and borewell).

Fig. 1
figure 1

Location map of Chittorgarh district of Rajasthan, India

The aquatic liquefier drainage system of the Chittorgarh district is a combination of the Berach, Wagan, Gambhiri, Chambal and Banas plains (CGWB, 2022). Ground water occurrences are controlled by topography, physiography, and the unconfined condition of rock formation. The movement of water in hard rock is governed by size, openness, interconnection, and continuity of weak structural planes. The leaching of the nuclide from soil and sediments to an aquatic liquefier system depends on the geochemistry and lithology of the area (IAEA, 2009). Chittorgarh exposes pre-Vindhyans represented by Hindoli Group intruded by Berach Granite and Vindhyan Supergroup. The Hindoli Group is represented by a mixed assemblage of volcanoclastics with tuffs, metagraywacke, slate, phyllite, quartzite and dolomite. Lot of uraniferous radioactivity is observed in Vindhyans. A regional scale boundary fault known as Great Boundary Fault traverse the contact of Vindhyans and Pre-Vindhyans. The geological set-up of the district is a collective mixer of various igneous and meta-sedimentary rocks (Roy & Purohit, 2015).

Sampling

Surface and ground water samples were collected from drinking water sources during winter and spring season (December 2020 to March 2021). The surface water samples were collected from river, dam, lake and pond. The ground water samples were further categorised into well water, handpump water and borewell water. A total of 87 water samples were collected. Out of 87 water samples, 17.2% represented surface water, and 82.8% expressed ground water. Out of 82.8% of ground water, well water, handpump water, and borewell water contain 8%, 42.5%, and 32.3%, respectively. Ground water samples were collected from different depths varying from 10 to 200 m. The average water table depth from the ground is 20–30 m, 60–90 m and 140–200 m for the well, handpump and borewell, respectively. Water samples were collected as per standard procedures for analysis of uranium and physicochemical parameters (IUGS, 2000; CPCB, 2008). 1.5 L of water samples were collected in cleaned and acidified plastic bottles (ISO, 20062018). Sample details like location name, depth of water source, rock types etc., were recorded.

Uranium concentration estimation

Uranium concentration in collected water samples was analysed using the fluorimetry technique (Sahoo et al., 2021). In the standard addition method, 5.4 mL of the collected water sample was used in a clean and dry cuvette, and 0.54 mL of 5% sodium pyrophosphate (Na4P2O7) (pH-7, adjusted using H3PO4) was added to it followed by 0.06 mL standard uranium source (1000 µgL−1). The fluorescence intensity of the water sample and uranium standard was measured at 510 nm wavelength. The instrument was calibrated using a certified standard uranium solution of1000 ppb uranium standard which was diluted from uranium of Plasma Emission Standard (ICP-MS-66N-0.01X-1 with concentration 100 µg/ml in 2.5% Nitric acid) supplied by traceable to National Institute of Standard and Technology Reference Materials, USA.

Physico-chemical parameters analysis

For the analysis of the physico-chemical parameters of water samples, the standard methods and instruments used are tabulated in Table 1.

Table 1 Detection method for estimation of uranium and physico-chemical parameters in water

Annual ingestion dose

The annual ingestion dose (Ig in Svy−1) for adults (male and female) via consumption of drinking water was estimated using the following equation;

$$\text{I}_\text{g} \left(\text{Svy}^{-1}\right)=\text{U}_\text{a} \left(\text{BqL}^{1}\right)*\text{W}_\text{in}\left(\text{Ly}^{-1}\right)*\text{DCF} \left(\text{SvBq}^{-1}\right)$$
(1)

where Ua is uranium activity in water (BqL−1), the conversion factor of uranium concentration to activity is 1µgL−1 = 25.28 mBqL−1 (ISO, 2015). Win is the annual water consumption (Ly−1) (DRIs, 2005). DCF = Dose conversion factor of ingestion, 4.5* 10–8 SvBq−1 (ICRP, 2012).

