Also see “How-to guide to criteria for siting wind turbines to prevent health risks from sound” by the same authors
Paper presented at Institute of Noise Control Engineering (INCE) NOISE-CON 2008, July 28-31, 2008
- George W. Kamperman, INCE Bd. Cert. Emeritus, INCE Kamperman Associates, Inc., firstname.lastname@example.org
- Richard R. James, E-Coustic Solutions, email@example.com
Rev. 1.0, July 27, 2008
Industrial scale wind turbines are a familiar part of the landscape in Europe, U.K., and other parts of the world. In the U.S., however, similar industrial-scale wind energy developments are just beginning operation. The presence of industrial wind projects will increase dramatically over the next few years given the push by the Federal and state governments to promote renewable energy sources through tax incentives and other forms of economic and political support. States and local governments in the U.S. are promoting what appear to be lenient rules for how industrial wind farms can be located in communities, which are predominantly rural and often very quiet. Studies already completed and currently in progress describe significant health effects associated with living in the vicinity of industrial grade wind turbines. This paper reviews sound studies conducted by consultants for governments, the wind turbine owner, or the local residents for a number of sites with known health or annoyance problems. The purpose is to determine if a set of simple guidelines using dBA and dBC sound levels can serve as the “safe” siting guidelines. Findings of the review and recommendations for sound limits will be presented. A discussion of how the proposed limits would have affected the existing sites where people have demonstrated pathologies apparently related to wind turbine sound will also be presented.
… Our review covered the community noise studies performed in response to complaints, research on health issues related to wind turbine noise, critiques of noise studies performed by consultants working for the wind developer, and research/technical papers on wind turbine sound immissions and related topics. The papers are listed in Tables 1-4.
Table 1 – List of Studies Related to Complaints
- Resource Systems Engineering, Sound Level Study – Ambient & Operations Sound Level Monitoring, Maine Department of Environmental Protection Order No. L-21635-26-A-N, June 2007
- ESS Group, Inc., Draft Environmental Impact Statement For The Dutch Hill Wind Power Project – Town of Cohocton, NY, November 2006
- David M. Hessler, Environmental Sound Survey and Noise Impact Assessment – Noble Wethersfield Wind Park – Towns of Wethersfield and Eagle NY, For: Noble Environmental Power, LLC, January 2007
- George Hessler, “Report Number 101006-1, Noise Assessment Jordanville Wind Power Project,” October 2006
- HGC Engineering, “Environmental Noise Assessment Pubnico Point Wind Farm, Nova Scotia, Natural Resources Canada Contract NRCAN-06-0046,” August 23, 2006
- John I. Walker, “Sound Quality Monitoring, East Point, Prince Edward Island” by Jacques Whitford, Consultants for Prince Edward Island Energy Corporation, May 28, 2007
Table 2 – List of Studies Related to Health
- Nina Pierpont, “Wind Turbine Syndrome – Abstract” from draft article and personal conversations. www.ninapierpont.com
- Nina Pierpont, “Letter from Dr. Pierpont to a resident of Ontario, Canada, re: Wind Turbine Syndrome,” Autumn 2007
- Amanda Harry, “Wind Turbine Noise and Health” (2007)
- Barbara J. Frey and Peter J. Hadden, “Noise Radiation from Wind Turbines Installed Near Homes, Effects on Health” (2007)
- Eja Pedersen, “Human response to wind turbine noise – Perception, annoyance and moderating factors, Occupational and Environmental Medicine,” The Sahlgrenska Academy, Gotenborg 2007
- Robin Phipps, “In the Matter of Moturimu Wind Farm Application, Palmerston North, Australia,” March 2007
- WHO European Centre for Environment and Health, Bonn Office, “Report on the third meeting on night noise guidelines,” April 2005
Table 3 – List of Studies That Review Siting Impact Statements
- Richard H. Bolton, “Evaluation of Environmental Noise Analysis for ‘Jordanville Wind Power Project’,” December 14, 2006 Rev 3
- Clifford P. Schneider, “Accuracy of Model Predictions and the Effects of Atmospheric Stability on Wind Turbine Noise at the Maple Ridge Wind Power Facility,” Lowville, NY – 2007
Table 4 – List of Research and Technical papers Included in Review Process
- Anthony L. Rogers, James F. Manwell, Sally Wright, “Wind Turbine Acoustic Noise,” Renewable Energy Research Laboratory, Dept. of ME and IE, U of Mass, Amherst, amended June 2006
- ISO. 1996. Acoustics – Attenuation of sound during propagation outdoors – Part 2: General method of calculation. International Organization of Standardization. ISO 9613-2. p. 18
