Noise Health Home 

[Download PDF]
Year : 2004  |  Volume : 6  |  Issue : 22  |  Page : 83--93

Nocturnal aircraft noise effects

M Basner, A Samel 
 Deutsches Zentrum für Luft- und Raumfahrt, Institut für Luft- und Raumfahrtmedizin, Cologne, Germany

Correspondence Address:
M Basner
Deutsches Zentrum für Luft- und Raumfahrt, Institut für Luft- und Raumfahrtmedizin, Linder Höhe, 51147 Cologne


Noise protection associated with the construction and extension of airports in the Federal Republic of Germany has been regulated by the law for protection against aircraft noise since 1971. This legislation is due for revision because of different aspects. One aspect is the growth of air traffic which has led many airports to the limits of their capacity and in search of new ways of adaptation to the increasing demand for flight services. Another aspect is the increasing concern of the population about noise effects which has to be addressed by better protection against the effects of aircraft noise. The framework conditions of policy in terms of society as a whole, its health and economic environment need to be put into effect by political action. Science can contribute to this goal by performing noise effects research and by providing recommendations to the political body. However, it remains controversial, what measures are necessary or adequate to assure effective protection of the population against aircraft noise. This is particularly true for the protection of rest and sleep at night. The problem of finding a common basis for adequate recommendations is associated with (1) the low number of primary studies, which also exhibited highly variable results and assessments, (2) the handling of acoustic or psycho-acoustic dimensions for quantifying psychological or physiological reactions, and (3) the conception of how far preventive measures have to go to prove effective. With this in mind, the DLR Institute for Aerospace Medicine is conducting a large-scale, multi-stage study for investigating the acute effects of nocturnal aircraft noise on human sleep. This enterprise is implemented in the framework of the HGF/DLR project «DQ»Quiet Air Traffic«DQ» for developing sustainable assessment criteria for human-specific effects of aircraft noise at night.

How to cite this article:
Basner M, Samel A. Nocturnal aircraft noise effects.Noise Health 2004;6:83-93

How to cite this URL:
Basner M, Samel A. Nocturnal aircraft noise effects. Noise Health [serial online] 2004 [cited 2021 Dec 6 ];6:83-93
Available from:

Full Text


The number of passengers being transported at German airports has increased approximately sevenfold since the middle of the sixties. In freight and mail transport an increase as high as nine-fold is registered, where the greater part of these services is executed during night time. In view of the capacity load of some airports, already reaching their limits during the day, the predicted increments of air traffic will presumably take place to a greater extent at night and in the twilight hours in future. The sound energy emitted by certain individual airplanes has indeed been drastically reduced in the past, particularly when loud and older aircraft have been taken out of service. Nonetheless, sometimes more than 100 planes are registered in a single night by aircraft noise monitoring stations close to residential areas, exhibiting a wide range of different outdoor levels, depending on their location. The effects of such a large number of nocturnal aircraft noise episodes on sleep and on the performance capacity of human beings have not been adequately investigated so far.

To quote Jansen et al., "The claim of achieving applicable research results in terms of 'noise and sleep', can only be assessed as not being completely accomplished" (Jansen et al. 1995). This statement particularly applies to the effects of nocturnal aircraft noise on sleep and its consequences on well-being and performance.

In her welcoming address to the scientific symposium entitled "Environmental Capacity of Airports - Aspects of Noise Effects Research -Targets and Measures of Protection", October 25/26, 1999 in the Berlin district of Koepenick, Mrs. Altmann (1999) asserted that "noise has been viewed far too little in terms of its medical, psychological and social aspects" and that the basis of the legislation of 1971 (Aircraft Noise Protection Act, 1971) is fully out-of-date. She called for science to develop new threshold values for protection areas for consideration by political bodies.

The German Aerospace Center (DLR) accepts this request and challenge. The project "Quiet Air Traffic" is being carried out for this purpose, in order to determine statistically confirmed and valid criteria for the effects of nocturnal aircraft noise on sleep using an innovative research approach (Basner et al. 2000, Samel 2000).

