Blow down a storm of protest

Concerns about noise in the environment were first raised officially by the Committee on the Problem of Noise who published the Wilson Report in 1963, named after its chair, Sir Alan Wilson. The report starts by defining noise as 'sound which is undesired by the recipient'. Noise is, therefore, subjective and its assessment more a matter of human perception than of precise physical measurement. These perception issues are highly complicated and may depend upon many factors which are independent of the physical qualities of the noise.

These observations, which are made at the beginning of the Wilson Report, remain as true today as they were in 1963. As a result of this, even the most complex of noise level measurements can only give an indication of the way sound is perceived by an individual. Noise measurements and noise limits are, however, necessary and essential in order to protect the majority of the population from unacceptable levels of noise and to give guidance to noise makers on how much noise is reasonable to make in a given situation.

Measurement and assessment

Noise is measured in decibels (dB), meaning a tenth of a 'bel' which is a term named in honour of Alexander Graham Bell, the Scottish inventor, to represent the ratio of two physical quantities. In the case of noise, it represents the ratio between the sound pressure being measured and that corresponding to the threshold of hearing. The normal audio range extends from 0dB (the quietest sounds which can be heard) to 120dB (the loudest sounds which can be tolerated without serious discomfort). A change in noise of 3dB is stated in government planning policy guidance document PPG24, 'Planning and noise', to be the minimum perceptible change under normal circumstances, while a change of 10dB corresponds to a doubling or halving of perceived loudness.

To fit the physical descriptor to people's perceptions, various corrections and adjustments are usually made to the measured decibel level, hence the use of the alternative dB(A) measurement index which allows for the response by the human ear to sounds of different frequency (pitch). An additional correction, up to a maximum of 5dB(A), may be added if the noise is considered to be 'tonal' or 'impulsive' or has any other characteristic 'likely to attract attention'. Measured noise level may be averaged over a given period, such as the 16-hour day (7am'“11pm) or the eight-hour night (11pm'“7am).

Noise limits can be set as absolute levels '“ for instance the WHO states that an external limit of 55dB(A), averaged over a 16-hour day, is required to prevent serious annoyance and a 45dB(A) external limit over the eight-hour night is required to prevent sleep disturbance. Or they can be set as levels relative to existing 'background' noise '“ for instance British Standard BS4142 states that complaints are likely when the difference between a new noise level and the original background noise is 10dB(A) or greater.

Wind turbine noise

Wind turbine noise has the unique characteristic of having a different source noise level depending on wind speed. Background noise also varies with wind speed, particularly in rural areas where, in the absence of any other noise sources, the effect of wind in trees and foliage and over buildings and land forms, becomes the dominant source of noise.

Noise from wind farms is assessed according to a methodology known as ETSU-R-97 which is specified as appropriate in government planning policy supplement PPS22, 'Renewable Energy'. This uses a hybrid of the 'absolute' and 'relative' approaches identified above, which takes into account the increase in background noise with wind speed. When background noise is low, the external noise level has to meet an absolute criterion of 35'“40dB(A) during the day (significantly lower than the WHO limit described above) and 43dB(A) at night (similar to the WHO external night time criterion). As wind speed, and hence background noise, increases, the maximum permissible limit changes to 5dB above the background noise, based on the results of measurements made over a period of about two weeks at each assessment location.

ETSU-R-97 acknowledges, in the setting of these noise limits, that the noise from a wind turbine site may contain an element of 'blade swish'. The noise level measured is, however, adjusted for any tonal components by a correction of up to +5dB depending on the level of tonal content.

Noise sources

The source noise from wind turbines consists of aerodynamic noise (from the air passing over the blades as they rotate around the central hub), and mechanical noise (from components in the nacelle at the top of the tower: usually gearbox, generator and auxiliary items such as fans and pumps). Tonal noise from these mechanical sources used to be a problem with older turbines but this has now been significantly reduced by good design including acoustic isolation and insulation and it is now the aerodynamic noise which is dominant.

Low frequency 'thumping' was a problem with older designs, not seen on larger turbines in the UK, where the turbine tower was upwind of the blades such that the blades cut through the wake left by the tower. This problem has been eliminated by modern computer-controlled 'active yaw' turbines, installed since large-scale wind power development started in the UK in 1991, whereby the blades are now located upwind of the tower.

Media hysteria

However there has been a certain amount of media hysteria which caused problems for the wind turbine industry, generated by claims of 'infrasound' from wind turbines which was supposedly causing health problems. Infrasound is sound with very low frequency (pitch), below that considered to be part of the audible range. This appears to have been generated by some misplaced guidance in a report for DEFRA on low frequency noise, which stated that wind farms were a source of possible low frequency noise. Because infrasonic signals are not considered to be part of the normal audible range, they have always held a certain fascination, particularly for the media, with inaudible infrasound being held to be responsible for any number of health issues. In fact, these effects only occur at very high levels of infrasound, orders of magnitude above that created by the active -yaw wind turbines.

Indeed, Geoff Leventhall, the highly respected author of a report on low frequency noise and effects for DEFRA has stated that: 'I can state quite categorically that there is no significant infrasound from current designs of wind turbines'. The same can be said of vibration with none being perceptible even at the base of a wind turbine tower. In a recent letter to the press, two authors of a report on vibration from wind turbines prepared for the MoD stated that, at residential distances, 'to put the level of vibration into context, they are ground vibrations with amplitudes of about one millionth of a millimetre. There is no possibility of humans sensing the vibration and absolutely no risk to human health'.

Low frequency noise

The issue of general low frequency noise rather than infrasound is, if anything, rather more complicated. Aerodynamic noise from wind turbines is essentially 'broadband' in nature meaning that the emitted sond contains energy at all frequencies equally within the normal audible range.

Because high frequency sounds are attenuated with distance more than low frequency sounds, due to atmospheric absorption, the net result, over a long distance, is that low frequency noise level gradually increases relative to high frequency sound. This can be observed with other broadband noise sources in the environment such as traffic noise or the sea and even amplified music. Normally, this increase in low relative to high frequencies only becomes significant at long distances from the turbines, when the overall level of noise is, in fact, below the threshold of hearing for all except the largest sites.

Noise perception and health

This issue of aerodynamic (noise) modulation highlights the effect of human perception on the assessment of noise. Going back to the Wilson Report, it is stated that 'a sound which most people would ignore in, say, the industrial part of a city, would be a disturbing noise in a quieter environment, for example at night in a residential area, or in the country.

The annoyance produced by a noise is often related to the information it conveys rather than to its actual intensity'. In the case of wind turbine noise, the noise may well bring an association with an undesired change in the environment which can be very important to those who live in rural areas.

This particularly applies to those who have chosen to live in such locations at considerable financial cost rather than those who have grown up there. This response can of course generate stress and when combined with lack of sleep resulting from anxiety over noise issues, can be described as a health effect. Whether or not it could be described as a direct effect of the noise is arguable.

Conclusion

It is a fact of commercial-scale wind energy projects that the benefit they bring in terms of fossil- and nuclear-free energy generation has the unfortunate consequence of audible noise in areas which may otherwise have been very, very quiet indeed, or subject only to the sounds of nature. Desirable as such locations are, however, it may be unrealistic, partcicularly for the rural economy, to try to protect these areas from new sources of audible noise from whatever source.

The noise limits, effectively set by government in PPS22 via ETSU-R-97, represent noise levels which are justifiable in terms of reasonableness in rural environment, even though they may represent a clearly audible change to the environment under some conditions. These may be unacceptable to some people due to the complexity of the issues identified in 1963 which are not about to go away soon.