Metrics for Assessing Environmental Noise
Selecting a metric for assessment of environmental noise is no simple task, because it must reflect the impact on people. No single metric can describe all responses in all situations.1 Context, expectations, and people’s experiences and circumstances all affect their responses. Hence, levels of community response (e.g., annoyance) may vary from community to community, just as individual responses vary from person to person, even if noise levels do not change. However, one consistent finding has been that changes in noise exposure do affect individual and community responses and that increases in man-made noise usually have a negative impact. This is illustrated by the Schultz curves later in this chapter.
Thus, it is important to understand which characteristics of noise elicit a negative response and how exposure to noise with those characteristics affects people’s lives. The metric chosen or developed for measuring community noise must reflect this human response and must be taken into account in making policy decisions.
Fifty years ago, when noise metrics were developed, the choices were based on simpler calculations and technologies and the acoustical quantities that could be predicted by sound propagation models used at the time. Although much more sophisticated measurements can be made today, many still consider these “older” metrics valid and continue to use them. However, with modern instruments (see Appendix E), much more accurate measurements and predictions can now be made of people’s reactions to noise.
A meaningful metric, or set of metrics, translates sound pressure-time history measurements into a prediction of the effects of noise, such as annoyance, sleep disturbance, changes in health, interference with understanding speech, and ability to learn. Ideally, this translation should be based on context, expectations, and personal situations and preferences, in addition to noise information, and should account for a distribution of responses, including responses of vulnerable populations, such as children. Unfortunately, a holistic model of community response is still beyond present capabilities.
One fundamental issue that must be considered in the choice of an environmental noise metric(s) is the purpose for which the metric will be used:
to implement public policy on noise immission from one or more sources
to provide information about noise exposures in a form understandable to the public
to assess a noise situation in terms of noise control engineering
The metrics to accomplish these purposes may differ, but all three relate directly to the impact of noise on the community. For example, a metric to inform decisions about noise control engineering strategies should result in reducing the noise impact, which would then be reflected in the policy metric(s) and the public information metric(s).
As new research results become available and accessible, they should influence the choice of metrics for the three purposes listed above. The results of such research may result in complex calculations that include many variables and may better quantify individual reactions to sound. Some modern procedures, such as calculation of loudness, are more complex than earlier methods, but available computational procedures make the results widely available.
Much of this chapter recounts the evolution of noise metrics and their applications to public policy. This history includes criteria originally used by the U.S. Environmental Protection Agency (EPA) to select a noise metric and the rationale Europeans have used for using a curve passed through highly variable data to determine what percentage of a population is “highly annoyed” by a given noise. In ad-
dition, alternative metrics are described that may be easier for the public to understand than the day-night average sound level (DNL).2
LOUDNESS AND A-WEIGHTING
Arguably the modern history of noise metrics began in the 1930s with the search for a way to describe the loudness of sound. This led to the definition of weighting networks for sound-level meters and, because of limitations on the capabilities of calculating sound pressure levels at that time, a single frequency-weighted value—either A-weighted or C-weighted—came into common usage.
In an early attempt to determine the loudness of sound (using discrete-frequency tones), Fletcher and Munson (1933) found that the loudness of a tone depends on both its amplitude and its frequency. Knowing this dependence, they were able to develop a set of equal-loudness curves. In modern terms the unit of loudness is the phon. For example, a 1,000-Hz tone with a sound pressure level of 40 dB has a loudness of 40 phon. At this loudness level the sound pressure level of tones between 1,000 and about 5,000 Hz is generally lower than 40 dB, and the sound pressure level of tones below 1,000 Hz and above about 5,000 Hz is higher than 40 dB.
The sound-level meter was standardized in the early 1930s when microphones and electronic circuits were being developed. Ideally, the standard sound-level meter would have a single-number description of the sound at a given point in space. The best description at the time came from the studies by Fletcher and Munson, who clearly showed that the shape of the equal-loudness curve was dependent on both the amplitude and the frequency of sound. Thus, using the linear electronic circuits of the time, a few curves had to be selected based on the amplitude of the sound. One of the curves selected, which is very close to the 40-phon curve, was designated as “A-weighting.” Another, which was nearly independent of frequency, was designated as “C-weighting.” A third curve, the “B-weighting” curve, which fell between the A and C curves, has long since fallen out of favor. A-weighting and C-weighting are still used today, although the shape of the curves has changed somewhat to provide a standardized mathematical description in terms of poles and zeros of a transmission network.
Work on improving the calculation of loudness based on measurement of the spectrum of sound continued. The best-known early work in the United States was by S. S. Stevens and in Germany by Eberhard Zwicker. Stevens’s Mark VI and Zwicker’s work on loudness were standardized by the International Organization for Standardization (ISO, 1975). Later work by Brian Glasberg and Brian Moore in the United Kingdom was the basis for the American National Standard on computation of loudness (ANSI, 2007).
