Outdoor ²²⁰Rn, ²²²Rn and terrestrial gamma radiation levels: investigation study in the thorium rich Fen Complex, Norway

Abstract

The present study was done in the Fen Complex, a Norwegian area rich in naturally occurring radionuclides, especially in thorium (²³²Th). Measurement of radioactivity levels was conducted at the decommissioned iron (Fe) and niobium (Nb) mining sites (TENORM) as well as at the undisturbed wooded sites (NORM), all open for free public access. The soil activity concentrations of ²³²Th (3280–8395 Bq kg⁻¹) were significantly higher than the world and the Norwegian average values and exceeded the Norwegian screening level (1000 Bq kg⁻¹) for radioactive waste, while radium (²²⁶Ra) was present at slightly elevated levels (89–171 Bq kg⁻¹). Terrestrial gamma dose rates were also elevated, ranging 2.6–4.4 μGy h⁻¹. Based on long-term surveys, the air concentrations of thoron (²²⁰Rn) and radon (²²²Rn) reached 1786 and 82 Bq m⁻³, respectively. Seasonal variation in the outdoor gamma dose rates and Rn concentrations was confirmed. Correlation analyses showed a linear relationship between air radiation levels and the abundance of ²³²Th in soil. The annual outdoor effective radiation doses for humans (occupancy 5 h day⁻¹) were estimated to be in the range of 3.0–7.7 mSv, comparable or higher than the total average (summarized indoor and outdoor) exposure dose for the Norwegian population (2.9 mSv year⁻¹). On the basis of all obtained results, this Norwegian area should be considered as enhanced natural radiation area (ENRA).

---

Environmental impact

Natural radiation, comprising cosmic and radiation from terrestrial radionuclides with half-lives comparable to the age of the earth, is ubiquitous. In enhanced natural radiation areas (ENRA), like the one investigated in the current study, the contribution of exposure to terrestrial gamma radiation and radon (^220,222Rn) levels could overcome other radiation exposures and pose a serious risk for people living there. The Fen Complex, in southern Norway, is an area significantly rich in thorium (Th) ore. Several assessment studies have previously shown the increased risk for humans correlated mainly with indoor exposure to radon (²²²Rn). However, the current study is the first one showing the outdoor levels of gamma radiation, ²²²Rn and thoron (²²⁰Rn), and their possible significant contribution to the total radiation exposure dose. The analyses of the magnitude of radiation parameters, their seasonal variation, and correlation with mother radionuclides in the soil (²³²Th and ²³⁸U) put additional light on the complex radiation exposure scenario in this area.

1. Introduction

Natural radiation, comprising cosmic and radiation from terrestrial radionuclides with half-lives comparable to the age of the earth, is ubiquitous. Specific radiation levels due to terrestrial background radiation are positively correlated with geological structure of the terrain and current geological conditions.¹ According to UNSCEAR,² exposure from naturally occurring radioactive materials gives the major contribution to the total effective radiation dose of the population. The exposure pathways include internal exposure, mainly due to the inhalation of radon (²²²Rn, half-life 3.8 d) and its progeny, and external exposure due to the gamma irradiation with radionuclides originated in primordial radionuclides of earth's crust (decay chains of thorium (²³²Th) and uranium (²³⁸U, ²³⁵U), as well as radionuclide potassium (⁴⁰K)).

The world average annual effective dose per capita is estimated to be 2.4 mSv of which 1.1 mSv results from inhalation of indoor ²²²Rn and its progeny.³ The indoor ²²²Rn levels, emanation processes, measurement methods and associated health risks have been heavily investigated.^4–7 In contrast, information on thoron (²²⁰Rn, half-life 56 s) is scarce, probably due to the short half-life, difficulties in measurement techniques and limited abundance in nature. Still, ²²⁰Rn has recently been recognized as a potential health hazard in worldwide areas rich in ²³²Th, and therefore, its determination and risk estimation are emphasized as being important.^3,8–10

The Fen Complex area, situated in the Norwegian county of Telemark, is well known for its specific bedrock of volcanic magmatic origin,^11,12 containing ²³²Th and ²³⁸U rich rock types (e.g., rödbergite, rauhaugite, sövite, fenite). Due to the elevated levels of a series of metals, mining of iron (Fe) ores and rare earth elements (e.g., niobium (Nb)) has been performed during the past centuries. Previous studies in the Fen Complex have demonstrated elevated levels of radionuclides within the mining areas, as well as elevated indoor radiation exposure and associated human risk.^13–16 The major health issue in terms of radiation exposure has been associated with the inhalation of ²²²Rn progeny.^14,17 The estimated annual doses in the Fen Complex area are reported up to four times higher than the Norwegian average effective dose of 2.9 mSv.¹³ However, the contribution of the outdoor exposure to the total radiation dose has not been previously determined.

Even though data on radiation doses due to outdoor occupancy are generally much less available compared to those due to indoor occupancy, such data are important for estimating human doses, especially in the areas of high natural radiation.³ With respect to that, the results from the present study could give a valuable insight into outdoor radiation exposure issues.

The objective of this study was to investigate the air concentration levels of ²²⁰Rn and ²²²Rn, as well as terrestrial gamma dose rates, at sites rich in naturally occurring radioactive materials (NORM) and technologically enhanced naturally occurring radioactive materials (TENORM). The sites are freely accessible to the public and significantly rich in ²³²Th and moderately in ²³⁸U. The seasonal variations and correlations between radiation variables were also evaluated. Based on the attained results, the possible annual effective radiation dose received by humans outdoor was estimated.

