Note: Where available, the PDF/Word icon below is provided to view the complete and fully formatted document
Senate Select Committee on Wind Turbines

HANSEN, Emeritus Professor Colin Henry, Private capacity

CHAIR: I declare open this in camera hearing. Welcome. I must advise you that it is not the intention of the committee to publish or present to the parliament all or part of the evidence you are about to give. However, you need to know that it is within the power of the committee to do so and the parliament has the authority to order the production and publication of undisclosed evidence. You should also note that an individual committee member may refer to in camera evidence in a dissenting report to the extent necessary to support the reasoning of the dissent. However, the committee would try to seek your view prior to any such proposed disclosure.

Could you please confirm that information on parliamentary privilege and the protection of witnesses and evidence has been provided to you.

Prof. Hansen: Yes, it has.

CHAIR: Thank you. The committee has your submission. I now invite you to make a brief opening statement. At the conclusion of your remarks I will invite members of the committee to put questions to you.

Prof. Hansen: Thanks for inviting me here today. By way of introduction, I have been consulting, teaching and researching in acoustics for all of my professional life, which is 40 years. I retired three years ago, but in 2012 I was awarded an Australian Research Council grant to investigate the effects of wind farm noise on rural communities. I first became interested in wind farm noise in 2010 when I was asked by legal counsel acting for residents if I would be willing to review a report on noise level predictions for a new wind farm. The report had been prepared by a consultant engaged by the wind farm industry, and I noticed that there was no mention of noise character in the report or any assessment of potential annoyance of residents or the accuracy of the noise predictions. This sort of reporting has continued on to the present day.

I became familiar with current wind farm noise guidelines as part of my work and I have read work from many other researchers who have assessed the effects of noise from various wind farms in a number of different countries. Based on my own experience measuring wind farm noise as part of my Australian Research Council grant and all the other literature I have seen, I have no doubt in my mind that wind farm noise has serious adverse health effects for some people, resulting in some of them having to leave their homes. Whether or not the noise causes direct physiological responses or any adverse effects are due to stress arising from audible noise causing sleep disruption and annoyance has not been definitively proven either way. However, there have been two separate theories published that could explain a direct physiological effect of infrasound produced by a wind farm. There is also a theory that could explain why low-level audible noise produced by wind farms is disturbing to some people.

The percentage of people affected by wind farm noise is dependent on the wind farm and seems to be very difficult to quantify; you see lots of different percentages. There is anecdotal evidence that suggests that, for affected people, the symptoms become worse over time and seem to disappear when they leave the vicinity of the wind farm for an extended period. It is not surprising that wind farm noise affects some people and not others, because hearing thresholds and sensitivity to infrasound can vary considerably from one person to another. It is interesting to note that hearing threshold values are measured for tonal noise, which has a small ratio of peak noise to average noise level. It is also published for the 50th percentile level of the population. If the hearing threshold follows a normal distribution—and many people think it does—that means between two and three per cent of people would have a threshold level 10-12 decibels lower than the published levels. Wind farm noise also has much higher peaks than normal tonal noise used for threshold-level measurements, so it is very likely that people can detect wind farm noise when the average level is below the threshold of detection for normal people.

Another thing we do not understand is what prolonged exposure to annoying sound does to some people’s hearing threshold or to their ability to detect infrasound without actually hearing it by experiencing symptoms such as pressure in their ears or nausea.

I would like to go now to World Health Organisation publications of 1999 and 2009. The night-time guidelines for Europe recommend much lower indoor night-time noise levels than are achieved by following current wind farm noise guidelines in Australia. The World Health Organisation recommendations for maximum night-time outdoor noise levels are based on traffic noise in urban and suburban areas in Europe and are designed for people living in these areas. The World Health Organisation document states that their recommended levels should be considerably reduced in situations where the noise is dominated by low frequencies, which is the case for wind farm noise.

An important point made on page 105 in that document is that guidelines for acceptable outdoor noise levels assume there is a 21 A-weighted decibel noise reduction from outside to inside when windows are open. The actual noise reduction is dependent on house construction as well as the frequency content of the noise incident on the house and may be approximately correct for urban noise in a European city. However, it is a gross overestimation of the noise reduction that we achieve with wind farm noise in rural communities in Australia.