Results and discussion

Physico-chemical analysis of surface and ground water sample

The results of physico-chemical analysis of water samples were tabulated in Table 2 and compared with acceptable and permissible limits of drinking water given by the Bureau of Indian Standards (BIS, 2012) in Table 3.

Table 2 Physico-chemical analysis of surface and ground water samples
Table 3 Frequency distribution of uranium and physico-chemical parameters as per drinking water standard [BIS: 10,500(2), 2012; WHO, 2017]

The temperature was recorded at 20.7 ± 4.5 °C in the surface water sample, while it was found 27.2 ± 1.0 °C in the ground water sample. In the sub-categories of the ground water sample, the temperature was noticed 26.3 ± 0.9 °C, 26.9 ± 1.1 °C, and 27.8 ± 1.0 °C in well water, handpump water, and borewell water samples, respectively. The pH was observed at 8.3 ± 0.4 in the surface water sample and 8.0 ± 0.4 in the ground water sample. In the sub-categories of the ground water sample, the pH was found 7.8 ± 0.3, 8.1 ± 0.4, and 8.1 ± 0.4 in well water, handpump water, and borewell water samples, respectively. The surface water and ground water of the area are alkaline in nature (pH > 7). The surface and ground water pH may vary due to the study area's geology. About 66.7% of surface water samples and 87.5% of ground water samples had pH levels in the range of 6.5–8.5, and only 33.3% of the surface water samples and 12.5% of the ground water samples exceeded the maximum permissible limit (> 8.5) of pH in drinking water. In the sub-categories of the ground water sample, 85.7% of the well water, 86.5% of the handpump water, and 89.2% of the borewell water samples had pH level in the range of 6.5–8.5, while 14.3% of the well water, 13.5% of the handpump water, and 10.8% of the borewell water samples exceeded the maximum permissible limit of pH in drinking water (BIS, 2012).

The electrical conductivity (EC) was recorded at 563.5 ± 185.6 µScm−1 in the surface water sample, while it was observed at 1214.9 ± 571.0 µScm−1 in the ground water sample. In the sub-categories of the ground water sample, the EC was noticed at 899.0 ± 500.0 µScm−1, 1269.0 ± 597.0 µScm−1, and 1220.0 ± 547.0 µScm−1 in well water, handpump water, and borewell water samples, respectively. The EC values were found higher, indicating a high amount of dissolved inorganic substances (solutes) in an ionised form. The total dissolved solids (TDS) concentration was noticed at 312.3 ± 117.3 mgL−1 in the surface water sample, while it was recorded at 970.5 ± 300.0 mgL−1 in the ground water sample (Fig. 2a). In the sub-categories of the ground water sample, the TDS concentration was observed at 485.0 ± 283.0 mgL−1, 698.0 ± 329.0 mgL−1, and 664.0 ± 285.0 mgL−1 in well water, handpump water, and borewell water samples, respectively (Fig. 2b). The elevated levels of EC and TDS in the groundwater samples can be attributed to the lithology of the study area, predominantly consisting of limestone and dolomite rocks. The presence of these rocks enhances the TDS and EC levels in ground water through the dissolution of minerals and ions. This process increases the concentration of dissolved solids, thereby elevating the TDS. Additionally, the dissolved ions contribute to the water's electrical conductivity, resulting in higher EC values. It is observed that the TDS concentration of 86.7% of surface water samples and 40.3% ground water samples were less than the acceptable limit (500 mgL−1) of drinking water. In comparison, 13.3% of surface water samples and 58.3% of ground water samples exceeded the acceptable limit but less than the permissible limit (2000 mgL−1), and only 1.4% of ground water samples exceeded the permissible limit of drinking water. In the sub-categories of the ground water sample, TDS values in 57.1% of the well water, 35.1% of the handpump water, and 42.9% of the borewell water samples were less than the acceptable limit of drinking water, while 42.9% of the well water, 62.2% of the handpump water, and 57.1% of the borewell water samples were exceeded the acceptable limit but less than the permissible limit of drinking water. Only 2.7% of handpump water samples exceeded the permissible limit of drinking water (BIS, 2012).

