Understanding The Evidence: Wind Turbine Noise The Expert Panel On Wind Turbine Noise and Human Health

Understanding
the
Evidence: Wind Turbine Noise
The Expert Panel on Wind Turbine Noise

and Human Health
Science Advice in the Public Interest
UNDERSTANDING THE EVIDENCE: WIND TURBINE NOISE

The Expert Panel on Wind Turbine Noise and Human Health
ii
Understanding the Evidence: Wind Turbine Noise
THE COUNCIL OF CANADIAN ACADEMIES

180 Elgin Street, Suite 1401, Ottawa, ON, Canada K2P 2K3
Notice: The project that is the subject of this report was undertaken with the approval
of the Board of Governors of the Council of Canadian Academies. Board members are
drawn from the Royal Society of Canada (RSC), the Canadian Academy of Engineering
(CAE), and the Canadian Academy of Health Sciences (CAHS), as well as from the
general public. The members of the expert panel responsible for the report were selected
by the Council for their special competencies and with regard for appropriate balance.
This report was prepared for the Government of Canada in response to a request
from the Minister of Health. Any opinions, findings, or conclusions expressed in this
publication are those of the authors, the Expert Panel on Wind Turbine Noise and
Human Health, and do not necessarily represent the views of their organizations of
affiliation or employment.
Library and Archives Canada Cataloguing in Publication

Understanding the evidence : wind turbine noise / the Expert Panel on Wind Turbine
Noise and Human Health. Includes bibliographical references. Electronic monograph
in PDF format.
ISBN 978-1-926522-07-4 (pdf)
1. Wind turbinesCanadaNoise. 2. Wind turbinesHealth aspectsCanada.
3. Wind turbinesTechnological innovations. I. Council of Canadian Academies, issuing
body II. Council of Canadian Academies.
Expert Panel on Wind Turbine Noise and Human Health, author
TJ828.U54 2015
621.45
C2015-901475-1
This report should be cited as: Council of Canadian Academies, 2015. Understanding the
Evidence: Wind Turbine Noise. Ottawa (ON): The Expert Panel on Wind Turbine Noise
and Human Health, Council of Canadian Academies.
Disclaimer: The internet data and information referenced in this report were correct,
to the best of the Councils knowledge, at the time of publication. Due to the dynamic
nature of the internet, resources that are free and publicly available may subsequently
require a fee or restrict access, and the location of items may change as menus and web
pages are reorganized.
2015 Council of Canadian Academies
Printed in Ottawa, Canada
This assessment was made possible with

the support of the Government of Canada.
The Council of Canadian Academies
iii
The Council of Canadian Academies

Science Advice in the Public Interest
The Council of Canadian Academies (the Council) is an independent,
not-for-profit organization that supports independent, science-based, authoritative
expert assessments to inform public policy development in Canada. Led by a
12-member Board of Governors and advised by a 16-member Scientific Advisory
Committee, the Councils work encompasses a broad definition of science,
incorporating the natural, social, and health sciences as well as engineering
and the humanities. Council assessments are conducted by independent,
multidisciplinary panels of experts from across Canada and abroad. Assessments
strive to identify emerging issues, gaps in knowledge, Canadian strengths, and
international trends and practices. Upon completion, assessments provide
government decision-makers, researchers, and stakeholders with high-quality
information required to develop informed and innovative public policy.
All Council assessments undergo a formal report review and are published and
made available to the public free of charge in English and French. Assessments
can be referred to the Council by foundations, non-governmental organizations,
the private sector, or any level of government.
The Council is also supported by its three founding Member Academies:
The Royal Society of Canada (RSC) is the senior national body of distinguished
Canadian scholars, artists, and scientists. The primary objective of the RSC is
to promote learning and research in the arts and sciences. The RSC consists
of nearly 2,000 Fellows men and women who are selected by their peers
for outstanding contributions to the natural and social sciences, the arts, and
the humanities. The RSC exists to recognize academic excellence, to advise
governments and organizations, and to promote Canadian culture.
The Canadian Academy of Engineering (CAE) is the national institution
through which Canadas most distinguished and experienced engineers provide
strategic advice on matters of critical importance to Canada. The Academy
is an independent, self-governing, and non-profit organization established
in 1987. Fellows are nominated and elected by their peers in recognition of
their distinguished achievements and career-long service to the engineering
profession. Fellows of the Academy, who number approximately 600, are
committed to ensuring that Canadas engineering expertise is applied to the
benefit of all Canadians.
iv
The Canadian Academy of Health Sciences (CAHS) recognizes individuals of

great achievement in the academic health sciences in Canada. Founded in 2004,
CAHS has approximately 400 Fellows and appoints new Fellows on an annual
basis. The organization is managed by a voluntary Board of Directors and a
Board Executive. The main function of CAHS is to provide timely, informed,
and unbiased assessments of urgent issues affecting the health of Canadians. The
Academy also monitors global health-related events to enhance Canadas state
of readiness for the future, and provides a Canadian voice for health sciences
internationally. CAHS provides a collective, authoritative, multidisciplinary
voice on behalf of the health sciences community.
www.scienceadvice.ca
@scienceadvice
Expert Panel on Wind Turbine Noise and Human Health
Expert Panel on Wind Turbine Noise and Human Health

Tee L. Guidotti (Chair), Fulbright Visiting Chair, Institute for Science, Society
and Policy, University of Ottawa (Ottawa, ON)
Hugh W. Davies, Associate Professor, Occupational and Environmental
Health Division, School of Population and Public Health, University of British
Columbia (Vancouver, BC)
Yves Gagnon, Professor of Engineering, Universit de Moncton (Edmundston, NB);
Adjunct Professor, Mechanical Engineering, cole de technologie suprieure
(Montral, QC)
Christian Gigure, Professor, Audiology and Speech-Language Pathology
Program, School of Rehabilitation Sciences, University of Ottawa (Ottawa, ON)
Sheryl Grace, Associate Professor, Department of Aerospace and Mechanical
Engineering, Boston University (Boston, MA)
Robert V. Harrison, Professor and Vice-Chair Research, Department of
Otolaryngology, Head and Neck Surgery; Professor, Department of Physiology,
Institute of Biomaterials and Biomedical Engineering and Institute of Medical
Science, University of Toronto; Senior Scientist, Program for Neuroscience and
Mental Health, Hospital for Sick Children (Toronto, ON)
Brian Howe, President, HGC Engineering (Toronto, ON)
David A. Johnson, Professor, Department of Mechanical and Mechatronics
Engineering, University of Waterloo (Waterloo, ON)
Kerstin Persson Waye, Professor, Occupational and Environmental Medicine,
University of Gothenburg (Gothenburg, Sweden)
Jennifer D. Roberts, Assistant Professor, Occupational and Environmental
Health Sciences, F. Edward Hbert School of Medicine, Department of Preventive
Medicine and Biometrics (Division of Occupational and Environmental Health
Sciences), Uniformed Services University of the Health Sciences (Bethesda, MD)
vi
Message from the Chair

Wind turbines are a relatively new addition to the Canadian landscape and
energy mix. Although wind power in the form of windmills is a common sight
on the farm and on the Prairies, wind turbines on a commercial scale are a
modern phenomenon. Their recent growth in both number and size has raised
questions regarding potential health impacts on nearby residents.
In response to public concern, the Government of Canada, through the Minister
of Health, asked the Council to determine if there is evidence to support a
causal association between exposure to wind turbine noise and health effects.
This report presents the expertise and contributions of a panel of 10 experts from
Canada and abroad, drawn from fields as diverse as engineering and medical
science, including myself as Chair. I am deeply grateful for my colleagues on
the Panel who contributed their substantial time and effort to ensure the depth
and quality of this report. I would also like to extend my appreciation to the
nine reviewers who assisted the Panel and whose efforts significantly improved
the earlier version of the report.
Before this Panel was assembled, Health Canada had started, in 2012, a large
cross-epidemiological study to measure potential health outcomes of exposure
to sound from wind turbines in areas of Canada where wind energy is used. The
preliminary results from this study became available as the Panel was concluding
its deliberations and finalizing this report (November 2014). Although results
from this study were not included in the body of evidence assessed by the Panel,
they are summarized and discussed in this report. I would like to assure readers
that Health Canada was not involved in or privy to the Panels deliberations
before publication, nor was the Department given access to drafts of this report.
Message from the Chair
vii
Finally, the Panel is grateful for the support it received from the staff members
of the Council of Canadian Academies who were assigned to this assessment.
They are a dedicated and accomplished team of scholars and professionals,
and it has been an honour and a pleasure to work with them.
I would like to extend my personal appreciation to the Panel members for
their cooperation, rigour, patience, and devotion to the task.
Tee L. Guidotti
Chair, Expert Panel on Wind Turbine Noise and Human Health
viii
Project Staff of the Council of Canadian Academies

Assessment Team:

Emmanuel Mongin, Associate Program Director

Stefan Jungcurt, Research Associate
Jonathan Whiteley, Research Associate
Tess Lin, Program Coordinator
Weronika Zych, Program Coordinator
Naomi Shuman, Intern
With assistance from:

Leela Steiner, Consultant

Carolyn Brown, Editor
Jody Cooper, Copyeditor
Accurate Design & Communication Inc., Report Design
Report Review
ix
Report Review
This report was reviewed in draft form by the individuals listed below a
group of reviewers selected by the Council of Canadian Academies for their
diverse perspectives, areas of expertise, and broad representation of academic,
industrial, policy, and non-governmental organizations.
The reviewers assessed the objectivity and quality of the report. Their
submissions which will remain confidential were considered in full by
the Panel, and many of their suggestions were incorporated into the report.
They were not asked to endorse the conclusions, nor did they see the final
draft of the report before its release. Responsibility for the final content of this
report rests entirely with the authoring Panel and the Council.
The Council wishes to thank the following individuals for their review of
this report:
Prudence Allen, Director and Associate Professor, National Centre for Audiology,
Western University (London, ON)
Franois Benoit, Scientific and Administrative Lead, National Collaboration
Centre for Healthy Public Policy, Institut national de sant publique du Qubec
(Montral, QC)
Arline L. Bronzaft, Consultant and Professor Emerita, City University of
New York (New York, NY)
Jeffrey M. Ellenbogen, Assistant Professor of Neurology, Johns Hopkins
University (Baltimore, MD)
Wendy Heiger-Bernays, Associate Professor of Environmental Health, Boston
University School of Public Health (Boston, MA)
Stefan Oerlemans, Senior Key Expert Aeroacoustics, Siemens Wind Power
(Brande, Denmark)
Stphane Perron, Physician and Clinical Adjunct Professor, Montral
Public Health Department, Universit de Montral and McGill University
(Montral, QC)
Bo Sndergaard, Senior Consultant, Department of Acoustics, Grontmij
(Aarhus, Denmark)
Jian Wang, Full Professor, School of Human Communication Disorders,
Dalhousie University (Halifax, NS)
The report review procedure was monitored on behalf of the Councils Board
of Governors and Scientific Advisory Committee by Susan A. McDaniel, FRSC,
Director, Prentice Institute; Canada Research Chair in Global Population and Life
Course; Prentice Research Chair in Global Population and Economy; Professor of
Sociology, University of Lethbridge (Lethbridge, AB). The role of the report review
monitor is to ensure that the Panel gives full and fair consideration to the submissions
of the report reviewers. The Board of the Council authorizes public release of an
expert panel report only after the report review monitor confirms that the Councils
report review requirements have been satisfied. The Council thanks Dr. McDaniel
for her diligent contribution as report review monitor.
Janet W. Bax, Interim President

Council of Canadian Academies
Executive Summary
xi
Executive Summary
Demand for renewable energy, including wind power, is expected to continue
to grow both in Canada and globally for the foreseeable future. The wind
energy sector in Canada has grown at an ever-increasing pace since the 1990s,
and Canada is now the fifth-largest market in the world for the installation
of new wind turbines. As the sector grows, the wind turbines being installed
are getting more powerful. The first megawatt-scale turbines were installed in
Canada in 2004, with 3 megawatt models arriving in 2008; larger models up
to 7.5 megawatt are currently being tested internationally. To produce this
power, turbines have also increased in size. As wind turbines become a more
common feature of the Canadian landscape, this new source of environmental
sound has raised concerns about potential health effects on nearby residents.
Determining whether wind power causes adverse health effects in people is
therefore important so that all Canadians can equitably share in the benefits
of this technology.
THE C HA R G E T O T H E PANE L
In response to growing public concern about the potential health effects of

wind turbine noise, the Government of Canada, through the Minister of Health
(the Sponsor), asked the Council of Canadian Academies (the Council) to
conduct an assessment of the question:
Is there evidence to support a causal association between exposure to wind turbine noise
and the development of adverse health effects?
The Charge also includes the following sub-questions:
Are there knowledge gaps in the scientific and technological areas that need to be
addressed in order to fully assess possible health impacts from wind turbine noise?
Is the potential risk to human health sufficiently plausible to justify further research
into the association between wind turbine noise exposure and the development
of adverse health effects?
How does Canada compare internationally with respect to prevalence and nature of
reported adverse health effects among populations living in the vicinity of commercial
wind turbine establishments?
Are there engineering technologies and/or other best practices in other jurisdictions that
might be contemplated in Canada as measures that may minimize adverse community
response towards wind turbine noise?
xii
The Panel defined health in a way that is consistent with the World Health
Organizations concept of health: a state of complete physical, mental
and social well-being and not merely the absence of disease or infirmity
(WHO, 1946). The Panel interpreted noise to include both objective measures
of acoustic signals in the environment (sound), as well as subjective perceptions
of sound sensations that are unwanted by the listener (noise). As there are a
variety of wind turbines available worldwide, with differing sound characteristics,
the Panel focused specifically on the type that constitutes almost all of the
installed turbines in Canada: modern, three-bladed, tower-mounted, utilityscale (500 kilowatt capacity or more), upwind, horizontal-axis wind turbines
that were land-based.
THE PA N E L S A P P R OACH
To respond to the Charge, the Panel used an evidence-based approach to

identify and review relevant research. First, the Panel identified more than
30 symptoms and health outcomes that have been attributed to exposure to
wind turbine noise, based on a broad survey of peer-reviewed and grey literature,
web pages, and legal decisions.
Empirical evidence related to any associations between these health outcomes
and exposure to wind turbine noise was then collected from several sources,
including peer-reviewed journal articles, conference papers, and grey literature.
More than 300 publications were found through a comprehensive search, and
these were narrowed down to 38 relevant studies related to the health effects of
wind turbine noise. The body of evidence concerning each health outcome was
appraised and assessed according to Bradford Hills guidelines for causation,
and summarized using standard terms adopted from the International Agency
for Research on Cancer (IARC). The major steps of the Panels approach are
illustrated in Figure 1.
KEY F I N D I N G S
Based on its expertise and review of empirical research, the Panel made findings
in the following areas:
Acoustic characteristics of wind turbine noise;
Evidence of causal relationships between exposure to wind turbine noise
and adverse health effects;
Knowledge gaps and further research; and
Promising practices to reduce adverse community response.
Other aspects of the Charge, such as the prevalence of adverse health outcomes
in Canada, could not be answered because of a lack of data.
xiii
Executive Summary
Literature Specific to Wind Turbines and Human Health

Environmental
Noise and Human
Health Literature
Empirical
Research
Critical
Appraisal
Reviews and
Discussions
Legal Decisions
(ERTs)
Web Pages
Reported Adverse
Health Effects
Conceptual Framework
Weight and Summary of Evidence
of a Causal Relationship
Sufficient
Inadequate
Limited
Lack of causality
Conceptual Framework:
Conclusions
Figure 1
Evidence Assessment Process
Brown lines show information used in defining potential health outcomes and building a model
of pathogenic mechanisms; blue lines show the literature review process with reference to causal
associations between wind turbine noise and each potential health effect.
A C O US TI C CH ARAC T E RI S T I C S OF W IND TU R B INE NOISE
1. Sound from wind turbines is complex and variable

Like sound from any source, wind turbine noise can be described by frequency
components (which determine pitch), sound pressure levels (which determine
loudness), and the way both of these change over time. Sound from wind turbines
is highly complex and variable, but has some characteristics that are similar
to other sources of community noise, such as road and airport traffic noise:
Sound from wind turbines is broadband, composed of sound over a broad
range of frequencies.
The overall sound pressure levels outdoors vary greatly depending on distance,
wind speed, and transmission from the source to the receiver.
However, higher frequencies tend to be reduced indoors and with increasing
distance, leading to an emphasis on lower frequencies.
It is amplitude modulated, with sound levels changing over time.
Wind turbines also emit sound with the following characteristics, which are
less common than other sources of community noise:
Sounds from wind turbines may extend down to the infrasonic range and,
in some cases, may include peaks or tonal components at low frequencies.
xiv
Sound emissions from a wind turbine increase with greater wind speed at
the height of the blades, up to the turbines rated wind speed (speed at which
it generates maximum power), above which sound does not increase.
Sound from wind turbines can exhibit periodic amplitude modulation, often
described as a swishing or thumping sound. The causes and consequences
of this periodic amplitude modulation are areas of ongoing research, as
wind turbine designers and manufacturers seek ways to reduce or mitigate it.
Most sound from wind turbines is produced by interactions between the
surface of the blade and the air flowing over it (aerodynamic processes), which
is strongest near but not at the blade tips. Mechanical noise from the
physical movements of the gearbox, generator, and other components produces
low-frequency tones in some cases.
2. Standard methods of measuring sound may not capture the
low-frequency sound and amplitude modulation characteristic
of wind turbine noise
Measurement of sound for health surveillance and research uses standard
methods. The most commonly used methods include A-weighting, which
emphasizes the frequencies according to human hearing sensitivity, and deemphasizes low and very high frequencies. Although A-weighted measurement
is an essential method, it may fail to capture the low-frequency components
of wind turbine sound. In addition, measurement is often averaged over time
(Leq), which does not convey changes in sound pressure levels occurring in
short periods (for example, within a second). Time-averaged measurement
may thus fail to capture amplitude modulation.
A-weighted measurements are an important first step in determining peoples
exposure to audible sound in most cases, but more detailed measurements may
be necessary in order for researchers to fully investigate the potential health
impact of specific sources of wind turbine noise. The metrics of sound exposure
most relevant to potential health outcomes are not completely understood,
however, and remain an important area for further research.
WIND TUR B I N E N O I S E AND ADV E R S E H E ALTH EFFEC TS
The relevant empirical evidence was reviewed and weighted in order to determine
the strength of evidence for a causal link between wind turbine noise and each
potential adverse health effect.
Executive Summary
xv
3. The evidence is sufficient to establish a causal relationship between

exposure to wind turbine noise and annoyance
The evidence consistently shows a positive relationship between outdoor wind
turbine noise levels and the proportion of people who report high levels of
annoyance. However, many factors can modify the strength of this relationship,
such as a persons attitudes toward wind turbines and any economic benefits the
person derives from them. As well, visual and noise effects of wind turbines are
difficult to isolate from each other. The current state of the evidence does not
allow for a definite conclusion about whether annoyance is caused by exposure
to wind turbine noise alone, or whether factors such as visual impacts and
personal attitudes modify the noise-annoyance relation and to what extent,
since the studies completed to date do not measure these factors independently
of each other. It is also unclear which sound characteristics contribute to longterm chronic annoyance, although low-frequency components and periodic
amplitude modulation have been investigated as likely candidates.
4. There is limited evidence to establish a causal relationship between
exposure to wind turbine noise and sleep disturbance
The available evidence suggests that a direct causal relationship or an indirect
(via annoyance) relationship between exposure to wind turbine noise and sleep
disturbance might exist. While sleep disruption has been investigated in several
studies, the resulting evidence base is smaller than that which examines the
relationship between wind turbine noise and annoyance.
5. The evidence suggests a lack of causality between exposure to wind
turbine noise and hearing loss
There is convincing evidence that exposure to wind turbine noise at typical levels
associated with regulated noise limits and setbacks (distance from structures)
does not cause loss of hearing, even over a lifetime of exposure.
6. The Panel found inadequate evidence of a direct causal relationship
between exposure to wind turbine noise and stress, although stress
has been linked to other sources of community noise
Available evidence suggests that a direct or indirect mechanism between exposure
to wind turbine noise and stress might exist, similar to the finding for sleep
disturbance, but the evidence lacks methodological and statistical strength.
Stress has been identified as a risk factor for a number of other diseases, such
as cardiovascular diseases, in the context of long-term exposure to community
noise from other sources, such as road, rail, and air traffic. The current evidence
related to exposure to wind turbine noise and stress is inconsistent, however.
xvi
7. For all other health effects considered (fatigue, tinnitus, vertigo,

nausea, dizziness, cardiovascular diseases, diabetes, etc.), the evidence
was inadequate to come to any conclusion about the presence or
absence of a causal relationship with exposure to wind turbine noise
Hypertension and other cardiovascular diseases, diabetes, tinnitus, cognitive or
task performance, psychological health, and health-related quality of life have
all been the subject of empirical, population-based, wind-turbine noise studies.
The evidence, however, was inconsistent or the studies had methodological
limitations preventing the determination of a causal relationship between these
effects and exposure to wind turbine noise. None of the other health effects
considered have been the subject of a population-level study or experiments
in the context of wind turbine noise. Therefore, the evidence for a causal
association is largely lacking for these other effects.
Factors Influencing Responses
External Stimuli
Sound Pressure Level
Frequency Characteristics
Tonal Components
Amplitude Modulation
Modifying Factors
Noise Sensitivity
Pre-Existing Conditions
Personal Attitudes
Visual Impacts
Mechanisms
Direct
ReceptorMediated
(hearing,
vestibular)
Health Outcomes
Nervous System
Cognitive
Performance
Tinnitus
Health-Related
Quality of Life
Sleep
Disturbance
Endocrine System
Diabetes
Stress
Perceptual
Cognitive and
Emotional
Response
Sufficient evidence of a causal relationship

Limited evidence of a causal relationship
Evidence from the environmental noise literature
Annoyance
Psychological
Health
Cardiovascular
System
Grey boxes: inadequate evidence of a causal

relationship for the specific health outcome
Unlinked boxes: no clear mechanism linking
the health outcome to wind turbine noise
Figure 2
Summary of Evidence for Causal Pathways Between Exposure to Wind Turbine Noise
and Adverse Health Effects
Executive Summary
xvii
Conclusions about causal relationships are therefore lacking for most of the health
effects postulated in a wide variety of sources reviewed by the Panel, mainly as a result
of lack of evidence or problems with the quality of evidence. However, research
on environmental noise has shown that annoyance can be a contributing factor or
precursor to adverse health effects such as sleep disturbance, stress and cardiovascular
diseases. The Panel thus developed a conceptual framework of pathways through
which sound from wind turbines could plausibly result in health outcomes.
Figure 2 shows this framework and summarizes the Panels findings on the potential
causal pathways between exposure to wind turbine noise and the development
of adverse health effects, or the exacerbation of existing health conditions.
KN O W LE D G E GAP S AND F U RT H E R RESEAR C H
8. Knowledge gaps prevent a full assessment of public health effects

The Panel identified specific knowledge gaps for each health condition studied,
where specific types of evidence would help clarify the strength of associations,
minimize bias, or eliminate possible confounding factors with respect to exposure
to wind turbine noise. For example, it is unclear whether the possible pathway
that could lead to sleep disturbance or stress is the direct result of exposure to
wind turbine noise or of annoyance as a mediating factor.
Most existing epidemiological studies of wind turbine noise lack sufficient
power to detect small changes in the risk of adverse health effects, or were
designed in a way that could not rule out bias in responses or adequately control
confounding factors. The Panel also identified an absence of longitudinal studies.
The Panel stresses that there is a paucity of research on sensitive populations,
such as children and infants and people affected by clinical conditions that
may lead to an increased sensitivity to sound.
The use of adequate methods and procedures for measuring and modelling
sound exposure from wind turbines, particularly indoors, would improve the
quality of future studies on adverse health effects (see Key Finding 2).
9. Research on long-term exposure to wind turbine noise would
provide a better understanding of the causal associations between
wind turbine noise exposure and certain adverse health effects
Chronic annoyance and sleep disturbance have been linked to stress responses
in studies of long-term exposure to other sources of noise, such as air and road
traffic. Furthermore, these health effects are themselves risk factors for other
diseases, such as cardiovascular diseases, which have previously been associated
xviii
with long-term exposure to other sources of community noise. Given the

burden of cardiovascular diseases on society and Canadas health care system,
further research on the long-term effects of exposure to wind turbine noise, in
particular on stress and sleep disturbance, would provide more data to assess
the health effects of wind turbine noise. Finally, the Panel stresses that the
available evidence does not allow conclusions with regard to the prevalence of
annoyance or other health effects within the population exposed to sound from
wind turbines in Canada. Further research and surveillance would provide a
better understanding of this prevalence, both in those exposed to wind turbine
noise and in the general population.
PROMI S I N G P R A C T I CE S AND T E C H NOL OGIES TO R EDU C E
ADVE R S E C O M M U NI T Y R E S P ONS E T O W I ND TU R B INE NOISE
10. Technological development is unlikely to resolve, in the short term,

the current issues related to perceived adverse health effects of wind
turbine noise
Wind turbine designs, modifications, and technology that could reduce sound
emissions are currently being explored by wind turbine manufacturers. Ongoing
technological development has contributed to lower sound emissions for
turbines of a given size over the previous generation of turbines, with further
improvements expected. Other factors such as power output favour larger
turbines, however, which can offset overall reductions in sound emissions per
kilowatt of electricity produced.
11. Impact assessments and community engagement provide
communities with greater knowledge and control over wind energy
projects and therefore help limit annoyance
Equity and fairness have been crucial for the acceptance of wind turbines in
many communities, with perceived loss of social justice and disempowerment
being significant barriers to acceptance in some cases. One important regulatory
approach is to conduct a noise impact assessment of any proposed project;
several Canadian provinces and other countries require such an assessment.
In some of the international practices reviewed by the Panel, wind energy
developers engaged in consultation and communication with local authorities
and residents beginning at an early stage of project development, through all
stages of implementation, and even after installation. Community engagement
helps to inform and educate local residents, as well as involve them in a wind
energy project with the goal of fostering social acceptance.
Executive Summary
xix
Wind turbines are a progressively familiar sight in Canada and contribute an

increasing share of the electricity consumed in Canada. Concerns over the health
effects of wind turbine noise have been expressed in many ways but rarely with
detailed, reproducible, and rigorous data sufficient to support a conclusion on
either causation or magnitude of any potential health effect. The Panels final
report is an attempt to objectively and rigorously review empirical research on
the causal link between wind turbine noise and adverse health effects, as well as
potential solutions to noise-related issues contemplated elsewhere, all of which
may help in addressing concerns about wind turbine noise in Canada. The
report is intended not only as a tool to inform decision-making and academic
research on the subject, but also to inform the continuing dialogue across
Canada and internationally, and across many sectors, about wind turbine noise
and adverse human health effects.
xx
Table of Contents
1
Introduction............................................................................. 1
1.1
1.2
1.3
1.4
Charge to the Panel..........................................................................3

Interpreting the Charge ..................................................................3
Health Canada Wind Turbine Noise and
Health Study.....................................................................................5
Organization of the Report .............................................................5
Wind Energy Context.............................................................. 7
2.1
2.2
2.3
2.4
2.5
Wind Turbine Design........................................................................9

Canadian and International Contexts...........................................14
Health Surveillance........................................................................17
Noise Limits and Setbacks..............................................................18
Chapter Summary...........................................................................19
Wind Turbine Acoustics ........................................................ 20
3.1
3.2
3.3
3.4
Fundamentals of Acoustics.............................................................22
Sources of Sound from Wind Turbines..........................................32
Wind Turbine Sound Levels and Spectra......................................34
Chapter Summary...........................................................................40
Hearing and Perception of Sound........................................ 42
4.1
4.2
4.3
4.4
4.5
The Auditory Pathway.....................................................................44

Non-Auditory Pathways...................................................................45
Basic Auditory Perception and Mechanisms.................................46
Sound Perception...........................................................................49
Chapter Summary...........................................................................50
Assessment of Evidence........................................................ 51
5.1
5.2
5.3
Recent Reviews of Wind Turbine Noise and Health.....................53

Proposed Adverse Health Effects...................................................54
Prevalence of Reported Health Effects and
Noise Complaints in Canada..........................................................57
Conceptual Framework for Hypothesized Health Effects ............58
Evidence Regarding Sound from Wind Turbines
and Human Health.........................................................................59
Weighing and Summarizing Evidence...........................................62
Chapter Summary...........................................................................63
5.4
5.5
5.6
5.7
Table of Contents
Current Evidence Related to Adverse

Health Effects of Wind Turbine Noise.................................. 65
6.1 Annoyance.......................................................................................67
6.2
Sleep Disturbance ..........................................................................74
6.3 Stress................................................................................................78
6.4
Cardiovascular Diseases .................................................................80
6.5 Diabetes...........................................................................................83
6.6
Effects on Hearing..........................................................................85
6.7 Tinnitus...........................................................................................86
6.8
Effects of Non-Audible Sound on the Inner Ear...........................87
6.9
Cognitive and Mental Performance...............................................90
6.10 Psychological Health......................................................................92
6.11 Quality of life..................................................................................93
6.12 Other Health Impacts ....................................................................93
6.13 Chapter Summary...........................................................................94
Overview of the Evidence, Knowledge Gaps,

and Research Needs.............................................................. 95
7.1
7.2
7.3
7.4
Summary of the Evidence...............................................................96

Quality of the Evidence Reviewed ...............................................102
Further Research .........................................................................104
Chapter Summary.........................................................................106
Promising Practices......................................................................... 107

8.1
8.2
8.3
8.4
Engineering Technologies............................................................108
Promising Practices to Monitor Noise.........................................111
Increasing Acceptance..................................................................114
Chapter Summary.........................................................................117
Conclusions..................................................................................... 119
9.1
Final Thoughts..............................................................................124
Glossary........................................................................................... 126
References...................................................................................... 132
xxi
Chapter 1 Introduction
1
Introduction
Charge to the Panel
Interpreting the Charge
Health Canada Wind Turbine Noise

and Health Study
Organization of the Report
Introduction
Wind energy is increasingly seen as a viable source of renewable energy that does
not emit greenhouse gases during operation and has a limited environmental
footprint. In 2013, the total wind energy capacity installed worldwide approached
320 gigawatts (GW), with more than 35 GW of new capacity installed in that year
alone (GWEC, 2014). In Canada, the total wind energy capacity installed was
9.2 GW at the end of 2014, distributed over 206 wind projects with more than
5,114 individual turbines (CanWEA, 2014). Renewable energy demand and wind
energy capacity are expected to continue to increase for the foreseeable future.
Wind energy capacity has increased in Canada since the 1990s, but in recent
years a growing number of individuals and organizations have expressed
concern that sound (see Box 1.1) emitted by wind turbines near peoples
homes may represent a risk to public health. Wind turbines are often located
in rural areas with low levels of background noise, making the noise of wind
turbines noticeable to some nearby residents. Sound from wind turbines is
characterized by:
periodic amplitude modulation (swishing or thumping characteristics);
unpredictable changes in sound levels depending on wind at the height of
the blades;
increased sound because of higher wind velocity due to increasing height
above the ground; and
a broad spectrum of frequencies.
Box 1.1
On the Use of Terms Noise and Sound in this Report
In the context of this report, sound* and noise are used largely interchangeably,
without affecting the interpretation of the meaning. However, where the Panel
describes a simple acoustic signal (e.g., in defining the properties of acoustics, or
an acoustic signal before it is cognitively processed), the neutral term sound is used.
Where the Panel describes a sound that is cognitively processed and that may be
perceived as unpleasant or unwanted, the term noise is used.
* Italicized words and terms are defined in the Glossary.
1 .1
C HA RGE T O T H E PANE L
In May 2012, the Government of Canada, through the Minister of Health (the
Sponsor), asked the Council of Canadian Academies (the Council) to conduct
an assessment of the question:
Is there evidence to support a causal association between exposure to wind turbine
noise and the development of adverse health effects?
The Sponsor also submitted the following sub-questions:
Are there knowledge gaps in the scientific and technological areas that need to be
addressed in order to fully assess possible health impacts from wind turbine noise?
Is the potential risk to human health sufficiently plausible to justify further research
into the association between wind turbine noise exposure and the development
of adverse health effects?
How does Canada compare internationally with respect to prevalence and nature
of reported adverse health effects among populations living in the vicinity of
commercial wind turbine establishments?
Are there engineering technologies and/or other best practices in other jurisdictions
that might be contemplated in Canada as measures that may minimize adverse
community response towards wind turbine noise?
To address these questions, the Council convened the Expert Panel on Wind
Turbine Noise and Human Health (the Panel), which included 10 Canadian and
international experts from relevant medical and engineering fields. This report
presents the results of their deliberations and analysis.
1 .2
I N TE R P R E T I NG T H E C H AR GE
A wide range of health impacts have been attributed to sound from wind turbines,
involving not only effects consistent with a narrow definition of health, but also
aspects of well-being and quality of life. The Panel therefore decided to take
a broad approach that would include the breadth of concerns that have been
raised by the public.
The Panel thus decided to guide its deliberations using the World Health
Organizations concept of health as a state of complete physical, mental and
social well-being and not merely the absence of disease or infirmity (WHO, 1946).
This understanding is also reflected in the approach taken by Health Canada,

which states that health is not an objective concept but varies among individuals
and societies (Health Canada, 2011). Health is therefore a concept that is open
to interpretation and discussion. The Panel recognized that health outcomes
include clinical symptoms and other outcomes that harm well-being, such as
annoyance or sleep disturbance.
This report considers both the incidence of health impacts following exposure to
sound from wind turbines, as well as the exacerbation of pre-existing conditions.
In many cases, the development of new health conditions post-exposure is
determined in part by a persons overall health condition before exposure,
particularly conditions that might lead to increased sensitivity to sound or noise.
First, the Panel developed a broad list of all health conditions, symptoms, and
other outcomes that have been attributed to sound from wind turbines, based
on a survey of sources that includes scientific literature, grey literature,1 media
reports, and legal documents, as well as self-published sources such as websites
and blogs. In a subsequent step, the Panel conducted a structured review of a
range of sources to assess the existing evidence on the relationship between
exposure to sound from wind turbines and the conditions identified in step
one. If published research had not yet considered a particular health outcome
in relation to wind turbine noise, or if available evidence was limited, the Panel
also considered analogous research on the health conditions with respect to
other sources of environmental noise. The assessment methodology applied
several concepts from evidence-based and evidence-informed public health: A
structured definition of the problem, systematic search of the literature, and
critical appraisal of the literature, as well as the use of Bradford Hill guidelines.
Those guidelines are commonly used in weighing a body of evidence to assess
causal relationships between an exposure to a potential hazard (e.g., sound
from wind turbines) and health outcomes (Bradford Hill, 1965).
The scope of the assessment was limited to health impacts that are thought to
be caused by, or associated with, exposure to sound emitted by wind turbines.
This focus excluded other potential sources of health impacts from these
turbines, such as shadow flicker, ground vibrations, and occupational health
risks for service personnel. The Panel also focused on modern upwind utilityscale wind turbines, which are the most common type used in the wind energy
industry in Canada and around the world.
Grey literature refers to documents in electronic and print formats that are not controlled by
commercial publishing; i.e., where publishing is not the primary activity of the producing body
(GreyNet International, 2014).
There have been several reviews of health effects of wind turbine noise over
the past 10 years. This report, however, differs from previous reviews in the
following ways:
a Canadian context and focus, with international comparisons where possible;
a review of claimed health outcomes attributed to wind turbine noise drawn
from a variety of sources (from peer-reviewed primary studies to web pages);
the use of a broad evidence base, including peer-reviewed literature but
also grey literature, a book, a graduate thesis and other sources of direct
evidence related to utility-scale wind turbines, as well as broader literature
on biological effects of noise;
identification of knowledge gaps and areas of research that are justifiable
because of the plausibility of potential adverse health effects; and
a consideration of promising practices to minimize adverse responses to
wind turbine noise.
1 .3
HE A LT H C ANADA W I ND T U RB INE NOISE AND

HE A LT H S T U DY
The Sponsor of this assessment, Health Canada, also conducted a separate study
of wind turbine noise and health, begun in July 2012. Preliminary results of
this epidemiological study were released in November 2014. As the Panel had
retrieved the primary evidence used to assess the health effects of wind turbine
noise in August 2014 (see Appendix B) and was finalizing this report when the
preliminary study results from Health Canada were released, the Panel was
unable to incorporate those results in its deliberations. However, the results of
the Health Canada study are presented in Box 7.1.
Like all Council assessments, this assessment was researched and written
independently by the Panel in response to the Sponsors questions. The Sponsor
was not involved in Panel deliberations or given access to drafts of the report.
1 .4
O R G A NI Z AT I ON OF T H E RE P ORT
Chapter 2 provides an overview of the Canadian and international contexts of

wind energy production and future growth in Canada. It introduces the general
framework for regulation of wind turbines in Canada. Chapter 3 introduces
key concepts and background information concerning acoustics and sound
measurement, as they apply to the emission and transmission of sound from
wind turbines. Chapter 4 clarifies the general principles of sound transduction
within the ear as well as other possible pathways of sound perception and how
it may vary among individuals. Chapter 5 explains the process used for the
evaluation of adverse health effects. Chapter 6 provides a detailed overview of

the scientific evidence available to assess causation between sound from wind
turbines and the health outcomes previously described. Chapter 7 summarizes
the findings from Chapter 6 and describes knowledge gaps that pose challenges
to understanding adverse health effects of wind turbine noise. Chapter 8 lists
promising technologies or practices identified in other jurisdictions that might
be considered in Canada to minimize adverse community response towards
wind turbine noise. Chapter 9 summarizes the main findings of the report in
response to the Charge.
Chapter 2 Wind Energy Context
2
Wind Energy Context
Wind Turbine Design
Canadian and International Contexts
Health Surveillance
Noise Limits and Setbacks
Chapter Summary
Wind Energy Context
Key Findings
Canada ranked fifth in the world for new wind energy capacity installed during
2013, and ninth in the world in terms of total installed wind energy capacity at
the end of 2013.
Most of Canadas installed wind energy capacity is located in Ontario, Quebec,
and Alberta.
The pace of wind energy development has varied among Canadian provinces.
Wind energy capacity in Alberta has grown slowly but steadily since 1993, driven
by market forces and green energy marketing programs by utilities, while capacity
in Ontario has expanded rapidly since 2006 as a result of provincial government
incentives. Capacity in Quebec has grown steadily since 1999 but accelerated
dramatically in 2013.
Wind energy development in Canada is subject to regulations at the federal,
provincial, and municipal levels, and risks associated with wind turbine noise are
managed differently in different jurisdictions. There are no national standards for
setbacks or noise limits in Canada, although the noise limits across Canadian
jurisdictions are comparable to those used internationally.
It is important to have some understanding about the development of wind

energy in Canada throughout any discussion about possible adverse health effects
created by this technology. While Canada has long had both experimental and
operational wind energy installations, these have grown rapidly in recent years,
with wind turbines becoming a familiar sight on the Canadian landscape. As
with many new technologies, the growth of wind energy has raised regulatory
and social issues that need to be addressed if Canadians are to rely increasingly
on this source of energy.
Currently, the most accurate source of information on wind energy capacity,
technology, and development is the wind energy industry itself. The Canadian
Wind Energy Association (CanWEA) is a national non-profit association of
wind turbine owners, operators, manufacturers, and service providers to the
industry. CanWEA maintains a database of wind energy projects in Canada,
including the number of wind turbines and their power ratings. These data are
collected directly from members and reported to international organizations
such as the Global Wind Energy Council (GWEC). The Panel used these data,
in combination with other published analyses where available, to describe the
wind energy sector in Canada.
Most wind energy projects in Canada are located in rural communities, yet most
of the electricity they produce feeds demand in larger urban centres. Individual
wind turbines occupy relatively little land area, requiring only enough land
for the foundation at the base of the turbine and for an access road. However,
turbines are tall structures, often exceeding a height of 80 metres at the hub
(see Chapter 3), and can have a visual and auditory presence in an otherwise
quiet rural landscape. Wind turbines usually do not prevent other uses of the
surrounding land, such as agriculture, hiking, hunting, or highways, provided
these uses are not negatively affected by wind turbine operation. However, the
construction phase may cause short-term disruptions. Some rural landowners
can supplement their income by leasing a small portion of land to a wind
turbine operator. Wind project operators argue that, because wind turbines
require regular maintenance and monitoring, they create technical jobs in the
communities where they are installed.
2 .1
W I N D T U RB I NE DE S I GN
Devices capable of converting the kinetic energy of the wind into electrical
energy are referred to as wind turbines. Wind turbines are classified by various
design characteristics, including axis of rotation, position of the blades (rotor
orientation), number of blades, rotor control, and alignment with the wind
(see Manwell et al., 2010).
In this report, a wind turbine refers to a device that is:
Modern: Able to convert kinetic energy in wind to electricity by mechanically
turning blades mounted on a rotor connected to a generator. Modern wind
turbines almost always have three blades.
Horizontal-axis wind turbine (HAWT): Having a main axis of rotation parallel
to the ground (like the axle of the wheels on a truck).
Upwind: Having blades that rotate in front (upwind) of a supporting tower.
Utility-scale: Connected to an electrical grid, and having a power rating of
at least 500 kilowatts (kW). Most utility-scale wind turbines currently sold
worldwide are rated at 1.5 megawatts (MW) or more, and designs are expected
to continue to increase in capacity and size. Some authors have described
these as industrial wind turbines.
Onshore: Located on land.
These are the characteristics of most modern wind turbines used in
the wind energy industry worldwide (Manwell et al., 2010) and describe
the vast majority of wind turbines installed for utility-scale electricity
generation in Canada. Figure 2.1 shows that all but a few turbines installed
in Canada have a rated power of at least 500 kW, while growth in capacity
since 2005 has been from turbines with increasingly larger power ratings
10
Number of Wind Turbines

in Canada
(at least 2 MW). Note that most utility-scale wind turbines sold today have a
power rating of at least 1.5 MW, but several important studies of health effects
of wind turbine noise are in the context of older and slightly smaller turbines,
all of which have a power rating of at least 500 kW.
4,000
3,000
2,000
1,000
Power Rating (MW)

