Hinnang

Critical Analysis and Scientific Review of the Report produced by the University of Tartu

in October 2025, titled:

Health effects of wind turbines: A systematic review of studies published in peer-reviewed scientific journals over the last fifteen years—Development of a methodology for interpreting the results of scientific studies on the potential health effects of wind farms and other energy production technologies in the Estonian context

Document IARO25-6

October 13, 2025

Review of Tartu Univeristy Report, Estonia

Page 2 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

International Acoustics Research Organization

IARO is an international group of researchers with a mission to investigate acoustical environments, especially with respect to features that affect humans and animals, and to publish the results. IARO holds the ethics approval for the CSI-ACHE, the Citizen Science Initiative into Acoustical Characterisation of Human Environments, the results of which are publicly disseminated.

Contacts:

IARO, 37 Weston Ave, Palmerston North, 4414, New Zealand

Tel: +64 21 033 6528

Email: [email protected]

Authors of this Report (alphabetical)

Mariana Alves-Pereira, Ph.D., Lusófona University, Lisbon, Portugal

Huub Bakker, Ph.D., IARO, Palmerston North, New Zealand

Richard Mann, Ph.D., Waterloo University, Canada

Rachel Summers, MSc., IARO, Palmerston North, New Zealand

Acknowledgements

The authors of this report would like to acknowledge the longstanding assistance of Dr Bruce Rapley of Sound Analytics. The authors would also like to acknowledge the many insights provided by Les Huson of L Huson & Associates and the vast experience in acoustics made available by Dr Philip Dickinson, Senior Researcher at IARO.

Review of Tartu Univeristy Report, Estonia

Page 3 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

CONTENTS

EXECUTIVE SUMMARY 4

A. INTRODUCTION 5 I. Background 5 II. Goal 6 III. Disclaimer 6 IV. International Acoustics Research Organization, IARO 7 V. Acronyms and Variables Used in IARO Reports 7

B. ORGANIZATION OF THIS REPORT 8 I. Sequential approach to various aspects 8 II. To the Authors of the TU Report 8

C. PURPOSE AND RESEARCH QUESTIONS UNDERLYING THE TU REPORT 10 I. Purpose of the TU Report Study 10 II. Research Questions 11

D. WHAT IS A NOISE-INDUCED HEALTH EFFECT? 15

E. WHY CURRENT NOISE MEASURING METHODOLOGIES ARE NON-APPLICABLE FOR WIND TURBINE NOISE 19

F. ‘WHAT YOU CAN’T HEAR, CAN’T HURT YOU’ 22

G. LUXURIES NOT AFFORDED TO SCIENTISTS—A GLIMPSE OF THE TEDIOUS WORK REQUIRED TO UPHOLD SCIENTIFIC RIGOUR. 24

H. A CANDID CONVERSATION AMONG SCIENTISTS 29 I. The ‘nocebo effect’ narrative 29 II. The questionnaire approach 32 III. Another ‘Scientific Authorship’ of another “Wind Turbine Health Impact Study”… 33

I. CONCLUSIONS 35

ANNEX A: English Translation of the TU report ANNEX B: Critical Review of Marshall et al. study (2023) ANNEX C: Critical Review of Maijala et al. study (2020) ANNEX D: Response to Massachusetts Independent Expert Panel (2012)

Review of Tartu Univeristy Report, Estonia

Page 4 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

EXECUTIVE SUMMARY

1. On 6 October, 2025, IARO scientists were contacted by Citizens Initiative Estonia [a non- -profit organization], with the request to provide an assessment of a report commissioned by the Ministry of Environment of Estonia, and produced by the University of Tartu (TU Report).

2. This IARO Critical Analysis Report is not an “oppositional document” to the TU Report. Rather, it has been prepared as a pedagogical document.

3. It is hoped that the authors of the TU Report, whom we view as fellow scientists, take this IARO Critical Analysis Report as an educational tool, contributing to their “Review Study Phase I,” rather than some gratuitous “attack document.”

4. The study documented by the TU Report has clearly been conducted properly in terms of how an analysis of published papers and reports should be undertaken, when those participating are not experts in the subject matter.

5. IARO Scientists have the distinct impression that these Estonian authors were pre- conditioned to believe that wind power plant sound emissions have no effect on public health.

6. This is further justified by the Recommendations made which are unfounded, skewed from reality, and not “evidence-based,” as promised by the authors of the TU Report.

7. Given the content of the Recommendations proffered by the TU Report, it seems probable that the Authorship of the TU Report has unwittingly succumbed to the unscientific practices promoted by governments and international special interest groups.

8. In the opinion of IARO Scientists, this study can only be regarded as, yet another, artificially constrained review of papers, with outcomes predetermined by politically generated questions, resulting in a report of low scientific standard.

Review of Tartu Univeristy Report, Estonia

Page 5 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

A. INTRODUCTION

I. Background

9. On 6 October, 2025, IARO scientists were contacted by Citizens Initiative Estonia [a non- -profit organization], and were requested to provide an assessment of a report commissioned by the Ministry of Environment of Estonia and produced by the University of Tartu (henceforth referred to as the TU Report).

10. For this purpose, IARO scientists received an English translation of the TU Report, included in this IARO Report as Annex A. References to page numbers of the TU Report correspond to those in this English version, provided in Annex A.

11. The TU Report states that it is related to Phase I of a Review Study titled: “Health effects of wind turbines: A systematic review of studies published in peer-reviewed scientific journals over the last fifteen years.” Within this context, the study of the TU Report is, more specifically, titled: Development of a methodology for interpreting the results of scientific studies on the potential health effects of wind farms and other energy production technologies in the Estonian context.

12. Figure 1 shows the Purpose of this Study as stated in the TU Report.

Review of Tartu Univeristy Report, Estonia

Page 6 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

Figure 1. Description of the Purpose of the Study of the TU Report and its Research Questions (p. 12)

II. Goal

13. To provide a scientific review of the TU Report, within the context of The Scientific Method, Evidence-based Medicine and Critical Analysis.

III. Disclaimer

a. The report provided herein has one, and only one, agenda; that of pure scientific inquiry.

b. The authors of this report are not party to anti-technology sentiments and do not harbour anti-wind-energy sentiments.

Review of Tartu Univeristy Report, Estonia

Page 7 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

c. In no way can or should this scientific review be construed as a document arguing for or against the implementation of wind power plants, or any other type of infrastructure or industrial complexes that generate acoustic pollution.

d. IARO members and authors of this report hold no financial interest in the SAM Technology.

IV. International Acoustics Research Organization, IARO

14. The International Acoustics Research Organization represents a group of scientists who, collectively, hold over 300 years of scientific experience in the field of infrasound and low frequency noise, and its effects of human health. Since 2016, IARO researchers have been recording and analysing acoustical data in and near homes located in the vicinity of onshore wind power plants, in the following countries (alphabetical): Australia, Canada, Denmark, England, France, Germany, Ireland, New Zealand, Northern Ireland, Portugal, Scotland, Slovenia, and The Netherlands. Prior to 2016, all IARO scientists were already working either in acoustics alone or in acoustics and health. All research conducted by IARO is part of the Citizen Science Initiative for Acoustic Characterization of Human Environments (CSI-ACHE).

V. Acronyms and Variables Used in IARO Reports

15. Table 1 lists the acronyms and variables used in IARO Reports.

Table 1. Acronyms and Variables that may appear in IARO Reports

dB Decibel unweighted (measure of sound pressure level) dBA Decibel A-weighted (measure of sound pressure level) dBC Decibel C-weighted (measure of sound pressure level) dBG Decibel G-weighted (measure of sound pressure level) Hz Hertz (units for measure of frequency)

ILFN Infrasound and Low Frequency Noise (≤200 Hz) IWT Industrial Wind Turbine LFN Low frequency noise (20-200 Hz) SPL Sound Pressure Level

WHO World Health Organization WPP Wind Power Plant

WTAS Wind Turbine Acoustic Signature

Review of Tartu Univeristy Report, Estonia

Page 8 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

B. ORGANIZATION OF THIS REPORT

I. Sequential approach to various aspects

16. Given the Authorship of the TU Report, the tone of this IARO Report is meant to be educational and not oppositional—IARO Scientists consider they are addressing fellow scientists.

17. The “Stated Purpose of the Study” and the “Research Questions” will be discussed first, in Section C.

18. A brief, science-based, educational approach is provided regarding ‘health effects,’ using annoyance as an example, in Section D.

19. A brief, science-based, educational approach is provided regarding the use of the A- weighting filter, and its appropriateness for measuring ‘wind turbine noise’ in Section E.

20. Section F demonstrates the fallacy of the notion ‘what you can’t hear, can’t hurt you,’ which wholly biases the TU Report.

21. Section G examines if Scientists have the luxury of accepting conclusions of meta-analyses or systematic reviews at face-value.

22. Section H discusses three topics that the Authors of the TU Report may find important for their own knowledge base.

23. Section I documents the Conclusions of this IARO Critical Analysis Report.

II. To the Authorship of the TU Report

24. With this Critical Analysis of the TU Report, in no way do IARO Scientists wish to offend or insult the authors of the TU Report, who are considered to be fellow scientists.

25. It is clear that a genuine effort has been made, within the context of systematic reviews, to adequately select published scientific papers, under the self-imposed exclusion criteria.

26. It has also become clear, however, that the Authors of the TU Report are unfamiliar with the deep complexities and intricacies of this particular subject, both in terms of acoustics and of biological sciences—This is entirely understandable, but errors (especially those arising from unfamiliarity with a particular subject) must be raised where they are made!

Review of Tartu Univeristy Report, Estonia

Page 9 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

27. As mentioned above (Parag. 14), IARO Scientists have been individually dedicated to studying the health effects caused by infrasound and low frequency noise for many decades, and from many different perspectives (biological, clinical, signal analysis, instrumentation, occupational and environmental settings, animal exposures, among others).

28. IARO Scientists hope that the authors of the TU Report view this IARO Report as an educational tool, contributing to their “Review Study Phase I,” rather than some gratuitous “attack document.”

29. Please see Section H: A Candid Conversation among Scientists.

Review of Tartu Univeristy Report, Estonia

Page 10 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

C. PURPOSE AND RESEARCH QUESTIONS UNDERLYING THE TU REPORT

I. Purpose of the TU Report Study

The aim of the study was to systematically analyse the evidence published in the scientific literature over the last fifteen years (2010–2025) on the health effects of wind turbines (p.12). (See Fig. 1)

30. While it is understood what is meant, this purpose is very badly worded, given the scientific credentials of the TU Report’s Authorship.

31. Medical Sciences classifies agents of disease into 4 categories: biological, chemical, physical and psychosocial.

32. In which category, then, would “wind turbines” be inserted, since they are allegedly producing health effects? The wind turbines do not cause health issues; the emissions from wind turbines may cause health issues.

33. IARO scientists would suggest the following re-wordings for scientific accuracy:

“…on the health effects associated with the proximity of wind turbines to residential areas,”

Or

“…on the health effects claimed to be associated with wind turbine emissions,”

Or

“The aim of the study was to systematically analyse the evidence published in the scientific literature over the last fifteen years (2010–2025) on the purported health effects due to wind power plant operations.”

34. This issue is not a trivial matter, as it may seem to some.

35. Instead, it reflects a deep misunderstanding of the matter at hand pertaining to the fundamental principles of Medical Sciences.

36. After all, the foremost concern here is the health of Estonian Citizens, is it not?

Review of Tartu Univeristy Report, Estonia

Page 11 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

II. Research Questions

37. Throughout the Research Questions, it is apparent that wind turbines are (erroneously) interpreted as an agent of disease.

38. This greatly curtails the expansion of questions into a more scientific realm. As all scientists are aware, asking the right question is of fundamental importance.1

Question 1: What are the main conclusions of existing studies on the health effects of wind turbines? (p.12, see Fig. 1)

39. This seems like an innocuous and purposeful question, but a closer inspection already reveals bias: is it presumed that the “health effects of wind turbines” are specific and exclusive to wind turbines—they are not!

40. Other industries can have similar emissions that bring about the same “health effects” as those allegedly developed by residents neighbouring wind power plants.

41. The agent of disease is not the wind turbine but its various emissions and, yes, one of those emissions is acoustical in nature.

42. Again, to the uniformed this may seem a trivial point, more related to semantics—It is not!

43. Imagine the following question:

What are the main conclusions of existing studies on the health effects of automobiles?

44. Is this a question that, taken alone, makes any sense?

Question 2: What is the overall quality of the existing evidence? Is there evidence in the scientific literature that wind turbines have a negative impact on human health? (p.12)

45. The “overall quality of existing evidence” is evaluated by reading the Methodology Section of each and every selected paper to ascertain if the conclusions reached are supported by the methodology used (see Section G).

46. Are the Authors of the TU Report qualified to evaluate whether the methodologies imposed by law to “measure noise” are fit-for-purpose when human health is a concern?

1 Back in the late 1800’s, the question was posed: “Is Light a particle OR a wave.” This question reduced physical

reality to a dichotomy, not open to the possibility that Light can be BOTH. Hence the fundamental importance for proper Scientists to ask the pertinent and insightful questions.

Review of Tartu Univeristy Report, Estonia

Page 12 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

47. Or have the Authors of the TU Report, instead, blindly relied upon the “noise measuring” methodologies as per legislated stipulations?

48. Many authors, unfamiliar with the matter at hand, do rely upon legislated methodologies.

49. However, given the stated “Purpose of the Study” and the scientific background of the authorship, can the Authors of the TU Report be afforded this luxury? (see Section E)

50. On the other hand, when the selected paper is referring to the evaluation of health endpoints, do the Authors of the TU Report have the expertise in Medical and Clinical Sciences to evaluate whether or not the selected health endpoint is pertinent and relevant? (See Section D)

51. The same can be pointed out regarding the second part of Question 2, “Is there evidence in the scientific literature that wind turbines have a negative impact on human health?” Whether there is or not, is the Authorship of the TU Report qualified to critically analyse the methods applied in these studies? (see Sections D and E)

Question 3: If wind turbines have negative health effects, what health effects are associated with wind turbines? (p.12)

52. This question trickles down from the prior questions. Again, it is not “wind turbines” that have negative health effects (unless the wind turbines themselves are becoming sick), but emissions from wind turbines that can act as agents of disease upon biological organisms.

53. Nevertheless, it is understood that the object of this question is to determine what kind of health effects have been documented as related to living in the proximity of wind power plants.

54. Do the Authors of the TU Report have the necessary expertise to evaluate the robustness of the methodologies used in papers that report health endpoints as related to residential proximity to wind power plants? (See Sections D, G and H-I)

Question 4: If wind turbines have negative health effects, what role do environmental factors such as noise, infrasound, shadow flicker, visual aspects, psychological factors (including general attitudes towards wind turbines, people's beliefs and perceptions of wind turbines), vibration and electromagnetic fields in causing these health effects? (p.12)

55. Let us dissect this question: “[W]hat role do environmental factors such as

noise, infrasound—a potential acoustical physical agent of disease,

shadow flicker—a potential optical physical agent of disease,

visual aspects—a potential optical physical and/or psychosocial agent of disease,

Review of Tartu Univeristy Report, Estonia

Page 13 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

psychological factors—a potential psychosocial agent of disease,

vibration—a potential vibratory physical agent of disease and

electromagnetic fields—a potential electromagnetic physical agent of disease.

56. Perhaps, laid out like this, the Authors of the TU Report might realize why this question is entirely inappropriate…unless it is broken up into 6 distinct questions, each warranting its own independent study and (very) complex evaluation.

57. For example, “shadow flicker” is a term that only appeared after the advent of wind energy—before, it was called the stroboscopic effect.

58. Therefore, as the Authors of the TU Report would certainly agree, a proper investigation into “shadow flicker” must include prior studies (at least a glimpse into them) on the stroboscopic effects on humans (for example, such as those related to military helicopter pilots). Similar prior studies would be needed for each of the other environmental factors.

Question 5: If wind turbines have health effects, under what conditions are these health effects more likely to occur (e.g. at what distance from the turbine, with powerful or tall turbines, etc.)? (p.12)

59. Again, the wording of this question does not do justice to the scientific credentials of the TU Report’s Authors. While it is understood what is being asked here, its formulation is most unscientific.

60. Suggested rewording of Question 5:

Question 5 (suggested rewording):

If it can be demonstrated that “health effects” develop in residents neighbouring wind power plants, what external physical conditions (e.g. distance to turbine(s), type and specifications of turbine(s), etc.) become significant factors for the onset and/or development of these “health effects”?

61. It is hoped that this rewording is self-explanatory.

Question 6: Are certain population groups more vulnerable to the potential health effects of wind turbines? (p.12)

62. This is a very interesting question to have at such an initial stage of the study.

63. Is it intended to point out that population groups known to be vulnerable, such as the elderly, the chronically ill, infants and children, and pregnant women, should be

Review of Tartu Univeristy Report, Estonia

Page 14 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

approached as separate populations to determine if they are (also) more vulnerable to “health effects of wind turbines,” if they exist?2

64. Given the vast experience of IARO Scientists, it seems that this question is most likely based on the prior supposition that some people ‘are more sensitive’ than others to the “health effects of wind turbines.” (See Section H-II).

Question 7: What evidence-based recommendations can be made to policymakers, industry stakeholders and affected communities to protect human health? (p.12)

65. This final question is, in and of itself, quite unscientific. “Evidence-based recommendations”—Are there any other type?

66. And yet, having read the Recommendations of the TU Report (and having pointed out their failings), it is now realized that, indeed, non-evidence-based Recommendations are, regrettably still (see Section H-III) a possibility from authors with significant Scientific Credentials.

2 If the TU Report were to include animal studies, then perhaps this question could refer to different types of animal

populations. Cows, sheep, rabbits and mink all react very differently when in the vicinity of wind power plants. Perhaps some are more vulnerable?

Review of Tartu Univeristy Report, Estonia

Page 15 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

D. WHAT IS A NOISE-INDUCED HEALTH EFFECT?

67. As stated by the World Health Organization:

An adverse effect of noise is defined as a change in the morphology and physiology of an organism that results in impairment of functional capacity, or an impairment of capacity to compensate for additional stress, or increases the susceptibility of an organism to the harmful effects of other environmental influences. 3

68. ‘Annoyance’ is commonly (yet erroneously) considered as a “health effect.”

69. Dutifully, the TU Report covers this subject. Here are some examples:

Several of the studies included in this review (Appendix 2, Table 4) investigated the extent to which one specific characteristic of wind turbine noise, amplitude modulation (AM), contributes to annoyance (Ioannidou et al., 2016; Lee et al., 2011; Schaffer et al., 2016, 2018). In addition, these studies also examined the effect of noise frequency distribution and source origin on annoyance. (p. 21)

A review article (McCunney et al., 2014) also concluded that wind turbine noise plays only a minor role in causing annoyance compared to other factors that influence people's willingness to experience annoyance in relation to wind turbines. Pohl et al. (2018) also found that noise-related annoyance was influenced to a small extent by the distance to the nearest wind turbine and the intensity of the sound, but was most influenced by the extent to which people felt that the wind turbine planning process had been conducted fairly and transparently. (p. 33)

In summary, the relationship between wind turbines and disturbance depends on several factors, such as expectations/knowledge of the health effects of wind turbines, perceived fairness and transparency of the planning process, economic benefits, visual aspects and noise. It is likely that a combination of all these factors causes annoyance, and reducing just one factor (e.g. noise) may not reduce annoyance. (p. 36)

70. Annoyance is also included in the Recommendations Section of the TU Report:

3 World Health Organization. (1999) Guidelines for community noise. Stockholm University & Karolinska Institute:

Stockholm, Sweden. pp. 21. https://www.who.int/publications/i/item/a68672

Review of Tartu Univeristy Report, Estonia

Page 16 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

We recommend that developers and researchers explore ways to reduce AM depth in order to reduce the annoyance of wind turbine noise (p.46). [AM = Amplitude Modulation]

71. Are the Authors of the TU Report acquainted with the formal definition of annoyance?

72. In the 2017 edition of Mosby’s Medical Dictionary,4 there were zero entries for the word ‘annoyance.’

73. In the 2018 edition of the Medical Dictionary published by the British Medical Association,5 there were also no instances of the word ‘annoyance.’

74. In the 2020 edition of the Oxford Medical Dictionary,6 one single entry is found for this word:

Glare n. the undesirable effects of scattered stray light on the retina, causing reduced contrast and visual performance as well as annoyance and discomfort.

75. Within the context of noise nuisance, perhaps the best definition for ‘annoyance’ is (still) the one given in 2000 by the European Commission Noise Team:

Annoyance is the scientific expression for the non-specific disturbance by noise, as reported in a structured field survey. Nearly every person that reports to be annoyed by noise in and around its home will also experience one or more of the following specific effects: Reduced enjoyment of balcony or garden; When inside the home with windows open: interference with sleep, communication, reading, watching television, listening to music and radio; Closing of bedroom windows in order to avoid sleep disturbance. Some of the persons that are annoyed by noise also experience one or more of the following effects: Sleep disturbance when windows and doors are closed; Interference with communication and other indoor activities when windows and doors are closed; Mental health effects; Noise-induced hearing impairment; Hypertension; Ischemic heart disease.7

4 O’Toole MT et al. (Eds). (2017) Mosby’s Medical Dictionary. 10th Ed. Elsevier: St Louis, MI, USA.

5 British Medical Association. (2018) Medical Dictionary. 4th Edition. Dorling Kindersley: London, UK.

6 Martin E, Law J. (Eds) (2020) Concise Colour Medical Dictionary. 7th Ed. Oxford University Press: Oxford, UK.

7 European Commission. (2000) The Noise Policy of the European Union—Year 2. Towards improving the urban environment and contributing to global sustainability. European Commission Noise Team: Luxembourg. https://www.europeansources.info/record/the-noise-policy-of-the-european-union-year-2-1999-2000-towards- improving-the-urban-environment-and-contributing-to-global-sustainability/

Review of Tartu Univeristy Report, Estonia

Page 17 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

76. This comprehensive definition of ‘annoyance’ clearly establishes it as a legitimate measure to be used within the realm of Psychoacoustic studies.

77. But it is far from being an appropriate health endpoint within the context of an “adverse effect of noise,” as defined by the World Health Organization (see Parag. 67).

78. Do the Authors of the TU Report have the necessary expertise to identify this issue, or will it be ‘business as usual’?8

79. For the edification of these Authors, in papers that have been excluded from their selection, annoyance has been linked to morphological changes in the auditory cilia and some medical professionals view self-reported ‘noise annoyance’ in their patients as a symptom of excessive prior noise exposure. (See Section H-II)

80. For the further edification TU Report’s Authors: The International Classification of Diseases (ICD-11), published by the World Health Organization, has specific codes for infrasound- induced vertigo—NF08.2Y (see Figure 2).

8 The exclusion criteria should have included all papers that have used ‘annoyance’ as a bona fide health endpoint.

Review of Tartu Univeristy Report, Estonia

Page 18 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

B

Figure 2. Results of the search for “infrasound” in the WHO ICD-11 (coding tool option). 9 (A) One instance of infrasound appears—Code NF08.2Y, covering infrasound-induced vertigo, under the heading of “other specified effects of vibration.” (B) Index terms covered by this Code differentiate between infrasound- and vibration-induced vertigo.10

9 World Health Organization. (2024) International Classification of Diseases-11 (ICD-11).

https://icd.who.int/ct/icd11_mms/en/release

10 World Health Organization. (2024) International Classification of Diseases-11 (ICD-11). https://icd.who.int/browse/2024-01/mms/en#621374492%2Fother

Review of Tartu Univeristy Report, Estonia

Page 19 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

E. WHY CURRENT NOISE MEASURING METHODOLOGIES ARE NON-APPLICABLE FOR WIND

TURBINE NOISE

81. In its Recommendations section, the TU Report states the following:

For living and sleeping areas, we recommend setting a limit for wind turbine noise of 30 dB(A) during the day and 25 dB(A) at night, similar to the existing limits for traffic noise and noise from technical equipment. (p. 48).

82. Presumably, then, a value of 28 dBA would, more or less, comply with this recommendation.

83. Which value of 28 dBA would the Authors of the TU Report consider acceptable in the following field-data situation, shown in Fig. 3:

84. The 28 dBA in Fig. 3A or the 28 dBA in Fig. 3B?

A

B

Figure 3. A: 28 dBA (red bars) and 89 dB (pink bars). B: 28 dBA (red bars) and 47 dB (pink bars).11

11 Data from urban field measurements (no wind turbines), published in a paper that was excluded from the TU Report

selection of papers. Pereira-Sousa P, Alves-Pereira M, Bakker H. (2025). Dose-Response Relationship in Occupational Noise Exposures: The Distorted Quantification of Dose that Misinforms the Medical Community. SHO 2025 – International Symposium on Occupational Safety and Hygiene. Proceedings Book. DOI: https://doi.org/10.24840/978-989-54863-7-3_0125-0132

Review of Tartu Univeristy Report, Estonia

Page 20 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

85. Do the Authors of the TU Report understand that noise measured in dBA cannot differentiate between these two, significantly different, acoustic environments?

86. Hence, the recommendation transcribed in Para. 81 is entirely skewed from the matter at hand.

87. This type of information was known by the World Health Organization in 1999:

A noise measure based only on energy summation and expressed as the conventional equivalent measure, LAeq, is not enough to characterize most noise environments. It is equally important to measure the maximum values of noise fluctuations, preferably combined with a measure of the number of noise events. If the noise includes a large proportion of low-frequency components, still lower values than the guideline values below will be needed. When prominent low-frequency components are present, noise measures based on A-weighting are inappropriate.12 [Emphasis added.]

88. In the TU Report, Fig. 1 is a very informative graph showing the frequency response curves of the different frequency-weighting filters that are imposed on noise measurements by legislated stipulations. This graph is reproduced here in Fig. 4.

Figure 4. Frequency response curves for A, C and G the frequency-weighting filters and for the absence of filter, Z. (p. 8, TU Report)

12 World Health Organization. (1999) Guidelines for community noise. Stockholm University & Karolinska Institute:

Stockholm, Sweden. pp. xiii. https://www.who.int/publications/i/item/a68672.

Review of Tartu Univeristy Report, Estonia

Page 21 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

89. Do the Authors of the TU Report realize that the application of any of these different filters (A, C and G) means that profound assumptions are being made, namely:

1) what you can’t hear can’t hurt you (see Section F), and

2) annoyance is a bona fide health endpoint (see Section D)?

3) noise only affects humans via the auditory pathway (see Section F).

90. Do the Authors of the TU Report realize that, for the purposes of the matter at hand, the act of “measuring noise” constitutes the quantification of the dose of the agent of disease?

91. Do the Authors of the TU Report understand that the Y axis of their Figure 1 indicates that the application of frequency weighting filters means that the measurements no longer reflect physical reality?

92. Does this begin to explain why legislated methodologies are scientifically irrelevant for measuring the types of environments where noise has significant lower frequency components, such as those generated by wind power plants?

93. Does this also suggest why a high-quality scientific investigation should ignore legislated methodologies in favour of evidence-based methodologies?

94. Real, scientific-grade information on the medical dose of noise is not obtained, if legislated procedures are applied, i.e., the mandatory use of A, C or G frequency- weighting filters. (See Fig. 3)

Review of Tartu Univeristy Report, Estonia

Page 22 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

F. ‘WHAT YOU CAN’T HEAR, CAN’T HURT YOU’

95. Is this what the Authors of the TU Report have been told? What you can’t hear, can’t hurt you?

96. The significant difference between the two 28-dBA environments shown in Fig. 3 will be summarily dismissed by those who believe this fallacy.

97. The Authors of the TU Report will be told that the real physical presence of the 47 and 89 dB difference (i.e., no filter is applied) is irrelevant for human health because it is occurring below the human auditory threshold.

98. Will these Authors, then, also believe that only environmental factors that can be readily perceived by all people are relevant for consideration in human health? In the same way that radioactivity is (not) readily perceived or carcinogenic chemicals are (not) readily perceived?

99. See Fig. 5, which shows an abstract of a paper from 1978 (!)13

Figure 5. Busnel RG, Lehmann AG (1978). Infrasound and sound: Differentiation of their psychophysiological effects through use of genetically deaf animals. Journal of the Acoustical Society of America14 (see text).

13 Certainly, way beyond the scope of the TU Report’s systematic review…being from 1978 and because it involves

animals.

14 Busnel RG, Lehmann AG (1978). Infrasound and sound: Differentiation of their psychophysiological effects through use of genetically deaf animals. Journal of the Acoustical Society of America, 63(3): 974-977. https://pubmed.ncbi.nlm.nih.gov/670562/

Review of Tartu Univeristy Report, Estonia

Page 23 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

100. In this 1978 study, genetically-deaf mice were used as study subjects, and infrasound had a deleterious effect on their performance. What you can’t hear can’t hurt you?

101. How, then, to explain the more recent scientific results shown in Figs. 6A and 6B, where an acoustic phenomenon, presumed to be inaudible to humans (below 20 Hz) was able to distress the residents during a sleepless night, to the point of compelling them to take medication?15

A

B

Figure 6.

A: Residents near wind power plants slept peacefully— 26 dBA and 67.3 dB, B: Same residents could not sleep and needed medication—26.5 dBA and 69.9 dB16

102. What you can’t hear, can’t hurt you…doesn’t really work very well, does it? Not for mice in 1978, nor for humans in 2023.

103. For the edification of the Authors of the TU Report, the sequence of peaks seen in Fig. 6B is called a wind turbine acoustic signature. Mathematically, it is a harmonic series whose fundamental frequency corresponds to the blade pass frequency of the corresponding wind turbine (see Fig. 7B in Section G).

104. Wind turbine acoustic signatures become invisible when legislated noise measuring methodologies are applied.

15 The residents, authors of the diary providing this information, were not privy to any acoustical information that was

being simultaneously recorded. Data presented here in Figure 6 are the result of post-processing analysis. Please see Footnote 16 for the full, peer-reviewed report on this case.

16 This paper was excluded from the selection of papers considered by the TU Report, as it is a Case Report. Bakker HHC, Alves-Pereira M, Mann R, Summers R, Dickinson P. (2023) Infrasound exposure: High resolution measurements near wind power plants. In: Suhanek M, Kevin Summers J. (Eds) Management of Noise Pollution. IntechOpen: London. DOI: 10.5772/intechopen.109047

Review of Tartu Univeristy Report, Estonia

Page 24 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

G. LUXURIES NOT AFFORDED TO SCIENTISTS—A GLIMPSE OF THE TEDIOUS WORK REQUIRED TO

UPHOLD SCIENTIFIC RIGOUR.

105. Scientists do not have the luxury of taking the conclusions of meta-analyses (or systematic reviews or literature reviews) of pre-existing papers, for granted or at face-value.

106. In contrast to laypersons, policy- and decision-makers, industry stakeholders and the general public, Scientists must evaluate the methodology of each and every paper included in a review.

107. A tedious exercise for sure, but a necessary one if scientific rigour is to be upheld.

108. How else can one scientifically vouch for the conclusions offered by the author of the meta-analysis, systematic review or literature review?

109. As a demonstrative exercise, let us explore an 11-year-old paper, quoted several times in the TU Report:

Basner, M., Babisch, W., Davis, A., Brink, M., Clark, C., Janssen, S., Stansfeld, S., 2014. Auditory and non-auditory effects of noise on health. Lancet 383, 1325– 1332.

110. This same reference justified the following statements, made by the Authors of the TU Report on page 7:

A decibel indicates how much louder the sound is than the reference value. In air, the reference value is an air pressure of 20 micropascals (20 μPa or 2×10−5 Pa), which is considered to be the human hearing threshold at a frequency of 1000 Hz – this is the quietest sound that the average person can still hear at this frequency. (p.7)

111. And on page 32:

Disturbance can also act as a mediating factor between other health effects, including influencing the development of more serious conditions such as cardiovascular disease through stress (p. 32)

112. And, under the heading “Audible noise [sic] generated by wind turbines and clinically manifested health effects,” (p. 40), the TU Report makes another statement justified by this same, 2014 reference:

Review of Tartu Univeristy Report, Estonia

Page 25 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

Disturbance and sleep disturbances caused by audible noise may contribute to the development of diagnosable diseases (p. 40). 17

113. Returning to the original 2014 reference, it states:

In this Review, we summarise knowledge and research related to noise exposure and both auditory and non-auditory health effects. (…)

These noise exposures have been linked to a range of non-auditory health effects including annoyance (Miedema & Oudshoorn, 200118), sleep disturbance (Muzet, 200719), cardiovascular disease (van Kempen & Babisch, 201220; Sorensen et al., 2012 21 ) and impairment of cognitive performance in children (Stansfeld & Matheson, 2003 22 ). 23 [The original numbered references were replaced with citations.]

114. Scientific ‘work’ involves reading each of these 5 references that are quoted in this 2014 review if and only if scientific rigour is to be maintained.

115. (It should be recalled that scientific rigour is not necessarily in the purview of laypersons, policy- and decision-makers, industry stakeholders and the general public.)

116. Just by reading the titles of these 5 papers we see that one is a meta-analysis, which eliminates it from this immediate consideration.

117. Let us look into the other four.

17 This last assertion is a truism (at least since the times of Ancient Rome) as it is referring to audible noise! Strictly

speaking, no reference would have been needed.

18 Miedema HME, Oudshoorn CGM. Annoyance from transportation noise: relationships with exposure metrics DNL and DENL and their confidence intervals. Environ Health Perspect. 2001; 109:409–16. https://pubmed.ncbi.nlm.nih.gov/11335190/.

19 Muzet A. Environmental noise, sleep and health. Sleep Med Rev. 2007; 11:135–42. https://pubmed.ncbi.nlm.nih.gov/17317241/

20 van Kempen E, Babisch W. The quantitative relationship between road traffic noise and hypertension: a meta- analysis. J Hypertens. 2012; 30:1075–86. https://pubmed.ncbi.nlm.nih.gov/22473017/

21 Sørensen M, Andersen ZJ, Nordsborg RB, et al. Road traffic noise and incident myocardial infarction: a prospective cohort study. PLoS One. 2012; 7:e39283. https://pubmed.ncbi.nlm.nih.gov/22745727/

22 Stansfeld SA, Matheson MP. Noise pollution: non-auditory effects on health. Br Med Bull. 2003; 68:243–57. https://pubmed.ncbi.nlm.nih.gov/14757721/

23 Basner M, Babisch W, Davis A, Brink M, Clark C, Janssen S, Stansfeld S. (2014) Auditory and non-auditory effects of noise on health. Lancet 383: 1325–1332. https://pubmed.ncbi.nlm.nih.gov/24183105/

Review of Tartu Univeristy Report, Estonia

Page 26 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

118. Miadema & Oudshoorn (2001): “Here we model the distribution of annoyance responses as a function of the noise exposure” for road, rail and air traffic noise. “Day–night level (DNL) and day–evening–night level (DENL) were used as noise descriptors.”

119. Wind turbine noise is not considered in the Miadema & Oudshoorn paper.

120. Annoyance, which Miadema & Oudshoorn used as a health endpoint, is not a bona fide health outcome (see Section D).

121. The noise parameters used to characterize “noise exposure” are inconsequential for the matter at hand (see Section E).24

122. Muzet (2007) is classified as a “Clinical Review” and uses sleep as a measure of a health— a bona fide health endpoint.

123. However, noise environments of the papers used by Muzet in his Clinical Review are still characterized in dBA, and wind turbine noise is not considered—this excludes any real scientific relevance to the matter at hand.

124. Sorensen et al. (2012), not a review paper, and a very scientifically robust health endpoint was chosen—ischemic heart disease (see Section D).

125. Sorensen et al. (2012) stated: ”Exposure to long-term residential road traffic noise was associated with a higher risk for MI, in a dose-dependent manner.” [MI=Myocardial Infarction, i.e., ischemic heart disease.]

126. As with the Miadema & Oudshoorn study, the ‘day–evening–night level,’ or Lden, was used to quantify the noise environment (see Section E and footnote 24), and wind turbine noise was not considered.

127. Stansfeld & Matheson (2003), yet another review, based on 86 references…

128. The tediousness of this exercise is an integral part of the Scientific process.

129. To finalize this Section, a last assertion is transcribed from the Introduction of the TU Report:

However, reviews of wind turbine noise conducted to date have not confirmed a link between wind turbine noise and clinically apparent health effects (Karasmanaki, 2022; Schmidt and Klokker, 2014; Teneler and Hassoy, 2023; van Kamp and van den Berg, 2021, 2018). (p. 4)

24 Although not explicitly indicated, the use of the Lden or DENL noise parameter implies the application of the A-

frequency-weighting filter (see Section E, Figs. 3 and 4).

Review of Tartu Univeristy Report, Estonia

Page 27 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

130. These five references, offered as (evidence-based?) justifications for this assertion, are all meta-analysis, systematic reviews or literature reviews.

131. Does this mean that the Authors of the TU Report vouch for the position taken by the author(s) of each one of these five reviews, and therefore guarantees that all the papers cited in reviews uphold the author(s)’ position? Can they vouch for the methodologies of all the papers when they have not seen them?

132. Of course not!

133. However, unlike others, Scientists, do not have the luxury of merely depending on the conclusions reached by the authors of these types of review papers, because they could include papers of dubious scientific integrity.

134. This is a part of what the scientific process is all about, is it not?

135. On page 26 of the TU Report, the following is stated:

The study concluded that wind turbine infrasound does not disturb people's sleep, does not cause symptoms of 'wind turbine syndrome', does not impair measured cardiovascular health indicators, and does not impair people's mental well-being (Marshall et al., 2023). The results of the study can be considered well- proven. (p.26)

136. Annex B provides a critical analysis of the Marshall et al. (2023) paper prepared by IARO Scientists in 2024.

137. On page 33 of the TU Report, the following is stated:

An experiment conducted in Finland showed that the audible sounds of a wind farm were more disturbing than the sounds of the ocean (Maijala et al., 2021).

138. Would the Authors of the TU Report care to know the scientific reason for why this is so?

139. Here is a comparison between ocean noise and wind turbine noise, as measured without the methodologies imposed by legislation:

Review of Tartu Univeristy Report, Estonia

Page 28 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

A

B

Figure 7. Characterization of acoustic environments (i.e., noise measurements) without the legislated-imposed methodologies. A. Beach, Rømo Island, Denmark, 13 Dec 2016 at 01:10H. B. Wind turbines acoustic signature, present in the acoustic environment corresponding to the night when residents could not sleep and were compelled to take medication—See Fig 3B in Section F.

140. All this important information becomes invisible when legislated methodologies are imposed.

141. Annex C provides a critical analysis of the Maijala et al. paper prepared by IARO Scientists in 2024.

Review of Tartu Univeristy Report, Estonia

Page 29 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

H. A CANDID CONVERSATION AMONG SCIENTISTS

142. Since the Authors of the TU Report are considered by IARO as fellow-scientists, an uncommon decision has been taken to speak directly to these Authors through this Critical Analysis Report.

143. This was deemed all the more appropriate since IARO Scientists have been informed that this Team of Estonian Scientists will proceed with more studies to monitor the development health effects among residents neighbouring wind power plants.

144. If this TU Report is any indication of the avenues of research that will be followed (particularly given its appalling Recommendations), then IARO’s position is simple—what a waste of time, money and brainpower!

I. The ‘nocebo effect’ narrative

145. As part of the Recommendations, the TU Report states:

The results of our study show that several factors other than wind turbine noise affect disturbance, and that noise reduction alone may not be sufficient to mitigate disturbance. Just as important as noise restrictions in preventing disturbance may be informing residents about the nocebo effect, the absence of negative expectations regarding the health effects of wind turbines, and understanding the positive characteristics of wind turbines (Crichton et al., 2015, 2014b, 2014a; Crichton and Petrie, 2015b, 2015a; Tonin et al., 2016).

146. By advocating this ‘nocebo effect narrative,’ the Authorship of the TU Report is taking a position that is absolutely indefensible in terms of Science.

147. The Authors of the TU Report, as Scientists, should be aware that a nocebo effect cannot be proven, as it is impossible to eliminate all environmental factors that might be a cause but are unmeasured.25 The Authors should be asking, where is the evidence for a nocebo effect?

148. Given the scientific credentials of the Authors of the TU Report, they should, instead, be inquiring into the studies that justified attributing the label of ‘nocebo effect’ to the collection of symptoms, self-reported people by all over the world.

25 The nocebo effect can never be proved, it can only fail to be disproved.

Review of Tartu Univeristy Report, Estonia

Page 30 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

149. This collection of symptoms is not specific to people who live in the proximity of wind power plants, but it is specific to people who live in infrasound-contaminated homes (whatever the source).

150. Are all these cases supposed to be the result of some collective psychosomatic disorder?

151. For the edification of these Estonian Scientists, the ‘nocebo effect’ is, in Clinical Medicine, considered to be of psychosomatic origin (or aetiology), falling under the category of pathology caused by psychosocial agents of disease (see Parag. 31).

152. Under the rules of Evidence-based Medicine, to claim that a collection of symptoms is a ‘nocebo effect,’ then, objective medical examinations must have been prescribed and no organic aetiology for the symptoms was found.

153. Has the TU Report’s Authors found a scientific justification for labelling this collection of symptoms as a ‘nocebo effect’?

154. Moreover, if, as Scientists, these Authors truly insist on standing by the ‘nocebo effect narrative,’ then they must be prepared to explain all the effects seen in animals living in proximity to wind power plants, such as:

Exposed cows in France registered a dramatic fall in milk output.26

Exposed cows in Korea are reported to have many cases of foetal death.27

In Poland, there was a negative effect on the stress parameters and productivity of exposed geese.28

In England, higher cortisol levels were found in exposed badgers and “these high levels may affect badgers’ immune systems, which could result in increased risk of infection and disease in the badger population.”29

26 Mulholland R. (2015) French farmer sues energy giant after wind turbines ‘make cows sick.’ The Telegraph, 18

September. https://www.telegraph.co.uk/news/worldnews/europe/france/11875989/French-farmer-sues- energy-giant-after-wind-turbines-make-cows-sick.html.

27 Se-hwan B. (2018). Wind turbines destroy local farming village. Rapid expansion of wind power facilities raises health and environmental concerns. The Korea Herald, 20 March. http://www.koreaherald.com/view.php?ud=20180320000768

28 Mikolajczak J, Borowski S, Marc-Pienlowska J, Odrowaz-Sypniewska G, Bernacki Z, Siodmiak J, Szterk P. (2013) Preliminary studies on the reaction of growing geese (Anser anser f. domestica) to the proximity of wind turbines. Polish Journal of Veterinary Sciences, 16(4):679-86. DOI: 10.2478/pjvs-2013-0096

29 Agnew RCN, Smith VJ, Fowkes RC. (2016) Wind turbines cause chronic stress in badgers (meles meles) in Great Britain. Journal of Wildlife Diseases, 52(3): 459-67. DOI: 10.7589/2015-09-231

Review of Tartu Univeristy Report, Estonia

Page 31 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

In a Polish study, the meat quality of exposed pigs decreased significantly.30

In Spain, a rabbit farm saw a 50% decrease in production after the WPP was installed. Exposed rabbits developed “problems of stress, infertility, death and behavioural changes,”31 and “a disproportionate increase in mortality rates.” This farm has since been closed down.32

Exposed frogs in Japan, “collected from paddy fields with wind power generators exhibited a faster call rate, higher salivary concentrations of corticosterone, and lower innate immunity (…) [This] can alter the disease epidemiology of local populations by regulating the balance between reproduction and immunity.”33

Exposed horses in Portugal developed flexural deformities and blood vessel walls revealed the characteristic, collagen-based thickening.34, 35

In Denmark, a mink farm was forced to close due to greatly increased aggressiveness, stillbirths and birth defects.36

30 Karwowska M, Milolajczak J, Dolatowski ZJ, Borowski S. (2015) The effect of varying distances from wind turbine

on meat quality of growing-finishing pigs. Annals of Animal Sciences, 15(4):1043-54. DOI: 10.1515/aoas-2015- 0051.

31 Ephe. (2023) [Union of Peasants of Castile and Leon denounces the ruin of one of the best farms in Spain by a wind farm] El Diário.es, 9 May. (In Spanish) https://www.eldiario.es/castilla-y-leon/union-campesinos-castilla-leon- denuncia-ruina-mejores-granjas-espana-parque-eolico_1_10189253.html

32 Fernández, JI. (2023) [A rabbit farm's fight against a wind farm: “We are in ruins”]. El Español, 9 May. [Article in Spanish] https://www.elespanol.com/castilla-y-leon/economia/el-campo/20230509/lucha-granja-conejos- parque-eolico-ruina/762423940_0.html

33 Park JK, Do Y. (2022) Wind turbine noise behaviourally and physiologically changes male frogs. Biology, 11, 516. DOI: 10.3390/biology11040516

34 Castelo Branco NAA, Costa e Curto T, Mendes Jorge L, Cavaco Faísca J, Amaral Dias L, Oliveira P, Martins dos Santos J, Alves-Pereira M. (2010) Family with wind turbines in close proximity to home: follow-up of the case presented in 2007. Proceedings of the 14th International Meeting on Low Frequency Noise, Vibration and Its Control. Aalborg, Denmark, 9-11 June, 31-40. https://www.researchgate.net/publication/290444702_Family_with_wind_turbines_in_close_proximity_to_home _follow-up_of_the_case_presented_in_2007

35 Costa e Curto TM. (2012) [Acquired flexural deformity of the distal interphalangic articulation in foals]. Master’s Thesis. Faculty of Veterinary Medicine, Technical University of Lisbon. [Thesis in Portuguese] https://www.repository.utl.pt/handle/10400.5/4847

36 Rapley, B. (2018). Conversation for a Small Planet Vol. 3-Biological Consequences of Low-Frequency Sound. Bouncing Koala Press, Palmerston North, New Zealand. (Chapter 8-Death in Denmark) Available from [email protected].

Review of Tartu Univeristy Report, Estonia

Page 32 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

155. Is it the position of the Authors of the TU Report that the adverse health effects observed in these animals, living in the proximity of wind power plants, are also caused by a ‘nocebo effect’ (i.e., psychosocial factors)?

156. Or is this one of the reasons why animal studies were excluded from the selection of papers chosen for this systematic review?

II. The questionnaire approach

157. In order to construct an appropriate questionnaire for people who live in proximity to wind power plants, yet another concept must be understood, regarding Medical Sciences and Physical Agents of Disease:

The health effects of physical agents of disease are cumulative.

158. This means the overall, prior noise exposure time (whatever the source!) is a parameter that must be considered, if a bona fide study based on questionnaires is desired.

159. This is true for vibration exposures, electromagnetic radiation exposures (where personal dosimeters are applied to actually quantify the cumulative exposure) and for noise exposures.

160. Stratification of study and control populations, as per prior noise exposures (severe, moderate and mild) must be made before any statistically valid study of health effects developed by citizens living in proximity to wind power plants can be properly obtained.

161. ‘Increased sensitivity’ can, therefore, merely be synonymous with significant, prior noise exposure, such as foetal exposures and/or prior occupational or residential exposures.

162. This particular topic has been extensively discussed elsewhere.37

37 The exclusion criteria applied by the Authorship of the TU report eliminated this paper from consideration. Alves-

Pereira M, Rapley B, Bakker H, Summers R. (2019) Acoustics and Biological Structures. In: Abiddine Fellah ZE, Ogam E. (Eds) Acoustics of Materials. IntechOpen: London. DOI: 10.5772/intechopen.82761.

Review of Tartu Univeristy Report, Estonia

Page 33 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

III. Another ‘Scientific Authorship’ of another “Wind Turbine Health Impact Study”…

163. In 2012 (13 years ago!), the Massachusetts Department of Environmental Protection and the Massachusetts Department of Public Health commissioned an Expert Independent Panel to conduct a “Wind Turbine Health Impact Study.”

164. IARO Scientists invite the Estonian Authors of the TU Report to read the Charge given to this Scientific Panel, shown in Figure 8.

Figure 8. Charge given to the Expert Independent Panel by the Massachusetts Department of Environmental Protection and the Massachusetts Department of Public Health38

38 Expert Independent Panel. (2012) Wind turbine health impact study. Massachusetts Department of Environmental

Protection and the Massachusetts Department of Public Health. https://www.mass.gov/files/documents/2016/08/th/turbine-impact-study.pdf

Review of Tartu Univeristy Report, Estonia

Page 34 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

165. How much does it differ from the Charge given to the Authors of the TU Report by the Ministry of Environment of Estonia?

166. In Annex D, please find the full Response from one of IARO’s Scientists to this 2012 Expert Independent Panel.

167. For the benefit of the Authors of the TU Report, excerpts taken from this Response are offered to our fellow Estonian Scientists in Fig. 9:

(p.4, Annex D)

(p.9, Annex D)

Figure 9. Excerpts from the Response to the Massachusetts Independent Expert Panel (Full Response Report is provided in Annex D).

Review of Tartu Univeristy Report, Estonia

Page 35 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

I. CONCLUSIONS

168. The systematic review documented in the TU Report has clearly been conducted properly in terms of how an analysis of published papers and reports should be undertaken when those participating are not experts in the subject matter.

169. The exclusion criteria applied to the selection of scientific papers for the study, blinds the Authors of the TU Report to a broader understanding of the matter at hand.

170. Given the non-evidence-based Recommendations proffered by the TU Report, it seems probable that the Authors of the TU Report have unwittingly succumbed to the unscientific practices promoted by governments and international special interest groups.

171. In the opinion of IARO Scientists, this study can only be regarded as, yet another, artificially constrained review of papers, with outcomes predetermined by politically generated questions, resulting in a report of low scientific standard.

REVIEW STUDY PHASE I REPORT

HEALTH EFFECTS OF WIND TURBINES: A SYSTEMATIC REVIEW OF STUDIES PUBLISHED IN PEER-REVIEWED SCIENTIFIC JOURNALS OVER THE LAST FIFTEEN

YEARS

Title of the study:

Development of a methodology for interpreting the results of scientific studies on the potential health effects of wind

farms and other energy production technologies in the Estonian context

Tartu 2025

2

Research team:

Triin Veber (MSc, MPH), University of Tartu, project manager and expert

Ene Indermitte (PhD, MPH), University of Tartu, expert

Kaja-Triin Laisaar (MD, MPH, PhD), University of Tartu, expert

Urmeli Katus (RN, MSc), University of Tartu, senior methodologist

Ele Kiisk (MA, MSc), University of Tartu, junior methodologist

Hans Orru (PhD, MPH), University of Tartu, research director

Commissioned by: Ministry of the Environment

3

Contents Introduction and scientific background of the study.........................................................................................................4

Noise and infrasound terms and limit values .................................................................................................................6

Aim of the study ................................................................................................................................................................12

Methods ..........................................................................................................................................................................13

Results .............................................................................................................................................................................16

The impact of audible noise from wind turbines on human health ............................................................................17

Audible noise from wind turbines and sleep ...........................................................................................................17

Audible noise and disturbance caused by wind turbines ........................................................................................21

Audible noise from wind turbines and clinically significant health effects .............................................................22

The impact of infrasound from wind turbines on human health ................................................................................25

The impact of visual aspects related to wind turbines on human health....................................................................29

Discussion of results ........................................................................................................................................................31

Audible noise from wind turbines and sleep................................................................................................................31

Disturbance caused by wind turbines ..........................................................................................................................32

The impact of infrasound from wind turbines on health .............................................................................................36

Claims about the health effects of infrasound in public debates and the media ...................................................38

Audible noise generated by wind turbines and clinically manifested health effects ..................................................40

Electromagnetic fields and vibrations caused by wind turbines..................................................................................43

Justification for the choice of methodology, strengths and weaknesses of the study ................................................44

Conclusions........................................................................................................................................................................46

Recommendations ...........................................................................................................................................................48

References .......................................................................................................................................................................50

Appendix 1. Search terms and search strategy used .......................................................................................................57

Appendix 2. Tables of results...........................................................................................................................................58

Table 1. Systematic reviews.....................................................................................................................................58

Table 2. Observational studies ................................................................................................................................59

Table 3. Experiments on infrasound ........................................................................................................................61

Table 4. Experiments involving audible noise and visual aspects of wind turbines................................................65

4

Introduction and scientific background of the study Global climate change poses a threat to health and well-being around the world (Lancet Countdown, 2024).

As greenhouse gases are the main factor influencing climate change, the European Union has agreed to

reduce gas emissions by at least 55% by 2030 compared to 1990 levels in order to mitigate the

consequences (European Commission, 2019). To achieve this goal, many countries are increasingly

developing renewable energy production methods, particularly wind energy. One of the most important

issues in the development of wind energy is ensuring the health safety of wind farms. Conclusions about

health safety cannot be based solely on selective scientific literature, but require a systematic review of the

literature. A systematic review of the literature provides a much more comprehensive and reliable picture of

the existing scientific knowledge than reading individual scientific articles and drawing conclusions based

on their results.

One of the most important health safety issues is the noise generated by wind turbines. Some studies have

shown that noise from wind turbines (wind generators, windmills, wind turbines) is considered more

disturbing than other sources of noise, such as transport or industrial noise (Teneler and Hassoy, 2023). It

has also been argued that wind turbine noise can affect sleep and disturb people (Schmidt and Klokker,

2014). However, the stress and sleep disturbances caused by disturbance can in turn lead to clinically

apparent health effects (Basner et al., 2014). For example, traffic noise has been shown to increase the risk

of cardiovascular disease (heart attack and stroke) and metabolic disease (diabetes and obesity) (van

Kempen et al., 2018; Wu et al., 2023). However, reviews of wind turbine noise conducted to date have not

confirmed a link between wind turbine noise and clinically apparent health effects (Karasmanaki, 2022;

Schmidt and Klokker, 2014; Teneler and Hassoy, 2023; van Kamp and van den Berg, 2021, 2018).

However, whether and what links between wind turbine noise and health indicators have been found in

more recent studies has not been systematically analysed in recent years.

Wind turbine noise is classified as industrial noise, and in Estonia, the limit values for industrial noise in

residential areas are 60 dB (A) during the day and 45 dB (A) at night. The target values for noise are 50 dB

(A) during the day and 40 dB

(A) at night (Regulation No. 71 of the Minister of the Environment, 2016). Currently, when planning new

wind farms in existing residential areas, wind turbine noise must not exceed the limit values. In practice,

however, impact assessors generally base their assessments on outdoor noise target values, meaning that

wind turbine noise should not exceed 40 dB (A) in residential areas at night (Ministry of Climate, 2025). It

is important to determine whether the current regulations are sufficient to ensure health safety or whether

the limits should be tightened.

5

Various review studies suggest different values for safe wind turbine noise levels. A systematic review by

Schmidt and Klokker (2014) recommends that wind turbine noise levels in residential areas should not

exceed LAeq 35 dB(A) in order to ensure the least disruptive environment possible for residents (Schmidt

and Klokker, 2014). Ellenbogen et al.'s review on sleep disturbance (Ellenbogen et al., 2024) concluded

that wind turbine noise measured outside residential buildings up to 46 dB (A) or modelled according to the

new standard (ANSI/ACP 2022) up to 49 dB (A) does not pose a risk to human sleep. The World Health

Organisation (WHO) recommends limiting wind turbine noise on the exterior facade of residential

buildings to a weighted sound pressure level of Lden 45 dB (A) and Lnight 40 dB (A) (Basner and McGuire,

2018; Clark and Paunovic, 2018; Guski et al., 2017; van Kempen et al., 2018; WHO, 2009). However, Lden

is not directly comparable with the noise indicators used in our legislation. The different noise indicators,

units and terms are described in more detail in the chapter ‘Noise and infrasound terms and limits’. Davy et

al. (2020) have suggested that Lden 45 dB (A) corresponds to a noise level of LAeq 38.6 dB (A) (Davy et al.,

2020).

An Australian review study found that the main indicator for setting a limit value for wind turbines could

be that no more than 10% of the population should feel significantly disturbed by the noise. Based on their

analysis, this limit appears to be in the range of 34–40 dB LAeq ( 10 min) at the exterior of a dwelling, with an

average value of 37 dB LAeq (10 min) ( Davy et al., 2020).

An important topic in public debates in many countries and in non-scientific literature has been the possible

health risks associated with exposure to infrasound (the inaudible part of wind turbine noise with a

frequency below 20 Hz) (Schmidt and Klokker, 2014; van Kamp and van den Berg, 2018). However, no

adverse health effects associated with wind turbine infrasound have been demonstrated to date (Knopper

and Ollson, 2011; Schmidt and Klokker, 2014; van Kamp and van den Berg, 2018). Questions have also

been raised about the possible health effects of low-frequency noise (20–200 Hz) from wind turbines

(Schmidt and Klokker, 2014).

Another significant problem associated with wind turbines is that they disturb many people for various

reasons. Although disturbance cannot be considered a clinically manifest health outcome, it affects people's

well-being and can therefore be considered part of the World Health Organisation (WHO) definition of

health, according to which health is "a state of complete physical, mental and social well-being and not

merely the absence of disease or infirmity" (WHO, 2025). The level of disturbance is usually measured

using questionnaires in which respondents are asked to assess their level of disturbance.

6

Studies conducted to date have shown that, in addition to noise, visual aspects of wind turbines such as

shadow flicker, flashing lights on the blades, obstruction of views, etc. can also contribute to disturbance

(Freiberg et al., 2019b; Knopper et al., 2014). Disturbance also depends on people's own knowledge,

attitudes and perceptions. People who have a generally negative attitude towards wind energy have been

shown to experience greater disturbance compared to those who have a neutral or positive attitude

(Karasmanaki, 2022). It has been shown that people who believe that wind turbines are harmful to health

complain more about disturbance and also experience more self-reported health problems caused by wind

turbines. This phenomenon is called the nocebo effect (Davy et al., 2020; Karasmanaki, 2022). To reduce

disturbance, it is important to know exactly which factors increase people's disturbance in order to

minimise it.

