Comparison of personal radio frequency electromagnetic field exposure in different urban areas across Europe

Environmental Research 110 (2010) 658–663

Contents lists available at ScienceDirect

Environmental Research journal homepage: www.elsevier.com/locate/envres

Comparison of personal radio frequency electromagnetic field exposure in different urban areas across Europe ¨ b,c, Gyorgy ¨ Wout Joseph a,n,1, Patrizia Frei b,c,1, Martin Roosli Thuro´czy d,h, Peter Gajsek e, Tomaz Trcek e, f a b,c ¨ Vermeeren , Evelyn Mohler , Pe´ter Juha´sz d, Viktoria Finta g, Luc Martens a John Bolte , Gunter a

Department of Information Technology, Ghent University/IBBT Gaston Crommenlaan 8, B-9050 Ghent, Belgium Swiss Tropical and Public Health Institute Basel, Switzerland c University of Basel, Switzerland d Department of Non-ionizing Radiation, National Research Institute for Radiobiology and Radiohygiene, Pf. 101, Budapest 1775, Hungary e Institute of Non-ionizing Radiation, 1000 Ljubljana, Slovenia f Laboratory for Radiation Research, National Institute for Public Health and the Environment (RIVM), Bilthoven, The Netherlands, g ´nd University, Faculty of Science, Institute of Physics, Department of Atomic Physics Address 1117 Budapest, Pa ´zma ´ny Pe´ter se´ta ´ny 1/A, Hungary E¨ otv¨ os Lora h French National Institute for Industrial Environment and Risks (INERIS), Verneuil en Halatte, France b

a r t i c l e in f o

a b s t r a c t

Article history: Received 26 February 2010 Received in revised form 1 June 2010 Accepted 28 June 2010

Background: Only limited data are available on personal radio frequency electromagnetic field (RF-EMF) exposure in everyday life. Several European countries performed measurement studies in this area of research. However, a comparison between countries regarding typical exposure levels is lacking. Objectives: To compare for the first time mean exposure levels and contributions of different sources in specific environments between different European countries. Methods: In five countries (Belgium, Switzerland, Slovenia, Hungary, and the Netherlands), measurement studies were performed using the same personal exposure meters. The pooled data were analyzed using the robust regression on order statistics (ROS) method in order to allow for data below the detection limit. Mean exposure levels were compared between different microenvironments such as homes, public transports, or outdoor. Results: Exposure levels were of the same order of magnitude in all countries and well below the international exposure limits. In all countries except for the Netherlands, the highest total exposure was measured in transport vehicles (trains, car, and busses), mainly due to radiation from mobile phone handsets (up to 97%). Exposure levels were in general lower in private houses or flats than in offices and outdoors. At home, contributions from various sources were quite different between countries. Conclusions: Highest total personal RF-EMF exposure was measured inside transport vehicles and was well below international exposure limits. This is mainly due to mobile phone handsets. Mobile telecommunication can be considered to be the main contribution to total RF-EMF exposure in all microenvironments. & 2010 Elsevier Inc. All rights reserved.

Keywords: Radio frequency electromagnetic fields (RF-EMF) Personal exposure meter RF measurement Exposure of general public Mobile phone base station DECT cordless phone

1. Introduction Very little is known about RF-EMF exposure in everyday life. Personal electromagnetic field exposure of the general public is nowadays assessed with personal exposure meters (exposimeters). Research recommendations about the feasibility

Abbreviations: DECT, digital enhanced cordless telecommunication, cordless phone; FM, frequency modulation (FM radio); RF-EMF, radio frequency electromagnetic field; ROS, regression on order statistics; TV, television; DAB, digital audio broadcast; W-LAN, wireless local area network; MHz, megahertz n Corresponding author. Fax: + 32 9 33 14899. E-mail address: [email protected] (W. Joseph). 1 Both contributed equally to the paper. 0013-9351/$ - see front matter & 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.envres.2010.06.009

of performing studies with exposimeters and concerning their limitations are given in (Neubauer et al., 2007; Radon et al., 2006; ¨ et al., 2010). In the Mann et al., 2005; Lehmann et al., 2006; Ro¨ osli last few years, several countries have performed measurement studies using exposimeters and some of the results have already been published (Bolte et al., 2008; Frei et al., 2009a; Joseph et al., ¨ 2008; Ro¨ osli et al., 2008; Thomas et al., 2008a, b; Thuro´czy et al., 2008; Trcek et al., 2007; Viel et al., 2009). In some of these studies, measurements were performed in different microenvironments such as offices or outdoor urban areas to characterize typical exposure levels in these places (microenvironmental studies). The other studies were population surveys where the personal exposure distribution in the population of interest was determined. The strategies for the recruitment of the study