Over the years, A-weighted levels were found to correspond reasonably well to human response, especially for noise spectra in typical offices. Single-number methods of rating noise in offices and other building spaces were also developed, including so-called noise rating curves (NR curves—a curve tangent method of obtaining a single number from an octave band spectrum) and ratings based on loudness and A-weighting.3
METRICS FOR MEASURING COMMUNITY REACTION TO NOISE
One early attempt to develop a metric for forecasting community response to noise was made by Stevens et al. (1955). Unlike the DNL, this metric included nonacoustical factors as well as noise levels and yielded a “composite noise rating.” This rating was then plotted against a scale of community responses—vigorous community action, threats of community action, widespread complaints, sporadic complaints, and no observed reaction. A few case studies showed a reasonable correlation between the measurement and response but with considerable scatter. Community noise levels were determined by measuring the average octave band levels in the community averaged in space and time. A curve tangent method was used to reduce the octave band data to a single-number rating.
Day-Night Average Sound Level
After EPA established the Office of Noise Abatement and Control and after passage of the Noise Control Act of 1972, EPA was faced with the task of developing a metric for community noise with the following characteristics (EPA, 1974):
The measure should be applicable to the evaluation of pervasive long-term noise in various defined areas and under various conditions over long periods of time.
The measure should correlate well with known effects of the noise environment on the individual and the public.
The measure should be simple, practical, and accurate. In principle, it should be useful for planning as well as for enforcement or monitoring purposes.
The required measurement equipment, with standardized characteristics, should be commercially available.
A-weighting is less useful for measuring human response to sound when the spectrum has a large low-frequency component, when high-amplitude peaks in the spectrum are in the 2- to 4-kHz range, and when the sound is tonal or impulsive.
The measure should be closely related to existing methods currently in use.
The single measure of noise at a given location should be predictable, within an acceptable tolerance, from knowledge of the physical events producing the noise.
The measure should lend itself to small, simple monitors that can be left unattended in public areas for long periods of time.
EPA also published its rationale for choosing A-weighting and for leaving open the possibility of using a different metric in the future (EPA, 1974; von Gierke, 1975):
With respect to both simplicity and adequacy for characterizing human response, a frequency-weighted sound level should be used for the evaluation of environmental noise. Several frequency weightings have been proposed for generalusein the assessment of response to noise, differing primarily in the way sounds at frequencies between 1000 and 4000 Hz are evaluated. The A-weighting, standardized in current sound level meter specifications, has been widely used for transportation and community noise description. For many noises, the A-weighted sound level has been found to correlate as well with human response as more complex measures, such as the calculated perceived noise level or the loudness level derived from spectral analysis. However, psychoacoustic research indicates that, at least for some noise signals, a different frequency weighting which increases the sensitivity to the 1000–4000 Hz region is more reliable. Various forms of this alternative weighting function have been proposed; they will be referred to here as the type “D-weightings.” None of these alternative weightings [have] progressed in acceptance to the point where a standard has been approved for commercially available instrumentation.
It is concluded that a frequency-weighted sound pressure level is the most reasonable choice for describing the magnitude of environmental noise. In order to use available standardized instrumentation for direct measurement, the A-frequency weighting is the only suitable choice at this time. The indication that a type D-weighting might ultimately be more suitable than the A-weighting for evaluating the integrated effects of noise on people suggests that at such time as a type D-weighting becomes standardized and available in commercial instrumentation, its value as the weighting for environmental noise should be considered to determine if a change from the A-weighting is warranted.
The decision to add 10 dB4 in measuring nighttime levels and the selection of a two-period (day-night) metric rather than a three-period metric (day-evening-night) was based on community reaction studies at the time and tests that showed little difference between a two-period and a three-period metric. Thus, the DNL (A-frequency weighting for both day-time and nighttime levels and a 10-dB increase in measuring system gain at night) came into being for the evaluation of community noise.
In the United States, DNL and the percentage of persons highly annoyed (discussed in the next section) are widely used, especially by the Federal Aviation Administration (FAA). The Federal Highway Administration uses A-weighting and the average sound pressure level during the busiest traffic hour as a measure of community impact. The difference between C-weighted and A-weighted levels is used as an indication of the low-frequency content of the sound, and the sound exposure level (see Appendix A) is used to evaluate sounds of finite duration—for example, an aircraft flyover.
Day-evening-night sound level is widely used in Europe. In some countries, Lday and Lnight, (average A-weighted sound pressure levels) are used in addition to or instead of a DNL-type metric. None of these metrics takes into account the time of night when the noise occurs, even though noise appears to cause greater sleep disturbance at the beginning and end of the night.