2. Materials and methods

2.1. Study area

The study area is located in a small village Fen, in south-eastern Norway, about 009° 18′ E and 59° 16′ N. The whole area is known as the Fen Complex and has unique bedrock geology of volcanic origin. Based on our previous work,¹⁸ two TENORM sites (Fen and Søve mining sites) and three NORM sites (Bolladalen, Gruvehaugen and Rullekoll) that had high terrestrial gamma dose rates and elevated rocks and soil ²³²Th and ²³⁸U activity concentrations were selected for the investigation of outdoor radiation levels (Fig. 1).

Fig. 1 Map showing the studied sites in the Fen Complex, Norway: Søve mining site (S), Fen mining site (F), Bolladalen (B), Gruvehaugen (G) and Rullekoll (R) (source of original map: Norwegian Mapping Authority).

The Søve site (TENORM) is an abandoned mining facility on the west part of the Fen Complex. The mining operation was based on limestone mineral sövite, present in abundance at this site. Sövite consists mainly (75–95%) of calcium carbonate and some minerals (e.g., pyrochlore, columbite, fersmite) rich in Nb (0.35% Nb₂O₅) and to a lesser degree radionuclides ²³²Th and ²³⁸U. The mining activities, related to ferro-niobium production, were conducted during 1953–1965. Large amounts of waste in the form of crushed rocks and slag, enriched with radioactive elements (²³²Th and ²³⁸U and their daughters), were left out in the area following the mining activities.^16,19 The site, in fact, was covered with sand layers in a remediation action conducted after the decommission. However, several points in this area have recently been investigated^16,18,20 revealing significantly inhomogeneous and elevated radionuclide concentrations in soil, as well as elevated terrestrial gamma radiation dose rates. Currently, a mechanical workshop is in operation at one part of this site.

Mining of Fe was conducted in the Fen Complex at several locations during 1650–1929. As a consequence, certain wooded zones are found to be with elevated radiation levels.¹⁸ Major Fen rock types, rauhaugite (magnesium–calcium–carbonate) and rödbergite (ferro-carbonate), contain little of Nb, but are enriched in ²³²Th, rare earth elements, and Fe. Rödbergite, in particular, is considered as a possible source for future Th mining and exploitation.¹⁶ The Fen site (TENORM) is situated in the north part of the Fen Complex, along the shores of Lake Norsjø where Fe mining waste rocks are still deposited.

Bolladalen and Gruvehaugen sites comprise the area in the central Fen Complex wooded zone. The investigated subsites within these sites were considered as NORM since no waste from the previous Fe mining was found there, although some mining holes and open tunnels were observed (limited and not accessible).

The Rullekoll site (NORM) is an undisturbed site located in the south of the Fen Complex, consisting of a small forest just above the human settlement area. Radionuclide ²³²Th and its progeny, significantly elevated in rödbergite rock found at this site, result in higher radiation levels in the soil and air.¹⁸

2.2 Study design

Five field expeditions (2008–2010) were organized in different seasons, allowing the investigation of the radiation parameters not only in stable spring–summer conditions (May–August 2008, 2009 and 2010), but also in rainy and snowy conditions (October and November 2009). Since four of the five investigated sites (F, B, G, R) (Fig. 1) were large and comprised of wooden areas, a certain number of subsites, within each of the main sites, were investigated in order to obtain more reliable data. The geographical positions of all sampling and measurement points within sites were recorded with the Global Positioning System (GPS, Garmin, USA). Additionally, the sites were photographed, enabling easy return in the following field expeditions.

To obtain all necessary data for this study, the following activities were conducted:

- Soil sampling for ex situgamma ray spectrometry,

- In situ measurements of gamma dose rates,

- Continuous measurements of ²²⁰Rn and ²²²Rn air concentrations (two seasonal, three months long surveys).

2.2.1. Soil sampling and ex situ radiometric analysis. Soil sampling was conducted simultaneously with the gamma radiation measurements and detector placement at GPS recorded sites. Samples were collected using the soil corer with a size diameter of 10 cm. At each site, soil corers were taken from at least 5 sampling points with a distance of 5 m between them, at depth up to a maximum of 20 cm. The abundance of rocks in soil layers was a limiting factor regarding the sampling depth. The soil was packed into the polyethylene bags and properly marked. In the laboratory, the collected soil was dried at 110 °C, crushed and sieved to less than 2 mm. The gamma spectrometry measurement of samples was done at the Norwegian Radiation Protection Authority (NRPA). The soil was prepared in cylindrical geometries, isolated in aluminium foil and kept for a month before measurement to ensure secular equilibrium. Measurements were carried out on three different coaxial p-type detectors: a Canberra GR2521-7500 with 45% relative efficiency and a FWHM of 1.9 keV; an Ortec GEM-40190-S with 40% relative efficiency and a FWHM of 1.8 keV and an Ortec GEM-33190-S with 33% relative efficiency and a FWHM of 1.8 keV. Samples were generally counted for one or two days depending on the activity level (minimum measuring time of 16 hours for the most active samples). Quantitative analysis of ⁴⁰K was based on primary photon emissions (1460 keV), while ²²⁶Ra was determined via²¹⁴Pb (352 keV) and ²¹⁴Bi (609 keV) and ²³²Thvia²²⁸Ac (911 keV). Spectra were obtained through an Ortec Maestro v6 and spectrum analysis was carried out using self-written Sampo-based software. The measurement uncertainties at the 1σ level were in the range 4–8, 1–6 and 5–9% for ²²⁶Ra, ²³²Th and ⁴⁰K, respectively. To ensure the quality of analysis, the control of detector efficiency calibration and background levels were in place, as well monitoring of energy calibration and laboratory climate.

