Of the many issues surrounding wind energy, noise continues to be a controversial topic. The industry and governments continue to insist that wind energy projects are appropriately sited – far enough from the neighbors so they are not a nuisance. However, around the world the health/nuisance complaints and abandonments indicate that whatever rules are in place are generally not adequate.
Dr. Nicholas Kouwen, a retired engineering professor, had the time and resources to examine the noise issue in some detail for Ontario. Starting in June of 2012 and going into November he took extended noise measurements at five residences in the Grey Highlands region; three of them within the Plateau Project and two “controls” at locations away from the turbines. The Ontario wind project noise regulations, pretty much unique in the world, allow more noise at higher wind speeds so he also recorded wind speeds. He then compared the actual readings with the Ontario limits and it should come as no surprise that those limits were routinely violated.
Links to Everything
Kouwen interview, Wind Wise Radio (from 20:00 to 54:00)
The clickable figure below shows just how egregious the violations are. It shows a 17-day period where the noise and wind speed were constantly monitored at a residence located at a distance that the MOE regulations said should be compliant. The black line is the measured noise and the red line is the Ontario noise limit, which varies depending on the measured wind speed. The blue lines at the bottom show when the noise was above the limit – 82% of the time!In order to properly appreciate the results of Kouwen’s study it is necessary to first take a look at how Ontario has set the noise limits for wind turbines. The following clickable chart provides a good summary.
The magenta line shows the rural Ontario limit, which starts at 40 dBA and stays there until the wind speed goes above 6 metres/second (21 kph or 13.4 mph) at which point the noise allowed increases until it reaches 51 dBA at 10 m/s. The dashed line shows the sans turbine ambient levels in a particularly quiet place, and the limit at wind speeds above 6 m/s is set at 7 dBA above that ambient. It would be interesting to discover how the MOE set the allowable increase at 7 dBA, as a 5 dB increase typically leads to complaints while a 10 dB increase leads to actions on the part of the neighbors, and a 15 db increase typically leads to lawsuits.
It is important to realize that the developer simply runs a computer model to indicate compliance and the MOE then presumes that the model is accurate until proven otherwise – a lengthy, expensive and uncertain process. As a professor Kouwen used modeling to study underground water movement, and a necessary step was the validation of the model. He naively expected the wind turbine noise models used to keep harm from the neighbors would also have been validated. Let me know when you stop laughing.
Apparently the MOE used an ideal non-populated “location” for their base (the dashed line above). Perhaps this explains why the MOE seems to have convinced itself that their regulations are adequate. After all, many rural Ontario locations routinely experience noise at or above the green line, even without turbines nearby. This ignores two very real problems: (1) that a wind turbine’s effect on humans can be adequately equated to a dBA number (the noise wind turbines make has been shown to be uniquely disturbing), and (2) that the turbines actually meet these limits.
So he started his quest to see just how close the models were to the reality. This requires measurements. Looking at the first chart above, the industry and the MOE would likely claim what Kouwen is measuring is really just the normal ambient noise level, and he has no “proof” that the turbines are to blame. Simply going out into the area and listening would quickly indicate that the turbines are the main contributor to the noise level, but the MOE is firmly entrenched in Toronto and shows little desire to venture outside. So the numbers themselves have to implicate the turbines.
He decided that the best way to proceed was to do some long-term measurements of residences both near to and far from wind turbine and see if the turbines created more noise than the models predicted and the regulations allowed. He obtained some pretty serious noise and wind-speed measuring equipment, costing thousands of dollars. He solved the problem of protecting the equipment, especially the microphone, from wind and weather over a long term with a very clever burlap-enclosed trailer. While his setup will be dissed by the MOE (as any setup that provided inconvenient results would be) it strikes me as very fit for purpose. More importantly, it is portable and provides consistent measurements at all locations.
He presents his results in a series of charts (like the one above) that first show that the noise is above the MOE’s limits – and it is, even at distances several times as large as Ontario’s setback of 550 m. He then presents another series of charts that compare typical rural Ontario ambient noise levels with those that exist within a wind energy project. In order to hopefully better present his findings, I’ve combined these charts into one that shows all the locations at once. Thank goodness it can be clicked on to enlarge.
The plum-colored line is the MOE’s ambient line from their chart above. Kouwen’s control measurements are in blue and his turbine-neighbor measurements are in yellow. The MOE’s noise limit is in green. Even though the MOE’s regulations are very concise and say that the green line doesn’t depend on any other ambient measurements, Kouwen went ahead and adjusted it upwards to compensate for his ambients being larger than the MOE’s. His first series of charts are all based on the adjusted limits, shown by the red line. As it happens, the exceedences were so large that it wouldn’t have made much difference which criterion he used.
Note the obvious – the turbine locations are consistently noisier than the control locations, and that the noise at all locations over longer periods of time is largely dependent on the wind speed. Also note that slope of the control’s lines agree with the MOE’s slopes – indicating that they agree on the wind’s contribution, even though the base might be different.
Given that these measurements ran over multiple weeks, it is difficult to assign the higher noise levels to anything but the turbines. The results are just too consistent. I’ve tried to think of an alternative plausible noisemaker that fits this data and so far haven’t come up with one. The turbine operators could settle this issue pretty easily by agreeing to turn the turbines on and off while the acousticians were recording. Needless to say, they won’t do this voluntarily, and apparently the MOE doesn’t have the ability or the desire to order them to do so. Australia doesn’t have this reluctance and it will be interesting to see their results.
The wind industry and the MOE will likely try to ignore this study. If they can’t I predict they’ll (a) attack Kouwen’s credentials (he’s merely an engineering professor and not an acoustician) and (b) point out that his study doesn’t meet MOE standards (as if meeting MOE’s standards somehow translates to more accurate). What they won’t do is produce any measurements that contradict his, nor will they offer any substantial critique of his methods. This is a solid piece of work, and will be valuable to anyone looking for an honest look at what is going on out in the field.
In addition to the measurements he discusses why the limits and the models are not sufficient in the first place, and that’s important enough for its own posting.