A09.03 Low Frequency Noise Effects: What Research Shows
Key Takeaways
- Scientific understanding of low frequency noise is built from multiple research approaches, including laboratory studies, community surveys, sleep studies and epidemiological research. No single study provides the complete answer.
- The strongest and most consistent evidence relates to annoyance and sleep disturbance. Broader health effects remain an active area of research.
- There is no universal low frequency noise threshold that guarantees comfort or disturbance. Human response depends on sound characteristics, exposure conditions, buildings, background environment and individual sensitivity.
- Conventional A-weighted measurements are useful for general environmental noise assessment but may not fully represent low frequency exposure because they reduce the contribution of low frequency sound.
- Research from industrial facilities, wind turbines and other mechanical sources shows that low frequency noise assessment often requires frequency-specific analysis beyond a single dBA value.
- Emerging neuroscience research suggests that very low frequency sound perception may involve more complex biological pathways than previously understood, although these findings remain under investigation.
- For data centres, considering low frequency noise during planning and design can help reduce future community concerns, improve stakeholder confidence and support sustainable development.
- Low frequency noise is best understood through evidence-based assessment that considers both acoustic measurements and human experience.
Low frequency noise has become one of the most widely discussed—and often misunderstood—topics in environmental acoustics (→A09 Understanding Low-Frequency Noise from Data Centres).
For decades, engineers, regulators and researchers have recognised that large industrial facilities, ventilation systems, compressors, cooling towers and electrical infrastructure can generate substantial amounts of low frequency sound. More recently, the rapid expansion of hyperscale data centres has brought renewed attention to this subject because many of the cooling systems required to remove large quantities of heat operate continuously throughout the day and night.
Public discussion, however, often presents two opposing narratives.
One suggests that low frequency noise presents limited concern because it is often difficult to detect, and because assessments frequently show compliance with existing environmental noise criteria.
The other argues that low frequency noise is responsible for a wide range of health effects, sometimes extending far beyond what has been established by scientific research.
Neither perspective accurately reflects the current state of knowledge.
Over the past four decades, researchers have conducted laboratory experiments, community surveys, sleep studies, epidemiological investigations and systematic evidence reviews to better understand how people respond to prolonged exposure to low frequency noise.
Some findings have become well established.
Others remain the subject of active scientific investigation.
The purpose of this article is not to promote or dismiss any particular viewpoint.
Instead, it reviews the current body of scientific evidence, explains what researchers know with reasonable confidence, identifies where uncertainty remains, and considers what these findings may mean for communities, industry and policymakers.
Throughout this article, the discussion focuses on low frequency noise rather than infrasound alone. Although the two are closely related, most environmental concerns surrounding industrial facilities and data centres arise from frequencies that remain within the lower end of the normal hearing range rather than frequencies entirely below it.
As in previous MDCO articles, the emphasis is placed on understanding the engineering and scientific evidence before drawing conclusions about its practical implications.
MDCO Insight: Scientific understanding of low frequency noise has advanced considerably over recent decades, but the strongest conclusions come not from individual studies, but from the collective weight of evidence accumulated across many different fields of research.
Human Response Is More Complex Than Simply “Can You Hear It?”
One of the most significant developments in environmental acoustics over the past several decades has been the recognition that human response to sound cannot be explained simply by asking whether a sound is audible (→E02.02 Low Frequency Noise: How Humans Hear Sound Beyond Decibels).
For much of the twentieth century, environmental noise assessment relied heavily on the assumption that:
human response was closely related to perceived loudness and audibility.
This assumption proved to be broadly successful for many common environmental noises because conventional environmental noise is dominated by frequencies where human hearing is most sensitive.
Low frequency noise, however, presented researchers with a more complicated picture.
Community surveys and field investigations have reported that some residents living near industrial facilities describe persistent humming, sleep disturbance or sensations of pressure, including in situations where measured sound levels complied with existing guidelines.
These observations prompted researchers to investigate whether human response to low frequency sound involved additional mechanisms beyond conventional hearing sensitivity.
Today, scientists recognise that several different processes contribute to the way people experience environmental sound.
These include:
- auditory perception, or whether a sound is consciously heard;
- annoyance, reflecting the psychological response to persistent or unwanted sound;
- sleep disturbance, particularly during quiet nighttime conditions;
- cognitive effects, such as distraction or reduced concentration, which may occur in some exposure conditions;
- physiological stress responses, which may occur indirectly through repeated sleep disruption or prolonged annoyance.
Importantly, these responses should not be confused with one another.
A sound may be clearly audible without causing significant annoyance.
Conversely, a sound that is only faintly perceptible may become increasingly disturbing if it is continuous, tonal or repeatedly interferes with sleep.
Similarly, the presence of annoyance does not necessarily imply physical injury or disease, although persistent annoyance and sleep disruption may themselves contribute to broader health outcomes over extended periods.
Understanding these distinctions is essential when interpreting scientific research.
Many apparent disagreements between different studies arise because researchers are investigating different aspects of human response rather than contradicting one another.
One study may examine hearing thresholds.
Another may investigate annoyance.
A third may focus on sleep quality.
All three may reach valid conclusions while addressing different scientific questions.
The challenge, therefore, is not to identify a single “effect” of low frequency noise, but to understand how different forms of human response relate to different types and levels of exposure.
MDCO Insight: Hearing a sound, noticing a sound, being annoyed by a sound and experiencing measurable health effects are related but scientifically distinct responses. Understanding low frequency noise requires recognising these important differences.
What Makes Low Frequency Noise Different?
Low frequency noise is governed by the same physical laws as every other sound.
It consists of pressure waves travelling through the air, generated whenever a vibrating object transfers energy into its surroundings.
