Definitions

Infrasound is defined as sound with a frequency of less than 20 Hz, and low frequency noise as sound with a frequency of less than 200 Hz.Reference Dommes, Bauknecht, Scholz, Rothemund, Hensel and Klingebiel^5–Reference Leventhall⁷ Low frequency noise is generated by many everyday sources such as household appliances, transportation and industrial machinery.Reference Salt and Hullar^8, Reference Takahashi⁹ There are natural sources of infrasound such as ocean waves and volcanoes, as well as internal sources such as the heartbeat and respiration. A child playing on a swing experiences levels of infrasound of approximately 110 dB.Reference Alves-Pereira and Castelo Branco¹⁰ Some large mammals may use infrasound for communication.¹¹

Infrasound is commonly misinterpreted as sound that is inaudible. There is no definite delineation between what is audible and what is not. Studies have shown a hearing threshold as low as 4 Hz can be measured in an acoustic chamber, and 1.5 Hz for earphone listening.Reference Leventhall¹² Infrasound, if presented at a high enough level, may indeed be audible.Reference Salt and Hullar^8, Reference Leventhall¹³ To become audible, the sound pressure level for a 20 Hz sound needs to be approximately 75 dB, compared with 4 dB for a 1000 Hz sound.Reference Salt and Hullar^8, Reference Leventhall¹³

Historically, infrasound and low frequency noise have attracted much adverse publicity mainly based on misunderstanding and media hype, with stories such as the worker who ‘fell dead on the spot’ and the ability of infrasound to ‘knock down buildings’.Reference Leventhall^12, Reference Leventhall¹³ Wind turbine syndrome has reinvigorated these anxieties about possible adverse health effects of infrasound.Reference Leventhall^12, Reference Leventhall¹³

Wind turbine noise

Wind turbines consist of a tower, a rotor (or hub), a set of rotating blades, and a nacelle which houses the generator and controls.Reference Punch, James and Pabst³ These tall and highly visible structures are often placed in open, rural areas – places where the level of background noise is typically low and the surroundings perceived as natural. Unsurprisingly, wind turbines are often viewed as both visual and auditory intruders.Reference Dommes, Bauknecht, Scholz, Rothemund, Hensel and Klingebiel⁵

There are two sources of noise from wind turbines: mechanical noise from the gearbox and generator; and aerodynamic noise produced by the flow of air around the blades.Reference Pedersen and Persson Waye¹⁴ The latter tends to be the main source of problems.Reference Punch, James and Pabst³

Under good conditions with low wind speeds, the sound of wind turbines can appear benign or ‘difficult to discern above the rustling of trees’, as described by the British Wind Energy Association.Reference Salt and Hullar⁸ However, in suboptimal conditions the aerodynamic noise varies as the blades turn with changes in the strength and direction of wind.Reference Punch, James and Pabst³ Several studies have indicated that this amplitude modulated sound, rather than being a constant, steady type of noise, is more easily perceivable, and hence more likely to cause annoyance or health symptoms.Reference Punch, James and Pabst³

We know that wind turbines produce broad-band noise; however, unweighted sound measurements have shown that this noise is indeed dominated by infrasound with frequencies of less than 10 Hz, and that levels of more than 90 dB are generated for sounds around 1 Hz.^11, Reference Salt and Kaltenback^15, Reference Salt and Lichtenhan¹⁶

Wind turbine syndrome

Some people have reported adverse symptoms which they have attributed to noise from wind turbines (Table I). Pierpont was the first to describe wind turbine syndrome.Reference Pierpoint^2, Reference Punch, James and Pabst³ Otological symptoms such as vertigo, tinnitus and ear pain have been reported, and have been proposed to be caused by prolonged exposure of the vestibulocochlear system to infrasound and low frequency noise.Reference Pierpoint² Other reported symptoms include sleep disturbance, headaches and concentration problems. This collection of symptoms has been colloquially termed wind turbine syndrome, but this is not currently a clinically recognised diagnosis.Reference Salt and Hullar⁸

Measuring noise

The World Health Organization has recognised that assessment measures for environmental noise are inadequate for evaluating noise with a large low frequency component, such as that produced by wind turbines.Reference Leventhall¹² This is for two reasons: firstly, the widespread use of A-weighting in sound measurement; and secondly, technical issues in the actual measurement process.