Although many review articles analyse the health effects of wind turbines or wind farms, there are several

shortcomings in the study design and methodological quality. Most reviews are narrative (Karasmanaki,

2022; Knopper and Ollson, 2011; van Kamp and van den Berg, 2021) and/or based on cross-sectional

studies, where we cannot establish a definite temporal relationship between exposure and health effects.

Narrative reviews are at greater risk of bias than systematic reviews. One of the few systematic reviews

conducted to date is the study by Schmidt and Klokker (2014) on the health effects of wind turbine noise.

At the time of writing, however, no long-term follow-up studies had been conducted on this topic, so the

review relied on case studies and cross-sectional studies, which, due to their study design, cannot provide

evidence of causal relationships. Cohort studies have since been published that allow conclusions to be

drawn about causal relationships between wind turbine noise and health indicators, for example (Bräuner et

al., 2018; Poulsen et al., 2019a), but their results have not yet been systematically analysed. Experimental

studies are also important for establishing causal relationships. For example, an experimental study on the

effects of wind turbine infrasound was published in 2023 (Marshall et al., 2023), but the results of

experimental studies on the health effects of wind turbines have not been systematically analysed.

Noise and infrasound terms and limits

Noise is defined as sound that disturbs people or harms their health and well-being (Regulation No. 42 of

the Minister of Social Affairs, 2002). Noise is therefore a subjective indicator – what seems like noise to

one person may be music to another. Noise is caused by sounds.

7

Sound is a vibration that propagates through the environment (e.g. air, water or solid matter) in waves.

When sound propagates through the air, it causes fluctuations in air pressure. Sound pressure varies in

relation to atmospheric pressure, sometimes being greater than the mean atmospheric pressure and

sometimes less (Lahti, 2010). The intensity of sound is determined by the extent of the pressure change and

is measured in decibels (dB). A decibel indicates how much louder the sound is than the reference value. In

air, the reference value is an air pressure of 20 micropascals (20 μPa or 2×10⁻⁵ Pa), which is considered to

be the human hearing threshold at a frequency of 1000 Hz – this is the quietest sound that the average

person can still hear at this frequency (Basner et al., 2014). Sound pressure level (SPL) Lp is a logarithmic

measure that indicates the effective pressure of a sound compared to a reference value (agreed hearing

threshold) and is expressed in decibels (dB) (Regulation No. 42 of the Minister of Social Affairs, 2002).

Humans can perceive sound pressure in the range from 0 dB (agreed hearing threshold) to over 140 dB

(pain threshold). The smallest change in sound level that humans can distinguish outdoors is 3 dB. This

change corresponds to a doubling of sound energy. A 10 dB increase is subjectively perceived as twice as

loud, although physically it represents a tenfold increase in sound energy (Ellenbogen et al., 2024).

In addition to sound intensity, sound is also characterised by its frequency. Frequency is the number of

sound wave amplitudes per second and is measured in hertz (Hz). The lower the sound, the lower the

frequency, and the higher the sound, the higher the frequency. The timbre of a sound is determined by the

frequency distribution of sound waves (Lahti, 2010).

Sound is conventionally divided into three categories according to frequency: infrasound (< 20 Hz), audible

sound (20–20,000 Hz) and ultrasound (> 20,000 Hz). Infrasound is generally lower than the human hearing

threshold and is usually inaudible to humans. However, there are no clear boundaries in nature – the

transition between audible and inaudible sound is gradual and smooth. At higher frequencies, infrasound is

audible or perceptible when it is very strong (Ellenbogen et al., 2024). Sources of infrasound include

natural phenomena such as wind, waves, heartbeats, etc. Blue whales (10–20 Hz) also generate and hear

infrasound. The most common man-made sources of infrasound are fans, motors, cars, aeroplanes, trains,

air source heat pumps, heating systems, etc. (Ellenbogen et al., 2024; Lahti, 2010).

Sources of audible noise include traffic noise, noise from neighbours' activities, and noise in the workplace.

Low-frequency sound is sometimes considered to be the lowest frequency range of audible sound, 20–200

Hz (Ellenbogen et al., 2024), but in some scientific articles and also according to Regulation (Regulation

No. 42, 2002) by the Minister of Social Affairs, , low-frequency noise

8

in the frequency range 10 Hz–200 Hz, which partially overlaps with the frequency of infrasound, but is also

audible.

Humans hear sounds of different frequencies with varying intensity, but cannot hear sounds that are too

high or too low at all. Hearing is most sensitive in the 2000–5000 Hz range. For example, the sensitivity of

hearing for sounds with a frequency of 100 Hz is approximately 20 dB lower (Lahti, 2010). This means that

if a sound contains more low or high frequencies, a higher sound pressure level (sound intensity) is required

to perceive it (Ellenbogen et al., 2024). Since the measuring device measures all frequencies equally, a

correction is made to better assess noise, which evaluates sound frequencies similarly to the human ear.

This is called A-correction (LpA). The terms A-sound level or A-weighted sound level may also be used. The

pressure symbol p can be omitted from the sound level symbol LpA. The symbol dB (A) is also often used

after the unit to express the A-weighted sound level (Lahti, 2010).

For measuring low-frequency sound and infrasound, C- or G-correction is used, or measurements are taken

without correction (Z). Z-correction is denoted by dB (Z) and actually means unweighted sound level

across the entire frequency range. Z comes from the English expression zero-weighted. Figure 1 shows how

different frequency corrections assess noise.

Figure 1. Frequency correction (frequency weighting) curves (Health Board, 2025a)

9

Since sound pressure (sound intensity) changes constantly over time, the time during which noise is

measured and averaged must also be taken into account when assessing environmental noise. For this

purpose, the equivalent sound pressure level LpA,eq,T is used, which characterises noise with a variable level

over a period of time T, using A-correction (Lahti, 2010; Regulation No. 42 of the Minister of Social

Affairs, 2002). The duration of noise measurement (T) is selected according to the measurement method

and the nature of the noise; for noise with varying levels, the minimum period is 10 minutes (Regulation

No. 42 of the Minister of Social Affairs, 2002).

In addition to the equivalent noise level, the maximum sound pressure level LpA,max is also used to

characterise noise, which indicates the maximum value of the sound pressure level measured during a

specified period (Lahti, 2010; Regulation No. 42 of the Minister of Social Affairs, 2002).

In Estonia, noise limits have been established for A-weighted sound pressure levels in the outdoor

environment (Regulation No. 71 of the Minister of the Environment, 2016). Indoors, both A-weighted noise

levels and C-weighted low-frequency noise levels are regulated, and there are also requirements for

assessing low-frequency noise without frequency correction (Regulation No. 42 of the Minister of Social

Affairs, 2002). Currently, the limit values for industrial noise in residential areas are 60 dB (A) during the

day and 45 dB

(A) in residential areas. The target noise levels are 50 dB (A) during the day and 40 dB (A) at night

(Regulation No. 71 of the Minister of the Environment, 2016). In bedrooms, traffic noise levels may be up

to 30 dB (A) at night and noise from technical equipment up to 25 dB (A), and 50 dB (C) both at night and

during the day (Regulation No. 42 of the Minister of Social Affairs, 2002).

A low-frequency noise frequency curve has been established to limit low-frequency sounds (Figure 2).

(Regulation No. 42 of the Minister of Social Affairs, 2002). These are values without frequency correction

in the frequency range 10–200 Hz. This frequency curve allows for higher uncorrected sound levels at

lower frequencies (10 Hz to 95 dB (Z)) and lower levels at higher frequencies (200 Hz to 32 dB (Z)). In

addition, Estonia has a regulation on infrasound and ultrasound, which sets the limit value for the G-

corrected equivalent sound pressure level LpG,eq,T at 85 dB (G) (Regulation No. 75 of the Minister of Social

Affairs, 2002).

10

Figure 2. Standard levels for sounds without frequency correction according to the annex to Regulation No.

42 of the Minister of Social Affairs of 4 March 2002 (Health Board, 2025b).

One characteristic feature of wind turbine noise is amplitude modulation (AM), i.e. a sound level that varies

over time with approximately regular periodicity. AM is a fluctuation in sound level whose cycle time

generally corresponds to the frequency of blade passage from the wind turbine tower. The fluctuations in

sound level are usually small, around 2–4 dB (McCunney et al., 2014), which is why wind turbine noise is

considered to be continuous rather than impulsive (Ellenbogen et al., 2024). In some rare cases, however,

fluctuations in wind turbine sound levels can be much greater – up to 10 dB. In groups of several wind

turbines, the modulations of individual generators can synchronise, causing periodic increases in

modulation intensity. There may also be periods when the modulations of individual generators balance

each other out, reducing the modulation strength. Sometimes the synchronisation of generators can last for

hours if the wind speed and direction remain constant. The AM level is not correlated with wind speed.

Most cases of so-called 'strong' AM are caused by unusual meteorological conditions. AM also varies

depending on location — in some places it occurs rarely, while in others it has been measured up to 30% of

the time (McCunney et al., 2014).

In addition, the following noise indicators are used to assess environmental noise (Regulation No. 71 of the

Minister of the Environment, 2016):

Lden – day-evening-night noise indicator. A long-term average sound pressure level determined on the basis

of the numerical values of all day, evening and night sound pressure levels throughout the year, which is a

general noise disturbance indicator.

11

disturbance indicator. When determining Lden, a correction factor of +5 dB is applied to evening noise and

+10 dB to night-time noise. The indicator is calculated on the basis of Lnight, Lday and Levening.

Lnight – night-time noise indicator. The long-term average sound pressure level determined on the basis of all

night-time hours of the year, which is an indicator of noise disturbing sleep and characterises sleep

disturbance between 23:00 and 7:00. Determined in accordance with standard ISO 1996-2: 1987.

Lday – daytime noise indicator. The long-term average sound pressure level determined on the basis of all

daytime hours of the year, which characterises the disturbing effect of noise during the day between 7:00

and 19:00. Determined in accordance with standard ISO 1996-2: 1987.

Levening – evening noise indicator. Long-term average sound pressure level determined on the basis of all

evening times of the year, which characterises the disturbing effect of noise in the evening between 19:00

and 23:00. Determined in accordance with standard ISO 1996–2: 1987.

12

Purpose of the study The aim of the study was to systematically analyse the evidence published in the scientific literature over the

last fifteen years (2010–2025) on the health effects of wind turbines.

Research questions:

1. What are the main conclusions of existing studies on the health effects of wind turbines?

2. What is the overall quality of the existing evidence? Is there evidence in the scientific literature

that wind turbines have a negative impact on human health?

3. If wind turbines have negative health effects, what health effects are associated with wind

turbines?

4. If wind turbines have negative health effects, what role do environmental factors such as noise,

infrasound, shadow flicker, visual aspects, psychological factors (including general attitudes

towards wind turbines, people's beliefs and perceptions of wind turbines), vibration and

electromagnetic fields in causing these health effects?

5. If wind turbines have health effects, under what conditions are these health effects more likely

to occur (e.g. at what distance from the turbine, with powerful or tall turbines, etc.)?

6. Are certain population groups more vulnerable to the potential health effects of wind turbines?

7. What evidence-based recommendations can be made to policymakers, industry stakeholders

and affected communities to protect human health?

13

Methods This study was conducted as a systematic literature review. In compiling the systematic literature review,

we used the principles of rapid review methodology (Garritty et al., 2024; King et al., 2022). A systematic

review is a scientific research method that aims to collect, evaluate and synthesise all relevant scientific

studies on a specific research question or topic using a systematic, transparent and repeatable process to

minimise the risk of bias in the conclusions drawn. To achieve this goal, specific criteria are agreed upon

for the inclusion and analysis of studies. Several members of the research team participate in each stage of

the study, checking each other's work. The included studies are compared with each other on the basis of

study quality, and the results of higher quality studies are given greater weight in the conclusions. Scientific

studies are the most reliable way to obtain accurate information, but it is not possible to conduct a scientific

study without limitations or the risk of obtaining inaccurate results. For example, a study may suffer from

"selection bias". This occurs when the participants in the study do not adequately represent the target group.

For example, people who are more health-conscious may be more likely to participate in a sleep study than

the general population we want to study. Self-reported data is subject to "recall bias"; for example, patients

with a disease may remember their exposure to noise better than healthy people. "Measurement bias"

occurs when exposure or outcome data are not measured accurately, for example, there is a measurement

error in noise measurements or different methodologies have been used to measure noise in comparable

groups. In observational studies, it is important to take into account "confounding factors". Confounding

factors occur when another factor is associated with both the exposure under study and the outcome. For

example, wealthier people may live in areas with less noise and also have better opportunities for healthier

lifestyles, and in fact, it is people's income that influences the onset of disease, even though the analysis

shows a link with noise. This can be avoided by adjusting for confounding factors (e.g. income, age,

gender). Adjustment ensures that only subjects with similar adjusted characteristics are compared.

Scientific journals tend to exhibit

"publication bias". Studies that find a link are more likely to be published than studies that do not find a

link. The most common risk of bias in experiments is "lack of blinding" – if participants or researchers

know which study group is receiving the placebo infrasound and which is receiving the real sound, this will

affect the results. People with certain characteristics, such as those who are sensitive to noise, may also

drop out of the experiment. There may also be errors in measuring the outcome. For example, a person may

have high blood pressure, but if a doctor has not diagnosed it

14

diagnosed it, they will be considered healthy in a registry-based study. At the same time, if a person reports

that they have high blood pressure, it is not known whether this is a temporary increase in blood pressure

(e.g. due to stress) or whether they have developed hypertension. We took all these potential risks of error

into account when drawing conclusions in our study.

We used the internationally recognised PECO (population, exposure, comparison, outcome) framework to

define the research questions. We included pre-reviewed scientific articles in the systematic literature

review that dealt with:

• Population: the general population (all people)

• Exposure to the following factors of onshore and offshore wind turbines or wind farms:

Acoustic factors: noise (frequency above 20 Hz), low-frequency noise (20 Hz to 200 Hz) and

infrasound (below 20 Hz) from wind turbines or wind farms; visual factors: visual disturbance

of the landscape, shadow flicker, flashing lights associated with wind turbines at night, direct

visibility from home windows; psychological factors: general attitude towards wind turbines,

people's beliefs and perceptions of wind turbines (as a disturbing factor), disturbance from

wind turbines; vibration; electromagnetic fields.

• Comparison group: people who have no exposure to wind turbines or less exposure than

others; placebo groups in experimental studies; the same people before and after exposure (i.e.

self-controls) in ‘before and after’ studies.

• Health outcomes: all possible health-related outcomes, including disturbance, sleep

disturbances, health symptoms (measured subjectively or objectively, measured physiological

parameters), health-related quality of life, psychological indicators.

Only studies with a study design that allows causal relationships to be identified were included. Studies

with the following study designs were included:

• Single studies: longitudinal studies (cohort studies, case-control studies); intervention studies

(comparisons before and after the installation of wind turbines), experimental studies/trials

• Systematic reviews

We did not include the following studies in this systematic review:

15

• Studies that did not address human health effects. Studies that only addressed technical aspects of wind turbines or studied animals, wildlife or laboratory animals

• Studies that dealt with exposure to wind turbines in the working environment, not in the living environment

• Studies that addressed the health effects of sound (noise), infrasound, electromagnetic fields, vibration and visual aspects, but were not related to wind turbines.

• Individual studies published before 2010 and systematic reviews published before 2015

• Narrative review studies, case studies, cross-sectional studies, letters to scientific journals,

comments in scientific journals, editorials in scientific journals, conference summaries

• Information sources that were not published in peer-reviewed scientific literature

• Non-English studies

If a comprehensive systematic review on the research question had been published in 2015 or later, only

individual studies published after the inclusion period of the review were included in the study. In other

cases, the research question was answered by including all individual studies that met the inclusion criteria

(Garritty et al., 2024; King et al., 2022).

A systematic literature search was conducted on the health effects of wind turbines (wind farms). The

search was conducted in the PubMed and Scopus databases for the period from 1 January 2010 to 22 April

2025. The PubMed database search strategy is presented in Appendix 1.

Endnote reference management software was used to remove duplicates, and Mendeley Reference Manager

software was used to manage references. The initial selection of studies (screening) was based on the title

and abstract of the study. Initially, two members of the research team independently screened 20

publications and then discussed the results to reach a common understanding regarding the inclusion of

studies. To ensure the quality of the review, two members of the research team screened 20% of the entries,

while the rest were screened by only one member of the research team.

The full texts of studies initially assessed as potentially suitable were retrieved and a second screening was

conducted on the basis of these. One member of the research team read through the full texts of potentially

suitable studies and, if he or she decided to exclude a study, another member of the research team read

through the full text and confirmed or rejected the decision to exclude it. However, studies deemed suitable

by the second member of the research team were discussed, and the decision to include or exclude them

was made by consensus. The data from the articles were entered into an MS Excel table.

16

Finding studies through database searches

Database searches (n = 1374): PubMed (n = 716) Scopus (n = 658)

Duplicate entries removed (n = 211)

Included in the review (n = 32), of which systematic reviews (n = 4)

Results The database search yielded 1,374 entries (scientific articles), of which 1,163 were assessed for compliance

with our inclusion criteria based solely on their titles and abstracts. The full texts were reviewed in 115

cases. In addition to the database search, one study was found through a manual search. Thirty-two studies

met the inclusion criteria and were used in the analysis of this study, of which four were systematic

reviews, 19 were experiments and n i n e were observational studies (Figure 3).

The most important results of all included articles are summarised in the tables (Appendix 2, Tables 1–3).

Figure 3. PRISMA flow diagram for describing scientific literature searches.

Full texts reviewed (n = 115)

Full texts searched (n = 115)

Found by browsing (n = 1):

Excluded (n = 84): study design (n = 55) no health impact (n = 15) included in systematic review (n = 8) Other (n = 6)

Full texts unavailable (n = 0)

Entries removed (n = 1048)

Initial screening of records based on title and abstract (n = 1163)

R ev

ie w

Se ar

ch In

cl us

io n

17

The impact of audible noise from wind turbines on human health

Audible noise from wind turbines and sleep

Two systematic reviews/meta-analyses (Godono et al., 2023; Liebich et al., 2021) (Appendix 2, Table 1),

two experiments in a sleep laboratory (Liebich et al., 2022a, 2022b) (Appendix 2, Table 4) and one long-

term follow-up study (Poulsen et al., 2019b) (Appendix 2, Table 2).

Both systematic reviews included in this review concluded that increased wind turbine noise increases the

risk of self-reported sleep disturbances and/or lower sleep quality (Godono et al., 2023; Liebich et al.,

2021). However, one of the included reviews (Liebich et al., 2021) found no effect of noise on objectively

measured sleep parameters (using polysomnography (PSG) and actigraphy). Another review study only

considered subjectively assessed sleep parameters and found that both wind turbine distance and sound

levels above 30 dB(A) affect self-reported sleep disturbances (Godono et al., 2023).

The aim of a systematic review and meta-analysis (Liebich et al., 2021) was to assess the impact of wind

turbine noise on sleep using only validated subjective and objective measures. Objective measures were

assessed using polysomnography (PSG) and actigraphy. Polysomnography is the 'gold standard' for

objective sleep measurement as it uses direct electroencephalography (EEG). An actigraph is a wrist-worn

motion sensor that detects sleep and wakefulness based on general body movements. Subjective measures

included sleep diaries and questionnaires. The review included nine studies that used widely accepted and

validated objective and subjective sleep assessment methods and were published after 2000. Five studies

were included in the meta-analysis, four of which used PSG and one of which used actigraphy ( Liebich et

al., 2021).

The review study described (Liebich et al., 2021) showed that wind turbine noise does not significantly

affect the main objective indicators of sleep: sleep onset latency (SOL), wake after sleep onset (WASO),

total sleep time (TST) and sleep efficiency. However, an impact was found on subjectively measured (self-

reported) sleep indicators. Based on the questionnaires, wind turbine noise affected the severity of insomnia

symptoms (Insomnia Severity Index, ISI), sleep quality (Pittsburgh Sleep Quality Index, PSQI) and daytime

sleepiness. There were also indications that higher amplitude modulation (AM) may increase wakefulness

and reduce the duration of deep sleep.

18

Another systematic review and meta-analysis on sleep (Godono et al., 2023) assessed the impact of wind

turbine noise only on self-reported subjective sleep indicators. The study found that both the distance from

wind turbines and sound intensity affect self-reported sleep disturbances. The lowest prevalence of sleep

disturbance was found at sound levels <30 dB(A) (31%) and increased as noise levels increased. The

prevalence of self-reported sleep disturbances at different distances (in metres) from the nearest wind

turbine in the wind farm was as follows in the study (Godono et al., 2023):

• <500 m: 79%

• 500–1000 m: 65%

• 1000–1500 m: 41%

• 1500–2000 m: 29%

• 2000–3000 m: 22%

Godono et al. (2023) reviewed studies from Europe, the US, Canada, Japan, and China published between

2004 and 2021. The distance to wind turbines in the included studies ranged from 495 to 3093 m, and the

capacity of the wind turbines ranged from 0.5 to 3.5 MW.

Wind turbines can be a source of stress, which can contribute to self-reported sleep disturbances. However,

it should be noted that self-reported sleep indicators are affected by recall bias, inaccuracy in sleep

perception, or misinterpretation of awakenings, and the study by Godono et al. (2023) only included cross-

sectional studies, which do not allow conclusions to be drawn about causal relationships. The authors of

Godono et al. (2023) rate the quality of most of the studies included as low, as they often did not adjust for

many necessary factors, such as air pollution, visual disturbance, air temperature and humidity, and the

economic situation of the subjects (Godono et al., 2023).

In Australia, two single-blind randomised controlled trials were conducted in a sleep laboratory, where

subjects were exposed to pre-recorded audible noise and infrasound from a wind farm. In the first study

(Liebich et al., 2022a), 68 subjects aged 18–80 slept for seven consecutive nights in a sleep laboratory,

where they were exposed to a wind turbine noise recording at a volume of 25 dB(A). The recording used

was a recording of real wind farm noise made indoors in a house located 3.3 km from a wind farm in South

Australia. The noise level of 25 dB(A) was chosen because it is similar to the median noise level of wind

turbines measured indoors throughout the year at a distance of 1–3 km from wind farms. This level was

also 6 dB (A) higher than the laboratory background noise and therefore clearly audible to participants with

normal hearing. The recordings contained infrasound from a frequency of 1 .6 Hz and perceptible

amplitude modulation ( AM). Objective sleep parameters

19

were assessed using polysomnography (PSG). Electromyography, electrooculography, electrocardiography,

pulse oximetry and leg movement signals were also recorded. Subjective sleep indicators were assessed

using a validated web-based ‘sleep diary’ questionnaire. This diary included questions about time spent in

bed, time spent out of bed, and minutes of sleep and wakefulness during the night, allowing for the

calculation of time spent in bed, sleep onset time, number and duration of awakenings, time of awakening,

and total sleep time. The subjects were divided into four groups: those who reported sleep disturbances

related to wind turbine noise (N=14, living <10 km from the wind farm); those without sleep disturbances

(N=18, living <10 km from the wind farm); traffic noise sleep disturbance (N=18), who reported sleep

disturbances related to traffic noise; control group (N=18), who lived in a quiet rural area. During the

experiment, four different noise scenarios were played to the subjects in random order on four nights after

an adaptation night: quiet control night with only laboratory background noise of 19 dB (A); continuous

wind turbine noise at 25 dB(A) throughout the night; wind turbine noise at 25 dB(A) only during

established sleep; and wind turbine noise at 25 dB(A) only during wakefulness or light sleep.

The results of the study (Liebich et al., 2022a) showed that there was no significant effect of wind turbine

noise on the sleep parameters studied. This means that at a level of 25 dB (A), wind turbine noise did not

significantly affect sleep efficiency, sleep latency, total sleep time, wake time after sleep onset, or different

sleep stages, regardless of the participant's previous exposure to noise or self-reported sleep disturbances.

Those who were more sensitive to noise slept worse on the control night than on the night without noise.

However, the sleep of those sensitive to noise did not differ from that of others on the nights with noise.

Another study (Liebich et al., 2022b) was conducted with louder recorded wind turbine noise and involved

23 subjects aged 18–29 who had not previously been exposed to wind turbine noise. The recorded wind

farm noise (which also included infrasound) was presented in a sleep laboratory at an intensity of 33 dB (A)

in random order, alternating with laboratory background noise of 23 dB (A). The wind farm noise

contained infrasound and noticeable AM at 46 Hz. The study showed that wind turbine noise at a level of

33 dB(A) does not prolong the time it takes to fall asleep, as measured objectively or subjectively, in

young, healthy people who have not previously been exposed to wind turbines.

These two high-quality randomised controlled trials (Liebich et al., 2022a, 2022b) showed that wind

turbine noise at a level of 25 dB (A) has no effect on objective or subjective sleep parameters, even in older

and noise-sensitive individuals. No effect was observed even in those who reported wind turbine-related

sleep disturbances. Based on these studies, it can be assumed that wind turbine noise levels below 25 dB

(A) indoors, which also includes amplitude modulation, are safe even for sensitive groups. Even wind

turbine noise levels of 33 dB (A) did not prolong sleep

20

, but this study only included young people and only looked at one sleep indicator, so no conclusions can be

drawn about other sleep indicators or older people.

However, the disadvantage of randomised controlled trials is that they do not show the long-term effects of

noise. The long-term effects of noise on sleep have been studied in a large cohort study conducted in

Denmark (Poulsen et al., 2019b). This study assessed the impact of night-time wind turbine noise in the

vicinity of residential buildings and the impact of low-frequency night-time noise indoors on an objectively

measured sleep disturbance indicator, which was the purchase of sleeping pills based on the Danish patient

register.

The study by Poulsen et al. (2019b) included 583,968 subjects who were not taking sleeping pills at the

start of the follow-up. They were followed for 17 years (1996 to 2013). Of these, 68,696 were taking

sleeping pills at the end of the follow-up period. The subjects were aged 25–84 (Poulsen et al., 2019b).

Wind turbine noise was modelled (only noise originating from wind turbines was assessed) for all Danish

dwellings located within a radius of up to 20 wind turbine tower heights (the study group). For example, if

the height of the wind turbine tower was 35 m, the noise from the wind turbines was modelled for

dwellings within a radius of 700 m; if the height of the tower was 100 m, the radius was 2000 m. Wind

turbine noise was also modelled for 25% of randomly selected dwellings located within a radius of 20 to 40

wind turbine tower heights (control group). Two noise indicators were modelled: A-weighted night-time

noise level near the dwelling in the yard at frequencies of 10–10,000 Hz (which, in addition to audible

sound, also included low-frequency sound and, in part, infrasound) and, separately, only low-frequency

noise ( also including, in part, infrasound) inside dwellings at frequencies of 10–160 Hz. Wind turbine

noise outdoors near residential buildings was divided into the following classes: less than 24; 24–30; 30–

36; 36–42 and over 42 dB (A). Low-frequency wind turbine noise indoors was divided into the following

classes: less than 5; 5–10, 10–15 and over 15 dB (A). The analysis used 1-year and 5-year average noise

indicators (Poulsen et al., 2019b).

Log-linear Poisson regression analysis was used to assess the associations between noise levels and the

consumption of the sleeping pills under investigation. All analyses were adjusted for gender, calendar year

and age. In addition, the analyses were adjusted for educational level, income, marital status, labour market

participation status, type of housing (farm, single-family house, other), traffic load within a 500 m radius of

the place of residence, and distance from the nearest road with more than 5,000 vehicles per day. The

analyses took into account changes in these characteristics over time during the follow-up period (Poulsen

et al., 2019b).

The study described by Poulsen et al. (2019b) found that, compared to those whose homes were in the

lowest noise class (less than 24 dB (A)), there was a statistically significant 3–8% higher risk of sleeping

pill use among those living in higher noise classes ( noise near residences

21

outdoors 24–30; 30–36; 36–42 dB (A)). Those who lived in the highest noise class (over 42 dB (A)) had a

14% higher risk of using sleeping pills compared to those who lived in the lowest noise class (less than 24

dB (A)), but this difference was not statistically significant. In order to find out whether people of different

ages may be affected differently by wind turbine noise, the study analysed the data by age group. It was

found that wind turbine noise affects older people more. Those over 65 years of age who lived near wind

turbines with an outdoor noise level of over 42 dB (A) had a 68% higher risk of using sleeping pills

compared to those who lived near wind turbines with an outdoor noise level of less than 24 dB (A). In the

over-65 age group, a higher risk of sleeping pill use was observed from a night-time noise level of 30

dB(A) outdoors near their homes. No effect of low-frequency noise indoors on sleeping pill use was found.

Audible noise and disturbance caused by wind turbines

Disturbance is an indicator measured by a questionnaire. The respondent assesses how disturbing they find

the noise on a given scale. The relationship between environmental noise and disturbance is discussed in a

review article (Guski et al., 2017) (Appendix 2, Table 1). This review focused on environmental noise,

assessing the relationship between traffic noise from roads, railways and air traffic and disturbance, in

addition to noise from wind turbines. Four cross-sectional articles published between 2000 and 2012 were

included from studies on wind turbine noise. The study showed that as wind turbine noise increases, the

likelihood of disturbance increases, but this relationship is not as clear as in the case of traffic noise.

Several of the studies included in this review (Appendix 2, Table 4) investigated the extent to which one

specific characteristic of wind turbine noise, amplitude modulation (AM), contributes to annoyance

(Ioannidou et al., 2016; Lee et al., 2011; Schäffer et al., 2016, 2018). In addition, these studies also

examined the effect of noise frequency distribution and source origin on annoyance.

A laboratory experiment (Ioannidou et al., 2016) was designed to investigate how AM depth (experiment

1), AM frequency (experiment 2) and the interaction between AM type and depth (experiment 3) affect

annoyance. The subjects were presented with sound of varying modulation depth (more uniform or more

variable) at an intensity of 60 dB(A) for 30 seconds. Modulation depth was defined as the difference

between the maximum and minimum sound levels and ranged up to 12 dB(A) in different sound stimuli.

Both recorded and generated wind turbine sounds with frequencies between 200 and 1200 Hz were used.

During the experiment, the subjects were asked to rate the annoyance of different sound samples on a 10-

point scale. The study found that annoyance is influenced by AM depth. This means that the smaller the

range in which the noise level fluctuates, the less disturbing the sound is. It was also found that the

frequency of

22

the noise level fluctuates. The authors of the study recommend taking measures to make the sound of wind

turbines more uniform.

Amplitude modulation is also the focus of a laboratory experiment (Lee et al., 2011) in which participants

were presented with noise at different modulation levels and intensities: 35, 40, 45, 50 and 55 dB

(A). Recordings of the audible noise from a real wind turbine (1.5 MW capacity with a rotor diameter of 72

m) at different distances from the turbine at frequencies of 250–8000 Hz were used. This study also showed

that disturbance increased with increasing AM depth. The study by Schäffer et al. (2016) also found that the

presence of AM increased disturbance. In addition, Schäffer et al. (2016) found that at the same sound

intensity, disturbance was greater for wind turbines than for traffic noise (Appendix 2, Table 4).

Another study conducted later by the same researcher (Schäffer et al., 2018) also found that AM depth

increases disturbance. In addition, it was found that artificially enhanced low-frequency component sound

was more disturbing than wind turbine noise. The disturbance was not related to the gender or noise

sensitivity of the participants, but was higher with increasing age and lower with a more positive attitude

towards wind farms (Schäffer et al., 2018).

Audible noise from wind turbines and clinically expressed health effects

All observational studies included in this review examined the effects of audible noise (including low-

frequency noise) on various objectively measured health indicators. Eight cohort studies (Bräuner et al.,

2019a, 2019b, 2018; Poulsen et al., 2019a, 2019b, 2018a, 2018b, 2018c) and one case-crossover study

(Poulsen et al., 2018d) (Appendix 2, Table 2). All observational studies were conducted in Denmark and

are based on data from two study cohorts (groups of people) – a cohort of Danish nurses and a register-

based cohort of the entire Danish population.

The Danish Nurses Cohort was established in 1993 when a questionnaire was sent to female members of

the Danish Nurses Organisation who were at least 44 years old at the time. Initially, the cohort consisted of

28,731 nurses. The Danish Population Register was used to obtain information on their places of residence.

The Danish Nurses Cohort was used to study the incidence of stroke (Bräuner et al., 2019a), myocardial

infarction (Bräuner et al., 2018) and cardiac arrhythmias (atrial fibrillation) (Bräuner et al., 2019b) in

relation to wind turbine noise. Morbidity was determined based on Danish patient and cause of death

registries.

In all studies, A-weighted wind turbine noise was modelled at frequencies of 10–10,000 Hz and the annual

average Lden was calculated at the exterior facade of the residence under investigation within a 6 km radius

of the nearest wind turbine. In all three studies, wind turbine noise levels were low, with only 3% of those

studied being exposed to

23

wind turbine noise on the exterior facade of their homes above 29.9 dB (A). The studies found no link

between wind turbine noise and the occurrence of myocardial infarction and stroke.

The results of the study (Bräuner et al., 2019b) showed that an average annual night-time wind turbine

noise level on the exterior facade of a residential building of Lnight above 20 dB (A) may increase the

incidence of atrial fibrillation compared to those with less than 20 dB of wind turbine noise at night. A

similar association was found with the Lday and Leavning indicators, but no such association was found with the

24-hour indicator Lden. The studies were adjusted for important factors such as age, calendar year (when the

cohort was recruited), employment status, smoking status, alcohol consumption, physical activity, body

weight, other diseases, marital status, etc. The study also examined how traffic noise and air pollution

(NO2) could affect the associations. Adjustment did not significantly change the associations found (or the

lack thereof).

Poulsen's studies (Poulsen et al., 2019a, 2019b, 2018d, 2018a, 2018b, 2018c) were based on data from a

nationwide register-based cohort in Denmark. The study included all wind turbines in Denmark and all

residents who had lived for one year or more between 1996 and 2013 within 20 wind turbine tower heights

of a wind farm, with a random sample of 25% of all subjects who had lived in dwellings located 20–40

wind turbine tower heights from the nearest wind turbine in the wind farm. To assess exposure, A-weighted

night-time (10 p.m. to 7 a.m.) wind turbine noise was modelled (calculated) taking into account the type,

height, location, wind direction and other weather conditions of the wind turbine. Noise near residential

buildings outdoors was modelled at frequencies of 10 to 10,000 Hz (including low-frequency noise) and

low-frequency noise indoors (10–160 Hz) was modelled separately. Night-time noise near residential

buildings outdoors was divided into classes: less than 24; 24–30; 30–36; 36–

42 and over 42 dB (A) and low-frequency noise indoors below 5; 5–10, 10–15 and over 15 dB (A). All

analyses were adjusted for gender, calendar year of inclusion in the study and age. In addition, the analysis

was adjusted in different models for the educational level of the subjects, personal income, marital status,

labour market participation, average income in the region, type of housing (apartment or house), distance

from the road (with ≥ 5000 vehicles per day), and traffic load within a 500 m radius of the dwelling. All

data were updated according to changes in the person's place of residence during the follow-up period.

The study found that long-term night-time exposure to wind turbine noise outdoors near homes or low-

frequency wind turbine noise indoors does not affect the risk of developing diabetes and high blood

pressure (Poulsen et al., 2018b, 2018a). Similarly, no evidence was found that wind turbine noise was

associated with any of the adverse birth outcomes studied: preterm birth, low birth weight at term, or low

birth weight for gestational age at term (Poulsen et al., 2018c).

24

However, it was found that high long-term night-time exposure to wind turbine noise (10 to 10,000 Hz)

audible outdoors near residential areas increases the risk of taking sleeping pills and antidepressants. Noise

above 42 dB(A) increased the risk of taking antidepressants by 17%. Those over 65 years of age were most

affected. Low-frequency noise indoors (10–160 Hz) did not affect the purchase of sleeping pills and

antidepressants (Poulsen et al., 2019b).

Studies on heart attacks and strokes do not provide clear answers. For example, a study by Poulsen et al.

(2019a) found conflicting evidence on the link between wind turbine noise and heart attacks or strokes.

Participants who were exposed to night-time noise levels of 24–30 dB(A) and 30–36 dB(A) outdoors near

their homes had a higher risk of heart attack and stroke than those exposed to noise levels below 24 dB(A).

However, the increased risk was not statistically significant for exposure to louder wind turbine noise of

36–42 dB

(A) and ≥42 dB (A), indicating that there is no dose-response relationship. No association was found

between low-frequency noise indoors and the occurrence of heart attacks and strokes.

Poulsen et al., 2018d investigated whether short-term changes in wind turbine noise could affect

hospitalisation and death due to heart attacks and strokes. Noise levels were examined four days prior to

illness or on a reference day. The results did not provide convincing evidence of a link between short-term

night-time noise and myocardial infarction or stroke. No statistically significant associations were found in

the main analysis. However, the study by Poulsen et al. (2018d) concluded, based on additional analyses,

that short-term exposure to higher low-frequency noise indoors may trigger heart attacks or strokes

(Poulsen et al., 2018d).

An experiment conducted in Taiwan (Chiu et al., 2021) also suggests that wind turbine noise may

contribute to the development of cardiovascular disease. The experiment involved volunteers who lived up

to 500 m away from wind turbines. The subjects were divided into two groups: one group spent 30 minutes

outdoors 20 metres from the nearest wind turbine in the wind farm, and the other group spent 30 minutes

indoors 500 metres from the nearest wind turbine in the wind farm. The heart rate and heart rate variability

of the subjects were measured during the experiment using a portable electrocardiogram (ECG) recorder. In

the test area 20 m from the nearest wind turbine, low-frequency noise (20–200 Hz) ranged from 38.3 dB

(A) to 57.1 dB (A). The results of the experiment showed that the higher the wind turbine noise, the lower

the heart rate variability (HRV). Higher HRV usually indicates better physical adaptability and stress

tolerance, while lower HRV may indicate stress, fatigue or health problems. Thus, this study shows that

unusually high wind turbine noise (people do not usually live so close to wind turbines) can increase the

risk of cardiovascular disease. In addition, the study measured low-frequency noise inside residential

buildings

25

, which were located 124–330 metres from the nearest wind turbine. The low-frequency noise levels

measured inside these homes at 20–200 Hz ranged from 30.7 to 43.4 dB(A).

The impact of infrasound generated by wind turbines on human health

Evidence of the potential impact of wind turbine infrasound on human health comes from 13 experiments

in our study (Appendix 2, Table 3). The main health outcomes analysed in relation to wind turbine

infrasound were sleep quality (Liebich et al., 2022a, 2022b; Marshall et al., 2023), mental health (Ascone et

al., 2021; Małecki et al., 2023; Rosciszewska et al., 2025), disturbance and self-reported symptoms (Ascone

et al., 2021; Maijala et al., 2021; Marshall et al., 2023). In a series of experiments, Crichton et al.

investigated whether wind turbine infrasound causes disturbance and various symptoms, such as headaches,

pressure in the ears, etc., and to what extent the disturbance and reporting of symptoms are influenced by

information circulating in the media and social media that infrasound is dangerous to health (Crichton et al.,

2015, 2014b, 2014a; Crichton and Petrie, 2015a, 2015b).

Five experiments investigating the effects of wind turbine infrasound alone (without accompanying audible

noise) were included in the present study (Ascone et al., 2021; Crichton et al., 2014b; Małecki et al., 2023;

Marshall et al., 2023; Tonin et al., 2016) (Appendix 2, Table 3).

The study with the highest quality of evidence among these is the experiment conducted in an Australian

sleep laboratory (Marshall et al., 2023). It was a randomised, double-blind trial conducted in an isolated

sleep laboratory under controlled conditions (noise levels and other factors are precisely known in the

laboratory), which was designed in the style of a studio apartment. The study included people who, based

on a questionnaire, considered themselves to be noise-sensitive. The study involved 37 noise-sensitive but

otherwise healthy adults aged 18–72, 51% of whom were women.

The subjects underwent three test periods. During each test period, which began around noon, the

participants remained in one of three noise conditions (infrasound, placebo, traffic noise) for 72 hours

(including three nights) without leaving the laboratory. The laboratory was furnished as a bedroom with a

private bathroom. After each test period, the subjects spent 10 days in their normal environment. Neither

the subjects nor the research team knew whether the subject was being exposed to infrasound or not. The

generation of infrasound or placebo sound was controlled from a separate room by engineers who did not

meet the participants. The statistical analysis was also performed by researchers who did not know which

condition corresponded to infrasound. The exposure conditions were labelled numerically (1 vs. 2 vs. 3).

26

The subjects slept in random order under the following conditions:

1. Infrasound (test condition). During the night, the subjects were exposed to infrasound with a

frequency of 1.6–20 Hz and a maximum intensity of 90 dB (Z), simulating wind turbine infrasound

at a higher than normal level. This sound was inaudible to the subjects.

2. Placebo (negative control). No additional sounds were presented to the subjects, but a loudspeaker

was placed in the room so that the subjects did not know whether sound was being generated or

not. However, the placebo group was exposed to background noise from the laboratory air

conditioning system. The average background noise during the night was 39 dB (A), which

corresponds to a noise level of 80–85 dB (Z). The background noise was dominated by frequencies

below < 1 Hz.

3. Traffic noise (positive control). The subjects were exposed to traffic noise with an average

intensity of 40–50 dB (A) during the night and a maximum intensity of 70 dB (A). This noise was

audible.

The physiological and psychological indicators measured included objectively and subjectively measured

sleep indicators, cardiovascular indicators (24-hour blood pressure and heart rate, endothelial function and

pulse wave velocity), stress hormone and insulin levels in blood and urine, neurobehavioural and

psychological indicators, questionnaires on wind turbine syndrome symptoms and mental well-being.

Exposure to infrasound did not worsen any subjective or objective measured indicators. For some

indicators, infrasound improved measured health indicators, but according to the researchers, these

statistically significant associations could be coincidental. For example, systolic blood pressure was lower

and the Warwick–Edinburgh Mental Well-being Index showed that the subjects felt better when exposed to

infrasound compared to the placebo.

The study concluded that wind turbine infrasound does not disturb people's sleep, does not cause symptoms

of 'wind turbine syndrome', does not impair measured cardiovascular health indicators, and does not impair

people's mental well-being (Marshall et al., 2023). The results of the study can be considered well-proven.

The longest-term wind turbine infrasound experiment was conducted in Germany with 38 participants

(Ascone et al., 2021). An infrasound-generating device or a placebo device was placed in the bedrooms of

the participants for 28 nights. The participants were randomly divided into two groups: the first group (23

people) was exposed to infrasound at a frequency of 6 Hz and an intensity of 80–90 dB (Z) during the

night, while the placebo group (15 people) was not exposed to infrasound. Somatic and psychiatric

symptoms, noise sensitivity, sleep quality, cognitive ability and brain structure (using MRI) were measured

before and after the intervention. Exposure to infrasound did not affect self-reported health, sleep quality or

mental

27

abilities. Changes in brain grey matter were observed, but these cannot be interpreted as either harmful or

beneficial, and this finding may not be related to infrasound (Ascone et al., 2021).

The aim of the study by Malecki et al. (2023) was to test whether infrasound amplitude modulation (AM)

affects students' mental performance. To this end, students participating in a university experiment were

divided into three groups, which were exposed to the following sounds: 1) Recorded wind turbine noise

filtered to an infrasound intensity of 83 dB (G)/47 dB (A) and an AM depth of 4 dB at 1 Hz; 2) Synthesised

infrasound with an intensity of 78 dB (G)/46 dB (A) at 5–20 Hz and no AM; 3) Background noise with an

intensity of 63 dB (G)/43 dB (A) (traffic noise, speech in an educational institution). The results of the

study did not show significant differences in cognitive test results or in the number of reported unpleasant

sensations or complaints between different sound conditions when men and women were analysed

separately. Women reported discomfort and various complaints more than men (Małecki et al., 2023).

Between 2014 and 2015, five experiments were conducted at the University of Auckland in New Zealand

to investigate how false information about infrasound disseminated through the media and the internet

affects the onset of symptoms and the experience of disturbance. (Crichton et al., 2015, 2014b, 2014a;

Crichton and Petrie, 2015a, 2015b) (Appendix 2, Table 3).

In the experiment by Crichton et al. (2014b), 54 subjects were randomly divided into two groups. One

group was led to expect that infrasound has harmful effects based on real information circulating on the

internet, while the other group was led to expect that infrasound is not harmful. To create a high negative

health impact expectation, videos available on the internet were shown, containing descriptions of

symptoms that people associated with the operation of wind farms. To create low expectations of negative

health effects, a video was shown featuring interviews with experts who presented the scientific view that

infrasound generated by wind farms does not cause symptoms. After expectations were formed, the

subjects were exposed to 10 minutes of generated infrasound (5 Hz, 40 dB) or placebo sound. All

participants were told that they were exposed to infrasound during both 10-minute sessions. Neither the

participants nor the researcher conducting the experiment knew when the real infrasound was being played

and when the placebo sound was being played. The participants rated the presence of symptoms on a seven-

point scale before and during exposure. The symptoms assessed were those commonly associated with

wind farms on the internet: headache; pressure in the ears; ringing in the ears; itchy skin; pressure in the

sinuses; dizziness; pressure in the chest; perceived vibration; heart palpitations; nausea; fatigue; weakness.

Other random symptoms were also assessed for control purposes. Blood pressure and heart rate were also

measured. In addition, participants were asked to assess the accuracy of the following statement

28

the following statement: "I am concerned about the health effects of the noise caused by wind turbines"

both before and after watching the video. The group with high expectations of negative health effects was

significantly more concerned about health effects and reported significantly more and more intense

symptoms during the experiment compared to the pre-experiment test, regardless of actual exposure to

infrasound. More symptoms were reported that the subjects had been informed were associated with

infrasound. In the low-expectation group, the number and intensity of symptoms reported did not change

compared to the pre-test. Infra-sound exposure did not affect heart rate or blood pressure in either group.

The study shows that real information circulating on the internet about the negative health effects of infra-

sound increases the occurrence of self-reported symptoms (Crichton et al., 2014b).

Similar studies (Crichton et al., 2015, 2014a) reported a placebo effect with positive expectations. The

subjects were divided into two groups – one group was given negative expectations that wind turbines have

a negative impact on health, while the other group was given positive expectations that wind turbines have

a positive impact on health. Both groups were exposed to generated infrasound (9 Hz, 50.4 dB) and audible

sound recorded 1 km away from the wind farm (43 dB) in 7-minute sessions. In the negative expectation

group, symptom reporting increased during the session, mood deteriorated, and disturbance increased. In

the positive expectations group, symptom reporting decreased, mood improved, and distress decreased

compared to what was reported before the session.

Crichton and Petrie (2015b) used a similar study design in their study, but after the experiments with the

positive and negative expectation groups, the information given to the groups was changed and the listening

tests were repeated. When the group with negative expectations was given positive information about the

benefits of wind turbines in the repeat test, they reported fewer symptoms and their mood improved.

Similarly, those who had heard the positive information first showed a worse mood and more symptoms

after receiving the negative information. The results show that the availability of positively worded health

information can reverse or reduce the impact of negative expectations arising from media warnings about

the health risks of wind turbines (Crichton and Petrie, 2015b).

Crichton and Petrie (2015a) investigated the effectiveness of providing a nocebo explanation in changing

expectations of negative health effects in their experiment. All participants were made to expect negative

health effects. The subjects were then randomly divided into two groups: 1) subjects were given

information that the health effects of infrasound are biologically based vs. 2) subjects were told that the

health effects are the result of a nocebo effect. After receiving the negative information, the number of

reported symptoms increased and intensity in both groups compared to baseline, which

showed

29

the effectiveness of manipulation in this experiment. In the biological explanation group, the increase in

symptom reporting persisted during the second session. In the nocebo explanation group, however, the

number and intensity of symptoms decreased and mood improved during the second session. The

experiment shows that false information found in the media increases the occurrence of symptoms and

concerns about health. It also shows that providing an explanation of the nocebo effect can reduce the

reporting of symptoms associated with wind turbines. Participants in both groups found the explanation

they were given to be understandable, reasonable, convincing and correct (Crichton and Petrie, 2015a).

A similar study to Crichton's experiments was conducted by Tonin et al. (2016) using stronger infrasound

and a larger sample (72 subjects aged 17–82). Variable wind turbine infrasound was simulated at a

frequency of 0.8–40 Hz with a maximum intensity of 91 dB (Z). The subjects were exposed to it for

for 23 minutes or were given placebo noise through headphones. The results also support the existence of

the nocebo effect. In the infrasound group, the reporting of symptoms even decreased during infrasound

exposure. However, no statistically significant correlation was found between the nature of the information

provided before the listening test (expectations of health effects or expectations that there would be no

health effects) and the results. However, the result depended on the previously formed opinion about the

health effects of infrasound. Those who believed that infrasound had a negative effect on health reported

more symptoms (Tonin et al., 2016).

Rosciszewska et al. (2025) investigated the impact of wind turbine noise on cognitive performance. The

subjects were randomly divided into three groups: 1) exposure to wind turbine noise. This experiment used

real recorded wind turbine noise from a 2 MW wind turbine 500 metres away, which contained both

infrasound and audible noise. The sound pressure level used was 65.4 dB (Z) ( corresponding to a sound

level of 38.5 dB (A)). The wind turbine noise contained AM (average frequency 0.8–1 Hz, depth ~6.9 dB);

2) exposure to traffic noise. Recorded road traffic noise with an intensity of 65.4 dB (Z) (corresponding to a

sound level of 56.8 dB (A)) was used; 3) exposure to background noise only. Their study showed that

short-term exposure to wind turbine noise (at a level corresponding to the actual situation at a distance of

500 m) did not have a statistically significant effect on the cognitive performance (brain functions,

attention, thinking) of the subjects. Similarly, no statistically significant differences were found in the

results of inductive reasoning tests (accuracy, test completion time, average reaction speed) between

different noise exposure conditions (Rosciszewska et al., 2025).

The impact of visual aspects of wind turbines on human health

This study included one systematic review (Freiberg et al., 2019b) and one experiment (Murcia et al., 2017)

on visual aspects (Appendix 2, Table 1 and Table 4).

30

The health effects associated with the visual aspects of wind turbines were addressed in a systematic review

published in 2018, which included all epidemiological studies without time or language restrictions

published by 2017 (Freiberg et al., 2019b). A total of 17 studies were included in the descriptive analysis

and six studies in the meta-analysis. The quality of the studies was rated as high for five studies, acceptable

for three studies, and low for the remaining studies. The review addressed the impact of the following

visual aspects of wind turbines on sleep quality and disturbance: direct visibility from the place of

residence; altered view of the landscape; flashing lights on the blades; obstacle markings; shadow flicker;

reflections from the blades.

Disturbance caused by direct visibility, shadow flicker and flashing lights was statistically significantly

associated with an increased risk of sleep disturbance. The study found that altered views of the landscape,

obstacle markings and light reflections from wind turbine blades can also disturb people (Freiberg et al.,

2019b).

Studies have shown that when wind turbines are audible but not visible from the home, this significantly

reduces disturbance. However, when wind turbines were both visible and audible, the noise was considered

more disturbing than the visual aspects of the wind turbines. The study found that visual disturbance may or

may not depend on the distance of the wind turbines from residential buildings. Two studies showed that

disturbance from the visual aspects of wind turbines decreased with distance, while two other studies found

no significant effect of distance on visual disturbance (Freiberg et al., 2019b).

In an experiment by Murcia et al. (2017), electroencephalographic (EEG) measurements were taken to

measure both objectively and subjectively the brain's reactions to landscape visuals. EEG records the

electrical impulses generated by nerve cell activity in the brain. Sixty different images were used as stimuli.

These were divided into three groups: images with wind turbines and the same images without them;

images with a solar park and without it; and images with a nuclear power plant and without it. Both the

brain activity measurements and the questionnaire responses showed that people were not more disturbed

by images with wind turbines and solar panels. However, clear and significant differences were found when

viewing landscapes that did or did not include a nuclear power plant. The nuclear power plant evoked

negative emotions according to both the questionnaires and EEG measurements. The results of the study

may have been influenced by the fact that only 14 subjects participated in the study and most of them had a

positive attitude towards renewable energy (Murcia et al., 2017).

31

Discussion of results

Audible noise from wind turbines and sleep

Previous review studies have found a link between audible noise from wind turbines and sleep indicators

(Ellenbogen et al., 2024; Karasmanaki, 2022; Schmidt and Klokker, 2014; Teneler and Hassoy, 2023).

However, the articles included in our study do not confirm this unequivocally. Our study shows that the

link with self-reported sleep disturbances is better proven than with objectively measured sleep

disturbances (Godono et al., 2023; Liebich et al., 2021). Liebich et al. (2021) did not find an effect of noise

on objectively measured sleep parameters in their review, but found that self-reported sleep parameters may

be affected (Liebich et al., 2021). Godono et al. (2023) found a link between self-reported sleep indicators

and wind turbine noise. The cohort study by Poulsen et al. (2019b) included in our study concluded that

there is no clear link between the use of sleeping pills and wind turbine noise in the general population, but

found that long-term exposure to audible wind turbine noise increased the risk of sleep medication use in

subjects over 65 years of age, starting at a night-time noise level of 30 dB outdoors near the home. In this

regard, there was also a dose-response relationship, which strengthens the validity of this result (Poulsen et

al., 2019b). However, it was not possible to adjust the analyses in this study for people's attitudes,

knowledge or perception of disturbance, and therefore the association found between wind turbine noise

and sleep disturbance may have been influenced by factors other than noise. The study did not take into

account differences in the sound insulation of residential buildings, which affects people's actual exposure

to wind turbine noise at night.

The impact of wind turbine noise on self-reported sleep disturbance may be direct, but it may also be

related to greater disturbance, which in turn may be influenced by other factors (Teneler and Hassoy,

2023). For example, a systematic review of the impact of visual aspects (Freiberg et al., 2019b) found that

the risk of sleep disturbance (e.g. insomnia or reduced sleep quality) increased when people could see wind

turbines from their homes, were affected by shadow flicker, or saw the night-time lights on the blades of

the wind turbines.

Studies included in our work: two systematic reviews (Godono et al., 2023; Liebich et al., 2021), a cohort

study (Poulsen et al., 2019b) and two experiments (Liebich et al., 2022a, 2022b), it can be concluded that

night-time wind turbine noise of up to 30 dB (A) outdoors and up to 25 dB (A) indoors does not increase

the risk of sleep disturbance, even in noise-sensitive and older people (Liebich et al., 2022a, 2022b, 2021;

Poulsen et al., 2019b). This conclusion is based on a very

32

a small number of studies. Further research is certainly needed to investigate the relationship between sleep

indicators and audible noise from wind turbines in order to verify the validity of this claim.

Measurements taken in the yards of residential buildings located near the Saarde wind farm in Estonia (at a

distance of 1060 to 3540 m from the nearest wind turbine) showed that the audible noise at night in the

yards near the residential buildings ranged from 27.7 to 40.5 dB (A), but these measurements did not

distinguish the noise from the wind turbines from other background noise (wind noise, rain, birdsong,

traffic, etc.). The maximum night-time noise level of 40.5 dB (A) was probably influenced by the nearby

river and its dam. Measurements taken indoors showed night-time noise levels of 15.6 dB (A) to 24.1 dB

(A) (Health Board, 2025b).

Noise levels of 33.8–37.6 dB (A) were measured at night in residential areas near the Sopi-Tootsi wind

farm, which were also affected by natural background noise that cannot be distinguished from wind turbine

noise in measurements. Night-time noise levels indoors ranged from 15.3 to 18.2 dB(A) (Health Board,

2025a).

Poulsen et al. 2019b also studied the effect of low-frequency noise (10–160 Hz) on sleep disturbances. No

effect of low-frequency noise indoors on the consumption of sleeping pills was found at any of the noise

levels studied (up to 20 dB (A)) (Poulsen et al. 2019b).

Disturbance caused by wind turbines

Although disturbance cannot be considered a clinically significant health effect, it depends on people's

well-being and can therefore be considered part of the WHO definition of health, according to which health

is "a state of complete physical, mental and social well-being and not merely the absence of disease or

infirmity" (WHO, 2025). Disturbance can also act as a mediating factor between other health effects,

including influencing the development of more serious conditions such as cardiovascular disease through

stress (Basner et al., 2014; Freiberg et al., 2019a).

Several previous review studies have shown that the louder the noise from wind turbines, the more

disturbed people feel (Knopper et al., 2014; Teneler and Hassoy, 2023; van Kamp and van den Berg, 2021,

2018). A systematic review included in our study (Guski et al., 2017) also showed that the likelihood of

disturbance increases with increasing wind turbine noise, but this relationship is not as clear as in the case

of traffic noise. At the same time, the prevalence of disturbance is very uneven, and the relationships

between wind turbine distance, noise, and other indicators vary greatly between studies.

A previous review study found that wind turbine noise was more disturbing than noise from other sources

(Teneler and Hassoy, 2023). The studies included in this work did not provide a single answer to this

question.

