Wind turbine noise and the abuse of technocratic regulation: Part 1

Imagine that you have moved from a city or surrounding suburbs in search of peace and to live in natural surroundings. Then, you learn that an energy company has taken an option on land owned by a neighbouring farm to develop a wind farm. You are contacted by someone working for the developer who wants to put some noise measurement equipment in your garden as the basis for estimating what they refer to as “background noise” at your property. On probing you learn that wind turbines emit a lot of noise at hub height but that there are “rules” which are supposed to set an upper limit on the increase in amount of noise at your property due to the development.

As someone of a sceptical disposition, you decide to investigate further. Specifically, you obtain copies of the environmental reports prepared for other wind farms and read the sections which present an assessment of the noise impacts of those projects. All of them claim to have applied an approach referred to as ETSU-R-97 published in 1997 with revisions or clarifications published by the Institute of Acoustics in 2013. As someone used to dealing with technical material, you read the noise assessments and the guideline on preparing. Much of the material seems to be either boiler-plate or gobbledegook, but in all cases the authors claim repeatedly and in bold that the ETSU guidelines will be met.

After a little more probing you talk to a planning officer, who eventually admits that they do not have the technical skills to verify noise assessments. They rely upon the opinions of Environmental Health Officers (EHOs), who have many other things to do. On enquiry, you discover that EHOs have little or no training in dealing with complex noise issues, so they simply rely on the statements in the noise assessments that the ETSU Guidelines will be met.

One thing that you notice is that the background noise values claimed in the noise assessments are derived from curves fitted to the data that the measuring equipment to be installed in your garden will collect. However, the curves are not fitted to all the data points. Many appear to have been excluded for a whole variety of reasons including rain, “directional filtering” (??), and “manual exclusions”. The statistical analysis has the feel of being entirely arbitrary.

Further, the fitted curves rely on something called “wind speed standardised to 10 metres”. That couldn’t be the wind speed at your property, since no measurements were taken. You learn that wind speeds were measured at a met mast installed at the proposed site for the wind farm. Was that at 10 metres above ground level? No: wind speeds were (a) measured at various heights, (b) converted to estimated wind speeds at turbine hub height, and (c) converted again to wind speeds at 10 metres above ground level. You learn that the conversion formulae used in these steps (b) and (c) are different and, in fact, fundamentally inconsistent. Finally, you are told that the formula used to convert from hub height to 10 metres is outdated fiction with no basis in what might be regarded as current scientific practice

As you try to understand the process, you come to the growing realisation that this is an Alice in Wonderland world, whose connection with actual noise at your or any other location as well as the impact of proposed wind farms is at best random. There are other things that you may learn. Notwithstanding data exclusions, the fitted curves used to estimate background noise are calculated using statistical methods that were out-of-date 30 years ago when the ETSU Guidelines were originally drafted. Statistical analysis has changed a lot since then, so again the background noise levels are as questionable as the standardised wind speeds on which they are based.

Still, in accordance with ETSU the developer’s noise assessment proposes that the turbine noise at your property should not exceed 36 dB for a standardised wind speed between 4 and 7 metres per second (m/s). What does that mean? Suppose the background noise at your property was estimated to be 27 dB (above average for a quiet rural area), so the increase is 9 dB. Noise measurement in decibels is based on a logarithmic scale. An increase of 3 dB corresponds to a doubling of noise intensity, so an increase of 9 dB implies an increase of 8 times in noise intensity. Not so small!

What was a quiet rural location will no longer be quiet. However, the consultants claim that this is a “conservative” estimate – i.e. noise levels should be less than 36 dB most of the time. Your immediate reaction is to ask how they can know that. What is “most of the time”? Suppose that, after the wind farm starts to operate, you buy a noise meter (there are plenty of decent ones available) and put it in your garden. You discover that, for example, the measured noise level exceeds 36 dB for an average of 25% of measurement periods (usually 10 minutes) during evenings and weekends when you are not at work.

You complain to your local authority. They are responsible for enforcing the noise limit specified in the noise assessment. What happens next? Is the wind farm instructed to ensure that noise limit is met by using what are called quiet operating modes for the turbines (though not so quiet in reality) or by turning off some turbines? Er … no. Instead, the wind farm operator is supposed to appoint a noise consultant – probably the same one who claimed that there would be no problem in meeting the noise limit at your property – to go through the same palaver of collecting noise measurements and wind speeds to repeat the original analysis.

Then, of course, they will claim that there is nothing to see and that you are imagining the noise problem. The fact that you can’t sleep at night because of an intermittent rumble from the direction of the wind farm or you can’t enjoy being in the garden on summer evenings is irrelevant. Even noise measurements made using professional equipment are rejected as immaterial. Without the magic authority of “qualified” acousticians all counter evidence is rejected.