Several issues have arisen from the use of DNL and the percentage of persons highly annoyed: no one actually “hears” a DNL; there is a high variability from study to study around a nominal Schultz curve; and in many situations “highly annoyed” is not an appropriate measure of human response. Although the percent highly annoyed and DNL approach has been widely endorsed, variability around a nominal Schultz curve is troubling, and there are reports that this approach is not sufficient to predict community response (Fidell, 2002). Attitudinal and personal variables impact people’s responses and are, to some extent, the reason for scatter (Fields, 1993; Flindell and Stallen, 1999; Miedema and Vos, 1999).
As shown in Figure 3-1, some researchers (Miedema and Oudshoorn, 2001) have found in their analyses of survey results that the nominal Schultz curve appears to depend on the noise source (e.g., aircraft, road traffic, rail traffic). In addition, DNL is a relatively insensitive measure of sleep disturbance and thus is not an appropriate metric for predicting awakenings in sleep disturbance studies. Finally, A-weighting is not the best weighting for measuring noises with unusual spectra (e.g., excessive high- or low-frequency noise or noise that has unusual peaks in its spectrum). For sounds with levels that evolve over time, the most appropriate
weighting should change with the level; typically, however, only one weighting is used.
Percentage of Persons Highly Annoyed
The next major event in the selection of a noise metric was a study by Schultz (1978) of surveys of community reaction to noise. Schultz went back to original data to estimate the percentage of the population “highly annoyed” as a function of DNL. Even at that time, it was recognized that, for a variety of reasons, there was considerable scatter in the data. Nevertheless, Schultz proposed that a single curve (the Schultz curve) drawn through the data should be used as a measure of community response. Later studies led to modifications of the Schultz curve (Fidell et al., 1991; Finegold et al., 1994). In the latter study, three curves were compared (see Figure 3-2), and a U.S. Air Force logistic curve was defined
The scatter in the highly annoyed response, compared to scatter in the average curve, was presented by Miedema and Vos (1998) and has been commented on by several subsequent researchers (e.g., Schomer, 2005). The first problem with scatter is that it causes great uncertainty in the prediction of community reaction. A second problem is that community reaction (percent highly annoyed) appears to depend on the source of the noise; for example, responses to aircraft noise, road traffic noise, and rail noise vary, even if the noises have the same DNL (see Figure 3-1). The question that must be answered is whether the variability in response is due to the nature of the noise source or reflects how the metric is calculated.
Consultants and other professionals are often asked to study community noise issues and recommend remedial action. Predictions of community response should not be based only on variations of the Schultz curve. It has been known for many years (Stevens et al., 1955) that nonacoustical factors influence community reaction to noise. Thus, at a minimum, temporal and spectral variations must also be taken into account.
Based on work by EPA, Schomer (2002) proposed modifications to DNL to account for tonality, impulsiveness, background noise, type of community, and other factors. Schomer also showed how this modified approach could be used to reduce variances in the survey data on which the Schultz curve is based.
The Federal Interagency Committee on Noise (FICON, 1992) endorsed the use of percent highly annoyed and DNL as metrics for assessing community noise around airports and recommended that the equation above be accepted as showing the definitive relationship between percent highly annoyed and DNL (see also Finegold and Finegold, 2002). Response curves for community annoyance have now been standardized nationally (ANSI, 2005) and internationally (ISO, 2003).
The science of measuring environmental noise has progressed rapidly in the past decade as computer technology has come on line to provide rapid data acquisition and analysis in small portable packages. The end result has been a revolution
in the type and complexity of measurements and calculations that can be made in analyzing environmental noise. This section provides a more detailed description of presently used metrics and a variety of alternative metrics that are well within the capabilities of modern instrumentation.5
A Different Frequency Weighting
An alternative to A-weighting (i.e., D-weighting) could be considered. As noted earlier, this weighting was considered by the EPA in 1974 but rejected because there was no standard shape for the curve.
Perceived Noise Level
Community reaction to noise from jet planes led to important events in the development of noise metrics. The problem, which arose in 1956, is described in an autobiography by Beranek (2008). According to measurements made with a standard sound-level meter, the noise produced by a Boeing 707 jet airplane and that by a propeller airplane (Super Constellation) were equal. However, subjective testing showed that the 707 was considered much noisier; by subjective measures, the A-weighted sound pressure levels of the 707 would have to be significantly reduced to be considered as noisy as the Super Constellation. This early test of the usefulness of A-weighted levels in judging noisiness led to further evaluations of the relative noisiness of propeller-driven and jet airplanes and the development of the concept of “perceived noisiness” (Kryter, 1960; Kryter and Pearsons, 1962, 1963).6
Perceived noise level (PNL) was used in the development of specifications of noise emissions from airplanes for regulatory purposes in 1969 and is still used to certify airplanes today. When the perceived noise-level metric was adopted, it was possible to compute it only with a large amount of equipment. Today, it can be done with a handheld sound-level meter. D-weighting simplifies the PNL calculation, but neither PNL nor D-weighting solves the decibel issue, which relates to explaining noise to the public.
