 | Noise, shadows and flicker
This page is taken from the book Available on-line at www.stacey-international.co.uk |
The sound of a wind turbine generating electricity is likely to be about the
same level as noise from a flowing stream about 50-100 metres away or
the noise of leaves rustling in a gentle breeze. (BWEA website)
E.ON has today announced that it no longer intends to continue to develop
an eight turbine wind farm near Ferndale because of concerns that the
project’s original design could potentially pose a noise nuisance to nearby
homes. (Press release, 2 July, 2008)
Not all wind farms cause a problem of noise. Many earlier ones were remote from homes and their sound irritated only a few walkers and other users of the countryside, though even then there were complaints that quite distant machines made sufficient noise to disrupt sleep and cause annoyance during daytime. The situation is changing, however. As the developers have grabbed (or been denied) the remote lands of Britain, so their flailing blades perforce creep closer to habitations. The E.ON development, Ty’n Tyle wind farm, cited in the chapter header, is about a kilometre from each of the South Wales’ valley communities of Ferndale and Ystrad. Many current developments will be at this sort of distance from homes, but denial is endemic: Gordon James, director of Friends of the Earth Cymru, dismissively responded to E.ON’s release, “Noise isn’t a problem . . . Modern wind turbines are very quiet.”
Sound attenuates with distance from source by an approximate inverse square law so, if distance is halved, perceived noise increases by about four times. The remarkable push for renewable energy which has been imposed by the regional planning advice notes will allow many more turbines to be built close to habitation and there will thus be a growing impact of noise on human health, happiness and prosperity. This will all be done under the umbrella guidelines of the cryptically named ETSU-R-971(1) which was prepared by The Working Group on Noise from Wind Turbines comprising developers, noise consultants, environmental health officers and others set up by the DTI (now DBERR).
There are two potential sources of noise: that from turbine blades passing through the air at the speed of a light aircraft, and from the gearbox and generator in the nacelle. According to the industry, blade design can reduce the first problem of aerodynamic sound, whilst gear design, sound insulation and isolation suppresses mechanical noise. This is to an extent a clever exercise in concealment.
It is true that engineering can suppress mechanical noise but an aerofoil blade, the size of a Jumbo’s wing, travelling at 150 mph and harvesting 0.6 MW of power or more, inevitably makes substantial sound! The air passing through the rotor is swept into turbulent wake-vortices, the source of much of the sound, and within a few feet encounters the obstruction of the tower. As a blade passes a tower every one to two seconds this imposes a pulsating quality to the aerodynamic sound which many people find deeply disturbing. Other periodic sounds arise as the blades sweep down into the region of wind shear so that the lowest blade position experiences not only different wind speed but also varying turbulence. It is a deliberate untruth that “Noise isn’t a problem” and as we shall see ETSU-R-97 is not a fit instrument to assess it.
In the case of wind farm clusters of turbines there are further possibilities of interaction of sound periodicity. As the rotors of different machines come into and go out of phase, they can create periodic “beat” sounds (aerodynamic or amplitude modulation) allowing the rhythmic “whoomph, whoomph” at one to two second intervals to rise and fall in loudness, an effect which so disturbs some people. This sound is of low but audible frequency – comparable to the base – “woofer” speaker output of a sound system.
In addition to normally audible sound, any machinery will generate a degree of low frequency sound (effectively mechanical vibration) which ranges from just audible “sub-woofer” frequencies (below 200 Hz) down to wavelengths which cannot be heard but are often sensed as bodily discomfort (below 20 Hz) and are often referred to as infrasound which is particularly difficult to measure instrumentally. The industry and several independent reports claim that it is not a problem but this is controversial.
The measurement of noise
Noise is measured in decibels (dB). The decibel is a measure of the sound pressure level, i.e., the magnitude of the pressure variations in the air, expressed as a ratio to a reference pressure. The scale is logarithmic so an increase of 3 dB is a doubling of sound pressure. Measurements of environmental noise are usually made in dB(A) which includes a correction for the frequencies (different pitches), best-heard by the human ear. Unfortunately the A-weighting tends to devalue the low frequency end of the spectrum and has been criticised where low frequencies are important – repetitive base notes in music and of course the pervasive aerodynamic “whoomph” of wind turbines. The C- weighting curve is more satisfactory as it is less selective against low frequency but ETSU-R-97 mandates dB(A) as do many other sound measuring conventions.