Fig. 2
figure 2

The mean concentration of TDS, alkalinity, chloride, hardness, and nitrate in surface and ground water and sub-categories of ground water

The oxidation–reduction potential (ORP) was detected at 261.2 ± 12.8 mV in the surface water sample and 239.0 ± 28.0 mV in the ground water sample. In the sub-categories of the ground water sample, the ORP was found at 255.0 ± 14.1 mV, 233.0 ± 31.2 mV, and 243.0 ± 23.7 mV in well water, handpump water, and borewell water samples, respectively. The dissolved oxygen (DO) was found at 10.8 ± 0.1 mgL−1 in the surface water sample, while it was observed at 10.7 ± 0.3 mgL−1 in the ground water sample. In the sub-categories of the ground water sample, the DO was observed at 10.7 ± 0.1 mgL−1, 10.7 ± 0.2 mgL−1, and 10.6 ± 0.3 mgL−1 in well water, handpump water, and borewell water samples, respectively. The DO concentration of ground water was observed almost similar to surface water which may be due to movement of groundnwater through the soil and rocks, where it may be come into contact with oxygen-rich areas. Oxygen from these areas can dissolve into the ground water through diffusion, leading to similar dissolved oxygen levels as in surface water.

The alkalinity concentration was detected at 478.4 ± 154.2 mgL−1 in the surface water sample. At the same time, it was found at 523.8 ± 189.8 mgL−1 in the ground water sample (Fig. 2a). In the sub-categories of the ground water sample, the alkalinity concentration was recorded at 368.0 ± 104.1 mgL−1, 527.1 ± 201.4 mgL−1, and 557.0 ± 188.1 mgL−1 in well water, handpump water, and borewell water samples, respectively (Fig. 2b). The prevalence of limestone and dolomite rocks in the Chittorgarh region can contribute to higher alkalinity levels in ground water due to the release of carbonate and bicarbonate ions during mineral dissolution. It is observed that the alkalinity concentration of 13.3% surface water samples and 12.5% of ground water samples were less than the acceptable limit (200 mgL−1) of drinking water. In comparison, 80.0% of surface water samples and 66.6% of ground water samples exceeded the acceptable limits but less than the permissible limit (600 mgL−1) of drinking water. Only 6.7% of surface water and 20.9% of ground water samples exceeded the permissible limit of drinking water. In the sub-categories of the ground water sample, alkalinity concentration in 28.6% of the well water, 8.1% of the handpump water, and 7.1% of the borewell water samples were within the acceptable limit of drinking water, while 71.4% of the well water, 54.1% of the handpump water, and 57.1% of the borewell water samples were exceeded the acceptable limit but less than the permissible limit of drinking water. Only 37.8% of handpump water and 35.8% of borewell water samples exceeded the permissible limit of drinking water (BIS, 2012).

The chloride concentration was recorded at 75.5 ± 28.4 mgL−1 in the surface water sample. In contrast, it was detected at 125.0 ± 96.4 mgL−1 in the ground water sample (Fig. 2a). In the sub-categories of the ground water sample, the chloride concentration was noticed at 61.8 ± 43.0 mgL−1, 120.0 ± 87.8 mgL−1, and 147.0 ± 116.0 mgL−1 in well water, handpump water, and borewell water sample, respectively (Fig. 2b). The elevated chloride levels in the ground water of study area may be attributed to the specific geological composition of the region. The presence of geological formations that contain minerals or deposits rich in chloride may result in the dissolution of chloride ions into the ground water, leading to higher chloride concentrations. It is observed that the chloride concentration of 100% of surface water samples and 87.5% of ground water samples were less than the acceptable limit (250 ppm) of drinking water. Only 12.5% of ground water samples exceeded the acceptable limit but less than the permissible limit (1000 mgL−1) of drinking water. In the sub-categories of the ground water sample, chloride concentration in 100% of well water, 91.9% of the handpump water and 75.0% of the borewell water samples were within the acceptable limit, while 8.1% of the handpump water, and 25.0% of borewell water samples exceeded the acceptable limits but less than the permissible limit of drinking water (BIS, 2012).

