ASSESSMENT OF LOW FREQUENCY NOISE COMPLAINTS

PRESENTED AT INTERNOISE '99

G.P. van den Berg, Science Shop for Physics, University of Groningen

W. Passchier-Vermeer, TNO Prevention and Health

 

INTRODUCTION

Complaints. The occurrence and annoyance of low frequency sounds have been described in a number of publications [e.g. 1,2,3]. In recent years a growing number of people in the Netherlands have brought complaints to authorities or to medical or acoustical experts about persistent low frequency (LF) sounds. To address this problem a number of studies have been executed in the Netherlands in 1998 and 1999. One study dealt with the personal characteristics of complainants themselves (such as age, gender, occupation, hearing threshold, sound sensitivity, time spent home, psychological and health status) [4]. The study shows that complaints may last for years, threatening the complainants’ quality of life and health.

Measurements. According to Dutch regulations, environmental noise is assessed by its A-weighted equivalent sound level, measured or calculated at a specified distance from the façade (or, in some cases, indoors). Due to the A-weighting and the time when measurements are usually carried out, such measurements do not indicate a LF sound unless it is loud. In some cases complainants assume the noise comes from a specific source such as an air conditioning unit, fan or from a nearby or more distant office or factory. In most cases the source is unknown. When the investigator does not perceive the LF sound, the investigation ends. This is often the case, leaving both parties dissatisfied. Therefore, in a second study measurements have been made of low frequency sound levels in dwellings of complainants at a time they are indeed perceived by the complainant. The results have been compared with measurements in dwellings of a control group [5].

Assessment. In the case of a complaint to authorities about LF sound it must be judged whether the measurement results do or do not indicate a LF sound. A third desk study therefore assessed a frequency dependent level above which LF sounds can reasonably be considered to be perceived and therefore act as a source of stress and annoyance [6].

Both measurement [5] and assessment [6] procedure have been included in a Guideline for LF Sound, published by the Dutch Noise Annoyance Foundation (NSG) and supported by the Dutch Ministry of Housing, Spatial Planning and the Environment [12]. The guideline does not specify a method to identify the sound source. This paper clarifies the procedures and considerations it is based on.

LOW FREQUENCY SOUND PERCEPTION

Perception. Complainants usually describe a perception of humming or engine-like sounds or a feeling of pressure or vibration. This may be perceptible for days or weeks, then be less intense or imperceptible and then appear again. Or, it may be perceptible more or less continuously for years. For all complainants it is an impairment to sleep. As such the unwanted exposure is a threat to complainants’ health: the long-term night time perception of LF sound is an important stressor at night, usually adverse feelings persist over day time, and the perception is related to an increased use of tranquilizers and sleeping drugs [4]. In many cases other people (house mates, visitors, the investigator) do not notice any specific LF sound, which is the more frustrating to complainants.

All complainants indicate that, during the period the sound is perceptible to them, it is continuous and constant or it varies in loudness with a period of several seconds (approximately at breathing frequency). Most authors state that a LF sound that raises complaints has tonal components [6]. In a suitably time-averaged LF spectrum, the LF sound should therefore be visible as a (local) maximum in the spectrum. To eliminate any uncertainty about the presence of the LF sound in the measured spectrum, it must be from a period the complainant positively stated the sound was perceptible.

Assessment method. Criteria to assess LF noise for regulations to protect the majority of a population with regard to annoyance have been proposed by the Swedish Socialstyrelsen [7], Vercammen [8], DIN [9] and ANSI [10]. In contrast, the method specified below seeks to assess whether it is possible to identify a LF sound possibly causing complaints at a time specified by a complainant.

Studies show that the impact of a LF sound is not strongly related to its level: whenever it is perceived, it may be annoying and cause complaints (for references, see [6]). Therefore, it is assumed that a LF sound should be considered potentially annoying when it is audible, i.e. above the complainants individual hearing threshold level at the frequency(ies) of the LF sound. As hearing threshold levels in the LF region are not usually assessed by common audiological test methods, the method specifies hearing threshold levels that are considered to be representative for the complainants. Most complainants are elder people (average age: 54 [4]). Therefore, the hearing threshold levels in the relevant frequency range of an otologically unselected population aged 50 - 60 years is chosen as criterion. To limit the possibility that it is incorrectly concluded that the complainant does not perceive a LF sound when he/she actually does, the binaural hearing threshold level for 10% (HTL10%) of the population mentioned is taken as a reference. For the population under consideration 90% of them actually has worse hearing. Therefore, the curve that presents the values of HTL10% (specified as 1/3 octave band sound pressure levels in dB) as a function of frequency is the curve separating the best hearing 10 % of the population from the other 90 %. Table 1 presents hearing threshold levels assessed from available information on hearing threshold levels in the lower frequency region (for references, see [6]). The available data did not allow the specification of hearing threshold levels at a percentage lower than 10%. It is estimated, however, that the HTL5% is about 2 dB below HTL10%. In the guideline [12] HTL10% between 20 and 100 Hz is used as the reference curve for the assessment of the audibility of LF sounds. It is 4.5 dB below the median hearing threshold level of young, otologically selected adults according to ISO 226 [12].
 