We have collected a large amount of data for rural residences in South Australia near wind farms and, when wind farm noise is the intruding noise, the outside-to-inside noise reduction is closer to seven A-weighted decibels with windows open compared to the World Health Organisation’s 21 and 15-18 A-weighted decibels when the windows are closed. We also have data that shows much smaller noise reductions at low frequencies and have even measured higher noise levels inside the house than outside the house for some infrasound frequencies. There is a technical reason for that which I can explain later if you wish.

On page 109 in the World Health Organisation document the recommended outdoor night-time noise level is 40 A-weighted decibels, and that assumes outdoor-to-indoor noise reduction of 21 with windows open. There is also an interim recommendation of 55 in situations where it is not possible to achieve 40. However, the World Health Organisation document points out that this interim level will not protect vulnerable groups of people. What is worse for rural residents living near wind farms in Australia is that background noise levels are much lower than they are in urban areas, so any additional noise source is going to be much more intrusive even if it is at very low levels.

There are a number of problems with existing wind farm guidelines. According to the South Australian guidelines, an intrusive noise can be 40 A-weighted decibels for rural communities because they are classified as rural industry. Anyone that sells any product is an industry, so they have a 40 A-weighted decibel level. If you live in a rural town environment, you have a 35 A-weighted decibel maximum allowed level. For the reasons that I have just mentioned, according to the World Health Organisation, these noise levels will result in interior noise levels that are much higher than are recommended.

Another problem with current guidelines is that they do not address the character of the noise. This includes the frequency content as well as the variability in a repeating periodic wave—often referred to as amplitude modulation. Noise varying in this way is far more annoying than randomly varying noise and, when it is infrasound, some researchers have postulated the effect on the vestibular system could be similar to motion sickness even at levels below the threshold of hearing. Noise dominated by low frequencies is more annoying than noise with a balanced spectrum, and low-frequency noise from wind farms is particularly intrusive in rural residences due to the lack of other background noise in the area.

Lastly, I would like to talk about compliance checking. According to guidelines, it is sufficient to acquire a large number of data points each representing a 10-minute average. Then the data are plotted on a graph of noise level versus wind speed and a line of best fit is drawn through these data. We have many data points that are 10-minute averages that are above the line, so there are many 10-minute periods where the noise levels exceed the guidelines. Not only that; a 10-minute average has lost all the peak levels. On top of that, we use an L90 level. L90 means a level that is exceeded 90 per cent of the time. If noise is going up and down all over the place, all these high levels are completely removed and all you are measuring is what is there 90 per cent of the time.

In summary, I think there is a case for a complete and independent review of wind farm noise guidelines in Australia. I would be happy to answer any questions on what I have just said or my submission.

CHAIR: Thank you. Senator Leyonhjelm.

Senator LEYONHJELM: Did you hear the evidence from the group of acoustics people this morning?

Prof. Hansen: No, I was not here this morning. I am sorry.

Senator LEYONHJELM: It will be on Hansard. It was not in camera, so you had better read it. They kind of disagreed with almost everything you just said, so there is clearly more than one school of thought in the acoustics world. Is your school in the majority or minority?

Prof. Hansen: I would have to say a minority. A lot of these acoustic consultants have been working for the wind farm industry for some time. I do not know why they are doing things how they are. Perhaps they are doing things how they are asked to do things rather than how they should be doing things.

Senator LEYONHJELM: You were involved in the NHMRC review, I think.

Prof. Hansen: That is right.

Senator LEYONHJELM: What was your role?

Prof. Hansen: I was asked to review the draft report.

Senator LEYONHJELM: What was your view on that?

Prof. Hansen: I have my review I submitted as part of my submission, but it is a bit hard to summarise; it is quite detailed. It is attached to the back of my submission.

Senator LEYONHJELM: Could you give me a quick verbal summary of it?

Senator CANAVAN: Could I follow up? In your submission you have said that many reports were not included in their literature review and others that were were poor in quality. Could you expand on that point?