The noise a wind turbine creates can be expressed in terms of its sound power level at source. This is a measure of the noise emitted by the machine and is also expressed in dB(A). BWEA claims that a single wind turbine usually emits between 90 and 100 dB(A) and creates a sound pressure level of 50-60 dB(A) at a distance of 40 metres from the turbine(2) and that:
| Ten such wind turbines, all at a distance of 500 metres would create a noise level of 35-45 dB(A) under the same conditions. With the wind blowing in the opposite direction the noise level would be about 10 dB lower | ||
| To put this in perspective some comparable noise sound pressure levels in dB(A) are: | ||
| Rural background | 20-40 | |
| Bedroom at night | 25 | |
| Quiet home interior | 35-40 | |
| Wind farm at 500 m | 35-45 | |
| Car at 40mph at 100 m | 55 | |
| Vestas V80 2 MW wind turbine close-up (wind 10 m/s) | 98-99 | |
| Jet aircraft take-off at 100 m | 125 | |
Use of the dB scale tends to confuse the lay person and this has been deliberately exploited in many of the wind industry’s planning applications. In recent years several of these have been ruled inadequate. It is useful to know that in the open air, a change of 3 dB is barely discernable but a 5 dB change will cause most people to comment and a 10 dB increase, perceived as an approximate doubling of noise, will result in complaints from most people.
The use of the dB(A) frequency scale, biased for human hearing, also implies that perception of sounds as unpleasant, neutral or pleasing, simply relates to loudness. This is not so, and one has to ask how a single noise-level reading relates to the range of subjective experiences described below which include periodic sounds, their variation in pitch (frequency) and vibrations close to the lowest audible frequency.
It has been noted that music and noise from discos and the like have a totally different sound character to either steady or sporadic sounds and an A- weighted level has been found to be “inappropriate to assessing the intrusion inside a dwelling from low frequency thumping bass” (noted in an acoustician’s response to a proposed wind power station at Bald Hills, Victoria, Australia).
Throughout the UK wind farm noise is assessed in planning applications using the prescribed methodology of ETSU-R-97, which, supposedly, can be used to define the noise from wind turbines and thus protect wind farm neighbours.
The first stage is to measure prevailing background noise levels during day and night time periods at a sample of representative properties – those where noise is expected to be a problem.
The second stage is to use those measurements to generate maximum permissible day-and night-time noise levels which are set at a prescribed margin above background level – normally 5 dB(A) (or, in low noise environments, at recommended fixed levels). The margin is prescribed in ETSU-R-97 and the required levels that emerge from this stage of the process are entirely dependent upon the results of the background noise measurements.
The third stage is to predict the likely noise emissions from the turbines at each of the representative properties – using manufacturers’ output specifications against local anemometric data, topography etc. This is supposed to provide assurance that the turbines will be capable of operating within the pre-established noise limits. They are thus produced solely for comparison with the background noise measurements. The comparison does help ensure that there will be adequate separation distances from places of habitation.
If the work is not conducted critically, problems arise. These include, not only the appropriateness of selected “representative” locations but also replicability. Measurements taken by the same person, using the same protocol, but on different dates, would very likely vary from each other. This is not a precise science.
The ease with which poor sound measurement can result in an injustice is
illustrated by the Judicial Review of an Inquiry’s findings concerning the Den
Brook Valley Wind Farm in Devonshire. Renewable Energy Systems’ (RES)
appeal against refusal of permission for the wind farm was conceded on noise
grounds alone but a Judicial Review in 2008 led the Secretary of State to quash
the unlawful planning permission on another ground, namely that the planning
inspector, as part of the overall planning balance must consider the likely
electricity contribution of the particular site. The case set a precedent that wind
farm neighbours have the right to raw environmental data and it is now beyond
doubt that RES’ noise assessment contained errors
www.denbrookvalley.co.uk
The appropriateness of ETSU-R-97 may be further judged from the comments by acoustic consultant Bowdler(3):
| The conclusions of ETSU-R-97 are so badly argued as to be laughable in parts (the daytime standard is based on the principle that it does not matter if people cannot get to sleep on their patio so long as they can get to sleep in their bedrooms). It is the only standard where the permissible night time level is higher than the permissible day time level . . . | |
| Bowdler further comments on Paragraph 1 of the Executive Summary of ETSU- R-97 that it clearly says: | |
| This document describes a framework for the measurement of wind farm noise and gives indicative noise levels thought to offer a reasonable degree of protection to wind farm neighbours, without placing unreasonable restrictions on wind farm development or adding unduly to the costs and administrative burdens on wind farm developers or local authorities. | |
Government is effectively saying “the debate is over” – an undemocratic and anti-science stance we have grown used to in other quarters of the climate change discussion. But the debate on noise does continue in technical publication, for example in van den Berg’s recent investigation of Dutch wind farm of variable speed turbines (10 to 22 rpm). Complaints about noise, especially at night, extended to 1.9 km whereas the developer had claimed there would be no problem over 0.5 km(4).
