- Email: [email protected]
Electrical stimulation vs thermal effects in a complex electromagnetic environment
Science of the Total Environment 407 (2009) 4717–4722
Contents lists available at ScienceDirect
Science of the Total Environment j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / s c i t o t e n v
Electrical stimulation vs thermal effects in a complex electromagnetic environment Jesús M. Paniagua ⁎, Montaña Rufo, Antonio Jiménez, Alicia Antolín, Miguel Sánchez Department of Applied Physics, Polytechnic School, University of Extremadura. Avda. de la Universidad s/n, 10071 Cáceres, Spain
a r t i c l e
i n f o
Article history: Received 15 December 2008 Received in revised form 31 March 2009 Accepted 21 April 2009 Available online 28 May 2009 Keywords: Electrical stimulation Exposure levels Radiation levels Spectral analysis Urban environment Thermal effects
a b s t r a c t Studies linking exposure to low levels of radiofrequencies with adverse health effects, notwithstanding their present apparent inconsistency, have contributed to a steady improvement in the quality of evaluating that exposure. In complex electromagnetic environments, with a multitude of emissions of different frequencies acting simultaneously, knowledge of the spectral content is fundamental to evaluating human exposure to non-ionizing radiation. In the present work, we quantify the most significant spectral components in the frequency band 0.5–2200 MHz in an urban area. The measurements were made with a spectrum analyzer and monopole, biconical, and log-periodic antennas. Power density levels were calculated separately for the medium wave, short wave, and frequency modulation radio broadcasting bands, and for the television and GSM, DCS, and UMTS mobile telephony bands. The measured levels were compared with the ICNIRP reference levels for exposure to multiple frequency sources for thermal effects and electrical stimulation. The results showed the criterion limiting exposure on the basis of preventing electrical stimulation of peripheral nerves and muscles to be stricter (exposure quotient 24.7 10− 4) than that based on thermal considerations (exposure quotient 0.16 10− 4). The bands that contribute most to the latter are short wave, with 46.2%, and mobile telephony with 32.6% of the total exposure. In a complex electromagnetic environment, knowledge of the radiofrequency spectrum is essential in order to quantify the contribution of each type of emission to the public's exposure. It is also necessary to evaluate the electrical effects as well as the thermal effects because the criterion to limit exposure on the basis of the effect of the electrical stimulation of tissues is stricter than that based on thermal effects. © 2009 Elsevier B.V. All rights reserved.
1. Introduction Environmental levels of electromagnetic radiation have been constantly increasing due to the continual implementation of new telecommunications systems. The technical evolution of mobile telephony, and the appearance of wifi and wireless systems, have generated radiofrequency (RF) emissions in addition to those that had existed for decades from radio and television broadcasting antennas. The sudden implementation of all these telecommunication devices, together with the constant stream of studies linking electromagnetic fields with adverse health effects, have led to concern among the public at large over the possible risks of exposure to nonionizing radiation. As a result, various national and international agencies, such as the International Commission on Non-ionizing Radiation Protection (ICNIRP), the Federal Communications Commission (FFC), and the Institute of Electrical and Electronics Engineers (IEEE), have prepared regulatory guidelines to limit exposure to electromagnetic fields (FCC, 1999; ICNIRP, 1998; IEEE, 2006) based primarily on criteria of thermal effects. There have been studies, however, that have detected other nonthermal effects. For example, Croft et al. (2008) in studies performed ⁎ Corresponding author. Tel.: +34 927257597; fax: +34 927257203. E-mail address: [email protected] (J.M. Paniagua). 0048-9697/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.scitotenv.2009.04.034
on volunteers find effects caused by cell-phones on the alpha rhythm in electroencephalograms. Kolodynski and Kolodynska (1996) find reduced development of memory, attention, and motor functions in children living near a radio station. And other studies report increased incidence of various types of cancer in people living near television towers (Hocking et al., 1996), TV and FM radio transmitters (Dolk et al., 1997), medium and short wave radio transmitters (Michelozzi et al., 2002), and mobile telephony antennas (Wolf and Wolf, 2004). A exhaustive review of epidemiological studies of the effects of RF fields on human health was published by Ahlbom et al. (2004), with reports of high rates of cancer in some cases. Those same authors indicate, however, that the results of epidemiological studies do not as yet provide consistent or convincing evidence of a causal relationship between RF exposure and adverse health effects, and furthermore that those studies have too many deficiencies for it to be possible to establish any association (Ahlbom et al., 2004). A key aspect in all studies of this type is the quality of the evaluation of RF exposure. There are often situations in which the determination of the degree of compliance with standards for protection against electromagnetic fields is difficult and cannot always be done directly. The spectral content, spatial and temporal patterns, and polarization are some of the factors in the electromagnetic environment that may be important in evaluating a biological effect. Despite the rapid growth of the new technologies,
4718
J.M. Paniagua et al. / Science of the Total Environment 407 (2009) 4717–4722
little is known about public exposure to radiofrequencies and even less about the relative importance of different sources. This is particularly relevant in complex electromagnetic environments in which a multitude of emissions of different frequencies are acting simultaneously. It is therefore necessary to improve the methods of measurement so that the spectral components can be quantified, and hence the measured field levels compared with the reference levels for different frequencies, since the potential effects on the organism will be different for different frequencies. Equally important is to use criteria for calculating exposure levels that take into account the possible additivity of the effects. For example, in its regulatory guidelines, ICNIRP (ICNIRP, 1998) states that additivity should be examined separately to limit electrical stimulation and thermal effects. The effects of electrical stimulation predominate over other possible effects in the low frequency region, up to 10 MHz, where the current density J is the dosimetry parameter that is applied. For frequencies between 100 kHz and 300 GHz, thermal effects predominate, and from 10 MHz to 10 GHz the specific absorption rate, SAR, and from 10 GHz to 300 GHz the incident power density are the most appropriate dosimetry parameters. In the present work, we evaluated the relative importance of different RF sources in the range 0.5–2200 MHz for the exposure of the population in an urban area. We used as reference levels the regulatory guidelines published by ICNIRP (ICNIRP, 1998) for exposure to multiple frequencies based on criteria for limiting electrical stimulation and thermal effects. The study was conducted in the city of Merida, capital of the Region of Extremadura in western Spain. The sources of electromagnetic radiation emissions that could potentially affect the city's inhabitants are mobile telephony antennas within the city limits, and radio and television transmitters at different distances outside the city. Measurements were made with a spectrum analyzer and monopole, biconical, and log-periodic antennas, determining the most significant spectral components of the medium wave, short
wave, and frequency modulation broadcasting bands, and of the television and GSM, DCS, and UMTS mobile telephony frequency bands. The signal measurements were converted to electric field strengths using the appropriate antenna calibration equation, and the magnetic field strengths and equivalent plane-wave power densities were calculated assuming a far-field regime. The electric and magnetic field intensities were used together with the reference levels to determine the exposure quotients corresponding to electrical stimulation and thermal effects. 2. Methods 2.1. Site description The study was conducted in the city of Merida, capital of the Autonomous Region of Extremadura and seat of its government institutions. In 1993 the city was declared a UNESCO World Heritage Site because of its historical monuments and archaeological importance. It is the third centre of population in size in Extremadura, with 53 915 inhabitants according to the 2007 census. It is located approximately in the geographic centre of the region (see Fig. 1), with geographic coordinates 38° 54′ N and 6° 20′ W. It is crossed by the rivers Guadiana and Albarregas, and its height above sea level is 217 m. Within the city, there are mobile telephony base stations responsible for providing the city with the necessary mobile communications coverage. These stations use GSM, DCS, and UMTS technologies. Their locations on the street map of Merida, taken from the information given by the Service of Information on Radioelectric Installations and Exposure Levels (SETSI, 2007), are shown in Fig. 1. The radio and television retransmission antennas provide the city the coverage needed to receive radio and TV channels. These antennas are located outside the urban area due to the high power of the transmitters. The sites with the most important radio and television antennas in terms of their power and proximity to the city are also
Fig. 1. The Region of Extremadura (lower left), with the numbers 1–5 indicating the location of the sites of radio and television broadcasting antennas whose characteristics are detailed in Table 1. On the right is a street plan of Merida, with triangles indicating sites with mobile telephony antennas and circles the sampling points.