Prof. Hansen: Certainly. I have read a lot of the literature, and there are many well-known researchers who have published papers. Some of the papers do not follow the gold standard of epidemiology studies. That means you have to have a control group and another group and that thousands of people should be involved. Because they did not follow that process, they would seem to be disregarded. There are lots of case studies where people have looked at individual cases. As you accumulate lots of case studies, you get an idea that there are not just one or two people; there are hundreds and hundreds of people who have been studied individually and have been affected. But none of that seems to have been accepted as good evidence by the NHMRC even though the researchers doing this are medical researchers who have good reputations. It seems that the NHMRC were a little bit stringent on what they accepted as evidence.

Senator LEYONHJELM: You are not the first to have said that. You said there needed to be a review of wind farm guidelines, which I think I probably agree with. The question would be: if you were to nominate a standard for noise, including low-frequency and infrasound, what would that standard be? The reason I ask is that we have heard very little agreement on what a ‘safe’—for want of a better word—level is.

Prof. Hansen: Yes. It depends on how you measure it, as you see. Thirty-five is way too high. Thirty A-weighted decibels is sort of okay provided there are other things like a maximum low-frequency limit. You would look at all the noise below 160 hertz, for example, and have a limit. You might even divide that up into frequency bands, have a limit for each frequency band below 160 hertz and say you cannot exceed that limit.

Senator LEYONHJELM: So 30 A-weighted decibels is for audible noise.

Prof. Hansen: That includes the entire frequency range. If you restrict the allowed level to 30 A-weighted decibels then it is likely that lower frequencies might be reduced to an acceptable level, but we do not really know that yet. We do not have enough data for that. We would really have to go through all the data we have recorded in the last three years, look at everything and come to some conclusion.

Senator LEYONHJELM: The acoustics consultants this morning did not think much of the fact that we were a little sceptical. Is measurement of infrasound and low-frequency sound a technical problem? Is there an issue with accurate measurement?

Prof. Hansen: No, there is no issue with measuring infrasound low-frequency sound, but there is an issue with measuring low-level sound. When consultants go out and do a background noise measurement, they have instrumentation that cannot measure below about 18 or 20 decibels. If you have electronic noise of 18 decibels in your instrumentation and you have actual real noise of 18, the noise that you will see on the dial will be 21. Any noise below 28 decibels will have a ½-decibel error or more. That means that, when you look at this plot of all this data of noise level versus wind speed, right down at the bottom end it starts to flatten out because their instruments cannot measure the low-level noise. Then they do a curve fit on their data. When you curve fit wrong data, the curve ends up being in the wrong place, so they do this very complicated polynomial fit for the data, and it is just wrong.

Senator LEYONHJELM: We heard from the Mayor of Goyder this morning, and one of his lines was, ‘If we were doing it again’—in relation to wind turbines—‘we wouldn’t do it the way we’ve done it.’ On that basis, given what we know now about wind turbines, if you were going to write national guidelines in relation to sound, what would you put in them?

Prof. Hansen: Right now I would say there could be no noise more than 30 A-weighted decibels and there should be a five-kilometre buffer zone. If it could be proven that noise levels between four and five kilometres were not going to be a problem, then maybe that is acceptable.

Senator LEYONHJELM: Would you have a different standard below 160 hertz?

Prof. Hansen: Yes. There is a Danish standard that says you should not exceed 20 decibels in that low-frequency range, but I am not sure whether or not that is adequate. There are no studies being done on that. There is also a DEFRA standard from the UK which also talks about allowable levels below that, but I cannot remember offhand what the number is.

Senator LEYONHJELM: Right.

Senator DAY: I must confess I have a bit of a science background—not quite of your stature in that area. I find it fascinating to listen to our opening panel and then hear you say almost the complete opposite of what they were saying.

Senator Leyonhjelm asked if you were in the majority or the minority. I recall the famous case when Albert Einstein was approached by 100 scientists to say that he was wrong, and they had signed a petition. He said: ‘What do you want 100 for? I only need one to prove that I’m wrong.’ Sometimes the opinion of the majority is not necessarily correct. We had evidence yesterday from a Dr Huson, and he talked about instrumentation and measurement using, not microphone, but microbarometers. Do you want to comment on whether you agree?

