A study of naturally occurring
radon in Swedish water
purification plants
VIKTORIA WIKING
KTH ROYAL INSTITUTE OF TECHNOLOGY
A study of naturally occurring
radon in Swedish water
purification plants
Viktoria Wiking
Degree Project in Environmental Engineering and Sustainable Infrastructure KTH Royal Institute of Technology
TRITA LWR Degree Project ISSN 1651-064X
Summary in Swedish
Radon är ett naturligt förekommande radioaktivt ämne. Exponering för radon, och specifik för isotopen radon-222, är kopplad till en ökad risk för att utveckla lungcancer. Det är därför viktigt att undersöka de platser där förhöjda radonhalter kan förekomma och där människor vistas regelbundet.
Det är känt att förekomst av radon i dricksvatten kan leda till förhöjda radonhalter i inomhusmiljön. I dricksvattenverk behandlas stora volymer vatten och därmed kan det finnas en risk att förhöjda radonhalter uppstår i inomhusluften. Syftet med studien är att undersöka vilken påverkan rening av grundvatten har på radonhalten. Detta för att se om det finns något samband mellan den uppmätta radonhalten i luften och mängden vatten som renas, eller mellan den uppmätta radonhalten i luften och radonhalten i vattnet. Dessutom har en uppskattning av exponeringen utförts för att undersöka om exponeringen kan anses problematisk för de arbetstagare som regelbundet vistas i dessa utrymmen.
Totalt ingår resultat från 39 grundvattenverk med bergborrade vattentäkter. Genom vattenprover, långtidsmätningar och korttidsmätningar har radonhalten i vatten och luft kartlagts. Två vattenprover, ett på råvattnet och ett på det renade vattnet, har hämtats från varje verk. På samtliga verk har även långtidsmätningar av radonhalten i inomhusluften utförts, med spårfilmsdetektorer. På två av verken har dessa data kompletterats med korttidsmätningar av radonhalten i inomhusluften, med hjälp av AlphaGUARDs. Vidare har en exponeringsundersökning utförts där deltagarna ombads att ange hur ofta verken besöktes och hur lång tid besöken tog. Detta möjliggjorde en kvantifiering av exponeringen då tidsuppskattningarna kunde kombineras med de uppmätta radonhalterna i inomhusluften.
Vattenproverna visar på en minskning av radonhalten under reningen av vattnet. Dock varierar omfattningen av denna minskning kraftigt, mellan 0-99 %. Spårfilmsmätningarna visar att 90 % av verken har halter som överstiger 200 Bq/m3. Liknande resultat uppmättes vid korttidsmätningarna
som vid spårfilmsmätningarna på samma verk. Dock uppvisade korttidsmätningarna en stor variation i den uppmätta radonhalten vid olika tidpunkter på dygnet.
Tekniker/ingenjörer är den yrkesgrupp som vistas mest på verken, även om den tid de spenderar på verken är begränsad. För 84 % av verken understiger tiden 100 timmar årligen, och samtliga verk besöks mindre än 1000 timmar årligen. På ett verk fanns indikationer på att exponeringen översteg gränsvärdena, men även i andra fall kan exponeringen vara problematisk. Till exempel besökte alla tekniker och ingenjörer även andra vattenverk under sin arbetstid vilket kan medföra en totalt sett högre exponering.
I princip inga tydliga samband kunde hittas mellan de parametrar som undersöktes. Till exempel kunde inget samband hittas mellan radonhalten i luft och volymen vatten som renades eller mellan radonhalten i luft och radonhalten i vatten. Även ett flertal andra samband undersöktes såsom samband mellan brunnsdjup och radonhalten i vatten. Vidare undersöktes också möjliga samband för de parametrar som uppmättes under korttidsmätningarna.
Acknowledgements
I would like to thank the Swedish Radiation Protection Agency for providing the funds necessary for this thesis. In addition, there are several persons I would like to thank for their support during the completion of this thesis. Firstly, I would like to thank my supervisor Kirlna Skeppström at the Swedish Radiation Protection Agency for the help and support during this thesis. I would also like to thank my examiner professor Bo Olofsson at KTH for his advices. Moreover, I appreciate the help from Magnus Ahnesjö and Erik Wåhlin at the Swedish Radiation Safety Authority with the AlphaGUARDs and some logistics. I also appreciate the help from SGU for providing the contact information to the water suppliers. Furthermore, I would like to thank Hedi Rasul at KTH who drove me to the sites for the short-term measurements. In addition, I would like to thank my family for all their invaluable support during this period.
Abstract
Radon dissolved in drinking-water can be transferred into the indoor air and is one of the main transfer pathways for radon. At water purification plants, large quantities of water are treated and there is a risk that radon degasses from the water and enters into the indoor air. Hence, there is a risk for elevated radon levels in the indoor air at these facilities. This study aims to investigate the general impact of water treatment processes on the radon concentration in water and its transfer into the indoor air. Moreover, the risk that radon exposure exceeds the regulatory limits at workplaces was investigated. In total, the results from 39 Swedish water purification plants are included in the study. The methodology includes long-term air measurements with alpha track detectors, and short-term air measurements with AlphaGUARDs. In addition, water samples were collected in order to analyze the radon concentration in the untreated and treated water.
The results show that several plants experience elevated radon levels in the indoor air and in some cases the exposure could be problematic. Several connections were investigated without finding apparent connections for those cases. For example, the relation between radon concentration in the water and radon level in the indoor air was investigated and the connection between the volume of water treated and the radon level in the indoor air. Calculations with transfer coefficients indicate that the transfer of radon into the indoor air is relatively small. However, there can also be contribution from other radon transfer pathways, such as soil and buildings, which may have an impact on the radon levels in the indoor air.
