Science, Technology, Environment and Resources Group
The acceptance of mobile phones in Australia has been phenomenal, a total of about four and a half million being presently in use. However, not so welcome for many people has been the sprouting of mobile telephone towers in unexpected places close to homes and schools. There are now about 2000 of them. It is reported that expanding phone companies in the US are hiding the antennae in church steeples, arena lighting, artificial trees and flagpoles. It is the newness and the close proximity of these towers that has made them more controversial than the established radio and TV towers. However, all transmit electromagnetic radiation (often referred to by officials as 'electromagnetic energy' in order to avoid the term 'radiation') which some scientists have implicated in increased incidence of cancer.
Undoubtedly there has been an aesthetic angle to the debate on mobile phone tower placement; some residents find them very ugly and likely to depress house values for that reason alone. But a Four Corners program in July 1995 alerted many Australians for the first time to the possible health effects not only of high-power transmitters but of mobile phone use. Anecdotal but still compelling accounts of cancer association with exposure to transmitters and mobile phone use featured in the program. A CSIRO report of the previous year(1) had urged that more research on health effects be carried out. Also in 1995, a preliminary study of cancer incidence in Sydney appeared to show an increase of childhood leukaemia in homes relatively close to TV transmitters(2). Meanwhile, there has been a controversial move to have the existing Australian radiation standard loosened by a factor of five in order to bring it into line with overseas standards.
This paper is intended to provide background on the two-year Australian debate on the possible hazards of electromagnetic radiation from transmitter towers. Of immediate importance is the prospect of looser national electromagnetic radiation standards, which raises questions as to the validity of the basis for such standards in terms of what laboratory or other results have been relied on for setting standards. The relative energy of radiation received from transmitter towers compared with hand-held mobile phones is relevant and is discussed. So also is the range of reported laboratory effects on test animals and cells observed at very low levels of radiation near the standard or less; are they meaningful? The paper concludes with a suggested approach to experimental work which may help us to determine whether Australian and world standards are soundly based or not.
For an understanding of the issues involved, it is necessary to have some knowledge of the range and nature of the electromagnetic radiation (EMR) spectrum. Electromagnetic radiation may be thought of in terms of waves in air which transmit energy but can also be modulated (controlled) through amplitude, pulsing, etc. to transmit speech, TV images and so on. These waves have a range or spectrum of frequency expressed in hertz, i.e. cycles per second. At the higher frequencies we have kilohertz, megahertz and gigahertz. The greater the frequency, the shorter the wavelength and the greater the energy transmitted.
A significant division within the EMR spectrum is the frequency at about 10 million gigahertz above which waves become ionising in nature, i.e. they are capable of knocking electrons out of atoms to form ions. Thus ultraviolet rays, X-rays and gamma radiation are ionising because they are of greater frequency than 10 million gigahertz. When directed at the body, such radiation is known to be capable of initiating cancer through damage to genetic material (DNA). Too much sunlight, too many X-rays or too much exposure to the gamma-radiating isotope cobalt-60 can cause cancer.
That part of the EMR spectrum of concern in this paper is non-ionising and is known as radiofrequency/microwave radiation (RF radiation for short). This is defined in the Australian Standard (AS 2772.-1990) as waves having frequencies from 100 kilohertz up to 300 gigahertz. The radiofrequency spectrum includes, in increasing order of energy, waves from AM radio, FM radio, TV (very high and ultra high frequency), mobile phones, police radar, microwave ovens and satellite stations.
All electromagnetic radiation involves an oscillating electric field and a magnetic field. Whereas at the extremely low frequency end of the spectrum (e.g. AC current at 50 or 60 hertz) the two fields can be measured and considered separately, in the radiofrequency spectrum they are measured together. The intensity ('power density') of the combined fields can be readily expressed in terms of a power unit relative to area (e.g. watts per square centimetre) which denotes the electric and magnetic fields as a multiple. Absorption of electromagnetic radiation energy by living organisms can be expressed in terms of watts per kilogram. This represents the dose, or more correctly, the specific absorption rate (SAR). The value for SAR is not always easy to calculate, especially in respect of individual organs or cell types.
