Introduction

Radon, the only natural radioactive gas, which is produced from the α-decay of radium-226, is part of the uranium-238 series (Cothern 2014). The half-life of radon is 3.82 days, and this is long enough for diffusive transport through the fissures in rocks and dissolution into ground water due to its appreciable solubility in water (0.01 mol kg−1 bar−1 at 293 K) (Rangaswamy et al. 2016; Lima-Flores et al. 2021). High radon content in groundwater is subject to the presence of shear zones and other geological features, and the presence of granite bedrocks—containing uranium and radium—through which the water passes (Singh et al. 2008; Kareem et al. 2020; Abdallah et al. 2007). In addition, abundance of radon in groundwater depends on the porosity, permeability and the moisture content in the sub-surface (Jaishi et al. 2014; Günay et al. 2019). The dissolved radon can escape into the atmosphere by diffusion (Kessongo et al. 2020). Average transfer coefficient from water to air for radon is 10−4 (National Research Council 1999). This indicates that the concentration of most of the indoor radon gas comes from the household usage of radon-rich water. Rn-222 has two progeny, namely, Po-214 having half-life of 160 µs and Po-218 having half-life of 3 min. They are responsible for the majority of the radioactivity resulting from radon gas.

Approximately half of total radiation dose received by the population globally comes from the inhalation or ingestion of radon (UNSCEAR 2000; Naskar et al. 2018). The alpha particles emitted from radon can enter into the lungs of human being by inhalation of radon-containing air, which may lead to lung cancer (Kessongo et al. 2020; Agoubi 2021).When groundwater with high radon concentration is used for drinking, there are additional risks of stomach and gastrointestinal cancer (Thabayneh 2015; Naskar et al. 2018; Khan 2021). According to a report published by the United States Environmental Protection Agency (USEPA), Rn-222 is the second major cause of lung cancer deaths every year, after smoking (EPA U 1991; EPA 1999; Sharma et al. 2017). Nayak et al. (2022) published a review work on groundwater radon distribution and associated health issues and its probable mitigation strategy. Çam-Kaynar and Parlak (2022) published their work on indoor radon measurement and the annual effective dose rates for different season dwellings in Manisa, Turkey. Therefore, assessment of radon in groundwater is absolutely essential for safe consumption and daily household usage, in addition to the usual estimation of Water Quality Index (WQI) for suitability for drinking purpose (Matta et al. 2020; Mohamed Amine et al. 2021).

Apart from all these conventional characteristics, due to high abundance of Rn-222 in groundwater, it is easy to evaluate the quantitative and qualitative groundwater study, knowing its path through the soil and rock formations. Due to the dissolution of several minerals, compounds and radioactive substances, Rn-222 activity in groundwater is much high on magnitude compared to the sea water (Nalukudiparambil et al. 2021; Singh et al. 2021). Therefore, the presence of Rn-222 in groundwater or soil can be used in the estimation of groundwater flow velocities, groundwater discharge rate, aquifer water residence time, detection of faults and fractures in the earth crust and as a tracer for earthquakes (Bertin and Bourg 1994; Hoehn and Von Gunten 1989; Naskar et al. 2022).

From research that has already been carried out all over the world, it is established that the concentration of radon in soil and groundwater undergoes wide variations due to various meteorological factors such as climatic conditions, soil permeability, soil porosity and soil temperature and due to geological features of a region such as active faults, fractures in the earth crust and types of rock present in that area (Ramola et al. 1998; Naskar et al. 2018; Shilpa et al. 2017; Arvela 1995; Kullab et al. 2001; Al-Khateeb et al. 2017; Rani et al. 2021). As the meteorological parameters change with seasons, it is necessary to conduct repeated measurements in all season to determine the seasonal variations of radon (Miklyaev and Petrova 2021; Rengan et al. 2022). Recent results of radon measurement across the world are enlisted in Table 1.

Table 1 Study of water radon activity around the world (AM arithmetic mean, Min minimum value, Max maximum value)
Full size table

In a previous research work by the present group, four groundwater samples in the Purulia municipal area of Purulia district, West Bengal, were reported to have significantly high radon concentration (Mitra et al. 2021). Now, in order to study the possible reason behind its occurrence and associated health effects due to the consumption of this radon-rich water, more samples were analysed from this municipal area. Additionally, dissolved Na+, K+, Ca2+ and Li+ ions and the pH levels of these groundwater samples have been analysed with an aim to study their possible correlations with radon present in the samples.