Prof. Hansen: Yes, I am familiar with his work. I agree that he is getting good data with his microbarometer survey. There is no reason why they would not work. In fact we have been using similar microbarometers, not the one he has designed and used. You can purchase them from the US for about $1,000. We have been using one of them to back-up our low-frequency measurements with microphones.

Senator DAY: We also had evidence from Andrew Bell who talked about measuring infrasound, recording it and then playing it back to people in a simulated kind of environment. He said that it is not possible to properly replicate or simulate infrasound when you are playing it back.

Prof. Hansen: I do not know that I agree with him that it is not possible, but is difficult. There are now two devices on the market that produce infrasound at levels that would be easily sufficient for wind farm noise investigations in a room, but it requires setting up the room. You cannot have the door open and get the same noise level or frequency content as when the door is shut, for example.

Senator DAY: The reason I ask is it seems that, with a lot of witnesses that have appeared before us, it would not matter how many hundreds or thousands of people we presented before them that had experienced severe adverse health effects, they will not believe them unless they can prove it and replicate it in a laboratory. It is a bit like the old economists’ joke that they see something in practice but do not believe it will work in theory.

Prof. Hansen: We are hoping to get the NHMRC to fund the project and actually do that through sleep studies. One thing that worries me a bit about that is that some problems develop over a period of time in people. Some people can put up with it for a night and not be bothered, but if it goes on for week after week they start to become sensitised to it and then their reaction becomes even worse.

Senator DAY: People living near wind farms year in and year out exhibit these symptoms of all types. Obviously there are not just connections with the ears but with the heart and other organs and so on. How one attempts to reproduce, replicate or test that on somebody in a laboratory or in some other kind of controlled experiment is beyond me.

Prof. Hansen: We could take some of these people who have been sensitised and put them in a laboratory, and test other people who have not been sensitised and put them in a laboratory and see the difference in response.

Senator DAY: One of the great principles of our jurisprudence and justice system is the fact that a person can testify, ‘This is what I experience.’ It is considered in Supreme Courts and in judicial forums around the world that when a person giving personal testimony says, ‘This is what I saw, this is what I felt, this is what I experienced,’ it is normally given credibility and I have been quite surprised. It is a bit like the blind man story in the Bible who was cured and the Pharisees would not accept it until he could explain what happened. He said, ‘Look, all I know is this. One thing I know is whereas once I was blind now I can see. Don’t ask me how it happened. I just know.’ When we hear testimony from people who say, ‘We were healthy, we welcomed these wind farms, and now we are all sick.’ I am interested in getting some of your ‘colleagues’ to try to see this and what your view is of how your colleagues have responded to this.

Prof. Hansen: I think they have the problem of, ‘If I don’t experience it when I go near a wind farm—and I have been measuring lots of wind farms—then it can’t be a problem’.

Senator DAY: Exactly, yes.

Prof. Hansen: They do not seem to be able to accept that everyone is different and that what is not a problem for some people is a problem for others and that, maybe, if you live there it gets worse.

Senator DAY: They would not live there. If you said, ‘Well, if it’s not a problem, go and live there yourself.’ Thank you.

Senator LEYONHJELM: You are with a project under NHMRC work?

Prof. Hansen: Yes, I am part of a team of nine people on a current NHMRC. Peter Catcheside is the leader. He is a sleep studies expert. I am just providing some acoustics expertise in terms of generating and reproducing the wind farm noise in the lab.

Senator LEYONHJELM: This was an issue that we wrestled with with the acoustics consultants this morning. You are confident that reproduction of the acoustic effect of a wind farm can occur in a room, are you?

Prof. Hansen: Yes.

Senator LEYONHJELM: You are quite sure of that.

Prof. Hansen: It is going to be a challenge. Put it this way, there are sound sources that are capable of it, but whether we can do it in a room without too much leaking out is yet to be seen. I am pretty sure we can otherwise I would not have applied.

Senator LEYONHJELM: What about the claim that a single turbine is not the issue, in fact multiple turbines are not necessarily the issue, it is only when they are rotating in synchrony that there is an issue and you can get hot spots.