Intense waves in the radiofrequency spectrum are readily able to raise the temperature of, say, a culture of cells brought near the source of radiation (the principle of the microwave oven) as wave energy is converted to heat energy on contact with the cells. This is known as a thermal effect. However, because the radiation is non-ionising there is no electron stripping of cellular DNA and therefore no direct initiation of cancer. Radiofrequency standards to protect health are totally based on avoiding thermal effects (see below).
The thermal or heating effects of radiofrequency radiation (including microwaves) on living organisms are well known, they are dose-related and they are mostly reproducible. These crucial characteristics have been regarded by many scientists as justifying the selection of thermal effects as a powerful and single basis for determining health standards. The following information has been adapted from information contained in the previously mentioned CSIRO review report.
Heating caused by RF radiation is caused mainly by water molecules lining up with the electric field imposed by the radiation. Since the field is oscillating very rapidly (wave frequency), the water molecules are rapidly swinging one way then another in sympathy, thus generating heat. Some biological molecules are also influenced by applied electric fields.
Exposure of people to a dose of radiofrequency radiation of less than about 4 watts per kilogram body weight is thought to give rise to an increase in body temperature of less than 1o Centrigrade and can be reasonably well tolerated for short periods. Higher induced temperatures are not tolerated, however, and have several well-known deleterious effects, depending on the precise location of radiation absorption. An effect observed at RF intensities sufficient to raise the rectal temperature of an experimental animal by 1o C or more is classified as thermal in nature. Such effects could be induced by any method designed to raise body temperature.
- Firstly, the skin can detect RF radiation but the sensation is much less than that from infrared radiation and is extremely dependent on frequency which determines penetration. In the range 0.5-100 gigahertz, skin detection is not regarded as a reliable warning mechanism.
- Heat effects on brain tissue are thought to be the reason why people can actually hear pulsed radiofrequencies between 200 megahertz and 6.5 gigahertz. The sound is described as 'buzzing, clicking, hissing or popping'.
- Thirdly, the eyes are felt to be peculiarly sensitive to RF radiation. Lens tissue has no blood supply to act as coolant, there is little self-repair at that site and thus damage and damage products tend to accumulate. At a threshold of about 41o C, exposed laboratory rabbits show cataract formation. Further work needs to be done on the susceptibility of primate eyes, which seem to be less sensitive.
- Fourthly, rat testes exposed to RF radiation leading to temperature increases of 1.5-3.5o C are damaged to the extent that there is temporary infertility and an altered division pattern of germ cells.
- Fifthly, the thermal disruption of behaviour by RF radiation, e.g. task learning and short term memory, has been demonstrated in the rat. Effects were observed at doses between 0.6 and 8 watts per kilogram.
- Sixthly, the circulatory and immune system in rodents shows some alterations in response to RF radiation. For example, blood cell counts decline in some experiments while the immune system appears to be stimulated. Once again, these effects appear to be thermally induced.
- One laboratory has reported symptoms similar to heat stroke leading to death in rats following exposure at three microwave frequencies.
- Lastly, a body temperature of 43o C in pregnant rats brought about by a dose of 11 watts per kilogram of RF radiation caused abnormalities and death of embryos. So long as there is a temperature increase of at least 2.5o C, birth defects can be expected to occur.
It has already been observed that RF standards are based on the prevention of thermal effects since these are well accepted in the scientific community and are generally reproducible. Two standards will be mentioned here, namely, the American National Standards Institute/American Institute of Electrical and Electronic Engineers (ANSI/IEEE) Standard C95.1-1991 and the Australian Standard 2772.1-1990 (Standards Australia). Both are designed for the RF/microwave spectrum (100 kilohertz to 300 gigahertz).
ANSI power density limits for members of the public vary within the RF range from a low of 0.2 milliwatts per square centimetre (mW/square cm) at 100 megahertz to a high of 10 mW/square cm from about 10 gigahertz. The ANSI standard at the frequency used for Australian mobile phones (800-1000 megahertz) is slightly less than 1 mW/square cm.