W. Joseph et al. / Environmental Research 110 (2010) 658–663

participants as well as the data analysis methods differed between these studies, and therefore, a direct comparison of their results is difficult. The objective of this paper is to compare personal radio frequency electromagnetic field exposure in different microenvironments between urban areas in five European countries by applying the same data analysis methods. The countries which are considered in this paper are Belgium (Joseph et al., 2008), ¨ Switzerland (Ro¨ osli et al., 2008; Frei et al., 2009a), Slovenia (Trcek et al., 2007; Valic et al., 2009), Hungary (Thuro´czy et al., 2008), and the Netherlands (Bolte et al., 2008). In each of these countries large measurement studies using exposimeters were performed. Data on mean exposure levels in the specific microenvironments were collected and the pooled data were analyzed in order to enable a comparison. Ethical approval for the conduct of the studies was received in each country from the authorized ethical committees, where needed.

2. Materials and methods 2.1. Definition of the microenvironments We defined a limited number of typical microenvironments for the general public in order to enable a comparison between the different countries. Table 1 lists these microenvironments together with a short name with the logic: ‘‘location–environment–time’’ (e.g., microenvironment outdoor–urban–during daytime). As ‘‘locations’’ we considered office, train, car/bus, at home, and outdoor (Table 1). To minimize heterogeneity among the studies, we defined the environments as follows: urban areas are areas with more than 400 persons per square km. With office we mean regular offices, i.e., working places of employees working at desks (not working places in factories). We define ‘‘daytime’’ as the period from 6 am to 6 pm (working hours) and ‘‘nighttime’’ is the period where most people do not work or are sleeping, thus the period after 6 pm and before 6 am. Daytime and nighttime are here thus defined with respect to working hours.

2.2. Measurement procedures in the different countries

Antennessa, Avenue de Norve ge 17, 91953 Courtaboeuf, France). In Hungary some of the measurements were done with an older version of the EME SPY and in the Netherlands a newer version of the device (EME SPY 121) was used. The exposimeter measures 12 frequency bands, namely, FM (frequency modulation, i.e., FM radio broadcast transmitters; 88–108 MHz), TV/DAB (television broadcast transmitters/digital audio broadcasting; 174-223 MHz (TV and DAB) and 470–830 MHz (TV)), Tetrapol (terrestrial trunked radio, i.e., mobile communication system for closed groups; 380–400 MHz), uplink in three frequency ranges (communication from mobile phone handset to base station; 880–915, 1710–1785, and 1920–1980 MHz), downlink in three frequency ranges (communication from mobile phone base station to handset; 925–960, 1805–1880, and 2110–2170 MHz), DECT (digital enhanced cordless telecommunications; 1880–1900 MHz), and W-LAN (wireless local area network). For all frequency bands, the exposimeter detects power flux densities S (shortly noted as power density) between 0.0067 and 66.3 mW/m2 (which corresponds to electric field strengths ranging from 0.05 to 5 V/m). The measurement procedures and number of samples and subjects are summarized in Tables 2 and 3, respectively. The measurements were performed in the period 2007–2009. In the studies of Belgium and the Netherlands, measurements were made by hired staff whereas the studies of Switzerland, Slovenia, and Hungary were based on volunteers from a population sample. In Belgium, the measurements in the different microenvironments were conducted by seven researchers of the author’s institute following a prespecified protocol. They had to perform measurements during at least 5 h per microenvironment. Of the 131 study participants in Switzerland, 8 were volunteers from the authors’ institute, 55 registered for participation on the study homepage, and 68 were friends of the authors or of the study participants. Each participant carried an exposimeter during 1 week. In Slovenia, a population survey based on 20 participants was conducted. Of these participants 18 were volunteers recruited from a call on the national radio station and 2 registered on the study homepage. Participants carried an exposimeter for 34 h. In Hungary, the 138 volunteers were derived from two different recruitment strategies: 90 subjects were recruited according to their occupation, which was restricted to teachers and elementary school workers, workers of daycare centers and kindergartens, and workers of municipal offices. The responsible persons of the given institution (director, department head, etc.) were first approached by the