Historically, the method of calculating PNL was similar to the method of calculating loudness. Today, several methods can be used to calculate loudness, all of them with a handheld sound-level meter. Loudness that exceeds some agreed-on value a given percentage of the time also can be calculated. On a linear scale (as opposed to a logarithmic scale), a doubling of the value of the calculated value corresponds to a doubling of the loudness. This may be easier to explain to the public than a metric that uses the phon (which uses a logarithmic scale) as a unit of loudness. For sounds in the midfrequency range, an increase in A-weighted level of 10 dB corresponds to a doubling of loudness.
Standard methods of calculating speech interference are available, and the values may be translated into effects that are easier for the public to understand than DNL. For example, the difficulty of communicating over a given distance between speaker and listener may be quantified in terms of percentage of speech likely to be understood. Speech interference can be affected by the fact that hearing loss increases with age and usually starts at high frequencies. Thus, the ability to distinguish consonants that have high-frequency content such as “s” and “th” is diminished.
Nighttime Sleep Disturbance
In Night Noise Guidelines for Europe, published by the World Health Organization (WHO, 2007), sleep disturbance is related to the nighttime level designated as Lnight, although researchers also use indoor LAmax and indoor A-weighted sound exposure level (ASEL) when investigating the relationship between awakenings and noise. The temporal pattern of noise at night, however, is known to influence sleep disturbance. This problem is addressed to some extent in a new American National Standard (ANSI, 2008), in which terms such as the likelihood of awaking, are used; the new standard may be more understandable to the public than the day-night average level or the nighttime level used in Europe.
METRICS FOR COMMUNICATING WITH THE PUBLIC
An often-cited shortcoming of DNL is that the public does not understand what it means. Over the years, various people have advocated using supplemental metrics that describe noise in ways that are more understandable to the majority of people (FICAN, 2002). Metrics used to supplement DNL include time above (a certain level), number of events above a given value of the ASEL, number of loud events above a certain ASEL in a given period, and single-event descriptors such as LAmax and ASEL. Most advocate using a group of metrics to give a fuller picture of the potential impact of the exposure and explaining that these measures supplement metrics such as DNL. The same argument can be made for using a group of metrics when addressing other measurements or predicting a variety of impacts (Eagan, 2007), such as the number of occurrences of speech interference; when noise levels inside buildings exceed recommended levels for a particular activity, such as learning in schools (ANSI, 2002); or the likelihood of being awakened based on predicted indoor single-event metrics (ANSI, 2008).
For a description of the instruments, see Appendix E.
For a general assessment of human reaction to aircraft noise, see Beranek et al., 1959.
The number of events has been recognized as an important factor in noise exposure, and it is included in metrics that are or have been used to predict annoyance; alternatives to DNL, such as the Noise Exposure Forecast (NEF) system used in Canada (Transport Canada, 2005) and elsewhere; and the Noise and Number Index (NNI) that was used prior to 1990 in the United Kingdom. The NEF metric is based on effective perceived noise level as well as the number of events; hence it takes into account some of the impact of tonal components and impulsiveness on annoyance. NNI is also based on a very basic loudness measure, perceived noise level in decibels and number of events. Analysis of data from a study at U.K. airports in 1982 and another study in 2005 showed that the relationship between annoyance and A-weighted equivalent level had changed. However, by combining a measure of average noise exposure with the number of events, it was possible to develop a metric that worked consistently for both studies (ANASE, 2007).
NOISE METRICS FOR RURAL/NATURALLY QUIET AREAS
Neither day-night average sound level nor percent highly annoyed is an appropriate metric for measuring noise in naturally quiet areas. Because of the logarithmic nature of the decibel, short-duration sounds of high amplitude compared with background noise can significantly increase the day-night level, even though the sound remains at the background level most of the time. As for percent highly annoyed, this is hardly the best measure of satisfaction for areas where quiet and solitude are valued. In addition, it can be difficult to measure very low sound pressure levels. A-weighted levels of 40 dB are at the upper end of the range, and lower levels can be at or even below the levels measurable with conventional sound-level meters.