RECOMMENDED MEASUREMENT AND ASSESSMENT PROCEDURE

The LF sound is recorded first and the recordings are then analysed.
Recording the sound. The purpose of the guideline is to determine whether the perceived LF sound can be detected and characterised by measurement. The measurement method should therefore be optimal for the detection of the LF sound, and does not necessarily have to conform to general acoustical practice. Except for situations in which a LF sound is obvious, it is strongly recommended that the complainant records the sound him/herself at a time and place the complainant deems relevant -which is mostly at night. To be able to test the reproducibility of the recordings, at least two are made when the LF sound is audible, preferably at different nights. A recording is also made, if possible, when the LF sound is not audible according to the complainant. The microphone is placed at a position where the sound is clearly audible according to the complainant. Preferably this is a standard position in a corner of the room where the LF sound is considered loud (usually the sleeping room), at a height of 1 to 1.5 m and 0.4 to 0.5 m from both walls. To record the sound a microphone and digital recorder are recommended with a flat frequency response above 20 Hz and an internal noise level well below (³ 10 dB) the reference curve.

During the recording the investigator’s presence is not necessary and in some instances even unwanted: he may not be able to recognize or hear the sound (it is not there, not loud enough or masked) or focus on other sounds, which may lead to confusion or misunderstanding. Although the complainant has to be instructed how to use the recorder, the absence of the investigator during the recordings is also cost-effective, the more so if the investigator would have to come back several times for (successful ?) measurements. Other procedures apply if it is important to witness the measurement (e.g. if practical or legal measures towards the source are considered). Then the investigator should be present; but this is usually after the characteristics of the LF sound have been identified by the procedure specified here.

The complainant is instructed verbally and in writing: a) to operate the recorder; b) to avoid disturbing sounds; c) to make at least two recordings of 10-15 minutes each when the LF sound is perceptible and, if possible, when it is not perceptible in comparable circumstances (nights); d) to give a description of the LF sound (such as: loud, noticeable, not audible, etc.) and, if not recorded automatically, to note the date and time of the recording. After a practical training of this procedure the equipment is left at the complainant’s dwelling until sufficient relevant recordings have been made.

Analysis. The first step in analyzing the recordings is to provide time histories of the overall A- and C-weighted sound levels. This is to indicate the background sound level and to identify disturbances that have to be left out of further analysis, such as passing cars, voices, footsteps, etc. Then the equivalent unweighted 1/3 octave band sound pressure levels of the part of the recordings without disturbances are determined. Modern instruments give 1/3 octave bands of all audio frequencies down to several hertz. The minimum frequency range to be considered is 20 - 100 Hz. If possible all 1/3 octave bands from 10 - 200 Hz are assessed to be able to see relevant maxima at frequencies adjacent to the frequency band of 20 - 100 Hz. Then the differences between the measured equivalent 1/3 octave sound pressure levels and the values of the reference curve (bold numbers in table 1) are calculated for the frequency range under consideration.

In cases where one or more 1/3 octave band sound pressure levels are above the reference curve, more precise information on the frequencies of the LF sound may be obtained through a 1/12 or 1/24 octave spectrum or a narrow bandwidth line spectrum. This may be useful to identify the source in a later stage of the investigation.

Assessment. The LF sound is considered audible and potentially annoying if the equivalent 1/3 octave sound pressure level is above the reference curve at one or more frequencies. The investigation has then identified the LF sound most likely causing the complaint. The different recordings may be compared between them and they may be compared with the loudness judgement of the complainant to support the conclusion.

If the equivalent 1/3 octave band sound pressure levels are all below the reference curve, the conclusion is that either the complainant is one of the 10 % best hearing (this can hold only if the difference with the reference curve is small), or perceives a sound of a lower (< 20 Hz) or higher (> 100 Hz) frequency, or perceives an internal sound (e.g. caused by blood flow or tinnitus).

REFERENCES

1. B. Berglund, P. Hassmén, R.F. Soames Job: Sources and effects of low-frequency noise, JASA Vol. 99 (5), mei 1996
2. K. Persson, R. Rylander: Disturbance from low-frequency noise in the environment: a survey among the local environmental health authorities in Sweden, JSoundVibr vol. 121 No. 2 (1988)
3. G.P. van den Berg: Sound exposure measurements in cases of low frequency noise complaints, proc. Internoise 1998, Christchurch
4. C. Gielkens-Sijstermans, T.H. Collijn, A.W. Jongmans-Liedekerken: Sensitivity for low frequency sounds: a study into possible factors (Gevoeligheid voor laagfrequent geluid; een studie naar mogelijke factoren), GGD Oostelijk Zuid-Limburg, 1998 (in Dutch).
   An abstract is published as: A.W. Jongmans-Liedekerken, C. Gielkens-Sijstermans: Susceptibility to low frequency noise, proceedings ISEE/ISEA’99, Athens
5. G.P. van den Berg: Case control study in low frequency sound measurements, proc. Internoise 1999, Fort Lauderdale
6. W. Passchier-Vermeer: Assessment of low frequency noise in dwellings (Beoordeling laag-frequent geluid in woningen), TNO Preventie en Gezondheid (rapport 98.028), 1998 (in Dutch)
7. Socialstyrelsen: Indoor Noise and High Sound-Levels: General Guidelines issued by the Swedish National Board of Health and Welfare (Socialstyrelsen), 1996
8. M.L.S. Vercammen: Low-Frequency Noise Limits, J Low Freq Noise Vib 11 no.1 pp.7-13, 1992
9. DIN: Measurement and assessment of residential low frequency sound immission (Messung und Bewertung tieffrequenter Geräuschimmissionen in der Nachbarschaft), Deutsches Institut für Normierung (DIN 45680), Berlijn, 1997 (in German)
10. ANSI: Quantities and procedures for description and measurement of environmental sound. Part 4 of Noise assessment and prediction of long-term community response, American National Standards Institute, 1997
11. ISO: ISO 226:Acoustics-Equal Loudness contours for otologically normal listeners, part 1
12. NSG: Nederlandse Stichting Geluidhinder: Guideline for Low Frequency Noise (RichtlijnLaagfrequent Geluid), Delft 1999 (in Dutch)