 Research efforts in the past

A series of studies are concerned with the physiological effects of road, rail or air traffic noise on human sleep. However, the world-wide number of primary studies using physiological methods of measurement (e.g. polysomnography, pulse amplitude, sampling of stress hormones) on people asleep and affected by aircraft noise is very modest. In the studies published so far, small sample sizes, confined age ranges as well as low numbers of acclimatization and noise nights are obvious (Linnemeier 1995, Maschke 1997). Therefore, the results of statistical analysis of these experiments provide only limited if any validity. In addition to this, deficits in the design of trials (e.g. no control groups) or in assessment of the data (e.g. observations of individual cases without statistical analysis) are often presented (cf. Linnemeier 1995, Maschke 1997, Jansen et al. 1995). Wysk (1998) noted that "the courts have to deal with specialist appraisals from noise effects research, an interdisciplinary range of sciences in continuous motion and finally with incomplete statements, which are hardly uncontroversial".

Another divisive discussion is taking place on the question of a threshold (by means of an equivalent sound level L eq or a maximum sound level L AS,max ) at which physiological reactions in form of sleep disturbances or increased vegetative hormonal secretions occur (Gottlob 1998). Jansen et al. (1995) assume first changes in sleep depth induced by noise events at a maximum level of 55 dB(A), and awakening reactions at more than 60 dB(A). These "cornerstones" are derived from observations of individual cases. The results of pulse amplitude recordings on the 10th trial night of a volunteer who had been subjected to several weeks of investigation are presented as an example (Jansen et al. 1995). Statistical analyses of several nights or subjects, however, are not shown. The awakening threshold of 60 dB(A) is assumed to be a "theoretical" value, which has yet to be confirmed by systematic studies (Jansen et al. 1995). Nonetheless, already in 1976 an attempt was made to find an average awakening threshold by a meta-analysis of results of various studies (Griefahn et al. 1976), which resulted in a mean value of 60 dB(A) and a standard deviation of 7 dB(A). Maschke et al. (2001) doubt this awakening threshold; their new calculation leads to a range between 0 dB(A) and 48 dB(A) where the beginning of noise induced awakenings is to be expected. These authors conclude from their new calculation that awakening is to be anticipated at 48 dB(A) with a probability of 95%. These (newly calculated) results are in contradiction to those derived by Maschke himself (1992), where he deduced that the lower threshold for sleep stage changes (i.e. reactions less pronounced than awakenings) should be set at a level, L eq , of 36 dB(A) (at L AS,max of 75 dB(A)), becoming particularly manifest at an L eq from 50 - 56 dB(A). These results were determined in 40 subjects over a period of five nights. A control group was missing in this study. A subsample of 8 persons from this group was exposed to sound over a period of 10 nights in order to study the stress induced by aircraft noise at night by determining catecholamine secretions in their overnight urine samples. The results show a higher adrenaline secretion at 65 dB(A) than at 75 dB(A) (Maschke 1992). The difficulties experienced in interpreting these results were attributed to the small size of the database and the incompleteness of the test design model.

In order to find out whether long-term exposure can lead to changes in stress hormone secretion, a study was conducted by Harder et al. (1999) in which 16 subjects were observed at home for six weeks. Subsequent to a control period of three nights, 32 episodes of aircraft noise with a maximum level of 65 dB(A) were played back, leading to an L eq of 32 dB(A) for eight hours at night (Harder et al. 1999, p. 41). Our own calculations, however, indicate an equivalent continuous sound level of 45.6 dB(A) derived from 32 × 65 dB(A) with an assumed noise duration of 42 sec per episode. No significant changes were noted in the catecholamine and cortisol secretions in the entire sample of all 16 subjects (Harder et al. 1999). The authors subdivided the collective into three subgroups and extrapolated the nocturnal cortisol excretion rates to 24-hour excretion rates. Both is scientifically inappropriate, the latter because the prominent rise in cortisol excretion as a result of circadian rhythms has to be taken into account. Nonetheless, the result is interpreted as a significant increase (or decline) for certain groups of subjects (number of subjects per group between 4 and 7).

The trend to reduce the expense per tested subject, in order to be able to access larger groups of subjects with the same overall effort, has been observed in the past years. The attempt to replace cost-intensive test methods such as sleep quality analysis using polysomnography, by cost-effective actimetry, consuming much less labour and time, has to be viewed as promising little success at present (cf. e.g. Jansen et al. 1995, Maschke 1997). Actimetry is a simple method of measuring body movements. The actimeter is worn on the wrist and records movements, if these generate an acceleration of more than 0.1 g. This procedure was used in a study involving 400 subjects to determine sleep disturbances as a result of aircraft noise (Ollerhead et al. 1992). In eight regions close to four airports in England 50 subjects were equipped with an actimeter for a period of 15 nights. A control group was not included. To validate the actimeter measurements, the sleep of a total of 48 persons was polysomnographically recorded over a period of four successive nights.