3
2.5 < 3.0
2.0 < 2.5
1.5 < 2.0
1.0 < 1.5
0.5 < 1.0
0 < 0.5
0
Total Installed Wind Power Capacity
in Canada (MW)
1995
2000
2005
2010
2005
2010
Year
8,000
6,000
4,000
2,000
0
1995
2000
Year
Data Source: CanWEA, personal communication, 2014
Figure 2.1
Number of Turbines and Total Installed Wind Power Capacity in Canada,
by Power Rating, 19922013
The vast majority of wind turbines (top panel) and installed wind power capacity (bottom panel) in
Canada consist of turbines with a rated power of 500 kW or more. Most growth in capacity since 2005
has come from the installation of larger turbines with a rated power of at least 2 MW.
This report will not consider wind turbines that have a vertical axis of rotation,
have a downwind design, are located offshore, or are designed to provide power
to a single dwelling (residential scale). Canada had no offshore wind energy
facilities at the time of this report writing. Other designs are not considered
because they are not used for utility-scale electricity generation, and some have
very different acoustical properties (Wagner et al., 1996; Oerlemans, 2011).
Downwind turbines, for example, are known to produce high levels of infrasound
and low-frequency sound, which is one reason why they have not become an
11
industry standard (Jakobsen, 2005; van den Berg, 2011). This report therefore
discusses the technologies currently used in Canada and technologies likely
to be used in the future.
The major components of a modern utility-scale wind turbine are illustrated
in Figure 2.2. The shape of the blades creates aerodynamic lift and drag when
wind flows around them, much like the wing of an airplane. On a wind turbine,
however, these forces are used to generate torque, which causes the blades to
spin the rotor on its axis, creating mechanical power that is converted into
electricity in a generator housed in the nacelle.
Blade Tip
Leading Edge
Trailing Edge
Nacelle
Hub
Blade
Rotor
(hub + blades)
Tower
Foundation
Figure 2.2
Components of a Typical Modern Utility-Scale Wind Turbine
The major components typical of a modern wind turbine, as discussed in this report, are shown
here in this image of a General Electric 1.5 MW wind turbine at the Melancthon Wind Facility near
Shelburne, Ontario. The rotor consists of three blades mounted on a central hub, which rotates as
a single unit in response to wind flow over the blades. Each blade has a leading edge and a trailing
edge behind. The hub connects the rotor to a generator and other mechanical components inside
the nacelle, which converts the mechanical energy of the rotor movements into electrical energy. The
nacelle and rotor are mounted on top of the tower, which carries cables that connect the generator
to an electricity grid. The entire turbine is anchored by the foundation.
12
Wind turbines are often described in terms of rotor diameter (the diameter
of the swept area of the blades, roughly equal to twice the blade length plus
the hub diameter), and hub height (the height of the hub on the tower). Hub
height is usually 1 to 1.5 times the rotor diameter (Manwell et al., 2010). For
example, one of the most common wind turbines installed in Canada is the
GE 1.5s, manufactured by General Electric. It has a rotor diameter of
70.5 metres (231 ft.), a hub height ranging from 65 to 100 metres (213 to
328 ft.), weighs 157 tonnes, and is capable of producing up to 1.5 MW of
electricity in 12 metres per second (m/s) winds (GE, 2004). The largest wind
turbine currently being manufactured for onshore use, the Enercon E-126,
has a hub height of 135 metres (443 ft.), a rotor diameter of 126 metres
(413 ft.), and a rated power of 7.5 MW it is not currently installed in Canada.
Technical specifications of some commonly used wind turbines are listed in
Table 2.1. These are examples for illustrative purposes only, and the table is
not intended to present a comprehensive list, nor an endorsement of available
wind turbine models.
% of
Canadian
Capacity (end
of 2013)
12
22
11.4
Vestas V80
80
60100
1,800
105.0
4.0
15
25
7.0
Vestas V47
47
4055
660
102.0
4.0
15
25
1.5
Rated Wind
Speed (m/s)
4.0
Cut-In Wind
Speed (m/s)
NA
Maximum
Sound Power
Level dB(A) *
1,500
Rated Power
(kW)
65100
Hub Height
(m)
70
Rotor
Diameter (m)
GE 1.5s
Manufacturer
and Model
Cut-Out Wind
Speed (m/s)
Table 2.1
Examples of Utility-Scale Wind Turbines and Their Specifications
Legacy In use in Canada, but no longer sold
Currently sold and used in Canada

Enercon
E-82 E2
82
78138
2,000
NA
2.0
12.5
34
9.9
Vestas
V90-3.0
90
65105
3,000
109.4
3.5
15
25
7.8
Vestas
V90-1.8
90
80105
1,800
NA
4.0
12
25
7.2
GE 1.5sle
77
6580
1,500
NA
3.5
12
25
6.8
Siemens
SWT-2.3-101
101
80
2,300
NA
34
1213
25
5.8
Senvion
MM92
92.5
68100
2,050
103.2
12
24
3.4
Enercon
E-70
71
57113
2,300
104.5
2.5
15
34
3.3
13
Cut-Out Wind
Speed (m/s)
% of
Canadian
Capacity (end
of 2013)
NA
Rated Wind
Speed (m/s)
2,000
Cut-In Wind
Speed (m/s)
80100
Maximum
Sound Power
Level dB(A) *
93
Rated Power
(kW)
Hub Height
(m)
DeWind
D9.2
Rotor
Diameter (m)
Manufacturer
and Model
4.5
12
25
0.1
Global Not in use in Canada, but sold worldwide

Senvion
6.2M126
126
100117
6,150
NA
3.5
14.5
25
Enercon
E-126
127
135
7,580
NA
3.0
16.5
34
Vestas
V164 **
164
site
specific
8,000
NA
4.0
13.5
25
Data Source: van den Berg et al., 2008; Bauer, n.d.; CanWEA, personal communication, 2014
The models listed are examples of common utility-scale wind turbines in use in Canada and/or in other
countries, and do not account for 100% of Canadian capacity; this list is not exhaustive. All models
and technical specifications are the same for those used in other countries (wind turbines are also
sold in a global marketplace).
* The sound power level of a wind turbine, measured in decibels (dB), varies with wind speed: this
column shows the highest estimated sound power level for wind speeds between 3 and 11 m/s at
10m height in a neutral atmosphere and a standard ground roughness (van den Berg et al., 2008).
Note that these are sound power levels, which estimate total sound emission by the wind turbine;
the sound pressure level (sound immission at a receiver) is a function of the power level, distance,
and other factors (see Section 3.1.1).
** Designed exclusively for offshore use.
For comparison, a Canadian football field (CFL) is 100.6 metres (110 yards)
between goal lines and 59.4 metres (65 yards) wide. Smaller wind turbines are
about as tall as a football field is long, when measured from the base of the tower
to the top of the spinning blades (hub height plus half the rotor diameter).
Most turbines, however, would be too large to fit inside a football field if lying
flat on the ground, and, for some larger turbines, the tower alone is about as
tall as a football field is long. A single blade from most wind turbines would
cover the width of a football field, while the largest would exceed this width.
All wind turbines require a minimum wind speed at hub height to generate
power, called the cut-in wind speed, and have a maximum cut-out wind speed;
between these two wind speeds, determined on the basis of engineering and
design constraints, the turbine will deliver power safely. When the wind speed
exceeds the cut-out rating, control systems adjust the pitch of the blades so that
the rotor stops turning, and power production is cut off to prevent damage to
the turbine. The maximum power output of the generator is the rated power or
nameplate capacity (in kW or MW), which occurs at the rated wind speed at hub
14
height. For example, the rated power of the GE 1.5s wind turbine is 1.5 MW,
which is the power produced at a rated wind speed of 12 m/s at hub height
(see Table 2.1).
Wind slows down near surfaces, due to friction, creating a steep gradient of
wind speed between the ground and increasing height, an effect called wind
shear (Manwell et al., 2010). Taller turbines can take advantage of higher wind
speeds farther above the ground to produce greater amounts of energy. Taller
turbines can also accommodate larger blades. The use of fewer turbines with
larger rotor diameters can produce more energy per unit of land area and
is generally more cost-effective than a larger number of smaller turbines
(EWEA, 2005). These factors create incentives to build increasingly larger and
taller turbines, which are limited in size primarily by available construction and
transportation technologies.
2 . 2
C A N A D I A N A ND I NT E RNAT I ONAL CONTEX TS
Wind turbines are expected to have an operating life of 20 to 25 years, and current
capacity in many countries represents many years of cumulative investment.
Five countries accounted for 72% of the global wind energy capacity as of the
end of 2013: China, United States, Germany, Spain, and India (GWEC, 2014)
(Figure 2.3). Canada ranked ninth in the world in terms of installed capacity,
contributing around 2.5% of global wind energy production. However, Canada
ranked fifth in terms of new capacity installed in 2013, making it one of the
fastest-growing markets. Only China, Germany, United Kingdom, and India
added more wind energy capacity than Canada during 2013.
Canada had 9,219 MW of installed wind energy capacity as of the end of 2014,
which meets about 4% of Canadas electricity demand (CanWEA, 2014).
CanWEA (2008) argues that wind energy could satisfy 20% of Canadas electricity
demand by 2025. This would require the installation of about 50,000 MW of
additional capacity (roughly six times the current capacity), occupying a total land
area about the size of Prince Edward Island, but distributed over 450locations
across Canada. However, the National Energy Board the independent
federal agency that regulates oil, gas, and electric industry development in
Canada has forecast that total wind capacity will likely reach only 16,000MW
by 2035 (NEB, 2013). Many provinces in Canada have targets for wind power
or for renewable energy including wind.
15
90,000
Installed Capacity End 2012

New Installed Capacity 2013
80,000
70,000
60,000
50,000
40,000
30,000
20,000
ld
ia
W
or
he
Re
st
o
ft
Ro
m
an
nd
zil
Po
la
en
Br
a
k
ar
ed
Sw
da
nm
De
nc
e
na
Ca
ly
Ita
Fra
In
dia
Ki
ng
do
m
d
ite
ain
Sp
an
at
rm
Ge
Un
Un
ite
St
Ch
es
10,000
ina
Installed Wind Energy Capacity (MW)
100,000
Data Source: GWEC, 2014
Figure 2.3
Total Installed Wind Energy Capacity Worldwide, 2013
The total height of the bars indicates the cumulative installed capacity for wind energy at the end
of 2013, with the blue section at the top of each bar indicating the amount of wind energy capacity
installed (and coming online) during 2013. Canada ranks ninth in the world in terms of total wind
energy capacity in 2013, but was fifth in terms of newly installed capacity that year.
As of the end of 2013, 77% of Canadas installed wind energy capacity was in
Ontario, Quebec, and Alberta, but these provinces have had differing histories
of wind energy development (Figure 2.4). The first wind project in Canada
was installed in Alberta in 1993, and consisted of a single 150 kW turbine
(CanWEA, personal communication, 2014; Ferguson-Martin & Hill, 2011).
Wind capacity in Alberta has grown slowly but steadily since 1998, whereas
capacity in Ontario expanded rapidly after 2005. As a result, Ontario has
quickly become the province with the most installed capacity in Canada
(Figure 2.4). In Quebec, wind energy was introduced in 1998, and capacity
increased sporadically, roughly keeping pace with Alberta, until 2013, when
capacity increased by 78%.
Wind energy development across Canada is determined only in part by how
windy an area is that is, by the availability of wind resources. Wind energy
projects are generally not pursued unless they are both economically viable
and socially acceptable (Toke et al., 2008; Ferguson-Martin & Hill, 2011).
Development appears to depend on a combination of institutional factors
16
Total Installed Wind Power Capacity (MW)
2,500
ON
QC
2,000
YT
NU
NT
1,500
NL
BC
AB
SK
MB
QC
PE
NB
ON
1,000
AB
NS
BC
NS
NB
MB
SK
PE
NL
NT
YT
500
0
1992
1994
1996
1998
2000
2002
2004
2006
2008
2010
2012
Data Source: CanWEA, personal communication, 2014
Figure 2.4
Installed Capacity for Wind Power by Province, 19922013
The size of the circles on the inset map is proportional to the total installed capacity in each province
or territory at the end of 2013. As of 2013, Ontario, Quebec, and Alberta together account for 76.8%
of Canadas total installed wind energy capacity. Although Alberta and Quebec have historically
had the greatest installed capacity for wind power, Ontarios capacity has grown rapidly since 2005,
exceeding that of other provinces since 2008. In contrast, installed capacity in Alberta has grown
slowly over a longer period of time, while capacity growth in Quebec has been intermittent but has
accelerated in recent years.
such as how the landscape is valued by stakeholders (for alternative uses) and
whether there are government policies (e.g., incentive programs, processes
for planning and approval of projects) (see Ferguson-Martin & Hill, 2011).
Wind facilities also have to be able to sell their electricity to a nearby electricity
grid that is physically compatible. The grid must be able to accept the multiple
smaller and variable inputs generated by wind turbines, in contrast to the fewer
large inputs from nuclear plants, hydroelectric dams, or other traditional sources
of electricity. The structure of the electricity market affects how competitive
wind energy is relative to alternatives such as nuclear, hydroelectric, or coal
sources. If a government offers tax incentives, or requires utilities to buy wind
energy at a minimum price, such factors can make wind energy more viable.
17
The regulatory frameworks of federal, provincial, and municipal governments

encompass several of these factors, and can therefore affect both the economic
viability and social acceptability of wind energy development.
Wind energy in Alberta has grown despite grid limitations and a lack of explicit
government incentive programs. Growth in wind power capacity appears to
have been driven by market and consumer demand, supported by green energy
marketing programs of private electrical utilities and municipalities, with
the provincial government taking no official position on wind or renewable
energy (Ferguson-Martin & Hill, 2011). This contrasts with Ontario, where
the provincial government has had explicit policies in support of wind energy
development, including capacity targets, economic incentives, and streamlined
approval processes. Despite these government policies, public attitudes
towards wind energy in Ontario have become sharply polarized, with many
grassroots organizations supportive of, or opposed to, wind energy development
(Ferguson-Martin & Hill, 2011).
2 .3
HE A LT H S U R V E I L L ANC E
Like development of all energy sources and utilities, effects of wind energy
development on the health of workers and nearby residents are regulated; they
fall within the purview of public health. Surveillance of any health effects is
complicated by the fact that responsibility for public health is divided among
various jurisdictions in Canada. The federal government has powers over spending
(e.g., provincial transfers and research funding), and various aspects of public
health and safety, whereas provinces have jurisdiction over the provision of
health care services, hospitals, and matters of a local nature. The Public Health
Agency of Canada (PHAC) was established in 2004 in part as a response to the
2002 SARS outbreak, which highlighted the need for better surveillance, data
sharing, and collaboration within and between jurisdictions (Health Canada,
2003b; PHAC, 2008). Health surveillance gathers critical information about
trends in the general population, often providing early warning signs of public
health emergencies, and of the effects of various factors on public health.
Health surveillance can be passive or active. Passive surveillance in Canada
often takes the form of compiling health records, reports, and vital statistics
(births and deaths) at the local level, by local and regional health units, which
are reported up to a provincial agency, and then shared at the national level
through agencies such as PHAC. Passive surveillance can capture high-level
trends in the population but is often poorly suited to tracking rare conditions
or to establishing a causal link to a novel factor in the environment. Active
surveillance, through information-gathering approaches such as surveys, requires
more resources but can better target specific conditions or factors that might
18
influence health. Canadas passive health surveillance system does not collect
information about exposure related to wind turbines, such as sound levels,
distance, or visual presence. The Panel also found no active surveys in Canada
that could be used to compare the prevalence of health conditions associated
with wind turbines to the prevalence in the general population, or to the
prevalence in different regions, in Canada or internationally (see Chapter 5).
2 . 4
N O I S E LI M I T S AND S E T B ACK S
The wind energy industry in Canada operates in a regulatory context involving

multiple orders of government. The federal government has limited jurisdiction
over the energy sector but has supported wind energy development nationally
through direct funding, such as the ecoENERGY for Renewable Power program
(CanWEA, 2011). Health Canada is responsible for regulating devices emitting
radiation, which includes acoustical waves (sound and noise), under the Radiation
Emitting Devices Act (GoC, 2004). As a result, Health Canada has expertise in
measuring noise and assessing health impacts of noise (Health Canada, 2012).
Because electricity falls under provincial jurisdiction, provincial government
policies have the most effect on market structure, grid infrastructure, financial
incentives, and the planning and approvals processes for wind energy projects
(see CanWEA, 2011 for a list of initiatives). Most provincial approval processes
also require an environmental impact assessment. Provincial governments have
primary jurisdiction over wind turbine noise regulations, although individual
projects must follow local municipal regulations when applicable, including
zoning, building permits, and noise limits. Regulations covering wind turbine
noise vary across Canada, in the absence of a national standard or criteria
for noise exposure from wind turbines. The variation in wind turbine noise
regulations in Canada, and internationally, also reveals the diversity of approaches
and socially acceptable exposure levels. Common policy tools for regulating
the placement of wind turbines are setbacks and noise limits.
A setback is the minimum distance between a wind turbine and the closest
building or residence, intended to limit nearby residents exposure to various
safety hazards, including ice thrown from wind turbines, mechanical failure,
and noise. In Ontario, wind turbines with a capacity over 50 kW require a
setback of at least at 550 metres from dwellings (Ontario Regulation, 2012).
New Brunswick, Manitoba, and Quebec also impose setbacks from residential
areas, with distances ranging from 500 to 550 metres (Haugen, 2011). These
setbacks are similar to those applied in other countries.
19
Noise levels can also be regulated by specifying limits on sound pressure

levels, as measured at nearby residential buildings, irrespective of distance.
Such levels are defined so that the majority of the population is not disturbed
by noise exceeding the limit defined by the regulation (EPA, 2011). Noise
limits vary between and within provinces, and are sometimes specified in
relation to background noise, wind speed, rural/residential areas, or time of day
(Haugen, 2011). Manitoba has adopted a noise limit scale that ranges from
40 decibels (A) (dB(A)) at wind speeds of 4 m/s to 53 dB(A) at 11 m/s. Ontario
has sound level limits in rural areas that range from 40 to 51 dB(A) at wind
speeds of 4 to 10 m/s. In New Brunswick, noise limits range from 40 dB(A) at
wind speeds below 7 m/s to 53 dB(A) at wind speeds above 10 m/s. The Alberta
Utilities Commission requires that night-time noise limits fall between 40 and
56 dB(A), while the daytime noise limits are 10 dB(A) above night-time limits
(Haugen, 2011). Absolute noise limits across provinces are comparable to those
in Denmark, Netherlands, Sweden, Australia, and New Zealand.
2 .5
C HA P T E R S U M M ARY
Wind energy has been developed in Canada since 1993, but capacity has increased
dramatically since 2006. The history of wind energy development has differed
among provinces, not only because of differences in available wind resources,
but also as a result of economic viability, electricity grid infrastructure, social
acceptability, and government policies and regulations at all levels. Ontario and
Quebec have dramatically increased capacity in recent years, while capacity in
Alberta has grown steadily over a longer period of time. Canadian capacity is
expected to increase further, with the industry estimating that it could satisfy
20% of electricity demand by 2025, although National Energy Board estimates
are more conservative. In 2013, Canada ranked fifth in the world in terms
of newly installed wind energy capacity, and ninth in terms of total installed
capacity. However, this fast growth has created new social situations. Wind
turbines have a visual and auditory presence in some rural areas, and are a
divisive and polarizing issue in some communities. Uncertainty around their
benefits and impacts, and how to measure them, has contributed not only to
public unease, but also to a lack of consistency in noise regulations. Sound
levels from wind turbines are regulated by setbacks and noise limits, which
vary across jurisdictions in Canada. This new energy source in the Canadian
economy thus creates novel challenges for the energy industry, regulatory
organizations, and communities.
20
3
Wind Turbine Acoustics
Fundamentals of Acoustics
Sources of Sound from Wind Turbines
Wind Turbine Sound Levels and Spectra
Chapter Summary
Chapter 3 Wind Turbine Acoustics
21
Wind Turbine Acoustics
Key Findings
Wind turbines produce sound through multiple mechanisms with varying
characteristics. These signals are then modified as they travel through the environment.
Measuring sound from any source is complex, involves many dimensions
(e.g., sound pressure level, frequency, frequency weighting, time weighting), and
requires special equipment and procedures.
Wind turbines are a particularly complex and distinctive source of sound, which
can span a wide range of frequencies including low-frequency tones. These in
turn can travel longer distances and are less impeded by building materials than
higher-frequency sound. Turbine sound is also characterized by amplitude modulation,
often described as a swishing or thumping sound.
Low-frequency components may not be captured properly by standard frequencyweighted measurements (e.g., dB(A)), and amplitude modulation is not captured
by time-averaged measurements (e.g., Leq). It is also difficult to separate out other
background sound such as the wind itself.
Simple metrics of sound are an important first step in determining audible sound
exposure in most cases, but more detailed measurements may be necessary in
order for researchers to fully investigate the potential health impact of specific
wind turbine noise sources.
In any discussion about evidence of wind turbine noise affecting human health, it
is helpful to describe the characteristics of the sound produced by such technology.
These characteristics underlie the possible effects of sound from wind turbines on
the human body, which will be explored further in later chapters.
This report provides only a brief overview of wind turbine acoustics, and readers
interested in additional details or technical aspects of the issues presented in this
chapter may refer to the sources cited. The information presented is current as of
this writing, but the pace of technology development in the wind energy industry
is rapid, and some details or design aspects may change over the coming years.
Although fundamental aspects of acoustics and sound measurement are likely to
stay the same, research into the mechanisms of sound production by wind turbines
continues to try to address many knowledge gaps, including some identified in
this report.
22
3 . 1
FUN D A M E N TAL S OF AC OU S T I C S
Sound is a pressure wave that travels through a medium; in the case of sound from
wind turbines, the medium is air (ANSI/ASA S1.1, 2013). As a pressure wave, sound
consists of tiny alternating increases and decreases around atmospheric pressure,
which travel (propagate) through the air. Sound is described by several features, such
as the frequency of pressure oscillations that make up the wave, and the amplitude
of the oscillations (sound pressure), both of which can change over time. Noise is
often used to describe sound that is unwanted, for a variety of reasons, depending
on the context in which it is perceived (see Chapter 4).
Measuring exposure to sound from wind turbines is complex because it is challenging
to (1) detect and record sound at low frequencies, (2) summarize information
about sound pressure levels over a period of time and a range of frequencies, and
(3) isolate acoustic signals from wind turbines in an environment with many other
sources of sound. Measurements must take all three challenges into account to
provide information that is relevant to potential health impacts, although, for some
health effects, the important characteristics remain uncertain (see Chapters 6 and 7).
3.1.1 Sound Pressure Level, Sound Power, and the Decibel Scale
Sound power indicates the total amount of sound being produced by a given source,
measured in watts (W, the standard unit of power). Sound pressure, on the other
hand, indicates the amplitude of pressure fluctuations in a sound wave travelling
through air at a particular location, called the receiver, and is measured in standard
units of Pascals (Pa). Sound power represents the acoustic output from a source in
all directions, whereas sound pressure describes only the portion that reaches the
receiver. Therefore, sound pressure is a relevant measure of exposure to sound.
Sound pressure depends in part on sound power, although other factors in the
environment play a role, as will be discussed in the next section.
Although the two concepts are related, it is important to keep in mind the differences
between sound power and sound pressure, particularly since both are typically
converted to levels measured in decibels (dB). The decibel scale is used for many
measures that are expressed as a level, by taking a measurement relative to a reference
value and converting it to a logarithmic scale. The reference value is always in
the same units as the raw measurement, and therefore different measures have
their own reference values. For sound pressure levels in air, the reference value of
20 Pa (2 10-5 Pa) is equivalent to 0 dB, which is close to the limit of normal
human hearing at a frequency of 1 kHz. The reference value for sound power levels
is 1 10-12 W, which is equivalent to 0 dB.
23
The decibel scale is logarithmic, compressing large linear ranges into smaller
numerical differences. A doubling of sound pressure translates to a 6 dB increase
in sound pressure level. When sound measurements from different sources are
combined, the resulting dB value is not a simple sum of the two values; each
must be converted to linear units before applying calculations, then converted
back to dB (see Crocker, 2007; Leventhall, 2011). In general, non-coherent
sounds (those that are not in perfect phase with each other) behave as follows:
Combining two sounds of equal pressure level (dB) increases the sound
pressure level by 3 dB.
If the difference between pressure levels of two sounds is greater than 15 dB,
adding the lower level has a negligible effect on the higher level.
(Manwell et al., 2010)
Sound pressure level is calculated as an average level over a time interval and
integrated over a specific range of frequencies. This produces a single number, which
can be convenient for comparing sounds with similar frequency components and
variation in time. However, information about frequency and time can be measured
and described a number of different ways, leading to different conclusions when
comparing sound exposure. Not all decibel values are comparable; some methods
might provide similar values for sounds that actually have very different qualities.
The challenge of any measurement technique is identifying which characteristics
are important and capturing and highlighting the important information.
The sound power level of a wind turbine is estimated using measurements
under controlled conditions, including distance and wind speed, specified
by the International Electrotechnical Commission (IEC) 61400-11 standard
(Hessler, 2011). The sound power level of utility-scale wind turbines is typically between
95 and 110 dB(A),2 depending on the size and design of the turbine
(van den Berg et al., 2008; Leventhall, 2011; Mller & Pedersen, 2011). Wind turbines
with a larger rotor diameter or rated power generally have a higher sound power
level, but there is variation among designs and models. Furthermore, overall sound
power levels have decreased in newer generations of wind turbines of the same
size (Hau, 2006); as wind turbine design and construction evolves, larger turbines
produce sound power levels similar to older, smaller turbines. There is some evidence
that sound produced by larger turbines has more low-frequency components, but,
again, this varies among designs and manufacturers (Mller & Pedersen, 2011).
See Section 3.1.3 for an explanation of A-weighted sound pressure levels (frequency weighting).
24
The sound pressure level at a receiver ultimately depends on how this overall sound
emission changes while travelling, or propagating, through the environment between
the source and the receiver. There are currently no international standards for
measuring sound from wind farms at a receiver, although there are many technical
considerations to take into account (Hessler, 2011).
3.1.2 Sound Propagation
The pathways of transmission between a source and receiver can modify the sound
characteristics in a variety of ways. Therefore, the movement of sound in the
environment is affected by the properties of the medium (air) as well as by physical
structures in the path between the source and receiver.
Under ideal, theoretical conditions, sound from a point source spreads out
in all directions, like the surface of an expanding sphere, with the sound
pressure declining by 6 dB for every doubling of distance (Bullmore, 2011).
This assumes that sound is emitted equally in all directions, that the air is of
uniform density, and that nothing will absorb or reflect the sound waves. In
reality, the propagation of sound depends on several factors, including:
distance between the source and receiver: this determines the extent to which
other factors influence sound during propagation;
meteorological conditions: wind speed, direction, wind shear, turbulence,
and temperature gradients;
air density, which is affected by air temperature and humidity; and
ground characteristics: topography, surface features, and other factors that
may absorb or reflect sound.
(Hau, 2006; Crocker, 2007; Manwell et al., 2010; Bullmore, 2011)
A source of sound can be described by its sound power level as a function of
frequency or over a frequency range. However, sound power level and frequency
range are modified by transmission, absorption, reflection, or refraction of sound
travelling to the receiver. Many of these factors depend on distance and frequency,
and they can interact.
As sound travels through air, some acoustical energy is absorbed by the atmosphere.
Higher frequencies are absorbed much faster than lower frequencies over the
same distance (Bullmore, 2011). Higher frequencies are also reflected or absorbed
to a greater degree by surfaces in the environment, including the ground.
Low frequencies, however, are diffracted to a greater degree around barriers
(Bullmore, 2011). Therefore, the sound pressure level of low-frequency sounds
declines less with distance than the level of high frequencies.
25
Furthermore, gradients in air density or wind speed create small local changes in the
speed of sound, causing sound waves to be refracted (Jakobsen, 2005; Crocker, 2007;
Bullmore, 2011). For example, high wind shear occurs when wind speed increases
quickly with height above the ground; with such a strong gradient in wind speed,
sound is refracted upwards at locations upwind of the source, and down towards
the ground at locations downwind of the source. This leads to lower sound pressure
levels at locations on the ground upwind, and higher levels downwind of a source
such as a wind turbine. At the same time, wind speeds at ground level are very low,
and do little to mask other sounds.
Building construction and materials also heavily modify sounds. Most residential
building materials, including windows and walls, affect the transmission of highfrequency sound, causing declines in sound pressure level. However, such materials
transmit low-frequency sound to the interior with little loss in pressure level
(HGC Engineering, 2010; Madsen & Pedersen, 2010; Sndergaard, 2011). Sound
waves can also induce vibrations in structures if the frequency of the sound matches
the fundamental acoustic resonance frequency of the materials. Sound frequency
is inversely related to wavelength (the distance between pressure fluctuations),
and wavelengths can be more than 10 metres long for frequencies below 30 Hz.
Sounds interact with physical structures, such as a house, very differently when
the wavelength exceeds the size of the structure. Rather than acting as a barrier to
sound, a structure smaller than the wavelength is effectively surrounded by oscillating
pressures, which can induce vibrations and sound within the structure at the same
frequency as the sound (Findeis & Peters, 2004).
Low-frequency indoor sound can resonate depending on the condition of building
materials, such as size and shape. This resonance creates a standing wave with
alternating areas of amplified and muted sound pressure levels within a room
(Hubbard, 1982; Findeis & Peters, 2004). Because sound levels can vary within
different areas of a room or dwelling, particularly when such resonance occurs, it is
difficult to predict or even measure average indoor sound levels to which a resident
might be exposed (Pedersen et al., 2007b).
In general, distance also results in a decline in sound pressure levels, which involves
higher frequencies more than lower frequencies. This effect is not unique to wind
turbine sound, but accentuates the low-frequency components of any source indoors
and over longer distances. However, distance alone is a poor predictor of overall
sound pressure levels (see Tachibana et al., 2014).
Current knowledge about the characteristics of sound emitted by wind turbines,
and the environment through which those sounds travel, can be combined to build
models of how sound from wind turbines travels to receivers, such as neighbouring
26
residents. Models are used to predict outdoor sound pressure levels for planned
wind turbines before they are built, in order to identify and avoid potential cases
of unacceptably high sound pressure levels at nearby residences (Box 3.1).
Box 3.1
Models of Exposure to Sound from Wind Turbines
Mathematical models can use information about a sound source, such as sound power
level, for a range of frequencies and environmental factors in order to predict the
propagation of sound and resulting sound pressure levels across a landscape. When
planning a wind energy project, developers use such models to predict sound levels
and identify potential noise problems before the turbines are built, as well as to
select locations that would minimize sound exposure for nearby residents. Planning
must involve models, because actual measurements cannot be taken until a turbine
is built and installed, which is an expensive endeavour.
Standard modelling approaches have been developed to predict sound propagation
in general (e.g., ISO 9613-2, Nord 2000, HARMONOISE 2002), and specifically
for wind turbines (e.g., VDI 2714, Concawe, DIN 45645-1) (Wagner et al., 1996;
Hau, 2006; Bullmore, 2011; Burton et al., 2011; Evans & Cooper, 2012). These models
vary in levels of detail, in their reliance on empirical data, and in analytical approaches
that take into account aspects of turbine design (Wagner et al., 1996; Oerlemans &
Schepers, 2009; Evans & Cooper, 2012). Models may not be appropriate if they are
based on data from older designs for wind turbines. Most prediction methods are
specific to outdoor sound levels, with indoor levels often estimated by adjusting
predicted outdoor levels. The greater challenge, however, is accounting for the large
degree of variability in wind and atmospheric conditions, which affect sound emission
and propagation and add uncertainty to models.
Models of sound exposure are also used in many studies of health effects (Pedersen &
Persson Waye, 2007; van den Berg et al., 2008; Pedersen et al., 2009). Taking
measurements across hundreds of survey respondents is expensive and requires
special expertise, whereas models can provide a more relevant measure of exposure
than distance alone, by taking into account source characteristics and propagation
in the environment.
27
3.1.3 Sound Measurement

Standard methods for measuring and representing sound have affected research
and knowledge in acoustics relevant to wind turbines and are detailed here to
provide background information for the following chapters.
Sound is generally measured using sound level meters, standard devices that use a
microphone to convert acoustic signals into electrical signals, which are analyzed
to provide various measures of sound. Different microphones are sensitive to
different ranges of frequencies and sound pressures. Infrasound poses a particular
challenge because of the long wavelengths. Specialized equipment, such as a large
underground shielded microphone, is needed to measure infrasound accurately
(HGC Engineering, 2010). Furthermore, humans have a relatively high hearing
threshold for infrasound (typically 80 to 110 dB); see Chapter 4. For these reasons,
infrasound is sometimes omitted from standard measurements of audible sounds,
although it is a component of various sounds in the environment, including wind
turbines, diesel engines, wind, and ocean surf.
A microphone responds to all sounds that reach it, so establishing the characteristics
of a particular source requires measurements under controlled conditions in which
other sources are minimized. For wind turbines, this often involves measuring sound
at a location while turbines are operating and while turbines are off, under similar
wind conditions. The difference between such measurements provides an estimate
of the sound levels produced by wind turbines relative to other background noise.
Wind itself poses additional challenges to measuring sound outdoors, by creating
turbulence around the microphone as well as false signals (HGC Engineering, 2010).
Once an acoustic signal is captured by a microphone, the information in that signal
must be described using a relevant summary of the physical properties of the signal.
The fundamental properties of sound are the pressure (amplitude) and frequency,
both of which vary over time at any given location. Summarizing the information
in an acoustic signal involves a trade-off between simplicity and complexity. Simple
metrics are useful for highlighting specific characteristics of a sound while overlooking
others. In cases where the relevant sound characteristics are unclear, including both
acoustics and health research, more detailed measurements and analyses are often
necessary to identify which characteristics are the most important. Measuring sound
is therefore a complex, multi-dimensional challenge.
Frequency Components of Sound (Frequency Spectra)
The frequency of sound refers to the number of pressure waves per second (measured
in Hertz), which determines a sounds tone or pitch. A sound measurement can
be broken down into intervals, called frequency bands, within the overall range
captured by a sound level meter. Measurements over a range of adjacent frequency
28
bands are collectively called a frequency spectrum. In acoustics, frequency bands

are typically labelled by the centre frequency, exactly halfway between the upper
and lower limits of a band on a logarithmic scale (the geometric mean of the upper
and lower limits). A musical octave is an example of a frequency band, where a
note is double the frequency of a note exactly one octave lower. One-third octave
bands follow a similar principle, except that the ratio between centre frequencies
is two and one-third, instead of double.
Narrower frequency bands (e.g., one-third octave bands vs. one-octave bands)
provide more detail about the frequency components of a sound, but also capture
less acoustical energy within each band. As a spectrum describing a signal is split
into narrower bands, the sound pressure level of each band is typically lower than
wider bands around the same centre frequency, as shown in Figure 3.1. Peaks in a
spectrum of narrower bands indicate possible tones at those frequencies.
70.0
Band Widths
octave
1/3 octave
1/6 octave
1/24 octave
1.3 Hz (FFT)
Hearing threshold
Sound Pressure Level per Frequency Band (dB)
60.0
50.0
40.0
30.0
20.0
10.0
0.0
-10.0
10
100
1,000
10,000
Centre Frequency (Hz)

Reproduced with permission from Madsen & Pedersen, 2010
Figure 3.1
Example Frequency Spectra of Sound from a Wind Turbine
The spectra in this figure are based on sound from a 1.3 MW wind turbine, measured at 70 metres. The
overall sound pressure level of the sound is 33dB(A). A spectrum of narrower frequency bands shows more
detail, with spikes indicating possible tones at those frequencies. The hearing threshold shown is for an
average person responding to tones in a quiet room with no background noise. Unlike the logarithmic
scale of octave bands, narrow-bands are the same width along the entire frequency range (1.3 Hz, in
this case, analyzed using a form of narrow-band frequency analysis called fast Fourier transform (FFT))
and provide highly detailed information about the frequency components of a sound measurement.
29
The frequency spectra presented in Figure 3.1 are based on real measurement
data for a wind turbine, but are representative of the conditions specific to that
measurement. As discussed in Section 3.1.1, many factors affect the frequency spectra
and overall sound pressure level at a given location, including the characteristics
of the source (turbines), distance from the source, atmospheric conditions, and
terrain. Therefore, different conditions and turbines might lead to slightly different
frequency spectra (see Section 3.2 for more examples).
Comparing frequency spectra to the threshold of human hearing is not always
straightforward, as suggested by Figure 3.1. Standard hearing thresholds for humans
are based on measurements made under highly controlled conditions and may
not reflect audibility in real-world conditions. These thresholds are also median
values for a population sampled, and there is individual variation at all frequencies
(see Section 4.3).
This report refers to frequencies between 200 Hz and 2 kHz as mid-range, while
high-frequency sound includes frequencies between 2 and 20 kHz. Sound frequencies
above 20 kHz are known as ultrasound and are generally inaudible to humans, due
to the sharp drop in sensitivity of the human ear to these frequencies.
At the low end of the frequency range, sound at frequencies below 20 Hz is typically
referred to as infrasound. Infrasound is often described as inaudible, although, as
discussed above, measurements indicate that it can be heard at sound pressure levels
above 70 to 100 dB by an average person (Watanabe & Mller, 1990). Although
human hearing does respond qualitatively differently to infrasound (<20 Hz) and
ultrasound (>20,000 Hz), the boundaries between these and other frequency ranges
are arbitrary, as the changes in hearing thresholds occur over a range of frequencies
around these boundaries (see Section 4.3).
Low-frequency sound has different meanings, depending on the context and application,
and has no standard definition. The upper limit ranges from 100 to 250 Hz or
sometimes even 500 Hz (Berglund et al., 1996; Persson Waye, 2005), while the
lower limit ranges from 20 Hz (the upper limit of infrasound) to 10 Hz or even
5 Hz (Leventhall, 2009). In this report, low-frequency sound refers to sound in
frequencies between 20 and 200 Hz, acknowledging that the exact range may vary
in individual studies.
Frequency Weighting
Sound levels are often summarized across the range of frequencies measured. This
value can be adjusted, within sound measurement equipment or after recording, to
provide greater weight to certain frequencies of interest. An overall sound pressure
level trades off the complexity of frequency components for the simplicity of a single
value. Simpler metrics make it easier to compare different sounds, but comparisons
30
based on a frequency-weighted value cannot take into account any differences in

frequency components. The choice of frequency weighting is a function of both
the characteristics of the source and the effects being studied.
The weighting profiles of common methods used to measure sounds from wind
turbines and other environmental sources are shown in Figure 3.2. A- and C-weighting
give less weight to low-frequency and very high-frequency (>10 kHz) sounds.
A-weighted sound pressure levels are indicated dBA or dB(A). G-weighting gives
greater weight to low-frequency sound and infrasound between 10 and 30 Hz.
Unweighted measurements are called Z-weighted (for zero) to explicitly indicate
an absence of any frequency weighting. Z-weighting may also be labelled dB SPL
(decibel sound pressure level) or dB alone. However, specifying the weighting avoids
possible confusion. The same sound can have different dB(Z), dB(A), or dB(G)
values, depending on the levels of different frequency components of the sound.
20
0
-20
Weighting (dB)
-40
-60
-80
-100
-120
-140
-160
1
10
100
Fequency (Hz)
1,000
10,000
Data Source: IEC 61672:2003 (A and C-weighting); ISO 7196:1995 (G-weighting)
Figure 3.2
Common Weighting Methods for Sound Pressure Level Measurement
Positive values on the vertical axis indicate that sound levels at that frequency are increased
(up-weighted) at the indicated frequency; negative values mean that sound levels at that frequency
are reduced (down-weighted) by the amount indicated on the vertical axis. Shaded areas indicate
infrasound (frequencies below 20 Hz) and ultrasound (frequencies above 20 kHz). Z-weighting indicates
linear, unweighted measurements in which sound levels are not adjusted based on frequency. A- and
C-weighting were originally intended to mimic the response of the human ear, by down-weighting
low-frequency sounds and infrasound. G-weighting emphasizes sound frequencies between 10 and
30 Hz. The G-weighting curve shown is an approximation on a log-scale.
31
A-weighting was originally intended to mimic human perception of loudness at

low sound pressure levels, and C-weighting was similarly intended for higher sound
levels. In practice, this distinction has been abandoned, and both are used at all
levels (Leventhall, 2011). A-weighted measurements are commonly provided by sound
level meters, and specified in regulations and guidelines, for all types of environmental
noise (Colby et al., 2009; Leventhall, 2011). As a result, it has become standard
practice to report environmental noise levels using A-weighted measurements,
which makes measurements comparable across studies in a way that is relevant to
human hearing responses. Comparisons based on A-weighted measurements may
not be valid, however, for sounds with markedly different frequency components,
especially low-frequency components that are down-weighted.
Time Weighting
In addition to frequency weighting, all sound measurements are averaged over
a given time interval. Even when sound is measured over long periods, such as
several hours or days, measurements are often recorded over many regular intervals,
typically 1 second (slow) or 1/8 of a second (fast). A series of measurements can
then be processed to calculate different metrics and analyze how sound levels
change over time. Some common indicators of environmental noise exposure are
described in Table 3.1.
Table 3.1
Sound Pressure Level Indicators and Their Notation
Leq, LAeq
Equivalent continuous sound level: A steady sound pressure level that would produce
an equivalent level over the period of measurement. There is no weighting given to
any portion of the time period measured, which is indicated in the subscript after a
comma. Frequency-weighting is indicated by the appropriate subscript (A, C, G, etc.).
E.g., LAeq,1hr is the equivalent continuous A-weighted sound pressure level over 1 hour.
Lf
Fast measurements, integrated over one-eighth of a second.
Ls
Slow measurements, integrated over one second.
Ldn
An average A-weighted sound pressure level over a year (LAeq,1yr), where 10 dB(A) is added
to night-time levels. Night-time is usually defined as 11 pm to 7 am, or 10 pm to 6 am.
Lden
Similar to the Ldn, except that 5 dB(A) are also added to levels during evening hours,
6 to 10 pm or 7 to 11 pm.
Lnight,outside A yearly average of night-time sound pressure levels (LAeq,night over a typical year),
measured outside. This is often used as a long-term indicator for night noise levels
in Europe.
Lmax
The maximum sound pressure level observed over the measurement period.
L10, L50,
L90
The level exceeded during a percentage of the measurement period, indicated by the
subscript number. For example, L10 indicates the sound level exceeded 10% of the time
and gives an idea of the upper extremes over this period.
(WHO, 2009; Mnzel et al., 2014)
32
The time interval of a measurement has several implications for the way it is
interpreted and compared with other measurements. Long-term averages can be
useful indicators of sound exposure relevant to health (WHO, 2009), but shorter
intervals can provide much more detail about how sound varies over time. In
general, shorter measurement intervals capture larger variations. However, longer
intervals such as Leq are often used, and these fail to capture short-term effects such
as amplitude modulation (see Figure 3.6).
3 . 2
S O UR C E S O F S OU ND F R OM W I ND T UR B INES
Wind turbines produce sound primarily from two sources:

