Palmer, W. Wind Turbines – A Changed Environment

6th International Meeting on Wind Turbine Noise Glasgow 20-23 April 2015

Wind Turbines – A Changed Environment

William K.G. Palmer TRI-LEA-EM, Canada.



This paper gives examples of the sound from wind turbines in the outdoor environment, and in the indoor environment. These are compared to other sounds occurring in the environment, such as road traffic or overhead aircraft, and to the sounds produced in a typical municipal library and by a typical refrigerator. In summary, the paper shows that wind turbines do alter the acoustic environment, both outside homes and inside homes presenting a greater difference at low frequencies than other sound sources normally met.

1. Introduction

Classical problem solving for systems suggests that when a working system experiences a failure, look for changes in its environment. As an example, if an engine that has worked well for some time suddenly experiences distress, look for what has changed. Was the oil change schedule altered? Has a bearing reached end of life? This paper applies a similar approach to look for what has changed when distress occurs in the human system.

When wind turbines are installed in the environment of humans, a common finding based on face-to-face interviews conducted by the author, with many people, is that some report discomfort, at varying degrees, ranging from mild annoyance, to severe adverse health impacts. The healthy human system experiences distress. Interviews reveal when people leave the wind turbine environment, their distress diminishes, but when they return, so does the impact. What change is causing the distress? Rather than trying to discount the discomfort, this paper looks for changes in the environment wind turbines create based on research into the sound and it’s special characteristics as received where the humans live.

The sound from wind turbines is compared to other sounds in the environment to examine the differences based on analyzing recordings of actual sound monitoring. We will look at sound from wind turbines, vehicular traffic, aircraft, wind, and people, to identify differences. We will look too at the way the sound is monitored, to see if that can have an impact. Finally, through examination of the special characteristics of the sound generated by wind turbines that are different from the sound from other sources, a reason for the discomfort people experience is offered. The links between the changes in the environment wind turbines create and the human condition is explored.

2. Background

2.1 What do we mean by “the sound” from a wind turbine? Questions of how much sound wind turbines emit, and how that sound compares to other sound sources has been around for a long time. Almost everyone who has been following the information cloud surrounding wind turbines has heard the common expressions:

  • The sound from wind turbines at your home is less than from your refrigerator or air conditioner.
  • The sound from a wind turbine is comparable to a quiet library.
  • The background noise of the wind “masks” the sounds emitted by wind turbines.
  • The sound level produced by typical wind farms is so low that it would not be noticeable in most residential areas.
  • So, what does it mean when we speak of “the sound” from wind turbines? It is often represented by a single value, representing the amplitude in the range our hearing is most sensitive, the A-weighted value. But, should we not also consider the “quality” of the sound and it’s special characteristics? The nature of human hearing is that we respond to a very wide range of sound inputs, and often it is differences in sound, and differences in the characteristics of the sound that gives them a recognizable signature. We can hear a whisper of a companion or a whimper from a restless baby. We can recognize the voice of someone we know in a crowded room. Most of us can whistle a familiar tune that sticks in our head. Our hearing responds better to differences than just to volume, and our mind responds to specific tones, or repetitive patterns more than to a random sound. Sound is far more than “volume.”

2.2. What have we learned already? From previous work, we have learned that:

  • At distances of more than 500 metres to a kilometer, the sound from wind turbines are rich in low frequency sound (sound less than 200 Hz) and infrasound (sound less than 20 Hz), while the higher frequencies are attenuated to be comparable to background,
  • low frequency sound travels longer distances than high frequency sound
  • low frequency easily passes through most building materials, even while higher frequency sounds are attenuated
  • WHO states low frequency sound warrants special consideration
  • the special characteristics of the sound from wind turbines makes them recognizable even when the volume is low

Further adding to the confusion is the fact that most regulators base sound level limits on A-weighted values, often found by considering only the octaves centred from 63 Hz to 8000 Hz. All this means sound levels at different octaves are adjusted as follows:

  • sound from octaves at frequencies below 63 Hz is ignored
  • 63 Hz – measured sound reduced by 26.2 dB
  • 125 Hz – measured sound reduced by 16.1 dB
  • 250 Hz – measured sound reduced by 8.6 dB
  • 500 Hz – measured sound reduced by 3.2 dB
  • 1000 Hz – measured sound considered as is
  • 2000 Hz – measured sound augmented by 1.2 dB
  • 4000 Hz – measured sound augmented by 1.0 dB
  • 8000 Hz – measured sound reduced by 1.1 dB
  • sound from octaves at frequencies above 8000 Hz is ignored

As a result low frequency and infrasound are reduced or ignored by most regulators, as are higher frequencies, on the assumption that those sounds are not considered part of the normal hearing range as used in spoken communication.

2.3 What will we not do?

What this document will resist doing is to identify a single value of the sound intensity for any of the sources. Generating a single value by somehow adding together the octave contribution across the spectrum of sound produced by a source neglects the impact of the special characteristics of the sound. If a sound is cyclical (displaying a repeating pattern) or tonal (with a discernable pitch at one or more frequencies) it is more recognizable than a sound that is constant, and evenly distributed across all frequencies. Generally regulators recognize that if a sound has special characteristics of recognizable tonality, a cyclical nature, or impulsiveness (like a hammer blow or a gunshot) then the sound is penalized, yet, some regulators specifically do not consider sound from wind turbines which have a recognizable repetitive “swoosh” which modulates the sound at all frequencies as being cyclical. Yet, it is the cyclical, repetitive nature of sound from wind turbines that seems to make them most recognizable. To those who hoped this document would produce a simple answer to the question about how the sound from a wind turbine compares to other sources, an apology must be given in advance. Instead, this document proceeds at a somewhat “pedestrian” rate, trying to give the reader a better understanding of why there is no simple answer, and why a response must be conditioned with “it depends.”