Key words
Table of contents
Sammanfattning ... VI
Acknowledgements ... VIII
Abstract ... X
Key words ... X
Table of contents ... XII
Table of figures ... XIV
Introduction ... 1
The aim of the study ...1
Background ... 2
Basic theory about radioactivity...2
Decay series and radon ...2
Health effects due to radon exposure ...3
Transfer of radon into the indoor air ...4
Soil ...4
Building materials ...4
Water ...4
Regulatory limits ...7
Directive from the European Council and its implementation ...7
Measurement and analyzing methods for radon in air and water...7
Air measurements ...7
Water analysis ...9
Methods
...9Selection and contact ...9
Collection and analysis of the data ... 10
Water samples and measurement of radon in water ... 10
Long-term air measurements of radon ... 11
Short-term air measurements of radon ... 12
Exposure survey ... 13
Statistical analysis ... 13
Ethics ... 14
Results
... 14Volume of water treated and radon concentrations in the water samples ... 15
Results from the long-term measurements of radon in the indoor air ... 16
Results from the short-term measurements of radon in the indoor air ... 16
Estimations of the time spent at each water purification plant ... 18
Estimations of the exposure to radon ... 19
Relation between parameters ... 20
Estimation of radon transfer by application of transfer coefficients... 24
Discussion
... 25Comparison of the results from the water samples to previous studies ... 25
Interprentation of the long-term air measurements ... 26
Impact from different radon transfer pathways ... 26
Short-term measurements ... 27
Investigation of possible relationships ... 27
Exposure estimations ... 28
Conclusions... 29
References ... 30
Published references ... 30
Other references ... 34
Appendix A: Water production ... 35
Appendix B: Results from the water samples ... 36
Appendix C: Results from the long-term air measurements ... 39
Appendix D: Results from the short-term measurements at W16 ... 44
Appendix E: Results from the short-term measurements at W31 ... 46
Appendix F: The estimated time spent at the water purification plants ... 48
Appendix G: The estimate annual exposure... 50
Appendix H: A translation of the time estimation questionnaire ... 51
Appendix I: Estimated contributions from the water to the indoor air ... 53
Table of figures
Figure 1: The elements in the uranium-238 series, their half-lives and if they emit alpha or beta radiation during their decay (ICRP, 2014). ... 3 Figure 2: The location of the water purification facilities that participated in this study. The figure is based on data from Lantmäteriet; their 10 million Sweden map (Lantmäteriet, 2016a) and their overview map (Lantmäteriet, 2016b). ... 10 Figure 3: The risk for high radon concentration in groundwater in Sweden and the location of the water supplies in this study. The map is from SGU based on data about the uranium concentration in the bedrock and data from radon samples of well water in Sweden (Strålsäkerhetsmyndigheten, 2009). ... 11 Figure 4: The appearance of an alpha track detector (Landauer Nordic, n.d.). ... 12 Figure 5: One of the AlphaGUARDs used in the short-term measurement of radon levels in the indoor air. .... 13 Figure 6: The daily average volume of water treated by each of the participating water purification plants. ... 15 Figure 7: The radon concentrations in the samples from the untreated water at the water purification plants. . 15 Figure 8: The radon concentrations in the samples from the treated water at the water purification plants. ... 16 Figure 9: The detected radon levels in the indoor air at the water purification facilities, where the treatment occurs.
... 17 Figure 10: The measured radon gas levels in the indoor air at the water purification plant W16 during the week the measurement took place. ... 17 Figure 11: The measured radon gas levels in the indoor air at the water purification plant W31 during the week the measurement took place. ... 18 Figure 12: The estimated time spent at each water purification plants by the technicians/engineers annually. 19 Figure 13: The estimated annual exposure for the personnel at the water purification plants. ... 19 Figure 14: Relation between the radon concentration in the untreated water and the radon level in the indoor air of the treatment rooms. ... 21 Figure 15: The relation between the quantity of water produced and the radon concentration in the untreated water. ... 21 Figure 16: The decrease in radon concentrations in the water versus the radon concentration in the untreated water. The samples that were below the detection limit (less than 20 Bq/l) are not included. ... 22 Figure 17: The measured radon levels in the indoor air at the water purification plants W16 and W31. ... 22 Figure 18: Estimated transfer from radon dissolved in water into the indoor air for the transfer coefficient 4.9*10 -3, for each of the plants, and comparison to the measured radon levels in the indoor air.. ... 24
Figure 19: Estimated transfer from the radon dissolved in water into the indoor air for the two transfer coefficients 10-4 and 1.96*10-4, for each of the water purification plants, and comparison to the measured radon levels in
Introduction
Exposure to the radon isotope radon-222 has been linked to an increased risk for lung cancer. In Sweden, it is estimated that exposure to radon causes approximately 500 deaths from lung-cancer annually (Strålsäkerhetsmyndigheten, 2009). As this isotope occurs naturally in the environment, it provides a continuous source of radiation. By controlling and altering the pathways of radon, the inflow to the indoor air can be minimized (ICRP, 2014).
One of pathways is the transfer of radon that is dissolved in drinking-water into the indoor air (ICRP, 2014). The radon concentration is often related to the origin of the drinking-water. Groundwater often have higher concentrations of radon compared to surface water (De Zuane, 1996). Moreover, water from private wells tend to have higher radon concentrations than water from public water supplies (Kulich et al., 1988). Often the radon concentrations in water from the public water supplies are low as a result of degassing and decay during the treatment processes (ICRP, 2014). The transfer of radon from water into the indoor air during treatment processes can potentially cause elevated radon levels in the indoor air at water purification plants. Several previous studies have detected elevated, and occasionally very high, radon levels in the indoor air at water purification plants (Trautmannsheimer et al., 2002; Körner et al., 2005; Ringer et al., 2008; Stietka et al., 2013). Moreover, it is important to investigate facilities that are suspected to have elevated radon levels in the indoor air due to the new radiation directive from the European Council, 2013/59/Euratom. The directive regulates the acceptable radon exposure (The Council of the European Union, 2014) and will in Sweden be implemented by a new radiation law. In the preliminary version, the radon levels in the indoor air at workplaces should be as low as reasonable achievable and optimized below the reference level of 200 Bq/m3 (Strålsäkerhetsmyndigheten, 2016).
This study is performed in collaboration with the Swedish Radiation Safety Authority. They have provided several resources, such as the sampling equipment and resources for laboratory analysis. The water purification plants that are participating in this study extract their drinking-water from groundwater in the bedrock. Most of the participating water purification plants are small-scale water producers. Some of the water purification plants have specific treatment for radon reduction while other are without specific treatment for radon removal.
The aim of the study
In Sweden, there are few studies that combine an analysis of the radon levels in indoor air and the radon concentrations in water for these types of facilities. Often, the radon concentration in drinking-water is assessed to estimate the concentrations distributed to the public. For the radon level in the indoor air, it is mainly assessed in order to estimate exposure for the workers. This study includes and combines both these perspectives. The study intends to investigate the impact that water treatment processes have on the transfer of radon from water into the indoor air. The aim is to detect possible relations between the radon levels in the indoor air and other parameters and to estimate the radon exposure for the workers at the participating water purification plants. This can be summarized into the following research questions:
Is there a relation between the radon levels in the indoor air and the radon concentration in the untreated water?
Background
The following sections provide more detailed information about radioactivity and especially radon. There is information about its origin, health effects and the Swedish regulatory limits and guidelines.
Basic theory about radioactivity
Some isotopes are instable and their number of protons and neutrons do not remain constant (Burchfield, 2009). Instead, these are transformed through different radioactive decay processes. During the radioactive decay energy-rich particles and radiation are emitted from the nucleuses of these instable atoms that often is referred to as radionuclides (Kunz, 2009). As a result of the radioactive decay process, an atoms transforms into a new atom with other properties than its parent compound (Airey et al., 2012).
Three common types of radioactive processes are: alpha decay, beta decay and gamma-ray emission. Different particles are emitted during these processes. For alpha decay, the particle is a helium nucleus that consists of two neutrons and two protons (Airey et al., 2012). There are several types of beta decay processes that emit different particles: electron capture, B- and B+ decay processes. An example of the emitted particles is the electron and antineutrino that is emitted during the B- decay process. For the gamma-ray emission, photons are emitted (Kratz and Lieser, 2013). These decay processes can be combined, and for some radionuclides, alpha and beta decay processes are followed by emission of gamma-ray radiation (Kunz, 2009).