Australian Standard 2772.1-1990 lists a constant limit of 0.2 mW/square cm (equal to 200 microwatts/square cm for members of the public at frequencies between 30 megahertz and 300 gigahertz. Thus, at Australian mobile phone frequencies our national standard is about five times stricter than the ANSI standard.
As is the case for many other US standards, the ANSI determination is influential here and there is a strong move for the Australian standard to be loosened by a factor of five in order to correspond to ANSI's limit. It is therefore important to be able to assess the basis of ANSI reckoning on RF safety.
According to the CSIRO, the US approach to its standard has been to consider thermal effects of RF radiation only, and to regard behavioural changes in experimental animals as the most sensitive of those effects. In contrast to ionising radiation, where adverse effects on people are well documented, RF effects on humans are inadequately described, which explains the need for animal results. Of course this raises the immediate question: can experimental animals, especially small animals, provide an adequate model?
Since it is always necessary to dose non-human primates with more than 4 watts per kilogram body weight for behavioural effects to appear, this has been taken by ANSI as the official threshold for humans. As mentioned earlier, 4 watts per kilogram is also the approximate threshold for human tolerance of the heat generated. A tenfold and a fifty-fold safety factor has been applied to the threshold for occupational and non-occupational exposure limits and the corresponding power density figure worked out. Thus, the five-fold stricter Australian (non-occupational) standard is 250 times (i.e. 50x5) below the experimental animal threshold for thermally induced behavioural changes.
In this paper it has been necessary to describe RF standards and their basis in some detail in order to assess emissions of radiation from TV and mobile phone towers. Note that both telecommunications carriers and broadcasting stations are required to adhere to Australian Standard 2772.1-1990.
TV towers have a much higher power rating-and thus give out more intense radiation- than mobile phone towers. For example, the TV transmitter on top of Black Mountain, Canberra, is rated at 300 kilowatts. A typical mobile phone tower is emitting only about 20 watts, i.e. 15 000 times weaker. Perhaps fortunately, most large TV towers are situated on hilltops which are relatively far from housing. It is the occasional exception, for example, on Sydney's North Shore, that deserves special attention.
Since radiation from both TV and mobile phone towers is not directed vertically downwards, there is not a simple relationship between the tower-observer distance and the strength of electric and magnetic fields combined as EMR. Take firstly the case of mobile phone towers. Between 0 and 10 metres from a digital mobile phone tower, levels of exposure are approximately the same. The level of radiation peaks at between 100 and 150 metres, intensity values ranging from 0.1 up to 1.0 microwatt/square cm, depending on how many telephones are in use at the time (note that one microwatt equals one-thousandth of a milliwatt). Further away than 10 metres, radiation intensity falls off rapidly, approximating the 'inverse square' law. Radiation from analogue mobile phone towers is slightly more intense, peaking at 4-6 microwatts. These figures have been supplied by the Australian Radiation Laboratory.
In comparison with the Australian Standard(3) (200 microwatts/square cm), a power density level of 6 microwatt/square cm from a mobile phone tower (said to be a maximum value) represents only 3% of the value of the maximum allowable power density. A more typical figure of 0.1 microwatt/square cm is only 0.05% of the standard.
Turning to larger TV broadcast towers, a person standing one kilometre away would expect to be exposed to a power density of 5-10 microwatts/square cm of radiation. At two kilometres this reduces greatly to about 0.5 microwatt/square cm. These figures are still far less than the prescribed limit of 200 microwatts/square cm.
Dr Bruce Hocking, a former Telstra medical director, has presented findings in a recent issue of The Medical Journal of Australia(4) linking leukaemia incidence with proximity to TV towers . Radiation levels of 8 microwatts/square cm were cited near the towers, decreasing to 0.2 microwatts/square cm at a radius of 4 kilometres and 0.02 microwatts/square cm at a radius of 12 kilometres.