J.M. Paniagua et al. / Science of the Total Environment 407 (2009) 4717–4722
4719
Table 1 Characteristics of the radio and TV broadcasting transmitters of the different groups of antennas shown in Fig. 1, and their distance, r, from the city of Mérida. Antenna group 1
r = 10 km
Antenna group 2
FM broadcasting
r = 40 km
TV
f(MHz)
Longitude(°)
Latitude(°)
ERP(kW)
Channels
Longitude(°)
Latitude(°)
ERP(kW)
87.9 90.4 95.6 100.4 103.6 107.4
06W2537 06W2537 06W2533 06W2200 06W1924 06W2034
38N5151 38N5146 38N5145 38N5700 38N5801 38N5507
5 1 1 2 1 0.5
23, 26, 29, 59, 62, 65 FM broadcasting f(MHz) 97.7 99.3 105.3
06W0830
39N1250
150
Longitude(°) 06W0700 06W0700 06W0700
Latitude(°) 39N1300 39N1300 39N1300
ERP(kW) 60 60 60
Antenna group 3
r = 50 km
Antenna group 4
AM broadcasting
r = 50 km
AM broadcasting
f(kHz)
Longitude(°)
Latitude(°)
ERP(kW)
f(kHz)
Longitude(°)
Latitude(°)
ERP(kW)
648 1008 1125 1269
06W5520 06W5558 06W5533 06W5610
38N5318 38N5125 38N5310 38N5155
10 10 10 10
774 1107
06W2025 06W2014
39N2055 39N2049
60 25
Antenna group 5
r = 60 km
AM broadcasting f(kHz)
Longitude(°)
Latitude(°)
ERP(kW)
900
006W2030
39N2726
10
ERP: effective radiated power.
shown in Fig. 1. Their most notable features are detailed in Table 1. As one observes in the table, the sites with AM, FM, and TV antennas are located at distances between 10 and 60 km from the city centre, and their powers (ERP) are between 10 (antenna group 5) and 330 kW (antenna group 2) (RD, 1988, 1993, 2006). 2.2. Materials and measurement procedure Spectrometry was performed using an Agilent EP4404-2B spectrum analyzer which is sensitive in the frequency range 30 Hz–3 GHz and has a dynamic range from − 153 dBm to 30 dBm. The antennas used with this analyzer were of three types: passive monopole for the frequency range 0.5–30 MHz, biconical (30–200 MHz), and logperiodic (200–2200 MHz). The configuration of the spectrum analyzer for measurements in the different frequency ranges is given in Table 2 (ECC, 2003). The duration of the measurements was 6 min, with the spectrum being collected during that time with the antenna at a height of 1.7 m above the ground. Sampling was carried out at 18 outdoor locations in the urban centre of the city (see Fig. 1) corresponding to so-called sensitive spaces — public parks, schools, and hospitals — and in squares and streets used by large numbers of pedestrians. The selection of the zones was independent of whether or not there existed mobile telephony antennas in the vicinity. At each sampling point, seven spectra were collected. Two were with the passive monopole antenna in the ranges 0.5–5 MHz and 5– 30 MHz, corresponding to the medium wave (MW) and short wave (SW) bands, respectively. Another spectrum was collected with the biconical antenna in the range 30–200 MHz corresponding to FM radio. The four remaining spectra were collected with the log-periodic antenna. One was for the range 200–890 MHz, corresponding to analogue and digital television broadcasting. This spectrum was taken with the log-periodic antenna pointing to group 2 of the transmitting Table 2 Configuration of the spectrum analyzer for measurements in the different frequency bands (ECC, 2003). Frequency range
Bandwith
Sweep time
9 kHz–30 MHz 30 MHz–300 MHz 300 MHz–3 GHz
10 kHz 100 kHz 100 kHz
50–100 ms 100 ms 700 ms–1 s
antennas (see Fig. 1 and Table 1), which is the site with the most powerful television transmitters in Extremadura. The other three spectra collected with the log-periodic antenna were for the frequency ranges 925–960 MHz, 1805–1880 MHz, and 1900–2200 MHz, corresponding to emissions from mobile telephony base stations that use GSM, DCS, and UMTS, respectively. These measurements were the most complicated to perform because the log-periodic antenna is directional, and it was necessary to seek the maximum of each spectral component by orienting the antenna and observing the spectrum in situ. After the spectra had been collected, the voltage signals V detected with the spectrum analyzer were converted to electric field strength E by applying Eq. (1): E½dBð μV = mÞ = K ½dBð1 = mÞ + V ½dBð μV Þ
ð1Þ
where K is the antenna factor given in the calibration certificate provided by the manufacturer. With the electric field strength expressed in SI units, the magnetic field strength H and power density S were calculated using the expressions H = E/Z0 and S = E2/Z0. This assumes a far-field regime, a reasonable assumption given that the AM radio transmitters (corresponding to the longest wavelengths involved in the study — between 236 m and 463 m for the frequencies of Table 1) were more than 50 km away from the sampling zone. The power density of each frequency band was calculated as the sum of the power densities of all the emissions of that band. To calculate the levels of exposure to electromagnetic fields to which people are subjected, it is necessary to compare the measured fields with reference levels. To this end, we used the ICNIRP reference levels for E-field and H-field strengths for the general public. Their Table 3 Reference levels for general public exposure to E-field strength and H-field strength in the frequency range 0.15 MHz–300 GHz (ICNIRP, 1998). Frequency range
E-field strength (V/m)
H-field strength (A/m)
0.15–1 MHz 1–10 MHz 10 MHz–400 MHz 400–2000 MHz 2–300 GHz
87 87/f1/2 28 1.375f1/2 61
0.73/f 0.73/f 0.073 0.0037f1/2 0.16
f: frequency (MHz).
4720
J.M. Paniagua et al. / Science of the Total Environment 407 (2009) 4717–4722
Fig. 2. Spectrum collected in the city of Mérida. The groups of emissions are classified as: MW medium wave; SW short wave; FM frequency modulation broadcasting; TV television; and GSM, DCS, and UMTS mobile telephony emissions.