Nevertheless, some quantification of noise impact is clearly needed in these areas as a basis for establishing public policy, which usually means regulatory action. The classic definition of noise is “unwanted sound,” so the source of sound must be identified, either as part of the natural soundscape or not. Thus, simple metrics like sound pressure level are clearly not appropriate. For example, an airplane overflight may have a much lower sound pressure level and shorter duration than sound from a rushing stream, but the former is considered noise and the latter is considered sound. The method of assessment of the noise environment should also take into account the likely long-term impact on animals that use, for example, very low level sounds (perhaps inaudible or unnoticed by people) to locate prey or predators.7
INTERNATIONAL ACTIVITIES RELATED TO NOISE METRICS
The International Commission for the Biological Effects of Noise holds meetings at five-year intervals. In 2008 the meeting was held in the United States, but most of the participants came from other countries, as did the presenters. Truls Gjestland of Norway presented a summary report on research in the past five years related to the effects of community noise, specifically annoyance. Although some research has been done in Japan, he said, not many significant projects had been undertaken. However, he noted that at least three different versions of the Schultz curve had been developed, all of them based generally on the same datasets (Gjestland, 2008). Around the same time Lawrence Finegold of the United States presented a review of major noise-related policy efforts around the world during the same time period (Finegold et al., 2008).
In 1996 the European Union (EU) published The Green Paper, which established new noise programs that are used to address noise issues today (EC, 1996). European Directives have been issued concerning noise emissions from consumer products, and an EU Environmental Noise Directive (END) in 2002 led to the development of noise-mapping and, in a few cases, action plans that require noise metrics (EC, 2002a). Related activities include the HARMONOISE and IMAGINE projects (http://www.imagine-project.org/).
European Metrics (Indicators)
A-frequency weighting for determining sound levels that have been standardized in the United States and internationally is widely used in Europe. However, as discussed elsewhere in this chapter, frequency weighting alone is not enough to define a metric. A Working Group (WG1) that produced a report in 2000, Position Paper on EU Noise Indicators, in support of future European noise policies, identified five criteria for selecting an indicator: validity, practical applicability, transparency, enforceability, and consistency. Although this report was not an official EU document, the metrics recommended therein are now widely used (EC, 2000).
WG1 recommended that two indicators, both based on A-frequency weighting, be used for reporting data on noise exposure. These indicators were designated LEU and LEUN but today they are widely known as the day-evening-night sound level, DENL, and the equivalent sound pressure level during the eight-hour nighttime period, Lnight. The group explained, and questioned, the rationale for using 5 dB as level weighting for the evening period and 10 dB for nighttime. Nighttime was nominally designated as eight hours, from 11:00 p.m. to 7:00 a.m.; daytime, 12 hours; and evening,
four hours (with some variation, depending on the country). For general purposes, the long-term average A-weighted sound pressure level, LAeq, was used. The WG1 report also recognized that the character of noise (impulsive, tonal, etc.) may affect human response. Thus, corrections to the metrics may be necessary, and A-frequency weighting may not be appropriate for measuring low-frequency noise.
The WG1 report was also the basis for metrics specified in the 2002 END that led to noise mapping. The directive also suggests supplemental metrics based on the WG1 report (EC, 2002a).
Dose-Effect Relationships (Exposure-Response Relationships)
Another Working Group (WG2) on Health and Socio-Economic Aspects of Noise also produced a report, again not official EU policy. In Position Paper on Dose-Response Relationships Between Transportation Noise and Annoyance (EC, 2002b), the group recommended that the percent highly annoyed (%HA) be used as a measure of community response to noise. Updated and modified Schultz curves, based on the work of Miedema and colleagues (e.g., Miedema and Oudshoorn, 2001) for the %HA as a function of day-evening-night sound level, are used to measure road traffic, rail traffic, and aircraft noise. WG2 also acknowledged the variability from study to study in the mean values in Schultz curves (e.g., Gjestland, 2008) but still supported the use of “norm” curves:
Substantial deviations from the predicted percentage [of] annoyed persons must be expected for limited groups at individual sites because random factors, individual and local circumstances and study characteristics affect the noise annoyance. However, in many cases the prediction on the basis of a “norm” curve that is valid for the entire population is a more suitable basis for policy than the actual annoyance of a particular individual or group. For example, a “norm” curve is useful when exposure limits for dwellings and noise abatement measures are discussed. Equity and consistency require that limits and abatement measures do not depend on the particularities of the persons and their actual circumstances. For similar reasons, a “norm” curve also can be used to estimate the number of annoyed persons in the vicinity of an airport, road, or railway when different scenarios concerning, e.g., extension of these activities or emission reductions are to be compared. That the norm curve does not take local circumstances or reactions to a change in exposure itself into account, is considered to be an advantage for many purposes. Equity and consistency of policy would not be served if in each case the actual annoyance is taken as the (only) basis for these evaluations. The use of “norm curves” or “norm thresholds,” which are valid for the entire population (or a particular sensitive subgroup), is common practice when exposures to other environmental pollutants, such as air pollutants or radiation, are evaluated. There they are used for the evaluation of an individual situation, irrespective of the population in that situation. It is recommended to take the same approach in the case of environmental noise and use the same curve irrespective of the population in the situation evaluated.