33

question. For example, Schäffer et al. (2018) used generated

"pink noise", wind turbine noise and noise with an artificially increased low-frequency component. Pink

noise was the least disturbing, followed by wind turbine noise, and noise with an artificially increased low-

frequency component was the most disturbing (Schäffer et al., 2018). An experiment conducted in Finland

showed that the audible sounds of a wind farm were more disturbing than the sounds of the ocean (Maijala

et al., 2021). Schäffer et al. (2016) showed in their experiment that, at the same sound level, the subjects

rated wind turbine noise as more disturbing than traffic noise. This experiment simulated the noise of a 2

MW Vestas V90 wind turbine, to which sounds with generated amplitude modulation were added (Schäffer

et al., 2016). However, in an experiment by Rosciszewska et al. (2025) with recorded wind turbine noise

(from a 2 MW wind turbine 500 metres away), which also contained AM, the subjects did not perceive the

wind turbine noise as more disturbing or stressful than the recorded road traffic noise. According to the

authors, this may have been due to the fact that the subjects did not know what kind of noise they were

exposed to. The participants did not know whether or what kind of noise they were being presented with,

nor did they know the purpose of the noise presentation. In a survey conducted after the experiment, no one

identified the wind turbine noise, most described it as "some noise", and some compared it to ocean waves

or an aeroplane. In the experiment by Rosciszewska et al. (2025), both road traffic noise and wind turbine

noise were unfiltered and had the same intensity of 65.4 dB (Z). When A-weighted, the same sound level

for wind turbines is 38.5 dB (A) and for road traffic noise 56.8 dB (A). The results of the experiment by

Rosciszewska et al. (2025) support the hypothesis of Crichton et al. (2015) that the disturbance is caused

not so much by the noise or infrasound itself, but rather by the negative expectation created by the media

that wind turbine noise is disturbing and hazardous to health. Crichton et al. (2015) divided the subjects into

two groups: one group was shown a video that created the expectation that the infrasound from wind

turbines is harmful, while the other group was given the expectation that it is beneficial. Both groups were

presented with both the generated infrasound and the audible sound recorded 1 km away from the wind

farm. Disturbance was assessed before and during the listening session. In the negative expectation group,

disturbance increased during the session, while in the positive expectation group, disturbance decreased,

regardless of the noise or infrasound presented. The study shows that negative information causes

disturbance from wind farms, while positive information reduces disturbance. A review article (McCunney

et al., 2014) also concluded that wind turbine noise plays only a minor role in causing annoyance compared

to other factors that influence people's willingness to experience annoyance in relation to wind turbines.

Pohl et al. (2018) also found that noise-related annoyance was influenced to a small extent by the distance

to the nearest wind turbine and the intensity of the sound, but was most influenced by the extent to which

people felt that the wind turbine planning process had been conducted fairly and transparently.

34

According to Knopper and Ollson (2011), disturbance may be more strongly related to the visual aspects of

wind turbines and people's attitudes towards them than to the noise they generate (Knopper and Ollson,

2011). Studies have shown that if wind turbines are audible but not visible from homes, this significantly

reduces disturbance. However, when wind turbines were both visible and audible, the noise was considered

more disturbing than the visual aspects of the wind turbines (Freiberg et al., 2019b).

One of the important visual aspects affecting disturbance is the presence of shadow flicker. Shadow flicker

is not continuous, but occurs at specific times: the sun must be low enough and the wind turbine rotor must

be in the right direction for the dwelling to be in the shadow cast by the wind turbine. Shadow flicker is

disturbing, but no clinically significant health effects have been found (Freiberg et al., 2018; Knopper et al.,

2014; Teneler and Hassoy, 2023). Wind turbines should preferably be located so that shadow flicker does

not disturb residents. If this is not possible, special software can be used to assess the extent and impact of

the shadow, and mitigation measures can be implemented. As a mitigation measure, it is recommended to

set time limits on the shadow. In residential areas, shadowing should not exceed 30 hours per year or 30

minutes per day in the worst case. Wind turbines can be programmed to stop at times when shadowing

limits may be exceeded (World Bank Group, 2015).

People living near wind turbines report various health symptoms and often

symptoms of 'wind turbine syndrome'. These health symptoms described by people may be due to

disturbance, which in turn causes stress. These symptoms may not be related to wind turbine noise

(Knopper and Ollson, 2011).

An article (Pohl et al., 2018) analysed whether the complaints and disturbance experienced by people living

near wind turbines are directly related to wind turbine noise or whether other factors are significant. The

conditions under which disturbance occurs were also investigated. People living near a wind farm (1.25 to

2.89 km from the nearest wind turbine) in Germany were surveyed in 2012 and 2014. A total of 212 people

participated in the first year and 133 in the second year. The subjects were exposed to wind turbine noise

near their homes outdoors at 25–30 dB(A) or 30–35 dB(A), which came from Enercon E-82 wind turbines

that were 150 m high and had a capacity of 2 MW.

Of all residents, 69.3% heard the noise from the wind turbines and 30.7% did not. Nearly half (53.6%) of

the subjects experienced disturbing noise once a week, 20.9% once a month and 13.6% almost every day.

18.4% were not disturbed at all. Only a small percentage of residents reported being severely disturbed by

the noise from the wind turbines, which decreased over time: in 2012, one tenth (9.9%) of residents were

severely disturbed, but two years later, only 6.8% were. It is noteworthy that the majority of the residents

who were severely disturbed

35

residents (75.0%) had already been opposed to the wind farm before it was built, either passively or

actively. They also felt that the wind turbine planning process was unfair and had little knowledge of how

to improve their situation themselves. At the time of the survey (three years after the wind turbines were

erected), there were slightly more supporters of the wind farm (40.2%) than opponents (35.8%) living in the

vicinity of the wind farm. Only a small proportion, 16.7%, were ambivalent, and 7.4% had no opinion

about the wind farm, but opponents were more active than supporters (Pohl et al., 2018).

In the study by Pohl et al. (2018), the participants were asked to observe when wind turbine noise disturbed

them the most and to record the disturbing noise. Disturbing noise occurred most frequently in the evenings

(33.6%) and at night (18.2%). The disturbance was most prevalent when people were sleeping (30.0%) or

resting (24.5%). This caused anger in 39.1% of cases. The disturbance occurred most often when there was

a westerly wind (68.2%) and in humid weather (30.9%). The most frequently used measures to reduce the

impact of noise were talking to family members, friends and neighbours (32.1%), closing windows

(25.9%), changing location (indoors 11.8% and outdoors 7.1%) and turning up the volume of the

radio/television (7.5%). The disturbing noise from wind turbines was mainly described as irregular and

fluctuating in volume. Of those disturbed, 71.6% described it as a pulsating hiss. Complaints were not

caused by the absolute loudness of the noise, but by the variation in sound intensity, i.e. amplitude

modulation (Pohl et al., 2018).

In a study by Pohl et al. (2018), more residents complained of physical and psychological symptoms caused

by traffic noise (16%) than by wind turbine noise (10%, 7% two years later). Both noise sources caused

similar symptoms: reduced work capacity and concentration, increased irritability/anger, negative mood

and disturbed sleep.

An important factor influencing disturbance is also the financial benefit derived from the wind farm (Taylor

and Klenk, 2019; van Kamp and van den Berg, 2018). Farmers who benefited financially from wind

turbines were very little disturbed by them and reported fewer health and sleep problems than the rest of the

study participants, even though they lived closer to the turbines and were exposed to higher noise levels

than the other respondents. However, economic benefits may not be the only factor in reducing disturbance.

Different attitudes, education and greater control over the location of wind turbines may also have played a

role (van Kamp and van den Berg, 2018).

Most of the experiments included in this study showed that disturbance is greater when the audible noise

from wind turbines contains amplitude modulation (AM). The deeper the AM, the greater the disturbance

(Ioannidou et al., 2016; Lee et al., 2011; Maijala et al., 2021; Schäffer et al.,

36

2018, 2016). Australian researchers also found in their review article that AM is an important characteristic

of wind turbine noise that can increase disturbance and recommended testing AM measurements and

setting limits for AM in Australia to reduce disturbance (Davy et al., 2020).

In summary, the relationship between wind turbines and disturbance depends on several factors, such as

expectations/knowledge of the health effects of wind turbines, perceived fairness and transparency of the

planning process, economic benefits, visual aspects and noise. It is likely that a combination of all these

factors causes annoyance, and reducing just one factor (e.g. noise) may not reduce annoyance.

Regardless of the reasons, a certain degree of disturbance among the population can be expected, as with

any other project involving changes to the local environment. The acceptable level of disturbance is a

political decision that should be made by weighing the benefits of wind energy against its negative effects

(Knopper and Ollson, 2011).

The impact of infrasound from wind turbines on health

Wind turbine noise always includes infrasound. Opinions circulating on the internet and in public

discussions suggest that this component of wind turbine noise may be harmful to health. We have compiled

a list of experiments (Appendix 2, Table 3) that have experimentally investigated the health effects of the

infrasound component of wind turbines, either separately or in combination with the audible noise of wind

turbines. The claims circulating on the internet and in the media have been the motivation for conducting

these experiments. Therefore, the study focused mainly on health indicators that are the subject of

widespread speculation: various symptoms associated with infrasound (e.g. pressure in the ears, headaches,

etc.), disturbance, sleep disorders and effects on cognitive ability. The experiments included in our study

showed that exposure to infrasound at any of the sound pressure levels studied (up to 91 dB (Z)) did not

impair any of the health indicators measured in the studies. No effects were found on sleep disturbances

and sleep quality (Ascone et al., 2021; Liebich et al., 2022a, 2022b; Marshall et al., 2023). No effect on

disturbance and mental well-being was found (Crichton et al., 2015; Maijala et al., 2021; Marshall et al.,

2023; Rosciszewska et al., 2025). No effect was found on cognitive performance (Ascone et al., 2021;

Małecki et al., 2023; Rosciszewska et al., 2025). No effect was found on cardiovascular parameters (heart

rate, blood pressure) (Maijala et al., 2021; Marshall et al., 2023). No effect was found on the occurrence of

symptoms that people themselves associate with wind turbine infrasound (Crichton et al., 2014b, 2014a;

Crichton and Petrie, 2015b, 2015a; Maijala et al., 2021; Małecki et al., 2023; Marshall et al., 2023). A

previous review of infrasound and low-frequency sound (van

37

Kamp and van den Berg, 2018) found that there is no evidence of a specific health impact from the

infrasound component of wind turbines.

According to measurements taken in Estonia, the unfiltered infrasound intensity at a frequency of 6.3 Hz

measured at night in the interiors of four residential buildings near the Saarde wind farm (2070 to 2530 m

from the nearest wind turbine) was 35–50 dB (Z) (Health Board, 2025b) and in the vicinity of the Sopi-

Tootsi wind farm (1400 to 2560 m from the nearest wind turbine) 37.7 to 58.2 dB (Z) (Health Board,

2025a). Ascone et al., 2021 did not find any significant effect in a long-term experiment (28 consecutive

nights) with a similar frequency (6 Hz) and intensity (80–90 dB (Z)) of infrasound. A more realistic wind

turbine infrasound level was used in the experiment (Crichton et al., 2014b). The subjects were exposed to

generated infrasound at a frequency of 5 Hz and an intensity of 40 dB for 10 minutes. Exposure to

infrasound did not affect symptoms such as headache, pressure in the ears, blood pressure and heart rate.

However, the same experiment found a significant effect on the occurrence of self-reported symptoms

based on information provided to the subjects prior to the study. The subjects who were shown a video

about the negative health effects of infrasound, based on real information circulating on the internet, began

to report the symptoms mentioned in the video, regardless of whether or not they had been exposed to

infrasound (Crichton et al., 2014b). It should also be noted that the measurements taken in Estonia do not

distinguish between infrasound from wind turbines and other possible sources of infrasound in the home

(fans, air heat pumps, etc.), which may have increased the measured infrasound level, and the actual level

of infrasound generated by wind turbines may be lower than measured in homes located near the Saarde

and Sopi-Tootsi wind farms.

Crichton's studies (Crichton et al., 2015, 2014b, 2014a; Crichton and Petrie, 2015b, 2015a) clearly show

that expectations can influence the reporting of symptoms and mood in both positive and negative

directions. Although actual exposure (e.g., to infrasound) may be harmless, the expectation or belief that it

is harmful causes people to experience real symptoms. This phenomenon is called the nocebo effect. There

is a lot of misinformation on the internet about the health effects of infrasound, and it is not possible to

correct or change this information. Crichton's research shows that explaining the nature of the nocebo effect

to people or providing them with positive information about wind turbines as a counterbalance reduces

disturbance and the onset of symptoms.

As no effects have been observed to date even at much higher infrasound levels, which are commonly

found in homes located near wind farms, there is no reason to assume, based on this study, that infrasound

from wind turbines that complies with the limits in force in Estonia

38

affect human health. Based on this study, there is also no reason to recommend changes to the current

infrasound limits.

Infrasound is common in our environment. For example, a study by Staniek et al. (2013) showed that

infrasound from the ventilation shaft of an operating coal mine was stronger than that from a wind farm 750

m away from the nearest wind turbine (Staniek et al., 2013). Infra-sound from large wind turbines (with a

capacity of over 2 MW) can range from 59 to 107 dB (G) at distances of 68 to 1000 metres. Similar

infrasound levels can also be found 350 metres from a gas-fired power plant (74 dB (G)), 70 metres from

major roads (76 dB (G)) and 25 metres from the coastline (75 dB (G)) (Schmidt and Klokker, 2014).

Claims about the health effects of infrasound are circulating in public debates and the media

One of the most common theories circulating on the internet is that of Nina Pierpont. According to Pierpont

(2009), infrasound can reach the inner ear and stimulate the balance organs (vestibular organs), especially

the otoliths – sensory structures that respond to movement and gravity. According to her, this unusual or

constant stimulation can cause symptoms such as dizziness, balance disorders, nausea and nystagmus

(involuntary eye movement). Nina Pierpont has coined the term 'wind turbine syndrome'. This is not a

medical diagnosis, but a term coined by Nina Pierpont to summarise people's complaints associated with

wind turbines (Pierpont, 2009). According to a review study (Schmidt and Klokker, 2014), Pierpont's

(Pierpont, 2009) source is a case series study, which is well suited for proposing hypotheses, but

methodologically, this approach does not allow for the identification of a causal link between the

influencing factor under investigation and the health outcome. Harrison (2015) has discussed whether the

effect of infrasound generated by wind turbines on the vestibular organs of the inner ear could be

biologically possible at all. The study concludes that acoustic activation of the vestibular system is possible

from a sound level of 110 dB (Z) and, based on animal experiments, from 120 dB

(Z). Comparing this with the actual infrasound levels in residential areas (approximately 60 dB (Z)), the

study finds that the impact of low-frequency and infrasound from wind turbines on the balance organs is

not biologically justified (Harrison, 2015).

Measurements taken inside Estonian homes (Health Board, 2025a, 2025b) show that the strongest

uncorrected wind turbine noise levels are in the 0.8–2 Hz range, mostly reaching 50 to 66 dB (Z). At higher

frequencies, the sound intensity decreases, as can be seen in Figure 4. The figure shows, for example, the

results of noise level measurements at all frequencies, both at night and during the day, in the residential

building closest (1400 m) to the Sopi-Tootsi wind farm.

39

Figure 4. Results of uncorrected sound level measurements inside a residential building 1400 m from the

Sopi-Tootsi wind farm (Health Board, 2025a).

The occurrence of wind turbine syndrome symptoms (e.g. headache, fatigue, dizziness, nausea, pressure in

the ears, etc.) in contact with infrasound was tested experimentally in studies (Crichton et al., 2014b,

2014a; Crichton and Petrie, 2015b, 2015a; Maijala et al., 2021; Małecki et al., 2023; Marshall et al., 2023).

None of these experiments found that exposure to infrasound caused the symptoms mentioned or any

symptoms at all (Appendix 2, Table 3).

Public discussions often refer to the hypothesis of a Portuguese research group that high levels of

infrasound and low-frequency sound cause “vibroacoustic disease” (VAD) (Alves-Pereira and Castelo-

Branco, 2007; Castelo-Branco and Alves-Pereira, 2004). Alves-Pereira and Castelo-Branco (2007) and

Castelo-Branco and Alves-Pereira (2004) have argued that VAD occurs in people who work in places with

high levels of infrasound and low-frequency sound, specifically in aeroplanes, trains, bars, discotheques,

underground railways and ordinary cars. They describe vibroacoustic disease as a whole-body pathology

that can manifest itself in a wide variety of diseases and symptoms, such as respiratory, digestive, nervous

system and cardiovascular diseases and symptoms, cancer, autoimmune diseases and endocrine disorders,

but its root cause is the abnormal proliferation of collagen and elastin in the intercellular matrix. VAD is

also not a recognised medical diagnosis (van Kamp and van den Berg, 2018). VAD is diagnosed and

discussed only by a small group of researchers who publish mostly in peer-reviewed journals, mainly

referencing each other (Chapman

40

and George, 2013). Most of the sources on which this concept is based are conference presentations and

very old, unreviewed publications (dating back to 1928), some of which are in Russian. VAD as a diagnosis

has so far remained a theoretical hypothesis that has not been confirmed by other researchers.

To our knowledge, no scientific research has been published in a peer-reviewed scientific journal

investigating the connection between infrasound or low-frequency sound generated by wind turbines and

VAD. To our knowledge, the main authors of this theory, Mariana Alves-Pereira and A.A Nuno Castelo-

Branco, have not published any articles in peer-reviewed scientific journals on the relationship between

infrasound or low-frequency sound generated by wind turbines and health. Therefore, there is no

scientifically accepted confirmation of the claims circulating among the population that Mariana Alves-

Pereira's research has shown that infrasound and/or low-frequency sound from wind turbines have

significant health effects.

Audible noise generated by wind turbines and clinically manifested health effects Disturbance and sleep disturbances caused by audible noise may contribute to the development of

diagnosable diseases (Basner et al., 2014). This has already been demonstrated in the case of traffic noise

(van Kempen et al., 2018). In the case of traffic noise, the best-documented link is between noise from roads

(trucks, motorcycles, trams, cars, etc.) and cardiovascular disease (van Kempen et al., 2018).

Similar links have not yet been found for the audible noise from wind turbines, but there are indications that

such links may exist. The results of Bräuner et al. (2019b) showed that night-time, daytime and evening

average wind turbine noise on the exterior facade of a dwelling Lnight, Lday, and Leavning above 20 dB (A) may

cause an increase in atrial fibrillation, but no similar association was found with the 24-hour indicator Lden

( Bräuner et al., 2019b). Studies with a cohort of Danish nurses found no association between wind turbine

noise and stroke and heart attack in women over 44 years of age (Bräuner et al., 2019a, 2018), but a cohort

covering the entire Danish population (Poulsen et al., 2019a) yielded conflicting results, making it difficult

to draw specific conclusions. References to a possible link between audible noise from wind turbines and

heart attacks and strokes were also found by Poulsen et al. (2018d), who investigated the effect of short-

term exposure to higher noise levels on the incidence of heart attacks and strokes (Poulsen et al., 2018d).

In a cohort covering the entire Danish population, it was found that greater long-term night-time exposure

to audible noise from wind farms (at frequencies between 10 and 10,000 Hz) near residential areas

increases the risk of taking sleeping pills and antidepressants, particularly among older people, starting at

noise levels of 30

41

dB (A) (Poulsen et al., 2019b). However, as this association was only addressed in one study, the level of

evidence for this finding cannot be considered high.

Compared to traffic noise levels, the noise levels in these studies conducted in Denmark are very low.

Traffic noise from motorways has been found to have an impact on cardiovascular disease from noise

levels of Lden 53 dB (A) (WHO, 2019). Most of the participants in the Danish nationwide cohort (79%) lived

in dwellings where the wind turbine noise level outdoors near their homes was below 24 dB(A). Although

the study included more than 700,000 people, few people were exposed to noise levels above 42 dB, and

the number of cases (47 heart attacks and 23 strokes) may have been too small to find statistically

significant associations (Poulsen et al., 2019a). Most wind turbines in Denmark are less than 100 m high,

but there are also taller ones. In the case of lower wind turbines (less than 35 m high), people in Denmark

live closer than 500 m to the wind turbines. Those who lived closest to the lowest wind turbines ( less than

35 m high) (closer than 500 m) made up the majority of the group exposed to the highest noise levels

(above 42 dB (A)). People living near taller wind turbines were mostly exposed to lower noise levels

(Poulsen et al., 2019a).

The Danish studies used noise levels that were not measured but modelled. The noise level was modelled

taking into account the noise emission of the wind turbine, weather conditions, the distance of the residence

under study and other relevant information, but it is still a calculated indicator. At the same time, this

indicator is more accurate than measured noise levels, as it is not possible to distinguish wind turbine noise

from other noise sources (background noise) such as normal natural sounds (wind noise, rain, birdsong),

traffic noise or noise caused by people themselves. The modelled noise level only shows the noise level

originating from wind turbines ( Lahti, 2010).

There is never absolute silence in the environment surrounding humans. Background noise itself can often

be close to 40 dB (A). Background noise is caused, for example, by household appliances, ventilation

systems, air conditioners and other technical equipment, natural sounds and, in urban environments, traffic

noise. In the tests included in this study, the background noise in soundproof laboratories was 19–23 dB

(A) in two tests by Liebich (Liebich et al., 2022a, 2022b) and 39 dB (A)/80–85 dB (Z) in the test by

Marshall et al. (2023), where it originated from air conditioning. Wind turbine noise was presented in the

test by Marshall et al. (2023) under controlled conditions in a soundproof laboratory, but the air

conditioning could not be turned off because otherwise the subjects would have been affected by unusual

temperatures. The background noise in the Malecki et al. (2023) experiment was 43 dB (A). This

experiment took place in a school building during class time.

42

The Danish studies were adjusted for many important characteristics, but it is still not possible to adjust the

analyses in such register-based studies for people's attitudes, negative expectations, political sense of justice

and other factors causing disturbance, as such data are not recorded in the registers. Cardiovascular disease,

sleep disorders and depression can all be affected by wind farm disturbance, which in turn can be caused by

a wide variety of factors (see chapter on Audible noise and disturbance from wind turbines).

One experiment (Chiu et al., 2021) also suggests that cardiovascular diseases may be affected by audible

noise from wind turbines. This experiment showed that the louder the noise from the wind turbines, the

lower the heart rate variability of the subjects. Heart rate variability (HRV) is a measure that describes

changes in the time interval between heartbeats. HRV reflects the activity and balance of the autonomic

nervous system. Higher HRV usually indicates better physical adaptability and stress tolerance, while lower

HRV may indicate stress, fatigue or health problems. In this study, the subjects spent 30 minutes at a

distance of 20 m from the nearest wind turbine, and the low-frequency noise (20-200 Hz) was 38.3-57.1

dB(A). Thus, this study shows that higher than normal wind turbine noise may increase the risk of

cardiovascular disease. As the study was conducted outdoors and indoors, rather than in a controlled

laboratory setting, other factors, such as noise from other sources, also influenced the results. The study

also did not take into account possible psychological stress and air pollution, which are also known to affect

HRV.

Long-term, large-scale, registry-based cohort studies conducted in Denmark showed that wind turbine noise

at none of the levels studied had an impact on the development of diabetes and high blood pressure, nor did

it worsen birth outcomes (Poulsen et al., 2018a, 2018b, 2018c).

In summary, no clear links have been found to date between audible wind turbine noise and clinically

significant health effects, but there are indications that audible noise may increase the risk of cardiovascular

disease and depression. However, too few long-term, high-quality follow-up studies have been conducted

to date to draw firm conclusions. Finding clear dose-response relationships is also limited by the fact that

there are no people who have long-term exposure to high wind turbine noise (above 50 dB (A)), as wind

turbines are not allowed to be built so close to residential areas that such noise exposure could occur.

Based on a cohort study conducted in Denmark, low-frequency noise indoors (10–160 Hz) was not

associated with heart attacks, strokes, diabetes, poor birth outcomes, high blood pressure,

43

depression and sleep disorders at any of the noise levels studied (up to 20 dB (A)) (Poulsen et al., 2019a,

2019b, 2018a, 2018b, 2018c).

Electromagnetic fields and vibrations caused by wind turbines

We did not find any scientific articles that met the inclusion criteria for this study that examined the health

effects of electromagnetic fields (EMF) or vibration caused by wind turbines. However, these factors have

been analysed by Knopper et al. (2014) and van Kamp and van den Berg (2018) in their review studies.

An electromagnetic field (EMF) is a physical field created by electric charges and electric currents and

consists of two interrelated components: an electric field and a magnetic field. An electric field occurs

when there is a difference in voltage (e.g. in an electrical conductor, battery or power line). A magnetic

field occurs when an electric current flows (e.g. when you switch on a device or an electric motor is

running) (WHO, 2016). EMF is present everywhere in our environment. Electric fields are created when

electrical charges accumulate in the atmosphere during thunderstorms, and the Earth's magnetic field causes

a compass needle to point north-south (WHO, 2016). In addition to natural sources, we are also surrounded

by man-made fields: X-rays, low-frequency EMFs associated with the flow of electricity from electrical

outlets, and higher-frequency radio waves (WHO, 2016).

Very high-intensity EMFs have been found to have both short-term direct and long-term health effects,

which is why Estonia, like many other countries, has established requirements and limits for

electromagnetic fields in the working and living environment (Regulation No. 38 of the Minister of Social

Affairs, 2002; Regulation No. 44 of the Government of the Republic, 2016). However, the field strengths of

most EMF sources (power lines, microwave ovens, mobile phones, etc.) are low, they are located in

households and workplaces at a sufficient distance from people or are encountered for short periods of time,

and therefore do not usually pose a health risk (WHO, 2016).

In 2011, the International Agency for Research on Cancer classified radiofrequency EMFs as a Group 2B

possible carcinogen for humans. This category is used when a causal relationship is considered possible but

there is no evidence from human studies (IARC, 2011). Some studies have found that low-frequency

EMFs, which come mainly from power lines, may increase the risk of leukaemia and brain and breast

tumours (Carpenter, 2019). However, the results of most studies are contradictory, mostly showing no

effect, and there is currently no conclusive evidence of the long-term health effects of EMFs (Bodewein et

al., 2019).

44

A 2014 review by Knopper et al. showed that EMF levels measured 2–3 metres from the base of a wind

turbine are similar to or lower than those generated by many common household appliances (e.g.

refrigerator, dishwasher, microwave oven, hair dryer) and significantly below the applicable regulatory

limits. Therefore, health effects from wind turbine electromagnetic fields are highly unlikely (Knopper et

al., 2014).

Vibration is the oscillation of a solid body (Regulation No. 78 of the Minister of Social Affairs, 2002). The

health effects of vibration are particularly evident in work environments where people are exposed to the

vibration of tools such as jackhammers or large machines such as tractors (Health Board, 2025c). In the

living environment, vibration can be caused by traffic (e.g. trains, heavy goods vehicles), which can cause

windows to rattle and cracks to appear in buildings. Limit values for vibration have also been established

for residential buildings (Regulation No. 78 of the Minister of Social Affairs, 2002).

Vibration measurements in the vicinity of wind turbines (less than 300 m from the turbines) have yielded

values that are close to zero (Ministry of Climate, 2025; Knopper et al., 2014; van Kamp and van den Berg,

2018). Therefore, the vibration caused by wind turbines is not a significant safety risk to people living near

wind turbines.

Justification for the choice of methodology, strengths and weaknesses of the study

We included only peer-reviewed scientific articles published in journals whose design allowed for the

identification of causal relationships. Randomised controlled trials are the most reliable. Therefore, we also

included all experiments that met the criteria in the study. However, it is not possible to conduct such

experiments to study long-term exposure. In order to study the long-term effects of an environmental

factor, long-term follow-up studies must be conducted. Such studies are cohort studies, which we also

included in our review. We also included systematic reviews and meta-analyses, which are the highest

quality type of studies. A systematic review is a synthesis of existing evidence that uses a clear, transparent

and systematic methodology to find, evaluate and present relevant evidence. We did not include narrative

reviews because they are subject to a high risk of error due to their subjective nature. We also did not

include cross-sectional studies because it is not possible to identify a causal relationship in these studies.

Cross-sectional studies examine the presence of an exposure (e.g. noise) and an outcome (e.g. health

problem) at the same point in time (e.g. during a survey). Therefore, it is not possible to know which came

first, the health problem or the influencing factor. For an influencing factor (e.g. noise) to cause a health

problem, it must be present before the health problem develops. This approach is possible with a cohort

study.

45

We only included scientific literature published in English in the study. In our opinion, this does not cause a

significant bias in our results, as English is currently the main language of science, in which all important

scientific results are published. It is unlikely that any very important results have been published only in

other languages. Furthermore, it is not reasonable to include literature published in all languages, as the

research team does not speak all languages and including studies in all languages would carry a high risk of

misinterpreting the results of the articles.

46

Conclusions Based on the scientific studies found in the systematic literature review, we can draw the following

conclusions:

• Scientific studies have not shown any negative health effects from electromagnetic fields,

vibrations and infrasound generated by wind turbines that remain within the limits. The evidence

for the absence of health effects from infrasound is based on thirteen experiments included in this

review.

• Based on this study, low-frequency noise from wind turbines indoors (10–160 Hz) was not

associated with heart attacks, strokes, diabetes, poor birth outcomes, high blood pressure,

depression or sleep disturbances at any of the noise levels studied (up to 20 dB A). As only one

study examined the relationship between noise exposure and each health outcome, the level of

evidence for this finding cannot be considered high.

• Long-term, large-scale, register-based studies conducted in Denmark found that audible noise from

wind turbines (including low-frequency noise) did not affect the development of diabetes and high

blood pressure, nor did it worsen birth outcomes (birth weight and premature birth) at any of the

noise levels studied (up to 50 dB (A)). As only one study addressed the relationship between noise

exposure and each health outcome, the level of evidence for this finding cannot be considered high.

• To date, no clear links have been found between audible wind turbine noise (including low-

frequency noise) and clinically significant health effects. However, there are indications that

audible noise may increase the risk of certain cardiovascular diseases (atrial fibrillation, heart

attack and stroke), increase the incidence of depression and impair sleep quality. To date, only two

large cohort studies have been conducted in Denmark, which means that the level of evidence for

these findings cannot be considered high. More long-term follow-up studies are needed to identify

health effects more precisely.

• There is clear evidence that wind turbines cause disturbance to residents. In addition to wind

turbine noise, other factors also contribute to the disturbance. Expectations/knowledge about the

health effects of wind turbines obtained from the media, the perceived fairness and transparency of

the planning process, economic benefits and visual aspects all play an important role. It is likely

that a combination of all these factors causes disturbance.

• Based on listening tests, audible wind turbine sounds with greater amplitude modulation depth (changes in sound levels) are more disturbing than sounds with lower

47

amplitude modulation depth. The greater amplitude modulation depth associated with audible wind

turbine noise may increase human disturbance from wind farms.

48

Recommendations Currently, the limit values for industrial noise (including wind turbines) in residential areas are 60 dB (A) during the day and 45 dB

(A) in residential areas. The target noise levels are 50 dB (A) during the day and 40 dB (A) at night

(Regulation No. 71 of the Minister of the Environment, 2016). Indoors, traffic noise levels may be up to 30

dB (A) during the day and noise from technical equipment up to 25 dB (A) at night (Regulation No. 42 of

the Minister of Social Affairs, 2002). The guidelines for assessing the environmental impact of wind farms

recommend using the strictest value for wind turbines in outdoor conditions, i.e. the noise from wind

turbines should not exceed 40 dB (A) at night in the vicinity of residential buildings (Ministry of Climate,

2025).

Based on our study, there is currently no reason to recommend a stricter limit value for residential areas

than that specified in the guidelines for assessing the environmental impact of wind farms (Ministry of

Climate, 2025). Currently, the regulation of wind farm noise in Estonia is confusing due to the difference

between limit values and target values and the lack of a specific limit value for wind turbines. We

recommend that wind turbine noise be regulated more clearly in legislation. Based on the studies included

in this review, we recommend establishing a limit value based on current knowledge so that wind turbine

noise in the immediate vicinity of residential buildings outdoors at night (23:00–07:00) does not exceed 40

dB (A), as already recommended in the guidelines for assessing the environmental impact of wind farms.

For living and sleeping areas, we recommend setting a limit for wind turbine noise of 30 dB(A) during the

day and 25 dB(A) at night, similar to the existing limits for traffic noise and noise from technical

equipment.

Based on this study, there is no reason to recommend stricter limits for infrasound, which are established in

Regulation No. 75 of the Minister of Social Affairs (2002), and low-frequency sound, which are established

in Regulation No. 42 of the Minister of Social Affairs (2002).

We recommend continuing with noise and infrasound measurements at wind farms in Estonia and with

studies on health effects and perceived disturbance. If new significant scientific research becomes available

after five years, we recommend reassessing the appropriateness of the limit values.

This study found that audible wind turbine sounds with greater amplitude modulation (AM) depth are more

disturbing than sounds with lower AM depth, based on listening tests, and that perceptible AM may

increase people's disturbance from wind farms. We recommend that developers and researchers explore

ways to reduce AM depth in order to reduce the annoyance of wind turbine noise. The more uniform the

noise level of wind turbines, the less disturbing it is. When measuring and modelling the noise generated by

wind farms, both the average and maximum noise levels should be highlighted separately, as well as the

AM depth, if possible.

49

The results of our study show that several factors other than wind turbine noise affect disturbance, and that

noise reduction alone may not be sufficient to mitigate disturbance. Just as important as noise restrictions in

preventing disturbance may be informing residents about the nocebo effect, the absence of negative

expectations regarding the health effects of wind turbines, and understanding the positive characteristics of

wind turbines (Crichton et al., 2015, 2014b, 2014a; Crichton and Petrie, 2015b, 2015a; Tonin et al., 2016).

The perceived openness and fairness of political processes (Pohl et al., 2018), economic benefits

(McCunney et al., 2014) and visual aspects (Freiberg et al., 2019b) also influence the emergence of wind

turbine-related disturbance. To prevent and mitigate disturbance, it is important to provide communities

with objective scientific information to counteract the impact of misinformation about infrasound

circulating on the internet. Positive experiences have been gained from early and informal involvement of

residents in wind turbine planning processes. It is also important that residents feel that their concerns are

being taken into account and that mitigation measures are being implemented even when everything

complies with the standards (e.g. the modelled noise level is 39.9 dB).

An important aspect of visual disturbance that can be reduced is shadowing. The relevant mitigation

measures have already been described in the guidelines for assessing the environmental impact of wind

farms (Ministry of Climate, 2025) and must always be implemented.

Based on this study, there is no clear basis for recommending a minimum distance from the nearest wind

turbine to residential buildings, as wind turbines can vary greatly in terms of noise emissions and other

factors that cause disturbance.

Existing studies show that wind turbines can be located safely for human health if they are placed in

accordance with the noise standards recommended in this study, if planning is carried out transparently,

taking into account the interests of the community and offering benefits to them, and if visual pollution is

minimised. It is important to note that the use of wind energy helps to mitigate climate change and reduces

health problems caused by air pollution. For example, in 2020, more than 1,000 people in Estonia died

prematurely due to air pollution (Orru et al., 2022).

50

References

Alves-Pereira, M., Castelo-Branco, N.A.A., 2007. Vibroacoustic disease: Biological effects of infrasound and low- frequency noise explained by mechanotransduction cellular signalling. Prog Biophys Mol Biol 93, 256–279. https://doi.org/10.1016/j.pbiomolbio.2006.07.011

Ascone, L., Kling, C., Wieczorek, J., Koch, C., Kühn, S., 2021. A longitudinal, randomised experimental pilot study to investigate the effects of airborne infrasound on human mental health, cognition, and brain structure. Sci Rep 11. https://doi.org/10.1038/S41598-021-82203-6

Basner, M., Babisch, W., Davis, A., Brink, M., Clark, C., Janssen, S., Stansfeld, S., 2014. Auditory and non-auditory effects of noise on health. Lancet 383, 1325–1332. https://doi.org/10.1016/S0140-6736(13)61613-X

Basner, M., McGuire, S., 2018. WHO Environmental Noise Guidelines for the European Region: A Systematic Review on Environmental Noise and Effects on Sleep. Int J Environ Res Public Health

Bodewein, L., Schmiedchen, K., Dechent, D., Stunder, D., Graefrath, D., Winter, L., Kraus, T., Driessen, S., 2019. Systematic review on the biological effects of electric, magnetic and electromagnetic fields in the intermediate frequency range (300 Hz to 1 MHz). Environ Res 171, 247–259. https://doi.org/10.1016/J.ENVRES.2019.01.015

Bräuner, E. V., Jørgensen, J.T., Duun-Henriksen, A.K., Backalarz, C., Laursen, J.E., Pedersen, T.H., Simonsen, M.K., Andersen, Z.J., 2019a. Association Between Long-Term Exposure to Wind Turbine Noise and the Risk of Stroke: Data From the Danish Nurse Cohort. J Am Heart Assoc 8. https://doi.org/10.1161/JAHA.119.013157

Bräuner, E. V., Jørgensen, J.T., Duun-Henriksen, A.K., Backalarz, C., Laursen, J.E., Pedersen, T.H., Simonsen, M.K., Andersen, Z.J., 2019b. Long-term wind turbine noise exposure and the risk of incident atrial fibrillation in the Danish Nurse cohort. Environ Int 130. https://doi.org/10.1016/J.ENVINT.2019.104915

Bräuner, E. V., Jørgensen, J.T., Duun-Henriksen, A.K., Backalarz, C., Laursen, J.E., Pedersen, T.H., Simonsen, M.K., Andersen, Z.J., 2018. Long-term wind turbine noise exposure and incidence of myocardial infarction in the Danish nurse cohort. Environ Int 121, 794–802. https://doi.org/10.1016/J.ENVINT.2018.10.011

Carpenter, D.O., 2019. Extremely low frequency electromagnetic fields and cancer: How source of funding affects results. Environ Res 178. https://doi.org/10.1016/J.ENVRES.2019.108688

Castelo-Branco, N.A.A., Alves-Pereira, M., 2004. Vibroacoustic disease. Noise Health 6(23), 3–20.

Chapman, S., George, A.S., 2013. How the factoid of wind turbines causing ‘vibroacoustic disease’ came to be ‘irrefutably demonstrated.’ Aust N Z J Public Health 37, 244–249. https://doi.org/10.1111/1753- 6405.12066

Chiu, C.H., Lung, S.C.C., Chen, N., Hwang, J.S., Tsou, M.C.M., 2021. Effects of low-frequency noise from wind turbines on heart rate variability in healthy individuals. Sci Rep 11. https://doi.org/10.1038/S41598-021- 97107-8

51

Clark, C., Paunovic, K., 2018. WHO Environmental Noise Guidelines for the European Region: A Systematic Review on Environmental Noise and Quality of Life, Wellbeing and Mental Health. Int J Environ Res Public Health 15. https://doi.org/10.3390/IJERPH15112400

Crichton, F., Dodd, G., Schmid, G., Gamble, G., Cundy, T., Petrie, K.J., 2014a. The power of positive and negative expectations to influence reported symptoms and mood during exposure to wind farm sound. Health Psychol 33, 1588–1592. https://doi.org/10.1037/HEA0000037

Crichton, F., Dodd, G., Schmid, G., Gamble, G., Petrie, K.J., 2014b. Can expectations produce symptoms from infrasound associated with wind turbines? Health Psychol 33, 360–364. https://doi.org/10.1037/A0031760

Crichton, F., Dodd, G., Schmid, G., Petrie, K.J., 2015. Framing sound: Using expectations to reduce environmental noise annoyance. Environ Res 142, 609–614. https://doi.org/10.1016/J.ENVRES.2015.08.016

Crichton, F., Petrie, K.J., 2015a. Health complaints and wind turbines: The efficacy of explaining the nocebo response to reduce symptom reporting. Environ Res 140, 449–455. https://doi.org/10.1016/J.ENVRES.2015.04.016

Crichton, F., Petrie, K.J., 2015b. Accentuate the positive: Counteracting psychogenic responses to media health messages in the age of the Internet. J Psychosom Res 79, 185–189. https://doi.org/10.1016/J.JPSYCHORES.2015.04.014

Davy, J.L., Burgemeister, K., Hillman, D., Carlile, S., 2020. A Review of the Potential Impacts of Wind Turbine Noise in the Australian Context. Acoust Aust 48, 181–197. https://doi.org/10.1007/S40857-020-00192-4/FIGURES/1

Ellenbogen, J.M., Kellam, C.B., Hankard, M., 2024. Noise-induced sleep disruption from wind turbines: scientific updates and acoustical standards. Sleep 47. https://doi.org/10.1093/SLEEP/ZSAD286

European Commission, 2019. The European Green Deal - European Commission [WWW Document]. URL https://commission.europa.eu/strategy-and-policy/priorities-2019-2024/european-green-deal_en (accessed 7.23.25).

Freiberg, A., Schefter, C., Girbig, M., Murta, V.C., Seidler, A., 2019a. Health effects of wind turbines on humans in residential settings: Results of a scoping review. Environ Res 169, 446–463. https://doi.org/10.1016/J.ENVRES.2018.11.032

Freiberg, A., Schefter, C., Girbig, M., Murta, V.C., Seidler, A., 2018. Health effects of wind turbines in working environments – A scoping review. Scand J Work Environ Health 44, 351–369. https://doi.org/10.5271/SJWEH.3711

Freiberg, A., Schefter, C., Hegewald, J., Seidler, A., 2019b. The influence of wind turbine visibility on the health of local residents: a systematic review. Int Arch Occup Environ Health 92, 609–628. https://doi.org/10.1007/S00420-019-01403-W

Garritty, C., Hamel, C., Trivella, M., Gartlehner, G., Nussbaumer-Streit, B., Devane, D., Kamel, C., Griebler, U., King, V.J., 2024. Updated recommendations for the Cochrane rapid review methods guidance for rapid reviews of effectiveness. BMJ 384. https://doi.org/10.1136/BMJ-2023-076335

52

Godono, A., Ciocan, C., Clari, M., Mansour, I., Curoso, G., Franceschi, A., Carena, E., De Pasquale, V., Dimonte, V., Pira, E., Dallapiccola, B., Normanno, N., Boffetta, P., 2023. Association between exposure to wind turbines and sleep disorders: A systematic review and meta-analysis. Int J Hyg Environ Health 254, 114273. https://doi.org/10.1016/J.IJHEH.2023.114273

Guski, R., Schreckenberg, D., Schuemer, R., 2017. WHO Environmental Noise Guidelines for the European Region: A Systematic Review on Environmental Noise and Annoyance. Int J Environ Res Public Health 14, 1539. https://doi.org/10.3390/IJERPH14121539

Harrison, R. V., 2015. On the biological plausibility of Wind Turbine Syndrome. Int J Environ Health Res 25, 463– 468. https://doi.org/10.1080/09603123.2014.963034

IARC, 2011. IARC classifies Radiofrequency Electromagnetic Fields as possibly carcinogenic to humans IARC [WWW Document]. URL https://www.iarc.who.int/pressrelease/iarc-classifies-radiofrequency-electromagnetic-fields- as-possibly-carcinogenic-to-humans/ (accessed 7.22.25).

Ioannidou, C., Santurette, S., Jeong, C.-H., 2016. Effect of modulation depth, frequency, and intermittence on wind turbine noise annoyance. J Acoust Soc Am 139, 1241–1251. https://doi.org/10.1121/1.4944570

Karasmanaki, E., 2022. Is it safe to live near wind turbines? Reviewing the impacts of wind turbine noise. Energy for Sustainable Development 69, 87–102. https://doi.org/10.1016/J.ESD.2022.05.012

Regulation No. 71 of the Minister of the Environment, 2016. Standard levels of noise in outdoor air and methods for measuring, determining and assessing noise levels – State Gazette [WWW Document]. URL https://www.riigiteataja.ee/akt/127052020002?leiaKehtiv (accessed 5.29.25).

King, V.J., Stevens, A., Nussbaumer-Streit, B., Kamel, C., Garritty, C., 2022. Paper 2: Performing rapid reviews. Syst Rev 11. https://doi.org/10.1186/S13643-022-02011-5

Ministry of Climate, 2025. Guidelines for assessing the environmental impact of wind farms, noise, vibration, shadowing [WWW Document]. URL https://kliimaministeerium.ee/sites/default/files/documents/2025- 03/Tuuleparkide%20keskkonnam%C3%B5ju%20hindamise%20juhend.pdf (accessed 7.28.25).

Knopper, L.D., Ollson, C.A., 2011. Health effects and wind turbines: A review of the literature. Environ Health 10, 1–10. https://doi.org/10.1186/1476-069X-10-78/COMMENTS

Knopper, L.D., Ollson, C.A., McCallum, L.C., Aslund, M.L.W., Berger, R.G., Souweine, K., McDaniel, M., 2014. Wind Turbines and Human Health. Front Public Health 2, 63. https://doi.org/10.3389/FPUBH.2014.00063

Lahti, T., 2010. Ökokratt - Environmental Noise Handbook [WWW Document]. URL https://www.okokratt.ee/index.php/et-ee/kaeimasolevad-projektid/muera/keskkonnamra-ksiraamat- mainmenu-339 (accessed 5.29.25).

Lancet Countdown, 2024. The 2024 Global Report of the Lancet Countdown [WWW Document]. URL https://lancetcountdown.org/2024-report/ (accessed 7.23.25).

Lee, Seunghoon, Kim, K., Choi, W., Lee, Soogab, 2011. Annoyance caused by amplitude modulation of wind turbine noise. Noise Control Eng J 59, 38–46. https://doi.org/https://doi.org/10.3397/1.3531797

53

Liebich, T., Lack, L., Hansen, K., Zajamšek, B., Lovato, N., Catcheside, P., Micic, G., 2021. A systematic review and meta- analysis of wind turbine noise effects on sleep using validated objective and subjective sleep assessments. J Sleep Res 30. https://doi.org/10.1111/JSR.13228

Liebich, T., Lack, L., Hansen, K., Zajamsek, B., Micic, G., Lechat, B., Dunbar, C., Nguyen, D.P., Scott, H., Catcheside, P., 2022a. An experimental investigation on the impact of wind turbine noise on polysomnography-measured and sleep diary-determined sleep outcomes. Sleep 45. https://doi.org/10.1093/SLEEP/ZSAC085

Liebich, T., Lack, L., Micic, G., Hansen, K., Zajamšek, B., Dunbar, C., Lechat, B., Scott, H., Lovato, N., Decup, F., Nguyen, D.P., Catcheside, P., 2022b. The effect of wind turbine noise on polysomnographically measured and self- reported sleep latency in wind turbine noise naïve participants. Sleep 45. https://doi.org/10.1093/SLEEP/ZSAB283

Maijala, P.P., Kurki, I., Vainio, L., Pakarinen, S., Kuuramo, C., Lukander, K., Virkkala, J., Tiippana, K., Stickler, E.A., Sainio, M., 2021. Annoyance, perception, and physiological effects of wind turbine infrasound. J Acoust Soc Am 149, 2238–2248. https://doi.org/10.1121/10.0003509

Małecki, P., Pawlaczyk-Łuszczyńska, M., Wszołek, T., Preis, A., Kłaczyński, M., Dudarewicz, A., Pawlik, P., Stępień, B., Mleczko, D., 2023. Does Stochastic and Modulated Wind Turbine Infrasound Affect Human Mental Performance Compared to Steady Signals without Modulation? Results of a Pilot Study. Int J Environ Res Public Health 20. https://doi.org/10.3390/IJERPH20032223

Marshall, N.S., Cho, G., Toelle, B.G., Tonin, R., Bartlett, D.J., D’rozario, A.L., Evans, C.A., Cowie, C.T., Janev, O., Whitfeld, C.R., Glozier, N., Walker, B.E., Killick, R., Welgampola, M.S., Phillips, C.L., Marks, G.B., Grunstein, R.R., 2023. The Health Effects of 72 Hours of Simulated Wind Turbine Infrasound: A Double-Blind Randomised Crossover Study in Noise-Sensitive, Healthy Adults. Environ Health Perspect 131. https://doi.org/10.1289/EHP10757

McCunney, R.J., Mundt, K.A., Colby, W.D., Dobie, R., Kaliski, K., Blais, M., 2014. Wind turbines and health: a critical review of the scientific literature. J Occup Environ Med 56, e108–e130. https://doi.org/10.1097/JOM.0000000000000313

Murcia, G.M.D., Ferrer, S.F., Sorinas, J., Ferrandez, J.M., Fernandez, E., 2017. Application of electroencephalographic techniques to the study of visual impact of renewable energies. J Environ Manage 200, 484–489. https://doi.org/10.1016/J.JENVMAN.2017.05.096

Orru, H., Teinemaa, E., Maasikmets, M., Keernik, H., Paju, M., Sikk, A., Tamm, T., Lainjärv, H.M., Kriit, H.K., Lõhmus Sundström, M., 2022. Comparison of the impact of outdoor air quality on human health in Estonia in 2010 and 2020 and a forecast of the health effects of air pollution for 2030 [WWW Document]. URL https://keskkonnaportaal.ee/et/valisohu-kvaliteedi-moju-vordlus-inimeste-tervisele-eestis-aastatel-2010-ja- 2020-ning-ohusaaste (accessed 7.30.25).

Pierpont, N., 2009. (PDF) Wind Turbine Syndrome: A Report on a Natural Experiment [WWW Document]. URL https://www.researchgate.net/publication/265247204_Wind_Turbine_Syndrome_A_Report_o n_a_Natural_Experiment (accessed 7.17.25).

Pohl, J., Gabriel, J., Hübner, G., 2018. Understanding stress effects of wind turbine noise – The integrated approach. Energy Policy 112, 119–128. https://doi.org/10.1016/J.ENPOL.2017.10.007

54

Poulsen, A.H., Raaschou-Nielsen, O., Peña, A., Hahmann, A.N., Nordsborg, R.B., Ketzel, M., Brandt, J., Sørensen, M., 2019a. Long-Term Exposure to Wind Turbine Noise and Risk for Myocardial Infarction and Stroke: A Nationwide Cohort Study. Environ Health Perspect 127. https://doi.org/10.1289/EHP3340

Poulsen, A.H., Raaschou-Nielsen, O., Peña, A., Hahmann, A.N., Nordsborg, R.B., Ketzel, M., Brandt, J., Sørensen, M., 2019b. Impact of Long-Term Exposure to Wind Turbine Noise on Redemption of Sleep Medication and Antidepressants: A Nationwide Cohort Study. Environ Health Perspect

Poulsen, A.H., Raaschou-Nielsen, O., Peña, A., Hahmann, A.N., Nordsborg, R.B., Ketzel, M., Brandt, J., Sørensen, M., 2018a. Long-term exposure to wind turbine noise at night and risk for diabetes: A nationwide cohort study. Environ Res 165, 40–45. https://doi.org/10.1016/J.ENVRES.2018.03.040

Poulsen, A.H., Raaschou-Nielsen, O., Peña, A., Hahmann, A.N., Nordsborg, R.B., Ketzel, M., Brandt, J., Sørensen, M., 2018b. Long-term exposure to wind turbine noise and redemption of antihypertensive medication: A nationwide cohort study. Environ Int 121, 207–215. https://doi.org/10.1016/J.ENVINT.2018.08.054

Poulsen, A.H., Raaschou-Nielsen, O., Peña, A., Hahmann, A.N., Nordsborg, R.B., Ketzel, M., Brandt, J., Sørensen, M., 2018c. Pregnancy exposure to wind turbine noise and adverse birth outcomes: a nationwide cohort study. Environ Res 167, 770–775. https://doi.org/10.1016/J.ENVRES.2018.09.011

Poulsen, A.H., Raaschou-Nielsen, O., Peña, A., Hahmann, A.N., Nordsborg, R.B., Ketzel, M., Brandt, J., Sørensen, M., 2018d. Short-term nighttime wind turbine noise and cardiovascular events: A nationwide case-crossover study from Denmark. Environ Int 114, 160–166. https://doi.org/10.1016/J.ENVINT.2018.02.030

Rosciszewska, A., Buszkiewicz, M., Dobrzynska-Kobylec, G., Klichowska, A., Przybyla, T., Nagy, B.B., Wicher, A., Klichowski, M., 2025. Cognitive neuroscience approach to explore the impact of wind turbine noise on various mental functions. Humanities and Social Sciences Communications 2025 12:1 12, 1–17. https://doi.org/10.1057/s41599-025-04645-x

Schäffer, B., Pieren, R., Schlittmeier, S.J., Brink, M., 2018. Effects of Different Spectral Shapes and Amplitude Modulation of Broadband Noise on Annoyance Reactions in a Controlled Listening Experiment. Int J Environ Res Public Health 15. https://doi.org/10.3390/IJERPH15051029

Schäffer, B., Schlittmeier, S.J., Pieren, R., Heutschi, K., Brink, M., Graf, R., Hellbrück, J., 2016. Short-term annoyance reactions to stationary and time-varying wind turbine and road traffic noise: A laboratory study. J Acoust Soc Am 139, 2949–2963. https://doi.org/10.1121/1.4949566

Schmidt, J.H., Klokker, M., 2014. Health Effects Related to Wind Turbine Noise Exposure: A Systematic Review. PLoS One 9, e114183. https://doi.org/10.1371/JOURNAL.PONE.0114183

Regulation No. 38 of the Minister of Social Affairs, 2002. Limit values for non-ionising radiation in residential and recreational areas,

residential buildings and public buildings, classrooms and measurement of non-ionising radiation levels [WWW Document]. URL https://www.riigiteataja.ee/akt/163816?leiaKehtiv (accessed 8.9.25).

55

Regulation No. 42 of the Minister of Social Affairs, 2002. Noise standards in residential and recreational areas, residential buildings and public buildings, and methods for measuring noise levels [WWW Document]. State Gazette. URL https://www.riigiteataja.ee/akt/163756 (accessed 7.11.25).

Regulation No. 75 of the Minister of Social Affairs, 2002. Limit values for ultra- and infrasound sound pressure levels and measurement of ultra- and infrasound sound pressure levels [WWW Document]. URL https://www.riigiteataja.ee/akt/129122020046?leiaKehtiv (accessed 5.29.25).

Regulation No. 78 of the Minister of Social Affairs, 2002. Limit values for vibration in residential and public buildings and methods for measuring vibration [WWW Document]. URL https://www.riigiteataja.ee/akt/110061?leiaKehtiv (accessed 8.9.25).

Staniek, A., Kompała, J., Bramorska, A., Bartmański, C., 2013. Comparison analysis of noise generated by wind turbines with other noise sources in outdoor environments. Vibrations in Physical Systems 2023, 2023108. https://doi.org/10.21008/j.0860-6897.2023.1.08

Taylor, J., Klenk, N., 2019. The politics of evidence: Conflicting social commitments and environmental priorities in the debate over wind energy and public health. Energy Res Soc Sci 47, 102–112. https://doi.org/10.1016/J.ERSS.2018.09.001

Teneler, A.A., Hassoy, H., 2023. Health effects of wind turbines: a review of the literature between 2010-2020. Int J Environ Health Res 33, 143–157. https://doi.org/10.1080/09603123.2021.2010671/ASSET/CB263B43-F0EB- 4EF2-94C8-593ADABE314A/ASSETS/IMAGES/CIJE_A_2010671_F0001_B.GIF

Health Board, 2025a. Report on noise measurements at the Sopi-Tootsi wind farm [WWW Document]. URL https://sopitootsipargid.ee/et/avaleht/ (accessed 7.21.25).

Health Board, 2025b. Saarde wind farm noise study [WWW Document]. URL https://utilitas.ee/failipank/saarde-tuulepargi-murauuring/ (accessed 7.21.25).

Health Board, 2025c. Noise and vibration [WWW Document]. URL https://www.terviseamet.ee/keskkonnatervis/fuusikalised-ohutegurid-elukeskkonnas/mura-ja- vibratsioon#kohtvibratsioon (accessed 8.9.25).

Tonin, R., Brett, J., Colagiuri, B., 2016. The effect of infrasound and negative expectations to adverse pathological symptoms from wind farms. Journal of Low Frequency Noise Vibration and Active Control 35, 77–90. https://doi.org/10.1177/0263092316628257/ASSET/D90140C8-16F5-4714-B165- 9A39FD096E84/ASSETS/IMAGES/LARGE/10.1177_0263092316628257-FIG13.JPG

Government Regulation No. 44, 2016. Occupational health and safety requirements for work environments affected by electromagnetic fields, exposure limits for electromagnetic fields and application values, and procedure for measuring electromagnetic fields [WWW Document]. URL https://www.riigiteataja.ee/akt/107042016004?leiaKehtiv (accessed 7.22.25).

van Kamp, I., van den Berg, F., 2021. Health effects related to wind turbine sound: An update. Int J Environ Res Public Health 18. https://doi.org/10.3390/IJERPH18179133

van Kamp, I., van den Berg, F., 2018. Health Effects Related to Wind Turbine Sound, Including Low-Frequency Sound and Infrasound. Acoust Aust 46, 31–57. https://doi.org/10.1007/S40857-017-0115-6/FIGURES/3

van Kempen, E., Casas, M., Pershagen, G., Foraster, M., 2018. WHO Environmental Noise Guidelines for the European Region: A Systematic Review on Environmental Noise and Cardiovascular and

56

Metabolic Effects: A Summary. International Journal of Environmental Research and Public Health 2018, Vol. 15, Page 379 15, 379. https://doi.org/10.3390/IJERPH15020379

WHO, 2025. Constitution of the World Health Organisation [WWW Document]. URL https://www.who.int/about/governance/constitution (accessed 7.24.25).

WHO, 2019. Environmental noise guidelines for the European Region [WWW Document]. URL https://www.who.int/europe/publications/i/item/9789289053563 (accessed 7.24.25).