The actimeter readings showed that no significant increase in the rates of awakenings was recorded, even with external sound levels of up to L AS,max = 87 dB(A). A reason for this very surprising result may in fact be the insensitivity of actimetry. The authors claim that by comparing EEG measurements with those of actimetry, 88% of the awakening phases registered by EEG matched those shown by actimetry, but by contrast only 40% of the reactions determined by actimetry matched those found by the EEG (Ollerhead et al. 1992). On the one hand, it is to question whether the reactions detected by actimetry really reflect the true state of sleep disturbance. On the other hand, only the outside levels were measured in this study. Detailed data on the position of the windows was not included in this report, and soundproof windows had already been fitted in the homes of 60% of the participants.

The following recommendations for limit values for the prevention of sleep disturbances are provided in the report published by the German Federal Environment Agency entitled "Aircraft Noise Effects" (Ortscheid and Wende 2001): L eq(3) AS,max AS,max and frequency of occurrence are played back in the pre-calibrated sleeping rooms while the physiological reactions of the subjects are recorded. The criteria ascertained from these laboratory studies are tested in two extensive field studies. Altogether 128 subjects are investigated in the laboratory and 64 volunteers in the field (i.e., in their own homes). An approximately equal distribution with regard to age (between 18 and 65 years of age), sex and prior exposure to aircraft noise is sought to ensure the validity of the results for this group of people. The methods used in the laboratory and field studies are identical to a large extent. The four laboratory studies are set up as a double blind trial in a crossover design. In each case, 32 healthy persons with normal hearing thresholds are chosen for each of the four laboratory studies. These 32 subjects are divided into 4 subgroups of eight persons each, who then undergo an investigative phase of 13 consecutive nights simultaneously (see [Table 1]).

The first night provides familiarization with laboratory conditions, the second night is used as the baseline night. In the nine subsequent nights, aircraft noise with a varying distribution of L AS,max and rate of occurrence is applied to the subjects. The noise is played back at equidistant intervals between 11:15 p.m. and 6:45 a.m. In a night with exposure to noise, 8 volunteers are each subjected to the same pattern of noise. The maximum level, L AS,max of an individual noise event lies between 50 and 80 dB(A) at the ear of the sleeper, the number of events is between 4 and 128 per night (corresponding to intervals between noise events between 2 hr and 3 min). These combinations are distributed over the 9 noise nights in a random fashion and lead to continuous sound levels L eq(3) between 31.2 and 52.6 dB(A), including the constant background level of 30 dB(A) caused by the air-condition system. If the individual noise episodes are averaged in terms of energy, the combinations of L AS,max and frequency of occurrence lead to L eq(3) values from 28.1 to 52.6 dB(A) (i.e.,

excluding the background noise level). The last two nights are also kept free of aircraft noise and used for additional comparative control purposes. [Table 2] shows the combinations of L AS,max and frequency of occurrence used in the noise nights of the first two laboratory studies.

The subjects only spend the time between approximately 7 p.m. and 8 a.m. in AMSAN, so as to disturb their normal daily routine as little as possible.

The acoustic equipment facilitates mobile two­channel recording of the noise signals with class­1 sound level meters and a maximum sampling rate of 48 kHz. An essential prerequisite for the assessment of physiological reactions to acoustic stimuli is the time-synchronous recording of noise signals with electrophysiological functions. The chronological coupling of acoustic and electrophysiological data recording permits a correlated assessment of sleep disturbances and noise events with a time resolution of 125 ms.

The aircraft noise events were recorded in the sleeping quarters of residents in the vicinity of airports with microphones at a position near to the ear of sleepers, either with closed or tilted windows. The quality of playback of aircraft noise in the sleeping laboratory is warranted as genuine as possible by the individual acoustic calibration of each sleep cabin.