Geological description of the study area

The present work focuses on the main urban centre of Purulia district, i.e. Purulia Municipality comprising 23 wards covering a total area around 14 square km (from 22°42′35″ N to 23°42″ N and from 85°49′ 25″ E to 86°57′ 37″E). According to the 2011 census, the municipality has a total population of 121,436 with a population density of about 6760 people per square km (District Census Handbook 2011; Chamling 2013). However, the population density as well as the population has increased in the last 10 years. A large portion of this population use tube-well water for domestic purposes. Geographically, the area is dominated by subtropical climate of high evaporation and low precipitation with a wide variation of temperature, minimum being 3.8 °C in winter while maximum being 52 °C in summer (Chamling 2013).

From the geological point of view, the study area lies in the Chotanagpur Granite Gneissic Complex (CGGC) of the eastern peninsular shield of India, which is well identified for mineral deposition (Sanyal and Sengupta 2012; Dolui et al. 2016; Mitra et al. 2021). The CGGC contains various stages of migmatites, granitoids, gneisses, amphibolites and granulites in different regions (Acharya and Nag 2013). Among them, exposed granites, Archaean-age gneisses and meta-sediments of pre-Cambrian age compose the bedrocks of the region (Sanyal and Sengupta 2012; Dolui et al. 2016). The composite rocks and soils and presence of a shear zone, namely, the South Purulia Shear Zone nearby, may raise the groundwater radon activity in this area (Otton 1992).

Sampling

A total of 120 groundwater samples were collected from deep tube-wells of depth ranging from about 20 m to about 60 m, from the 23 wards of Purulia municipality in such a way that there were at least two samples from each ward. Large population density and high daily usage of tube-well water for domestic purposes were considered for the selection of tube-wells in the municipal area. In order to ensure that the groundwater came from the desired depth, the tube-wells were first pumped for at least 5 − 10 min (Naskar et al. 2018; Mitra et al. 2021). Then, the mouths of the tube-wells were sealed using a football bladder and then pumped until the water rose up to the top of the tube-well, and it was kept for at least 30 min. Next, using 200-ml syringe water samples were collected from it and transferred into a 200-ml air-tight glass bottle to minimise radon loss during transportation. The glass bottles were then properly sealed and labelled. Additionally, water samples were taken into another 250 ml plastic bottle for measurement of dissolved Na+, K+, Ca2+, Li+ and properly sealed and labelled. Then, all the samples were brought into the laboratory for measurement. Water radon content of the samples were measured on the same day as the collection (delay time < 4 h). A global positioning system (GPS) meter was used to determine the sampling location.

Experimental methods

The temperatures and pHs of the water samples were measured in situ using a thermometer and a digital pH meter (HM_PH80) respectively. For the assessment of radon content in ground water samples, an online radon monitoring system composed of the ionisation chamber AlphaGuard DF2000 PRO and its accessory AquaKIT was used. This device is capable of measuring radon concentration in water from 0.01 Bq/l to 2 × 106 Bq/l with a maximum instrumental error of ± 0.3% (AlphaGuard Manual 2020; AquaKIT Manual 2017).

The following equation prescribed by the manufacturer (Naskar et al. 2018; Chowdhury et al. 2019) has been used to calculate radon concentration in water samples:

$${C}_{water}=\frac{{C}_{air}\left(\frac{{V}_{system}-{V}_{sample}}{{V}_{sample}}+k\right)-{C}_{0}}{1000}$$
(1)

where Cwater is the radon concentration in water sample (Bq/l), Cair is radon concentration (Bq/m3) in the measuring set-up, C0 is the radon concentration in the measuring set-up before sample injection (Bq/m3), Vsystem is interior volume of the measurement setup (ml), Vsample is volume of the water sample (ml) and k is the radon diffusion coefficient. Diffusion coefficient value can be determined by using the following equation:

$$k=0.106+0.405 {e}^{(-0.05T)}$$
(2)

where T is the temperature of the water (°C)

The measurements were performed at room temperature so that the value of the diffusion coefficient was same as that provided by the manufacturer, and no more corrections were necessary.

