REF has responded to the Insitute of Acoustics' consultation on their discussion document entitled “A Good Practice Guide to the Application of ETSU-R-97 for Wind Turbine Noise Assessment”.
1.1 The Renewable Energy Foundation (REF) has taken a particular interest in the issue of wind farm noise assessment and has produced three information notes relevant to this subject:
(i) Freedom of Information data obtained from University of Salford Report on wind farm amplitude modulation (AM) noise complaints 
(ii) An assessment of the Den Brook AM condition 
(iii) A Critique of the IoA Treatment of Background Noise for Wind Farm Noise Assessments 
1.2 In each of these documents we have stated that the existing wind farm noise guidance document, ETSU-R-97, is out of date, contains errors and omissions, and is not fit for purpose.
1.3 The current exercise (of which this consultation is part) has been undertaken by the Institute of Acoustics, under commission by DECC, in order to develop good practice guidance based on the existing ETSU-R-97 standards.
1.4 However, it is misguided. There are inherent problems from the outset, including the fact that the indicative noise limits suggested in ETSU-R-97 may not be questioned. Furthermore, the character of turbine noise is also off limits and the actual environmental impacts of the noise standards in terms of audibility and likelihood of complaints are not addressed. Moreover, the guidance diverges from ETSU-R-97 in a number of key areas resulting in a reduction in protection from noise for wind farm neighbours.
1.5 We also believe that the IoA document, like ETSU-R-97, is not fit for purpose. The major deficiencies are as follows.
1.6 The purpose of the IoA document is to help ensure that local planning authorities and planning inspectors receive reliable information on the noise impacts of a proposed wind farm in order that a robust planning decision may be made.
1.7 The IoA guide does not appear to recognise that it is a legal requirement that a noise assessment forming part of an Environmental Statement must supply “the data required to identify and assess the main effects which the project is likely to have on the environment” , and that the “direct effects and any indirect, secondary, cumulative, short, medium and long-term, permanent and temporary, positive and negative effects of the project must be described”. 
1.8 A noise assessment is required to describe the ‘levels and effects of noise from the development’.  There is an obligation that the ‘democratic right of a member of the public to make representations must be meaningful and therefore the information which is made available must be sufficient to enable a member of the public:
a) to respond to the significant effects on the environment to which it is suggested the project may give rise;
b) to examine the project to see whether it is likely to give rise to significant effects which have not been identified.’ 
(Emphasis – bold – added.)
1.9 The problem with the IoA Discussion Document is that it does not require a noise assessment to include any interpretation of the myriad of technical acoustic detail such that the actual environmental effects can be understood in any “meaningful” way, as is required by law.
1.10 For example, there is no requirement to quantify or to discuss whether the wind farm development will be audible at neighbouring dwellings, either indoors or outside and for what duration. That is to say, the IoA does not require that a developer assess and describe the likely impacts of the noise in the context of the existing soundscape.
1.11 As a consequence of this fundamental flaw, a developer who followed the guidance indicated in the IoA document would nevertheless fail to comply with over-arching legal requirements, and neither the neighbours of the proposed application, nor the decision maker, would have access to non-technical noise information that would allow them to understand the effects that would result. This is obviously an absurd outcome, and the IoA document must rectify the failing.
1.12 Acousticians acting for developers routinely exceed their area of expertise in noise assessments; their reports often contain claims in relation to wind farm power output, meteorological factors or impacts of noise on sleep and health of neighbours. The IoA document also errs in this way. The acoustician’s role is to do no more than gather and interpret the necessary acoustic data, providing the public and decision makers with a clear and accessible description of the noise impacts.
1.13 It is not the acoustician’s role to make value judgements about the merits or otherwise of the planning application. At times, the IoA document strays into planning territory and makes statements which reveal an inappropriate support for wind generation. For example paragraph 4.3.4 comments that a conservative choice of noise prediction parameters will negatively impact wind energy generation levels. The discussion on increased noise limits for dwellings which may or may not house people with a financial interest in a wind farm is not appropriate for a document on acoustic practices (paragraphs 3.3.21-22).