Mechanical, produced by physical movements of the gearbox, generator,
and other components in and around the nacelle, which is mostly tonal in
character (dominated by a narrow range of frequencies), but can also have
a broadband character, meaning it is composed of sound over a wide range
of frequencies.
Aerodynamic, caused by air flowing over the blades, usually producing
broadband sound at levels proportional to the relative velocity of the air flow.
(Hau, 2006; Manwell et al., 2010; Oerlemans, 2011)
Appropriate design and manufacturing have reduced sound levels from mechanical
sources; as a result, aerodynamic noise from the blades is the dominant type of
sound for modern utility-scale wind turbines (Hau, 2006; Oerlemans, 2011). Lowfrequency tones associated with mechanical sources are distinctly audible for some
turbines, however (Sndergaard & Madsen, 2008; Madsen & Pedersen, 2010).
The aerodynamic mechanisms that produce sound from wind turbines involve
interactions between air flow and different parts of the blade. These interactions
depend on the speed and turbulence of the incoming wind, the shape of the blade
(in cross-section and outline), the angle between the blade and the relative wind
velocity flowing over the blade (the angle of attack), and the distance from the hub
(Wagner et al., 1996; Oerlemans, 2011). In particular, parts of the blade closer to
the tip are moving faster than those closer to the hub, leading to faster relative air
velocities. Higher relative velocities typically lead to higher sound levels, especially
at higher frequencies (Oerlemans, 2011). Therefore, most aerodynamic sound
is produced near (but not at) the blade tips (Oerlemans & Mndez Lpez, 2005;
Oerlemans, 2009). This is partly why turbines with longer blades have a higher
sound power level: longer blades mean higher relative velocities near the blade
tips for the same rotation rate.
33
Different aerodynamic mechanisms produce sounds with different characteristics,

however. Low-frequency sound may be created as a result of inflow turbulence in the
wind along the leading edge of the turbine blade (Oerlemans, 2011). This varies
considerably from site to site, depending on local conditions. For example, turbines
located in the wake of another turbine are particularly likely to experience inflow
turbulence. Sound produced by air flow at the blade tip (tip noise) is generally
broadband with high-frequency components, but only in some cases contributes to
overall sound emissions (Wagner et al., 1996; Oerlemans, 2011). Sound produced
by air flow at the trailing edge of the blade is usually the dominant source of sound
from wind turbines, and has an overall broadband character (Oerlemans, 2011).
As blades degrade because of erosion and buildup of debris, however, they can
produce more sound than normal, due to turbulence created by air flow over
a blade surface that is not as smooth as intended for the blade design
(Oerlemans et al., 2007).
Aerodynamic sources of sound from wind turbine blades also have a strong
directional component, radiating mainly downwind, upwind, or even perpendicular,
depending on the dominant mechanism (Oerlemans, 2011). Therefore, the
amplitude of the acoustic signal at a particular location, measured by the sound
pressure level, can vary depending on the direction, speed, and turbulence of the
wind. Furthermore, as the rotor turns, the orientation of each blade changes from
the perspective of a stationary receiver. As a result, the pressure level of sound from
a given turbine reaching a receiver also varies as the blades rotate, resulting in
periodic amplitude modulation regular changes in sound pressure level over time
(van den Berg & Bowdler, 2011; RenewableUK, 2013).
Since aerodynamic mechanisms account for the majority of sound emissions
from wind turbines, sound power levels increase with wind speed up to the
maximum sound power level at the rated wind speed (Wagner et al., 1996;
Sndergaard, 2011). Furthermore, the relevant wind speed affecting sound power
levels is that at the height of the blades, which can differ from that at ground level.
Given rotor diameters of 50 to 100 metres or more, uneven wind speeds within the
swept area of the wind turbine rotor can also add to sound production, as well as
to amplitude modulation of emitted sound.
34
Apparent Sound Power Level (LWA) (dB relative to 10-12 W)
120
110
100
90
80
100 kW
1 MW
10 MW
Rated Power
Reproduced with permission from Mller & Pedersen, 2011; Mller et al., 2011
Figure 3.3
Apparent Sound Power Level (LWA) of Wind Turbines as a Function of Rated Power
These data are based on the measurements of 61 wind turbines from a range of models in Europe,
downwind at 8 m/s, according to IEC 61400-11 standards. Each turbine is shown as a diamond symbol.
A regression line is shown, with 90% confidence intervals shown as dashed lines. Turbines with a higher
rated power tend to have a higher sound power level, but there is variation in sound power level
among models of the same rated power. The slope is close to 10 dB per decade (9.7) corresponding
to equal acoustic power per electrical power.
3 . 3
W I N D TUR B I NE S OU ND L E V E L S AND SPEC TR A
Characterizing wind turbine sound is complicated by the variation in mechanisms

that produce sound, the changes that occur during transmission across a range of
distances, and a lack of measurement standards (Hessler, 2011). See Figure 3.3 for
the sound power level of wind turbines as a function of rated power.
Although there is a standard method for measuring the sound power level of
wind turbines (IEC 61400-11), methods for measuring the sound pressure level
(exposure) at a receiver must be tailored to individual circumstances and technical
factors. Questions such as how loud are wind turbines? are difficult to answer,
because the response depends on local wind and atmospheric conditions, distance
to the turbine, and intervening topography. Large numbers of measurements in
recent years, however, have started to shed light on the range of variation and the
characteristics of sound emissions from wind turbines.
35
90
1/3 Octave Band Sound Pressure Level (dB)
80
Hearing thresholds for
pure tones (ISO 389-7
70
60
50
-4
40
dB
/oc
tav
30
20
10
0
16
31.5
63
125
Frequency (Hz)
250
500
1k
2k
4k
Adapted with permission from Tachibana et al., 2014 and INCE/USA
Figure 3.4
One-Third Octave Band Unweighted Sound Pressure Level Measurements
Around 29 Wind Facilities in Japan
One-third octave band spectra are overlaid for 164 outdoor locations around 29 wind facilities across
Japan, from about 100 to 1,000 metres from the nearest wind turbine. Each spectrum is based on
a 10-minute measurement period at night, and sound pressure levels are unweighted (dB(Z)). Data
from four coastal sites were excluded because wind turbine sounds could not be detected above
background noise from waves and wind. Spikes in individual spectra suggested the presence of
low-frequency tones in some cases. These unweighted spectra corresponded to average night-time
levels between 25 and 50 dB(A), with most being between 35 and 45 dB(A). Levels of infrasound
appeared to be below the median threshold of human hearing (based on laboratory experiments);
infrasound was measured at a range of 46 to 75 dB(G), with the peak of the distribution around
61 to 65 dB(G). Hearing thresholds (ISO 389-7:2005) are shown for comparison, but this threshold is
based on measurements using pure tones and varies among people; frequency components above
this line are not necessarily audible, nor are frequencies just below this line necessarily inaudible.
36
Tachibana et al. (2014) report outdoor measurements from multiple locations around
29 wind facilities, ranging from 100 to 1,000 metres from the nearest wind turbine.
The combined spectra are presented in Figure 3.4, which shows some variation in
levels at all frequencies, but an overall consistent shape, with levels dropping by
about 4 dB per octave. Peaks in the low-frequency and infrasonic ranges (<250 Hz)
suggest low-frequency tones present in a few cases. The average level for 10-minute
night-time periods ranged between 25 to 50 dB(A) across all outdoor measurements,
with most being between 36 and 45 dB(A) (Tachibana et al., 2014).
Other sounds with similar A-weighted levels include an average living room, quiet
office, light car traffic more than 20 m away, an air conditioning unit more than
60 m away, or the wind itself (Wagner et al., 1996; Colby et al., 2009; Fortin et al., 2013),
although these sounds may vary over time in different ways than wind turbines, or
have different frequency components. In some cases, background noise from wind
and waves can be at similar or higher levels, making it difficult to isolate the sound
from wind turbines (Hepburn, 2006; Tachibana et al., 2014). Two sounds with the
same A-weighted overall pressure level may not be associated with the same health
effects, however, as other characteristics may be more relevant, such as maximum
levels, long-term exposure, or interpretation of the sound by its hearer.
Although pressure levels of sound from wind turbines tend to decline with increasing
frequency, the mid-range and high frequencies are often the most audible, due to
the sensitivity of the human ear. As mentioned above, assessing audibility based on
spectra such as those in Figure 3.5 is difficult because of differences in the way the
sounds and hearing threshold are measured (see Chapter 4 for additional details).
Infrasound (<20 Hz) levels in the Japanese measurements were below human
hearing thresholds, based on laboratory studies using recordings of the sounds
(Tachibana et al., 2014). Above 20 Hz, the same lab studies found that the recorded
sounds from these wind turbines were indeed audible. This is consistent with many
other studies that have concluded that average infrasound levels from modern wind
turbines at outdoor locations are below the human hearing threshold, even as close
as 100 m from the turbine (Jakobsen, 2005; HGC Engineering, 2010; Madsen &
Pedersen, 2010). Hepburn (2006) found that infrasound levels were not significantly
above background levels at 1,000 metres from wind turbines at the Castle River
Wind Farm in Alberta; infrasound levels ranged from 53 to 62 dB(Z) in low wind
conditions to 76 to 82 dB(Z) in high wind conditions, which is consistent with the
measurements by Tachibana et al. (2014).
37
The general spectra shown in Figure 3.5 include measurements at a range of distances.
The spectrum from an individual turbine is expected to shift toward dominance of
lower frequencies at increasing distance, combined with an overall decline in sound
pressure level, as discussed in Section 3.1.2. Figure 3.5 shows predicted A-weighted
spectra at increasing distance from a wind turbine, based on a measured overall
sound pressure level of 45 dB(A) at 300 metres. At further distances, the frequencies
likely to dominate what is heard (i.e., A-weighted) shift towards lower frequencies,
while the overall sound pressure level decreases.
40
LpA,1/3-octave (dB re 20 uPa)
30
20
10
80
10
0
12
5
16
0
20
0
25
0
31
5
40
0
50
0
63
0
80
1, 0
00
1, 0
25
1, 0
60
2, 0
00
2, 0
50
3, 0
15
4, 0
00
5, 0
00
6, 0
30
8, 0
0
10 00
,0
00
63
50
25
31
.5
40
-10
-20
1/3-octave band (Hz)
300 m
500 m
1,000 m
3,000 m
5,000 m
10,000 m
Wind farm
Data Source: Sndergaard, 2011. Reproduced with permission from Multi-Science Publishing Company
Figure 3.5
Predicted Sound Spectra at Various Distances from a Wind Turbine
Estimated sound spectra were generated using the Nord2000 turbine, assuming flat farmland and
a wind speed of 8 m/s. Predictions were based on a measured sound spectrum of 45 dB(A) at 300 m
from a typical wind turbine with a hub height of 90 m (solid bold line in the figure). The dotted
line labelled Wind farm shows a spectrum from a wind facility with multiple turbines that also has
an overall sound pressure level of 45 dB(A). These model predictions show how dominant A-weighted
frequencies tend to shift toward lower frequencies at increasing distance from a wind turbine.
38
As described in Section 3.2, the rotating blades on a wind turbine can lead to
sound emissions at levels that fluctuate periodically from the perspective of a
stationary observer a phenomenon called periodic amplitude modulation. Amplitude
modulation is described by modulation depth, the difference between the lowest
and highest sound pressure levels, and the modulation frequency, which is the rate
at which the sound pressure level alternates between the lowest and highest levels.
Modulation frequency should not be confused with the frequency range of the
sound in which pressure levels are changing. Although both may be described in
Hz, amplitude modulation typically fluctuates at a similar rate to the blade-pass
frequency, which is around 1 Hz (meaning there is one second between each
blade passing the same point), whereas such amplitude modulation can affect
a range of sound frequencies produced by wind turbines. While other sources
of environmental noise, such as road traffic, also exhibit amplitude modulation
(the sound pressure level changes over time), sound from wind turbines is
distinct in that the amplitude modulation repeats regularly with the rotation
of the blades it is periodic and often described as swishing or thumping,
depending on the sound frequencies involved.
There are few published measurements of amplitude modulation from wind
turbines, and modulation depth is sometimes measured using different metrics
(RenewableUK, 2013; Tachibana et al., 2014) (see Figure 3.6). Available data
suggest that modulation depth ranges from non-existent to about 5 dB, with
most being between 2 and 3 dB (Oerlemans & Schepers, 2009; van den Berg &
Bowdler, 2011; Tachibana et al., 2014). A modulation depth of 2 dB or more is
considered audible (van den Berg & Bowdler, 2011), so amplitude modulation
in wind turbine sounds is likely perceptible in many cases, but not all. Most
amplitude modulation occurs at mid-range sound frequencies between 200
and 1,000 Hz, which is often described as a swishing sound (Bowdler, 2008;
van den Berg & Bowdler, 2011; RenewableUK, 2013).
RenewableUK a not-for-profit energy trade association for renewable energy
companies in the United Kingdom has published a collection of research
reports that distinguish between normal and enhanced or other amplitude
modulation, which exhibits greater modulation depth, lower frequencies, or
other characteristics outside the normal range (RenewableUK, 2013). Normal
amplitude modulation can be explained by air flow at the trailing edge and
blade movement, as described above, and tends to be strongest perpendicular to
the wind direction, in the same plane as the wind turbine rotor (Oerlemans &
Schepers, 2009; van den Berg & Bowdler, 2011; RenewableUK, 2013). Other
amplitude modulation is thought to be due to wind gusts, local turbulence, or
other non-uniform wind conditions that interact with turbine blades rotating in a
periodic manner (RenewableUK, 2013). Enhanced amplitude modulation tends
39
to occur for brief periods, but the modulation depth can approach 10 dB, often
at low frequencies (1 to 2 kHz), and the sound can travel for longer distances
downwind than normal amplitude modulation (Bowdler, 2008; van den Berg
& Bowdler, 2011; RenewableUK, 2013). Enhanced amplitude modulation is
sometimes described as thumping, and distinguished from swishing by greater
modulation depth and lower frequency components. The causes and types of
amplitude modulation are a topic of ongoing research, including investigation
into how the modulation character is propagated outdoors (van den Berg &
Bowdler, 2011; RenewableUK, 2013).
49
LAs
LAf
1s Leq
1s LAsmax
1s LAfmax
Sound Pressure Level (dB(A))
48
47
46
45
44
43
42
41
20:00
20:01
20:02
20:03
20:04
20:05
20:06
Time (mm:ss)
20:07
20:08
20:09
20:10
Data Source: Kaliski, 2014. Reproduced and adapted with permission from RSG, Inc.,
(funded by MassCEC and co-managed with MassCEC by MassDEP)
Figure 3.6
A-Weighted Sound Metrics for a Wind Turbine Over a 10-Second Period
This graph shows several metrics for the same sound measured 260 metres crosswind from a single
wind turbine (size or model not specified). All metrics are A-weighted. LAs indicates slow measurements,
integrated over one second intervals, whereas LAf indicates fast measurements integrated over oneeighth second intervals: these intervals overlap, so that there is more than one instantaneous LAs
measurement per second of observation (each LAs measurement integrates over the previous second).
The major difference is that fast measurements respond to shorter changes in sound pressure level.
The Leq and maximum levels for LAs and LAf are also indicated for each one-second interval over the
measurement period. These values illustrate the ability of shorter intervals (LAf) to capture larger
variations and therefore affect the magnitude of other metrics, such as LAmax, for the same sound.
The time series also shows regular amplitude modulation of wind turbine sound, which appears to
be more extreme when using fast measurements (LAf) but would not be apparent to short- or longterm equivalent averages (Leq).
40
3 . 4
C HA P TE R S U M M ARY
Wind turbines produce sound through multiple mechanisms with varying

characteristics. These signals are then modified during propagation in the
environment. Sound from wind turbines is therefore highly complex and variable,
but has some characteristics that are similar to other sources of community
noise, such as road and airport traffic noise:
It is broadband, composed of sound over a range of frequencies, with levels
typically dropping by about 4 dB per octave (HGC Engineering, 2010;
Tachibana et al., 2014).
The overall sound pressure levels outdoors vary greatly depending on
distance, wind speed, and transmission from the source to the receiver
(Tachibana et al., 2014).
Higher frequencies tend to be reduced indoors and with increasing distance,
leading to an emphasis on lower frequencies (HGC Engineering, 2010;
Sndergaard, 2011). Low-frequency sound can cause acoustic resonance
within structures.
It is amplitude modulated, with sound levels changing over time.
Wind turbines also emit sound with the following characteristics that are less
common to other sources of community noise:
Sounds from wind turbines may extend down to the infrasonic range and,
in some cases, may include peaks or tonal components at low frequencies
(Sndergaard & Madsen, 2008; HGC Engineering, 2010; Madsen & Pedersen,
2010; Tachibana et al., 2014).
The sound power level increases with the wind speed, up to the rated
wind speed of the turbine, at the height of the blades, which is often 60 to
100 metres or more above ground level (Wagner et al., 1996; Sndergaard, 2011).
Sound from wind turbines can have periodic amplitude modulation, fluctuating
regularly at the same rate as the blade pass frequency (around 1 Hz), which
gives it a swishing or thumping character, depending on the dominant
sound frequencies at which modulation occurs. Modulation depth is less than
5 dB in most cases, but in rare cases may approach 10 dB (van den Berg &
Bowdler, 2011; RenewableUK, 2013; Tachibana et al., 2014).
Sound measurement is limited by available equipment and methods, but
typically includes sound pressure level and frequency components. These
measurements are often weighted by frequency to reflect human hearing
responses (A-weighting), and are summarized by an average over a period of
time (equivalent sound pressure level, Leq).
41
The Panel acknowledges that a standard A-weighted average sound level (LAeq) is
not intended to capture the full complexity of sound, but often provides a useful
first approximation of long-term sound exposure. Comparisons based on such
overall average levels, however, can fail to take into account important differences
in frequency content or amplitude modulation. When sound conditions are
unusual or of particular concern, more detailed measurements may be needed
to identify problems. The most appropriate sound metric is a function of both
the source and the sound characteristics of interest.
42
4
Hearing and Perception of Sound
The Auditory Pathway
Non-Auditory Pathways
Basic Auditory Perception and Mechanisms
Sound Perception
Chapter Summary
Chapter 4 Hearing and Perception of Sound
43
Hearing and Perception of Sound
Key Findings
Sound can affect the human body through auditory and non-auditory pathways.
The auditory pathway involves sounds reaching the cochlea through air and bone
conduction, where they are converted into neural signals by the hair cells. Sounds
can also stimulate the vestibular system and/or cause whole-body vibration under
certain conditions. At commonly specified setback distances from wind turbines,
the dominant sound pathway is through air conduction.
Absolute hearing thresholds have been standardized for a reference population of
young adults. These represent the lowest sound pressure levels at which the average
young adult with normal hearing can hear a pure tone sound signal at different
frequencies in controlled quiet conditions. However, hearing thresholds can vary
widely in the general population as a result of age, prior exposure to loud sounds
or ototoxic agents, medical conditions, and many other factors.
In everyday environments where there are background noise signals, the detection
of sounds also depends on the levels of competing masking sounds. In the context
of wind turbine noise, it is appropriate to consider masked hearing threshold. This
is the sound pressure level at which a sound can be heard in the presence of other
sound (competing acoustic signals). The perceptual attributes of sounds heard in
the environment depends on their spectral content (e.g., tonal versus broadband),
temporal characteristics (e.g., continuous or modulated), loudness (subjective
perception of sound intensity), and many other factors.
Sound may have emotional and psychological effects that also need to be taken
into account.
To explore how sound from wind turbines can affect individuals, it is important
to understand how it is processed by the human body and how the same sound
may affect people differently depending on physiological and psychological
factors or pre-existing medical conditions. This chapter describes how sound
is processed by the body through the auditory pathway, and how it can activate
non-auditory body structures such as the vestibular system. Finally, the chapter
explores how individuals may be affected differently by sound and provides an
overview of non-acoustic modifiers of sound perception.
44
4 . 1
THE A UD I TORY PAT H WAY
In order to be processed by the brain, sound has to reach the inner ear, which
converts sound into neural signals (Figure 4.1). The human ear can be divided
into three main sections:
The outer ear, composed of the pinna and the auditory canal, directs airborne
acoustic signals to the tympanic membrane (eardrum).
The middle ear, including the eardrum and delicate ossicular chain (ossicles
or tiny bones), converts acoustic vibrations into mechanical vibrations that
are transmitted to fluid-filled chambers in the inner ear.
The inner ear is composed of the cochlea and vestibular end-organs. Hair
cells in the cochlea detect waves created by the mechanical signals from the
middle ear, and convert them into electrical signals that are transmitted to
the brain via the cochleovestibular (VIIIth) cranial nerve.
Ossicles
Incus
Malleus
Stapes
Pinna
Cochlea
Auditory canal
Auditory nerve
Eustachian tube
Eardrum
Outer ear
Middle ear
Inner ear
Reproduced with permission from the University of Ottawa/Universit dOttawa
Figure 4.1
Sound Pathway from the Environment to the Brain
Airborne sound enters the ear through the outer ear and vibrates the tympanic membrane of the
middle ear. The ossicular chain transmits the mechanical vibrations into fluid-filled chambers of the
inner ear. Hair cells in the cochlea of the inner ear convert vibrations in the fluid into electrical signals
that are carried to the brain by the auditory nerve. Mechanical vibrations in the middle ear can also
result from skull vibrations conducted by bone from other parts of the body.
45
Airborne sound causes vibration of the eardrum and ossicles in the middle
ear. In the inner ear, the fluid within the cochlea moves in response to these
vibrations. The sound pathway described is called air conduction but is not the
only pathway for a sound to be perceived by the brain. Acoustic signals can also
cause vibrations that are conducted to the inner ear through bones. This bone
conduction pathway is also involved in hearing; when we speak, we hear our own
voice largely through bone-conducted sound (Hansen & Stinson, 1998). Bone
conduction is an important pathway when the head (or body) is connected to
a vibrating object, either internally (e.g., vocal apparatus) or externally (e.g.,
mechanical stimulator placed on the skull). By contrast, in response to external
sounds, bone conduction is 40 to 60 dB lower than airborne conduction in
normal-hearing individuals (Berger et al., 2003). As a result, bone conduction
is a significant pathway for external sounds only when the airborne pathway is
blocked (e.g., by use of a high-attenuation hearing protector). These conditions
do not apply to sound exposure from wind turbines.
Auditory signals are processed at many levels in the auditory pathway, including
at the brainstem and midbrain levels, but are not perceived consciously as
sound until there is activation of the auditory cortex.
4 .2
N O N - AU DI T ORY PAT H WAY S
Sound may also activate other parts of the body that are not part of the auditory
pathway, such as the vestibular system, or cause whole-body vibration through
direct mechanical stimulation of tissues.
4.2.1 Vestibular System
The vestibular system is part of the inner ear that contributes to balance and
spatial orientation. It may be activated by certain sound signals and is therefore
relevant in the context of possible effects of sound. The vestibular system is
composed of semicircular canals, which detect rotational movements (angular
accelerations) of the head, and the otoliths, which detect gravitational pull and
linear movements of the head. The vestibular system has some very important
functions that we rarely notice until things go wrong. For example, the stability
of the eyes is controlled by detection of head movements via the semicircular
canals. Balance is achieved in large part by input signals from the vestibular
system, which are combined with visual inputs and signals from the somatosensory
system to allow us to stand upright and generally orient ourselves in space.
In their normal balance function, the hair cells of the vestibular system are
typically activated by much slower movements of inner ear fluids than hair cells
in the cochlea. As a result, the vestibular hair cells are generally sensitive to
relatively low vibration frequencies. For example, the otolith organs are most
46
sensitive to acoustic signals around 100 Hz (Todd et al., 2008). The input from
the vestibular system is transmitted to the vestibular cortex of the brain, which
processes the information.
4.2.2 Whole-Body Vibration
Typically, to achieve whole-body vibration, direct mechanical stimulation from a
vibratory source (e.g., working with power tools) is required. However, acoustic
exposure at high sound pressure levels can also induce some amount of wholebody vibration (Takahashi, 2011). The levels of acoustic energy generated
by wind turbines, at commonly specified setbacks and/or sound levels (see
Chapters 2 and 3), are much lower than typically needed to affect body tissues.
4 . 3
B A S I C A UD IT ORY P E R CE P T I ON AND MEC H ANISMS
Perception of sound depends on many factors, including not only physiological

sensitivity but also psychological and external factors. Basic auditory perception
can be described by the following critical components:
Auditory sensitivity
Masked hearing threshold
Loudness
4.3.1 Auditory Sensitivity
Hearing thresholds reflect the combined sensitivity of a persons ear and the
persons ability to notice an environmental sound. (Other factors affecting
perception of sound are discussed in Section 4.4.) The absolute hearing threshold
is the lowest sound pressure level of a pure tone that a person can hear in a
very quiet environment with focused concentration. This threshold varies
depending on the frequency of the tone. Determining this threshold is one
way of measuring auditory sensitivity across frequencies in human subjects.
The healthy human ear is most sensitive between 2 and 5 kHz (roughly the
highest octave on a piano; middle C is 261 Hz), and is less sensitive to very low
and very high frequencies. Figure 4.2 shows the absolute hearing threshold
curve in a reference population of young adults with normal hearing. It
shows the differences in threshold as a function of sound frequency. For
example, at 1,000 Hz, the average threshold level is 4 dB, whereas at low
frequencies (<20 Hz), sound is perceived only at sound pressure levels above
90 to 100 dB. Note that, even within this reference population of young
normal-hearing adults, hearing thresholds typically vary by approximately
20 dB between individuals, as indicated in Figure 4.2 by the variance plots.
47
140
Source
ISO 389: 2004
Watanabe & Mller, 1990
2.56 sd (99%)
Sound Pressure Level (dB)
120
Variation (from Kurakata & Mizunami, 2008)

1 sd (68%)
2 sd (95%)
2.56 sd (99%)
100
80
60
40
20
0
10
20
40 60 100
1,000
10,000
Frequency (Hz)
Data Source: ISO 389: 2004; Watanabe & Mller, 1990; Kurakata & Mizunami, 2008
Figure 4.2
Absolute Human Hearing Thresholds
The central solid or dashed line indicates the sound pressure level (dB SPL) at which a typical young
adult can hear a pure tone under controlled conditions, at the frequency along the bottom axis. The
line indicates the median threshold: half of the population are more sensitive and can hear sounds at
lower pressure levels, while the other half are less sensitive and can hear sounds only at higher pressure
levels. The shaded blue bands around the line indicate the amount of variation: the darkest blue band
indicates the range of sound pressure levels at which approximately 68% of people can hear a tone
at a given frequency (one standard deviation) (see Leventhall, 2009). The lower the frequency, the
higher the sound pressure level needed for the sound to be heard. For example, infrasound (typically
a sound at a frequency under 20 Hz) can be heard only above 70 to 100 dB (Watanabe & Mller, 1990).
In the general population, hearing thresholds vary considerably and can be

much higher than the reference audiometric thresholds in Figure 4.2 as a
result of aging, prior exposure to loud sounds or ototoxic agents, medical
conditions, and a wide range of other factors. In audiological practice, the
hearing level of individual patients is expressed as the amount of threshold
elevation (measured in decibels hearing level [dB HL]) with respect to the
reference absolute threshold curve. For example, an individual with a hearing
level of 40 dB HL has absolute hearing thresholds that are 40 dB above that
shown in Figure 4.2. The amount of threshold elevation often varies with
frequency. There are various schemes to classify the severity of a threshold
elevation or hearing loss. Under the classification of the American Speech and
48
Hearing Association, normal hearing corresponds to hearing levels equal to

or below 15 dB HL, while hearing loss is classified on a scale ranging from mild
loss (16-25 dB HL) to severe loss (91+ dB HL) (ASHA, 1981). Box 4.1 provides
details on particular conditions affecting auditory sensitivity.
Box 4.1
Sensitive Groups and Pathologies
When considering the effects of acoustic signals through hearing or vestibular
pathways, it is important to be aware that certain individuals may have a lower hearing
threshold than normal and may thus be more sensitive to sound. As noted above,
even among people with normal hearing, thresholds can vary by as much as 20 dB.
There are also some conditions involving a heightened sensitivity to sound and a
lowering of hearing thresholds. For example individuals with hyperacusis can have an
unusual intolerance to ordinary environmental sounds that are normally tolerated
by the majority of people (Baguley, 2003). This condition may be relatively common,
with prevalence estimates ranging from 8 to 15% of the population. In regard to
possible sound activation of the vestibular system, some pathological conditions are
known to lower vestibular activation thresholds (e.g., enlarged vestibular aqueduct,
perilymphatic fistula, or superior semi-circular canal dehiscence). However, even in
these cases, the sound pressure level required for vestibular system activation is
much higher than that present in wind turbine noise (Harrison, 2014).
4.3.2 Masked Hearing Threshold

An absolute hearing threshold is rarely appropriate in the context of
environmental sound, as other sounds and noises prevent detection of sounds
at pressure levels near that threshold. Thus, in considering environmental
sound, the masked hearing threshold is often used. This is the sound pressure
level at which a sound can be heard in the presence of other sound (competing
acoustic signals). In the context of wind turbine noise, it is appropriate to
consider masked hearing threshold.
49
Box 4.2
Sound, Noise, and Loudness
Sound pressure is an objective measure of the amplitude of a sound wave. Loudness
is the subjective perception of the intensity of a signal after it has been detected
by the ear and processed in the auditory system. Noise can be described as an
unwanted sound. It may be a competing acoustic signal in the environment that
masks a sound of interest, or it can be a sound percept that individuals are trying
to separate out or ignore. In this respect, noise is subjective, and depends on the
individual and the context.
4.3.3 Loudness
When a sound or noise reaches the human ear and is above the masked
hearing threshold, the magnitude of the sensation in response to that sound
is termed loudness. Hence, loudness is the intensity of an auditory sensation;
it is related to, but distinct from, the physical acoustic signal that generated
it. In everyday terms, sounds are usually ordered on a scale extending from
soft to loud (Olson, 1972). Loudness is a subjective perception of the intensity
of a signal and depends not only on the characteristics of sound but also its
duration, its perception by the brain, and the presence of other sounds. Like
variations in the threshold of hearing, perception of loudness also varies from
person to person.
4 .4
S O UN D P E R CE P T I ON
Sound may have an effect on the brain through direct and indirect mechanisms.
The primary auditory pathway projects to the core auditory cortex and is
responsible for hearing (i.e., the percept of sound). The indirect or non-primary
pathway is a parallel ascending system mainly activating cortical association
areas and the limbic system (e.g., amygdala). This pathway is responsible
for the emotional response to sound stimuli. When defining the impact of
exposure to a sound, and the subsequent physiological stress reactions, both
mechanisms must be considered. The physical acoustic signal characteristics,
such as spectral (e.g., tonal versus broadband) and temporal (e.g., continuous
or modulated) characteristics, are perceived via the direct pathway, and the
emotional content of sounds activating the limbic system through non-primary
pathways (Spreng, 2000; Babisch, 2002; Mnzel et al., 2014). When a sound
is perceived as unpleasant, intrusive, or disturbing, it is sometimes referred
to as noise.
50
4 . 5
Sound reaches the inner ear through air conduction and bone conduction.
Signals are then converted into neural signals by hair cells in the inner ear and
are processed by different parts of the brain, both consciously and subconsciously.
Acoustic signals, especially those at low frequencies, can also stimulate the
vestibular system, and this stimulation occurs at lower levels for people with
certain conditions. Direct contact with a vibratory source or higher sound pressure
levels are typically needed to induce significant amounts of whole-body vibration.
Auditory perception can be described by auditory sensitivity, masked hearing
threshold, and loudness. Hearing thresholds reflect a persons auditory sensitivity
and ability to notice a sound in the environment. Absolute hearing thresholds
(in quiet surroundings with focused concentration) vary greatly in the general
population, by as much as 20 dB. Masked hearing thresholds tend to be
higher than absolute thresholds, and reflect the ability of people to notice
a particular sound in the presence of other competing sounds. Loudness is
the intensity of the auditory sensation, and its perception varies from one
individual to another. Some people may be more sensitive to sound than the
general population, either due to higher auditory sensitivity (lower hearing
threshold) or sound perception.
Sound signals are processed in various parts of the brain, through direct and
indirect mechanisms. In reviewing the evidence regarding potential health
effects of wind turbine noise, the Panel was informed by a schema proposed
by Babisch (2002) suggesting that both the direct and indirect effects of noise
lead to physiological arousal (see Chapter 5).
51
Chapter 5 Assessment of Evidence
5
Assessment of Evidence
Recent Reviews of Wind Turbine Noise

and Health
Proposed Adverse Health Effects
Prevalence of Reported Health Effects

and Noise Complaints in Canada
Conceptual Framework for Hypothesized

Health Effects
Evidence Regarding Sound from Wind Turbines

and Human Health
Weighing and Summarizing Evidence
Chapter Summary
52
Assessment of Evidence
Key Findings
The Panel identified 32 reported symptoms and health conditions attributed to
exposure to sound from wind turbines, based on a broad survey of peer-reviewed
and grey literature, web pages, and legal decisions.
The adverse health effects most widely attributed to wind turbine sound are
annoyance, sleep disturbance, and stress-related symptoms. This does not provide
evidence of causation, but does indicate a level of concern related to these possible
health effects.
The empirical research available to assess the support for a causal relationship
between wind turbine noise and adverse health effects is composed of a great
variety of sources, including peer-reviewed articles, conference papers, a graduate
thesis, a book, and grey literature.
More than 300 publications were found through a comprehensive search, and
these were narrowed down to 38 relevant studies.
To respond to the Charge, the Panel developed an approach based on the concepts
of evidence-informed public health proposed by the National Collaborating Centre
for Methods and Tools at McMaster University in 2008 (Ciliska et al., 2008). Broadly
speaking, when considering a public health issue, existing evidence concerning
the problem and its solution is found and weighed according to the strength of
its methods and findings; conclusions are based on the weight of evidence. In
using this approach, the Panel followed three major steps:
Define the issue: The Panel began by compiling, from a broad range of
sources, a list of health effects that have been attributed to wind turbines.
The health effects identified provided key terms used to search for relevant
empirical research.
Search and appraisal: The Panel searched for and appraised empirical
research papers discussing the effect of wind turbine noise on human health.
Synthesis: The Panel used Bradford Hills guidelines as a basis to weigh the
body of evidence and screen for plausibility of causation between exposure
to sound from wind turbines and specific health effects (Bradford Hill,
1965). In addition to empirical research specific to wind turbines, review
articles and results from research on other environmental noise sources were
used when appropriate, and when empirical evidence from wind turbines
was lacking. The Panels findings concerning causation, and the state of
evidence in each case, were summarized using standard language (HCN, 1994;
IARC, 2006). Figure 5.1 illustrates the major steps of the review process
undertaken by the Panel.
53
Literature Specific to Wind Turbines and Human Health

Environmental
Noise and Human
Health Literature
Empirical
Research
Critical
Appraisal
Reviews and
Discussions
Legal Decisions
(ERTs)
Web Pages
Reported Adverse
Health Effects
Conceptual Framework
Weight and Summary of Evidence
of a Causal Relationship
Sufficient
Inadequate
Limited
Lack of causality
Conceptual Framework:
Conclusions
Figure 5.1
Evidence Assessment Process
Brown lines show information used in defining potential health outcomes and in building models
of pathogenic mechanisms; blue lines show the literature review process with reference to causal
associations between wind turbine noise and each potential health effect. Empirical research, grey
literature, and sources such as legal decisions and web pages guided the Panel in listing health effects
possibly linked to wind turbine noise. Consideration of these health effects served the development
of a conceptual framework and search for empirical research specific to the effect of wind turbine
noise on human health. The empirical evidence resulting from this search was critically appraised and
constituted along with broader literature on the health effects of environmental noise. The body of
evidence for each proposed adverse health effect was reviewed based on Bradford Hills guidelines.
Findings were summarized in the report using language adapted from the International Agency
for Research on Cancer (IARC). The conceptual framework was updated, taking into account these
findings, and presented in Figure 7.1.
5 .1
R E C E NT RE V I E W S OF W I ND T UR B INE NOISE
A N D H E ALT H
Many reviews have been published on wind turbines and health, several
in the past five years alone. Among these recent reviews are those
commissioned by various orders of governments (HGC Engineering, 2010;
NHMRC, 2010, 2014; Masotti & Hodgetts, 2011; Ellenbogen et al., 2012;
Rod, 2012; Brisson et al., 2013; Hodgetts & OConnor, 2013; Merlin et al., 2013),
nonprofit organizations (e.g., Sierra Club Canada, 2011), and industry
organizations (e.g., Colby et al., 2009). Other reviews were either published in
peer-reviewed journals (e.g., Roberts & Roberts, 2013) or self-published online
(e.g., Frey & Hadden, 2012). In general, these reviews were inconsistent with
regard to their findings on the effects of wind turbine noise on human health
54
and were not used in the critical evaluations. However the Panel used them to
help guide and frame issues in the current context and in ensuring that key
findings were communicated clearly.
5 . 2
P R O P O S E D ADV E R S E H E ALT H E F F E C T S
The Panel began by identifying health conditions and symptoms that have
commonly been attributed to wind turbine noise. The Panel took a broad
approach to assembling such a list, including a review of the scientific peerreviewed literature related to wind turbine noise, as well as a sample of lay
literature such as web pages, self-published reports or books, and legal decisions.
Each source was scanned for mentions of symptoms attributed to wind turbine
noise. These documents were not assessed by the Panel, and the evidence
supporting claims of association was not evaluated at this stage.
Twenty internet sites (36 web pages) were identified from references in recent
reviews, empirical literature, and a bibliography compiled from submissions
to Health Canada during consultations on the design of a wind turbine noise
and health study (Health Canada, 2013a). The Google search engine was
used to search each site (web domain) for pages published since 2009 that
discuss wind turbine sound and human health.3 These pages included blog
posts, individual accounts, and other types of web documents. Some sites also
had pages devoted primarily to summarizing health effects of wind turbines,
easily identifiable from the home page, and those were also scanned as part
of the process. In addition, decisions from Environmental Review Tribunal
(ERT) hearings in Ontario, Canada, were scanned for health effects claimed
by appellants or described by expert witnesses.
Only health effects attributed to wind turbines reported in the last five years
(from 2009 to 2014) were included, to ensure that they are current and relevant
to modern utility-scale wind turbines. Older wind turbine models and designs
produce higher levels of sound, including infrasound and low-frequency sound,
than modern wind turbines (see Section 2.1). In addition, sound propagation
factors differ between the new and old turbines owing to the height of the
wind turbines.
3 Search terms used: (wind turbine* or wind farm*) and (sound or noise) and (health or
sick or symptom*). If no results were found, a second search was attempted using the terms:
wind turbine noise health.
55
Table 5.1 presents 32 health conditions that have been attributed4 (proposed
but unconfirmed) to wind turbine noise in various types of sources (e.g.,
self-reported individual case, compiled list). This list was the starting point
for the Panel to assess the evidence for causal relationships between these
outcomes and wind turbine noise but does not constitute in any way evidence
for causal relationship.
Peer-Reviewed
Review
Conference
Proceedings
Grey
Literature
Legal
Decision
20
13
Annoyance
Sleep disturbance
Stress, tension
Health-related quality of life
Condition or Symptom
Number of Sources Reviewed
Vibroacoustic disease
Web Page
Peer-Reviewed
Study
Table 5.1
Proposed, but Unconfirmed, Adverse Health Effects Attributed to Wind Turbine Noise
in at Least Three Documents Reviewed
36
Cardiovascular System
Cardiovascular disease
High blood pressure (hypertension)