2.2.2. Gamma dose rate measurements. In situ spatial surveys of gamma dose rates (μGy h⁻¹) were carried out by a portable gamma detector (Automess, Radiacmeter 6150 AD 4 LF) calibrated with ¹³⁷Cs source. The response range of the instrument was 0.01 μGy h⁻¹ to 9.99 mGy h⁻¹. Measurements were conducted at 1 m above the ground surface. A regular grid 10 × 10 m was used at the subsites, and readings were repeated until constant signal. The dose rate in the air at each of the sites was obtained as the arithmetic mean of all measurements at the subsites. The data collection was performed for different months (May, September, November and June), allowing us to identify the seasonal variation. Besides the studied sites, several readings at public places (road, vicinity of the houses, schoolyard) were also recorded, providing the background dataset of dose rates for comparison purposes.

2.2.3. Measurements of ²²⁰Rn and ²²²Rn air concentrations. Continuous measurements of ²²⁰Rn and ²²²Rn concentrations (Bq m⁻³) in the air were performed in two separate periods, during September–November 2009 and June–August 2010. Passive integrating ²²²Rn–²²⁰Rn discriminative detectors, type RADUET (RadoSys Co., Ltd., Budapest, Hungary), were used.²¹ The detectors, consisting of two different diffusion chambers, were developed and evaluated at the National Institute of Radiological Studies (NIRS), Chiba, Japan.²² These detectors have been widely used in surveys throughout the world.^9,23

In the current study, the detectors were fixed to the trees, lying 1 m above the ground surface. A total of 86 detectors were placed at five chosen sites, and 82 were recovered. After collecting the detectors, they were sent to the Japan Chemical Analysis Center, Chiba, for counting. The detailed performance has been described by Zhuo et al.²¹ and Tokonami et al.²⁴

2.3. Data analyses

Statistical analyses were performed using a Minitab 16 (Minitab Inc.). Normality was assessed with the Anderson–Darling test. The ²²⁰Rn and ²²²Rn concentrations were logarithmic transformed prior to statistical tests. Difference analyses of two measurement sets (²²⁰Rn and ²²²Rn) from different seasons (summer and fall) were done with the Student's t-test. One-way ANOVA, followed by Tukey's post hoc description, was used in difference analysis of ²²⁰Rn, ²²²Rn and gamma dose rates measured at the investigated sites, as well as in comparison of gamma dose rates measured in four different months. Correlation analyses were performed using the Pearson correlation coefficient. For all the analyses, the p-value lower than 0.05 is considered statistically significant.

In addition to the measurements, the gamma dose rates in the air were calculated on the basis of soil gamma spectrometry results, following the guidelines of UNSCEAR:²

D(nGy h⁻¹) = 0.042C_K + 0.604C_Th + 0.462C_Ra (1)

where D is the gamma dose rate; 0.042, 0.604 and 0.462 are conversion factors expressed in nSv h⁻¹ per activity unit and C_K, C_Th and C_Ra are soil activity concentrations (Bq kg⁻¹) of ⁴⁰K, ²³²Th and ²²⁶Ra, respectively. Contributions of other soil radionuclides, e.g., ⁹⁰Sr, ¹³⁷Cs and ²³⁵U, were considered as insignificant and were not taken into the calculation of the dose rates.²⁵ Calculated gamma dose rates were compared to the measurements obtained with the portable detectors using the Student's t-test.

The annual outdoor effective doses from external gamma radiation were determined using the following equation:²

H_{gamma rad.} (mSv) = D (nGy h⁻¹) × 8760 h × 0.2 × 0.7 Sv Gy⁻¹ × 10⁻⁶ (2)

where H is the effective dose, D is the measured gamma dose rate, 0.2 is the outdoor occupancy factor and 0.7 Sv Gy⁻¹ is the conversion coefficient from absorbed dose in air to human effective dose equivalent.

Effective doses from ²²⁰Rn and ²²²Rn outdoor were estimated as:²

H_i (mSv) = C_i × F_i × t × DCF_i × 10⁻⁶ i = (²²⁰Rn or ²²²Rn) (3)

where H_i is the effective dose, C_i is the ²²⁰Rn (²²²Rn) air concentration (Bq m⁻³), F_i is the equilibrium factor 0.003 for ²²⁰Rn (ref. 2 and 3) and 0.6 for ²²²Rn (ref. 2); t is the exposure time (1752 hours) and DCF is the dose conversion factor 40 nSv Bq⁻¹ equivalent equilibrium concentration (EEC) h m⁻³ for ²²⁰Rn and 9 nSv Bq⁻¹ EEC h m⁻³ for ²²²Rn.

Finally, total outdoor effective dose was obtained as sum:

H_tot = H_{gamma rad} + H_220Rn + H_222Rn (4)

3. Results and discussion

3.1. Soil analysis

The activity concentrations of the terrestrial radionuclides in soil samples from different Fen Complex sites are given in Table 1. The world average soil activity concentrations of ²²⁶Ra, ²³²Th and ⁴⁰K are estimated to be 32, 45 and 420 Bq kg⁻¹, respectively.^2,26 The average Norwegian soil values are somewhat higher, i.e., 50, 45 and 850 Bq kg⁻¹ for ²²⁶Ra, ²³²Th and ⁴⁰K, respectively.^2,26 The median values of ²³²Th soil concentrations in the present work (3280–8395 Bq kg⁻¹) were two orders of magnitude higher, exceeding the screening value (1000 Bq kg⁻¹) for radioactive materials given by the Norwegian Pollution Act.²⁷ Wide ranges, showing the difference in ²³²Th concentration between close sampling spots, suggested significantly inhomogeneous distribution at each of the investigated sites. This distribution could be considered as a consequence of bedrock weathering, formation of space localized radionuclides rich soil and existence of ²³²Th rich soil particles. No statistically significant difference in ²³²Th concentration was found regarding NORM and TENORM sites. The relationship analysis has previously shown the positive correlation (r = 0.78, p < 0.001) between gamma dose rates in the air and ²³²Th concentration in the soil.¹⁸