1.14 Similarly, there is an extensive discussion on what noise limit should be set for a locality which takes into account the power output of a wind farm. Calculating power output does not lie within the area of expertise of acousticians and should not form part of their deliberations.
1.15 It is our position that acousticians should concentrate on quantifying the “likely duration and level of exposure” (paragraph 3.3.2), as we have done, by calculating the percentage of time in a year that complaints would be likely or marginal based on the BS4142 metric. This information could then be used by a planning inspector or local authority to decide if a proposal is satisfactory and what noise limits would be acceptable given the site-specific results.
1.16 ETSU-R-97 is clear in describing how to measure the dependence of background noise data on wind speeds measured at 10m height. The IoA document seeks to change the way this is done and to endorse a new methodology that relies on the concept of a ‘standardised 10m wind speed’. This is not the actual 10m wind speed, but a hypothetical wind speed that only occurs in certain meteorological conditions.
1.17 The ‘standardised 10m wind speed’ method is demonstrably less protective of wind farm neighbours. In our previously published critique of this new methodology, we demonstrated by use of real wind speed data, that this methodology
- Increases the uncertainty of the background noise curves, contrary to ETSU-R-97 guidance, and consequently, reduces the reliability of noise conditions based on them, 
- Produces different noise conditions depending on the dates chosen for the baseline background noise survey, thus yielding differences in permitted wind farm noise levels of as much as 5dB for the same site , and
- Can even result in permitting site layouts where the wind farm noise would breach ETSU-R-97 conditions known to be unacceptable. 
1.18 A further problem is that noise conditions based on a ‘standardised 10m wind speed’ require accurate measurement of hub height wind speed in any noise compliance testing. As can be seen in paragraphs 8.1.10 of the IoA document, this is problematic: nacelle anemometers need to be accurately and verifiably calibrated and the wind speeds measured by the nacelle anemometers on each turbine will be affected by the wake of adjacent turbines. Even if a free-standing meteorological mast is part of the site application, there will be ambiguity about the accuracy of wind speeds when the mast is within the wake of the turbines. The IoA document is silent on how these issues could be resolved.
1.19 No valid justification has been presented by the IoA for changing this part of ETSU-R-97 guidance. In view of the fact that it decreases the protection of wind farm neighbours, results in unreasonably variable noise conditions, and increases the difficulty of achieving an enforceable noise condition, it is hard to believe that any such justification could exist.
Assertions unsubstantiated by data
1.20 There are a significant number of unsubstantiated assertions in the IoA documents.
1.21 For example, it is stated that “there is no evidence that atypical wind shear conditions during the survey period can have other than a minor effect on the derived background noise levels for the site” (paragraph 2.3.4). On the contrary, the empirical data that REF produced in its Critique provides such evidence, and moreover, the assertion is weakened by the admission at paragraph 3.1.8 that “there is at present limited information on the potential significance of the variations of wind shear conditions during typical survey periods.”
1.22 The document states that there is no reason to suggest that background noise measurements need be taken at more than one time of the year and produces no evidence to justify disregarding seasonal variations. This ignores evidence in the public domain which demonstrates seasonal variations in background noise measurement; for example, the noise assessment for the Linton Wind Farm, undertaken by the Hayes McKenzie Partnership, includes data for winter and summer surveys and demonstrates substantial variation in noise levels between the two seasons.
1.23 It is claimed that realistic predictions of turbine noise levels can be achieved with particular prescribed parameters (paragraph 4.3.5). No data is available in the public domain by which this assertion can be validated and although some of the authors of the document clearly have access to such data, they have routinely refused to publish that data.
1.24 These unsubstantiated assertions are in themselves fatal to the credibility of the document, but they are only a selection. There can be no confidence in good practice guidance unless it is rigorous and its claims are capable of independent verification using publically accessible data.