Prof. Hansen: You can. The hot spots will depend on which direction the wind is blowing as well. There will be times when it will be a hot spot and it will be loud for a while and then it will die off for a while, because you have very low-frequency, in-phase noise.

Senator LEYONHJELM: Two questions arise out of that. How do you replicate a hot spot in a room with recorded sound?

Prof. Hansen: You just record the sound at the hot spot and then play that in the room.

Senator BACK: It was put to us this morning by, I think, Mr Turnbull from the acoustical consultants—and I wrote it down—who said, ‘The level of infrasound from a wind farm is below the level of human perception.’ Do you agree with that statement?

Prof. Hansen: No, because the infrasound he has been measuring is random infrasound from a number of different sources. He produced a report showing infrasound from a number of different sources and then showing infrasound from wind farms. His wind farm infrasound seems to be within that range he measured.

The problem is that wind farm infrasound is periodically varying, regularly varying. It seems to be more perceptible to some people than random infrasound. That is why I do not necessarily agree that just because it is at a low level it is not detectible.

We have measured infrasound levels in people’s houses when the wind farms are going. They are much higher than the levels he presented. He presented levels outside of someone’s house. We have measured inside and outside, and we found in both cases at Waterloo that levels have been higher than what he showed.

Senator BACK: The work of Cooper—and I regard Cooper’s work as having been only a pilot, nothing more than a pilot—had cooperation from Pacific Hydro, with terms dictated by Pacific Hydro: three dwellings, six people. Do you think there is enough in Cooper’s pilot conclusions to suggest that the work should be extended further, including other wind farm operators and including other physical locations, so that we get to a level of statistical significance whose absence Mr Turnbull and others are so critical of Cooper for?

Prof. Hansen: Yes, I think it would be very helpful to extend it to other wind farms, but we also need to be cognisant of the fact that at the same time as infrasound is affecting people there is also maybe audible low-frequency sound. So, we still do not know whether it is the infrasound causing the problem or the audible low-frequency sound causing annoyance and causing lack of sleep and then causing the stress and health problems. I think the more data we have the better off we are going to be, and we would need to try to separate out those two effects where we can.

Senator BACK: That certainly is a view that I have formed on this committee. I do not know about others. You mentioned that a question was asked of you about the NHMRC review of literature. It was put to us yesterday that on the review panel there was one audiologist, an acoustician and three epidemiologists. What confidence can we have, because some of us have been associated with the move to make sure there are some funds to properly independently do this work, that the party or consortium selected to do this work is going to be independent?

Prof. Hansen: I think the NHMRC has very good procedures to make sure things are independent. They get independent reviews from people, a range of reviews. And I am sure that the committee that is deciding on who gets the funding is a medically based committee; it is not going to be anyone in my discipline. So, I do not see how it is going to be prejudiced from the acoustics point of view.

Senator LEYONHJELM: Medically based, did you say?

Prof. Hansen: Yes, medically based.

Senator BACK: So people with the capacity. I am encouraged by that, because I have spoken more than once in the Senate about my concerns about the bias of people who have been involved on review panels. And I do not want to go through this exercise, be it $2½ million or more—and I do not care where the bias is, whether it is one side or the other—and then see a situation in which there is a credible argument at the end of the day that the group that gets to do the work was biased one way or the other.

Again, it was put to us this morning that there has been no new credible scientific evidence in the last few years so as to even bother looking at the possible adverse health impacts of infrasound—certainly in the audible sound, but not infrasound. Do you agree with that summary?

Prof. Hansen: I find it puzzling, that summary, because there is no evidence, according to their strict guidelines on what constitutes good evidence, but there is no evidence the other way either. So I do not know how anyone can come to that conclusion that it is not worth doing any more study if you say that the evidence that is there is not good enough to point to a problem—

Senator BACK: ‘We haven’t found it, so stop looking!’

Prof. Hansen: Yes. I do not understand that.

Senator BACK: I think Senator Day was getting at the same point a little bit earlier. Thank you, professor Hansen; it has been very interesting to listen to you.

Proceedings suspended from 1 5:20 to 16 : 02