4.0 Concluding Observations

The charts in this document show the sound from wind turbines is indeed rich in low frequency, exceeding the low frequency contribution received from the wind in the environment, of a helicopter flying directly overhead, of refrigerators, or libraries, and of most highway traffic. The charts show that the sound from wind turbines shows tonal characteristics. The charts also show that inside homes, room conditions cause a greater variation across a room than in the outdoor environment, and result in intensity increases at room mode frequencies, a function of the room size, and exciting source.

What the charts cannot show is the duration of the wind turbine sound, which can continue for hours at a time, particularly at night, when meteorological conditions favour higher wind turbine output, and be significantly greater than other sound sources in the environment. Neither do the charts properly identify the cyclical nature of the sound, rising and falling repeatedly, which makes them particularly recognizable. These durations and repetitive patterns (amplitude modulation) are apparent from the audio recordings that form the basis for this document though, which were made in a rigorous manner.

It is suggested that the information displayed in these charts provides a firm argument that use of A-weighting and of considering only octaves from 63 Hz to 8000 Hz does not provide an adequate regulatory environment for wind turbines.

Some regulators recognize that special audible characteristics of wind turbines should be addressed. New Zealand Standard NZS6808-2010 states, “5.4.2. Wind turbine sound levels with special audible characteristics (such as, tonality, impulsiveness, and amplitude modulation) shall be adjusted by arithmetically adding up to +6dB to the measured noise at a noise sensitive location. This adjustment is a penalty to account for the adverse subjective response likely to be aroused by sounds containing such characteristics.”  

While the issues are identified, the regulations do not necessarily deal effectively with the issue. For example New Zealand Standard NTZ-6808-2010 states, “C5.5.2 The World Health Organization recognizes that adverse noise effects can be increased by sound with a large proportion of low frequency components.” However, all stakeholders, including the wind turbine industry, influence development of regulations.

The Standard goes on to diminish the issue. “Measurements show that wind turbine sound does not contain a large proportion of low frequency components. As sound propagates from a wind farm (or any other source) the higher frequency components are attenuated quicker than the low frequency components. At a distance from any sound source it is often lower frequency components that are audible, albeit at a low sound level, Wind farm low frequency sound at a noise location which is tonal or has amplitude modulation would be penalized for special audible characteristics.”

This document does give evidence that the wind turbine contribution to the environmental noise at low frequencies particularly is indeed above other sound sources, suggesting that they provide a basis for recognition in the New Zealand Standard.

The Ontario Noise Guidelines for Wind Farms (2008) state, “the information (for acoustic emission of wind turbines) must include the sound power levels, frequency spectra in octave bands (63 to 8000Hz), and tonality at integer wind speeds from 6 to 10 m/s.” The guidelines go on to note, “Should the manufacturer’s data indicate that the wind turbine acoustic emissions are tonal, the acoustic emissions must be adjusted by 5 dB for tonality … otherwise the prediction should assume that the wind turbine noise requires no adjustments for special quality of sound.”

This document shows that measurement confirms that tonal acoustic emissions do occur; even if they are not indicated by the manufacturers data, suggesting that they may develop over time due to blade wear as an example. The Ontario guidelines specifically exclude cyclical sound from wind turbines, noting, “No special adjustments are necessary to address the variation in wind turbine sound level (swishing sound) due to the blade rotation, see Section 4. This temporal characteristic is not dissimilar to other sounds to which no adjustments are applied. It should be noted that the adjustments for special quality of sound described in Publication NPC-104, Reference [1], were not designed to apply to sounds exhibiting such temporal characteristic.”

This document shows that using only the octave bands from 63 to 8000 Hz as well as A-weighting those sounds results in a regulatory environment that is not protective, as both A-weighting, and restricting the octave bands does not address the large proportion of low frequency components that is specifically identified by the World Health Organization as a source of adverse noise effects, which is noted in the New Zealand Standard. The audio recordings made to support this document demonstrate that the cyclical nature of the wind turbines is not similar to other sounds to which no adjustments are applied, it is indeed a “signature” quite specific to wind turbines, and the cyclical penalty from Publication NPC-104 should apply. The exclusion of the variation in sound from wind turbines should be revisited. This document demonstrates that a revision to the regulatory environment for wind turbines is justified. The basis for neglecting the low frequency components and the cyclical (or amplitude modulation) nature of the sound by regulators that they are minimal has been proven to be faulty. The document also provides a part of the information called for by the Conclusion on Page 10 or the report of the Ontario Chief Medical Officer of Health, which states, “The review also identified that sound measurements at residential areas around wind turbines and comparisons with sound levels around other rural and urban areas, to assess actual ambient noise levels prevalent in Ontario, is a key data gap that could be addressed.”

This fact that thousands of complaints have been filed with the Ontario Ministry of the Environment regarding noise from wind turbines and adverse impacts on scores of citizens have been reported is confirmation of the result of regulations that are not protective. This document is provided in good faith, as a demonstration of the evidence that shows there are reasons of changed conditions brought about by wind turbines that are not addressed by current regulations, which call for their review.

Download the complete paper →