Table 3 Number of samples per microenvironment and number of subjects (population survey) or number of environments (microenvironmental measurement) for different countries. Country

Samples Microenvironment

In each of the countries, measurements were performed using the selective isotropic personal exposure meter of type DSP120 EME SPY of SATIMO (former

Out-urban Office Belgium

Table 1 Considered microenvironments for comparison. Logical name

Microenvironment

Description

Out-urban

Outdoor–urban–day

Office Train Car/bus Urban-home

Indoor–office–daytime Indoor–train–daytime Indoor–car/bus–daytime Indoor–urban home–day/ night

Outdoor walking, standing, sitting in urban environment during daytime, 4400 persons/km2 Office environment (employees) Exposure in train Exposure in car or bus: driving In home in urban area 4400 persons/km2

659

#Sampl #Env Switzerland #Sampl #Subj Slovenia #Sampl #Subj Hungary #Sampl #Subj The Netherlands #Sampl #Env

4539 16 35211 127 5115 15   9999 51

Train Car/bus Urban-home

5111 3116 2914 4 9 21 49557 5362 15339 58 46 97 20832  4038 16  15 34304  8058 29  42 6237 5316 6219 3 11 19

8100 11 537331 131 37679 13 81840 42 3544 19

#Sampl¼number of samples, #env¼ number of microenvironments (microenvironmental study). #Subj¼ number of subjects (population survey).

Table 2 Measurement procedures applied in the different countries. Country

Methods described in

Selection of participants

Study area

Time of measurem. collection

Measurement interval (s)

Length of measurem.

Number of participants

Belgium

Joseph et al. (2008)

Researchers

Ghent, Brussels

Aug 07–Jan 08

10

7

Switzerland

Frei et al. (2009a)

Volunteers

Basel

Apr 07–Feb 08

90

Slovenia

Trcek et al. (2007) Valic et al. (2009) Thuro´czy et al. (2008) Bolte et al. (2008)

Volunteers

Ljubljana

20

Volunteers

Budapest

Hired employee

Utrecht and Netherlands

May 08–Nov 08 May 07–Sept 09 Spring 08

Min. 5 h per environment 1 week per person 34 h per person 48 h per person All 13.5 h per environment

Hungary The Netherlands

30 4

131 20 138 5

660

W. Joseph et al. / Environmental Research 110 (2010) 658–663

authors either via email or phone. If they agreed to collaborate, the authors went to the institution and asked who wanted to collaborate. The remaining 48 subjects were university students, teachers, and lecturers from the authors’ institute. The university students and teachers were directly approached by the authors. The volunteers had to be from three specific districts of Budapest. The volunteers carried the exposimeter during 48 h. In the Netherlands, five persons were hired for conducting the measurements (microenvironmental study). Master students who were contacted using a call at Utrecht University were instructed to collect the measurements according to a prespecified protocol. Each student carried an exposimeter for 13.5 h per environment. The measurement interval was between 4 and 90 s (Table 2). In all population surveys, participants were allowed to use their own mobile and cordless phones. Hired persons were not allowed to use their mobile phones. In all studies, activity diaries were filled in during the measurements that allowed allocation of the measurements to specific activities or places. In Belgium, Switzerland, and Slovenia, the exposimeter was worn at the belt, in a backpack or shoulder bag when moving, and when stationary it was standing alone in the vicinity of the person. A similar approach was used in Hungary and in the Netherlands, but during the day while being stationary, the exposimeter was also worn on the body, and at night (sleeping) it was standing alone close to the volunteer.