The properties of the emitted particles determine their ability to travel through matter. According to Burchfield (2009) particles with higher charge and size have less ability than smaller particles to penetrate materials. For example, alpha particles can be stopped by a piece of paper. Gamma-ray radiation on the other hand requires very dense materials, such as lead, to be hindered to some extent (Kunz, 2009).
Moreover, the product from a radioactive decay is not necessarily stable and subsequent radioactive decay may occur. So called decay series describe the stages of radioactive decay, until a stable isotope is created (Burchfield, 2009).
Decay series and radon
Radon has several isotopes, where radon-222, radon-220 and radon-219 occur naturally. It is common to refer to these isotopes as radon (radon-222), thoron (radon-220) and actinon (radon-219) (Martin and McBride, 2012). Generally, radon is perceived as the most important isotope (Martin, 2013). In most cases, thoron exposure is limited and do not pose any problem and the radiation exposure from actinon is insignificant due to their short half-lives (ICRP, 2014).
Radon is a part of the decay series for the radionuclide uranium-238 (Martin, 2013). There are several stages in the U-238 decay series where radon is the direct product of the radionuclide radium-226 (Fig. 1). For the main series, there are several alpha and beta decay processes between the initial uranium-238 and the final lead-206 isotope (Airey et al., 2012).
Figure 1: The elements in the uranium-238 series, their half-lives and if they emit alpha or beta radiation during their decay (ICRP, 2014).
Health effects due to radon exposure
Exposure to radon has been linked to an increased risk for lung-cancer. In Europe, it is estimated that 2 % of the deaths in cancer are due to radon exposure. The main risk groups are smokers and recent non-smokers. The risk for lung-cancer is estimated to be 25 times larger for smokers compared to the risk for non-smokers (Darby et al., 2005). In Sweden, it is estimated that 500 deaths from lung-cancer each year can be linked to radon exposure (Strålsäkerhetsmyndigheten, 2009).
Transfer of radon into the indoor air
Radionuclides can have a natural or an anthropogenic origin. Uranium-238 originates from before the formation of the Earth (Kratz and Lieser, 2013). Uranium-238 and radium-226 have a natural origin and exist in bedrocks and soils (ICRP, 2010). So does radon, which occur naturally in all soils, to varying extent (Martin, 2013). Henceforth, radium-226 is referred to as radium and uranium-238 as uranium. This section describes the main radon transfer pathways to the indoor air.
Soil
In some cases, radon located in the soil can transfer into buildings where it may cause elevated radon levels (ICRP, 2014). Radon is the only radionuclide in its series that occurs in gaseous form. Therefore, radon may migrate from the soil and its parent compounds (Martin, 2013). High radon concentrations are often measured in the soil, where it is primarily transported by diffusion or convection, depending on the properties of the soil (ICRP, 2014). For instance, the permeability of the soil and the water contents in the pores affect the transportation through the soil. A higher water content in the pores decreases the potential for radon to transfer through the soil (Tell et al., 1994). A dilution quickly occurs for the radon that is transferred from the soil to the soil surface and atmosphere (ICRP, 2014).
There are several factors that influence the indoor radon levels. Generally, these can be divided into three categories; spatial factors, temporal factors and specific properties of the investigated building. An example of spatial variations is site-specific geology (Kropat et al., 2014). A high uranium concentration of the bedrock has been linked to high radon levels in the indoor air (Buchli and Burkart, 1985; Tell et al., 1994). For the temporal variations, an example is climatic variations (Kropat et al., 2014). Climatic variations that impact the indoor radon levels are the wind and the outdoor temperature (Buchli and Burkart, 1985; Arvela et al., 2016). For instance, higher radon levels in the indoor air were detected at less windy days. Moreover, a relationship was detected between the outdoor temperature and the radon level in the indoor air (Buchli and Burkart, 1985). Generally, the indoor and outdoor temperature difference is a major cause of the radon transfer to a building. The difference causes a pressure gradient that leads to increased transfer of (radon-rich) soil air to the building (ICRP, 2014). Moreover, the outflow of indoor radon is linked to the air-exchange of a building that is affected both by wind and the outdoor temperature (Arvela et al., 2016).
Also the buildings’ construction and properties have an impact on the in- and outflows of radon (Arvela et al., 2016). For instance, the occurrence of cracks and pipes can affect the air flow into the building (ICRP, 2014). Moreover, the ventilation and variations in the ventilation can impact the radon level in the indoor air. In some cases, the ventilation is only running during the work day and not during evenings and weekends. Hence, the radon level in the indoor air can increase during that time. At some places, there is therefore a risk for overestimating the indoor radon levels (Annanmäki et al., 1996; Kávási et al., 2006)
Building materials
The radon emissions from building materials are important to consider in certain cases. In Sweden, concrete produced from alum shale with high uranium concentrations has been linked to elevated radon levels in the indoor air. It was produced between the years 1929-1975 and there was a widespread use of this material (Åkerblom et al., 2005). For other building materials, emission of radon can be prevented by selection of materials with low radium concentrations (ICRP, 2014). Water
Factors that affect the radon concentration in water
The radon concentration in untreated water is related to its origin. Generally, surface water has a low concentration of radon. Groundwater on the other hand can have high concentrations (De Zuane, 1996). Generally, groundwater from bedrocks tend to have a higher radon concentration compared to soil dug wells (Kulich et al., 1988).
Groundwater extracted from the bedrock is affected by the subsurface environment. Several factors affect the radon concentration: radon emanation, concentration of uranium and radium in the bedrock and contact surface between the water and the bedrock (Kulich et al., 1988). When radium decay, radon can be contained in the mineral if the mineral is sufficiently dense. Then the radon remains in the rock until it subsequently decays. The ratio between the amount radon that leaves the mineral and the total amount produced is often referred to as the emanation coefficient. In bedrocks, this coefficient is often lower than 50 % although it can be higher in tectonic areas. It is lower in regions where the bedrock contains less fissures and cracks. In some cases, this coefficient is more important for the radon concentration in groundwater than the radium concentration detected in the rocks (Przylibski, 2000). However, the fractures need to be water-bearing in order for the groundwater to transport radon (Strålsäkerhetsmyndigheten, 2009).
It is not necessarily a link between the uranium concentration in the bedrock and the radon concentration in the groundwater. In some areas, high radon concentrations in groundwater has been linked to the occurrence of bedrock with elevated uranium levels. In contrast, high radon concentrations in groundwater have also occurred at sites with low uranium concentrations in the bedrock (Kulich et al., 1988; Skeppström and Olofsson, 2006). This can be explained by the properties of some of the elements in the uranium-238 series. Several compounds in the series have low solubility, such as uranium and radium (Sun and Semkow, 1998). These elements can precipitate on the surfaces of the fractures and there provide a continuous radon source. This process could explain the variations in uranium concentration in the bedrock and radon concentration in the groundwater. These elements could be transported significant distances before precipitation. Subsequently, the emanation of radon could cause transportation of the radon even further from the initial bedrock. Hence, the process could explain the occurrence of elevated radon levels in areas with low-uranium containing bedrocks (Knutsson and Olofsson, 2002). This process could also explain the weak relations that have been detected between the radon concentration in groundwater and the concentration of its precursors in groundwater. For instance, relatively weak correlation occurred between radon and uranium and between radon and radium in groundwater (Ek et al., 2008). Also Skeppström (2005) only detected weak correlations between the radon and radium concentration and between the radon and uranium concentration in groundwater.