Table 1 Radionuclide activity concentration of the Fen Complex soil (median and range (in parenthesis) are given)

Site	N	^226Ra/Bq kg^−1	^232Th/Bq kg^−1	^40K/Bq kg^−1
Bolladalen	5	127 (66–234)	8395 (1740–15 500)	654 (496–921)
Fen mining site	5	122 (88–160)	3280 (2360–3940)	404 (366–429)
Gruvehaugen	8	110 (65–272)	8020 (5100–11 500)	500 (304–637)
Rullekoll	5	171 (38–376)	6655 (5560–9270)	439 (343–640)
Søve mining site	5	89 (77–101)	5650 (5130–6170)	551 (489–614)

---

Soil activity concentrations of ²²⁶Ra were only slightly enhanced (89–171 Bq kg⁻¹), which was expected since the mother radionuclide (²³⁸U) was present in rocks and soil to a significantly less degree than ²³²Th.^16,18 The lowest level was obtained at site Søve, but no significant differences between sites were confirmed. Similar results, regarding both ²³²Th and ²²⁶Ra, have been published for a high natural background radiation area (monazite and zircon rich) in India.²⁸

The concentration of ⁴⁰K (404–654 Bq kg⁻¹) was in the normal variation range for Norway given by UNSCEAR.^2,26

3.2. Gamma dose rates

Gamma dose rates, recorded with a portable detector at NORM and TENORM sites, were in the range of 2.6–4.4 μGy h⁻¹ (Table 2). Generally, all gamma dose rates were 30–80 times higher than the Norwegian average of 0.073 μGy h⁻¹ and the world average of 0.058 μGy h⁻¹ given by UNSCEAR,^2,26 as well as higher than those obtained for different reference sites worldwide.^29–31 However, results were comparable and confirmed the previous Fen Complex measurements obtained by Dahlgren,³² Heincke et al.,¹⁵ and IFE.²⁰ Significant differences (p = 0.0005) were found between some of the investigated sites. Although it indicated the site specific radioactivity enrichment, the comparison of soil activity concentrations at TENORM and NORM sites did not demonstrate any significant correlation between the enrichment and former mining activities. The highest mean gamma dose rate (4.4 μGy h⁻¹) was obtained at a TENORM site, i.e., at decommissioned Nb mining site Søve. However, the gamma dose rate obtained for NORM site Gruvehaugen had almost the same mean value (4.3 μGy h⁻¹), while another NORM site (Rullekoll) showed the lowest gamma dose rate (2.6 μGy h⁻¹). Furthermore, Fen, the former mining site (TENORM), showed similar dose rates (2.8 μGy h⁻¹) to those of the Rullekoll site (NORM). Based on these results, the variation of the gamma dose rates is more likely related to the variation in the abundance of the certain rock types at chosen sites than to the presence or absence of former mining activities at the investigated sites.

Table 2 Gamma dose rates directly measured and calculated from soil activity concentrations (mean ± standard deviation); percentage contribution of ²²⁶Ra, ²³²Th and ⁴⁰K to calculated gamma dose rates

Site	Measured gamma dose rate/μGy h^−1	Calculated gamma dose rate/μGy h^−1	^226Ra (%)	^232Th (%)	^40K (%)
Bolladalen	3.2 ± 0.8	5.2 ± 0.4	1.1	98.3	0.5
Fen mining site	2.8 ± 0.3	2.1 ± 0.4	2.8	96.1	1.0
Gruvehaugen	4.3 ± 1.8	4.9 ± 1.6	1.0	98.5	0.4
Rullekoll	2.6 ± 0.4	4.1 ± 1.1	1.9	97.6	0.4
Søve mining site	4.4 ± 2.8	3.5 ± 0.4	1.2	98.1	0.7

---

The gamma dose rates, recorded outside the forest area (within the Fen living areas), with a mean value of 0.18 μGy h⁻¹, were statistically lower (p = 0.0005) than those recorded at NORM and TENORM sites. Still, these public sites had also the slightly enhanced gamma radiation dose rates in comparison to the world average.^2,26

In addition to measurements, a theoretical approach of gamma dose rate estimation, based on soil activity concentrations of radionuclides (eqn (1)), was used and obtained results are shown in Table 2.

Computed gamma dose rates in the air due to naturally occurring radionuclides varied in the range of 2.1–5.2 μGy h⁻¹. The highest radionuclide contribution (>96%) to the outdoor gamma dose rates was from radionuclide ²³²Th and its short lived gamma progeny, at all investigated sites. The comparison of gamma dose rates estimated from soil radionuclide activity concentrations and gamma dose rates directly measured showed no significant differences (p = 0.4).

Observed seasonal change of measured gamma dose rates is shown in Fig. 2. Statistically significant differences (p < 0.001, for each of the investigated sites) between gamma dose rate values in different months confirmed the seasonal variation at each of the investigated sites. The maximal mean value was obtained in early September (2009) at TENORM site Søve, showing the considerable difference (p < 0.05) from values obtained at other locations (Fig. 2). High readings at this site would give the annual absorbed gamma dose in the air up to 38.5 mGy (exposure time 8760 h). These considerably high but spatially limited gamma radiation measurements at Søve corresponded to inhomogeneous distribution of soil radioactivity, previously reported at the same location.¹⁶ Gamma dose rates, similar to present findings, have been previously reported for the Søve site by IFE²⁰ and NGI.¹⁶ For three of the four investigated sites, the mean value of gamma dose rate was highest in September, while lowest in November. The dry weather with no wind, recorded during the expedition in early September, and in contrast, strong wind, rain and snow cover of approximately 10 cm at the end of November, could provide the explanation for obtained readings. However, it is emphasized that no common seasonal variation pattern applicable to all sites was observed. The detailed recording of all atmospheric and weather conditions and other factors (e.g., soil humidity and permeability) is essential to obtain a much accurate explanation of variation in gamma dose rates.

