Resource Documents: Noise (549 items)
Documents presented here are not the product of nor are they necessarily endorsed by National Wind Watch. These resource documents are provided to assist anyone wishing to research the issue of industrial wind power and the impacts of its development. The information should be evaluated by each reader to come to their own conclusions about the many areas of debate.
Author: Stelling, Keith; and Multi-Municipal Wind Turbine Working Group
Typically, regulating authorities have not required the measurement of infrasound (sound below 20 Hz in frequency) and low frequency (LFN) (generally sound from 200 Hz to 20 Hz) inside homes adjacent to wind turbines as a condition of their installation and operational monitoring. The health risk of infrasound from wind turbines has been dismissed by the wind industry as insignificant. It has maintained that since the typical loudness and frequency of wind turbine sound within a home is not audible, it cannot have any effect on human health.
Noise measurements for most studies and environmental assessments have been limited to the measurement of audible sound outside homes– using dBA weighted monitoring which is insensitive to infrasound frequencies. Some studies and environmental assessments have even relied on projected audible sound averages from computer produced models.
Such observations and projections fail to take appropriate account of the distinguishing signature of the sound from a wind turbine. Unlike the more random naturally occurring sounds (such as wind or lake waves which may themselves have an infrasound component), the sound from wind turbines displays characteristics that produce a pattern that the ear and audio processing in the brain recognize. Our hearing is strongly influenced by pattern recognition. (This is why we can pick out the sound of a familiar voice even in a crowded room with many people speaking).
One recognizable wind turbine pattern is a tonal signal of sharply rising and falling pulses in the infrasound range, (typically about 0.75 Hz, 1.5 Hz, 2.25 Hz, 3.0 Hz, and so on). It is produced by the blade passing the tower. At this frequency these pulses may be “felt or sensed” more than “heard” by the ears. Research by Dr. Alec Salt and others has demonstrated that subaudible infrasound does result in a physiological response from various systems within the body.
The second recognizable pattern is the amplitude modulation. This is the typical “swoosh” rising and falling that is audible.
A third recognizable pattern of sound from wind turbines results from the equipment in the nacelle (such as the gearbox if the turbine has one) and ventilating fans. Although in some cases this third sound source may become predominant, it is usually of lesser effect that the first two.
We now know that subaudible pulsating infrasound can be detected inside homes near operating wind turbines. It can also be identified up to 10 kilometres distant. We know also that very low levels of infrasound and LFN are registered by the nervous system and affect the body even though they cannot be heard. The research cited in this report implicates these infrasonic pulsations as the cause of some of the most commonly reported “sensations” experienced by many people living close to wind turbines including chronic sleep disturbance, dizziness, tinnitus, heart palpitations, vibrations and pressure sensations in the head and chest etc.
Similarly, there is medical research (also cited below) which demonstrates that pulsating infrasound can be a direct cause of sleep disturbance. In clinical medicine, chronic sleep interruption and deprivation is acknowledged as a trigger of serious health problems.
Author: Swinbanks, Malcolm; and Australia Senate Select Committee on Wind Turbines
Dr Swinbanks: Just briefly, I will review the submission that I made. I addressed four separate issues: first of all, the physical mechanisms for generating low-frequency sound and infrasound; secondly, the mechanisms by which people can perceive such infrasound; thirdly, I commented on the health effects and, in particular, two reports relating to these supposed health effects or the absence of them; and, finally, I gave an account of my own personal experience of adverse effects I have encountered when taking measurements near to a wind turbine installation.
If I could start off with the generation of infrasound, it is not often realised that NASA, in the early 1980s, actually carried out research on upwind rotor turbines. That is the modern configuration where the rotor is upwind of the supporting the tower, rather than downwind. Wind developers have often dismissed NASA’s work, saying it was not relevant because it related only to downwind turbines, but this is completely inaccurate. NASA had in fact identified the benefits of going to the upwind configuration at a very early stage. They also examined the effects of multiple turbines operating together and the effects of the separation between those turbines. They found that seven to 10 diameters separation was the ideal requirement for a turbine located downwind of its neighbours. But, in recent years, some wind developers have compromised on that spacing and have reduced it even to as little as three diameters in some cases, and that is asking for trouble, because the increased turbulence leads to increased low-frequency sound and infrasound.
The other effect that has to be considered is that as wind farm arrays are made larger and larger, the rate of attenuation as you move away from the wind farm is reduced. The result is that the setbacks from the boundaries have to be much greater to achieve the same reduction in sound. In recent years, people have stated that they have problems at distances of as much as three miles, and that is entirely consistent with the effects of increasing the size of the wind farms. Finally, I would point out that under conditions where the temperature profile is what is known as a temperature inversion, the low-frequency noise and what would I call the ‘silent thump’ of wind farms can carry over distances of three miles or more.
I would like to turn to how people perceive infrasound. The conventional method of hearing is through what are known as the inner hair cells of the cochlea. The effects of infrasound can be measured by a G-weighting scale, which is very similar to the A-weighting scale. It is effectively an extension of it, although the exact values do not correspond directly. Many people have evaluated whether or not the effects of infrasound are perceptible by simply comparing spectra with the hearing threshold and stating that the spectra are well below the threshold values and therefore the sound cannot possibly be perceived. That is not correct. At very low frequencies, it is the combination of different frequency components adding together which defines the total level of the infrasound, and that can be significantly greater than is observed simply by looking at the power spectrum.
People have reported having significant problems believed to be due to infrasound at distances from wind turbines. In that context, there are three different mechanisms which may be contributing to enhanced sensitivity. I have analysed a specific effect relating to the interaction with the threshold as a result of low and high frequencies being present simultaneously. In America, Dr Alec Salt has identified that the outer hair cells of the cochlea are actually much more sensitive at very low frequencies. He believes that there is some input to the nervous system resulting from them. Most recently, Paul Schomer, also in America, has considered the possible effects of sound pressure on the vestibular organs, which are the balance organs, and those effects could give the person on the receiving end a sensation of apparent motion, even though they are actually stationary.
I would like to make a further addition, which is just related to my own experience. Lying in bed, at a distance of three miles from a wind farm, my wife and I have on occasions been disturbed by the wind turbine noise. The most marked feature is that when you have gusts of wind, the turbine noise is masked by the gust and you get a huge sense of relief, only to find that when the wind subsides, the turbine noise returns and you again find yourself subject to the relentless sound.
The point is that when the wind gusts rise it is very much like the effect of when you come out of a tunnel into the light — a huge sense of relief. The sound levels of the turbines under those circumstances are probably less than the average sound levels of the wind, but nevertheless they are far more disturbing. This is noted also at higher frequencies, where people have identified that the annoyance from turbines at 35 dB(A) can be comparable to the annoyance of other more conventional sources at 55 dB(A). One commonly sees statements made that wind turbine noise is no different from any other noise, but the fact is it is different. It is clearly more perceptible at lower levels, and criteria relating to more conventional noise do not necessarily apply.
Turning to the health effects of wind turbines, there was an early report in 2009, which was an American Wind Energy Association funded report. This was the first time that experts had been brought together from both the medical profession and the acoustics profession. That report has been regarded as a definitive baseline report, and subsequent reports have tended to draw on it because of the qualifications of its authors. I consider that report to have been extremely biased. It failed to mention at all two of the most important aspects of wind turbine perception. Firstly, that in rural areas the hearing threshold is much reduced compared to the threshold when you are in urban areas and consequently you are much more sensitive to additional noise. Secondly, there is increasing sensitivity with continuing exposure. Some authors have described this a learned aversion. I have also experienced that at firsthand myself 30 years ago when working on natural gas compressor installations, which are effectively jet engines driving a compressor into an exhaust. In those circumstances, I found that over time, ultimately a period of two years, I had become very sensitive indeed to the low frequency noise and I could detect it under circumstances where previously I could not detect it at all.