Irregular heartbeat (cardiac dysrhythmia,
tachycardia)
Endocrine System
Diabetes
Immune System
Impaired immunity
Musculoskeletal System
Back pain
Joint pain
Muscle pain (myalgia)
Shaking (palsy)
continued on next page
This list is not exhaustive and is meant to capture the major issues only. A single source document
may contribute to multiple conditions and symptoms (dots) in the table.
56
Fatigue
Headache
Nausea
Pressure in the chest
Sensation of internal vibration
Vertigo, dizziness
Web Page
Legal
Decision
Disturbances of skin sensation
Grey
Literature
Peer-Reviewed
Review
Cognitive or task performance
Conference
Proceedings
Peer-Reviewed
Study
Nervous System (General)
Vision problems
Nervous System (Auditory)

Communication interference
Ear pressure or pain
Anxiety
Depression
Hearing loss
Tinnitus
Psychological Health
Irritability
Psychological distress
Respiratory System
Nosebleed
This list was used as a starting point for the Panel to assess the evidence for causal relationships
between these health outcomes and wind turbine noise. It includes health effects attributed to wind
turbine noise, without assessing the strength of evidence. It includes self-reported individual reports
and compiled lists from various sources such as web pages, and grey and peer-reviewed literature. A
dot corresponds to one or more mentions for the given type of document, but only symptoms
mentioned in at least three separate documents of any type are presented. Symptoms associated with
wind turbine syndrome (Pierpont, 2009) are considered individually in the list above.
5 .3
57
P R E VAL E NCE OF R E P ORT E D H EALTH EFFEC TS

A N D NOI S E C OM P L AI NT S I N C ANADA
The Panels ability to assess the prevalence of adverse health effects related
to wind turbine noise in Canada was limited by a lack of available data. Public
health surveillance for chronic diseases falls under the responsibility of federal,
provincial, and territorial governments (Health Canada, 2003a). However, no
formal processes or surveillance programs exist in Canada to capture potential
health outcomes resulting from wind turbine noise. Neither did the Panel identify,
through its literature review, research studies in Canada or internationally that
had estimated prevalence. In the absence of such data, estimates of disease
burden attributable to wind turbine noise could be derived from data on (1) the
diseases causally associated with wind turbine noise; (2) the incidence of each
such disease in the general population; (3) the proportion of the population
exposed to wind turbine noise; and (4) the incremental risk of developing the
disease associated with exposure to wind turbine noise. However, calculating
such estimates is currently challenging because of a paucity of data with respect
to all four of the necessary parameters.
In 2013, the Pembina Institute studied documented formal noise complaints
concerning wind turbines in Alberta made to various authorities. This study
includes noise complaints to regulatory bodies, municipalities, and wind
farm operators, covering approximately 90% of the wind energy capacity in
the province (Thibault et al., 2013). This study found five noise complaints
related to wind turbine operations between 2007 and 2011 across Alberta. They
concluded that noise complaints to authorities related to wind turbines had
been infrequent and measurably fewer than complaints related to conventional
energy activities such as oil and gas operations (Thibault et al., 2013).
There are limitations to using complaints to estimate prevalence of health
outcomes. Alberta may not represent the whole of Canada, and the situation
may be different in other parts of the country. For example, there are likely
many more noise complaints in Ontario, given the many cases before the
Environmental Review Tribunal. No systematic review by independent parties
of noise complaints across jurisdictions in Canada has been performed, likely
because of the lack of consistency in reporting mechanisms for such complaints
or health reports, the collection of complaint information, and whether
complaints are even related to health (as opposed to economic disparity or
visual impacts, for example). However, there are studies of noise complaints in
other countries, such as Australia, where policies require that noise complaints
made to wind project developers are collated and made available to the public.
Such policies allow for an analysis of these complaints, showing uneven patterns
of complaints focused on specific projects within the country, several years
58
following initial construction. In Australia, from 2006 to 2012, 129 complaints

were received by wind turbine developers corresponding to about 0.4% of
the population living within 5 km of a wind turbine (Chapman et al., 2013).
However, conclusions on the probability of complaints in Canada cannot be
reached based on other countries studies of complaints.
5 . 4
C O N C E P TUAL F RAM E W OR K F OR H Y POTH ESIZED

HE A LTH E FFECT S
The Panel developed a framework (Figure 5.2) that included the proposed but
unconfirmed adverse health effects of wind turbine noise described in Table
5.1, as well as possible mechanisms relating noise and health to the various
component characteristics of wind turbine noise. This framework was used to
guide the search for, and evaluation of, relevant evidence concerning these
causal relationships between wind turbine noise and the health outcomes in
Table 5.1. The analysis is presented in Chapter 6, and an updated figure (after
review of the evidence and the final conclusion of the Panel) can be found
in Chapter 7.
Exposure
Mechanisms
External Stimuli
E.g.,
Low Frequency Sound
ReceptorMediated
(inner ear,
vestibular)
and
Other Physical
Effects
(not mediated
by auditory
receptors)
Modifying Factors
E.g.,
Noise Sensitivity
Health Condition
Activity
Coping Ability
Adverse Health Effects
Sleep
Disturbance
Annoyance
Physiological
and
Psychological
Changes
E.g.,
Stress
Reactions
Blood Pressure
Blood Glucose
Mood
Diseases
E.g.,
Auditory
Cardiovascular
Immune
Endocrine
Nervous
Psychological
Respiratory
Cognitive and
Emotional
Response
Figure 5.2
Proposed Elements of Potential Relationships Between Wind Turbine Noise and
Adverse Health Effects
This framework includes proposed physical mechanisms, some of which are mediated by effects on
auditory and vestibular receptors, as well as proposed effects mediated by a persons cognitive and
emotional response to sound.
5 .5
59
E VI D ENCE R E GARDI NG S OU ND FR OM WIND TU R B INES

A N D H U M AN H E ALT H
The empirical research used to assess causal associations between exposure

to sound from wind turbines and the development of adverse health effects
was found in peer-reviewed literature and conference papers. The Panel also
considered relevant grey literature, such as technical reports, and reports
published by not-for-profit organizations. All documents presenting results of
empirical studies were reviewed (see Appendix B), and only studies assessing the
effects of modern utility-scale wind turbines5 (500 kW or more) were considered.
5.5.1 What Can Scientific Studies Tell Us About Causality?
The ability to draw valid conclusions from a health study depends largely on
the quality of that study. In an experimental study, all aspects of the study are
controlled by the researcher and are thus modifiable and reproducible. In
most population health studies, such control is not possible. Factors such as
peoples location, age, when the wind turbine is installed, and exposure to
noise cannot be controlled. This does not mean that observational studies
(as epidemiologists call studies of people in the real world) are necessarily
of a lower quality than experimental studies. Carefully designed, well-executed
studies that consider the impact of all factors that may influence relationships
between an exposure and outcome are very powerful (Howick et al., 2009).
High-quality observational studies have informed many of the major public
health advances in the past century, such as the link between lung cancer and
smoking, and asbestos and mesothelioma.
Few if any studies are perfect, meeting every test of quality, but that does not
mean they are without value. It does mean, however, that a determination of
causality is more reliable when that determination is based on the weight of
evidence provided by multiple studies, because of factors such as consistency
of findings across studies of different designs, different analysis techniques,
different populations, etc. A consistent finding of a relationship across several
studies strengthened the Panels confidence that the observed relationship is
causal and not a simple association.
5 All wind turbines in Canada are modern (in the last 20 years) and few (1.6%) are not
utility-scale (<500 kW). See Figure 2.1.
60
5.5.2 Assessing the Methodological Quality of a Study

Study validity is threatened by poor design, errors in measurement, or failure
to take into account confounders (i.e., factors that muddy the relationship
between a cause and effect, such as between wind turbine noise and sleep).
Below is a brief review of some of the main issues (or biases) that commonly
affect overall methodological quality and study validity:
Study Design
Study design strongly influences methodological quality (see Appendix A).
A randomized controlled trial or experiment is generally considered to be the
highest-quality study design, but these are rarely feasible in population-based
studies. Two types of longitudinal study designs cohort and case-control
studies follow a study population over time. These designs analyze natural
experiments, since the researcher does not control the conditions under
which the study population lives. Cohort and case-control studies may use
sophisticated design and analysis tools to improve validity. At the other end
of the spectrum are case reports; while these have some usefulness in initial
identification of associations between exposures and diseases, they are prone
to many forms of bias, affecting validity.
Sample Size
To reduce the impact of random differences between disease risk in exposed
and unexposed groups, a certain number of participants or study subjects is
required. Generally, studies of larger groups have more statistical power than
those of smaller groups to detect effects that are less pronounced (i.e., larger
studies increase confidence that any observed difference in outcome between
two groups is real, and not just random chance). If studies are also looking
at exposed sub-groups (e.g., males and females, several age groups) and also
multiple confounders, then study size may need to be very large.
Confounding Factors
Scientists are always concerned that, when they identify a possible relationship
between an exposure and an outcome, the relationship might have been caused
by some other unknown factor. For example, if it were found that gamblers
were more likely to get lung cancer, researchers would be cautious to label this
a causal relation, because casinos are smoky places; perhaps it is smoking, and
not gambling, that is to blame for the excess rate of lung cancer in the subject
population. These factors are called confounders. However, there are several
techniques scientists can use to control the impact of confounders on a study.
61
Misclassification of Exposure or Outcome

The quality of the measurement of the exposure (such as sound from wind
turbines) or the outcome (such as blood pressure) is very important to the
quality of a study and to its validity. If a study does not accurately measure these
two variables, then it may misclassify exposed subjects as being unexposed and
vice versa. Similarly, without a good measure of an outcome, a study may be
lumping people who do not have a disease in with those who do, thus diluting
the study group. In both examples of misclassification (of exposure or of
outcomes), the apparent relationship between exposure and outcome may
be diminished relative to the true relationship, which thus becomes obscured.
In reviewing studies, a key concern is how accurately the study measures exposure.
Studies on noise, for instance, can be highly complex because noise has several
parameters (amplitude, frequency, variation in time) that make up the acoustic
signal to which a subject is exposed. In the case of an outcome or disease, a case
definition (standard criteria for deciding whether a person has a particular disease
or condition (Columbia University, n.d.)) that is systematic and ideally based on
an unbiased test, such as a blood pressure examination, is needed. Self-reported
outcomes are generally considered more prone to misclassification or bias.
Selection Bias
Scientists must be very careful in deciding how individuals are recruited into
studies to avoid biases that could skew the results. Ideally, a study includes a
random selection of individuals who are either exposed or unexposed (for a
cohort study), or those with a disease and those in whom the disease is absent
(in a case-control study). Self-selection bias may occur if, for example, a scientist
asks for volunteers to participate in a study; volunteers are often those with a
special interest in the subject, and may differ in important ways from the general
population. A biased selection of subjects is not representative of the general
population, and the study results may therefore be skewed away from the truth.
Summary
Observational epidemiological studies can be very powerful tools in discovering
and explaining causal associations that is, does a specific exposure cause a
certain disease outcome? However, most studies have strengths and weaknesses,
and must be reviewed carefully in order to identify problems that may affect
the studys validity, to assess the body of knowledge (i.e., multiple studies), and
to consider the weight of evidence. Other factors such as biological plausibility
and experimental evidence must be looked at, too, when assessing causality.
A short summary of the evidence reviewed is given in Appendix B, which is
intended to be used as a supplement to Chapter 6.
62
5 . 6
W E I G HI N G A ND S U M M ARI Z I NG E V I D ENC E
5.6.1 Searching for Empirical Research

The aim of the search for evidence was to be as comprehensive as possible. The
Panel searched the SCOPUS and PubMed databases using keywords related
to wind turbines, general health, and specific health conditions attributed to
wind turbine noise (see Appendix B). Search results from both databases were
combined and duplicates removed, producing over 300 articles published
in peer-reviewed journals and conference proceedings between 1966 and
2014. Abstracts and titles of papers from all of these sources were scanned to
ensure relevance, which narrowed the number of articles and reports to 33. The
Panel also considered evidence submitted to Health Canada during
consultation on the design of a wind turbine noise and health study (Health
Canada, 2013b), literature cited in key reviews, and Panel members input.
Through this process, eight additional studies, including a graduate thesis, a
book, and grey literature, were added to the evidence base. In total 38 studies
were therefore included and critically appraised. No time limitation was applied
to studies for inclusion; however, the relevance of the wind turbine design was
verified. The quality and weight of the evidence presented in an empirical study
was critically reviewed using a questionnaire based on assessment factors used
by the U.S. Environmental Protection Agency (EPA, 2003). This critical review
informed the Panel in describing the evidence.
The Panel also extended the search for empirical literature discussing the
health effects of specific sound characteristics produced by wind turbines,
such as infrasound, low-frequency sound, and amplitude modulation. This was
undertaken to look for additional evidence on the possible mechanism that
could cause adverse health effects, where evidence was lacking. The Panel also
used additional evidence from peer-reviewed environmental noise literature.
5.6.2 Assessing Causality
Epidemiological studies cannot determine the cause of an outcome in a given
individual, or even the mechanism responsible, but they can establish an
association between a given exposure and the frequency of an outcome in a
population (e.g., many epidemiological studies assessed the association between
smoking and lung cancer). Causation can be inferred, usually based on several
factors, such as the strength and consistency of the association, mechanistic
plausibility, as well as the temporal sequence and biological gradient or doseresponse relationship of the exposure and the outcome (Bradford Hill, 1965;
Howick et al., 2009). Along with coherence, specificity (a reliable association
between exposure and a line of outcomes), and evidence from experiments and
63
analogy, these guidelines were proposed by Bradford Hill (1965) for evaluating
the plausibility of causal relationships based on observational studies, even in
the absence of randomized controlled experiments.
A critical appraisal guided the Panel in assessing and assigning weight to
the evidence linking wind turbine noise to health effects. The Bradford Hill
guidelines were used to guide Panel deliberations and to structure the summaries
of evidence (Chapter 6), keeping in mind that they are not intended to be
strict guidelines, and should be applied to a body of evidence rather than to
individual studies. The final determination of causality was ultimately based
on the Panels judgment of the findings.
5.6.3 Summarizing Findings
The Panel further adopted standard language to summarize the findings of causal
relationships, following a framework similar to that used by the International
Agency for Research on Cancer (IARC, 2006), also adopted by the Health
Council of the Netherlands (HCN, 1994). According to this scheme, the overall
strength of evidence for a causal relationship falls into one of four categories:
Sufficient evidence of a causal relationship: A relationship was observed
between exposure to sound from wind turbines and a specific health effect
in studies in which chance, bias, and confounding factors can be ruled out
with reasonable confidence.
Limited evidence of a causal relationship: An association was found between
exposure to sound from wind turbines and a health effect for which causal
interpretation is considered by the Panel to be plausible, but for which chance,
bias, and confounding factors cannot be ruled out with reasonable confidence.
Inadequate evidence of a causal relationship: The available studies are of
insufficient quality, or lack the consistency or statistical power to permit a
conclusion about the presence or absence of a causal relationship.
Evidence suggesting lack of causality: Several adequate studies covering
the full range of exposure are available that are mutually consistent in not
showing a positive association between exposure and effect at any observed
level of exposure.
5 .7
The Panel considered a range of evidence sources in various ways:

to identify health outcomes that have been attributed, claimed, or reported
to be linked to wind turbine noise;
to develop a conceptual framework of hypothesized pathways linking wind
turbine noise to the identified outcomes; and
to weigh evidence for these causal relationships.
64
The Panel identified 32 potential adverse health effects attributed to wind turbine
noise, with sleep disturbance, annoyance, and stress effects being mentioned
the most often in the sources reviewed. Literature on the outcomes identified
was searched and reviewed systematically. A total of 38 empirical studies were
selected and assessed for methodological quality, and these were supplemented
by relevant peer-reviewed articles from the environmental noise literature. The
Panel used the causation guidelines of Bradford Hill to inform and structure the
review, and a framework adapted from the International Agency for Research
on Cancer to summarize the findings for evidence of causal relationships.
Chapter 6 Current Evidence Related to Adverse Health Effects of Wind Turbine Noise
6
Current Evidence Related to Adverse Health
Effects of Wind Turbine Noise

Annoyance
Sleep Disturbance
Stress
Cardiovascular Diseases
Diabetes
Effects on Hearing
Tinnitus
Effects of Non-Audible Sound on the Inner Ear
Cognitive and Mental Performance
Psychological Health
Quality of Life
Other Health Impacts
Chapter Summary
65
66
Current Evidence Related to Adverse Health Effects

of Wind Turbine Noise
Key Findings
Primary evidence is available for some, but not all, health effects attributed to wind
turbine noise. The following findings are based on available primary evidence, mainly
from population-based studies and in some cases experimental studies:
The current state of the evidence is sufficient to conclude that a causal relationship
between exposure to wind turbine noise and annoyance can be established;
however, knowledge gaps remain with regard to the influence of specific sound
characteristics, such as amplitude modulation, low frequency content, or visual
aspects of wind turbines, which are difficult to study in isolation.
The current evidence is limited with regard to the relationship between exposure
to wind turbine noise and sleep disturbance. The available evidence suggests that
a direct or indirect mechanism might exist, but confounding factors cannot be ruled
out with reasonable certainty.
The available evidence is inadequate to establish the presence or absence of a
causal relationship between exposure to wind turbine noise and stress. Similar to
sleep disturbance, available evidence suggests that a direct or indirect mechanism
might exist; however, the evidence lacks methodological and statistical strength.
The Panel concluded that there is evidence of no causal relationship between
exposure to wind turbine noise and hearing loss.
While statistically significant associations between exposure to wind turbine noise
and diabetes and tinnitus have been found, these associations are not consistent
across studies and evidence remains inadequate to determine the presence or
absence of causal relationships.
Primary research has also addressed cardiovascular diseases, effects of non-audible
sound, cognitive and mental performance, psychological health, and general
quality of life; however, evidence remains inadequate to determine the presence
or absence of causal relationships between exposure to wind turbine noise and
any of these health effects.
Using its search and assessment process (Chapter 5), the Panel reviewed
evidence on adverse health effects, and assessed and summarized the current
body of evidence. This chapter reviews the potential health effects identified
in the preliminary scan of health outcomes (Table 5.1) and offers the Panels
conclusions about the state of the evidence regarding each outcome. Table
B.2 (in Appendix B) summarizes the evidence from empirical research on
67
health impacts of wind turbine noise. The Panel identified the datasets that
have been used in the papers reviewed in this report and gave each dataset a
code identifying the country of a study and the year the study was undertaken
(e.g., NL-07), as presented in Table B.2. This table also provides an overview of
the key methodological aspects of each study. The following sections provide
a detailed review of the evidence for all health outcomes that have been the
subject of population-based studies or laboratory experiments. The Panels
findings with regard to these outcomes are summarized in Chapter 7. The
Panel also considered possible causal mechanisms linking exposure to wind
turbine noise and other health outcomes that have been mentioned in case
series or other sources, but have not yet been the subject of empirical research.
6 .1
A N N O YANC E
Noise annoyance can be defined as a feeling of displeasure evoked by a

noise and any feeling of resentment, displeasure, discomfort and irritation
occurring when a noise intrudes into someones thoughts and moods or
interferes with activity (Passchier-Vermeer & Passchier, 2006). Annoyance is
the most common and most studied effect of noise on individuals. Whether
a person becomes annoyed may depend on the sounds characteristics (e.g.,
how intense it is, how it varies with time, and what frequencies it contains),
and on factors related to the individual (e.g., noise sensitivity, attitude towards
the noise source, physiological and psychological state), and on situational or
contextual factors (e.g., activities performed or intended to be performed,
expectations of disturbance, noise source-related factors such as controllability of
the noise source, fear, and permanence). The Panel used this broad descriptive
definition, but stresses that in most studies described below, researchers used
questionnaire data in which respondents replied based on their own perception
and recollection of annoyance.
Pedersens (2011) re-analysis of three study datasets (SWE-00, SWE-05,
NL-07) showed a statistically significant association between exposure to wind
turbine noise and annoyance. Pedersen found that an increase in the estimated
A-weighted sound pressure level was associated with an increased proportion of
participants being annoyed. In these three datasets, the proportion of respondents
annoyed outdoors ranged from 5 to 20% at 35-40 dB(A) and from 10 to 45%
at 40-45 dB(A) (Persson Waye, 2009). Annoyance was the only health effect of
wind turbine noise that was consistently associated with estimates of A-weighted
sound levels in all three datasets. Preliminary analyses of studies in Poland
(POL-13) and Japan (JAP-10) found similar exposure-response relationships
between A-weighted sound levels and annoyance (Kuwanoet al.,2013; Yano
et al.,2013; Pawlaczyk-uszczyska et al., 2013; Pawlaczyk-uszczyska et al.,2014).
In the Polish study (Pawlaczyk-uszczyska et al., 2013), the percentage of
68
respondents who were annoyed (i.e., those who reported being rather annoyed,
annoyed, or extremely annoyed on a five-point verbal scale) by exposure
to wind turbine noise outdoors increased from 27.1% at 35-40 dB(A) to 63.6%
at 45-50 dB(A). Similarly, the percentage of subjects annoyed by exposure to
wind turbine noise indoors increased from 18.6% at 35-40 dB(A) to 23.4% at
40-45 dB(A), but decreased to 18.2% at 45-50 dB(A). In the Japanese study
(Yano et al., 2013), the share of respondents who were very or extremely
annoyed increased from 9.7% to 22.6% as night-time exposure levels increased
from 30 to 45 dB(A).6
To date, few studies have used actual measurements of noise exposure rather
than estimated exposure levels. Sound measurements in the field are costly, timeconsuming, and difficult to conduct consistently across large population groups.
Therefore, studies using sound measurements have been field experiments
rather than population-based studies. Bockstael et al. (2012) conducted regular
sound measurements at eight households located between 270 metres and
approximately 750 metres from the closest of three wind turbines. The authors
found that noise exposure and annoyance depended on wind speed, wind
direction (i.e., angle of the rotating turbine to the exposed households), as
well as the energy output of the wind turbines. Magari et al. (2014) conducted
short-term indoor and outdoor sound measurements at 52 households located
approximately 400 to 800 metres from the closest turbine of a large wind park.
Average LAeq measures were 47 dB(A) (standard deviation: 11.5 dB(A)) indoors
and 45.3 dB(A) (standard deviation: 8.2 dB(A)) outdoors. The researchers
also surveyed residents, and their survey results did not support an exposureresponse relationship between short-term indoor or outdoor noise exposure
and self-reported annoyance. However, the data did show correlations between
measured noise exposure and concern about health effects, and between noise
exposure and the prevalence of sleep disturbance and stress. The authors
noted that larger cohort studies with sound measurements taken indoors
and outdoors are necessary to verify the divergent results from studies using
calculated exposure measures. Zajamsek et al. (2014) presented a method to
simultaneously record time-series noise data and corresponding annoyance
ratings submitted by exposed subjects. The authors tested the method at two
homes at a distance of 2.5 km and 8 km from the nearest wind turbine. While
the recordings showed sound patterns that correlated with the wind turbines
energy output, the overall noise level and annoyance of residents at these
6 Although all studies used a five-point verbal scale, the terminology used is different among
papers. For example extremely and very annoyed in Yano et al. (2013) corresponds to
extremely annoying and annoying in Zajamsek et al. (2014) and very annoyed and rather
annoyed in Pedersen (2011). There are therefore differences due to language but the results are
considered comparable.
69
distances were better explained by the prevailing wind speeds at the residences
(Zajamsek et al., 2014). In population-based studies with large sample sizes
included in this review, the relative risk of being annoyed by wind turbine
noise increased with estimated outdoor sound pressure levels, suggesting that
annoyance follows an exposure-response relationship (see Figure 6.1).
Similar exposure-response relationships based on sound pressure level
were also found in laboratory experiments involving humans. Among these,
Lee et al. (2011) conducted an experiment to identify the role of amplitude
modulation on annoyance, using recorded samples of sound from wind turbines
to simulate amplitude modulated exposure with modulation depth between
5 and 12 dB at equivalent sound pressure levels between 35 and 55 dB(A). The
results showed that annoyance increased with both the equivalent sound pressure
level of exposure and the amplitude modulation depth of wind turbine sound.
Seong et al. (2013) investigated annoyance using sound samples representing
exposure at different distances and angular positions relative to the direction
of the wind hitting a wind turbine, with LAeq ranging approximately from
25 to 50 dB(A). Maffei et al. (2013) and Ruotolo et al. (2012) used audiovisual simulations of distance, including recordings of real-world exposure, as
auditory stimuli. Fastl and Menzel (2013) showed that annoyance decreased
with sound pressure level in subjects exposed to amplitude modulated
sound stimuli between 50 and 38 dB(A). While none of these experiments
(Ruotolo et al., 2012; Fastl & Menzel, 2013; Maffei et al., 2013) were specifically
designed to investigate the exposure-response relationship, the results consistently
showed that increasing levels of sound from wind turbines are associated with
higher levels of annoyance among those exposed.
Annoyance can be caused by a multitude of factors, including several that often
occur together with exposure to wind turbine noise, such as visual impacts
of wind turbines. The studies reviewed by the Panel controlled for different
combinations of contributing factors, including self-identified noise sensitivity
(Pedersen & Persson Waye, 2004, 2007; Pawlaczyk-uszczyska et al., 2013);
background noise from road traffic (Bakker et al., 2012); possible masking of wind
turbine noise from other noise sources (Pedersen et al., 2010; Van Renterghem
et al., 2013); self-identified personal attitudes (Pedersen & Persson Waye, 2004;
Pawlaczyk-uszczyska et al., 2013); the influence of background sound on
annoyance from wind turbine noise (Bolin et al., 2012), participation in economic
benefits from wind turbine operation (Pedersen & Persson Waye, 2004; Janssen
et al., 2011); and characteristics of living environments (Pedersen & Larsman, 2008).
The observed association between wind turbine noise and annoyance generally
remained after controlling for these factors. Regardless, while cross-sectional
studies can control for these factors when assessing the relationship between
70
wind turbine noise and annoyance, they suffer from an inability to determine
the temporality of the relation (i.e., whether exposure to wind turbine noise
leads to negative attitudes or whether negative attitudes affect noise perception).
A factor in perception of and annoyance due to wind turbine noise is the
visual impact of turbines. In several studies, residents in direct line of sight
of a wind turbine were more likely to be annoyed than those who could not
see a wind turbine (Pedersen & Persson Waye, 2004; Pawlaczyk-uszczyska
et al., 2013). Similarly, residents who valued the visual appeal and quietness of
rural landscapes were more likely to be annoyed than those who viewed the
countryside as a place for economic opportunity (Pedersen & Persson Waye,
2004; Mulvaney et al., 2013). Laboratory experiments using short-term exposure
to infrasound have shown that positive and negative expectations can influence
self-reported symptoms and impacts on mood in positive and negative directions,
respectively (Crichton et al., 2014a, 2014b). Bockstael et al. (2013) found that
subjects who could recognize sound from wind turbines among other types of
environmental sound were more likely to be annoyed than subjects who noticed
the sound without recognizing its source. The authors suggested these results
could support a hypothetical causal pathway in which noticing a wind turbine
is followed by an appraisal step, the outcome of which influences the degree
of annoyance (Bockstael et al., 2013).
Pedersen et al. (2009) noted that their study design could not exclude the
possibility that negative attitudes to wind turbines are caused in part by noise
exposure or that annoyance is more strongly associated with visual intrusion than
with sound exposure (see also Ellenbogen et al., 2012). Pedersen and Larsman
(2008) used a series of models to simultaneously account for different aspects of
attitude including visual attitude and other impacts of wind turbines, and found
that the association between wind turbine noise and annoyance is significant
even when other factors are considered, but that a negative visual attitude
would enhance the risk for noise annoyance for people living on flat terrain.
Janssen et al. (2011) used the responses from the Swedish and the Dutch studies
(SWE-00, SWE-05, NL-07) to compare the exposure-response relationship for
annoyance from wind turbine noise with annoyance associated with air, road,
and rail transportation noise. The authors found that annoyance from wind
turbine noise occurred at relatively low sound pressure levels of 40 to 45 dB(A).
The model predicted that the percentage of persons annoyed by wind turbine
noise indoors was higher than the percentage annoyed by other sources of
sound at the same sound pressure level and grew faster when sound pressure
levels increased (Figure 6.1).
71
50
50
40
40
Air
% Highly Annoyed
% Annoyed
Road
30
20
Wind
30
20
10
10
Rail
25
30
35
40
45
L den
50
55
60
65
25
30
35
40
45
L den
50
55
60
65
Reproduced with permission from Janssen, S. A., Vos, H., Eisses, A. R., & Pedersen, E. (2011). A comparison
between exposure-response relationships for wind turbine annoyance and annoyance due to other noise sources.
The Journal of the Acoustical Society of America, 130, 3746-3753. Copyright 2015, Acoustical Society of America
Figure 6.1
Comparison of Annoyance Due to Wind Turbine Noise and Transportation Noise
Comparison of the percentage of residents annoyed (left) or highly annoyed (right) indoors due
to wind turbine noise (wind) and due to traffic noise (air, road, rail), based on data from datasets
SWE-00, SWE-05, and NL-07. For comparison, sound exposure measures are expressed as Lden values
calculated using the A-weighted immission levels determined in the original studies in accordance
with the European Union environmental noise guidelines. Lden (day/evening/night sound level, also
referred to as community noise equivalent level or CNEL) expresses the average sound level over a
24-hour period with a penalty of 5 dB added for the evening hours (7 pm to 10 pm) and a penalty of
10 dB added for the night-time hours (10 pm to 7 am). The calculation of Lden is based on a complex
protocol that includes correction factors for specific conditions affecting noise exposure at the location
of each respondent, such as prevailing wind speeds and direction or topography.
Earlier research suggested that the higher risk of annoyance was related to
the specific sound characteristics of wind turbine noise, including amplitude
modulation and tonality. For example, participants in an experimental study
described wind turbine noise as lapping, swishing, or whistling, and perceived
these sound characteristics as annoying (Persson Waye & hrstrm, 2002).
In another experiment, Lee et al. (2011) used stimuli based on recorded
sound samples that were processed to simulate sounds with different levels
of modulation depth. They showed that the amplitude modulation of sound
from wind turbines is perceived as more annoying than continuous sound with
the same frequency and average sound level. Based on a similar experiment,
Seong et al. (2013) suggested that a measure of weighted maximum sound
level that captures the peaks of amplitude modulation (LAFmax), rather than
daily averages (LAeq), is the best predictor for the risk of annoyance. A recent
72
report prepared by the University of Salford for RenewableUK, a wind energy

industry group, included a series of listening tests with 20 participants using
sound stimuli from real-world recordings of wind turbine noise with increasing
modulation depth at constant LAeq (RenewableUK, 2013). The objective was to
explore how much higher the level of an unmodulated sound would have to be
in order to be perceived as equally annoying to a modulated sound. The results
showed that increasing modulation depth led to increasing annoyance ratings.
On average, the adjustments necessary to express the perceived difference in
loudness of a sound with amplitude depth varying from 0 dB to 12 dB were
3.5 dB for a 30 dB(A) test sound and 1.7 dB for a 40 dB(A) test sound. In other
words, a modulated sound of LAeq of 30 dB(A) was perceived to be as loud as
an unmodulated sound of 33.5 dB(A) (von Hnerbein et al., 2013).
A similar study by Yokoyama et al. (2013) used 52 low-frequency (1-250Hz)
stimuli with amplitude modulation that was generated with sound recordings
from wind turbines, in order to test the subjective noise perceived by 10 subjects.
This study found that perceived noise increased with increased amplitude
modulation depth, although the time-averaged sound pressure level remained
the same (Yokoyama et al., 2013). While the samples in these experiments
were too small to generalize the findings, the results suggested that amplitude
modulation may lead to higher annoyance than an unmodulated sound at the
same equivalent sound pressure level. Furthermore, higher annoyance ratings
could be linked to modulation depth in an exposure-response relationship
(Yokoyama et al., 2013). Von Hnerbein et al. (2013) also noted that, in realworld conditions, spectral characteristics (tonality) vary along with amplitude,
with both effects likely to affect annoyance. These findings are consistent with
results of studies on other types of noise with amplitude modulation, such as
exhaust noise (Kantarelis & Walker, 1988), or noise from heating, venting, and
air-conditioning systems (Bradley, 1994).
Some researchers have suggested that the higher degree of annoyance associated
with wind turbine noise could be related to its low-frequency components. Earlier
research showed that low-frequency noise produced by fans was perceived as
more annoying than other types of sound at reference levels over 40 dB(A)
(Persson Waye & Bjrkman, 1988). However, to date no population-based
studies have been conducted to measure exposure to low-frequency sound,
mainly because taking appropriate measures of low-frequency sound is costly
and technically challenging (see Chapter 3 for more information on the
adequacy of sound measurements).
73
Annoyance is the most prevalent community response to environmental noise