2.3. Data analysis In all countries, all measurements taken in each microenvironment were combined and analyzed. Since a large proportion of the measurements was censored, i.e., below the lower detection limit of the exposimeter (0.0067 mW/m2), we applied the robust regression on order statistics (ROS) method ¨ proposed by Ro¨ osli et al. (2008) to determine the mean values of the power density S (mW/m2) for each microenvironment. ROS is a method that fits a normal distribution (or log-normal if logs are used) to the data above the detection limit (detected values). In its robust form, the modeled censored values are then combined with the detected values to obtain summary statistics. A full description of the method can be found in Helsel (2005). By combining the detected values with the modeled censored values, this method is more resistant to any nonnormality errors and may thus be particularly applicable for exposimeter data with a large proportion of censored data. Using the ROS method for personal RF-EMF measurements, generally appropriate results are obtained for data files with more than 3000 observations, ¨ even if the proportion of censored values is larger than 80% (Ro¨ osli et al., 2008). Exposimeter data of all participating countries were processed in exactly the same way using the ROS algorithm with the statistical software R (www.r-project.org), which is discussed in detail in Helsel (2005).

The total exposure Stot (mW/m2) was calculated by summing up the mean power density values of all frequency bands. Table 3 lists the number of measurements per scenario for each microenvironment. In addition, the number of subjects (population surveys) or the number of measured microenvironments (microenvironmental studies), are provided in Table 3.

3. Results 3.1. General comparison of total RF-EMF Table 4 summarizes the mean power densities for all the microenvironments and countries. All mean values are well below the ICNIRP exposure guidelines (ICNIRP, 1998), which are the basis for exposure limits in the considered countries. Also, the more restrictive limits for Belgium (4 times lower than the ICNIRP guidelines and additionally 3 V/m at 900 MHz) and for specific situations in Slovenia and Switzerland (10 times lower than the ICNIRP guidelines at places of sensitive use (homes, offices)) are not exceeded. Fig. 1 shows Stot (mW/m2) for mean exposures for all microenvironments and all countries. Exposure in all countries is of the same order of magnitude. Table 4 and Fig. 1 demonstrate that highest exposure occurs in transportation vehicles (train and car/bus) for all countries except for the Netherlands, where the highest exposure levels were measured in offices. In all countries except for the Netherlands, the same pattern can be observed: highest exposures were measured in transportation vehicles (train, car/bus), followed by outdoor urban exposure, offices, and urban homes. Exposure is lowest in urban homes in all countries. 3.2. Comparison of contributions from different sources Fig. 2 shows the contribution of the different sources to the exposure in the five microenvironments for all countries. In all microenvironments, exposure to mobile telecommunication

Table 4 Mean values of S (mW/m2) for all microenvironments and all countries. Microenvironment

Country

FM

TV/DAB

Tetrapol

Uplink

Downlink

DECT

W-LAN

Total

Outdoor-urban

Belgium Switzerland Slovenia Hungary Netherlands Belgium Switzerland Slovenia Hungary Netherlands Belgium Switzerland Slovenia Hungary Netherlands Belgium Switzerland Slovenia Hungary Netherlands Belgium Switzerland Slovenia Hungary Netherlands

0.014 0.007 0.004 NA 0.024 0.096 0.003 0.008 0.009 0.001 0.007 0.001 NA NA 0.019 0.040 0.002 0.005 0.013 0.014 0.031 0.003 0.005 0.010 0.001

0.015 0.017 0.002 NA 0.003 0.089 0.002 0.001 0.005 0.001 0.001 0.002 NA NA 0.004 0.005 0.004 0.001 0.003 0.006 0.014 0.003 0.001 0.004 0.001

0.002 0.000 0.001 NA 0.002 0.000 0.001 0.000 0.001 0.000 0.000 0.000 NA NA 0.000 0.000 0.000 0.000 0.001 0.001 0.001 0.001 0.001 0.002 0.000

0.003 0.097 0.418 NA 0.091 0.041 0.062 0.212 0.077 2.122 0.892 0.983 NA NA 0.718 0.252 0.239 1.162 0.246 1.012 0.029 0.014 0.054 0.036 0.135

0.334 0.081 0.131 NA 0.345 0.024 0.024 0.088 0.027 0.005 0.020 0.073 NA NA 0.016 0.052 0.053 0.073 0.115 0.036 0.020 0.019 0.025 0.014 0.004

0.003 0.005 0.011 NA 0.009 0.024 0.058 0.060 0.001 0.056 0.001 0.002 NA NA 0.001 0.005 0.004 0.250 0.014 0.001 0.000 0.046 0.015 0.012 0.000

