WSP Parsons Brinckerhoff were commissioned by DECC in 2015 to review the evidence around Amplitude Modulation (AM) in wind turbine noise with a view to recommending how excess AM could be controlled through a planning condition. The resulting Report was published by DECC’s successors, the Department for Business, Energy and Industrial Strategy (DBEIS), on 26 October 2016[i].
This review considers the recommendations of the report, from a practical, implementation perspective. For further advice, please contact Michael Reid, Associate Director – Acoustics, Arcus Consultancy Services Ltd, on 0141 221 9997 or email@example.com.
AM refers to noise which varies (modulates) in its level (amplitude) in a cyclical manner. In the case of wind turbines, this occurs at the rate that the turbines blades pass a fixed point such as the tower, i.e., blade passing frequency.
The report concludes that there is sufficient evidence of increased annoyance from wind turbine noise due to AM to justify its control through planning conditions. It discusses the elements that should form such a condition and provides an outline example, though it emphasises that the wording of planning conditions is a matter for the relevant planning authority and that legal advice would be required in relation to any specific condition.
It is notable that the Report states that the proposed control mechanism and outline condition could be applied to new consents and that the investigation of nuisance is outwith the scope of the Report. This precludes the retrospective application of the method and outline condition to existing wind energy developments.
It recommends that AM is quantified using the method recently published by the Institute of Acoustics (IOA)[ii]. It then goes on to recommend a penalty scheme, which is that proposed by Renewable UK[iii] (RUK) in 2013. In addition, the report states that the planning condition should address the following:
- It should relate to periods of complaint due to AM;
- Analysis should be made on individual 10-minute periods, with the appropriate penalty applied to each, in addition to any relevant penalty for tonality;
- An additional penalty should be applied at night, equal to the difference between daytime and night-time ETSU-R-97 noise limits, but only where the night-time limit is greater than that for the day;
- A (wind speed) bin analysis should then be carried out; and
- Enforcement of the condition should be based on professional judgement in relation to the frequency and duration of ‘breaches identified’.
The outline condition provided reads:
“During periods of complaint, the IOA metric should be applied to the data collected[iv] to derive the reconstructed AM values for consecutive 10-minute periods. For each period with an AM value of equal to or greater than 3 dB, a penalty should be assigned in accordance with Figure 11, and added to the absolute level of noise. Each summed value of Overall average level (corrected for background where necessary) + AM penalty + Tonal Penalty (if applicable) should be binned into wind speeds of 1 m/s intervals over the range of the data for when the turbine is operating and complaints occurring. Where the number of 10-minute breaches at any given wind speed during the period of complaint is considered to be unacceptable, the operator should be required to submit details of a scheme describing proposals for suitable mitigation of the unacceptable AM periods to reduce the number of breaches during the operational conditions giving rise to the complaint, to that considered acceptable by the relevant authority.”
A breakdown of the condition is presented below in relation to the well-established tests for a valid planning condition, i.e. that the condition should be:
- Relevant to planning;
- Relevant to the development to be permitted;
- Precise; and
- Reasonable in all other respects.
We believe that tests 1 to 3 are met by the condition. Our comments on compliance with tests 4 to 6 are set out below.
“During periods of complaint, the IOA metric should be applied to the data collected to derive the reconstructed AM values for consecutive 10-minute periods.”
The precision of this part of the condition could be improved by providing a specific reference to the ‘IOA metric’. It is important that planning conditions ‘stand alone’ i.e., that they provide all of the information necessary to their interpretation.
“For each period with an AM value of equal to or greater than 3 dB, a penalty should be assigned in accordance with Figure 11, and added to the absolute level of noise.”
Figure 11, within the report, is a reproduction of the penalty scheme graph from the RUK planning condition. The report also contains, as Figure 12, a similar chart of the author’s own creation and titled ‘Proposed Level Penalty Regime’ which would have been a more appropriate reference. We would recommend that the chart in question is attached to any condition applied in practice.
“Each summed value of Overall average level (corrected for background where necessary) + AM penalty + Tonal Penalty (if applicable) should be binned into wind speeds of 1 m/s intervals over the range of the data for when the turbine is operating and complaints occurring.”
