Natural radionuclides in drinking water and annual effective dose to infant, child, and adult

Water is essential for human life and plays a crucial role in maintaining health and well-being (Guidelines for drinking-water quality 2022). The importance of drinking water for humans cannot be overstated, as it is involved in almost all bodily functions (Riveros-Perez & Riveros 2018), including digestion, circulation, and temperature regulation. Adequate hydration is vital for maintaining cognitive function, physical performance, and overall health (Popkin et al. 2010). Moreover, ensuring that water consumption aligns with recommended guidelines is essential to prevent dehydration, which can lead to various health issues (Medicine et al. 2005; EFSA Panel on Dietetic Products, Nutrition, and Allergies (NDA) 2010). Conversely, excessive water intake can also have negative consequences, highlighting the importance of monitoring water consumption. Additionally, water is essential for nature, serving as a habitat for various organisms and playing a key role in ecosystems. It supports plant growth and helps maintain biodiversity.

Access to safe water is crucial for improving livelihoods, fostering economic growth, and reducing health risks in communities (UNESCO 2009; United Nations Development Programme (UNDP) 2015).

Regular monitoring of drinking water can help detect any potential contamination and ensure that appropriate measures are taken to protect public health. There are a large number of parameters that need to be measured in drinking water, among which the radiological parameters occupy an important place (European Council 2013; Guidelines for drinking-water quality 2022). The sources of radioactive contamination in water are diverse, ranging from natural processes like the decay of radioactive elements in soil and rocks to human activities such as nuclear power generation, industrial processes, and improper disposal of radioactive waste.

Understanding the characteristics of radioactive elements, their behavior in aquatic environments, and the potential pathways through which they can enter water supplies is crucial for assessing the associated risks and implementing effective mitigation strategies. Moreover, the impact of radioactivity on ecosystems, aquatic life, and human health necessitates continuous monitoring and regulation of water quality to ensure its safety for consumption and other uses.

Drinking water, in particular, must be meticulously controlled to meet stringent safety standards, protecting individuals from potential health hazards associated with exposure to radioactive contaminants. The European Union has approved Directive 2013/51/Euratom (European Council 2013), which specifies the radiological parameters that must be evaluated in drinking water. Additionally, it establishes the parametric levels for radon (Rn-222) in water, which must be less than 100 Bq L⁻¹, tritium (H-3), the indicative dose, which must be less than 0.1 mSv/year, gross alpha activity, which must be less than 0.1 Bq L⁻¹, and gross beta activity, which must be less than 1 Bq/L. From the UNSCEAR 2008 (2011) Report, it can be understood that on average, individuals receive approximately 0.29 mSv of radiation each year from naturally occurring radionuclides from ingestion (food and drinking water). The drinking water contribution to the ingested dose is typically 0.05 mSv.

So far, there have been several scientific publications regarding the level of radioactivity in Kosovo. The presence of natural and artificial radioactivity has been confirmed in soil (Hasani et al. 2014; Kadiri et al. 2022), and honey samples (Dizman et al. 2020), with findings indicating no risk to the population or the environment. Additionally, published scientific literature contains data on radon concentration in drinking water (Marković et al. 2020; Vučković et al. 2023), and spas (Bahtijari et al. 2022), with some papers presenting data from specific areas without providing comprehensive coverage across the country. It also mentions the studies conducted to assess the risk of depleted uranium (DU) in drinking water (Berisha & Goessler 2013), as NATO forces used DU against Serbian forces during the Kosovo War (UNEP 2001). Publicly available data on radionuclide concentration levels in drinking water are generally insufficient, and information on exposure levels of infants, children, and adults from drinking water is especially poor. As a result, it is crucial to carry out research of this kind, which will first draw attention to the current situation of the country and then act as a baseline for tracking any potential changes during the next years. In alignment with the goals of this study, all nations should maintain transparent information of this kind to ensure that consumers are kept appropriately informed. According to earlier research, Kosovo's soil contains naturally occurring radioactive elements in amounts comparable to those of its neighbors. Nonetheless, it has been verified that the artificial isotope Cs-137 has been detected in the soil (Kadiri et al. 2022). Despite the extremely low quantities found, it is unlikely that this element has been transferred to water. If verified, the effect on customers should be evaluated. Therefore, the amount of tritium in drinking water will receive particular attention.