Radon emission from water to the air
There are several factors that affect the transfer of radon from water into the indoor air. In similarity with its precursors, radon has a low solubility in water (ICRP, 2014). Its solubility in water is temperature dependent and decrease at higher temperatures (Battino, 1979). The transfer process depends on the radon concentration in the water, the flow of the water and the properties of the premise such as the total indoor volume and the properties of the air-exchange (Tayyeb et al., 1998). Larger flow rates can increase the risk of radon transfer into the air (Stietka et al., 2013). However, there are not necessarily an apparent correlation between the indoor radon levels and the flow rate of the water or between the radon concentrations in water and the indoor radon levels (Stietka et al., 2014).
Treatment processes often have an impact on the radon concentration in water and may influence its transfer into the air. Basically all form of mechanic movement of the water may cause a release of radon from the water (Reichelt, 1996). Often the radon concentrations in water from public water supplies are low, as a result of decay or degassing of radon during its treatment (ICRP, 2014). Especially vaporizers and cascade processes have been linked to an increased risk for radon transfer from the water (Ringer et al., 2008). Cascade mixture tanks cause aeration, which increase the risk for radon in the indoor air (Stietka et al., 2013).
Some processes are implemented specifically for radon removal at water purification plants. Generally, aeration methods are very efficient and examples of methods that can be used are packed tower aerators and diffused air systems. Also granular activated carbon (GAC) can be used, although it is not considered as efficient as the previous mentioned aeration methods (Twort et al., 2000). Transfer coefficient
The radon transfer coefficient can be used to estimate the transfer from radon dissolved in water into the indoor air (Ongori et al., 2015). For households with a normal consumption, the standard value is 10.4 (Vinson et al., 2008). For radon concentrations of 10 000 Bq/m3 in the water (i.e. 10 Bq/l) that
results in an indoor increase of 1 Bq/m3 (UNSCEAR, 1993). At thermal spas, transfer coefficient has
also been estimated. In those cases, it has been estimated to 4.9 * 10-3 (Radolić et al., 2005) and 1.96
* 10.4 (Sainz et al., 2016).
Previous studies on radon exposure at water purification facilities
Previous studies at water purification plants have detected high radon levels in the indoor air. In some cases, the radon levels in the indoor air exceeded 30 kBq/m3 (Ringer et al., 2008; Stietka et al., 2013).
There are studies where even higher radon levels in the indoor air have been measured, such as maximum levels at 1 000 kBq/m3 (Reichelt, 1996; Körner et al., 2005) and at 271 kBq/m3
(Trautmannsheimer et al., 2002).
Generally, the median radon level in these studies is significantly lower than the maximum radon level. One of the studies had regional medians that ranged between 500 Bq/m3 to 1200 Bq/m3
(Trautmannsheimer et al., 2002). Ringer et al., (2008) observed that more than half of the water purification facilities had radon levels below 1000 Bq/ m3 and 91 % below 5000 Bq/m3. Also Stietka
et al., (2013) detected a large variation in the measured radon levels that ranged from less than a 100 Bq/m3 to more than 30 000 Bq/m3.
However, there are other factors that can impact the exposure. The exposure increase for workers who visit many plants during their work day and an individual assessment is recommended to manage those situations (Arbetsmiljöverket, 2002). Moreover, the exposure could vary between different professions due to variations in the frequency and length of their visits. According to Reichelt (1996) the so called attendant of the facility is the person that is most at risk for exposure, although workers from other professions also visit the plants and are at risk for exposure.
Regulatory limits
There are recommendations and regulations for radon levels in indoor air and concentrations in drinking-water. In Sweden, the maximum allowed radon level at workplaces, not classified as underground work, is 0.36 * 10-6 Bq h/m3 during a year. It is defined as a total allowed exposure that
for 1800 work hours equals a radon level of approximately 200 Bq/m3. Other guidelines apply for
underground work, such as for mining or underground construction work (Arbetsmiljöverket, 2015). There are guideline values regulating the radon concentration in drinking-water for water suppliers. Radon concentrations exceeding 1000 Bq/l are considered unfit for consumption. A concentration exceeding 100 Bq/l can be consumed but actions are needed in order to decrease the radon concentration (Livsmedelsverket, 2015a). These guidelines apply for the concentration at the consumer’s tap. Therefore, these guideline values consider the concentration after distribution from the water purification facilities, by for instance pipes and pumps (Livsmedelsverket, 2015b). Certain criteria need to be met for these guidelines to apply to the drinking-water. These guidelines only apply for water suppliers who supply water to 50 persons or more, or treats 10 m3/day or more. They also
apply for those cases where the water is used for commercial or public purposes, regardless the size of the business (Livsmedelsverket, 2015b). For example, it is classified as public if it the treatment is administrated by the municipality (Socialstyrelsen, 2006).
Directive from the European Council and its implementation
There is an updated radiation directive in the European council, 2013/59/Euratom (The Council of the European Union, 2014). In Sweden, this directive is implemented by a new radiation law. In the preliminary version, it is suggested that the radon levels should be as low as reasonably achievable for work places. The level should be optimized below the reference level of 200 Bq/m3 for work. In
accordance with the requirement from the directive, radon concentrations which continue to exceed reference level despite remediation measurements should be reported. When there is a risk that the radiation dose due to radon exceeds 6 mSv annually, the exposure situation should be managed as a planned exposure situation (Strålsäkerhetsmyndigheten, 2016). For drinking-water, there is another council directive that regulates the concentration of radon, 2013/51/Euratom. As regards to radon, the prevailing limits in Sweden did not need any revision (Livsmedelsverket, 2015d).
Measurement and analyzing methods for radon in air and water
There are several methods that can be used to analyze the radon concentration in air and water. The following sections describe common methods for radon analysis that are relevant for this study. Air measurements
Alpha Track Detectors (ATD) can be used to measure the radon levels in the indoor air at work places. These often have detector material placed inside a closed detector with a filtered entrance or with a limited diffusion possibility (Hagberg et al., 2008). Filtered entrances on the detectors are often used to prevent the entry of radon progeny into the detectors (Bochicchio, 2005). This type of detector measures the alpha radiation from the radon gas that has diffused inside by investigating the damage on the detector material (Hagberg et al., 2008). By etching with a reagent, these damages can become visible and observed in a microscope (Kropat et al., 2015). From this the radon levels can be determined; based on the amount of traces per unit surface and the exposure time. Other parameters are also considered, such as the material of the detector, the construction of the detectors and the methodology during the etching stage (Hagberg et al., 2008). For instance, CR-39 or LR 115 are examples of two types of detector materials (Bochicchio, 2005).