The hardness concentration was recorded at 268.0 ± 77.5 mgL−1 in the surface water sample, while it was noticed at 447.3 ± 161.3 mgL−1 in the ground water sample (Fig. 2a). In the sub-categories of the ground water sample, the hardness concentration was found 448.0 ± 175.1 mgL−1, 440.0 ± 158.0 mgL−1 and 456.0 ± 161.2 mgL−1 in well water, handpump water, and borewell water samples, respectively (Fig. 2b). The elevated hardness levels in ground water, as opposed to surface water, may be attributed to the specific geological composition of the study area. The geological formations in the Chittorgarh region consist primarily of limestone and dolomite rocks, which contain minerals contributing to water hardness. Ground water's prolonged residence time beneath the surface allows for an extended interaction with these mineral-rich rocks, leading to enhanced dissolution of minerals and subsequently higher levels of hardness in the ground water. It was observed that the hardness concentration of 29.3% of surface water samples and 8.3% of ground water samples were less than the acceptable limit (200 mgL−1) of drinking water. In comparison, 70.7% of surface water samples and 72.2% of ground water samples exceeded the acceptable limits but less than the permissible limit (600 mgL−1) of drinking water. Only 19.5% of ground water samples exceeded the permissible limit of drinking water. In the sub-categories of the ground water sample, the hardness concentration in 14.3% of the well water, 5.4% of the handpump water, and 7.1% of the borewell water samples were less than the acceptable limit, while 71.4% of the well water, 73.0% of the handpump water, and 71.4% of the borewell water samples were more than acceptable limit but less than the permissible limit of drinking water. In comparison, 14.3% of the well water, 21.6% of the handpump water, and 21.5% of borewell water samples exceeded the acceptable limits of drinking water (BIS, 2012).

The nitrate concentration was detected at 49.4 ± 30.2 mgL−1 in the surface water sample. At the same time, it was observed at 86.5 ± 35.8 mgL−1 in the ground water sample (Fig. 2a). In the sub-categories of the ground water sample, the nitrate concentration was found at 63.0 ± 27.0 mgL−1, 76.5 ± 33.0 mgL−1, and 106.0 ± 31.6 mgL−1 in well water, handpump water and borewell water samples, respectively (Fig. 2b). It is observed that the nitrate concentration of 58.3% of surface water samples and 20.8% ground water samples were less than the acceptable limit (45 ppm) of drinking water. In comparison, 41.7% of surface water and 79.2% of ground water samples exceeded the permissible limit of drinking water. In the sub-categories of ground water, nitrate concentration in 28.6% of the well water, 27.0% of the handpump water and 14.7% of the borewell water samples were within the acceptable limit of drinking water, while 71.4% of the well water, 73.0% of the handpump water and 85.3% of the borewell water samples were exceeded the acceptable limits of drinking water.

The fluoride concentration was found at 0.5 ± 0.3 mgL−1 in the surface water sample and at 0.7 ± 0.4 mgL−1 in the ground water sample. In the sub-categories of the ground water sample, the fluoride concentration was noticed at 0.3 ± 0.2 mgL−1, 0.6 ± 0.3 mgL−1 and 0.9 ± 0.4 mgL−1 in well water, handpump water, and borewell water samples, respectively. The higher fluoride levels in ground water compared to surface water in the study area, can be attributed to the presence of fluoride-rich geological formations and minerals. Fluoride-containing minerals like fluorite, apatite, amphiboles, and mica in the aquifers can release fluoride into the ground water. It was observed that the fluoride concentration of 93.3% of surface water samples and 81.9% of ground water samples were less than the acceptable limit (1 mgL−1) of drinking water. In comparison, 6.7% of surface water samples and 11.1% of ground water samples exceeded the acceptable limit but less than the permissible limit (1.5 mgL−1) of drinking water. Only 7.0% of ground water samples exceeded the permissible limit of drinking water. In the sub-categories of ground water, fluoride concentration in 100% of well water, 83.8% of the handpump water, and 78.6% of the borewell water samples were within the acceptable limit of drinking water. In contrast, 10.8% of the handpump water, and 10.7% of borewell water exceeded the acceptable limit but less than the permissible limit of drinking water. Only 10.7% of borewell water samples exceeded the permissible limit of drinking water (BIS, 2012).

