How do jurisdictions come up with the rules in the first place? Neither Canada nor the U.S. have any national standards for noise control, so at best it is left up to the states/provinces. Most states have at least basic frameworks for noise limits, but their standards and how rigidly they are pressed onto local governments varies widely. So far it has mostly fallen on local governments to set standards for their communities. Without local standards the state standards (which are typically minimums) are used.
A typical scenario – a developer comes into a rural area waving a bunch of money and promoting green energy. This is the first major industrial project in some of these communities and the local council members have no idea what questions to even ask. They are pretty much at the mercy of the well-prepared developers and often accept at face value the assurances that there will be no noise, so there’s no need to take the trouble to pass regulations restricting the noise levels. Or maybe they use a nearby town’s rules, or fall back on the usually looser state rules. The developer may know that there will be significant noise and that nearby residents will probably complain, but as long as the noise is below what the agreed rules, there’s nothing the residents can do. If needed, additional tactics might be used where needed, like appeals, legal blackmail and court fights. A common tactic is to demonize the opposition as either nimbys and/or jealous. Even when the locals make a good-faith effort to determine what regulations might be appropriate, they get so much conflicting information that they are stymied.
Initially the above strategy was quite successful. The first wind projects in an area usually sent in with very little oversight. Pubnico in Nova Scotia, Mars Hill in Maine, and Lincoln Township in Wisconsin were early projects that ended up creating large numbers of complaints. Other towns got wind (so to speak, yuk yuk) of the troubles and became more protective. The wind industry responded by lobbying at the state/provincial level, trying to get the state to pass legislation to override local control. The state says they do this in the name of “standardizing” and “streamlining” the process, but in every case the rules favor the wind projects and weaken local protections. Ontario, Wisconsin and New York are all in the process (as of March 2009) of passing such laws. There are also efforts at the national level, in places like England and the US.
The topics below cover in more detail different aspects of the rules commonly contained in noise regulations with the usual emphasis on wind turbine noise and Ontario.
Noise Limits and Setbacks
Governments generally use some combination of two techniques to control noise at wind farms. The first way is the most direct – establish upper noise limits (in decibels, usually annotated dBA). While this seems to be the most obvious solution, in practice it gets complicated and oft times allows circumvention. The second way is to establish setbacks, and while indirect it is quite difficult to circumvent. The noise regulations of most non-local governments use the first method, usually because the noise regulations pre-dated wind farms.
Local councils, often enacting rules about specific wind farms, may use a combination of both methods, with the first method reflecting the legal mandate and the second used in siting decisions. Even for the same project both methods are used. During the planning phase a computer modeling of the noise levels is generally done, providing the regulators with assurances that the applicable decibel-related noise limits will not be exceeded. But when it comes to actually citing the turbines, most of the discussion and negotiation is about setbacks. And then, once the project is in operation, any complaints or any legal action would probably be judged against the noise level regulations, not the setbacks.
Within decibel-related schemes there are also two techniques: (1)limits above pre-project ambient levels and (2)absolute levels. Often they are used in an and/or combination. Ambient-related rules typically measure or assume an ambient noise level and then the allowance is added to arrive at the upper limit. If measuring, most jurisdictions use L90 as the ambient. Measuring requires professional equipment, set up by a professional, left in several places for several days each, in all sorts of weather. This gets expensive. So more often the area is generalized as i.e. “rural” and assigned an ambient noise level of i.e. 30dB. The allowance typically varies anywhere from 5 to 10 dB. So a typical example would be an assumed ambient of 30 plus the allowance of 10 for a maximum of 40 dB. Absolute levels are just that. They typically range anywhere from 40 to 50 dB. Many jurisdictions add penalties to these limits for certain types of more intrusive noise, i.e. it varies in some sort of pattern (like beats) or has some sort of tone to it.
Although I’ve used the words “maximum” and “upper limit” I’m not aware of any rules that specify an instantaneous maximum, or even an L10 maximum. Almost without exception they use Leq. Because Leq must be taken over a time period they may specify the minimum time, i.e. 10 minutes. Of course this allows some amount of noise above the maximum, as long as the average stays below the limit.
Most jurisdictions further define the limit to use one of the standard frequency weightings. I’m not aware of anyone who uses anything but the A weighting. So the final expression of the maximum might be something like LAeq10.
While larger governments (and court systems) tend to regulate noise by rules that specify noise limits in dBA, local governments and other interested parties tend to use setbacks. Most of these setbacks were created empirically, typically set where complaints stopped or were at least minimized. Most developers maintain that large fixed setbacks are unnecessary, as they claim to have the tools to customize each installation to minimize any problems. Unfortunately, historical experience does not support this claim. On the other hand, fixed setbacks pretty much relieve the developer of any legal problems with noise. So most developers can live with fixed setbacks, as long as they don’t limit the number of turbines too much.