(Basner et al., 2014). The evidence for annoyance from wind turbine noise shows
findings consistent with those from research on other types of environmental
noise. Strong associations between traffic or aircraft noise and annoyance based
on exposure-response relationships have been established in numerous studies
(Babisch et al., 2003; Bakker et al., 2012). Annoyance response to noise from
other sources has also been shown to depend on sound characteristics of the
source in question, including frequency, complexity, and duration of sound,
and its meaning to the subject (Stansfeld, 2003). Different noise sources have
been shown to elicit different levels of community response: at equivalent sound
levels, aircraft noise, for example, has been shown to be more annoying than
road traffic noise, while railway noise was associated with less annoyance than
road traffic (Miedema & Oudshoorn, 2001; Yano et al., 2012). Like annoyance
from wind turbine noise, annoyance from other sources is also influenced by
individual traits, such as noise sensitivity, which can moderate the perception
of annoyance independently of the sound pressure level of noise exposure
(van Kamp et al., 2004).
6.1.1 Summary of the Evidence on Annoyance
The Panel determined that the evidence available is sufficient to establish a
causal relationship between exposure to wind turbine noise and annoyance.
The evidence linking annoyance and wind turbine noise consistently showed an
association between annoyance and wind turbine noise exposure that follows
an exposure-response relationship: higher sound pressure levels lead to an
increasing risk of being annoyed. The findings from several cross-sectional
studies with large sample sizes were consistent across different environments and
exposure conditions and supported by evidence from other types of research
designs (e.g., short-term laboratory studies).
A large body of research has established an exposure-response relationship
between sound levels of various sources of community noise, such as road
and rail traffic, and airports. The evidence for annoyance from wind turbine
noise showed similar types of exposure-response relationships and associations
with these other sources of noise. Furthermore, wind turbine noise was more
annoying than other types of noise at equivalent sound pressure levels. The
specific response pattern appeared to correspond to the acoustic characteristics
of wind turbine noise, including amplitude modulation, as well as to the
particular factors that affected wind turbine noise perception, such as noise
sensitivity, visual impact, and personal attitude.
74
Based on available evidence, the influence of confounding factors can be

ruled out with reasonable certainty. Knowledge gaps remain with regard to the
influence of visual impacts on the perception of sound from wind turbines. The
current state of the evidence does not allow for a definite conclusion about
whether annoyance is caused by exposure to wind turbine noise alone, or
whether factors such as visual impacts and personal attitudes modify the noiseannoyance relation and to what extent, since the studies completed to date
do not measure these factors independently of each other. Furthermore, little
is known so far about the baseline prevalence of annoyance, the magnitude of
the effect, and the thresholds for the perception of wind turbine noise under
different environmental and topographic conditions.
6 . 2
S LE E P D I S TU R BANC E
Sleep disturbance may be defined as any deviation, measurable or

subjectively perceived, from an individuals habitual or desired sleep behavior
(Perron et al., 2012). Indicators of sleep disturbance include the following:
sleep quality, sleep medication taken, total sleep time and time spent in specific
sleep stages, sleep stage changes, arousal, and awakenings (Perron et al.,2012).
As stated by the WHO, sleep is a biological necessity, and disturbed sleep
is associated with a number of adverse impacts on health (WHO, 2009).
Despite the established importance of sleep, sleep disturbance is a common
phenomenon in all populations. In an international survey, only 43% of Canadians
stated that they have a good nights sleep every or almost every night, while
23% said so rarely or never (NSF, 2013).
The auditory system is never completely disengaged. As a result, noise can
have a direct effect on sleep (Griefahn, 2002; Persson Waye et al., 2003;
WHO, 2009). Sleep disruption can also occur in an unconscious state, without
the subject noticing the disturbance (Basner et al., 2011), for example by
activating the autonomic nervous system (Mathias et al., 2008). This is in contrast
to annoyance, which is caused only by consciously perceived noise. However,
most studies on environmental noise and sleep disturbance that were assessed
by the Panel relied on self-reported sleep impacts (namely, awakenings or
daytime sleepiness linked to perceived reduction in sleep quality).
Sleep disturbance can negatively affect physiological functions. The circadian
rhythm a key mechanism that regulates daily rhythms, activities, and rest
cycles can be affected by sleep disturbance, and its disruption by external
influences (e.g., jet lag or shift work) can cause sleep problems (Persson Waye
et al., 2003). Environmental noise can disrupt regular sleep cycles and cause
arousals, which include vegetative arousal and the release of stress hormones
such as cortisol (Maschke & Hecht, 2004). Acute activation, or full awakening,
75
represents a major physiological response, but increased cortisol levels have also
been found in individuals who have been exposed to increased noise during
sleep without conscious awakening (Ising et al., 1999). Similarly, the effects of
noise on sleep can lead to the stimulation of the sympathetic nervous system
(a division of the autonomic nervous system), an unconscious response that
may not be noticed by the person experiencing the disturbance. This can result
in the release of hormones such as adrenaline (epinephrine), noradrenaline
(norepinephrine), and cortisol, and can precipitate further sleep disturbance in
the long run (Maschke & Hecht, 2004). Long-term daytime noise exposure, for
example, leads to an increase in night-time minimum levels of cortisol, which
means a reduction in sleep quality (time spent in deep sleep) and increases
the chance of night-time arousal (Fruhstorfer et al., 1985). The term sleep
disturbance thus describes two different mechanisms: the acute awakening
caused by noise exposure, which is accompanied by night-time release of stress
hormones; and long-term effects of daytime or night-time noise exposure that
change the pattern and intensity of stress hormone release, and thus affect
sleep quality even in the absence of acute disturbance at night.
Two out of the three data sets analyzed in Pedersens (2011) re-analysis (SWE-00
and NL-07) showed a statistically significant association between exposure to
sound from wind turbines and self-reported incidences of sleep interruption.
All three studies also showed a significant association between annoyance
and sleep interruption. Furthermore, sleep interruption was associated more
strongly with annoyance due to noise experienced indoors across the three
studies (Pedersen, 2011). Study SWE-05, in which the turbines were not always
visible as a result of differentiated topography, showed a weaker increase in
annoyance with increasing sound pressure levels outdoors, but a stronger
increase in annoyance with increasing sound pressure levels indoors, as well
as a stronger association between sleep interruption and annoyance indoors
(Pedersen, 2011). These findings suggested that sleep interruption was related
to annoyance rather than exposure to the sound from wind turbines directly.
Thus, sleep disturbance could both be a cause and a consequence of annoyance
related to exposure to sound from wind turbines.
Using data from the Dutch study (NL-07), Bakker et al. (2012) found that
one-half of the respondents reported sleep disturbance at a sound pressure
level of 45 dB(A) and above. Pedersen (2011) and Bakker et al. (2012) both
noted that it was unclear whether sleep disturbance was a direct consequence
of exposure to wind turbine noise, or an indirect consequence of the state of
annoyance caused by exposure to wind turbine noise and other contributing
factors in the vicinity of a subjects dwelling.
76
Nissenbaum et al. (2012) used the Pittsburgh Sleep Quality Index (PSQI)
and the Epworth Sleepiness Score (ESS) to investigate whether there
was an association between the residence distance from wind turbines
(<1.5 km vs. >3-7 km) and sleep quality or daytime sleepiness. While the authors
concluded that the sleep of residents living closer to wind turbines was negatively
affected, other researchers have suggested that the studys limitations, including
possible selection bias, small sample size, and poor exposure data, did not allow
for such a conclusion (Ollson et al., 2013). Bearing these limitations in mind,
the Panel noted that the study did show an association between the exposure
to sound from wind turbines and impacts on sleep in a different region from
the Dutch and Swedish studies, and using a different methodology. The Panel
found one study (unpublished at the time of writing) investigating the impact
of exposure to wind turbine noise on a population in rural Ontario. Based on
sleep diaries and actigraphy-derived measures of sleep quality, the study found
that residents living between 474 metres to 1085 metres from the closest wind
turbine had slightly lower sleep efficiency, longer sleep onset latency, and
longer wake after sleep onset; however, none of these differences was statistically
significant. The study was limited by small sample size (23 participants and
110 person-night observations) and relatively low wind speeds and wind turbine
performance during the observation period (Lane, 2013). The WindVOiCe
self-reported survey reported what the authors described as a moderately
significant association (p-value = 0.0778, which is greater than conventional
statistical significance where p = 0.05) between sleep disturbance and distance
to wind turbines (Krogh et al., 2011).
The Panel found no sleep studies or experiments that measured the effect of
wind turbine noise on sleep physiology using standard methods such as brain
wave measurements (electroencephalography or EEG).
Other types of environmental noise have been linked to impacts on sleep. For
example, sleep disturbance has been shown to be a direct consequence of exposure
to traffic and aircraft noise (hrstrm & Rylander, 1982; hrstrm, 1989; Perron
et al., 2012). In a review of existing evidence, WHO (2009) found that exposure
to sound pressure levels as low as 32 dB(A) had biological effects on sleep quality.
In a review of noise and sleep, Muzet (2007) noted that external sound stimuli
triggered autonomic responses, such as heart rate change and vasoconstriction,
even when the person remained asleep. In a laboratory study, Basner et al., (2011)
(N=72 participants) showed that exposure to rail, road, and air traffic noise
led to statistically significant changes in sleep continuity (measured as a
combination of time to sleep onset, number of arousals, number of awakenings,
and number of sleep state changes). Furthermore, there was a difference in
measured impacts on sleep structure and continuity and subjective evaluations
77
of different types of traffic noise on sleep quality. While road traffic noise
led to the most prominent changes in sleep structure and continuity, air and
rail traffic noise exposure at night were perceived to be more disturbing
(Basner et al., 2011). In research on the effect of the temporal variability of
environmental noise, continuous noise was found to have a statistically smaller
effect on sleep quality than intermittent noise (hrstrm & Rylander, 1982;
Persson Waye et al., 2003). Intermittent noise mainly affects deep sleep, while
continuous noise at exposure levels between 36 and 55 dB(A) has been shown
to decrease the time spent in REM sleep (Eberhardt et al., 1987).
As noted above, there are currently no studies that have investigated the specific
impacts of exposure to low-frequency sound emitted by wind turbines. The Panel
found two field studies in the context of other environmental noise that have
explored whether exposure to low-frequency sound leads to specific impacts on
sleep (Nagai et al., 1989; Persson Waye & Rylander, 2001). Nagai et al. (1989)
described effects on people exposed to low-frequency sound from road traffic
at levels between 72 and 85 dB(A). Subjects were also exposed to shaking and
rattling windows caused by a superhighway. The combination of these disturbances
led to insomnia and excessive tiredness (Nagai et al., 1989). Another study
showed that people who were exposed to sound from heat pump or ventilation
installations in their homes (26-36 dB(A), corresponding to 49-60 dB(C))
were significantly more likely to report annoyance and experience disturbed
concentration and rest than residents exposed to noise from similar sources
but without the low-frequency components (Persson Waye & Rylander, 2001).
6.2.1 Summary of the Evidence on Sleep Disturbance
An association between wind turbine noise and sleep interruption or reduced
sleep quality has been found in some observational studies, but not consistently
across all studies reviewed. A direct association between exposure to wind
turbine noise and sleep disturbance was observed in only two population-based
studies, whereas sleep disturbance was consistently associated with annoyance
due to noise exposure in three studies. Furthermore, the exposure-response
relationship is currently unclear. The Panel therefore concluded that the
current evidence for a causal relationship is limited. The evidence related to
wind turbines is generally in line with findings from studies on the impacts of
other sources of noise on sleep. While sleep disruption has been investigated in
several studies, the resulting evidence base is smaller (fewer and less consistent
studies, fewer participants per study) than for the relationship between wind
turbine noise and annoyance.
78
The main knowledge gap concerns the nature of the mechanism. It is unclear
whether sleep disruption can result directly from exposure to wind turbine
noise, and what proportion of the observed sleep disruption is an indirect
consequence of annoyance. There are currently no experiments or sleep
studies available that demonstrate the impact of sound from wind turbines on
the brain physiology of sleep, as gauged by traditional sleep measures such as
EEG. Further knowledge gaps include the impact of specific characteristics of
sound emitted by wind turbines, such as low-frequency components or periodic
amplitude modulation, on sleep.
6 . 3
S TR E S S
The term stress is broad and generally includes conditions of a physical, biological,
or psychological nature that strain the adaptive capacity of a person up to or
beyond his or her limits (Welford, 1974; Gemmert & Van Galen, 1997). Stress
conditions for humans include factors such as extreme temperatures, heavy
workloads, noise, or social pressures (Broadbent, 1971; Van Gemmert & Van
Galen, 1997). These stressors are typically classified as emotional (caused by
emotions or relating to personality traits), cognitive (related to mental load when
the person is faced with complex tasks), or physical (related to physical loads
originating in the physical environment) (Van Gemmert & Van Galen, 1997).
The same stressor can have multiple effects. For example, noise may interfere
with cognitive activity and trigger an emotional response to the source of noise.
Environmental noise is a common physical stressor for people in urban areas
or in other areas exposed to high levels of community noise (Van Gemmert &
Van Galen, 1997). Noise exposure can lead to stress directly as a physical
effect, or indirectly as a consequence of annoyance. Babisch (2002) provided a
comprehensive review of a direct mechanism for a stress response to noise. Noise
exposure activates the sympathetic nervous and endocrine systems, resulting
in increased levels of stress hormones such as epinephrine, norepinephrine,
and cortisol. These hormones affect metabolism and antibody immunity,
and act as mediators along the pathway from noise to stress-related disease
(Babisch, 2002). Changes in levels of these hormones also result in physiological
effects, including changes in blood pressure or blood clotting factors
(see Babisch et al., 2003 for a summary of research on the general stress model).
Noise can also cause stress indirectly through annoyance. Kalveram et al.
(1999) suggested that noise, in combination with information about the noise
source, leads to annoyance. This can convey a possible loss of fitness signal,
signifying fitness would decline if the individual stays in the same situation
79
(Kalveramet al., 1999). Such a signal can motivate retreat, aggression, stand-by,
or coping behaviour. While stress responses are normal and generally beneficial,
they are thought to become pathological if chronic or frequently repeated,
especially without appropriate resolution. In line with this model, Babisch
(2002) suggested that noise annoyance leads to physiological arousal and stress.
None of the three datasets analyzed by Pedersen (2011) showed a statistically
significant association between exposure to estimated A-weighted sound
pressure levels of wind turbine noise and self-reported perceptions of stress;
however, all three studies reported a positive association between annoyance
from wind turbine noise outdoors and stress (Pedersen, 2011). The study
designs did not allow conclusions about cause and effect: annoyance may be a
consequence of stress, or vice versa. Using data from the Dutch study (NL-07),
Bakker et al. (2012) found that psychological distress was indirectly correlated
with sound exposure in respondents who noticed wind turbine noise, with
noise annoyance acting as a mediator.
In a review article, Baqtasch et al. (2006) noted that stress related to wind
turbines was among the variables related to annoyance. Growing turbine sizes
have also raised concerns that the sound spectrum emitted has shifted toward
lower frequencies (Mller & Pedersen, 2011); however, Shepherd et al. (2011)
also highlighted links among turbine noise, stress, and annoyance, speculating
that these links were exacerbated because chronic noise exposure is a potent
psychosocial stressor, leading to chronic stress (Shepherd et al., 2011).
Pedersen (2011) suggested that cognitive stress theory could explain the
association between annoyance from wind turbine noise and stress, (see also
Lazarus & Folkman, 1984). According to this theory, an individual in an
already stressful situation would evaluate wind turbine noise as an additional
threat to restoration. Since the source of the noise is beyond the individuals
control, the response is manifested as annoyance. It is worth noting, however,
that people who reported being annoyed by wind turbine noise also reported
more symptoms related to stress and resulting secondary effects, such as feeling
less well-rested, and considered their environment as less suited to rest and
restoration (Pedersen et al., 2007a; Pedersen & Persson Waye, 2008).
While the association between annoyance from wind turbine noise and stress has
been found consistently in all studies that investigated stress-related variables,
stress has not been associated directly with exposure to wind turbine noise. The
Panel did not find any studies or experiments that explored whether stress can
result directly from the specific characteristics of wind turbine noise.
80
Indirect stress response via annoyance has also been studied in the context of
other sources of environmental noise, namely traffic noise exposure. Annoyance
and chronic stress have been linked in numerous studies (most often involving
occupational environmental noise or traffic noise); however, a causal relationship
has not been shown. Rylanders (2004) review noted, for instance, that there
are no data available on the relationship between annoyance and cortisol
in saliva (an indicator of stress) under conditions of either acute or chronic
environmental noise exposure.
6.3.1 Summary of the Evidence on Stress
The Panel found no evidence of a direct association between wind turbine
noise and stress. However, several studies showed indirect associations between
annoyance due to wind turbine noise and stress. Stress was associated with
annoyance due to wind turbine noise exposure outdoors in three studies, but
due to wind turbine noise exposure indoors in only one study. Furthermore,
stress has not been the main outcome investigated by existing studies. The
datasets that presented results related to stress (SWE-00, SWE-05, NL-07), and
the resulting studies that investigated the relationship with stress (Pedersen
& Persson Waye, 2008; Pedersen, 2011; Bakker et al., 2012), were limited to
self-reported data, which are treated inconsistently across different studies.
Available studies are of insufficient quality with regard to stress and lack the
statistical power to permit a conclusion regarding the presence or absence of a
causal relationship between wind turbine noise and stress. The Panel therefore
concluded that the current state of the evidence of a causal relationship
is inadequate.
The Panel noted, however, that the patterns of self-reported stress reactions
are generally consistent with associations between stress reactions observed in
a large body of research on the impacts of other environmental noise. These
associations have not yet been tested for wind turbine noise using comparable
methods. It is therefore unknown whether wind turbine noise has effects on
stress comparable to those of other types of noise. Further knowledge gaps
involve the nature of the underlying mechanism, in particular whether stress
can be caused directly through exposure to wind turbine noise or whether it
is an indirect consequence of annoyance, sleep disturbance, or both.
6 . 4
C A R D I O VA S C U L AR DI S E AS E S
Cardiovascular diseases are a group of conditions affecting the heart and blood
vessels (e.g., coronary artery disease). Cardiovascular diseases also result from
multiple causes related to risk factors such as diet, physical inactivity, tobacco
use, the environment, or hereditary preconditions (WHO, 2013). These diseases
have high prevalence in the general population. For example, cardiovascular
81
diseases were responsible for 32% of all premature deaths across Canada
in 2004, and were the second largest contributor to health care costs in 2000
(PHAC, 2009). The complex relationship between multiple risk factors and
effects makes it difficult to associate the risk of cardiovascular disease with a
single cause such as wind turbine noise. No high-quality studies designed to
examine cardiovascular disease outcomes and their relationship to exposure
to wind turbine noise have been conducted to date.
It is known that noise can act on individuals as a physiological stressor, inducing
a vegetative response (such as blood pressure elevations), causing somatic (such
as hypertension) and psychosomatic responses, which in turn, may affect the
risk of disease (van Kempen & Babisch, 2012). However, evidence from studies
of road traffic noise showed that there appeared to be no increased risk of
cardiovascular disease resulting from exposure to sound pressure levels below
60 dB(A) during the day (Babisch, 2008) or 55 dB(A) at night (Babisch, 2011).
Research on other types of environmental noise has shown that cardiovascular
disease might also be indirectly linked to noise exposure via chronic annoyance
and stress. For instance, chronic annoyance has been used as an indicator
of increased risk for persistent imbalance in the physiological stress system
(Barregard et al., 2009; Pedersen, 2011). This imbalance can lead to high blood
pressure and other manifestations of cardiovascular disease if the exposure is
prolonged (Barregard et al., 2009). The pathway through which noise affects
blood pressure is not yet fully understood. Stansfeld (2003) proposed that
the effect could be mediated through an intermediate response such as noise
annoyance; however, current evidence is insufficient to support this pathway.
Willich et al. (2006) found an association between chronic noise burden and
the risk of myocardial infarction. They noted that this finding was consistent
with a causal mechanism based on a conceptual stress model in which sound
pressure levels and/or annoyance by noise may enhance psychological stress.
Suls (2013) reviewed evidence for an association between annoyance and anger
or hostility, which are known risk factors for stress.
Furthermore, there is evidence that sleep disturbance caused by traffic noise
can lead to cardiovascular effects. For example, a laboratory experiment
(number of participants =38) simulating freight train noise exposure (at
40 dB(A) and 50 dB(A)) showed that railway noise at night had an effect on the
cardiovascular system of sleeping subjects (Tassi et al., 2010; Croy et al., 2013).
Pedersen (2011) found no statistically significant association between wind
turbine noise exposure or annoyance and either self-reported high blood
pressure or self-reported cardiovascular disease in any of the three studies that
82
she re-analyzed. Using the two Swedish studies (SWE-00, SWE-05), Pedersen
and Persson Waye (2008) examined the relationship between noise annoyance
from wind turbines and cardiovascular disease. Their findings suggested no
differences in self-reported rates of cardiovascular diseases between respondents
reporting that they were annoyed by wind turbine noise and other respondents
(Pedersen & Persson Waye, 2008). The sample sizes of the studies analyzed
in Pedersen (2011) and Pedersen and Waye (2008) are relatively small, given
the need to examine what would be expected to be minor increases in relative
risk, and potentially large misclassification errors and other sources of bias.
The cardiovascular effects of various sources of noise, such as road traffic
noise, community noise, and occupational noise, have been extensively
researched (Babisch, 2008; van Kempen & Babisch, 2012). A meta-analysis of
43 epidemiologic studies published between 1970 and 1999 investigating the
relationship between noise exposure and blood pressure or ischemic heart
disease found small increases in blood pressure as a result of occupational
noise exposure (van Kempen et al., 2002). The review also showed a significant
association between exposure to either occupational or air traffic noise at
45 to 75 dB(A) (LAeq,16hr) and hypertension (van Kempen et al., 2002). Another
meta-analysis of effects of road traffic noise on annoyance and health found
a positive and significant association between noise annoyance and the risk
of arterial hypertension (Ndrepepa & Twardella, 2011).
Studies by Bakker (2012) and Shepherd et al. (2011) suggested that annoyance
acts as a mediator or pathway in a causal chain between exposure to wind turbine
noise and impacts on health or quality of life. This is supported by the finding
that annoyance was a significant effect modifier with respect to the association
between aircraft noise and hypertension (Babisch et al., 2003). Sound pressure
level and noise annoyance were found to be equally good predictors of the impact
of aircraft noise on health. Another related study suggested that sound pressure
level and noise annoyance may both serve as explanatory variables for the link
between chronic noise exposure and cardiovascular disease, although sound
pressure level is often a stronger predictor (Babisch et al., 2013). Annoyance
and disturbance due to road traffic noise have also been associated with a
higher incidence of ischemic heart disease; however, when study subjects had
conditions predisposing them to cardiovascular disease, noise exposure did
not change the risk of ischemic heart disease (Babisch et al., 2003).
Chronic environmental noise exposure in general has been associated with an
increased risk of myocardial infarction, consistent with the hypothesis that there
is an association between long-term noise exposure and risk of cardiovascular
83
disease. Willich et al. (2006) suggested that chronic noise exposure is

the equivalent of an exogenous risk factor contributing to the development
of cardiovascular disease.
Although low-frequency sound is a common component of wind turbine noise,
little is known about its specific effects on cardiovascular disease, especially
in comparison with other types of noise exposure with complex and difficulttomeasure characteristics, such as impulse noise (e.g., explosions and gunshots)
(Berglund et al., 1996). Schust (2004) reported a study by Danielsson and
Landstrom (1985) that showed the effects of infrasound ranging in frequency
(6, 12, or 16 Hz) and sound pressure level (95, 110, or 125 dB(Z)) on blood
pressure, pulse rate, and serum cortisol levels. This study showed an induced
peripheral vasoconstriction with increased blood pressure and concluded
that environmental infrasound may be of importance for the development of
essential hypertension in predisposed individuals (Danielsson & Landstrom,
1985). However, exposure levels reported were much higher than those
observed in the context of wind turbine noise. A Portuguese research group
has argued that infrasound and low-frequency noise from wind turbines may
cause vibroacoustic disease, a hypothesized syndrome including cardiovascular
effects such as increased risk of coronary artery surgery, which may be associated
with long-term exposure to sound with high sound pressure levels and low-frequency
components (Alves-Pereira & Branco, 2007). To date, independent research
has failed to support the existence of vibroacoustic disease (Ksin et al., 2012).
Furthermore, there is no evidence that wind turbines emit infrasound or
low-frequency sound at pressure levels comparable to those studied in the
context of vibroacoustic disease (Bolin et al., 2011).
6.4.1 Summary of the Evidence on Cardiovascular Disease
The Panel found no evidence suggesting a direct association between exposure
to wind turbine noise and cardiovascular disease. Several of the studies reviewed
used case definitions of specific cardiovascular diseases in self-reported data,
but none of these revealed a statistically significant association. Therefore, the
Panel concluded that available evidence is inadequate to permit a conclusion
regarding the presence or absence of a causal relationship. Prior research
on other sources of environmental noise has repeatedly found associations
between noise exposure and cardiovascular effects, but at exposure levels that
are much higher than those encountered in the context of wind turbine noise.
6 .5
D I A BE T E S
Diabetes is a chronic disease in which the body either cannot produce

(type 1 diabetes) or properly use (type 2 diabetes) insulin, a hormone that
controls the amount of glucose in the blood. Type 1 diabetes develops in
84
early childhood or adolescence and is generally thought to be the result of a

hereditary genetic predisposition. Type 2 diabetes develops mainly in adults and
is associated with a number of risk factors, including stress, sleep disturbances,
high body weight and physical inactivity (Srensen et al., 2012; WHO, 2014).
Researchers have investigated whether exposure to traffic noise is associated
with a higher risk of incidence of type 2 diabetes. Similar to research on
cardiovascular disease, research on diabetes is challenged by the multitude of
risk factors associated with this disease. Furthermore, diabetes develops only
after long-term exposure to those risk factors.
None of the studies reviewed by the Panel focused on diabetes mellitus;
however, the three studies analyzed in Pedersen (2011) did record self-reported
cases of diabetes among the study populations. One study (SWE-05) showed
a statistically significant association between A-weighted sound pressure levels
at participants homes and cases of diabetes. The study did not record the
duration of exposure, however, nor was it designed to control for confounding
factors specific to diabetes, such as the rate of pre-existing or type 1 diabetes,
diet, physical activity, current and previous smoking, work strain, or general
health status.
Two recent studies have investigated whether road traffic noise or long-term
exposure to aircraft noise affect the incidence of type 2 diabetes in exposed
residents. Srensen et al. (2012) used data from a cohort study on road traffic
noise exposure for 57,053 subjects who were followed for an average of
9.6 years. During the follow-up period, 2,752 cases of diabetes were recorded.
The study showed that a higher average exposure of 10 dB at the time of
diagnosis was associated with a higher risk of diabetes with an incidence rate
ratio (IRR) of 1.117 (95% confidence interval (CI) 1.03 to 1.19), whereas
exposure in the five years preceding diagnosis was associated with an
IRR of 1.14 (95% CI 1.06 to 1.22) after adjusting for confounding factors.
Eriksson et al. (2014) investigated the effects of long-term aircraft noise on body
mass index (BMI), waist circumference, and type 2 diabetes using data from
a cohort study with 5,156 participants. The results did not show a significant
association between noise exposure (based on an ordinal-scale noise variable)
and either BMI or type 2 diabetes. A 5 dB increase in aircraft noise level was
associated with a greater-than-average increase in waist circumference of 1.51 cm
(95% CI 1.13 to 1.89).
7 This means that for every 100 cases that would be expected in the general (unexposed)
population based on the average rate of type 2 diabetes cases, there were 111 actual cases
found in the exposed group.
85
Both studies used exposure assessment scales that considered residents to

be unexposed if Lden was <50 dB(A) for road traffic noise and <42 dB(A)
for aircraft noise. The highest associations occurred at sound level categories
>60 dB (road traffic) and >55 dB (aircraft). These exposure levels are generally
higher than those that have been measured for the subjects considered in
studies of wind turbines (see Chapter 3).
6.5.1 Summary of the Evidence on Diabetes
The Panel found only one study that showed an association between wind
turbine noise and cases of diabetes; however, the study design did not identify
the timeframe over which these cases developed or whether other factors may
have confounded the association. Therefore, the Panel concluded that available
evidence is inadequate to permit a conclusion about the presence or absence of
a causal relationship. Studies have found that diabetes may be related to other
sources of environmental noise; however, knowledge gaps persist regarding the
exposure levels and duration of exposure that lead to increased incidence of
type 2 diabetes. With regard to wind turbine noise, there is a general knowledge
gap concerning the effects of long-term exposure. This gap is partly due to the
fact that wind turbine noise is a fairly recent phenomenon.
6 .6
E FFE C T S ON H E AR I NG
Noise-induced hearing loss is a partial or total inability to hear caused by

sensorineural damage affecting the inner ear (e.g., hair cells) or the auditory
nerve, leading to higher thresholds for detecting sounds or permanent
and irreversible loss of hearing. Exposure to noise at high sound pressure
levels in the short term can result in reversible temporary threshold shift
(Nelson et al., 2005). However, permanent noise-induced hearing loss is
common, for example, in those who have long-term occupational exposures
to broadband noise with a sound pressure level over 80 to 85 dB(A) that leads
to damage of outer hair cells.
No studies or experiments have been conducted to date to specifically investigate
whether exposure to wind turbine noise can lead to hearing loss. Noise-induced
hearing loss generally only occurs at exposure levels above those observed in
the proximity of wind turbines. Exposure to environmental noise at levels of
75 dB(A) or lower is not expected to lead to hearing loss, even after a lifetime
of exposure (WHO, 2011).
86
6.6.1 Summary of the Evidence on Hearing Loss

Previous research on noise-induced hearing loss provides evidence of no causal
relationship between exposure to broadband noise at sound pressure levels
associated with wind turbines at any distance and hearing loss. Previous studies
of other sources of noise are mutually consistent in failing to show a positive
association between human exposure to sound at pressure levels associated
with wind turbines and any symptom of hearing loss.
6 . 7
TI N N I TUS
Tinnitus is the general term for perceived sound perception (for instance,
roaring, hissing, or ringing) that cannot be attributed to an external source
(WHO, 2011). Previous research has identified a wide range of possible causes
of tinnitus, including noise-induced hearing loss, neurological disorders, stress,
metabolic disorders, and psychiatric disorders. The specific pathways that lead
to tinnitus are not yet fully understood, despite relatively high prevalence in
many populations (Henry et al., 2005).
One of the three studies analyzed in Pedersen (2011) showed a statistically
significant association between exposure to wind turbine noise and self-reported
cases of tinnitus (SWE-00). No association was found with annoyance due to
wind turbine exposure outdoors or indoors. In research on other sources from
environmental noise, approximately 10% of patients with tinnitus recorded in the
Oregon Tinnitus Clinic Database self-reported long-duration noise as the cause
of their tinnitus (Henry et al., 2005). The WHO (2011) suggests that, globally,
3% of the population could suffer from tinnitus caused by environmental noise
exposure. Tinnitus is also associated with hearing loss. In one study conducted in
Britain, 16.1% of men who reported severe difficulties in hearing also reported
having tinnitus, compared with 5% in those with slight or no difficulties in
hearing. The respective numbers among women were 33.1% and 2.6% (Palmer
et al., 2002). In a literature review, Henry et al. (2005) showed that the majority
of tinnitus patients are also affected by some degree of hearing loss, with rates
ranging from 50 to 80% depending on study design and age of the subjects
studied. On the other hand, tinnitus is also common among people who do
not have any measurable hearing loss (Eggermont, 2005).
Most studies of noise-induced tinnitus have focused on impulse noise
(e.g., explosions or gunshots) or short-term exposure to intense sound
(e.g., loud music, occupational noise), which are at much higher sound pressure
levels than wind turbine noise. No studies could be found of associations
between tinnitus and low-frequency noise or infrasound.
87
6.7.1 Summary of the Evidence on Tinnitus

The Panel concluded that the state of evidence is inadequate to determine the
presence or absence of a causal link between wind turbine noise and tinnitus.
The Panel noted considerable uncertainty over the general causes of tinnitus
and knowledge gaps with regard to associations between long-term exposure
to environmental noise and tinnitus.
6 .8
E FFE C T S OF NON-AU DI B L E S O U ND ON TH E INNER EAR
Several reports have suggested that symptoms, such as vertigo, nausea, ear
pressure, or vision problems, could be caused by non-audible sound emitted
by wind turbines, such as infrasound and low-frequency sound below a persons
hearing threshold (Pierpont, 2006; 2009; Jeffery et al., 2013; Ambrose et al., 2012).
The Panel found no evidence from epidemiological studies or experiments that
specifically investigated impacts of non-audible sound emitted by wind turbines
on the ear. However, research has explored possible pathways through which
infrasound or low-frequency sound might affect the ear, including damage to
the inner ear, stimulation of the cochlea, or stimulation of the vestibular system.
Stimulation of the Cochlea
The hair cells of the cochlea are the main organ of human sound perception.
Sound waves stimulate the inner hair cells, which transform the sound into
a neural signal that is perceived by the brain as sound. The inner hair cells
are known to be less sensitive to infrasound ( 20 Hz) or low-frequencies, and
rather high sound pressure levels are needed to provoke a sensation especially
for the infrasound range (see Chapter 4); however, it is unclear whether there
are other pathways of stimulation.
Using cochlear monitoring, Hensel et al. (2007) showed that cochlear processing
was altered after exposure to infrasound of 6 Hz at a sound pressure level of
130 dB. Based on this and other research, Salt and Hullar (2010) suggested
that low-frequency sound can stimulate the outer hair cells at sound pressure
levels 40 dB below the threshold of hearing, when the inner hair cells are not
stimulated. The study showed that cochlear potentials generated by hair cell
transduction in guinea pigs can be recorded in response to 5 Hz infrasound,
although these potentials are not perceived by the brain as sound because they
are transmitted via a different neural pathway. In a subsequent animal study,
Salt et al. (2013) showed that infrasound at 5 Hz can produce an electrical
response in the cochlea larger than that produced by tones at other frequencies
between 50 and 1,000 Hz at levels as low as 60 to 65 dB. Based on these results,
the authors argued that the apical region (tip) of outer hair cells may be more
88
sensitive than previously estimated and could provide a mechanism through

which infrasound stimulates the brain that does not involve hearing. They
further suggested that some clinical conditions, such as Mnires disease,
superior canal dehiscence, or asymptomatic cases of endolymphatic hydrops
could further increase this sensitivity (Salt & Hullar, 2010). This would imply
that some individuals may be hypersensitive to infrasound. In an earlier paper,
Salt and Kaltenbach (2011) also speculated that infrasound could lead to
perceptions of fullness, ear pressure, or tinnitus. This work suggests that there
may be a response of the cochlea to suprathreshold levels of infrasound, but
the effect on the brain remains unknown (see Ellenbogen et al., 2012).
Stimulation of the Vestibular System
The vestibular system is the apparatus that helps a person maintain balance,
spatial orientation, and visual fixation (see Chapter 4). No epidemiological
studies or experiments that investigated the activation of the human vestibular
system through air-conducted non-audible sound were found. The evidence
is currently limited to studies of human subjects using acoustic stimulation
through bone conduction and of animal subjects using exposure to air-conducted
low-frequency sound.
In an experiment using mice, Tamura et al. (2012) showed that low-frequency
sound of 100 Hz at 70 dB caused impaired balance due to partial loss of hair
cells and increased levels of oxidative stress in the vestibule. The authors noted
that it is unclear how this level of sound exposure can directly cause loss of hair
cells, but that the finding indicates that further study is required including
an extrapolation to humans. In human studies, Todd et al. (2008) showed that
acoustic signals transmitted by bone conduction can activate the otolith organs
of the inner ear. The threshold for triggering the vestibulo-ocular reflex, a reflex
in which the head movement is compensated by a change of position of the eyes,
was lowest at a frequency of 100 Hz. Acoustic stimulation is a well-established
technique for testing vestibular system function (vestibular evoked myogenic
potential (VEMP)) (e.g., Colebatch & Halmagyi, 1992; Colebatchet al., 1994;
Robertson & Ireland, 1995; Curthoys et al., 2006; Zhang et al., 2012).
These studies suggest that air-conducted sound at sound pressure levels below
100 dB are unlikely to stimulate the otoliths (the most sensitive part of the
vestibular end organs) in a healthy person. There is evidence, however, that
in some pathological conditions acoustic signals could activate the vestibular
system at lower levels. Tullio syndrome (Tullio, 1929), characterized by vertigo
or an abnormal vestibular ocular reflex, results from changes in the fluid
pathways between the cochlear and vestibular parts of the inner ear caused
by various anatomical abnormalities (e.g., enlarged vestibular aqueduct or
89
perilymphatic fistula). Superior (semicircular) canal dehiscence syndrome (SCDS)

(Minor et al., 1998; Minor, 2000) is an observable fenestration (opening) of a
semicircular canal, which is correlated with sound-induced vestibular activity
(see Chapter 4). Under these pathologic conditions, acoustic signals can be
more effective in activating the vestibular system. A number of studies have
used VEMP testing to compare the sensitivity of the vestibular system to sound
in patients with various degrees of canal dehiscence (Pfammatter et al., 2010)
or in patients before and after surgery to plug superior canal dehiscence
(Welgampola et al., 2008). These papers noted a 20 to 30 dB reduction in
the sound stimulation threshold for patients affected by SCDS. Russo (2014)
estimated that SCDS may affect up to 5% of the population. Other studies
found that 1.2% of normal adults have definite or possible dehiscence (Erdogan
et al., 2011). Hagiwara et al. (2012) reported 3% prevalence in adults and 27%
prevalence in children less than two years of age.
A relatively large number of adults, and many more infants, are more sensitive
to acoustic activation of the vestibular system than the general population
(Hagiwara et al., 2012). But even these individuals are unlikely, in the Panels
opinion, to experience stimulation of the vestibular system by the acoustic
signals from wind turbines. The normal threshold for sound activation of the
vestibular system, as judged from VEMP testing, is a sound pressure level of
around 110 dB. Subjects with a pre-existing condition such as canal dehiscence
may be sensitive at 85 to 90 dB. However, at the standard 4050 dB(A) noise
level regulations commonly in place for wind turbine exposure, the broadband
components of wind turbine noise do not approach levels high enough to
activate the vestibular system (Harrison, 2014). If the reduced emphasis of
low-frequency components by the use of A-weighted sound levels is taken into
account, and signal levels of low frequencies are estimated based on sound
spectrum data, peaks in low-frequency levels rarely exceed 70 dB. Therefore,
based on the existing evidence, the Panel could not make conclusions about
whether vestibular symptoms such as vertigo, dizziness, or nausea are the direct
result of exposure to wind turbine noise (Harrison, 2014).
6.8.1 Summary of the Evidence on Effects of Non-Audible Sound
Based on the evidence reviewed above, the Panel found:
Infrasound (5 Hz) can stimulate the outer hair cells of the cochlea at moderate
sound pressure levels (>60 dB) in certain animal models; however, the effect
on the brain remains unknown, and this research has not been extrapolated
to humans.
Infrasound can stimulate the human vestibular system at high sound pressure
levels (>110 dB for a person with normal sensitivity and possibly >85 dB for
a person with higher sensitivity).
90
The thresholds for infrasound or low-frequency sound associated with damage

to the inner ear and vestibular activation are far above the levels of exposure
associated with wind turbines. The threshold for cochlear activation in animal
models could be achieved by sound from wind turbines; however, since neither
the threshold in humans nor the impact on the brain are known, it is not
possible to associate health impacts with cochlear activation through infrasound.
It is plausible that certain pre-existing clinical conditions could decrease the
threshold of stimulation of the cochlear or the vestibular system and thus
lead to higher sensitivities to non-audible sound in affected individuals,
in particular young children; however, no specific thresholds have been
established to date. It thus remains unclear whether wind turbine noise
could exceed such thresholds.
In view of these findings, the Panel concluded that the current evidence is
inadequate to determine the presence or absence of causal links between
exposure to non-audible sound emitted by wind turbines and symptoms such
as vertigo, nausea, ear pressure, or vision problems. Knowledge gaps include
measurement and understanding of low-frequency sound from wind turbines,
thresholds for stimulation or damage of the cochlear and vestibular systems in
humans, and lower thresholds in sensitive populations.
6 . 9
C O G N I TI VE AND M E NTAL P E RF OR M ANC E
Cognitive abilities develop early in life (Diamond, 2002). Stansfeld et al. (2005)
hypothesized that environmental noise could impact the mental performance of
young children. As well, ambient environmental noise could distort awareness
of speech sounds (Bryant & Bradley, 1985). Several early studies conducted
by Hockey and colleagues identified other potential pathways through which
noise exposure may influence cognitive performance in adults (Hockey, 1984;
Robert & Hockey, 1997). These studies suggested that noise may affect the
performance of specific tasks and that, while noise increases levels of alertness,
it does not increase mental performance speed, but rather reduces cognitive
performance accuracy and short-term memory performance.
The Panel found no epidemiological studies of impacts of wind turbine noise
on cognitive performance. Evidence is currently limited to two laboratory
studies. Ruotolo et al. (2012) showed that executive control and semantic
memory deteriorate with auditory and visual stimuli that simulate increasing
proximity to a wind energy facility. Visual features appeared to amplify the
negative impact on executive control. No impact was found on short-term
verbal memory. These results suggested that wind turbines may have selective
effects on cognitive performance, with stronger impacts on tasks that demand
high levels of executive control.
91
Alimohammadi et al. (2013) conducted an experiment to test whether

low-frequency noise from different sources, including wind turbines, causes
annoyance or has an impact on mental performance, and whether these impacts
depend on personality type. Participants exposed to low-frequency noise had
increased mental performance both with respect to speed and accuracy in the
completion of tests designed to evaluate executive control. The results also
revealed that introverted participants were significantly more annoyed and
performed tests with less accuracy than extroverted participants. These results
contradict research using higher-frequency sound from ventilation noise,
which suggested that noise at levels above 51 dB(A) disrupted performance
(Kjiellberg & Wide, 1988). But the literature on performance is very complex
and the divergence in results indicates that the impact of noise on cognitive
performance likely depends on the type of task subjects have to perform.
Past studies of environmental noise have shown that it can have an impact on
cognitive performance and learning, particularly in children (Clark&Srqvist,2012;
Klatte et al., 2013). For example, Hughes and Jones(2001) presented a review of
laboratory studies showing that sound affects cognitive processing and disrupts
performance. Several studies have shown that children exposed to noise at school
experience cognitive impairments (Clark&Stansfeld,2007). Tasks that were
impaired involve central processing and language, namely, reading comprehension,
memory, and attention (Evans&Maxwell, 1997; Haineset al.,2001). A set of
key studies examined the impacts of the relocation of an airport in Munich,
Germany, on childrens cognition (Evans et al., 1995,1998). High noise exposure
was associated with deficits in memory and reading comprehension, but after
the noise exposure ceased, the deficits disappeared, indicating that effects may
be reversible.
A large-scale study in the Netherlands, Spain, and the United Kingdom conducted
by Stansfeld et al. (2005) found a linear exposure-effect association between
chronic aircraft noise exposure and impairment of reading comprehension and
recognition memory; the researchers highlighted aircraft noise as an auditory
stressor that is detrimental to a healthy educational environment.
6.9.1 Summary of the Evidence on Cognitive and Mental Performance
To date no population-based studies and very limited experimental evidence are
available on the specific impact of wind turbine noise on cognitive performance.
Research on other sources of environmental noise has shown a possible link;
however, the results are inconclusive, showing both beneficial and adverse
effects on cognitive and mental performance. Therefore, the Panel found that
the evidence is inadequate to determine the presence or absence of a causal
relationship between wind turbine noise and cognitive and mental performance.
92
Knowledge gaps include clear-case definitions and measurements of mental

performance under noise exposure; a lack of understanding of how noise
characteristics may stimulate or adversely affect cognitive performance; and
how these impacts are influenced by characteristics of the subjects exposed
and the type of cognitive task performed.
6 . 10 P S YC HO LO G I CAL H E ALT H
Mental health refers to a state of emotional and psychological well-being