Measurement methods for radon in air should consider certain factors. The equipment should be suitable for the conditions at the measurement site. For instance, they should be adapted for conditions where there can be high humidity and limited access to electricity (Ringer et al., 2008). CR-39 detectors have been used previously for other studies at water purification plants. There, the usage of a closed cup surrounding the detector makes it more less affected by for example high humidity or dust (Stietka et al., 2013).
There are several types of continuous radon gas monitors. The AlphaGUARD consists of a so called, pulse-counting ionization chamber (SAPHYMO, 2015). This means that it measures and analyzes the alpha decay that occur in the air and that is caused by the occurrence of radon. The air either diffuses or is pumped into the chamber where the radon level is analyzed at certain time intervals. The radon progeny is not analyzed as it is separated from the radon by a filter (Hagberg et al., 2008). Moreover, the equipment is appropriate for places with high humidity (SAPHYMO, 2015).
The time period for the measurements is important to consider. For work places, orienting measurements with ATDs can be performed during two months and generates an annual average. Also so called follow-up measurements can be performed using continuous radon gas monitoring. Follow-up measurements are often carried out during a shorter time period. For work places, both orientating and follow-up measurements are recommended to be measured during the so called heating period (Hagberg et al., 2008).
The heating period takes place from the beginning of October to the end of April. It is recommended due to that the radon level in the indoor air often is lower and experience more variations in the summer (Arvela et al., 2016). The heating period is considered finalized by April 30th for all
measurements at work places. During this period, it is assumed that the temperatures fall below 10 degrees Celsius on a daily average. Hence, the natural ventilation is at a maximum as a result of the temperature difference between the indoor and outdoor air. In the summer months the natural ventilation decrease, which affect the inflow of radon into the building and air flow within the building. All buildings are affected by this process, as it occur both in buildings with or without mechanical ventilation (Hagberg et al., 2008). Moreover, the indoor radon levels can also be affected by different behavior patterns. For instance, it is more likely that the windows are open in the summer (Arvela, 2005).
Water analysis
There are two methods for analyzing the radon concentration in water samples described by the Swedish Radiation Safety Authority. These are gamma-ray spectrometry and liquid scintillation counting (Strålsäkerhetsmyndigheten, 2013). The first method detects the gamma-rays emitted by two of the radon progeny (lead-214 and bismut-214). By doing so, it is possible to determine the concentration of radon. The second method, liquid scintillation counting, uses the light that is emitted when alpha particles interact with the scintillation liquid to determine the radon levels in water samples. These alpha particles are emitted by both radon and its progeny (polonium-218 and polonium-214) (Talha et al., 2008). Therefore, it is possible to measure radon when it is at equilibrium with its progeny (Cassette et al., 2006).
Methods
There methodology consists of several steps: selection and contact with the water suppliers, collection of water samples and analysis, measurement of radon in indoor air and evaluation of data including statistical analysis.
Selection and contact
Selection of water purification plants was based on data from the SGU’s Water Archive and included small-scale water suppliers who use groundwater as a source of drinking water. A hundred water purification facilities were contacted in total. A letter with information about the study, a reply form and contact information to the ones responsible for the study were sent to the water purification facilities (Appendix J). Additional information (via email and phone calls) was provided to those who were willing to participate in the study. In some cases, the water purification facilities were not used anymore, no answers were received or they simply did not wish to participate. A total of 43 water purification facilities decided to participate.
However, some samples and detectors were not received by the laboratory, which caused some plants to be excluded from the study. In total, four plants have been excluded due to the lack of results. There were no results from the long-term measurements in the indoor air for two of these water purification plants and there were no samples of the untreated water from the other two plants. Follow-up measures are needed to conclude the reason that these samples were not received by the laboratory, but it was not possible within the scope of this study. Three other water purification plants were not excluded although there were no results from the treated water samples for these plants. They were included in the study, but not included in the analyses that required this particular sample. For instance, they were not included when the difference between the radon concentration in the untreated and treated water samples was determined.
Figure 2: The location of the water purification facilities that participated in this study. The figure is based on data from Lantmäteriet; their 10 million Sweden map (Lantmäteriet, 2016a) and their overview map (Lantmäteriet, 2016b).
Collection and analysis of the data
The data collection included collection of water samples for analysis of radon. Long-term and short-term measurements of radon levels in the indoor air were also performed. Collection of water samples and long-term measurements of radon in indoor air were performed at all of the plants. The short-term measurements which were based on time-resolved measurement of the radon level in the indoor air were carried out only at two of the plants (W16 and W31).
Water samples and measurement of radon in water
Figure 3: The risk for high radon concentration in groundwater in Sweden and the location of the water supplies in this study. The map is from SGU based on data about the uranium concentration in the bedrock and data from radon samples of well water in Sweden (Strålsäkerhetsmyndigheten, 2009).
Most of the water samples were analyzed at Landauer Nordic’s laboratory. Landauer Nordic is accredited for using gamma-ray spectroscopy to analyze the radon content in water samples (SWEDAC, 2015). This analysis is based on the method description by the Swedish Radiation Safety Authority (Strålsäkerhetsmyndigheten, 2013).
A few of the water samples were analyzed by ALcontrol laboratories, which were the standard laboratory used by those water purification plants. They use liquid scintillation instead of gamma-ray spectrometry to analyze the radon concentrations in water samples (ALcontrol Laboratories, 2016). The methodology follows a method description by the Swedish Radiation Safety Authority (Suomela, 1993).
Long-term air measurements of radon
Long-term measurements of the radon levels in the indoor air were performed at all the plants. Alpha Track Detectors (ATDs) of the type Radtrak2 were used for the long-term measurements (Fig. 4). They
have the detector material CR-39 and closed detectors with filters (Landauer Nordic, n.d.).
Two ATDs were used at each plant for long-term measurement of the radon level in indoor air. The measurement period was two months during the heating period, which range from the 1st of October
Figure 4: The appearance of an alpha track detector (Landauer Nordic, n.d.).
The ATDs were placed in areas of the plants where elevated levels of radon are expected and where some categories of workers can potentially be exposed. For most of the plants, all the treatment steps took place in one room and there was no additional working area that the staff visited. Therefore, the detectors were placed in this area.
At three of the water purification plants an additional ATD was also placed on the premise. Two of the ATDs were placed in the room where the treatment process occurred. The additional detector was placed in rooms in the vicinity of the treatment room. For these plants, there was an increased risk for radon exposure. W10 had previously measured radon levels exceeding the guidelines. At W9, the contact person stated that there were very high radon concentrations in the untreated water and there was a risk that the concentrations had increased. The water purification plant W39 was located in the basement of a public facility that was frequently visited. Moreover, the laboratory provided instructions about the placement of the ATDs (Landauer Nordic, 2013a). All the detectors were analyzed at Landauer Nordic’s laboratory.
Short-term air measurements of radon
At two water purification plants short-term measurements of radon were performed to study how radon levels vary daily. The radon levels were measured every hour for one week (04-25 to 2016-05-02) by two AlphaGUARDs (PQ2000 Pro and PQ2000), one at each plant (Fig. 5). The manufacturer is SAPHYMO, previously Genitrion Instruments (SAPHYMO, 2015).