What makes low frequency noise different is not the way it is produced, but the way it behaves after it is generated.
Because low frequency sound has much longer wavelengths than higher-frequency sound, it interacts with the environment in fundamentally different ways.
Long wavelengths generally experience lower atmospheric absorption than higher frequencies, allowing low frequency sound to remain detectable over longer distances.
They also bend, or diffract, around buildings, walls and other obstacles more effectively than shorter wavelengths, making them more difficult to shield using conventional noise barriers.
These same long wavelengths can also interact with buildings in distinctive ways.
Instead of being completely blocked by external walls, low frequency sound is more capable of penetrating building envelopes, sometimes producing audible indoor humming or causing certain rooms to resonate at particular frequencies.
Unlike high-frequency environmental sounds, which often vary rapidly over time, many industrial sources of low frequency noise operate continuously.
Large cooling towers, compressors, pumps and transformers may generate relatively stable acoustic signatures that remain present throughout both day and night.
This persistence is one reason why researchers have devoted particular attention to low frequency noise in studies of environmental annoyance and sleep disturbance.
Another important characteristic is that low frequency noise is frequently dominated by discrete tonal components rather than broad-band random sound.
A cooling tower fan, for example, may produce identifiable blade-pass frequencies, while a 50 Hz power transformer generates a stable 100 Hz magnetic hum that changes very little with time.
Such tonal sounds are often more noticeable than random background noise, even when their overall sound level is relatively modest.
Finally, advances in neuroscience have suggested that the human response to low frequency sound may involve more complex biological pathways than previously recognised. Although many aspects of these mechanisms remain under active investigation, emerging evidence indicates that the auditory system, vestibular system and central nervous system may together contribute to the perception and processing of very low frequency sound.
These emerging findings do not overturn the established understanding of hearing.
Rather, they expand it by suggesting that the relationship between physical sound and human experience may be more complex than earlier models assumed.
The following sections examine how researchers have investigated these questions and what decades of scientific evidence reveal about the effects of prolonged exposure to low frequency noise.
MDCO Insight: Low frequency noise differs not because it is a different kind of sound, but because its long wavelengths, persistence and interaction with the human body create challenges that conventional environmental noise assessments were not originally designed to address.
How Scientists Study Low Frequency Noise
One of the most common misconceptions about low frequency noise research is that all scientific studies provide the same type of evidence.
In reality, researchers use different methods to answer different questions.
Some studies investigate how the human body responds to carefully controlled sound exposure. Others examine how entire communities experience environmental noise over many years. Large population studies explore whether long-term exposure is associated with broader health outcomes, while detailed investigations of individual communities help explain why particular locations experience persistent problems.
Each approach has its own strengths and limitations. None provides a complete answer on its own.
Scientific confidence increases when different methods, conducted by independent researchers in different countries, consistently point towards similar conclusions.
Laboratory Studies
Laboratory studies investigate how people respond to sound under carefully controlled conditions.
Participants are exposed to precisely reproduced sounds inside specialised acoustic chambers where researchers control variables such as sound pressure level, frequency, exposure duration and background conditions.
This allows researchers to isolate the effect of the sound itself while minimising interference from unrelated factors.
Laboratory experiments have been widely used to investigate:
- hearing thresholds;
- perceived loudness;
- annoyance;
- concentration;
- cognitive performance;
- physiological responses; and
- brain activity.
Their principal advantage is high experimental control, allowing researchers to identify cause-and-effect relationships with considerable confidence.
However, laboratory studies usually involve relatively short exposure periods and cannot fully reproduce the experience of living near a continuous environmental noise source for months or years.
Community Surveys
Community surveys investigate how environmental noise affects people under real-world conditions.
Rather than studying volunteers in a laboratory, researchers examine residents living near actual sources of low frequency noise, including:
- industrial plants;
- cooling towers;
- ventilation systems;
- compressors;
- wind turbines; and
- other large mechanical installations.
Residents are asked about issues such as:
- whether they notice the noise;
- annoyance;
- sleep disturbance;
- effects on daily activities; and
- overall quality of life.
These responses are then compared with measured or modelled sound levels.
Community surveys provide valuable insight into long-term lived experience, although individual perception is also influenced by personal sensitivity, attitudes, previous experiences and many other social factors that require careful statistical interpretation.
Sleep Laboratory Studies
Because sleep disturbance is one of the most consistently reported consequences of environmental noise, many researchers study sleep directly.
Participants sleep under controlled acoustic conditions while physiological measurements are continuously recorded.
Typical measurements include:
- electroencephalography (EEG);
- sleep stages;
- awakenings and micro-awakenings;
- heart rate;
- heart rate variability;
- breathing patterns; and
- stress-related hormones such as cortisol.
These studies provide objective evidence of how noise influences sleep architecture, although they cannot fully reproduce the experience of repeated exposure over many years in a person’s own home.
Epidemiological Studies
Epidemiological studies examine environmental noise across large populations.
Instead of measuring individual responses in detail, researchers compare long-term noise exposure with health outcomes such as:
- sleep disorders;
- hypertension;
- cardiovascular disease;
- mental wellbeing;
- medication use; and
- hospital admissions.
Advanced statistical methods are used to account for other influences such as age, smoking, socioeconomic status and air pollution.
Although epidemiological studies cannot by themselves prove that noise causes a particular disease, they are invaluable for identifying consistent long-term patterns across thousands of people.
Case Investigations
Some of the most detailed engineering knowledge comes from investigations of specific communities experiencing persistent low frequency noise.
These investigations typically combine:
- long-term acoustic monitoring;
- frequency spectrum analysis;
- vibration measurements;
- meteorological observations;
- building inspections; and
- interviews with residents.
Individual case studies cannot automatically be generalised to every location because every site has different geography, buildings and industrial equipment.
Nevertheless, they often identify practical issues that later become the focus of broader scientific research and improvements in engineering practice.
Taken together, these five approaches provide complementary evidence. Laboratory experiments explain how people respond to sound, community surveys describe real-world experience, sleep studies reveal physiological changes, epidemiology identifies long-term population trends, and case investigations help explain why particular problems occur.
MDCO Insight: Confidence grows not from one study, but when different scientific methods consistently point towards the same conclusions.
How Scientific Understanding Has Evolved
The modern understanding of low frequency noise did not emerge from a single breakthrough.
Instead, it has developed over almost five decades as successive studies addressed different pieces of the puzzle.
Some researchers questioned whether conventional noise measurements adequately described low frequency sound. Others investigated why certain communities continued to report problems despite complying with existing regulations. More recent research has begun exploring how the human brain itself processes very low frequency sound.
Together, these landmark studies illustrate how scientific understanding has progressively matured.
Broner (1978): The First Challenge to Conventional Noise Measurement
One of the earliest influential investigations was conducted by Norman Broner, who examined persistent community complaints associated with industrial low frequency noise.
At the time, environmental assessments relied almost entirely on A-weighted sound levels. If a facility complied with the relevant dBA limit, it was generally assumed that nearby residents would not experience significant problems.
Broner observed that this assumption was often inconsistent with real community experience.
Residents sometimes reported persistent humming, vibration or annoyance even though measured dBA levels appeared relatively modest.
He proposed that conventional A-weighting could substantially reduce the apparent contribution of low frequency sound, allowing two environments with very different low frequency characteristics to produce similar dBA values.
This work became one of the foundations for later low frequency assessment methods.
Leventhall (2003–2004): Bringing Low Frequency Noise into Mainstream Environmental Acoustics
Few researchers have shaped modern understanding of low frequency noise more than Geoff Leventhall.
Drawing together engineering measurements, psychoacoustics and community observations, his work demonstrated that human response depends on more than overall sound level.
His research consistently found that:
- tonal sounds are generally more disturbing than broadband noise;
- indoor environments may increase annoyance through room resonance;
- individual sensitivity varies considerably; and
- sleep disturbance and long-term annoyance are the most consistently reported community impacts.
Leventhall’s publications remain among the most widely cited references in environmental low frequency noise research and continue to influence engineering practice worldwide.
DEFRA NANR45: Learning from Real Communities
The United Kingdom’s DEFRA Project NANR45 examined hundreds of genuine community complaints relating to low frequency noise.
Investigators repeatedly found that residents sometimes experienced persistent disturbance even when environmental noise complied with existing A-weighted limits.
Rather than concluding that all low frequency noise is problematic, the project demonstrated that conventional measurements may not always explain community experience where significant low frequency or tonal components are present.
Its findings led to practical assessment procedures that continue to influence environmental investigations internationally.
Wind Turbine Research: Understanding Long-Term Exposure
Extensive research conducted in Denmark, Sweden, the Netherlands and other countries examined communities living near wind turbines.
Although wind turbines differ from data centres, both rely on large rotating machinery capable of generating substantial low frequency sound.
Across numerous studies, researchers found that:
- annoyance generally increased with increasing exposure;
- indoor environments often became the primary location where residents noticed persistent low frequency sound; and
- outdoor A-weighted sound levels alone did not fully explain reported responses.
These findings reinforce the importance of considering frequency content rather than relying solely on overall dBA values.
WHO Environmental Noise Guidelines (2018): Reviewing the Global Evidence
By 2018, environmental noise research had expanded sufficiently for the World Health Organization (WHO) to conduct one of the largest evidence reviews ever undertaken.
Rather than performing new experiments, the WHO evaluated the collective findings from hundreds of independent studies.
Among the health outcomes reviewed, the strongest evidence related to:
- annoyance;
- sleep disturbance; and
- increased long-term cardiovascular risk associated with chronic environmental noise exposure.
Although the WHO guidelines are not specific to industrial low frequency noise, they reinforce an important principle: protecting sleep is central to protecting long-term health.
Recent Neuroscience Research: A New Direction
Recent neuroscience studies published between 2023 and 2025 have opened a new area of investigation.
Using advanced brain imaging and controlled laboratory experiments, researchers have observed measurable neural responses to infrasound and very low frequency sound even when participants reported little or no conscious hearing.
These findings suggest that very low frequency sound may be processed through neural pathways that extend beyond the classical model of hearing.
At present, these studies should be regarded as emerging evidence rather than established scientific consensus.
They do not overturn decades of previous research, nor do they demonstrate previously unknown health effects.
Instead, they suggest that the biological processing of very low frequency sound may be more complex than earlier models assumed, providing an important direction for future research.
Taken together, these landmark studies illustrate the progressive development of scientific understanding. Early investigations questioned conventional measurement methods, later research documented community experience and physiological responses, while the latest neuroscience is beginning to explore the underlying biological mechanisms.
The overall picture has become progressively clearer—not because one study provided all the answers, but because decades of independent research have converged on many of the same conclusions.
MDCO Insight: Decades of complementary research consistently show that low frequency noise deserves assessment beyond conventional dBA measurements, while emerging neuroscience suggests there may still be aspects of human perception that science has yet to fully explain.
What Effects Have Been Consistently Observed?
One of the most common questions asked about low frequency noise is whether it affects human health.
The answer depends on how “effect” is defined.
Some effects, such as annoyance and sleep disturbance, have been investigated extensively over several decades and are consistently supported by scientific evidence.
Other reported effects—including headaches, pressure sensations and reduced concentration—have also been documented in many studies, although the strength of the evidence varies.
Research into broader long-term health outcomes, including cardiovascular disease, continues to develop. While some statistical associations have been observed, scientists remain careful to distinguish between an observed association and proof that low frequency noise directly causes a particular medical condition.
This distinction is important.
Scientific research rarely asks whether low frequency noise is simply “safe” or “dangerous”. Instead, it investigates how the probability of different outcomes changes as exposure increases, recognising that individuals respond differently under similar acoustic conditions.
Annoyance
Among all human responses to environmental low frequency noise, annoyance is the most consistently demonstrated effect.
In everyday conversation, annoyance usually means mild irritation.
In environmental health research, however, the term has a much broader meaning.
Annoyance describes the degree to which environmental noise interferes with a person’s comfort, wellbeing and normal enjoyment of daily life. It may include persistent awareness of the sound, difficulty relaxing, frustration, reduced enjoyment of one’s home and concern about continued exposure.
It does not imply that the person is simply being oversensitive.
Instead, annoyance is recognised internationally as a legitimate environmental health outcome because persistent annoyance itself can reduce quality of life and contribute to chronic stress.
Numerous studies have reached similar conclusions.
Leventhall: Understanding Why Some Sounds Become More Annoying
Through a series of influential reviews during the early 2000s, Geoff Leventhall examined engineering measurements together with reports from communities exposed to environmental low frequency noise.
Rather than asking whether low frequency noise could be measured, his work focused on why some communities experienced greater disturbance than conventional sound levels appeared to predict.
His review identified several recurring observations.
Persistent tonal sounds tended to produce greater annoyance than broadband mechanical noise.
Indoor environments often became the primary location where annoyance was experienced because rooms could reinforce certain low frequencies through resonance.
Individual sensitivity also varied considerably. While many residents experienced little difficulty, a smaller proportion consistently reported substantially greater disturbance under similar acoustic conditions.
Perhaps most importantly, Leventhall concluded that persistent annoyance and sleep disturbance were the dominant community outcomes, rather than direct physical injury.
These conclusions continue to influence environmental acoustic practice internationally.
Persson Waye: Measuring Community Response
Professor Kerstin Persson Waye and colleagues have conducted some of the world’s most influential research into community response to environmental low frequency noise.
Their studies investigated residents living near industrial facilities and wind turbines, comparing measured sound exposure with reported annoyance.
A consistent pattern emerged.
As low frequency exposure increased, the proportion of residents reporting annoyance also increased.
Importantly, annoyance was influenced not only by overall sound level, but also by factors such as tonal characteristics, nighttime exposure and indoor sound conditions.
These findings reinforced an important principle that now underpins many environmental assessments:
Two locations with similar A-weighted sound levels may produce very different community responses if their low frequency content differs substantially.
DEFRA: When Measurements and Experience Did Not Match
The United Kingdom’s DEFRA NANR45 investigation reached similar conclusions after examining hundreds of genuine low frequency noise complaints.
Researchers found that many complaints occurred even though conventional environmental noise limits had been met.
Rather than dismissing these complaints, the investigation concluded that conventional A-weighted measurements sometimes failed to describe the characteristics that residents actually experienced.
This work helped establish the need for supplementary low frequency assessment methods in appropriate situations.
Taken together, these studies consistently demonstrate that annoyance is the most robustly established human response to environmental low frequency noise.
Sleep Disturbance
After annoyance, sleep disturbance has the strongest and most consistent scientific evidence.
Sleep is one of the body’s most important biological processes. Even relatively small disturbances, when repeated night after night, may gradually affect wellbeing.
Researchers have investigated sleep using both laboratory experiments and community studies.
Sleep laboratory studies measure objective physiological responses, including:
- brain activity (EEG);
- awakenings and micro-awakenings;
- sleep stage transitions;
- heart rate;
- breathing patterns; and
- stress-related hormones.
Community studies examine how residents living near environmental noise sources describe their own sleep quality over longer periods.
Across both types of research, several findings appear repeatedly.
People exposed to environmental noise are more likely to report:
- difficulty falling asleep;
- repeated awakenings;
- lighter sleep;
- poorer perceived sleep quality; and
- feeling less refreshed the following day.
Basner and Colleagues: Measuring Sleep Directly
Among the leading researchers in environmental sleep science is Professor Mathias Basner.
His work combines controlled laboratory experiments with physiological monitoring to examine how environmental noise influences sleep.
Rather than relying solely on people’s memories of whether they woke during the night, these studies measure objective biological responses.
The research demonstrates that environmental noise can alter normal sleep architecture even when individuals do not always remember being awakened.
This has become an important finding in environmental health because repeated disruption of normal sleep may accumulate over time.
The WHO Evidence Review
The 2018 WHO Environmental Noise Guidelines reviewed hundreds of published studies relating environmental noise to human health.
Among all health outcomes considered, the evidence for sleep disturbance was judged to be one of the strongest.
The WHO therefore places particular emphasis on reducing unnecessary nighttime environmental noise exposure because protecting sleep is regarded as an important public health objective.
Reduced Quality of Life
Many studies report that persistent environmental low frequency noise can reduce overall quality of life even when measured sound levels remain relatively modest.
Residents commonly describe:
- reduced enjoyment of outdoor spaces;
- reluctance to open windows;
- constant awareness of background humming;
- difficulty relaxing at home; and
- feeling unable to escape the sound.
These effects are closely related to annoyance but represent broader impacts on everyday living rather than simply the perception of noise itself.
Reduced Concentration
Evidence that low frequency noise affects concentration is more limited than evidence for annoyance or sleep disturbance.
Some laboratory studies suggest that persistent low frequency sound may interfere with sustained attention, reading or mentally demanding tasks, particularly where the sound is tonal or continuously noticeable.
However, findings between studies are less consistent.
For this reason, reduced concentration is generally regarded as a possible effect supported by moderate evidence, rather than one of the most firmly established outcomes.
Headaches and Pressure Sensations
Residents living near sources of environmental low frequency noise sometimes report symptoms such as:
- headaches;
- a feeling of pressure in the ears;
- a sensation of fullness;
- vibration sensations; or
- awareness of sound that is difficult to describe.
These reports appear frequently in community investigations.
However, an important scientific distinction must be made.
The existence of reported symptoms does not necessarily establish the biological mechanism responsible for those symptoms.
Current research has not reached a scientific consensus explaining exactly how low frequency noise may contribute to these experiences.
Accordingly, these symptoms are best described as commonly reported but not yet fully understood.
Stress Response
Persistent annoyance and repeated sleep disturbance may contribute to increased psychological stress.
Researchers have observed that chronic environmental noise exposure may be associated with elevated stress responses, including changes in stress-related hormones such as cortisol.
Importantly, these effects are generally understood as secondary consequences of prolonged disturbance rather than direct physiological effects of the sound itself.
In other words, the pathway is often considered to be:
Persistent environmental noise → annoyance and disrupted sleep → chronic stress → broader health consequences.
This distinction is widely recognised in environmental health research.
Cardiovascular Disease
Research over the past two decades has increasingly examined whether long-term environmental noise exposure may be associated with cardiovascular disease.
Large epidemiological studies have reported statistical associations between chronic environmental noise exposure and outcomes such as hypertension and ischaemic heart disease.
However, these findings require careful interpretation.
An association does not prove that low frequency noise directly causes cardiovascular disease.
Many factors influence cardiovascular health, including age, smoking, diet, physical activity, socioeconomic conditions and air pollution.
Researchers therefore use sophisticated statistical methods to account for these influences, but absolute proof of causation remains difficult.
Accordingly, organisations such as the WHO describe the evidence as supporting an association with long-term environmental noise exposure, while recognising that research into the underlying mechanisms continues.
MDCO Insight: The strongest scientific evidence consistently links environmental low frequency noise with annoyance and sleep disturbance, while research into broader health effects continues to evolve.
At What Sound Levels Do These Effects Become More Likely?
One of the questions most frequently asked by communities is:
“At what sound level does low frequency noise become a problem?”
Surprisingly, there is no single universally accepted threshold.
Human response does not suddenly change at one particular sound level.
Instead, scientific research shows that the probability of noticing, being annoyed by or being disturbed by low frequency noise generally increases as exposure increases.
Some people remain unaffected at sound levels that others find disturbing, while particularly sensitive individuals may notice sounds that most people scarcely perceive.
This variability explains why modern environmental health research focuses on increasing probability of response, rather than a single “safe” or “unsafe” value.
Why Conventional Noise Limits Use dBA
Most environmental noise regulations around the world are based on A-weighted sound pressure levels (dBA).
A-weighting reflects the sensitivity of the human ear to ordinary sounds and has been highly successful in protecting people from hearing damage and excessive environmental noise across most frequencies.
For most transportation and industrial noise, dBA remains an appropriate and practical assessment metric.
However, as discussed in the previous article, A-weighting substantially reduces the contribution of frequencies below approximately 200 Hz.
Consequently, two environments with similar dBA levels may contain very different amounts of low frequency energy.
This limitation has led several countries to develop supplementary guidance specifically for low frequency noise.
Low Frequency Assessment Approaches
Recognising the limitations of relying solely on A-weighted measurements, several organisations have introduced additional methods for assessing low frequency noise.
Examples include:
- DIN 45680 (Germany), which evaluates octave-band sound pressure levels across the low frequency range;
- DEFRA NANR45 (United Kingdom), which provides practical procedures for investigating community complaints;
- Nordic guidance documents, particularly those developed in Denmark and Sweden for wind turbine assessments; and
- Japanese environmental guidance, which also incorporates octave-band analysis for low frequency sound.
Although these approaches differ in detail, they share a common principle:
Low frequency noise should be assessed using frequency-specific measurements rather than relying solely on overall dBA values.
Relating Research to the 63 Hz Octave Band
Throughout this MDCO series, the 63 Hz octave band has been used as a representative indicator of low frequency noise because it captures one of the principal frequency ranges generated by large cooling towers and other mechanical infrastructure.
Based on a synthesis of published engineering guidance, psychoacoustic research and environmental studies, the following table provides a practical interpretation of 63 Hz octave-band sound pressure levels.
These values are not regulatory limits and should not be interpreted as fixed thresholds.
Instead, they summarise the approximate ranges within which the probability of perception and community response generally increases.
| 63 Hz Octave-Band Sound Pressure Level | General Interpretation |
|---|---|
| Below 25 dB | Usually below perception indoors for most people under typical conditions. |
| 25–35 dB | May become audible in very quiet environments, particularly at night or for sensitive individuals. |
| 35–40 dB | Increasing likelihood of awareness; some individuals may begin reporting persistent humming or tonal perception. |
| 40–50 dB | Increasing probability of annoyance and sleep disturbance among susceptible individuals, particularly during quiet nighttime conditions. |
| Above 50 dB | Higher probability of community complaints and increased likelihood of sleep disturbance, especially where exposure is prolonged or tonal components are present. |
These approximate ranges correspond well with the engineering modelling presented earlier in this series (→A09.02 Estimating Low-Frequency Noise from a Typical Data Centre).
For example:
- A representative 100 MW data centre under neutral atmospheric conditions is estimated to produce 63 Hz octave-band sound pressure levels that generally decrease below 40 dB at distances approaching 1 kilometre and beyond. Under favourable nighttime propagation conditions, however, levels above 40 dB may persist at greater distances, with the upper range remaining measurable beyond 1–2 kilometres, potentially increasing the likelihood of perceptibility or annoyance for some sensitive receptors.
- A larger 500 MW facility, with greater cooling and mechanical infrastructure, may produce higher 63 Hz sound levels. Under strong favourable propagation conditions, estimated levels may approach or exceed 50 dB at distances around 2 kilometres, although actual community response would depend on site-specific factors including system design, background sound conditions, building response and individual sensitivity.
These comparisons do not predict that every resident will experience disturbance.
Rather, they illustrate why larger facilities, favourable atmospheric conditions and proximity to residential areas deserve careful consideration during planning and environmental assessment.
Understanding where predicted sound levels fall within the ranges identified by decades of research allows engineers, developers, regulators and communities to have more informed discussions about mitigation, monitoring and site design.
MDCO Insight: Scientific research shows that the probability of annoyance increases as low frequency sound levels rise, but no single sound level guarantees either comfort or disturbance.
Why Some People Are More Sensitive Than Others
One of the most interesting findings in environmental noise research is that people exposed to the same sound do not necessarily experience it in the same way.
This observation has been reported repeatedly across laboratory experiments, community surveys and epidemiological studies.
In almost every study, some individuals report little or no disturbance, while others describe the same sound as highly intrusive.
This does not mean that one group is “right” and the other is “wrong”.
Rather, it reflects the fact that human response to environmental noise is influenced by far more than the sound itself.
Noise Sensitivity
Among all personal characteristics, noise sensitivity is one of the strongest predictors of annoyance.
Noise sensitivity is regarded as a relatively stable personal characteristic rather than simply a dislike of particular sounds.
People with higher noise sensitivity are generally more likely to notice environmental sounds, become annoyed by persistent noise and report greater disruption to daily activities under similar acoustic conditions.
Importantly, noise sensitivity is not considered a medical disorder.
Instead, it is recognised as a normal variation in human perception, much like differences in vision, taste or tolerance to bright light.
For this reason, modern environmental health research routinely considers noise sensitivity when interpreting community responses.
Expectations and Previous Experience
How people interpret a sound also influences how they respond to it.
Research has shown that expectations, previous experiences and attitudes towards a nearby facility can all affect annoyance.
For example, a sound that is understood, expected and regarded as temporary may be tolerated more readily than an unfamiliar sound whose source is uncertain.
Likewise, communities that receive clear information about industrial activities often report different perceptions from communities that feel excluded from planning or communication.
This does not imply that concerns are purely psychological.
Rather, it reflects the reality that environmental noise is experienced by people within a broader social and environmental context.
Existing Health and Wellbeing
Individual health also influences how environmental noise is experienced.
People already experiencing poor sleep, chronic stress, anxiety or certain medical conditions may be more vulnerable to additional disturbance.
Conversely, people who are well rested and living in otherwise quiet environments may find the same sound less intrusive.
Researchers therefore recognise that environmental noise interacts with existing physical and psychological conditions rather than affecting everyone identically.
Buildings and Indoor Acoustics
Perhaps surprisingly, people often notice low frequency noise more indoors than outdoors.
Buildings do not simply block sound.
They also alter it.
Depending on room dimensions, construction materials and window openings, buildings may reinforce particular low frequencies through resonance while reducing others.
This means that two neighbouring houses exposed to the same outdoor sound may experience noticeably different indoor acoustic environments.
Indoor conditions therefore play an important role in explaining why community responses sometimes vary within the same neighbourhood.
Background Sound
The surrounding acoustic environment also matters.
During the daytime, traffic, conversation, wind and other environmental sounds may partially mask low frequency noise.
At night, however, background sound levels often decrease substantially.
As the environment becomes quieter, persistent low frequency sounds that were previously unnoticed may become more apparent.
This is one reason why environmental noise assessments place particular emphasis on nighttime conditions.
Time of Day and Sleep
The same sound level can be experienced very differently depending on when it occurs.
During the day, people are generally engaged in work, conversation and other activities that compete for attention.
At night, particularly during periods of rest or sleep, persistent low frequency sounds may become much more noticeable.
This is consistent with the scientific evidence discussed earlier, which identifies sleep disturbance as one of the most consistently demonstrated outcomes of environmental noise exposure.
Age
Age may also influence response to environmental noise, although the relationship is complex.
As people grow older, sensitivity to higher frequencies often declines naturally.
Low frequency hearing, however, tends to be preserved more effectively.
At the same time, sleep patterns often become lighter with increasing age, potentially making nighttime environmental noise more noticeable.
Current research suggests that age is one of several interacting factors rather than a single dominant predictor of community response.
Taken together, these findings explain why environmental noise assessments cannot predict exactly how every individual will respond.
Engineering calculations can estimate sound exposure with considerable accuracy.
Human experience, however, depends on the interaction between the sound itself, the surrounding environment and the characteristics of each individual.
MDCO Insight: Environmental noise is measured objectively, but experienced individually, which is why the same sound may be barely noticeable to one person yet highly disturbing to another.
What Current Research Still Cannot Explain
Scientific understanding of low frequency noise has advanced considerably over the past five decades.
Researchers now have strong evidence linking environmental low frequency noise with annoyance and sleep disturbance, and a growing body of evidence relating long-term environmental noise exposure to broader health outcomes.
Nevertheless, important questions remain unanswered.
Recognising these uncertainties is an essential part of evidence-based science.
Why Are Some People Much More Sensitive?
Perhaps the greatest unresolved question is why some individuals appear substantially more sensitive to low frequency noise than others.
Differences in noise sensitivity, health, sleep quality and indoor acoustic conditions explain part of the variation.
However, they do not appear to explain all reported differences.
Some individuals consistently perceive or are disturbed by low frequency sounds that others scarcely notice under apparently similar conditions.
Researchers continue to investigate the biological and psychological factors that may contribute to this variability.
How Is Very Low Frequency Sound Processed?
The classical understanding of hearing describes sound as travelling through the outer, middle and inner ear before being interpreted by the brain.
Recent neuroscience research suggests that the biological processing of very low frequency sound may be more complex.
Studies published in recent years have observed measurable neural responses to infrasound and very low frequency sound even when participants reported little or no conscious hearing.
These findings have prompted researchers to investigate whether additional sensory pathways may contribute to the perception of very low frequency sound.
Several possibilities are currently being explored, including:
- responses within the cochlea beyond conscious hearing;
- contributions from the vestibular system, which is responsible for balance and spatial orientation;
- interactions between multiple sensory systems; and
- alternative neural processing pathways within the brain.
At present, however, these mechanisms remain areas of active scientific investigation rather than established fact.
What Happens During Long-Term Exposure?
Most laboratory experiments expose participants to sound for hours or, at most, several days.
By contrast, residents living near major industrial facilities may experience environmental noise over many years.
Scientists therefore continue to investigate how prolonged exposure influences adaptation, annoyance, sleep quality and overall wellbeing.
Some individuals appear to become accustomed to persistent environmental noise over time.
Others report increasing annoyance despite relatively stable sound levels.
Understanding why these different patterns occur remains an important research challenge.
How Should Individual Susceptibility Be Considered?
Modern environmental regulation is generally based on protecting the majority of the population.
However, scientific evidence consistently shows that individual susceptibility varies.
One of the continuing challenges for researchers and policymakers is determining how environmental guidelines should balance population-wide protection with the needs of particularly sensitive individuals.
There is no universally accepted answer.
Future advances in neuroscience, psychoacoustics and environmental epidemiology may improve understanding of why susceptibility differs and how assessment methods can better reflect these differences.
For now, most researchers agree on an important principle:
Scientific uncertainty should not be confused with scientific ignorance.
Many aspects of low frequency noise are already well understood, while others continue to be investigated using increasingly sophisticated techniques.
As new evidence emerges, engineering practice and environmental guidance will continue to evolve.
MDCO Insight: Modern research has answered many questions about low frequency noise, but important scientific uncertainties remain regarding individual susceptibility and long-term exposure.
What These Findings Mean for Different Stakeholders
The research reviewed throughout this article demonstrates that low frequency noise is neither a purely engineering issue nor solely a public health issue.
It sits at the intersection of acoustics, environmental science, psychology, planning and community engagement.
Understanding this evidence benefits every stakeholder (→E04 The Stakeholders Behind Every Data Centre) involved in the development, regulation and operation of large infrastructure projects.
Communities
For communities, scientific understanding provides an evidence-based framework for discussing environmental concerns (→A09.01 — Why Low-Frequency Noise Becomes a Community Conflict).
Residents who understand how low frequency noise is measured, assessed and studied are better equipped to:
- distinguish established scientific findings from speculation;
- communicate concerns using objective information;
- understand why some people may experience greater disturbance than others; and
- participate more constructively in planning and consultation processes.
Where concerns arise, careful documentation of observations, measurement data and environmental conditions is generally more valuable than relying solely on subjective impressions.
Data Centre Operators
For operators, low frequency noise should be viewed as a design consideration (→E03 Why Data Centres Are Built This Way) rather than simply a compliance issue.
Experience from many industries shows that addressing acoustic performance during the earliest stages of design is usually more effective—and significantly less expensive—than attempting to resolve community complaints after operations have commenced.
Practical measures may include:
- selecting quieter equipment;
- optimising cooling system layouts;
- incorporating acoustic attenuation;
- evaluating low frequency performance in addition to overall dBA compliance; and
- considering nearby residential areas during site planning.
Beyond technical performance, proactive noise management can also:
- reduce the likelihood of future complaints;
- minimise regulatory disputes;
- strengthen environmental, social and governance (ESG) performance;
- protect corporate reputation; and
- support long-term business continuity.
Consultants
For acoustic consultants, the research reinforces the importance of adopting assessment methods that reflect the characteristics of low frequency noise.
Where appropriate, this may include:
- octave-band analysis;
- assessment of tonal components;
- indoor as well as outdoor measurements;
- evaluation of nighttime conditions; and
- consideration of the 63 Hz octave band, where large cooling infrastructure often contributes significant sound energy.
Looking beyond overall dBA values may provide a more complete understanding of potential community exposure.
Regulators
For regulators, the scientific evidence supports a balanced and proportionate approach.
Conventional A-weighted limits remain appropriate for many environmental noise situations.
However, where substantial low frequency components are expected, supplementary assessment methods may improve the ability to identify and manage potential impacts.
Evidence-based regulation should seek to protect communities while remaining practical, transparent and technically defensible.
Policymakers
For policymakers, the continuing evolution of low frequency noise research presents an opportunity to strengthen future planning frameworks.
Areas worthy of continued attention include:
- planning guidance for major industrial developments;
- environmental monitoring requirements;
- transparency of acoustic assessments;
- community engagement during project development; and
- periodic review of technical guidance as scientific understanding advances.
As data centres continue to increase in scale and importance, integrating current scientific knowledge into planning and governance can help reduce future conflicts while supporting sustainable infrastructure development.
Ultimately, the value of this research lies not in creating unnecessary concern, but in enabling better decisions.
Communities gain greater understanding.
Operators gain clearer design objectives.
Consultants gain more effective assessment tools.
Regulators gain stronger evidence for balanced decision-making.
And policymakers gain a sounder foundation for developing future guidance.
MDCO Insight: Understanding low frequency noise benefits every stakeholder because preventing problems through evidence-based design is generally more effective than resolving disputes after facilities begin operation.
The Observatory Perspective
Scientific research has moved well beyond the simple question of whether people can hear low frequency noise.
Decades of laboratory experiments, community surveys, sleep studies, epidemiological investigations and engineering case studies have produced a substantial body of evidence describing how people respond to prolonged exposure under different environmental conditions.
The picture that emerges is more nuanced than either of the extreme positions often encountered in public discussions.
The evidence does not support the view that all low frequency noise is harmless simply because conventional A-weighted noise limits are satisfied.
Equally, it does not support the view that every source of low frequency noise inevitably produces serious health effects.
Instead, modern research shows that human response depends on the interaction of many factors, including:
- the acoustic characteristics of the sound;
- the duration and timing of exposure;
- indoor and outdoor environments;
- individual sensitivity;
- sleep;
- and the broader environmental context.
Among all reported outcomes, the scientific evidence is strongest and most consistent for annoyance and sleep disturbance.
Evidence relating to broader physiological and long-term health effects continues to develop, with several important questions remaining the subject of active international research.
For data centres, these findings do not imply that every facility will create unacceptable impacts.
Nor do they suggest that low frequency noise is fundamentally different from other engineering challenges associated with modern infrastructure.
Rather, they reinforce a practical conclusion.
Low frequency noise should be treated as an engineering design parameter—one that deserves the same systematic attention given to energy efficiency, water consumption, electrical reliability and structural safety.
Considering low frequency noise early in the planning and design process allows engineers, operators and regulators to identify potential issues before they become community concerns.
For communities, understanding the scientific evidence provides a more objective basis for interpreting environmental noise and participating in discussions about future developments.
As Malaysia continues to develop increasingly large and sophisticated data centres, a sound understanding of low frequency noise provides a stronger foundation for better engineering, more informed regulation and greater confidence among all stakeholders.
MDCO Insight: The question is no longer whether low frequency noise exists, but how scientific evidence can be used to understand, assess and manage it responsibly.
Selected References
International Guidelines and Standards
- World Health Organization (WHO) Regional Office for Europe. Environmental Noise Guidelines for the European Region. A comprehensive review of scientific evidence relating environmental noise exposure to annoyance, sleep disturbance and broader health outcomes. The guidelines provide evidence-based recommendations for managing environmental noise and protecting public health. https://www.who.int/publications/i/item/9789289053563
- International Organization for Standardization (ISO). ISO 1996 Series: Acoustics — Description, Measurement and Assessment of Environmental Noise. International standards establishing principles for describing, measuring and assessing environmental noise, including sound pressure levels at receiving locations and methods for evaluating acoustic environments. https://www.iso.org/standard/59765.html, https://www.iso.org/standard/59766.html
- Deutsches Institut für Normung (DIN). DIN 45680: Measurement and Assessment of Low Frequency Noise Immissions. A widely referenced technical standard for evaluating low frequency noise exposure using frequency-specific assessment methods. https://www.din.de
- UK Department for Environment, Food & Rural Affairs (DEFRA). NANR45: Procedure for the Assessment of Low Frequency Noise Complaints. A major UK research programme investigating real community complaints and developing practical procedures for assessing low frequency noise. https://www.gov.uk/government/organisations/department-for-environment-food-rural-affairs
Foundational Low Frequency Noise Research
- Leventhall, H. G. A Review of Published Research on Low Frequency Noise and Its Effects (2003) and Low Frequency Noise and Annoyance (2004). Foundational reviews bringing together engineering measurements, psychoacoustic research and community response studies. These works remain among the most widely cited references in environmental low frequency noise research.
- Broner, N. (1978). Early investigations into community complaints associated with industrial low frequency noise, highlighting the limitations of relying solely on A-weighted measurements where significant low frequency energy is present.
- Persson Waye, K., and colleagues. Research studies examining community response, annoyance and sleep effects associated with environmental low frequency noise, including exposure from wind turbines and industrial sources.
Sleep and Environmental Health Research
- Basner, M., Babisch, W., Davis, A., Brink, M., Clark, C., Janssen, S., & Stansfeld, S. Auditory and Non-Auditory Effects of Noise on Health. A comprehensive review of environmental noise research covering auditory effects, annoyance, sleep disturbance and broader health outcomes. https://doi.org/10.1016/s0140-6736(13)61613-x
- European Environment Agency (EEA). Environmental Noise in Europe. Periodic assessments of environmental noise exposure, population impact and associated health implications across Europe. https://www.eea.europa.eu
Emerging Neuroscience
- Recent neuroscience and vestibular research (2020–2025). Emerging studies using neuroimaging and physiological measurements have investigated how very low frequency sound and infrasound may be processed by the human nervous system. These studies suggest that perception of very low frequency sound may involve more complex biological pathways than previously recognised. However, this remains an active area of scientific investigation, and the findings should be regarded as emerging evidence rather than established scientific consensus.
Citation
Malaysia Data Centre Observatory (MDCO). A09.03 Low Frequency Noise Effects: What Research Shows. MDCO Analyse Series.
MDCO Note
This article forms part of the Malaysia Data Centre Observatory (MDCO) Analyse Series, which aims to improve public understanding of data centre development through evidence-based, accessible and balanced analysis. It is intended for educational and informational purposes only and does not constitute legal, engineering, acoustic, environmental, medical or professional advice.
Low-frequency noise is a multidisciplinary subject spanning engineering, acoustics, environmental science, medicine, psychology, planning and public policy. MDCO does not advocate for or against any particular stakeholder, project, technology or regulatory position. Its role is to facilitate transparency, reduce information asymmetry and support informed discussion by explaining complex issues through publicly verifiable information and multiple complementary perspectives.