The term A-weighting refers to a filtering process which aims to simulate human hearing in the measurement of sound. A-weighting is recognised internationally and used in the vast majority of noise-related studies and legislation.Reference Alves-Pereira and Castelo Branco^{10, 11} Humans are considered able to perceive sound between 20 and 20 000 Hz, but not uniformly. The A-filter de-emphasises all auditory energy with frequencies of less than 500 Hz, and completely ignores all auditory energy of less than 20 Hz, in an effort to estimate the noise thought to be actually processed by the ear.Reference Alves-Pereira and Castelo Branco¹⁰ Hence, much of the noise produced by a wind turbine is effectively ignored. The A-filter is not used in physiological studies of cochlear function.Reference Salt and Hullar⁸

There are alternative weighting systems available which may be better suited to measuring sound with a predominantly low frequency component, namely: unweighted measurements, C-weighting (designed for 50–4000 Hz) and G-weighting (designed specifically for infrasound).^11, Reference Salt and Kaltenback¹⁵ Studies support the view that these may better predict the effects of low frequency sound.Reference Pawlaczyk-Luszczynska, Dudarewicz, Szymczak and Sliwinska-Kowalska¹⁷

The second problem in evaluating noise generated by wind turbines relates to a technical issue. Suitable instruments for measuring infrasound and low frequency noise are not widely available.Reference Salt and Hullar⁸ Most commonly used types of apparatus are insensitive to infrasound components such as those produced by wind turbines.Reference Salt and Hullar⁸

Sound level meters also tend to measure noise by averaging the sound levels over a period of time. Hence, for fluctuating sound such as wind turbine noise, information about the quality of the noise and its corresponding perceived sound character may be lost.¹¹

Studies tend to assess sound pressure levels under specific meteorological conditions (generally winds of 8 m/second at 10 m height).Reference Pedersen, van den Berg, Bakker and Bouma¹⁸ Again, this may mean that specific characteristics present under different conditions are ignored.

The location is also important. As sound travels away from its source, the sound pressure level is affected both by absorption from ground cover and by a dilution effect as sound energy spreads out from its source.¹¹ Both these effects may play an important part in the subjective assessment of noise, and may be useful topics for future research.Reference Leventhall¹²

Regulations regarding noise

In contrast to ultrasound, there are no clear protection standards established for infrasound.Reference Dommes, Bauknecht, Scholz, Rothemund, Hensel and Klingebiel⁵ Infrasound and low frequency noise are not currently recognised as disease agents, and therefore there is no legislation regarding permissible exposure levels and dose–response relationships.Reference Punch, James and Pabst³

Guidelines for noise management have been produced, such as the World Health Organization 1999 guidelines on community noise.Reference Pedersen, van den Berg, Bakker and Bouma¹⁸ Unfortunately, these guidelines contain no specific reference to wind turbine noise.¹¹ The World Health Organization night noise guidelines (2007) do comment on wind turbine noise, and state that observable effects of night-time, outdoor wind turbine noise do not occur at levels of less than 30 dBA;Reference Pedersen, van den Berg, Bakker and Bouma¹⁸ they also state that there is evidence to link prolonged exposure to adverse health effects if levels are greater than 40 dBA. This has led to a proposed sound level limit of 45 dBA for new wind farms in Canada.Reference Pedersen, van den Berg, Bakker and Bouma¹⁸ A number of other countries have agreed upper limits of 80 dBA at 20 Hz frequencies and 20–40 dBA at 100 Hz.Reference Leventhall¹²

Wind turbines are a new source of community noise to which relatively few people have been exposed. As the number of wind turbines increases, with a corresponding increase in the number of people exposed, guidelines will be needed to indicate the maximum safe exposure to infrasound and low frequency noise.Reference Pedersen, van den Berg, Bakker and Bouma¹⁸ Since infrasound also occurs regularly in other environments (e.g. industrial and military sites), appropriate recognition of adverse health effects is a relevant public health issue.Reference Dommes, Bauknecht, Scholz, Rothemund, Hensel and Klingebiel⁵

What is known about the effects of infrasound and low frequency noise on the ear?

Sounds presented below the typical human audible range have previously been presumed to have no influence on the ear.Reference Salt and Hullar⁸ However, Alves-Pereira and Castelo Branco have utilised comparison with the electromagnetic spectrum: light is perceived by the eye at certain frequencies but not others (e.g. X-ray frequencies), however, despite not being perceived as light, we do of course know that excessive exposure to X-rays can cause severe biological damage.Reference Alves-Pereira and Castelo Branco¹⁰ Perhaps the situation is similar with sound: although low frequency noise may not be audibly perceptible, there may be physiological effects which are as yet unknown.Reference Salt and Hullar⁸ Responses to infrasound are not well documented, but there is some evidence that low frequency noise may indeed affect inner ear function.Reference Dommes, Bauknecht, Scholz, Rothemund, Hensel and Klingebiel^5, Reference Salt and Lichtenhan¹⁶