Electrophysiological reactions of the subjects are recorded in both the control nights and the noise nights by means of polysomnographic parameters (EEG, EOG, EMG), electrocardiogram (ECG), finger pulse amplitude (plethysmography) and respiration. Supplementary physical indicators such as the position in bed and the wrist activity during sleep, episode markers and sound level patterns of the individual noise events in the sleeping quarters as well as the light conditions in the sleep cabin are measured. In addition, activities during the entire day are recorded (including the time spent outside AMSAN).

Furthermore, all night urine is collected (8-h samples) and the following parameters are determined:

* Urine volume and exact period of collection,

* Flow rates of adrenaline, noradrenaline, cortisol, magnesium and potassium.

Computerized tests are used to determine mental and psychomotor activity in the morning and evening (Santucci et al. 1989). The test battery consists of two memory-and-search tasks with four or six letters (MST 4 or 6), an unstable tracking task (UTT) and a reaction time task (SRT), and lasts about 20 minutes. To maintain an evenly balanced level of performance and to prevent learning effects from the computer tests during the study, the subjects have 40 training sessions prior to the beginning of the studies.

Standardized questionnaires serve for the objective determination of subjectively encountered sleep experiences, as well as fatigue, feelings of well-being, stress and relaxation during the day and at night. For this purpose, a multidimensional mood questionnaire is used (MDBF, Steyer et al. 1997), together with a recuperation and stress questionnaire (EBF, Kallus 1995), a fatigue questionnaire (FAT) and a flight-noise questionnaire (FN-L).

For the field studies, designed to validate the results obtained from the laboratory studies, 64 subjects are chosen and subjected to the same test methods at home. The noise level in- and outdoors, which is time-synchronized with the electrophysiological parameters, is recorded for the purpose of establishing any relationship between aircraft noise and physiological reactions.

 Experience gained from the first two laboratory studies

The first two laboratory studies were successfully completed in the fall of 1999 and the early summer of 2000. Thus, data of 832 subject nights are available, the majority of which has been evaluated. The preliminary results are summarized in the DLR research report 2001-26 (Basner et al. 2001). Since only one third of the overall amount of data has been analysed so far, final conclusions can and should only be drawn after the data of the other 64 subjects to be studied in the laboratory have been analysed, and, additionally, validated by 64 subjects in the field. The process of data acquisition and analysis should have been completed by the end of 2003. In this report, some of the achieved results from the first two laboratory studies are presented.

 Examples of reactions during sleep to aircraft noise events

[Figure 2] (65 dB L AS,max with arousal), [Figure 3] (75 dB L AS,max without arousal) and [Figure 4] (arousal without noise) are intended to serve as illustrations of the very different reactions displayed by different individuals during sleep.

 Sleep stage distribution

[Table 3] summarizes the results of the comparison of baseline and noise nights irrespective of the intensity and frequency of noise events. Both nights were standardized to the shorter sleep period time (SPT). TST decreased non-significantly by an average difference of 3.1 min in the noise nights. Slow wave sleep (SWS, sleep stages 3 and 4) was significantly reduced by 9.1 min, whereas sleep stage 1 increased significantly by 3.8 min.

The amounts of sleep stage 2 did not change at all. Wake and REM-sleep were both non­significantly increased by 3.1 min and 1.9 min respectively in the noise nights.

Thus, although TST was only little reduced in the noise nights compared to baseline nights, changes in sleep architecture were obvious. Amounts of SWS were significantly reduced in favor of higher amounts of sleep stage 1 and wake. Hence aircraft noise lead to a shallower sleep. With actimetry alone it would not have been possible to detect these subtle changes.

 Event-correlated awakenings

In noise-free baseline nights 21.8 (SD 8.6) spontaneous awakenings were detected. More than half (56.7%) of the awakenings lasted only one epoch (30 sec); in 7% the awakenings lasted 4 min or longer. In noisy nights the average number of awakenings was 25.1 (SD 8.4). In these nights, short awakenings ( AS,max of the noise event (see [Figure 5]), but decreased with the number of noise events per night (see [Figure 6]).