mathematical expressions are given in Table 3 for the range of frequencies used in the present work (ICNIRP, 1998). Having quantified the field strengths and their reference levels, we applied the criteria for limiting exposure to electromagnetic radiation from multiple-frequency sources to prevent electrical stimulation and thermal effects. For the former, which are relevant up to 10 MHz, the exposure quotients should satisfy Eqs. (2) and (3): 1MHz X
Ei + E i = 1Hz L;i
10MHz X
Ei V1 a N 1MHz
i
65kHz X
Hj + H j = 1Hz L; j
ð2Þ
10MHz X i
Hj V1 b N 65kHz
ð3Þ
where Ei is the electric field strength at frequency i, EL,i is the electric field reference level at frequency i, Hj is the magnetic field strength at frequency j, HL,j is the magnetic field reference level at frequency j, and a = 87 V m− 1 and b = 5 A m− 1 both for general public exposure. For thermal considerations, relevant above 100 kHz, the following two requirements (Eqs. (4) and (5)) should be applied to the field levels: 2 Ei + c i = 100kHz 1MHz X
2 Hj + d j = 100kHz 1MHz X
300GHz X i N 1MHz
Ei EL;i
300GHz X i N 1MHz
!2
Hi HL;j
V1
ð4Þ
!2 V1
ð5Þ
figure, we have indicated the various groups of signals: MW (medium wave) in which are found the AM radio emissions indicated in Table 1, SW (short wave) with emissions of transmitters normally used for international broadcasting by governments and private organizations, FM (FM broadcast services), TV (UHF television transmitters), and GSM, DCS, and UMTS (systems providing mobile telephony coverage in the city). The spectrum shown in Fig. 2 is representative of the general trend of the spectra collected in the city. One observes that the signals in the MW, SW, and mobile telephony bands are more intense than those of FM radio and TV. The detection of signals of these last two bands depends on the existence of a direct line of sight between the transmitting antenna and the measuring equipment, so that in general they were sharply reduced within the city due to the presence of buildings and other urban structures. Medium wave and short wave signals, in contrast, propagate primarily as ground and ionospheric waves, respectively (Seybold 2005), and are therefore less affected by attenuation inside the city than when propagation is by line of sight. Unlike the foregoing, the mobile telephony base stations are inside the city. The levels detected within the telephony bands presented greater spatial variability because their detection depended heavily on such factors as the existence of a direct line of sight to the placement of an antenna, the position of the detector with respect to the main radiation lobe, reflections from the ground and walls of buildings, etc. The maximum electric field levels detected were 0.132, 0.203, 0.083, and 0.046 V/m for the peaks of the MW, SW, FM, and TV bands, respectively. These values are relatively low compared with those reported in the literature (see for example the compilation of levels of radiofrequencies by Mantiply et al. (1997), the work on medium-wave and short-wave antennas by Al-Ruwais (1998), and that on FM radio and television antennas by Burch et al. (2006)). The reason for such low electric field levels is that in the present case the radio and television transmitters are located far from the city, while the radiofrequency levels produced by mobile telephony systems are commonly found inside cities (Bergqvist et al. 2001). The maximum value detected in the present study for these telephony bands was 1.14 V/m. In order to present a synthesis of the information provided by the different spectra as simply and clearly as possible, we calculated the equivalent plane-wave power density for each frequency band as described in the Methods section. The measured radiation levels were in the range [16.7–4977.1] µW/m2, with a median value of 80.2 µW/m2. The frequency bands that most contributed to these levels were the three of mobile telephony (henceforth denominated TF) with a 34.8%, SW with 29.4% and MW with 28.4%. Those of FM and TV contributed to a lesser extent with 6.5% and 0.9%, respectively. Fig. 3 shows a box and whisker plot of the power density levels for each band. In this plot,
where Ei is the electric field strength at frequency i, EL,i is the electric field reference level at frequency i, Hj is the magnetic field strength at frequency j, HL,j is the magnetic field reference level at frequency j, and c = 87/f1/2 V m− 1 and d = 0.73/f A m− 1 both for general public exposure (ICNIRP, 1998). 3. Results and discussion 3.1. Radiation levels In the Methods section, it was indicated that seven spectra were collected at each sampling point, with the three antennas covering different frequency ranges. These spectra were merged to form a single spectrum in the range 0.5–2.2 MHz, and thus allow the display of all the emissions simultaneously so as to evaluate their relative importance. Fig. 2 shows by way of example one of these spectra for the electric field taken at a point in the urban centre of Mérida. In this
Fig. 3. Box and whisker plot of the power density levels inside the city according to the different bands: MW medium wave; SW short wave; FM frequency modulation radio; TV television; and TF emissions from GSM, DCS, and UMTS mobile telephony considered together. The maximum value on the Y axis is 150 µW/m2 for better visualization of the figure.
J.M. Paniagua et al. / Science of the Total Environment 407 (2009) 4717–4722
Fig. 4. Normalized spectrum: the electric field values,\ E, of Fig. 2 have been divided by the electric field reference levels, Er, of Table 3.
the central box covers the middle 50% of the data, the lower and upper edges of the box are the lower and upper quartiles, respectively, and the horizontal line drawn through the box is the median. The whiskers extend out to the lowest and highest values of the data (the range). This figure therefore represents a synthesis of the statistics of the power density levels at all the sampling points in the city, showing the minimum and maximum values, median, and 25 and 75 percentiles. One observes in the figure what was noted above, i.e., that the MW, SW, and TF bands are the main contributors to the total radiation levels. In particular, the power density values in the MW band are in the range 2.3–53.1 µW/m2 with median 17.5 µW/m2, in the SW band are in the range 0.5–124.5 µW/m2 with median 10.6 µW/m2, and in the TF band are 0.2–4959.0 µW/m2 with median 18.9 µW/m2. While the medians of these three cases are fairly similar to each other, the maxima are very different, particularly that of the TF band which reaches a much higher value than the other two due to the aforementioned greater spatial variability in power density levels produced by mobile telephony systems. On the contrary, the FM and TV bands have lower power density values, their medians being 1.8 and b0.1 µW/m2 and maxima 25.6 and 20.3 µW/m2, respectively. 3.2. Exposure levels As one observes, the spectral components that most contribute to the radiation levels are those of the TF, MW, and SW bands. Their relative weights in the levels of exposure will be different from those in the raw values of the radiation levels, however, due to the different biological effects of fields of different frequencies. The contribution of each radioelectric emission to the level of exposure will be calculated from the ratio between the measured field level and the reference
4721
level. For example, Fig. 4 shows the electric field levels of Fig. 2 divided by the electric field reference levels of Table 3. In this normalized spectrum, the effect of the 10–400 MHz band emissions on the levels of exposure is enhanced because the corresponding reference level is the lowest, 28 V/m. On the contrary, the effect of emissions in the 0.5– 1 MHz band is reduced because the corresponding reference level is the highest, 87 V/m. One aspect that stands out in Fig. 4 is that the ratios E/Er are well below unity, so that the individual emissions thus meet the basic restrictions set out by the ICNIRP. Nevertheless, according to the recommendations of the ICNIRP, the evaluation of a complex electromagnetic environment with exposure to sources with multiple frequencies must consider independently the thermal and electrical criteria given by Eqs. (2)–(5). Table 4 presents the resulting exposure quotients obtained with those equations for the electric and magnetic fields. With respect to the effects of electrical stimulation, one observes in the table that the coefficients for the electric field vary between 9.6 10− 4 and 37.4 10− 4. The value of the median is 24.7 10− 4, representing 0.247% of the guideline level, i.e., about 400 times lower. The same calculations for the magnetic field strength give a median of 1.14 10− 4, nearly 9000 times below the guideline. The criterion limiting the electric field strength is therefore stricter than that for the magnetic field in avoiding the electrical stimulation of tissues. In both cases, E and H, 87% of the exposure corresponds to the AM radio emissions in the MW band, while the remaining 13% belong to electromagnetic emissions in the SW band whose frequencies are below 10 MHz. The calculation of the thermal effects involves all the emissions between 100 kHz and 300 GHz. The values of the exposure quotient for the electric field vary between 0.03 10− 4 and 10.6 10− 4. The median is 0.16 10− 4, i.e., some 62,500 times below the guideline level. The results for the magnetic field are the same. Fig. 5 shows the percentage contributions of the different frequency bands to this coefficient together with the corresponding percentages of their contributions to the radiation levels. One observes that the SW and TF bands are those with the greatest contributions to the exposure coefficient, with 46.4% and 32.6%, respectively, followed by the FM band with 13.3%, and finally the MW and TV bands with 6.5% and 1.2%, respectively. Hence, the weight of each frequency band in the levels of radiation is different from that in the levels of exposure since the reference levels depend on the frequency. This is especially important in the MW, SW, and FM bands. The first of these contributes much less to the exposure levels than to the radiation levels because its corresponding reference levels are the highest (Table 3). The opposite is the case with the SW and FM bands, for which the reference levels are the lowest (Table 3), and hence the percentage of the exposure level exceeds that of the radiation level. For the TF and TV bands, there is little difference between the levels of exposure and radiation.
Table 4 Exposure quotient calculated according to the ICNIRP standard for electric and magnetic fields using Eqs. (2)–(5) to prevent electrical stimulation and thermal effects. Electrical stimulation Exposure quotient Electric field strength, E (×10− 4) Median 24.7 Minimum 9.6 Maximum 37.4 SD 9.2 Magnetic field strength, H (×10− 4) Median 1.14 Minimum 0.45 Maximum 1.73 SD 0.42 Size 18 SD standard deviation.