A recent functional magnetic resonance imaging study assessing the effects of low frequency noise on healthy volunteers found that auditory cortex activation was induced by low frequency exposure.Reference Dommes, Bauknecht, Scholz, Rothemund, Hensel and Klingebiel⁵ This was dependent on high sound pressure levels (i.e. thresholds of 90–110 dB in normal hearing adults). The cortical activation patterns suggested that low frequency noise was processed through acoustic pathways in a similar way to noise frequencies within the main hearing range.Reference Dommes, Bauknecht, Scholz, Rothemund, Hensel and Klingebiel⁵

A review by Salt and Hullar has described in detail the possible effects of infrasound as mediated by the inner ear.Reference Salt and Hullar⁸ These authors have described how hearing perception by the inner hair cells of the cochlea, the main hearing pathway in mammals, is indeed insensitive to infrasound. However, other structures of the inner ear (e.g. the outer hair cells of the cochlea) are more sensitive to infrasound and low frequency noise, and can be stimulated at levels below those normally perceived as audible.Reference Salt and Hullar⁸

The outer hair cells of the cochlea send information to the brain via type two afferent fibres, nerves which were previously thought to be unresponsive to sound but which seem to respond to the static position of the organ of Corti. The outer hair cells are stimulated via direct contact with the tectorial membrane as it moves in response to sound. This displacement of the tectorial membrane remains sensitive at lower frequencies. Therefore, the outer hair cells are more sensitive at lower frequencies, including those below the audible range.Reference Salt and Hullar⁸

The outer hair cells of the cochlea have been shown to respond to sound at frequencies and levels such as those produced by wind turbines, suggesting that this type of low frequency noise may well influence inner ear physiology. However, whether this actually alters function or causes symptoms is currently unknown.Reference Salt and Hullar⁸

Aural pain is another proposed symptom of wind turbine syndrome. This usually arises due to mechanical displacement of the middle ear beyond its comfortable limits: this requires high sound pressure levels, such as 165 dB at 2 Hz or 145 dB at 20 Hz.Reference Salt and Kaltenback^15, Reference Salt and Lichtenhan¹⁶ It has been postulated that noise from a wind turbine may produce sensations of fullness, pressure or tinnitus; however, there is currently no firm evidence to support this.Reference Salt and Kaltenback^15, Reference Salt and Lichtenhan¹⁶

What is known is that the variation in human response is wide.¹¹ There is no doubt that some humans exposed to infrasound and low frequency noise experience abnormal ear symptoms that are stressful, and the effects of this can be far-reaching.¹¹

What is known about the effects of infrasound and low frequency noise on other systems?

Infrasound and low frequency noise have been controversially described as potential health hazards, causing various somatic and psychosomatic disorders such as chronic fatigue, sleep disorders, hypertension and vibroacoustic disease.Reference Dommes, Bauknecht, Scholz, Rothemund, Hensel and Klingebiel⁵ There has been ongoing debate about whether acoustic phenomena can cause extra-auditory effects in humans.Reference Alves-Pereira and Castelo Branco¹⁰ The role of alternative receptors has been discussed in the literature, with questions arising about the possibility of a hierarchy of sound receptors some of which may be more important at lower frequencies.Reference Leventhall¹² The remote possibility of a currently unknown, highly sensitive receptor remains, but there is no clear evidence of this.Reference Leventhall¹²

What is known is that wind turbine noise is annoying.Reference Pedersen, van den Berg, Bakker and Bouma¹⁸ Any unwanted noise leads to annoyance, a symptom which can be moderated by multiple personal and social characteristics.Reference Leventhall¹² Wind turbine noise appears to be more annoying than other noise at similar levels, such as transportation or industrial noise, possibly due to specific sound qualities and lack of night-time abatement.Reference Punch, James and Pabst^3, Reference Pedersen, van den Berg, Bakker and Bouma¹⁸ The low frequency nature of the sound also means it is less likely to be masked by other environmental noise.Reference Punch, James and Pabst³ Pederson et al. have demonstrated a dose–response relationship between A-weighted sound pressure levels from wind turbines and annoyance.Reference Pedersen and Persson Waye^14, Reference Pedersen, van den Berg, Bakker and Bouma¹⁸ Other interesting factors noted to be significant are visibility of the turbine and whether or not the responder benefits economically from the wind turbine.Reference Pedersen, van den Berg, Bakker and Bouma^18, Reference Pedersen and Persson Waye¹⁹