This research presents the first effort at a national level to evaluate the level of radioactivity and assess the health risks from natural and artificial radioactivity in Kosovo's drinking water. Given that this study encompasses the entire territory of Kosovo at the national level, it can serve as a pivotal reference for subsequent research. Additionally, it aims to enhance awareness, facilitate informed decision-making, and protect the invaluable resource of water for current and future generations. The aim of this work is to determine the annual effective dose from the ingestion of radionuclides for members of the public, with a particular focus on different age categories: infants, children, and adults. Additionally, the measured radioactivity levels have been compared with the limit values recommended by international organizations, including the European Commission and the World Health Organization.

Kosovo, a small country in the Western Balkan region with a territory of approximately 10,908 km² (‘Learn About Kosova,’ n.d.), experiences a continental climate characterized by significant variations in average temperatures and regular rainfall and snowfall. The climate features a notable temperature amplitude between seasons and high precipitation levels during spring and autumn. Kosovo receives an average annual precipitation of about 900–1,300 mm (‘Fillimi – Instituti Hidrometeorologjik i Kosovës,’ n.d.), which influences the availability and usability of water resources for local communities. Relief efforts in Kosovo focus on improving water infrastructure and access, contributing to better living conditions for its residents. Officially in 2011, the population of Kosovo was approximately 1.73 million inhabitants (ASK n.d.). Water supply and treatment systems are crucial to ensure sustainable and safe access to drinking water, with infrastructure collecting and distributing water from natural sources such as rivers, lakes, aquifers, and reservoirs. Most residents rely on these systems, while in some villages, drinking water is still provided through private wells. Additionally, a significant portion of the community obtains drinking water from commercially bottled sources distributed through market networks.

In this study, a total of 44 drinking water samples were systematically collected from various locations across the country. In each case, a distilled 1 L plastic bottle was fully filled with water at the sample place. The sample site was strategically chosen to represent the primary sources of drinking water, covering diverse geographical areas and various water sources. The geographic position of each sampling site was accurately recorded using GPS technology. The collected samples were then categorized into three distinct groups based on their origins.

The first group consists of 15 samples taken from the public drinking water systems. This group includes four samples collected directly from surface water lakes used for public water supply and 11 samples obtained from the public distribution network fed by these lakes. The second group of samples consists of 22 samples collected from private wells, with 15 of these originating from Stublla Village. Particular attention was given to Stublla village, due to previous studies confirming higher natural radioactivity levels of this village compared to the rest of the country. The third group of samples comprises seven samples of bottled water available in local stores, produced by different local water factories, and packaged in plastic bottles.

A Liquid Scintillation Counter (LSC) (Perkin Elmer, LSC Tricarb 2910 TR) was employed to quantify tritium concentrations in the water samples. The LSC technique is extensively utilized for the precise determination of tritium in environmental matrices. To establish the detection efficiency of the LSC, a laboratory-derived standard was prepared using a certified liquid tritium reference standard (Eckert & Ziegler, P.O. No.: P700723, Source No.: 1676-44). Both the prepared standards and the water samples were subjected to LSC analysis. For more detailed information regarding the procedure for measuring tritium in water, please refer to the publications by Dizman & Mukhtarli (2021) and Dizman et al. (2023).

Radiation levels were analyzed using a 55% efficiency high-purity germanium (HPGe) detector (ORTEC, model GEM55P4-95) with a resolution of 1.0 keV for 57Co and 1.9 keV for 60Co. The detector was shielded with a 10 cm thick lead shield to reduce background radiation. Efficiency was calibrated using a 152Eu source. Activity concentrations for 226Ra were averaged from 214Pb and 214Bi measurements, while 232Th concentrations were averaged from 212Pb, 228Ac, and 208Tl. The activity of 40 K and 137Cs was determined from their respective gamma rays. Background radiation was measured every 3 days with an empty beaker of the same geometry as the sample containers. Elsewhere, you can find more information regarding the procedure for measuring concentration levels (Dizman et al. 2020).