Thermal effects Times below
Exposure quotient
Times below
400 1040 270 –
0.16 0.03 10.6 2.5
62500 333000 940 –
8800 22,000 5800 –
0.16 0.03 10.6 2.5 18
62,500 333,000 940 – Fig. 5. Percentage contribution of the different frequency bands to the levels of radiation and of exposure according to the ICNIRP guidelines for thermal effects.
4722
J.M. Paniagua et al. / Science of the Total Environment 407 (2009) 4717–4722
It should be noted that the contribution of the mobile telephony systems represents approximately one third of the exposure due to thermal effects. In absolute terms, the exposure quotients for the three telephony systems are in the range [3.3 10− 8–1.1 10− 3], which in percentages is 3.3 10− 6% to 0.11%. This range is compatible with the [0.002–2]% that telephony base stations represent of the limits set out in the international guidelines according to Fact Sheet 304 of the World Health Organization (WHO, 2006). 4. Conclusions We have described a spectral analysis of the levels of electromagnetic radiation in an urban environment. The mobile telephony antennas located within the city, and the radio and television broadcasting antennas outside, were found to be major sources of non-ionizing radiation that could potentially affect the city's inhabitants. This environment could be described as one of low-level electromagnetic radiation, since the more powerful transmitters are not sufficiently close to the city to produce high levels of radiofrequencies. The radiation levels measured in the city are produced primarily by mobile telephony, short wave, and medium wave emissions, with relative contributions of 34.8%, 29.4%, and 28.4%, respectively. There were more modest contributions from the FM radio (6.5%) and television (0.9%) frequency bands. For the exposure levels corresponding to multiple-frequency sources, considering thermal criteria, the bands with greatest weights were short wave (46.4%) and mobile telephony (32.6%), while the contributions of the FM radio, MW, and television bands were 13.3%, 6.5%, and 1.2%, respectively. The weight of the mobile telephony systems in the exposure levels was approximately one-third of the total. When there exist emissions with frequencies below 10 MHz, AM radio for example, it is necessary to evaluate the exposure levels for multiple-frequency sources using the limiting criterion for avoidance of the electrical stimulation of tissues. The exposure quotient obtained in the present study using this criterion, 24.7 10− 4, was much closer to unity than that using the thermal criterion, 0.16 10− 4. As usual in this type of study, the exposure quotient for thermal effects was thousands of times (in our case 62,500 times) below the reference level, while the same coefficient evaluated to prevent electrical effects was only 400 times lower. The limiting criterion to avoid electrical stimulation is therefore stricter than the thermal criterion, and it is important to take it into consideration if there exist frequencies below 10 MHz. Therefore, in characterizing a complex electromagnetic environment accessible to the general public, knowledge of the content of the radiofrequency spectrum is essential in order to quantify the contribution of each type of emission to the public's exposure. It is not enough to calculate the radiation levels. One has to calculate the exposure levels since they take into account the possible effects on living organisms of the different emission frequencies. It is also necessary to evaluate the electrical effects as well as the thermal effects because the criterion to limit exposure on the basis of the effect of the electrical stimulation of tissues is stricter than that based on thermal effects.
References Ahlbom A, Green A, Kheifets L, Savitz D, Swerdlow A. Epidemiology of health effects of radiofrequency exposure. Environ Health Perspect 2004;112(17):1741–54. Al-Ruwais AS. Measurements of RF radiation near MW and SW radio broadcast stations. IEEE Trans Broadcast 1998;44(4):470–7. Bergqvist U, Friedrich G, Hamnerius Y, Martens L, Neubauer G, Thuroczy G, et al. Mobile telecommunication base stations — exposure to electromagnetic fields. Report of a Short Term Mission within COST 244bis; 2001. Available: http://www.cost281.org (Accessed July 2008). Burch J, Clark M, Yost M, Fitzpatrick C, Bachand A, Ramaprasad J, et al. Radio frequency nonionizing radiation in a community exposed to radio and television broadcasting. Environ Health Perspect 2006;114(2):248–53. Croft RJ, Hamblin DL, Spong J, Wood AW, McKenzie RJ, Stough C. The effect of mobile phone electromagnetic fields on the alpha rhythm of human electroencephalogram. Bioelectromagnetics 2008;29(1):1-10. Dolk H, Shaddick G, Walls P, Grundy C, Thakrar B, Kleinschmidt I, et al. Cancer incidence near radio and television transmitters in Great Britain. Am J Epidemiol 1997;145(1):1–9. ECC (Electronic Communication Committee). Recommendation (02)04: Measuring non-ionising electromagnetic radiation (9 kHz–300 GHz). Working Group “Frequency Management” (FM), Electronic Communications Committee (ECC), European Conference of Postal and Telecommunications Administrations (CEPT). 2003. FCC (Federal Communications Commission). Questions and answers about biological effects and potential hazards of radiofrequency electromagnetic fields. OET Bulletin Washington, DC: Office of Engineering and Technology; 1999. Hocking B, Gordon IR, Grain HL. Cancer incidence and mortality and proximity to TV towers. Med J Aust 1996;165(11–12):599–600. ICNIRP (International Commission on Non-Ionising Radiation Protection). Guidelines for limiting exposure to time-varying electric, magnetic, and electromagnetic fields (up to 300 GHz). Health Phys 1998;74(4):494–522. IEEE (Institute of Electrical and Electronics Engineers). IEEE standard for safety levels with respect to human exposure to radio frequency electromagnetic fields, 3 kHz to 300 GHz. USA: IEEE; 2006. Std. C95.1–2005. Kolodynski A, Kolodynska V. Motor and psychological functions of school children living in the area of the Skrunda radio location station in Latvia. Sci Total Environ 1996;180(1):87–93. Mantiply E, Pohl K, Poppoell S, Murphy J. Summary of measured radiofrequency electric and magnetic fields (10 kHz to 30 GHz) in the general and work environment. Bioelectromagnetics 1997;18:563–77. Michelozzi P, Capon A, Kirchmayer U, Forastiere F, Biggeri A, Barca A, et al. Adult and childhood leukemia near a high-power radio station in Rome, Italy. Am J Epidemiol 2002;155(12):1096–103. RD (Real Decreto). 1988. Real Decreto 1362/1988, de 11 de noviembre, por el que se aprueba el Plan Técnico Nacional de la Televisión Privada. B.O.E. num 275 de 16 de noviembre de 1988. In Spanish. RD (Real Decreto). 1993. Real Decreto 765/1993, de 21 de mayo, por el que se aprueba el Plan Técnico Nacional de Radiodifusión Sonora en Ondas Medias (hectométricas). B.O.E. de 17 de julio de 1993. In Spanish. RD (Real Decreto). 2006. Real Decreto 964/2006, de 1 de septiembre, por el que se aprueba el Plan Técnico Nacional de Radiodifusión Sonora en ondas métricas con modulación de frecuencia. B.O.E. num 233 de 18 de septiembre de 2006. In Spanish. SETSI (Secretaría de Estado de Telecomunicaciones y para la Sociedad de la Información). Servicio de Información sobre Instalaciones Radioeléctricas y Niveles de Exposición. Ministerio de Industria, Turismo y Comercio; 2007. Available: http://www.mityc.es/nivelesexposicion. (Accessed July 2008). In Spanish. Seybold JS. Introduction to RF propagation. Hoboken, New Jersey: John Wiley & Sons, Inc; 2005. 349 pp. WHO (World Health Organization). Electromagnetic fields and public health. Base stations and wireless technologies. Fact sheet n° 304; 2006. May, Available: http://www.who. int/mediacentre/factsheets/fs304/en/index.html (Accessed July 2008). Wolf R, Wolf D. Increased incidence of cancer near a cell-phone transmitter station. Int J Cancer Prev 2004;1(2):123–8.