Analysis of uranium concentration in the surface and ground water samples

The uranium concentration was observed at 2.5 ± 1.9 µgL−1 in the surface water sample, while it was observed at 16.5 ± 1.4 µgL−1 in the ground water sample (Table 4 and Fig. 3a). In sub-categories of the ground water sample, the highest uranium concentration was found in a borewell water sample (23.3 ± 17.0 µgL−1), followed by handpump water sample (13.5 ± 9.1 µgL−1) and well water sample (6.0 ± 5.5 µgL−1) (Table 4 and Fig. 3b). The frequency distribution of uranium concentration in water samples is given in Table 3. 100% of surface water samples and 88.9% of ground water samples carried uranium concentration within the acceptable limit (30 µgL−1) in drinking water and only 11.1% of ground water samples exceeded the acceptable limit of drinking water (WHO, 2017). In the sub-categories of the ground water sample, uranium concentration in 100% of well water, 94.6% of the handpump water, and 78.6% of the borewell water samples were within the acceptable limit of drinking water, while 5.4% of the handpump water, and 21.4% of borewell water samples exceeded the acceptable limit of drinking water,

Table 4 Uranium concentration in surface water, ground water, and sub-categories of ground water
Fig. 3
figure 3

Mean uranium concentration in surface water, ground water, and sub-categories of ground water

The skewness coefficient value was found as 1.0 for the surface water sample and 3.1 for the ground water sample. In the sub-categories of the ground water sample, the skewness coefficients value was noticed as 1.6, 4.5, and 2.1 for well water, handpump water and borewell water samples, respectively (Table 4). The kurtosis coefficient value was obtained at -0.2 for the surface water sample, while it was noticed at 10.7 for the ground water sample. In the sub-categories of the ground water sample, the kurtosis coefficient value was found 2.2, 24.5, and 4.3 for well water, handpump water and borewell water samples, respectively (Table 4). The 25th and 75th percentile was obtained at 0.8 and 3.3, respectively, for surface water sample and 4.4. and 18.5, respectively, for the ground water sample. In the sub-categories of the ground water sample, the 25th and 75th percentile was noticed at 1.2 and 7.9, respectively, for the well water sample, 3.9 and 15.5, respectively, for handpump water sample and 9.2 and 26.0, respectively, for borewell water sample (Table 4).

A significant correlation (R2 = 0.6369) was observed between uranium concentration and depth of ground water (Fig. 4). A significant correlation (R2 = 0.6211) was observed in the surface water sample between uranium concentration and nitrate, while in the ground water sample, it was poorly correlated (R2 = 0.3605) (Fig. 5a and b). The correlation between uranium concentration and electrical conductivity was noticed significantly in the surface water sample (R2 = 0.681) and the ground water sample (R2 = 0.5636) (Fig. 6a and b). The same trend was also observed between uranium concentration and TDS. It was significantly correlated in the surface water sample (R2 = 0.652) and ground water sample (R2 = 0.5495) (Fig. 7a and b). Other physicochemical parameters like temperature, pH, ORP, alkalinity, chloride, and hardness were very poorly correlated with the concentration of uranium in surface and ground water samples.