AlphaGUARD can be used for continuous measurement of radon levels and can measure from a range of 2 Bq/m3 to 2 000 000 Bq/m3. Moreover, the equipment simultaneously measures other
Figure 5: One of the AlphaGUARDs used in the short-term measurement of radon levels in the indoor air.
Exposure survey
A survey was performed in order to evaluate the annual exposure (Appendix I). The participants were asked about the professions of those who visited the plants, the frequency of their visits and the length of time they spent at the plants. This enabled a quantification of the time the workers spent at these facilities. These estimations were combined with the representative radon levels in the indoor air to estimate the annual exposure. The representative radon level in the indoor air was those radon levels detected in the rooms where the treatment occurred.
Moreover, the participants were asked about the ventilation at these plants (Appendix H). The survey took places in form of a structured interview as the questions and their order were defined in advance (Saunders et al., 2012). The intention was to ask questions that were short and straightforward, in order to eliminate any ambiguity.
The survey was in most cases performed by phone calls to the participants. Some water purification plants had the same contact person. Therefore, a total of 20 persons were interviewed for the exposure survey. The questionnaire was followed in order to eliminate external influence in their answered. The questions were so called category questions (appendix I) that were filled in during the interview (Saunders et al., 2012). There were two exceptions. In one case, a participant preferred to receive the questionnaire by email and subsequently answered the questions by sending the survey back (W10). Another contact person also wanted to see the questions in advance although that person subsequently answered the question by phone.
In some cases, there were several sub-categories for the different professions. For instance, one facility was visited by three different types of technicians and one type of engineers who all spent different amount of times at the facility. For these cases, the exposure was estimated for the sub-category that spent most hours at plant.
Statistical analysis
Also the relationships between different parameters were investigated. For example, it was investigated if there was a relationship between the depth of the wells and the measured radon concentrations in the water samples. This was done by the use of a linear regression model for bivariate datasets. This means that for two datasets, a line was fitted between the dependent and independent variable (Retherford and Choe, 1993).
The fit of the two datasets to the line determines if there is a relationship and how apparent it is. Two measures used to determine the fit of the datasets were the correlation coefficient and the coefficient of determination. The correlation coefficient ranges from 1 to -1, where zero indicates that the two datasets are uncorrelated and 1 or -1 indicates perfect correlation but that it is positive or negative, respectively. The correlation coefficient, r, (Eq. 1) is calculated from the two datasets and the mean of these, 𝑋̅ and 𝑌̅ (Retherford and Choe, 1993).
𝐶𝑜𝑟𝑟𝑒𝑙𝑎𝑡𝑖𝑜𝑛 𝑐𝑜𝑒𝑓𝑓𝑖𝑐𝑖𝑒𝑛𝑡 = ∑𝑛𝑖=1(𝑋𝑖−𝑋̅)(𝑌𝑖−𝑌̅)
√∑𝑛𝑖=1(𝑋𝑖−𝑋̅)∑𝑛𝑖=1(𝑌𝑖−𝑌̅)
(Eq. 1)
The coefficient of determination describes the how much of the variation in the dependent variable that can be explained by the other variable. For the case of linear regression, the coefficient of determination (R2) is the square of the correlation coefficient (r2) (Barrett, 2000). For the coefficient
of determination, certain intervals are can be used to estimate the relationship between variables. To evaluate the coefficient of determination in the water samples, the same intervals used by Ek et al., (2008) are applied.
Ethics
Several ethics aspects mentioned by Hartman (2003) were considered in the study. Transparent information was given to the participants; i.e. they received written information about the purpose of the study, the names of those responsible and the connection to different institutions (the Royal Institute of technology, the Swedish Radiation Safety Authority and the Swedish Geological Survey). Another aspect mentioned is the voluntarily principle, which was applied both for the participating in the study and for the subsequent short-term measurements that took place.
Also the demand of confidential management of the results were implemented (Saunders et al., 2012). This means that the data is treated in a way that eliminates any possibility for unauthorized persons to get access of the data (Hartman, 2003). Moreover, the results are managed confidentially and are anonymously stated in the report. I.e. no result can be directly related to a specific plant. The names of any participating plant or contact person stated are not stated. Instead, the plants are referred to as W1 – W39. These numbers are randomly assigned to the plants in order to guarantee the anonymity of the participating water purification plants.
Results
Volume of water treated and radon concentrations in the water samples
Generally, most of the water purification plants included in this study treated only a small volume of water daily (Fig. 6). Almost all the plants treat less than 100 m3/day and approximately half treat less
than 10 m3/day (Appendix A). There was a great variation in the radon concentrations for untreated
groundwater at different water purification plants. In 13 % of the 39 plants the radon concentrations in the groundwater exceeded 1000 Bq/l with a maximum concentration of 3585 Bq/l (Appendix B). Most of the plants had radon concentrations between 100 Bq/l and 1000 Bq/l (Fig. 7) and one plant had a radon concentration below the detection limit of 20 Bq/l (Appendix B).
The radon concentrations in the treated water were lower than in the untreated water. For most of the plants the concentrations were less than 100 Bq/l (Fig. 8). A few of the samples had higher concentrations, with a maximum of 170 Bq/l (Appendix B). The decrease in radon concentration during treatment varied between different plants and at most reached 99 % of the initial concentration in the untreated water. At one plant there was no decrease between the radon concentration in the untreated and treated water samples (W39). In this case, the radon concentration was 80 Bq/m3 in both samples.
Figure 6: The daily average volume of water treated by each of the participating water purification plants.
Figure 8: The radon concentrations in the samples from the treated water at the water purification plants.
Results from the long-term measurements of radon in the indoor air
The radon levels in the indoor air varied significantly between different water purification plants according to the long-term air measurements. It ranged from 30 Bq/m3 to 10080 Bq/m3 in the
treatment room (Appendix C). An even higher radon level was detected in basement at one of the plants where it reached 29 kBq/m3. Most of the plants had radon levels in the indoor air that exceeded
200 Bq/m3 and several of the plants have values that exceeded 1000 Bq/m3 (Fig. 9).
There were some variations in the radon levels measured by different detectors at the same plants. In some cases, there were no major variations and the variations were within the stated uncertainty intervals (Appendix C). In other cases, there were more significant variations. This was the case for the detectors placed in the basement of two of the plants, W9 and W33. These detectors measured significantly high radon levels in the air compared to the detectors located at other places in the plants.
Results from the short-term measurements of radon in the indoor air
The results from the short-term measurement showed variations in the radon levels of the indoor air at the two water purification plants (Fig. 10 and 11). For W16, the radon level in the indoor air exceeded 200 Bq/m3 most of the time. The exception was in the beginning of the measurements when
it was below 200 Bq/m3. During the measurement period the radon level in the indoor air varied
significantly and reached a maximum of 3800 Bq/m3 at one point. The measurements had a median
of 1740 Bq/m3 and slightly lower arithmetic and geometric means (Appendix D).