Contents lists available at ScienceDirect
Science of the Total Environment j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / s c i t o t e n v
Electrical stimulation vs thermal effects in a complex electromagnetic environment Jesús M. Paniagua ⁎, Montaña Rufo, Antonio Jiménez, Alicia Antolín, Miguel Sánchez Department of Applied Physics, Polytechnic School, University of Extremadura. Avda. de la Universidad s/n, 10071 Cáceres, Spain
a r t i c l e
i n f o
Article history: Received 15 December 2008 Received in revised form 31 March 2009 Accepted 21 April 2009 Available online 28 May 2009 Keywords: Electrical stimulation Exposure levels Radiation levels Spectral analysis Urban environment Thermal effects
a b s t r a c t Studies linking exposure to low levels of radiofrequencies with adverse health effects, notwithstanding their present apparent inconsistency, have contributed to a steady improvement in the quality of evaluating that exposure. In complex electromagnetic environments, with a multitude of emissions of different frequencies acting simultaneously, knowledge of the spectral content is fundamental to evaluating human exposure to non-ionizing radiation. In the present work, we quantify the most significant spectral components in the frequency band 0.5–2200 MHz in an urban area. The measurements were made with a spectrum analyzer and monopole, biconical, and log-periodic antennas. Power density levels were calculated separately for the medium wave, short wave, and frequency modulation radio broadcasting bands, and for the television and GSM, DCS, and UMTS mobile telephony bands. The measured levels were compared with the ICNIRP reference levels for exposure to multiple frequency sources for thermal effects and electrical stimulation. The results showed the criterion limiting exposure on the basis of preventing electrical stimulation of peripheral nerves and muscles to be stricter (exposure quotient 24.7 10− 4) than that based on thermal considerations (exposure quotient 0.16 10− 4). The bands that contribute most to the latter are short wave, with 46.2%, and mobile telephony with 32.6% of the total exposure. In a complex electromagnetic environment, knowledge of the radiofrequency spectrum is essential in order to quantify the contribution of each type of emission to the public's exposure. It is also necessary to evaluate the electrical effects as well as the thermal effects because the criterion to limit exposure on the basis of the effect of the electrical stimulation of tissues is stricter than that based on thermal effects. © 2009 Elsevier B.V. All rights reserved.
1. Introduction Environmental levels of electromagnetic radiation have been constantly increasing due to the continual implementation of new telecommunications systems. The technical evolution of mobile telephony, and the appearance of wifi and wireless systems, have generated radiofrequency (RF) emissions in addition to those that had existed for decades from radio and television broadcasting antennas. The sudden implementation of all these telecommunication devices, together with the constant stream of studies linking electromagnetic fields with adverse health effects, have led to concern among the public at large over the possible risks of exposure to nonionizing radiation. As a result, various national and international agencies, such as the International Commission on Non-ionizing Radiation Protection (ICNIRP), the Federal Communications Commission (FFC), and the Institute of Electrical and Electronics Engineers (IEEE), have prepared regulatory guidelines to limit exposure to electromagnetic fields (FCC, 1999; ICNIRP, 1998; IEEE, 2006) based primarily on criteria of thermal effects. There have been studies, however, that have detected other nonthermal effects. For example, Croft et al. (2008) in studies performed ⁎ Corresponding author. Tel.: +34 927257597; fax: +34 927257203. E-mail address: [email protected] (J.M. Paniagua). 0048-9697/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.scitotenv.2009.04.034
on volunteers find effects caused by cell-phones on the alpha rhythm in electroencephalograms. Kolodynski and Kolodynska (1996) find reduced development of memory, attention, and motor functions in children living near a radio station. And other studies report increased incidence of various types of cancer in people living near television towers (Hocking et al., 1996), TV and FM radio transmitters (Dolk et al., 1997), medium and short wave radio transmitters (Michelozzi et al., 2002), and mobile telephony antennas (Wolf and Wolf, 2004). A exhaustive review of epidemiological studies of the effects of RF fields on human health was published by Ahlbom et al. (2004), with reports of high rates of cancer in some cases. Those same authors indicate, however, that the results of epidemiological studies do not as yet provide consistent or convincing evidence of a causal relationship between RF exposure and adverse health effects, and furthermore that those studies have too many deficiencies for it to be possible to establish any association (Ahlbom et al., 2004). A key aspect in all studies of this type is the quality of the evaluation of RF exposure. There are often situations in which the determination of the degree of compliance with standards for protection against electromagnetic fields is difficult and cannot always be done directly. The spectral content, spatial and temporal patterns, and polarization are some of the factors in the electromagnetic environment that may be important in evaluating a biological effect. Despite the rapid growth of the new technologies,
4718
J.M. Paniagua et al. / Science of the Total Environment 407 (2009) 4717–4722
little is known about public exposure to radiofrequencies and even less about the relative importance of different sources. This is particularly relevant in complex electromagnetic environments in which a multitude of emissions of different frequencies are acting simultaneously. It is therefore necessary to improve the methods of measurement so that the spectral components can be quantified, and hence the measured field levels compared with the reference levels for different frequencies, since the potential effects on the organism will be different for different frequencies. Equally important is to use criteria for calculating exposure levels that take into account the possible additivity of the effects. For example, in its regulatory guidelines, ICNIRP (ICNIRP, 1998) states that additivity should be examined separately to limit electrical stimulation and thermal effects. The effects of electrical stimulation predominate over other possible effects in the low frequency region, up to 10 MHz, where the current density J is the dosimetry parameter that is applied. For frequencies between 100 kHz and 300 GHz, thermal effects predominate, and from 10 MHz to 10 GHz the specific absorption rate, SAR, and from 10 GHz to 300 GHz the incident power density are the most appropriate dosimetry parameters. In the present work, we evaluated the relative importance of different RF sources in the range 0.5–2200 MHz for the exposure of the population in an urban area. We used as reference levels the regulatory guidelines published by ICNIRP (ICNIRP, 1998) for exposure to multiple frequencies based on criteria for limiting electrical stimulation and thermal effects. The study was conducted in the city of Merida, capital of the Region of Extremadura in western Spain. The sources of electromagnetic radiation emissions that could potentially affect the city's inhabitants are mobile telephony antennas within the city limits, and radio and television transmitters at different distances outside the city. Measurements were made with a spectrum analyzer and monopole, biconical, and log-periodic antennas, determining the most significant spectral components of the medium wave, short
wave, and frequency modulation broadcasting bands, and of the television and GSM, DCS, and UMTS mobile telephony frequency bands. The signal measurements were converted to electric field strengths using the appropriate antenna calibration equation, and the magnetic field strengths and equivalent plane-wave power densities were calculated assuming a far-field regime. The electric and magnetic field intensities were used together with the reference levels to determine the exposure quotients corresponding to electrical stimulation and thermal effects. 2. Methods 2.1. Site description The study was conducted in the city of Merida, capital of the Autonomous Region of Extremadura and seat of its government institutions. In 1993 the city was declared a UNESCO World Heritage Site because of its historical monuments and archaeological importance. It is the third centre of population in size in Extremadura, with 53 915 inhabitants according to the 2007 census. It is located approximately in the geographic centre of the region (see Fig. 1), with geographic coordinates 38° 54′ N and 6° 20′ W. It is crossed by the rivers Guadiana and Albarregas, and its height above sea level is 217 m. Within the city, there are mobile telephony base stations responsible for providing the city with the necessary mobile communications coverage. These stations use GSM, DCS, and UMTS technologies. Their locations on the street map of Merida, taken from the information given by the Service of Information on Radioelectric Installations and Exposure Levels (SETSI, 2007), are shown in Fig. 1. The radio and television retransmission antennas provide the city the coverage needed to receive radio and TV channels. These antennas are located outside the urban area due to the high power of the transmitters. The sites with the most important radio and television antennas in terms of their power and proximity to the city are also
Fig. 1. The Region of Extremadura (lower left), with the numbers 1–5 indicating the location of the sites of radio and television broadcasting antennas whose characteristics are detailed in Table 1. On the right is a street plan of Merida, with triangles indicating sites with mobile telephony antennas and circles the sampling points.