(Van Kamp & Davies, 2008). In keeping with the definition of health used
in this report, mental health means the absence of mental illness as well as
unpleasant feelings or emotions (psychological distress) that may affect a
persons level of functioning. Environmental noise exposure has been linked
to effects on mental health under a variety of circumstances. However, in most
cases environmental noise leads to psychological distress rather than produce
serious mental illness (Van Kamp & Davies, 2008).
With regard to wind turbine noise, one study by Bakker et al. (2012) found a
positive association between exposure and psychological distress, with annoyance
appearing to act as an intermediary variable. The study further found that
psychological distress is linked to sleep disturbance, regardless of exposure to
sound from wind turbines.
Whether exposure to environmental noise can directly lead to mental illness is
still unclear. In a literature review, Stansfeld et al. (2000) found that exposure
to environmental noise did not appear to lead to diagnosed psychological
disorders, but was sometimes associated with mental health symptoms such
as depression or anxiety. The authors further noted that self-reported noise
sensitivity did not appear to interact with noise exposure in leading to mental
illness. Noise sensitivity was, however, associated with certain mental health
symptoms. In the same vein, Stansfeld et al. (2009) noted that transport-related
noise was unlikely to cause serious mental illness but may be responsible for
psychological symptoms. Hardoy et al. (2005) found a relationship between
exposure to aircraft noise and generalized anxiety disorder or anxiety disorder
not otherwise specified. In a study of road traffic noise, Sygna et al. (2014) found
a positive association between noise exposure and psychological distress among
subjects with poor sleep quality, but the association was not statistically significant.
6.10.1 Summary of the Evidence on Psychological Health
The current state of the evidence on noise exposure and mental health does not
allow conclusions as to whether noise causes mental illness or whether it only
contributes to psychological distress or exacerbates symptoms in people who already
have a mental illness. Evidence reviewed by the Panel does not permit conclusions
93
about whether there is a specific causal relationship linking psychological distress to

exposure to wind turbine noise. The Panel found, therefore, that current evidence
is inadequate to determine the presence or absence of a causal relationship. The
Panel further noted that, given the inconclusive nature of the impact of noise on
general mental health, knowledge gaps in this area are not specific to sound from
wind turbines. A better understanding of the effect of general noise exposure on
psychological disorders is needed, and the effects of specific characteristics of
sound from wind turbines remain even more uncertain.
6 .1 1 Q UA L I T Y OF L I F E
Two studies have investigated the impact of wind turbine noise on health-related
quality of life. Shepherd et al. (2011) conducted a cross-sectional study with a
control group in New Zealand using the short version of the WHOs quality
of life survey. The results supported findings from cross-sectional studies of an
exposure-response relationship between wind turbine noise and annoyance,
and showed that several domains of quality of life were negatively affected for
respondents living close to wind turbines (perceived sleep quality, energy levels,
environmental quality of life, and amenity). A two-year follow-up study by McBride
et al. (2013) showed that the difference in quality of life between the two groups
remained stable over time. Mroczek et al. (2012) conducted a randomized
cross-sectional study using the SF-36 General Health Questionnaire. The results of
this study did not show a significant difference in health-related aspects of quality
of life between residents living closer to wind turbines and those living further
away. The authors noted, however, that the study design did not include a number
of factors that influence quality of life, such as economic opportunity provided
by wind turbines and socioeconomic differences between areas. Such omissions
may lead to observed differences that are not related to wind turbine noise. The
WindVOiCe self-reported survey did not find that quality of life was significantly
altered for people living closer to wind turbines (Krogh et al., 2011).
6.11.1 Summary of the Evidence on Quality of Life
To date, very limited evidence is available on the specific impact of wind turbine
noise on quality of life. Moreover, the different studies available reach different
conclusions. The Panel found, therefore, that current evidence is inadequate
to determine the presence or absence of a causal link.
6 .1 2 O THER H E ALT H I M PAC T S
The previous sections in this chapter have discussed health outcomes that have been
the subject of population-based studies or laboratory experiments. Anumber of
other health impacts have been frequently attributed to exposure to wind turbine
noise in case series, informal surveys, complaints, the media, and other sources.
94
In research on other environmental noise, some of these health outcomes were

indirectly associated with noise via annoyance, but not directly associated with
exposure to wind turbine noise. These outcomes included diabetes (see Section
6.5), chronic disease (unspecified), undue tiredness/fatigue, and headache.
Fatigue and headache are very common, unspecific health outcomes that could
be secondary impacts of annoyance, sleep disruption, or stress. They can also
be a consequence of many other factors and influences that often cooccur with
exposure to wind turbine noise and are therefore difficult to isolate.
For all other health outcomes considered (see Table 5.1), the Panel did not
find any research involving population-based studies or experiments that could
provide evidence of the presence or absence of causal links. In these cases,
the Panel therefore considered the use of the IARC framework imperfect, as
these specific health effects have not yet been the subject of primary empirical
research. The Panel noted, however, that many of these health effects could be
secondary consequences of other effects that have been addressed in primary
empirical research related to wind turbines.
6 . 13 C HA P TE R S U M M ARY
The Panel reviewed the evidence selected according to the process outlined in
Chapter 5, and reported on all health outcomes for which at least one populationbased study found a statistically significant association or for which at least one
laboratory experiment has been published. The main health outcomes that
emerged were annoyance, sleep disturbance, and stress:
The Panel concluded that the current evidence is sufficient to establish a
causal relationship between exposure to wind turbine noise and annoyance.
The Panel concluded that current evidence for a causal relationship is limited
to establish a causal relationship between exposure to wind turbine noise
and sleep disturbance.
The current evidence of a causal relationship is inadequate to reach a conclusion
concerning the presence or absence of a relationship between exposure to wind
turbine noise and stress, although the effect may be indirect, via annoyance.
The Panel concluded that there is evidence of no causal relationship between
hearing loss and exposure to broadband noise at sound pressure levels
associated with wind turbines, at any distance.
The Panel concluded that the current evidence is inadequate to determine the
presence or absence of causal links between exposure to wind turbine noise
and the other health outcomes listed in Table 5.1. A more detailed description
summary is given in Chapter 7.
95
Chapter 7 Overview of the Evidence, Knowledge Gaps, and Research Needs
7
Overview of the Evidence, Knowledge
Gaps, and Research Needs

Summary of the Evidence
Quality of the Evidence Reviewed
Further Research
Chapter Summary
96
Overview of the Evidence, Knowledge Gaps,

and Research Needs
Key Findings
Much of the evidence reviewed by the Panel suffered from methodological limitations:
inadequate control for selection bias and confounding factors; small sample size
that limited statistical power; lack of longitudinal studies; and lack of measurement
to assess exposure.
The Panel identified specific gaps in knowledge for each health condition studied,
such as the visual impact of wind turbines on annoyance or the possible pathways
leading to sleep disturbance or stress.
There is a lack of longitudinal studies as well as a paucity of research on sensitive
populations.
In measurement of wind turbine sound, long-term equivalent A-weighted levels are
most often used, but this measure fails to capture wind turbine sound characteristics
such as amplitude modulation and low frequencies.
This chapter reviews the main findings related to wind turbine noise and health
outcomes and describes the knowledge gaps and research needs for relevant
adverse health effects. The Panels findings with regard to each adverse health
effect are summarized in Tables 7.1 and 7.2.
7 . 1
S UM M A RY OF T H E E V I DE NCE
More than one pathway or mechanism might plausibly link exposure to

wind turbine noise and the adverse health effects examined in this report.
Figure 7.1 summarizes the most probable causal pathways linking exposure to
sound from wind turbines to adverse health effects. The evidence is sufficient
to establish a causal relationship between exposure to wind turbine noise and
annoyance. However, knowledge gaps remain on the question of whether
factors such as visual impacts and personal attitudes modify the noise-annoyance
relation. Long-term annoyance resulting from other types of noise, such as
road traffic noise, has been shown to be a potential precursor or contributing
factor to other adverse health effects, including stress, sleep disturbance,
and cardiovascular diseases, but the exact mechanisms are incompletely
understood (Babisch, 2002, 2008; Persson Waye et al., 2003; Barregard et al.,2009;
vanKempen & Babisch, 2012).
97
In addition, the Panel found no evidence that wind turbines routinely produce
infrasound at levels significantly higher than other environmental sources,
such as the wind itself, or at levels associated with the known health effects
of infrasound. However, in some cases and in certain types of dwellings, lowfrequency sound may be more pronounced indoors because low frequencies
are less attenuated by walls and windows and because of structure resonance.
The Panel also found that exposure to wind turbine noise is unlikely to cause
hearing loss.
Factors Influencing Responses
External Stimuli
Frequency Characteristics
Tonal Components
Modifying Factors
Noise Sensitivity
Pre-Existing Conditions
Personal Attitudes
Visual Impacts
Mechanisms
Direct
ReceptorMediated
(hearing,
vestibular)
Health Outcomes
Nervous System
Cognitive
Performance
Tinnitus
Health-Related
Quality of Life
Sleep
Disturbance
Endocrine System
Diabetes
Stress
Perceptual
Cognitive and
Emotional
Response
Sufficient evidence of a causal relationship

Limited evidence of a causal relationship
Evidence from the environmental noise literature
Annoyance
Psychological
Health
Cardiovascular
System
Grey boxes: inadequate evidence of a causal

relationship for the specific health outcome
Unlinked boxes: no clear mechanism linking
the health outcome to wind turbine noise
Figure 7.1
Summary of Evidence for Causal Pathways Between Exposure to Wind Turbine Noise
and Adverse Health Effects
As discussed in Chapter 6, there is sufficient evidence to support a causal relationship for annoyance
only. The current evidence is limited with regard to the relationship between exposure to wind turbine
noise and sleep disturbance. The available evidence is inadequate with regard to stress; however,
general evidence suggests that stress may be aggravated by annoyance and sleep disturbance. Research
on the health effects of noise in general also suggests a strong relationship between long-term
stress and cardiovascular or other diseases. Nevertheless, the causal mechanisms remain uncertain,
particularly for wind turbine noise. Multiple pathways may be responsible for individual outcomes and
population-level rates of adverse health effects. The absence of a causal relationship between wind
turbine noise and hearing loss is not represented in this figure. See Table 7.1 for more information.
98
Knowledge gaps for the various conditions or symptoms discussed in Chapter 6

are listed in Table 7.1. Health effects that are not addressed in epidemiological
studies or experiments specific to wind turbines and/or for which no plausible
mechanisms can be suggested are listed in Table 7.2. Advancing understanding
about whether or not these health effects are linked to noise exposure in general
will also help determine if these are a concern in the context of wind turbine
noise and if additional, more specific research is needed.
Table 7.1
Overview of Findings with Regard to Adverse Health Effects Addressed in Empirical
Population-Based Research on Exposure to Wind Turbine Noise
Condition or
Symptom
Level of
Evidence
(IARC)
Possible Pathways
Knowledge Gaps
Annoyance
Sufficient
Direct exposure to wind

turbine noise can lead to
annoyance; however, the
effect may be modified
by factors such as visual
impact and attitudes.
Role of visual impact and

attitudes on perception
of wind turbines.
Prevalence of annoyance
in exposed populations,
gravity of effect, and
thresholds under
different conditions.
Role of specific sound
characteristics (amplitude
modulation, low
frequency noise).
Sleep
Disturbance
Limited
Direct and indirect (via

annoyance or stress response
or both) pathways are possible;
however, wind turbine noise is
likely only one among many
factors affecting sleep quality.
Nature of the mechanism

(direct, indirect, or both)
and the relative prevalence
and magnitude of the
effect for each.
Impacts of specific sound
characteristics (including
low-frequency sound)
on sleep.
Long-term effects of
wind turbine noise on
sleep disturbance.
99
Condition or
Symptom
Level of
Evidence
(IARC)
Possible Pathways
Knowledge Gaps
Stress
Inadequate
Direct and indirect (via

annoyance or sleep disturbance
or both) pathways are possible;
however, no evidence for a
direct association was found.
Wind turbine noise could be
one among many factors
contributing to stress response.

(direct, indirect, or both) and
relative prevalence and
magnitude of the effect.
Unclear whether mechanism or
stress response is comparable
with other sources of
environmental noise.
Impact of specific sound
characteristics on stress.
Long-term effects of wind
turbine noise on stress.
Cardiovascular
System and
Diseases
(including
hypertension,
cardiac
dysrhythmia,
tachycardia)
Inadequate
Analogous research suggests that

direct and indirect (via annoyance
and stress or sleep disturbance or
both) pathways are possible;
however, no evidence for an
association with wind turbine
noise was found.
Adequate epidemiological
evidence.
Effects of long-term exposure.
Nature of the mechanism.
Diabetes
Inadequate
An indirect pathway (via

stress, sleep disruption, or
combinations) is plausible;
however, the evidence linking
these to noise from wind
turbines was not consistent.
Adequate epidemiological
evidence.
Effects of long-term exposure.
and comparability to the
effect of other types of
environmental noise.
Hearing
Impairment
Evidence of
no causal
relationship
Sufficient evidence was found

in research on other types of
noise to conclude that
permanent noise exposure
below 75 dB(A) does not lead
to hearing loss, even after
lifelong exposure.
Tinnitus
Inadequate
Research on tinnitus suggests

that an indirect pathway via
stress is possible.
General uncertainty over the

causes of tinnitus and links
to other conditions such as
impaired hearing.
100
Condition or
Symptom
Level of
Evidence
(IARC)
Possible Pathways
Knowledge Gaps
Cognitive
or Task
Performance
Inadequate
Research on other types of

noise suggests that noise
exposure can affect cognitive
performance; however, the
character of the impact
(positive or negative) and
its strength vary with many
factors, including sound
characteristics and the type
of task used to test cognitive
performance.
Understanding of how noise

exposure affects different
types of cognitive or task
performance, including
clear case definitions
and measurement of
cognitive performance.
Psychological
Health
(anxiety,
depression,
psychological
distress)
Inadequate
Noise exposure could be a

contributing factor to the
development or aggravation
of psychological disorders.
General understanding
of possible links between
noise exposure and
psychological disorders.
Health-Related
Quality of Life
Inadequate
No mechanism identified or
postulated. Exposure to wind
turbine noise affects several
categories that are used to
measure quality of life, many
of which overlap with the
health impacts reviewed here.
Research focusing on the

relative impacts of wind
turbine noise compared
to other factors that affect
quality of life.
The conditions and symptoms listed here are those attributed to wind turbine noise from various
sources (see Table 5.1), for which the Panel found empirical research specific to wind turbine noise.
Table 7.2
Health Effects that Have Not Yet Been the Subject of Empirical Population-Based
Research on Wind Turbine Noise
Possible Pathways
Immune System
Impaired immunity
An indirect pathway via stress response is possible; however, none of the

studies assessed used variables that could test for associations between
wind turbine noise and impaired immunity.
Musculoskeletal System
Back pain
Joint pain
Muscle pain (myalgia)
Shaking (palsy)
Indirect pathways via stress are possible; these symptoms could be

caused by a large number of factors and medical conditions.
No evidence regarding association with wind turbine noise exposure
was found in the literature.
101
Possible Pathways
Nervous System
Disturbances of skin
sensation
No pathways were described or proposed in the literature reviewed.
Effects of non-audible
sound (inner ear, cochlea,
vestibular system)
Within the normal hearing frequency range, sounds have to be at a very

low level to be non-audible, and at those levels would not contribute
to any type of known hearing dysfunction. For low-frequency signals,
including infrasound, activation of the vestibular system is possible if
the signal levels are at high intensity. For both normal subjects and even
for individuals with certain medical conditions that result in lowered
vestibular activation thresholds (e.g., Tullio syndrome, superior canal
dehiscence, perilymphatic fistula, or enlarged vestibular aqueduct),
wind turbine signals are unlikely to reach the activation threshold. It
is therefore unlikely that wind turbine noise could directly cause any
symptoms associated with vestibular dysfunction, such as vertigo,
dizziness, vision problems, or nausea.
Fatigue
Daytime fatigue is a common consequence of sleep disruption.
Headache
Headaches are sometimes associated with noise annoyance, but not

consistently in the available literature.
Several pathways are possible (via annoyance, sleep disruption, or
stress, or combinations); however, the symptom is associated with
many possible causal pathways.
Nausea
A pathway based on stimulation of the vestibular system through

non-audible sound in populations with certain medical conditions
has been proposed; however, no evidence supporting its existence
has been found.
Pressure in the chest
Sensation of
internal vibration
Vertigo, dizziness
See effects of non-audible sound.
Vision problems
See effects of non-audible sound.
Nervous System: Auditory

Communication
interference
This could be a result of annoyance, and is often an indication of

noise annoyance. No evidence testing the association has been
found, however.
Ear pressure or pain
A pathway based on activation of the cochlea by sound in populations

with certain medical conditions has been proposed; however, no evidence
supporting its existence has been found.
Respiratory System
Nosebleed
The conditions and symptoms listed here are those attributed to wind turbine noise from
various sources (see Table 5.1), for which the Panel found no empirical research specific to
wind turbine noise. In each case, the evidence is inadequate with respect to a causal link
with wind turbine noise.
102
Furthermore, the Panel noted a paucity of research regarding the health

effects of wind turbine noise on sensitive populations, including children and
infants as well as populations affected by clinical conditions that may lead to an
increased sensitivity to sound. This includes research on the potential impacts
of noise exposure on cognitive development and learning in children. This is
also true of research on the health effects of environmental noise in general.
7 . 2
Q UA LI TY O F T H E E V I DE NC E RE V I E W ED
In view of the relative novelty of research on health effects of wind turbine noise,
the Panel found it useful to review the general quality of the evidence available
in comparison with research on the health impacts of other environmental
stressors, and to provide an overview of limitations and general knowledge gaps.
Selection Bias
All population-based studies, and most of the experiments reviewed by this
Panel, were based on self-reported data on health impacts. Self-reported surveys
of environmental stressors are susceptible to selection bias (i.e., people who
are affected by an environmental impact are more likely to respond to a survey
than those who are not) (Rief et al., 2011). In order to reduce selection bias in
self-reported cross-sectional surveys, researchers may hide the true intent of
surveys from respondent to the extent possible. The three largest cross-sectional
studies reviewed (SWE-00, SWE-05, NL-07) stated that the questionnaires used
were masked as general surveys on quality of life. Other studies noted that they
used similar methodologies, but do not explicitly state whether questionnaires
were sufficiently masked to avoid selection bias. The Panel therefore noted that
the current evidence base may be affected by some degree of selection bias.
Confounding Factors
Factors other than noise may be responsible for a health outcome. If those same
factors are also associated with exposure to wind turbine noise (for example,
if those living closer to wind turbines happen to be older than average), it is
possible that a confounder (in this example, age) is causing the increase in
adverse health effects, and not the sound from wind turbines. Age, health
history, pre-existing medical conditions, and attitudes towards wind turbines
or visual impacts are factors that might be considered potential confounders
and need to be controlled in the study analysis. While the cross-sectional
studies reviewed for this assessment made reasonable efforts to control for
those potential confounding factors, it is unlikely that they were able to control
them completely.
103
Sample Size
In observational epidemiological studies, the number of people in the study
(sample size) is important, for a number of reasons. The actual number
required depends on the study design, but, generally speaking, studies must
have adequate numbers of participants to ensure that different exposure levels
are adequately represented (e.g., exposed and control groups) as well as, for
example, all demographic categories (old and young, male and female, etc.) in
order to control for potential confounding. Researchers must also ensure that
the sample size is sufficient to rule out a result occurring by random chance. This
is called statistical significance and depends on a number of factors, including
the anticipated increase in risk for disease among the exposed persons. In
the studies reviewed by the Panel, sample size was sufficiently large to capture
annoyance and sleep disturbance, while sample sizes may have been too small
to detect other health effects such as cardiovascular diseases, for which the
anticipated increase in risk among the exposed would be quite small.
Exposure Monitoring
The Panel observed that most of the studies estimated exposure to indoor
or outdoor wind turbine noise using computer-based models, rather than
direct observations. A combination of measurements, modelled exposure, and
outcomes is needed to settle the question of how to estimate exposure and its
effects. In addition, the Panel noted that studies most often used a long-term
equivalent A-weighted level (LAeq) to describe sound exposure. However, such
approaches do not capture sound characteristics such as amplitude modulation.
Identifying specific wind turbine sound characteristics may help to better
understand the cause of certain health outcomes such as annoyance. This is
discussed in further detail in Section 8.2.
Most epidemiological studies of wind turbine noise lacked sufficient power
to detect meaningful health effects, or were designed in a way that could not
convincingly rule out bias in responses or confounding factors. As a result,
the evidence for a causal relationship between exposure to wind turbine
noise and many health effects was inconsistent or unclear. In addition to the
issues described above (selection bias, confounding factors, sample size, and
exposure monitoring), the Panel also identified an absence of longitudinal
health studies. However, the Panel recognizes, in the context of research on
the health effects of wind turbine noise exposure, the difficulty of recruiting
large samples of participants and following them over time.
104
7 . 3
FURTHE R R ES E ARC H
Evidence shows that annoyance is likely to be caused by exposure to wind

turbine noise. However, little is known in Canada about the prevalence of
annoyance within the population exposed to wind turbine noise. In addition,
the evidence reviewed suggested an association might exist between wind
turbine noise and sleep disturbance or stress, either caused directly by noise
or indirectly mediated by annoyance. Further research and surveillance would
provide a better understanding of the prevalence in the general and exposed
populations. In particular, research is needed to gain knowledge of (1) the
incidence of the health outcome in the general population, (2) the proportion
of the population exposed to wind turbine noise; and (3) the incremental risk
of developing the disease that is associated with exposure to wind turbine noise.
Chronic annoyance and sleep disturbance are linked to stress responses in
the context of long-term exposure to other sources of noise, such as air and
road traffic. Furthermore, these are risk factors for other health effects, such
as cardiovascular disease, which are also associated with long-term exposure
to other sources of community noise. Further research on stress and sleep
disturbance would provide more input in assessing a causal relationship between
exposure to wind turbine noise and those two conditions. Useful research
would involve studies with appropriate designs (addressing sample bias and
confounding factors in particular) and sample sizes with adequate statistical power.
Such research may help to better assess the causality between wind turbine
noise and those conditions. At the time this report was being finalized
(November 2014), preliminary results from a large-scale study from
Health Canada became available (see Box 7.1).
105
Box 7.1
Health Canadas Wind Turbine Noise and Health Study
In 2012, Health Canada started a large-scale cross-sectional epidemiological study
involving approximately 2,000 dwellings. The aim of this study was to measure
potential health outcomes in areas exposed to sound from wind turbines. This
study was developed by Health Canada in collaboration with Statistics Canada to
provide an evidence base and inform policies and practices in Canada regarding the
development of wind energy projects. Among the outcomes measured by the Health
Canada study were:
Sleep disturbance, measured with a sleep watch that gauges sleep onset, sleep time,
and efficiency. Self-reported sleep quality was also assessed with a questionnaire.
Stress, measured by cortisol concentration in hair samples, blood pressure, and
heart rate, as well as by a questionnaire (perceived stress).
Self-reporting annoyance (indoor and outdoor), measured by self-reporting.
Quality of life, measured using a questionnaire.
Health Canadas preliminary findings were made publicly available in November 2014.
The Panel reviewed those findings but, as they were preliminary, it could not integrate
this research into the evidence considered in Chapter 6. However, the Panel observed
that the findings from this study were mainly concordant with its own findings. The
main results are presented below.
Of the dwellings selected, 1,238 households (78.9%) agreed to participate in the
study. Both self-reported conditions and physiological measurements were described
in the preliminary results. Regarding self-reported conditions, the study did not find
associations between wind turbine noise exposure and self-reported sleep, self-reported
illnesses (such as dizziness, tinnitus, and headaches), and chronic diseases (such as
heart disease, high blood pressure, and diabetes). The study also did not find any
association between noise exposure and self-reported perceived stress and quality of
life. However, it found an association between increasing levels of wind turbine noise
and annoyance towards wind turbine characteristics (noise, shadow flicker, blinking
lights, vibrations, and visual impacts). Health Canada also captured physiological
measures related to stress and sleep quality and found that the measures (e.g., hair
cortisol levels, blood pressure, sleep watch) were consistent with self-reported results
(no association between cortisol concentration, blood pressure, sleep efficiency, and
exposure to wind turbine noise was found).
(Health Canada, 2014a, 2014b)
106
7 . 4
After its review of the evidence, the Panel returned to the framework proposed
in Chapter 5 that outlined possible relationships between wind turbine noise
exposure and health effects. Based on the literature reviewed, including studies
of other types of environmental noise, the Panel proposes a complex model of
causal pathways (Figure 7.1), in which characteristics of wind turbine noise (lowfrequency, tonality, and amplitude modulation) are modified by visual impact,
personal attitudes, noise sensitivity, and existing medical conditions. These
can lead to direct health effects via hearing, and to cognitive and emotional
responses via processing in the brain.
To confirm possible pathways, knowledge gaps need to be filled. For each
health outcome, knowledge gaps are outlined to help inform the research
agenda. For many other health symptoms for which no studies specific to wind
turbines were found, plausible pathways are suggested based on the literature.
Many of the studies reviewed had limitations or design weaknesses that affected
the quality of the evidence. Population-based studies that relied on self-reported
surveys were susceptible to selection bias, but many did not explain any steps
taken to avoid this.
Relationships among noise exposure, chronic annoyance, sleep disturbance,
and stress are a particularly fruitful area for further research to elucidate
pathways. Future research into specific health outcomes is needed to understand
the general incidence and prevalence of the health outcome, the population
exposed to wind turbine noise, and the risks associated with exposure.
107
Chapter 8 Promising Practices
8
Promising Practices
Engineering Technologies
Promising Practices to Monitor Noise
Increasing Acceptance
Chapter Summary
108
Promising Practices
Key Findings
Wind turbine designs that may lessen sound production are being explored;
however, technological development is unlikely to resolve, in the short term, the
current issues related to perceived adverse health effects of wind turbine noise.
Sound pressure levels can vary widely within a structure even within the same
room because of interference of sound waves reflected from walls. Improved
measurement protocols have been proposed to more accurately account for this
variation in indoor sound.
Further studies focused on amplitude modulation are needed to better understand
the potential impact of wind turbine noise on human health. Locals in some countries
have recommended the use of measures that capture this characteristic.
Impact assessments and community engagement give communities greater knowledge
and control over wind energy projects.
The previous chapters demonstrate the complex nature of wind turbine sound
production, propagation, and potential effects on humans. The Sponsor
specifically requested that the Panel consider practices in engineering and other
fields that could minimize adverse community response to wind turbine noise.
In light of its findings, the Panel has identified several promising strategies,
which are presented in this chapter:
Reducing sound emission at the source through technological improvements;
Better monitoring and understanding of the characteristics of sound from
wind turbines through adequate measurement methods, both indoors
and outdoors.
Implementing adequate impact assessments and community
engagement activities.
8 . 1
E N G I N E E R I NG T E CH NOL OGI E S
Modern utility-scale wind turbines produce sound from mechanical and

aerodynamic sources (see Chapter 3). As discussed in that chapter, mechanical
sound is not usually significant in modern turbines, as it has been greatly
reduced through high-quality manufacturing and design elements such as
soundproofing and insulation between mechanical components and the nacelle
109
structure (Hau, 2006). These and other advances in design and manufacturing
can generally reduce mechanical noise below the level of aerodynamic noise
from the blades (Oerlemans, 2011).
Aerodynamic noise is produced by air flow interacting with different parts of
the blade (see Section 3.1.2). No single technology was identified that will
drastically reduce the sound output of modern utility-scale wind turbines,
but the Panel did identify noise reduction technologies and practices that are
actively being studied (Barone, 2011; Oerlemans, 2011):
pitch control optimization;
blade modifications; and
technologies to manage curtailment to minimize both noise and loss of power.
8.1.1 Pitch Control Optimization
Modern utility-scale wind turbines use information from wind sensors on the
nacelle to automatically make physical adjustments to parts of the turbine, such
as the blade pitch and rotational speed, in order to generate the maximum power
possible (Manwell et al., 2010). Pitch control systems allow the blades to be rotated
around their long axis (from the hub to the blade tip). This allows a degree of
control over the angle of attack, or the angle at which air hits the leading edge
of the turbine blade. Reducing the angle of attack can reduce noise emissions,
but only by sacrificing some power production (Oerlemans, 2011). Theoretical
and empirical research continues to identify optimal pitch control settings
that strike a balance between power and sound output (Manwell et al., 2010;
Bakker et al., 2012). Applying these settings in response to continually changing
wind speed and direction, however, poses additional challenges.
8.1.2 Trailing Edge Modifications
Even with optimal blade pitch, the most common source of aerodynamic
noise is interaction between air flow and the trailing edge of the turbine
blades (see Section 3.1.2). Modifying the physical shape or properties of the
blades trailing edge can therefore affect the level and characteristics of sounds
produced. Aeroacoustic theory suggests that serrations or brushes along the
trailing edge of the blade would decorrelate the sound and make it propagate
less effectively, and would thus reduce aerodynamic sound emissions. In fullscale tests, trailing edge serrations were most effective at high wind speeds,
reducing noise by up to 5 dB in 10 m/s winds (Barone, 2011; Oerlemans, 2011).
Noise levels were reduced on average by about 3 dB, but at low wind speeds,
additional high-frequency noise was produced, reducing the effectiveness of
the blade serrations. Although trailing edge serrations have shown potential
for reducing wind turbine noise, the mechanisms responsible are still not well
understood (Barone, 2011).
110
8.1.3 Curtailment
Curtailment generally refers to sacrificing some power generation to reduce
sound output to below an acceptable threshold. As described above, this might
include adjusting blade pitch to reduce the angle of attack, or reducing the
rotation speed of the turbines, if possible (Oerlemans, 2011). In extreme cases,
wind turbines may have to be shut off to prevent sound emission.
Curtailment is often a last resort if other adjustments cannot reduce noise
emissions below applicable limits. It is more commonly used at night, or when
the wind is blowing from certain directions or above certain speeds, in order
to reduce exposure levels for residents at particular locations (EPA, 2011).
Identifying conditions under which curtailment is necessary can help reduce
the need to apply it. Therefore, better sensors and automated systems offer
another approach to limiting the amount of power that must be sacrificed to
reduce noise levels. If certain levels or types of turbulence are known to cause
enhanced amplitude modulation, meteorological sensors (e.g., LIDAR or
SODAR) could be used to detect such weather conditions before they reach
a turbine, and operations could be adjusted accordingly, by changing blade
pitch or reducing power output (RenewableUK, 2013).
8.1.4 Continual Technological Improvement in a Global Market
The technologies described above are examples of promising methods for
reducing noise levels in some situations, but the Panel notes that no single
solution is likely to address all noise concerns in the short term. Nevertheless,
ongoing research in wind turbine design and manufacture means that there
is continual technological development in this area. Over 20 years, many small
improvements have combined to reduce the overall sound power level of wind
turbines for the same rotor diameter. This allows the use of larger turbines
with the same level of noise emissions.
The Panel also notes that wind turbine manufacturers sell their products
on a global market. An individual country such as Canada would therefore
have a limited ability to influence technology trends through turbine design
standards, since wind turbine manufacturers sell their products around the world
(see Chapter 2). Most standards for wind turbine design, manufacture, and sound
measurement are set by international organizations, such as the International
Electrotechnical Commission (IEC) and the International Organization for
Standardization (ISO).
Canadian regulations could affect which wind turbine technologies are used
here. Noise limits and regulations are often set at national, provincial, and even
local levels. Therefore, while regulations and noise limits may not be able to
directly affect technology development, they may determine which technologies
are used, and how they are applied.
8 .2
111
P R O M I S I NG P RAC T I CE S T O M O NITOR NOISE
As discussed in Chapter 2, noise limits are an important and necessary tool to

limit exposure to sound from wind turbines. Adequate measurement of sound
from wind turbines is paramount to properly inform such regulations but also
to better understand the effect of wind turbine sound for research purposes.
The Panel identified four criteria that are important to consider in terms of
noise guidelines and measurement: background noise, indoor measurement,
night-time measurement, and measurement of amplitude modulation.
8.2.1 Background Noise
Background noise is an important element in sound perception. A specific
sound is more noticeable if the competing sound signals are at lower pressure
levels; that is, if background noise is absent or low. This effect is captured by
the masked hearing threshold (see Chapter 4), the sound pressure level at
which a sound can be heard in the presence of competing sound signals such
as background noise. The effect of background noise has been observed in
research on sound from wind turbines (see Chapter 6). Studies showed that,
for the same exposure, reported annoyance was more prevalent in quieter rural
areas than in noisier urban areas (Pedersen & Persson Waye, 2007).
Wind can contribute to background noise as well as noise produced by wind
turbines. As discussed in Chapter 3, the wind speed at the elevation of the blades
may be much higher than that at ground level, an effect called wind shear. As
a result, the high wind speed at the blade elevation leads to increased noise,
but the lower wind speed at the ground reduces the masking effect of natural
wind sound for residents. For example, in Sweden, particular attention is paid
to houses in wind-sheltered areas, where the wind speed is about half of that at
the wind turbine site. In such sheltered areas, the sound from wind turbines
is more prominent than in wind-exposed locations (Persson Waye, personal
communication). When regulating noise, taking into account background
noise creates a level of complication, as background noise changes depending
on factors that vary over time (e.g., wind speed).
8.2.2 Indoor Measurement
Indoor measurement is particularly relevant for environmental sound exposure,
including wind turbine noise, as this sound may affect people in their homes,
particularly at night when they are sleeping. As well, the lower frequency
sound associated with wind turbines loses less energy than higher frequencies
when transmitted through walls and windows (see Chapter 3). Furthermore,
interference among sound waves may create large variations in sound pressure
level in the same room.
112
There are specific guidelines to measure low-frequency sound indoors

(Jakobsen, 2001; Pedersen et al., 2007b; HGC Engineering, 2010). However,
because of the variations in sound pressure level noted above, it is challenging
to measure sound indoors. The variations result from reflection of sound
waves by walls, causing interference with the original wave. The pattern of
high and low sound pressure levels produced is called a standing wave pattern:
at low frequencies, it may create large variations in the same room. Despite
this challenge, there are methods to accurately measure the sound pressure
levels indoors. For example, Pedersen et al. (2007b) suggested an improved
measurement protocol to more accurately assess indoor sound (see Box 8.1).
However, such measurements are not always possible. Some countries, such as
Denmark, do not use indoor measurements and rely instead on computer-based
simulations to assess compliance with regulations (DME, 2011).
Box 8.1
Improved Indoor Sound Pressure Measurement
Because of interference among sound waves within a structure, the sound pressure
level at low frequencies may vary as much as 20 to 30 dB. Adequate measurement
of areas of the room with high sound pressure levels is critical for assessing the
causes of annoyance. Pedersen et al. (2007b) studied typical indoor measurement
methods and concluded that these methods are not optimal and suggested a new
methodology. This methodology is based on the average of measurements of sound
pressure levels taken in four corners of a room at a maximum distance of 0.1 metres
from the room boundaries (Pedersen et al., 2007b).
8.2.3 Night-Time Measurements

In contrast to traffic noise, which usually subsides at night, wind turbine noise
may carry on overnight when background noise is lower. Noise is therefore often
more noticeable at night and more likely to affect sleep. WHO has suggested
that, in residential areas, sound pressure levels should not exceed 40 dB(A)
at night (outside) (WHO, 2009). Ellenbogen et al. (2012) recommended
night-time noise limits of 37 dB(A) in residential area (6 m/s wind) and
42 dB(A) (6 m/s wind) in sparsely populated areas. There are various standards
for reporting day and night sound pressure levels, with a penalty added for
exposure during night-time. For example, Ldn adds a 10 dB(A) adjustment to
night-time exposure (see Chapter 3). The Panel recognizes the usefulness of
measuring night-time exposure to sound from wind turbines, which is currently
not widely undertaken in Canada.
113
8.2.4 Measurement Methods to Better Assess Possible Health Effects

of Wind Turbine Noise
There are several ways to describe and measure sound exposure, and each
emphasizes some and deemphasizes other sound characteristics (e.g., pressure
level, frequency, amplitude modulation; see Chapter 3). As noted in the WHOs
Guidelines for Community Noise, there is a very complex multidimensional
relationship between the various characteristics of the environmental noise and
the effects it has on people (WHO, 1999). Proper measurement and reporting
standards are important, because they allow sound levels to be comparable
across studies, and they can help identify and clarify the sound characteristics
associated with reported adverse health outcomes.
As noted earlier in this report, sound from wind turbines has the following
characteristics: (1) it is broadband and composed of a range of frequencies
including low-frequency sound; (2) higher frequencies tend to be reduced
more indoors and with increasing distance, leading to an emphasis on lowfrequency components; (3) it can have periodic amplitude modulation; and
(4) the sound power level varies with wind speed at hub height. The first and
second characteristics are common among many sources of environmental
sound, such as the sound from the wind.
In reviewing literature specific to wind turbines, the Panel noted that most
measurements in research papers were presented as A-weighted, time-averaged
measurements (LAeq; see Table 3.1). Although A-weighting best approximates
human hearing and is a well-adopted measure, it may not adequately capture
the low-frequency components of sound from wind turbines. In addition,
periodic amplitude modulation (swishing or thumping), a characteristic
specific to sound from wind turbines, is not captured by time-averaged
indicators (e.g., Leq) (RenewableUK, 2013). Periodic amplitude modulation
and dominance of low-frequency components may contribute to health effects
such as annoyance, as documented in Chapter 6. These characteristics are critical
to better understanding the effects of sound from wind turbines on humans.
A practice to better measure the amplitude modulation of sound from wind
turbines is presented in Box 8.2.
114
Box 8.2
Improved Capturing of Amplitude Modulation
A recent report reviewed the various methods of sound measurement to best
characterize the sound from wind turbines (Kaliski, 2014). The report was prepared
for the Massachusetts Wind Turbine Noise Technical Advisory Group (WNTAG), which
provides advice to the Massachusetts Department of Environmental Protection
(MassDEP) concerning noise regulations and policies for wind turbines (CoM, 2014).
As part of its activities, WNTAG has logged about 120 million wind turbine sound
records captured in 150 different conditions (e.g., meteorological, operational),
which constitute an important base for better understanding the characteristics of
sound from wind turbines.
The report suggested that the best measure to capture amplitude modulated sound
is the fast response sound level (LAf),* which is already used by MassDEP to assess
short duration repetitive sound. Using 1 second LAfmax allows for comparison of
background and operating sound levels during similar wind speeds, without ignoring
amplitude modulation peaks (Kaliski, 2014). Use of such short-period measurements
to capture amplitude modulation has also been suggested in conclusions from primary
research (RenewableUK, 2013). Such measurements have been shown to be a good
predictor of the risk of annoyance (Seong et al., 2013) (see Chapter 6).
* For more information on time-weighted sound metrics, see Figure 3.6.
8 . 3
I N C R E A S I N G ACC E P TANCE
In Chapter 6, the Panel concluded that annoyance results from combinations of

contributing factors, including personal attitude. Studies showed that residents
with negative attitudes towards wind energy were more likely to be annoyed
than those with positive attitudes. Negative attitudes towards wind energy can
be triggered by a process that is perceived as unfair. For example, Krogh (2011)
illustrated the relationship between wind energy development, health impacts,
and the loss of social justice. The author argued that loss of social justice arises
from several factors associated with wind energy development, including the
lack of fair process, loss of rights, and associated disempowerment. The author
also proposed that the lack of consultation with communities, as well as the
construction of many wind turbines over short periods of time in certain parts
of Canada, increased the feeling of injustice and resulted in negative social
impact. In this section, the Panel describes practices pertaining to impact
assessments and community engagement.
115
8.3.1 Environmental and Health Impact Assessment

In order to meet noise regulations at a prospective site, depending on the
jurisdiction, size, and capacity of a planned wind facility, wind energy developers
may need to undertake an environmental impact assessment. In Canada, there
are no requirements at the federal level except for locations within federal
jurisdiction (Haugen, 2011). Rules vary among provinces and territories. For
example, under the Alberta Utilities Commissions rules, applicants to set up
a new energy facility need to submit a noise impact assessment and predict
the impact of noise under normal conditions at the most affected dwelling(s)
(AUC, 2013). The Ontario Ministry of the Environment also requires a Noise
Assessment Report for all proposed wind energy projects and provides detailed
guidelines to conduct noise impact assessments (MOE, 2008). The Quebec
Ministre du Dveloppement durable, Environnement et Lutte contre les
changements climatiques also provides instructions on how to conduct a noise
impact assessment (MDDELCC, 2006). Outside of Canada, other countries have
similar requirements. For example, the Irish Department of the Environment
has developed particularly comprehensive guidelines to assess the impact of
wind turbine development (Box 8.3).
Box 8.3
Assessing the Impact of Wind Turbine Development:
the Irish Example
In the Republic of Ireland, the Department of the Environment, Community and Local
Government (DECLG), the regulatory body responsible for wind energy development,
requires developers to conduct full impact assessments to evaluate the areas of safety
(setback), environment (heritage of site, noise, shadow flicker, and visibility), ecology,
as well as social and economic sustainability. These guidelines cover activities from
pre-application consultations to decommissioning and reinstatement. In particular, the
DECLG suggests that the developer engages in consultations with the local community
at the early stage of the planning process, allowing the communities to have an input.
Appropriate mitigation and compensatory measures must be established before
construction to prepare for impacts or changes that were previously unanticipated.
(DEHLG, 2006)
116
8.3.2 Community Engagement