To calibrate the efficiency of the liquid scintillation counting (LSC) detector, a quenching calibration curve method was utilized to determine the detection efficiency. Ten standard tritium samples with different quenching levels were prepared from a certified standard tritium source (Eckert & Ziegler, P.O. No.: P700723, Source No.: 1676-44). The same tritium concentration was used in all the samples. To obtain different quenching effects, various volumes of the quenching agent (carbon tetrachloride) were added to vials. Then, vials were counted in the tritium (0–18.6 keV) counting windows, and the efficiency graph (Figure 1) for tritium in the measurement detector was obtained by using the quenching parameter.

Statistical analyses were performed using Minitab. A one-way ANOVA test was applied to determine whether there was a statistically significant difference between the studied water groups (bottled water, public water system, and wells) at the p < 0.05 significance level. A statistically significant difference was observed between these groups in terms of radioisotopes only for Ra-226 (p < 0.05). For other radioisotopes, no statistically significant difference was observed between these groups (p > 0.05). Then, a post-hoc test (Tukey) was applied to evaluate which of these three groups had statistically significant differences for the Ra-226 radioisotope.

To assess the health impact of the consumption of drinking water, the collected samples were homogenized and then divided into two containers. In one container, the concentration of tritium was measured, while the concentration of radioelements was determined in the other container using gamma spectrometry. The gamma spectrometry measurements primarily identified the elements Ra-226, Th-232, and K-40. According to the European Directive (European Council 2013), the levels of potassium-40 (40 K) do not need to be considered in assessing health risks from radionuclides in drinking water because potassium is a key element in regulating many body functions, and the potassium content of the body is maintained at a constant level by various physiological processes. The European Directive 2013/51/Euratom has established safety criteria for radionuclides in drinking water. Regarding the parameters we have in our study, the directive specifies that the tritium level should be below 100 Bq/L, and the indicative dose from radioelements should be below 0.1 mSv/year. The levels of radon, gross alpha, and beta are not the focus of this study.

The effects of ionizing radiation on the human body from drinking water consumption were quantified by calculating the annual effective dose, expressed in sieverts per year (Sv/year). This metric, defined by European Council (2013), represents the committed effective dose over one year of ingestion from all detected radionuclides in water intended for human consumption, including those of natural and artificial origin. Notably, tritium, potassium-40, radon, and short-lived radon decay products are excluded from this calculation. In Annex I of the same directive, it is noted that elevated levels of tritium may serve as an indicator of the presence of other artificial radionuclides. If the tritium concentration surpasses the specified parametric value of 1,000 Bq/L, further analysis is required to assess the presence of additional artificial radionuclides.

Initial screening involves measuring both gross alpha and gross beta activity concentrations. If these concentrations are below the screening levels of 0.5 Bq/L for gross alpha activity and 1 Bq/L for gross beta activity – no further action is required; otherwise, the measurements for specific radionuclides must be followed. For further guidance on actions to take when screening levels or guidance levels are exceeded, refer to Annex 1 of the Management of Radioactivity in Drinking Water (‘Management of radioactivity in drinking-water,’ n.d.).

If the sum exceeds one for a single sample, the IDC of 0.1 mSv/year would only be surpassed if exposure to the same concentrations persisted throughout an entire year. Therefore, such a result does not automatically indicate that the water is unsuitable for consumption.

The effective dose coefficients (CF) for ingestion of radionuclides by members of the public, published in the Annex F of ICRP 119 (Eckerman et al. 2013).

Using gamma spectrometry on water samples, the radionuclides Ra-226, Th-232, and K-40 were detected. Table 1 presents the values of the factors: f₁, CF, and R for the category's infant, child, and adult, corresponding to the radioisotopes present in the water samples. Often the factor of fractional absorption in the gastrointestinal tract is neglected (Pintilie-Nicolov et al. 2021).

R represents the annual consumption rate of drinking water and beverages (L year⁻¹), assumed to be 0.75 L day⁻¹ for infants, 1 L day⁻¹ for children, and 2 L day⁻¹ for adults (Guidelines for drinking-water quality 2022). From these values, the annual drinking water consumption is calculated to be 273 L years⁻¹ for infants, 365 L years⁻¹ for children, and 730 L years⁻¹ for adults, respectively.