That same health report misrepresented guidance which had been given in America by the Environmental Protection Agency as long ago as 1974 — that is 40 years ago — and they have failed to indicate that the presently permitted sound levels in the USA are too high and can lead to sleep disruption. The most recent health report that has been produced, again, funded by CanWEA, the Canadian Wind Energy Association, finally acknowledges the excessive permitted levels in the United States and the resultant consequences for sleep disturbance, but it does not highlight this. The statement is effectively buried in 25 pages of closely-spaced text. Now I believe a lot of the problems have been created as a result of that report and some of its successors, because it has completely understated the nature of the problem and has led, undoubtedly, to people being exposed to higher levels than they should be exposed to.
At the same time, it is common practice to place the burden of the effects of wind turbines onto the homeowners by stating that it is annoyance on the part of the homeowners and nocebo effects. By placing the burden on the homeowners, the apparent responsibility of the wind developer is reduced but, at the same time, this ignores completely the fact that the noise and, indeed, the infrasound can represent a significant intrusion into a rural home, particularly at night when people are trying to sleep. So I believe the correct terminology is to say that people suffer annoyance. It is something which is imposed on them.
I would also comment with respect to the nocebo effect that many communities welcomed wind turbines — this was particularly true of one island community in Vermont — but once the turbines started they discovered that there were some significant adverse effects. That is the very opposite of a nocebo effect. A nocebo effect is when there is prior anticipation of a problem, not when the problem is noted after the event. In that sense, I would like to make a brief comment that NASA, as long ago as 1982, presented a curve which showed the levels of infrasound that could cause adverse reactions by occupants. This showed that the levels of infrasound could be very much lower than the nominal threshold of hearing. People debate whether or not this is due to effects of vibration on a house structure — this is for people inside a house — or whether it is a true perception of infrasound; but that does not really matter. The fact is that, at octave levels as low as 60 decibels, which is a very low level for infrasound, there can be adverse reactions from occupants. That data goes back almost 35 years.
Finally, I would like to —
CHAIR: Excuse me, Dr Swinbanks —
Dr Swinbanks: comment briefly on my own personal experience of wind turbine health effects. I was asked by some friends of mine to help them measure the infrasound levels in the basement of their home at the wind farm at Ubly in Michigan. It is noteworthy that this particular wind farm had been designed in 2005, at which time Dr Nina Pierpont, a doctor in New York state, had been opposing that wind developer because of concerns she had relating to the likely noise environment of a wind farm. She has been roundly criticised all around the world for supposedly promoting scare stories. But in fact the wind farm that was developed at Ubly by exactly the same developer has proven to demonstrate all of the adverse effects that Dr Nina Pierpont warned about. Indeed, 10 families ultimately took legal action against that wind farm. The matter was settled out of court. But the important point is that I myself experienced directly many of the effects that Dr Nina Pierpont warned about, and she certainly was not making it up. The fact is that these effects can occur.
In my particular case, I was working on a very calm evening when wind turbines were operating but there was very little wind at ground level and you could not hear the turbines at all inside the house. I actually had to keep going outside to check that they were still running. After three hours in the house I began to feel ill and I found that I was lethargic and losing concentration, but it was not until sometime afterwards that I began to realise that it was the wind turbines that were likely to be responsible. The level of infrasound that I was measuring was a level that I considered to be very low and definitely not a problem. After five hours in the house I was only too glad to leave, and I thought, ‘At last I’m getting away from this,’ only to find that, when I started driving, my driving ability was completely compromised. The front of the car seemed to sway around as I consistently oversteered. I had lost coordination and I had difficulty judging speed and distance. When I arrived home, my wife observed immediately that I was ill; she could see that straight off. And it took me a further five hours to finally recover and for the effects to wear off.
The important point about that incident was that I had considered that the conditions — a nice calm evening at ground level, but with the turbines still running — were extremely benign, and I had wondered whether I would even get any results. So I certainly was not anxious about infrasound. Similarly, when I got —
CHAIR: Excuse me, Dr Swinbanks —
Dr Swinbanks: Yes?
CHAIR: We have got very short time. Would you mind if we go to questions now?
Dr Swinbanks: Yes, that is fine. In fact, I had effectively completed, so that is fine.
CHAIR: We will start with Senator Urquhart.
Senator URQUHART: Thanks, Dr Swinbanks. I picked up, I think, from your opening statement that you live near an operating wind farm — is that right?
Dr Swinbanks: Yes. We have a farmhouse in Michigan, and the county, Huron County, in which we live decided that they were going to install very large numbers of wind turbines. They installed a first set at two locations in the interior of the region where we are, and significant problems developed at one of those wind farms, but since then they have been installing progressively more wind turbines. We have an installation three miles south of us, which affects us only when the wind blows from a southerly direction and then only under certain weather conditions. But the intention is to install many times more turbines, and, essentially, the whole county will be covered in turbines if this situation continues as it is.
Senator URQUHART: Have you published any articles on infrasound from wind turbines in any peer-reviewed journals?
Dr Swinbanks: Not in peer-reviewed journals. I have presented, at conferences, the work that I have done, and it has represented a sequence of work. But I believed that it was better simply to get the information out into the public domain.
Senator URQUHART: In your submission you mention Steven Cooper’s study from the Cape Bridgewater wind farm. Do you believe this was a scientifically valid study equipped to make conclusions about the link between participant sensations and infrasound?
Dr Swinbanks: I believe that in a situation where people are reporting the effects that they observe while at the same time the operating characteristics of the wind farm are being monitored remotely, if you find that there is then a close correlation between those two situations, when they are well separated and there is no communication between the relevant parties, that does imply that there is a significant link and that people are reacting to real events.
Senator URQUHART: We heard from the Association of Australian Acoustical Consultants. They had done a small statistical analysis of Mr Cooper’s work. In this they found that Mr Cooper did not meet his hypothesis 63 per cent of the time. Do you think it is reasonable to suggest causality when a hypothesis is not meeting close to two-thirds of the event occurrences?
Dr Swinbanks: I would point out that I am not a statistician. I do not approach my own work from a statistical point of view. What I prefer to do is go and find out for myself what it is all about, and from my own experience I believe that what Steven Cooper has observed is entirely credible.
Senator URQUHART: Here in Australia we have had a population level study done that found no difference in the prescriptions that Australians had been given regardless of the distance that they lived from wind farms. Are you aware of any population level studies internationally that have found otherwise?
Dr Swinbanks: I am not aware of such studies. But I do know a lot of families whose life has been made pretty miserable by the wind turbines, and I find that every bit as impressive as the statistics that people collect. It is a characteristic of the medical profession that they operate hands-off and perform their evaluations entirely on a statistical basis. In the engineering profession, whenever possible we go and find out what it is like and subject ourselves to those conditions to gain an appreciation for ourselves. Sometimes I read documents from people who clearly have no direct experience. It is apparent from what they say. In this particular instance, occurrences are so comparatively rare amongst the general population that it is very easy to end up with a large number of negative responses and only a very small number of positive responses; yet the fact is that those positive responses can be directly associated with real problems.
Senator URQUHART: We are going to hear in a minute from Dr Leventhall. He has put forward in his submission that a much higher correlation in Mr Cooper’s work could be found between audible noise and sensations rather than infrasound and sensations. Do you agree with Dr Leventhall that the correlation that Mr Cooper found is statistically much higher with audible noise than infrasound?
