Bad Vibrations — Where’s the Science?

bad vibration

Roger Watt, Waterloo, Ontario, Canada – Updated July 2009


Sorry, Beach Boys, but when it comes to the big industrial-strength wind turbines that are spreading across Ontario’s landscape, there are no “good vibrations”.

In the Source Water Protection program resulting from Ontario’s Clean Water Act of 2006, the Ministry of the Environment mandates local committees to develop science-based plans to protect the sources of drinking water. Requiring a scientific basis for environmental regulations is a superb idea. MOE should have a science-based justification for its regulations on wind turbines, too.

The new regulations that MOE proposed on June 9, 2009, are an improvement in some respects; for example, the 550m minimum setback and the table of setbacks given on page 12 of the Proposed Content for the Renewable Energy Approval Regulation […] document.

However, that document creates an opportunity for proponents to attempt to circumvent those setbacks by conducting their own noise-estimation analyses. How can that be? If the MOE regulations are based on science and if MOE’s computations adhere to worst-case scenarios, and if the proponent has to adhere to that same science and to the same worst-case scenarios, then one must wonder how the proponent could achieve results that differ from MOE’s and be allowed to use lesser setbacks. That would be like hiring the fox to design the chicken coop.

It seems to me that MOE’s “science” is out of date or completely lacking. I believe MOE’s regulations do not protect a sufficient percentage of the people who are subjected to the emissions from wind turbines. And the regulations focus on frequencies to which the human ear is most sensitive, ignoring the impact that vibrations at other frequencies (particularly, low frequencies) have on other parts of the body.

The people of Ontario have a right to regulations based on science to protect them, and MOE has a responsibility to publish that science so that MOE’s computations can be subjected to public scrutiny. The literature abounds with cases demonstrating that annoyance from wind turbines can lead to mental and physical health problems. The June 9 documents fail to provide any scientific references to validate MOE’s claim that its 40 dB(A) limit protects the rural communities that are subjected to it. There is mounting evidence that it does not.

I end this article with an analysis based on two reports from Japan that, I hope, could lead to better, more people-friendly regulations.

Why do I care?

I am a seasonal resident of Ashfield-Colborne-Wawanosh Township in Huron County, with a cottage on the shore of Lake Huron. The closest wind-farm turbines are EPCOR’s (now Capital Power Corporation) Vestas V80 wind turbines are 5km inland. On quiet summer nights, I am sometimes aware of the zinging sound that tires make when cars go over the warning strips approaching the highway intersection 3km away. I am never aware of the wind turbines. But I have met people who live closer, and some of them are not so fortunate. That is what got me interested in this topic. And the more I learn, the more I fear that MOE is one day going to allow wind turbines in Lake Huron to be closer that 10-15km from the shoreline.

Where’s the hype?

Despite the growing body of case reports to the contrary, the wind-industry lobby continues to proclaim that there is no conclusive peer-reviewed scientific evidence that the vibrations from wind turbines have a negative impact on human health.

That is an interesting choice of words. It is the same argument that the tobacco lobby used for decades in the previous century, and that the cellphone industry is still using. To quote Carl Sagan, “absence of evidence is not evidence of absence”. It is also valid to say that there is no conclusive peer-reviewed scientific evidence that wind turbines are safe.

Where’s the problem?




In 2006, I downloaded EPCOR’s 681-page “Environmental Screening Report” (sadly, no longer available online) for the Kingsbridge Wind Power Project, and used its noise-estimation data [pp282-288] to graph the predicted noise at the receptor locations for wind speeds at or below 6 meters/second. I added a best-fit logarithmic curve, plus lines for the MOE 40 dB(A) rural threshold and two others.

  • France has wind-turbine regulations stating that the noise may not be more than 3 dB(A) above the ambient level during the night, and not more than 5 dB(A) above the ambient level during the day.
  • Denmark has a special scale that uses the dB(A) weights, but involves only the low frequencies.

Both of these produce setbacks from residences that are twice the distance that results from MOE’s regulations. And 2x distance reduces the impact by 6 dB.

There are impact-assessment noise studies indicating that rural areas remote from well-traveled roads often have ambient noise levels in the range 25-30 dB(A), or even less, on quiet summer nights. There are scientific reports that the noise from wind-turbine farms under certain atmospheric situations on quiet summer nights can be much greater and carry up to three times farther than predicted by most of the noise-estimating software used by the wind industry today.

On pages 35 and 47 of the 2007 Ontario Municipal Board report on the hearing on the Kincardine/Enbridge wind farm, there is expert testimony that the MOE guidelines “are designed to mitigate noise impacts for about 85% of the population”, but that they “would not protect all people, as individuals perceive noise differently”.

On page 56 of the 2007 Final Report of the Noise Expert on Wind Turbine Facilities Noise Issues commissioned by MOE, the author concluded that “the MOE process has provided a balanced approach between noise impact and the need for wind farms, based on currently available scientific data”, but did not cite that scientific data (and I will resist the temptation to elaborate on the fact that there is no scientific evidence proving that wind farms are an effective alternate-energy “solution”). On page 50, the report also mentions an old ISO document that lists the expected community response to wind-turbine noise that exceeds the ambient sound level: sporadic complaints at 5 dB, widespread complaints at 10 dB, threats of community action at 15 dB, and vigourous community action at 20 dB.