The concentrations of dissolved Na+, K+, Ca2+ and Li+ ions in the water samples were measured using a micro controller-based Flame Photometer (G-301 Series, HPG Instruments, India) following the standard calibration procedure prescribed by the manufacturer. Firstly, the instrument was calibrated with two known activity solutions of the cations (Na+, K+, Ca2+, Li+). Calibration was done by using 10 mg/l and 50 mg/l standard solutions of each cation. Then, the instrument was set for the measurement mode for particular cations (Na+, K+, Ca2+, Li+) and 50 ml water sample was taken in a glass beaker and aspirated through a compressor to the flame. The flame colour changes according to the cation present in the water sample, and the attached micro-controller detector detects the activity in comparison to the standard solution, and the result is displayed. After aspirating with distilled water, the flame resets. This process was repeated four times, and the average of the data was considered.

Estimation of annual effective dose by inhalation and ingestion

The Rn-222 content in drinking water is comparatively more harmful than most other elements or ions (Cothern 2014). Radon can enter the human body in two ways: firstly, from consuming radon-containing water, i.e. ingestion, and secondly, from the release of dissolved radon from water and its inhalation and cause serious health issues (Binesh et al. 2010; Duggal et al. 2020). Therefore, to assess the risks regarding health issues in humans, it is necessary to evaluate the annual effective doses due to the inhalation and ingestion of radon. Doses associated with radon for adults, children and infants are calculated with the help of following equations.

The annual effective dose due to inhalation has been estimated by the following:

$${D}_{Inh}\left(mSv/y\right)= {C}_{RnW}\times {R}_{aW}\times I\times DCF\times F\times {10}^{-3}$$
(3)

where RaW is the coefficient of transfer of radon from water to air (10−4), CRnW is the radon concentration in water (Bq/l), I is the average annual indoor occupancy time per individual (7000 h/y), F is the equilibrium factor between radon and its progeny (0.4) and DCF is the dose conversion factor for radon exposure (9 nSv/Bq/h/m3) (UNSCEAR 2000; UNSCER 2008; Naskar et al. 2018; Duggal et al. 2020).

The annual effective dose due to ingestion has been estimated using the following equation:

$${D}_{Ing}\left(mSv/y\right)={C}_{RnW}\times {C}_{W}\times EDC\times {10}^{-6}$$
(4)

where CRnW is the radon concentration in water (Bq/l), CW is the average annual water intake (l/y) and EDC is the effective dose coefficient for ingestion (3.5 nSv/Bq) (UNSCEAR 2000; UNSCER 2008; Mittal et al. 2016).

Result and discussion

The pH and the concentrations of radon and some major cations in 120 groundwater samples have been estimated in Purulia municipality of Purulia district, West Bengal, India. The measured values of pH of the water samples were found to range from 6.4 to 9.0 with an average of 7.70, indicating that the samples were mildly acidic to alkaline in nature. It has been observed that these ground water samples contain widely varying concentration of radon ranging from 10.44 to 403.56 Bq/l with an average of 108.70 Bq/l. The sampling locations along with the distribution of radon concentrations have been depicted in the bubble plot of Fig. 1. The concentration profiles of radon as well as cations (Na+, K+, Ca2+, Li+) and pH have been portrayed in the bar plot of Fig. 2. The wide variety of radon values, as reflected in these numbers and bar plots can be inferred to be a result of distribution of rocks containing different kinds of minerals, nature of the bedrocks, the presence of geological faults and fractures in the study area, as well as the depths of the tube-wells. From Fig. 1, it can be observed that the apparently high values of radon only appear in the southern part of Purulia Municipality where the tube-wells are deep. This may also show some indication of an unknown geological fault line which can provide a pathway for radon from deep inside the earth’s crust to travel to the upper layers and dissolve into the aquifer waters (Chowdhury et al. 2019; Mitra et al. 2021). The sample density in this region with high radon activity has been increased to understand the proper distribution of radon as well as certain cations (Na+, K+, Ca2+, Li+) and pH. As discussed earlier, the bedrocks of this region are mainly composed of granitic rocks, which are capable of trapping radon in their pores. This trapped radon may also dissolve into groundwater and may be another reason for high radon occurrence. On the other hand, the low concentrations of groundwater radon in certain locations may be due to water coming from less depth.