Fig. 2 Monthly variation of gamma dose rates in the Fen Complex (mean value ± standard error); site Rullekoll is not shown because of high measurement uncertainty.

3.3. Continuous measurements of ²²⁰Rn and ²²²Rn air concentrations

The values for ²²⁰Rn and ²²²Rn concentrations in the air, obtained in summer and fall surveys, are given in Tables 3 and 4 respectively.

Table 3 Air ²²⁰Rn concentrations in the Fen Complex

Site	Summer survey				Fall survey
	Arithmetic mean/Bq m^−3	Standard deviation/Bq m^−3	Geometric mean/Bq m^−3	Range/Bq m^−3	Arithmetic mean/Bq m^−3	Standard deviation/Bq m^−3	Geometric mean/Bq m^−3	Range/Bq m^−3
a One detector recovered at site Søve.
Bolladalen	1294	863	1029	284–2801	743	320	707	516–969
Fen mining site	1442	1155	1230	765–4996	445	259	406	262–628
Gruvehaugen	1786	860	1579	495–3495	741	359	668	339–1029
Rullekoll	1231	339	1189	709–2047	1000	138	994	896–1156
Søve mining site	91	90	67	24–362	7^a	—	7^a	—

---

Table 4 Air ²²²Rn concentrations in the Fen Complex

Site	Summer survey				Fall survey
	Arithmetic mean/Bq m^−3	Standard deviation/Bq m^−3	Geometric mean/Bq m^−3	Range/Bq m^−3	Arithmetic mean/Bq m^−3	Standard deviation/Bq m^−3	Geometric mean/Bq m^−3	Range/Bq m^−3
a Only one detector showed value > LOD. b One detector recovered at site Søve.
Bolladalen	73	40	59	8–157	9^a	—	—	—
Fen mining site	82	56	64	9–210	<LOD	—	—	—
Gruvehaugen	61	31	53	21–120	5^a	—	—	—
Rullekoll	47	21	41	9–72	8	3	7	4–11
Søve mining site	29	5	28	24–37	13^b	—	—	—

---

UNSCEAR ^2,3 suggests the typical outdoor concentration of ²²⁰Rn is of the order 10 Bq m⁻³, with the range from 1 to 100 Bq m⁻³. The air ²²⁰Rn concentrations from the summer continuous survey were found to be enhanced in comparison to the average world value, with the arithmetic and geometric mean in the range of 91–1786 and 67–1579 Bq m⁻³, respectively. The fall survey demonstrated values with the arithmetic and geometric mean in the range of 7–1000 and 7–994 Bq m⁻³, respectively.

Summer survey ranges were wider at each of the investigated sites than those recorded in fall, indicating the greater differences between subsites. However, these differences could be due to a higher number of detectors placed in the summer survey, giving more representative information, than due to the actual decrease in variation between subsites in the fall period. The demonstrated variation corresponded also to the significant inhomogeneous distribution of ²³²Th in the soil of same subsites and suggested strong dependence of air levels on the soil concentration.

The decrease in the values of the fall ²²⁰Rn concentration was noticed at all sites in comparison with summer values. However, except for the Søve site, the concentration values were still higher than the world average value. Sparse publications on outdoor ²²⁰Rn have shown much lower air concentration in comparison to this study.^33,34

Difference analysis showed that the ²²⁰Rn air concentration at mining site Søve was significantly lower (p = 0.0005) than ²²⁰Rn air concentrations at other investigated sites in the summer survey. The lack of detectors at Søve in the fall survey did not allow the statistical analysis of difference, but the lowest value in the fall survey was also recorded for the Søve site. The possible explanation is the sand covering placed on the soil surface of the site to reduce the public exposure. No other significant differences in ²²⁰Rn air concentrations between NORM and TENORM sites were obtained, suggesting no ²²⁰Rn air enrichment in terms of former mining.

The presence of much less ²²²Rn than ²²⁰Rn in the air in both seasons was demonstrated, reflecting again the bedrock and soil abundance of ²³²Th and moderate levels of ²³⁸U. The values of outdoor ²²²Rn in the summer period (Table 4) were in the range of 8–210 Bq m⁻³, with the arithmetic and geometric mean in the range 29–82 Bq m⁻³ and 28–64 Bq m⁻³, respectively. These measurements were in agreement with the wide outdoor background range (1–100 Bq m⁻³) given by UNSCEAR,^2,3 but still higher than the world average^2,3 of 10 Bq m⁻³ and higher than results published in the similar studies worldwide.^35–37 A significant decrease of ²²²Rn in the air was observed in the fall survey at all sites, giving non-measurable values at the majority of exposed detectors. No statistical differences were found in the ²²²Rn air concentration between sites in the summer survey.

Seasonal (summer and fall) variation in ²²⁰Rn and ²²²Rn air concentrations is presented in Fig. 3. Student's t-test showed significant seasonal difference for both ²²⁰Rn (p = 0.0039) and ²²²Rn (p = 0.0054). In the present study, the summer values of both ²²⁰Rn and ²²²Rn were higher than those obtained in fall, in contrast to their characteristic ‘high in winter and low in summer’ behaviour.^36,37 It seems that the Scandinavian weather conditions could be the reasons for these results. Decreased values of investigated variables in the fall could be explained with the low emanation and exhalation processes due to the significantly increased moisture content, snow coverage and frozen soil in fall months of 2009 (late October and November). The influence of soil humidity and snow coverage, on air ²²⁰Rn and ²²²Rn levels, has been previously studied and they have been suggested as the diminishing factors for ²²⁰Rn and ²²²Rn emanation.^38,39 Additionally, more wind, rain and snow conditions in the air might act as the removal processes that contributed to lower diffusion and concentration of radon gases.