Dr Swinbanks: There are both components of sound present. The definition of infrasound, according to Dr Leventhall himself, is that there is a very fuzzy boundary between infrasound and low-frequency noise. He has stated that that often causes confusion. In reports that he has written his definition of infrasound versus low-frequency sound, which is generally considered to be audible sound: he has defined 20 hertz on some occasions as being the boundary between the two effects and sometimes 16 hertz. In a different report he talked of 10 hertz to 200 hertz. Finally he even proposed 5 hertz to 200 hertz in a 2006 paper. So the point is that this definition of where you are between infrasound and audible noise is a very flexible definition. I do not consider that it is particularly important whether the noise is truly audible or just perceived as a sensation. The important effect is that people do detect something; they detect a sensation and can tell that something is happening. I learnt this 30 years ago when I was working on a gas turbine installation. Initially, I was very insensitive to the sound but, ultimately, I could drive up in my car and detect that the gas turbines were running even before the car engine had been turned off. There was a very marked increase in sensitivity. So I do not really think that it is important whether it is audible noise or inaudible noise that gives rise to the sensation. The fact is: people do experience real sensation, and these sensations can be very unpleasant.
Senator BACK: You mentioned size of wind farms. Were you referring to numbers of turbines or the actual physical size of the individual turbines, or both, when you made your comments in that regard?
Dr Swinbanks: I am referring primarily to the number of turbines. That is obviously related to the overall dimensions of the wind farm. But I have in mind, in particular, the Macarthur wind farm, which has very closely spaced turbines. It has a very large number — something in excess of 140. People are, I understand, experiencing adverse effects at distances of three miles. I believe that is a consequence of a large, closely spaced wind farm. Whether the effects would be as severe if the spacing of the turbines is made greater, I believe that would relieve some of the effects. But I think the main issue is the sheer number of turbines.
Senator BACK: You mentioned about what the 2009 American Wind Energy Association report had failed to take account of. You made the reference to increased sensitivity over time — increased exposure — and you gave an example of your own situation with gas turbines. One of the witnesses who has appeared before us, Dr Tonin, from this Association of Australian Acoustical Consultants, put to the committee that you could undertake this testing for infrasound using a pneumatic signal attached to hearing protectors effectively in a quiet room for a limited period of time. I think he mentioned 15 or 20 minutes. Could you comment on how much value you regard such testing would be in trying to come to terms with our situation?
Dr Swinbanks: My attention was drawn to that paper, and I have read it. I have two immediate comments. Firstly, he was attempting to distinguish whether symptoms were due to actual infrasound or due to nocebo effects. The important point is: there are two different outcomes which could distinguish between those effects, but, in fact, there are many more than two possible outcomes from the experiment. There are up to 16 outcomes of which only two are definitive outcomes relating to nocebo effects or infrasound effects. When I looked at the data, the most impressive correlation that I could see from the data was that the sheer action of putting on the headphones appeared to have increased the symptoms of the people being studied by at least 44 per cent. This was an experiment in which putting on the headphones had a measurable effect. I would argue that we do not yet know what exactly the mechanism causing people to suffer adverse effects. As I indicated, NASA, 30-odd years ago, had shown that people could experience adverse reactions at what are nominally very low levels of infrasound, but that was in houses where there was possibly vibration from the structures — and we do not know whether people are sensing anything through their body rather than their ears, because people often report in low-frequency noise or infrasound environments that they can feel —
Senator BACK: Can I stop you there. We need to get the answers fairly quickly so that all of us can have a go. You made reference to the circumstances of your own experience, where the wind was gusting and then was not gusting and the sound of the turbines with each. Some people have put to us the idea that an average sound or an average level is adequate. You in your paper have suggested that the use of an averaging technique may be missing cumulative pressure fluctuations and, in particular, peak pressure. Could you briefly explain that further and whether or not there is a value in averaged sound or averaged levels of infrasound decibels, please.
Dr Swinbanks: My immediate comment is that there is no value at all in an averaged level. In that example I gave, if you average it all, you find that the wind turbine level if anything would be less than the gusting level and you would then conclude that the wind turbines are not significant, whereas in fact it is very clear that they are significant. But the other important point is that there is a very well acknowledged paper that was written in 2004 by two authors, Moller and Pedersen, where they made it very clear that for the very low frequencies it is the actual shape and time history and peaks of the waveform that are important. In fact, Dr Leventhall, in an expert witness statement a couple of years ago, criticised me for supposedly not having read that report properly, but what I was doing was studying directly what the report recommended — namely, the time history and shape of the waveforms rather than long-term averaged versions of the waveforms.
Senator BACK: Thank you.
Senator LEYONHJELM: Dr Swinbanks, I have several questions. I hope we have time for them. Dr Leventhall was giving evidence in 2013 to a Vermont Senate hearing on the adverse health impacts of wind turbine operations in which he said they were ‘made-up, make-believe’, ‘hoo-hah’ and ‘a propaganda technique’. I understand he also dismissed some of your work on impulsive infrasound. Has he communicated those concerns to you?
Dr Swinbanks: He has not communicated the concerns directly. I have known Dr Leventhall for 40 years, but until very recently I had not seen him for 20 years. I was quite surprised, when I met him, that he appeared to have a very different perspective on the noise conditions in America from the perspective he gave at that Vermont meeting. When he was in the UK, he told me that he thought the sound levels in America were disgraceful. At the Vermont conversation, he attributed problems to ‘hysterical reaction’. The point is that permitted noise levels in the United States are significantly higher than in other countries and certainly higher than in Australia, so it is hardly surprising that there is what he called ‘hysterical reaction’. You would certainly expect that, if people are subjected to more adverse conditions, they are going to react and respond more strongly.
But it most certainly is not hoo-ha. I can say that from my own experience. There is no question that there are some significant effects. We do not know precisely what the mechanisms are. But people did not know what the mechanisms for seasickness were for many hundreds of years, and they still recognised the existence of seasickness.
Senator LEYONHJELM: In the NASA work in the 1980s, Kelley describes in detail the physical sensations resulting from infrasound. Are his descriptions consistent with what residents are now describing as the physical impacts of wind turbine sound?
Dr Swinbanks: Yes, I believe they are consistent. These symptoms have been known for a long time. Dr Leventhall says they are entirely consistent with his knowledge of low-frequency noise. He does not find it surprising, but he argues that it is not due to infrasound. As I have indicated, Dr Leventhall has even defined low-frequency noise as being from five hertz up to 200 hertz, which overlaps very substantially a region that most people tend to call infrasound. So we have a situation where, for frequencies around 12, 13 and 14 hertz, do you say, ‘That’s infrasound. That can’t be a problem,’ or do you say, ‘That’s low-frequency sound. The symptoms are perfectly understandable’? The fact is it is a very fuzzy distinction and you can place yourself either side of that boundary dependent on precisely how you choose to define the boundary. I believe that the symptoms are consistent. They are certainly consistent with low-frequency noise. It is a moot point whether or not people are subconsciously hearing something. They are aware of something. I have no doubts about the nature of the symptoms.
Senator LEYONHJELM: I just want to ask you a few technical questions. Your submission had some graphs that showed the pressure fluctuations and frequency. Mr Cooper’s report points out the need for narrowband measurements and not one-third octave bands for dB(A) or dB(G) when looking at infrasound and low frequency. Do you agree with that?
Dr Swinbanks: Certainly. I would not even dream of using one-third octaves or even averaging, over extended periods of time, just the pure spectrum levels. A proper analysis is both a narrowband frequency analysis coupled with a temporal analysis to look at the time history, as I commented earlier. If you go out to sea in a small boat, you do not worry about the spectrum of the waves; you worry about the shape of the next wave. This is what happens as you go down in lower and lower frequencies. For frequencies like 20 hertz and upwards, you tend to be more concerned with the blurring overall effect, but, as you get down to the very lowest frequencies, it is the shape of the individual waveforms that influences you. So one certainly should not be using these long-term averaging techniques.