In 2004, WG2 produced Position Paper on Dose-Effect Relationships for Nighttime Noise, again not an official EU document. In this paper the metric used was Lnight, as defined above as the measure for sleep disturbance. Based on questionnaires, curves similar to Schultz curves were developed, the ordinate being the percent highly disturbed and the abscissa being the nighttime noise level. An effort was made to relate single events to the nighttime sound level (EC, 2004).
Annoyance and the Microstructure of Noise Exposure
Several studies have been published, mostly in connection with the EU-funded SILENCE project (www.silence-ip.org), on the importance of the microstructure of a noise exposure situation. The argument is that equivalent levels do not “tell the full story.” Different traffic noise situations with the same equivalent level may be assessed differently with respect to annoyance. This is important information for decisions about how to reduce the negative impact of road noise through traffic management measures. Laboratory experiments have provided several examples:
An even flow of traffic causes the same annoyance as when vehicles are clustered, but an even flow is more damaging to mental performance than clustered traffic.
A large difference between equivalent level and Lmax is more annoying than a small difference.
Trams should receive a 3-dB “bonus” over buses.
Different noises from a rail yard at equal equivalent levels may have a subjective difference of as much as 5 dB.
Recommendations for Future Research in Europe
Research for a Quieter Europe in 2020, a report produced under the auspices of the CALM Network (2007), provides a strategy for future noise research in the EU. The report includes an excellent review of EU activities related to noise and covers a wide variety of future needs, including noise emissions from various sources and the need for perception-based research into the effects of noise. There is one short section on metrics (indicators).
European Versus Japanese Results on Transportation Noise
A recent Japanese study by Yano et al. (2007) compared the effects of transportation noise in Japan with the EU
results. The effects of road traffic noise are similar, but the effects of railway noise were quite different (see Figure 3-3). The authors suggest that the differences may be attributable to the proximity of Japanese homes to railroad tracks (where they are subject to vibration as well as noise). Differences in the construction of homes may also be a factor.
Japanese data for aircraft noise are based on one dataset of 410 responses around Kumamoto, a small airport, and may not be representative of noise around Japanese airports in general (Yano et al., 2007). There was also an active anti-noise group near Kumamoto airport. However, considering the scatter from study to study (e.g., Yano et al., 2007), the results of the Kumamoto study may be representative.
SUMMARY FINDINGS AND RECOMMENDATIONS
Established and New Environmental Noise Metrics
Use of the DNL metric has helped policy makers, road planners, airport managers, the public, and others understand potential noise impacts on communities and has helped guide noise mitigation efforts around airports, roadways, and rail systems. However, DNL has both strengths and weaknesses as a measure of noise.
The strengths of DNL are that it has become familiar over time, its calculation has been standardized, through experience it has become well understood, and it is now embedded in software used for planning. DNL has made it possible to communicate evaluations of noise to the public to provide people with a better understanding of how noise policy decisions are made and how changes in transportation systems, or choosing to live near an airport or a busy highway, might affect them. DNL has also been a mechanism by which people could be protected and systematically helped to address problems with environmental noise exposure fairly and equitably.
DNL also has some drawbacks. First, there is a great deal of variability from study to study in the percentage of the population believed to be “highly annoyed” as a function of DNL, which predicts only part of a community’s response to noise. Efforts to develop metrics that can provide a more definitive assessment of community impact are still a topic for research and policy debate.
Many limitations of a DNL-type metric based on the average A-weighted sound pressure level used to assess environmental noise have been noted:
DNL is insensitive to the impact of very loud, isolated events.
Fewer loud events can have the same DNL as many quieter events; thus, the impacts of very different soundscapes are described as equal.
DNL is insensitive to the time when an event occurs (e.g., noise early in the night causes different sleep disturbance than noise early in the morning).
The only strong argument for using night and evening weightings in DNL is based on the fact that average nighttime ambient levels are lower than those during the day.
Other metrics such as speech interference level and nighttime levels provide a better measure of annoyance with speech interference and conscious awakenings.
DNL is an outdoor noise measure that may not reflect differences between outdoor sounds and the same sounds heard indoors.
A-weighting does not reflect the results of research studies in psychoacoustics over the past 40 years.
DNL does not take into account other sound characteristics (e.g., tonality and rate of loudness onset) that can influence annoyance and sleep disturbance levels.