Fig. 3 Seasonal variation of ²²⁰Rn (left) and ²²²Rn (right) air concentrations. Box plot: the horizontal lines show the median; crossed circles average values; the top and bottom of the box show the 75^th and 25^th percentiles. The top and bottom of the whiskers show the maximum and minimum values.

Seasonal behaviour similar to behaviour obtained in the current study has previously been published.^6,40 For a better understanding of these seasonal differences, a detailed knowledge on precipitation levels, moving of air masses, temperature and atmospheric pressure, presence of clouds, humidity of air and soil moisture is needed.

3.4. Correlation analyses

Based on the present results, a scatter diagram, examining the correlation between ²²⁰Rn and ²²²Rn concentrations in the air, is presented in Fig. 4. The regression equation with a coefficient of R² = 0.49, an intercept of 462 and a slope of 16 was obtained (p = 0.0005). The correlation coefficient found in this study (0.7) is higher than previously published (0.009–0.42).^7,9,41 The obtained y-axis intercept (462) and ratio (16) between ²²⁰Rn and ²²²Rn air concentrations suggested highly relevant prevalence of ²²⁰Rn. It was expected since a significant soil enrichment of ²³²Th has been previously reported¹⁸ and confirmed in the current study.

Fig. 4 Correlation between ²²⁰Rn and ²²²Rn air concentrations.

Correlation analysis demonstrated a moderate positive relationship between gamma dose rates and ²²⁰Rn air concentrations (r = 0.56, p = 0.001), as well between gamma dose rates and ²²²Rn air concentrations (r = 0.64, p = 0.001). The correlations, similar to those we obtained, have been published elsewhere.^14,36

Furthermore, a positive linear correlation between air ²²⁰Rn concentrations and soil ²³²Th activity concentrations was observed (Fig. 5).

Fig. 5 Correlation between ²²⁰Rn concentrations in air and the corresponding soil ²³²Th activity concentrations.

The Pearson correlation coefficient of 0.66 (p = 0.001) suggested the dependence of air ²²⁰Rn on geological composition of terrain. However, the value of slope (0.14) implicated limited ²²⁰Rn in the air in comparison to what might be expected from ²³²Th concentrations in soil. The spatial variation and possible significant decrease of air ²²⁰Rn with distance from the soil are the most reasonable explanation. In addition, the emanation of ²²⁰Rn and its diffusion through soil, which is highly dependent on soil conditions, might also be limited and hence affect the results.

3.5. Effective doses related to outdoor exposure to radiation

The inhalation of ²²⁰Rn, ²²²Rn and consequently the deposition of their stable progenies inside the human bodies together with external irradiation with gamma emitting radionuclides contribute significantly to the exposure dose of humans.

The total annual effective doses were calculated in the present study using the following relatively rough assumptions:

- Summer radiation dataset (²²⁰Rn, ²²²Rn and gamma dose rates) was considered as constant for the period of six months with relatively stable atmospheric conditions. Therefore, they were used in calculation of effective doses for a half-year exposure period (eqn (2) and (3)).

- Fall radiation dataset (²²⁰Rn, ²²²Rn and gamma dose rates) was considered as constant for the period of six months with unstable conditions (snow, rain, wind). Effective doses were derived from it for the other half-year period (eqn (2) and (3)).

Total annual effective doses were then obtained by summarizing the effective doses from two periods (eqn (4)). Although the calculations with these assumptions do not give completely accurate values of doses, they are useful as they provide the upper limits of possible doses and allow preliminary risk estimation.

Total exposure doses to the population due to outdoor exposures and comparison with the ICRP⁴² annual constraint value of 1 mSv are given in Fig. 6. The estimated total annual outdoor effective doses were in the range 3.0–7.7 mSv. The lowest dose was obtained at NORM site (Rullekoll), while the highest at TENORM site (Søve). The major contributor to the total outdoor effective doses at all investigated sites was the dose from terrestrial gamma radiation (82.4–97.3%), both in stable and unstable weather conditions. The contribution of ²²⁰Rn and ²²²Rn was significantly lower, 0.1–7.8% and 2.5–8.0%, respectively. Estimated summer month doses were higher (up to 3 times) than fall at all sites. In all the cases, the outdoor doses were higher than the constraint value of 1 mSv given by ICRP⁴² for total (summarized outdoor and indoor) exposure of humans. Values of indoor radiation doses, previously published, could provide an additional insight into total annual exposure doses for the Fen population. The indoor concentrations of radon exceeding 200 and 400 Bq m⁻³ have been found in 37 and 11% of investigated Fen Complex dwellings.¹⁴ The effective dose from indoor gamma radiation has been found in the range of 0.2–3 mSv year⁻¹. If all available data for the Fen Complex were roughly considered together, total exposure dose could exceed 10 mSv year⁻¹ for certain limited parts of the Fen Complex population. Similar results have previously been published.¹³

Fig. 6 Total annual outdoor effective dose for Fen Complex sites; the dashed line presents maximum constraint of radiation for the general public (ICRP).⁴²

However, some uncertainties must be highlighted:

- An exposure period of 20% of the year (1752 hours) might not be the most realistic exposure scenario since the majority of measuring points were in a wooded area. Instead, an exposure period of 350 hours per year would give approximately 5 times lower outdoor doses, in the range of 0.6–1.5 mSv.

- The assumption of uniform distribution of ²²⁰Rn was probably not justified because of its very short half-life. Spatial changes^2,43 which were not taken into account in this paper, could lead to actually much different (both lower and higher) ²²⁰Rn doses, especially under different atmospheric conditions.