Senator LEYONHJELM: Following up on from a question from Senator Back earlier in relation to peaks and averages, could you comment on whether or not it is possible to take a recording of infrasound or low-frequency sound — whatever you like — from a wind turbine and replicate it in a laboratory under controlled conditions in order to measure whether or not there is an adverse effect to it?
Dr Swinbanks: Yes, it is possible to do so, but the way in which people have been doing it so far, to me, seems a bit back to front. What they should be doing is, first of all, testing people who are known to be sensitive to wind turbines to try to find out what conditions enable an accurate replication of the effects. I do not see the point in just setting up an experiment in a laboratory and saying, ‘We didn’t observe anything’ if you have not first established, for a person who does suffer ill-effects, whether or not they actually respond to that test. There are real questions about what exactly are the important effects and what exactly should be reproduced in a laboratory. For example, I have quoted the NASA work of 30 years ago. People consider that, possibly, it was the vibration of the structures that people were sensing rather than the physical pressure variations of the infrasound. We do not know exactly what gives rise to the adverse effects. One has to validate any laboratory testing by being satisfied that people who are sensitive and have reported adverse effects can indeed experience those effects under the test conditions.
Senator CAMERON: Thank you for being here, Dr Swinbanks. You are three miles from the wind farm — is that correct?
Dr Swinbanks: That is correct.
Senator CAMERON: Was your house there before the wind farm was built?
Dr Swinbanks: Yes. I must make it clear that I am not directly complaining about those noise levels because at the moment the effects occur only when the wind is blowing from the south, which is only five per cent of the time. They only occur under circumstances of very severe temperature inversion. So it is a very occasional event. The point is simply that it can occur, and people who are in a position where they are encountering those sorts of conditions more frequently could also be expected to encounter such effects at such distances. The point that I am making is that such effects can be detected at these distances, not that those effects are a significant intrusion at the moment. But I would point out that in the future they are proposing to build turbines not just to the south of us, but to the west and the north-west, in which case those conditions may prevail for 35 or 40 per cent of the time. The fact is that modest numbers of turbines at sensible distances are not generally a consistent problem. Large wind farms operating under adverse circumstances can indeed be a significant problem at those sorts of distances.
Senator CAMERON: So when the turbines started to be built, was there an opposition group formed in your area?
Dr Swinbanks: There was never an opposition group as such, but there were a significant number of people who were making known their concerns. There was not a formal opposition group, but people were making known their concerns. The fact that there were two wind farms built at an early stage meant that people had some experience of what could be happening. The interesting feature was that you might say that those two wind farms, if you looked at them initially, looked pretty similar and pretty comparable; but one of them gave rise to very severe problems, while the other one did not appear to give rise to anything like as many complaints. The skill of constructing a good quiet wind farm is still pretty well lacking. It is very much a trial and error process, unless people obey sensible guidelines like ensuring that the separation between the turbines is of a sensible size and they are not choosing to mount turbines in locations, for example, on ridges where there can be a significantly distorted wind pattern and shear flow effects. The point is that there is a difference between a well constructed wind farm with sensible spacings and numbers and a poorly constructed wind farm.
Senator CAMERON: You also indicated that an inversion caused problems, and you gave evidence in relation to one night when it was not windy, and you had to keep going outside to check if the turbines were operating, then you became lethargic, you were losing concentration, you lost coordination when you were driving. Were you the only one in your household who had these symptoms?
Dr Swinbanks: It was not my household, it was the house belonging to some people who lived at the wind farm, who had asked me to take the measurements for them. Those people have experienced adverse effects to the extent that they actually had to rent alternative accommodation and go and sleep in the alternative accommodation at night. They initially tried to look at weather forecasts and decide if they could sleep in their own house or not, but they ultimately decided that the wind conditions could change during the night, and it could go from a benign night to a bad night. Therefore they began to sleep away from the property routinely and regularly. The particular point that I should like to make is that I was extremely surprised to experience these symptoms. I thought it was a non-event. But one particular point was that I was using a computer very extensively, and if there is a relation to motion sickness, I would certainly comment that if I am in a motor car and I try to use the computer or read — assuming I am not driving — I can very quickly become ill. I wondered whether this was purely conjecture, whether the fact that I was concentrating on using a computer actually enhanced the severity of the effects.
Senator CAMERON: Are you aware of the study that was done by Fiona Crichton, George Dodd, Gian Schmid, Greg Gamble, and Keith J. Petrie, titled ‘Can expectations produce symptoms from infrasound associated with wind turbines?’ It was a peer reviewed analysis reported in Health Psychology. They indicated that if there were high expectancy that you would get sick from infrasound then you would become sick. They did work with infrasound and sham infrasound, and it really did support the analysis that the psychogenesis and nocebo effect were real. Have you had a look at that?
Dr Swinbanks: Yes, I am familiar with that and I wrote a criticism of that document at the time. The point was that the difference between their sham infrasound and their real infrasound was essentially negligible. The real infrasound was at a level of 40 decibels, which is very low, and not surprisingly there was no difference in the response of any of the people between the sham and the actual infrasound. The other point is that the duration was only 10 minutes. In the effects that I described it took five hours for the full effects to become apparent.
I have related that whole situation to sea sickness. It used to be the case, in the 1970s, when I did a lot of sailing, that one would frequently encounter people who considered that seasickness was just psychological. Very often, they learned the hard way that it is not. But the point is that, if you wanted to test two groups of people for seasickness, you would not put two separate groups into two separate boats and put them on a flat, calm lake for 10 minutes and then announce that any reactions prove that seasickness was caused by a nocebo effect. That would actually be regarded as a joke. So I am afraid that I consider that that particular experiment was more an experiment in a pretty obvious psychology than anything relating to the validity of whether infrasound represents a real problem or not.
Senator CAMERON: So many questions, so little time. Thank you.
CHAIR: Dr Swinbanks, is the sound pressure level important when considering biological effects of infrasound and low frequencies, or could it be the frequency via acoustic resonance?
Dr Swinbanks: I think I should make it clear that I am not a biology specialist, so anything I say is amateur in that context. But I believe that the long exposure times can be a factor in inducing effects in people. Again, drawing a parallel with seasickness, it was not uncommon to go to sea for eight, 12 or even 24 hours and think, well, you are not going to get seasick this time, only to discover suddenly at the end that you do in fact start to succumb. In that context you can find that the onset of the symptoms can seem to be very rapid, even although you have been exposed for a long duration. So I think there are important considerations relating to duration of exposure.
I point out briefly that Dr Alec Salt, who is an expert on the characteristics of the cochlea, has suggested there is a phenomenon known as temporary endolymphatic hydrops, which is a progressive swelling and blockage of the little pressure relief hole at the end of the cochlea. If that becomes blocked then you can become very much more sensitive to infrasound. So it is quite possible to hypothesise that long-duration exposure is causing a blockage to progressively develop, and when it becomes severe then the person will start to experience much more extreme effects from the sound pressure than they would if there were no blockage.
So you could imagine in those circumstances that there might be a protracted period where there was no effect and then a comparatively rapid onset of effects. It would then take time after the exposure for those effects to clear, so you would then have persistence for some time afterwards. This is a whole area that requires a great deal more study. One of the conclusions, though, of the original 2009 AWEA report was that there was no need for any further research. I would completely disagree with that. I think it is apparent that people are now taking the issue seriously and at last people are beginning to investigate more thoroughly exactly what may be happening.
CHAIR: From what you have told me, I take it that the level of sound pressure is less important?
Dr Swinbanks: There are several factors that are important and when they come together they can effectively reinforce one another. I am not certain that you can take out one specific component and reject the rest. It is a combination of different contributions that can ultimately lead to the end condition. But the obvious conditions are length of exposure, sound pressure levels but also the frequency and the nature and character of the time history of the wave forms.