The success of wind power depends on how well the wind industry
learns to include the public in decisions, both for the opportunities
this allows for broader dissemination of information about wind power
and for the suggestions the public can contribute to the discussion of
their concerns and how to accommodate them.
(Pasqualetti, 2002)
Residents want the opportunity to be involved in the decision-making process
for the location of wind turbines, since equity and fairness are crucial for
community acceptance (Wolsink, 2007). Collaborative approaches can give
local communities an opportunity to discuss and address their concerns
(Toke et al., 2008). Various best practices exist to guide wind turbine developers
and policy-makers in community engagement (RABDTI, 2007; CanWEA, n.d.).
The objective of community engagement is to inform, educate, and involve the
local residents with the wind energy project, specifically highlighting its benefits
and taking into account concerns, with the goal of achieving social acceptance.
Public participation techniques include processes such as referenda, public
hearings, public surveys, and negotiated rule-making (Jami & Walsh, 2014).
In addition to implementing setbacks and noise limits, most international
best practices strongly recommend that the wind energy developer engage in
consultation and communication with the local authorities and residents at an
early stage of project development. For example, the Australian government
recognizes that community consultation is an important two-way process: informing
the stakeholders and actively encouraging local people to offer feedback and ask
questions (EPHC, 2010) (see Box 8.4). A new model for environmental noise
annoyance is relevant in the context of community engagement: soundscape
a concept introduced to comprehensively measure the potential effect of
noise on quality of life (Schulte-Fortkamp, 2014). Soundscape techniques take
into account both the context and the perception of sound and consider that
human response is not a single factor but is related to the human perception
of the situation (Schomer et al., 2010).
117
Box 8.4
Guidelines for Engaging Communities
The Australian Environment Protection and Heritage Council (EPHC) has developed
a series of guidelines to assist developers in addressing stakeholder participation in
wind energy development. The goals of these guidelines are to provide:
Principles and considerations for wind energy developers for planning and delivering
stakeholder participation activities.
Recommendations for stakeholder input to the assessment and how to manage
key study areas (including noise).
These guidelines cover the various stages of wind energy project development:
site selection, project feasibility, planning application, construction, operation, and
decommissioning. For each project stage, the EPHC recommends specific milestones
be met, such as continuously seeking feedback and input from stakeholders at the
project feasibility and planning stages to better understand their concerns and
opinions. Regarding noise, the guidelines recommend measuring background noise
at the project feasibility stage and the construction stage, as well as conducting
noise monitoring at identified stakeholder properties. Although these guidelines
have been developed in Australia, the general principles of risk management and
communication can easily be applied to the Canadian context.
(EPHC, 2010)
With the combination of public inquiries and environmental impact assessments,

many European countries have been able to make wind energy development
less controversial, in large part because communities are engaged in and benefit
directly from this process (CSE, 2009).
8 .4
Technological improvements in wind turbine design have resulted in larger and

more powerful turbines over the past 20 years. While turbines now produce less
noise for the same size than previous-generation turbines, this effect is offset
by their growth in size and production capacity. There are some promising
technologies being studied to lessen sound production, but no silver-bullet
technology is expected to resolve all problems of sound from wind turbines.
The regulatory measures (noise limits and setbacks) currently used in Canadian
118
and international jurisdictions will therefore continue to be necessary. Many

of these measures have taken into account the need to adjust to different wind
speeds, to daytime and night-time differences, and to background noise.
These adjustments are important to any future regulatory measures, but they
all rely on sound measurement. Measurement techniques and standards need
to be established, as current use of A-weighted time-averaged measurement
does not capture some of the aspects of sound from wind turbines, in particular
amplitude modulation. Measuring indoor sound is important to capture sound
affecting residents, but it is challenging due to the wide variation in sound
pressure levels within a single room.
Fortunately, there are many international practices in these areas that can inform
future directions in Canada, including choice of measure, and methods for
measuring indoor sound pressure levels. However, because perception of wind
turbines has to do with aspects other than sound, health and environmental
impact assessment and community engagement are also critical. Several
international studies have found that environmental assessment, from the
proposal stage to after installation, helps lead to successful, well-accepted
wind energy projects. Engaging communities results in greater acceptance
and involvement. These measures might also mitigate health effects through
feedback, information, and positive attitudes.
119
Chapter 9 Conclusions
9
Conclusions
Final Thoughts
120
Conclusions
This chapter synthesizes the findings and evidence of the previous chapters
to summarize the Panels response to the Charge. The main question and
sub-questions asked by the Sponsor are answered individually.
Is there evidence to support a causal association between exposure to

wind turbine noise and the development of adverse health effects?
The Panel found sufficient evidence to establish a causal relationship between

exposure to wind turbine noise and annoyance, and limited evidence to establish a
causal relationship between exposure to wind turbine noise and sleep disturbance.
The Panel found evidence suggesting a lack of causality between exposure to
wind turbine noise and hearing loss. The available evidence was inadequate to
draw any conclusion regarding causation for all other health effects considered
(see Chapter 6 for details, summarized in Tables 7.1 and 7.2). Although
evidence concerning the association between wind turbine noise and stress is
inconsistent and not of high enough quality to permit firm conclusions, stress
has been linked to many other environmental sources of noise, often with
links to annoyance and sleep disturbance. Nevertheless, causal relationships
between these three effects are often unclear: each may lead to another, with
more than one causal pathway likely accounting for observed cases of each
health effect in a population. Other health effects commonly attributed to
wind turbine noise, including headache and communication interference,
have been attributed to many factors. However, no empirical research exists
to establish a causal relationship between these effects and wind turbine noise.
Some additional causal mechanisms have been proposed that could link wind
turbine noise to symptoms such as vertigo, dizziness, nausea, or vision problems.
Such symptoms would be consistent with stimulation of the vestibular system,
which can be activated by sound at levels that are above normal human hearing
thresholds for infrasound and low-frequency sound. Sensitive individuals with
certain clinical conditions might have lower activation thresholds, below normal
hearing thresholds, but whether such activation thresholds are within typical
sound levels from wind turbines remains unclear, as does the prevalence of
these conditions in populations exposed to wind turbine noise. However, the
Panel concluded that wind turbine signals are unlikely to reach activation
threshold in individuals.
121
Are there knowledge gaps in the scientific and technological areas

that need to be addressed in order to fully assess possible health
impacts from wind turbine noise?
The Panel identified specific knowledge gaps for each health condition studied,
where specific types of evidence would help clarify the strength of associations,
avoid bias, or eliminate possible confounding factors with respect to exposure
to wind turbine noise. For example, it is unclear whether annoyance is caused
by wind turbine noise alone, or whether factors such as visual impacts and
personal attitudes modify the noise-annoyance relation. The possible pathway
that could lead to sleep disturbance or stress is also unclear. It is similarly
unclear whether the possible pathway that could lead to sleep disturbance or
stress is the direct result of exposure to wind turbine noise or is the result of
annoyance as a mediating factor. The Panel also noted that knowledge gaps
concerning potentially sensitive populations such as children and infants, as
well as individuals with certain clinical conditions, complicate any attempts
to assess the level of health risk associated with exposure to sound from wind
turbines. In addition, the Panel noticed the lack of data on adverse health effects
at the population level, which could be gathered through rigorous research
designs to minimize bias and confounding factors, and could link outcomes
to exposure measurement.
Reviewing the evidence, the Panel observed that most of the cross-sectional studies
used standard modelling approaches to predict sound levels. Combinations
of measurements, modelled exposure, and outcomes are needed to settle the
question of how to estimate exposure and its effects. In addition, modelling
and some measurement approaches do not capture many sound characteristics
relevant to wind turbines, such as amplitude modulation. Being able to link
those specific wind turbine sound characteristics to effects may help researchers
to better understand the cause of certain health outcomes such as annoyance.
Is the potential risk to human health sufficiently plausible to justify

further research into the association between wind turbine noise
exposure and the development of adverse health effects?
122
In reviewing the literature, the Panel identified many research and knowledge
gaps concerning pathways and causality. However, the Panel notes that it was not
within its mandate to assess whether specific research should be prioritized over
other public health research, or to provide recommendations on any specific
type of research. The Panel can comment on lines of research that would be
useful in addressing existing knowledge gaps.
For many adverse health effects, questions and uncertainties remain. Of the
wide range of reported health effects attributed to wind turbine noise exposure,
many are possible secondary outcomes of long-term sleep disturbance or
stress, but may have other causes. In particular, cardiovascular diseases, a set
of conditions responsible for almost one third of the deaths in Canada, are
known to be affected by long-term annoyance, sleep disruption, and stress. It
would therefore be beneficial to improve understanding about the potential
risks arising from long-term exposure to wind turbine noise, and to further
study the causal relationship between exposure to wind turbine noise and sleep
disturbance and stress.
Although sound from wind turbines appears to be at levels similar to other
common sounds in the environment, some characteristics are more common in
sound from wind turbines than in sound from other sources, such as periodic
amplitude modulation (causing a swishing or thumping sound), and
variations linked to wind speed at the height of the blades (which can be 60 to
100 metres or more above the ground). Continued research could help identify
measurements of sound exposure that are most relevant to possible health
effects, as well as improve our understanding of the mechanisms responsible
for annoyance or other plausible health effects of exposure to sound from
wind turbines or other sources. In particular, the Panel noted that periodic
amplitude modulation may be a critical component of sound from wind turbines
that triggers annoyance.
How does Canada compare internationally with respect to prevalence

and nature of reported adverse health effects among populations
living in the vicinity of commercial wind turbine establishments?
The Panel found insufficient data available for Canada to answer this question.
However, in terms of the nature of health effects, there is no reason to believe
that potential health effects for our population would be significantly different
than those for other nations. The conclusions concerning health effects drawn
from studies worldwide would therefore be applicable to Canada. In terms of
123
the prevalence of these effects, none of the population-based studies cited in

this report were conducted in Canada, so their conclusions offer no insight into
how prevalence of adverse health effects might compare with other countries.
At the time of the writing of this report, Health Canada released preliminary
results from the first-ever study to combine subjective and objective health
measures with measured and modelled sound levels from wind turbines.
Previous population studies of health effects of wind turbine noise have relied
almost entirely on health conditions self-reported by survey participants. The
preliminary results of the Health Canada study suggest differences in the
prevalence of self-reported annoyance between the two provinces examined.
Therefore, regional differences in the prevalence of effects are certainly
possible, either due to differences in the perception of wind turbines, or in other
as-yet-unknown factors that affect individual sensitivity. The full peer-reviewed
results of the Health Canada study were not available before the Panels report
went to publication.
Other countries, such as Australia, have set up reporting systems for the consistent
collection of noise and health complaints, which has allowed systematic analyses
of complaints related to wind turbines. Such complaints are, at best, a weak
surrogate for self-reported health effects. The only comparable analysis of
complaints in Canada was carried out in Alberta, and found overall low levels,
in contrast with pockets of high levels of complaints in Australia. However, this
finding may be attributable to the fact that Canadas wind energy industry is
relatively new compared to the industries in other countries.
Are there engineering technologies and/or other best practices

in other jurisdictions that might be contemplated in Canada as
measures that may minimize adverse community response towards
wind turbine noise?
The Panel identified three areas of current technological development that

show the most promise in reducing the sound output of modern utility-scale
wind turbines: pitch control optimization, blade modifications (serrations or
brushes), and technologies to manage curtailment under conditions known to
produce excessive noise. Each of these developments attempts to reduce sound
output while minimizing loss of power output, thus striking a balance between
environmental impacts and economic benefits. Regardless of whether these
particular technologies may ultimately be viable, research and development
has produced a steady stream of improvements that have continually led
124
to reductions in sound power level for wind turbines of a given size. Such
continual technological improvement is expected to lead to further reductions
in sound output.
However, no silver-bullet technology is expected to resolve all problems
of sound from wind turbines, especially in the short term. Therefore, other
measures will continue to be relevant.
Wind turbines are sold on a global market, and regulations in a single country
such as Canada are unlikely to provide sufficient incentive to further reduce
their sound power. However, there is a global initiative across the wind energy
industry to continually reduce the sound power of wind turbines in order to
increase social acceptability across all markets. As well, regulatory measures
taken in a particular jurisdiction (country, province) determine which available
technologies and practices can be used to achieve local goals.
Thus, regulation is an important tool to control exposure to wind turbine
noise and, consequently, its potential health impact on the public. The Panel
reviewed best practices for improved measurement of background noise, indoor
sound, night-time sound, and amplitude modulation. The Panel suggests that a
combination of those practices would provide a better understanding of noise
exposure and inform noise guidelines or regulation by the various orders of
governments (federal, provincial, and municipal).
Finally, the Panel reviewed best practices for impact assessment and community
engagement. In the past, failure to engage with the community has increased
negative attitudes among the public toward wind projects and the wind industry.
By contrast, health and environmental impact assessment and community
engagement, from the inception of projects throughout the project lifecycle
including after installation, can provide an opportunity to adapt to community
feedback and concern.
9 . 1
FI N A L THO U GH T S
Wind turbines have become a contentious source of energy, in Canada and

internationally. Although some hail them as a source of renewable, clean energy,
others have expressed concerns about undesirable environmental and human
health impacts. Assessing the possible human health impacts of exposure to
wind turbine noise poses several scientific and technical challenges.
125
This report has found sufficient evidence that exposure to wind turbine noise
(possibly in combination with other factors) can contribute to annoyance, and
that such exposure does not lead to hearing loss. However, for many other
adverse health effects investigated, there was inadequate evidence to reach
conclusions concerning the absence or presence of a causal link.
The Panel stresses that, given the nature of the sound produced by wind
turbines and the limited quality of available evidence (small sample sizes, small
number of studies available, lack of comprehensive exposure measurement),
the health impacts of wind turbine noise cannot be comprehensively assessed
at this time. Furthermore, in noting the challenges of undertaking research on
health impacts caused by multiple factors (large cohort studies, longitudinal
studies, double-blind experiments), the Panel emphasizes that providing
high-quality evidence would require a major research effort.
126
Glossary
Glossary
127
Glossary
Absolute hearing threshold is the lowest sound pressure level at which
a person can hear different frequencies in a very quiet environment with
focused concentration.
Air conduction of sound is the conveyance of sound through the external
auditory canal and middle ear to the inner ear (Dorland, 2011).
Amplitude modulation refers to changes in sound pressure over time. It has
two characteristics: the difference between the highest and lowest sound pressure
levels (modulation depth), and the frequency range in which variations in sound
pressure occur (modulation frequency, not to be confused with the frequency
of the sound itself) (van den Berg & Bowdler, 2011). Amplitude modulation can
also be periodic, with changes in sound pressure occurring at regular intervals.
In the case of sound from wind turbines, the amplitude is modulated at the
same rate as the rotation of the blades (the blade pass frequency).
Association describes a relationship between two variables that makes them
statistically dependent. An association does not necessarily mean that one
variable is causing the other, as the association may be due to another, unknown
variable; it may also be unclear which variable is affecting the other.
Bone conduction of sound is the conveyance of sound to the inner ear through
the bones of the skull (Dorland, 2011).
Broadband see Frequency.
Case definition is a set of standard criteria for deciding whether a person has
a particular disease or health-related condition (Columbia University, n.d.).
Cut-in wind speed is the minimum wind speed for a wind turbine to
generate power.
Cut-out wind speed is a maximum wind speed for a wind turbine to
generate power.
128
dB(A), dB(C), dB(G) all refer to sound pressure level measurements that have been
adjusted according to a weighting filter, indicated by the letter in parentheses
following dB (decibel). For example, dB(A) or dBA indicates A-weighted
sound pressure levels. This weighting de-emphasizes sounds at low and very
high frequencies in a way that approximates the sensitivity of the human ear
to sound at low pressure levels (Leventhall, 2011). C-weighting de-emphasizes
low frequency sound less than A-weighting, whereas G-weighting emphasizes
infrasound more than other frequencies. See Section 3.1.1 for a more detailed
discussion of sound measurement and weighting filters.
Epidemiology is the science that studies the patterns of diseases and factors
associated with outcomes in a population.
Exposure is the condition of being subjected to something (e.g., sound) that
could have a detrimental effect. In this report, the type of exposure is always
mentioned (e.g., exposure to wind turbine sound).
Frequency of sound is the number of pressure waves per second, measured in
Hertz (Hz). Frequency is perceived as tone or pitch, with higher frequencies
having a higher tone: Middle C is 261.6 Hz. The human ear is most sensitive
in the 2 kHz to 5 kHz range, which is roughly the range of the highest octave
on a piano. Sound at a single frequency is called a pure tone, whereas sound
composed of a wide range of frequencies is described as broadband.
Health is a state of complete physical, mental and social well-being and not
merely the absence of disease or infirmity (WHO, 1946).
Infrasound is sound at frequencies below 20 Hz; when physiological effects are
discussed, a lower bound of 1 Hz is often used (ANSI/ASA S1.1, 2013). Infrasound
is audible at high sound pressure levels, typically above 70 to 100 dB, with higher
thresholds at lower frequencies (Watanabe & Mller, 1990).
Latency period is the time period between exposure and development of disease.
Loudness refers to a persons subjective sound perception or magnitude of
resulting sensation when a sound impinges on the ear. By contrast, sound pressure
is an objective measure of the energy in a sound wave. Sound and noise are often
measured by sound pressure level, but this is not always a reliable indicator
of loudness how these sounds are perceived by people since loudness
is affected not only by sound pressure but also by personal physiological and
psychological factors.
Glossary
129
Low-frequency sound, as used in this report, refers to sound at frequencies in

the range of 20 to 200 Hz (Persson Waye, 2005, 2011; Pedersen et al., 2007b). The
boundaries between infrasound, low-frequency sound, and other frequencies
are somewhat arbitrary, however, and vary in the literature depending on the
context and application.
Masked hearing threshold is the sound pressure level at which a sound can
be heard in the presence of competing sound signals.
Nameplate capacity: see Rated power.
Noise: Two different definitions of noise are used in this report one technical
definition and one that describes the perception of sound by an individual. In the
field of acoustics, noise is a competing acoustic signal that masks a sound of interest.
Noise also defines an unpleasant or unwanted sound that causes disturbance.
The perception of noise depends on the individual and the context (i.e., the
same sound may be considered noise by one individual and not by another).
Noise limit is a maximum sound power level (at the source) or sound pressure
level (at a receiver) recommended or regulated to prevent noise-related
problems in an environment.
Rated power or nameplate capacity is the maximum power output of a wind
turbine generator.
Rated wind speed is the lowest speed at which a wind turbine will generate
the maximum power (the rated power).
Receiver: A location at which sound is measured or detected. A receiver can
also refer to a recording instrument measuring sound, or a person exposed to
sound at a given location. A receiver is often exposed to sound from multiple
sources in the environment, although one source (e.g., one or more wind
turbines) is often of particular interest.
Selection bias is a statistical bias that may arise from the method used to select
the subjects of a study; if not corrected for during analysis, selection bias may
lead to incorrect conclusions. Examples of selection bias include self-selection
of subjects or non-random selection of subjects for a study. In these cases, the
subjects may have different characteristics from those not selected (age and
income are common ones), which may affect outcomes of the study. Another
example is attrition bias, in which subjects who do not complete the study may
have poorer outcomes that are not included in the analysis.
130
Setback, in land use, is the distance that a building or structure is set back from
any place that is deemed to need protection; in the case of wind turbines, it is
distance to the closest building or residence. Setbacks are commonly used to
limit noise exposure and other hazards from wind turbines.
Sound is a pressure wave (oscillation) that travels through a medium such
as air (ANSI/ASA S1.1, 2013). Sound can also refer to an auditory sensation
evoked by the pressure wave.
Sound power level (LW) is a measure of sound produced at a source, in
contrast to the sound pressure level, which is measured at a receiver location. Both
are measured in decibels (dB), which represent power relative to a reference
value, on a logarithmic scale. The reference level (0 dB) for sound power is
1 10-12 W. Sound power is a characteristic of the source, used to predict the
propagation of sound in an environment (Crocker, 2007; Leventhall, 2011).
Sound pressure level (sound level; acoustic pressure; SPL; Lp) is an objective
measure of the amplitude of sound: the magnitude of fluctuations in pressure
around atmospheric pressure. The value at a given location depends on the
characteristics of the source, and the transmission of sound between the source
and the receiver, where the sound pressure level is measured. Sound pressure
levels are measured in decibels (dB), which represent pressure fluctuations in
Pascals (Pa), relative to a reference value, on a logarithmic scale. The reference
level (0 dB) for sound pressure is 20 Pa (2 10-5 Pa), which is close to the
human hearing threshold at a frequency of 1 kHz.
Stress is defined as conditions of a physical, biological, or psychological nature
that strain the adaptive capacity of a person up to or beyond his or her limits
(Welford, 1974; Gemmert & Van Galen, 1997).
Superior canal dehiscence is a health condition resulting from a loss of
bone that covers the superior canal of the vestibular system. It may allow
communication between the superior canal and the cranial cavity, which can
cause loss of air-conducted hearing in low and middle frequencies ( 2,000 Hz);
in some cases it can also increase the threshold of bone-conducted hearing,
which may increase sound sensitivity.
Glossary
131
Wind energy facilities (e.g., wind energy projects, wind farms) consist of
multiple wind turbines in a defined geographic area, owned and operated to
generate electricity at utility scales, and connected to the main electrical grid.
Wind turbines, as used in this report, refer to modern utility-scale devices that
generate electricity from the kinetic energy of the wind. These typically consist
of three blades, each roughly 40 metres long or more, attached to a rotor and
horizontal nacelle, mounted on a tower roughly 80 metres tall or more, with
the blades spinning in front (upwind) of the tower. The focus of this report
is onshore turbines connected to an electricity grid; offshore wind turbines,
or smaller devices used to provide electricity to a single building, are outside
the scope of this report.
132
References
References
133
References
Alimohammadi, I., Sandrock, S., & Gohari, M. R. (2013). The effects of low
frequency noise on mental performance and annoyance. Environmental
Monitoring and Assessment, 185, 7043-7051. doi: 10.1007/s10661-013-3084-8.
Alves-Pereira, M. & Branco, N. A. A. C. (2007). Vibroacoustic disease:
Biological effects of infrasound and low-frequency noise explained by
mechanotransduction cellular signalling. Progress in Biophysics and Molecular
Biology, 93, 256-279. doi: 10.1016/j.pbiomolbio.2006.07.011.
Ambrose, S. E., Rand, R. W., & Krogh, C. M. E. (2012). Wind turbine acoustic
investigation: Infrasound and low-frequency noiseA case study. Bulletin of
Science, Technology & Society, 32, 128-141. doi: 10.1177/0270467612455734.
ANSI/ASA S1.1 (The American National Standards Institute/The Acoustical
Society of America). (2013). Acoustical Terminology. Melville (NY): The
American National Standards Institute, Inc. Accredited Standards
Committee S1, Acoustics.
ASHA (American Speech-Language-Hearing Association). (1981). Degree of
Hearing Loss. Retrieved November 2014, from http://www.asha.org/
public/hearing/Degree-of-Hearing-Loss/.
AUC (Alberta Utilities Commission). (2013). Rule 12: Noise Control. Calgary
(AB): AUC.
Babisch, W. (2002). The noise/stress concept, risk assessment and research
needs. Noise & Health, 4(16), 1-11.
Babisch, W., Ising, H., & Gallacher, J. E. J. (2003). Health status as a potential
effect modifier of the relation between noise annoyance and incidence of
ischaemic heart disease. Occupational and Environmental Medicine, 60(10),
739-745. doi: 10.1136/oem.60.10.739.
Babisch, W. (2008). Road traffic noise and cardiovascular risk. Noise & Health,
10(38), 27-33.
Babisch, W. (2011). Cardiovascular effects of noise. Noise & Health, 13(52), 201.
Babisch, W., Pershagen, G., Selander, J., Houthuijs, D., Breugelmans, O.,
Cadum, E., ... Hansell, A. L. (2013). Noise annoyanceA modifier of the
association between noise level and cardiovascular health? Science of the
Total Environment, 452-453(C), 50-57. doi: 10.1016/j.scitotenv.2013.02.034.
Baguley, D. M. (2003). Hyperacusis. Journal of the Royal Society of Medicine,
96(12), 582-585.
Bakker, R. H., Pedersen, E., van den Berg, G. P., Stewart, R. E., Lok, W., &
Bouma, J. (2012). Impact of wind turbine sound on annoyance, selfreported sleep disturbance and psychological distress. Science of the Total
Environment, 425, 42-51. doi: 10.1016/j.scitotenv.2012.03.005.
Baqtasch, M., Van Dam, J., Sndergaard, B., & Rogers, A. (2006). Wind turbine
noiseAn overview. Canadian Acoustics, 34(2), 7-15.
134
Barone, M. (2011). Survey of Techniques for Reduction of Wind Turbine Blade Trailing
Edge Noise. Albuquerque (NM): Sandia National Laboratories, operated
for the United States Department of Energy by Sandia Corporation.
Barregard, L., Bonde, E., & hrstrm, E. (2009). Risk of hypertension from
exposure to road traffic noise in a population-based sample. Occupational
and Environmental Medicine, 66(6), 410-415. doi: 10.1136/oem.2008.042804.
Basner, M., Mller, U., & Elmenhorst, E. M. (2011). Single and combined
effects of air, road, and rail traffic noise on sleep and recuperation. Sleep,
34(1), 11-23.
Basner, M., Babisch, W., Davis, A., Brink, M., Clark, C., Janssen, S., & Stansfeld,
S. (2014). Auditory and non-auditory effects of noise on health. The Lancet,
383(9925), 1325-1332. doi: 10.1016/S0140-6736(13)61613-X.
Bauer, L. (nd). Welcome to wind-turbine-models.com. Retrieved June 2014,
from http://en.wind-turbine-models.com/.
Berger, E. H., Kieper, R. W., & Gauger, D. (2003). Hearing protection: Surpassing
the limits to attenuation imposed by the bone-conduction pathways. The
Journal of the Acoustical Society of America, 114(4), 1955-1967.
Berglund, B., Hassmn, P., & Job, R. S. (1996). Sources and effects of lowfrequency noise. The Journal of the Acoustical Society of America, 99, 2985-3002.
Bockstael, A., Dekoninck, L., Can, A., Oldoni, D., De Coensel, B., & Botteldooren,
D. (2012). Reduction of wind turbine noise annoyance: An operational
approach. Acta Acustica united with Acustica, 98, 392-401. doi: 10.3813/
AAA.918524.
Bockstael, A., Van Renterghem, T., De Weirt, V., & Botteldooren, D. (2013).
Exploring Underlying Mechanisms for Human Response to Wind Turbine Noise.
Paper presented at INTER-NOISE 2013, the 42nd International Congress
and Exposition on Noise Control Engineering, Innsbruck, Austria.
Bolin, K., Bluhm, G., Eriksson, G., & Nilsson, M. E. (2011). Infrasound
and low frequency noise from wind turbines: Exposure and health
effects. Environmental Research Letters, 6(3), 1-6. doi: 10.1088/17489326/6/3/035103.
Bolin, K., Kedhammar, A., & Nilsson, M. E. (2012). The Influence of Background
Sounds on Loudness and Annoyance of Wind Turbine Noise. Acta Acustica
united with Acustica, 98(5), 741-748. doi: 10.3813/aaa.918555.
Bowdler, D. (2008). Amplitude modulation of wind turbine noise. A review of
the evidence. Institute of Acoustics Bulletin, 33(4).
Bradford Hill, A. (1965). The environment and disease: Association or causation?
Proceedings of the Royal Society of Medicine, 58, 295-300.
Bradley, J. S. (1994). Annoyance caused by constant-amplitude and amplitudemodulated sounds containing rumble. Noise Control Engineering Journal,
42(6), 203-208.
References
135
Brisson, G., Gervais, M.-C., Martin, R., Blackburn, D., Chagnon, M., Martel, K., ...
Tardiff, I. (2013). oliennes et Sant Publique: Synthse des Connaissances
Mise Jour. Quebec (QC): Institut national de Sant Publique du Qubec,
Direction de la sant environnementale et de la toxicologie.
Broadbent, D. E. (1971). Decision and Stress. London, United Kingdom: Academic
Press Inc.
Bryant, P. & Bradley, L. (1985). Childrens Reading Problems: Psychology and
Education. Oxford, United Kingdom: Wiley-Blackwell.
Bullmore, A. (2011). Chapter 3. Sound Propagation from Wind Turbines. In
D. Bowdler & G. Leventhall (Eds.), Wind Turbine Noise. Essex, United
Kingdom: Multi-Science Publishing Company, Ltd.
Burton, T., Jenkins, N., Sharpe, D., & Bossanyi, E. (2011). Wind Turbine
Installations and Wind Farms. In T. Burton, N. Jenkins, D. Sharpe &
E. Bossanyi (Eds.), Wind Energy Handbook (2 ed.). West Sussex, United
Kingdom: John Wiley & Sons, Ltd.
CanWEA (Canadian Wind Energy Association). (2008). Wind Vision 2025:
Summary. Ottawa (ON): CanWEA.
CanWEA (Canadian Wind Energy Association). (2011). Federal / Provincial
Initiatives on Wind Energy. Ottawa (ON): CanWEA.
CanWEA (Canadian Wind Energy Association). (2014). Why Wind Works.
Retrieved January 2015, from http://windfacts.ca/why-wind-works.
CanWEA (Canadian Wind Energy Association). (n.d.). Best Practices for Community
Engagement and Public Consultation. Ottawa (ON): CanWEA.
Chapman, S., St George, A., Waller, K., & Cakic, V. (2013). The pattern of
complaints about Australian wind farms does not match the establishment
and distribution of turbines: Support for the psychogenic, communicated
disease hypothesis. PLOS ONE, 8(10), e76584. doi: 10.1371/journal.
pone.0076584.t002.
Ciliska, D., Thomas, H., & Buffett, C. (2008). An Introduction to Evidence-Informed
Public Health and A Compendium of Critical Appraisal Tools for Public Health
Practice. Hamilton (ON): National Collaborating Centre for Methods
and Tools.
Clark, C. & Stansfeld, S. A. (2007). The effect of transportation noise on health
and cognitive development: A review of recent evidence. International
Journal of Comparative Psychology, 20(2).
Clark, C. & Srqvist, P. (2012). A 3 year update on the influence of noise on
performance and behavior. Noise & Health, 14(61), 292-296.
Colby, W. D., Dobie, R., Leventhall, G., Lipscomb, D. M., McCunney, R. J., Seilo,
M. T., & Sndergaard, B. (2009). Wind Turbine Sound and Health Effects:
An Expert Panel Review (AWEA and CWEA). Ottawa (ON): Canadian Wind
Energy Association.
136
Colebatch, J. G. & Halmagyi, G. M. (1992). Vestibular evoked potentials in

human neck muscles before and after unilateral vestibular deafferentation.
Neurology, 42(8), 1635-1636.
Colebatch, J. G., Halmagyi, G. M., & Skuse, N. F. (1994). Myogenic potentials
generated by a click-evoked vestibulocollic reflex. Journal of Neurology,
Neurosurgery, and Psychiatry 57(2), 190-197.
Columbia University. (n.d.). Glossary of epidemiological terms. Retrieved
January 2015, from http://www.cs.columbia.edu/digigov/LEXING/
CDCEPI/gloss.html.
CoM (Commonwealth of Massachusetts). (2014). MassDEP Wind Turbine Noise
Technical Advisory Group. Retrieved October 2014, from http://www.
mass.gov/eea/agencies/massdep/news/advisory-committees/wntag.html.
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 Psychology, 33(12), 1588-1592. doi: 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 Psychology, 33(4), 360-364. doi: 10.1037/a0031760.
Crocker, M. J. (2007). Handbook of Noise and Vibration Control. Hoboken (NJ):
John Wiley & Sons.
Croy, I., Smith, M. G., & Waye, K. P. (2013). Effects of train noise and vibration
on human heart rate during sleep: An experimental study. BMJ Open, 3, 1-8.
CSE (Centre for Sustainable Energy). (2009). Delivering Community Benefits
from Wind Energy Development: A Toolkit. London, United Kingdom: CSE.
Curthoys, I. S., Kim, J., McPhedran, S. K., & Camp, A. J. (2006). Bone conducted
vibration selectively activates irregular primary otolithic vestibular neurons
in the guinea pig. Experimental Brain Research, 175(2), 256-267.
Danielsson, A. & Landstrom, U. (1985). Blood pressure changes in man during
infrasonic exposure. An experimental study. Acta Medica Scandinavica,
217(5), 531-535.
DEHLG (Department of Environment, Heritage and Local Government).
(2006). Wind Energy Development. Dublin, Ireland: DEHLG.
Diamond, A. (2002). Normal Development of Prefrontal Cortex from Birth to
Young Adulthood: Cognitive Functions, Anatomy, and Biochemistry. In
D. T. Stuss & R. T. Knight (Eds.), Principles of Frontal Lobe Function. New
York (NY): Oxford University Press.
DME (Danish Ministry of the Environment). (2011). Statutory Order on Noise
from Wind Turbines. Copenhagen, Denmark: DME.
Dorland, W. A. N. (2011). Dorlands Illustrated Medical Dictionary (32 ed.).
Philadelphia (PA): Elsevier Health Sciences.
References
137
Eberhardt, J. L., Strle, L. O., & Berlin, M. H. (1987). The influence of continuous
and intermittent traffic noise on sleep. Journal of Sound and Vibration,
116(3), 445-464.
Eggermont, J. J. (2005). Tinnitus: Neurobiological substrates. Drug Discovery
Today, 10(19), 1283-1290.
Ellenbogen, J. M., Grace, S., Heiger-Bernays, W. J., Manwell, J. F., Mills, D. A.,
Sullivan, K. A., & Weisskopf, M. G. (2012). Wind Turbine Health Impact
Study: Report of Independent Expert Panel. Prepared for Massachusetts
Department of Environmental Protection, Massachusetts Department
of Public Health.
EPA (United States Environmental Protection Agency). (2003). A Summary of
General Assessment Factors for Evaluating the Quality of Scientific and Technical
Information. Washington (DC): U.S. Environmental Protection Agency.
EPA (Ireland Environmental Protection Agency). (2011). Guidance Note on Noise
Assessment of Wind Turbine Operations at EPA Licensed Sites (NG3). Wexford,
Ireland: Environmental Protection Agency Ireland.
EPHC (Environment Protection and Heritage Council). (2010). National Wind
Farm Development Guidelines Draft. Adelaide, Australia: EPHC.
Erdogan, N., Songu, M., Akay, E., Mete, B. D., Uluc, E., Ona, l. K., & Oyar, O.
(2011). Posterior semicircular canal dehiscence in asymptomatic ears.
Acta Oto-Laryngologica 131(1), 4-8. doi: 10.3109/00016489.2010.502184.
Eriksson, C., Hilding, A., Pyko, A., Bluhm, G., Pershagen, G., & stenson, C.-G.
(2014). Long-term aircraft noise exposure and body mass index, waist
circumference, and type 2 diabetes: A prospective study. Environmental
Health Perspectives, 122(7), 693.
Evans, G. W., Hygge, S., & Bullinger, M. (1995). Chronic noise and psychological
stress. Psychological Science, 6(6), 333-338. doi: 10.1111/j.1467-9280.1995.
tb00522.x.
Evans, G. W. & Maxwell, L. (1997). Chronic noise exposure and reading deficits:
The mediating effects of language acquisition. Environment and Behavior,
29(5), 638-656.
Evans, G. W., Bullinger, M., & Hygge, S. (1998). Chronic noise exposure and
physiological response: A prospective study of children living under
environmental stress. Psychological Science, 9(1), 75-77.
Evans, T. & Cooper, J. (2012). Comparison of predicted and measured wind
farm noise levels and implications for assessments of new wind farms.
Acoustics Australia, 40(1), 28-36.
EWEA (European Wind Energy Association). (2005). Prioritising Wind Energy Research:
Strategic Research Agenda of the Wind Energy Sector. Brussels, Belgium: EWEA.
Fastl, H. & Menzel, D. (2013). Psychoacoustic Aspects of Noise from Wind Turbines.
138
Ferguson-Martin, C. J. & Hill, S. D. (2011). Accounting for variation in wind

deployment between Canadian provinces. Energy Policy, 39(3), 1647-1658.
Findeis, H. & Peters, E. (2004). Disturbing effects of low frequency sound
immissions and vibrations in residential buildings. Noise & Health, 6(23),
29-35.
Fortin, P., Rideout, K., Copes, R., & Bos, C. (2013). Wind Turbines and Health.
Vancouver (BC): National Collaborating Centre for Environmental Health.
Frey, B. J. & Hadden, P. J. (2012). Wind Turbines and Proximity to Homes:
The Impact of Wind Turbine Noise on Health.
Fruhstorfer, B., Fruhstorfer, H., Grass, P., Milerski, H. G., Sturm, G., Wesemann,
W., & Wiesel, D. (1985). Daytime noise stress and subsequent night sleep:
Interference with sleep patterns, endocrine and neurocrine functions.
International Journal of Neuroscience, 26(3-4), 301-310.
GE (General Electric Company). (2004). 1.5sle/1.5sl/1.5s/1.5se Wind Turbines.
Amsterdam, The Netherlands: General Electric.
GoC (Government of Canada). (2004). Radiation Emitting Devices Act (R.S.C.,
1985, c. R-1). Retrieved November 2014, from http://laws-lois.justice.gc.ca/
eng/acts/R-1/page-1.html.
GreyNet International. (2014). Grey Literature Network Service. Retrieved
September 2014, from http://www.greynet.org/home/aboutgreynet.html.
Griefahn, B. (2002). Sleep disturbances related to environmental noise.
Noise & Health, 4(15), 57-60.
GWEC (Global Wind Energy Council). (2014). Global Wind Report: Annual
Market Update 2013. Brussels, Belgium: GWEC.
Hagiwara, M., Shaikh, J. A., Fang, Y., Fatterpekar, G., & Roehm, P. C. (2012).
Prevalence of radiographic semicircular canal dehiscence in very young
children: An evaluation using high-resolution computed tomography of
the temporal bones. Pediatric Radiology 42(12), 1456-1464. doi: 10.1007/
s00247-012-2489-9.
Haines, M. M., Stansfeld, S. A., Job, R. S., Berglund, B., & Head, J. (2001). Chronic
aircraft noise exposure, stress responses, mental health and cognitive
performance in school children. Psychological Medicine, 31(2), 265-277.
Hansen, M. O. & Stinson, M. R. (1998). Air conducted and body conducted
sound produced by own voice. Canadian Acoustics, 26(2), 11-19.
Hardoy, M. C., Carta, M. G., Marci, A. R., Carbone, F., Cadeddu, M., Kovess, V., ...
Carpiniello, B. (2005). Exposure to aircraft noise and risk of psychiatric
disorders: The Elmas surveyaircraft noise and psychiatric disorders.
Social Psychiatry and Psychiatric Epidemiology, 40(1), 24-26.
Harrison, R. V. (2014). On the biological plausibility of wind turbine
syndrome. International Journal of Environmental Health Research, 8, 1-6.
doi: 10.1080/09603123.2014.963034.
References
139
Hau, E. (2006). Wind Turbines Fundamentals, Technologies, Application, Economics

(2nd ed.). New York (NY): Springer.
Haugen, K. M. B. (2011). International Review of Policies and Recommendations
for Wind Turbine Setbacks from Residences: Setbacks, Noise, Shadow Flicker, and
Other Concerns. Saint Paul (MN): Minnesota Department of Commerce:
Energy Facility Permitting.
HCN (Health Council of the Netherlands). (1994). Committee Noise and Health:
Noise and Health. The Hague, Netherlands: HCN.
Health Canada. (2003a). Chronic Disease Surveillance in Canada: A Background
Paper. Ottawa (ON): Population and Public Health Branch, Health Canada.
Health Canada. (2003b). Learning from SARS. Ottawa (ON): Health Canada.
Health Canada. (2011). Canada Health ActAnnual Report 2010-2011. Ottawa
(ON): Health Canada.
Health Canada. (2012). Health Canada Announces Wind Turbine Noise and
Health Study. Retrieved July 2014, from http://www.hc-sc.gc.ca/ahc-asc/
media/nr-cp/_2012/2012-109-eng.php.
Health Canada. (2013a). Health Canada Wind Turbine Noise and Health Study
Design Consultation. Retrieved August 2014, from http://www.hc-sc.
gc.ca/ewh-semt/consult/_2013/wind_turbine-eoliennes/index-eng.php.
Health Canada. (2013b). Health Canada Wind Turbine Noise and Health Study
Design Consultation. Retrieved August 2014, from http://www.hc-sc.
gc.ca/ewh-semt/consult/_2013/wind_turbine-eoliennes/index-eng.php.
Health Canada. (2014a). Wind Turbine Noise and Health Study: Summary of
Results. Retrieved November 2014, from http://www.hc-sc.gc.ca/ewhsemt/noise-bruit/turbine-eoliennes/summary-resume-eng.php.
Health Canada. (2014b). Health Impacts and Exposure to Sound From Wind
Turbines: Updated Research Design and Sound Exposure Assessment.
Retrieved November 2014, from http://www.hc-sc.gc.ca/ewh-semt/
consult/_2013/wind_turbine-eoliennes/research_recherche-eng.php.
Henry, J. A., Dennis, K. C., & Schechter, M. A. (2005). General review of
tinnitus: Prevalence, mechanisms, effects, and management. Journal of
Speech, Language & Hearing Research, 48(5), 1204-1235.
Hensel, J., Scholz, G., Hurttig, U., Mrowinski, D., & Janssen, T. (2007). Impact
of infrasound on the human cochlea. Hearing Research, 233(1), 67-76.
Hepburn, H. G. (2006). Acoustic and geophysical measurement of infrasound
from wind farm turbines. Canadian Acoustics, 34(2), 51.
Hessler, D. (2011). Chapter 7. Measuring and Analyzing Wind Turbine Noise.
In D. Bowdler & G. Leventhall (Eds.), Wind Turbine Noise. Essex, United
HGC Engineering. (2010). Low Frequency Noise and Infrasound Associated with
Wind Turbine Generator Systems: A Literature Review. Toronto (ON): Ontario
Ministry of the Environment.
140
Hockey, G. R. J. (1984). Varieties of Attentional State: The Effects of Environment.