Fig. 4
figure 4

Correlation between uranium concentration and depth (m) of ground water

Fig. 5
figure 5

Correlation between uranium concentration and nitrate concentration

Fig. 6
figure 6

Correlation between uranium concentration and electrical conductivity

Fig. 7
figure 7

Correlation between uranium concentration and TDS concentration

The concentration of uranium in ground water samples was found significantly higher compared to surface water samples. In addition to the higher concentration of uranium in ground water samples compared to surface water, the study found a significant correlation between the depth of the ground water source and uranium concentration, particularly within sub-categories of ground water. The water table depth followed the trend of Borewell water (140–200 m) > Handpump water (60–90 m) > Well water (20–30 m), and the same order was observed for the concentration of uranium [Borewell water sample (23.3 ± 17.0 µgL−1) > Handpump water sample (13.5 ± 9.1 µgL−1) > Well water sample (6.0 ± 5.5 µgL−1)]. This difference in uranium concentration can be attributed to variations in the rock types and geological conditions present beneath the earth's surface. Several factors were found to impact the uranium concentration in water, including the uranium content in the host aquifer rock and its chemical constituents, the presence of oxygen and its complexing agents, chemical reactions with ions in solution, and the nature of the contact between uraniferrous minerals and water (Hess et al., 1985). The study area Chittorgarh, Rajasthan displays a diverse geology and lithology, with mineral pockets including dolomite, alluvium, sandstone, shale, and limestone. These formations belong to the Hindoli Group, which is intruded by the Berach Granite and part of the Vindhyan Supergroup. The varied lithologies suggest a complex geological history involving volcanic activity, sedimentation, metamorphism, and intrusive igneous events. The diverse geology of Chittorgarh creates favorable conditions for the formation of minerals like dolomite and uranium-bearing minerals, leading to an increased concentration of uranium in ground water.

The physico-chemical parameters of water also affect the concentration of uranium in water (Selvi et al., 2016). A significant correlation was observed between nitrate and uranium concentration in surface water. A literature study indicated that nitrate might play an essential role in the dissolution of uranium as uranyl nitrate [UO2 (NO3)2] (Riedel & Kubeck, 2017). Hence it can be concluded that uses of nitrogenous fertilizers like urea, ammonium sulphate, ammonium nitrate etc., in agricultural land might be the carrier of uranium in surface water (Nolan & Weber, 2015; Saikia et al., 2021).

The electrical conductivity and TDS were also correlated significantly with uranium concentration in the surface and ground water samples. Uranium concentration increased with increased electrical conductivity and TDS of the water sample. The possible reason behind the increase in uranium concentration in ground water could be the excessive use of fertilizers and pesticides in intensive agricultural practices. These chemicals tend to leach down to ground water and raise the total dissolved solids (TDS) concentration, forming more uranium salts. Duggal et al. (2016) also found similar observations. Similar studies have been conducted on the on arid region of Jalore and Sirohi district Rajasthan by Pandit et al. (2022) and significant correlation among TDS, electrical conductivity and uranium concentration was observed. It was observed due to the association of high uranium content with Na-Cl water type. Parimalarenganayaki et al. (2019) suggested that the ground water samples with higher TDS values may contain various ionic species that can interact with uranium, causing its dissolution and ultimately increasing radioactivity.

The range of skewness and kurtosis coefficients between -2 to 2 are accepted for normal distribution (Kim, 2013; Asakereh et al., 2020). The proximity also offers a significant skewness and kurtosis (Livingston, 2004). Hence, in our study, surface water samples had a normal distribution of uranium concentration, while ground water samples had heterogeneous distribution in ground water samples. Due to the heterogeneous distribution of the uranium in ground water, the mean and standard deviation was found close together in the sub-categories of ground water, i.e. well water (6.0 ± 5.5), handpump water (13.5 ± 9.1), and borewell water (23.3 ± 17.0). The highly heterogeneous distribution of uranium content in sub-categories of ground water, i.e. well water, handpump water, and borewell water, signifies the influence of the underlying regional lithography, geology, and geochemistry at various depths on uranium distribution.

Uranium concentration is highly variable in the ground water of India. In India, the concentration of uranium was reported at 0.8–71.9 µgL−1 in Central Tamilnadu (Adithya et al., 2019); 0.3–1442.9 µgL−1 in Kolar, Karnataka (Babu et al., 2008); 0.2–68.0 µgL−1 in Nalgonda, Andhra Pradesh (Brindha et al., 2011); 0.2–20.8 µgL−1 in Jammu and Kashmir (Kumar et al., 2016); 0.20–8.9 µgL−1 in Banswara Rajasthan (Menaria et al., 2021); 1.2–45.4 µgL−1 in Punjab and 0.6–10.1 µgL−1 in Himachal Pradesh (Rani et al., 2013); 0.3–10.3 µgL−1 in Nalbari District, Assam (Saikia et al., 2021) and 13.9–172.8 µgL−1 in Malwa region, Punjab (Tripathi et al., 2013) (Table 5).