CHAIR: Thank you. We are running over time. If there are no further questions —
Senator BACK: I have one, but it will have to go on notice.
CHAIR: Dr Swinbanks, there may be further questions placed on notice by senators. We would appreciate it if you accept those and respond.
Dr Swinbanks: Certainly.
CHAIR: Thank you for your appearance before the committee.
Dr Swinbanks: Thank you for giving me the opportunity to speak. I am very grateful for that.
CHAIR: Thank you.
—SWINBANKS, Dr Malcolm Alexander, Private capacity
Monday, 23 June 2015, Canberra
(Evidence was taken via teleconference)
Author: Swinbanks, Malcolm; and Australia Senate Select Committee on Wind Turbines
Q1. You mention the NASA wind turbine research of the 1980s. Is that relevant to the type of wind turbines used today?
Research into very large (multi-megawatt) wind turbines began at NASA in 1975. Much of this work was undertaken by very competent aero-acousticians, drawing on experience gained in the context of propeller and jet-engine development, and which has successfully resulted in substantial improvements in aero-engine noise. They identified at an early stage why the existing “downwind rotor” turbines were so noisy, and in 1979 commenced theoretical and practical evaluation of the first very large “upwind-rotor” turbine, the 2.5 MW “MOD-2”. In this context, in 1981 they confirmed the predicted reduced noise characteristics, while also investigating the adverse power generation and noise effects associated with close spacings between wind-turbines. They subsequently identified additional circumstances under which the low-frequency and infrasound generation of such upwind-rotor turbines could be compromised, and performed important studies on the human perception of low-frequency noise and infrasound. The latter investigations initially concentrated on the noise characteristics of the earlier downwind-rotor turbines, but the underlying physics governing hearing perception relate also to the upwind-rotor configuration.
Over the intervening 25-35 years, the basic physics of aerodynamic noise generation has not changed, the adverse effects of unduly close-spaced wind-turbine interaction remain the same, and the characteristics of human hearing have not changed. These aspects all continue to have immediate relevance to modern wind-turbine installations, yet this research has often been dismissed as old-fashioned and irrelevant by the wind-development community.
Q2. How do wind turbines produce infrasound and is this hazardous to humans if they cannot hear it?
The infrasound is generated by the aerodynamic lift forces on the blades, which are necessary to provide the driving torque to rotate the blades and generate electrical power. Newton’s law requires that there are corresponding forces of reaction on the air passing over the blades. Although these forces may be comparatively “steady”, the constantly changing position of the blades means that the resultant force pattern acting on the surrounding air is also changing so that, inevitably, infrasound is generated. Additional factors, such as the difference between blades encountering slow moving air at the bottom of their cycle, and faster moving air higher up, causes further changes to the blade-lift and force pattern, which can result in higher intensity and more impulsive infrasound.
It has been considered that the levels of infrasound generated by wind turbines are too low to cause adverse health effects, but such opinions have often relied on a mistaken interpretation of the “threshold of hearing”. There have been numerous instances over the last 40 years where people have reported adverse effects at sound pressure levels which are too low for people to “hear” it, while recent research is starting to identify possible mechanisms by which this process may take place.
Q3. There seems to be a somewhat semantic argument regarding “sensitization” and “annoyance” related to wind turbine exposure. Can you explain what “sensitization” is?
This question is perhaps most easily answered from my own experience. From 1979 to 1981, I worked directly on the low-frequency noise and infrasound from a ground-based gas-turbine compressor installation in a rural area, which was causing complaints and sleep-disturbance for residents up to 1 mile away. When I initially started working at the site, I did not consider the noise to be excessive and in conditions of a brisk breeze it was barely perceptible even at 100 yards. After two years, however, I found that if I drove up to the site having just travelled on a noisy motorway,while still in the motor car with the engine running I could “sense” directly whenever the gas-turbines were operating. This was not an immediately audible effect, nor was it in any way associated with “annoyance”, but this “sensing” of whether or not the turbines were operating nevertheless prove to be unerringly accurate. This was a process of enhanced perception which had developed during my time spent on site, and which became increasingly apparent as time progressed.
Q4. You make comment regarding the inter-turbine spacing of wind farms and how this affects the noise produced. Could you explain this further?
When conducting ground-based noise testing of large fan-jet aero-engines, it is usual to suspend the engine on a gantry and mount an extremely large porous “golf-ball” enclosure over the inlet of the engine. Without this, the ingestion of turbulent airflow from ground-effect wind-shear can result in very substantially increased noise generation from the large ducted fan at the inlet of the engine. The porous “golf-ball” smoothes out this turbulence, to reproduce the much cleaner airflow characteristics similar to those encountered by an engine operating forward of the wing when the aircraft is in flight.
Without such measures, turbulent ingestion can unrealistically increase the measured noise levels of the aero-engine by as much as 15dB, which is completely unacceptable for precision noise research and certification.
Exactly the same situation now relates to the operation of wind-turbines. Closely spaced wind turbines result in the turbulent wake from an upstream turbine becoming incident on a wind turbine immediately downstream, and the resultant unsteady airflow over the blades of the latter causes increased fluctuating lift forces giving enhanced noise generation and a reduced blade life due to fatigue. As indicated above, the need to guard against such effects has been well-known to the aero-acoustics community for many years.
Q5. Would you care to elaborate on your comments concerning the change in human hearing threshold between noisy and quiet environments and explain how this effects people living in close proximity to wind turbines?
I first became aware of the effects of the variable hearing threshold when performing experiments in the active control of low frequency sound in the late 1970’s and early 1980’s. Such experiments involved generating low-frequency noise for extended periods of time, while frequently alternating between the unsilenced condition and then turning on the active silencing to yield significantly reduced noise levels. I found that there could be very different perception of the magnitude of the apparent change when going from long periods of the quiet condition (sensitive threshold of hearing) to the loud condition, compared to going from a protracted period of noisy exposure (raised threshold of hearing) to the quiet condition.
The hearing tests carried out by NASA in 1982 (referred to in my answer to Q1) confirm the relevance to wind-turbine noise. Tests of simulated wind-turbine impulsive low-frequency noise and infrasound under very quiet conditions showed perception could take place at a level significantly below the nominal threshold of hearing. Introducing an increased ambient background of 35dB required the simulated sound level to be increased by approximately 10dB in order for the sound to be perceived. With an ambient level of 45dB the level required for perception became even higher, being increased by a further 6dB.
Some acousticians have argued that since infrasound levels directly comparable to wind-turbine infrasound are not a problem in urban environments (55dBA), then they should not be a problem in rural environments (25dBA). This argument completely fails to take into account the very significant increase in hearing sensitivity in the rural environment. Perception and response in these two very different sound environments cannot be equated in this manner. An immediate consequence is that such levels can give rise to significantly greater sleep disturbance in the rural environment.
Q6. It appears that continued exposure to low-frequency noise and infrasound can result in progressively more acute physical sensitivity to the sensations. Could you elaborate on this for us?
I referred in my answer to Q3 to my own experience of increased perception and sensing ability, which for me appeared to have evolved entirely naturally. In a 2004 publication , Dr Leventhall stated “ If complainants spend a great deal of time listening to, and listening for, their particular noise, it is possible that they may develop enhanced susceptibility to this noise. Enhanced susceptibility is therefore a potential factor in long-term low frequency noise annoyance.” The only aspect of this statement with which I would disagree is that it places responsibility on the victim by implying that their own actions and behaviour have led to the enhanced susceptibility. But if the victim cannot escape the imposed disturbance, it is entirely likely that they will also be unable to escape the enhanced susceptibility.