In summary, children under 15 years of age living in three Sydney suburbs within 4 kilometres of TV towers (North Sydney, Willoughby and Lane Cove) appear more likely to suffer from leukaemia than similarly aged children from Ryde, Kuringai and Wahroonga, localities more distant from TV towers. The data was retrieved from the NSW Cancer Registry(5) between 1972 and 1990. A similar type of study found increased levels of cancer in Honolulu, Hawaii, among people living near TV towers(6).
Dr Hocking stresses that his results are preliminary but they show that further research is warranted. The association between TV towers and cancer is certainly not proven but can be regarded as 'hypothesis-generating'. Dr Hocking also regards the results as unexpected because the measured radiation levels (up to 8 microwatts/square cm) are so far below the Australian Standard of 200 microwatts/square cm.
Opposition to mobile phone towers placed near houses can only increase in response to this preliminary finding of a cancer link in respect of TV transmitters. People tend to feel that sites near to schools are particularly undesirable because children are exposed throughout the day, yet have no choice in the matter and derive no benefit. This is in spite of the fact that mobile phone towers are of very low power. Mobile phone users have a much greater exposure to radiation but at least they get the benefit of the calls as well as being able to control their exposure by shortening conversations.
What is the RF exposure from personal mobile phone use as compared with exposure to a mobile phone tower? As described above, such towers radiate very small power densities of not more than about 6 microwatts but more typically 0.1 microwatt/square cm at close range. By contrast, an analogue phone is said to generate a power density of 0.27 milliwatts/square cm at a distance of 5 centimetres. This can be calculated as between 45 and 2700 times greater than radiation intensity from a mobile phone tower. Much discussion has centred on the actual dose to the head resulting from normal use of an analogue or digital phone. In terms of power density, however, the radiation generated is clearly of the same order of magnitude as set out in the Australian Standard for members of the public. This suggests that there may be some pressure from manufacturers of mobile phones to have the Australian Standard relaxed somewhat.
There are three levels of power densities (watts/square centimetre readings) in relation to heating effects on tissue. They are:
- High power densities, generally greater than 10 milliwatts/square cm, at which distinct thermal effects predominate (as listed earlier in this paper).
- Medium power densities, between 1 and 10 milliwatts/square cm, where weak but noticeable thermal effects exist, and
- Low power densities, below 1 milliwatt/square cm (the Australian upper limit for occupational exposure) where thermal effects do not appear to exist but other effects have been reported.
This section of the paper deals with the claimed non-thermal effects which have been reported at low and medium power densities, and discusses the reasons why these effects have been discounted, rightly or wrongly, as a basis for Australian and overseas standards.
Possible behavioural changes or indirect promotion of cancer is a principal focus of low-power radiofrequency (microwave) studies. As stated earlier, the RF spectrum is not energetic enough to cause mutation damage to cell genetic material (DNA) and thus directly initiate cancer. However, among the hundreds of reports of RF effects there are some which can be interpreted as possibly assisting the spread of cancer.
Firstly, some experiments (e.g. Ref. 7) have indicated radiation-caused changes in the so-called blood-brain barrier. The healthy brain is an exclusive organ which does not admit entry of many types of chemical and biochemical substances. The research has measured abnormal passage across the blood-brain barrier of protein-bound dyes, radioactively labelled sugars or peroxidase enzyme in irradiated rats and hamsters.
Secondly, there are examples of disturbances to foetal development (teratogenic effects) in mice, chicks and rats at low RF power. Retarded development (low birth weight), eye malformations, reduction in organ weight and embryonic death have been observed.
Experiments with RF radiation and cultured cells are thought by some scientists to demonstrate low power (non-thermal) effects on the cell membrane. The best-known work, that of Professor Ross Adey, has shown a consistent increase of calcium loss from brain tissue. This indicates that the membrane permeability has been changed. Calcium is known to be a highly significant biochemical regulator, e.g. it controls the division of certain cells. The RF waves may be creating free radicals or changing the physical characteristics of fats in the cell membrane.