Table 5 Comparison of uranium concentration of present study with other studies

Annual ingestion dose

Ingestion dose was calculated using intake rate (3.7 Lday−1 for males and 2.7 Lday−1 for females) (ATSDR, 1999; DRIs, 2005). The annual ingestion dose was found at 3.8 µSvy−1 (for males) and 2.8 µSvy−1 (for females) in surface water and 25.4 µSvy−1 (for males) and 18.5 µSvy−1 (for females) in ground water (Fig. 8a). In the sub-categories of the ground water sample, the ingestion dose was observed at 9.2 µSvy−1 (for males) and 6.7 µSvy−1 (for females) in well water, 20.7 µSvy−1 (for males) and 15.1 µSvy−1 (for females) in handpump water and 35.8 µSvy−1 (for males) and 26.1 µSvy−1 (for females) in borewell water samples (Fig. 8b). The results show that people using ground water as a drinking water source are getting higher doses of uranium than those drinking surface water as potable water in the area. However, the annual ingestion dose of surface and ground water and sub-categories of ground water like well water, handpump water, and borewell water was found very low than the acceptable limit (100 µSvy−1) of drinking water given by WHO (2017).

Fig. 8
figure 8

Mean annual ingestion dose in surface water, ground water, and sub-categories of ground water

Conclusions

The investigation conducted in Chittorgarh, Rajasthan revealed that the uranium concentration in ground water samples was significantly higher compared to surface water samples. The mean uranium concentration in ground water was 16.5 ± 1.4 µgL−1, with the highest concentrations observed in borewell water (23.3 ± 17.0 µgL−1), followed by handpump water (13.5 ± 9.1 µgL−1), and well water (6.0 ± 5.5 µgL−1). In contrast, the mean uranium concentration in surface water was much lower at 2.5 ± 1.9 µgL−1. The study also observed a correlation between the depth of the ground water table and uranium concentration, indicating higher levels in deeper water sources (Well water ˂ Handpump water ˂ Borewell water). This correlation suggests that the depth of the water source plays a role in the uranium levels found in the ground water. The distribution of uranium content and water quality parameters exhibited significant heterogeneity, reflecting the influence of regional lithography, geology, and geochemistry on uranium concentration. Additionally, uranium concentration was found to be significantly correlated with electrical conductivity, total dissolved solids, and nitrate concentration. Importantly, despite the higher uranium concentrations in ground water, a large proportion of the samples (88.9%) were within the acceptable limit of 30 µgL−1 set by the World Health Organization (WHO). All surface water samples (100%) were within this limit, indicating that both surface water and ground water sources in the region generally adhere to the recommended uranium concentration levels.

The annual ingestion dose resulting from consuming ground water was higher compared to surface water. The annual ingestion dose was found to be 25.4 µSvy−1 for males and 18.5 µSvy−1 for females in ground water, while it was 3.8 µSvy−1 for males and 2.8 µSvy−1 for females in surface water. Within the sub-categories of ground water, the annual ingestion dose followed the trend: 35.8 µSvy−1 for males (borewell water) > 20.7 µSvy−1 for males (handpump water) > 9.2 µSvy−1 for males (well water), and 26.1 µSvy−1 for females (borewell water) > 15.1 µSvy−1 for females (handpump water) > 6.7 µSvy−1 for females (well water). It is crucial to note that the annual ingestion doses of both surface water and ground water were found to be much lower than the acceptable limit of 100 µSvy−1 set by the WHO in 2017. This indicates that the usage of both surface water and ground water as drinking water sources in Chittorgarh does not pose any unacceptable health effects on the population.

In conclusion, the study confirmed that while uranium concentrations in ground water exceeded those in surface water, the overall uranium levels observed were within acceptable limits. Therefore, the presence of higher uranium concentrations in ground water samples of Chittorgarh, Rajasthan does not pose significant health risks to the population.