The mean concentration of tritium in drinking water was determined to be 1.84 Bq L⁻¹. Specifically, the average tritium concentrations in the public water system, wells, and bottled water were found to be 2.00, 1.75, and 1.67 Bq L⁻¹, respectively. These observed results were far lower than the standard for drinking water, which is 100 Bq L⁻¹ (European Council 2013; Guidelines for drinking-water quality 2022; ‘Management of radioactivity in drinking-water’ n.d.). Table 2 presents the statistical summary of the radionuclide concentrations detected during the measurements. The radioelement levels of the water sources are quite similar. The corresponding standard deviations for all samples of Ra-226, Th-232, K-40, and tritium are 0.41, 0.26, 3.33, and 0.52 Bq/L, respectively. This indicates that the range of results is quite well-defined and that increasing the number of samples will get comparable results. As a result, the quantity of samples seems to be typical of this occurrence, showing constancy in the results, even if they came from different origins.

Overall, the results reported in this study fall within the range of values for radioelements found in the publications of other authors. When comparing the measurement results from this study with values reported in a study conducted in Nigeria, we find that the concentration levels of Ra-226 and Th-232 in this study are seven and six times lower, respectively, than those in Nigeria (Muhammad et al. 2022). Similar levels of tritium concentration in drinking water have been reported by various authors. Other studies have documented the mean of tritium concentrations in tap water as follows: 2.38 Bq L⁻¹ in Bulgaria (Damianova et al. 2016), 7.0 Bq L⁻¹ in Adana, Turkey (Gören et al. 2014), and 6.2 Bq L⁻¹ in Mersin, Turkey (Karataşlı et al. 2017). Additionally, tritium levels were reported to be up to 0.9 Bq L⁻¹ in the spring water of Spain (Palomo et al. 2007b).

Tritium is generally not present in spring waters unless influenced by external sources. The contribution of tritium to the effective dose from drinking water is minimal. Therefore, it has not been included in the final calculation of the total effective dose. This exclusion is due to the effective dose coefficients for ingestion, which are in the order of 10⁻¹¹, as shown in the first column of Table 1. Potassium-40 (K-40) is excluded from health risk assessments under European Directive 2013/51/Euratom because it is a naturally occurring isotope present at stable levels in biological tissues. The directive focuses on radionuclides that pose more variable and direct health risks, often due to contamination. While K-40 contributes to the total dose, its consistent presence in the environment and diet justifies its exclusion from the directive's scope, emphasizing the monitoring of more variable radionuclides.

The activity concentrations of Ra-226, Th-232, and K-40 are presented in Table 2. The results indicate that the concentrations of these radionuclides varied minimally across different samples. Specifically, in surface lake water, the activity concentrations ranged from 0.37 to 0.98 Bq L⁻¹ for Ra-226, 0.27 to 0.48 Bq L⁻¹ for Th-232, and 2.47 to 9.35 Bq L⁻¹ for K-40. Similar values were observed in other groupings of source water.

To provide a clearer understanding, the following symbols were used in Figures 2 and 4: the red circle represents the mean value, the box indicates the 50% range, the vertical red line denotes the 90% range, the horizontal red line inside the box marks the median, and extreme black lines outside the box illustrate the minimum and maximum values.

To assess the impact of radionuclides on the annual effective dose, the average values of each radioelement across three age categories were determined. The results indicate that Ra-226 contributes the most to the annual effective dose, followed by K-40, and Th-232 contributes the least.

This suggests that there may be regional or environmental variables influencing water quality, as the well water resources in the Stublla region show unique characteristics when compared to other places. Interestingly, no significant changes were seen between other pairwise comparisons, suggesting that the water quality profiles of well water from non-Stublla locations and the distribution network are similar.

The concentration levels of radionuclides should not be undervalued, as has been shown in numerous research studies on the risk of cancer linked to drinking water. According to Karabıdak et al. (2019), a person who drinks such drinking water sources over a 70-year period has a cumulative cancer risk of about 1 in 100,000. Calculated average levels of 11.6, 2.8, and 102.8 Bq L⁻¹ for 226Ra, 232Th, and 40 K, respectively, served as the basis for their conclusions.