0.000 0.002 0.001 NA 0.000 0.018 0.014 0.002 0.004 0.001 0.001 0.002 NA NA 0.000 0.000 0.001 0.003 0.002 0.002 0.013 0.002 0.002 0.006 0.004

0.372 0.209 0.569 NA 0.474 0.293 0.164 0.372 0.125 2.185 0.923 1.063 NA NA 0.757 0.355 0.304 1.493 0.394 1.071 0.109 0.086 0.104 0.085 0.145

Office

Train

Car/bus

Urban-home

NA¼ not available. FM¼ frequency modulation (radio broadcast), TV ¼ television broadcast, DAB ¼digital audio broadcasting, Tetrapol ¼ terrestrial trunked radio, i.e., mobile communication system for closed groups, uplink¼ communication from mobile phone to base station, downlink¼ communication from base station to mobile phone, DECT ¼digital enhanced cordless telecommunications, W-LAN: wireless local area network.

W. Joseph et al. / Environmental Research 110 (2010) 658–663

(downlink and uplink) is important and in most cases clearly dominating. In Hungary, Slovenia, and the Netherlands mobile telecommunication exposure is the highest in all environments. In most microenvironments mainly mobile phone handset exposure is relevant for all countries: in the Netherlands, uplink exposure due to mobile phone calls even dominates in all environments except outdoor-urban. For outdoor-urban environments downlink exposure due to mobile phone base stations is important for all countries and dominating for Belgium and the Netherlands. Exposure in transportation vehicles (train and car/bus) is mainly due to radiation from mobile phone handsets (uplink, with

2.5 Belgium Switzerland Slovenia Hungary The Netherlands

S (mW/m2)

2 1.5 1 0.5

661

percentages of 92.5–96.6% for trains and 62.5–94.4% in cars/ busses (Fig. 2). Exposure contributions in car/bus are similar to the contributions in trains. In urban homes, exposure contributions are very different among different countries. The most important contributions per country are DECT in Switzerland (53.2%), Slovenia, and Hungary (14.7%), FM radio in Belgium (28.7%) and Hungary (11.9%), mobile phone base stations in Belgium, Switzerland, Slovenia, and Hungary (16.8–23.7%), and mobile phones in the Netherlands (92.8%), Slovenia (52.5%), and Hungary (48.2%). W-LAN is present in urban homes with a contribution ranging from 2.0% to 12.2% in the different countries. Total exposure in homes is similar among the different countries and is somewhat higher in the Netherlands. In Fig. 2, it can be observed that Tetrapol, W-LAN, and TV/DAB are sources of minor importance in most of the microenvironments. The contributions of TV/DAB are less than 5% of the total exposure except in homes and offices in Belgium (13.3% and 30.3%, respectively), and outdoor-urban in Switzerland (8.0%). FM radio is only relevant in offices and homes (mainly in Belgium and Hungary).

4. Discussion

0 Outdoor-urban Office

Train

car/bus Urban-home

Fig. 1. Mean total exposures (mW/m2) for all considered microenvironments in all countries.

Our study offers a comparison of mean personal RF-EMF exposure levels between urban areas in different European countries (Belgium, Switzerland, Slovenia, Hungary, and the Netherlands). Five

Urban-home

Office

100%

100%

80%

80%

60%

60%

40%

40%

20%

20% 0%

0% Belgium

Switzerland

Slovenia

Hungary

Belgium

The Netherlands

Switzerland

Slovenia

Hungary

The Netherlands

Hungary

The Netherlands

Outdoor-urban 100%

Train 100%

80%

80%

60%

60%

40%

40%

20%

20% 0%

0% Belgium

Switzerland

Slovenia

Hungary

The Netherlands

Belgium

Switzerland

Slovenia

Car/bus 100% W-LAN DECT Downlink Uplink Tetrapol TV FM

80% 60% 40% 20% 0% Belgium

Switzerland

Slovenia

Hungary

The Netherlands

Fig. 2. Comparison of the contribution of the different sources of exposure for all countries.

662

W. Joseph et al. / Environmental Research 110 (2010) 658–663

relevant microenvironments were defined, which allowed to compare mean exposure levels as well as the contributions of different exposure sources in these microenvironments.