J.M. Paniagua et al. / Science of the Total Environment 407 (2009) 4717–4722
4719
Table 1 Characteristics of the radio and TV broadcasting transmitters of the different groups of antennas shown in Fig. 1, and their distance, r, from the city of Mérida. Antenna group 1
r = 10 km
Antenna group 2
FM broadcasting
r = 40 km
TV
f(MHz)
Longitude(°)
Latitude(°)
ERP(kW)
Channels
Longitude(°)
Latitude(°)
ERP(kW)
87.9 90.4 95.6 100.4 103.6 107.4
06W2537 06W2537 06W2533 06W2200 06W1924 06W2034
38N5151 38N5146 38N5145 38N5700 38N5801 38N5507
5 1 1 2 1 0.5
23, 26, 29, 59, 62, 65 FM broadcasting f(MHz) 97.7 99.3 105.3
06W0830
39N1250
150
Longitude(°) 06W0700 06W0700 06W0700
Latitude(°) 39N1300 39N1300 39N1300
ERP(kW) 60 60 60
Antenna group 3
r = 50 km
Antenna group 4
AM broadcasting
r = 50 km
AM broadcasting
f(kHz)
Longitude(°)
Latitude(°)
ERP(kW)
f(kHz)
Longitude(°)
Latitude(°)
ERP(kW)
648 1008 1125 1269
06W5520 06W5558 06W5533 06W5610
38N5318 38N5125 38N5310 38N5155
10 10 10 10
774 1107
06W2025 06W2014
39N2055 39N2049
60 25
Antenna group 5
r = 60 km
AM broadcasting f(kHz)
Longitude(°)
Latitude(°)
ERP(kW)
900
006W2030
39N2726
10
ERP: effective radiated power.
shown in Fig. 1. Their most notable features are detailed in Table 1. As one observes in the table, the sites with AM, FM, and TV antennas are located at distances between 10 and 60 km from the city centre, and their powers (ERP) are between 10 (antenna group 5) and 330 kW (antenna group 2) (RD, 1988, 1993, 2006). 2.2. Materials and measurement procedure Spectrometry was performed using an Agilent EP4404-2B spectrum analyzer which is sensitive in the frequency range 30 Hz–3 GHz and has a dynamic range from − 153 dBm to 30 dBm. The antennas used with this analyzer were of three types: passive monopole for the frequency range 0.5–30 MHz, biconical (30–200 MHz), and logperiodic (200–2200 MHz). The configuration of the spectrum analyzer for measurements in the different frequency ranges is given in Table 2 (ECC, 2003). The duration of the measurements was 6 min, with the spectrum being collected during that time with the antenna at a height of 1.7 m above the ground. Sampling was carried out at 18 outdoor locations in the urban centre of the city (see Fig. 1) corresponding to so-called sensitive spaces — public parks, schools, and hospitals — and in squares and streets used by large numbers of pedestrians. The selection of the zones was independent of whether or not there existed mobile telephony antennas in the vicinity. At each sampling point, seven spectra were collected. Two were with the passive monopole antenna in the ranges 0.5–5 MHz and 5– 30 MHz, corresponding to the medium wave (MW) and short wave (SW) bands, respectively. Another spectrum was collected with the biconical antenna in the range 30–200 MHz corresponding to FM radio. The four remaining spectra were collected with the log-periodic antenna. One was for the range 200–890 MHz, corresponding to analogue and digital television broadcasting. This spectrum was taken with the log-periodic antenna pointing to group 2 of the transmitting Table 2 Configuration of the spectrum analyzer for measurements in the different frequency bands (ECC, 2003). Frequency range
Bandwith
Sweep time
9 kHz–30 MHz 30 MHz–300 MHz 300 MHz–3 GHz
10 kHz 100 kHz 100 kHz
50–100 ms 100 ms 700 ms–1 s
antennas (see Fig. 1 and Table 1), which is the site with the most powerful television transmitters in Extremadura. The other three spectra collected with the log-periodic antenna were for the frequency ranges 925–960 MHz, 1805–1880 MHz, and 1900–2200 MHz, corresponding to emissions from mobile telephony base stations that use GSM, DCS, and UMTS, respectively. These measurements were the most complicated to perform because the log-periodic antenna is directional, and it was necessary to seek the maximum of each spectral component by orienting the antenna and observing the spectrum in situ. After the spectra had been collected, the voltage signals V detected with the spectrum analyzer were converted to electric field strength E by applying Eq. (1): E½dBð μV = mÞ = K ½dBð1 = mÞ + V ½dBð μV Þ
ð1Þ
where K is the antenna factor given in the calibration certificate provided by the manufacturer. With the electric field strength expressed in SI units, the magnetic field strength H and power density S were calculated using the expressions H = E/Z0 and S = E2/Z0. This assumes a far-field regime, a reasonable assumption given that the AM radio transmitters (corresponding to the longest wavelengths involved in the study — between 236 m and 463 m for the frequencies of Table 1) were more than 50 km away from the sampling zone. The power density of each frequency band was calculated as the sum of the power densities of all the emissions of that band. To calculate the levels of exposure to electromagnetic fields to which people are subjected, it is necessary to compare the measured fields with reference levels. To this end, we used the ICNIRP reference levels for E-field and H-field strengths for the general public. Their Table 3 Reference levels for general public exposure to E-field strength and H-field strength in the frequency range 0.15 MHz–300 GHz (ICNIRP, 1998). Frequency range
E-field strength (V/m)
H-field strength (A/m)
0.15–1 MHz 1–10 MHz 10 MHz–400 MHz 400–2000 MHz 2–300 GHz
87 87/f1/2 28 1.375f1/2 61
0.73/f 0.73/f 0.073 0.0037f1/2 0.16
f: frequency (MHz).
4720
J.M. Paniagua et al. / Science of the Total Environment 407 (2009) 4717–4722
Fig. 2. Spectrum collected in the city of Mérida. The groups of emissions are classified as: MW medium wave; SW short wave; FM frequency modulation broadcasting; TV television; and GSM, DCS, and UMTS mobile telephony emissions.