The results indicated that, on average, for each type of drinking water source, the highest annual effective dose was observed for infants, followed by children, and the lowest for adults. This trend is most clearly illustrated in Figure 4. Generally, the level of annual effective dose across all drinking water sources in Kosovo was low and relatively similar. However, there were some variations, with the highest values found in well water, followed by bottled water and public drinking water systems, with average values of 0.367, 0.334, and 0.284 mSv year⁻¹, respectively.

The variation in the annual effective dose by age category was consistent across all drinking water sources, with the values ranging from highest to lowest in the following order: infants, children, and adults. This situation was analogous to the explanation previously provided for age categories in Figure 2. Thus, the factors f1 (fractional absorption) and conversion factor (CF) caused this distribution in the results.

When comparing the average values of drinking water from distribution networks and lakes, both of which were components of the public water supply system, it was observed that their averages were nearly identical, with a difference of only 0.3%. This similarity was expected, given that water from lakes was conveyed through pipes within the distribution network. Measurement uncertainties explained the minor differences in average values between the two water sources.

Since 80% of the population had access to the public water system in 2015, the values for this category most accurately represented the annual effective dose level for the population of Kosovo. On average, this annual effective dose was 0.286 mSv year⁻¹, with values ranging from 0.036 to 1.230 mSv year⁻¹.

According to previous publications and ongoing research in the village of Stubell in the municipality of Viti, it was confirmed that the level of natural radiation was higher than in other parts of the country. Therefore, special attention was given to this region. The residents of this village were supplied with drinking water from local wells, which is why comparisons were made with wells from other areas. The results obtained in this study indicated that the average annual effective dose from well water used for drinking in Stubell was 1.47 mSv year⁻¹ (ranging from 0.074 to 1.75 mSv year⁻¹). In contrast, the average annual effective dose from well water in other parts of the country was significantly lower, with a mean value of 0.245 mSv year⁻¹ (ranging from 0.042 to 0.877 mSv year⁻¹).

It has been shown that the annual effective dose primarily comes from the presence of Ra-226. Therefore, various methods can be used to reduce the level of radium in drinking water, such as ion exchange and reverse osmosis. Ion exchange is more cost-effective and offers a good solution for small to medium-scale water treatment, especially in residential or community-scale applications. The use of methods to reduce effective annual doses should be applied, especially to infants.

While this study sheds light on the concentrations of radionuclides in drinking water, it should be noted that there are certain limitations. One example is that seasonal weather variations, such as changes in precipitation, temperature, and indoor activities, can concentrate radionuclides in water. For example, heavy rainfall might exacerbate runoff from contaminated soil. Inadequate sampling throughout the year can lead to an incomplete picture of radioactive levels. Furthermore, the sample size and geographic dispersion may not accurately reflect local changes in water sources, particularly in areas with complex geological and hydrological characteristics.

Finally, uncontrolled confounding factors such as changes in water treatment procedures, industrial discharges, and geological characteristics were not systematically assessed, but they could have a considerable impact on radioactive concentrations. Addressing these limitations in future studies, such as seasonal sampling, larger sample sizes, and the incorporation of confounding variables, will provide a more robust understanding of radionuclide concentrations in drinking water and improve the findings' applicability to public health risk assessments and interventions. Also, future studies should include cancer risk assessments from drinking water.

The study confirmed that the annual effective dose from drinking water across Kosovo remained within national and international standards. Tritium levels were well below regulatory limits, and its contribution to the effective dose was negligible. The research highlighted that infants, due to higher fractional absorption and dose coefficients, received the highest annual effective dose from drinking water. To lower the annual effective dose for infants, one can apply any techniques to reduce radionuclide levels in drinking water, with distillation being one of the simplest and most effective methods. For children and adults, no additional measures are deemed necessary. Regional differences were noted, with Stubell exhibiting a significantly higher dose compared to other areas. The observed variations were attributed to natural radiation levels and measurement uncertainties. Overall, the annual effective dose from drinking water was consistent with safety standards, providing reassurance regarding public health.

No funding was received for this research

All relevant data are included in the paper or its Supplementary Information.

The authors declare there is no conflict.