How Much Noise Is Too Much Noise?

by Federico Miyara (fmiyara@unrctu.edu.ar)
August 1997

Prof. Miyara teaches Linear Electronics, Acoustics and Psychoacoustics, Noise Control, and Sound Systems at the National University of Rosario (Argentina).


One of the first questions people ask when told that noise can be harmful is: how much noise is actually necessary before human beings are adversely affected?

The answer isn't simple, because the effects of noise involve multiple aspects of people's health and welfare.

From a philosophical point of view, since noise may be defined as any unwanted acoustical stimulus which interferes with human activity or rest, any amount of noise affects people (otherwise it wouldn't be noise). Often, however, noise is better tolerated if it is judged to be unavoidable. The noise of rain, for instance, is much more acceptable than that coming from the isolated but steadily repetitive drops from a leaking tap. Generally speaking, periodical noises are more annoying than random ones.

But when objectively measurable effects are considered, research has shown that there exists a strong correlation between the physical intensity of the stimulus and the extent of the effect. That is particularly true as regards hearing loss. Extensive research has been done for decades on different groups of industry workers, and several criteria have been developed in order to assess specific situations.

One of those criteria, namely that contained in the ISO 1999 Standard, starts by defining "hearing loss" as a permanent increase of the auditory threshold (the minimum audible sound level) affecting the intelligibility of speech. This amounts to some 25 dB for middle-frequency tones. Then a double entry table is used to assess the percentage risk of "acquiring" hearing loss if exposed in the workplace to a certain average sound level for 8 hours per day, 6 days a week for a given number of years:

dBAYears of exposure
51015202530354045
80000000000
851356789107
9041014161618202115
9571724282931322933
100122937424344444135
105184253586062615441
110265571787877726245
115367183878481756447

Whenever working hours are halved, the criterion is applied subtracting 3 dBA from the actual sound level.

Let's take, for example, the case of a disc jockey working at a night-club 3 nights a week exposed to an average sound level of 106 dBA during a 4-hour shift. We have half the days and half the time per day, so 3 dBA must be subtracted twice, yielding 100 dBA. Suppose, further, that the disc jockey starts at the age of 15 and stays in this trade till 30 -- that is, a 15-year exposure. The table reveals a 37 percent risk of suffering speech intelligibility handicap. That is, almost 4 out of 10 such persons will experience difficulties in understanding normal speech -- at the age of 30.

Labor regulations usually allow exposure to levels of 85 dBA or even 90 dBA, but also compel employers to carry out hearing conservation programs, which typically include periodic hearing tests to screen out highly susceptible workers from hazardous noise environments.

A closer look at the table reveals that at 80 dBA the risk is 0% for any extension of working life. Does this mean that 80 dB is the safe ceiling? Not at all. This is because in the definition of "risk" there is another element which for simplicity was not mentioned earlier. As this table is intended to predict the risk of workplace exposure, the percentage of people suffering from hypoacusis just by aging (i.e. presbycusis) has already been subtracted.

There are several studies that reveal that what was formerly considered to be presbycusis is actually "sociocusis" -- that is, hearing loss due to exposure to social or community noise.

In the seventies the U.S. Environmental Protection Agency (EPA) was committed to providing "information on levels of environmental noise requisite to protect public health and welfare with an adequate margin of safety." The task was not easy, because though a large amount of information on workplace exposure to noise was readily available, very little information regarding non-workplace exposure had been published. The approach was thus to extrapolate the available data under a number of reasonable asumptions. The results were published in a famous paper known as "The Levels Documents."

The EPA criterion regarding hearing conservation states that in order to protect virtually all the population against hearing loss, the average sound level should not be greater than 70 dBA during a 24-hour day. Equivalently, it should not be greater than 75 dBA for an 8-hour working day, provided that for the rest of the day the level is kept considerably below that figure.

This criterion differs substantially from the ISO's in that it considers that hearing loss has occurred when the auditory threshold rises only 5 dB instead of 25 dB. It is thus a far stricter criterion.

It should be noted that the EPA limit is an average, meaning that much larger levels can usually be tolerated for brief periods of time. Indeed, each time the duration of the exposure is halved, the safe level increases by 3 dBA, so it would be safe to be exposed to 78 dBA for 4 hours, 81 dBA for 2 hours, 84 dBA for 1 hour, and so on. However, exposure to levels above 100 dBA even for very short periods is not recommended, since some susceptible persons may suffer immediate and irreversible hearing impairment.

As seen above, occupational limits tend to be rather permissive as compared to environmental ones, the reason being that they take into account other factors appart from safety and welfare, such as the technical or economical feasibility of decreasing noise levels at the workplace.

Finally, the EPA document addresses not only the hearing impairment issue but also issues of annoyance and interference with activities. In this case, it states that the average outdoor noise level should be no greater than 55 dBA in the daytime and 45 dBA at night. Indoors, the recommended limits are 45 dBA and 35 dBA respectively. period of time.

Actual environmental noise laws and regulations tend to be somewhat more permissive, usually allowing for corrections under given circumstances. For instance, when the noise source affects the surroundings of an industrial area, a correction of as much as 25 dB is often applied to the basic EPA criterion. Noise from motor vehicles is generally treated in a different way: since each vehicle passes by a given location only for a short period of time, much higher limits are imposed for individual vehicles, and little attention is paid to the cumulative noise level arising from continuous traffic flow.


Return to Readings or Right to Quiet home page