WHO, 2016. Radiation: Electromagnetic fields [WWW Document]. URL https://www.who.int/news- room/questions-and-answers/item/radiation-electromagnetic-fields (accessed 7.22.25).

WHO, 2009. Night noise guidelines for Europe [WWW Document]. URL https://iris.who.int/handle/10665/326486 (accessed 7.28.25).

World Bank Group, 2015. Environmental, Health, and Safety Guidelines for Wind Energy [WWW Document]. URL https://documents1.worldbank.org/curated/en/498831479463882556/pdf/110346-WP-FINAL-Aug-2015- Wind-Energy-EHS-Guideline-PUBLIC.pdf (accessed 7.28.25).

Wu, S., Du, W., Zhong, X., Lu, J., Wen, F., 2023. The association between road traffic noise and type 2 diabetes: a systematic review and meta-analysis of cohort studies. Environmental Science and Pollution Research 30, 39568–39585. https://doi.org/10.1007/S11356-023-25926-5/TABLES/5

57

Appendix 1. Search terms and search strategy used

Search strategy in the PubMed database

(("wind"[MeSH Terms] OR wind turbine[Title/Abstract] OR wind turbines[Title/Abstract] OR wind farms[Title/Abstract] OR wind parks[Title/Abstract] OR wind power plants[Title/Abstract] OR wind mill[Title/Abstract] OR wind generators[Title/Abstract]) AND ("noise"[MeSH Terms] OR "sound"[MeSH Terms:noExp] OR noise[Title/Abstract] OR infrasound[Title/Abstract] OR low-frequency noise[Title/Abstract] OR sound[Title/Abstract] OR vibration[Title/Abstract] OR visibility[Title/Abstract] OR visual[Title/Abstract] OR shadow flickering[Title/Abstract] OR flicker[Title/Abstract] OR electromagnetic field[Title/Abstract] OR "Electromagnetic Fields"[Mesh] OR infrasonic[Title/Abstract] OR "low frequency"[Title/Abstract] OR light flickering[Title/Abstract] OR stroboscopic effect[Title/Abstract] OR blinking lights[Title/Abstract] OR reflections[Title/Abstract] OR horizon pollution[Title/Abstract] OR light effects[Title/Abstract])) AND (("1 January 2010"[Date - Publication] : "22 April 2025"[Date - Publication]))

58

Appendix 2. Tables of results

Table 1. Systematic reviews Source Exposure Health outcomes Study designs Publication

date Included studies number

Assessment of study quality Study results

Liebich et al 2021

Noise dB (A) LAeq

Objective sleep disturbance indicators: WASO1, SOL2, TST3, sleep efficiency

Experiments, pre- post studies, cross-sectional

2000–2020 9 qualitative analysis; 5 meta- analyses

The reporting quality of individual studies was assessed using an adaptation of the STROBE4 checklist. The overall reporting quality of the studies was low. No tools were used to assess the risk of bias. , the assessment based on the limitations and biases of the studies as follows: 4 studies with high, 4 with moderate (some concerns) and 1 with low risk of bias

Objectively measured indicators of sleep macrostructure were not significantly affected in those exposed to wind turbine noise compared to controls not exposed to wind turbine noise. An effect was found on subjectively measured sleep indicators

Guski et al 2017

Noise dB (A), Lden

Disturbance Cross-sectional 2000–2012 4 The GRADE5 methodology assessed the level of evidence as low to moderate.

Wind turbine noise is associated with disturbance even at levels below 40 dB Lden.

Godono et al 2023

Distance from wind turbines

Noise dB (A)

Self-reported sleep quality

Cross-sectional 2004–2021 15 The methodological quality of the studies was assessed using the US National Institutes of Health Quality Assessment Tool: 2 high, 5 moderate and 8 low quality studies

The prevalence of sleep disturbance decreased with increasing distance from wind turbines and increased with higher sound pressure levels.

Freiberg et al 2019b

Visual aspects (visibility from the place of residence, shadowing, flashing lights, etc.)

Disturbance, sleep quality, quality of life

Cohort, cross-sectional

Up to = 2017

17 qualitative analysis; 6 meta- analysis

The quality of cross-sectional studies was assessed using the Appraisal tool for Cross-Sectional Studies (AXIS) tool, and cohort studies were assessed using a combination of tools from the Scottish Intercollegiate Guidelines Network (SIGN) and the Critical Appraisal Skills Programme (CASP) tools. Five studies were rated as high quality, three acceptable and 7 low quality

Disturbance from direct visibility, glare and flashing lights was statistically significantly associated with an increased risk of sleep disturbance. Changes in the view of the landscape, obstacle markings and light reflection from wind turbine blades may also disturb people .

1WASO - wake after sleep onset 2SOL - sleep onset latency 3TST - total sleep time 4STROBE – principles for reporting observational studies (STrengthening the Reporting of OBservational studies in Epidemiology) 5GRADE - Grading of Recommendations Assessment, Development and Evaluation

59

Table 2. Observational studies Source Study design and

subjects Sample size

Age Follow- up Self time

Exposure Health outcome Method of measurement method

Study results

Bräuner et al 2018

Danish nurses Cohort study

23,994 ≥44 1982– 2013

Modelled A-weighted wind turbine noise at frequencies 10 Hz–10,000 Hz. Calculated annual average Lden at the residences under study within a 6 km radius of the nearest wind turbine.

686 heart attacks (1.7 new cases per 1,000 person-years)

Danish patient and cause of death registries

Long-term exposure to wind turbine was not associated with the occurrence of heart attacks in women aged 44 and older.

Bräuner et al 2019b

Danish nurses cohort study

23,912 ≥44 1982– 2013

Modelled A-weighted wind turbine noise at frequencies of 10 Hz–10,000 Hz. Calculated annual average Lden at the residences under study within a 6 km radius of the nearest wind turbine. 1-, 5-, and 11-year average

1097 strokes (2.7 new cases per 1,000 person-years)

Danish patient registry

Long-term exposure to wind turbine was not associated with stroke in women aged 44 and older.

Bräuner et al 2019a

Danish nurses cohort study

24,137 ≥44 1982– 2013

Modelled A-weighted wind turbine noise at frequencies of 10 Hz–10,000 Hz. Calculated annual average Lden at the residences under study within a 6 km radius of the nearest wind turbine. 1-, 5-, and 11-year average

1430 homes atrial fibrillation (3.5 new cases per 1,000 person-years)

Danish patient and cause of death registries

The study results found no evidence between wind turbine noise and the occurrence of atrial fibrillation and the occurrence of atrial fibrillation , but no clear statistically significant association in women aged 44 and older.

Poulsen et al 2018a

Danish nationwide registry-based cohort study

614,731 25 1996 2012

Modelled A-weighted night-time noise levels in residential yards at frequencies of 10–10,000 Hz and low-frequency noise in indoor spaces at frequencies of 10–160 Hz. 1-year and 5-year averages

25,148 diabetes cases Danish Diabetes Register

Long-term exposure to wind turbine noise at night in residential yards and low-frequency noise indoors was not associated with the development of diabetes.

Poulsen et al 2018b

Danish nationwide register-based cohort study

535,675 25–84 1996 2013

Modelled A-weighted night-time noise levels in residential yards at frequencies of 10–10,000 Hz and low-frequency noise in indoor spaces at frequencies of 10–160 Hz. 1-year and 5-year averages

83,729 people purchased oral hypertension medication

Data from the Danish Prescription Centre

Long-term exposure to wind turbine noise at night in residential yards and low-frequency noise indoors was not associated with the purchase of oral hypertension medications Poulsen et al 2018c

Poulsen et al 2018c

Danish nationwide register-based cohort study

135,795 single birth births,

122,792 temporary single nitust

NA1 1996– 2013

Modelled night-time A-weighted noise level in the yard of a residential building at frequencies 10–10,000 Hz and low-frequency noise in indoor spaces at frequencies 10–160 Hz. 1-year and 5-year averages

13,003 premature births; 12,220 births with low birth weight for gestational age; 1,127 preterm births with low birth weight

Denmark birth register

Exposure to night-time noise in the outdoor area of residential buildings and low-frequency noise indoors during pregnancy was not associated with adverse birth outcomes.

60

Poulsen et al 2019a

Danish nationwide register-based cohort study

711,249 heart attack 712,401 Stroke

25–84 1996 2013

Modelled A-weighted night-time noise level in residential yards at frequencies of 10–10,000 Hz and low-frequency noise in indoor spaces at frequencies of 10–160 Hz. 1-year and 5-year averages

19,145 heart attacks, 18,064 strokes

Danish patient and cause of death registries

The study found evidence that long-term exposure to night-time noise from wind turbines in residential areas may increase the risk of heart attack and stroke. Low-frequency noise indoors was not associated with the occurrence of heart attacks and strokes .

Poulsen et al 2019b

Danish nationwide register-based cohort study

583,968 sample of sleep medicatio ns;

584,891 sample of antidepr essants

25 1996 2013

Modelled night-time A-weighted noise level in residential yards at frequencies of 10–10,000 Hz and low-frequency noise in indoor spaces at frequencies of 10–160 Hz. 1-year and 5-year averages

68,696 people purchased sleeping pills

82,373 people bought antidepressants

Data from the Danish Prescription Centre

Long-term exposure to night-time wind turbine noise in residential areas was associated with the purchase of sleeping pills and antidepressants among subjects over 65 years of age. Long-term exposure to low-frequency noise indoors was not associated with the purchase of sleeping pills and antidepressants Poulsen et al 2018c

Poulsen et al 2018c

Danish nationwide register-based case- crossover study

15,092 infarction (13,343 people)

14,623 strokes (13,026 people)

≥18 1982 2013

Modelled night-time A-weighted noise level in the yard of the dwelling at frequencies 10–10,000 Hz and low- frequency noise indoors at frequencies 10–160 Hz during the 4 days prior to the onset of illness or the reference day

Heart attack, stroke Danish patient and cause of death registers

Short-term exposure to wind turbine noise at night in residential yards and low-frequency noise indoors was not associated with the occurrence of heart attacks and strokes in the main analysis. Additional analyses found evidence that high levels of low-frequency noise indoors may trigger heart attacks or stroke.

1 NA – no data

61

Table 3. Experiments on infrasound Source Country Study design Sample

size Age Exposure Health outcome Method of measuring

outcome Method

Study results

Marshall et al 2023

Australia Randomised, controlled, double- blind study with three research groups. Only noise-sensitive individuals were included, who participated in three trials. Each trial lasted 72 consecutive hours.

37 18–72 Infra-sound with a frequency of 1.6– 20 Hz and a maximum intensity of 90 dB (Z), simulating the infrasound of wind turbines. The control group was exposed to traffic noise during the night (22:00–07:00) with an average intensity of 40–50 dB (A) and a maximum intensity of 70 dB, or no noise was generated (average background noise during the night 39 dB (A) / 80–85 dB (Z) originating from air conditioning).

Various physiological and psychological indicators were measured in all subjects. Sleep indicators, cardiovascular indicators, psychological and mental well-being indicators, stress indicators from blood samples. The subjects were also asked to assess the presence of symptoms of "wind turbine syndrome".

PSG1 and questionnaires.

Exposure to infrasound did not impair any of the measured health indicators. Traffic noise prolonged the time it took to fall asleep.

Ascone et al 2021

Germany Randomised controlled, unidirectionally blinded long-term exposure (1 month) study. Two groups: infrasound vs. placebo.

38 18 Generated infrasound (6 Hz, 80–90 dB (Z)) or placebo sound in the bedrooms of the subjects for 28 consecutive nights.

Self-reported symptoms, sleep quality, mental functioning.

Questionnaires, mental performance tests, MRI (magnetic

resonance imaging).

Exposure to infrasound did not affect self-reported health, sleep quality or mental abilities. Changes in brain grey matter were observed, but these cannot be interpreted as either harmful or beneficial.

Crichton et al 2014a

New Zealand

Randomised, controlled double-blind provocation study, two study groups: 1) expectations that infrasound is harmful were created based on real information circulating on the internet vs. 2) expectations that infrasound is not harmful were created.

54 Super - learne rs

Generated infrasound (5Hz, 40 dB) and placebo sound in 10-minute sessions.

24 different self-reported symptoms, such as headache, pressure in the ears, dizziness, nausea. Blood pressure and heart rate measured.

Symptoms reported on a scale before and during the sessions.

The group with high expectations of adverse effects reported significant increases in the number and intensity of symptoms compared to the pre-exposure assessment during both the infrasound and placebo sound sessions. No changes were observed in the group with low expectations of adverse effects. No effect of infrasound on blood pressure or heart heart rate. The study shows that real information circulating on the Internet about the negative health effects of infrasound increases the occurrence of self-reported symptoms .

62

Crichton et al 2014b

New Zealand

Randomised, controlled, two study groups: 1) expectation that infrasound is harmful vs 2) expectation that infrasound is beneficial.

60 Super - learne rs

Generated infrasound (9 Hz, 50.4 dB ) and audible sound from a wind farm 1 km away (43 dB) were presented simultaneously in 7-minute sessions.

Twenty-four different self- reported symptoms, plus 12 positive mood expressions and 12 negative mood expressions.

Symptoms and mood were assessed on a scale before each session and during the sessions.

In the negative expectations group, symptom reporting increased and mood worsened during the session, while in the positive expectations group, symptom reporting decreased and mood improved compared to what was reported before the session. The study shows that positive information has a placebo effect.

Crichton et al 2015

New Zealand

Randomised, controlled, two study groups: 1) expectation that infrasound is harmful vs 2) expectation that infrasound is beneficial.

60 Super - learne rs

Generated infrasound (9 Hz, 50.4 dB) and audible sound from a wind farm 1 km away (43 dB) were presented simultaneously in 7-minute sessions.

Disturbance, 12 positive mood expressions and 12 negative mood expressions, plus self-rated noise sensitivity.

Disturbance, mood, and noise sensitivity were assessed on a scale before each session and during the sessions.

The positive expectations group reported less disturbance during the session than the negative expectations group. The study shows that negative information causes annoyance from wind farms, while positive information reduces annoyance, and even among those who are sensitive to noise.

Crichton and Petrie 2015b

New Zealand

Randomised, controlled, two study groups: 1) an expectation was created that infrasound is harmful vs 2) expectation that infrasound is beneficial was created. A repeat experiment was conducted in which the information provided was changed.

64 17–56 Generated infrasound (9 Hz, 50.4 dB) and sound audible to birds 1 km away from the wind farm (43 dB) were presented simultaneously in 10- minute sessions.

Self-reported 24 different symptoms, 12 positive mood expressions and 12 negative mood expressions.

Symptoms and mood were assessed on a scale before each session and during the sessions.

In the negative expectations group, symptom increased and mood deteriorated compared to what was reported before the session, and vice versa. When the negative expectation group was given positive information about the benefits of wind turbines in a repeat experiment, they reported fewer symptoms and their mood improved. Similarly, those who had heard the positive information first experienced a deterioration in mood and more symptoms after receiving the negative information. The results show that positively worded health information can reverse or reduce the impact of negative expectations created by warnings about the health risks of wind turbines published in the media warnings about the health risks of wind turbines.

Crichton and Petrie 2015a

New Zealand

All of the were given negative health effects. The subjects were then randomly divided into two groups: 1) the subjects were given information that infrasound health effects are biologically justified vs 2)

66 17–70 Generated infrasound (9 Hz, 50.4 dB) and audible sound from a wind farm 1 km away (43 dB) were presented simultaneously in 14-minute sessions.

Self-reported 24 different symptoms, 12 positive mood expressions and 12 negative mood expressions.

Symptoms and mood were assessed on a scale before each session and during the sessions. The credibility and comprehensibility of the information shared was also assessed.

After receiving negative information, the number and intensity of reported symptoms increased in both groups compared to the baseline. In the biological explanation group, the increase in symptoms persisted during the second session. In the nocebo explanation group, however, the number and intensity of symptoms decreased and mood improved. The experiment shows that false information found in the media increases the occurrence of symptoms and concern about health . It also shows that the nocebo effect

63

was explained to the subjects that the health effects were the result of the nocebo effect

providing an explanation may reduce the reporting of symptoms associated with wind turbines. Participants in both groups found the explanation they were given to be understandable, reasonable, convincing and correct.

Tonin et al 2016

Australia Randomised double-blind trial of infrasound from headphones 4 groups

72 17 Simulated variable infrasound from wind turbines 0.8–40 Hz with a maximum intensity of 91 dB (Z) 23 minutes or placebo sound

Participants were led to expect that the infrasound would have a harmful effect or that there would be no harmful effects.

Self-reported 24 different symptoms, Concern about the health effects of wind turbines health effects.

Questionnaires before and after the listening session

In the infrasound group, the occurrence of symptoms decreased. It was not possible to shape expectations among the participants. A statistically significant worsening of symptoms was

observed among participants who believed before the experiment that infrasound affects health, regardless of whether they heard infrasound or a placebo sound. The results support the nocebo effect hypothesis.

Liebich et al 2022a

Australia Randomised controlled single- blind trial in a sleep laboratory

68 18 Recorded wind farm noise/infrasound was presented in a sleep laboratory at an intensity of 25 dB(A) on seven consecutive nights. The background noise in the laboratory was 19 dB (A). The wind farm noise contained infrasound from 1.6 Hz and amplitude modulation at frequencies of 31.5 and 63 Hz.

Objectively measured and self- reported sleep quality

PSG1, sleep diary, questionnaires on sleep and noise sensitivity questionnaires

Wind turbine noise at a level of 25 dB (A) indoors has no measurable effect on objective or subjective sleep indicators. No effect was observed even in those who reported wind turbine-related sleep disturbances.

Liebich et al 2022b

Australia Randomised controlled single- blind trial in a sleep laboratory

23 18 Recorded wind farm noise/infrasound was presented in a sleep laboratory at an intensity of 33 dB(A) in random order, alternating with laboratory background noise (23 dB(A)). Wind farm noise contained infrasound and noticeable amplitude modulation at 46 Hz .

Time taken to fall asleep (sleep latency)

PSG1, sleep diary, questionnaires on sleep and noise sensitivity The study shows that wind turbine noise at a level of 33

The study shows that wind turbine noise at a level of 33 dB (A) does not prolong the time it takes to fall asleep, as measured objectively or subjectively in young, healthy people who have not previously been exposed to wind turbines.

Maijala et al 2021

Finland Randomised controlled double-blind trial

26 30 Recorded sound in the wind farm area 200 m from the turbine (47–57 dB (A), 52–77 dB (Z)), in a yard 1.5 km from the turbine (42–59 dB (A)) and indoors (41–43 (A)). The highest sound pressure level and highest amplitude modulation

Measured autonomic nervous system responses (heart rate, heart rate variability, skin conductance).

on the disturbance assessment scale; measured heart rate , heart rate variability,

The study shows that the infrasound levels used in the experiment did not affect the disturbance or autonomic nervous system responses of the subjects, even though the experimental conditions corresponded acoustically to actual wind farms. The subjects did not distinguish between presentations containing infrasound and those without. The or non-presentation did not affect

64

deep recordings. The sound of the ocean shore (34–45 dB (A)) was used as a control. Subjects were exposed or not exposed to infrasound (20 Hz) that had been filtered out of the wind farm sound.

Subjective disturbance assessments

skin conductance disturbance level. The audible sounds from the wind farm were more disturbing than the ocean sounds. Sounds with greater amplitude modulation were also more disturbing. Subjects who reported health concerns related to infrasound prior to the study did not differ in their reactions to infrasound from other .

Małecki et al 2023

Poland Randomised trial, 3 groups, conditions not controlled (conducted in a classroom)

129 21 Different exposures: *Recorded and filtered wind turbine infrasound with an intensity of 83 dB (G) /47 dB (A). *Synthetic infrasound with an intensity of 5–20 Hz at 78 dB (G) / 46 dB (A) and no amplitude modulation or deviation. *Background noise with an intensity of 63 dB (G) / 43 dB (A).

Cognitive functions, especially attention. Feelings and symptoms

Cognitive ability tests, questionnaires before and after the session

The study results showed no significant differences in cognitive test results or in the number of reported unpleasant sensations or complaints between different sound conditions when men and women were analysed separately. Women reported discomfort and various complaints more than men.

Rosciszewska et al 2025

Poland Randomised controlled single-blind trial, 3 groups

45 18 Recorded wind turbine noise from a 2 MW wind turbine at a distance of 500 metres. The sound intensity used was 65.4 dB (Z) / 38.5 dB (A). Wind turbine noise contained amplitude modulation (average frequency 0.8–1 Hz, depth ~6.9 dB). Recorded road traffic noise 65.4 dB (Z) / 56.8 dB(A). There was background noise in the control.

Cognitive functions, disturbance, stress, depression

EEG2 measurements, cognitive ability tests, questionnaires disturbance, depression, anxiety and stress about

Short-term exposure to wind turbine noise did not affect the cognitive (measured by brain functions such as attention and thinking). Wind turbine noise was not perceived as significantly more disturbing or stressful than traffic noise. The participants did not know the source of the noise, which may be the reason why wind turbine noise was not perceived as more disturbing.

1PSG – polysomnography 2EEG – electroencephalography is a method of measuring the electrical activity of the brain

65

Table 4. Experiments involving audible noise and visual aspects of wind turbines Source Country Study design Sample

size Age Exposure Health outcome Method of measuring

outcome Method

Study results

Chiu et al 2021

Taiwan Randomised trial, two groups, but no control conditions.

29 22–75 Low-frequency wind turbine noise (20–200 Hz) was measured 20 m from the nearest wind turbine (38.3–57.1 dB(A)) outdoors and 500 m from the nearest wind turbine indoors (32.2–52.5 dB (A)).

Heart rate and heart rate variability

Portable electrocardiogra m (ECG) recorder

The test showed that exposure to wind turbine noise can reduce heart rate . This may increase the risk of cardiovascular disease.

Ioannidou et al 2016

Denmark Controlled experiment, all subjects listened to and rated sounds with different AM1

sounds.

19 23–28 Sounds with different AM1 frequencies ranging from 200 to 1200 Hz and an intensity of 60 dB (A), based on wind farm recordings, but with the AM1 artificially modified for the experiment artificially for the experiment.

Disturbance Self-assessment on a scale of 1–10

Disturbance is affected by the depth of AM1. The smaller the range in which the noise level fluctuates, the less disturbing the sound is.

Lee et al 2011 South Korea

Controlled experiment, all subjects listened to sounds of varying loudness and varying AM1 sounds.

30 20 Recorded noise from a single wind turbine (250 Hz– 8000 Hz). Participants were presented with noise levels of 35 dB, 40 dB, 45 dB, 50 dB and 55 dB (A) for each AM1 depth level.

Disturbance Self-assessment on a scale of 1–11

The greater the AM1 depth, the greater the disturbance. Disturbance also increased with louder sounds.

Schäffer et al 2018

Switzerland Controlled experiment, all subjects listened to sounds with different frequency distributions and different AM1 values.

52 18 Generated sounds in the range of 16 Hz to 16 kHz with an intensity of 40 dB (A). The tests used pink noise, wind turbine noise (simulated 2 MW Vestas V90 type wind turbine) and noise with an increased low-frequency component.

Disturbance Self-assessment on a scale of 1–11, questionnaire for background data

Disturbance increased with increasing AM1 depth. Disturbance was higher in situations with random AM1 than in situations without AM1 sounds. The sound with an increased low-frequency component was more disturbing than the wind turbine noise. Disturbance was not related to the gender or noise sensitivity of the participants, but was higher with increasing age and lower with a more positive attitude towards wind farms.

Schäffer et al 2016

Switzerland Controlled experiment, all subjects listened to sounds with different characteristics.

60 18 Generated sound with an intensity of 35 and 60 dB(A), simulating a 2 MW Vestas V90 wind turbine operating in strong wind conditions. Sounds with different AM1 were also generated (without AM1, periodic and random AM1). A total of 30 different sound stimuli were used in the study. These represented different situations involving wind turbine and traffic noise, varying in sound pressure level, source type and AM1. Each sound stimulus lasted 25 seconds.

Disturbance Self-assessment on a scale of 1–11, questionnaire for background data

At the same sound intensity, disturbance was greater for wind turbines than for traffic noise. The presence of AM1 increased disturbance.

66

Murcia et al 2017

Spain Experiment 14 18 Sixty images were used as stimuli. These were divided into three groups: images with wind turbines and the same images without them; images with and without solar parks; and images with and without nuclear power plants.

Disturbance, emotions

EEG2, questionnaires The images were scored on a scale from 9 (very pleasant) to 1 (very unpleasant).

Both objective and subjective measurements showed that disturbance and emotions did not differ when the subjects were shown images of landscapes with or without wind turbines and with or without solar panels. However, images with nuclear power plants evoked negative emotions.

1AM – amplitude modulation 2EEG – electroencephalography is a method of measuring the electrical activity of the brain

This is an excerpt from IARO Report (IARO24-5):

“Health Report on a Rural Sheep Farm in Scotland.”

[Critical Review of Marshall N, Cho G, Toelle BG, Tonin R, Bartlett DJ, et al. (2023) The Health Effects of 72 Hours of Simulated Wind Turbine Infrasound: A Double-Blind Randomised

Crossover Study in Noise-Sensitive, Health Adults. Environmental Health Perspectives, 131(3): 1-10. https://pubmed.ncbi.nlm.nih.gov/36946580/ ]

(All Figures and Paragraphs referred to, but not included, in this excerpt can be found in the Full Report, available at iaro.org.nz)

Health Report on a Rural Sheep Farm in Scotland

Page 64 of 169

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 http://IARO.org.nz

6. Other studies cited in the Letter

The dearth of knowledge on the matter at hand continues to be demonstrated by the signatory of the Letter:

“In addition to the impacts of audible noise itself, the contribution from low frequency infrasound to health effects has also been postulated although findings from recent studies have suggested that this is not supported. 297,298 Similarly, Turunen et al. whilst unable to assess a causal relationship due to the cross-sectional nature of the study, suggested that interpretations of symptoms are affected by other factors in addition to the actual exposure.299”

213. For educational purposes,300 a brief review is conducted of the three studies cited above by the NHS-Highland medical representative.

I. Immediate effects of infrasound exposure

214. In the 2023 study by Marshall et al.,301, 302 the objective is stated as follows:

297 Footnote 5 of the Letter. Marshall N, Cho G, Toelle BG, Tonin R, Bartlett DJ, et al. (2023) The Health Effects of 72 Hours of Simulated Wind Turbine

Infrasound: A Double-Blind Randomised Crossover Study in Noise-Sensitive, Health Adults. Environmental Health Perspectives, 131(3): 1-10. https://pubmed.ncbi.nlm.nih.gov/36946580/ [website added]

298 Footnote 6 of the Letter. Maijala PP, Kurki I, Vainio L, Pakarinen S, Kuuramo C, et al. (2021) Annoyance, perception, and physiological effects of wind turbine infrasound. Journal of the Acoustical Society of America, 149(4): 2238-2248. https://pubmed.ncbi.nlm.nih.gov/33940893/ [website added]

299 Footnote 7 of the Letter. Turunen AW, Tittanen P, Yli-Tuomi T, Taimisto P, Lanki T. (2021) Symptoms intuitively associated with wind turbine infrasound. Environmental Research, 192: 1-9. https://pubmed.ncbi.nlm.nih.gov/33131679/ [website added]

300 As indicated in Paragraphs 37 and 40, the primary reason for such a comprehensive approach to this IARO Health Report is to provide an educational and instructive document for the NHS-Highland medical staff, with the ultimate purpose of benefiting the Scottish Citizen.

301 Marshall N, Cho G, Toelle BG, Tonin R, Bartlett DJ, et al. (2023) The Health Effects of 72 Hours of Simulated Wind Turbine Infrasound: A Double-Blind Randomised Crossover Study in Noise-Sensitive, Health Adults. Environmental Health Perspectives, 131(3): 1-10. https://pubmed.ncbi.nlm.nih.gov/36946580/

302 Disclaimer included in the 2023 Marshall et al. paper: “All of the authors have superannuation accounts which are compulsory in Australia and these accounts may contain investments in both traditional and renewable energy, including wind turbines. R.T. is the founding principal of Renzo Tonin Associates who have previously worked as consultants for the NSW Department of Planning on several wind farms in NSW, Australia. None of the investigators have any other pecuniary interest or academic conflicts of interest in the outcomes of this study.“

Health Report on a Rural Sheep Farm in Scotland

Page 65 of 169

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 http://IARO.org.nz

We aimed to test the effects of 72 h of infrasound (1.6–20 Hz at a sound level of ∼ 90 dB pk re 20 microPa, [303 , 304 ] simulating a wind turbine infrasound signature) exposure on human physiology, particularly sleep.

215. In Medical Sciences, this type of study purports to investigate the immediate effects of exposure, as opposed to long-term effects:

Our principal hypothesis was that exposure to infrasound in healthy individuals, at a level of ∼ 90 dB pk re 20 microPa compared with the sham infrasound, increases WASO [305] —a measure of sleep disturbance—and worsens other measures of sleep quality, mood, WTS [306] symptoms, and other electrophysio- logical measures. In addition, as a positive control, we also tested whether audible traffic noise, a mixture of road (motorbike, truck, car) and aircraft noise (at a sound level of 40–50 dB LAeq; night and 70 dB LAFmax transient maxima)

had an adverse impact on these same outcomes, when compared with sham infrasound.307

216. The conclusions of this study were:

Our study found no evidence that 72 h of exposure to a sound level of ∼ 90 dB pk re 20 microPa of simulated wind turbine infrasound in double-blind conditions perturbed any physiological or psychological variable. None of the 36 people exposed to infrasound developed what could be described as WTS. Our study is unique because it measured the effects of infrasound alone on sleep. This study suggests that the infrasound component of WTN [wind turbine noise] is unlikely to be a cause of ill-health or sleep disruption, although this observation should be independently replicated.

217. The dose presented to these subjects “simulating a wind turbine infrasound signature” was questioned by IARO scientists, and correspondence with co-author R. Tonin was exchanged (in May 2023) to ascertain what “simulated wind turbine infrasound” meant.

303 See Appendix 1—Medical Sciences: IV. How is noise quantified?

304 See Appendix 2—Physics of Acoustics: I. What is Sound?

305 WASO = Wakefulness After Sleep Onset is the total number of minutes that an individual is awake after having initially fallen asleep.

306 WTS = Wind Turbine Syndrome. See: Pierpont N. (2009) Wind Turbine Syndrome: A Report on a Natural Experiment. K-Selected Books: Santa Fe, New Mexico, USA. https://www.researchgate.net/publication/265247204_Wind_Turbine_Syndrome_A_Report_on_a_Natural_Experiment

307 Marshall N, Cho G, Toelle BG, Tonin R, Bartlett DJ, et al. (2023) The Health Effects of 72 Hours of Simulated Wind Turbine Infrasound: A Double-Blind Randomised Crossover Study in Noise-Sensitive, Health Adults. Environmental Health Perspectives, 131(3): 1-10. https://pubmed.ncbi.nlm.nih.gov/36946580/ [Footnotes contained in the original text are not included.]

Health Report on a Rural Sheep Farm in Scotland

Page 66 of 169

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 http://IARO.org.nz

218. Regrettably, the material provided by co-author R. Tonin was regarded by IARO scientists as unsatisfactory, if “simulating a wind turbine infrasound signature” was the objective.308

219. Nevertheless, for the sake of scientific discussion, it will be temporarily accepted that the subjects of this study were actually presented with a properly simulated wind turbine infrasound signature.

220. The idea seems to have been to investigate immediate responses to the simulated wind turbine infrasound signature, but as measured by parameters that, perhaps, were not so relevant for assessing immediate responses.309, 310, 311, 312, 313, 314, 315

221. Another questionable practice was the selection of the “healthy individuals” as study subjects. To the understanding of IARO scientists, no evaluation was made regarding prior exposures 316 to infrasound and low frequency noise.317, 318

222. Marshall et al. explain the viewpoint that foundationally justifies their study:

People who suffer from WTS [Wind Turbine Syndrome 319] report that their symptoms begin quickly when they are exposed to infrasound from wind

308 The acoustic pattern used to simulate the wind turbine signal had a sawtooth profile, not the short-duration pulses of WTAS,

see Figure 3. A sawtooth-shaped wave has a quick onset, a slow decay, and only locally oscillates the air. WTAS has a rapid onset and decay, and ‘pumps the air’ (as proposed by Dr Stephan Kaula, Germany), rather than only causing the local oscillations that are typically seen in airborne, acoustic propagation phenomena.

309 See Appendix 4—Clinical & Biological Matters, Section 3-Occupational and Residential Exposures: I. Why are occupational exposures important to understand environmental exposures?

310 See Appendix 4—Clinical & Biological Matters, Section 3-Occupational and Residential Exposures: II. What extra-auditory medical conditions do noise-exposed workers develop?

311 See Appendix 4—Clinical & Biological Matters, Section 3-Occupational and Residential Exposures: III. Do the extra-auditory medical conditions seen in noise-exposed workers also emerge in residential infrasonic exposures?

312 Mohr GC, Cole JJN, Guild E, von Gierke HE. (1965) Effects of low-frequency and infrasonic noise on man. Aerospace Medicine, 36: 817-24.

313 Ponomarkov VI, Tysik A, Kudryavtseva VI, Barer AS. (1969) Biological action of intense wide-band noise on animals. Problems of Space Biology NASA TT F-529, 7(May): 307-9.

314 Castelo Branco NAA, Gomes-Ferreira P, Monteiro E, Costa e Silva A, Reis Ferreira J, Alves-Pereira M. (2003) Respiratory epithelia in Wistar rats after 48 hours of continuous exposure to low frequency noise. Journal of Pneumology, formerly Revista Portuguesa Pneumologia, IX (6): 474-79. https://pubmed.ncbi.nlm.nih.gov/15190432/

315 Castelo Branco NAA, Reis Ferreira J, Alves-Pereira M. (2007). Respiratory pathology in vibroacoustic disease: 25 years of research. Journal of Pneumology, formerly Revista Portuguesa Pneumologia, XIII (1): 129-135. https://pubmed.ncbi.nlm.nih.gov/17315094/

316 Including, foetal, childhood and young adult exposures in residential, occupational, and leisurely settings. See Appendix 1— Medical Sciences: II. What parameters are important when investigating the biological effects of exposures to physical agents of disease.

317 See Appendix 1—Medical Sciences: X. How are control populations selected for noise studies.

318 See Appendix 1—Medical Sciences: XI. What happens when control populations are incorrectly selected?

319 Pierpont N. (2009) Wind Turbine Syndrome: A Report on a Natural Experiment. K-Selected Books: Santa Fe, New Mexico, USA. https://www.researchgate.net/publication/265247204_Wind_Turbine_Syndrome_A_Report_on_a_Natural_Experiment

Health Report on a Rural Sheep Farm in Scotland

Page 67 of 169

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 http://IARO.org.nz

turbines and are then sustained.[320] Our scientifically robust study provides evidence to address this claim. The Australian NHMRC [National Health and Medical Research Council] report that gave rise to our study made note of this “absence of evidence” rather than concluding an “evidence of absence” owing to the lack of any laboratory-controlled double-blind experiments of sufficient duration and intensity to hypothetically induce WTS in a human.321

223. “Induce WTS in a human”? 322 As far as is understood by IARO scientists, WTS is not commonly viewed as an immediate effect of the exposure to this agent of disease.323

224. The expression “laboratory-controlled double-blind experiments of sufficient duration and intensity” as applied to the matter at hand is simultaneously unethical, dangerous, and unnecessary.324, 325

225. Is it the desire of the Australian NHMRC to expose subjects to a toxic agent—which is very difficult, if not impossible, to reproduce in laboratory settings—until some clearly severe health endpoint is observed? While tens of thousands of citizens are sitting in real- life laboratories being ‘accused’ of developing psychosomatic disorders? 326

226. This methodology is considered by IARO scientists to reflect sub-standard practices of Scientific Inquiry.

320 See Appendix 4—Clinical & Biological Matters, Section 1-Cellular and Tissue Biology. III. Biological tissues are viscoelastic—What does this

mean?

321 Marshall N, Cho G, Toelle BG, Tonin R, Bartlett DJ, et al. (2023) The Health Effects of 72 Hours of Simulated Wind Turbine Infrasound: A Double-Blind Randomised Crossover Study in Noise-Sensitive, Health Adults. Environmental Health Perspectives, 131(3): 1-10. https://pubmed.ncbi.nlm.nih.gov/36946580/ [Footnotes contained in the original text are not included.]

322 “The causes of this syndrome have been the subject of substantial international controversy. Proponents have contended that the symptoms that compose this syndrome are caused by low frequency subaudible infrasound generated by wind turbines. Critics have argued that these symptoms are psychological in origin and are attributable to nocebo effects. The Australian National Health and Medical Research Council Wind Farms and Human Health Reference Group concluded that the available evidence was not sufficient to establish which, if either, of these explanations is correct.” See: Marshall N, Cho G, Toelle BG, Tonin R, Bartlett DJ, et al. (2023) The Health Effects of 72 Hours of Simulated Wind Turbine Infrasound: A Double-Blind Randomised Crossover Study in Noise-Sensitive, Health Adults. Environmental Health Perspectives, 131(3): 1-10. https://pubmed.ncbi.nlm.nih.gov/36946580/

323 Pierpont N. (2009) Wind Turbine Syndrome: A Report on a Natural Experiment. K-Selected Books: Santa Fe, New Mexico, USA. https://www.researchgate.net/publication/265247204_Wind_Turbine_Syndrome_A_Report_on_a_Natural_Experiment

324 What kind of “laboratory-controlled double-blind experiments of sufficient duration and intensity” were conducted for asbestos contamination leading to asbestosis? Or for issues related to second-hand smoking, use of glyphosates, etc?

325 Alves-Pereira M, Rapley B, Bakker H, Summers R. (2019) Acoustics and Biological Structures. In: Abiddine Fellah ZE, Ogam E. (Eds) Acoustics of Materials. IntechOpen: London. DOI: 10.5772/intechopen.82761.

326 In the opinion of IARO scientists, had this study been performed on 3 groups of people, differentiated by the extent of their prior exposures (mild, moderate, or extensive), and, abiding by appropriate selection criteria of the study population, then, perhaps, statistically useful numbers could have been obtained, and scientifically useful results could have been achieved. The inability to reproduce ‘wind turbine infrasound’ under laboratorial conditions, however, would still render this study as irremediably flawed, while its overall design could be deemed ethically questionable.

Health Report on a Rural Sheep Farm in Scotland

Page 68 of 169

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 http://IARO.org.nz

227. In conclusion, in the opinion of IARO scientists, the effort expended by these authors to conduct this study is laudable (particularly given the position of the Australian NHMRC), even though, scientifically, within the realm of Medical Sciences and dose-response relationships, its results are inconsequential.

II. The Government-Sponsored Finnish Study

228. The 2021 study by Maijala et al.327 is based on the 169-page 2020 Governmental Report on a Research Project carried out by Maijala et al.328

229. The main objective was “to find out whether wind turbine infrasound has harmful effects on human health.”329

230. Table 3 lists the specific objectives of this 2020 Research Project.

Table 3. Specific objectives of the 2020 Research Project sponsored by the Government of Finland.330

A. To characterize wind turbine noise as an exposure

1 What are the full spectrum sound levels, down to 0.1 Hz, inside houses near the wind power plants?

2 What are the characteristics of the sound, both audible and inaudible infrasound?

B. To describe symptoms that are intuitively associated with infrasound from wind turbines, i.e., wind turbine infrasound related symptoms.

3 What is the prevalence of wind turbine infrasound related symptoms in the vicinity of wind power plants?

327 Maijala PP, Kurki I, Vainio L, Pakarinen S, Kuuramo C, et al. (2021) Annoyance, perception, and physiological effects of wind turbine infrasound. Journal

of the Acoustical Society of America, 149(4): 2238-2248. https://pubmed.ncbi.nlm.nih.gov/33940893/

328 Maijala P, Turunen A, Kurki I, Vainio L, Pakarinen S, et al. (2020) Infrasound does not explain symptoms related to wind turbines. Publications of the Finnish Government’s Analysis, Assessment and Research Activities, 2020:34. Prime Minister’s Office: Helsinki. https://julkaisut.valtioneuvosto.fi/handle/10024/162329

329 Maijala P, Turunen A, Kurki I, Vainio L, Pakarinen S, et al. (2020) Infrasound does not explain symptoms related to wind turbines. Publications of the Finnish Government’s Analysis, Assessment and Research Activities, 2020:34. Prime Minister’s Office: Helsinki. pp. 6. https://julkaisut.valtioneuvosto.fi/handle/10024/162329.

330 Maijala P, Turunen A, Kurki I, Vainio L, Pakarinen S, et al. (2020) Infrasound does not explain symptoms related to wind turbines. Publications of the Finnish Government’s Analysis, Assessment and Research Activities, 2020:34. Prime Minister’s Office: Helsinki. pp. 6-7. https://julkaisut.valtioneuvosto.fi/handle/10024/162329.

This is an excerpt from IARO Report (IARO24-5):

“Health Report on a Rural Sheep Farm in Scotland.”

[Critical Review of Maijala P, Turunen A, Kurki I, Vainio L, Pakarinen S, et al. (2020) Infrasound does not explain symptoms related to wind turbines. Publications of the Finnish Government’s Analysis, Assessment and Research Activities, 2020:34. Prime Minister’s Office: Helsinki.

https://julkaisut.valtioneuvosto.fi/handle/10024/162329]

(All Figures and Paragraphs referred to, but not included, in this excerpt can be found in the Full Report, available at iaro.org.nz)

Health Report on a Rural Sheep Farm in Scotland

Page 68 of 169

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 http://IARO.org.nz

227. In conclusion, in the opinion of IARO scientists, the effort expended by these authors to conduct this study is laudable (particularly given the position of the Australian NHMRC), even though, scientifically, within the realm of Medical Sciences and dose-response relationships, its results are inconsequential.

II. The Government-Sponsored Finnish Study

228. The 2021 study by Maijala et al.327 is based on the 169-page 2020 Governmental Report on a Research Project carried out by Maijala et al.328

229. The main objective was “to find out whether wind turbine infrasound has harmful effects on human health.”329

230. Table 3 lists the specific objectives of this 2020 Research Project.

Table 3. Specific objectives of the 2020 Research Project sponsored by the Government of Finland.330

A. To characterize wind turbine noise as an exposure

1 What are the full spectrum sound levels, down to 0.1 Hz, inside houses near the wind power plants?

2 What are the characteristics of the sound, both audible and inaudible infrasound?

B. To describe symptoms that are intuitively associated with infrasound from wind turbines, i.e., wind turbine infrasound related symptoms.

3 What is the prevalence of wind turbine infrasound related symptoms in the vicinity of wind power plants?

327 Maijala PP, Kurki I, Vainio L, Pakarinen S, Kuuramo C, et al. (2021) Annoyance, perception, and physiological effects of wind turbine infrasound. Journal

of the Acoustical Society of America, 149(4): 2238-2248. https://pubmed.ncbi.nlm.nih.gov/33940893/

328 Maijala P, Turunen A, Kurki I, Vainio L, Pakarinen S, et al. (2020) Infrasound does not explain symptoms related to wind turbines. Publications of the Finnish Government’s Analysis, Assessment and Research Activities, 2020:34. Prime Minister’s Office: Helsinki. https://julkaisut.valtioneuvosto.fi/handle/10024/162329

329 Maijala P, Turunen A, Kurki I, Vainio L, Pakarinen S, et al. (2020) Infrasound does not explain symptoms related to wind turbines. Publications of the Finnish Government’s Analysis, Assessment and Research Activities, 2020:34. Prime Minister’s Office: Helsinki. pp. 6. https://julkaisut.valtioneuvosto.fi/handle/10024/162329.

330 Maijala P, Turunen A, Kurki I, Vainio L, Pakarinen S, et al. (2020) Infrasound does not explain symptoms related to wind turbines. Publications of the Finnish Government’s Analysis, Assessment and Research Activities, 2020:34. Prime Minister’s Office: Helsinki. pp. 6-7. https://julkaisut.valtioneuvosto.fi/handle/10024/162329.

Health Report on a Rural Sheep Farm in Scotland

Page 69 of 169

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 http://IARO.org.nz

4 What factors are associated with wind turbine infrasound related symptoms?

C. To study how infrasound produced by wind turbines affects humans, in particular, perception, annoyance, and physiological responses

5 Can low-frequency and infrasound wind turbine noise be perceived at typical and at extreme noise levels?

6 What is the dependence between the depth of amplitude modulation and annoyance at low frequencies?

7 Does infrasound increase reported annoyance and psychophysiological responses?

8 What is the reactivity of the autonomic nervous system (ANS) to audible wind turbine sounds and its infrasound?

9 Are individuals who attribute their symptoms to wind turbines more sensitive to infrasound? Are they more able to detect infrasound and do they experience more annoyance compared to controls?

231. Objectives A1 and A2 were accomplished, and Figure 7 shows a representative example of the identified ‘dose.’

Health Report on a Rural Sheep Farm in Scotland

Page 70 of 169

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 http://IARO.org.nz

Figure 7. Representative example of the noise characterization (Raahe, indoors, 600-second sample). 331 LZ levels refer to unweighted dB values. LG refers to G-weighted values.332 LA refers to A-weighted values. Maximum and minimum LZ values are shown as curves.

232. Figure 7 shows a one-third-octave-band segmentation of the acoustic spectrum (similar to that shown in Figure 2). The solid black curve (LZ max) shows the highest sound pressure levels measured in unweighted dB.

233. There is no cut-off of spectral data as was seen in Figure 6 (i.e., the lower limiting frequency is 0.1 Hz and not 10 Hz), but there is also no recognition of a “wind turbine infrasound signal” as in the previous Marshall et al. study (see Paragraph 214). It was however recognized that “the most important frequencies were less than 2 Hz.”333

234. Objectives B3 and B4 (see Table 3) were more difficult to achieve, as “infrasound related symptoms” were established by questionnaires and telephone calls. While these types of surveys may have a certain usefulness, their direct results cannot be considered as a measure of Response within the realm of the Medical Sciences’ dose-response

331 Maijala P, Turunen A, Kurki I, Vainio L, Pakarinen S, et al. (2020) Infrasound does not explain symptoms related to wind turbines. Publications of the

Finnish Government’s Analysis, Assessment and Research Activities, 2020:34. Prime Minister’s Office: Helsinki. pp. 21. https://julkaisut.valtioneuvosto.fi/handle/10024/162329.

332 See Appendix 2—Physics of Acoustics: V. Can infrasound be measured in dBC or dBG?

333 Maijala P, Turunen A, Kurki I, Vainio L, Pakarinen S, et al. (2020) Infrasound does not explain symptoms related to wind turbines. Publications of the Finnish Government’s Analysis, Assessment and Research Activities, 2020:34. Prime Minister’s Office: Helsinki. pp. 77. https://julkaisut.valtioneuvosto.fi/handle/10024/162329.

Health Report on a Rural Sheep Farm in Scotland

Page 71 of 169

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 http://IARO.org.nz

relationship,334 nor as per the WHO definition of noise-induced adverse health effects (see Paragraph 189).

235. Furthermore, there seems to not have been any stratification of the study population regarding prior noise exposure histories.335

236. Objectives C5 through C9 used “provocation experiments” conducted in an “infrasound chamber” whereby “systematically selected samples from real wind turbine sounds from wind power plant areas where inhabitants report symptoms associated with wind turbine infrasound or sound were used as stimuli.”336

237. As with the study by Marshall et al. (Paragraphs 224 to 226), it is not entirely understood why there is a perceived need to subject individuals in laboratory to a potentially noxious agent (which is very difficult, if not impossible, to reproduce under laboratorial conditions), while tens of thousands of individuals are living in ‘real-life laboratories,’ awaiting an objective, clinical observational study on behalf of the competent authorities.337

III. Intuitive symptoms

238. In the third study of this series, the goal of Turunen et al.338 was to assess “the prevalence and severity of these wind turbine infrasound related symptoms:”

No matter what the true cause for the symptoms is, it is clear that symptoms are real and lead to worry, decreased quality of life, and potentially further to deteriorated health. High prevalence of this kind of phenomenon could be a serious threat to public health. The aim of this questionnaire study was to describe symptoms intuitively associated with infrasound from wind turbines.339

334 See Appendix 1—Medical Sciences: VIII. How is ‘Response’ measured?

335 See Appendix 1—Medical Sciences: II. What parameters are important when investigating the biological effects of exposures to physical agents of disease?

336 Maijala P, Turunen A, Kurki I, Vainio L, Pakarinen S, et al. (2020) Infrasound does not explain symptoms related to wind turbines. Publications of the Finnish Government’s Analysis, Assessment and Research Activities, 2020:34. Prime Minister’s Office: Helsinki. pp. 36 and 40. https://julkaisut.valtioneuvosto.fi/handle/10024/162329.

337 Although it is unclear to IARO scientists who (or what agency) could be classified as ‘the competent authorities.’

338 Please note that the authors of this study are the same as those of the Finnish Governmental study by Maijala et al. (see Paragraph 228), and the data collected through questionnaires and telephone calls in the Maijala et al. study are the same data used in this study. See: Turunen AW, Tittanen P, Yli-Tuomi T, Taimisto P, Lanki T. (2021) Symptoms intuitively associated with wind turbine infrasound. Environmental Research, 192: 1-9. https://pubmed.ncbi.nlm.nih.gov/33131679/

339 Turunen AW, Tittanen P, Yli-Tuomi T, Taimisto P, Lanki T. (2021) Symptoms intuitively associated with wind turbine infrasound. Environmental Research, 192: 1-9. https://pubmed.ncbi.nlm.nih.gov/33131679/

Review of

Wind Turbine Health Impact Study:

Report of Independent Expert Panel

as prepared for

Massachusetts Department of Environmental Protection

Massachusetts Department of Public Health

By

Mariana Alves-Pereira, Associate Professor

Faculty of Economics and Management

School of Health Sciences

Universidade Lusofona

Lisbon, Portugal

March 2012

Page 2 of 9

Background

As a leading expert on the biological response to low frequency noise exposure (see brief

biographical background offered at the end of this document), I was requested to provide a

review of the Jan 2012 Report of Independent Expert Panel, prepared for the

Massachusetts Department of Environmental Protection (MassDEP) and Department of

Public Health (MDPH), titled "Wind Turbine Health Impact Study".

Disclaimer

a) The author of this review is not party to anti-technology sentiments;

b) Wind turbines are considered by this author as welcome additions to modern

technological society;

c) The review provided herein has one, and only one, agenda - that of pure scientific

inquiry;

d) In no way can or should this review be construed as a document arguing for or against

the implementation of wind turbines;

e) There are no commercial, financial or professional agreements (contractual or

otherwise) between the author of this review and any persons or parties involved in the

wind turbine sector or persons or parties who stand against the implementation of wind

turbines;

f) This review was provided pro bono.

Goal

To provide a review of the aforementioned Report, within the author’s area of expertise

and therefore, exclusively focused on the infrasound and low frequency noise health

issues claimed to be associated with wind turbines (WT) operations.

Panel Charge

The Panel who authored the Report was charged with several tasks, the first of which is

succinctly stated as follows:

"Identify and characterize attributes of concern (eg noise, infrasound, vibrations) (...) and

identify any scientifically documented or potential connection between health impacts

associated with [land-based] wind energy turbines" (p.vi).

While identification and characterization of the attributes of concern might be a fairly easy

task to accomplish, finding scientifically documented connections between health impacts

and WT operations is almost an impossible task - not because such health impacts are

Page 3 of 9

non-existent, but rather because scientifically sound studies on this subject are sparse for

reasons discussed ahead.

A second charge of this Panel was:

"Evaluate and discuss information (...) on the nature and type of health complaints

commonly reported by individuals who reside near existing wind farms".

Noise annoyance seems to be the most consistent parameter associated with the

acoustical phenomena purportedly emanated by WT. Sleep disturbances and decreased

quality of life are also outcomes that have been assessed in populations living in the

vicinity of WT. While noise annoyance, sleep disturbances and decreased quality of life go

hand in hand with health deterioration, these parameters do not constitute objective clinical

data.

This is an unfortunate situation for the Panel since it limits the evaluation and discussion to

subjective parameters, known to vary in accordance with psychosocial factors. Negative or

positive health impacts due to any situation usually require confirmation, or at least

corroboration from clinical data. Questionnaires with self-reported symptoms provide a

type of subjective data that is usually considered insufficient to clearly establish a positive

or negative health effect.

Why Annoyance?

Despite the lack of scientific objectivity, determining annoyance levels seems to be the

preferential method to evaluate the health effects of individuals living in the vicinity of WT.

There may be several reasons for this:

1. In 1977, the U.S. Office of Noise Assessment established the relationship between

noise exposure level and the proportion of the community that is highly annoyed by

noise1. Through direct measurement based on numerous studies of large populations,

the annoyance parameter was determined to be useful as a noise predictor.

Annoyance rapidly achieved importance because it quickly replaced the term

"nuisance". In terms of legal jargon, "nuisance" can imply liability, while annoyance

usually does not.

2. Annoyance is easily evaluated through appropriate questionnaires. No clinical

physician is required to assess levels of annoyance among a noise-exposed

population. Acousticians are therefore qualified to assess the "health effects" (i.e.

annoyance), while no objective clinical data is actually gathered.

1 Office of Noise Abatement and Control. (1977). The urban noise survey. Environmental Protection Agency: Washington D.C.

Page 4 of 9

3. Grants evaluating annoyance among a noise-exposed populations are generally

reviewed by public health experts and epidemiologists, and only rarely by clinical

physicians. Claiming that health effects are being ascertained merely through

questionnaires evaluating (subjective) levels of noise annoyance would indeed surprise

any clinician.

As a consequence of these situations, health effects due to the proximity of WT to

residential areas are, essentially, unknown to peer-reviewed science - not because they

are inexistent, but because they are not the object of scientific study.

Loaded dice

In a way, this Panel was charged with the task of rolling loaded dice. Peer-reviewed

studies investigating the impact on human health of WT noise exposure practically do not

exist. Those that claim to study just that, fail when objective clinical outcomes are non-

existent end-points. Hence the Panel's charge, more than difficult is quite near impossible.

Literature survey

As stated by the Panel: "Because peer-reviewed literature (...) was relatively limited, we

also examined several non-peer reviewed papers, reports and books that discussed health

effects of wind turbines" (p.15).

As a result, 8 studies were reviewed, 4 of which were peer-reviewed:

Authors Parameter(s)

Peer-reviewed

Pederson et al. 2004 Annoyance questionniare + dBA

Pederson et al. 2007 Annoyance questionniare + dBA

Pederson et al. 2009 Mailed surveys + dBA

Shepard et al. 2011 Quality of life questionnaire

Non-Peer-Reviewed

Van den Berg et al. 2008 General health questionnaire + dBA

Phipps 2007 Survey

Pierpont 2009 Survey

Nissenbaum et al. 2011 Questionnaire + sleep disturbances

All these studies purport to study health effects through questionnaires, surveys and

queries. None provide corroborating clinical evidence. Moreover, of the 8 studies, 4 can be

considered to be authored by the same team.

It would seem that a precious and scientifically useful source of information was

overlooked - scientific conferences. Perhaps it would have been helpful to the Panel if

scientific research papers included in conference proceedings had not been excluded.

Page 5 of 9

Although papers presented at conferences are not considered to be peer-reviewed, they

are subjected to scientific scrutiny and might have provided the Panel with a broader

background, potentially useful for carrying out its charge. The Wind Turbine Noise

Conference and the International Conference on the Biological Effects of Noise are but two

examples of such sources.

Standing with these 8 studies and with the aforementioned charge is not a position one

would eagerly seek to be in.

Human hearing threshold and the dBA unit

Classically speaking, the impact of acoustical phenomena on humans has been limited to

the segment of the acoustical spectrum where the combination of pressure and frequency

allow the acoustical phenomenon to be perceived by humans.

This limitation is what justifies the use of the dBA unit when assessing noise among

human populations. The A-weighting system simulates human hearing, measuring the

loudness of acoustical phenomena.

The continued use of this same dBA unit to acoustically assess environments that are

suspected of being ILFN-rich is, however, scientifically indefensible. Hence, studies

purporting to characterize acoustical environments suspected of being rich in ILF

components, but presented entirely in dBA units are not scientifically valid.

As stated by the World Health Organization:

Noise measurements based solely on LAeq values do not adequately characterize

most noise environments and do not adequately assess the health impacts of noise

on human well-being. (…) If the noise included 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.

When prominent low-frequency components are present, measures based on A-

weighting are inappropriate. However, the difference between dBC (or dBLin) and

dBA will give crude information about the presence of low-frequency components in

noise. If the difference is more than 10 dB, it is recommended that a frequency

analysis of the noise be performed.2

2 World Health Organization. (1999). Guidelines for community noise. Berglund, B., Lindvall, T. and Schwela, D.H. (eds). World Health Organization, Geneva.

Page 6 of 9

Wrong assumptions and flawed study designs

The use of the dBA unit and the focus on human hearing threshold values are justified

however, by the assumption that acoustical phenomena are only harmful if perceived by

the human being.

• Can acoustical phenomena that are not perceived by the human auditory system be

detrimental to human health?

Once this question is set forth, results of studies where subjective parameters are the sole

outcome become moot.

• Does an agent of disease have to be perceived by the host for it to have a

pathogenic effect on the host?

• Does an agent of disease have to cause annoyance in order for it to have a

pathogenic effect on the host?

Clearly the answer is no.

