Wind turbine noise and the ETSU-R-97 condition

Noise from a wind turbine is assessed in England under ETSU-R-97, a method published in 1997 by a working group convened by the then energy department, and national Planning Practice Guidance on renewable and low carbon energy still points local planning authorities to it. It works differently from every other noise regime you will meet. Instead of comparing a rating level with a background level at a fixed point, it derives a limit curve that varies with wind speed, because the turbine and the wind that masks it are driven by the same thing.

For a farm considering a single turbine, that difference shapes the whole evidence base. You need background measurements across a range of wind speeds before you can say what the limit even is.

What the method sets out to do

ETSU-R-97 balances the protection of amenity at nearby dwellings against the practical reality that a turbine only runs when the wind blows, and that wind itself generates noise in vegetation and around buildings. It uses the LA90,10min descriptor rather than LAeq, on the reasoning that LA90 discounts extraneous events such as birdsong, passing vehicles and rustling leaves.

The output is a set of noise limits expressed against standardised wind speed at 10 metres height, applied at the nearest noise sensitive receptors. Compliance is then measured against those limits once the turbine is operating.

Day and night limits

The framework sets separate day and night limits, and the derivation of each is different.

Quiet waking hours are defined narrowly: evenings from 18:00 to 23:00 on all days, Saturday afternoons from 13:00 to 18:00, and Sundays from 07:00 to 18:00. Where the daytime lower fixed limit falls within the 35 to 40 dB range depends on the number of dwellings affected, the effect on energy generation and the duration and level of exposure, and choosing a figure at the bottom of that range without justification will be challenged.

The 43 dB night figure is not arbitrary. It comes from an internal sleep disturbance guideline with an allowance for attenuation through a partly open window, which is why arguments about the figure usually become arguments about that allowance.

Deriving the limits from background against wind speed

This is the part that takes time on site. The assessment requires simultaneous measurement of background LA90,10min at the receptor and wind speed, over a period long enough to cover the range of speeds at which the turbine will operate, usually several weeks.

The steps:

1. Install class 1 monitoring equipment at the receptor, at a position representative of outdoor amenity, with weather protection.
2. Log LA90 in 10 minute periods, alongside wind speed and direction measured on the site mast and standardised to 10 metres height.
3. Filter out rain periods and any data contaminated by extraneous sources such as agricultural operations.
4. Fit a regression curve of background level against standardised wind speed, separately for day and night, and separately by wind direction where the receptors sit in different sectors.
5. Derive the limit curve as the greater of background plus 5 dB and the applicable fixed limit.
6. Predict turbine noise at each receptor across the same wind speed range and compare.

The Institute of Acoustics published a good practice guide in 2013 setting out how these steps should be carried out, with supplementary notes, and most acoustic reports now follow it. Councils and their consultants expect to see it referenced explicitly.

Amplitude modulation and the model condition

Amplitude modulation is the periodic variation in level as each blade passes, the swish that becomes a thump under certain conditions. Ordinary blade swish is accounted for within ETSU-R-97. What the original method did not address is excess or other amplitude modulation, which occurs intermittently and is the source of most sustained complaints.

The Institute of Acoustics published a reference method for rating amplitude modulation in 2016, and a model planning condition built on it has become common practice. The structure is: measure the modulation depth using the reference method, and where it exceeds a stated threshold, apply a penalty in decibels which is added to the measured level before comparison with the ETSU limit.

If a condition of that kind is offered or imposed, read the detail. The threshold, the penalty scale, the measurement location and the trigger for investigation all vary between drafts, and an operator who accepts a condition without checking the measurement protocol can find compliance testing impractical.

An illustrative example

Illustrative only: a single 500 kW turbine proposed on a livestock farm, with the nearest third party dwelling 550 metres to the north east and the applicant's own farmhouse at 300 metres.

Background monitoring over five weeks produces a night-time regression showing background rising from around 26 dB at low wind speeds to the high thirties at higher speeds. The derived night limit therefore sits at the 43 dB fixed floor across most of the operating range. Predicted immission at the third party dwelling peaks several decibels below that, so the scheme passes with margin, and the farmhouse is treated as financially involved.

The margin is what makes the application straightforward. A scheme predicted at 42 dB against a 43 dB limit invites objection, an argument about uncertainty in the prediction, and a condition regime nobody enjoys.

Compliance measurement and complaints

A typical condition requires the operator, on receiving a written request from the local planning authority following a complaint, to carry out measurements at the complainant's property within a defined period and to supply supporting data including wind speed, wind direction and turbine power output. Some conditions allow the authority to require the turbine to be shut down for short periods so background can be separated from turbine noise.

Two practical points for an owner. Keep the supervisory control data, because without synchronised power output and wind records the measurements cannot be interpreted. And deal with the first complaint properly, because a pattern of unanswered complaints can lead to statutory nuisance action under Part III of the Environmental Protection Act 1990, which runs alongside planning rather than being displaced by it.

Where the method is criticised

ETSU-R-97 was written when commercial turbines were a fraction of the height of modern machines, and the criticisms are worth stating plainly rather than skating over. The use of LA90 is said to under-represent the character of the sound. The night-time limit is argued by some to be too permissive against more recent health guidance. Background regression practice, particularly the treatment of data at low wind speeds, produces results that vary with analyst judgement.

Against that, the method has been tested repeatedly at inquiry and remains the framework national guidance points to. The honest position is that it is an ageing method applied with a modern good practice overlay, and that the amplitude modulation condition exists precisely because the original method left a gap.

Wind schemes almost always need a landscape case alongside the acoustic one, and a Landscape and Visual Impact Assessment usually does more to determine the outcome than noise does. Construction phase noise is handled separately under the approach described in construction noise, BS 5228 and section 61 consent.

Councils expect a suitably qualified acoustician for work of this kind, and professional review before submission is strongly advised. PlanScope can draft the Noise Impact Assessment structure and policy framing for your acoustician to complete, and route it for sign-off through the professional directory.

This is general guidance, not legal advice; check the position with your local planning authority or a qualified professional before relying on it. PlanScope provides AI-assisted draft reports for planning applications. Reports should be reviewed before submission. We do not guarantee planning approval.