- G.P. van den Berg, “The Sounds of High Winds – the effect of atmospheric stability on wind turbine sound and microphone noise,” Ph.D. thesis, 2006
- Fritz van den Berg, “Wind Profiles over Complex Terrain,” Proceedings of Second International Meeting on Wind Turbine Noise, Lyons, France, Sept. 2007
- William K. G. Palmer, “Uncloaking the Nature of Wind Turbines – Using the Science of Meteorology,” Proceedings of Second International Meeting on Wind Turbine Noise, Lyons, France, Sept. 2007
- Soren Vase Legarth, “Auralization and Assessment of Annoyance from Wind Turbines,” Proceedings of Second International Meeting on Wind Turbine Noise, Lyons, France, Sept. 2007
- Julian T. and Jane Davis, “Living with aerodynamic modulation, low frequency vibration and sleep deprivation – how wind turbines inappropriately placed can act collectively and destroy rural quietitude,” Proceedings of Second International Meeting on Wind Turbine Noise, Lyons, France, Sept. 2007
- James D. Barnes, “A Variety of Wind Turbine Noise Regulations in the United States – 2007,” Proceedings of Second International Meeting on Wind Turbine Noise, Lyons, France, Sept. 2007
- M. Schwartz and D. Elliott, “Wind Shear Characteristics at Central Plains Tall Towers,” NREL 2006
- IEC 61400 “Wind turbine generator systems, Part 11: Acoustic noise measurement techniques,”.rev:2002
After reviewing the materials in the tables; we have arrived at our current understanding of wind turbine noise and its impact on the host community and its residents. The review showed that some residents living as far as 3 km (two (2) miles) from a wind farm complain of sleep disturbance from the noise. Many residents living one-tenth this distance (300 m. or 1000 feet) from a wind farm are experiencing major sleep disruption and other serious medical problems from nighttime wind turbine noise. The peculiar acoustic characteristics of wind turbine noise immissions cause the sounds heard at the receiving properties to be more annoying and troublesome than the more familiar noise from traffic and industrial factories. Limits used for these other community noise sources do not appear to be appropriate for siting industrial wind turbines. The residents who are annoyed by wind turbine noise complain of the approximately one (1) second repetitive swoosh-boom-swoosh-boom sound of the turbine blades and “low frequency” noise. It is not apparent to these authors whether the complaints that refer to “low frequency” noise are about the audible low frequency part of the swoosh-boom sound, the one-hertz amplitude modulation of the swoosh-boom sound, or some combination of both acoustic phenomena.
To assist in understanding the issues at hand, the authors developed the “conceptual” graph for industrial wind turbine sound (Figure 1). This graph shows the data from one of the complaint sites plotted against the sound immission spectra for a modern 2.5 MWatt wind turbine; Young’s threshold of perception for the 10% most sensitive population (ISO 0266); and a spectrum obtained for a rural community during a three hour, 20 minute test from 11:45 pm until 3:05 am on a windless June evening in near Ubly, Michigan a quiet rural community located in central Huron County. (Also called: Michigan’s Thumb.) It is worth noting that this rural community demonstrates how quiet a rural community can be when located at a distance from industry, highways, and airport related noise emitters.
During our review we posed a number of questions to ourselves related to what we were learning. The questions (italics) and our answers are:
Do National or International or local community Noise Standards for siting wind turbines near dwellings address the low frequency portion of the wind turbine’s sound immissions? No! State and Local governments are in the process of establishing wind farm noise limits and/or wind turbine setbacks from nearby residents, but the standards incorrectly presume that limits based on dBA levels are sufficient to protect the residents.
Do wind farm developers have noise limit criteria and/or wind turbine setback criteria that apply to nearby residents? Yes! But the Wind Industry recommended residential wind turbine noise levels (typically 50-55 dBA) are too high for the quiet nature of the rural communities and may be unsafe for the nearest residents. An additional concern is that some of the methods for implementing pre-construction computer models may predict sound levels that are too low. These two factors combined can lead to post-construction complaints and health risks.
Are all residents living near wind farms equally affected by wind turbine noise? No, children, people with pre-existing medical conditions, especially sleep disorders, and the elderly are generally the most susceptible. Some people are unaffected while some nearby neighbors develop serious health effects caused by exposure to the same wind turbine noise.