mathematical expressions are given in Table 3 for the range of frequencies used in the present work (ICNIRP, 1998). Having quantified the field strengths and their reference levels, we applied the criteria for limiting exposure to electromagnetic radiation from multiple-frequency sources to prevent electrical stimulation and thermal effects. For the former, which are relevant up to 10 MHz, the exposure quotients should satisfy Eqs. (2) and (3): 1MHz X
Ei + E i = 1Hz L;i
10MHz X
Ei V1 a N 1MHz
i
65kHz X
Hj + H j = 1Hz L; j
ð2Þ
10MHz X i
Hj V1 b N 65kHz
ð3Þ
where Ei is the electric field strength at frequency i, EL,i is the electric field reference level at frequency i, Hj is the magnetic field strength at frequency j, HL,j is the magnetic field reference level at frequency j, and a = 87 V m− 1 and b = 5 A m− 1 both for general public exposure. For thermal considerations, relevant above 100 kHz, the following two requirements (Eqs. (4) and (5)) should be applied to the field levels: 2 Ei + c i = 100kHz 1MHz X
2 Hj + d j = 100kHz 1MHz X
300GHz X i N 1MHz
Ei EL;i
300GHz X i N 1MHz
!2
Hi HL;j
V1
ð4Þ
!2 V1
ð5Þ
figure, we have indicated the various groups of signals: MW (medium wave) in which are found the AM radio emissions indicated in Table 1, SW (short wave) with emissions of transmitters normally used for international broadcasting by governments and private organizations, FM (FM broadcast services), TV (UHF television transmitters), and GSM, DCS, and UMTS (systems providing mobile telephony coverage in the city). The spectrum shown in Fig. 2 is representative of the general trend of the spectra collected in the city. One observes that the signals in the MW, SW, and mobile telephony bands are more intense than those of FM radio and TV. The detection of signals of these last two bands depends on the existence of a direct line of sight between the transmitting antenna and the measuring equipment, so that in general they were sharply reduced within the city due to the presence of buildings and other urban structures. Medium wave and short wave signals, in contrast, propagate primarily as ground and ionospheric waves, respectively (Seybold 2005), and are therefore less affected by attenuation inside the city than when propagation is by line of sight. Unlike the foregoing, the mobile telephony base stations are inside the city. The levels detected within the telephony bands presented greater spatial variability because their detection depended heavily on such factors as the existence of a direct line of sight to the placement of an antenna, the position of the detector with respect to the main radiation lobe, reflections from the ground and walls of buildings, etc. The maximum electric field levels detected were 0.132, 0.203, 0.083, and 0.046 V/m for the peaks of the MW, SW, FM, and TV bands, respectively. These values are relatively low compared with those reported in the literature (see for example the compilation of levels of radiofrequencies by Mantiply et al. (1997), the work on medium-wave and short-wave antennas by Al-Ruwais (1998), and that on FM radio and television antennas by Burch et al. (2006)). The reason for such low electric field levels is that in the present case the radio and television transmitters are located far from the city, while the radiofrequency levels produced by mobile telephony systems are commonly found inside cities (Bergqvist et al. 2001). The maximum value detected in the present study for these telephony bands was 1.14 V/m. In order to present a synthesis of the information provided by the different spectra as simply and clearly as possible, we calculated the equivalent plane-wave power density for each frequency band as described in the Methods section. The measured radiation levels were in the range [16.7–4977.1] µW/m2, with a median value of 80.2 µW/m2. The frequency bands that most contributed to these levels were the three of mobile telephony (henceforth denominated TF) with a 34.8%, SW with 29.4% and MW with 28.4%. Those of FM and TV contributed to a lesser extent with 6.5% and 0.9%, respectively. Fig. 3 shows a box and whisker plot of the power density levels for each band. In this plot,
where Ei is the electric field strength at frequency i, EL,i is the electric field reference level at frequency i, Hj is the magnetic field strength at frequency j, HL,j is the magnetic field reference level at frequency j, and c = 87/f1/2 V m− 1 and d = 0.73/f A m− 1 both for general public exposure (ICNIRP, 1998). 3. Results and discussion 3.1. Radiation levels In the Methods section, it was indicated that seven spectra were collected at each sampling point, with the three antennas covering different frequency ranges. These spectra were merged to form a single spectrum in the range 0.5–2.2 MHz, and thus allow the display of all the emissions simultaneously so as to evaluate their relative importance. Fig. 2 shows by way of example one of these spectra for the electric field taken at a point in the urban centre of Mérida. In this
Fig. 3. Box and whisker plot of the power density levels inside the city according to the different bands: MW medium wave; SW short wave; FM frequency modulation radio; TV television; and TF emissions from GSM, DCS, and UMTS mobile telephony considered together. The maximum value on the Y axis is 150 µW/m2 for better visualization of the figure.
J.M. Paniagua et al. / Science of the Total Environment 407 (2009) 4717–4722
Fig. 4. Normalized spectrum: the electric field values,\ E, of Fig. 2 have been divided by the electric field reference levels, Er, of Table 3.
the central box covers the middle 50% of the data, the lower and upper edges of the box are the lower and upper quartiles, respectively, and the horizontal line drawn through the box is the median. The whiskers extend out to the lowest and highest values of the data (the range). This figure therefore represents a synthesis of the statistics of the power density levels at all the sampling points in the city, showing the minimum and maximum values, median, and 25 and 75 percentiles. One observes in the figure what was noted above, i.e., that the MW, SW, and TF bands are the main contributors to the total radiation levels. In particular, the power density values in the MW band are in the range 2.3–53.1 µW/m2 with median 17.5 µW/m2, in the SW band are in the range 0.5–124.5 µW/m2 with median 10.6 µW/m2, and in the TF band are 0.2–4959.0 µW/m2 with median 18.9 µW/m2. While the medians of these three cases are fairly similar to each other, the maxima are very different, particularly that of the TF band which reaches a much higher value than the other two due to the aforementioned greater spatial variability in power density levels produced by mobile telephony systems. On the contrary, the FM and TV bands have lower power density values, their medians being 1.8 and b0.1 µW/m2 and maxima 25.6 and 20.3 µW/m2, respectively. 3.2. Exposure levels As one observes, the spectral components that most contribute to the radiation levels are those of the TF, MW, and SW bands. Their relative weights in the levels of exposure will be different from those in the raw values of the radiation levels, however, due to the different biological effects of fields of different frequencies. The contribution of each radioelectric emission to the level of exposure will be calculated from the ratio between the measured field level and the reference
4721
level. For example, Fig. 4 shows the electric field levels of Fig. 2 divided by the electric field reference levels of Table 3. In this normalized spectrum, the effect of the 10–400 MHz band emissions on the levels of exposure is enhanced because the corresponding reference level is the lowest, 28 V/m. On the contrary, the effect of emissions in the 0.5– 1 MHz band is reduced because the corresponding reference level is the highest, 87 V/m. One aspect that stands out in Fig. 4 is that the ratios E/Er are well below unity, so that the individual emissions thus meet the basic restrictions set out by the ICNIRP. Nevertheless, according to the recommendations of the ICNIRP, the evaluation of a complex electromagnetic environment with exposure to sources with multiple frequencies must consider independently the thermal and electrical criteria given by Eqs. (2)–(5). Table 4 presents the resulting exposure quotients obtained with those equations for the electric and magnetic fields. With respect to the effects of electrical stimulation, one observes in the table that the coefficients for the electric field vary between 9.6 10− 4 and 37.4 10− 4. The value of the median is 24.7 10− 4, representing 0.247% of the guideline level, i.e., about 400 times lower. The same calculations for the magnetic field strength give a median of 1.14 10− 4, nearly 9000 times below the guideline. The criterion limiting the electric field strength is therefore stricter than that for the magnetic field in avoiding the electrical stimulation of tissues. In both cases, E and H, 87% of the exposure corresponds to the AM radio emissions in the MW band, while the remaining 13% belong to electromagnetic emissions in the SW band whose frequencies are below 10 MHz. The calculation of the thermal effects involves all the emissions between 100 kHz and 300 GHz. The values of the exposure quotient for the electric field vary between 0.03 10− 4 and 10.6 10− 4. The median is 0.16 10− 4, i.e., some 62,500 times below the guideline level. The results for the magnetic field are the same. Fig. 5 shows the percentage contributions of the different frequency bands to this coefficient together with the corresponding percentages of their contributions to the radiation levels. One observes that the SW and TF bands are those with the greatest contributions to the exposure coefficient, with 46.4% and 32.6%, respectively, followed by the FM band with 13.3%, and finally the MW and TV bands with 6.5% and 1.2%, respectively. Hence, the weight of each frequency band in the levels of radiation is different from that in the levels of exposure since the reference levels depend on the frequency. This is especially important in the MW, SW, and FM bands. The first of these contributes much less to the exposure levels than to the radiation levels because its corresponding reference levels are the highest (Table 3). The opposite is the case with the SW and FM bands, for which the reference levels are the lowest (Table 3), and hence the percentage of the exposure level exceeds that of the radiation level. For the TF and TV bands, there is little difference between the levels of exposure and radiation.
Table 4 Exposure quotient calculated according to the ICNIRP standard for electric and magnetic fields using Eqs. (2)–(5) to prevent electrical stimulation and thermal effects. Electrical stimulation Exposure quotient Electric field strength, E (×10− 4) Median 24.7 Minimum 9.6 Maximum 37.4 SD 9.2 Magnetic field strength, H (×10− 4) Median 1.14 Minimum 0.45 Maximum 1.73 SD 0.42 Size 18 SD standard deviation.