Van den Berg showed that the night time wind speed at hub height is up to 2.6 times higher than expected from the conventional extrapolation from wind speed measured at 10 m height. The higher wind speed causes faster rotation and up to 15 dB higher sound levels, relative to the same 10 m reference wind speed in daytime. His paper also concluded that day or night, the background noise did not effectively mask the thumping sound of the blades passing the tower and interestingly observes that the thumping is not perceptible close to the turbines – only at a distance.
That discussion still continues is illustrated by a very recent publication from the independent consultants who sat on the 2006-7 DTI and BERR Noise Working Group(5). Their paper attempts to arrive at a procedure for coping with the 10 m wind speed problem and site-specific wind shear (as described by van den Berg). In concurrence with van den Berg, this paper also concluded that the mismatch between 10 m and hub height wind measurements can cause significant errors in noise calculation. The Noise Working Group also revisited the matter of infra-sound and repeated the views which have come from others, that infrasound from wind turbines (less than 20 Hz) is not sufficiently energetic to be sensed by the human body. This is discussed further below.
Possibly the most publicised case of a wind turbine noise problem in Britain is that of the Davies family of Spalding in Lincolnshire.(6) When the construction of Deeping St Nicholas wind farm was proposed, just 930 m from their farmhouse, Julian and Jane Davis initially had no objection. However, after the eight 2.0 MW turbines became operational in summer 2007 the Davis’s discovered that pervasive noise was intolerable.
| By May 2007 we were forced to abandon our home as a place in which to sleep and live. | |
| The problem has been recognised as rendering the house valueless: | |
| I am not able to place a current market value on the property as I do not believe any prospective purchaser would want to inhabit the property, or, indeed in the current climate, whether any mortgage lender would be prepared to lend . . . (Munton & Russel, Estate Agents Spalding, April 2008) | |
The noise also triggered a Council Tax reduction by the Lincolnshire Valuation Tribunal which reduced the Grays Farm and adjacent Farmhouse from Band B to Band A with effect from June 2006 on the grounds of “Noise pollution externally and internal low frequency noise pollution from new wind farm 930m.” Parliament has now been told that “proximity of a electricity generating wind turbine” can be the reason for a discretionary discount on Council Tax (Hansard, 13 May, 2008: Column 1442W ff). So much for “Noise isn’t a problem . . .” – and when did silent “elegant and beautiful structures” last qualify property for a rates reduction?
I have spent some time on the Davis’ case but there are many more examples. A victim of a wind farm in New Zealand wrote “… worst of all is the beat. An insidious, low-frequency vibration that’s more a sensation than a noise. It defeats double-glazing and ear plugs, coming up through the ground, or through the floors of houses, and manifesting itself as a ripple up the spine, a thump on the chest or a throbbing in the ears. Those who feel it say it’s particularly bad at night. It wakes them up or stops them getting to sleep.” (Hawkes Bay Today [NZ], 18 February, 2006).
Many of these responses appear to be triggered by what has come to be known as aerodynamic modulation (AM) and the Davis’s believe their experience at Deeping St Nicholas has that cause:
| We now know that we suffer from aerodynamic or amplitude modulation created by the noise from the wind turbine array. [which] is not fully understood by scientists. This means that no developer can categorically state that there will not be a noise problem.(6) |
The Hayes-McKenzie report of 2006, commissioned by the DTI to investigate low frequency effects, noted that the main cause of complaints was not low frequency sound but Aerodynamic Modulation(8). This can loosely be defined as audible modulation of aerodynamic noise, i.e. aerodynamic noise which displays a greater degree of fluctuation than usual and occasionally occurring in ways not anticipated by ETSU-R-97. The causes are not well understood but include the interaction of sound from more than one turbine and consequent reinforcement of sound pulses from the passing frequency of blades with both the tower and the wind shear zone lower in the swept circle of the rotor.
