Fact check: Does infrasound from wind farms make people sick?

It’s no secret that wind farms appear to have fallen out of favour with the Coalition Government.

Prime Minister Tony Abbott recently called them “visually awful”, “ugly” and “noisy” and issued a directive to the Clean Energy Finance Corporation, which receives $2 billion a year in government funding to invest in renewable and low emissions technologies, to stop funding any new wind power projects.

Liberal Democrats senator David Leyonhjelm has been another vocal opponent, instigating an inquiry into the regulation of wind farms.

In an interim report, the Senate Select Committee on Wind Turbines recommended that a commissioner to resolve complaints from residents living near wind turbines be appointed.

The Abbott Government has since agreed to appoint a windfarm commisioner.

Senator Leyonhjelm wrote in an article for The Australian that “it is well established that inappropriate levels of infrasound, regardless of the source, cause adverse health impacts”.

Expert analysis and data is provided here by researchers at the University of Melbourne School of Population and Global Health, and School of Mathematics and Statistics, who worked with ABC Fact Check to scrutinise the claims Senator Leyonhjelm made in the article.

What noises do wind farms emit?

Infrasound is low frequency noise that is defined by the American National Standards Institute as sounds at frequencies less than 20Hz.

A report on wind farms, sound and health by the Victorian Department of Health talks about two types of sounds emitted from wind farms.

The first is aerodynamic sound which is the main source of sound from wind farms and is “generated by the rotation of turbine blades through the air”.

The other type is the mechanical sounds that come from the internal machinery of the turbines.

The report notes that it is difficult to generalise about wind farm sound because it depends on a number of factors including the model, size and number of turbines, as well as the distance from the turbines.

Noise is also influenced by wind speed and direction and the topography of the land.

“A mid to high frequency intermittent “swish” is the main sound heard within approximately 300 metres of a wind turbine”, however low frequency sounds “may become more noticeable” further away from the turbine, the report says.

Most people can’t hear infrasound

Claim: “It is beyond dispute wind turbines emit infrasound and low frequency noise, much of which is inaudible to most people.”

Researchers from the University of Melbourne told Fact Check that “there is no controversy as to the production of such emissions per se”.

Such emissions can also result from other natural and man-made environmental sources, including but not limited to: large ventilator fans, traffic noise, the ocean, and walking – due to the impact of the foot on the floor.

These emissions are inaudible to the majority of the population according to a report by the Victorian Department of Health which relies on data from 1990 by researchers Toshio Watanabe and Henrik Moller.

This data, which is relatively uncontroversial, shows that at frequencies below 16-20Hz, human auditory sensitivity to sound is very low.

There is also the issue of feeling as opposed to hearing the low frequency sound.

Researchers from the University of Melbourne told Fact Check this is an empirical question that is yet to be addressed.

Verdict: Correct

A direct link between infrasound and adverse health effects?

Claim: “We have known this since 1987, when Neil Kelley identified a direct causal link between impulsive infrasound and low frequency noise, and adverse effects on people.”

The Kelley (1987) study is a United States department of energy report and conference paper on the relationship between simulated turbine sound and levels of annoyance that was not reviewed before it was published.

Researchers at the University of Melbourne told Fact Check that while the sound generation appears quite comprehensive and controlled, the behavioural measurement is lacking, with too few subjects and a uncontrolled subjective measure of acute “annoyance”.

There are no measurements of, or comments made on, any “health” effects in this study, while this may be adequate for an internal report, it would not pass peer-review, the researchers said.

Even if the study was well conducted, it would only identify an association, not causation, and with just seven self-selected subjects, no conclusions about the population as a whole can be drawn, no matter how carefully the study was conducted, they said.

Researchers told Fact Check that the Kelley study suggests the plausibility of a link between turbine sound and levels of “annoyance”, but does not constitute evidence in the epidemiological sense. Epidemiological evidence is the evidence any public health decision must be made on.

They said that to draw population level inferences, a well-designed large scale epidemiological study of the population is required as is set out in the National Health and Medical Research Council recommendations, and that the study, and others like it, warrant further rigorous epidemiological and behavioural investigation.