This clause is particularly problematic for the following reasons:
- ‘Overall average level’ is undefined. The previous sentence refers to the absolute level of noise within each 10-minute period. Under the ETSU-R-97[v] method, this is not an average level, but the level exceeded for 90% of the measurement period, i.e. LA90,10min. These levels are then ‘averaged’ by plotting them on a X-Y chart with wind speed on the x-axis, noise on the y-axis and a regression (best-fit) line applied. Alternatively, they are sometime sorted into ‘bins’ by wind speed of 1 m/s intervals as discussed, however in the outline condition this occurs after the ‘averaging’. If the condition is intended to refer to the ‘average’ noise level within a 10-minute period, this would be correctly expressed as the equivalent noise level, LAeq,period, which would be a significant and potentially unreasonable departure from the ETSU methodology.
- The penalties for AM and tonality are additive. This is not in itself unreasonable as, for example, BS4142:2014 applied penalties for different acoustics characteristics such as tonality and impulsivity in an additive manner. However, in the ETSU-R-97 method, the penalty for tonality is not applied directly to 10-minute measurements, but to the noise level derived through regression or bin analysis for each wind speed.
- The outline condition makes no reference to the additional night-time penalty recommended elsewhere within the report.
The clause is ambiguous and therefore not sufficiently precise. Arguably, it may also be unreasonable to apply the ETSU tonality penalty in a different manner to which it is intended.
“Where the number of 10-minute breaches at any given wind speed during the period of complaint is considered to be unacceptable, the operator should be required to submit details of a scheme describing proposals for suitable mitigation of the unacceptable AM periods to reduce the number of breaches during the operational conditions giving rise to the complaint, to that considered acceptable by the relevant authority.”
The above clause contains the following issues:
- It refers to 10-minute breaches, but does not define a ‘breach’, which constitutes a lack of precision;
- If it is intended that a breach is when the sum of the ‘Overall average level (corrected for background where necessary) + AM penalty + Tonal Penalty’ is greater than the ETSU limit for the corresponding wind speed, then this a significant departure from the ETSU methodology, in which the noise limits relate to ‘average’ levels for each wind speed, as discussed above. It would be unreasonable to apply ETSU limits in a manner in which they were not originally intended;
- Alternatively, it may be intended that ‘breach’ refers to an AM value of 3 dB or greater. If this were the case, then it would not be necessary to add penalties to make this assessment.
- There is no criteria provided for when the number of ‘breaches’ becomes ‘unacceptable’. This is a lack of precision and potentially unreasonable. Elsewhere within the report it is stated that the assessment of acceptability in relation to the frequency and duration of AM is a matter for professional judgement. It would likely therefore result in developers’ consultants arguing that a certain amount is acceptable with local authority EHOs taking a different view. This would not be in anyone’s interests as it is likely to lead to delays in resolving complaints, which is contrary to the intentions of the report and the condition.
In conclusion, the outline condition fails the tests of precision and reasonableness in a number of aspects. These may be sufficient to render the condition unenforceable. This is a cause for concern as there is a real possibility of the outline condition being uncritically applied in the absence of recognised alternatives. This could lead to protracted disputes between operators and planning authorities in the event of attempted enforcement.
Notwithstanding the above concerns, the Report represents a useful first step towards a control mechanism for AM. The recommendation of a penalty scheme is useful. The Report itself recommends further research into AM and the effectiveness of the condition.
Associate Director – Acoustics, Arcus Consultancy Services Ltd
2 November 2016
[iv] Data should be collected in accordance with the IOA Supplementary Guidance Note 5 at
[v] ETSU-R-97 The Assessment and Rating of Noise from wind Turbines is the primary standard for wind turbine noise assessment in the UK
Update March 2017
An article in the March / April edition of the Institute of Acoustics (www.ioa.org.uk) publication Acoustics Bulletin, by the authors of the DECC / DBEIS AM Report sets out to clarify the control scheme and outline planning condition presented within the Report. The article provides a welcome clarification as to how the control scheme should be implemented in practice. The control scheme itself appears to have merit. However, concerns remain regarding the outline planning condition. Any planning condition should have sufficient clarity and precision for it to be understood in its own right. However, in this case, the planning condition requires reference to its source document, and the new article in order for it to be properly understood. It therefore remains the case, in my opinion, that further work is required before such a condition is applied in practice. A preferable approach may be to include a condition requiring that at the reasonable request of the planning authority, following a complaint about wind turbine noise, the operator should employ an independent acoustic consultant to assess the level of AM within the wind turbine noise in accordance the recommendations of the DECC / DBEIS AM Report.