4.1. Strengths and limitations This paper is the first one where a comparison of personal RF-EMF exposure between urban areas in different countries is made. Till date, a comparison of the results from different measurement studies was limited due to different methods of analysis. Our approach enabled us to compare data from two different types of exposure studies. Analysis was performed by the same person in order to ensure the comparability of the results. Applying the ROS method ensures generally appropriate results ¨ even if the number of censored values is large (Ro¨ osli et al., 2008). The differences in study design and the selection of study participants in the population surveys is one of the limitations of this study. Another limitation is that in the microenvironmental studies some of the results are based on only a limited number of different microenvironments. For example for offices, the mean values in the Netherlands and Belgium are based on only 3 and 4 offices, respectively. Such results may thus not be representative for the whole study area. Including random population samples for the personal measurements would have been appealing. However, carrying an exposimeter and completing an activity diary is a demanding task and thus, participation rate and data quality is expected to be insufficient in a random population sample. In order to ensure a reliable matching between the activity diary data and the measurements, the aim of the recruitment strategy in the different countries was to include motivated people by either hiring them or by including persons who voluntarily registered for participation. Lack of a random population sample is not considered to be an important limitation in this study, because we did not aim at characterizing the exposure distribution in population samples, rather we determined exposure levels in specified microenvironments. Thus, various study participants were selected for the sole reason to visit a broad range of different microenvironments representing typical everyday behavior. As still little is known about the variability of RF-EMF in the everyday environment, it is problematic to determine exactly how many different microenvironments have to be included to be representative. In our study, many different types of urban outdoor and urban home microenvironments were included, whereas office and train measurements were based on fewer locations in most of the countries. There are some aspects to be considered when interpreting exposimeter measurements. Exposimeters cannot realistically reflect exposure from close to body sources (Inyang et al., 2008), as measurements during calls with mobile phones or DECT phones strongly depend on the distance between the device and the exposimeter (Frei et al., 2009a). In the population surveys included in this paper (Switzerland, Slovenia, and Hungary), the measurements of the user’s own mobile phone were included. Exposure to uplink for the microenvironments in these countries might therefore be overestimated in comparison to the microenvironmental studies (Belgium and the Netherlands), where it was not allowed to use own mobile phones during the measurements. Moreover, shielding of the human body when exposimeters are carried on the body might have played a role (Knafl et al., 2008; Neubauer et al., 2008). In all studies the exposimeter was worn on the body when moving. When stationary, it was standing alone in the vicinity of the person (Belgium, Switzerland, and Slovenia), where shielding is expected to be a minor issue. In the Netherlands and Hungary, the

exposimeters were worn on the body all of the time (except during night). For the microenvironments such as urban homes or offices, where the test subject is usually stationary, mean values might be underestimated for the Netherlands and Hungary, as shielding of the body is expected to lead to underestimations (Neubauer et al., 2008). For the other microenvironments, the exposimeter was handled in a similar way across the different countries. 4.2. Interpretation As the considered countries are spread over Europe, the results of this study give indications about the RF-EMF exposure distribution in European countries. We observed quite consistently in all countries that the highest exposure contributions were from mobile phone handsets. Contributions from handsets were particularly dominant in transportation vehicles (train, car/ bus). Mobile phone use causes higher exposure for mobile scenarios (inside a train and car/bus) than for outdoor or stationary exposure due to changing environmental conditions, handovers from one base station to another, and higher required transmitted signal power from mobile phones to overcome penetration loss through (metallized) windows. Additionally, the concentration of people in public transport is usually higher than in other environments, leading to a higher probability of uplink communication and therefore higher exposures. This was also observed in Belgian, Swiss, and French studies (Frei et al., 2009a; Joseph et al., 2008; Viel et al., 2009). Interestingly, in the Netherlands mobile telecommunication exposure is the highest in all environments, despite the fact that the persons carrying the exposimeter were not allowed to make phone calls. Exposure at home is particularly relevant in terms of cumulative exposure, as one spends most of the time at home. In general, exposure levels were quite similar in the different countries but somewhat higher in the Netherlands. Interestingly, exposure contributions were very different. The only pattern showing is that exposure due to mobile telecommunication is important in all countries. However, the exposure contributions in urban homes have to be interpreted with caution. Different recruitment strategies and the limited numbers of participants or microenvironments make a direct comparison difficult. The high proportion of uplink and mobile telecommunication in all countries is undeniable as its impact on exposure levels is clearly visible in all microenvironments in all studies. In all countries, exposure in offices was higher than in urban homes. One explanation for this might be that only daytime was considered for offices as most people are at their offices during daytime, whereas in urban homes nighttime measurements were also considered. It was shown that daytime values are in general higher than nighttime values (Frei et al., 2009b; Joseph et al., 2008). For outdoor urban environments, mobile phone base stations (all countries) and mobile phone handsets dominate the exposure because of their omnipresence in today’s world. Finally, exposure in all countries is of the same order of magnitude (Table 4). In all these countries there is a similar standard of living and the same RF-EMF sources and technologies (mobile phones, base stations, radio, TV/DAB) are present in similar amounts and densities.