Nevertheless, where acoustical phenomena are concerned, this is an established

assumption of a vast number of researchers and scientists who study "health effects" of

noise exposure. The idea "what you can't hear won't hurt you" is responsible for numerous

biased study designs which, in turn, have been leading to inconclusive or invalid results

(even if peer-reviewed). This has been true for noise studies whether or not they involve

WT, and has further justified the use of the dBA unit.

This wrong assumption which permeates throughout the area of science studying the

health effects of noise exposure justifies ignoring that noise-exposure effects are

cumulative. As a result, noise-exposure histories (including fetal exposures) which could

provide crucial information for establishing dose-responses are not obtained.

Lessons from ILFN-rich occupational environments.

Scientists with expertise in Environmental, Public or Occupational Health are well aware

that excessive exposure to physical agents is often first seen in occupational

environments. The health effects observed in workers have often been later observed in

populations exposed to the same physical agent, but continuously and at a lower level.

"The workplace is a unique environment. (...) Environmentally induced diseases

have (...) not uncommonly first been seen in working populations. The

Page 7 of 9

appearance of these illnesses may provide a warning to the general population

of the toxicity of environmental substances”.3

After several readings of this Report, it would seem that the Panel has, at times,

misunderstood the distinction between noise and vibration where human health is

concerned (p. ES-5, 45, 54).

Noise versus vibration within the context of human health

Infrasound and low frequency noise are airborne acoustical phenomena.

Infrasound is internationally classified as non-ionizing radiation.

Vibration is considered to be the transmission of mechanical vibrations when

the human is in direct physical contact with the vibrating structure, such as a

jackhammer (hand-arm vibration) or a vibrating platform (whole body vibration).

Airborne acoustical phenomena (which may or may not be audible to humans)

can cause vibration in structures existing along its propagation pathway,

depending on numerous variables. Similarly, a vibrating structure can originate

the emanation of airborne pressure waves (which may or may not be audible to

humans).

Vibroacoustic disease (VAD) does not "require a very clear coupling to large

vibration sources such as jackhammers and heavy equipment" (p. 45). The

physical agent of disease responsible for the development of VAD is airborne

acoustical phenomena, and not vibrations, as they are defined within the scope

of human health effects.

Lessons learned with VAD bring the possibility of objective clinical data being gathered

among populations residing in the vicinity of WT. Moreover, if the agent of disease

responsible for the development of VAD in occupational environments had been more

thoroughly explored (and understood) perhaps the "Panel's efforts (...) to examine the

biological plausibility or basis for the health effects of turbines" (p.ES-3) would have been

greatly improved.

An organic response to ILFN exposure has been consistently identified in ILFN-exposed

workers, animal models, and dwellers in ILFN-rich environments not generated by WT:

abnormal proliferation of collagen in the absence of an inflammatory process4. This

feature, however, cannot be evaluated through questionnaires.

3 Baker DB, Landrigan PJ. (1990). Occupationally related disorders. Environmental Medicine, 74, 441-60. 4 Alves-Pereira M, Castelo Branco NAA. (2007). Vibroacoustic disease: Biological effects of infrasound and low frequency noise explained by mechanotransduction cellular signaling. Progress Biophysics & Molecular Biology, 93, 256-79.

Page 8 of 9

Moreover, to design a study that adequately investigates the ILFN-induced pathology

potentially being developed among populations living in the vicinity of WT requires

knowledge not only in acoustics and clinical medicine, but also histology and cellular

mechanics. Clearly, not an easy task

Commentary on the Panel's findings regarding health impacts of noise and vibration

There is, indeed, "limited evidence suggesting an association between exposure to wind

turbines and annoyance" (p.ES-5, 54) because there are only 3 or 4 reported studies on

the subject, and not all of them agree.

The notion of the noise annoyance parameter being "independent from the effects of

seeing a wind turbine and vice-versa" (p.ES-5, 54), clearly emphasizes the inadequacy

and inappropriateness of selecting this parameter to evaluate "health effects". In terms of

both field work and research grant submission procedures however, it is evidently more

convenient to apply questionnaires to a study population than to provide objective medical

diagnostic tests.

Regarding sleep disruptions, although a definitive predictor for severe health problems, the

underlying rationale remains flawed: disruptions are caused by the audible portion of

acoustical phenomena. ILFN-exposed works suffer sleep disruptions even though they are

not exposed to ILFN during their sleep time. Most likely, individual cumulative effects of

ILFN-exposures play a crucial role in sleep patterns.

Unsurprisingly, "there is insufficient evidence that the noise from wind turbines is directly

(...) causing health problems or disease" (p.ES-6, 55). While this is true because no

studies exist, it could be erroneously interpreted as meaning that existing studies provide

insufficient evidence.

By "measures of psychological distress or mental health" (p.ES-7, 56), it is meant the

result of surveys and questionnaires. Given the nature of the agent of disease - airborne

pressure waves - it stands to reason that organic lesions may occur before measures of

psychological distress and mental health reach levels considered problematic. By the time

they do, lesions will most likely be irreversible.

It is not the charge of this Panel to recommend future studies, and yet it was charged with

"identifying documented best practices that could reduce potential human health impacts"

(p.vi). Considering that human health impacts associated with living in the vicinity of WT

are not the object of any of the 8 studies reviewed by the Panel, the usefulness of the best

practices as provided by the Panel regarding noise (p.59-61) can only be questionable.

Page 9 of 9

In conclusion

The Panel's charge is not an enviable one since it is nearly impossible to carry out. The

health impacts on populations living in the vicinity of WT are, simply put, not documented.

Health impacts are not scientifically evaluated through questionnaires and surveys.

Instead, objective clinical data are required which, in this case, do not exist.

The authorities who requested this Report (MassDEP and MDPH) will most likely not find it

very useful if their priority is the health of populations living near WT. However, if other

agendas exist, this Report may become relevant.

Mariana Alves-Pereira

Brief Biographical Background for the author of this Review:

Mariana Alves-Pereira holds a B.Sc. in Physics (State University of New York at Stony Brook), a M.Sc. in

Biomedical Engineering (Drexel University) and a Ph.D. in Environmental Sciences (New University of

Lisbon).She joined the multidisciplinary research team investigating the biological response to infrasound

and low frequency noise in 1988, and has been the team’s Assistant Coordinator since 1999. Recipient of

three scientific awards, and author and co-author of over 50 scientific publications (including peer-reviewed

and conference presentations), Dr. Alves-Pereira is currently Associate Professor at Lusófona University

teaching Biophysics and Biomaterials in health science programs (nursing and radiology), as well as Physics

and Hygiene in workplace safety & health programs. Mariana Alves-Pereira is a U.S. citizen and can be

readily reached at: [email protected].

Critical Analysis and Scientific Review of the Report produced by the University of Tartu

in October 2025, titled:

Health effects of wind turbines: A systematic review of studies published in peer-reviewed scientific journals over the last fifteen years—Development of a methodology for interpreting the results of scientific studies on the potential health effects of wind farms and other energy production technologies in the Estonian context

Document IARO25-6

October 13, 2025

Review of Tartu Univeristy Report, Estonia

Page 2 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

International Acoustics Research Organization

IARO is an international group of researchers with a mission to investigate acoustical environments, especially with respect to features that affect humans and animals, and to publish the results. IARO holds the ethics approval for the CSI-ACHE, the Citizen Science Initiative into Acoustical Characterisation of Human Environments, the results of which are publicly disseminated.

Contacts:

IARO, 37 Weston Ave, Palmerston North, 4414, New Zealand

Tel: +64 21 033 6528

Email: [email protected]

Authors of this Report (alphabetical)

Mariana Alves-Pereira, Ph.D., Lusófona University, Lisbon, Portugal

Huub Bakker, Ph.D., IARO, Palmerston North, New Zealand

Richard Mann, Ph.D., Waterloo University, Canada

Rachel Summers, MSc., IARO, Palmerston North, New Zealand

Acknowledgements

The authors of this report would like to acknowledge the longstanding assistance of Dr Bruce Rapley of Sound Analytics. The authors would also like to acknowledge the many insights provided by Les Huson of L Huson & Associates and the vast experience in acoustics made available by Dr Philip Dickinson, Senior Researcher at IARO.

Review of Tartu Univeristy Report, Estonia

Page 3 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

CONTENTS

EXECUTIVE SUMMARY 4

A. INTRODUCTION 5 I. Background 5 II. Goal 6 III. Disclaimer 6 IV. International Acoustics Research Organization, IARO 7 V. Acronyms and Variables Used in IARO Reports 7

B. ORGANIZATION OF THIS REPORT 8 I. Sequential approach to various aspects 8 II. To the Authors of the TU Report 8

C. PURPOSE AND RESEARCH QUESTIONS UNDERLYING THE TU REPORT 10 I. Purpose of the TU Report Study 10 II. Research Questions 11

D. WHAT IS A NOISE-INDUCED HEALTH EFFECT? 15

E. WHY CURRENT NOISE MEASURING METHODOLOGIES ARE NON-APPLICABLE FOR WIND TURBINE NOISE 19

F. ‘WHAT YOU CAN’T HEAR, CAN’T HURT YOU’ 22

G. LUXURIES NOT AFFORDED TO SCIENTISTS—A GLIMPSE OF THE TEDIOUS WORK REQUIRED TO UPHOLD SCIENTIFIC RIGOUR. 24

H. A CANDID CONVERSATION AMONG SCIENTISTS 29 I. The ‘nocebo effect’ narrative 29 II. The questionnaire approach 32 III. Another ‘Scientific Authorship’ of another “Wind Turbine Health Impact Study”… 33

I. CONCLUSIONS 35

ANNEX A: English Translation of the TU report ANNEX B: Critical Review of Marshall et al. study (2023) ANNEX C: Critical Review of Maijala et al. study (2020) ANNEX D: Response to Massachusetts Independent Expert Panel (2012)

Review of Tartu Univeristy Report, Estonia

Page 4 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

EXECUTIVE SUMMARY

1. On 6 October, 2025, IARO scientists were contacted by Citizens Initiative Estonia [a non- -profit organization], with the request to provide an assessment of a report commissioned by the Ministry of Environment of Estonia, and produced by the University of Tartu (TU Report).

2. This IARO Critical Analysis Report is not an “oppositional document” to the TU Report. Rather, it has been prepared as a pedagogical document.

3. It is hoped that the authors of the TU Report, whom we view as fellow scientists, take this IARO Critical Analysis Report as an educational tool, contributing to their “Review Study Phase I,” rather than some gratuitous “attack document.”

4. The study documented by the TU Report has clearly been conducted properly in terms of how an analysis of published papers and reports should be undertaken, when those participating are not experts in the subject matter.

5. IARO Scientists have the distinct impression that these Estonian authors were pre- conditioned to believe that wind power plant sound emissions have no effect on public health.

6. This is further justified by the Recommendations made which are unfounded, skewed from reality, and not “evidence-based,” as promised by the authors of the TU Report.

7. Given the content of the Recommendations proffered by the TU Report, it seems probable that the Authorship of the TU Report has unwittingly succumbed to the unscientific practices promoted by governments and international special interest groups.

8. In the opinion of IARO Scientists, this study can only be regarded as, yet another, artificially constrained review of papers, with outcomes predetermined by politically generated questions, resulting in a report of low scientific standard.

Review of Tartu Univeristy Report, Estonia

Page 5 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

A. INTRODUCTION

I. Background

9. On 6 October, 2025, IARO scientists were contacted by Citizens Initiative Estonia [a non- -profit organization], and were requested to provide an assessment of a report commissioned by the Ministry of Environment of Estonia and produced by the University of Tartu (henceforth referred to as the TU Report).

10. For this purpose, IARO scientists received an English translation of the TU Report, included in this IARO Report as Annex A. References to page numbers of the TU Report correspond to those in this English version, provided in Annex A.

11. The TU Report states that it is related to Phase I of a Review Study titled: “Health effects of wind turbines: A systematic review of studies published in peer-reviewed scientific journals over the last fifteen years.” Within this context, the study of the TU Report is, more specifically, titled: Development of a methodology for interpreting the results of scientific studies on the potential health effects of wind farms and other energy production technologies in the Estonian context.

12. Figure 1 shows the Purpose of this Study as stated in the TU Report.

Review of Tartu Univeristy Report, Estonia

Page 6 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

Figure 1. Description of the Purpose of the Study of the TU Report and its Research Questions (p. 12)

II. Goal

13. To provide a scientific review of the TU Report, within the context of The Scientific Method, Evidence-based Medicine and Critical Analysis.

III. Disclaimer

a. The report provided herein has one, and only one, agenda; that of pure scientific inquiry.

b. The authors of this report are not party to anti-technology sentiments and do not harbour anti-wind-energy sentiments.

Review of Tartu Univeristy Report, Estonia

Page 7 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

c. In no way can or should this scientific review be construed as a document arguing for or against the implementation of wind power plants, or any other type of infrastructure or industrial complexes that generate acoustic pollution.

d. IARO members and authors of this report hold no financial interest in the SAM Technology.

IV. International Acoustics Research Organization, IARO

14. The International Acoustics Research Organization represents a group of scientists who, collectively, hold over 300 years of scientific experience in the field of infrasound and low frequency noise, and its effects of human health. Since 2016, IARO researchers have been recording and analysing acoustical data in and near homes located in the vicinity of onshore wind power plants, in the following countries (alphabetical): Australia, Canada, Denmark, England, France, Germany, Ireland, New Zealand, Northern Ireland, Portugal, Scotland, Slovenia, and The Netherlands. Prior to 2016, all IARO scientists were already working either in acoustics alone or in acoustics and health. All research conducted by IARO is part of the Citizen Science Initiative for Acoustic Characterization of Human Environments (CSI-ACHE).

V. Acronyms and Variables Used in IARO Reports

15. Table 1 lists the acronyms and variables used in IARO Reports.

Table 1. Acronyms and Variables that may appear in IARO Reports

dB Decibel unweighted (measure of sound pressure level) dBA Decibel A-weighted (measure of sound pressure level) dBC Decibel C-weighted (measure of sound pressure level) dBG Decibel G-weighted (measure of sound pressure level) Hz Hertz (units for measure of frequency)

ILFN Infrasound and Low Frequency Noise (≤200 Hz) IWT Industrial Wind Turbine LFN Low frequency noise (20-200 Hz) SPL Sound Pressure Level

WHO World Health Organization WPP Wind Power Plant

WTAS Wind Turbine Acoustic Signature

Review of Tartu Univeristy Report, Estonia

Page 8 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

B. ORGANIZATION OF THIS REPORT

I. Sequential approach to various aspects

16. Given the Authorship of the TU Report, the tone of this IARO Report is meant to be educational and not oppositional—IARO Scientists consider they are addressing fellow scientists.

17. The “Stated Purpose of the Study” and the “Research Questions” will be discussed first, in Section C.

18. A brief, science-based, educational approach is provided regarding ‘health effects,’ using annoyance as an example, in Section D.

19. A brief, science-based, educational approach is provided regarding the use of the A- weighting filter, and its appropriateness for measuring ‘wind turbine noise’ in Section E.

20. Section F demonstrates the fallacy of the notion ‘what you can’t hear, can’t hurt you,’ which wholly biases the TU Report.

21. Section G examines if Scientists have the luxury of accepting conclusions of meta-analyses or systematic reviews at face-value.

22. Section H discusses three topics that the Authors of the TU Report may find important for their own knowledge base.

23. Section I documents the Conclusions of this IARO Critical Analysis Report.

II. To the Authorship of the TU Report

24. With this Critical Analysis of the TU Report, in no way do IARO Scientists wish to offend or insult the authors of the TU Report, who are considered to be fellow scientists.

25. It is clear that a genuine effort has been made, within the context of systematic reviews, to adequately select published scientific papers, under the self-imposed exclusion criteria.

26. It has also become clear, however, that the Authors of the TU Report are unfamiliar with the deep complexities and intricacies of this particular subject, both in terms of acoustics and of biological sciences—This is entirely understandable, but errors (especially those arising from unfamiliarity with a particular subject) must be raised where they are made!

Review of Tartu Univeristy Report, Estonia

Page 9 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

27. As mentioned above (Parag. 14), IARO Scientists have been individually dedicated to studying the health effects caused by infrasound and low frequency noise for many decades, and from many different perspectives (biological, clinical, signal analysis, instrumentation, occupational and environmental settings, animal exposures, among others).

28. IARO Scientists hope that the authors of the TU Report view this IARO Report as an educational tool, contributing to their “Review Study Phase I,” rather than some gratuitous “attack document.”

29. Please see Section H: A Candid Conversation among Scientists.

Review of Tartu Univeristy Report, Estonia

Page 10 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

C. PURPOSE AND RESEARCH QUESTIONS UNDERLYING THE TU REPORT

I. Purpose of the TU Report Study

The aim of the study was to systematically analyse the evidence published in the scientific literature over the last fifteen years (2010–2025) on the health effects of wind turbines (p.12). (See Fig. 1)

30. While it is understood what is meant, this purpose is very badly worded, given the scientific credentials of the TU Report’s Authorship.

31. Medical Sciences classifies agents of disease into 4 categories: biological, chemical, physical and psychosocial.

32. In which category, then, would “wind turbines” be inserted, since they are allegedly producing health effects? The wind turbines do not cause health issues; the emissions from wind turbines may cause health issues.

33. IARO scientists would suggest the following re-wordings for scientific accuracy:

“…on the health effects associated with the proximity of wind turbines to residential areas,”

Or

“…on the health effects claimed to be associated with wind turbine emissions,”

Or

“The aim of the study was to systematically analyse the evidence published in the scientific literature over the last fifteen years (2010–2025) on the purported health effects due to wind power plant operations.”

34. This issue is not a trivial matter, as it may seem to some.

35. Instead, it reflects a deep misunderstanding of the matter at hand pertaining to the fundamental principles of Medical Sciences.

36. After all, the foremost concern here is the health of Estonian Citizens, is it not?

Review of Tartu Univeristy Report, Estonia

Page 11 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

II. Research Questions

37. Throughout the Research Questions, it is apparent that wind turbines are (erroneously) interpreted as an agent of disease.

38. This greatly curtails the expansion of questions into a more scientific realm. As all scientists are aware, asking the right question is of fundamental importance.1

Question 1: What are the main conclusions of existing studies on the health effects of wind turbines? (p.12, see Fig. 1)

39. This seems like an innocuous and purposeful question, but a closer inspection already reveals bias: is it presumed that the “health effects of wind turbines” are specific and exclusive to wind turbines—they are not!

40. Other industries can have similar emissions that bring about the same “health effects” as those allegedly developed by residents neighbouring wind power plants.

41. The agent of disease is not the wind turbine but its various emissions and, yes, one of those emissions is acoustical in nature.

42. Again, to the uniformed this may seem a trivial point, more related to semantics—It is not!

43. Imagine the following question:

What are the main conclusions of existing studies on the health effects of automobiles?

44. Is this a question that, taken alone, makes any sense?

Question 2: What is the overall quality of the existing evidence? Is there evidence in the scientific literature that wind turbines have a negative impact on human health? (p.12)

45. The “overall quality of existing evidence” is evaluated by reading the Methodology Section of each and every selected paper to ascertain if the conclusions reached are supported by the methodology used (see Section G).

46. Are the Authors of the TU Report qualified to evaluate whether the methodologies imposed by law to “measure noise” are fit-for-purpose when human health is a concern?

1 Back in the late 1800’s, the question was posed: “Is Light a particle OR a wave.” This question reduced physical

reality to a dichotomy, not open to the possibility that Light can be BOTH. Hence the fundamental importance for proper Scientists to ask the pertinent and insightful questions.

Review of Tartu Univeristy Report, Estonia

Page 12 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

47. Or have the Authors of the TU Report, instead, blindly relied upon the “noise measuring” methodologies as per legislated stipulations?

48. Many authors, unfamiliar with the matter at hand, do rely upon legislated methodologies.

49. However, given the stated “Purpose of the Study” and the scientific background of the authorship, can the Authors of the TU Report be afforded this luxury? (see Section E)

50. On the other hand, when the selected paper is referring to the evaluation of health endpoints, do the Authors of the TU Report have the expertise in Medical and Clinical Sciences to evaluate whether or not the selected health endpoint is pertinent and relevant? (See Section D)

51. The same can be pointed out regarding the second part of Question 2, “Is there evidence in the scientific literature that wind turbines have a negative impact on human health?” Whether there is or not, is the Authorship of the TU Report qualified to critically analyse the methods applied in these studies? (see Sections D and E)

Question 3: If wind turbines have negative health effects, what health effects are associated with wind turbines? (p.12)

52. This question trickles down from the prior questions. Again, it is not “wind turbines” that have negative health effects (unless the wind turbines themselves are becoming sick), but emissions from wind turbines that can act as agents of disease upon biological organisms.

53. Nevertheless, it is understood that the object of this question is to determine what kind of health effects have been documented as related to living in the proximity of wind power plants.

54. Do the Authors of the TU Report have the necessary expertise to evaluate the robustness of the methodologies used in papers that report health endpoints as related to residential proximity to wind power plants? (See Sections D, G and H-I)

Question 4: If wind turbines have negative health effects, what role do environmental factors such as noise, infrasound, shadow flicker, visual aspects, psychological factors (including general attitudes towards wind turbines, people's beliefs and perceptions of wind turbines), vibration and electromagnetic fields in causing these health effects? (p.12)

55. Let us dissect this question: “[W]hat role do environmental factors such as

noise, infrasound—a potential acoustical physical agent of disease,

shadow flicker—a potential optical physical agent of disease,

visual aspects—a potential optical physical and/or psychosocial agent of disease,

Review of Tartu Univeristy Report, Estonia

Page 13 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

psychological factors—a potential psychosocial agent of disease,

vibration—a potential vibratory physical agent of disease and

electromagnetic fields—a potential electromagnetic physical agent of disease.

56. Perhaps, laid out like this, the Authors of the TU Report might realize why this question is entirely inappropriate…unless it is broken up into 6 distinct questions, each warranting its own independent study and (very) complex evaluation.

57. For example, “shadow flicker” is a term that only appeared after the advent of wind energy—before, it was called the stroboscopic effect.

58. Therefore, as the Authors of the TU Report would certainly agree, a proper investigation into “shadow flicker” must include prior studies (at least a glimpse into them) on the stroboscopic effects on humans (for example, such as those related to military helicopter pilots). Similar prior studies would be needed for each of the other environmental factors.

Question 5: If wind turbines have health effects, under what conditions are these health effects more likely to occur (e.g. at what distance from the turbine, with powerful or tall turbines, etc.)? (p.12)

59. Again, the wording of this question does not do justice to the scientific credentials of the TU Report’s Authors. While it is understood what is being asked here, its formulation is most unscientific.

60. Suggested rewording of Question 5:

Question 5 (suggested rewording):

If it can be demonstrated that “health effects” develop in residents neighbouring wind power plants, what external physical conditions (e.g. distance to turbine(s), type and specifications of turbine(s), etc.) become significant factors for the onset and/or development of these “health effects”?

61. It is hoped that this rewording is self-explanatory.

Question 6: Are certain population groups more vulnerable to the potential health effects of wind turbines? (p.12)

62. This is a very interesting question to have at such an initial stage of the study.

63. Is it intended to point out that population groups known to be vulnerable, such as the elderly, the chronically ill, infants and children, and pregnant women, should be

Review of Tartu Univeristy Report, Estonia

Page 14 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

approached as separate populations to determine if they are (also) more vulnerable to “health effects of wind turbines,” if they exist?2

64. Given the vast experience of IARO Scientists, it seems that this question is most likely based on the prior supposition that some people ‘are more sensitive’ than others to the “health effects of wind turbines.” (See Section H-II).

Question 7: What evidence-based recommendations can be made to policymakers, industry stakeholders and affected communities to protect human health? (p.12)

65. This final question is, in and of itself, quite unscientific. “Evidence-based recommendations”—Are there any other type?

66. And yet, having read the Recommendations of the TU Report (and having pointed out their failings), it is now realized that, indeed, non-evidence-based Recommendations are, regrettably still (see Section H-III) a possibility from authors with significant Scientific Credentials.

2 If the TU Report were to include animal studies, then perhaps this question could refer to different types of animal

populations. Cows, sheep, rabbits and mink all react very differently when in the vicinity of wind power plants. Perhaps some are more vulnerable?

Review of Tartu Univeristy Report, Estonia

Page 15 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

D. WHAT IS A NOISE-INDUCED HEALTH EFFECT?

67. As stated by the World Health Organization:

An adverse effect of noise is defined as a change in the morphology and physiology of an organism that results in impairment of functional capacity, or an impairment of capacity to compensate for additional stress, or increases the susceptibility of an organism to the harmful effects of other environmental influences. 3

68. ‘Annoyance’ is commonly (yet erroneously) considered as a “health effect.”

69. Dutifully, the TU Report covers this subject. Here are some examples:

Several of the studies included in this review (Appendix 2, Table 4) investigated the extent to which one specific characteristic of wind turbine noise, amplitude modulation (AM), contributes to annoyance (Ioannidou et al., 2016; Lee et al., 2011; Schaffer et al., 2016, 2018). In addition, these studies also examined the effect of noise frequency distribution and source origin on annoyance. (p. 21)

A review article (McCunney et al., 2014) also concluded that wind turbine noise plays only a minor role in causing annoyance compared to other factors that influence people's willingness to experience annoyance in relation to wind turbines. Pohl et al. (2018) also found that noise-related annoyance was influenced to a small extent by the distance to the nearest wind turbine and the intensity of the sound, but was most influenced by the extent to which people felt that the wind turbine planning process had been conducted fairly and transparently. (p. 33)

In summary, the relationship between wind turbines and disturbance depends on several factors, such as expectations/knowledge of the health effects of wind turbines, perceived fairness and transparency of the planning process, economic benefits, visual aspects and noise. It is likely that a combination of all these factors causes annoyance, and reducing just one factor (e.g. noise) may not reduce annoyance. (p. 36)

70. Annoyance is also included in the Recommendations Section of the TU Report:

3 World Health Organization. (1999) Guidelines for community noise. Stockholm University & Karolinska Institute:

Stockholm, Sweden. pp. 21. https://www.who.int/publications/i/item/a68672

Review of Tartu Univeristy Report, Estonia

Page 16 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

We recommend that developers and researchers explore ways to reduce AM depth in order to reduce the annoyance of wind turbine noise (p.46). [AM = Amplitude Modulation]

71. Are the Authors of the TU Report acquainted with the formal definition of annoyance?

72. In the 2017 edition of Mosby’s Medical Dictionary,4 there were zero entries for the word ‘annoyance.’

73. In the 2018 edition of the Medical Dictionary published by the British Medical Association,5 there were also no instances of the word ‘annoyance.’

74. In the 2020 edition of the Oxford Medical Dictionary,6 one single entry is found for this word:

Glare n. the undesirable effects of scattered stray light on the retina, causing reduced contrast and visual performance as well as annoyance and discomfort.

75. Within the context of noise nuisance, perhaps the best definition for ‘annoyance’ is (still) the one given in 2000 by the European Commission Noise Team:

Annoyance is the scientific expression for the non-specific disturbance by noise, as reported in a structured field survey. Nearly every person that reports to be annoyed by noise in and around its home will also experience one or more of the following specific effects: Reduced enjoyment of balcony or garden; When inside the home with windows open: interference with sleep, communication, reading, watching television, listening to music and radio; Closing of bedroom windows in order to avoid sleep disturbance. Some of the persons that are annoyed by noise also experience one or more of the following effects: Sleep disturbance when windows and doors are closed; Interference with communication and other indoor activities when windows and doors are closed; Mental health effects; Noise-induced hearing impairment; Hypertension; Ischemic heart disease.7

4 O’Toole MT et al. (Eds). (2017) Mosby’s Medical Dictionary. 10th Ed. Elsevier: St Louis, MI, USA.

5 British Medical Association. (2018) Medical Dictionary. 4th Edition. Dorling Kindersley: London, UK.

6 Martin E, Law J. (Eds) (2020) Concise Colour Medical Dictionary. 7th Ed. Oxford University Press: Oxford, UK.

7 European Commission. (2000) The Noise Policy of the European Union—Year 2. Towards improving the urban environment and contributing to global sustainability. European Commission Noise Team: Luxembourg. https://www.europeansources.info/record/the-noise-policy-of-the-european-union-year-2-1999-2000-towards- improving-the-urban-environment-and-contributing-to-global-sustainability/

Review of Tartu Univeristy Report, Estonia

Page 17 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

76. This comprehensive definition of ‘annoyance’ clearly establishes it as a legitimate measure to be used within the realm of Psychoacoustic studies.

77. But it is far from being an appropriate health endpoint within the context of an “adverse effect of noise,” as defined by the World Health Organization (see Parag. 67).

78. Do the Authors of the TU Report have the necessary expertise to identify this issue, or will it be ‘business as usual’?8

79. For the edification of these Authors, in papers that have been excluded from their selection, annoyance has been linked to morphological changes in the auditory cilia and some medical professionals view self-reported ‘noise annoyance’ in their patients as a symptom of excessive prior noise exposure. (See Section H-II)

80. For the further edification TU Report’s Authors: The International Classification of Diseases (ICD-11), published by the World Health Organization, has specific codes for infrasound- induced vertigo—NF08.2Y (see Figure 2).

8 The exclusion criteria should have included all papers that have used ‘annoyance’ as a bona fide health endpoint.

Review of Tartu Univeristy Report, Estonia

Page 18 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

B

Figure 2. Results of the search for “infrasound” in the WHO ICD-11 (coding tool option). 9 (A) One instance of infrasound appears—Code NF08.2Y, covering infrasound-induced vertigo, under the heading of “other specified effects of vibration.” (B) Index terms covered by this Code differentiate between infrasound- and vibration-induced vertigo.10

9 World Health Organization. (2024) International Classification of Diseases-11 (ICD-11).

https://icd.who.int/ct/icd11_mms/en/release

10 World Health Organization. (2024) International Classification of Diseases-11 (ICD-11). https://icd.who.int/browse/2024-01/mms/en#621374492%2Fother

Review of Tartu Univeristy Report, Estonia

Page 19 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

E. WHY CURRENT NOISE MEASURING METHODOLOGIES ARE NON-APPLICABLE FOR WIND

TURBINE NOISE

81. In its Recommendations section, the TU Report states the following:

For living and sleeping areas, we recommend setting a limit for wind turbine noise of 30 dB(A) during the day and 25 dB(A) at night, similar to the existing limits for traffic noise and noise from technical equipment. (p. 48).

82. Presumably, then, a value of 28 dBA would, more or less, comply with this recommendation.

83. Which value of 28 dBA would the Authors of the TU Report consider acceptable in the following field-data situation, shown in Fig. 3:

84. The 28 dBA in Fig. 3A or the 28 dBA in Fig. 3B?

A

B

Figure 3. A: 28 dBA (red bars) and 89 dB (pink bars). B: 28 dBA (red bars) and 47 dB (pink bars).11

11 Data from urban field measurements (no wind turbines), published in a paper that was excluded from the TU Report

selection of papers. Pereira-Sousa P, Alves-Pereira M, Bakker H. (2025). Dose-Response Relationship in Occupational Noise Exposures: The Distorted Quantification of Dose that Misinforms the Medical Community. SHO 2025 – International Symposium on Occupational Safety and Hygiene. Proceedings Book. DOI: https://doi.org/10.24840/978-989-54863-7-3_0125-0132

Review of Tartu Univeristy Report, Estonia

Page 20 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

85. Do the Authors of the TU Report understand that noise measured in dBA cannot differentiate between these two, significantly different, acoustic environments?

86. Hence, the recommendation transcribed in Para. 81 is entirely skewed from the matter at hand.

87. This type of information was known by the World Health Organization in 1999:

A noise measure based only on energy summation and expressed as the conventional equivalent measure, LAeq, is not enough to characterize most noise environments. It is equally important to measure the maximum values of noise fluctuations, preferably combined with a measure of the number of noise events. If the noise includes a large proportion of low-frequency components, still lower values than the guideline values below will be needed. When prominent low-frequency components are present, noise measures based on A-weighting are inappropriate.12 [Emphasis added.]

88. In the TU Report, Fig. 1 is a very informative graph showing the frequency response curves of the different frequency-weighting filters that are imposed on noise measurements by legislated stipulations. This graph is reproduced here in Fig. 4.

Figure 4. Frequency response curves for A, C and G the frequency-weighting filters and for the absence of filter, Z. (p. 8, TU Report)

12 World Health Organization. (1999) Guidelines for community noise. Stockholm University & Karolinska Institute:

Stockholm, Sweden. pp. xiii. https://www.who.int/publications/i/item/a68672.

Review of Tartu Univeristy Report, Estonia

Page 21 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

89. Do the Authors of the TU Report realize that the application of any of these different filters (A, C and G) means that profound assumptions are being made, namely:

1) what you can’t hear can’t hurt you (see Section F), and

2) annoyance is a bona fide health endpoint (see Section D)?

3) noise only affects humans via the auditory pathway (see Section F).

90. Do the Authors of the TU Report realize that, for the purposes of the matter at hand, the act of “measuring noise” constitutes the quantification of the dose of the agent of disease?

91. Do the Authors of the TU Report understand that the Y axis of their Figure 1 indicates that the application of frequency weighting filters means that the measurements no longer reflect physical reality?

92. Does this begin to explain why legislated methodologies are scientifically irrelevant for measuring the types of environments where noise has significant lower frequency components, such as those generated by wind power plants?

93. Does this also suggest why a high-quality scientific investigation should ignore legislated methodologies in favour of evidence-based methodologies?

94. Real, scientific-grade information on the medical dose of noise is not obtained, if legislated procedures are applied, i.e., the mandatory use of A, C or G frequency- weighting filters. (See Fig. 3)

Review of Tartu Univeristy Report, Estonia

Page 22 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

F. ‘WHAT YOU CAN’T HEAR, CAN’T HURT YOU’

95. Is this what the Authors of the TU Report have been told? What you can’t hear, can’t hurt you?

96. The significant difference between the two 28-dBA environments shown in Fig. 3 will be summarily dismissed by those who believe this fallacy.

97. The Authors of the TU Report will be told that the real physical presence of the 47 and 89 dB difference (i.e., no filter is applied) is irrelevant for human health because it is occurring below the human auditory threshold.

98. Will these Authors, then, also believe that only environmental factors that can be readily perceived by all people are relevant for consideration in human health? In the same way that radioactivity is (not) readily perceived or carcinogenic chemicals are (not) readily perceived?

99. See Fig. 5, which shows an abstract of a paper from 1978 (!)13

Figure 5. Busnel RG, Lehmann AG (1978). Infrasound and sound: Differentiation of their psychophysiological effects through use of genetically deaf animals. Journal of the Acoustical Society of America14 (see text).

13 Certainly, way beyond the scope of the TU Report’s systematic review…being from 1978 and because it involves

animals.

14 Busnel RG, Lehmann AG (1978). Infrasound and sound: Differentiation of their psychophysiological effects through use of genetically deaf animals. Journal of the Acoustical Society of America, 63(3): 974-977. https://pubmed.ncbi.nlm.nih.gov/670562/

Review of Tartu Univeristy Report, Estonia

Page 23 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

100. In this 1978 study, genetically-deaf mice were used as study subjects, and infrasound had a deleterious effect on their performance. What you can’t hear can’t hurt you?

101. How, then, to explain the more recent scientific results shown in Figs. 6A and 6B, where an acoustic phenomenon, presumed to be inaudible to humans (below 20 Hz) was able to distress the residents during a sleepless night, to the point of compelling them to take medication?15

A

B

Figure 6.

A: Residents near wind power plants slept peacefully— 26 dBA and 67.3 dB, B: Same residents could not sleep and needed medication—26.5 dBA and 69.9 dB16

102. What you can’t hear, can’t hurt you…doesn’t really work very well, does it? Not for mice in 1978, nor for humans in 2023.

103. For the edification of the Authors of the TU Report, the sequence of peaks seen in Fig. 6B is called a wind turbine acoustic signature. Mathematically, it is a harmonic series whose fundamental frequency corresponds to the blade pass frequency of the corresponding wind turbine (see Fig. 7B in Section G).

104. Wind turbine acoustic signatures become invisible when legislated noise measuring methodologies are applied.

15 The residents, authors of the diary providing this information, were not privy to any acoustical information that was

being simultaneously recorded. Data presented here in Figure 6 are the result of post-processing analysis. Please see Footnote 16 for the full, peer-reviewed report on this case.

16 This paper was excluded from the selection of papers considered by the TU Report, as it is a Case Report. Bakker HHC, Alves-Pereira M, Mann R, Summers R, Dickinson P. (2023) Infrasound exposure: High resolution measurements near wind power plants. In: Suhanek M, Kevin Summers J. (Eds) Management of Noise Pollution. IntechOpen: London. DOI: 10.5772/intechopen.109047

Review of Tartu Univeristy Report, Estonia

Page 24 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

G. LUXURIES NOT AFFORDED TO SCIENTISTS—A GLIMPSE OF THE TEDIOUS WORK REQUIRED TO

UPHOLD SCIENTIFIC RIGOUR.

105. Scientists do not have the luxury of taking the conclusions of meta-analyses (or systematic reviews or literature reviews) of pre-existing papers, for granted or at face-value.

106. In contrast to laypersons, policy- and decision-makers, industry stakeholders and the general public, Scientists must evaluate the methodology of each and every paper included in a review.

107. A tedious exercise for sure, but a necessary one if scientific rigour is to be upheld.

108. How else can one scientifically vouch for the conclusions offered by the author of the meta-analysis, systematic review or literature review?

109. As a demonstrative exercise, let us explore an 11-year-old paper, quoted several times in the TU Report:

Basner, M., Babisch, W., Davis, A., Brink, M., Clark, C., Janssen, S., Stansfeld, S., 2014. Auditory and non-auditory effects of noise on health. Lancet 383, 1325– 1332.

110. This same reference justified the following statements, made by the Authors of the TU Report on page 7:

A decibel indicates how much louder the sound is than the reference value. In air, the reference value is an air pressure of 20 micropascals (20 μPa or 2×10−5 Pa), which is considered to be the human hearing threshold at a frequency of 1000 Hz – this is the quietest sound that the average person can still hear at this frequency. (p.7)

111. And on page 32:

Disturbance can also act as a mediating factor between other health effects, including influencing the development of more serious conditions such as cardiovascular disease through stress (p. 32)

112. And, under the heading “Audible noise [sic] generated by wind turbines and clinically manifested health effects,” (p. 40), the TU Report makes another statement justified by this same, 2014 reference:

Review of Tartu Univeristy Report, Estonia

Page 25 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

Disturbance and sleep disturbances caused by audible noise may contribute to the development of diagnosable diseases (p. 40). 17

113. Returning to the original 2014 reference, it states:

In this Review, we summarise knowledge and research related to noise exposure and both auditory and non-auditory health effects. (…)

These noise exposures have been linked to a range of non-auditory health effects including annoyance (Miedema & Oudshoorn, 200118), sleep disturbance (Muzet, 200719), cardiovascular disease (van Kempen & Babisch, 201220; Sorensen et al., 2012 21 ) and impairment of cognitive performance in children (Stansfeld & Matheson, 2003 22 ). 23 [The original numbered references were replaced with citations.]

114. Scientific ‘work’ involves reading each of these 5 references that are quoted in this 2014 review if and only if scientific rigour is to be maintained.

115. (It should be recalled that scientific rigour is not necessarily in the purview of laypersons, policy- and decision-makers, industry stakeholders and the general public.)

116. Just by reading the titles of these 5 papers we see that one is a meta-analysis, which eliminates it from this immediate consideration.

117. Let us look into the other four.

17 This last assertion is a truism (at least since the times of Ancient Rome) as it is referring to audible noise! Strictly

speaking, no reference would have been needed.

18 Miedema HME, Oudshoorn CGM. Annoyance from transportation noise: relationships with exposure metrics DNL and DENL and their confidence intervals. Environ Health Perspect. 2001; 109:409–16. https://pubmed.ncbi.nlm.nih.gov/11335190/.

19 Muzet A. Environmental noise, sleep and health. Sleep Med Rev. 2007; 11:135–42. https://pubmed.ncbi.nlm.nih.gov/17317241/

20 van Kempen E, Babisch W. The quantitative relationship between road traffic noise and hypertension: a meta- analysis. J Hypertens. 2012; 30:1075–86. https://pubmed.ncbi.nlm.nih.gov/22473017/

21 Sørensen M, Andersen ZJ, Nordsborg RB, et al. Road traffic noise and incident myocardial infarction: a prospective cohort study. PLoS One. 2012; 7:e39283. https://pubmed.ncbi.nlm.nih.gov/22745727/

22 Stansfeld SA, Matheson MP. Noise pollution: non-auditory effects on health. Br Med Bull. 2003; 68:243–57. https://pubmed.ncbi.nlm.nih.gov/14757721/

23 Basner M, Babisch W, Davis A, Brink M, Clark C, Janssen S, Stansfeld S. (2014) Auditory and non-auditory effects of noise on health. Lancet 383: 1325–1332. https://pubmed.ncbi.nlm.nih.gov/24183105/

Review of Tartu Univeristy Report, Estonia

Page 26 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

118. Miadema & Oudshoorn (2001): “Here we model the distribution of annoyance responses as a function of the noise exposure” for road, rail and air traffic noise. “Day–night level (DNL) and day–evening–night level (DENL) were used as noise descriptors.”

119. Wind turbine noise is not considered in the Miadema & Oudshoorn paper.

120. Annoyance, which Miadema & Oudshoorn used as a health endpoint, is not a bona fide health outcome (see Section D).

121. The noise parameters used to characterize “noise exposure” are inconsequential for the matter at hand (see Section E).24

122. Muzet (2007) is classified as a “Clinical Review” and uses sleep as a measure of a health— a bona fide health endpoint.

123. However, noise environments of the papers used by Muzet in his Clinical Review are still characterized in dBA, and wind turbine noise is not considered—this excludes any real scientific relevance to the matter at hand.

124. Sorensen et al. (2012), not a review paper, and a very scientifically robust health endpoint was chosen—ischemic heart disease (see Section D).

125. Sorensen et al. (2012) stated: ”Exposure to long-term residential road traffic noise was associated with a higher risk for MI, in a dose-dependent manner.” [MI=Myocardial Infarction, i.e., ischemic heart disease.]

126. As with the Miadema & Oudshoorn study, the ‘day–evening–night level,’ or Lden, was used to quantify the noise environment (see Section E and footnote 24), and wind turbine noise was not considered.

127. Stansfeld & Matheson (2003), yet another review, based on 86 references…

128. The tediousness of this exercise is an integral part of the Scientific process.

129. To finalize this Section, a last assertion is transcribed from the Introduction of the TU Report:

However, reviews of wind turbine noise conducted to date have not confirmed a link between wind turbine noise and clinically apparent health effects (Karasmanaki, 2022; Schmidt and Klokker, 2014; Teneler and Hassoy, 2023; van Kamp and van den Berg, 2021, 2018). (p. 4)

24 Although not explicitly indicated, the use of the Lden or DENL noise parameter implies the application of the A-

frequency-weighting filter (see Section E, Figs. 3 and 4).

Review of Tartu Univeristy Report, Estonia

Page 27 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

130. These five references, offered as (evidence-based?) justifications for this assertion, are all meta-analysis, systematic reviews or literature reviews.

131. Does this mean that the Authors of the TU Report vouch for the position taken by the author(s) of each one of these five reviews, and therefore guarantees that all the papers cited in reviews uphold the author(s)’ position? Can they vouch for the methodologies of all the papers when they have not seen them?

132. Of course not!

133. However, unlike others, Scientists, do not have the luxury of merely depending on the conclusions reached by the authors of these types of review papers, because they could include papers of dubious scientific integrity.

134. This is a part of what the scientific process is all about, is it not?

135. On page 26 of the TU Report, the following is stated:

The study concluded that wind turbine infrasound does not disturb people's sleep, does not cause symptoms of 'wind turbine syndrome', does not impair measured cardiovascular health indicators, and does not impair people's mental well-being (Marshall et al., 2023). The results of the study can be considered well- proven. (p.26)

136. Annex B provides a critical analysis of the Marshall et al. (2023) paper prepared by IARO Scientists in 2024.

137. On page 33 of the TU Report, the following is stated:

An experiment conducted in Finland showed that the audible sounds of a wind farm were more disturbing than the sounds of the ocean (Maijala et al., 2021).

138. Would the Authors of the TU Report care to know the scientific reason for why this is so?

139. Here is a comparison between ocean noise and wind turbine noise, as measured without the methodologies imposed by legislation:

Review of Tartu Univeristy Report, Estonia

Page 28 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

A

B

Figure 7. Characterization of acoustic environments (i.e., noise measurements) without the legislated-imposed methodologies. A. Beach, Rømo Island, Denmark, 13 Dec 2016 at 01:10H. B. Wind turbines acoustic signature, present in the acoustic environment corresponding to the night when residents could not sleep and were compelled to take medication—See Fig 3B in Section F.

140. All this important information becomes invisible when legislated methodologies are imposed.

141. Annex C provides a critical analysis of the Maijala et al. paper prepared by IARO Scientists in 2024.

Review of Tartu Univeristy Report, Estonia

Page 29 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

H. A CANDID CONVERSATION AMONG SCIENTISTS

142. Since the Authors of the TU Report are considered by IARO as fellow-scientists, an uncommon decision has been taken to speak directly to these Authors through this Critical Analysis Report.

143. This was deemed all the more appropriate since IARO Scientists have been informed that this Team of Estonian Scientists will proceed with more studies to monitor the development health effects among residents neighbouring wind power plants.

144. If this TU Report is any indication of the avenues of research that will be followed (particularly given its appalling Recommendations), then IARO’s position is simple—what a waste of time, money and brainpower!

I. The ‘nocebo effect’ narrative

145. As part of the Recommendations, the TU Report states:

The results of our study show that several factors other than wind turbine noise affect disturbance, and that noise reduction alone may not be sufficient to mitigate disturbance. Just as important as noise restrictions in preventing disturbance may be informing residents about the nocebo effect, the absence of negative expectations regarding the health effects of wind turbines, and understanding the positive characteristics of wind turbines (Crichton et al., 2015, 2014b, 2014a; Crichton and Petrie, 2015b, 2015a; Tonin et al., 2016).

146. By advocating this ‘nocebo effect narrative,’ the Authorship of the TU Report is taking a position that is absolutely indefensible in terms of Science.

147. The Authors of the TU Report, as Scientists, should be aware that a nocebo effect cannot be proven, as it is impossible to eliminate all environmental factors that might be a cause but are unmeasured.25 The Authors should be asking, where is the evidence for a nocebo effect?

148. Given the scientific credentials of the Authors of the TU Report, they should, instead, be inquiring into the studies that justified attributing the label of ‘nocebo effect’ to the collection of symptoms, self-reported people by all over the world.

25 The nocebo effect can never be proved, it can only fail to be disproved.

Review of Tartu Univeristy Report, Estonia

Page 30 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

149. This collection of symptoms is not specific to people who live in the proximity of wind power plants, but it is specific to people who live in infrasound-contaminated homes (whatever the source).

150. Are all these cases supposed to be the result of some collective psychosomatic disorder?

151. For the edification of these Estonian Scientists, the ‘nocebo effect’ is, in Clinical Medicine, considered to be of psychosomatic origin (or aetiology), falling under the category of pathology caused by psychosocial agents of disease (see Parag. 31).

152. Under the rules of Evidence-based Medicine, to claim that a collection of symptoms is a ‘nocebo effect,’ then, objective medical examinations must have been prescribed and no organic aetiology for the symptoms was found.

153. Has the TU Report’s Authors found a scientific justification for labelling this collection of symptoms as a ‘nocebo effect’?

154. Moreover, if, as Scientists, these Authors truly insist on standing by the ‘nocebo effect narrative,’ then they must be prepared to explain all the effects seen in animals living in proximity to wind power plants, such as:

Exposed cows in France registered a dramatic fall in milk output.26

Exposed cows in Korea are reported to have many cases of foetal death.27

In Poland, there was a negative effect on the stress parameters and productivity of exposed geese.28

In England, higher cortisol levels were found in exposed badgers and “these high levels may affect badgers’ immune systems, which could result in increased risk of infection and disease in the badger population.”29

26 Mulholland R. (2015) French farmer sues energy giant after wind turbines ‘make cows sick.’ The Telegraph, 18

September. https://www.telegraph.co.uk/news/worldnews/europe/france/11875989/French-farmer-sues- energy-giant-after-wind-turbines-make-cows-sick.html.

27 Se-hwan B. (2018). Wind turbines destroy local farming village. Rapid expansion of wind power facilities raises health and environmental concerns. The Korea Herald, 20 March. http://www.koreaherald.com/view.php?ud=20180320000768

28 Mikolajczak J, Borowski S, Marc-Pienlowska J, Odrowaz-Sypniewska G, Bernacki Z, Siodmiak J, Szterk P. (2013) Preliminary studies on the reaction of growing geese (Anser anser f. domestica) to the proximity of wind turbines. Polish Journal of Veterinary Sciences, 16(4):679-86. DOI: 10.2478/pjvs-2013-0096

29 Agnew RCN, Smith VJ, Fowkes RC. (2016) Wind turbines cause chronic stress in badgers (meles meles) in Great Britain. Journal of Wildlife Diseases, 52(3): 459-67. DOI: 10.7589/2015-09-231

Review of Tartu Univeristy Report, Estonia

Page 31 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

In a Polish study, the meat quality of exposed pigs decreased significantly.30

In Spain, a rabbit farm saw a 50% decrease in production after the WPP was installed. Exposed rabbits developed “problems of stress, infertility, death and behavioural changes,”31 and “a disproportionate increase in mortality rates.” This farm has since been closed down.32

Exposed frogs in Japan, “collected from paddy fields with wind power generators exhibited a faster call rate, higher salivary concentrations of corticosterone, and lower innate immunity (…) [This] can alter the disease epidemiology of local populations by regulating the balance between reproduction and immunity.”33

Exposed horses in Portugal developed flexural deformities and blood vessel walls revealed the characteristic, collagen-based thickening.34, 35

In Denmark, a mink farm was forced to close due to greatly increased aggressiveness, stillbirths and birth defects.36

30 Karwowska M, Milolajczak J, Dolatowski ZJ, Borowski S. (2015) The effect of varying distances from wind turbine

on meat quality of growing-finishing pigs. Annals of Animal Sciences, 15(4):1043-54. DOI: 10.1515/aoas-2015- 0051.

31 Ephe. (2023) [Union of Peasants of Castile and Leon denounces the ruin of one of the best farms in Spain by a wind farm] El Diário.es, 9 May. (In Spanish) https://www.eldiario.es/castilla-y-leon/union-campesinos-castilla-leon- denuncia-ruina-mejores-granjas-espana-parque-eolico_1_10189253.html

32 Fernández, JI. (2023) [A rabbit farm's fight against a wind farm: “We are in ruins”]. El Español, 9 May. [Article in Spanish] https://www.elespanol.com/castilla-y-leon/economia/el-campo/20230509/lucha-granja-conejos- parque-eolico-ruina/762423940_0.html

33 Park JK, Do Y. (2022) Wind turbine noise behaviourally and physiologically changes male frogs. Biology, 11, 516. DOI: 10.3390/biology11040516

34 Castelo Branco NAA, Costa e Curto T, Mendes Jorge L, Cavaco Faísca J, Amaral Dias L, Oliveira P, Martins dos Santos J, Alves-Pereira M. (2010) Family with wind turbines in close proximity to home: follow-up of the case presented in 2007. Proceedings of the 14th International Meeting on Low Frequency Noise, Vibration and Its Control. Aalborg, Denmark, 9-11 June, 31-40. https://www.researchgate.net/publication/290444702_Family_with_wind_turbines_in_close_proximity_to_home _follow-up_of_the_case_presented_in_2007

35 Costa e Curto TM. (2012) [Acquired flexural deformity of the distal interphalangic articulation in foals]. Master’s Thesis. Faculty of Veterinary Medicine, Technical University of Lisbon. [Thesis in Portuguese] https://www.repository.utl.pt/handle/10400.5/4847

36 Rapley, B. (2018). Conversation for a Small Planet Vol. 3-Biological Consequences of Low-Frequency Sound. Bouncing Koala Press, Palmerston North, New Zealand. (Chapter 8-Death in Denmark) Available from [email protected].

Review of Tartu Univeristy Report, Estonia

Page 32 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

155. Is it the position of the Authors of the TU Report that the adverse health effects observed in these animals, living in the proximity of wind power plants, are also caused by a ‘nocebo effect’ (i.e., psychosocial factors)?

156. Or is this one of the reasons why animal studies were excluded from the selection of papers chosen for this systematic review?

II. The questionnaire approach

157. In order to construct an appropriate questionnaire for people who live in proximity to wind power plants, yet another concept must be understood, regarding Medical Sciences and Physical Agents of Disease:

The health effects of physical agents of disease are cumulative.

158. This means the overall, prior noise exposure time (whatever the source!) is a parameter that must be considered, if a bona fide study based on questionnaires is desired.

159. This is true for vibration exposures, electromagnetic radiation exposures (where personal dosimeters are applied to actually quantify the cumulative exposure) and for noise exposures.

160. Stratification of study and control populations, as per prior noise exposures (severe, moderate and mild) must be made before any statistically valid study of health effects developed by citizens living in proximity to wind power plants can be properly obtained.

161. ‘Increased sensitivity’ can, therefore, merely be synonymous with significant, prior noise exposure, such as foetal exposures and/or prior occupational or residential exposures.

162. This particular topic has been extensively discussed elsewhere.37

37 The exclusion criteria applied by the Authorship of the TU report eliminated this paper from consideration. Alves-

Pereira M, Rapley B, Bakker H, Summers R. (2019) Acoustics and Biological Structures. In: Abiddine Fellah ZE, Ogam E. (Eds) Acoustics of Materials. IntechOpen: London. DOI: 10.5772/intechopen.82761.

Review of Tartu Univeristy Report, Estonia

Page 33 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

III. Another ‘Scientific Authorship’ of another “Wind Turbine Health Impact Study”…

163. In 2012 (13 years ago!), the Massachusetts Department of Environmental Protection and the Massachusetts Department of Public Health commissioned an Expert Independent Panel to conduct a “Wind Turbine Health Impact Study.”

164. IARO Scientists invite the Estonian Authors of the TU Report to read the Charge given to this Scientific Panel, shown in Figure 8.

Figure 8. Charge given to the Expert Independent Panel by the Massachusetts Department of Environmental Protection and the Massachusetts Department of Public Health38

38 Expert Independent Panel. (2012) Wind turbine health impact study. Massachusetts Department of Environmental

Protection and the Massachusetts Department of Public Health. https://www.mass.gov/files/documents/2016/08/th/turbine-impact-study.pdf

Review of Tartu Univeristy Report, Estonia

Page 34 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

165. How much does it differ from the Charge given to the Authors of the TU Report by the Ministry of Environment of Estonia?

166. In Annex D, please find the full Response from one of IARO’s Scientists to this 2012 Expert Independent Panel.

167. For the benefit of the Authors of the TU Report, excerpts taken from this Response are offered to our fellow Estonian Scientists in Fig. 9:

(p.4, Annex D)

(p.9, Annex D)

Figure 9. Excerpts from the Response to the Massachusetts Independent Expert Panel (Full Response Report is provided in Annex D).

Review of Tartu Univeristy Report, Estonia

Page 35 of 35

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 iaro.org.nz

I. CONCLUSIONS

168. The systematic review documented in the TU Report has clearly been conducted properly in terms of how an analysis of published papers and reports should be undertaken when those participating are not experts in the subject matter.

169. The exclusion criteria applied to the selection of scientific papers for the study, blinds the Authors of the TU Report to a broader understanding of the matter at hand.

170. Given the non-evidence-based Recommendations proffered by the TU Report, it seems probable that the Authors of the TU Report have unwittingly succumbed to the unscientific practices promoted by governments and international special interest groups.

171. In the opinion of IARO Scientists, this study can only be regarded as, yet another, artificially constrained review of papers, with outcomes predetermined by politically generated questions, resulting in a report of low scientific standard.

REVIEW STUDY PHASE I REPORT

HEALTH EFFECTS OF WIND TURBINES: A SYSTEMATIC REVIEW OF STUDIES PUBLISHED IN PEER-REVIEWED SCIENTIFIC JOURNALS OVER THE LAST FIFTEEN

YEARS

Title of the study:

Development of a methodology for interpreting the results of scientific studies on the potential health effects of wind

farms and other energy production technologies in the Estonian context

Tartu 2025

2

Research team:

Triin Veber (MSc, MPH), University of Tartu, project manager and expert

Ene Indermitte (PhD, MPH), University of Tartu, expert

Kaja-Triin Laisaar (MD, MPH, PhD), University of Tartu, expert

Urmeli Katus (RN, MSc), University of Tartu, senior methodologist

Ele Kiisk (MA, MSc), University of Tartu, junior methodologist

Hans Orru (PhD, MPH), University of Tartu, research director

Commissioned by: Ministry of the Environment

3

Contents Introduction and scientific background of the study.........................................................................................................4

Noise and infrasound terms and limit values .................................................................................................................6

Aim of the study ................................................................................................................................................................12

Methods ..........................................................................................................................................................................13

Results .............................................................................................................................................................................16

The impact of audible noise from wind turbines on human health ............................................................................17

Audible noise from wind turbines and sleep ...........................................................................................................17

Audible noise and disturbance caused by wind turbines ........................................................................................21

Audible noise from wind turbines and clinically significant health effects .............................................................22

The impact of infrasound from wind turbines on human health ................................................................................25

The impact of visual aspects related to wind turbines on human health....................................................................29

Discussion of results ........................................................................................................................................................31

Audible noise from wind turbines and sleep................................................................................................................31

Disturbance caused by wind turbines ..........................................................................................................................32

The impact of infrasound from wind turbines on health .............................................................................................36

Claims about the health effects of infrasound in public debates and the media ...................................................38

Audible noise generated by wind turbines and clinically manifested health effects ..................................................40

Electromagnetic fields and vibrations caused by wind turbines..................................................................................43

Justification for the choice of methodology, strengths and weaknesses of the study ................................................44

Conclusions........................................................................................................................................................................46

Recommendations ...........................................................................................................................................................48

References .......................................................................................................................................................................50

Appendix 1. Search terms and search strategy used .......................................................................................................57

Appendix 2. Tables of results...........................................................................................................................................58

Table 1. Systematic reviews.....................................................................................................................................58

Table 2. Observational studies ................................................................................................................................59

Table 3. Experiments on infrasound ........................................................................................................................61

Table 4. Experiments involving audible noise and visual aspects of wind turbines................................................65

4

Introduction and scientific background of the study Global climate change poses a threat to health and well-being around the world (Lancet Countdown, 2024).

