Community Response to High-Energy Impulsive Sounds: An Assessment of the Field Since 1981 (1996)

The working group's major conclusions and recommendations are as follows:

1. The weight of accumulated evidence since publication of CHABA (1981) makes it inadvisable to continue to rely on the dosage-response relationship recommended in that earlier report. However, information currently available does not support construction of a single dosage-response relationship that is fully appropriate in all circumstances of impulsive noise exposure.

2. Two alternative procedures are therefore recommended, for use in circumstances described in Table 4:

Method 1 (adaptation of Green and Fidell, 1991, model): The prevalence of a consequential degree of annoyance due to high-energy impulsive noise exposure may be predicted from CDNL values as shown in Eq. 24

where D is the constant 61.1 chosen to minimize the variance of the set of available field observations of the annoyance of high-energy impulsive noise exposure. Note that CSEL values of individual events that are summed to form CDNL must exceed 85 dB (75 dB at night) to avoid overpredicting community response to nonhigh-energy impulsive sounds.

Method 2 (level-dependent summation): The prevalence of a consequential degree of annoyance due to high-energy impulsive noise exposure may alternately be predicted by cumulating CSELs of individual impulsive sounds in a manner consistent with the annoyance weighting identified in Chapter 3, in the section on the level-dependent summation approach.

3. Community response to sonic booms can be assessed using either of the above two procedures. Method 2 is preferable for assessment of community response to explosive noises from artillery or mining noise, especially when more detailed information is available about the distribution of CSEL values of individual high-energy noise events.

4. Since neither of the above predictive relationships is known to be causal, both should be regarded as interim expedients.

The most obvious remedy for the paucity of information about community response to high-energy impulsive noise exposure is the conduct of additional large-scale social surveys. This is an unlikely remedy for several reasons. First, opportunities for conducting large-scale field studies involving intentional exposure of entire communities to sonic booms or other high-energy impulsive sounds have been all but foreclosed in the United States since passage of the National Environmental Policy Act of 1969.

Second, opportunities to conduct surveys of adventitious exposure to high-energy impulsive noise are limited. Because no civil aircraft fly supersonically over land, and because the military confines its overland supersonic operations to areas of relatively low population density, only small populations have yet experienced relatively high levels of exposure to sonic booms on a regular, long-term basis. Blasting, artillery training, and other sources of nonaircraft, high-energy impulsive noise are also generally experienced only by localized populations.

Third, it is unlikely that information about effects of sonic booms in the high-exposure ranges of CDNL can ever be collected. CDNL values in studies of community response to high-energy impulsive noise exposure that have been conducted to date are all confined to the region below L_Cdn = 70 dB. No community experiences the numbers of daily supersonic flights necessary to produce greater long-term impulsive noise exposure, nor is it practical to create or credibly simulate such exposure in the long term.

Uncertainties about the characterization of high-energy impulsive noise exposure are one of the major limitations on the ability of purely acoustic variables to account for large amounts of variance in community response. Future studies would profit greatly from:

• Direct empirical measurements of noise exposure in preference to predictive modeling of exposure;

• Collection of information about the distribution of individual impulsive noise events in addition to CDNL values;

• Measurements of characteristic indoor noise levels associated with time-averaged cumulations of outdoor exposure; and

• To the extent possible, characterization of personal as well as place-oriented noise exposure.

Several measures would improve the utility of data collected in studies in which test participants are asked to judge the immediate annoyance of individual noise events:

• Careful attention should be given to sound reproduction capabilities and listening environments so that the very low-frequency content of high-energy impulsive sounds is accurately presented for judgment; so that spectra of test signals covary appropriately with level; and so that valid estimates of at-ear sound levels can be made;

• Secondary emissions (rattling sounds) should be fully controlled and accurately measured at test participants' ears; and

• Judgments of the annoyance of both impulsive and nonimpulsive sounds should be solicited over a wide range of levels.

These recommendations should be reviewed as additional information about the annoyance of high-energy impulsive noise exposure (preferably from field studies) becomes available. They should also be reviewed from other theoretical perspectives as advances are made in understanding the origins of noise-induced annoyance. It is possible, for example, that nonacoustic predictor variables might account for useful amounts of variance in community response to infrequent, individually memorable episodes of high-energy impulsive noise exposure.