Thermal effects Times below
Exposure quotient
Times below
400 1040 270 –
0.16 0.03 10.6 2.5
62500 333000 940 –
8800 22,000 5800 –
0.16 0.03 10.6 2.5 18
62,500 333,000 940 – Fig. 5. Percentage contribution of the different frequency bands to the levels of radiation and of exposure according to the ICNIRP guidelines for thermal effects.
4722
J.M. Paniagua et al. / Science of the Total Environment 407 (2009) 4717–4722
It should be noted that the contribution of the mobile telephony systems represents approximately one third of the exposure due to thermal effects. In absolute terms, the exposure quotients for the three telephony systems are in the range [3.3 10− 8–1.1 10− 3], which in percentages is 3.3 10− 6% to 0.11%. This range is compatible with the [0.002–2]% that telephony base stations represent of the limits set out in the international guidelines according to Fact Sheet 304 of the World Health Organization (WHO, 2006). 4. Conclusions We have described a spectral analysis of the levels of electromagnetic radiation in an urban environment. The mobile telephony antennas located within the city, and the radio and television broadcasting antennas outside, were found to be major sources of non-ionizing radiation that could potentially affect the city's inhabitants. This environment could be described as one of low-level electromagnetic radiation, since the more powerful transmitters are not sufficiently close to the city to produce high levels of radiofrequencies. The radiation levels measured in the city are produced primarily by mobile telephony, short wave, and medium wave emissions, with relative contributions of 34.8%, 29.4%, and 28.4%, respectively. There were more modest contributions from the FM radio (6.5%) and television (0.9%) frequency bands. For the exposure levels corresponding to multiple-frequency sources, considering thermal criteria, the bands with greatest weights were short wave (46.4%) and mobile telephony (32.6%), while the contributions of the FM radio, MW, and television bands were 13.3%, 6.5%, and 1.2%, respectively. The weight of the mobile telephony systems in the exposure levels was approximately one-third of the total. When there exist emissions with frequencies below 10 MHz, AM radio for example, it is necessary to evaluate the exposure levels for multiple-frequency sources using the limiting criterion for avoidance of the electrical stimulation of tissues. The exposure quotient obtained in the present study using this criterion, 24.7 10− 4, was much closer to unity than that using the thermal criterion, 0.16 10− 4. As usual in this type of study, the exposure quotient for thermal effects was thousands of times (in our case 62,500 times) below the reference level, while the same coefficient evaluated to prevent electrical effects was only 400 times lower. The limiting criterion to avoid electrical stimulation is therefore stricter than the thermal criterion, and it is important to take it into consideration if there exist frequencies below 10 MHz. Therefore, in characterizing a complex electromagnetic environment accessible to the general public, knowledge of the content of the radiofrequency spectrum is essential in order to quantify the contribution of each type of emission to the public's exposure. It is not enough to calculate the radiation levels. One has to calculate the exposure levels since they take into account the possible effects on living organisms of the different emission frequencies. It is also necessary to evaluate the electrical effects as well as the thermal effects because the criterion to limit exposure on the basis of the effect of the electrical stimulation of tissues is stricter than that based on thermal effects.
References Ahlbom A, Green A, Kheifets L, Savitz D, Swerdlow A. Epidemiology of health effects of radiofrequency exposure. Environ Health Perspect 2004;112(17):1741–54. Al-Ruwais AS. Measurements of RF radiation near MW and SW radio broadcast stations. IEEE Trans Broadcast 1998;44(4):470–7. Bergqvist U, Friedrich G, Hamnerius Y, Martens L, Neubauer G, Thuroczy G, et al. Mobile telecommunication base stations — exposure to electromagnetic fields. Report of a Short Term Mission within COST 244bis; 2001. Available: http://www.cost281.org (Accessed July 2008). Burch J, Clark M, Yost M, Fitzpatrick C, Bachand A, Ramaprasad J, et al. Radio frequency nonionizing radiation in a community exposed to radio and television broadcasting. Environ Health Perspect 2006;114(2):248–53. Croft RJ, Hamblin DL, Spong J, Wood AW, McKenzie RJ, Stough C. The effect of mobile phone electromagnetic fields on the alpha rhythm of human electroencephalogram. Bioelectromagnetics 2008;29(1):1-10. Dolk H, Shaddick G, Walls P, Grundy C, Thakrar B, Kleinschmidt I, et al. Cancer incidence near radio and television transmitters in Great Britain. Am J Epidemiol 1997;145(1):1–9. ECC (Electronic Communication Committee). Recommendation (02)04: Measuring non-ionising electromagnetic radiation (9 kHz–300 GHz). Working Group “Frequency Management” (FM), Electronic Communications Committee (ECC), European Conference of Postal and Telecommunications Administrations (CEPT). 2003. FCC (Federal Communications Commission). Questions and answers about biological effects and potential hazards of radiofrequency electromagnetic fields. OET Bulletin Washington, DC: Office of Engineering and Technology; 1999. Hocking B, Gordon IR, Grain HL. Cancer incidence and mortality and proximity to TV towers. Med J Aust 1996;165(11–12):599–600. ICNIRP (International Commission on Non-Ionising Radiation Protection). Guidelines for limiting exposure to time-varying electric, magnetic, and electromagnetic fields (up to 300 GHz). Health Phys 1998;74(4):494–522. IEEE (Institute of Electrical and Electronics Engineers). IEEE standard for safety levels with respect to human exposure to radio frequency electromagnetic fields, 3 kHz to 300 GHz. USA: IEEE; 2006. Std. C95.1–2005. Kolodynski A, Kolodynska V. Motor and psychological functions of school children living in the area of the Skrunda radio location station in Latvia. Sci Total Environ 1996;180(1):87–93. Mantiply E, Pohl K, Poppoell S, Murphy J. Summary of measured radiofrequency electric and magnetic fields (10 kHz to 30 GHz) in the general and work environment. Bioelectromagnetics 1997;18:563–77. Michelozzi P, Capon A, Kirchmayer U, Forastiere F, Biggeri A, Barca A, et al. Adult and childhood leukemia near a high-power radio station in Rome, Italy. Am J Epidemiol 2002;155(12):1096–103. RD (Real Decreto). 1988. Real Decreto 1362/1988, de 11 de noviembre, por el que se aprueba el Plan Técnico Nacional de la Televisión Privada. B.O.E. num 275 de 16 de noviembre de 1988. In Spanish. RD (Real Decreto). 1993. Real Decreto 765/1993, de 21 de mayo, por el que se aprueba el Plan Técnico Nacional de Radiodifusión Sonora en Ondas Medias (hectométricas). B.O.E. de 17 de julio de 1993. In Spanish. RD (Real Decreto). 2006. Real Decreto 964/2006, de 1 de septiembre, por el que se aprueba el Plan Técnico Nacional de Radiodifusión Sonora en ondas métricas con modulación de frecuencia. B.O.E. num 233 de 18 de septiembre de 2006. In Spanish. SETSI (Secretaría de Estado de Telecomunicaciones y para la Sociedad de la Información). Servicio de Información sobre Instalaciones Radioeléctricas y Niveles de Exposición. Ministerio de Industria, Turismo y Comercio; 2007. Available: http://www.mityc.es/nivelesexposicion. (Accessed July 2008). In Spanish. Seybold JS. Introduction to RF propagation. Hoboken, New Jersey: John Wiley & Sons, Inc; 2005. 349 pp. WHO (World Health Organization). Electromagnetic fields and public health. Base stations and wireless technologies. Fact sheet n° 304; 2006. May, Available: http://www.who. int/mediacentre/factsheets/fs304/en/index.html (Accessed July 2008). Wolf R, Wolf D. Increased incidence of cancer near a cell-phone transmitter station. Int J Cancer Prev 2004;1(2):123–8.