It is also known that high levels of low frequency noise induce body vibrations, most prominently chest resonance vibration in the range of 50–80 Hz.Reference Leventhall¹³ The lower a sound's frequency the longer its wavelength, which means resonance may occur in an enclosed or partially open space such as a room in a house, as well as within the body.¹¹ This vibratory sensation is one of the main contributors of the human response to low frequency noise, and may well be perceived as an ill effect.Reference Takahashi⁹

Sleep disturbance is another symptom commonly reported in case studies of low frequency noise. A large body of evidence now exists to suggest that wind turbines do disturb sleep and impair health at distances and sound pressure levels that are permitted in the UK.Reference Hanning and Evans²⁰

Alves-Pereira and Castelo Branco have described a multisystem disorder occurring secondary to excessive exposure to infrasound and low frequency noise, which they have named vibroacoustic disease.Reference Alves-Pereira and Castelo Branco¹⁰ This is characterised by the abnormal proliferation of collagen and elastin in people exposed to low frequency noise, in the absence of an inflammatory process.Reference Punch, James and Pabst³ These authors' work was initially based on observations of aircraft technicians regularly exposed to infrasound and low frequency noise. The manifestations described are wide, ranging from mild behavioural disturbance to epilepsy, arrhythmias and malignancies.Reference Punch, James and Pabst^3, Reference Leventhall¹² Vibroacoustic disease, like wind turbine syndrome, remains unproven and has not found general acceptance within the scientific community.Reference Leventhall¹²

Variation in response

Certain individuals may have more sensitive perception of low frequency noise than others.Reference Dommes, Bauknecht, Scholz, Rothemund, Hensel and Klingebiel^5, Reference Leventhall¹³ Such noise may be perceived as sound or vibration, or not heard at all.Reference Punch, James and Pabst³

The hearing threshold itself is a median value for an otologically normal young adult. Therefore, some individuals will be more sensitive and some less sensitive, such that a noise that is inaudible to one person may well be troublesome to others.Reference Leventhall¹² In addition, the dynamic range of the auditory system at low frequencies is such that a very slight increase in sound pressure level at a low frequency can change the perceived loudness from barely audible to very loud.Reference Dommes, Bauknecht, Scholz, Rothemund, Hensel and Klingebiel^5, Reference Leventhall^13, Reference Pawlaczyk-Luszczynska, Dudarewicz, Szymczak and Sliwinska-Kowalska¹⁷ Clinical conditions involving changes to endolymph volume regulation, such as Ménière's disease, may well further enhance an individual's sensitivity to low frequency noise.Reference Salt and Hullar⁸

It is possible that long term exposure to unwanted sound can lead to enhanced perception of that sound.Reference Leventhall^7, Reference Leventhall¹² Studies have shown the brain has plasticity, exemplified by the findings of a study of London taxi drivers, who were shown to have an enlarged posterior hippocampus in response to their time spent navigating around London.Reference Leventhall¹² A few people with highly sensitive hearing have been discovered among low frequency noise plaintiffs investigated; however, usually there is little difference between the hearing thresholds of those who complain about low frequency noise and those who do not, suggesting higher centre involvement.Reference Leventhall¹²

Wind turbine noise affects people differently. Public concerns about low frequency noise tend to be based around work on subjective effects.Reference Leventhall¹² Subjective and personal issues are bound to be involved; for example, people living in rural areas may well have moved there for the peace and quiet, and hence a negative reaction to wind turbines is unsurprising.Reference Pedersen and Persson Waye¹⁹ Indeed, some have suggested that personal and social factors can be so variable that it may not be feasible to develop a national noise policy.Reference Leventhall¹²

What is not known?

There is an abundance of information available on the internet describing the possibility of wind turbine syndrome. However, the majority of this information is based on purely anecdotal evidence.Reference Pierpoint² Whilst it is biologically and physically plausible that low frequency noise generated by wind turbines could affect people, there is insufficient evidence on which to base conclusions.¹ The fact that the ear may respond to low frequency noise at the frequency and levels generated by wind turbines does not necessarily mean that such noise will be perceived or will disturb function.Reference Salt and Hullar⁸

Another important factor is length of exposure. There are currently no reported studies which come close to replicating the long term exposure to low frequency noise experienced by individuals who live near wind turbines.Reference Salt and Kaltenback¹⁵ This may be a topic for future study.

A better understanding is needed in order to deal with low frequency noise, to establish why some people are more sensitive than others, and to allow modelling of acceptable exposure levels.Reference Pedersen and Persson Waye^14, Reference Salt and Kaltenback¹⁵ The complexity of the ear's response to low frequency noise suggests that there are many aspects which should be better understood before the influence of wind turbine noise on the ear can be dismissed as insignificant.Reference Salt and Kaltenback^15, Reference Salt and Lichtenhan¹⁶