 Discussion and conclusions

This paper intends to show the methodological approach and the design how the problem of acute sleep disturbances caused by aircraft noise can be tackled. It is necessary to use cost- and personnel-intensive acoustic, electrophysiologi­cal, biochemical, psychological and performance measuring methods and a large sample size - with respect to subjects of different ages and other individual factors, and with respect to the number of nights investigated, to achieve reliable and valid results. A novel approach is also the simultaneous recording of acoustical and electrophysiological data which allow an event correlated analysis of sleep disturbances and the allocation of changes in sleep to external noise events. Since the methods of data sampling and analysis are both identically applied in the laboratory and in the field, the resulting data will add a large amount of information to the field of aircraft noise dependent effects on human sleep. The design also includes the measurement of control groups and baseline nights (in both the control and the experimental groups), in order to differentiate between naturally (or usually) occurring changes of sleep stages and awakenings and those generated by noise events. An advantage of the chosen design is the comparison of data acquired from laboratory and field studies, because it can be expected that results may differ. In the controlled environment of the laboratory, results may be more pronounced than in field trials, because confounding factors (e.g. indoor noise, external noise stemming from other sources than aircraft, which are carefully recorded, and various individual factors) are influencing the data.

Since not all stages of the investigation are completed and the process of data analysis is still ongoing, only a small amount of the results can be shown in this report. These preliminary results indicate that it is essential for the interpretation of data to differentiate between endogenous and exogenous reasons that lead to changes in sleep. If and when aircraft noise affects sleep could be derived from the first two laboratory studies as well as first estimates of the degree of sleep disturbances, depending on the number of occurrences and the maximum level of aircraft noise events. Since a variety of different levels of L AS,max and numbers of events have been presented to subjects of different gender and age, the study will be likely to provide a database for the forecast of the amount of sleep disturbances. For this purpose, a model has been developed that will be further refined as additional data will become available from the remaining studies, using this broad database. The model should be able to predict sleep disturbances by a dose-response curve as it is available for annoyance by different traffic modes (e.g., Miedema and Vos 1996).[20]