Also for W31 the lowest radon levels occurred in the beginning of the measurements. At all other times the radon levels in the indoor air exceed 200 Bq/m3. In similarity with W16, the radon level
varied significantly although the level generally was lower compared to the other plant. For this plant, W31, the maximum measured radon level was 1528 Bq/m3. The median was 814 Bq/m3 and the
arithmetic and geometric means were slightly lower (Appendix E).
During the placement of the instrument for the short-time measurement, the workers mentioned that these plants did not produce water at a constant rate. Instead they produced water when the water level in the reservoirs fell below a certain level. For W16, the plant produced water about 3 times every day while W31 produced water approximately 5-6 times every day. For the ventilation, neither of the plants had mechanical ventilation. Only natural ventilation occurred at the plants (Appendix C). W16 had treatment specifically for radon removal unlike W31 that had no specific treatment for radon removal (Appendix C).
0 500 1000 1500 2000 2500 3000 3500 4000 4500 2 0 1 6 -0 4 -2 5 0 0 :0 0 2 0 1 6 -0 4 -2 6 0 0 :0 0 2 0 1 6 -0 4 -2 7 0 0 :0 0 2 0 1 6 -0 4 -2 8 0 0 :0 0 2 0 1 6 -0 4 -2 9 0 0 :0 0 2 0 1 6 -0 4 -3 0 0 0 :0 0 2 0 1 6 -0 5 -0 1 0 0 :0 0 2 0 1 6 -0 5 -0 2 0 0 :0 0 2 0 1 6 -0 5 -0 3 0 0 :0 0 Radon
(Bq/m3)
Radon levels in the indoor air
Figure 10: The measured radon gas levels in the indoor air at the water purification plant W16 during the week the measurement took place.
Figure 11: The measured radon gas levels in the indoor air at the water purification plant W31 during the week the measurement took place.
Estimations of the time spent at each water purification plant
Some members of the staff spent more time at the plants than others. The plants were visited most frequently and for the greatest length of time by technicians/engineers. However, the time spent at each plant is relatively limited. None of the technicians/engineers spent 1000 hours or more at a plant and at 84 % of the 39 water purification plants the technicians/engineers spent less than 100 hours annually.
The estimated time spent in the facilities each year varied significantly between different plants, from one hour to several hundred hours (Fig. 12). The variations were related to the frequency, routines and purposes of the visits. In some cases, a visit could take 10 minutes while in other cases the visits could take more than an hour. Moreover, some facilities were only visited twice a month while others were visited several times each week. There were some differences in routines as some water suppliers had rotating schedules for the technicians/engineers. Although some plants were visited on a weekly basis the rotating schedule could imply that one individual only visited that specific plant a few times each year. At other plants the same individual visited a plant most of the time, with the exception of vacation or sickness and was estimated to account for 90 % of the total visits.
In contrast, the time spent at the plants by workers in other professions was very limited in most cases. Two plants were visited by cleaners who annually spent twelve hours and one hour at each plant, respectively. Only one plant was visited by a janitor for approximately one hour annually. The time spent at the facility for craftsmen was very dependent on the need of the facility. This affected both the frequency and the duration of their visits. Most of the plants were visited by craftsmen between one and four times a year. Some plants stated that the visits were even more infrequent. Also the time could vary depending on the work needed; from a few hours to a few days. Furthermore, it was not necessarily the same person who visited a plant at each visit. In total, the time that craftsmen were estimated to visit each plant was estimated to range between zero to 24 hours annually. Moreover, a municipal monitoring unit visited the plants once a year or less frequent. Also laboratory workers and bosses could visit the plants. However, they only spent a small amount of time at the plants and less time than the technicians/engineers (Appendix F).
0 200 400 600 800 1000 1200 1400 1600 1800 2 0 1 6 -0 4 -2 5 0 0 :0 0 2 0 1 6 -0 4 -2 6 0 0 :0 0 2 0 1 6 -0 4 -2 7 0 0 :0 0 2 0 1 6 -0 4 -2 8 0 0 :0 0 2 0 1 6 -0 4 -2 9 0 0 :0 0 2 0 1 6 -0 4 -3 0 0 0 :0 0 2 0 1 6 -0 5 -0 1 0 0 :0 0 2 0 1 6 -0 5 -0 2 0 0 :0 0 2 0 1 6 -0 5 -0 3 0 0 :0 0 Radon
Figure 12: The estimated time spent at each water purification plants by the technicians/engineers annually.
Estimations of the exposure to radon
For most of the cases, the estimated exposure for the technicians/engineers was lower than the guideline of 0.36 MBq h/year (Fig. 13). An exception is W35 where the exposure is estimated to be more than this guideline. There are also relatively high exposures at other plants as the exposure is estimated to exceed 0.1 MBq h/year at seven plants (Appendix G). The exposure is estimated from the time the engineers/technicians spend at the plant each year and the radon levels registered during the long-term air measurements. The values are representative for the rooms where the treatments occur are used. For instance, very high radon levels were detected in the basements of two of the plants. Thus, additional time spent in those basements could lead to more exposure. Moreover, variations from the routine could cause more exposure. At W37 a renovation was planned that would take approximately 180 hours and be performed by a technician at the facility, which could lead to more exposure for that individual.
Moreover, all the technicians/engineers who visited the water purification plants in this study also visited other water purification plants during their work hours. Thus, the total exposure to radon during the work day can be more than these estimated exposures.
Relation between parameters
Possible relationships between parameters were investigated (Table 1). According to the coefficient of determination many of the investigated relations were weak, ranging from very weak to relatively weak. Visually, most of the graphs do not indicate any form of relationship either (Fig. 14, 15 and 16). For instance, there was no apparent relation between the radon concentration in the water and the radon level in the indoor air (Fig. 14). Moreover, there was not an apparent relation between the treated volume of water and the radon level in the indoor air (Fig. 15). The exception was for the relationship between the radon decrease during treatment and the radon concentration in the incoming water. The graph indicated that there could be a logarithmic relationship between these two datasets (Fig. 16).
For the short-term measurements, the relationship between the indoor radon levels and the other measured parameters were weak in most cases. The relationship between the radon levels in the indoor air and the other parameters at W16 was very weak or weak in most cases. The coefficient of determination could at most be classified as relatively weak (Table 2). For W31, the coefficient could be considered as moderate in one case, between the indoor relative humidity and the radon levels in the indoor air. For all other cases, the coefficient indicated weak relationships; ranging from very weak to relatively weak (Table 3).
Moreover, the graphs at the two plants W16 and W31 showed no obvious similarity. Their peaks and dips did not occur at the same occasions (Fig. 17) and there were no apparent links between the graphs and different weekdays or time during the day. Moreover, there was no apparent connection between the treatment processes (that occurred approximately three times a day for W16 and five times a day at W31) and the peaks.
Table 1: Coefficient of determination for different datasets.