- Numerous studies worldwide^3,8,44–48 have published the ²²⁰Rn equilibrium factor values in the range of 0.003–0.1.

With respect to the differences in the equilibrium factor for ²²⁰Rn, the additional dose estimation was done applying a significantly higher equilibrium factor of 0.1. Total annual effective doses were then found to be in the range of 8.0–14.3 mSv.

The contribution of ²²⁰Rn to total dose increased in this case to 74%, ten times higher than the values we derived in the first calculation (with equilibrium factor 0.003). Stranden¹³ reported inhaled ²²⁰Rn progenies as the main contributors to the effective dose at mining sites in the Fen Complex area. Considering that and our recalculations, the estimated total effective doses for Fen population could easily exceed 17 mSv year⁻¹.

Based on all given facts, an accurate estimation of outdoor radiation dose should comprise the direct measurement of ²²⁰Rn decay products and different exposure scenarios reflecting the realistic behaviour of people living in the area.

UNSCEAR ²⁶ provided a list of worldwide ‘enhanced natural radiation areas’ (ENRA), although the specific criteria to characterize an area as ENRA are still needed. Review of the current list showed that these areas mainly have high absorbed gamma dose rates in the air (>300 nGy h⁻¹), enhanced soil radionuclides activity concentrations and ²²²Rn concentrations in the air (indoor ²²²Rn > 150 Bq m⁻³). According to results of the present study, the Fen Complex in Norway should be considered as an ENRA. These locations are of interest to illustrate chronic human exposure to elevated natural radiation levels, as well as in studies of possible effects due to low dose exposure.²⁶

4. Conclusion

The survey of outdoor radiation levels in a ²³²Th rich Norwegian area was done in the present study. In addition, the seasonal variation and exposure doses due to outdoor radiation were evaluated. Significantly high air concentrations of ²²⁰Rn and high terrestrial gamma dose rates in the air were observed for all investigated sites. These high values were attributed to the primordial radiation in the bedrock of the area. Based on long-term surveys, the air concentrations of thoron (²²⁰Rn) and radon (²²²Rn) reached 1786 and 82 Bq m⁻³, respectively. Regression analysis results suggested highly relevant prevalence of ²²⁰Rn in the air. It corresponded to the significant soil enrichment with ²³²Th, found by gamma spectrometric analysis. Statistically significant seasonal variation was obtained regarding ²²⁰Rn, ²²²Rn and gamma dose rates in the air. The effective radiation doses from outdoor exposure (5 h day⁻¹) were estimated to be in the range of 3.0–7.7 mSv year⁻¹. More conservative exposure of spending one hour daily at these locations would give lower doses in the range 0.6–1.5 mSv. However, calculations using a higher equilibrium factor for ²²⁰Rn (F = 0.1) would give doses up to 14.3 mSv, although with high associated uncertainty.

In general, outdoor doses exceeded the value of the average dose constraint of 1 mSv for public exposure (both outdoor and indoor).⁴² Based on the results presented in this study, the Fen Complex area should be considered as an ENRA.

Acknowledgements

This project was financed by Norwegian University of Life Sciences. The authors are grateful to Prof. Dr Petar Stegnar and the Jozef Stefan Institute, Ljubljana, Slovenia, for suggestions and help in ²²⁰Rn and ²²²Rn surveys. Similarly, we are thankful to the Norwegian Radiation Protection Authority for helping with gamma spectrometric analysis of the soil. Also thanks to our colleagues Marit N. Pettersen and Merethe Kleiven for their valuable assistance during the field expeditions.