In R. Parasuraman & D. R. Davies (Eds.), Varieties of Attention. New York
(NY): Academic Press.
Hodgetts, M. & OConnor, K. (2013). Wind Turbines and Health: Summary of
a Scoping Review. Kingston (ON): Research and Evaluation Division of
Kingston, Frontenac and Lennox & Addington Public Health.
Howick, J., Glasziou, P., & Aronson, J. K. (2009). The evolution of evidence
hierarchies: What can Bradford Hills guidelines for causation contribute?
Journal of the Royal Society of Medicine, 102(5), 186-194. doi: 10.1258/
jrsm.2009.090020.
Hubbard, H. H. (1982). Noise induced house vibrations and human perception.
Noise Control Engineering Journal, 19(2), 49-55.
Hughes, R. & Jones, D. M. (2001). The intrusiveness of sound: Laboratory
findings and their implications for noise abatement. Noise & Health,
4(13), 51-70.
IARC (International Agency for Research on Cancer, World Health Organization).
(2006). IARC Monographs on the Evaluation of Carcinogenic Risks to Humans:
Preamble. Lyon, France: IARC.
Ising, H., Babisch, W., & Kruppa, B. (1999). Noise-induced endocrine effects
and cardiovascular risk. Noise & Health, 1(4), 37-48.
Jakobsen, J. (2001). Danish guidelines on environmental low frequency noise,
infrasound and vibration. Journal of Low Frequency Noise, Vibration and
Active Control, 20(3), 141-148.
Jakobsen, J. (2005). Infrasound emission from wind turbines. Journal of Low
Frequency Noise, Vibration and Active Control, 24(3), 145-155.
Jami, A. A. N. & Walsh, P. R. (2014). The role of public participation in identifying
stakeholder synergies in wind power project development: The case
study of Ontario, Canada. Renewable Energy, 68, 194-202. doi: 10.1016/
j.renene.2014.02.004.
Janssen, S. A., Vos, H., Eisses, A. R., & Pedersen, E. (2011). A comparison
between exposure-response relationships for wind turbine annoyance
and annoyance due to other noise sources. The Journal of the Acoustical
Society of America, 130, 3746-3753. doi: 10.1121/1.3653984.
Jeffery, R. D., Krogh, C., & Horner, B. (2013). Adverse health effects of industrial
wind turbines. Canadian Family Physician, 59, 473-475.
Kaliski, K. (2014). Wind Turbine Research Study Interim Modeling Report for WNTAG.
White River Junction (VT): RSG, Inc.
Kalveram, K. T., Dassow, J., & Vogt, J. (1999). How information about the source
influences noise annoyance. The Journal of the Acoustical Society of America,
105, 942. doi: 10.1121/1.425715.
References
141
Kantarelis, C. & Walker, J. (1988). The identification and subjective effect of

amplitude modulation in diesel engine exhaust noise. Journal of Sound
and Vibration, 120(2), 297-3032. doi: 10.1016/0022-460X(88)90438-5.
Ksin, J. I., Kjellevand, T. O., Kjekshus, J., Nesheim, G. B., & Wagstaff, A.
(2012). CT examination of the pericardium and lungs in helicopter
pilots exposed to vibration and noise. Aviation, Space, and Environmental
Medicine, 83(9), 858-864.
Kjiellberg, A. & Wide, P. (1988). Effects of Simulated Ventilation Noise on Performance
of Grammatical Reasoning Task. Paper presented at Noise as a Public Health
Problem, Stockholm, Sweden.
Klatte, M., Bergstroem, K., & Lachmann, T. (2013). Does noise affect learning?
A short review on noise effects on cognitive performance in children.
Frontiers in Psychology, 4, 578. doi: 10.3389/fpsyg.2013.00578.
Krogh, C. M. E. (2011). Industrial wind turbine development and loss of
social justice? Bulletin of Science, Technology & Society, 31(4), 321-333. doi:
10.1177/0270467611412550.
Krogh, C. M. E., Gillis, L., Kouwen, N., & Aramini, J. (2011). WindVOiCe, a
self-reporting survey: Adverse health effects, industrial wind turbines,
and the need for vigilance monitoring. Bulletin of Science, Technology &
Society, 31(4), 334-345. doi: 10.1177/0270467611412551.
Kurakata, K. & Mizunami, T. (2008). The statistical distribution of normal
hearing thresholds for low-frequency tones. Low Frequency Noise, Vibration
and Active Control, 27(2), 97-104.
Kuwano, S., Yano, T., Kageyama, T., Sueoka, S., & Tachibana, H. (2013). Social
Survey on Community Response to Wind Turbine Noise. Paper presented at
INTER-NOISE 2013, the 42nd International Congress and Exposition on
Noise Control Engineering, Innsbruck, Austria.
Lane, J. (2013). Association Between Industrial Wind Turbine Noise and Sleep Quality
in a Comparison Sample of Rural Ontarians. Waterloo (ON): University of
Waterloo.
Lazarus, R. S. & Folkman, S. (1984). Stress, Appraisal and Coping. New York (NY):
Springer Publishing Company.
Lee, S., Kim, K., Choi, W., & Lee, S. (2011). Annoyance caused by amplitude
modulation of wind turbine noise. Noise Control Engineering Journal, 59(1),
38-46. doi: 10.3397/1.3531797.
Leventhall, G. (2009). Low Frequency Noise. What we know, what we do not
know, and what we would like to know. Low Frequency Noise, Vibration and
Active Control, 28(2), 79-104. doi: 10.1260/0263-0923.28.2.79.
Leventhall, G. (2011). Chapter 1. Basic Acoustics. In D. Bowdler & G. Leventhall
(Eds.), Wind Turbine Noise. Essex, United Kingdom: Multi-Science Publishing
Company, Ltd.
142
Madsen, K. D. & Pedersen, T. H. (2010). EFP-06 Project: Low Frequency Noise from
Large Wind Turbines (www.madebydelta.com search for EFP-06). Hrsholm,
Denmark: Delta, performed for the Danish Energy Authority.
Maffei, L., Iachini, T., Masullo, M., Aletta, F., Sorrentino, F., Senese, V. P., &
Ruotolo, F. (2013). The effects of vision-related aspects on noise perception
of wind turbines in quiet areas. International Journal of Environmental
Research and Public Health, 10(5), 1681-1697. doi: 10.3390/ijerph10051681.
Magari, S. R., Smith, C. E., Schiff, M., & Rohr, A. C. (2014). Evaluation of
community response to wind turbine-related noise in Western New York
State. Noise & Health, 16(71), 228.
Manwell, J. F., McGowan, J. G., & Rogers, A. L. (2010). Wind Energy Explained:
Theory, Design And Application (2nd ed.). Chichester, United Kingdom:
John Wiley & Sons Ltd.
Maschke, C. & Hecht, K. (2004). Stress hormones and sleep disturbances electrophysiological and hormonal aspects. Noise & Health, 6(22), 49-54.
Masotti, P. & Hodgetts, M. (2011). Wind Turbines and Health: A Modified Scoping
Review. Kingston (ON): Kingston, Frontenac, Lennox & Addington
(KFL&A) Public Health.
Mathias, B., Uwe, M., Eva-Maria, E., Gtz, K., & Barbara, G. (2008). Aircraft
noise effects on sleep: A systematic comparison of EEG awakenings and
automatically detected cardiac activations. Physiological Measurement,
29(9), 1089.
McBride, D., Shepherd, D., Welch, D., & Dirks, K. N. (2013). A Longitudinal
Study of the Impact of Wind Turbine Proximity on Health Related Quality of Life.
MDDELCC (Ministre du Dveloppement durable, Environnement et Lutte
contre les changements climatiques). (2006). Traitement des plaintes sur le
bruit et exigences aux entreprises qui le gnrent. Qubec (QC): MDDELCC.
Merlin, T., Newton, S., Ellery, B., Milverton, J., & Farah, C. (2013). Systematic
Review of the Human Health Effects of Wind Farms. Adelaide, Australia:
National Health and Medical Research Council (NHMRC).
Miedema, H. M. & Oudshoorn, C. G. (2001). Annoyance from transportation
noise: Relationships with exposure metrics DNL and DENL and their
confidence intervals. Environmental Health Perspectives, 109(4), 409-419.
Minor, L. B. (2000). Superior canal dehiscence syndrome. American Journal of
Otology, 21(1), 9-19.
Minor, L. B., Solomon, D., Zinreich, J. S., & Zee, D. S. (1998). Sound- and/
or pressure-induced vertigo due to bone dehiscence of the superior
semicircular canal. Archives of Otolaryngology - Head and Neck Surgery
124(3), 249-258.
References
143
MOE (Ministry of Environment). (2008). Noise Guidelines for Wind Farms.

Toronto (ON): MOE.
Mller, H. & Pedersen, C. S. (2011). Low-frequency noise from large wind
turbines. The Journal of the Acoustical Society of America, 129, 3727-3744.
doi: 10.1121/1.3543957.
Mller, H., Pedersen, C. S., & Pedersen, S. (2011). Lavfrekvent stj fra store
vindmller-opdateret 2011: Acoustics, Department of Electronic Systems,
Aalborg University.
Mroczek, B., Kurpas, D., & Karakiewicz, B. (2012). Influence of distances between
places of residence and wind farms on the quality of life in nearby areas.
Annals of Agricultural and Environmental Medicine, 19(4), 692-696.
Mulvaney, K. K., Woodson, P., & Prokopy, L. S. (2013). Different shades of green:
A case study of support for wind farms in the rural midwest. Environmental
Management, 51, 1012-1024. doi: 10.1007/s00267-013-0026-8.
Mnzel, T., Gori, T., Babisch, W., & Basner, M. (2014). Cardiovascular effects
of environmental noise exposure. European Heart Journal. doi: 10.1093/
eurheartj/ehu030.
Muzet, A. (2007). Environmental noise, sleep and health. Sleep Medicine Reviews,
11(2), 135-142. doi: 10.1016/j.smrv.2006.09.001.
Nagai, N., Matsumoto, M., Yamasumi, Y., Shiraishi, T., Nishimura, K., Matsumoto,
K., ... Takeda, S. (1989). Process and emergence on the effects of infrasonic
and low frequency noise on inhabitants. Journal of Low Frequency Noise
and Vibration, 8, 87-99.
Ndrepepa, A. & Twardella, D. (2011). Relationship between noise annoyance
from road traffic noise and cardiovascular diseases: A meta-analysis.
Noise & Health, 13(52), 251.
NEB (National Energy Board). (2013). Canadas Energy Future 2013: Energy
Supply and Demand Projections to 2035. Calgary (AB): NEB.
Nelson, D. I., Nelson, R. Y., Concha-Barrientos, M., & Fingerhut, M. (2005).
The global burden of occupational noise-induced hearing loss. American
Journal of Industrial Medicine, 48(6), 446-458.
NHMRC (National Health and Medical Research Council). (2010). Wind
Turbines and Health: A Rapid Review of the Evidence. Canberra, Australia:
National Health and Medical Research Council.
NHMRC (National Health and Medical Research Council). (2014). NHMRC
Draft Information Paper: Evidence on Wind Farms and Human Health. Canberra,
Australia: NHMRC.
Nissenbaum, M. A., Aramini, J. J., & Hanning, C. D. (2012). Effects of industrial
wind turbine noise on sleep and health. Noise & Health, 14, 237-243.
NSF (National Sleep Foundation). (2013). Bedroom Poll: Summary of Findings.
Washington (DC): NSF.
144
Oerlemans, S. & Mndez Lpez, M. (2005). Acoustic Array Measurements on a Full

Scale Wind Turbine. Madrid, Spain: National Aerospace Laboratory NLR.
Oerlemans, S., Sijtsma, P., & Mndez Lpez, B. (2007). Location and
quantification of noise sources on a wind turbine. Journal of Sound and
Vibration, 299(45), 869-883.
Oerlemans, S. (2009). Detection of Aeroacoustic Sound Sources on Aircraft and Wind
Turbines. Thesis. Zwolle, The Netherlands: University of Twente, Enschede.
Oerlemans, S. & Schepers, J. (2009). Prediction of wind turbine noise and validation
against experiment. International Journal of Aeroacoustics, 8(6), 555-584.
Oerlemans, S. (2011). Chapter 2. Primary Noise Sources. In D. Bowdler &
G. Leventhall (Eds.), Wind Turbine Noise. Essex, United Kingdom:
Multi-Science Publishing Company, Ltd.
hrstrm, E. (1989). Sleep distrubance, psycho-social and medial symptoms:
A pilot survey among persons exposed to high-levels of road traffice noise.
Journal of Sound and Vibration 133(1), 117-128.
hrstrm, E. & Rylander, R. (1982). Sleep disturbance effects of traffic noise
A laboratory study on after effects. Journal of Sound and Vibration, 84(1), 87-103.
Ollson, C. A., Knopper, L. D., McCallum, L. C., & Whitfield-Aslund, M. L.
(2013). Letter to Editor: Are the findings of Effects of industrial wind
turbine noise on sleep and health supported? Noise & Health, 15, 148-151.
Olson, H. F. (1972). The measurement of loudness. Audio Magazine, 18-22.
Ontario Regulation. (2012). Environmental Protection Act. Ontario Regulation
359/09.
Palmer, K., Griffin, M., Syddall, H., Davis, A., Pannett, B., & Coggon, D. (2002).
Occupational exposure to noise and the attributable burden of hearing
difficulties in Great Britain. Occupational and environmental medicine,
59(9), 634-639.
Pasqualetti, M. J. (2002). Living with Wind Power in a Hostile Landscape. In
M. J. Pasqualetti, P. Gipe & R. W. Righter (Eds.), Wind Power in View:
Energy Landscapes in a Crowded World. San Diego (CA): Academic Press.
Passchier-Vermeer, W. & Passchier, W. F. (2006). Environmental Noise, Annoyance
and Sleep Disturbance. In P. Nicolopoulou-Stamati, L. Hens & V. Howard
(Eds.), Environmental Health Impacts of Transport and Mobility. Dordrecht,
The Netherlands: Springer Science & Business Media.
Pawlaczyk-uszczyska, M., Dudarewicz, A., Zaborowski, K., Zamojska, M., &
Waszkowska, M. (2013). Assessment of annoyance due to wind turbine
noise. Proceedings of Meetings on Acoustics, 19(1), 1-9.
Pawlaczyk-uszczyska, M., Dudarewicz, A., Zaborowski, K., Zamojska-Daniszewska,
M., & Waszkowska, M. (2014). Evaluation of annoyance from the wind
turbine noiseA pilot study. International journal of occupational medicine
and environmental health, 27(3), 364-388.
References
145
Pedersen, E. & Persson Waye, K. (2004). Perception and annoyance due to wind
turbine noise: A dose-response relationship. The Journal of the Acoustical
Society of America, 116, 3460-3470. doi: 10.1121/1.1815091.
Pedersen, E., Hallberg, L.-M., & Persson Waye, K. (2007a). Living in the vicinity
of wind turbines A grounded theory study. Qualitative Research in
Psychology, 4, 49-63. doi: 10.1080/14780880701473409.
Pedersen, E. & Persson Waye, K. (2007). Wind turbine noise, annoyance and
self-reported health and well-being in different living environments.
Occupational and Environmental Medicine, 64, 480-486. doi: 10.1136/
oem.2006.031039.
Pedersen, E. & Larsman, P. (2008). The impact of visual factors on noise
annoyance among people living in the vicinity of wind turbines. Journal
of Environmental Psychology, 28, 379-389.
Pedersen, E. & Persson Waye, K. (2008). Wind turbineslow level noise sources
interfering with restoration? Environmental Research Letters, 3, 015002. doi:
10.1088/1748-9326/3/1/015002.
Pedersen, E., van den Berg, F., Bakker, R., & Bouma, J. (2009). Response to
noise from modern wind farms in The Netherlands. The Journal of the
Acoustical Society of America, 126, 634-643. doi: 10.1121/1.3160293.
Pedersen, E., van den Berg, F., Bakker, R., & Bouma, J. (2010). Can road traffic
mask sound from wind turbines? Response to wind turbine sound at
different levels of road traffic sound. Energy Policy, 38(5), 2520-2527. doi:
10.1016/j.enpol.2010.01.001.
Pedersen, E. (2011). Health aspects associated with wind turbine noise Results
from three field studies. Noise Control Engineering Journal, 59, 47-53.
Pedersen, S., Mller, H., & Persson Waye, K. (2007b). Indoor Measurements
of Noise at Low Frequencies - Problems and Solutions. Journal of
Low Frequency Noise, Vibration and Active Control, 26(4), 249-270. doi:
10.1260/026309207783571389.
Perron, S., Ttreault, L.-F., King, N., Plante, C., & Smargiassi, A. (2012). Review
of the effect of aircraft noise on sleep disturbance in adults. Noise &
Health, 14(57), 58-67.
Persson Waye, K. & Bjrkman, M. (1988). Annoyance due to low frequency
noise and the use of the dB (A) scale. Journal of Sound and Vibration,
127(3), 491-497.
Persson Waye, K. & Rylander, R. (2001). The prevalence of annoyance and
effects after long-term exposure to low-frequency noise. Journal of Sound
and Vibration, 240(3), 483-497. doi: 10.1006/jsvi.2000.3251.
Persson Waye, K. & hrstrm, E. (2002). Psycho-acoustic characters of relevance
for annoyance of wind turbine noise. Journal of Sound and Vibration,
250(1), 65-73.
146
Persson Waye, K., Clow, A., Edwards, S., Hucklebridge, F., & Rylander, R.
(2003). Effects of nighttime low frequency noise on the cortisol response
to awakening and subjective sleep quality. Life Sciences, 72(8), 863-875.
Persson Waye, K. (2005). Adverse effects of moderate levels of low frequency noise
in the occupational environment. ASHRAE Transactions, 111(1), 672-683.
Persson Waye, K. (2009). Perception and Environmental Impact of Wind Turbine
Noise. Paper presented at INTER-NOISE and NOISE-CON Congress and
Conference Proceedings, Ottawa (ON).
Persson Waye, K. (2011). Effects of Low Frequency Noise and Vibrations:
Environmental and Occupational Perspectives. In J. Nriagu (Ed.),
Encyclopedia of Environmental Health (Vol. 2). Ann Arbor (MI): Burlington.
Pfammatter, A., Darrouzet, V., Grtner, M., Somers, T., Van Dinther, J.,
Trabalzini, F., ... Linder, T. A. (2010). Superior semicircular canal
dehiscence syndrome multicenter study: Is there an association between
size and symptoms? Ontology & Neurotology 31(3), 447-454. doi: 10.1097/
MAO.0b013e3181d27740.
PHAC (Public Health Agency of Canada). (2008). History. Retrieved September,
2014, from http://www.phac-aspc.gc.ca/about_apropos/history-eng.php.
PHAC (Public Health Agency of Canada). (2009). 2009 Tracking Heart
Disease and Stroke in Canada. Retrieved November 2014, from
http://www.phac-aspc.gc.ca/publicat/2009/cvd-avc/report-rapport-eng.php.
Pierpont, N. (2006). Wind Turbine Syndrome: Noise, Shadow Flicker, and Health.
Pierpont, N. (2009). Wind Turbine Syndrome: A Report on a Natural Experiment.
Santa Fe (NM): K-Selected Books.
RABDTI (Renewables Advisory Board and DTI). (2007). The Protocol for Public
Engagement with Proposed Wind Energy Developments in England. London, United
Kingdom: Centre for Sustainable Energy with BDOR and Peter Capener.
RenewableUK. (2013). Wind Turbine Amplitude Modulation: Research to Improve
Understanding as to its Cause and Effect. London, United Kingdom:
RenewableUK.
Rief, W., Barsky, A. J., Glombiewski, J. A., Nestoriuc, Y., Glaesmer, H., & Braehler,
E. (2011). Assessing general side effects in clinical trials: Reference data
from the general population. Pharmacoepidemiology and Drug Safety, 20(4),
405-415. doi: 10.1002/pds.2067.
Robert, G. & Hockey, J. (1997). Compensatory control in the regulation of human
performance under stress and high workload: A cognitive-energetical
framework. Biological Psychology, 45(1), 73-93.
Roberts, J. D. & Roberts, M. A. (2013). Wind turbines: Is there a human health
risk? Journal of Environmental Health, 75, 8-13, 16-17.
Robertson, D. D. I. & Ireland, D.J. (1995). Vestibular evoked myogenic potentials.
Journal of Otolaryngology 24(1), 3-8.
References
147
Rod, J. (2012). Health & Safety Impacts from Large-Scale Wind Turbines. Kentville
(NS): Municipality of the County of Kings.
Ruotolo, F., Senese, V. P., Ruggiero, G., Maffei, L., Masullo, M., & Iachini,
T. (2012). Individual reactions to a multisensory immersive virtual
environment: The impact of a wind farm on individuals. Cognitive Processing,
13 Suppl 1, S319-323. doi: 10.1007/s10339-012-0492-6.
Russo, J. E., Crowson, M. G., DeAngelo, E. J., Belden, C. J., & Saunders, J. E.
(2014). Posterior semicircular canal dehiscence: CT prevalence and
clinical symptoms. Ontology & Neurotology, 35(2), 310-314.
Rylander, R. (2004). Physiological aspects of noise-induced stress and annoyance.
Journal of Sound and Vibration, 277(3), 471-478.
Salt, A. N. & Hullar, T. E. (2010). Responses of the ear to low frequency sounds,
infrasound and wind turbines. Hearing Research, 268, 12-21. doi: 10.1016/
j.heares.2010.06.007.Responses.
Salt, A. N. & Kaltenbach, J. A. (2011). Infrasound from wind turbines could
affect humans. Bulletin of Science, Technology & Society, 31, 296-302. doi:
10.1177/0270467611412555.
Salt, A. N., Lichtenhan, J. T., Gill, R. M., & Hartsock, J. J. (2013). Large
endolymphatic potentials from low-frequency and infrasonic tones in the
guinea pig. The Journal of the Acoustical Society of America, 133(3), 1561-1571.
Schomer, P., Brown, A., De Coensel, B., Genuit, K., Gjestland, T., Jeon, J. Y., ...
Watts, G. (2010). On Efforts to Standardize a Graphical Description of the
Soundscape Concept. Paper presented at Proceedings of the International
Congress on Noise Control Engineering (internoise), Lisbon, Portugal.
Schulte-Fortkamp, B. (2014). Balancing the research approach on wind turbine
effects through improving psychological factors that affect community
response. The Journal of the Acoustical Society of America, 136(4), 2204-2204.
Schust, M. (2004). Effects of low frequency noise up to 100 Hz. Noise & Health,
6(23), 73-85.
Seong, Y., Lee, S., Gwak, D. Y., Cho, Y., Hong, J., & Lee, S. (2013). An Experimental
Study on Rating Scale for Annoyance due to Wind Turbine Noise. Paper presented
at INTER-NOISE 2013, the 42nd International Congress and Exposition
on Noise Control Engineering, Innsbruck, Austria.
Shepherd, D., McBride, D., Welch, D., Dirks, K. N., & Hill, E. M. (2011).
Evaluating the impact of wind turbine noise on health-related quality of life.
Noise & Health, 13, 333-339.
Sierra Club Canada. (2011). The Real Truth About Wind Energy: A Literature Review
on Wind Turbines in Ontario. Ottawa (ON): Sierra Club Canada.
Sndergaard, B. & Madsen, K. D. (2008). Low Frequency Noise from Large Wind
Turbines: Results from Sound Power Measurements. Venlighedsvej, Denmark:
DELTA (Danish Electronics, Light & Acoustics).
148
Sndergaard, B. (2011). Chapter 4. Wind Turbine Noise at the Receiver.

In D. Bowdler & G. Leventhall (Eds.), Wind Turbine Noise. Essex,
UnitedKingdom: Multi-Science Publishing Company, Ltd.
Srensen, M., Andersen, Z. J., Nordsborg, R. B., Becker, T., Tjonneland, A.,
Overvad, K., & Raaschou-Nielsen, O. (2012). Long-term exposure to road
traffic noise and incident diabetes: A cohort study. Environmental Health
Perspectives, 121(2), 217-222.
Spreng, M. (2000). Central nervous system activation by noise. Noise & Health,
2(7), 49.
Stansfeld, S., Haines, M., Burr, M., Berry, B., & Lercher, P. (2000). A review of
environmental noise and mental health. Noise & Health, 2(8), 1.
Stansfeld, S. A. (2003). Noise pollution: Non-auditory effects on health.
British Medical Bulletin, 68(1), 243-257. doi: 10.1093/bmb/ldg033.
Stansfeld, S. A., Berglund, B., Clark, C., Lopez-Barrio, I., Fischer, P., hrstrm, E., ...
Van Kamp, I. (2005). Aircraft and road traffic noise and childrens cognition
and health: A cross-national study. The Lancet, 365(9475), 1942-1949.
Stansfeld, S. A., Haines, M. M., Berry, B., & Burr, M. (2009). Reduction of road
traffic noise and mental health: An intervention study. Noise & Health,
11(44), 169.
Suls, J. (2013). Anger and the heart: Perspectives on cardiac risk, mechanisms
and interventions. Progress in Cardiovascular Diseases, 55(6), 538-547.
Sygna, K., Aasvang, G. M., Aamodt, G., Oftedal, B., & Krog, N. H. (2014). Road
traffic noise, sleep and mental health. Environmental Research, 131, 17-24.
Tachibana, H., Yano, H., Fukushima, A., & Sueoka, S. (2014). Nationwide field
measurements of wind turbine noise in Japan. Noise Control Engineering
Journal, 62(2), 90-101.
Takahashi, Y. (2011). A study on the contribution of body vibrations to the
vibratory sensation induced by high-level, complex lowfrequency noise.
Noise & Health, 13(50), 2-8.
Tamura, H., Ohgami, N., Yajima, I., Iida, M., Ohgami, K., Fujii, N., ... Kato, M. (2012).
Chronic exposure to low frequency noise at moderate levels causes impaired
balance in mice. PloS One, 7(6), e39807.
Tassi, P., Saremi, M., Schimchowitsch, S., Eschenlauer, A., Rohmer, O., &
Muzet, A. (2010). Cardiovascular responses to railway noise during sleep
in young and middle-aged adults. European Journal of Applied Physiology,
108(4), 671-680. doi: 10.1007/s00421-009-1270-8.
Thibault, B., Angen, E., & Weis, T. (2013). Survey of Complaints Received by Relevant
Authorities Regarding Operating Wind Energy in Alberta. Calgary (AB): The
Pembina Institute.
Todd, N. P. M., Rosengren, S. M., & Colebatch, J. G. (2008). Tuning and sensitivity
of the human vestibular system to low-frequency vibration. Neuroscience
Letters, 444, 36-41. doi: 10.1016/j.neulet.2008.08.011.
References
149
Toke, D., Breukers, S., & Wolsink, M. (2008). Wind power deployment outcomes:
How can we account for the differences? Renewable and Sustainable Energy
Reviews, 12(4), 1129-1147.
Tullio, P. (1929). Das Ohr und die Entstehung der Sprache und Schrift. Berlin,
Germany: Urban & Schwarzenberg.
van den Berg, F., Pedersen, E., Bouma, J., & Bakker, R. (2008). WINDFARMperception:
Visual and Acoustic Impact of Wind Turbine Farms on Residents. Groningen,
The Netherlands: University of Groningen, University of Gotenburg,
Science Shop for Medicine and Public Health Applied Health Research.
van den Berg, F. (2011). Chapter 6. Effects of Sound on People. In D. Bowdler
& G. Leventhall (Eds.), Wind Turbine Noise. Essex, United Kingdom: MultiScience Publishing Company, Ltd.
van den Berg, F. & Bowdler, D. (2011). Chapter 5. Amplitude Modulation. In
D. Bowdler & G. Leventhall (Eds.), Wind Turbine Noise. Essex, United
Van Gemmert, A. W. & Van Galen, G. P. (1997). Stress, neuromotor noise, and
human performance: A theoretical perspective. Journal of Experimental
Psychology: Human Perception and Performance, 23(5), 1299-1313.
van Kamp, I., Job, R. F. S., Hatfield, J., Haines, M., Stellato, R. K., & Stansfeld,
S. A. (2004). The role of noise sensitivity in the noise-response relation:
A comparison of three international airport studies. The Journal of the
Acoustical Society of America, 116(6), 3471-3479.
Van Kamp, I. & Davies, H. (2008). Noise as a Public Health Problem. Paper presented
at Proceedings of 9th Congress of the International Commission on the
Biological Effects of Noise (ICBEN), Foxwoods (CT).
van Kempen, E. & Babisch, W. (2012). The quantitative relationship between
road traffic noise and hypertension: A meta-analysis. Journal of Hypertension,
30(6), 1075-1086.
van Kempen, E. E., Kruize, H., Boshuizen, H. C., Ameling, C. B., Staatsen, B. A.,
& de Hollander, A. E. (2002). The association between noise exposure and
blood pressure and ischemic heart disease: A meta-analysis. Environmental
Health Perspectives, 110(3), 307-317.
Van Renterghem, T., Bockstael, A., De Weirt, V., & Botteldooren, D. (2013).
Annoyance, detection and recognition of wind turbine noise. The Science of
the Total Environment, 456-457, 333-345. doi: 10.1016/j.scitotenv.2013.03.095.
von Hnerbein, S., King, A., Piper, B., & Cand, M. (2013). Work Package B2:
Development of an AM Dose-Response Relationship. In RenewableUK
(Ed.), Wind Turbine Amplitude Modulation: Research to Improve Understanding
as to its Cause and Effects. London, United Kingdom: University of Salford,
Acoustics Research Centre.
150
Wagner, S., Barei, R., & Guidati, G. (1996). Wind Turbine Noise. Berlin,
Germany: Springer.
Watanabe, T. & Mller, H. (1990). Low frequency hearing thresholds in free
field. Journal of Low Frequency Noise and Vibration, 9, 106-115.
Welford, A. T. (1974). Stress and Performance. In A. T. Welford (Ed.),
Man Under Stress. New York (NY): Halsted Press.
Welgampola, M. S., Oluwaseun, A. M., Minor, L. B. Carey, J.P. (2008). Vestibularevoked
myogenic potential thresholds normalize on plugging superior canal
dehiscence. Neurology, 70(6), 464-472.
WHO (World Health Organization). (1946). Constitution of the World
Health Organization.
WHO (World Health Organization). (1999). Guidelines for Community Noise.
WHO (World Health Organization). (2009). Night Noise Guidelines for Europe.
Copenhagen, Denmark: WHO.
WHO (World Health Organization). (2011). Burden of Disease from Environmental Noise.
WHO (World Health Organization). (2013). Cardiovascular diseases (CVDs):
Fact Sheet No. 317. Geneva, Switzerland: WHO.
WHO (World Health Organization). (2014). Diabetes: Fact sheet N312. Geneva,
Switzerland: WHO.
Willich, S. N., Wegscheider, K., Stallmann, M., & Keil, T. (2006). Noise burden
and the risk of myocardial infarction. European Heart Journal, 27(3), 276-282.
Wolsink, M. (2007). Wind power implementation: The nature of public attitudes:
Equity and fairness instead of backyard motives. Renewable and Sustainable
Energy Reviews, 11, 1188-1207. doi: 10.1016/j.rser.2005.10.005.
Yano, T., Lee, S., & Gjestland, T. (2012). Community response to noise.
Noise & Health, 14(61), 303. doi: 10.4103/1463-1741.104898.
Yano, T., Kuwano, S., Kageyama, T., Sueoka, S., & Tachibana, H. (2013).
Dose-response Relationships for Wind Turbine Noise in Japan. Paper presented
at INTER-NOISE 2013, the 42nd International Congress and Exposition
on Noise Control Engineering, Innsbruck, Austria.
Yokoyama, S., Sakamoto, S., & Tachibana, H. (2013). Study on the Amplitude
Modulation of Wind Turbine Noise: Part 2 Auditory Experiments. Paper
presented at INTER-NOISE 2013, the 42nd International Congress and
Exposition on Noise Control Engineering, Innsbruck, Austria.
Zajamsek, B., Moreau, D. J., & Doolan, C. J. (2014). Characterising noise and
annoyance in homes near a wind farm. Acoustics Australia, 42(1), 15.
Zhang, A. S., Govender, S., & Colebatch, J. G. (2012). Tuning of the ocular
vestibular evoked myogenic potential to bone-conducted sound
stimulation. Journal of Applied Psychology, 112(8), 1279-1290. doi: 10.1152/
japplphysiol.01024.2011.
Assessments of the Council of Canadian Academies
151
Assessments of the Council of Canadian Academies

The assessment reports listed below are accessible through the Councils
website (www.scienceadvice.ca):
Understanding the Evidence: Wind Turbine Noise (2015)
Accessing Health and Health-Related Data in Canada (2015)
Leading in the Digital World: Opportunities for Canadas Memory
Institutions (2015)
Policing Canada in the 21st Century: New Policing for New Challenges (2014)
Energy Prices and Business Decision-Making in Canada: Preparing for the Energy
Future (2014)
Improving Medicines for Children in Canada (2014)
Science Culture: Where Canada Stands (2014)
Enabling Sustainability in an Interconnected World (2014)
Environmental Impacts of Shale Gas Extraction in Canada (2014)
Aboriginal Food Security in Northern Canada: An Assessment of the State of
Knowledge (2014)
Ocean Science in Canada: Meeting the Challenge, Seizing the Opportunity (2013)
The Health Effects of Conducted Energy Weapons (2013)
The State of Industrial R&D in Canada (2013)
Innovation Impacts: Measurement and Assessment (2013)
Water and Agriculture in Canada: Towards Sustainable Management of Water
Resources (2013)
Strengthening Canadas Research Capacity: The Gender Dimension (2012)
The State of Science and Technology in Canada (2012)
Informing Research Choices: Indicators and Judgment (2012)
Integrating Emerging Technologies into Chemical Safety Assessment (2012)
Healthy Animals, Healthy Canada (2011)
Canadian Taxonomy: Exploring Biodiversity, Creating Opportunity (2010)
Honesty, Accountability, and Trust: Fostering Research Integrity in
Canada (2010)
Better Research for Better Business (2009)
The Sustainable Management of Groundwater in Canada (2009)
Innovation and Business Strategy: Why Canada Falls Short (2009)
Vision for the Canadian Arctic Research Initiative: Assessing the
Opportunities (2008)
152
Energy from Gas Hydrates: Assessing the Opportunities and Challenges for
Canada (2008)
Small Is Different: A Science Perspective on the Regulatory Challenges of the
Nanoscale (2008)
Influenza and the Role of Personal Protective Respiratory Equipment: An
Assessment of the Evidence (2007)
The State of Science and Technology in Canada (2006)
The assessments listed below are in the process of expert panel deliberation:
STEM Skills for the Future
The Potential for New and Emerging Technologies to Reduce the Environmental
Impacts of Oil Sands Development
RISK: Is the Message Getting Through?
Energy Use and Climate Change: A Synthesis of the Latest Evidence
Board of Governors of the Council of Canadian Academies
153
Board of Governors of the Council of

Canadian Academies*
Margaret Bloodworth, C.M., Chair, Former Federal Deputy Minister and
National Security Advisor (Ottawa, ON)
Graham Bell, FRSC, President, Royal Society of Canada; Research Director,
James McGill Professor, Chair, Department of Biology, McGill University
(Montral, QC)
John Cairns, FCAHS, President, Canadian Academy of Health Sciences; Professor
of Medicine, University of British Columbia (Vancouver, BC)
Henry Friesen, C.C., FRSC, FCAHS, Vice Chair, Distinguished Professor
Emeritus and Senior Fellow, Centre for the Advancement of Medicine, Faculty
of Medicine, University of Manitoba (Winnipeg, MB)
Carol P. Herbert, FCAHS, Professor of Family Medicine, Western University
(London, ON)
Claude Jean,** Executive Vice President and General Manager, Foundry
Operation, Semiconductor, Teledyne DALSA (Bromont, QC)
Peter MacKinnon, O.C., Former President and Vice-Chancellor, University of
Saskatchewan (Saskatoon, SK)
Jeremy McNeil, FRSC, Helen Battle Professor of Chemical Ecology, Western
University (London, ON)
Axel Meisen, C.M., FCAE, Former Chair of Foresight at Alberta
Innovates Technology Futures (AITF) (Edmonton, AB)
Lydia Miljan, Associate Professor of Political Science and Chair of the Arts
and Science Program, University of Windsor (Windsor, ON)
Ted Morton, Executive Fellow at the School of Public Policy, Professor of
Political Science, University of Calgary (Calgary, AB)
P. Kim Sturgess, FCAE, CEO and Founder, Alberta WaterSMART (Calgary, AB)
* Affiliations as of March 2015

** Resigned in FY 20142015
154
Scientific Advisory Committee of the Council of

Canadian Academies*
Susan A. McDaniel, FRSC, Chair, Director, Prentice Institute; Canada Research
Chair in Global Population and Life Course; Prentice Research Chair (Tier1)
in Global Population and Economy; Professor of Sociology, University of
Lethbridge (Lethbridge, AB)
Lorne Babiuk, O.C., FRSC, FCAHS, Vice President (Research), University of
Alberta (Edmonton, AB)
Murray S. Campbell, Senior Manager, AI and Optimization, IBM T.J. Watson
Research Center (Yorktown Heights, NY)
Clarissa Desjardins, CEO, Clementia Pharmaceuticals Inc. (Montral, QC)
Jean Gray, C.M., FCAHS, Professor of Medicine (Emeritus), Dalhousie University
(Halifax, NS)
John Hepburn, FRSC, Vice-President, Research and International, University
of British Columbia (Vancouver, BC)
Gregory S. Kealey, FRSC, Professor, Department of History, University of
New Brunswick (Fredericton, NB)
Daniel Krewski, Professor of Epidemiology and Community Medicine and
Scientific Director of the McLaughlin Centre for Population Health Risk
Assessment, University of Ottawa (Ottawa, ON)
Avrim Lazar, Former President and CEO, Forest Products Association of
Canada (Ottawa, ON)
Norbert Morgenstern, C.M., FRSC, FCAE, University Professor (Emeritus),
Civil Engineering, University of Alberta (Edmonton, AB)
Sarah P. Otto, FRSC, Professor and Director of the Biodiversity Research
Centre, University of British Columbia (Vancouver, BC)
* Affiliations as of March 2015
Council of Canadian Academies

180 Elgin Street, Suite 1401
Ottawa, ON K2P 2K3
Tel: 613-567-5000
www.scienceadvice.ca

Understanding The Evidence: Wind Turbine Noise The Expert Panel On Wind Turbine Noise and Human Health

Uploaded by

Copyright:

Available Formats

Understanding The Evidence: Wind Turbine Noise The Expert Panel On Wind Turbine Noise and Human Health

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Understanding The Evidence: Wind Turbine Noise The Expert Panel On Wind Turbine Noise and Human Health

Uploaded by

Copyright:

Available Formats

Understanding

Evidence: Wind Turbine Noise

The Expert Panel on Wind Turbine Noise

Science Advice in the Public Interest

UNDERSTANDING THE EVIDENCE: WIND TURBINE NOISE

Understanding the Evidence: Wind Turbine Noise

THE COUNCIL OF CANADIAN ACADEMIES

Library and Archives Canada Cataloguing in Publication

This assessment was made possible with

The Council of Canadian Academies

The Council of Canadian Academies

Understanding the Evidence: Wind Turbine Noise

The Canadian Academy of Health Sciences (CAHS) recognizes individuals of

Expert Panel on Wind Turbine Noise and Human Health

Expert Panel on Wind Turbine Noise and Human Health

Understanding the Evidence: Wind Turbine Noise

Message from the Chair

Message from the Chair

Understanding the Evidence: Wind Turbine Noise

Project Staff of the Council of Canadian Academies

Emmanuel Mongin, Associate Program Director

With assistance from:

Leela Steiner, Consultant

Understanding the Evidence: Wind Turbine Noise

Janet W. Bax, Interim President

In response to growing public concern about the potential health effects of

Understanding the Evidence: Wind Turbine Noise

To respond to the Charge, the Panel used an evidence-based approach to

Literature Specific to Wind Turbines and Human Health

A C O US TI C CH ARAC T E RI S T I C S OF W IND TU R B INE NOISE

1. Sound from wind turbines is complex and variable

Understanding the Evidence: Wind Turbine Noise

3. The evidence is sufficient to establish a causal relationship between

Understanding the Evidence: Wind Turbine Noise

7. For all other health effects considered (fatigue, tinnitus, vertigo,

Sufficient evidence of a causal relationship

Grey boxes: inadequate evidence of a causal

8. Knowledge gaps prevent a full assessment of public health effects

Understanding the Evidence: Wind Turbine Noise

with long-term exposure to other sources of community noise. Given the

10. Technological development is unlikely to resolve, in the short term,

Wind turbines are a progressively familiar sight in Canada and contribute an

Understanding the Evidence: Wind Turbine Noise

Charge to the Panel..........................................................................3

Wind Energy Context.............................................................. 7

Wind Turbine Design........................................................................9

Wind Turbine Acoustics ........................................................ 20

Hearing and Perception of Sound........................................ 42

The Auditory Pathway.....................................................................44

Recent Reviews of Wind Turbine Noise and Health.....................53

Current Evidence Related to Adverse

Overview of the Evidence, Knowledge Gaps,

Summary of the Evidence...............................................................96

Promising Practices......................................................................... 107

Charge to the Panel

Interpreting the Charge

Health Canada Wind Turbine Noise

Organization of the Report

Understanding the Evidence: Wind Turbine Noise