As greenhouse gases are the main factor influencing climate change, the European Union has agreed to

reduce gas emissions by at least 55% by 2030 compared to 1990 levels in order to mitigate the

consequences (European Commission, 2019). To achieve this goal, many countries are increasingly

developing renewable energy production methods, particularly wind energy. One of the most important

issues in the development of wind energy is ensuring the health safety of wind farms. Conclusions about

health safety cannot be based solely on selective scientific literature, but require a systematic review of the

literature. A systematic review of the literature provides a much more comprehensive and reliable picture of

the existing scientific knowledge than reading individual scientific articles and drawing conclusions based

on their results.

One of the most important health safety issues is the noise generated by wind turbines. Some studies have

shown that noise from wind turbines (wind generators, windmills, wind turbines) is considered more

disturbing than other sources of noise, such as transport or industrial noise (Teneler and Hassoy, 2023). It

has also been argued that wind turbine noise can affect sleep and disturb people (Schmidt and Klokker,

2014). However, the stress and sleep disturbances caused by disturbance can in turn lead to clinically

apparent health effects (Basner et al., 2014). For example, traffic noise has been shown to increase the risk

of cardiovascular disease (heart attack and stroke) and metabolic disease (diabetes and obesity) (van

Kempen et al., 2018; Wu et al., 2023). However, reviews of wind turbine noise conducted to date have not

confirmed a link between wind turbine noise and clinically apparent health effects (Karasmanaki, 2022;

Schmidt and Klokker, 2014; Teneler and Hassoy, 2023; van Kamp and van den Berg, 2021, 2018).

However, whether and what links between wind turbine noise and health indicators have been found in

more recent studies has not been systematically analysed in recent years.

Wind turbine noise is classified as industrial noise, and in Estonia, the limit values for industrial noise in

residential areas are 60 dB (A) during the day and 45 dB (A) at night. The target values for noise are 50 dB

(A) during the day and 40 dB

(A) at night (Regulation No. 71 of the Minister of the Environment, 2016). Currently, when planning new

wind farms in existing residential areas, wind turbine noise must not exceed the limit values. In practice,

however, impact assessors generally base their assessments on outdoor noise target values, meaning that

wind turbine noise should not exceed 40 dB (A) in residential areas at night (Ministry of Climate, 2025). It

is important to determine whether the current regulations are sufficient to ensure health safety or whether

the limits should be tightened.

5

Various review studies suggest different values for safe wind turbine noise levels. A systematic review by

Schmidt and Klokker (2014) recommends that wind turbine noise levels in residential areas should not

exceed LAeq 35 dB(A) in order to ensure the least disruptive environment possible for residents (Schmidt

and Klokker, 2014). Ellenbogen et al.'s review on sleep disturbance (Ellenbogen et al., 2024) concluded

that wind turbine noise measured outside residential buildings up to 46 dB (A) or modelled according to the

new standard (ANSI/ACP 2022) up to 49 dB (A) does not pose a risk to human sleep. The World Health

Organisation (WHO) recommends limiting wind turbine noise on the exterior facade of residential

buildings to a weighted sound pressure level of Lden 45 dB (A) and Lnight 40 dB (A) (Basner and McGuire,

2018; Clark and Paunovic, 2018; Guski et al., 2017; van Kempen et al., 2018; WHO, 2009). However, Lden

is not directly comparable with the noise indicators used in our legislation. The different noise indicators,

units and terms are described in more detail in the chapter ‘Noise and infrasound terms and limits’. Davy et

al. (2020) have suggested that Lden 45 dB (A) corresponds to a noise level of LAeq 38.6 dB (A) (Davy et al.,

2020).

An Australian review study found that the main indicator for setting a limit value for wind turbines could

be that no more than 10% of the population should feel significantly disturbed by the noise. Based on their

analysis, this limit appears to be in the range of 34–40 dB LAeq ( 10 min) at the exterior of a dwelling, with an

average value of 37 dB LAeq (10 min) ( Davy et al., 2020).

An important topic in public debates in many countries and in non-scientific literature has been the possible

health risks associated with exposure to infrasound (the inaudible part of wind turbine noise with a

frequency below 20 Hz) (Schmidt and Klokker, 2014; van Kamp and van den Berg, 2018). However, no

adverse health effects associated with wind turbine infrasound have been demonstrated to date (Knopper

and Ollson, 2011; Schmidt and Klokker, 2014; van Kamp and van den Berg, 2018). Questions have also

been raised about the possible health effects of low-frequency noise (20–200 Hz) from wind turbines

(Schmidt and Klokker, 2014).

Another significant problem associated with wind turbines is that they disturb many people for various

reasons. Although disturbance cannot be considered a clinically manifest health outcome, it affects people's

well-being and can therefore be considered part of the World Health Organisation (WHO) definition of

health, according to which health is "a state of complete physical, mental and social well-being and not

merely the absence of disease or infirmity" (WHO, 2025). The level of disturbance is usually measured

using questionnaires in which respondents are asked to assess their level of disturbance.

6

Studies conducted to date have shown that, in addition to noise, visual aspects of wind turbines such as

shadow flicker, flashing lights on the blades, obstruction of views, etc. can also contribute to disturbance

(Freiberg et al., 2019b; Knopper et al., 2014). Disturbance also depends on people's own knowledge,

attitudes and perceptions. People who have a generally negative attitude towards wind energy have been

shown to experience greater disturbance compared to those who have a neutral or positive attitude

(Karasmanaki, 2022). It has been shown that people who believe that wind turbines are harmful to health

complain more about disturbance and also experience more self-reported health problems caused by wind

turbines. This phenomenon is called the nocebo effect (Davy et al., 2020; Karasmanaki, 2022). To reduce

disturbance, it is important to know exactly which factors increase people's disturbance in order to

minimise it.

Although many review articles analyse the health effects of wind turbines or wind farms, there are several

shortcomings in the study design and methodological quality. Most reviews are narrative (Karasmanaki,

2022; Knopper and Ollson, 2011; van Kamp and van den Berg, 2021) and/or based on cross-sectional

studies, where we cannot establish a definite temporal relationship between exposure and health effects.

Narrative reviews are at greater risk of bias than systematic reviews. One of the few systematic reviews

conducted to date is the study by Schmidt and Klokker (2014) on the health effects of wind turbine noise.

At the time of writing, however, no long-term follow-up studies had been conducted on this topic, so the

review relied on case studies and cross-sectional studies, which, due to their study design, cannot provide

evidence of causal relationships. Cohort studies have since been published that allow conclusions to be

drawn about causal relationships between wind turbine noise and health indicators, for example (Bräuner et

al., 2018; Poulsen et al., 2019a), but their results have not yet been systematically analysed. Experimental

studies are also important for establishing causal relationships. For example, an experimental study on the

effects of wind turbine infrasound was published in 2023 (Marshall et al., 2023), but the results of

experimental studies on the health effects of wind turbines have not been systematically analysed.

Noise and infrasound terms and limits

Noise is defined as sound that disturbs people or harms their health and well-being (Regulation No. 42 of

the Minister of Social Affairs, 2002). Noise is therefore a subjective indicator – what seems like noise to

one person may be music to another. Noise is caused by sounds.

7

Sound is a vibration that propagates through the environment (e.g. air, water or solid matter) in waves.

When sound propagates through the air, it causes fluctuations in air pressure. Sound pressure varies in

relation to atmospheric pressure, sometimes being greater than the mean atmospheric pressure and

sometimes less (Lahti, 2010). The intensity of sound is determined by the extent of the pressure change and

is measured in decibels (dB). A decibel indicates how much louder the sound is than the reference value. In

air, the reference value is an air pressure of 20 micropascals (20 μPa or 2×10⁻⁵ Pa), which is considered to

be the human hearing threshold at a frequency of 1000 Hz – this is the quietest sound that the average

person can still hear at this frequency (Basner et al., 2014). Sound pressure level (SPL) Lp is a logarithmic

measure that indicates the effective pressure of a sound compared to a reference value (agreed hearing

threshold) and is expressed in decibels (dB) (Regulation No. 42 of the Minister of Social Affairs, 2002).

Humans can perceive sound pressure in the range from 0 dB (agreed hearing threshold) to over 140 dB

(pain threshold). The smallest change in sound level that humans can distinguish outdoors is 3 dB. This

change corresponds to a doubling of sound energy. A 10 dB increase is subjectively perceived as twice as

loud, although physically it represents a tenfold increase in sound energy (Ellenbogen et al., 2024).

In addition to sound intensity, sound is also characterised by its frequency. Frequency is the number of

sound wave amplitudes per second and is measured in hertz (Hz). The lower the sound, the lower the

frequency, and the higher the sound, the higher the frequency. The timbre of a sound is determined by the

frequency distribution of sound waves (Lahti, 2010).

Sound is conventionally divided into three categories according to frequency: infrasound (< 20 Hz), audible

sound (20–20,000 Hz) and ultrasound (> 20,000 Hz). Infrasound is generally lower than the human hearing

threshold and is usually inaudible to humans. However, there are no clear boundaries in nature – the

transition between audible and inaudible sound is gradual and smooth. At higher frequencies, infrasound is

audible or perceptible when it is very strong (Ellenbogen et al., 2024). Sources of infrasound include

natural phenomena such as wind, waves, heartbeats, etc. Blue whales (10–20 Hz) also generate and hear

infrasound. The most common man-made sources of infrasound are fans, motors, cars, aeroplanes, trains,

air source heat pumps, heating systems, etc. (Ellenbogen et al., 2024; Lahti, 2010).

Sources of audible noise include traffic noise, noise from neighbours' activities, and noise in the workplace.

Low-frequency sound is sometimes considered to be the lowest frequency range of audible sound, 20–200

Hz (Ellenbogen et al., 2024), but in some scientific articles and also according to Regulation (Regulation

No. 42, 2002) by the Minister of Social Affairs, , low-frequency noise

8

in the frequency range 10 Hz–200 Hz, which partially overlaps with the frequency of infrasound, but is also

audible.

Humans hear sounds of different frequencies with varying intensity, but cannot hear sounds that are too

high or too low at all. Hearing is most sensitive in the 2000–5000 Hz range. For example, the sensitivity of

hearing for sounds with a frequency of 100 Hz is approximately 20 dB lower (Lahti, 2010). This means that

if a sound contains more low or high frequencies, a higher sound pressure level (sound intensity) is required

to perceive it (Ellenbogen et al., 2024). Since the measuring device measures all frequencies equally, a

correction is made to better assess noise, which evaluates sound frequencies similarly to the human ear.

This is called A-correction (LpA). The terms A-sound level or A-weighted sound level may also be used. The

pressure symbol p can be omitted from the sound level symbol LpA. The symbol dB (A) is also often used

after the unit to express the A-weighted sound level (Lahti, 2010).

For measuring low-frequency sound and infrasound, C- or G-correction is used, or measurements are taken

without correction (Z). Z-correction is denoted by dB (Z) and actually means unweighted sound level

across the entire frequency range. Z comes from the English expression zero-weighted. Figure 1 shows how

different frequency corrections assess noise.

Figure 1. Frequency correction (frequency weighting) curves (Health Board, 2025a)

9

Since sound pressure (sound intensity) changes constantly over time, the time during which noise is

measured and averaged must also be taken into account when assessing environmental noise. For this

purpose, the equivalent sound pressure level LpA,eq,T is used, which characterises noise with a variable level

over a period of time T, using A-correction (Lahti, 2010; Regulation No. 42 of the Minister of Social

Affairs, 2002). The duration of noise measurement (T) is selected according to the measurement method

and the nature of the noise; for noise with varying levels, the minimum period is 10 minutes (Regulation

No. 42 of the Minister of Social Affairs, 2002).

In addition to the equivalent noise level, the maximum sound pressure level LpA,max is also used to

characterise noise, which indicates the maximum value of the sound pressure level measured during a

specified period (Lahti, 2010; Regulation No. 42 of the Minister of Social Affairs, 2002).

In Estonia, noise limits have been established for A-weighted sound pressure levels in the outdoor

environment (Regulation No. 71 of the Minister of the Environment, 2016). Indoors, both A-weighted noise

levels and C-weighted low-frequency noise levels are regulated, and there are also requirements for

assessing low-frequency noise without frequency correction (Regulation No. 42 of the Minister of Social

Affairs, 2002). Currently, the limit values for industrial noise in residential areas are 60 dB (A) during the

day and 45 dB

(A) in residential areas. The target noise levels are 50 dB (A) during the day and 40 dB (A) at night

(Regulation No. 71 of the Minister of the Environment, 2016). In bedrooms, traffic noise levels may be up

to 30 dB (A) at night and noise from technical equipment up to 25 dB (A), and 50 dB (C) both at night and

during the day (Regulation No. 42 of the Minister of Social Affairs, 2002).

A low-frequency noise frequency curve has been established to limit low-frequency sounds (Figure 2).

(Regulation No. 42 of the Minister of Social Affairs, 2002). These are values without frequency correction

in the frequency range 10–200 Hz. This frequency curve allows for higher uncorrected sound levels at

lower frequencies (10 Hz to 95 dB (Z)) and lower levels at higher frequencies (200 Hz to 32 dB (Z)). In

addition, Estonia has a regulation on infrasound and ultrasound, which sets the limit value for the G-

corrected equivalent sound pressure level LpG,eq,T at 85 dB (G) (Regulation No. 75 of the Minister of Social

Affairs, 2002).

10

Figure 2. Standard levels for sounds without frequency correction according to the annex to Regulation No.

42 of the Minister of Social Affairs of 4 March 2002 (Health Board, 2025b).

One characteristic feature of wind turbine noise is amplitude modulation (AM), i.e. a sound level that varies

over time with approximately regular periodicity. AM is a fluctuation in sound level whose cycle time

generally corresponds to the frequency of blade passage from the wind turbine tower. The fluctuations in

sound level are usually small, around 2–4 dB (McCunney et al., 2014), which is why wind turbine noise is

considered to be continuous rather than impulsive (Ellenbogen et al., 2024). In some rare cases, however,

fluctuations in wind turbine sound levels can be much greater – up to 10 dB. In groups of several wind

turbines, the modulations of individual generators can synchronise, causing periodic increases in

modulation intensity. There may also be periods when the modulations of individual generators balance

each other out, reducing the modulation strength. Sometimes the synchronisation of generators can last for

hours if the wind speed and direction remain constant. The AM level is not correlated with wind speed.

Most cases of so-called 'strong' AM are caused by unusual meteorological conditions. AM also varies

depending on location — in some places it occurs rarely, while in others it has been measured up to 30% of

the time (McCunney et al., 2014).

In addition, the following noise indicators are used to assess environmental noise (Regulation No. 71 of the

Minister of the Environment, 2016):

Lden – day-evening-night noise indicator. A long-term average sound pressure level determined on the basis

of the numerical values of all day, evening and night sound pressure levels throughout the year, which is a

general noise disturbance indicator.

11

disturbance indicator. When determining Lden, a correction factor of +5 dB is applied to evening noise and

+10 dB to night-time noise. The indicator is calculated on the basis of Lnight, Lday and Levening.

Lnight – night-time noise indicator. The long-term average sound pressure level determined on the basis of all

night-time hours of the year, which is an indicator of noise disturbing sleep and characterises sleep

disturbance between 23:00 and 7:00. Determined in accordance with standard ISO 1996-2: 1987.

Lday – daytime noise indicator. The long-term average sound pressure level determined on the basis of all

daytime hours of the year, which characterises the disturbing effect of noise during the day between 7:00

and 19:00. Determined in accordance with standard ISO 1996-2: 1987.

Levening – evening noise indicator. Long-term average sound pressure level determined on the basis of all

evening times of the year, which characterises the disturbing effect of noise in the evening between 19:00

and 23:00. Determined in accordance with standard ISO 1996–2: 1987.

12

Purpose of the study The aim of the study was to systematically analyse the evidence published in the scientific literature over the

last fifteen years (2010–2025) on the health effects of wind turbines.

Research questions:

1. What are the main conclusions of existing studies on the health effects of wind turbines?

2. What is the overall quality of the existing evidence? Is there evidence in the scientific literature

that wind turbines have a negative impact on human health?

3. If wind turbines have negative health effects, what health effects are associated with wind

turbines?

4. If wind turbines have negative health effects, what role do environmental factors such as noise,

infrasound, shadow flicker, visual aspects, psychological factors (including general attitudes

towards wind turbines, people's beliefs and perceptions of wind turbines), vibration and

electromagnetic fields in causing these health effects?

5. If wind turbines have health effects, under what conditions are these health effects more likely

to occur (e.g. at what distance from the turbine, with powerful or tall turbines, etc.)?

6. Are certain population groups more vulnerable to the potential health effects of wind turbines?

7. What evidence-based recommendations can be made to policymakers, industry stakeholders

and affected communities to protect human health?

13

Methods This study was conducted as a systematic literature review. In compiling the systematic literature review,

we used the principles of rapid review methodology (Garritty et al., 2024; King et al., 2022). A systematic

review is a scientific research method that aims to collect, evaluate and synthesise all relevant scientific

studies on a specific research question or topic using a systematic, transparent and repeatable process to

minimise the risk of bias in the conclusions drawn. To achieve this goal, specific criteria are agreed upon

for the inclusion and analysis of studies. Several members of the research team participate in each stage of

the study, checking each other's work. The included studies are compared with each other on the basis of

study quality, and the results of higher quality studies are given greater weight in the conclusions. Scientific

studies are the most reliable way to obtain accurate information, but it is not possible to conduct a scientific

study without limitations or the risk of obtaining inaccurate results. For example, a study may suffer from

"selection bias". This occurs when the participants in the study do not adequately represent the target group.

For example, people who are more health-conscious may be more likely to participate in a sleep study than

the general population we want to study. Self-reported data is subject to "recall bias"; for example, patients

with a disease may remember their exposure to noise better than healthy people. "Measurement bias"

occurs when exposure or outcome data are not measured accurately, for example, there is a measurement

error in noise measurements or different methodologies have been used to measure noise in comparable

groups. In observational studies, it is important to take into account "confounding factors". Confounding

factors occur when another factor is associated with both the exposure under study and the outcome. For

example, wealthier people may live in areas with less noise and also have better opportunities for healthier

lifestyles, and in fact, it is people's income that influences the onset of disease, even though the analysis

shows a link with noise. This can be avoided by adjusting for confounding factors (e.g. income, age,

gender). Adjustment ensures that only subjects with similar adjusted characteristics are compared.

Scientific journals tend to exhibit

"publication bias". Studies that find a link are more likely to be published than studies that do not find a

link. The most common risk of bias in experiments is "lack of blinding" – if participants or researchers

know which study group is receiving the placebo infrasound and which is receiving the real sound, this will

affect the results. People with certain characteristics, such as those who are sensitive to noise, may also

drop out of the experiment. There may also be errors in measuring the outcome. For example, a person may

have high blood pressure, but if a doctor has not diagnosed it

14

diagnosed it, they will be considered healthy in a registry-based study. At the same time, if a person reports

that they have high blood pressure, it is not known whether this is a temporary increase in blood pressure

(e.g. due to stress) or whether they have developed hypertension. We took all these potential risks of error

into account when drawing conclusions in our study.

We used the internationally recognised PECO (population, exposure, comparison, outcome) framework to

define the research questions. We included pre-reviewed scientific articles in the systematic literature

review that dealt with:

• Population: the general population (all people)

• Exposure to the following factors of onshore and offshore wind turbines or wind farms:

Acoustic factors: noise (frequency above 20 Hz), low-frequency noise (20 Hz to 200 Hz) and

infrasound (below 20 Hz) from wind turbines or wind farms; visual factors: visual disturbance

of the landscape, shadow flicker, flashing lights associated with wind turbines at night, direct

visibility from home windows; psychological factors: general attitude towards wind turbines,

people's beliefs and perceptions of wind turbines (as a disturbing factor), disturbance from

wind turbines; vibration; electromagnetic fields.

• Comparison group: people who have no exposure to wind turbines or less exposure than

others; placebo groups in experimental studies; the same people before and after exposure (i.e.

self-controls) in ‘before and after’ studies.

• Health outcomes: all possible health-related outcomes, including disturbance, sleep

disturbances, health symptoms (measured subjectively or objectively, measured physiological

parameters), health-related quality of life, psychological indicators.

Only studies with a study design that allows causal relationships to be identified were included. Studies

with the following study designs were included:

• Single studies: longitudinal studies (cohort studies, case-control studies); intervention studies

(comparisons before and after the installation of wind turbines), experimental studies/trials

• Systematic reviews

We did not include the following studies in this systematic review:

15

• Studies that did not address human health effects. Studies that only addressed technical aspects of wind turbines or studied animals, wildlife or laboratory animals

• Studies that dealt with exposure to wind turbines in the working environment, not in the living environment

• Studies that addressed the health effects of sound (noise), infrasound, electromagnetic fields, vibration and visual aspects, but were not related to wind turbines.

• Individual studies published before 2010 and systematic reviews published before 2015

• Narrative review studies, case studies, cross-sectional studies, letters to scientific journals,

comments in scientific journals, editorials in scientific journals, conference summaries

• Information sources that were not published in peer-reviewed scientific literature

• Non-English studies

If a comprehensive systematic review on the research question had been published in 2015 or later, only

individual studies published after the inclusion period of the review were included in the study. In other

cases, the research question was answered by including all individual studies that met the inclusion criteria

(Garritty et al., 2024; King et al., 2022).

A systematic literature search was conducted on the health effects of wind turbines (wind farms). The

search was conducted in the PubMed and Scopus databases for the period from 1 January 2010 to 22 April

2025. The PubMed database search strategy is presented in Appendix 1.

Endnote reference management software was used to remove duplicates, and Mendeley Reference Manager

software was used to manage references. The initial selection of studies (screening) was based on the title

and abstract of the study. Initially, two members of the research team independently screened 20

publications and then discussed the results to reach a common understanding regarding the inclusion of

studies. To ensure the quality of the review, two members of the research team screened 20% of the entries,

while the rest were screened by only one member of the research team.

The full texts of studies initially assessed as potentially suitable were retrieved and a second screening was

conducted on the basis of these. One member of the research team read through the full texts of potentially

suitable studies and, if he or she decided to exclude a study, another member of the research team read

through the full text and confirmed or rejected the decision to exclude it. However, studies deemed suitable

by the second member of the research team were discussed, and the decision to include or exclude them

was made by consensus. The data from the articles were entered into an MS Excel table.

16

Finding studies through database searches

Database searches (n = 1374): PubMed (n = 716) Scopus (n = 658)

Duplicate entries removed (n = 211)

Included in the review (n = 32), of which systematic reviews (n = 4)

Results The database search yielded 1,374 entries (scientific articles), of which 1,163 were assessed for compliance

with our inclusion criteria based solely on their titles and abstracts. The full texts were reviewed in 115

cases. In addition to the database search, one study was found through a manual search. Thirty-two studies

met the inclusion criteria and were used in the analysis of this study, of which four were systematic

reviews, 19 were experiments and n i n e were observational studies (Figure 3).

The most important results of all included articles are summarised in the tables (Appendix 2, Tables 1–3).

Figure 3. PRISMA flow diagram for describing scientific literature searches.

Full texts reviewed (n = 115)

Full texts searched (n = 115)

Found by browsing (n = 1):

Excluded (n = 84): study design (n = 55) no health impact (n = 15) included in systematic review (n = 8) Other (n = 6)

Full texts unavailable (n = 0)

Entries removed (n = 1048)

Initial screening of records based on title and abstract (n = 1163)

R ev

ie w

Se ar

ch In

cl us

io n

17

The impact of audible noise from wind turbines on human health

Audible noise from wind turbines and sleep

Two systematic reviews/meta-analyses (Godono et al., 2023; Liebich et al., 2021) (Appendix 2, Table 1),

two experiments in a sleep laboratory (Liebich et al., 2022a, 2022b) (Appendix 2, Table 4) and one long-

term follow-up study (Poulsen et al., 2019b) (Appendix 2, Table 2).

Both systematic reviews included in this review concluded that increased wind turbine noise increases the

risk of self-reported sleep disturbances and/or lower sleep quality (Godono et al., 2023; Liebich et al.,

2021). However, one of the included reviews (Liebich et al., 2021) found no effect of noise on objectively

measured sleep parameters (using polysomnography (PSG) and actigraphy). Another review study only

considered subjectively assessed sleep parameters and found that both wind turbine distance and sound

levels above 30 dB(A) affect self-reported sleep disturbances (Godono et al., 2023).

The aim of a systematic review and meta-analysis (Liebich et al., 2021) was to assess the impact of wind

turbine noise on sleep using only validated subjective and objective measures. Objective measures were

assessed using polysomnography (PSG) and actigraphy. Polysomnography is the 'gold standard' for

objective sleep measurement as it uses direct electroencephalography (EEG). An actigraph is a wrist-worn

motion sensor that detects sleep and wakefulness based on general body movements. Subjective measures

included sleep diaries and questionnaires. The review included nine studies that used widely accepted and

validated objective and subjective sleep assessment methods and were published after 2000. Five studies

were included in the meta-analysis, four of which used PSG and one of which used actigraphy ( Liebich et

al., 2021).

The review study described (Liebich et al., 2021) showed that wind turbine noise does not significantly

affect the main objective indicators of sleep: sleep onset latency (SOL), wake after sleep onset (WASO),

total sleep time (TST) and sleep efficiency. However, an impact was found on subjectively measured (self-

reported) sleep indicators. Based on the questionnaires, wind turbine noise affected the severity of insomnia

symptoms (Insomnia Severity Index, ISI), sleep quality (Pittsburgh Sleep Quality Index, PSQI) and daytime

sleepiness. There were also indications that higher amplitude modulation (AM) may increase wakefulness

and reduce the duration of deep sleep.

18

Another systematic review and meta-analysis on sleep (Godono et al., 2023) assessed the impact of wind

turbine noise only on self-reported subjective sleep indicators. The study found that both the distance from

wind turbines and sound intensity affect self-reported sleep disturbances. The lowest prevalence of sleep

disturbance was found at sound levels <30 dB(A) (31%) and increased as noise levels increased. The

prevalence of self-reported sleep disturbances at different distances (in metres) from the nearest wind

turbine in the wind farm was as follows in the study (Godono et al., 2023):

• <500 m: 79%

• 500–1000 m: 65%

• 1000–1500 m: 41%

• 1500–2000 m: 29%

• 2000–3000 m: 22%

Godono et al. (2023) reviewed studies from Europe, the US, Canada, Japan, and China published between

2004 and 2021. The distance to wind turbines in the included studies ranged from 495 to 3093 m, and the

capacity of the wind turbines ranged from 0.5 to 3.5 MW.

Wind turbines can be a source of stress, which can contribute to self-reported sleep disturbances. However,

it should be noted that self-reported sleep indicators are affected by recall bias, inaccuracy in sleep

perception, or misinterpretation of awakenings, and the study by Godono et al. (2023) only included cross-

sectional studies, which do not allow conclusions to be drawn about causal relationships. The authors of

Godono et al. (2023) rate the quality of most of the studies included as low, as they often did not adjust for

many necessary factors, such as air pollution, visual disturbance, air temperature and humidity, and the

economic situation of the subjects (Godono et al., 2023).

In Australia, two single-blind randomised controlled trials were conducted in a sleep laboratory, where

subjects were exposed to pre-recorded audible noise and infrasound from a wind farm. In the first study

(Liebich et al., 2022a), 68 subjects aged 18–80 slept for seven consecutive nights in a sleep laboratory,

where they were exposed to a wind turbine noise recording at a volume of 25 dB(A). The recording used

was a recording of real wind farm noise made indoors in a house located 3.3 km from a wind farm in South

Australia. The noise level of 25 dB(A) was chosen because it is similar to the median noise level of wind

turbines measured indoors throughout the year at a distance of 1–3 km from wind farms. This level was

also 6 dB (A) higher than the laboratory background noise and therefore clearly audible to participants with

normal hearing. The recordings contained infrasound from a frequency of 1 .6 Hz and perceptible

amplitude modulation ( AM). Objective sleep parameters

19

were assessed using polysomnography (PSG). Electromyography, electrooculography, electrocardiography,

pulse oximetry and leg movement signals were also recorded. Subjective sleep indicators were assessed

using a validated web-based ‘sleep diary’ questionnaire. This diary included questions about time spent in

bed, time spent out of bed, and minutes of sleep and wakefulness during the night, allowing for the

calculation of time spent in bed, sleep onset time, number and duration of awakenings, time of awakening,

and total sleep time. The subjects were divided into four groups: those who reported sleep disturbances

related to wind turbine noise (N=14, living <10 km from the wind farm); those without sleep disturbances

(N=18, living <10 km from the wind farm); traffic noise sleep disturbance (N=18), who reported sleep

disturbances related to traffic noise; control group (N=18), who lived in a quiet rural area. During the

experiment, four different noise scenarios were played to the subjects in random order on four nights after

an adaptation night: quiet control night with only laboratory background noise of 19 dB (A); continuous

wind turbine noise at 25 dB(A) throughout the night; wind turbine noise at 25 dB(A) only during

established sleep; and wind turbine noise at 25 dB(A) only during wakefulness or light sleep.

The results of the study (Liebich et al., 2022a) showed that there was no significant effect of wind turbine

noise on the sleep parameters studied. This means that at a level of 25 dB (A), wind turbine noise did not

significantly affect sleep efficiency, sleep latency, total sleep time, wake time after sleep onset, or different

sleep stages, regardless of the participant's previous exposure to noise or self-reported sleep disturbances.

Those who were more sensitive to noise slept worse on the control night than on the night without noise.

However, the sleep of those sensitive to noise did not differ from that of others on the nights with noise.

Another study (Liebich et al., 2022b) was conducted with louder recorded wind turbine noise and involved

23 subjects aged 18–29 who had not previously been exposed to wind turbine noise. The recorded wind

farm noise (which also included infrasound) was presented in a sleep laboratory at an intensity of 33 dB (A)

in random order, alternating with laboratory background noise of 23 dB (A). The wind farm noise

contained infrasound and noticeable AM at 46 Hz. The study showed that wind turbine noise at a level of

33 dB(A) does not prolong the time it takes to fall asleep, as measured objectively or subjectively, in

young, healthy people who have not previously been exposed to wind turbines.

These two high-quality randomised controlled trials (Liebich et al., 2022a, 2022b) showed that wind

turbine noise at a level of 25 dB (A) has no effect on objective or subjective sleep parameters, even in older

and noise-sensitive individuals. No effect was observed even in those who reported wind turbine-related

sleep disturbances. Based on these studies, it can be assumed that wind turbine noise levels below 25 dB

(A) indoors, which also includes amplitude modulation, are safe even for sensitive groups. Even wind

turbine noise levels of 33 dB (A) did not prolong sleep

20

, but this study only included young people and only looked at one sleep indicator, so no conclusions can be

drawn about other sleep indicators or older people.

However, the disadvantage of randomised controlled trials is that they do not show the long-term effects of

noise. The long-term effects of noise on sleep have been studied in a large cohort study conducted in

Denmark (Poulsen et al., 2019b). This study assessed the impact of night-time wind turbine noise in the

vicinity of residential buildings and the impact of low-frequency night-time noise indoors on an objectively

measured sleep disturbance indicator, which was the purchase of sleeping pills based on the Danish patient

register.

The study by Poulsen et al. (2019b) included 583,968 subjects who were not taking sleeping pills at the

start of the follow-up. They were followed for 17 years (1996 to 2013). Of these, 68,696 were taking

sleeping pills at the end of the follow-up period. The subjects were aged 25–84 (Poulsen et al., 2019b).

Wind turbine noise was modelled (only noise originating from wind turbines was assessed) for all Danish

dwellings located within a radius of up to 20 wind turbine tower heights (the study group). For example, if

the height of the wind turbine tower was 35 m, the noise from the wind turbines was modelled for

dwellings within a radius of 700 m; if the height of the tower was 100 m, the radius was 2000 m. Wind

turbine noise was also modelled for 25% of randomly selected dwellings located within a radius of 20 to 40

wind turbine tower heights (control group). Two noise indicators were modelled: A-weighted night-time

noise level near the dwelling in the yard at frequencies of 10–10,000 Hz (which, in addition to audible

sound, also included low-frequency sound and, in part, infrasound) and, separately, only low-frequency

noise ( also including, in part, infrasound) inside dwellings at frequencies of 10–160 Hz. Wind turbine

noise outdoors near residential buildings was divided into the following classes: less than 24; 24–30; 30–

36; 36–42 and over 42 dB (A). Low-frequency wind turbine noise indoors was divided into the following

classes: less than 5; 5–10, 10–15 and over 15 dB (A). The analysis used 1-year and 5-year average noise

indicators (Poulsen et al., 2019b).

Log-linear Poisson regression analysis was used to assess the associations between noise levels and the

consumption of the sleeping pills under investigation. All analyses were adjusted for gender, calendar year

and age. In addition, the analyses were adjusted for educational level, income, marital status, labour market

participation status, type of housing (farm, single-family house, other), traffic load within a 500 m radius of

the place of residence, and distance from the nearest road with more than 5,000 vehicles per day. The

analyses took into account changes in these characteristics over time during the follow-up period (Poulsen

et al., 2019b).

The study described by Poulsen et al. (2019b) found that, compared to those whose homes were in the

lowest noise class (less than 24 dB (A)), there was a statistically significant 3–8% higher risk of sleeping

pill use among those living in higher noise classes ( noise near residences

21

outdoors 24–30; 30–36; 36–42 dB (A)). Those who lived in the highest noise class (over 42 dB (A)) had a

14% higher risk of using sleeping pills compared to those who lived in the lowest noise class (less than 24

dB (A)), but this difference was not statistically significant. In order to find out whether people of different

ages may be affected differently by wind turbine noise, the study analysed the data by age group. It was

found that wind turbine noise affects older people more. Those over 65 years of age who lived near wind

turbines with an outdoor noise level of over 42 dB (A) had a 68% higher risk of using sleeping pills

compared to those who lived near wind turbines with an outdoor noise level of less than 24 dB (A). In the

over-65 age group, a higher risk of sleeping pill use was observed from a night-time noise level of 30

dB(A) outdoors near their homes. No effect of low-frequency noise indoors on sleeping pill use was found.

Audible noise and disturbance caused by wind turbines

Disturbance is an indicator measured by a questionnaire. The respondent assesses how disturbing they find

the noise on a given scale. The relationship between environmental noise and disturbance is discussed in a

review article (Guski et al., 2017) (Appendix 2, Table 1). This review focused on environmental noise,

assessing the relationship between traffic noise from roads, railways and air traffic and disturbance, in

addition to noise from wind turbines. Four cross-sectional articles published between 2000 and 2012 were

included from studies on wind turbine noise. The study showed that as wind turbine noise increases, the

likelihood of disturbance increases, but this relationship is not as clear as in the case of traffic noise.

Several of the studies included in this review (Appendix 2, Table 4) investigated the extent to which one

specific characteristic of wind turbine noise, amplitude modulation (AM), contributes to annoyance

(Ioannidou et al., 2016; Lee et al., 2011; Schäffer et al., 2016, 2018). In addition, these studies also

examined the effect of noise frequency distribution and source origin on annoyance.

A laboratory experiment (Ioannidou et al., 2016) was designed to investigate how AM depth (experiment

1), AM frequency (experiment 2) and the interaction between AM type and depth (experiment 3) affect

annoyance. The subjects were presented with sound of varying modulation depth (more uniform or more

variable) at an intensity of 60 dB(A) for 30 seconds. Modulation depth was defined as the difference

between the maximum and minimum sound levels and ranged up to 12 dB(A) in different sound stimuli.

Both recorded and generated wind turbine sounds with frequencies between 200 and 1200 Hz were used.

During the experiment, the subjects were asked to rate the annoyance of different sound samples on a 10-

point scale. The study found that annoyance is influenced by AM depth. This means that the smaller the

range in which the noise level fluctuates, the less disturbing the sound is. It was also found that the

frequency of

22

the noise level fluctuates. The authors of the study recommend taking measures to make the sound of wind

turbines more uniform.

Amplitude modulation is also the focus of a laboratory experiment (Lee et al., 2011) in which participants

were presented with noise at different modulation levels and intensities: 35, 40, 45, 50 and 55 dB

(A). Recordings of the audible noise from a real wind turbine (1.5 MW capacity with a rotor diameter of 72

m) at different distances from the turbine at frequencies of 250–8000 Hz were used. This study also showed

that disturbance increased with increasing AM depth. The study by Schäffer et al. (2016) also found that the

presence of AM increased disturbance. In addition, Schäffer et al. (2016) found that at the same sound

intensity, disturbance was greater for wind turbines than for traffic noise (Appendix 2, Table 4).

Another study conducted later by the same researcher (Schäffer et al., 2018) also found that AM depth

increases disturbance. In addition, it was found that artificially enhanced low-frequency component sound

was more disturbing than wind turbine noise. The disturbance was not related to the gender or noise

sensitivity of the participants, but was higher with increasing age and lower with a more positive attitude

towards wind farms (Schäffer et al., 2018).

Audible noise from wind turbines and clinically expressed health effects

All observational studies included in this review examined the effects of audible noise (including low-

frequency noise) on various objectively measured health indicators. Eight cohort studies (Bräuner et al.,

2019a, 2019b, 2018; Poulsen et al., 2019a, 2019b, 2018a, 2018b, 2018c) and one case-crossover study

(Poulsen et al., 2018d) (Appendix 2, Table 2). All observational studies were conducted in Denmark and

are based on data from two study cohorts (groups of people) – a cohort of Danish nurses and a register-

based cohort of the entire Danish population.

The Danish Nurses Cohort was established in 1993 when a questionnaire was sent to female members of

the Danish Nurses Organisation who were at least 44 years old at the time. Initially, the cohort consisted of

28,731 nurses. The Danish Population Register was used to obtain information on their places of residence.

The Danish Nurses Cohort was used to study the incidence of stroke (Bräuner et al., 2019a), myocardial

infarction (Bräuner et al., 2018) and cardiac arrhythmias (atrial fibrillation) (Bräuner et al., 2019b) in

relation to wind turbine noise. Morbidity was determined based on Danish patient and cause of death

registries.

In all studies, A-weighted wind turbine noise was modelled at frequencies of 10–10,000 Hz and the annual

average Lden was calculated at the exterior facade of the residence under investigation within a 6 km radius

of the nearest wind turbine. In all three studies, wind turbine noise levels were low, with only 3% of those

studied being exposed to

23

wind turbine noise on the exterior facade of their homes above 29.9 dB (A). The studies found no link

between wind turbine noise and the occurrence of myocardial infarction and stroke.

The results of the study (Bräuner et al., 2019b) showed that an average annual night-time wind turbine

noise level on the exterior facade of a residential building of Lnight above 20 dB (A) may increase the

incidence of atrial fibrillation compared to those with less than 20 dB of wind turbine noise at night. A

similar association was found with the Lday and Leavning indicators, but no such association was found with the

24-hour indicator Lden. The studies were adjusted for important factors such as age, calendar year (when the

cohort was recruited), employment status, smoking status, alcohol consumption, physical activity, body

weight, other diseases, marital status, etc. The study also examined how traffic noise and air pollution

(NO2) could affect the associations. Adjustment did not significantly change the associations found (or the

lack thereof).

Poulsen's studies (Poulsen et al., 2019a, 2019b, 2018d, 2018a, 2018b, 2018c) were based on data from a

nationwide register-based cohort in Denmark. The study included all wind turbines in Denmark and all

residents who had lived for one year or more between 1996 and 2013 within 20 wind turbine tower heights

of a wind farm, with a random sample of 25% of all subjects who had lived in dwellings located 20–40

wind turbine tower heights from the nearest wind turbine in the wind farm. To assess exposure, A-weighted

night-time (10 p.m. to 7 a.m.) wind turbine noise was modelled (calculated) taking into account the type,

height, location, wind direction and other weather conditions of the wind turbine. Noise near residential

buildings outdoors was modelled at frequencies of 10 to 10,000 Hz (including low-frequency noise) and

low-frequency noise indoors (10–160 Hz) was modelled separately. Night-time noise near residential

buildings outdoors was divided into classes: less than 24; 24–30; 30–36; 36–

42 and over 42 dB (A) and low-frequency noise indoors below 5; 5–10, 10–15 and over 15 dB (A). All

analyses were adjusted for gender, calendar year of inclusion in the study and age. In addition, the analysis

was adjusted in different models for the educational level of the subjects, personal income, marital status,

labour market participation, average income in the region, type of housing (apartment or house), distance

from the road (with ≥ 5000 vehicles per day), and traffic load within a 500 m radius of the dwelling. All

data were updated according to changes in the person's place of residence during the follow-up period.

The study found that long-term night-time exposure to wind turbine noise outdoors near homes or low-

frequency wind turbine noise indoors does not affect the risk of developing diabetes and high blood

pressure (Poulsen et al., 2018b, 2018a). Similarly, no evidence was found that wind turbine noise was

associated with any of the adverse birth outcomes studied: preterm birth, low birth weight at term, or low

birth weight for gestational age at term (Poulsen et al., 2018c).

24

However, it was found that high long-term night-time exposure to wind turbine noise (10 to 10,000 Hz)

audible outdoors near residential areas increases the risk of taking sleeping pills and antidepressants. Noise

above 42 dB(A) increased the risk of taking antidepressants by 17%. Those over 65 years of age were most

affected. Low-frequency noise indoors (10–160 Hz) did not affect the purchase of sleeping pills and

antidepressants (Poulsen et al., 2019b).

Studies on heart attacks and strokes do not provide clear answers. For example, a study by Poulsen et al.

(2019a) found conflicting evidence on the link between wind turbine noise and heart attacks or strokes.

Participants who were exposed to night-time noise levels of 24–30 dB(A) and 30–36 dB(A) outdoors near

their homes had a higher risk of heart attack and stroke than those exposed to noise levels below 24 dB(A).

However, the increased risk was not statistically significant for exposure to louder wind turbine noise of

36–42 dB

(A) and ≥42 dB (A), indicating that there is no dose-response relationship. No association was found

between low-frequency noise indoors and the occurrence of heart attacks and strokes.

Poulsen et al., 2018d investigated whether short-term changes in wind turbine noise could affect

hospitalisation and death due to heart attacks and strokes. Noise levels were examined four days prior to

illness or on a reference day. The results did not provide convincing evidence of a link between short-term

night-time noise and myocardial infarction or stroke. No statistically significant associations were found in

the main analysis. However, the study by Poulsen et al. (2018d) concluded, based on additional analyses,

that short-term exposure to higher low-frequency noise indoors may trigger heart attacks or strokes

(Poulsen et al., 2018d).

An experiment conducted in Taiwan (Chiu et al., 2021) also suggests that wind turbine noise may

contribute to the development of cardiovascular disease. The experiment involved volunteers who lived up

to 500 m away from wind turbines. The subjects were divided into two groups: one group spent 30 minutes

outdoors 20 metres from the nearest wind turbine in the wind farm, and the other group spent 30 minutes

indoors 500 metres from the nearest wind turbine in the wind farm. The heart rate and heart rate variability

of the subjects were measured during the experiment using a portable electrocardiogram (ECG) recorder. In

the test area 20 m from the nearest wind turbine, low-frequency noise (20–200 Hz) ranged from 38.3 dB

(A) to 57.1 dB (A). The results of the experiment showed that the higher the wind turbine noise, the lower

the heart rate variability (HRV). Higher HRV usually indicates better physical adaptability and stress

tolerance, while lower HRV may indicate stress, fatigue or health problems. Thus, this study shows that

unusually high wind turbine noise (people do not usually live so close to wind turbines) can increase the

risk of cardiovascular disease. In addition, the study measured low-frequency noise inside residential

buildings

25

, which were located 124–330 metres from the nearest wind turbine. The low-frequency noise levels

measured inside these homes at 20–200 Hz ranged from 30.7 to 43.4 dB(A).

The impact of infrasound generated by wind turbines on human health

Evidence of the potential impact of wind turbine infrasound on human health comes from 13 experiments

in our study (Appendix 2, Table 3). The main health outcomes analysed in relation to wind turbine

infrasound were sleep quality (Liebich et al., 2022a, 2022b; Marshall et al., 2023), mental health (Ascone et

al., 2021; Małecki et al., 2023; Rosciszewska et al., 2025), disturbance and self-reported symptoms (Ascone

et al., 2021; Maijala et al., 2021; Marshall et al., 2023). In a series of experiments, Crichton et al.

investigated whether wind turbine infrasound causes disturbance and various symptoms, such as headaches,

pressure in the ears, etc., and to what extent the disturbance and reporting of symptoms are influenced by

information circulating in the media and social media that infrasound is dangerous to health (Crichton et al.,

2015, 2014b, 2014a; Crichton and Petrie, 2015a, 2015b).

Five experiments investigating the effects of wind turbine infrasound alone (without accompanying audible

noise) were included in the present study (Ascone et al., 2021; Crichton et al., 2014b; Małecki et al., 2023;

Marshall et al., 2023; Tonin et al., 2016) (Appendix 2, Table 3).

The study with the highest quality of evidence among these is the experiment conducted in an Australian

sleep laboratory (Marshall et al., 2023). It was a randomised, double-blind trial conducted in an isolated

sleep laboratory under controlled conditions (noise levels and other factors are precisely known in the

laboratory), which was designed in the style of a studio apartment. The study included people who, based

on a questionnaire, considered themselves to be noise-sensitive. The study involved 37 noise-sensitive but

otherwise healthy adults aged 18–72, 51% of whom were women.

The subjects underwent three test periods. During each test period, which began around noon, the

participants remained in one of three noise conditions (infrasound, placebo, traffic noise) for 72 hours

(including three nights) without leaving the laboratory. The laboratory was furnished as a bedroom with a

private bathroom. After each test period, the subjects spent 10 days in their normal environment. Neither

the subjects nor the research team knew whether the subject was being exposed to infrasound or not. The

generation of infrasound or placebo sound was controlled from a separate room by engineers who did not

meet the participants. The statistical analysis was also performed by researchers who did not know which

condition corresponded to infrasound. The exposure conditions were labelled numerically (1 vs. 2 vs. 3).

26

The subjects slept in random order under the following conditions:

1. Infrasound (test condition). During the night, the subjects were exposed to infrasound with a

frequency of 1.6–20 Hz and a maximum intensity of 90 dB (Z), simulating wind turbine infrasound

at a higher than normal level. This sound was inaudible to the subjects.

2. Placebo (negative control). No additional sounds were presented to the subjects, but a loudspeaker

was placed in the room so that the subjects did not know whether sound was being generated or

not. However, the placebo group was exposed to background noise from the laboratory air

conditioning system. The average background noise during the night was 39 dB (A), which

corresponds to a noise level of 80–85 dB (Z). The background noise was dominated by frequencies

below < 1 Hz.

3. Traffic noise (positive control). The subjects were exposed to traffic noise with an average

intensity of 40–50 dB (A) during the night and a maximum intensity of 70 dB (A). This noise was

audible.

The physiological and psychological indicators measured included objectively and subjectively measured

sleep indicators, cardiovascular indicators (24-hour blood pressure and heart rate, endothelial function and

pulse wave velocity), stress hormone and insulin levels in blood and urine, neurobehavioural and

psychological indicators, questionnaires on wind turbine syndrome symptoms and mental well-being.

Exposure to infrasound did not worsen any subjective or objective measured indicators. For some

indicators, infrasound improved measured health indicators, but according to the researchers, these

statistically significant associations could be coincidental. For example, systolic blood pressure was lower

and the Warwick–Edinburgh Mental Well-being Index showed that the subjects felt better when exposed to

infrasound compared to the placebo.

The study concluded that wind turbine infrasound does not disturb people's sleep, does not cause symptoms

of 'wind turbine syndrome', does not impair measured cardiovascular health indicators, and does not impair

people's mental well-being (Marshall et al., 2023). The results of the study can be considered well-proven.

The longest-term wind turbine infrasound experiment was conducted in Germany with 38 participants

(Ascone et al., 2021). An infrasound-generating device or a placebo device was placed in the bedrooms of

the participants for 28 nights. The participants were randomly divided into two groups: the first group (23

people) was exposed to infrasound at a frequency of 6 Hz and an intensity of 80–90 dB (Z) during the

night, while the placebo group (15 people) was not exposed to infrasound. Somatic and psychiatric

symptoms, noise sensitivity, sleep quality, cognitive ability and brain structure (using MRI) were measured

before and after the intervention. Exposure to infrasound did not affect self-reported health, sleep quality or

mental

27

abilities. Changes in brain grey matter were observed, but these cannot be interpreted as either harmful or

beneficial, and this finding may not be related to infrasound (Ascone et al., 2021).

The aim of the study by Malecki et al. (2023) was to test whether infrasound amplitude modulation (AM)

affects students' mental performance. To this end, students participating in a university experiment were

divided into three groups, which were exposed to the following sounds: 1) Recorded wind turbine noise

filtered to an infrasound intensity of 83 dB (G)/47 dB (A) and an AM depth of 4 dB at 1 Hz; 2) Synthesised

infrasound with an intensity of 78 dB (G)/46 dB (A) at 5–20 Hz and no AM; 3) Background noise with an

intensity of 63 dB (G)/43 dB (A) (traffic noise, speech in an educational institution). The results of the

study did not show significant differences in cognitive test results or in the number of reported unpleasant

sensations or complaints between different sound conditions when men and women were analysed

separately. Women reported discomfort and various complaints more than men (Małecki et al., 2023).

Between 2014 and 2015, five experiments were conducted at the University of Auckland in New Zealand

to investigate how false information about infrasound disseminated through the media and the internet

affects the onset of symptoms and the experience of disturbance. (Crichton et al., 2015, 2014b, 2014a;

Crichton and Petrie, 2015a, 2015b) (Appendix 2, Table 3).

In the experiment by Crichton et al. (2014b), 54 subjects were randomly divided into two groups. One

group was led to expect that infrasound has harmful effects based on real information circulating on the

internet, while the other group was led to expect that infrasound is not harmful. To create a high negative

health impact expectation, videos available on the internet were shown, containing descriptions of

symptoms that people associated with the operation of wind farms. To create low expectations of negative

health effects, a video was shown featuring interviews with experts who presented the scientific view that

infrasound generated by wind farms does not cause symptoms. After expectations were formed, the

subjects were exposed to 10 minutes of generated infrasound (5 Hz, 40 dB) or placebo sound. All

participants were told that they were exposed to infrasound during both 10-minute sessions. Neither the

participants nor the researcher conducting the experiment knew when the real infrasound was being played

and when the placebo sound was being played. The participants rated the presence of symptoms on a seven-

point scale before and during exposure. The symptoms assessed were those commonly associated with

wind farms on the internet: headache; pressure in the ears; ringing in the ears; itchy skin; pressure in the

sinuses; dizziness; pressure in the chest; perceived vibration; heart palpitations; nausea; fatigue; weakness.

Other random symptoms were also assessed for control purposes. Blood pressure and heart rate were also

measured. In addition, participants were asked to assess the accuracy of the following statement

28

the following statement: "I am concerned about the health effects of the noise caused by wind turbines"

both before and after watching the video. The group with high expectations of negative health effects was

significantly more concerned about health effects and reported significantly more and more intense

symptoms during the experiment compared to the pre-experiment test, regardless of actual exposure to

infrasound. More symptoms were reported that the subjects had been informed were associated with

infrasound. In the low-expectation group, the number and intensity of symptoms reported did not change

compared to the pre-test. Infra-sound exposure did not affect heart rate or blood pressure in either group.

The study shows that real information circulating on the internet about the negative health effects of infra-

sound increases the occurrence of self-reported symptoms (Crichton et al., 2014b).

Similar studies (Crichton et al., 2015, 2014a) reported a placebo effect with positive expectations. The

subjects were divided into two groups – one group was given negative expectations that wind turbines have

a negative impact on health, while the other group was given positive expectations that wind turbines have

a positive impact on health. Both groups were exposed to generated infrasound (9 Hz, 50.4 dB) and audible

sound recorded 1 km away from the wind farm (43 dB) in 7-minute sessions. In the negative expectation

group, symptom reporting increased during the session, mood deteriorated, and disturbance increased. In

the positive expectations group, symptom reporting decreased, mood improved, and distress decreased

compared to what was reported before the session.

Crichton and Petrie (2015b) used a similar study design in their study, but after the experiments with the

positive and negative expectation groups, the information given to the groups was changed and the listening

tests were repeated. When the group with negative expectations was given positive information about the

benefits of wind turbines in the repeat test, they reported fewer symptoms and their mood improved.

Similarly, those who had heard the positive information first showed a worse mood and more symptoms

after receiving the negative information. The results show that the availability of positively worded health

information can reverse or reduce the impact of negative expectations arising from media warnings about

the health risks of wind turbines (Crichton and Petrie, 2015b).

Crichton and Petrie (2015a) investigated the effectiveness of providing a nocebo explanation in changing

expectations of negative health effects in their experiment. All participants were made to expect negative

health effects. The subjects were then randomly divided into two groups: 1) subjects were given

information that the health effects of infrasound are biologically based vs. 2) subjects were told that the

health effects are the result of a nocebo effect. After receiving the negative information, the number of

reported symptoms increased and intensity in both groups compared to baseline, which

showed

29

the effectiveness of manipulation in this experiment. In the biological explanation group, the increase in

symptom reporting persisted during the second session. In the nocebo explanation group, however, the

number and intensity of symptoms decreased and mood improved during the second session. The

experiment shows that false information found in the media increases the occurrence of symptoms and

concerns about health. It also shows that providing an explanation of the nocebo effect can reduce the

reporting of symptoms associated with wind turbines. Participants in both groups found the explanation

they were given to be understandable, reasonable, convincing and correct (Crichton and Petrie, 2015a).

A similar study to Crichton's experiments was conducted by Tonin et al. (2016) using stronger infrasound

and a larger sample (72 subjects aged 17–82). Variable wind turbine infrasound was simulated at a

frequency of 0.8–40 Hz with a maximum intensity of 91 dB (Z). The subjects were exposed to it for

for 23 minutes or were given placebo noise through headphones. The results also support the existence of

the nocebo effect. In the infrasound group, the reporting of symptoms even decreased during infrasound

exposure. However, no statistically significant correlation was found between the nature of the information

provided before the listening test (expectations of health effects or expectations that there would be no

health effects) and the results. However, the result depended on the previously formed opinion about the

health effects of infrasound. Those who believed that infrasound had a negative effect on health reported

more symptoms (Tonin et al., 2016).

Rosciszewska et al. (2025) investigated the impact of wind turbine noise on cognitive performance. The

subjects were randomly divided into three groups: 1) exposure to wind turbine noise. This experiment used

real recorded wind turbine noise from a 2 MW wind turbine 500 metres away, which contained both

infrasound and audible noise. The sound pressure level used was 65.4 dB (Z) ( corresponding to a sound

level of 38.5 dB (A)). The wind turbine noise contained AM (average frequency 0.8–1 Hz, depth ~6.9 dB);

2) exposure to traffic noise. Recorded road traffic noise with an intensity of 65.4 dB (Z) (corresponding to a

sound level of 56.8 dB (A)) was used; 3) exposure to background noise only. Their study showed that

short-term exposure to wind turbine noise (at a level corresponding to the actual situation at a distance of

500 m) did not have a statistically significant effect on the cognitive performance (brain functions,

attention, thinking) of the subjects. Similarly, no statistically significant differences were found in the

results of inductive reasoning tests (accuracy, test completion time, average reaction speed) between

different noise exposure conditions (Rosciszewska et al., 2025).

The impact of visual aspects of wind turbines on human health

This study included one systematic review (Freiberg et al., 2019b) and one experiment (Murcia et al., 2017)

on visual aspects (Appendix 2, Table 1 and Table 4).

30

The health effects associated with the visual aspects of wind turbines were addressed in a systematic review

published in 2018, which included all epidemiological studies without time or language restrictions

published by 2017 (Freiberg et al., 2019b). A total of 17 studies were included in the descriptive analysis

and six studies in the meta-analysis. The quality of the studies was rated as high for five studies, acceptable

for three studies, and low for the remaining studies. The review addressed the impact of the following

visual aspects of wind turbines on sleep quality and disturbance: direct visibility from the place of

residence; altered view of the landscape; flashing lights on the blades; obstacle markings; shadow flicker;

reflections from the blades.

Disturbance caused by direct visibility, shadow flicker and flashing lights was statistically significantly

associated with an increased risk of sleep disturbance. The study found that altered views of the landscape,

obstacle markings and light reflections from wind turbine blades can also disturb people (Freiberg et al.,

2019b).