If MOE continues to hold to its unfounded 40 dB(A) basis, it will continue to perpetuate situations in which the community response will range from widespread complaint to vigorous action.

For an excellent treatment of what remains to be addressed, I refer you to the 2008 article The Public Health Issue by John Adams. There are still many failures in the new regulations that MOE proposed on June 9, 2009.

Where’s the science?

Is there any scientific data from which one can compute limits to wind-turbine noise that would be acceptable? Yes. Two recent scientific works from Japan seem relevant. The first verifies that people’s ability to detect vibration follows a normal distribution, for all frequencies from 25 Hz to 18,000 Hz. The second lists the dB levels that can be detected by various percentages of the population at each frequency:

I believe these reports can be used to compute people-friendly limits. Consider the following “study in methodology” example. (I am not a wind or vibro-acoustic expert, and it may well be that my treatment is flawed. If so, I invite correction and clarification.)

  1. From the second report, we can take the dB values for the mid-point frequencies that are used in computing dB(A) values, and derive the lowest dB(A) levels that various percentages of the population can detect.
       percent:  1   5  10  20  30  40  50  60  70  80  90  95  99
       dB(A):    7  12  13  16  18  20  21  23  25  27  30  33  38
  2. At a public forum held by Ashfield-Colborne-Wawanosh Township Council concerning objections to EPCOR‘s “Kingsbridge 1” project, an expert from EPCOR stated that passage through walls and open windows will reduce the exterior noise by 10 dB(A). [This is not true for low-frequency emissions, but MOE acknowledges that as a matter for further regulatory study.] So, adding 10 to the above gives us the lowest outside dB(A) that can be detected inside.
       percent:  1   5  10  20  30  40  50  60  70  80  90  95  99
       dB(A):   17  22  23  26  28  30  31  33  35  32  40  43  48

    [As an “aside”, I object to assumptions that it is acceptable to force people to go into their homes and close their windows on nice summer evenings in an attempt to lessen the impact of wind-farm noise.]

  3. The ISO document mentioned above states that an increase of 5 dB(A) above the ambient level will yield sporadic complaints from the community, and an increase of 10 dB(A) will produce widespread complaints. So, by adding another 5 to the above, we get the dB(A) at which we can expect a wind-turbine farm to begin causing complaints from various percentages of the community.
       percent:  1   5  10  20  30  40  50  60  70  80  90  95  99
       dB(A):   22  27  28  31  33  35  36  38  40  42  45  48  53
  4. By reversing the top row, we get the percentage of the community that should find the dB(A) level acceptable.
       percent: 99  95  90  80  70  60  50  40  30  20  10   5   1
        dB(A):  22  27  28  31  33  35  36  38  40* 42  45  48  53

As indicated by the * above, the 40 dB(A) basis for MOE’s regulations meets the needs of only 30% of the rural community.

There are reports in the literature that are consistent with this. For example, in Public Health Impacts of Wind Turbines (Minnesota Department of Health, May 2009), on page 17, the authors cite two studies from Sweden …

[…] when noise measurements were greater than 40 dB(A), about 50% of the people […] reported annoyance. When noise measurements were between 35 and 40 dB(A), about 24% reported annoyance. Noise annoyance was more likely in areas that were rated as quiet […]

The diagram on page 20 shows that 30% are “highly annoyed” by exposure to 40 dB(A), but that the percentage of “highly annoyed” approaches zero at 32 dB(A).


Regulations derived from ambient background noise seem much more appropriate to quiet rural environments in which summer night-time lows of 25 dB(A) are common. For example, the above derivations from the Japan reports suggest that France’s regulation of not more than 3 dB(A) above the lowest night-time ambient level would address the needs of 90% of the rural population. [As an “aside”, I do not believe that a regulation that meets the needs of only 90% of the community is acceptable; 99% should be the minimum objective.]

An ambient low of 25 dB(A) would result in 28 dB(A) as the limit. That is 12 db(A) less than MOE’s 40 dB(A). A doubling of distance results in a 6 dB(A) decrease in noise. Therefore, a rural limit of 28 dB(A) would require a quadrupling of MOE’s new proposed setbacks. Thus, the minimum setback of 550m becomes 2.2km, and the table becomes …

Number of Wind Turbines Setback in meters (m) from closest Point of Reception corresponding to wind turbine Sound Power Levels in dB(A)
102 103-104 105 106-107 >107
1-5 turbines 2.2km 2.4km 3.4km 3.8km Noise study required
6-10 turbines 2.6km 2.8km 4.0km 4.8km
11-25 turbines 3.0km 3.4km 5.0km 6.0km
26+ turbines Noise study required

In addition, MOE’s table needs to be expanded on the left to include categories down to 90 dB(A) to encompass evolving technology, and MOE’s regulations need to be based on peak wind-turbine noise (Lmax or L10), rather than hourly-averaged noise (Leq), in accordance with MOE’s directive that projections must be based on predictable worst-case impacts. It is not the averages that people perceive, it is the peaks.

What’s the conclusion?

The resolution of this must fall to our elected representatives. For what percentage of the population is it acceptable to sacrifice health and well-being in order to meet green-energy goals for wind turbines?

To quote Gordon Miller, Environmental Commissioner of Ontario, in his “State of the Great Lakes” keynote address at the 2008 It All Ends Up In The Lake conference … Make waves.