The Hayes-McKenzie report also stated that concerns about this phenomenon have been expressed in relation to only five out of 126 wind farms in the UK. Despite this, the DTI commissioned a further study of AM from the University of Salford which reported in 2007 that the incidence was low but even so, concluded:
| On the other hand, since AM cannot be fully predicted at present, and its causes are not understood we consider that it might be prudent to carry out further research to improve understanding in this area(8) | |
| The official response to this advice whilst the Davis’s languished in rented accommodation was: | |
| Government does not consider there to be a compelling case for further work into AM and will not carry out any further research at this time | |
A recent report from the Noise Association(9) (NA) certainly recognises a generalised noise problem for some people, living less than 1 to 1.5 miles from a turbine, and that not all individuals are equally affected.
| For people who cannot shut out the noise, the problem can be exacerbated by the rotating blades and the dancing shadows of turbines so that noise from turbines can be much more intrusive that other noises of a similar decibel level. |
The report also notes that, for those who are sensitive, the impact of turbines can be overwhelming and a particular problem in quiet rural areas. Low-frequencies may form an audible but not major part of turbine noise and can create additional problems for some but the infrasound content of wind turbine noise is too low to affect most people. However, low frequency may be underestimated because of the persistent use of “A” weighting in measuring the noise, rather than taking “C” weighted measurements
Some medical reports include persistent complaints from people saying they not only hear the noise from wind turbines, but can “feel” disturbance in their bodies, leading to symptoms similar to those associated with vibro-acoustic disease (VAD). This latter observation is in slight conflict with the NA’s previous suggestion that infrasound content is too low to have clinical effect and it is worth looking further at the industry’s and Government’s current stance on low frequency noise.
Sounds below the frequency range for human hearing have been the subject of ongoing controversy in the context of wind power. The BWEA has consequently felt need to provide a web-page on the subject www.bwea.com/ref/lowfrequencynoise.html which reports:
| Dr Geoff Leventhall, Consultant in Noise Vibration and Acoustics and author of the Defra Report on Low Frequency Noise and its Effects, says: ‘I can state quite categorically that there is no significant infrasound from current designs of wind turbines.’ | |
| However, a report from Keele University on infrasound(10) says: | |
| We have clearly shown that both fixed speed and variable speed turbines generate low frequency vibrations which are multiples of blade passing frequencies and can be detected by seismometers buried in the ground | |
Detection was possible at distances up to many kilometres and in the presence of background seismic noise. In the absence of peer-reviewed medical evidence concerning low frequency sound from wind turbines, these two statements make uncomfortable bedfellows and so, as with many other aspects of this industry, we have a “Catch 22” in which proof of a problem can only come when it is too late.
However it is significant that the few medical workers looking at low-frequency noise from wind turbines on three continents are in agreement to the extent of christening the health consequence “Wind Turbine Syndrome”(11) and now in a forthcoming book of eponymous title. The syndrome includes sleep disturbance, headache, dizziness, nausea, rapid heart rate, panic attacks and significantly, if the families Pierpont studied moved away from the turbines (sometimes abandoning their homes), the symptoms, significantly, went away.
The BWEA followed-up the publication of the Keele University report with a rebuttal of any suggestion that infrasound was a health issue. In this rebuttal two of the original authors, Styles and Toon, wrote:
| To put the level of vibration into context, they are ground vibrations with amplitudes of about one millionth of a millimetre. There is no possibility of humans sensing the vibration and absolutely no risk to human health. |
However, a more recent development has been the publication in 2008 of a study which has shown
| for the first time that the human vestibular system is also extremely sensitive to low-frequency and infrasound vibrations by making use of a new technique for measuring vestibular activation.(12) |
Perturbation of the vestibular apparatus is a core response underlying Pierpont’s Wind Turbine Syndrome. This demonstration of extreme sensitivity to low frequency vibration suggests that Styles and Toon’s dismissive “no possibility of humans sensing the vibration” may not be correct and that government’s refusal to commission further work is, at the least, premature. Coupled with recent findings by Alves-Pereira & Branco that “In-Home Wind Turbine Noise Is Conducive to Vibroacoustic Disease”(13) this appears to support the former Dean of Medicine at the University of Western Ontario who is calling for health studies into the wind turbine farms being built in the State and suggests that if there is enough evidence, a formal epidemiological study should be made.(14) Alves-Pereira & Branco’s conclusion was that infrasound and low frequency noise generated by WT blades can lead to severe health problems, specifically, VAD, and efficient zoning for WT must be scientifically determined, and quickly adopted, in order that Public Health may be properly protected.