References

1. N. W. El-Dine, Appl. Radiat. Isot., 2008, 66, 80–85.
2. UNSCEAR (United Nations Scientific Committee on the Effects of Atomic Radiation), Sources, Effects and Risks of Ionizing Radiation, Report to the General Assembly with Scientific Annexes, United Nations, New York, 2000.
3. UNSCEAR (United Nations Scientific Committee on the Effects of Atomic Radiation), Sources, Effects and Risks of Ionizing Radiation, Report to the General Assembly with Scientific Annexes, United Nations, New York, 2006.
4. A. H. Amer, S. Shawky, M. I. Hussein and M. L. Abd El-Hady, J. Environ. Monit., 2002, 4, 583–587.
5. Z. S. Zunic, I. Kobal, J. Vaupotic, K. Kozak, J. Mazur, A. Birovljev, M. Janik, I. Celikovic, P. Ujic, A. Demajo, G. Krstic, B. Jakupi, M. Quarto and F. Bochicchio, J. Environ. Radioact., 2006, 89, 249–260.
6. D. Arnold, A. Vargas and X. Ortega, Appl. Radiat. Isot., 2009, 67, 833–838.
7. N. Sulekha Rao and D. Sengupta, Appl. Radiat. Isot., 2010, 68, 28–32.
8. Y. Yamada, Q. Sun, S. Tokonami, S. Akiba, W. Zhuo, C. Hou, S. Zhang, T. Ishikawa, M. Furukawa and K. Fukutsu, J. Toxicol. Environ. Health, Part A, 2006, 69, 723–734.
9. Z. Zunic, M. Janik, S. Tokonami, N. Veselinovic, I. Yarmoshenko, M. Zhukovsky, T. Ishikawa, R. C. Ramola, G. Ciotoli, P. Jovanovic, K. Kozak, J. Mazur, I. Celikovic, P. Ujic, A. Onischenko, S. K. Sahoo and F. Bochicchio, J. Radiat. Res., 2009, 50, 355–361.
10. D. B. Chambers, Radiat. Prot. Dosim., 2010, 141(4), 351–356.
11. W. C. Brøgger, Die Eruptivgesteiene des Kristianiagebietes IV. Das Fengebiet in Telemark, Norwegen. Skrifter, Det Norske videnskap—akademi i Oslo I, Mat—naturv., 1921, vol. 9, p. 408, in Norwegian.
12. T. F. W. Barth and I. B. Ramberg, Carbonat. Intersci., 1966, 225–257.
13. E. Stranden, SIS Rapport, 1982, in Norwegian.
14. A. V. Sundal and T. Strand, J. Environ. Radioact., 2004, 77, 175–189.
15. B. H. Heincke, M. A. Smethurst, A. Bjørlykke, S. Dahlgren, J. S. Rønning and J. O. Mogaard, Geology for Society, ed. T. Slagstad, Geological Survey of Norway Special Publication, 2008, vol. 11, pp. 131–143.
16. NGI, Søvegruver report (in Norwegian), 2010.
17. H. M. Solli, A. Andersen, E. Stranden and S. Langård, Scandinavian Journal of Work, Environment & Health, 1985, 1, 7–13.
18. J. Mrdakovic Popic, B. Salbu, T. Strand and L. Skipperud, J. Environ. Monit., 2011, 13, 1730–1738.
19. S. Dahlgren, Miljøgeologisk undersøkelse av lavradioaktivt slagg fra ferroniobproduksjonen ved Norsk Bergverk på Søve 1956–1965, 2005, in Norwegian.
20. IFE, Rapport IFE/KR/F, 2006, in Norwegian.
21. W. Zhuo, S. Tokonami, H. Yonehara and Y. Yamada, Rev. Sci. Instrum., 2002, 73, 2877–2881.
22. S. Tokonami, H. Takahashi, Y. Kobayashi, W. Zhuo and E. Hulber, Rev. Sci. Instrum., 2005, 76, 113505.
23. G. Suzuki, I. Yamaguchi, H. Ogata, H. Sugiyama, H. Yonehara, F. Kasagi, S. Fujiwara, Y. Tatsukawa, I. Mori and S. Kimura, J. Radiat. Res., 2010, 51, 683–689.
24. S. Tokonami, W. Zhuo, H. Ryuo, H. Yonehara, Y. Yamada and M. Shimo, Radiat. Prot. Dosim., 2003, 103, 69–72.
25. K. C. Leung, S. Y. Lau and C. B. Poon, J. Environ. Radioact., 1990, 11, 279–290.
26. UNSCEAR (United Nations Scientific Committee on the Effects of Atomic Radiation), Sources, Effects and Risks of Ionizing Radiation, Report to the General Assembly with Scientific Annexes, United Nations, New York, 2008.
27. Forskrift om forurensningslovens anvendelse på radioaktiv forurensning og radioaktivt avfall FOR 2010-11-01 nr 1394 (In Norwegian).
28. A. K. Mohanty, D. Sengupta, S. K. Das, S. K. Saha and K. V. Van, J. Environ. Radioact., 2004, 75, 15–33.
29. A. T. Termizi Ramli, A. W. M. A. Hussein and A. Khalik Wood, J. Environ. Radioact., 2005, 80, 287–304.
30. N. G. S. Prasad, N. Nagaiah, G. V. Ashok and N. Karunakara, Health Phys., 2008, 94, 264–271.
31. S. Almgren and M. Isaksson, J. Environ. Radioact., 2009, 100, 599–604.
32. S. Dahlgren, Naturlig radioaktivitet i berggrunnen, gammastrålingskart, Fensfeltet, Prosjekt temakart, Telemark (In Norwegian), 1983.
33. J. Porstendörfer, A. Reineking, G. Butterweck and J. Kesten, Messung von Radon und Radon-Folgeprodukten, Verlag TÜV Rheinland, Köln, 1991.
34. M. Doi and S. Kobayashi, Health Phys., 1994, 67, 385–392.
35. D. J. Steck, W. Field and C. Lynch, Environ. Health Perspect., 1999, 107, 123–127.
36. S. Oikawa, N. Kanno, T. Sanada, N. Ohashi, M. Uesugi, K. Sato, J. Abukawa and H. Higuchi, J. Environ. Radioact., 2003, 65, 203–213.
37. S. W. Chan, C. W. Lee and K. C. Tsui, J. Environ. Radioact., 2010, 101, 494–503.
38. C. Dueñas, M. C. Fernandez, J. Carretero, E. Liger and M. Perez, Ann. Geophys., 1997, 15, 124–133.
39. R. Winkler and K. Aehlig, Radiat. Environ. Biophys., 1998, 37, 35–39.
40. A. C. Baciu, J. Environ. Radioact., 2005, 83, 171–189.
41. M. P. Chougaonkar, K. P. Eappen, T. V. Ramachandran, P. G. Shetty, Y. S. Mayya, S. Sadasivan and V. Venkat Raj, J. Environ. Radioact., 2004, 71, 275–297.
42. ICRP, Recommendations of the International Commission on Radiological Protection, ICRP Publication, 2007, vol. 103.
43. V. Urosevic, D. Nikezic and S. Vulovic, J. Environ. Radioact., 2008, 99, 1829–1833.
44. P. Chittaporn, N. H. Harley, R. Medora and R. Merrill, Health Phys., 2001, 80, 171.
45. H. Stewart and D. Steck, Proceedings of the American association of radon scientists and technologists, International Symposium Las Vegas NV, September 14–18 2008.
46. K. Kant, R. Sonkawade, G. Sharma, R. Chauhan and S. Chakarvarti, Iranian J. Radiat. Res., 2009, 7, 79–84.
47. J. Chen, D. Moir, A. Sorimachi and S. Tokonami, Radiat. Environ. Biophys., 2011, 50, 85–89.
48. N. Harley, P. Chittaporn, R. Medora and R. Merril, Radiat. Prot. Dosim., 2010, 141, 357–362.

---