How does wind turbine noise impact nearby residents? Initially, the most common problem is chronic sleep deprivation during nighttime. According to the medical research documents, this may develop into far more serious physical and psychological problems
What are the technical options for reducing wind turbine noise immission at residences? There are only two options: 1) increase the distance between source and receiver; and/or 2) reduce the source sound power immission. Either solution is incompatible with the objective of the wind farm developer to maximize the wind power electrical generation within the land available.
Is wind turbine noise at a residence much more annoying than traffic noise? Yes, researchers have found that “Wind turbine noise was perceived by about 85% of the respondents even when the calculated A-weighted SPL were as low as 35.0-37.5 dB. This could be due to the presence of amplitude modulation in the noise, making it easy to detect and difficult to mask by ambient noise.” [JASA 116(6), December 2004, pgs 3460-3470, “Perception and annoyance due to wind turbine noise-a dose-relationship” Eja Pedersen and Kerstin Persson Waye, Dept of Environmental Medicine, Goteborg University, Sweden]
Why do wind turbine noise immissions of only 35 dBA disturb sleep at night? This issue is now being studied by the medical profession. The affected residents complain of the middle to high frequency swooshing sounds of the rotating turbine blades at a constant repetitive rate of about 1 hertz plus low frequency noise. The amplitude modulation of the swooshing sound changes continuously. The short time interval between the blade’s swooshing sounds described by residents as sometimes having a thump or low frequency banging sound that varies in amplitude up to 10 dBA. This may be a result of phase changes between turbine emissions, turbulence, or an operational mode.. The assumptions about wall and window attenuation being 15 dBA or more may not be sufficiently protective considering the relatively high amplitude of the wind turbine’s low frequency immission spectra.
What are the typical wind farm noise immission criteria or standards? Limits are not consistent and may vary even within a particular country. Example criteria include:
- Australia: the higher of 35 dBA or L90 + 5 dBA
- Denmark: 40 dBA
- France: L90 + 3 (night) and L90 + 5 (day)
- Germany: 40 dBA
- Holland: 40 dBA
- United Kingdom: 40 dBA (day) and 43 dBA (night) or L90 + 5 dBA
- Illinois: 55 dBA (day) and 51 dBA (night)
- Wisconsin: 50 dBA
- Michigan: 55 dBA
Note: Illinois statewide limits are expressed only in nine contiguous octave frequency bands with no mention of A-weighting for the hourly leq limits. Typically, wind turbine noise just meeting the octave band limits would read 5 dB below the energy sum of the nine octave bands after applying A-weighting. So the Illinois limits are approximately 50 dBA (daytime 7 AM to 10 PM) and 46 dBA at night, assuming a wind farm is a Class C Property Line Noise Source.
What is a reasonable wind farm sound immission limit to protect the health of residences? We are proposing an immission limit of 35 dBA or L90A + 5 dBA, whichever is lower, and a C-weighted criteria to address the affected resident’s complaints of wind turbine low frequency noise: For the proposed criteria the dBC sound level at a receiving property shall not exceed L90A + 20 dB. In other words, the dBC operating immission limit shall not be more than 20 dB above the measured dBA (L90A) pre-construction nighttime background sound level. A maximum not-to-exceed limit of 50 dBC is also proposed. … The World Health Organization and others have determined a sound emitter’s noise that results in a difference between the dBC and dBA value greater than 20 dB will be an annoying low frequency issue.
Is not L90A the minimum dBA background noise level? This is correct, but it is very important to establish the statistical average background noise environment outside a potentially affected residence during the quietest (10 pm to 4 am) sleeping hours of the night. This nighttime sleep disturbance has generated the majority of the wind farm noise complaints throughout the world. The basis for a community’s wind turbine sound immission limits would be the minimum 10 minute nighttime L90A plus 5 dB for the time period of 10 pm to 7 am. This would become the Nighttime Immission Limits for the proposed wind farm. This can be accomplished with one or several 10 minute measurements during any night when the atmosphere is classified stable with a light wind from the area of the proposed wind farm. The Daytime Limits (7 am to 7 pm) could be set 10 dB above the minimum nighttime L90A measured noise, but the nighttime criteria will always be the limiting sound levels.
A nearby wind farm meeting these noise immission criteria will be clearly audible to the residents occasionally during nighttime and daytime. Compliance with this noise standard would be determined by repeating the initial nighttime minimum nighttime L90A tests and adding the dBC (LeqC) noise measurement with the turbines on and off. If the nighttime background noise level (turbines off) was found to be slightly higher than the measured background prior to the wind farm installation, then the results with the turbines on must be corrected to determine compliance with the pre-turbine established sound limits. …
Including wind as a masking source in the criteria is one method for elevating the permissible limits. Indeed the background noise level does increase with surface wind speed. When it does occur, it can be argued that the increased wind noise provides some masking of the wind farm turbine noise emission. However, in the middle of the night when the atmosphere is defined as stable (no vertical flow from surface heat radiation) the layers of the lower atmosphere can separate and permit wind velocities at the turbine hubs to be 2 to 2.5 times the wind velocity at the 10m high wind monitor but remain near calm at ground level. The result is the wind turbines can be operating at or close to full capacity while it is very quiet outside the nearby dwellings.