Perception
In no part of the confrontation between the wind power industry and people have there been more attempts at misrepresentation than in relation to noise and visual intrusion. The following quotations from Pedersen & Waye’s (2005) paper to the 1st International Meeting on Wind Turbine Noise sums up the subjective feelings of countless people, that exposure to wind turbine noise, shadows and the rotating movement of the rotor blades, were an intrusion into the “private domain”.
| The wind turbine noise was by some of the informants perceived as intruding into private domain, physically into the garden and the home, but also as intruder into themselves. | |
| The experience of lacking control, being subjected to injustice, lacking influence, and/or not being believed. | |
| The noise . . . was to those who could not mentally shut it out, an obstacle to pleasant experiences decreasing the joy of daily life at home . . . creating a feeling of violation that was expressed as anger, uneasiness, and tiredness. |
That such feelings are not amenable to interpretation by noise metering is the crux of the problem. A dripping tap making a sound near the lower threshold of hearing can be more infuriating than the continuous hum of traffic on a nearby road.
| As the science of acoustics has developed, sociological surveys have become an important aspect in developing noise criteria. These surveys, combined with accurate measurements of the noise, enable a reliable assessment of the percentage of people likely to be annoyed to be developed.(15) |
The setting of levels in ETSU-R-97 in no way made this sort of approach. It does appear to be true that a significant proportion of people are much more seriously affected by noise than others. In our twenty-first century society a similar proportion of people also suffer from a range of disabilities which reduce the quality of their lives and Government has been wise in making it a legal requirement that such unfortunates should not be prevented from leading a normal life. However, the same government has engendered a subsidy system without which wind turbines could not be built. The consequence is that a sensitive minority may be tormented by the legal, but in my view quite unreasonable, activity of wind power developers.
Unfortunately, despite 20 years of complaint about noise, most of the evidence is still dismissed by government as apocryphal and it will remain true that there is little clinical evidence until proper independent research is financed. The repeated reference to compliance with ETSU-R-97 which appears in countless government statements and planning documents is little more than an escape clause. The noise problem of wind turbines, both modulation effects and low frequency sound, is not addressed by the provisions of ETSU-R-97 and the document seems not to be a fit instrument for purpose.
I close this account of wind farm noise with sympathy for the thousands of people, worldwide, who could write, as has the Marton, Askam & Ireleth Windfarm Action Group (MAIWAG):
| The windfarm is noisy, it is a visual blight, it does create shadow flicker, it has resulted in very little benefit to the local economy, it has not resulted in an increase in tourism and negotiating with PowerGen Renewables and Wind Prospect to try to resolve the problems has been a most unpleasant experience for all those involved. Simply put, we want our quality of life back. |
Shadow flicker, reflection and silhouetting
Shadow flicker occurs when the sun passes behind the hub of a wind turbine and shadows of the rotating blades pass repeatedly over neighbouring properties. The seasonal timing and duration of flicker can be calculated from the geometry of the turbine, its orientation relative to nearby houses and the latitude of the site. Quite detailed information and a calculator are given on the Danish Wind Industry Association (DWIA) website.(16) Wind power developers claim to use commercially available software to minimise the risk of shadow flicker affecting homes. According to the DBERR website, flicker “has only been recorded occasionally at one site in the UK. The effect must depend to some extent on regulation of “set-back” – the permitted distance between turbines and homes which is a matter for the local planning authority in the UK.
Investigations of light flicker and photosensitive epilepsy suggest that there is a relatively low risk, as the flicker rate from a typical three-blade turbine is below three per second, i.e., sixty revolutions per minute, but the public should be protected from viewing interacting blades where the shadows cast by one turbine on another cause a cumulative flicker rate exceeding three per second and also to reduce this risk, turbine blades should not be reflective. Flicker rate is not affected by distance so any risk does not decrease significantly until it exceeds about 100 times the hub height – about 10 km for a big turbine.(17)
DBERR’s comment on flicker is more cavalier:
| At a distance of 10 rotor diameters (equivalent to 400 to 800 metres) a person should not perceive a wind turbine to be chopping through sunlight.(18) |
These findings are however supported by Epilepsy Action which reports that it has never received a call from anyone who believes they have had a seizure as a result of a wind turbine.