I know sufferers from wind-turbine noise who report that the effects can be felt as pressure pulsations in the chest. One farmer has told me that he could even sense the wind-turbines while riding his tractor. So the overall nature of the sensations can amount to more than a simple impact on hearing.
A further comment is that I know of couples for whom the continuing stress of exposure to wind turbines has ultimately threatened the stability of their marriage. The long term consequences can result in more than just immediate stress-related health problems.
Q7. You comment on the differences between setback distances between Australia and the USA. Would you like to comment on what you consider to be the safe setback distances for wind turbines?
When I first became familiar with the problems of wind-turbines, it was immediately apparent that the setbacks such as 1000 feet and 1320 feet, which I encountered in the USA, were completely unacceptable. I then met people in the UK who were experiencing problems at 3000 feet, so I formed the opinion that a safe distance would be in the order of 5000-6000 feet. It should be noted that in the UK, the “ETSU” procedure is not to define a fixed setback distance, but rather to define permitted sound pressure levels and then to place the onus on the wind-developer to show that his proposed windfarm layout will not exceed these permitted levels. The overall assessment process takes substantial account of measured ambient sound pressure levels at various dwellings, coupled with anticipated projection of measured ground level wind speeds to the wind-turbine hub height, and estimation of the resultant wind-turbine noise generation under those ambient conditions.
Subsequently, however, my wife and I have experienced ( limited) sleep disturbance from the newly erected windfarm 3 miles to the south of us, under certain weather conditions. I have already described in my written testimony, the measurements that I had obtained at Ubly on the occasion that I became severely ill. At a later date, the Australian acoustician Les Huson sent me his infrasound measurements taken at the Macarthur windfarm, where residents were encountering severe adverse effects at a distance of 3 miles. I compared his infrasound signals to my Ubly measurements, and found them to be very similar indeed, so it came as no surprise that residents were reporting similar problems. It is quite apparent that for some windfarms, a distance even of 3 miles can be insufficient.
The problem is twofold. First, there is very little experience of the accurate prediction of the generation of infrasound and low frequency noise by windfarms, particularly if the wind-turbines are badly positioned with respect to each other. Secondly, the subsequent propagation of low frequency and infrasound is highly dependent on the geometry of the terrain, the temperature profile, and the wind-profile. These factors can be difficult to predict with a high level of confidence.
Thus, I regret that at this stage I cannot give a firmer recommendation as to appropriate setbacks, but to emphasize that as the size of windfarms is increased, so the need for greater setbacks becomes increasingly apparent.
Q8. There is a lot of controversy surrounding what is called the Nocebo Effect. Can you comment on whether or not you think that this could be an explanation for the adverse health effects reported by residents living in close proximity to wind farms?
My personal opinion is that attributing the problems suffered by people in close proximity to wind farms to the “Nocebo Effect” is simply a convenient “get-out”. There have been a sufficient number of examples where communities have welcomed the introduction of wind-turbines, only to discover the adverse effects after they commenced operation. Moreover, there are also common factors with respect to overall noise levels, or low-frequency noise levels and infrasound levels to indicate that such adverse effects are not of a completely random nature. In many circumstances they are only to be expected. Unfortunately, failure to acknowledge these problems, or attempts to dismiss them as “Nocebo Effects” can have the opposite effect, namely of strengthening opposition to wind farms. During the 1970’s and 1980’s, when I encountered noise problems of a similar nature, it was quite customary to acknowledge the problems and attempt to address them. Now the policy is often a complete refusal to acknowledge them.
Q9. What were your experiences when you investigated a wind farm in Ubly. How can you be sure the effects you experienced were due to the wind turbines and have you ever experienced this reaction again?
The sequence of events at Ubly started early in the evening, at a house located downwind of six turbines, with the nearest at 1500 feet. There was little or no wind at ground levels, but there was clearly wind at the height of the turbines which were running steadily but with little sound. Inside the house it was impossible to hear or feel anything. We set up instrumentation, and started taking 30-minute measurements which I then would analyse before undertaking the next recording using modified instrumentation parameters. I certainly did not consider that the measured infrasound levels were sufficient to be of significance. After 1 hour I began to feel lethargic with reduced concentration, after 3 hours I felt distinctly nauseous, but only after 3.5 hours did I start to attribute the problem to the wind turbines. After 5 hours I was extremely relieved to leave, only to find that my ability to drive was severely compromised, to an extent that I have never previously encountered.
The effects had started when I was working amongst the wind turbines, and finally abated 5 hours after I had left. Although I have encountered some adverse effects in the past from low frequency noise and infrasound, I have never before encountered anything of this severity.
On a few subsequent visits to this house under similar conditions I have again experienced the preliminary indications of this process, but I have always made a point of leaving promptly, before any effects have had time to build up.
Q10. A number of residents report sleep disturbance when turbines are operating. Do you think this could be due to the wind turbines and if so, are these effects serious?
The low-frequency levels and infrasound levels for some residents near to windfarms are such that it is only to be expected that they have experienced sleep disturbance. It is my understanding that protracted sleep disturbance can ultimately lead to more complex and serious medical effects. Dr Chris Hanning, a retired specialist from the sleep research laboratory in Leicester, England, whom I believe has given testimony to this Senate Hearing, has often commented on such medical effects.
Q11. Dr. Leventhall is somewhat critical of your work and conclusions. Can you give us your side of the story?
In written testimony prepared for the Kent Breeze 2011 hearings in Ontario, Canada, and in verbal testimony in 2013 in Alberta, Dr Leventhall has commented on my work. Several of his comments have resulted from his inaccurate recollection and description of my papers and presentations. But Dr Leventhall also stated that I had misunderstood 1982 NASA work on the perception of simulated low-frequency wind-turbine noise, arguing that it related only to old-fashioned downwind-rotor turbines. He had failed to appreciate that the relevant issue lies in the basic physics of the impulsive process. This physics applies not just to the impulsive low-frequency sound of the early wind-turbines, but also to the impulsive infrasound emissions from modern upwind-rotor turbines. Failure to quantify this correctly has led to many instances where modern-day acousticians have examined infrasonic wind-turbine power spectra from a superficial perspective and consequently have completely underestimated the likelihood of its perception.
In his 2013 testimony, Dr Leventhall stated of my work “he tends to overcomplicate things. He can’t do things in a sort of simple broad-brush clear way”. He also took me to task for having apparently failed to read properly the definitive 2004 paper of Moller and Pedersen  I had read and cited this paper in my very first wind-turbine submission in December 2009 – yet the specific issue which Dr Leventhall claimed that I “obviously did n’t read” is not mentioned in it. His comments related to my 2011 paper  in which I had started out by referring to an acknowledged problem for which no-one could offer any solution. I proposed a solution, and showed how it endorsed a commonly applied rule-of-thumb. My intention was to demonstrate that my investigations using rigorous methodology yielded a start point which could be confirmed, namely a result consistent with existing empirical evidence. I then developed this methodology further, drawing directly on Moller & Pedersen’s argument that at extremely low frequencies, it is the time history of the infrasonic waveform that is important. In this way, I could investigate methodically what consequences might follow-on from their statements.
It should be remarked that in the early 1970’s, it was extremely difficult to assimilate the vast amounts of data that often are associated with acoustic analysis. In those circumstances, there was often no alternative but to adopt a “broad-brush” approach. The advent of mini-computers and powerful digital analysis equipment later in that same decade transformed the nature of acoustics research and analysis, so that it was no longer necessary to rely on informed guesswork. This revolution became extremely important in such areas as the development of quiet aero-engines, or very quiet submarine design, together with efficient underwater sound detection.