Fig. 1
figure 1

(Source: Geological Survey of India and modified after Sanyal and Sengupta 2012)

Bubble plot of the measured radon concentration in water samples with locations of the study area

Full size image
Fig. 2
figure 2

Concentration profile of measured radon as well as cations (Na+, K+, Ca2+, Li+) and pH of the water samples

Full size image

In order to compare the water radon activity data obtained in this study with earlier studies carried out in different parts of the world, radon activity data from some of those studied are presented in Table 1. These data give an idea about how groundwater radon concentration changes from one place to another depending upon geological and meteorological parameters. The data in Table 1 also show that the average radon concentration in the study area is much higher than that in many parts of the world.

As for the cations, the concentration of Ca2+ ion was found to vary from 24.4 to 691.0 mg/l with an average of 164.7 mg/l, with 77% of the samples exceeding the maximum permissible limit of Ca2+ in drinking water (75 mg/l) prescribed by WHO (Kumar and Puri 2012). The high abundance of Ca2+ may be due to the presence of gypsum and carbonate in the underlying sedimentary bed rocks (Telahigue et al. 2018). Moreover, cation exchange reactions can raise the concentration of Ca2+, while water flows upward from deep aquifer (Maoui et al. 2010).

Na+ was found to vary between 12.8 and 452.0 mg/l in the groundwater samples, with an average of 87.7 mg/l. In this case, all the samples exceed the maximum permissible limit of 20 mg/l (WHO 1996; NYS:DoH 2019) except two samples (W-69, W-55). The major source of Na+ in the water samples can be the cation exchange reactions, water–rock interaction and mixing of saline water deep down the aquifer (Daniele et al. 2013; Telahigue et al. 2018).

The samples were not found to contain significant amounts of K+ and Li+. The concentrations of K+ dissolved in the water samples were found to lie between 0.3 and 100.0 mg/l with an average of 5.8 mg/l. Thirteen percent of the samples contain K+ above the prescribed consumption limit of 8 mg/l (WHO 1996). The concentrations of Li+ in 19 samples were found to be below detection limit (BDL), and in the rest of the samples the maximum was 0.70 mg/l.

For the statistical analysis of the dataset, initially all the values have been taken into consideration. Statistical parameters such as mean, median, standard deviation and correlations between the parameters have been estimated and listed in Table 2. It has been observed that when all the data was included, the Pearson correlation coefficient between radon and Na+, K+, Li+, Ca2+ and pH are − 0.18, − 0.11, − 0.04, − 0.21 and 0.06, respectively, as depicted in Fig. 3. It can be concluded from Fig. 3 that there is almost no correlation between radon and Li+ and pH, while slight negative correlations between radon and Na+, K+ and Ca2+ were seen. In order to find the variation of all the measured cations (Na+, K+, Ca2+, Li+) and pH with each other, linear fit curves have been plotted and shown in Fig. 4. From the Pearson correlation coefficients for all the plots of Fig. 4, it is observed that there is moderately positive correlation between Na+ and Ca2+, Na+ and Li+ and pH and Li+, whereas slightly negative correlation has been found between pH and Ca2+.

Table 2 Statistical Analysis of the measured quantity for all samples
Full size table
Fig. 3
figure 3

Correlation analysis between a Na+ and radon, b K+ and radon, c Li+ and radon, d Ca2+and radon and e pH and radon concentration of all samples

Full size image
Fig. 4
figure 4

Correlation analysis between a K+ and Na+, b Li+ and Na+, c Ca2+ and Na+, d pH and Na+, e Li+and K+, f Ca2+ and Li+, g pH and K+, h Ca2+ and K+, i pH and Li+ and j pH and Ca2+ concentration of all samples