Studies have shown that when wind turbines are audible but not visible from the home, this significantly

reduces disturbance. However, when wind turbines were both visible and audible, the noise was considered

more disturbing than the visual aspects of the wind turbines. The study found that visual disturbance may or

may not depend on the distance of the wind turbines from residential buildings. Two studies showed that

disturbance from the visual aspects of wind turbines decreased with distance, while two other studies found

no significant effect of distance on visual disturbance (Freiberg et al., 2019b).

In an experiment by Murcia et al. (2017), electroencephalographic (EEG) measurements were taken to

measure both objectively and subjectively the brain's reactions to landscape visuals. EEG records the

electrical impulses generated by nerve cell activity in the brain. Sixty different images were used as stimuli.

These were divided into three groups: images with wind turbines and the same images without them;

images with a solar park and without it; and images with a nuclear power plant and without it. Both the

brain activity measurements and the questionnaire responses showed that people were not more disturbed

by images with wind turbines and solar panels. However, clear and significant differences were found when

viewing landscapes that did or did not include a nuclear power plant. The nuclear power plant evoked

negative emotions according to both the questionnaires and EEG measurements. The results of the study

may have been influenced by the fact that only 14 subjects participated in the study and most of them had a

positive attitude towards renewable energy (Murcia et al., 2017).

31

Discussion of results

Audible noise from wind turbines and sleep

Previous review studies have found a link between audible noise from wind turbines and sleep indicators

(Ellenbogen et al., 2024; Karasmanaki, 2022; Schmidt and Klokker, 2014; Teneler and Hassoy, 2023).

However, the articles included in our study do not confirm this unequivocally. Our study shows that the

link with self-reported sleep disturbances is better proven than with objectively measured sleep

disturbances (Godono et al., 2023; Liebich et al., 2021). Liebich et al. (2021) did not find an effect of noise

on objectively measured sleep parameters in their review, but found that self-reported sleep parameters may

be affected (Liebich et al., 2021). Godono et al. (2023) found a link between self-reported sleep indicators

and wind turbine noise. The cohort study by Poulsen et al. (2019b) included in our study concluded that

there is no clear link between the use of sleeping pills and wind turbine noise in the general population, but

found that long-term exposure to audible wind turbine noise increased the risk of sleep medication use in

subjects over 65 years of age, starting at a night-time noise level of 30 dB outdoors near the home. In this

regard, there was also a dose-response relationship, which strengthens the validity of this result (Poulsen et

al., 2019b). However, it was not possible to adjust the analyses in this study for people's attitudes,

knowledge or perception of disturbance, and therefore the association found between wind turbine noise

and sleep disturbance may have been influenced by factors other than noise. The study did not take into

account differences in the sound insulation of residential buildings, which affects people's actual exposure

to wind turbine noise at night.

The impact of wind turbine noise on self-reported sleep disturbance may be direct, but it may also be

related to greater disturbance, which in turn may be influenced by other factors (Teneler and Hassoy,

2023). For example, a systematic review of the impact of visual aspects (Freiberg et al., 2019b) found that

the risk of sleep disturbance (e.g. insomnia or reduced sleep quality) increased when people could see wind

turbines from their homes, were affected by shadow flicker, or saw the night-time lights on the blades of

the wind turbines.

Studies included in our work: two systematic reviews (Godono et al., 2023; Liebich et al., 2021), a cohort

study (Poulsen et al., 2019b) and two experiments (Liebich et al., 2022a, 2022b), it can be concluded that

night-time wind turbine noise of up to 30 dB (A) outdoors and up to 25 dB (A) indoors does not increase

the risk of sleep disturbance, even in noise-sensitive and older people (Liebich et al., 2022a, 2022b, 2021;

Poulsen et al., 2019b). This conclusion is based on a very

32

a small number of studies. Further research is certainly needed to investigate the relationship between sleep

indicators and audible noise from wind turbines in order to verify the validity of this claim.

Measurements taken in the yards of residential buildings located near the Saarde wind farm in Estonia (at a

distance of 1060 to 3540 m from the nearest wind turbine) showed that the audible noise at night in the

yards near the residential buildings ranged from 27.7 to 40.5 dB (A), but these measurements did not

distinguish the noise from the wind turbines from other background noise (wind noise, rain, birdsong,

traffic, etc.). The maximum night-time noise level of 40.5 dB (A) was probably influenced by the nearby

river and its dam. Measurements taken indoors showed night-time noise levels of 15.6 dB (A) to 24.1 dB

(A) (Health Board, 2025b).

Noise levels of 33.8–37.6 dB (A) were measured at night in residential areas near the Sopi-Tootsi wind

farm, which were also affected by natural background noise that cannot be distinguished from wind turbine

noise in measurements. Night-time noise levels indoors ranged from 15.3 to 18.2 dB(A) (Health Board,

2025a).

Poulsen et al. 2019b also studied the effect of low-frequency noise (10–160 Hz) on sleep disturbances. No

effect of low-frequency noise indoors on the consumption of sleeping pills was found at any of the noise

levels studied (up to 20 dB (A)) (Poulsen et al. 2019b).

Disturbance caused by wind turbines

Although disturbance cannot be considered a clinically significant health effect, it depends on people's

well-being and can therefore be considered part of the WHO definition of health, according to which health

is "a state of complete physical, mental and social well-being and not merely the absence of disease or

infirmity" (WHO, 2025). Disturbance can also act as a mediating factor between other health effects,

including influencing the development of more serious conditions such as cardiovascular disease through

stress (Basner et al., 2014; Freiberg et al., 2019a).

Several previous review studies have shown that the louder the noise from wind turbines, the more

disturbed people feel (Knopper et al., 2014; Teneler and Hassoy, 2023; van Kamp and van den Berg, 2021,

2018). A systematic review included in our study (Guski et al., 2017) also showed that the likelihood of

disturbance increases with increasing wind turbine noise, but this relationship is not as clear as in the case

of traffic noise. At the same time, the prevalence of disturbance is very uneven, and the relationships

between wind turbine distance, noise, and other indicators vary greatly between studies.

A previous review study found that wind turbine noise was more disturbing than noise from other sources

(Teneler and Hassoy, 2023). The studies included in this work did not provide a single answer to this

question.

33

question. For example, Schäffer et al. (2018) used generated

"pink noise", wind turbine noise and noise with an artificially increased low-frequency component. Pink

noise was the least disturbing, followed by wind turbine noise, and noise with an artificially increased low-

frequency component was the most disturbing (Schäffer et al., 2018). An experiment conducted in Finland

showed that the audible sounds of a wind farm were more disturbing than the sounds of the ocean (Maijala

et al., 2021). Schäffer et al. (2016) showed in their experiment that, at the same sound level, the subjects

rated wind turbine noise as more disturbing than traffic noise. This experiment simulated the noise of a 2

MW Vestas V90 wind turbine, to which sounds with generated amplitude modulation were added (Schäffer

et al., 2016). However, in an experiment by Rosciszewska et al. (2025) with recorded wind turbine noise

(from a 2 MW wind turbine 500 metres away), which also contained AM, the subjects did not perceive the

wind turbine noise as more disturbing or stressful than the recorded road traffic noise. According to the

authors, this may have been due to the fact that the subjects did not know what kind of noise they were

exposed to. The participants did not know whether or what kind of noise they were being presented with,

nor did they know the purpose of the noise presentation. In a survey conducted after the experiment, no one

identified the wind turbine noise, most described it as "some noise", and some compared it to ocean waves

or an aeroplane. In the experiment by Rosciszewska et al. (2025), both road traffic noise and wind turbine

noise were unfiltered and had the same intensity of 65.4 dB (Z). When A-weighted, the same sound level

for wind turbines is 38.5 dB (A) and for road traffic noise 56.8 dB (A). The results of the experiment by

Rosciszewska et al. (2025) support the hypothesis of Crichton et al. (2015) that the disturbance is caused

not so much by the noise or infrasound itself, but rather by the negative expectation created by the media

that wind turbine noise is disturbing and hazardous to health. Crichton et al. (2015) divided the subjects into

two groups: one group was shown a video that created the expectation that the infrasound from wind

turbines is harmful, while the other group was given the expectation that it is beneficial. Both groups were

presented with both the generated infrasound and the audible sound recorded 1 km away from the wind

farm. Disturbance was assessed before and during the listening session. In the negative expectation group,

disturbance increased during the session, while in the positive expectation group, disturbance decreased,

regardless of the noise or infrasound presented. The study shows that negative information causes

disturbance from wind farms, while positive information reduces disturbance. A review article (McCunney

et al., 2014) also concluded that wind turbine noise plays only a minor role in causing annoyance compared

to other factors that influence people's willingness to experience annoyance in relation to wind turbines.

Pohl et al. (2018) also found that noise-related annoyance was influenced to a small extent by the distance

to the nearest wind turbine and the intensity of the sound, but was most influenced by the extent to which

people felt that the wind turbine planning process had been conducted fairly and transparently.

34

According to Knopper and Ollson (2011), disturbance may be more strongly related to the visual aspects of

wind turbines and people's attitudes towards them than to the noise they generate (Knopper and Ollson,

2011). Studies have shown that if wind turbines are audible but not visible from homes, this significantly

reduces disturbance. However, when wind turbines were both visible and audible, the noise was considered

more disturbing than the visual aspects of the wind turbines (Freiberg et al., 2019b).

One of the important visual aspects affecting disturbance is the presence of shadow flicker. Shadow flicker

is not continuous, but occurs at specific times: the sun must be low enough and the wind turbine rotor must

be in the right direction for the dwelling to be in the shadow cast by the wind turbine. Shadow flicker is

disturbing, but no clinically significant health effects have been found (Freiberg et al., 2018; Knopper et al.,

2014; Teneler and Hassoy, 2023). Wind turbines should preferably be located so that shadow flicker does

not disturb residents. If this is not possible, special software can be used to assess the extent and impact of

the shadow, and mitigation measures can be implemented. As a mitigation measure, it is recommended to

set time limits on the shadow. In residential areas, shadowing should not exceed 30 hours per year or 30

minutes per day in the worst case. Wind turbines can be programmed to stop at times when shadowing

limits may be exceeded (World Bank Group, 2015).

People living near wind turbines report various health symptoms and often

symptoms of 'wind turbine syndrome'. These health symptoms described by people may be due to

disturbance, which in turn causes stress. These symptoms may not be related to wind turbine noise

(Knopper and Ollson, 2011).

An article (Pohl et al., 2018) analysed whether the complaints and disturbance experienced by people living

near wind turbines are directly related to wind turbine noise or whether other factors are significant. The

conditions under which disturbance occurs were also investigated. People living near a wind farm (1.25 to

2.89 km from the nearest wind turbine) in Germany were surveyed in 2012 and 2014. A total of 212 people

participated in the first year and 133 in the second year. The subjects were exposed to wind turbine noise

near their homes outdoors at 25–30 dB(A) or 30–35 dB(A), which came from Enercon E-82 wind turbines

that were 150 m high and had a capacity of 2 MW.

Of all residents, 69.3% heard the noise from the wind turbines and 30.7% did not. Nearly half (53.6%) of

the subjects experienced disturbing noise once a week, 20.9% once a month and 13.6% almost every day.

18.4% were not disturbed at all. Only a small percentage of residents reported being severely disturbed by

the noise from the wind turbines, which decreased over time: in 2012, one tenth (9.9%) of residents were

severely disturbed, but two years later, only 6.8% were. It is noteworthy that the majority of the residents

who were severely disturbed

35

residents (75.0%) had already been opposed to the wind farm before it was built, either passively or

actively. They also felt that the wind turbine planning process was unfair and had little knowledge of how

to improve their situation themselves. At the time of the survey (three years after the wind turbines were

erected), there were slightly more supporters of the wind farm (40.2%) than opponents (35.8%) living in the

vicinity of the wind farm. Only a small proportion, 16.7%, were ambivalent, and 7.4% had no opinion

about the wind farm, but opponents were more active than supporters (Pohl et al., 2018).

In the study by Pohl et al. (2018), the participants were asked to observe when wind turbine noise disturbed

them the most and to record the disturbing noise. Disturbing noise occurred most frequently in the evenings

(33.6%) and at night (18.2%). The disturbance was most prevalent when people were sleeping (30.0%) or

resting (24.5%). This caused anger in 39.1% of cases. The disturbance occurred most often when there was

a westerly wind (68.2%) and in humid weather (30.9%). The most frequently used measures to reduce the

impact of noise were talking to family members, friends and neighbours (32.1%), closing windows

(25.9%), changing location (indoors 11.8% and outdoors 7.1%) and turning up the volume of the

radio/television (7.5%). The disturbing noise from wind turbines was mainly described as irregular and

fluctuating in volume. Of those disturbed, 71.6% described it as a pulsating hiss. Complaints were not

caused by the absolute loudness of the noise, but by the variation in sound intensity, i.e. amplitude

modulation (Pohl et al., 2018).

In a study by Pohl et al. (2018), more residents complained of physical and psychological symptoms caused

by traffic noise (16%) than by wind turbine noise (10%, 7% two years later). Both noise sources caused

similar symptoms: reduced work capacity and concentration, increased irritability/anger, negative mood

and disturbed sleep.

An important factor influencing disturbance is also the financial benefit derived from the wind farm (Taylor

and Klenk, 2019; van Kamp and van den Berg, 2018). Farmers who benefited financially from wind

turbines were very little disturbed by them and reported fewer health and sleep problems than the rest of the

study participants, even though they lived closer to the turbines and were exposed to higher noise levels

than the other respondents. However, economic benefits may not be the only factor in reducing disturbance.

Different attitudes, education and greater control over the location of wind turbines may also have played a

role (van Kamp and van den Berg, 2018).

Most of the experiments included in this study showed that disturbance is greater when the audible noise

from wind turbines contains amplitude modulation (AM). The deeper the AM, the greater the disturbance

(Ioannidou et al., 2016; Lee et al., 2011; Maijala et al., 2021; Schäffer et al.,

36

2018, 2016). Australian researchers also found in their review article that AM is an important characteristic

of wind turbine noise that can increase disturbance and recommended testing AM measurements and

setting limits for AM in Australia to reduce disturbance (Davy et al., 2020).

In summary, the relationship between wind turbines and disturbance depends on several factors, such as

expectations/knowledge of the health effects of wind turbines, perceived fairness and transparency of the

planning process, economic benefits, visual aspects and noise. It is likely that a combination of all these

factors causes annoyance, and reducing just one factor (e.g. noise) may not reduce annoyance.

Regardless of the reasons, a certain degree of disturbance among the population can be expected, as with

any other project involving changes to the local environment. The acceptable level of disturbance is a

political decision that should be made by weighing the benefits of wind energy against its negative effects

(Knopper and Ollson, 2011).

The impact of infrasound from wind turbines on health

Wind turbine noise always includes infrasound. Opinions circulating on the internet and in public

discussions suggest that this component of wind turbine noise may be harmful to health. We have compiled

a list of experiments (Appendix 2, Table 3) that have experimentally investigated the health effects of the

infrasound component of wind turbines, either separately or in combination with the audible noise of wind

turbines. The claims circulating on the internet and in the media have been the motivation for conducting

these experiments. Therefore, the study focused mainly on health indicators that are the subject of

widespread speculation: various symptoms associated with infrasound (e.g. pressure in the ears, headaches,

etc.), disturbance, sleep disorders and effects on cognitive ability. The experiments included in our study

showed that exposure to infrasound at any of the sound pressure levels studied (up to 91 dB (Z)) did not

impair any of the health indicators measured in the studies. No effects were found on sleep disturbances

and sleep quality (Ascone et al., 2021; Liebich et al., 2022a, 2022b; Marshall et al., 2023). No effect on

disturbance and mental well-being was found (Crichton et al., 2015; Maijala et al., 2021; Marshall et al.,

2023; Rosciszewska et al., 2025). No effect was found on cognitive performance (Ascone et al., 2021;

Małecki et al., 2023; Rosciszewska et al., 2025). No effect was found on cardiovascular parameters (heart

rate, blood pressure) (Maijala et al., 2021; Marshall et al., 2023). No effect was found on the occurrence of

symptoms that people themselves associate with wind turbine infrasound (Crichton et al., 2014b, 2014a;

Crichton and Petrie, 2015b, 2015a; Maijala et al., 2021; Małecki et al., 2023; Marshall et al., 2023). A

previous review of infrasound and low-frequency sound (van

37

Kamp and van den Berg, 2018) found that there is no evidence of a specific health impact from the

infrasound component of wind turbines.

According to measurements taken in Estonia, the unfiltered infrasound intensity at a frequency of 6.3 Hz

measured at night in the interiors of four residential buildings near the Saarde wind farm (2070 to 2530 m

from the nearest wind turbine) was 35–50 dB (Z) (Health Board, 2025b) and in the vicinity of the Sopi-

Tootsi wind farm (1400 to 2560 m from the nearest wind turbine) 37.7 to 58.2 dB (Z) (Health Board,

2025a). Ascone et al., 2021 did not find any significant effect in a long-term experiment (28 consecutive

nights) with a similar frequency (6 Hz) and intensity (80–90 dB (Z)) of infrasound. A more realistic wind

turbine infrasound level was used in the experiment (Crichton et al., 2014b). The subjects were exposed to

generated infrasound at a frequency of 5 Hz and an intensity of 40 dB for 10 minutes. Exposure to

infrasound did not affect symptoms such as headache, pressure in the ears, blood pressure and heart rate.

However, the same experiment found a significant effect on the occurrence of self-reported symptoms

based on information provided to the subjects prior to the study. The subjects who were shown a video

about the negative health effects of infrasound, based on real information circulating on the internet, began

to report the symptoms mentioned in the video, regardless of whether or not they had been exposed to

infrasound (Crichton et al., 2014b). It should also be noted that the measurements taken in Estonia do not

distinguish between infrasound from wind turbines and other possible sources of infrasound in the home

(fans, air heat pumps, etc.), which may have increased the measured infrasound level, and the actual level

of infrasound generated by wind turbines may be lower than measured in homes located near the Saarde

and Sopi-Tootsi wind farms.

Crichton's studies (Crichton et al., 2015, 2014b, 2014a; Crichton and Petrie, 2015b, 2015a) clearly show

that expectations can influence the reporting of symptoms and mood in both positive and negative

directions. Although actual exposure (e.g., to infrasound) may be harmless, the expectation or belief that it

is harmful causes people to experience real symptoms. This phenomenon is called the nocebo effect. There

is a lot of misinformation on the internet about the health effects of infrasound, and it is not possible to

correct or change this information. Crichton's research shows that explaining the nature of the nocebo effect

to people or providing them with positive information about wind turbines as a counterbalance reduces

disturbance and the onset of symptoms.

As no effects have been observed to date even at much higher infrasound levels, which are commonly

found in homes located near wind farms, there is no reason to assume, based on this study, that infrasound

from wind turbines that complies with the limits in force in Estonia

38

affect human health. Based on this study, there is also no reason to recommend changes to the current

infrasound limits.

Infrasound is common in our environment. For example, a study by Staniek et al. (2013) showed that

infrasound from the ventilation shaft of an operating coal mine was stronger than that from a wind farm 750

m away from the nearest wind turbine (Staniek et al., 2013). Infra-sound from large wind turbines (with a

capacity of over 2 MW) can range from 59 to 107 dB (G) at distances of 68 to 1000 metres. Similar

infrasound levels can also be found 350 metres from a gas-fired power plant (74 dB (G)), 70 metres from

major roads (76 dB (G)) and 25 metres from the coastline (75 dB (G)) (Schmidt and Klokker, 2014).

Claims about the health effects of infrasound are circulating in public debates and the media

One of the most common theories circulating on the internet is that of Nina Pierpont. According to Pierpont

(2009), infrasound can reach the inner ear and stimulate the balance organs (vestibular organs), especially

the otoliths – sensory structures that respond to movement and gravity. According to her, this unusual or

constant stimulation can cause symptoms such as dizziness, balance disorders, nausea and nystagmus

(involuntary eye movement). Nina Pierpont has coined the term 'wind turbine syndrome'. This is not a

medical diagnosis, but a term coined by Nina Pierpont to summarise people's complaints associated with

wind turbines (Pierpont, 2009). According to a review study (Schmidt and Klokker, 2014), Pierpont's

(Pierpont, 2009) source is a case series study, which is well suited for proposing hypotheses, but

methodologically, this approach does not allow for the identification of a causal link between the

influencing factor under investigation and the health outcome. Harrison (2015) has discussed whether the

effect of infrasound generated by wind turbines on the vestibular organs of the inner ear could be

biologically possible at all. The study concludes that acoustic activation of the vestibular system is possible

from a sound level of 110 dB (Z) and, based on animal experiments, from 120 dB

(Z). Comparing this with the actual infrasound levels in residential areas (approximately 60 dB (Z)), the

study finds that the impact of low-frequency and infrasound from wind turbines on the balance organs is

not biologically justified (Harrison, 2015).

Measurements taken inside Estonian homes (Health Board, 2025a, 2025b) show that the strongest

uncorrected wind turbine noise levels are in the 0.8–2 Hz range, mostly reaching 50 to 66 dB (Z). At higher

frequencies, the sound intensity decreases, as can be seen in Figure 4. The figure shows, for example, the

results of noise level measurements at all frequencies, both at night and during the day, in the residential

building closest (1400 m) to the Sopi-Tootsi wind farm.

39

Figure 4. Results of uncorrected sound level measurements inside a residential building 1400 m from the

Sopi-Tootsi wind farm (Health Board, 2025a).

The occurrence of wind turbine syndrome symptoms (e.g. headache, fatigue, dizziness, nausea, pressure in

the ears, etc.) in contact with infrasound was tested experimentally in studies (Crichton et al., 2014b,

2014a; Crichton and Petrie, 2015b, 2015a; Maijala et al., 2021; Małecki et al., 2023; Marshall et al., 2023).

None of these experiments found that exposure to infrasound caused the symptoms mentioned or any

symptoms at all (Appendix 2, Table 3).

Public discussions often refer to the hypothesis of a Portuguese research group that high levels of

infrasound and low-frequency sound cause “vibroacoustic disease” (VAD) (Alves-Pereira and Castelo-

Branco, 2007; Castelo-Branco and Alves-Pereira, 2004). Alves-Pereira and Castelo-Branco (2007) and

Castelo-Branco and Alves-Pereira (2004) have argued that VAD occurs in people who work in places with

high levels of infrasound and low-frequency sound, specifically in aeroplanes, trains, bars, discotheques,

underground railways and ordinary cars. They describe vibroacoustic disease as a whole-body pathology

that can manifest itself in a wide variety of diseases and symptoms, such as respiratory, digestive, nervous

system and cardiovascular diseases and symptoms, cancer, autoimmune diseases and endocrine disorders,

but its root cause is the abnormal proliferation of collagen and elastin in the intercellular matrix. VAD is

also not a recognised medical diagnosis (van Kamp and van den Berg, 2018). VAD is diagnosed and

discussed only by a small group of researchers who publish mostly in peer-reviewed journals, mainly

referencing each other (Chapman

40

and George, 2013). Most of the sources on which this concept is based are conference presentations and

very old, unreviewed publications (dating back to 1928), some of which are in Russian. VAD as a diagnosis

has so far remained a theoretical hypothesis that has not been confirmed by other researchers.

To our knowledge, no scientific research has been published in a peer-reviewed scientific journal

investigating the connection between infrasound or low-frequency sound generated by wind turbines and

VAD. To our knowledge, the main authors of this theory, Mariana Alves-Pereira and A.A Nuno Castelo-

Branco, have not published any articles in peer-reviewed scientific journals on the relationship between

infrasound or low-frequency sound generated by wind turbines and health. Therefore, there is no

scientifically accepted confirmation of the claims circulating among the population that Mariana Alves-

Pereira's research has shown that infrasound and/or low-frequency sound from wind turbines have

significant health effects.

Audible noise generated by wind turbines and clinically manifested health effects Disturbance and sleep disturbances caused by audible noise may contribute to the development of

diagnosable diseases (Basner et al., 2014). This has already been demonstrated in the case of traffic noise

(van Kempen et al., 2018). In the case of traffic noise, the best-documented link is between noise from roads

(trucks, motorcycles, trams, cars, etc.) and cardiovascular disease (van Kempen et al., 2018).

Similar links have not yet been found for the audible noise from wind turbines, but there are indications that

such links may exist. The results of Bräuner et al. (2019b) showed that night-time, daytime and evening

average wind turbine noise on the exterior facade of a dwelling Lnight, Lday, and Leavning above 20 dB (A) may

cause an increase in atrial fibrillation, but no similar association was found with the 24-hour indicator Lden

( Bräuner et al., 2019b). Studies with a cohort of Danish nurses found no association between wind turbine

noise and stroke and heart attack in women over 44 years of age (Bräuner et al., 2019a, 2018), but a cohort

covering the entire Danish population (Poulsen et al., 2019a) yielded conflicting results, making it difficult

to draw specific conclusions. References to a possible link between audible noise from wind turbines and

heart attacks and strokes were also found by Poulsen et al. (2018d), who investigated the effect of short-

term exposure to higher noise levels on the incidence of heart attacks and strokes (Poulsen et al., 2018d).

In a cohort covering the entire Danish population, it was found that greater long-term night-time exposure

to audible noise from wind farms (at frequencies between 10 and 10,000 Hz) near residential areas

increases the risk of taking sleeping pills and antidepressants, particularly among older people, starting at

noise levels of 30

41

dB (A) (Poulsen et al., 2019b). However, as this association was only addressed in one study, the level of

evidence for this finding cannot be considered high.

Compared to traffic noise levels, the noise levels in these studies conducted in Denmark are very low.

Traffic noise from motorways has been found to have an impact on cardiovascular disease from noise

levels of Lden 53 dB (A) (WHO, 2019). Most of the participants in the Danish nationwide cohort (79%) lived

in dwellings where the wind turbine noise level outdoors near their homes was below 24 dB(A). Although

the study included more than 700,000 people, few people were exposed to noise levels above 42 dB, and

the number of cases (47 heart attacks and 23 strokes) may have been too small to find statistically

significant associations (Poulsen et al., 2019a). Most wind turbines in Denmark are less than 100 m high,

but there are also taller ones. In the case of lower wind turbines (less than 35 m high), people in Denmark

live closer than 500 m to the wind turbines. Those who lived closest to the lowest wind turbines ( less than

35 m high) (closer than 500 m) made up the majority of the group exposed to the highest noise levels

(above 42 dB (A)). People living near taller wind turbines were mostly exposed to lower noise levels

(Poulsen et al., 2019a).

The Danish studies used noise levels that were not measured but modelled. The noise level was modelled

taking into account the noise emission of the wind turbine, weather conditions, the distance of the residence

under study and other relevant information, but it is still a calculated indicator. At the same time, this

indicator is more accurate than measured noise levels, as it is not possible to distinguish wind turbine noise

from other noise sources (background noise) such as normal natural sounds (wind noise, rain, birdsong),

traffic noise or noise caused by people themselves. The modelled noise level only shows the noise level

originating from wind turbines ( Lahti, 2010).

There is never absolute silence in the environment surrounding humans. Background noise itself can often

be close to 40 dB (A). Background noise is caused, for example, by household appliances, ventilation

systems, air conditioners and other technical equipment, natural sounds and, in urban environments, traffic

noise. In the tests included in this study, the background noise in soundproof laboratories was 19–23 dB

(A) in two tests by Liebich (Liebich et al., 2022a, 2022b) and 39 dB (A)/80–85 dB (Z) in the test by

Marshall et al. (2023), where it originated from air conditioning. Wind turbine noise was presented in the

test by Marshall et al. (2023) under controlled conditions in a soundproof laboratory, but the air

conditioning could not be turned off because otherwise the subjects would have been affected by unusual

temperatures. The background noise in the Malecki et al. (2023) experiment was 43 dB (A). This

experiment took place in a school building during class time.

42

The Danish studies were adjusted for many important characteristics, but it is still not possible to adjust the

analyses in such register-based studies for people's attitudes, negative expectations, political sense of justice

and other factors causing disturbance, as such data are not recorded in the registers. Cardiovascular disease,

sleep disorders and depression can all be affected by wind farm disturbance, which in turn can be caused by

a wide variety of factors (see chapter on Audible noise and disturbance from wind turbines).

One experiment (Chiu et al., 2021) also suggests that cardiovascular diseases may be affected by audible

noise from wind turbines. This experiment showed that the louder the noise from the wind turbines, the

lower the heart rate variability of the subjects. Heart rate variability (HRV) is a measure that describes

changes in the time interval between heartbeats. HRV reflects the activity and balance of the autonomic

nervous system. Higher HRV usually indicates better physical adaptability and stress tolerance, while lower

HRV may indicate stress, fatigue or health problems. In this study, the subjects spent 30 minutes at a

distance of 20 m from the nearest wind turbine, and the low-frequency noise (20-200 Hz) was 38.3-57.1

dB(A). Thus, this study shows that higher than normal wind turbine noise may increase the risk of

cardiovascular disease. As the study was conducted outdoors and indoors, rather than in a controlled

laboratory setting, other factors, such as noise from other sources, also influenced the results. The study

also did not take into account possible psychological stress and air pollution, which are also known to affect

HRV.

Long-term, large-scale, registry-based cohort studies conducted in Denmark showed that wind turbine noise

at none of the levels studied had an impact on the development of diabetes and high blood pressure, nor did

it worsen birth outcomes (Poulsen et al., 2018a, 2018b, 2018c).

In summary, no clear links have been found to date between audible wind turbine noise and clinically

significant health effects, but there are indications that audible noise may increase the risk of cardiovascular

disease and depression. However, too few long-term, high-quality follow-up studies have been conducted

to date to draw firm conclusions. Finding clear dose-response relationships is also limited by the fact that

there are no people who have long-term exposure to high wind turbine noise (above 50 dB (A)), as wind

turbines are not allowed to be built so close to residential areas that such noise exposure could occur.

Based on a cohort study conducted in Denmark, low-frequency noise indoors (10–160 Hz) was not

associated with heart attacks, strokes, diabetes, poor birth outcomes, high blood pressure,

43

depression and sleep disorders at any of the noise levels studied (up to 20 dB (A)) (Poulsen et al., 2019a,

2019b, 2018a, 2018b, 2018c).

Electromagnetic fields and vibrations caused by wind turbines

We did not find any scientific articles that met the inclusion criteria for this study that examined the health

effects of electromagnetic fields (EMF) or vibration caused by wind turbines. However, these factors have

been analysed by Knopper et al. (2014) and van Kamp and van den Berg (2018) in their review studies.

An electromagnetic field (EMF) is a physical field created by electric charges and electric currents and

consists of two interrelated components: an electric field and a magnetic field. An electric field occurs

when there is a difference in voltage (e.g. in an electrical conductor, battery or power line). A magnetic

field occurs when an electric current flows (e.g. when you switch on a device or an electric motor is

running) (WHO, 2016). EMF is present everywhere in our environment. Electric fields are created when

electrical charges accumulate in the atmosphere during thunderstorms, and the Earth's magnetic field causes

a compass needle to point north-south (WHO, 2016). In addition to natural sources, we are also surrounded

by man-made fields: X-rays, low-frequency EMFs associated with the flow of electricity from electrical

outlets, and higher-frequency radio waves (WHO, 2016).

Very high-intensity EMFs have been found to have both short-term direct and long-term health effects,

which is why Estonia, like many other countries, has established requirements and limits for

electromagnetic fields in the working and living environment (Regulation No. 38 of the Minister of Social

Affairs, 2002; Regulation No. 44 of the Government of the Republic, 2016). However, the field strengths of

most EMF sources (power lines, microwave ovens, mobile phones, etc.) are low, they are located in

households and workplaces at a sufficient distance from people or are encountered for short periods of time,

and therefore do not usually pose a health risk (WHO, 2016).

In 2011, the International Agency for Research on Cancer classified radiofrequency EMFs as a Group 2B

possible carcinogen for humans. This category is used when a causal relationship is considered possible but

there is no evidence from human studies (IARC, 2011). Some studies have found that low-frequency

EMFs, which come mainly from power lines, may increase the risk of leukaemia and brain and breast

tumours (Carpenter, 2019). However, the results of most studies are contradictory, mostly showing no

effect, and there is currently no conclusive evidence of the long-term health effects of EMFs (Bodewein et

al., 2019).

44

A 2014 review by Knopper et al. showed that EMF levels measured 2–3 metres from the base of a wind

turbine are similar to or lower than those generated by many common household appliances (e.g.

refrigerator, dishwasher, microwave oven, hair dryer) and significantly below the applicable regulatory

limits. Therefore, health effects from wind turbine electromagnetic fields are highly unlikely (Knopper et

al., 2014).

Vibration is the oscillation of a solid body (Regulation No. 78 of the Minister of Social Affairs, 2002). The

health effects of vibration are particularly evident in work environments where people are exposed to the

vibration of tools such as jackhammers or large machines such as tractors (Health Board, 2025c). In the

living environment, vibration can be caused by traffic (e.g. trains, heavy goods vehicles), which can cause

windows to rattle and cracks to appear in buildings. Limit values for vibration have also been established

for residential buildings (Regulation No. 78 of the Minister of Social Affairs, 2002).

Vibration measurements in the vicinity of wind turbines (less than 300 m from the turbines) have yielded

values that are close to zero (Ministry of Climate, 2025; Knopper et al., 2014; van Kamp and van den Berg,

2018). Therefore, the vibration caused by wind turbines is not a significant safety risk to people living near

wind turbines.

Justification for the choice of methodology, strengths and weaknesses of the study

We included only peer-reviewed scientific articles published in journals whose design allowed for the

identification of causal relationships. Randomised controlled trials are the most reliable. Therefore, we also

included all experiments that met the criteria in the study. However, it is not possible to conduct such

experiments to study long-term exposure. In order to study the long-term effects of an environmental

factor, long-term follow-up studies must be conducted. Such studies are cohort studies, which we also

included in our review. We also included systematic reviews and meta-analyses, which are the highest

quality type of studies. A systematic review is a synthesis of existing evidence that uses a clear, transparent

and systematic methodology to find, evaluate and present relevant evidence. We did not include narrative

reviews because they are subject to a high risk of error due to their subjective nature. We also did not

include cross-sectional studies because it is not possible to identify a causal relationship in these studies.

Cross-sectional studies examine the presence of an exposure (e.g. noise) and an outcome (e.g. health

problem) at the same point in time (e.g. during a survey). Therefore, it is not possible to know which came

first, the health problem or the influencing factor. For an influencing factor (e.g. noise) to cause a health

problem, it must be present before the health problem develops. This approach is possible with a cohort

study.

45

We only included scientific literature published in English in the study. In our opinion, this does not cause a

significant bias in our results, as English is currently the main language of science, in which all important

scientific results are published. It is unlikely that any very important results have been published only in

other languages. Furthermore, it is not reasonable to include literature published in all languages, as the

research team does not speak all languages and including studies in all languages would carry a high risk of

misinterpreting the results of the articles.

46

Conclusions Based on the scientific studies found in the systematic literature review, we can draw the following

conclusions:

• Scientific studies have not shown any negative health effects from electromagnetic fields,

vibrations and infrasound generated by wind turbines that remain within the limits. The evidence

for the absence of health effects from infrasound is based on thirteen experiments included in this

review.

• Based on this study, low-frequency noise from wind turbines indoors (10–160 Hz) was not

associated with heart attacks, strokes, diabetes, poor birth outcomes, high blood pressure,

depression or sleep disturbances at any of the noise levels studied (up to 20 dB A). As only one

study examined the relationship between noise exposure and each health outcome, the level of

evidence for this finding cannot be considered high.

• Long-term, large-scale, register-based studies conducted in Denmark found that audible noise from

wind turbines (including low-frequency noise) did not affect the development of diabetes and high

blood pressure, nor did it worsen birth outcomes (birth weight and premature birth) at any of the

noise levels studied (up to 50 dB (A)). As only one study addressed the relationship between noise

exposure and each health outcome, the level of evidence for this finding cannot be considered high.

• To date, no clear links have been found between audible wind turbine noise (including low-

frequency noise) and clinically significant health effects. However, there are indications that

audible noise may increase the risk of certain cardiovascular diseases (atrial fibrillation, heart

attack and stroke), increase the incidence of depression and impair sleep quality. To date, only two

large cohort studies have been conducted in Denmark, which means that the level of evidence for

these findings cannot be considered high. More long-term follow-up studies are needed to identify

health effects more precisely.

• There is clear evidence that wind turbines cause disturbance to residents. In addition to wind

turbine noise, other factors also contribute to the disturbance. Expectations/knowledge about the

health effects of wind turbines obtained from the media, the perceived fairness and transparency of

the planning process, economic benefits and visual aspects all play an important role. It is likely

that a combination of all these factors causes disturbance.

• Based on listening tests, audible wind turbine sounds with greater amplitude modulation depth (changes in sound levels) are more disturbing than sounds with lower

47

amplitude modulation depth. The greater amplitude modulation depth associated with audible wind

turbine noise may increase human disturbance from wind farms.

48

Recommendations Currently, the limit values for industrial noise (including wind turbines) in residential areas are 60 dB (A) during the day and 45 dB

(A) in residential areas. The target noise levels are 50 dB (A) during the day and 40 dB (A) at night

(Regulation No. 71 of the Minister of the Environment, 2016). Indoors, traffic noise levels may be up to 30

dB (A) during the day and noise from technical equipment up to 25 dB (A) at night (Regulation No. 42 of

the Minister of Social Affairs, 2002). The guidelines for assessing the environmental impact of wind farms

recommend using the strictest value for wind turbines in outdoor conditions, i.e. the noise from wind

turbines should not exceed 40 dB (A) at night in the vicinity of residential buildings (Ministry of Climate,

2025).

Based on our study, there is currently no reason to recommend a stricter limit value for residential areas

than that specified in the guidelines for assessing the environmental impact of wind farms (Ministry of

Climate, 2025). Currently, the regulation of wind farm noise in Estonia is confusing due to the difference

between limit values and target values and the lack of a specific limit value for wind turbines. We

recommend that wind turbine noise be regulated more clearly in legislation. Based on the studies included

in this review, we recommend establishing a limit value based on current knowledge so that wind turbine

noise in the immediate vicinity of residential buildings outdoors at night (23:00–07:00) does not exceed 40

dB (A), as already recommended in the guidelines for assessing the environmental impact of wind farms.

For living and sleeping areas, we recommend setting a limit for wind turbine noise of 30 dB(A) during the

day and 25 dB(A) at night, similar to the existing limits for traffic noise and noise from technical

equipment.

Based on this study, there is no reason to recommend stricter limits for infrasound, which are established in

Regulation No. 75 of the Minister of Social Affairs (2002), and low-frequency sound, which are established

in Regulation No. 42 of the Minister of Social Affairs (2002).

We recommend continuing with noise and infrasound measurements at wind farms in Estonia and with

studies on health effects and perceived disturbance. If new significant scientific research becomes available

after five years, we recommend reassessing the appropriateness of the limit values.

This study found that audible wind turbine sounds with greater amplitude modulation (AM) depth are more

disturbing than sounds with lower AM depth, based on listening tests, and that perceptible AM may

increase people's disturbance from wind farms. We recommend that developers and researchers explore

ways to reduce AM depth in order to reduce the annoyance of wind turbine noise. The more uniform the

noise level of wind turbines, the less disturbing it is. When measuring and modelling the noise generated by

wind farms, both the average and maximum noise levels should be highlighted separately, as well as the

AM depth, if possible.

49

The results of our study show that several factors other than wind turbine noise affect disturbance, and that

noise reduction alone may not be sufficient to mitigate disturbance. Just as important as noise restrictions in

preventing disturbance may be informing residents about the nocebo effect, the absence of negative

expectations regarding the health effects of wind turbines, and understanding the positive characteristics of

wind turbines (Crichton et al., 2015, 2014b, 2014a; Crichton and Petrie, 2015b, 2015a; Tonin et al., 2016).

The perceived openness and fairness of political processes (Pohl et al., 2018), economic benefits

(McCunney et al., 2014) and visual aspects (Freiberg et al., 2019b) also influence the emergence of wind

turbine-related disturbance. To prevent and mitigate disturbance, it is important to provide communities

with objective scientific information to counteract the impact of misinformation about infrasound

circulating on the internet. Positive experiences have been gained from early and informal involvement of

residents in wind turbine planning processes. It is also important that residents feel that their concerns are

being taken into account and that mitigation measures are being implemented even when everything

complies with the standards (e.g. the modelled noise level is 39.9 dB).

An important aspect of visual disturbance that can be reduced is shadowing. The relevant mitigation

measures have already been described in the guidelines for assessing the environmental impact of wind

farms (Ministry of Climate, 2025) and must always be implemented.

Based on this study, there is no clear basis for recommending a minimum distance from the nearest wind

turbine to residential buildings, as wind turbines can vary greatly in terms of noise emissions and other

factors that cause disturbance.

Existing studies show that wind turbines can be located safely for human health if they are placed in

accordance with the noise standards recommended in this study, if planning is carried out transparently,

taking into account the interests of the community and offering benefits to them, and if visual pollution is

minimised. It is important to note that the use of wind energy helps to mitigate climate change and reduces

health problems caused by air pollution. For example, in 2020, more than 1,000 people in Estonia died

prematurely due to air pollution (Orru et al., 2022).

50

References

Alves-Pereira, M., Castelo-Branco, N.A.A., 2007. Vibroacoustic disease: Biological effects of infrasound and low- frequency noise explained by mechanotransduction cellular signalling. Prog Biophys Mol Biol 93, 256–279. https://doi.org/10.1016/j.pbiomolbio.2006.07.011

Ascone, L., Kling, C., Wieczorek, J., Koch, C., Kühn, S., 2021. A longitudinal, randomised experimental pilot study to investigate the effects of airborne infrasound on human mental health, cognition, and brain structure. Sci Rep 11. https://doi.org/10.1038/S41598-021-82203-6

Basner, M., Babisch, W., Davis, A., Brink, M., Clark, C., Janssen, S., Stansfeld, S., 2014. Auditory and non-auditory effects of noise on health. Lancet 383, 1325–1332. https://doi.org/10.1016/S0140-6736(13)61613-X

Basner, M., McGuire, S., 2018. WHO Environmental Noise Guidelines for the European Region: A Systematic Review on Environmental Noise and Effects on Sleep. Int J Environ Res Public Health

Bodewein, L., Schmiedchen, K., Dechent, D., Stunder, D., Graefrath, D., Winter, L., Kraus, T., Driessen, S., 2019. Systematic review on the biological effects of electric, magnetic and electromagnetic fields in the intermediate frequency range (300 Hz to 1 MHz). Environ Res 171, 247–259. https://doi.org/10.1016/J.ENVRES.2019.01.015

Bräuner, E. V., Jørgensen, J.T., Duun-Henriksen, A.K., Backalarz, C., Laursen, J.E., Pedersen, T.H., Simonsen, M.K., Andersen, Z.J., 2019a. Association Between Long-Term Exposure to Wind Turbine Noise and the Risk of Stroke: Data From the Danish Nurse Cohort. J Am Heart Assoc 8. https://doi.org/10.1161/JAHA.119.013157

Bräuner, E. V., Jørgensen, J.T., Duun-Henriksen, A.K., Backalarz, C., Laursen, J.E., Pedersen, T.H., Simonsen, M.K., Andersen, Z.J., 2019b. Long-term wind turbine noise exposure and the risk of incident atrial fibrillation in the Danish Nurse cohort. Environ Int 130. https://doi.org/10.1016/J.ENVINT.2019.104915

Bräuner, E. V., Jørgensen, J.T., Duun-Henriksen, A.K., Backalarz, C., Laursen, J.E., Pedersen, T.H., Simonsen, M.K., Andersen, Z.J., 2018. Long-term wind turbine noise exposure and incidence of myocardial infarction in the Danish nurse cohort. Environ Int 121, 794–802. https://doi.org/10.1016/J.ENVINT.2018.10.011

Carpenter, D.O., 2019. Extremely low frequency electromagnetic fields and cancer: How source of funding affects results. Environ Res 178. https://doi.org/10.1016/J.ENVRES.2019.108688

Castelo-Branco, N.A.A., Alves-Pereira, M., 2004. Vibroacoustic disease. Noise Health 6(23), 3–20.

Chapman, S., George, A.S., 2013. How the factoid of wind turbines causing ‘vibroacoustic disease’ came to be ‘irrefutably demonstrated.’ Aust N Z J Public Health 37, 244–249. https://doi.org/10.1111/1753- 6405.12066

Chiu, C.H., Lung, S.C.C., Chen, N., Hwang, J.S., Tsou, M.C.M., 2021. Effects of low-frequency noise from wind turbines on heart rate variability in healthy individuals. Sci Rep 11. https://doi.org/10.1038/S41598-021- 97107-8

51

Clark, C., Paunovic, K., 2018. WHO Environmental Noise Guidelines for the European Region: A Systematic Review on Environmental Noise and Quality of Life, Wellbeing and Mental Health. Int J Environ Res Public Health 15. https://doi.org/10.3390/IJERPH15112400

Crichton, F., Dodd, G., Schmid, G., Gamble, G., Cundy, T., Petrie, K.J., 2014a. The power of positive and negative expectations to influence reported symptoms and mood during exposure to wind farm sound. Health Psychol 33, 1588–1592. https://doi.org/10.1037/HEA0000037

Crichton, F., Dodd, G., Schmid, G., Gamble, G., Petrie, K.J., 2014b. Can expectations produce symptoms from infrasound associated with wind turbines? Health Psychol 33, 360–364. https://doi.org/10.1037/A0031760

Crichton, F., Dodd, G., Schmid, G., Petrie, K.J., 2015. Framing sound: Using expectations to reduce environmental noise annoyance. Environ Res 142, 609–614. https://doi.org/10.1016/J.ENVRES.2015.08.016

Crichton, F., Petrie, K.J., 2015a. Health complaints and wind turbines: The efficacy of explaining the nocebo response to reduce symptom reporting. Environ Res 140, 449–455. https://doi.org/10.1016/J.ENVRES.2015.04.016

Crichton, F., Petrie, K.J., 2015b. Accentuate the positive: Counteracting psychogenic responses to media health messages in the age of the Internet. J Psychosom Res 79, 185–189. https://doi.org/10.1016/J.JPSYCHORES.2015.04.014

Davy, J.L., Burgemeister, K., Hillman, D., Carlile, S., 2020. A Review of the Potential Impacts of Wind Turbine Noise in the Australian Context. Acoust Aust 48, 181–197. https://doi.org/10.1007/S40857-020-00192-4/FIGURES/1

Ellenbogen, J.M., Kellam, C.B., Hankard, M., 2024. Noise-induced sleep disruption from wind turbines: scientific updates and acoustical standards. Sleep 47. https://doi.org/10.1093/SLEEP/ZSAD286

European Commission, 2019. The European Green Deal - European Commission [WWW Document]. URL https://commission.europa.eu/strategy-and-policy/priorities-2019-2024/european-green-deal_en (accessed 7.23.25).

Freiberg, A., Schefter, C., Girbig, M., Murta, V.C., Seidler, A., 2019a. Health effects of wind turbines on humans in residential settings: Results of a scoping review. Environ Res 169, 446–463. https://doi.org/10.1016/J.ENVRES.2018.11.032

Freiberg, A., Schefter, C., Girbig, M., Murta, V.C., Seidler, A., 2018. Health effects of wind turbines in working environments – A scoping review. Scand J Work Environ Health 44, 351–369. https://doi.org/10.5271/SJWEH.3711

Freiberg, A., Schefter, C., Hegewald, J., Seidler, A., 2019b. The influence of wind turbine visibility on the health of local residents: a systematic review. Int Arch Occup Environ Health 92, 609–628. https://doi.org/10.1007/S00420-019-01403-W

Garritty, C., Hamel, C., Trivella, M., Gartlehner, G., Nussbaumer-Streit, B., Devane, D., Kamel, C., Griebler, U., King, V.J., 2024. Updated recommendations for the Cochrane rapid review methods guidance for rapid reviews of effectiveness. BMJ 384. https://doi.org/10.1136/BMJ-2023-076335

52

Godono, A., Ciocan, C., Clari, M., Mansour, I., Curoso, G., Franceschi, A., Carena, E., De Pasquale, V., Dimonte, V., Pira, E., Dallapiccola, B., Normanno, N., Boffetta, P., 2023. Association between exposure to wind turbines and sleep disorders: A systematic review and meta-analysis. Int J Hyg Environ Health 254, 114273. https://doi.org/10.1016/J.IJHEH.2023.114273

Guski, R., Schreckenberg, D., Schuemer, R., 2017. WHO Environmental Noise Guidelines for the European Region: A Systematic Review on Environmental Noise and Annoyance. Int J Environ Res Public Health 14, 1539. https://doi.org/10.3390/IJERPH14121539

Harrison, R. V., 2015. On the biological plausibility of Wind Turbine Syndrome. Int J Environ Health Res 25, 463– 468. https://doi.org/10.1080/09603123.2014.963034

IARC, 2011. IARC classifies Radiofrequency Electromagnetic Fields as possibly carcinogenic to humans IARC [WWW Document]. URL https://www.iarc.who.int/pressrelease/iarc-classifies-radiofrequency-electromagnetic-fields- as-possibly-carcinogenic-to-humans/ (accessed 7.22.25).

Ioannidou, C., Santurette, S., Jeong, C.-H., 2016. Effect of modulation depth, frequency, and intermittence on wind turbine noise annoyance. J Acoust Soc Am 139, 1241–1251. https://doi.org/10.1121/1.4944570

Karasmanaki, E., 2022. Is it safe to live near wind turbines? Reviewing the impacts of wind turbine noise. Energy for Sustainable Development 69, 87–102. https://doi.org/10.1016/J.ESD.2022.05.012

Regulation No. 71 of the Minister of the Environment, 2016. Standard levels of noise in outdoor air and methods for measuring, determining and assessing noise levels – State Gazette [WWW Document]. URL https://www.riigiteataja.ee/akt/127052020002?leiaKehtiv (accessed 5.29.25).

King, V.J., Stevens, A., Nussbaumer-Streit, B., Kamel, C., Garritty, C., 2022. Paper 2: Performing rapid reviews. Syst Rev 11. https://doi.org/10.1186/S13643-022-02011-5

Ministry of Climate, 2025. Guidelines for assessing the environmental impact of wind farms, noise, vibration, shadowing [WWW Document]. URL https://kliimaministeerium.ee/sites/default/files/documents/2025- 03/Tuuleparkide%20keskkonnam%C3%B5ju%20hindamise%20juhend.pdf (accessed 7.28.25).

Knopper, L.D., Ollson, C.A., 2011. Health effects and wind turbines: A review of the literature. Environ Health 10, 1–10. https://doi.org/10.1186/1476-069X-10-78/COMMENTS

Knopper, L.D., Ollson, C.A., McCallum, L.C., Aslund, M.L.W., Berger, R.G., Souweine, K., McDaniel, M., 2014. Wind Turbines and Human Health. Front Public Health 2, 63. https://doi.org/10.3389/FPUBH.2014.00063

Lahti, T., 2010. Ökokratt - Environmental Noise Handbook [WWW Document]. URL https://www.okokratt.ee/index.php/et-ee/kaeimasolevad-projektid/muera/keskkonnamra-ksiraamat- mainmenu-339 (accessed 5.29.25).

Lancet Countdown, 2024. The 2024 Global Report of the Lancet Countdown [WWW Document]. URL https://lancetcountdown.org/2024-report/ (accessed 7.23.25).

Lee, Seunghoon, Kim, K., Choi, W., Lee, Soogab, 2011. Annoyance caused by amplitude modulation of wind turbine noise. Noise Control Eng J 59, 38–46. https://doi.org/https://doi.org/10.3397/1.3531797

53

Liebich, T., Lack, L., Hansen, K., Zajamšek, B., Lovato, N., Catcheside, P., Micic, G., 2021. A systematic review and meta- analysis of wind turbine noise effects on sleep using validated objective and subjective sleep assessments. J Sleep Res 30. https://doi.org/10.1111/JSR.13228

Liebich, T., Lack, L., Hansen, K., Zajamsek, B., Micic, G., Lechat, B., Dunbar, C., Nguyen, D.P., Scott, H., Catcheside, P., 2022a. An experimental investigation on the impact of wind turbine noise on polysomnography-measured and sleep diary-determined sleep outcomes. Sleep 45. https://doi.org/10.1093/SLEEP/ZSAC085

Liebich, T., Lack, L., Micic, G., Hansen, K., Zajamšek, B., Dunbar, C., Lechat, B., Scott, H., Lovato, N., Decup, F., Nguyen, D.P., Catcheside, P., 2022b. The effect of wind turbine noise on polysomnographically measured and self- reported sleep latency in wind turbine noise naïve participants. Sleep 45. https://doi.org/10.1093/SLEEP/ZSAB283

Maijala, P.P., Kurki, I., Vainio, L., Pakarinen, S., Kuuramo, C., Lukander, K., Virkkala, J., Tiippana, K., Stickler, E.A., Sainio, M., 2021. Annoyance, perception, and physiological effects of wind turbine infrasound. J Acoust Soc Am 149, 2238–2248. https://doi.org/10.1121/10.0003509

Małecki, P., Pawlaczyk-Łuszczyńska, M., Wszołek, T., Preis, A., Kłaczyński, M., Dudarewicz, A., Pawlik, P., Stępień, B., Mleczko, D., 2023. Does Stochastic and Modulated Wind Turbine Infrasound Affect Human Mental Performance Compared to Steady Signals without Modulation? Results of a Pilot Study. Int J Environ Res Public Health 20. https://doi.org/10.3390/IJERPH20032223

Marshall, N.S., Cho, G., Toelle, B.G., Tonin, R., Bartlett, D.J., D’rozario, A.L., Evans, C.A., Cowie, C.T., Janev, O., Whitfeld, C.R., Glozier, N., Walker, B.E., Killick, R., Welgampola, M.S., Phillips, C.L., Marks, G.B., Grunstein, R.R., 2023. The Health Effects of 72 Hours of Simulated Wind Turbine Infrasound: A Double-Blind Randomised Crossover Study in Noise-Sensitive, Healthy Adults. Environ Health Perspect 131. https://doi.org/10.1289/EHP10757

McCunney, R.J., Mundt, K.A., Colby, W.D., Dobie, R., Kaliski, K., Blais, M., 2014. Wind turbines and health: a critical review of the scientific literature. J Occup Environ Med 56, e108–e130. https://doi.org/10.1097/JOM.0000000000000313

Murcia, G.M.D., Ferrer, S.F., Sorinas, J., Ferrandez, J.M., Fernandez, E., 2017. Application of electroencephalographic techniques to the study of visual impact of renewable energies. J Environ Manage 200, 484–489. https://doi.org/10.1016/J.JENVMAN.2017.05.096

Orru, H., Teinemaa, E., Maasikmets, M., Keernik, H., Paju, M., Sikk, A., Tamm, T., Lainjärv, H.M., Kriit, H.K., Lõhmus Sundström, M., 2022. Comparison of the impact of outdoor air quality on human health in Estonia in 2010 and 2020 and a forecast of the health effects of air pollution for 2030 [WWW Document]. URL https://keskkonnaportaal.ee/et/valisohu-kvaliteedi-moju-vordlus-inimeste-tervisele-eestis-aastatel-2010-ja- 2020-ning-ohusaaste (accessed 7.30.25).

Pierpont, N., 2009. (PDF) Wind Turbine Syndrome: A Report on a Natural Experiment [WWW Document]. URL https://www.researchgate.net/publication/265247204_Wind_Turbine_Syndrome_A_Report_o n_a_Natural_Experiment (accessed 7.17.25).