Coefficient of determination R2
Very weak Weak Relatively
weak Moderate < 0.1 0.1 - 0.2 0.2 - 0.3 0.3 - 0.4 Radon concentration in
untreated water (Bq/l) and in treated water (Bq/l)
0,019 Radon concentration in
untreated water (Bq/l) and decrease of radon concentration during treatment (%)
0,27
Depth of the well (m) and radon concentration in the untreated water (Bq/l)
0,065 Radon concentration in
untreated water (Bq/l) and radon level in the air (Bq/m3)
0,012 Volume of treated water each
day and (m3/day) and radon
concentration in untreated water samples (Bq/l)
Figure 14: Relation between the radon concentration in the untreated water and the radon level in the indoor air of the treatment rooms.
Figure 15: The relation between the quantity of water produced and the radon concentration in the untreated water. 0 2000 4000 6000 8000 10000 12000 0 500 1000 1500 2000 2500 3000 3500 4000 Rad o n l e v e l in th e i n d o o r a ir (Bq /m 3 )
Radon concentration in the untreated water (Bq/l)
Relation between the radon level in the indoor air and radon
concentration in the untreated water
0 2000 4000 6000 8000 10000 12000 0 50 100 150 200 250 300 350 400 Rad o n l e v e l in th e i n d o o r a ir (Bq /m 3 )
Volume of water treated (m3/day)
Figure 16: The decrease in radon concentrations in the water versus the radon concentration in the untreated water. The samples that were below the detection limit (less than 20 Bq/l) are not included.
Figure 17: The measured radon levels in the indoor air at the water purification plants W16 and W31. 0 10 20 30 40 50 60 70 80 90 100 0 500 1000 1500 2000 2500 3000 3500 4000 Dec re a s e o f ra d o n c o n c e n tra ti o n d u ri n g tre a tm e n t (% )
Radon concentration in the raw water (Bq/l)
Decrease in radon concentration during treatment
0 500 1000 1500 2000 2500 3000 3500 4000 4500 Radon
(Bq/m3)
Radon levels in the indoor air
in W18 and W33
Table 2: The coefficient of determination between the measured radon levels in the indoor air and other parameters at W16. The parameters include both those that were measured indoors and those that have been derived from SMHI (SMHI, 2016).
Coefficient of determination R2
Very weak Weak Relatively weak Moderate Comparison between the
radon level measured in the indoor air and:
< 0.1 0.1 - 0.2 0.2 - 0.3 0.3 - 0.4
The indoor temperature 0,0044 The indoor air pressure 0,0043 The indoor relative
humidity 0,056
The outdoor temperature 0,13 The outdoor wind direction 0,017
The outdoor wind speed 0,016
The outdoor humidity 0,21
The outdoor air pressure 0,0020 The difference between the
indoor and outdoor temperature
0,14 the difference between the
indoor and outdoor air pressure
0,030
Table 3: The coefficient of determination between the measured radon levels in the indoor air and other parameters at W31. The parameters include both those that were measured indoors and those that have been derived from SMHI (SMHI, 2016).
Coefficient of determination
Very weak Weak Relatively weak Moderate Comparison between the
radon level measured in the indoor air and:
< 0.1 0.1 - 0.2 0.2 - 0.3 0.3 - 0.4 The indoor temperature 0,16
The indoor air pressure 0,28
The indoor relative
humidity 0,39
The outdoor temperature 0,025 The outdoor wind
direction 0,15
The outdoor wind speed 0,00052 The outdoor humidity 0,0037
The outdoor air pressure 0,22
The difference between the indoor and outdoor temperature
0,012 the difference between the
indoor and outdoor air pressure
Estimation of radon transfer by application of transfer coefficients
Transfer coefficient can be used to estimate the radon transfer from water to the indoor air. Different coefficients have been stated: 4.9 * 10-3 (Radolić et al., 2005), 1.96 * 10-4 (Sainz et al., 2016) and 10-4
(Vinson, Campbell and Vengosh, 2008). The first two are for thermal baths and the last is applied for household consumption. The transfer coefficients give different indications of the contribution from water. Application of the transfer coefficient 4.9 * 10-3 generated results that exceeded the measured
indoor levels in several of the plants (Fig. 18). The other coefficients generated a contribution lower than the radon levels in the indoor air (Fig. 19). According to these two coefficients, the contribution from the water to the indoor air was significantly lower than the radon level in the indoor air (Appendix I).
Figure 18: Estimated transfer from radon dissolved in water into the indoor air for the transfer coefficient 4.9*10-3, for each of
the plants, and comparison to the measured radon levels in the indoor air.
Figure 19: Estimated transfer from the radon dissolved in water into the indoor air for the two transfer coefficients 10-4 and
1.96*10-4, for each of the water purification plants, and comparison to the measured radon levels in the indoor air.
0 2000 4000 6000 8000 10000 12000 0 5000 10000 15000 20000
Radon level in the indoor air (Bq/m3) Transfer from
water (Bq/m3)
Estimated transfer and radon levels
4,9E-03 Measured Rn level
0 2000 4000 6000 8000 10000 12000 0 100 200 300 400 500 600 700 800
Radon level in the indoor air (Bq/m3) Transfer from
water (Bq/m3)
Estimated transfer and radon levels
Discussion
Comparison of the results from the water samples to previous studies
The concentration of radon in the untreated water is relatively high compared to other studies. In this study, the median is 300 Bq/l for the incoming water. In another study, the median of samples collected from groundwater wells were 194 and 232 Bq/l, where 232 Bq/l was the result of a sampling process directed specifically to areas with high uranium content (Ek et al., 2008). In the Swedish Water Archive, groundwater has a median of 90 Bq/l (Thunholm and Whitlock, 2014). It could be related to the fact this study is not as comprehensive as the data in the Swedish Water Archive and the other study. Another potential explanation is that the water suppliers in radon prone areas were more interested in participating in this study.
In similarity with the data in the Swedish Water Archive, the radon concentration in the water decrease during treatment. In this study, the median radon concentration of the treated water is 60 Bq/l while their median were 40 Bq/l for treated groundwater (Thunholm and Whitlock, 2014).
The results from the study show that the radon concentrations in the treated water are generally lower than in the untreated water. At most of the water purification plants the radon concentration is lower in the treated water compared to the untreated water. However, the decrease in radon concentration can vary significantly. At several plants the decrease between the untreated and treated water is more than 90 % of the radon concentration in the untreated water. There are also other plants where the decrease in radon concentration between the untreated and treated water is less than 50 % of the radon concentration in the untreated water. In fact, at W39 the radon concentration of 80 Bq/l was detected in both the untreated and treated water. However, at most plants a decrease in the radon concentration can be noted (Appendix B).
Treatment processes could have an impact on the decrease in the measured radon concentration in the untreated water compared to the treated water. In several plants there are treatment directed specifically at radon removal. In the case of W35, with no decrease in radon concentration between the untreated and treated water, there is no treatment specifically for radon removal (Appendix C). However, radon removal can occur indirectly even if there is no specifically for radon removal. Radon can be removed as a result of treatment to remove other substances or as a result of the degassing during the mechanical transport of water. In fact, at most of the water purification plants with no specific treatment for radon removal there is a decrease in the radon concentration between the untreated and treated water.