1.25 Wind farm developers are often reluctant to allow neighbours access to the raw background noise, turbine noise and wind speed data necessary to independently determine the noise impacts of a proposed wind farm. Members of the public still experience the extraordinary situation of being refused access to background noise measurements taken on their land with their permission.
1.26 This is contrary to European and UK environmental law, which gives the right to the public concerned to be given early and effective opportunities to participate in the environmental decision-making procedures.  Without access to the relevant data, it is impossible for neighbours, or an acoustic expert acting on their behalf, to test claims made by a developer.
1.27 It is surprising, indeed it is disgraceful, that the IoA should be party to this tendency to withhold data. For example, we note that this document suggests an offer to provide background noise measurements data to the landowner is contingent upon the developer being in agreement (paragraph 2.2.2).
1.28 We would remind the IoA of its own Code of Conduct which includes the principle that “Members shall show proper regard for the sanctity of data. In particular members will:[…] ensure that primary data used in any publication or report are available in a form that would allow for independent scrutiny and that sufficient details of any experiments, by which the data were derived, are available to allow others to replicate such experiments.” 
1.29 We believe that it is a professional duty of members of the IoA to make sure that residents are properly informed of the implications of a noise assessment and that this includes being given all relevant data and being made aware that a considerable amount of information about their property will be put into the public domain including location and often photographs of their house and garden.
Reliance on noise conditions unfounded
1.30 As a result of developers producing noise assessments for candidate, rather than specific, turbine models or relying on noise-capped, reduced-power, turbines (e.g. paragraph 4.5.1), it seems that it is implicitly accepted that noise conditions are now being used as a means of constraining noise contrary to the assumption in ETSU-R-97.
1.31 This necessarily means that noise conditions must be robust and capable of very rapid enforcement. Experience has shown that this is not the case, and the evidence in the IoA document is that wind farm noise conditions are over-complex, utterly dependent on data in the developer’s control and open to interpretation.
1.32 It is unreasonable to recommend that noise conditions should be based on ‘standardised 10m wind speeds’ derived from measured hub height wind speeds when measuring those wind speeds is flawed as demonstrated by the document itself.
1.33 The tonal part of the noise condition is clearly inadequate if it cannot protect neighbours from tonal noise with a ‘droning character’ (Paragraph 8.2.6).
1.34 There is no apparent requirement for, or evidence of historic, monitoring of outcomes of noise complaints or noise compliance testing. We are aware that there continue to be noise complaints made about many wind farms, very few if any of which seem to be resolved to the complainant’s satisfaction. As far as we are aware, there is no compliance measurement data in the public domain.
1.35 In view of the fact that such data must exist, and that this data could be used to validate some of the assumptions made in the IoA document and the efficacy of current noise conditions, we believe that the IoA should insist that the data is made publically available.
6. ^ Environmental impact assessment: guide to procedures DCLG http://www.communities.gov.uk/documents/planningandbuilding/pdf/157989.pdf
7. ^ Newman J. in R(Burkett) v London Borough of Hammersmith and Fulham,  EWHC 1031 para 8 (vii)
8. ^ See “Quantifying Noise Impact on Neighbouring Dwellings” in A Critique of the IoA Treatment of Background Noise for Wind Farm Noise Assessments http://www.ref.org.uk/publications/255-ioa-critique
9. ^ Appendix C of ETSU-R-97 describes how noise impacts on neighbours are worse when background noise levels are so scattered that they cannot be accurately represented by a single fitted line. Figure 3 in the REF Critique (http://www.ref.org.uk/publications/255-ioa-critique) demonstrates that the ‘standardised 10m method’ results in a third of background noise points falling 4dB or more lower than the line used to derive the noise condition.
10. ^ See table 1 in the Critique http://www.ref.org.uk/publications/255-ioa-critique
11. ^ See table 2 in the Critique http://www.ref.org.uk/publications/255-ioa-critique
13. ^ A1.5 IoA Code of Conduct http://www.ioa.org.uk/membership/code-of-conduct.asp