Full size image

Since the sample size is 120, and some of the samples have radon and/or cation concentrations much higher than the average, statistical analysis is distorted. Therefore, it is necessary to discard the comparatively higher values for a detailed statistical analysis using a standard statistical method of elimination. The values higher than the limit of mean plus twice the standard deviation (M + 2σ) have been discarded, and the descriptive statistical analysis is performed and tabulated in Table 3. It should be noted that a positive value of skewness indicates that the distribution has a long tail while a negative value of skewness implies a tail at the beginning of the distribution (Press et al. 2007). Another important statistical parameter is kurtosis, which describes the peakedness of the distribution. From the value of the parameter excess kurtosis (kurtosis – 3), an idea about the distribution can be obtained. If the value of the excess kurtosis is exactly zero, the distribution is called normal distribution (Kim 2013). In this study, the value of the excess kurtosis is negative for radon, Na+, Ca2+ and pH, which implies that the distribution is flat-topped curve and possesses broad distribution. On the other hand, value of the excess kurtosis is positive for K+ which corresponds to high peak and a narrow distribution (Sharma and Ojha 2020). The Pearson’s correlation coefficients between radon and Na+, K+, Ca2+, Li+ and pH are − 0.16, − 0.19, − 0.18, 0.03 and 0.16 respectively, and are depicted in Fig. 5. Thus, low negative correlation has been found between radon and all the cations except Li+ and pH. Similar correlation coefficients have been estimated for the cations and pH also. These coefficients obtained for different combination are depicted in Fig. 6. From the Pearson’s correlation coefficient of the linear fit curve (Fig. 6), it is observed that there is a positive correlation between Ca2+ and Na+ and K+, and moderately negative correlation is found between Ca2+ and pH. Remaining correlation coefficients show no significant relation except a low positive correlation between Li+ and K+.

Table 3 Descriptive statistical analysis of the measured quantity after excluding the high value samples
Full size table
Fig. 5
figure 5

Correlation analysis between a Na+ and radon, b K+ and radon, c Ca2+and radon, d Li+ and radon and e pH and radon concentration after excluding the high value samples

Full size image
Fig. 6
figure 6

Correlation analysis between a K+ and Na+, b Li+ and Na+, c Ca2+ and Na+, d pH and Na+, e Li+ and K+, f Ca2+ and K+, g pH and K+, h Ca2+ and Li+, i pH and Li+ and j pH and Ca2+ concentration after excluding the high value samples

Full size image

Dissolved radon in drinking water is the principal source of radiation doses to the population. The doses due to inhalation and ingestion have been evaluated using Eqs. (3) and (4), and these two have been added to get the total Annual Effective Dose (AED). These values are portrayed in Fig. 7 for different age groups, i.e. adults, children and infants. For the calculation of the ingestion dose associated with radon, the amount of water consumption is of great importance. Since the area of study lie in a substantially dry and humid region, an adult usually consumes ~ 4.5 L of water per day. The average water intake for children and infants are considered 2 and 0.75 L per day respectively (Ademola and Oyeleke 2017; Mitra et al. 2021). The total AED for adults, children and infants has been found to vary in the range of 0.06 to 2.42 mSv/y, 0.03 to 1.13 mSv/y and 0.01 to 0.49 mSv/y respectively, with an average of 0.65 mSv/y, 0.30 mSv/y and 0.13 mSv/y respectively. The RDL for annual effective dose recommended by European Union Commission and WHO is 0.1 mSv/y (Water, Sanitation and Health Team and World Health Organization 2004 and EU 2005). Now, from Fig. 7, it can be seen that, though there is a wide variation among the doses in the study area, 21% of the samples exceed the RDL of 0.1 mSv/y, and are even at ~ 10 − 20 times higher than the RDL. This may be due to the fact that the area from where the samples were collected could be granite-rich, and the tube-wells may also be deeper, which lead to the high radon concentration in groundwater and the corresponding high doses.

Fig. 7
figure 7

Distribution of annual effective dose received by adults, children and infants

Full size image

Conclusion

The concentration of radon in 34% of the groundwater samples under study has been found to exceed the value of 100 Bq/l for which controls to be implemented as per guidelines prescribed by World Health Organization and European Union Commission (Water, Sanitation and Health Team and World Health Organization ‎‎2004; EU commission 2001), while all samples surpass the limit of 11.1 Bq/l radon concentration in drinking water set by USEPA (United States Environmental Protection Agency) 1999. In addition, it has been observed that the total annual effective doses due to water radon in adults (≥ 17 years), children (2 − 17 years) and infants (≤ 1 year) are significantly higher than the RDL of 0.1 mSv/y prescribed by WHO and European Council. To mitigate some of the health consequences, the local people may be suggested to boil the water before drinking, as boiling expels dissolved radon from water.

Regarding the presence of cations in groundwater, correlations between radon and pH and some cations (Na+, K+, Ca2+) have been found to be weakly negative. This work would provide a baseline data for further study of water radon in the region, radon mapping of the study area, and may also be helpful in determining the Water Quality index (WQI). However, it is necessary to perform this type work repeatedly in all seasons in order to estimate the seasonal variation and the exact health risk to the inhabitants of this study area.