Although DNL has limitations, it has served as the major environmental noise metric since the early 1970s. Despite the variability in community response, it is clear that the percentage of the population highly annoyed for a DNL of 65 dB is considerably greater than the corresponding percentage for a DNL of 55 dB. This supports the findings of EPA in the 1970s (EPA, 1974) that a DNL of 55 dB is the level necessary to protect the public health and welfare with an adequate margin of safety.
When new metrics are developed and values selected as a matter of public policy, the goal should be to protect a larger fraction of the population than is protected under the value now widely used—the DNL = 65 dB criterion. Many steps would have to be taken before a different metric (or set of metrics) could be recommended to policymakers. Changing to another metric would entail significant effort and cost
(e.g., in conducting surveys and development of databases) and would be of limited value unless the new metric offers significant benefits over DNL, most importantly in providing a more transparent and definitive connection between noise level and annoyance or other effects on people’s lives. Unfortunately, because of a lack of “real-world” data to test the performance of metrics, it is difficult to establish their advantages and disadvantages. The situation with respect to DNL has been recognized by the FAA, and two meetings have been held—one in August 2009 and one in December 2009—to discuss a “roadmap” to improve the situation regarding noise metrics.
A set of metrics, rather than a single metric, to describe different types of outcomes of environmental noise (e.g., number of interruptions of speech, learning impairment in schools, number of additional awakenings) would provide a multidimensional picture of noise impact and may be the best approach to informing the public. Supplementary metrics could make possible predictions of noise from transportation in sufficient detail to enable the development of noise maps.
When communicating with the public, it might be useful to translate metric values into words (e.g., categories such as no observed reaction, sporadic complaints, widespread complaints, threats of community action, vigorous community action) that can be more easily understood than DNL and other numerical metrics.
The ability to predict direct health effects of noise (e.g., hypertension, speech interference, cognitive impairment, sleep disturbance) and the relationship between these effects and annoyance requires further study in order to develop new metrics that account for health effects.
Recommendation 3-1: The federal government (e.g., agencies of the U.S. Department of Transportation with responsibilities related to noise and the U.S. Department of Housing and Urban Development) should adopt as a goal the 1974 recommendation of the Environmental Protection Agency (EPA, 1974) to limit the day-night average sound level (DNL) to 55 decibels (dB) to protect the public health and welfare. Currently, DNL (DENL in Europe), the accepted metric for characterizing the impact of community noise, shows that a large proportion of the population is highly annoyed at a DNL of 65 dB or higher.
Recommendation 3-2: Relevant agencies of the federal government (e.g., agencies of the U.S. Department of Transportation with responsibilities related to noise, the Environmental Protection Agency, and the U.S. Department of Housing and Urban Development) should fund the development of environmental noise metrics that are more transparent and more reflective of the impact of noise on an affected population than DNL. This will require improved tools for predicting community sound pressure time histories and the development of metrics that accurately reflect the sounds people hear. A more holistic model of annoyance is also needed that incorporates situational variables that can be used to generate predictions for overall response, as well as responses of vulnerable populations (e.g., elderly people, sick people, children, and noise-sensitive individuals). International cooperation in this effort will facilitate the development of national and international standards for calculating metrics and should include open-source code to facilitate broad implementation of the metrics. Certain measures should be taken to facilitate this development:
The international noise control engineering community should develop an open, collaborative data-sharing environment in which researchers can deposit and access data from community noise surveys (e.g., data from surveys of acoustic, environmental, community, and transportation systems to support comparisons of metrics and predictions by models).
Policy agencies should conduct extensive surveys around at least six U.S. airports to generate high-quality data to populate the database. These surveys should serve as models of good survey practices, including data recording and archiving to ensure that they are useful for future studies.
Noise Metrics for Quiet Environments
The impact of man-made noise in national parks and other quiet environments is another parameter that is not well modeled by the metrics used to assess the impact of noise around airports or roads. Detection of the sound and distinguishing between man-made and natural sounds are important because human reactions to man-made and natural sounds differ. If one goal of the national parks is to preserve places of natural beauty, then the natural soundscape of a park, which is an aspect of its beauty, should also be preserved.
In addition, predicting the impact of noise on wildlife in national parks may require a different kind of metric that reflects animals’ hearing systems. Preserving wildlife is essential to preserving the ecostructure of a park. But wildlife preservation will require that animals’ hearing also be preserved and protected, because many animals depend on their hearing to hunt and to detect potential predators. The U.S. Department of the Interior should fund the development of metrics to support noise management decisions in national parks and other quiet environments.
ANASE (Attitudes to Noise from Aviation Sources in England). 2007. Attitudes to Noise from Aviation Sources in England (ANASE). Final Report. Prepared for United Kingdom Department for Transport, by P. Le Masurier, J. Bates, J. Taylor, I. Flindell, D. Humpheson, C. Pownall, and A. Woolley. Available online at http://www.dft.gov.uk/pgr/aviation/environmentalissues/Anase.