But what is this qualification? Since noise assessment is based on large amounts of statistical data, do “qualified” acousticians have any significant expertise in statistics and data analysis? No. Most haven’t learned anything beyond first year undergraduate level. In addition, they rely upon packaged software that is nearly 3 or 4 decades out of date. Well, perhaps they have detailed training covering the perception and impact of noise on listeners? No, that is for audio engineers and acoustic designers. What they are expert at is noise measurement and the use of noise dispersion models. Whether the latter are of any use for wind farms is open to question, since it is difficult or impossible to check them empirically.

Any consultant with even a limited training in statistics ought to know that the procedure for assessing whether the noise limit has been exceeded is complete statistical nonsense. Noise measurements are “noisy” in statistical terms – i.e. they tend follow a distribution with a high level of variability - so estimates of mean of the distribution are very imprecise. This uncertainty is exacerbated by the systematic exclusion of data points due to factors like rain, wind direction, the dawn chorus of birds, etc.

A large part of the problem lies in the fact that a noise limit as interpreted under the current (and proposed future guidelines) is no such thing. Suppose you move to live a village which claims to have a speed limit of 30 mph. Then you learn that the “speed limit” is interpreted as requiring that any specific driver’s average speed when travelling through the village must be less than 30 mph over a 3-month period. Quite apart from the obvious difficulty of collecting data to enforce such a rule, it is absurd as a mechanism to protect villagers from the effects of excessive speeds. Of course, we can agree that there may be some uncertainty about measurements of vehicle speeds as well as special cases such as fire or rescue vehicles and ambulances on emergency calls, but none of this alters the basic principle that a speed limit is exactly that for 90% or 95% of all vehicles observed.

What makes all this worse is the reliance on a priesthood of “qualified” – i.e. technocratic - consultants to interpret and apply the rules. A large majority of these consultants rely primarily on assignments from wind farm developers and operators for their income and continuing employment. Thus, irrespective of their competence and independence, many of those who wish to dispute noise assessments for proposed developments and later analyses of turbine noise feel that the priesthood is significantly compromised by knowing full well on which side their bread is buttered.

The last straw for you and anyone else at the tender mercy of this technocratic elite might be to learn that the UK government department responsible for overseeing the wind industry has put out a consultation draft of a new set of noise regulations. These regulations were prepared by a project team drawn from three prominent consultancies that are employed by wind farm developers to prepare noise assessments and deal with noise complaints. The term “marking your own homework” hardly begins to capture the potential conflicts of interest. The draft regulations propose to increase noise limits (referred to as Lower Limit Values) by up to 5 dB – an increase in noise intensity of over 3 times – depending on the capacity of the wind farm and the number of properties affected.

Put plainly, these draft regulations are not about environmental protection nor about reconciling potential conflicts between the competing interests of neighbouring land and property owners. Some of these may want to protect their existing amenity rights including peace and quiet, while others may wish to generate rents by hosting wind turbines on their land. Instead, the draft regulations are an open invitation to wind farm developers to engage in oppressive and abusive behaviour that overrides existing amenity and property rights.

There is a story behind the whingeing of wind farm developers and the disingenuous enthusiasm of DESNZ bureaucrats to waive reasonable environmental protections. Most noise problems at proposed wind farms arise because of overdevelopment of the site. Wind turbines are placed too close together as well as too close to site boundaries, thus increasing the potential noise levels at neighbouring properties.

These design decisions are driven by simple greed: the goal is to claim higher capacity and thus obtain greater investment funding and, sometimes, subsidies. However, though policymakers seem to be entirely ignorant of the issue, overdevelopment can be self-defeating. It leads to what are called wake effects by which wind turbulence created by upwind turbines reduces the output from downwind and neighbouring. The losses from wake effects can exceed 15% of potential output, so adding, say, 2 turbines to a design with 10 turbines may reduce total output from the wind farm. Hence, the desire of DESNZ to promote onshore wind developments by oppressing their neighbours is not only morally indefensible but it is probably counter-productive.

In summary, official bodies and their technocratic consultants are promoting a set of regulations that are both abusive and foolish. There is a lesson to be learned here. Technocrats are less competent than they think they are – or, at least, than they present themselves to be. Noise consultants know about measuring noise and modelling how noise emissions disperse. Beyond that they know almost nothing about noise impacts and how these should be assessed or how noise limits should be enforced. Bureaucrats and planners understand even less but that does not excuse the standard practice of taking technocratic assessments as unchallengeable. The fact that a topic is unfamiliar and technically complex have never been a reason to treat the opinions of experts as gospel.

The impact of noise from wind turbines is a complex issue. In this article I have presented what you may consider to be the case for the prosecution against existing procedures and proposals to change the regulatory guidelines. In the next article, I will try to explain a case in mitigation. I am not going to try to defend the ETSU-R-97 guidelines. They have always been flawed, even technically absurd. Still, there are genuine problems with finding a good way to establish and enforce reasonable noise limits. In the final article in the sequence, I will try to outline an alternative approach that is closer to other kinds of environmental regulation.