I decided to apply similar methodology to the process of hearing perception, using no more than the established macro-level hearing characteristics which are well known to audiologists, but then examining what features emerge when one conducts dynamic simulation of these processes. Using this approach, I concluded that perception of infrasound may be possible to very much lower frequencies and sound pressure levels than has hitherto been considered to be the case. Moller & Pedersen’s paper states “Generally low-frequency and infrasonic sounds from everyday life are not pure tones alone, but rather combinations of different random noises and tonal components. It is however, impossible to make thresholds for all imaginable combinations of sounds that exist ….. “.
The unique and all-pervading noise characteristics of wind-turbines represents a situation where it is essential to make constructive inroads into this apparently “impossible” task. This necessarily requires a much more rigorous approach than simple “broad-brush” procedures.
Q12. What level of infrasound, low frequency noise would be safe in your view?
As a result of increasing familiarity with these problems, I have had to continually revise and lower my estimates of levels that would be considered to be safe or acceptable. I very recently met with a family who have been suffering for almost 6 years with the problem of sleep disturbance from an adjacent windfarm. Initially they could not sleep in their bedrooms. They converted the underground basement of their house into sleeping quarters, and found some immediate relief moving to this location. But then with the passage of time, they found that this was no longer satisfactory and they appeared to be becoming increasingly sensitive. So they constructed a dedicated sleeping chamber within the basement, with improved sound proofing which yielded measurably lower infrasound and low-frequency sound levels. Moving into this additional chamber was found to give further relief, but now in the longer term, even in this lower sound environment they are again finding their sleep patterns to be unsatisfactory. The noise characteristics of wind turbines are essentially unique, which results in them causing annoyance at much lower levels than more conventional noise sources (e.g. motor vehicles, railways and aircraft). For higher frequency audible components of the spectrum, external sound pressure levels of 35dBA would appear to be acceptable for the majority of people, but placing firm figures on safe levels of low-frequency noise and infrasound is still a difficult question given the potential for enhanced long-distance propagation under adverse weather conditions.
Q13. What research should be carried out as a priority to progress our understanding of this whole issue of adverse human effects from wind turbine emissions?
In the early 1980’s, NASA published a report containing a figure indicating that adverse effects had been reported in houses where the measured infrasound levels were significantly below the threshold of hearing. They commented that the effects might be due to subsequent induced vibration of the house structure by the pressure variations. At present, it is not known whether the effects are due to pure infrasound in the absence of any other sound components, or vibration induced by infrasound, or (according to my own analysis) by the interaction of low-level higher frequency noise causing the infrasound to become perceptible.
A first priority is to identify which of these mechanisms can successfully reproduce the symptoms in people who have already reported adverse effects from wind-turbines. Hearing processes and associated neurological response are extremely complex. It is well known that when people enter a well-designed anechoic chamber, which is intended to suppress all acoustic reflection and reproduce an entirely “dead” acoustic environment, they can often feel that they are being suffocated and suffering from lack of air. So even the complete absence of sound and reflections can result in strongly adverse physical reactions !;
It has even been considered that the adverse effects from wind-turbines might be induced by some electromagnetic influence. This would seem unlikely, however, because similar effects were reported back in the 1970’s from aero-engine test-beds, where there is no accompanying electromagnetic effect. Moreover, the long-distance effects appear to correlate with specific atmospheric sound propagation conditions, which would suggest that some aspect of sound generation is responsible.
Q14. What factors do you consider should be included in the development of appropriate setback distances for proposed windfarms?
I consider that it is important to obtain better assessment and prediction techniques for infrasound and low frequency noise from windfarms. Until comparatively recently, such measurements have only been made on a limited basis, and the incorrect use of long averaging times and third octave measurements has substantially compromised their usefulness. It is important to obtain consistent time-data records using instrumentation having appropriate characteristics for measurements at these frequencies. Deployment of low cost, very low frequency microbarometers, coupled with more conventional higher frequency microphones can be useful in providing a detailed overall perspective.
Setback distances are more appropriately defined according to likely expected noise levels, rather than being defined by single all-embracing distances which may not be relevant for extremely large windfarms.
There have been recent arguments  that dBA levels provide an adequate prediction of wind turbine low-frequency noise levels, resulting from a supposedly close correlation between dBA levels and low-frequency noise. This cannot be correct under all circumstances, because the effects of temperature gradient and wind-shear can significantly modify the propagation characteristics of low-frequency wind-turbine noise, whereas the frequency components which dominate dBA measurement are generally much less sensitive to these effects.
Q15. Do you consider we have enough information about the factors to be able to make assessments that ensure public safety?
There is still much to be learned about both the operating characteristics of large windfarms, and the detailed factors which can compromise health and well-being. I believe the most constructive improvement will come about when the already-known problems that residents can face are formally acknowledged. This would permit the adoption of mechanisms by which people who experience proven difficulties can receive reasonable compensation or redress, without having to undertake solitary and protracted legal action against heavyweight, subsidised organizations.
Q16. How do you propose that councils ensure they are able to assess the quality of information presented in windfarm consent applications?
This is a difficult question which I cannot answer fully. While there are many acousticians who appear to possess the necessary formal qualifications to address these issues, there can still be a considerable lack of real experience. Given that the deployment of wind-turbine technology continues to give rise to unsatisfactory installations, conventional qualifications alone do not yet guarantee a competent outcome. Pressure to conform to political or financial pressures relating to specific windfarm installations can also lead to compromise recommendations which may ultimately prove to be less than satisfactory.
Q17. You comment that low-frequency noise could cause a blockage in the helicotrema. If that is true, could this result in less of the low-frequency signal reaching the underside of the basilar membrane, thus it would not stimulate hearing so readily? However, could this cause an increase in pressure in the endolymph, which is connected to the utricle and saccule of the vestibular system? Could this pressure imbalance actually cause vestibular effects as the endolymph is connected to the utricle and saccule? Could this cause displacement of the otoliths and generate motion sickness and vertigo? Could you please comment on this?
This question Q17 suggests a credible overall explanation for the adverse effects. But the initial statement “less of the low-frequency signal reaching the underside of the basilar membrane, thus it would not stimulate hearing so readily” is incorrect. The purpose of the helicotrema is to provide pressure relief across the basilar membrane at very low frequencies. Natural background infrasound tends to rise in level as the frequency reduces, and it is important to prevent the basilar membrane from being unduly displaced by such natural pressure variations at these very low frequencies. Fluid motion through the open helicotrema ensures that the pressure tends to equalize on both sides of the membrane, thus minimizing displacement and reducing hearing sensitivity at such frequencies. But if the helicotrema is blocked, pressure balancing is prevented, and there is higher pressure on the upper side resulting in unwanted increased membrane excursions and resultant increased excitation of the cochlea hair cells. This in turn causes significantly increased sensitivity to very low-frequency sound (by as much as 20dB according to Dr Alec Salt.)
Q17 then considers whether such a one-sided pressure distribution can cause increased pressure within the endolymph – the separate central fluid which communicates with the utricle and saccule of the vestibular system. This quite possibly may be the case, so the same process which gives rise to increased hearing sensitivity may perhaps give rise to greater excitation of the otolith organs. If such enhanced excitation does indeed occur, it would then be expected to give rise to motion sickness and vertigo.
Q18. Is the sound pressure level important when considering biological effects of infrasound and low frequencies or could it be the frequency via acoustic resonance? Would this mean the level is less important?
I am not aware of any specific acoustic resonances within the conventional hearing organs which would account for extremely low frequency perception. At the higher frequencies, travelling waves along the basilar membrane “bunch together” at specific locations, giving a maximum resonant response to specific frequencies. Different locations along the basilar membrane are tuned to respond preferentially to different parts of the frequency range, with the highest frequencies concentrating at the input end of the cochlea (stapes) and the lowest frequencies concentrating towards the apex of the cochlea. At the very lowest frequencies, the entire basilar membrane tends to be excited.