1Altmann (1999) Welcoming address from the German Federal Ministry for the Environment, Natural Conservation and Reactor Safety. In Minutes of the scientific symposium entitled "Umweltkapazitat von Flughafen, Aspekte der Larmwirkungsforschung, Schutzziele und Schutzmassnahmen (Environmental Capacity of Airports, Aspects of Noise Effects Research, Targets and Measures of Protection)", October 25/26, 1999 in the Berlin District of Kopenick, pp7-8
2Basner, M., Buess, H., Linke-Hommes, A., Luks, N., Maass, H., Mawet, L., Mueller, E.W., Mueller, U., Piehler, C., Plath, G., Rey, E., Samel A., Schulze, M., Vejvoda, M., Wenzel, J. (2000) Wirkung von Nachtfluglaerm auf den Schlaf - ein neuer Forschungsansatz (Effects of Nocturnal Aircraft Noise on Sleep - a New Approach to Research). Z. Larmbekampf. 47: 201-205
3Basner, M., Buess, H., Luks, N., Maass, H., Mawet, L., Mueller, E.W., Mueller, U., Piehler, C., Plath, G., Quehl, J., Rey, E., Samel A., Schulze, M., Vejvoda, M., Wenzel, J. (2001) Nachtfluglaermwirkungen - eine Teilauswertung von 64 Versuchspersonen in 832 Schlaflabornaechten (Effects of Nocturnal Aircraft Noise - Preliminary Results Based on 64 Subjects and 832 Laboratory Study Nights). Forschungsbericht 2001-26, Deutsches Zentrum fuer Luft­und Raumfahrt, Institut fuer Luft- und Raumfahrtmedizin, Koeln, ISSN 1434-8454
4Gottlob, D. (1999) Ergebnisbericht der Arbeitsgruppe 1 (Laermwirkungen). In: Tagungsband des wissenschaftlichen Symposiums "Umweltkapazitaet von Flughaefen, Aspekte der Laermwirkungsforschung, Schutzziele und Schutzmassnahmen (Environmental Capacity of Airports, Aspects of Noise Effects Research, Targets and Measures of Protection)", October 25/26, 1999 in the Berlin District of Kopenick, pp12-17
5Gesetz zum Schutz gegen Fluglaerm (Law for protection against aircraft noise (Aircraft Noise Protection Act)) (1971) BGBl. I, 282 (last amendments 25 September 1990, BGBl. I, 2106)
6Griefahn, B., Jansen, G., Klosterkoetter, W. (1976) Zur Problematik laermbedingter Schlafstoerungen - eine Auswertung von Schlafliteratur (The Problems of Noise­Induced Sleep Disturbances - an Assessment of the Literature on Sleep). Umweltbundesamt, Bericht 4/76
7Harder, J., Maschke, C., Ising, H. (1999) Laengsschnittstudie zum Verlauf von Stressreaktionen unter Einfluss von naechtlichem Fluglaerm (Longitudinal Study of the Course of Stress Rreactions Induced by Aircraft Noise at Night). Institut fuer Wasser-, Boden- und Lufthygiene des Umweltbundesamtes, Berlin, Bericht Nr. 4/99
8Jansen, G., Linnemeier, A., Nitzsche M. (1995) Methodenkritische Ueberlegungen und Empfehlungen zur Bewertung von Nachtfluglaerm (Considerations and Recommendations Critical of Current Methodology for the Assessment of Nocturnal Aircraft Noise). Z. Larmbekampf. 42: 91-106
9Kallus, K.W. (1995) Erholungs-Belastungs-Fragebogen (EBF): Handanweisung (Relaxation and Stress Questionnaire: Pocket Guideline). Frankfurt, Swets Test Services
10Linnemeier, A. (1995) Die Auswirkungen von Laerm auf den natuerlichen Nachtschlaf des Menschen. Versuch einer Klassifikation und Bewertung von Veroeffentlichungen von 1984 bis 1993 (The Effects of Noise on the Natural Night's Sleep of Human Beings. Attempt to Classify and Assess the Literature from 1984 to 1993). Dissertation, Heinrich-Heine-Universitaet Duesseldorf
11Maschke, C. (1992) Der Einfluss von Nachtfluglaerm auf den Schlafverlauf und die Katecholaminausscheidung (The Influence of Nocturnal Aircraft Noise on the Progress of Sleep and Catecholamine Secretion). Dissertation, Technische Universitaet Berlin
12Maschke, C. (1997) Kriterien fuer schaedliche Umwelteinwirkungen: Beeintraechtigung des Schlafes durch Laerm (Criteria with Respect to Harmful Environmental Effects: the Impairment of Sleep by Noise). Technische Universitaet Berlin, Forschungsbericht im Auftrag des Umweltbundesamtes
13Maschke, C., Hecht, K., Wolf, U. (2001) Naechtliches Erwachen durch Fluglaerm (Nocturnal Awakenings Caused by Aircraft Noise). Bundesgesundheitsblatt - Gesundheitsforschung - Gesundheitsschutz 44: 1001­1010
14Miedema, H.M.E., Vos, H. (1998) Exposure-Response Relationships for Transportation Noise. J. Acoust. Soc. Am. 104 (6): 3432-3445
15Ollerhead, J.B., Jones, C.J., Cadoux, R.E., Woodley, A., Atkinson, B.J., Horne, J.A., Pankhurst, F., Reyner, L., Hume, K.I., Van, F., Watson, A., Diamond, I.D., Egger, P., Holmes, D., McKean, J. (1992) Report of a Field Study of Aircraft Noise and Sleep Disturbance. UK Department of Transport, London
16Ortscheid, J., Wende, H. (2001): Fluglaermwirkungen (Aircraft Noise Effects). Bericht des Umweltbundesamts, Report of the German Federal Environment Agency, revised edition of August 14, 2001
17Samel, A. (1999) Der Laerm und seine Folgen (Noise and Its Consequences). DLR-Nachrichten 96, 22-27
18Santucci, G., Boer, L., Farmer, E., Goeters, K.M., Grissett, J., Schwartz, E., Wetherwall, A., Wilson, G., Yates, R. (1989) Human Performance Methods. Neuilly-sur-Seine, AGARDograph N° 309, NATO-AGARD
19Steyer, R., Schwenkmezger, P., Notz, P., Eid, M. (1997) Der Mehrdimensionale Befindlichkeitsfragebogen MDBF (Multidimensional Feelings Questionnaire MDBF). Goettingen, Hofgrefe-Verlag
20Wysk, P. (1998): Ausgewaehlte Probleme zum Rechtsschutz gegen Fluglaerm III (Selected Problems of Legal Protection Against Aircraft Noise). Z. Larmbekampf. 47: 456-489