This is the heart of the wind turbine noise problem for residents within 3 km (approx. two miles) of a wind farm. When the turbines are producing the sound from operation it is quietest outside the surrounding homes. The PhD thesis of P.G. van den Berg “The Sounds of High Winds” is very enlightening on this issue. See also the letter by John Harrison in Ontario “On Wind Turbine Guidelines.” …
The simple fact that so many residents complain of low frequency noise from wind turbines is clear evidence that the single A-weighted (dBA) noise descriptor used in most jurisdictions for siting turbines is not adequate. The only other simple audio frequency weighting that is standardized and available on all sound level meters is the C-weighting or dBC. A standard sound level meter set to measure dBA is increasingly less sensitive to low frequency below 500 Hz (one octave above middle-C). The same sound level meter set to measure dBC is equally sensitive to all frequencies above 32 Hz (lowest note on grand piano). It is well known that dBC readings are more predictive of perceptual loudness than dBA readings if low frequency sounds are significant.
We are proposing to use the commonly accepted dBA criteria that is based on the preexisting background sound levels plus a 5 dB allowance for the wind turbine’s immissions (e.g. L90A +5) for the audible sounds from wind turbines. But, to address the lower frequencies that are not considered in A-weighted measurements, we are proposing to add limits based on dBC. The Proposed Sound Limits are presented in the text box at the end of this paper. For the current industrial grade wind turbines in the 1.5 to 3 MWatt range, the addition of the dBC requirement will result in an increased distance between wind turbines and the nearby residents. For the generalized graphs shown in Figure 1, the distances would need to be approximately double the current distance. This will result in setbacks in the range of 1 km or greater for the current generation of wind turbines if they are to be located in rural areas where the L90A background sound levels are 30 dBA or lower. In areas with higher background sound levels, turbines could be located somewhat closer, but still at a distance greater than the 305 m (1000 ft.) or less setbacks commonly seen in U.S. based wind turbine standards set by many states and used for wind turbine developments.
Proposed Wind Turbine Siting Sound Limits
- Audible Sound Limit
- No Wind Turbine or group of turbines shall be located so as to cause an exceedance of the pre-construction/operation background sound levels by more than 5 dBA. The background sound levels shall be the L90A sound descriptor measured during a pre-construction noise study during the quietest time of evening or night. All data recording shall be a series of contiguous ten (10) minute measurements. L90A results are valid when L10A results are no more than 15 dBA above L90A for the same time period. Noise sensitive sites are to be selected based on wind development’s predicted worst-case sound emissions (in LeqA and LeqC) which are to be provided by the developer.
- Test sites are to be located along the property line(s) of the receiving nonparticipating property(s).
- A 5 dB penalty is applied for tones as defined in IEC 61400-11.
- Low Frequency Sound Limit
The LeqC and L90C sound levels from the wind turbine at the receiving property shall not exceed the lower of either:
- LeqC − L90A greater than 20 dB outside any occupied structure, or
- A maximum not-to-exceed sound level of 50 dBC (L90C) from the wind turbines without other ambient sounds for properties located at one mile or more from State Highways or other major roads or 55 dBC (L90C) for properties closer than one mile. These limits shall be assessed using the same nighttime and wind/weather conditions required in 1.a. Turbine operating sound immissions (LeqA and LeqC) shall represent worst case sound immissions for stable nighttime conditions with low winds at ground level and winds sufficient for full operating capacity at the hub.
- General Clause
Not to exceed 35 dBA within 30 m. (approx. 100 feet) of any occupied structure.
- All instruments must meet ANSI or IEC Precision integrating sound level meter performance specifications.
- Procedures must meet ANSI S12.9 and other applicable ANSI standards.
- Measurements must be made when ground level winds are 2m/s (4.5 mph) or less. Wind shear in the evening and night often results in low ground level wind speed and nominal operating wind speeds at wind turbine hub heights.
- IEC 61400-11 procedures are not suitable for enforcement of these requirements except for the presence of tones.
This article is the work of the author(s) indicated. Any opinions expressed in it are not necessarily those of National Wind Watch.