Non-thermal treatment which increases the rate of division of cell lines or increases cancers in whole animals is of particular interest. Lymphocytes, a line of white blood cells, have been reported to proliferate more rapidly under what are claimed to be non-thermal conditions of irradiation. Spontaneous mammary cancers and artificially induced lung and skin cancers in mice have been said to increase under low power RF radiation applied over varying periods up to ten months. Another study has found that the number of spontaneous cancers in irradiated rats increases significantly.
The above examples, plus many others in the scientific literature, are sufficient to arouse concern over possible health consequences of non-thermal RF irradiation in the same range of intensity or less than Australian and overseas standards. Why then are non-thermal effects disregarded in the current standards?
The truth is that there is no scientific consensus on non-thermal effects, and the literature overall reveals a highly unsatisfactory state of affairs. The effects listed above represent the most positive results; however, lack of confirmation is a chronic problem. Many laboratories simply cannot replicate the results of others, and negative results are difficult to have published. One of the difficulties with this type of research is that the experimental variables, e.g. radiation frequency, orientation, method of modulation, etc. are numerous and very few scientists seem to try hard enough to standardise others' experimental conditions. Also, experiments which are claimed to be non-thermal can be judged to involve local temperature changes or irrelevant stress conditions. Non-thermal effects are frequently not dose-dependent and therefore lack scientific credibility. Lastly, there is still no universally accepted physical or chemical mechanism to explain how RF radiation can interfere with animal metabolism apart from heating effects. For example, the role of the magnetic component as distinct from the electric field component, if any, can only be guessed at.
At the Federal level there is a committee and a program dealing with radiofrequency radiation and health.
The Committee on Electromagnetic Energy Public Health Issues is located in the Department of Communications and the Arts. It is made up of representatives from that Department, the Department of Health and Family Services, The Australian Radiation Laboratory, the Spectrum Management Agency, the Therapeutic Goods Administration, AUSTEL and the CSIRO. The Committee's role is to coordinate the $4.5 million Radiofrequency Electromagnetic Energy Program announced by the Government on 15 October 1996. The Program has three parts, namely:
- public education on radiofrequency health issues
- Australian participation in a World Health Program
- the setting up of a research program in Australia.
With regard to the research program, the Committee is preparing a priorities paper which is intended to be released for public discussion. When the priorities are finalised, it will be the responsibility of the National Health and Medical Research Council (NH&MRC) to manage the research, in the first instance by calling for specific proposals.
Under the circumstances, the best approach for the NH&MRC would be to encourage good quality research at low power (non-thermal) radiation levels. Much more scientific effort has to be invested in making the RF field respectable. While there is no convincing evidence as yet that low power RF sources such as mobile phone towers can increase the incidence of cancer, some caution is warranted given that existing health standards are based on rather narrow criteria, and that cancers often have a long lead time (as for example, with asbestos and mesothelioma). Since the sum of less than $4.5 million for research will not go far, a small levy on every mobile telephone sold would help to speed up our understanding in this area.
1. Barnett, S. B. CSIRO Report on the Status of Research on the Biological Effects and Safety of Electromagnetic Radiation: Telecommunications Frequencies. CSIRO Division of Radiophysics, June 1994.
2. Hocking, B., Gordon, I. R., Grain, H. L. and Hatfield, G. E. Cancer incidence and mortality and proximity to TV towers. Med. J. Aust. December 1996, p. 601.
3. Australian Standard AS 2772.1 Radiofrequency Radiation Part 1: Maximum Exposure Levels-100kHz to 300 GHz. Sydney: Standards Australia, 1990.
4. Hocking et al., loc. cit.
5. HealthWiz. National health database. Commonwealth Department of Human Services and Health. 1991-1996. Canberra: Prometheus Pty Ltd, 1996.
6. Maskarinec, G., Cooper, J. and Swygert, L. Investigation of increased incidence in childhood leukaemia near radio towers in Hawaii: preliminary observations. J. Environ. Pathol. Toxicol. Oncol. 1994: 13:33.
7. Salford, L.G., Brun, A., Eberhardt, J. L., Malmgren, L. and Persson, R.R. in: Interaction Mechanism of Low-Level Electromagnetic Fields in Living Systems. C. Ramel and B. Norden, eds. Oxford University Press, 1992.