When turbines are geographically placed so they are visible from homes or valuable view-points, with either the early morning or the low evening sun behind them, they may become very much more prominent by silhouetting, and in these circumstances attempts to reduce their landscape impact such as DWIA’s recommendation of grey painting are valueless. It is also in these circumstances that wind turbines draw the eye and become an irritation in the landscape. The Inspector at a Guestwick (Norfolk) planning Inquiry acknowledged this:
I consider that the dominance of the turbine towers and the visual prominence of the blades, particularly when in motion, would have a much more serious adverse effect than the appellants’ witness claimed. A much earlier planning inquiry into a proposed wind farm at Jordanston, Pembrokeshire (2000) included this:
| The movement of WTGs [wind turbine generators]… has a discordant effect on the eye. The rotation of the blades of WTGs in a cluster, while in the same direction, is not synchronised and gives a constant restless quality to the overall experience of a landscape. Especially when several overlapping WTGs are in view at one time. |
Reflective flashing or enhanced visibility occurs when the sun is behind or slightly to the side of the observer and again accentuates the irritation caused by the moving blades.
To suggest, as does DWIA, that wind turbines are “natural elements in the landscape” or as journalist Polly Toynbee once wrote, “the gentle turning of these silent white wings delights more people than it offends” is beyond belief for any lover of the countryside who has the slightest inkling that wind turbines do virtually nothing for the supposed objective of stabilising or reducing atmospheric CO2. They are money factories which industrialise the landscape for no other significant purpose.
References and Notes
| 1 | ETSU-R-97 (Assessment and Rating of Noise from Wind Farms) (1997). Prepared for Government by The Working Group on Noise from Wind Turbines. Department of Trade and Industry’s Energy Technology Support Unit (ETSU). | |
| 2 | BWEA website www.bwea.com/ref/noise.html | |
| 3 | (a) Bowdler, R. (2005) ETSU-R-97 Why it is Wrong (Whinash Inquiry). (b) Bowdler, R. (2007) Moorsyde Wind Farm Comments on the Noise Section of the Environmental Statement and other Documents. | |
| 4 | Van den Berg, G. P. (2004) Effects of the wind profile at night on wind turbine sound. Journal of Sound and Vibration, 277, 955–970 | |
| 5 | Bowdler, R, Bullmore A et al (2009) Prediction and assessment of wind turbine noise. Acoustics Bulletin, March-April. | |
| 6 | Davis, J & J (2008) evidence to the House of Lords House of Lords Select Committee on Economic Affairs 4th Report of Session 2007–08 The Economics of Renewable Energy Volume II: Evidence. | |
| 7 | Hayes McKenzie (2006) The Measurement of Low Frequency Noise at Three UK Wind Farms www.berr.gov.uk/whatwedo/energy/sources/renewables/explained/wind/onshore-offshore/page31267.html | |
| 8 | University of Salford (2007) Research into Aerodynamic Modulation of Wind Turbine Noise. | |
| 9 | Noise Association (2006). Location, location, location. | |
| 10 | Styles, Simpson, et al (2005) Microseismic and Infrasound Monitoring of Low Frequency Noise and Vibration from Windfarms. Keele University. | |
| 11 | Pierpont, N. (2006). Wind Turbine Syndrome. And a forthcoming book of the same name www.windturbinesyndrome.com/?page_id=932 | |
| 12 | Todd, Rosengren & Colebatch (2008) Tuning and sensitivity of the human vestibular system to low-frequency vibration. Neuroscience Letters 444 36–41. | |
| 13 | Mariana Alves-Pereira & Nuno A. A. Castelo Branco (2007) In-Home Wind Turbine Noise Is Conducive to Vibroacoustic Disease. Second International Meeting on Wind Turbine Noise. Lyon, France. | |
| 14 | The London Free Press (1 February, 2009) Doctor calls for health studies on windmill farms lfpress.ca/newsstand/News/2009/02/01/8228966.html | |
| 15 | Evidence from Graeme E Harding & Associates, retained by the Tarwin Valley Coastal Guardians to advise on proposed wind power station at Bald Hills. | |
| 16 | Danish Wind Industry Association (DWIA) Guided Tour www.windpower.org/ | |
| 17 | Harding et al (2008) Wind turbines, flicker, and photosensitive epilepsy: Characterizing the flashing that may precipitate seizures and optimizing guidelines to prevent them www3.interscience.wiley.com/journal/120084200/abstract?CRETRY=1&SRETRY=0 | |
| 18 | DBERR Flicker www.berr.gov.uk/energy/sources/renewables/planning/onshore-wind/shadow-flicker/page18736.html |