5. Conclusions Our study offers for the first time a comparison of mean RF-EMF exposure levels and contribution of different sources in urban environments among different European countries.

W. Joseph et al. / Environmental Research 110 (2010) 658–663

Exposure in all countries is of the same order of magnitude. In all environments, exposure to mobile telecommunication is important and mostly dominating. Consistently in all countries, exposure is lowest in urban homes and highest exposure contributions are obtained from mobile phone handsets, which are particularly dominant in transportation vehicles (trains). Future research should be the planning and execution of large common measurement studies among different countries. A joint study in which an equal study design is used in all the countries would enable a more detailed comparison. The different microenvironments could be defined according to the method proposed in this paper.

Conflicts of interest There are no potential conflicts of interest related to this work.

Funding sources and Acknowledgments W. Joseph is a Post-Doctoral Fellow of the FWO-V (Research Foundation – Flanders). The Swiss study was funded by the Swiss National Science Foundation (Grant 405740-113595). The Hungarian study was granted by Ministry of Health ETT-037/2006. ¨ G. Thuro´czy thanks Edit Sa´rkozi for her technical help in the data recording and evaluation. J. Bolte thanks the Netherlands Organization for Health Research and Development (ZonMw) for funding the Dutch study (http://www.zonmw.nl/en/programmes/ all-programmes/electromagnetic-fields-and-health-research/). References Bolte, J., Pruppers, M., Kramer, J., Van der Zande, G., Schipper, C., Fleurke, S., Kluwer, T., Van Kamp, I., Kromhout, J., 2008. The Dutch exposimeter study: developing an activity exposure matrix. Epidemiology 19 (6), S78–79. Frei, P., Mohler, E., Neubauer, G., Theis, G., Burgi, A., Frohlich, J., Braun-Fahrlander, ¨ C., Bolte, J., Egger, M., Roosli, M., 2009a. Temporal and spatial variability of personal exposure to radiofrequency electromagnetic fields. Environmental Research 109, 779–785. ¨ Frei, P., Mohler, E., Burgi, A., Frohlich, J., Neubauer, G., Braun-Fahrlander, C., Roosli, M., 2009b. A prediction model for personal radio frequency electromagnetic field exposure. Sci. Total Environ. 408, 102–108 QUALIFEX team. Helsel, D.R., 2005. In: Scott, M., Barnett, V. (Eds.), Nondetects and Data Analysis. John Wiley & Sons Inc., New Jersey. International Commission on Non-ionizing Radiation Protection (ICNIRP), 1998. Guidelines for limiting exposure to time-varying electric, magnetic, and electromagnetic fields (up 300 GHz). Health Phys. 74 (4), 494–522. Inyang, I., Benke, G., McKenzie, R., Abramson, M., 2008. Comparison of measuring instruments for radiofrequency radiation from mobile telephones in