Verdict: Irrelevant

Some people are more sensitive to noise than others

Claim: “It is also clear that 10 to 15 per cent of the population is more sensitive to noise, including low frequency noise and infrasound.”

Researchers from the University of Melbourne were unable to identify where the 10-15 per cent value has been drawn from.

A spokesman for Senator Leyonhjelm referred Fact Check to Daniel Shepherd who has a doctorate in psychoacoustics and lectures at Auckland University of Technology.

Dr Shepherd’s expertise lies in the human response to sound.

Dr Shepherd told Fact Check it would be difficult to dispute Senator Leyonhjelm’s figures.

He said he had obtained similar numbers in New Zealand and provided a chart from research he conducted in 2010 showing about 10 per cent of city residents and 20 per cent of rural residents self-reported a high degree of noise sensitivity.

The chart did not refer to low frequency noise or infrasound.

Dr Shepherd said that in recent peer-reviewed papers he had reported similar results, and had referred to higher numbers obtained in European studies.

He did not identify the papers.

Researchers from the University of Melbourne responded by saying there was a natural variation in the population with regard to noise sensitivity and there will be a modest proportion of the population who are particularly sensitive to noise, possibly in certain frequency ranges.

They said Dr Shepherd and colleagues have measured perceived quality life and proximity to wind turbines “but not in conjunction with noise sensitivity”.

Verdict: Unsubstantiated

People attribute various health problems to wind turbines

Claim: “I have met some of those affected. They tell me they mainly suffer from chronic sleep deprivation, but some also suffer sinus pressure, tinnitus, pains in the chest, headaches, nausea, and vertigo.”

The researchers from the University of Melbourne told Fact Check that such data is no doubt available and real and “we accept that those individuals are suffering from those symptoms”.

But they said that anecdotal data, while suggestive, does not constitute epidemiological evidence.

To understand what actually causes the health problems requires proper epidemiological investigation.

They also said the listed symptoms are those of generalised anxiety, which could arise from multiple sources.

Verdict: Anecdotal

Infrasound can make people sick

Claim: “It is also well established that inappropriate levels of infrasound, regardless of the source, cause adverse health impacts.”

The researchers Fact Check spoke to said that an examination of the current published research found the available “literature does not support this strong claim”.

While there is suggestive evidence and anecdotal reports that exposure is associated with adverse health impacts, there is no definitive evidence to prove that this is a causal link; this remains an unanswered empirical question.

There is also evidence suggesting that such impacts are able to be psychologically manipulated through expectation and context.

The NHMRC review and Targeted Research Call (2015) says no epidemiological studies have ever been performed that are able to definitively confirm or reject the hypothesis that exposure to infrasound causes adverse health outcomes. While the link may be plausible, it has not been proven.

The assumption that infrasound causes illness does not have a strong basis in scientific evidence.

“It may be correct. It may not be correct. It is not known,” the University of Melbourne researchers said.

Verdict: Unproven

The Verdict

In the article Senator Leyonhjelm also said “some argue the evidence linking wind turbines to adverse health effects is too tenuous to warrant action. It is true that it is not yet well understood.”

He then went onto say that “by the time further studies are published in recognised journals following peer review, many more people will have suffered. The fact we are not yet at that stage is no excuse for inaction.”

So, while Senator Leyonhjelm acknowledges that the evidence wind turbines are harmful is not conclusive, his key claim is still that it is well established that inappropriate levels of infrasound, regardless of the source, cause adverse health impacts.

That claim is unproven.

Sources

• David Leyonhjelm, The Australian, June 10, 2015
• Victorian Health Department, Wind farms, sound and health, April 2013
• N.D.Kelley, A proposed metric for assessing the potential of community annoyance from wind turbine low-frequency noise emissions, November 1987
• NHMRC Statement and information paper: Evidence on wind farms and human health, 2015

Editor’s note: This Fact Check sought expert comment from a number of researchers at the University of Melbourne’s School of Population and Global Health, and School of Mathematics and Statistics. All commentary is attributable to these schools.