Bill Palmer wrote this nice explanation on setbacks, along with this chart comparing European and Canadian setbacks. At the Picton Conference Orville Walsh presented the latest compilation of standards.
Below is a sampling of setbacks that are either part of a local government’s zoning rules, or a recommendation from an interested party, or a negotiated settlement with a developer.
- World Health Organization – 1500 metres
- France, Academy of Medicine – 1500 metres
- England, UK Noise Association – 1 mile
- Riverside, California, city council – 2 miles
- Cartier, Manitoba, city council – 2000 metres
- Ontario MOE – 350 metres, updated to 550 metres mid-2009
- Wolfe Island – 400 metres
- Ripley – 700 metres
- Kincardine/Enbridge – 450 metres
- Kittitas county, Washington – 1640 feet
- Frey – 2000 metres
If there’s any organization that is (a)competent, (b)disinterested and (c)experienced in matters relating to health, it would be the World Health Organization, or WHO. Their current noise recommendations have been around for quite some time, predating modern wind turbines, but they still are useful in comparing what the world’s experts are recommending with what, as an example, Ontario requires. Anything italicized is from their publication, the WHO Guidelines for Community Noise, published in 1995, updated in 1999.
Section 4, Guideline Values
Part 4.3, Specific Environments
Noise measures based solely on LAeq values [LAeq is a logrithmic average, using the A weighting] do not adequately characterize most noise environments and do not adequately assess the health impacts of noise on human well-being. It is also important to measure the maximum noise level and the number of noise events when deriving guideline values. If the noise includes a large proportion of low-frequency components, values even lower than the guideline values will be needed, because low-frequency components in noise may increase the adverse effects considerably.
In other words, if the noise in question has either a large low-frequency component or a maximum level that is above the LAeq average – and the noise from wind farms has both – then you should use lower limits. How much lower is not specified. Since just about everyone, including Ontario, uses LAeq (even the WHO in the following sections) the numbers below should be even smaller than they are.
Part 4.3.1, Dwellings
In dwellings, the critical effects of noise are on sleep, annoyance and speech interference. To avoid sleep disturbance, indoor guideline values for bedrooms are 30 dB LAeq for continuous noise and 45 dB LAmax for single sound events. Lower levels may be annoying, depending on the nature of the noise source. The maximum sound pressure level should be measured with the instrument set at “Fast”.
To protect the majority of people from being seriously annoyed during the daytime, the sound pressure level on balconies, terraces and outdoor living areas should not exceed 55 dB LAeq for a steady, continuous noise. To protect the majority of people from being moderately annoyed during the daytime, the outdoor sound pressure level should not exceed 50 dB LAeq. These values are based on annoyance studies, but most countries in Europe have adopted 40 dB LAeq as the maximum allowable level for new developments (Gottlob 1995). Indeed, the lower value should be considered the maximum allowable sound pressure level for all new developments whenever feasible.
At night, sound pressure levels at the outside façades of the living spaces should not exceed 45 dB LAeq and 60 dB LAmax, so that people may sleep with bedroom windows open. These values have been obtained by assuming that the noise reduction from outside to inside with the window partly open is 15 dB.
For outdoors, the important numbers (all in LAeq) are 55 causes serious annoyance, 50 is moderate, and most European countries have adopted 40, which is also what this Guideline recommends. At night the levels must be consistent with an ability to sleep, which translates to an outside LAeq of 45 and an LAmax of 60, both of which assume a house will produce a reduction of 15. The reduction of 15 may be optimistic – in other words, many houses have walls that will not provide this much reduction. This is especially true if windows are open and with low-frequency sounds. In my actual case with the outside noise measuring at 40dbA and the window closed my bedroom measures 25dbA, which is very quiet. But then, I have very-well-insulated 6″ walls. With my one smallish window open, the noise goes up to 27dbA, which is still very quiet.
LAmax is generally not available, so the next best measurement is LA10, which would be the level that is exceeded 10% of the time. This is probably the best measurement available for the “swooshes” that windmills produce. But Ontario still uses LAeq.
- There are many ways in addition to what I’ve covered here in which the Ontario regulations are looser than the WHO guidelines. John Harrison has written a report that details these ways.
- The WHO Guidelines for Community Noise, section 4 “Guideline Values” is here.
It is well understood that wind turbines produce more noise in certain directions and in certain conditions. So many jurisdictions add language to the rules that specify “worse predictable case” or something to that effect.
So far these rules seem fairly straight-forward. But the devil is always in the details, and I’ll have a great deal more to say about the computer modeling and taking measurements in the Noise Regulation Problems post.