663

epidemiological studies: implications for exposure assessment. J. Expo. Sci. Environ. Epidemiol. 18, 134–141. Joseph, W., Vermeeren, G., Verloock, L., Heredia, M.M., Martens, L., 2008. Characterization of personal RF electromagnetic field exposure and actual absorption for the general public. Health Phys. 95 (3), 317–330. Knafl, U., Lehmann, H., Riederer, M., 2008. Electromagnetic field measurements using personal exposimeters. Bioelectromagnetics 29, 160–162. Lehmann, H., Biner, J., Eicher, B., Fritschi, P., Hermann, U., Knafl, U., Rubin, M., 2006. Benchmarking personal radiofrequency exposimeters. In: Proceedings of the International Conference and COST 281 Workshop on Emerging Technologies, Potential Sensitive Groups and Health, Graz. Mann, S.M., Addison, D.S., Blackwell, R.P., Khalid, M., 2005. Personal dosimetry of RF radiation: laboratory and volunteer trials of an RF personal exposure meter. Health Protection Agency HPA-RPD-008. Neubauer, G., Feychting, M., Hamnerius, Y., Kheifets, L., Kuster, N., Ruiz, I., Schuz, J., ¨ Uberbacher, R., Wiart, J., Ro¨ osli, M., 2007. Feasibility of future epidemiological studies on possible health effects of mobile phone base stations. Bioelectromagnetics 28, 224–230. ¨ ¨ Neubauer, G., Cecil, S., Giczi, W., Petric, B., Preiner, P., Frohlich, J., Ro¨ osli, M., 2008. Final report on the Project C2006-07, Evaluation of the correlation between RF dosimeter reading and real human exposure. ARC-Report ARC-IT-0218. Austrian Research Centers GmbH—ARC. Available at /http://www.mobile-r esearch.ethz.ch/var/SB_Neubaue_pre25f.pdfS [Accessed 9 February 2010]. Radon, K., Spegel, H., Meyer, N., Klein, J., Brix, J., Wiedenhofer, A., Eder, H., Praml, G., Schulze, A., Ehrenstein, V., Von Kries, R., Nowak, D., 2006. Personal dosimetry of exposure to mobile telephone base stations? An epidemiologic feasibility study comparing the Maschek dosimeter prototype and the Antennessa DSP-090 system. Bioelectromagnetics 27 (1), 77–81. ¨ R¨oosli, M., Frei, P., Mohler, E., Braun-Fahrl¨ander, C., Burgi, A., Fr¨ohlich, J., Neubauer, G., Theis, G., Egger, M., 2008. Statistical analysis of personal radiofrequency electromagnetic field measurements with nondetects. Bioelectromagnetics 29 (6), 471–478. ¨ Ro¨ osli, M., Frei, P., Bolte, J., Neubauer, G., Cardis, E., Feychting, M., Gajsek, P., Heinrich, S., Joseph, W., Mann, S., Martens, L., Mohler, E., Parslow, R., Poulsen, ¨ J., Thuroczy, G., Viel, J.-F., Vrijheid, M., 2010. Conduct of A.H., Radon, K., Schuz, a personal radiofrequency electromagnetic field measurement study: proposed study protocol. Environ. Health 9, 23. ¨ Thomas, S., Kuhnlein, A., Heinrich, S., Praml, G., Von Kries, R., Radon, K., 2008a. Exposure to mobile telecommunication networks assessed using personal dosimetry and well-being in children and adolescents: the German MobilEestudy. Environ. Health 7 (1), 54. ¨ Thomas, S., Kuhnlein, A., Heinrich, S., Praml, G., Nowak, D., Von Kries, R., Radon, K., 2008b. Personal exposure to mobile phone frequencies and well-being in adults: a cross-sectional study based on dosimetry. Bioelectromagnetics 29 (6), 463–470. Thuro´czy, G., Molna´r, F., Ja´nossy, G., Nagy, N., Kubinyi, G., Bakos, J., Szabo´, J., 2008. Personal RF exposimetry in urban area. Ann. Telecommun. 63, 87–96. Trcek, T., Valic, B., Gajsek, P., 2007. Measurements of background electromagnetic fields in human environment. IFMBE Proceedings 11th Mediterranean Conference on Medical and Biomedical Engineering and Computing 2007, Ljubljana, Slovenia MEDICON, vol. 16, pp. 222–225. Valic, B., Trcek, T., Gajsek, P., 2009. Personal exposure to high frequency electromagnetic fields in Slovenia. In: Joint Meeting of the Bioelectromagnetics Society and the European BioElectromagnetics Association, 14–19 June 2009. Davos, Switzerland [Online 7 December 2009 /http://bioem2009.org/uploads/ Abstract%20Collections.pdfS]. Viel, J.-F., Cardis, E., Moissonnier, M., De Seze, R., Hours, M., 2009. Radiofrequency exposure in the French general population: band, time, location and activity variability. Environ. Int. 35 (8), 1150–1154. doi:10.1016/j.envint.2009.07.007 [Online 7 July 2009].