Pohl, J., Gabriel, J., Hübner, G., 2018. Understanding stress effects of wind turbine noise – The integrated approach. Energy Policy 112, 119–128. https://doi.org/10.1016/J.ENPOL.2017.10.007

54

Poulsen, A.H., Raaschou-Nielsen, O., Peña, A., Hahmann, A.N., Nordsborg, R.B., Ketzel, M., Brandt, J., Sørensen, M., 2019a. Long-Term Exposure to Wind Turbine Noise and Risk for Myocardial Infarction and Stroke: A Nationwide Cohort Study. Environ Health Perspect 127. https://doi.org/10.1289/EHP3340

Poulsen, A.H., Raaschou-Nielsen, O., Peña, A., Hahmann, A.N., Nordsborg, R.B., Ketzel, M., Brandt, J., Sørensen, M., 2019b. Impact of Long-Term Exposure to Wind Turbine Noise on Redemption of Sleep Medication and Antidepressants: A Nationwide Cohort Study. Environ Health Perspect

Poulsen, A.H., Raaschou-Nielsen, O., Peña, A., Hahmann, A.N., Nordsborg, R.B., Ketzel, M., Brandt, J., Sørensen, M., 2018a. Long-term exposure to wind turbine noise at night and risk for diabetes: A nationwide cohort study. Environ Res 165, 40–45. https://doi.org/10.1016/J.ENVRES.2018.03.040

Poulsen, A.H., Raaschou-Nielsen, O., Peña, A., Hahmann, A.N., Nordsborg, R.B., Ketzel, M., Brandt, J., Sørensen, M., 2018b. Long-term exposure to wind turbine noise and redemption of antihypertensive medication: A nationwide cohort study. Environ Int 121, 207–215. https://doi.org/10.1016/J.ENVINT.2018.08.054

Poulsen, A.H., Raaschou-Nielsen, O., Peña, A., Hahmann, A.N., Nordsborg, R.B., Ketzel, M., Brandt, J., Sørensen, M., 2018c. Pregnancy exposure to wind turbine noise and adverse birth outcomes: a nationwide cohort study. Environ Res 167, 770–775. https://doi.org/10.1016/J.ENVRES.2018.09.011

Poulsen, A.H., Raaschou-Nielsen, O., Peña, A., Hahmann, A.N., Nordsborg, R.B., Ketzel, M., Brandt, J., Sørensen, M., 2018d. Short-term nighttime wind turbine noise and cardiovascular events: A nationwide case-crossover study from Denmark. Environ Int 114, 160–166. https://doi.org/10.1016/J.ENVINT.2018.02.030

Rosciszewska, A., Buszkiewicz, M., Dobrzynska-Kobylec, G., Klichowska, A., Przybyla, T., Nagy, B.B., Wicher, A., Klichowski, M., 2025. Cognitive neuroscience approach to explore the impact of wind turbine noise on various mental functions. Humanities and Social Sciences Communications 2025 12:1 12, 1–17. https://doi.org/10.1057/s41599-025-04645-x

Schäffer, B., Pieren, R., Schlittmeier, S.J., Brink, M., 2018. Effects of Different Spectral Shapes and Amplitude Modulation of Broadband Noise on Annoyance Reactions in a Controlled Listening Experiment. Int J Environ Res Public Health 15. https://doi.org/10.3390/IJERPH15051029

Schäffer, B., Schlittmeier, S.J., Pieren, R., Heutschi, K., Brink, M., Graf, R., Hellbrück, J., 2016. Short-term annoyance reactions to stationary and time-varying wind turbine and road traffic noise: A laboratory study. J Acoust Soc Am 139, 2949–2963. https://doi.org/10.1121/1.4949566

Schmidt, J.H., Klokker, M., 2014. Health Effects Related to Wind Turbine Noise Exposure: A Systematic Review. PLoS One 9, e114183. https://doi.org/10.1371/JOURNAL.PONE.0114183

Regulation No. 38 of the Minister of Social Affairs, 2002. Limit values for non-ionising radiation in residential and recreational areas,

residential buildings and public buildings, classrooms and measurement of non-ionising radiation levels [WWW Document]. URL https://www.riigiteataja.ee/akt/163816?leiaKehtiv (accessed 8.9.25).

55

Regulation No. 42 of the Minister of Social Affairs, 2002. Noise standards in residential and recreational areas, residential buildings and public buildings, and methods for measuring noise levels [WWW Document]. State Gazette. URL https://www.riigiteataja.ee/akt/163756 (accessed 7.11.25).

Regulation No. 75 of the Minister of Social Affairs, 2002. Limit values for ultra- and infrasound sound pressure levels and measurement of ultra- and infrasound sound pressure levels [WWW Document]. URL https://www.riigiteataja.ee/akt/129122020046?leiaKehtiv (accessed 5.29.25).

Regulation No. 78 of the Minister of Social Affairs, 2002. Limit values for vibration in residential and public buildings and methods for measuring vibration [WWW Document]. URL https://www.riigiteataja.ee/akt/110061?leiaKehtiv (accessed 8.9.25).

Staniek, A., Kompała, J., Bramorska, A., Bartmański, C., 2013. Comparison analysis of noise generated by wind turbines with other noise sources in outdoor environments. Vibrations in Physical Systems 2023, 2023108. https://doi.org/10.21008/j.0860-6897.2023.1.08

Taylor, J., Klenk, N., 2019. The politics of evidence: Conflicting social commitments and environmental priorities in the debate over wind energy and public health. Energy Res Soc Sci 47, 102–112. https://doi.org/10.1016/J.ERSS.2018.09.001

Teneler, A.A., Hassoy, H., 2023. Health effects of wind turbines: a review of the literature between 2010-2020. Int J Environ Health Res 33, 143–157. https://doi.org/10.1080/09603123.2021.2010671/ASSET/CB263B43-F0EB- 4EF2-94C8-593ADABE314A/ASSETS/IMAGES/CIJE_A_2010671_F0001_B.GIF

Health Board, 2025a. Report on noise measurements at the Sopi-Tootsi wind farm [WWW Document]. URL https://sopitootsipargid.ee/et/avaleht/ (accessed 7.21.25).

Health Board, 2025b. Saarde wind farm noise study [WWW Document]. URL https://utilitas.ee/failipank/saarde-tuulepargi-murauuring/ (accessed 7.21.25).

Health Board, 2025c. Noise and vibration [WWW Document]. URL https://www.terviseamet.ee/keskkonnatervis/fuusikalised-ohutegurid-elukeskkonnas/mura-ja- vibratsioon#kohtvibratsioon (accessed 8.9.25).

Tonin, R., Brett, J., Colagiuri, B., 2016. The effect of infrasound and negative expectations to adverse pathological symptoms from wind farms. Journal of Low Frequency Noise Vibration and Active Control 35, 77–90. https://doi.org/10.1177/0263092316628257/ASSET/D90140C8-16F5-4714-B165- 9A39FD096E84/ASSETS/IMAGES/LARGE/10.1177_0263092316628257-FIG13.JPG

Government Regulation No. 44, 2016. Occupational health and safety requirements for work environments affected by electromagnetic fields, exposure limits for electromagnetic fields and application values, and procedure for measuring electromagnetic fields [WWW Document]. URL https://www.riigiteataja.ee/akt/107042016004?leiaKehtiv (accessed 7.22.25).

van Kamp, I., van den Berg, F., 2021. Health effects related to wind turbine sound: An update. Int J Environ Res Public Health 18. https://doi.org/10.3390/IJERPH18179133

van Kamp, I., van den Berg, F., 2018. Health Effects Related to Wind Turbine Sound, Including Low-Frequency Sound and Infrasound. Acoust Aust 46, 31–57. https://doi.org/10.1007/S40857-017-0115-6/FIGURES/3

van Kempen, E., Casas, M., Pershagen, G., Foraster, M., 2018. WHO Environmental Noise Guidelines for the European Region: A Systematic Review on Environmental Noise and Cardiovascular and

56

Metabolic Effects: A Summary. International Journal of Environmental Research and Public Health 2018, Vol. 15, Page 379 15, 379. https://doi.org/10.3390/IJERPH15020379

WHO, 2025. Constitution of the World Health Organisation [WWW Document]. URL https://www.who.int/about/governance/constitution (accessed 7.24.25).

WHO, 2019. Environmental noise guidelines for the European Region [WWW Document]. URL https://www.who.int/europe/publications/i/item/9789289053563 (accessed 7.24.25).

WHO, 2016. Radiation: Electromagnetic fields [WWW Document]. URL https://www.who.int/news- room/questions-and-answers/item/radiation-electromagnetic-fields (accessed 7.22.25).

WHO, 2009. Night noise guidelines for Europe [WWW Document]. URL https://iris.who.int/handle/10665/326486 (accessed 7.28.25).

World Bank Group, 2015. Environmental, Health, and Safety Guidelines for Wind Energy [WWW Document]. URL https://documents1.worldbank.org/curated/en/498831479463882556/pdf/110346-WP-FINAL-Aug-2015- Wind-Energy-EHS-Guideline-PUBLIC.pdf (accessed 7.28.25).

Wu, S., Du, W., Zhong, X., Lu, J., Wen, F., 2023. The association between road traffic noise and type 2 diabetes: a systematic review and meta-analysis of cohort studies. Environmental Science and Pollution Research 30, 39568–39585. https://doi.org/10.1007/S11356-023-25926-5/TABLES/5

57

Appendix 1. Search terms and search strategy used

Search strategy in the PubMed database

(("wind"[MeSH Terms] OR wind turbine[Title/Abstract] OR wind turbines[Title/Abstract] OR wind farms[Title/Abstract] OR wind parks[Title/Abstract] OR wind power plants[Title/Abstract] OR wind mill[Title/Abstract] OR wind generators[Title/Abstract]) AND ("noise"[MeSH Terms] OR "sound"[MeSH Terms:noExp] OR noise[Title/Abstract] OR infrasound[Title/Abstract] OR low-frequency noise[Title/Abstract] OR sound[Title/Abstract] OR vibration[Title/Abstract] OR visibility[Title/Abstract] OR visual[Title/Abstract] OR shadow flickering[Title/Abstract] OR flicker[Title/Abstract] OR electromagnetic field[Title/Abstract] OR "Electromagnetic Fields"[Mesh] OR infrasonic[Title/Abstract] OR "low frequency"[Title/Abstract] OR light flickering[Title/Abstract] OR stroboscopic effect[Title/Abstract] OR blinking lights[Title/Abstract] OR reflections[Title/Abstract] OR horizon pollution[Title/Abstract] OR light effects[Title/Abstract])) AND (("1 January 2010"[Date - Publication] : "22 April 2025"[Date - Publication]))

58

Appendix 2. Tables of results

Table 1. Systematic reviews Source Exposure Health outcomes Study designs Publication

date Included studies number

Assessment of study quality Study results

Liebich et al 2021

Noise dB (A) LAeq

Objective sleep disturbance indicators: WASO1, SOL2, TST3, sleep efficiency

Experiments, pre- post studies, cross-sectional

2000–2020 9 qualitative analysis; 5 meta- analyses

The reporting quality of individual studies was assessed using an adaptation of the STROBE4 checklist. The overall reporting quality of the studies was low. No tools were used to assess the risk of bias. , the assessment based on the limitations and biases of the studies as follows: 4 studies with high, 4 with moderate (some concerns) and 1 with low risk of bias

Objectively measured indicators of sleep macrostructure were not significantly affected in those exposed to wind turbine noise compared to controls not exposed to wind turbine noise. An effect was found on subjectively measured sleep indicators

Guski et al 2017

Noise dB (A), Lden

Disturbance Cross-sectional 2000–2012 4 The GRADE5 methodology assessed the level of evidence as low to moderate.

Wind turbine noise is associated with disturbance even at levels below 40 dB Lden.

Godono et al 2023

Distance from wind turbines

Noise dB (A)

Self-reported sleep quality

Cross-sectional 2004–2021 15 The methodological quality of the studies was assessed using the US National Institutes of Health Quality Assessment Tool: 2 high, 5 moderate and 8 low quality studies

The prevalence of sleep disturbance decreased with increasing distance from wind turbines and increased with higher sound pressure levels.

Freiberg et al 2019b

Visual aspects (visibility from the place of residence, shadowing, flashing lights, etc.)

Disturbance, sleep quality, quality of life

Cohort, cross-sectional

Up to = 2017

17 qualitative analysis; 6 meta- analysis

The quality of cross-sectional studies was assessed using the Appraisal tool for Cross-Sectional Studies (AXIS) tool, and cohort studies were assessed using a combination of tools from the Scottish Intercollegiate Guidelines Network (SIGN) and the Critical Appraisal Skills Programme (CASP) tools. Five studies were rated as high quality, three acceptable and 7 low quality

Disturbance from direct visibility, glare and flashing lights was statistically significantly associated with an increased risk of sleep disturbance. Changes in the view of the landscape, obstacle markings and light reflection from wind turbine blades may also disturb people .

1WASO - wake after sleep onset 2SOL - sleep onset latency 3TST - total sleep time 4STROBE – principles for reporting observational studies (STrengthening the Reporting of OBservational studies in Epidemiology) 5GRADE - Grading of Recommendations Assessment, Development and Evaluation

59

Table 2. Observational studies Source Study design and

subjects Sample size

Age Follow- up Self time

Exposure Health outcome Method of measurement method

Study results

Bräuner et al 2018

Danish nurses Cohort study

23,994 ≥44 1982– 2013

Modelled A-weighted wind turbine noise at frequencies 10 Hz–10,000 Hz. Calculated annual average Lden at the residences under study within a 6 km radius of the nearest wind turbine.

686 heart attacks (1.7 new cases per 1,000 person-years)

Danish patient and cause of death registries

Long-term exposure to wind turbine was not associated with the occurrence of heart attacks in women aged 44 and older.

Bräuner et al 2019b

Danish nurses cohort study

23,912 ≥44 1982– 2013

Modelled A-weighted wind turbine noise at frequencies of 10 Hz–10,000 Hz. Calculated annual average Lden at the residences under study within a 6 km radius of the nearest wind turbine. 1-, 5-, and 11-year average

1097 strokes (2.7 new cases per 1,000 person-years)

Danish patient registry

Long-term exposure to wind turbine was not associated with stroke in women aged 44 and older.

Bräuner et al 2019a

Danish nurses cohort study

24,137 ≥44 1982– 2013

Modelled A-weighted wind turbine noise at frequencies of 10 Hz–10,000 Hz. Calculated annual average Lden at the residences under study within a 6 km radius of the nearest wind turbine. 1-, 5-, and 11-year average

1430 homes atrial fibrillation (3.5 new cases per 1,000 person-years)

Danish patient and cause of death registries

The study results found no evidence between wind turbine noise and the occurrence of atrial fibrillation and the occurrence of atrial fibrillation , but no clear statistically significant association in women aged 44 and older.

Poulsen et al 2018a

Danish nationwide registry-based cohort study

614,731 25 1996 2012

Modelled A-weighted night-time noise levels in residential yards at frequencies of 10–10,000 Hz and low-frequency noise in indoor spaces at frequencies of 10–160 Hz. 1-year and 5-year averages

25,148 diabetes cases Danish Diabetes Register

Long-term exposure to wind turbine noise at night in residential yards and low-frequency noise indoors was not associated with the development of diabetes.

Poulsen et al 2018b

Danish nationwide register-based cohort study

535,675 25–84 1996 2013

Modelled A-weighted night-time noise levels in residential yards at frequencies of 10–10,000 Hz and low-frequency noise in indoor spaces at frequencies of 10–160 Hz. 1-year and 5-year averages

83,729 people purchased oral hypertension medication

Data from the Danish Prescription Centre

Long-term exposure to wind turbine noise at night in residential yards and low-frequency noise indoors was not associated with the purchase of oral hypertension medications Poulsen et al 2018c

Poulsen et al 2018c

Danish nationwide register-based cohort study

135,795 single birth births,

122,792 temporary single nitust

NA1 1996– 2013

Modelled night-time A-weighted noise level in the yard of a residential building at frequencies 10–10,000 Hz and low-frequency noise in indoor spaces at frequencies 10–160 Hz. 1-year and 5-year averages

13,003 premature births; 12,220 births with low birth weight for gestational age; 1,127 preterm births with low birth weight

Denmark birth register

Exposure to night-time noise in the outdoor area of residential buildings and low-frequency noise indoors during pregnancy was not associated with adverse birth outcomes.

60

Poulsen et al 2019a

Danish nationwide register-based cohort study

711,249 heart attack 712,401 Stroke

25–84 1996 2013

Modelled A-weighted night-time noise level in residential yards at frequencies of 10–10,000 Hz and low-frequency noise in indoor spaces at frequencies of 10–160 Hz. 1-year and 5-year averages

19,145 heart attacks, 18,064 strokes

Danish patient and cause of death registries

The study found evidence that long-term exposure to night-time noise from wind turbines in residential areas may increase the risk of heart attack and stroke. Low-frequency noise indoors was not associated with the occurrence of heart attacks and strokes .

Poulsen et al 2019b

Danish nationwide register-based cohort study

583,968 sample of sleep medicatio ns;

584,891 sample of antidepr essants

25 1996 2013

Modelled night-time A-weighted noise level in residential yards at frequencies of 10–10,000 Hz and low-frequency noise in indoor spaces at frequencies of 10–160 Hz. 1-year and 5-year averages

68,696 people purchased sleeping pills

82,373 people bought antidepressants

Data from the Danish Prescription Centre

Long-term exposure to night-time wind turbine noise in residential areas was associated with the purchase of sleeping pills and antidepressants among subjects over 65 years of age. Long-term exposure to low-frequency noise indoors was not associated with the purchase of sleeping pills and antidepressants Poulsen et al 2018c

Poulsen et al 2018c

Danish nationwide register-based case- crossover study

15,092 infarction (13,343 people)

14,623 strokes (13,026 people)

≥18 1982 2013

Modelled night-time A-weighted noise level in the yard of the dwelling at frequencies 10–10,000 Hz and low- frequency noise indoors at frequencies 10–160 Hz during the 4 days prior to the onset of illness or the reference day

Heart attack, stroke Danish patient and cause of death registers

Short-term exposure to wind turbine noise at night in residential yards and low-frequency noise indoors was not associated with the occurrence of heart attacks and strokes in the main analysis. Additional analyses found evidence that high levels of low-frequency noise indoors may trigger heart attacks or stroke.

1 NA – no data

61

Table 3. Experiments on infrasound Source Country Study design Sample

size Age Exposure Health outcome Method of measuring

outcome Method

Study results

Marshall et al 2023

Australia Randomised, controlled, double- blind study with three research groups. Only noise-sensitive individuals were included, who participated in three trials. Each trial lasted 72 consecutive hours.

37 18–72 Infra-sound with a frequency of 1.6– 20 Hz and a maximum intensity of 90 dB (Z), simulating the infrasound of wind turbines. The control group was exposed to traffic noise during the night (22:00–07:00) with an average intensity of 40–50 dB (A) and a maximum intensity of 70 dB, or no noise was generated (average background noise during the night 39 dB (A) / 80–85 dB (Z) originating from air conditioning).

Various physiological and psychological indicators were measured in all subjects. Sleep indicators, cardiovascular indicators, psychological and mental well-being indicators, stress indicators from blood samples. The subjects were also asked to assess the presence of symptoms of "wind turbine syndrome".

PSG1 and questionnaires.

Exposure to infrasound did not impair any of the measured health indicators. Traffic noise prolonged the time it took to fall asleep.

Ascone et al 2021

Germany Randomised controlled, unidirectionally blinded long-term exposure (1 month) study. Two groups: infrasound vs. placebo.

38 18 Generated infrasound (6 Hz, 80–90 dB (Z)) or placebo sound in the bedrooms of the subjects for 28 consecutive nights.

Self-reported symptoms, sleep quality, mental functioning.

Questionnaires, mental performance tests, MRI (magnetic

resonance imaging).

Exposure to infrasound did not affect self-reported health, sleep quality or mental abilities. Changes in brain grey matter were observed, but these cannot be interpreted as either harmful or beneficial.

Crichton et al 2014a

New Zealand

Randomised, controlled double-blind provocation study, two study groups: 1) expectations that infrasound is harmful were created based on real information circulating on the internet vs. 2) expectations that infrasound is not harmful were created.

54 Super - learne rs

Generated infrasound (5Hz, 40 dB) and placebo sound in 10-minute sessions.

24 different self-reported symptoms, such as headache, pressure in the ears, dizziness, nausea. Blood pressure and heart rate measured.

Symptoms reported on a scale before and during the sessions.

The group with high expectations of adverse effects reported significant increases in the number and intensity of symptoms compared to the pre-exposure assessment during both the infrasound and placebo sound sessions. No changes were observed in the group with low expectations of adverse effects. No effect of infrasound on blood pressure or heart heart rate. The study shows that real information circulating on the Internet about the negative health effects of infrasound increases the occurrence of self-reported symptoms .

62

Crichton et al 2014b

New Zealand

Randomised, controlled, two study groups: 1) expectation that infrasound is harmful vs 2) expectation that infrasound is beneficial.

60 Super - learne rs

Generated infrasound (9 Hz, 50.4 dB ) and audible sound from a wind farm 1 km away (43 dB) were presented simultaneously in 7-minute sessions.

Twenty-four different self- reported symptoms, plus 12 positive mood expressions and 12 negative mood expressions.

Symptoms and mood were assessed on a scale before each session and during the sessions.

In the negative expectations group, symptom reporting increased and mood worsened during the session, while in the positive expectations group, symptom reporting decreased and mood improved compared to what was reported before the session. The study shows that positive information has a placebo effect.

Crichton et al 2015

New Zealand

Randomised, controlled, two study groups: 1) expectation that infrasound is harmful vs 2) expectation that infrasound is beneficial.

60 Super - learne rs

Generated infrasound (9 Hz, 50.4 dB) and audible sound from a wind farm 1 km away (43 dB) were presented simultaneously in 7-minute sessions.

Disturbance, 12 positive mood expressions and 12 negative mood expressions, plus self-rated noise sensitivity.

Disturbance, mood, and noise sensitivity were assessed on a scale before each session and during the sessions.

The positive expectations group reported less disturbance during the session than the negative expectations group. The study shows that negative information causes annoyance from wind farms, while positive information reduces annoyance, and even among those who are sensitive to noise.

Crichton and Petrie 2015b

New Zealand

Randomised, controlled, two study groups: 1) an expectation was created that infrasound is harmful vs 2) expectation that infrasound is beneficial was created. A repeat experiment was conducted in which the information provided was changed.

64 17–56 Generated infrasound (9 Hz, 50.4 dB) and sound audible to birds 1 km away from the wind farm (43 dB) were presented simultaneously in 10- minute sessions.

Self-reported 24 different symptoms, 12 positive mood expressions and 12 negative mood expressions.

Symptoms and mood were assessed on a scale before each session and during the sessions.

In the negative expectations group, symptom increased and mood deteriorated compared to what was reported before the session, and vice versa. When the negative expectation group was given positive information about the benefits of wind turbines in a repeat experiment, they reported fewer symptoms and their mood improved. Similarly, those who had heard the positive information first experienced a deterioration in mood and more symptoms after receiving the negative information. The results show that positively worded health information can reverse or reduce the impact of negative expectations created by warnings about the health risks of wind turbines published in the media warnings about the health risks of wind turbines.

Crichton and Petrie 2015a

New Zealand

All of the were given negative health effects. The subjects were then randomly divided into two groups: 1) the subjects were given information that infrasound health effects are biologically justified vs 2)

66 17–70 Generated infrasound (9 Hz, 50.4 dB) and audible sound from a wind farm 1 km away (43 dB) were presented simultaneously in 14-minute sessions.

Self-reported 24 different symptoms, 12 positive mood expressions and 12 negative mood expressions.

Symptoms and mood were assessed on a scale before each session and during the sessions. The credibility and comprehensibility of the information shared was also assessed.

After receiving negative information, the number and intensity of reported symptoms increased in both groups compared to the baseline. In the biological explanation group, the increase in symptoms persisted during the second session. In the nocebo explanation group, however, the number and intensity of symptoms decreased and mood improved. The experiment shows that false information found in the media increases the occurrence of symptoms and concern about health . It also shows that the nocebo effect

63

was explained to the subjects that the health effects were the result of the nocebo effect

providing an explanation may reduce the reporting of symptoms associated with wind turbines. Participants in both groups found the explanation they were given to be understandable, reasonable, convincing and correct.

Tonin et al 2016

Australia Randomised double-blind trial of infrasound from headphones 4 groups

72 17 Simulated variable infrasound from wind turbines 0.8–40 Hz with a maximum intensity of 91 dB (Z) 23 minutes or placebo sound

Participants were led to expect that the infrasound would have a harmful effect or that there would be no harmful effects.

Self-reported 24 different symptoms, Concern about the health effects of wind turbines health effects.

Questionnaires before and after the listening session

In the infrasound group, the occurrence of symptoms decreased. It was not possible to shape expectations among the participants. A statistically significant worsening of symptoms was

observed among participants who believed before the experiment that infrasound affects health, regardless of whether they heard infrasound or a placebo sound. The results support the nocebo effect hypothesis.

Liebich et al 2022a

Australia Randomised controlled single- blind trial in a sleep laboratory

68 18 Recorded wind farm noise/infrasound was presented in a sleep laboratory at an intensity of 25 dB(A) on seven consecutive nights. The background noise in the laboratory was 19 dB (A). The wind farm noise contained infrasound from 1.6 Hz and amplitude modulation at frequencies of 31.5 and 63 Hz.

Objectively measured and self- reported sleep quality

PSG1, sleep diary, questionnaires on sleep and noise sensitivity questionnaires

Wind turbine noise at a level of 25 dB (A) indoors has no measurable effect on objective or subjective sleep indicators. No effect was observed even in those who reported wind turbine-related sleep disturbances.

Liebich et al 2022b

Australia Randomised controlled single- blind trial in a sleep laboratory

23 18 Recorded wind farm noise/infrasound was presented in a sleep laboratory at an intensity of 33 dB(A) in random order, alternating with laboratory background noise (23 dB(A)). Wind farm noise contained infrasound and noticeable amplitude modulation at 46 Hz .

Time taken to fall asleep (sleep latency)

PSG1, sleep diary, questionnaires on sleep and noise sensitivity The study shows that wind turbine noise at a level of 33

The study shows that wind turbine noise at a level of 33 dB (A) does not prolong the time it takes to fall asleep, as measured objectively or subjectively in young, healthy people who have not previously been exposed to wind turbines.

Maijala et al 2021

Finland Randomised controlled double-blind trial

26 30 Recorded sound in the wind farm area 200 m from the turbine (47–57 dB (A), 52–77 dB (Z)), in a yard 1.5 km from the turbine (42–59 dB (A)) and indoors (41–43 (A)). The highest sound pressure level and highest amplitude modulation

Measured autonomic nervous system responses (heart rate, heart rate variability, skin conductance).

on the disturbance assessment scale; measured heart rate , heart rate variability,

The study shows that the infrasound levels used in the experiment did not affect the disturbance or autonomic nervous system responses of the subjects, even though the experimental conditions corresponded acoustically to actual wind farms. The subjects did not distinguish between presentations containing infrasound and those without. The or non-presentation did not affect

64

deep recordings. The sound of the ocean shore (34–45 dB (A)) was used as a control. Subjects were exposed or not exposed to infrasound (20 Hz) that had been filtered out of the wind farm sound.

Subjective disturbance assessments

skin conductance disturbance level. The audible sounds from the wind farm were more disturbing than the ocean sounds. Sounds with greater amplitude modulation were also more disturbing. Subjects who reported health concerns related to infrasound prior to the study did not differ in their reactions to infrasound from other .

Małecki et al 2023

Poland Randomised trial, 3 groups, conditions not controlled (conducted in a classroom)

129 21 Different exposures: *Recorded and filtered wind turbine infrasound with an intensity of 83 dB (G) /47 dB (A). *Synthetic infrasound with an intensity of 5–20 Hz at 78 dB (G) / 46 dB (A) and no amplitude modulation or deviation. *Background noise with an intensity of 63 dB (G) / 43 dB (A).

Cognitive functions, especially attention. Feelings and symptoms

Cognitive ability tests, questionnaires before and after the session

The study results showed no significant differences in cognitive test results or in the number of reported unpleasant sensations or complaints between different sound conditions when men and women were analysed separately. Women reported discomfort and various complaints more than men.

Rosciszewska et al 2025

Poland Randomised controlled single-blind trial, 3 groups

45 18 Recorded wind turbine noise from a 2 MW wind turbine at a distance of 500 metres. The sound intensity used was 65.4 dB (Z) / 38.5 dB (A). Wind turbine noise contained amplitude modulation (average frequency 0.8–1 Hz, depth ~6.9 dB). Recorded road traffic noise 65.4 dB (Z) / 56.8 dB(A). There was background noise in the control.

Cognitive functions, disturbance, stress, depression

EEG2 measurements, cognitive ability tests, questionnaires disturbance, depression, anxiety and stress about

Short-term exposure to wind turbine noise did not affect the cognitive (measured by brain functions such as attention and thinking). Wind turbine noise was not perceived as significantly more disturbing or stressful than traffic noise. The participants did not know the source of the noise, which may be the reason why wind turbine noise was not perceived as more disturbing.

1PSG – polysomnography 2EEG – electroencephalography is a method of measuring the electrical activity of the brain

65

Table 4. Experiments involving audible noise and visual aspects of wind turbines Source Country Study design Sample

size Age Exposure Health outcome Method of measuring

outcome Method

Study results

Chiu et al 2021

Taiwan Randomised trial, two groups, but no control conditions.

29 22–75 Low-frequency wind turbine noise (20–200 Hz) was measured 20 m from the nearest wind turbine (38.3–57.1 dB(A)) outdoors and 500 m from the nearest wind turbine indoors (32.2–52.5 dB (A)).

Heart rate and heart rate variability

Portable electrocardiogra m (ECG) recorder

The test showed that exposure to wind turbine noise can reduce heart rate . This may increase the risk of cardiovascular disease.

Ioannidou et al 2016

Denmark Controlled experiment, all subjects listened to and rated sounds with different AM1

sounds.

19 23–28 Sounds with different AM1 frequencies ranging from 200 to 1200 Hz and an intensity of 60 dB (A), based on wind farm recordings, but with the AM1 artificially modified for the experiment artificially for the experiment.

Disturbance Self-assessment on a scale of 1–10

Disturbance is affected by the depth of AM1. The smaller the range in which the noise level fluctuates, the less disturbing the sound is.

Lee et al 2011 South Korea

Controlled experiment, all subjects listened to sounds of varying loudness and varying AM1 sounds.

30 20 Recorded noise from a single wind turbine (250 Hz– 8000 Hz). Participants were presented with noise levels of 35 dB, 40 dB, 45 dB, 50 dB and 55 dB (A) for each AM1 depth level.

Disturbance Self-assessment on a scale of 1–11

The greater the AM1 depth, the greater the disturbance. Disturbance also increased with louder sounds.

Schäffer et al 2018

Switzerland Controlled experiment, all subjects listened to sounds with different frequency distributions and different AM1 values.

52 18 Generated sounds in the range of 16 Hz to 16 kHz with an intensity of 40 dB (A). The tests used pink noise, wind turbine noise (simulated 2 MW Vestas V90 type wind turbine) and noise with an increased low-frequency component.

Disturbance Self-assessment on a scale of 1–11, questionnaire for background data

Disturbance increased with increasing AM1 depth. Disturbance was higher in situations with random AM1 than in situations without AM1 sounds. The sound with an increased low-frequency component was more disturbing than the wind turbine noise. Disturbance was not related to the gender or noise sensitivity of the participants, but was higher with increasing age and lower with a more positive attitude towards wind farms.

Schäffer et al 2016

Switzerland Controlled experiment, all subjects listened to sounds with different characteristics.

60 18 Generated sound with an intensity of 35 and 60 dB(A), simulating a 2 MW Vestas V90 wind turbine operating in strong wind conditions. Sounds with different AM1 were also generated (without AM1, periodic and random AM1). A total of 30 different sound stimuli were used in the study. These represented different situations involving wind turbine and traffic noise, varying in sound pressure level, source type and AM1. Each sound stimulus lasted 25 seconds.

Disturbance Self-assessment on a scale of 1–11, questionnaire for background data

At the same sound intensity, disturbance was greater for wind turbines than for traffic noise. The presence of AM1 increased disturbance.

66

Murcia et al 2017

Spain Experiment 14 18 Sixty images were used as stimuli. These were divided into three groups: images with wind turbines and the same images without them; images with and without solar parks; and images with and without nuclear power plants.

Disturbance, emotions

EEG2, questionnaires The images were scored on a scale from 9 (very pleasant) to 1 (very unpleasant).

Both objective and subjective measurements showed that disturbance and emotions did not differ when the subjects were shown images of landscapes with or without wind turbines and with or without solar panels. However, images with nuclear power plants evoked negative emotions.

1AM – amplitude modulation 2EEG – electroencephalography is a method of measuring the electrical activity of the brain

This is an excerpt from IARO Report (IARO24-5):

“Health Report on a Rural Sheep Farm in Scotland.”

[Critical Review of Marshall N, Cho G, Toelle BG, Tonin R, Bartlett DJ, et al. (2023) The Health Effects of 72 Hours of Simulated Wind Turbine Infrasound: A Double-Blind Randomised

Crossover Study in Noise-Sensitive, Health Adults. Environmental Health Perspectives, 131(3): 1-10. https://pubmed.ncbi.nlm.nih.gov/36946580/ ]

(All Figures and Paragraphs referred to, but not included, in this excerpt can be found in the Full Report, available at iaro.org.nz)

Health Report on a Rural Sheep Farm in Scotland

Page 64 of 169

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 http://IARO.org.nz

6. Other studies cited in the Letter

The dearth of knowledge on the matter at hand continues to be demonstrated by the signatory of the Letter:

“In addition to the impacts of audible noise itself, the contribution from low frequency infrasound to health effects has also been postulated although findings from recent studies have suggested that this is not supported. 297,298 Similarly, Turunen et al. whilst unable to assess a causal relationship due to the cross-sectional nature of the study, suggested that interpretations of symptoms are affected by other factors in addition to the actual exposure.299”

213. For educational purposes,300 a brief review is conducted of the three studies cited above by the NHS-Highland medical representative.

I. Immediate effects of infrasound exposure

214. In the 2023 study by Marshall et al.,301, 302 the objective is stated as follows:

297 Footnote 5 of the Letter. Marshall N, Cho G, Toelle BG, Tonin R, Bartlett DJ, et al. (2023) The Health Effects of 72 Hours of Simulated Wind Turbine

Infrasound: A Double-Blind Randomised Crossover Study in Noise-Sensitive, Health Adults. Environmental Health Perspectives, 131(3): 1-10. https://pubmed.ncbi.nlm.nih.gov/36946580/ [website added]

298 Footnote 6 of the Letter. Maijala PP, Kurki I, Vainio L, Pakarinen S, Kuuramo C, et al. (2021) Annoyance, perception, and physiological effects of wind turbine infrasound. Journal of the Acoustical Society of America, 149(4): 2238-2248. https://pubmed.ncbi.nlm.nih.gov/33940893/ [website added]

299 Footnote 7 of the Letter. Turunen AW, Tittanen P, Yli-Tuomi T, Taimisto P, Lanki T. (2021) Symptoms intuitively associated with wind turbine infrasound. Environmental Research, 192: 1-9. https://pubmed.ncbi.nlm.nih.gov/33131679/ [website added]

300 As indicated in Paragraphs 37 and 40, the primary reason for such a comprehensive approach to this IARO Health Report is to provide an educational and instructive document for the NHS-Highland medical staff, with the ultimate purpose of benefiting the Scottish Citizen.

301 Marshall N, Cho G, Toelle BG, Tonin R, Bartlett DJ, et al. (2023) The Health Effects of 72 Hours of Simulated Wind Turbine Infrasound: A Double-Blind Randomised Crossover Study in Noise-Sensitive, Health Adults. Environmental Health Perspectives, 131(3): 1-10. https://pubmed.ncbi.nlm.nih.gov/36946580/

302 Disclaimer included in the 2023 Marshall et al. paper: “All of the authors have superannuation accounts which are compulsory in Australia and these accounts may contain investments in both traditional and renewable energy, including wind turbines. R.T. is the founding principal of Renzo Tonin Associates who have previously worked as consultants for the NSW Department of Planning on several wind farms in NSW, Australia. None of the investigators have any other pecuniary interest or academic conflicts of interest in the outcomes of this study.“

Health Report on a Rural Sheep Farm in Scotland

Page 65 of 169

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 http://IARO.org.nz

We aimed to test the effects of 72 h of infrasound (1.6–20 Hz at a sound level of ∼ 90 dB pk re 20 microPa, [303 , 304 ] simulating a wind turbine infrasound signature) exposure on human physiology, particularly sleep.

215. In Medical Sciences, this type of study purports to investigate the immediate effects of exposure, as opposed to long-term effects:

Our principal hypothesis was that exposure to infrasound in healthy individuals, at a level of ∼ 90 dB pk re 20 microPa compared with the sham infrasound, increases WASO [305] —a measure of sleep disturbance—and worsens other measures of sleep quality, mood, WTS [306] symptoms, and other electrophysio- logical measures. In addition, as a positive control, we also tested whether audible traffic noise, a mixture of road (motorbike, truck, car) and aircraft noise (at a sound level of 40–50 dB LAeq; night and 70 dB LAFmax transient maxima)

had an adverse impact on these same outcomes, when compared with sham infrasound.307

216. The conclusions of this study were:

Our study found no evidence that 72 h of exposure to a sound level of ∼ 90 dB pk re 20 microPa of simulated wind turbine infrasound in double-blind conditions perturbed any physiological or psychological variable. None of the 36 people exposed to infrasound developed what could be described as WTS. Our study is unique because it measured the effects of infrasound alone on sleep. This study suggests that the infrasound component of WTN [wind turbine noise] is unlikely to be a cause of ill-health or sleep disruption, although this observation should be independently replicated.

217. The dose presented to these subjects “simulating a wind turbine infrasound signature” was questioned by IARO scientists, and correspondence with co-author R. Tonin was exchanged (in May 2023) to ascertain what “simulated wind turbine infrasound” meant.

303 See Appendix 1—Medical Sciences: IV. How is noise quantified?

304 See Appendix 2—Physics of Acoustics: I. What is Sound?

305 WASO = Wakefulness After Sleep Onset is the total number of minutes that an individual is awake after having initially fallen asleep.

306 WTS = Wind Turbine Syndrome. See: Pierpont N. (2009) Wind Turbine Syndrome: A Report on a Natural Experiment. K-Selected Books: Santa Fe, New Mexico, USA. https://www.researchgate.net/publication/265247204_Wind_Turbine_Syndrome_A_Report_on_a_Natural_Experiment

307 Marshall N, Cho G, Toelle BG, Tonin R, Bartlett DJ, et al. (2023) The Health Effects of 72 Hours of Simulated Wind Turbine Infrasound: A Double-Blind Randomised Crossover Study in Noise-Sensitive, Health Adults. Environmental Health Perspectives, 131(3): 1-10. https://pubmed.ncbi.nlm.nih.gov/36946580/ [Footnotes contained in the original text are not included.]

Health Report on a Rural Sheep Farm in Scotland

Page 66 of 169

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 http://IARO.org.nz

218. Regrettably, the material provided by co-author R. Tonin was regarded by IARO scientists as unsatisfactory, if “simulating a wind turbine infrasound signature” was the objective.308

219. Nevertheless, for the sake of scientific discussion, it will be temporarily accepted that the subjects of this study were actually presented with a properly simulated wind turbine infrasound signature.

220. The idea seems to have been to investigate immediate responses to the simulated wind turbine infrasound signature, but as measured by parameters that, perhaps, were not so relevant for assessing immediate responses.309, 310, 311, 312, 313, 314, 315

221. Another questionable practice was the selection of the “healthy individuals” as study subjects. To the understanding of IARO scientists, no evaluation was made regarding prior exposures 316 to infrasound and low frequency noise.317, 318

222. Marshall et al. explain the viewpoint that foundationally justifies their study:

People who suffer from WTS [Wind Turbine Syndrome 319] report that their symptoms begin quickly when they are exposed to infrasound from wind

308 The acoustic pattern used to simulate the wind turbine signal had a sawtooth profile, not the short-duration pulses of WTAS,

see Figure 3. A sawtooth-shaped wave has a quick onset, a slow decay, and only locally oscillates the air. WTAS has a rapid onset and decay, and ‘pumps the air’ (as proposed by Dr Stephan Kaula, Germany), rather than only causing the local oscillations that are typically seen in airborne, acoustic propagation phenomena.

309 See Appendix 4—Clinical & Biological Matters, Section 3-Occupational and Residential Exposures: I. Why are occupational exposures important to understand environmental exposures?

310 See Appendix 4—Clinical & Biological Matters, Section 3-Occupational and Residential Exposures: II. What extra-auditory medical conditions do noise-exposed workers develop?

311 See Appendix 4—Clinical & Biological Matters, Section 3-Occupational and Residential Exposures: III. Do the extra-auditory medical conditions seen in noise-exposed workers also emerge in residential infrasonic exposures?

312 Mohr GC, Cole JJN, Guild E, von Gierke HE. (1965) Effects of low-frequency and infrasonic noise on man. Aerospace Medicine, 36: 817-24.

313 Ponomarkov VI, Tysik A, Kudryavtseva VI, Barer AS. (1969) Biological action of intense wide-band noise on animals. Problems of Space Biology NASA TT F-529, 7(May): 307-9.

314 Castelo Branco NAA, Gomes-Ferreira P, Monteiro E, Costa e Silva A, Reis Ferreira J, Alves-Pereira M. (2003) Respiratory epithelia in Wistar rats after 48 hours of continuous exposure to low frequency noise. Journal of Pneumology, formerly Revista Portuguesa Pneumologia, IX (6): 474-79. https://pubmed.ncbi.nlm.nih.gov/15190432/

315 Castelo Branco NAA, Reis Ferreira J, Alves-Pereira M. (2007). Respiratory pathology in vibroacoustic disease: 25 years of research. Journal of Pneumology, formerly Revista Portuguesa Pneumologia, XIII (1): 129-135. https://pubmed.ncbi.nlm.nih.gov/17315094/

316 Including, foetal, childhood and young adult exposures in residential, occupational, and leisurely settings. See Appendix 1— Medical Sciences: II. What parameters are important when investigating the biological effects of exposures to physical agents of disease.

317 See Appendix 1—Medical Sciences: X. How are control populations selected for noise studies.

318 See Appendix 1—Medical Sciences: XI. What happens when control populations are incorrectly selected?

319 Pierpont N. (2009) Wind Turbine Syndrome: A Report on a Natural Experiment. K-Selected Books: Santa Fe, New Mexico, USA. https://www.researchgate.net/publication/265247204_Wind_Turbine_Syndrome_A_Report_on_a_Natural_Experiment

Health Report on a Rural Sheep Farm in Scotland

Page 67 of 169

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 http://IARO.org.nz

turbines and are then sustained.[320] Our scientifically robust study provides evidence to address this claim. The Australian NHMRC [National Health and Medical Research Council] report that gave rise to our study made note of this “absence of evidence” rather than concluding an “evidence of absence” owing to the lack of any laboratory-controlled double-blind experiments of sufficient duration and intensity to hypothetically induce WTS in a human.321

223. “Induce WTS in a human”? 322 As far as is understood by IARO scientists, WTS is not commonly viewed as an immediate effect of the exposure to this agent of disease.323

224. The expression “laboratory-controlled double-blind experiments of sufficient duration and intensity” as applied to the matter at hand is simultaneously unethical, dangerous, and unnecessary.324, 325

225. Is it the desire of the Australian NHMRC to expose subjects to a toxic agent—which is very difficult, if not impossible, to reproduce in laboratory settings—until some clearly severe health endpoint is observed? While tens of thousands of citizens are sitting in real- life laboratories being ‘accused’ of developing psychosomatic disorders? 326

226. This methodology is considered by IARO scientists to reflect sub-standard practices of Scientific Inquiry.

320 See Appendix 4—Clinical & Biological Matters, Section 1-Cellular and Tissue Biology. III. Biological tissues are viscoelastic—What does this

mean?

321 Marshall N, Cho G, Toelle BG, Tonin R, Bartlett DJ, et al. (2023) The Health Effects of 72 Hours of Simulated Wind Turbine Infrasound: A Double-Blind Randomised Crossover Study in Noise-Sensitive, Health Adults. Environmental Health Perspectives, 131(3): 1-10. https://pubmed.ncbi.nlm.nih.gov/36946580/ [Footnotes contained in the original text are not included.]

322 “The causes of this syndrome have been the subject of substantial international controversy. Proponents have contended that the symptoms that compose this syndrome are caused by low frequency subaudible infrasound generated by wind turbines. Critics have argued that these symptoms are psychological in origin and are attributable to nocebo effects. The Australian National Health and Medical Research Council Wind Farms and Human Health Reference Group concluded that the available evidence was not sufficient to establish which, if either, of these explanations is correct.” See: Marshall N, Cho G, Toelle BG, Tonin R, Bartlett DJ, et al. (2023) The Health Effects of 72 Hours of Simulated Wind Turbine Infrasound: A Double-Blind Randomised Crossover Study in Noise-Sensitive, Health Adults. Environmental Health Perspectives, 131(3): 1-10. https://pubmed.ncbi.nlm.nih.gov/36946580/

323 Pierpont N. (2009) Wind Turbine Syndrome: A Report on a Natural Experiment. K-Selected Books: Santa Fe, New Mexico, USA. https://www.researchgate.net/publication/265247204_Wind_Turbine_Syndrome_A_Report_on_a_Natural_Experiment

324 What kind of “laboratory-controlled double-blind experiments of sufficient duration and intensity” were conducted for asbestos contamination leading to asbestosis? Or for issues related to second-hand smoking, use of glyphosates, etc?

325 Alves-Pereira M, Rapley B, Bakker H, Summers R. (2019) Acoustics and Biological Structures. In: Abiddine Fellah ZE, Ogam E. (Eds) Acoustics of Materials. IntechOpen: London. DOI: 10.5772/intechopen.82761.

326 In the opinion of IARO scientists, had this study been performed on 3 groups of people, differentiated by the extent of their prior exposures (mild, moderate, or extensive), and, abiding by appropriate selection criteria of the study population, then, perhaps, statistically useful numbers could have been obtained, and scientifically useful results could have been achieved. The inability to reproduce ‘wind turbine infrasound’ under laboratorial conditions, however, would still render this study as irremediably flawed, while its overall design could be deemed ethically questionable.

Health Report on a Rural Sheep Farm in Scotland

Page 68 of 169

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 http://IARO.org.nz

227. In conclusion, in the opinion of IARO scientists, the effort expended by these authors to conduct this study is laudable (particularly given the position of the Australian NHMRC), even though, scientifically, within the realm of Medical Sciences and dose-response relationships, its results are inconsequential.

II. The Government-Sponsored Finnish Study

228. The 2021 study by Maijala et al.327 is based on the 169-page 2020 Governmental Report on a Research Project carried out by Maijala et al.328

229. The main objective was “to find out whether wind turbine infrasound has harmful effects on human health.”329

230. Table 3 lists the specific objectives of this 2020 Research Project.

Table 3. Specific objectives of the 2020 Research Project sponsored by the Government of Finland.330

A. To characterize wind turbine noise as an exposure

1 What are the full spectrum sound levels, down to 0.1 Hz, inside houses near the wind power plants?

2 What are the characteristics of the sound, both audible and inaudible infrasound?

B. To describe symptoms that are intuitively associated with infrasound from wind turbines, i.e., wind turbine infrasound related symptoms.

3 What is the prevalence of wind turbine infrasound related symptoms in the vicinity of wind power plants?

327 Maijala PP, Kurki I, Vainio L, Pakarinen S, Kuuramo C, et al. (2021) Annoyance, perception, and physiological effects of wind turbine infrasound. Journal

of the Acoustical Society of America, 149(4): 2238-2248. https://pubmed.ncbi.nlm.nih.gov/33940893/

328 Maijala P, Turunen A, Kurki I, Vainio L, Pakarinen S, et al. (2020) Infrasound does not explain symptoms related to wind turbines. Publications of the Finnish Government’s Analysis, Assessment and Research Activities, 2020:34. Prime Minister’s Office: Helsinki. https://julkaisut.valtioneuvosto.fi/handle/10024/162329

329 Maijala P, Turunen A, Kurki I, Vainio L, Pakarinen S, et al. (2020) Infrasound does not explain symptoms related to wind turbines. Publications of the Finnish Government’s Analysis, Assessment and Research Activities, 2020:34. Prime Minister’s Office: Helsinki. pp. 6. https://julkaisut.valtioneuvosto.fi/handle/10024/162329.

330 Maijala P, Turunen A, Kurki I, Vainio L, Pakarinen S, et al. (2020) Infrasound does not explain symptoms related to wind turbines. Publications of the Finnish Government’s Analysis, Assessment and Research Activities, 2020:34. Prime Minister’s Office: Helsinki. pp. 6-7. https://julkaisut.valtioneuvosto.fi/handle/10024/162329.

This is an excerpt from IARO Report (IARO24-5):

“Health Report on a Rural Sheep Farm in Scotland.”

[Critical Review of Maijala P, Turunen A, Kurki I, Vainio L, Pakarinen S, et al. (2020) Infrasound does not explain symptoms related to wind turbines. Publications of the Finnish Government’s Analysis, Assessment and Research Activities, 2020:34. Prime Minister’s Office: Helsinki.

https://julkaisut.valtioneuvosto.fi/handle/10024/162329]

(All Figures and Paragraphs referred to, but not included, in this excerpt can be found in the Full Report, available at iaro.org.nz)

Health Report on a Rural Sheep Farm in Scotland

Page 68 of 169

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 http://IARO.org.nz

227. In conclusion, in the opinion of IARO scientists, the effort expended by these authors to conduct this study is laudable (particularly given the position of the Australian NHMRC), even though, scientifically, within the realm of Medical Sciences and dose-response relationships, its results are inconsequential.

II. The Government-Sponsored Finnish Study

228. The 2021 study by Maijala et al.327 is based on the 169-page 2020 Governmental Report on a Research Project carried out by Maijala et al.328

229. The main objective was “to find out whether wind turbine infrasound has harmful effects on human health.”329

230. Table 3 lists the specific objectives of this 2020 Research Project.

Table 3. Specific objectives of the 2020 Research Project sponsored by the Government of Finland.330

A. To characterize wind turbine noise as an exposure

1 What are the full spectrum sound levels, down to 0.1 Hz, inside houses near the wind power plants?

2 What are the characteristics of the sound, both audible and inaudible infrasound?

B. To describe symptoms that are intuitively associated with infrasound from wind turbines, i.e., wind turbine infrasound related symptoms.

3 What is the prevalence of wind turbine infrasound related symptoms in the vicinity of wind power plants?

327 Maijala PP, Kurki I, Vainio L, Pakarinen S, Kuuramo C, et al. (2021) Annoyance, perception, and physiological effects of wind turbine infrasound. Journal

of the Acoustical Society of America, 149(4): 2238-2248. https://pubmed.ncbi.nlm.nih.gov/33940893/

328 Maijala P, Turunen A, Kurki I, Vainio L, Pakarinen S, et al. (2020) Infrasound does not explain symptoms related to wind turbines. Publications of the Finnish Government’s Analysis, Assessment and Research Activities, 2020:34. Prime Minister’s Office: Helsinki. https://julkaisut.valtioneuvosto.fi/handle/10024/162329

329 Maijala P, Turunen A, Kurki I, Vainio L, Pakarinen S, et al. (2020) Infrasound does not explain symptoms related to wind turbines. Publications of the Finnish Government’s Analysis, Assessment and Research Activities, 2020:34. Prime Minister’s Office: Helsinki. pp. 6. https://julkaisut.valtioneuvosto.fi/handle/10024/162329.

330 Maijala P, Turunen A, Kurki I, Vainio L, Pakarinen S, et al. (2020) Infrasound does not explain symptoms related to wind turbines. Publications of the Finnish Government’s Analysis, Assessment and Research Activities, 2020:34. Prime Minister’s Office: Helsinki. pp. 6-7. https://julkaisut.valtioneuvosto.fi/handle/10024/162329.

Health Report on a Rural Sheep Farm in Scotland

Page 69 of 169

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 http://IARO.org.nz

4 What factors are associated with wind turbine infrasound related symptoms?

C. To study how infrasound produced by wind turbines affects humans, in particular, perception, annoyance, and physiological responses

5 Can low-frequency and infrasound wind turbine noise be perceived at typical and at extreme noise levels?

6 What is the dependence between the depth of amplitude modulation and annoyance at low frequencies?

7 Does infrasound increase reported annoyance and psychophysiological responses?

8 What is the reactivity of the autonomic nervous system (ANS) to audible wind turbine sounds and its infrasound?

9 Are individuals who attribute their symptoms to wind turbines more sensitive to infrasound? Are they more able to detect infrasound and do they experience more annoyance compared to controls?

231. Objectives A1 and A2 were accomplished, and Figure 7 shows a representative example of the identified ‘dose.’

Health Report on a Rural Sheep Farm in Scotland

Page 70 of 169

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 http://IARO.org.nz

Figure 7. Representative example of the noise characterization (Raahe, indoors, 600-second sample). 331 LZ levels refer to unweighted dB values. LG refers to G-weighted values.332 LA refers to A-weighted values. Maximum and minimum LZ values are shown as curves.

232. Figure 7 shows a one-third-octave-band segmentation of the acoustic spectrum (similar to that shown in Figure 2). The solid black curve (LZ max) shows the highest sound pressure levels measured in unweighted dB.

233. There is no cut-off of spectral data as was seen in Figure 6 (i.e., the lower limiting frequency is 0.1 Hz and not 10 Hz), but there is also no recognition of a “wind turbine infrasound signal” as in the previous Marshall et al. study (see Paragraph 214). It was however recognized that “the most important frequencies were less than 2 Hz.”333

234. Objectives B3 and B4 (see Table 3) were more difficult to achieve, as “infrasound related symptoms” were established by questionnaires and telephone calls. While these types of surveys may have a certain usefulness, their direct results cannot be considered as a measure of Response within the realm of the Medical Sciences’ dose-response

331 Maijala P, Turunen A, Kurki I, Vainio L, Pakarinen S, et al. (2020) Infrasound does not explain symptoms related to wind turbines. Publications of the

Finnish Government’s Analysis, Assessment and Research Activities, 2020:34. Prime Minister’s Office: Helsinki. pp. 21. https://julkaisut.valtioneuvosto.fi/handle/10024/162329.

332 See Appendix 2—Physics of Acoustics: V. Can infrasound be measured in dBC or dBG?

333 Maijala P, Turunen A, Kurki I, Vainio L, Pakarinen S, et al. (2020) Infrasound does not explain symptoms related to wind turbines. Publications of the Finnish Government’s Analysis, Assessment and Research Activities, 2020:34. Prime Minister’s Office: Helsinki. pp. 77. https://julkaisut.valtioneuvosto.fi/handle/10024/162329.

Health Report on a Rural Sheep Farm in Scotland

Page 71 of 169

International Acoustics Research Organization

37 Weston Ave, Palmerston North, New Zealand T +64 21 033 6528 http://IARO.org.nz

relationship,334 nor as per the WHO definition of noise-induced adverse health effects (see Paragraph 189).

235. Furthermore, there seems to not have been any stratification of the study population regarding prior noise exposure histories.335

236. Objectives C5 through C9 used “provocation experiments” conducted in an “infrasound chamber” whereby “systematically selected samples from real wind turbine sounds from wind power plant areas where inhabitants report symptoms associated with wind turbine infrasound or sound were used as stimuli.”336

237. As with the study by Marshall et al. (Paragraphs 224 to 226), it is not entirely understood why there is a perceived need to subject individuals in laboratory to a potentially noxious agent (which is very difficult, if not impossible, to reproduce under laboratorial conditions), while tens of thousands of individuals are living in ‘real-life laboratories,’ awaiting an objective, clinical observational study on behalf of the competent authorities.337

III. Intuitive symptoms

238. In the third study of this series, the goal of Turunen et al.338 was to assess “the prevalence and severity of these wind turbine infrasound related symptoms:”

No matter what the true cause for the symptoms is, it is clear that symptoms are real and lead to worry, decreased quality of life, and potentially further to deteriorated health. High prevalence of this kind of phenomenon could be a serious threat to public health. The aim of this questionnaire study was to describe symptoms intuitively associated with infrasound from wind turbines.339

334 See Appendix 1—Medical Sciences: VIII. How is ‘Response’ measured?

335 See Appendix 1—Medical Sciences: II. What parameters are important when investigating the biological effects of exposures to physical agents of disease?

336 Maijala P, Turunen A, Kurki I, Vainio L, Pakarinen S, et al. (2020) Infrasound does not explain symptoms related to wind turbines. Publications of the Finnish Government’s Analysis, Assessment and Research Activities, 2020:34. Prime Minister’s Office: Helsinki. pp. 36 and 40. https://julkaisut.valtioneuvosto.fi/handle/10024/162329.

337 Although it is unclear to IARO scientists who (or what agency) could be classified as ‘the competent authorities.’

338 Please note that the authors of this study are the same as those of the Finnish Governmental study by Maijala et al. (see Paragraph 228), and the data collected through questionnaires and telephone calls in the Maijala et al. study are the same data used in this study. See: Turunen AW, Tittanen P, Yli-Tuomi T, Taimisto P, Lanki T. (2021) Symptoms intuitively associated with wind turbine infrasound. Environmental Research, 192: 1-9. https://pubmed.ncbi.nlm.nih.gov/33131679/

339 Turunen AW, Tittanen P, Yli-Tuomi T, Taimisto P, Lanki T. (2021) Symptoms intuitively associated with wind turbine infrasound. Environmental Research, 192: 1-9. https://pubmed.ncbi.nlm.nih.gov/33131679/

Review of

Wind Turbine Health Impact Study:

Report of Independent Expert Panel

as prepared for

Massachusetts Department of Environmental Protection

Massachusetts Department of Public Health

By

Mariana Alves-Pereira, Associate Professor

Faculty of Economics and Management

School of Health Sciences

Universidade Lusofona

Lisbon, Portugal

March 2012

Page 2 of 9

Background

As a leading expert on the biological response to low frequency noise exposure (see brief

biographical background offered at the end of this document), I was requested to provide a

review of the Jan 2012 Report of Independent Expert Panel, prepared for the

Massachusetts Department of Environmental Protection (MassDEP) and Department of

Public Health (MDPH), titled "Wind Turbine Health Impact Study".

Disclaimer

a) The author of this review is not party to anti-technology sentiments;

b) Wind turbines are considered by this author as welcome additions to modern

technological society;

c) The review provided herein has one, and only one, agenda - that of pure scientific

inquiry;

d) In no way can or should this review be construed as a document arguing for or against

the implementation of wind turbines;

e) There are no commercial, financial or professional agreements (contractual or

otherwise) between the author of this review and any persons or parties involved in the

wind turbine sector or persons or parties who stand against the implementation of wind

turbines;

f) This review was provided pro bono.

Goal

To provide a review of the aforementioned Report, within the author’s area of expertise

and therefore, exclusively focused on the infrasound and low frequency noise health

issues claimed to be associated with wind turbines (WT) operations.

Panel Charge

The Panel who authored the Report was charged with several tasks, the first of which is

succinctly stated as follows: