Home Email this page Print this page Bookmark this page Decrease font size Default font size Increase font size
Noise & Health  
Similar in PUBMED
   Search Pubmed for
   Search in Google Scholar for
 Related articles
Email Alert *
Add to My List *
* Registration required (free)  

   Article Figures
   Article Tables

 Article Access Statistics
    PDF Downloaded336    
    Comments [Add]    

Recommend this journal


REVIEW ARTICLE Table of Contents   
Year : 1999  |  Volume : 1  |  Issue : 4  |  Page : 11-36
Exposure to nocturnal road traffic noise: Sleep disturbance its after effects

Department of Engineering Acoustics, Lund Institute of Technology, P.O. Box 118, S-221 00 Lund, Sweden

Click here for correspondence address and email

In this paper the disturbances to sleep caused by road traffic noise are reviewed in the light of the latest published findings. First, a short presentation is made of what noise is in general. Then an exposition is made of the different characteristics of road traffic noise and how it may be measured and rated with various descriptors. In general terms, the continuous exposure of people to road traffic noise leads to suffering various kinds of discomfort, thereby reducing the number of well-being elements. However, this conclusion is made more complex to attain when non-acoustical factors such as socio-economic situation, age and gender are taken into account. In broad terms, nocturnal road traffic activity leads to difficulties in falling asleep for people and to a reduction of their sleep quality. This, however, depends strongly on physical measures of noise like for instance the intermittency of loud single noise events, their level relative to that of the background noise and the frequency and time of their occurrence. Several studies have also confirmed the fact that mood, too, is strongly affected after spending a night with significant noise exposure. Other psychological and physiological functions affected by night-time exposure to road traffic noise, such as performance the following day and cardiovascular reactivity are also reviewed.

Keywords: Road Traffic Noise, SleepDistrubance, Non-acoustical factors

How to cite this article:
Ouis D. Exposure to nocturnal road traffic noise: Sleep disturbance its after effects. Noise Health 1999;1:11-36

How to cite this URL:
Ouis D. Exposure to nocturnal road traffic noise: Sleep disturbance its after effects. Noise Health [serial online] 1999 [cited 2018 May 21];1:11-36. Available from: http://www.noiseandhealth.org/text.asp?1999/1/4/11/31726

  Introduction Top

In modern times, noise is recognised as a serious health problem. Annoyance caused by noise has been known since antiquity, but it is only during recent times that the importance of environmental factors is taken into account in transport planning decisions. This is obviously necessary when facing the growing pollution problems caused by different agents, among which noise stands for a non-negligible share. In fact, of the environmental pollution factors that are affected by the use of transportation means, noise is perhaps the most commonly cited. This problem is exacerbated when one knows that nowadays the number of vehicles circulating in the urban network of roads is steadily increasing while at the same time the number of quiet hours during night-time has a tendency to diminish, although at a somewhat slower rate. Hence, expressed in terms of social cost, the adverse effects on people of noise in general and of traffic noise in particular result in a reduction of their elements of well-being. The technical problems associated with the design of quiet vehicles are still incompletely solved, and unfortunately so is the confinement of the noise emitted by these vehicles within the limits of what is bearable. Another important element to be taken into consideration is the complexly subjective human sensitivity to noise which requires the achievement of considerable noise reductions before its benefits are felt. For instance, halving the acoustic power of a sound source results in only a 3 dB reduction of the sound level, and this is scarcely noticeable to the hearing of an average listener.

In general, every noise problem involves a system of three basic elements: a source, a transmission path and a receiver. When possible, the best way to remedy the exposure to undesirable noise, both economically and aesthetically, is to control the noise emission at the source itself. But for an already existing source, the most corrective measure is making changes in the path. Moreover, different noise sources may have different acoustical characteristics. While some generate a pure tone, others may radiate a random noise with a more or less known spectrum. So in this respect the definition of the noise problem is important. For traffic noise, an automobile generally has several noise generating sources. Because of commercial limitations, the very competitive car industry has serious difficulties in the matter of noise reduction cost (see for instance Aspinall, 1970; Mills and Aspinall, 1968; Rathe et al., 1973). But despite this, in recent years much has been achieved in the car industry for the construction of quieter engines and better mufflers. However, the dominant noise source in the automobile remains that one due to the tyre, which is situated at the ground level. For traffic noise, not only the source is important; it has been known for many years that driving habits affect the degree of annoyance to a no less important degree. Data from urban traffic experiments have shown, for instance, that sudden accelerations have negative effects on traffic noise control, and good road planning with beneficial effects would require the setting of speed limits and the smoothing of traffic flow (Jones and Hothersall, 1979; Lewis, 1973; Waters, 1970). Tyre noise reduction depends more on the absorption properties of road materials than on the road-tyre contact mechanisms (Nelson and Abott, 1987; Watts 1996).

An automobile has a typical normalised noise spectrum as shown in [Figure - 1]. This feature of the spectrum, with a heavier frequency content at around 1 kHz, is also shared by the spectra of lightweight trucks. On the other hand, heavy trucks show different noise emitting behaviour because of their larger number of noise sources. Besides the flow noise and the noise due to tyres, a heavy truck has several other noise generators (see e.g. Alexandre et al., 1975; Olson, 1972; Priede, 1971; SOU, 1974) which among others are summarised in the often highly placed exhaust, the engine (at a height of about 1m) and the side panels. The combination of all these sources results in a noise spectrum which lies higher and is much broader than that of an automobile, see [Figure - 1]. It is worthwhile to note that from the spectra shown in this figure one cannot see that a typical heavy truck may cause a noise level 90 dB more than that of an automobile (Hayek, 1990).

Facts and concepts about noise.

Before entering into the subject of the effects of road traffic noise, and also for later reference, it is useful to review some important facts related to sound and its quantification and to present some notions which describe noise in general and traffic noise in particular.

The nature of sound

Sound is the result of the propagation of a disturbance from a source in an elastic medium, usually air. The radiant nature of sound makes it obey the law of radiation, i.e. that its intensity decreases in proportion to the square distance from the source. Usually sound does not propagate from a single source, but is more likely to originate from a distributed source like a machine. It is sometimes possible to trace the larger source to a collection of simple sources, but at comparatively large distances from it, sound can be regarded as if originating from a single source. Sound propagates in air with the typical properties of longitudinal waves, that is the particles of the medium move in the same direction as that of wave propagation. This motion is characterised by a wave speed c and a wavelength 'λ. The speed of sound may be calculated according to

where P is the ambient pressure, γ the ratio of specific heat at constant pressure to the specific heat at constant volume of the gas (equal to 1.4 for air) and ρ its density. The accepted value of c for air at normal conditions is c = 340 m/s.

If a source is radiating sound at a single frequency f (in Hz), then the wavelength 'λ of the radiated sound which is the distance between two points having similar particle states is given by:

Sound pressure level

The physical quantity that is generally of interest in noise quantification is sound pressure, which is the incremental pressure due to the passage of the wave and which oscillates above and below the ambient pressure. The sound pressure is, even in the limits of ear pain, very small compared to the static air pressure. A quantity fluctuating with time, like sound pressure, is made up of a series of rapidly varying positive and negative values. These are usually measured by an instrument which is intended to present a statistical value, the so-called rms (root mean square) sound pressure prms without considering its instantaneous time variation. This is given according to

where T is a period of time long enough to permit the build-up of the statistical process. It is this rms value of the pressure that is usually of concern, and the rms index is thus omitted from the pressure symbol. The sound pressure is not given by its absolute value, but rather with reference to some quantity which is conventionally the pressure at the average limit of audibility of a normal healthy subject at 1 kHz. Hence, the sound pressure level Lp is defined as

where p ref =2.10 -5 Pa and the logarithm is taken to cover the large scale of human pressure sensitivity.

[Figure - 2] gives a presentation of some of the most usual noise sources arranged according to their frequency content and their position on a sound level scale.

Noise ratings

Over the years, several noise scales have been introduced in an attempt to give a more or less qualitative assessment of exposure to noise. In fact, as the problem of noise disturbance is a subjective matter, the first task is then to translate and process the measured quantities according to the perceptive sensitivity of human hearing. (For equal sound pressures, higher frequencies tend to appear much louder than lower frequencies.) This is accomplished by the A-filtering of the pressure, a procedure which gives a lighter weight to the lower frequencies over the higher ones as measured by a linear microphone. When sound pressures have been measured according to the A-filtering mode, their level is given in dB(A). [Figure - 3] presents the curves of equal loudness contours for pure tones and the different international standard weighting curves. Regarding traffic noise, the definition of sound pressure level, SPL, has been somewhat refined to take into account the time­dependent character of the traffic flow (whether it is occurring by day or by night) and also on the duration of the noise itself. The A-weighted SPL, however, is the basic in most of these measures, and this is marked by a p A instead of the pressure p. The Equivalent Continuous Sound Level, ECSL, or L eq , is one of the simplest of these measures and is the A-weighted SPL over a specified time of measurement T.

where T is the averaging time which is usually taken as 1, 8, 12 or 24 hours.

The Day-Night Average Sound Level, DNL, L dn is equivalent to Leq for a 24-four day with an extra 10 dB weighting for noise occurring between 10 p.m. and 7 a.m. to account for the extra nocturnal noise annoyance. Formally DNL is given by

where L d and L n are respectively the 15-hour daytime and 9-hour night-time A-weighted equivalent sound levels.

The Statistical Level, SL, L N is suitable for a stationary random noise which, in the case of traffic noise, is satisfied only in the event of a free flow of vehicles. An example is illustrated in [Figure - 4]. A time-varying noise level measured in dB(A) can be described in terms of its cumulative distribution. From [Figure - 4]-right one can determine the level which is exceeded for a particular percentage of the total time. Usual values are L 10, L 50 , and L 90 which are respectively the levels exceeded for 10, 50 and 90% of the time. These features are shown in [Figure - 4]-left.

With some restrictions on the vehicle flow rate and on the measurement time, a relatively simple empirical relationship has been found between L 10 and L eq , namely (Nelson, 1987)

There are numerous other noise ratings serving different purposes and which had their originated in different countries, but although used for road traffic noise they have mainly been adapted to aircraft noise rating.

Acoustical characteristics of road traffic noise

In the built-up environment of the city, road traffic is the main source of noise. Under the plausible assumption that small vehicles are in numbers the greatest contributor to urban traffic, the noise generated by a small vehicle can be thought to have four different origins: the engine, the exhaust, the tyres and the air turbulence. Other sources of noise like fan and structure are likely to be less important in our case; these are reviewed in (Alexandre et al., 1975, Nelson, 1987). Noise due to tyre-road contact becomes significant only for relatively high driving speeds, so this can be neglected at the limited car speeds in town. For the same reason, turbulence also contributes little, and one is thus left only with the noise from the engine and from the exhaust of the vehicles. Traffic noise can in general take two ways to reach subjects living in an apartment with windows facing the street: one way is that the waves generated by the vehicle are transmitted directly through the air to the windows of the residence or set the building into vibration, and the other way is that the rolling tyres of the vehicle induce vibrations to the road beneath which, when transmitted to the building through structural contact, generate sound waves inside the building, [Figure - 5].

In practice, people have different feelings for building vibrations, ranging from sensing the vibrations of the structure to aural perceptions such as the sound resulting from the rattling of ornaments, or a sensitivity to low frequency noise at certain levels.

Traffic noise is from a mixture composed of different vehicles, light and heavy ones, running randomly in a city street or in the different lanes of a highway. The usual sounds. For the coherent summation where the addition operation is instead made on the pressures, the result of the summation would be 6 dB.

Road traffic noise is made of two components: the noise generated by the whole stream of cars and that generated by each individual car. It is important to make a distinction between these two types of noises because in the overall noise level, the peaks (which may be defined by L10 or L1) are due to the passing of individual vehicles. These individual noise sources behave as simple point sources, i.e. their corresponding peak sound level decreases at the rate of 6 dB per each doubling of the distance from them (the sound pressure is inversely proportional to the distance). On the other hand, the whole stream of cars has similarities with a line source of sound. Its corresponding noise level, which may be considered as the background noise and which can be described by L 90 or L 99, decreases only by 3 dB for each doubling of the distance (the sound pressure is inversely proportional to the square root of the distance). This is illustrated in [Figure - 6] where the variation with distance of LN for different values of N is shown.

From extensive measurements on different car models, it has been confirmed by Rathe et al. that in principle the noise level from the engine and the exhaust increase in proportionality to the logarithm of the speed of the vehicle and that a twofold increase in the speed implies an increase of about 10 dB(A) in the maximum noise level (Rathe et al., 1973). From this last study one also finds that noise increases as a function of load, and that it increases with acceleration at low speeds of the car particularly when starting to move. For the directivity characteristics, and due to the ground reflection, the sound radiation is predominantly along 20 to 40 degrees above the horizontal.

The measurement of traffic noise from streams of vehicles is more complicated than the measurement from single vehicles as it involves many operations, like statistical analyses and integration procedures which require longer to be performed. The distribution of noise levels in the case of heavy and steady traffic approaches nearly that of a Gaussian distribution. From the knowledge of two parameters, for instance the median level L50 and the standard deviation 6, one can make an estimate of different noise parameters (Alexandre et al., 1975). As regards the inclusion of heavy vehicles in the traffic stream, several curves have been proposed from different studies for the correction of the noise level due to light traffic as a function of the extra percentage number in trucks. In [Figure - 7], two traffic noise time histories are shown to demonstrate the difference between the noises from typical city centre traffic and traffic from a typical highway. One can see clearly the effects of congestion causing acceleration in the emergence of occasional high peaks in the case of urban traffic noise, whereas the highway noise has a steadier level because of the constant cruising speed and the relatively short time between successive vehicles in the stream.

The analysis of the noise level distribution also reveals a difference between the two noises, namely that the distribution of the highway noise is more symmetrical and narrower than that of the city centre (Hassal and Zaveri, 1979).

The frequency composition of the noise is also important in the matter of sound insulation. Unfortunately, traffic noise has a large contribution in low frequencies, especially at about 60 Hz, and in this range of low frequencies sound insulation is most difficult to achieve in both structural and airborne sound insulation. In [Figure - 8] one sees typical spectra of traffic noise registered outside and inside a building, and one can see that large differences between the indoor and outdoor noise levels occur only at relatively high frequencies.

It might be interesting to point out that one direct effect of the low frequency noise on people is the excitation of body resonances. It has in fact been found that depending on individual characteristics, people have a chest resonance lying in the range 30-90 Hz. More precisely, a study of a group of male and female subjects resulted in an estimation of the average chest resonance of 64 Hz and 74 Hz for females and males respectively. Another study using subjects breathing helium/oxygen instead of air showed no shift of the resonance frequency, indicating that the chest resonance is a structural effect rather than an effect due to air-filled cavities. The effect of chest resonance excitation is best conveyed in the common experience of pedestrians walking alongside heavy vehicles, especially during start-up, but for people staying indoors, although chest vibrations could be perceptible, noise level is the more general complaint (Nelson, 1987).

Interference of noise with daily human activity

Generally, the term noise is used to designate any undesirable sound. As people may not experience all undesirable situations as dangerous, a definition containing the word undesirable becomes ambiguous. In acoustics one distinguishes between two kinds of sounds: noise and signal. The latter, which is of more interest in measurements, has particular characteristics both in time and frequency which are usually lacking in the former. But in the case of traffic noise this ambiguity persists because of the fact that although random, traffic noise has its distinguished spectral and temporal uniformities. This problem of definition pertains also at the biophysical level. A radio turned to a high volume may convey perfectly meaningful information to the involuntary listener who nevertheless perceives it as an annoying event. Thus with this signal-noise consideration it seems difficult to set a line of demarcation between what is considered as a useful and desirable signal and what could be considered as a useless and disturbing noise. In the complex realm of stimuli where people are the receptors, the differences between individuals in assessing the noise effects are made more difficult by many acoustical and non-acoustical factors.

Among the important acoustical factors one can cite pressure level, duration of exposure, frequency spectrum, impulsive character and level fluctuations, whereas the non-acoustical factors may include time of day, time of year and past experience. One can also add the physiological and psychological states of the person. This domain becomes even fuzzier when one takes into consideration people's social attitudes and their belonging to different cultural groups (Alexandre et al., 1975; Griffiths and Langdon, 1980; Relster, 1975). So, with respect to taking into account the many different components of the reaction to noise, it is important to realise that the relationship between measured noise level and its effect on people is not easy to determine in a systematic way so as to be able to build a sort of "annoyancemeter". The amount of literature written on the subject of the various negative effects of traffic noise on people whether in the form of articles, books or reports, has grown appreciably over the years. Broadly speaking, the effects of noise on people can be divided into three main categories: psychological, social and physiological. This can be illustrated by a simple imaginative experiment where, as the SPL of a noise is increased, its effect ranges from attitudinal to behavioural and lastly to physiological effects. Thus at levels above 130 dB, noise can cause temporary deafness, intense pain or, in the extreme case, damage of the inner hearing system. The auditory effects of noise on people have been quite well-known for some decades ago (see for instance Kryter, 1985) but, becoming a relatively accessible personal need, cars are invading the urban landscape more and more and contributing to a higher level of noise pollution than any other man-powered engine. Hence, most of today's research on noise control is confined to that from transportation with special emphasis on that of urban traffic. One should note that until the late seventies or so, active research on the hazardous effects of traffic noise on people was focusing most on auditory­related topics without however neglecting non­auditory health effects.

[Figure - 9] presents a simplified model for the main relationships between noise, its effects and the social context of people (some other models may also be found in (Job, 1996) and in (Lercher, 1996b)).

From the diagram in this figure, noise is represented as causing some direct effects or more delayed reactions in the form of annoyance. The set of personal characteristics may also cause some annoyance but not necessarily physiological reactions. In its turn, annoyance may lead to some obvious change in one's actions, e.g. shutting the window to isolate the source of noise, or it may engender less evident effects in the form of emotional reactions.

It is a well-known fact in all societies worldwide that noise is a serious environmental pollutant. Governments, especially in the industrialised world, aware of this danger up to a level of litigation (Bryan and Tempest, 1973; Fidell, 1996), have set organisations and commissions operating for the regulation of people's exposition to noise. In the case of traffic, not only inhabitants of residential areas are subject to noise nuisance but also vehicle drivers are exposed, both to the surrounding traffic noise and to the noise from their own vehicle. However, it is not an easy matter to predict a community's reaction to a specific street or highway noise traffic level based on simple quantitative measures. From the late sixties and for more than a decade thereafter, many publications have presented the results of studies on traffic noise annoyance in the main large cities of Europe and the USA (it is worth noting in passing that Sweden was among the leading countries in this important large-scale project (Fog and Jonsson, 1968)). The major aim of collecting data over the different physical characteristics of urban noise was to assess the importance of the problem and to process a suitable noise rating which would give subjective judgements from simple objective measures. As mentioned earlier, this mission is not easy to fulfil in the case of noise annoyance when so many non-acoustical variables (age, sex, social status, education, etc.) have been found to play roles that cannot be neglected.

Sleep disturbances caused by nocturnal road traffic noise

Noise in general can cause interference with many human activities. An important part of time of people's diurnal cycle is filled with sleep, which takes on average one third of the day. It is thus important, especially for working people, to enjoy a good night's sleep to recover from the tiredness of the day gone by and to ensure efficacy and good performance for the coming day. Besides darkness and the right amount of heat, silence is a necessary element for a restful night. It is therefore usually normal for people dwelling in lively city districts to close their windows before going to bed (sometimes also guaranteed by the cold of the winter). In apartment blocks, internal regulation forbids tenants from turning their radios or television sets above some level after a certain hour in the evening. Moreover, people living in noisy quarters of the city or in the vicinity of airports and railroads often complain of irregular sleeping habits or sleep disturbances. This has been proven by the results of recent studies where the residents of such areas have a tendency to suffer from various physiological symptoms like headaches and nervous stomach disorders (Ohrstrom, 1989) and to consume more sleeping medicaments than do people in quieter areas (Lercher, 1996a and Wanner et al., 1977). [Table - 1] which is taken from (Ohrstrom, 1991) summarises the comparisons between some different noise-related parameters as registered in two different areas from the same city, one noisy and one quiet.

In Sweden, the environmental authorities at both the local and the national levels are aware of this problem and of its effect on the well-being of the citizens to a point that they have generously encouraged research in this field since the early eighties. One can mention the research teams at the Departments of Environmental Health at the Universities of Lund and Gothenburg and the efforts of whom have led to the publication of numerous valuable information on the various aspects of sleep disturbances caused by traffic noise. In other parts of the world, one can also mention Carter and his co-researchers in Australia, as well as others who for many years have surveyed several different health aspects of noise in general and of traffic noise on sleep in particular (Brown, 1994; Carter and Beh, 1987; Carter and Lockington, 1991; Carter et al., 1976; Carter et al., 1994). In the USA, Fidell, Pearsons et al. also reported on results from field studies conducted mainly on airport neighbours (Fidell and Teffeteller, 1981; Fidell et al., 1980; Fidell et al., 1995; Horonjeff et al., 1982), and on the synthesis of other studies in the hope of developing some reliable quantitative model for the prediction of noise-induced sleep disturbances (Pearsons et al., 1995). It would also be worthwhile mentioning other research investigators, either as individuals or in teams, in Germany (Griefahn, 1991; Griefahn and Gros, 1986; Griefahn et al., 1993), France (Vallet et al., 1983; Vallet et al., 1988, Vernet, 1983) and more recently in Japan (Izumi and Yano, 1991, Kabuto and Kageyama, 1994; Kageyama et al., 1997, Osada, 1991; Suzuki et al., 1997, Yoshida et al., 1997). The primary objective from these and related investigations is to find the characteristics of traffic noise which adversely affect sleep quality in view of setting requirements and limitations in the matter of nocturnal traffic activity.

Main characteristics of sleep and experimentation on sleep

A sleeping subject cannot be asked to answer a questionnaire or to perform a specific task. One relies instead on registering the physiological reactions using a sleep polygraph comprising appropriate sensors like the Electroencephalogram, EEG, and recording REM, Rapid Eye Movement, and muscle tone under the effect of some external action (Carskadon and Rechtschaffen, 1989; Carter, 1996; Carter, 1998; Naganuma et al., 1991). It is only under extreme excitations for arousing the subject he or she is asked to fulfil a specific action, thus assuring of his full wakefulness (Thiessen, 1978; Ohrstrom and Rylander, 1982). When asleep, the peripheral sense organs of a person appreciably reduce their sensitivity. This means that the organism increases the threshold of the sensations, and it is well-known that to awaken a sleeper, a more intense sound is needed than that needed to initiate a reaction while the person is awake. However, both the central and the autonomous nervous systems can react to external stimuli without causing awakening. Another important aspect of human reaction to sound stimuli under sleeping state is that organs may react very differently and this applies particularly to cardiovascular responses. Indeed, diNinsi et al. found that under sleep, cardiovascular responses are more frequent and are of larger amplitudes than under waking state. These differences were further enhanced by the fact that the noise level used in their experiments was reduced by 15 dB(A) under sleep conditions, and that the amplitude of the heart rate response, HRR, was not affected in the case of female subjects. This increase of the cardiovascular sensitivity to noise during sleep may be attributed to some induced autonomic response. Hence, during the waking state, noise is partly processed by the brain in term of its meaning, whereas during sleeping this mental processing is attenuated (diNisi et al., 1990).

During a typical sleep period at night, one can distinguish in general between five different stages of sleep based on observations of the EEG and of certain eye movements. These stages tend to occur sequentially and periodically as illustrated in [Figure - 10]. Stage 1 is a short hypnagogic period of time between the awake state and stage 2, which is the most dominant one and occupies about half of the sleeping time. Stages 3 and 4 combine in what is called the Delta stage and constitute deep sleep, which fills roughly the first third of the sleeping period. Lastly, the REM stage is usually associated with dreaming. This last stage tends to occur more frequently and with longer periods of time during the last half of the sleeping period.

A short well-documented review of the outcomes of noise on sleep in general may be found in Lukas' article (Lukas, 1975). The results of the studies which are presented were based on middle-aged or young healthy individuals during periods of normally eight hours' sleep at night. The experiments were made either in the field in the form of pilot studies or under laboratory conditions where the noise from daily traffic is presented artificially through loudspeakers. It is important to note that the results of tests conducted under laboratory or under field conditions may be very different. In fact, from reviewing and comparing several such studies, Pearsons et al., (1995) could synthesise large and systematic differences in findings which may be attributed only to the different character of the studies. Moreover and probably due to problems of habituation to new environments, laboratory conditions introduce some uncertainty to the validity of results on sleep experiments (Thiesen and Lapointe, 1978), and noise disturbs sleep in the laboratory studies to a larger degree than it does in the field studies (Pearsons et al., 1995). From early studies, it was believed that the noise-induced disturbances on sleep at night could be classified under two types of reactions: primary effects which are recorded directly after the onset of the stimulation and secondary effects which comprise a variety of psychophysiological effects appearing the following day or later (Griefahn and Muzet, 1978). This differentiation between the two effects is based on the observation that sounds which awaken a sleeper or delay the onset of sleep have measurable effects on task performance and social well-being (Alexandre et al., 1975). As a first important outcome of the reviewed work, the World Health Organisation, WHO, recommends a noise level at night of no more than Leq=35 dB(A) (WHO, 1980). It is in general a fact that several studies support this specification, though many of them suggest other details. A late study by Berglund and Lindvall contains a recommendation for limiting the level of indoor continuous noise at the lower value of L eq =30 dB(A) (Berglund and Lindvall, 1995).

Effects on sleep onset

For a person whose bedroom windows face the noise coming from a heavily trafficked street, the time needed to fall asleep is one of the first symptoms of a restless night. This aspect of a noise-disturbed sleep may be monitored in the affected population by the need of using ear plugs or taking sleeping pills (Lambert et al., 1984; Lercher, 1996a, Wanner et al., 1977; Ohrstrom, 1991). Thus, the reasonable complaint is often made that traffic noise increases the time needed to fall asleep (Thiesen and Lapointe, 1983). The time required to fall asleep is also in close correlation with the cessation of movements as sleep sets in (Ohrstrom and Rylander, 1982) and it is also correlated to sleep quality (Ohrstrom and Bjorkman, 1988). Ohrstrom and Rylander, exposing sleeping subjects to 4, 8, 16, and 64 noise events per night found that the time to fall asleep increased monotonically with the number of events, [Table - 2], (Ohstrom and Rylander, 1990).

In a later study using instead 16, 32, 64 and 128 events per night, the time to fall asleep increased also up to when the number of events reached 64. At this number the sleep latency and the difficulty in falling asleep increased significantly. At 128 events per night no significant changes were found for these variables than in the case of 64 events (Ohrstrom, 1995). Lambert et al. further point out that nocturnal noise appears to have a much more pronounced effect on waking people up than in preventing them from sleeping with this latter symptom being also closely related to the age of the subjects (Lambert et al., 1984). Several other studies, some of which were performed for evaluating the effect of reducing indoor noise through double glazing the windows of the bedroom showed that reducing noise in the bedroom, leads in fact to a shorter time needed to fall asleep for both adults or young adults (Griefahn and Gros, 1983; Ohrstrom, 1989, Ohrstrom, 1991; Ohrstrom et al., 1988) and for children (Eberhardt, 1988). Road traffic noise during the early hours of a night also tends also to disturb sleep more than in later hours of the night (Eberhardt, 1988). A quite recent study shows moreover that living too near densely trafficked roads is a risk factor for insomnia, and that this risk increases linearly with the traffic volume (Kageyama et al., 1997).

Influence on sleep stage pattern

Although sleep may be defined as a condition of isolation and detachment from the outside world, the sleeping brain is extremely sensitive to acoustic stimuli. In the awake state, sounds do not become noise until processed by the brain in terms of their meaning to the subject. However, the physiological responses of a sleeping subject to sounds reflect the magnitude of the auditory stimuli (Terzano et al., 1993). In the literature review made by Lukas on the variables that appear to affect human responsiveness to noise during sleep, there seems to be some evidence on the importance of age on the specific distribution of responses. This finds support from Griefahn and Gros' research results, who stress the general importance of personal factors like age and sex (Griefahn and Gros, 1986). Carter and Lockington even find evidence of differences for both EEG pattern and sleep stage responses between individuals within the same age group (Carter and Lockington, 1991). As to the architecture and cyclical character of the sleep stages throughout the night, and because stage Delta occurs infrequently during the late hours of sleep [Figure - 10], the relative responsiveness to noise during the different stages is confounded with the time of night and number of hours of accumulated sleep (Lukas, 1975). Hence, investigating people living near a noisy road, Vallet et al. found that noisy conditions at night affect both the total sleep time and the durations of the various sleep stages, and that these changes may be chronic (Vallet et al., 1983). Taking the age factor into consideration, young adults may suffer mainly from stage 3 and 4 deficits, whereas in older subjects a REM sleep deficit is more to be noticed (Vallet et al., 1983). In an attempt to validate a physiological measure of sleep quality, Wilkinson and Campbell also confirmed the increase of mainly stage 4 sleep of their subjects when the noise level in their bedrooms was lowered by 5.8 dB(A) from Leq =46.6 dB(A) after double glazing the windows (Wilkinson and Campbell, 1984). All these studies were carried out on people in their homes where they had been living for more than one year. Eberhardt et al. conducted a study on young adults under laboratory conditions and concluded that continuous traffic noise, even though of relatively low level (45 dB(A)), negatively affects mainly REM sleep, whilst intermittent noise influences slow wave sleep (SWS, stage 3 and 4). Thus the occurrence of changes in sleep stage distribution seems to be influenced more by the noise level than by the emergence of noise peaks from the background noise (Eberhardt et al., 1987). Thiesen and Lapointe, who experimented exclusively on deep sleep, found that an exposure to eight noise events per night resulted in an average 3% reduction of deep sleep (Thiesen and Lapointe, 1978). In a later investigation, this time using continuous traffic noise, the effect of noise was instead an increase in the duration of deep sleep. The subjects were moved from a quiet laboratory environment (32 dB(A)) to a noisier one, and the fraction of deep sleep increased on the average by 2.5% and 4.6% respectively for 47 dB(A) and 60 dB(A) noise levels (Thiesen and Lapointe, 1983). Kawada et al. experimented on 20-year­old subjects with truck noise of different levels and concluded that the decreases of stage 1 and frequency of awakenings during exposure later at night are an indication of evidence of relative increase in deep sleep (Kawada et al., 1997). Eberhardt and Akselsson also demonstrated the beneficial effect of sound insulation on the sleep of young adults in their homes through an increase in the duration of SWS with emphasis on a shorter latency of stage 4 combined with its later occurrence during the night (Eberhardt and Akselsson, 1987).

Level and intermittency of nocturnal road traffic noise

Eberhardt, experimenting on groups of different ages, states that a continuous noise of 36 dB(A) had no effects on the sleep of his subjects (Eberhardt, 1988), whereas in the case of discontinuous noise he proposes a supplementation of this criterion by a maximum level in the form of some L 1 for instance (Eberhardt et al., 1987). This proposal is qualitatively supported by Ohrstrom and Rylander, who assert that continuous noise has smaller effects on sleep quality than intermittent noise (Ohrstrom and Rylander, 1982). The prominence of peaks in the time history of the noise has been anticipated as a major cause of sleep disturbances. However, not only the peak level is important in this respect. It has also been noticed that longer individual noises with lower levels affect sleep quality negatively more than the shorter booms of higher levels (Thiessen, 1978). The important factors to take into consideration in this context are the number of isolated events and their level relative to that of the background noise. Considering this latter case, it has been found that at the extreme of producing arousal reactions in sleeping subjects, the emergence of peaks from the background noise is definitely more important than the peak level. Moreover, continuous and intermittent noises, although of equal L eq, have different effects on the different stages of the sleep cycle. While continuous noise affects REM activity, discontinuous noise causes mainly stage 3 and 4 deficits (Eberhardt et al., 1987). Vallet et al. confirmed the necessity of a 35 dB(A) maximum noise level in the bedroom as recommended by the WHO in 1980, as they found that sleep quality becomes impaired for a continuous noise level of Leq =37 dB(A) or if the maximum level of the single events in the discontinuous noise exceeds 45 dB(A) (Vallet et al., 1983).

From [Figure - 10], and noting that deep sleep in stages 3 and 4 is essential for sleep quality, one can easily conclude that generally discontinuous noise is more unfavourable for sleep. This is in complete agreement with Eberhardt's finding that road traffic noise during the first hours of a night's sleep disturbs sleep more then if the disturbance occurs in the late night hours (Eberhardt, 1988). In a companion paper, Eberhardt and Akselsson, after studying cases at their homes, report that if passing cars produce a level 50 dB(A) higher than the quiet level in the bedrooms, then there is a good correlation between the number of disturbing vehicles and the relative worsening of sleep quality (Eberhardt and Akselsson, 1987). This result was also confirmed by the outcome of a related laboratory study, using this time recorded truck noise (Ohrstrom, 1995) which is diagrammatically presented in [Figure - 11]. In situations where awakening may be caused by intrusive noises in the bedroom, several studies conducted under normal settings confirm the importance of noise detectability, i.e. the relative level of noise event to background noise. Moreover, the temporal character of the exposure has a major effect on awakening, and this gives some credibility to also considering the energy content of sleep disrupting noises (Horonjeff et al., 1982), although noise exposure based only on energy considerations may give erroneous assessments of negative effects (Berglund and Lindvall, 1995). A recent study made by Fidell et al. on a large population sample exposed to different environmental noises suggests considering the influence of the noise level rather than long term noise exposure levels on awakening reactions(Fidell et al.,

Through exposing sleeping subjects to different numbers of stimulations with varying levels, sleep quality has decreased significantly as a result of highly noisy events -60 dB(A)- whereas at 50 dB(A) no correlation between number-of­events and sleep quality was found. These same observations were also made by Naganuma et al. using rectangular intermittent noise on their test subjects (Naganuma et al., 1991). Ohrstrom et al. found moreover that there is a threshold of noise level for single events to have an effect on sleep quality (Ohrstrom et al., 1988). From these and related results, it has been suggested that sleep quality could be monitored by the number of single events exceeding a certain level (Thiessen and Lapointe, 1978; Ohrstrom and Rylander, 1990). All these papers support Tulen et al.'s earlier qualitative judgements (Tulen et al., 1986).

Body movements

In normal subjects, sleep quality can also be assessed through the motional activity of the sleeper. This activity is usually recorded by means of accelerometers either fastened to the sleeping facility or worn on the sleeper's wrist. It is already known from early research on sleep that a close connection exists between the frequency of body movements and sleep stages. With reference to quieter nights, it has been found in general terms that the total number of body movements increases during noisier nights (Eberhardt and Akselsson, 1987; Eberhardt et al., 1987; Naganuma et al., 1991; Ohrstrom, 1995; Ohrstrom and Rylander, 1982). [Figure - 12] represents a typical time variation of the frequency of body movements during two nights, one quiet and one with an intermittent noise at 70 dB(A).

Ohrstrom and Bjorkman studied the opposite effect, and found that their subjects exhibited less body movements under quieter nights after double glazing the windows of their bedrooms (Ohrstrom and Bjorkman, 1983). However, drawing conclusions for sleep quality from simply counting body movements can be misleading, as different methodologies are used in different investigations and supplementation from other indices is necessary (Ohrstrom and Rylander, 1982). For instance, from comparing the results of in-situ and laboratory experiments, laboratory experiments have led to the observation of no significant differences in the total number of body movements between noisy and quiet nights. However, the number of body movements during noise exposure sessions or at the time of falling asleep was significantly higher in noisy situations (Ohrstrom and Bjorkman, 1988; Ohrstrom, 1995; Ohrstrom and Rylander, 1990). Nonetheless, a quite recent study gives convictions about the importance of peak noise levels to the number of body movements and sleep quality (Ohrstrom et al., 1988). Although sleep disturbance can be partly assessed in some cases through quantifying the number of nocturnal body movements, this information is not equivalent to other EEG data which may still persist even in situations where motor response diminishes due to adaptation or habituation (Eberhardt et al., 1987). Another aspect of body movement is the full awakening of the sleeper. Several studies have in fact led to the common conclusion that at higher numbers of disturbing isolated events, the subject presents more resistance to falling asleep again (Alexandre et al., 1975; Griefahn and Gros, 1986; Ohrtstrom, 1995).

Performance of tasks the following day

Often, after the test night, test subjects are also required to fulfil a performance test in association with filling in a questionnaire. These tests are usually short and are aimed in determining the vigilance of the subjects. Most common is the reaction test, RT, consisting of asking the subject to respond as quickly as possible to a specific stimulus (Carter, 1996). Updating the review of literature on the general after effects of sleep disturbances on health, Griefahn and Muzet conclude that sleep deficits accumulate with persisting disturbances and that, at a certain point, they lead to a diminution in performance. These diminutions may in turn gradually cause functional diseases (Griefahn and Muzet, 1978). Furthermore, recuperation during sleep definitely improves performance, and intermittent noise exceeding certain levels negatively affects recovering from tiredness (Ohrstrom, 1995; Ohrstrom and Rylander, 1982). For wakened subjects submitted to not too intense noise (below 90 dB), there has in general been found no impairment in the performance of tasks. A similar trend has been found for subjects during the day following a noisy night, namely that no, or at least a slightly significant, change in the performance of tasks can be attributed to noise exposure at night (see for instance (Griefahn and Gros, 1986)). At the other extreme, subjects who have been deprived of sleep perform tasks less well than rested subjects (Alexandre et al., 1975).

Although the results from some different studies on this topic seem controversial, due to known or unknown reasons (an interesting discussion on this issue may be found in (Nelson, 1987)), Ohrstrom and colleagues report on the poorer performance of their test subjects the day following a noisy night in the laboratory (Ohrstrom and Bjorkman, 1988; Ohstrom et al., 1988). Similarly, other research groups assure that the reaction time to a psychomotor performance is palpably shorter after a quiet night as compared to a noisy one, see [Figure - 13] for instance (Vallet et al., 1983, Wilkinson and Campbell, 1984).

Ohrstrom and Rylander exposed young adults to different numbers of noises from heavy vehicles with different maximum levels. They found no significant differences in the mean time of the RT test for either of the 50 or 60 dB(A) maximum levels, whereas RT decreased when the number of noise events with a maximum level of 60 dB(A) increased from 16 to 64 (Ohrstrom and Rylander, 1990). To complete this short overview, it is known that some stimuli can increase SWS the following night suggesting perhaps some "cerebral compensation" for this activity. It is possible that SWS deprivation by reducing the SWS available for cerebral a noise when subject to a prolonged exposure to the noise (Berglund and Lindvall, 1995). The outcomes of many studies suggest that there is no evidence of complete or even less appreciable adaptation and habituation to noise in general (Weinstein, 1982; Griefahn, 1991) or to traffic noise in particular, either in the long term (Eberhardt, 1987; Eberhardt, 1988; Eberhardt and Akselsson, 1987; Griefahn and Gros, 1986; Hofman et al., 1995; Vallet et al., 1983; Ohrstrom, 1989) or in the short term (Fidell et al., 1995; Thiessen, 1978). This applies particularly well to psychophysiological reactions such as finger pulse and heart and respiration rates. Considering awakening effects and body movements, it is to be observed on the other hand that some kind of habituation occurs during the same night, or over several nights. This is explained by the tendency of less frequent awakenings and a relatively lesser number of body movements with increasing number of noise events (Berglund and Lindvall, 1995; LeVere et al., 1972; Ohrstrom and Rylander, 1990). Other factors like the periodicity and duration of noise events may also contribute to habituation (Kawada et al., 1997).

During experiments, especially in a laboratory, researchers sometimes explain their unexpected results as due to causes related to adaptation problems of their subjects to new conditions. However, it normally takes only a few days (Thiesen and Lapointe, 1983), usually two days are sufficient (Ohrstrom and Rylander, 1982), so until habituation to laboratory conditions takes place. The progression of the number of wakings per night under a laboratory experiment is shown in [Figure - 14]. The curves show clearly how this number tends to be constant as the number of nights increases, which gives evidence to the fact that the persons get familiar to the experimental environment.

Ohrstrom and Bjorkman performed a study under laboratory conditions with the specific goal of studying the habituation of their subjects to artificial noise from heavy vehicles. After a two-week experimentation period, the subjects, independently of their sensitivity to noise, experienced a worsened sleep quality, nor was habituation found either for mood (Ohrstrom and Bjorkman, 1988). On the other hand, several other studies, not particularly conducted with the aim of investigating the special issue of human adaptation to noise induced sleep disturbances, have also concluded, with almost general agreement, that habituation is the factor of least correlation with noise exposure. Whether the subjects have been living for many years in noisy areas or whether they have been under the exposure of artificial noise less than a week or so, habituation ceases after only a few days at most (Eberhardt, 1988; Eberhardt and Akselsson, 1987; Griefahn and Gros, 1986; Thiessen and Lapointe, 1978; Ohrstrom et al., 1988).

Comparison with the effects of other transportation noises

The comparison of sleep disturbances caused by different noise sources is as yet incomplete. An attempt to achieve this goal has been made recently by Carter (1996), and since then the number of publications on this theme has been rather limited. In the early eighties Hall and colleagues reviewed the different available data published on the community responses to different traffic noise sources in view of ranking these noise sources according to their various adverse effects on human beings. This included the major environmental noise sources due to transportation, aircraft, rail and road traffic. An important finding from these surveys and other original works is that there are differences in the community responses and that these differences are mainly attributed to the noise sources, while other factors like age, sex and attitude towards the source cannot be altogether neglected (Hall, 1984; Hall et al., 1981; Osada, 1991, Ahrlin, 1988). Most of these comparative studies were concerned with only a few response variables: annoyance, interference with other activities and sleep disturbance see [Table - 3].

Ahrlin found that road traffic noise primarily disturbs rest and sleep (Ahrlin, 1988) and Vernet could conclude that for the same L eq , road traffic noise causes three times more sleep disturbances than railroad noise (Vernet, 1979). In a noisy environment Vernet also finds that the crucial factor for instantaneous sleep reaction is the duration of the noise event, while in a quiet context it is its emergence (Vernet, 1983). These results once again stress the importance of the number of noise occurrences and the unreliability of the Leq measure if considered alone for quantifying intermittent noise (Griefahn et al., 1993). It may be stated from the outcomes of other studies in various parts of the world that, in general and with regard to annoyance, road traffic noise is more annoying than railroad traffic noise (Hall, 1984). Few studies, although simultaneously investigating the effects of road and aircraft noises, permit no direct comparison between these two noises (Carter and Lockington, 1991), but it is generally believed that aircraft noise causes less sleep disturbances that road traffic noise (Rohrman et al., 1980). In the case of air and railroad transportation, noise from freight traffic is found to cause more annoyance than that from passenger traffic ( deJong and Miedema, 1996; Taylor et al., 1981)

Some other aspects of sleep disturbances

Mood-which is a state of feelings-has also been found to diminish significantly after noise exposure during the night. This factor is a set of four parameters: activity, relaxation, extroversion and pleasantness. Activity and extroversion have been measured as decreasing after a succession of noisy nights, and then increasing after a return to quiet nights (Ohrstrom, 1995; Ohrstrom et al., 1988).

As to problems regarding individual differences, noise has in general different impacts on people of different age, sex, sensitivity to noise or socioeconomic situation. The effects of nightly occurring traffic noise on some sleep parameters of people when grouped under different categories is shown in the diagram of [Figure - 15]. This figure summarises some results of an extensive study conducted on a population living in a city with a relatively high traffic activity by night. The experiment consisted of recording the sleep of subjects in their homes under twelve days. The subjects slept under their usual conditions from the first to the fifth night and then again under the eleventh and twelfth nights. In between, at the nights six to ten, the subjects underwent an experimental phase and they were divided into several groups where for instance with reference to [Figure - 15], the earplugs group consisted of subjects who were asked to use earplugs during sleep (Griefahn and Gros, 1986).

In his review of the importance of gender in sleep disturbance caused by different noise signals, Lukas states that women have lower arousal thresholds than men (Lukas, 1975). This result is also validated in the case of road traffic noise (Griefahn and Gros, 1986; Langdon and Buller, 1977), although some studies could not give a clear-cut answer on this issue (Griefahn and Muzet, 1978), or they could find no distinctive reactions between men and women (Wilkinson, 1984, Ohrstrom, 1989). Regarding age, elderly people in general have a greater probability for awakening reactions. This is an almost straightforward conclusion from several investigations (Eberhardt, 1988; Griefahn and Gros, 1986; Griefahn and Muzet, 1978; Lukas, 1975; Wilkinson, 1984). Thiessen, on the other hand, subdividing his subjects in groups with different age ranges, found that people over 55 years of age had nearly the same response as young adults (under 25 years) whereas middle­aged people had a sensitivity to noise around 15 dB lower (Thiessen, 1978). Finally, it is worth noting that people with a higher sensitivity to noise tend to be more adversely affected by road traffic noise at night than people who are less sensitive to noise (Langdon and Buller, 1977; Stansfeld et al., 1993; Ohrstrom, 1995; Ohrstrom et al., 1988). [Figure - 16] represents the relationship between noise sensitivity and the degree of annoyance caused by road traffic noise in a population. These results are reported from a study on residents in a medium sized city where a significant part -about 25%- of the randomly chosen persons showed in fact some sensitivity towards noise (Matsumara and Rylander, 1991).

Studies on cardiac reactivity to traffic noise during sleep show that cardiac reactivity is due to stimulus -related responses and that peaks due to traffic noise cause an increase in heart rate. Reducing noise level by double glazing the widows did not diminish the magnitude of the cardiac response which suggests the possibility that cardiac reactivity during sleep is more affected by the number of sound events rather than by the sound level (Hofman et al., 1995). This is further evidence of the importance of the emergence of disturbing noise events from the background noise. Heart rate and cardiac arrythmia were also not found to be correlated to such metrics such as L Aeq, L Amax or L 1 (Carter et al., 1994).

  Conclusions Top

This paper presented the diverse effects of nocturnal road traffic noise on the sleep of human beings. In the introductory part, a description of noise and the different metrics used to quantify it have been succinctly overviewed with special consideration to road traffic noise. Research in the subject of noise­induced sleep disturbance has accumulated during the past three decades into a large amount of valuable information. However, results are still streaming in from several parts of the industrialised world with the goal of gathering and comprehending as much information as possible on the factors that negatively affect the quality of human sleep when subject to external auditory stimuli. The overall effect of nocturnal traffic noise on sleep is not easy to describe, and one of the major reasons behind these difficulties is due to the characteristics of both traffic noise and sleep. Intrusive traffic noise in the dwelling of a sleeping subject has specific frequency and level characteristics as well as a time-variational dimension. The normal daily sleep cycle also depends on a set of time varying psychophysiological factors. In the long range, there are still inconsistencies in the research results that prohibit to drawing firm conclusions on the health effects of noise exposure (Lercher, 1996b). On the other hand, the non-auditory influence of noise on people in general and on their sleep in particular depends on a variety of factors like the noise metric chosen, the disturbance metric chosen and other psychological and social factors (Fields, 1993; Osada, 1991; Pearsons et al., 1995). The non­ noise factors can further enhance the effects of the noise level factor and may be a demographical modifier like age, a lifestyle variable like shiftwork, or of a pure psychological character like attitude towards the noise source (Carter, 1998). Consequently, and based on the fact that nocturnal noise may differently affect the various sleep stages of humans, research is needed to develop means for measuring the meaning of noise to the exposed population in terms of its specific social and psychological context more than its environmental one (Job, 1996; Lercher, 1996b). However, the strongest determinant of sleep disturbance remains the noise level. It is also important to specify whether the study is being conducted under normal or under laboratory conditions, because the relevant results indicate that laboratory results may not be capable of predicting sleep disturbance effects in community settings (Fidell et al, 1995; Pearsons et al., 1995). Moreover, most of the past studies have been related to indoor noise levels, whereas standards and recommendations require guidelines in terms of outdoor noise level (Carter, 1998).

In attempting to set regulations for nocturnal noise, it is imperative to distinguish between the continuous and the non-continuous components of night-time noise exposure. Concerning continuous noise, the outcomes of the studies reviewed in this work support the old WHO recommendation of 35 dB(A) indoor noise level during sleep for assuring sleep of good quality, although the lower value of 30 dB(A) as proposed lately by Berglund and Lindvall (1995) could achieve even better results.

As night traffic activity is not continuous, three factors seem to be of most importance for assessing overall sleep quality. First, the background noise level if predominant should not exceed a certain level, and the metric L eq may be used in this respect with a value not exceeding 30dB(A). Noise exposure may cause changes in the sleep stage architecture, and this value ensures the least effect especially on REM sleep. When single events such as passing vehicles come into play, the L eq becomes unreliable when used alone. Instead, the emergence of the isolated events from the background noise becomes important. This is the second factor of noise-disturbed sleep. Noise ratings based on energy considerations are still a subject of controversy among researchers, but the maximum level has been adopted as a reliable descriptor, and every single event should at least not induce awakening or arousal reactions in the sleeper. In order to avoid this extreme response and other minor effects, there is some agreement that the peak level of the isolated events not be in excess of 15 dB(A) from the background noise. Equivalently, this maximum value must be around L max=45 dB(A) for a background noise satisfying the WHO recommendation, but Lmax may even have to be lower in a quieter environment. The third important factor to be considered in the case of intermittent noise is the number of events or their frequency. It is an almost unanimous conclusion that an increased number of disturbing noise events prolongates the time required to fall asleep and increases the number of awakenings. However, most of the studies pertinent to this subject have been conducted in laboratories and thus may not be generalised to community settings without taking some precautions. Updating the results reached from these studies and considering single events of 45 dB(A) maximum level, the critical number of single events can be recommended to lie somewhere between 16 and 32 for a normal sleeping period. For noise level above 50 dB(A), this number must be below 16 (Ohrstrom, 1995)

  Acknowledgement Top

The author is much indebted to Professor Sven G. Lindblad for valuable help in the achievement of this work. This work was financially supported by the Swedish Environmental Protection Agency, and this is gratefully acknowledged.[92]

  References Top

1.Alexandre A., Barde J. P., Lamure C. and Langdon F. J., 1975 ROAD TRAFFIC NOISE Applied Science Publishers, London.  Back to cited text no. 1    
2.Ahrlin U., 1988 "Activity Disturbances Caused by Different Environmental Noises" Journal of Sound and Vibration 127, p. 599.  Back to cited text no. 2    
3.Aspinall D. T., 1970 "Control of Road Noise by Vehicle Design" Journal of Sound and Vibration 13, p. 435.  Back to cited text no. 3    
4.Berglund B. and Lindvall T., 1995 "Community Noise" Archives of the Center for Sensory Research Vol. 2, Issue 1.  Back to cited text no. 4    
5.Bryan M. E. and Tempest W., 1973 "Are our Noise Laws Adequate" Applied Acoustics 6, p. 219.  Back to cited text no. 5    
6.Brown A. L., 1994 "Exposure of the Australian Population to Road Traffic Noise" Applied Acoustics 43, p. 169.  Back to cited text no. 6    
7.Carskadon M. A. and Rechtschaffen A., 1989 "Monitoring and Staging Human Sleep" in PRINCIPLES AND PRACTICES OF SLEEP MEDICINE, Kryger M. H., Roth T. and Dement W. C. Editors, Saunders, Philadelphia.  Back to cited text no. 7    
8.Carter N. L., 1996 "Transportation Noise, Sleep, and Possible After Effects" Environment International 22(1), p. 105.  Back to cited text no. 8    
9.Carter N. L., 1998 "Some Issues in Noise-Induced Sleep Disturbance" Acoustics Australia 26(2), p. 49.  Back to cited text no. 9    
10.Carter N. L. and Beh H. C., 1987 "The Effect of Intermitent Noise on Vigilance Performance" Journal of the Acoustical Society of America 82, p. 1334.  Back to cited text no. 10    
11.Carter N. L. and Lockington G., 1991 "Sleep Stage Response to Aircraft and Traffic Noise" Proceedings of Inter-Noise 91, Vol II, p. 835.  Back to cited text no. 11    
12.Carter N. L., Macsween L, Bulteau V. G., Gray C. H. and Ferris M., 1976 "Environmental and Pure Tone Thresholds" Journal of Sound and Vibration 46, p. 259.  Back to cited text no. 12    
13.Carter N. L., Ingham P., Tran K. and Hunyor S. N., 1994 "A Field Study of the Effects of Traffic Noise on Heart Rate and Cardiac Arrhythmia during Sleep" Journal of Sound and Vibration 169, p. 211.  Back to cited text no. 13    
14.deJong R. G. and Miedema M. E., 1996 "Is Freight Traffic Noise More Annoying than Passenger Traffic Noise" Journal of Sound and Vibration 193, p. 35.  Back to cited text no. 14    
15.diNisi J, Muzet A. and Libert J. P., 1990 "Cardiovascular Responses to Noise During Waking and Sleeping in Man" Proceedings of the 5th International Congress on Noise as a Public Health Problem Vol. 5, part II, p. 47.  Back to cited text no. 15    
16.Eberhardt J. L., 1987 The Influence on Sleep of Noise and Vibrations Caused by Road Traffic PhD Dissertation, University of Lund, Lund.  Back to cited text no. 16    
17.Eberhardt J. L., 1988 "The Influence of Road Traffic Noise on Sleep" Journal of Sound and Vibration 127, p. 449.  Back to cited text no. 17    
18.Eberhardt J. L. and Akselsson K. R., 1987 "The Disturbance by Road Traffic Noise of the Sleep of Young Male Adults as Recorded in the Home" Journal of Sound and Vibration 114, p. 417.  Back to cited text no. 18    
19.Eberhardt J. L., Strale L. O. and Berlin M. H., 1987 "The Influence of Continuous and Intermittent Traffic Noise on Sleep" Journal of Sound and Vibration 116, p. 445.  Back to cited text no. 19    
20.Fields J., 1993 "Effect of Personal and Situational variables on Noise Annoyance in residential Areas" Journal of the Acoustical Society of America 93, p. 2753.  Back to cited text no. 20    
21.Fidell S., 1996 "Some Policy and regulatory Implications of Recent Findings of Field Studies on Noise-Induced Sleep Disturbance" proceedings of Internoise 96, Book 5, p. 2261.  Back to cited text no. 21    
22.Fidell S. A. and Teffetteller S., 1981 "Scaling the Annoyance of Intrusive Sounds" Journal of Sound and Vibration 78, p. 291.  Back to cited text no. 22    
23.Fidell S., Horonjeff R. D., Teffetteller S. and Pearsons K., 1980 "Adaptation to Changes in Aircraft Noise Exposure" Journal of the Acoustical Society of America, 67, p. S34.  Back to cited text no. 23    
24.Fidell S., Pearsons K., Tabachnick B., Howe R., Silvati L. and Barber D. S., 1995 "Field Study of Noise-Induced Sleep Disturbance" Journal of the Acoustical Society of America 98, p. 1025.  Back to cited text no. 24    
25.Fog H. and Jonsson E., 1968 "Traffic Noise in Residential Areas" Report No E:36/68. The National Swedish Institute for Building Research.  Back to cited text no. 25    
26.Griefahn B., 1991 "Environmental Noise and Sleep. Review-Need for Further research" Applied Acoustics 32, p. 255.  Back to cited text no. 26    
27.Griefahn B. and Gros E., 1986 "Noise and Sleep at Home, a Field Study on Primary and After-Effects" Journal of Sound and Vibration 105, p. 373.  Back to cited text no. 27    
28.Griefahn B. and Muzet A., 1978 "Noise-Induced Disturbances and Their Effects on Health" Journal of Sound and Vibration 59, p. 99.  Back to cited text no. 28    
29.Griefahn B., Brode P. and Schwarzenau P., 1993 "The Equivalent Sound Pressure Level - A Reliable Predictor for Human Responses to Impulse Noise ?" Applied Acoustics 38, p. 255.  Back to cited text no. 29    
30.Griffiths I. D., Langdon F. J. and Swan M. A., 1980 "Subjective Effects of Traffic Noise Exposure: Reliability and Seasonal Effects" Journal of Sound and Vibration 71, p. 227.  Back to cited text no. 30    
31.Hall F. L., 1984 "Community Response to Noise: Is All Noise the Same ?" Journal of the Acoustical Society of America 76, p. 1161.  Back to cited text no. 31    
32.Hall F. L., Birnie S. E., Taylor S. M. and Palmer J. E., 1981 "Direct Comparison of Community Response to Road Traffic and to Aircraft Noise" Journal of the Acoustical Society of America 70, p. 1690.  Back to cited text no. 32    
33.Hassall J. R. and Zaveri K., 1979 ACOUSTIC NOISE MEASUREMENTS 2nd edition Brael & Kjxr Publication, Nxrum.  Back to cited text no. 33    
34.Hayek S., 1990 "Mathematical Modeling of Absorbent Highway Noise Barriers" Applied Acoustics 31, p. 77.  Back to cited text no. 34    
35.Hofman W. F., Kumar A. and Tulen J. H. M., 1995 "Cardiac Reactivity to Traffic Noise During Sleep in Man" Journal of Sound and Vibration 179, p.577.  Back to cited text no. 35    
36.Horonjeff R. D., Fidell S., Teffetteller S. and Green D. M., 1982 "Behavioral Awakening as Functions of Duration and Detectability of Noise Intrusions in the Home" Journal of Sound and Vibration , p.327.  Back to cited text no. 36    
37.Izumi K. and Yano T., 1991 "Community Response to Road Traffic Noise: Sicial Surveys in Three Cities in Hokkaido" Journal of Sound and Vibration 151, p. 505.  Back to cited text no. 37    
38.Job R. F. S., 1996 "The Influence of Subjective Reactions to Noise on Health Effects of the Noise" Environment International 22, p. 93.  Back to cited text no. 38    
39. Jones R. R. K. and Hothersall D. C., 1979 "Effect of Operating Parameters on Noise Emission from Individual Road Vehicles" Journal of Sound and Vibration 13, p. 121.  Back to cited text no. 39    
40.Kabuto M. and Kageyama T., 1994 "Nighttime Road Traffic Noise and Sleep Quality" Proceedings of Internoise 94, Vol II, p. 1041.  Back to cited text no. 40    
41.Kageyama T., Kabutto M., Nitta N., Kurowaka Y., Taira  Back to cited text no. 41    
42.K., Suzuki S. and Takemoto T., 1997 "A Cross -Sectional Study on Insomnia Among japanese Adult Women in Relation to Night-Time Road Traffic Noise" Journal of Sound and Vibration 205, p.405.  Back to cited text no. 42    
43.Kawada T., Sasazawa Y., Kiryu Y. and Suzuki S., 1997 "Daily sleep changes in a noisy environment assessed by subjective and polygraphic sleep parameters " Journal of Sound and Vibration, 205, p.405.  Back to cited text no. 43    
44.Kryter K. D., 1985 THE EFFECTS OF NOISE ON MAN, 2nd ed. Academic Press, New York.  Back to cited text no. 44    
45.Lambert J., Simonet F. and Vallet M., 1984 "Patterns of Behaviour in Dwellings Exposed to Road Traffic Noise" Journal of Sound and Vibration 92, p. 159.  Back to cited text no. 45    
46.Langdon F. J. and Buller I. B., 1977 "Road Traffic Noise and Disturbance to Sleep" Journal of Sound and Vibration 50, p. 13.  Back to cited text no. 46    
47.Lercher P., 1996a "Road Traffic Noise, Self Medication, and Prescriptions: A Community Study" Proceedings of Internoise 96, p. 2171.  Back to cited text no. 47    
48.Lercher P., 1996b "Environmental Noise and Health: an Integrated Research Perspective" Environment International 22, p. 117.  Back to cited text no. 48    
49.LeVere T. E., Bartus R. T., Hart F. D., 1972 "Electroencephalographic and Behavioral Effects of Nocturnally Occuring jet Aircraft Sounds" Aerospace Medicine 43, p. 384.  Back to cited text no. 49    
50.Lukas J., 1975 "Noise and Sleep: A literature Review and a Proposed Criterion for Assessing Effects" Journal of the Acoustical Society of America 58, p. 1232.  Back to cited text no. 50    
51.Matsumara Y. and Rylander R., 1991 "Noise Sensitivity and Road Traffic Annoyance in a Population Sample", Journal of Sound and Vibration 151, p. 415.  Back to cited text no. 51    
52.Mills C. H. G. and Aspinall D. T., 1968 "Some Aspects of Commercial Vehicle Design" Applied Acoustics 1, p. 47.  Back to cited text no. 52    
53.Naganuma S., Kawada T., Aoki S. and Suzuki S., 1991 "Instantaneous Effects of Road Traffic Noise on Sleep Assessed by Computerized Polygraphy" Proceedings of Internoise 91, p. 851  Back to cited text no. 53    
54.Nelson P. M., Editor 1987 TRANSPORTATION NOISE REFERENCE BOOK, Butterworths, Cambridge.  Back to cited text no. 54    
55.Nelson P. M. and Abbott P. G., 1987 "Low Noise Road Surfaces" Applied Acoustics 21, p. 119.  Back to cited text no. 55    
56.Olson N., 1972 "Survey of Motor Vehicle Noise" Journal of the Acoustical Society of America 52, p. 1291.  Back to cited text no. 56    
57.Ohrstrom E., 1989 "Sleep Disturbance, Psycho-Social and  Back to cited text no. 57    
58.Medical Symptoms- A Pilot Survey Among Persons Exposed to High Levels of Road Traffic Noise" Journal of Sound and Vibration 133, p. 117.  Back to cited text no. 58    
59.Ohrstrom E., 1991 "Psycho-Social Effects of Traffic Noise Exposure " Journal of Sound and Vibration 151, p. 513.  Back to cited text no. 59    
60.Ohrstrom E., 1995 "Effects of Low Levels of Road Traffic Noise During the Night: A Laboratory Study on Number of Events, Maximum Noise Levels and Noise Sensitivity" Journal of Sound and Vibration 179, p. 603.  Back to cited text no. 60    
61.Ohrstrom E. and Bjorkman M., 1983 "Sleep Disturbance Before and After Traffic Noise Attenuation in an Apartment Building" Journal of the Acoustical Society of America 73, p. 877.  Back to cited text no. 61    
62.Ohrstrom E. and Bjorkman M., 1988 "Effects of Noise­Disturbed Sleep- A Laboratory Study on Habituation and Subjective Noise Sensitivity" Journal of Sound and Vibration 122, p. 277.  Back to cited text no. 62    
63.Ohrstrom E. and Rylander R., 1982 "Sleep Disturbance Effects of Traffic Noise- A Laboratory Study on After Effects" Journal of Sound and Vibration 84, p. 87.  Back to cited text no. 63    
64.Ohrstrom E. and Rylander R., 1990 "Sleep Disturbance by Road Traffic Noise- A Laboratory Study on Number of Noise Events" Journal of Sound and Vibration 143, p. 93.  Back to cited text no. 64    
65.Ohrstrom E., Rylander R. and Bjorkman M., 1988 "Effects of Night Time Road Traffic Noise- An Overview of Laboratory and Field Studies on Noise Dose and Subjective Sensitivity" Journal of Sound and Vibration 127, p. 441.  Back to cited text no. 65    
66.Osada Y., 1991 "Comparison of Community Reactions to Road Traffic Noise" Journal of Sound and Vibration 151, p. 479.  Back to cited text no. 66    
67.Pearsons K. S., Barber D. S., Tabachnick B. G. and Fidell S., 1995 "Predicting Noise-Induced Sleep Disturbance" Journal of the Acoustical Society of America 97, p. 331.  Back to cited text no. 67    
68.Priede T., 1971 "Origins of Automotive Engine Noise" Journal of Sound and Vibration 15, p. 1.  Back to cited text no. 68    
69.Rathe E. J., Casula F., Hartwig H. and Mallet H., 1973 "Survey of the Exterior Noise of Some Passenger Cars" Journal of Sound and Vibration 29, p. 483.  Back to cited text no. 69    
70.Relster E., 1975 TRAFFIC NOISE ANNOYANCE Polyteknisk Forlag, Lyngby.  Back to cited text no. 70    
71.Rohrman B., Finke H. O. an Guski R., 1980 "Analysis of Reactions to Different Environmental Noise Sources in Residential Areas" Proceedings of the 3rd International Congress on Noise as a Public Health Problem, p. 548.  Back to cited text no. 71    
72.SOU, 1974 (Statens Offentliga Utredningar SOU 1974:60)  Back to cited text no. 72    
73.TRAFIKBULLER Del I "Vagtrafikbuller", (road traffic noise) Stockholm. (In Swedish).  Back to cited text no. 73    
74.Stansfeld S. A. et al., 1993 "Road Traffic Noise, Noise Sensitivity and Psychological Disorder" Psychological Medicine 23, p. 977.  Back to cited text no. 74    
75.Suzuki S., Kawada T., Kiryu Y., Sazawa Y. and Tamura Y., 1997 "Transient Effect of the Noise of Passing Trucks on Sleep EEG" Journal of Sound and Vibration 205, p. 411.  Back to cited text no. 75    
76.Taylor S. M., Hall F. L. and Birnie S. E., 1981 "A Comparison of Community Response to Aircraft Noise at Toronto International and Oshawa Municipal Airports" Journal of Sound and Vibration 77, p. 233.  Back to cited text no. 76    
77.Terzano M. G., Parrino L., Spaggiari M. C., Buccino G. P., Fioriti G. and Depoortere H., 1993 "Assessment of Noise­Induced Sleep Fragility in Two Age Ranges by Means of Polysomnographic Microstucture" Journal of Sound and Vibration 162, p. 345.  Back to cited text no. 77    
78.Thiessen G. J., 1978 "Disturbance of Sleep by Noise" Journal of the Acoustical Society of America 64, p. 216.  Back to cited text no. 78    
79.Thiessen G. J. and Lapointe A. C., 1978 "Effect of Intermittent Truck Noise on Percentage of Deep Sleep" Journal of the Acoustical Society of America 64, p. 1078.  Back to cited text no. 79    
80.Thiessen G. J. and Lapointe A. C., 1983 "Effect of Continuous Traffic Noise on Percentage of Deep Sleep, Waking and Sleep Latency" Journal of the Acoustical Society of America 73, p. 225.  Back to cited text no. 80    
81.Tulen J. H., Kumar A. and Jurriens A. A., 1986 "Psychophysiological Acoustics of Indoor Sound due to Traffic Noise During Sleep" Journal of Sound and Vibration 110, p. 110.  Back to cited text no. 81    
82. Vallet M., Gagneux G. M, Blanchet V., Favre B. and Labiale G., 1983 "Long Term Sleep Disturbance Due to Traffic Noise" Journal of Sound and Vibration 90, p. 173.  Back to cited text no. 82    
83.Vallet M., Olivier D., Laurens J. F. and Clairet J. M., 1988 "Effects of Road Traffic Noise on Pulse Rate During Sleep" Proceedings of the 5th International Congress on Noise as a Public Health Problem, Vol. 5, Part II, p. 21.  Back to cited text no. 83    
84.Vernet M., 1979 "Effect of Train Noise on Sleep for People Living in Houses Bordering the Railway Line" Journal of Sound and Vibration 66, p. 331.  Back to cited text no. 84    
85.Vernet M., 1983 "Comparison Between Train Noise and Road Noise Annoyance During Sleep" Journal of Sound and Vibration 87, p. 331.  Back to cited text no. 85    
86.Wanner H. U., Wehrli B., Bakke P., Nemecek J. Turrian V. and Granjean E., 1977 "Effects of Road Traffic Noise on Residents" Proceedings of Internoise 77, p. B 698.  Back to cited text no. 86    
87.Waters P. E., 1970 "Control of Road Noise by Vehicle Operation" Journal of Sound and Vibration 13, p. 445.  Back to cited text no. 87    
88.Weinstein N. D., 1982 "Community Noise Problems: Evidence Against Adaptation" Journal of Environmental Psychology 2, p.87.  Back to cited text no. 88    
89.WHO, 1980 "Environmental Health Criteria 12-Noise" WHO, Geneva.  Back to cited text no. 89    
90.Wilkinson R. T., 1984 "Disturbance of Sleep by Noise: Individual Differences" Journal of Sound and Vibration 95, p. 55.  Back to cited text no. 90    
91.Wilkinson R. and Campbell K. B., 1984 "Effects of Traffic Noise on Quality of Sleep: Assessment by EEG, Subjective Report, or Performance the Next Day" Journal of the Acoustical Society of America 75, p. 468.  Back to cited text no. 91    
92.Yoshida T., Osada Y., Kawaguchi T., Hoshiyama Y., Yoshida K. and Yamamoto K. 1997 "Effects of Road Traffic Noise on Inhabitants of Tokyo" Journal of Sound and Vibration 205, p. 517.  Back to cited text no. 92    

Correspondence Address:
D Ouis
Department of Engineering Acoustics, Lund Institute of Technology, P.O. Box 118, S-221 00 Lund
Login to access the Email id

Source of Support: None, Conflict of Interest: None

PMID: 12689487

Rights and PermissionsRights and Permissions


  [Figure - 1], [Figure - 2], [Figure - 3], [Figure - 4], [Figure - 5], [Figure - 6], [Figure - 7], [Figure - 8], [Figure - 9], [Figure - 10], [Figure - 11], [Figure - 12], [Figure - 13], [Figure - 14], [Figure - 15], [Figure - 16]

  [Table - 1], [Table - 2], [Table - 3]