The sensitivity of the vestibular organs which has been demonstrated by response to motion sickness testing does however show a broad bandwidth “resonant” type of behavior, with maximum response over the frequency range 0.07Hz to 0.7Hz. This corresponds to motions with period 1.4 seconds to 14 seconds. Over this frequency range, frequency can have as much or more importance in defining the response than the level. The fundamental blade rates of the largest, most modern wind-turbines do start to encroach on the upper end of this frequency range. It may be relevant that the adverse effects of wind-turbines appear to have become more apparent as the overall size has increased, and corresponding blade-rate frequencies have reduced. Until the precise mechanisms governing adverse health effects from wind turbines have been fully identified, such commentary is largely speculative, but may still represent relevant contributory information.
1. H.G. Leventhall. Low frequency noise and annoyance Noise & Health, Volume 6, 23 pp 59-72, 2004
2. H. Moller & C.S. Pedersen. Hearing at Low & Infrasonic Frequencies, Noise & Health, Volume 6, Issue 23, April-June 2004
3. M.A. Swinbanks. The Audibility of Low Frequency Wind Turbine Noise. Fourth International Meeting on Wind Turbine Noise Rome Italy 12-14 April 2011
4. R.G. Berger, P.Ashtiani, C.A. Ollson, M.W. Aslund, L.C. McCallum, H.G. Leventhall, L.D. Knopper. Health-based audible noise guidelines account for infrasound and low-frequency noise produced by wind turbines. Frontiers in Public Health, Volume 3, Article 31, February 2015
Additional Questions on Notice from Senator Anne Urquhart
AU1. What symptoms do you think are directly attributable to the operation of wind farms if there are any? What research are you basing this on?
For many years, it has been well-known and acknowledged that low-frequency noise can give rise to such symptoms as nausea, dizziness, headaches, feelings of pressure in the chest, while at nighttime it can give rise to very significant sleep disturbance. Over the years that I have worked with such noise, I have experienced several of these effects myself; these effects are not generally considered to represent an issue which is disputed in the acoustics community.
The transition between low-frequency noise and infrasound is often regarded as a “fuzzy boundary”, so it is not unreasonable to conclude that such effects would also be experienced in portions of the infrasound regime. This depends, however, on the existence of appropriate “perception”, so the thrust of my own research has been to investigate whether the present conventions for assessing infrasound perception are adequate. I have been able to show that these conventions are not adequate, and moreover that this is consistent with hitherto unexplained experimental reports in the peer-reviewed acoustics literature.
AU2. What proportion of the population do you think are susceptible to health impacts from wind farms if there are any? What factors do you think make people more susceptible than others? What research are you basing this on?
Based on the proportion of complaints arising from wind-farms, which are closely dependent on the setback distance, I would venture that around 5-10% of people in the immediate vicinity may suffer from direct physical effects. A much more common complaint is that of sleep disturbance, which over a long period of time can give rise to a very wide range of adverse physical effects.
One of the factors which is often not adequately taken into account is the fact that people cannot escape or gain relief from the exposure, nor can they be certain whether it will continue for hours or days at a time. This greatly enhances the psychological stress. Any noise which is purely transient, and is clearly going to come to an end ( for example the noise of a nearby combine harvester or large agricultural machine) is much more tolerable because it will only be of limited duration.
In a 2013 paper , Paul Schomer set out a convincing statistical argument indicating a strong correlation between people who suffer from motion sickness and those who experience adverse physical effects from wind turbines. But I have also been told of people who are not necessarily sensitive to motion, yet who have nevertheless experienced adverse effects from wind turbines.
AU3. How long do you think it takes from wind farm operations to manifest if there are any? What research are you basing this on?
I have read reports of people becoming adversely affected within 20 minutes in cases of nearby wind turbines. I know personally people who simply can no longer tolerate being in certain areas of their property and who react almost immediately, as a result of continuing exposure accumulated over several years. They do not “get used to it”. In my own case, I first experienced mild symptoms which I did not immediately recognize as due to wind turbines, after 1 hour. After 3 hours the symptoms were very clear and unpleasant, and after 5 hours I was only too glad to leave. I have based my conclusions on over 5 years of discussion and interaction with windfarm residents, both in the USA and the UK, together with my own direct experience.
AU4. What do you believe is an appropriate seback distance for wind farms? What research do you base this opinion on? In my answer to Q7 in the preceding set of questions, I remarked that with the passage of time and experience I have consistently revised and increased my assessment of an adequate setback distance. I believe the issue becomes increasingly important as the size of windfarms (ie number, land area, and individual size of turbines) is increased. I do not believe that sufficient consideration is being given to the cumulative effect of large numbers of turbines, or of the consequences of constructing multiple windfarms in close proximity.
AU5. You said in your testimony: “I believe that in a situation where people are reporting the effects that they observe while at the same time the operating characteristics of the wind farm are being monitored remotely, if you find that there is then a close correlation between those two situations, that does imply that there is a significant link and that people are reacting to real events.” Could you direct the committee to any peer-reviewed research published in an indexed medical journal that has found people’s perceived effects have been closely correlated to wind farm operating characteristics when they were operating within prescribed guidelines? If so, could you confirm what the statistical strength of this correlation was?
Unfortunately, I don’t think that there is to date any direct, peer-reviewed medical research which has addressed this specific issue. The recent work of Steven Cooper , an acoustician, represents probably the first quantitative study relating to this aspect. Such studies require close cooperation between the community and the windfarm operator.
In oral questions from the committee, I was asked (the equivalent of) whether I considered a success rate of 1/3rd in this particular context represented an adequate measure of statistical confidence. If one tosses a coin many times, one would expect to guess correctly “heads or tails” on average 50% of the time. In that particular context, a success rate of only 1/3rd would be a poor result, implying a problem with either the coin or the process. But suppose you have one person placed in an enclosed, windowless sound-proof room, who over an extended time interval tosses a coin many times at completely random well-separated instants, and allows it to land each time on the floor. Another person outside the room is asked to press a button whenever he thinks the coin has just landed. Under such circumstances, a success rate of 1/3rd would be a result indicating that some form of perception is likely to be taking place.
5. Schomer P, Edreich J, Boyle J, Pamidighantam P (2013). A proposed theory to explain some adverse physiological effects of the infrasonic emissions at some wind farm sites. 5th International Conference on Wind Turbine Noise 28-30, August 2013
Public Hearing, 23 June 2015
Parliament House, Canberra, ACT
Author: Castelo Branco, Nuno; Alves-Pereira, Mariana; et al.
In November 2006, 4 Industrial Wind Turbines (IWT) were installed in the vicinity of a residential dwelling in Portugal. In March 2007, this team was contacted by the family requesting assistance in dealing with their Infrasound & Low Frequency Noise (ILFN) problem that they claimed was being generated by the IWT. The family began legal proceedings for the removal of the IWT, and in September 2007, this team’s first report was presented at the 2nd International Meeting on Wind Turbine Noise. In June 2010, a follow-up report of this case was presented at the 14th International Meeting on Low Frequency Noise and Vibration and its Control, wherein ILFN-induced pathology was confirmed through histology in this family’s thoroughbred horses. The goal of this report is to provide second follow-up to this case, five years later.
Nuno A. A. Castelo Branco, M.D., Senior Surgical Pathologist
Mariana Alves-Pereira, Ph.D., Biomedical Engineer, Lusófona University
Augusto Martinho Pimenta, M.D., Senior Neurologist, Julio de Matos Hospital
José Reis Ferreira, M.D., Senior Pneumologist, Clínica Doentes Pulmonares
Presented at EuroNoise 2015, 31 May–3 June, Maastricht, The Netherlands