ANSI (American National Standards Institute). 2002. American National Standard: Acoustical Performance Criteria, Design Requirements, and Guidelines for Schools. ANSI/ASA S12.60-2002 (R2009). New York: ANSI.
ANSI. 2005. American National Standard: Quantities and Procedures for Description and Measurement of Environmental Sound—Part 4: Noise Assessment and Prediction of Long-Term Community Response. ANSI S12.9-2005/Part 4. New York: ANSI.
ANSI. 2007. American National Standard for the Computation of Loudness and Steady Sounds. ANSI S3.4-2007. New York: ANSI.
ANSI. 2008. Methods for the Estimation of Awakenings Associated with Outdoor Noise Events Heard in Homes. American National Acoustical Society of America Standard S12.19-2008, Part 6. Melville, NY: Acoustical Society of America.
Beranek, L.L. 2008. Riding the Waves: A Life in Sound, Science, and Industry. Cambridge, MA: MIT Press.
Beranek, L.L., K.D. Kryter, and L.M. Miller. 1959. Reaction of people to exterior aircraft noise. Noise Control 5:23–60.
CALM Network. 2007. Research for a Quieter Europe in 2020. An Updated Strategy Paper of the CALM II Network. Available online at http://www.calm-network.com/SP_2020_final07.pdf.
Eagan, M.E. 2007. Supplemental metrics to communication aircraft noise effects. Transportation Research Record: Journal of the Transportation Research Board 2011:175–183.
EC (European Commission). 1996. Green Paper on Future Noise Policy. COM(96) 540. Available online at http://europa.eu/documents/comm/green_papers/com96_540/summary_en.htm. See also http://ec.europa.eu/environment/noise/greenpap.htm.
EC. 2000. Position Paper on EU Noise Indicators. A Report prepared for the European Commission, Directorate-General Environment. Available online at http://ec.europa.eu/environment/noise/pdf/noiseindicators.pdf.
EC. 2002a. Directive 2002/49/EC of the European Parliament and of the Council of 25 June, 2002 Relative to the Assessment and Management of Environmental Noise. Official Journal of the European Communities, L189/12. Available online at http://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=OJ:L:2002:189:0012:0025:EN:PDF.
EC. 2002b. EU’s Future Noise Policy, WG2—Dose/Effect. Position Paper on Dose-Response Relationships Between Transportation Noise and Annoyance. Available online at http://ec.europa.eu/environment/noise/pdf/noise_expert_network.pdf.
EC. 2004. Position Paper on Dose-Noise Effect Relationships for Night Time Noise. European Commission Working Group on Health and Socio-Economic Effects. Available online at http://ec.europa.eu/environment/noise/pdf/positionpaper.pdf.
EPA (U.S. Environmental Protection Agency). 1974. Information on Levels of Environmental Noise Requisite to Protect Public Health and Welfare with an Adequate Margin of Safety. Document 550/9-74-004. Available online at http://www.nonoise.org/library/levels74/levels74.htm.
FICAN (Federal Interagency Committee on Aviation Noise). 2002. The Use of Supplemental Noise Metrics in Aircraft Noise Analyses. Available online at http://www.wylelabs.com/content/global/documents/Metrics.pdf.
FICON (Federal Interagency Committee on Noise). 1992. Federal Agency Review of Selected Airport Noise Analysis Issues. Available online at http://www.wylelabs.com/services/arc/documentlibrary/federalandlocalguidanceonoise/ficon.html.
Fidell, S. 2002. Reliable Prediction of Community Response to Noise: Why You Can’t Get There from Here. Proceedings of INTER-NOISE 02, 2002 International Congress and Exposition on Noise Control Engineering, Dearborn, MI, August 19–12. Available online at http://www.bookmasters.com/marktplc/00726.htm.
Fidell, S., and L. Silvati. 2004. Parsimonious alternative to regression analysis for characterizing prevalence rates of aircraft noise annoyance. Noise Control Engineering Journal 52(2):56–68.
Fidell, S., D.S. Barber, and T.J. Schultz. 1991. Updating a dose-effect relationship for the prevalence of annoyance due to general transportation noise. Journal of the Acoustical Society of America 89:221–233.
Fidell, S., D.S. Barber, and T.J. Schultz, 1989. Updating a Dosage-Effect Relationship for the Prevalence of Annoyance Due to General Transportation Noise. (Technical Report HSD-TR-89-009), Human System Division (HSD/YAH), U.S. Air Force Systems Command, Brooks AFB, TX.
Fields, J.M. 1993. Effect of personal and situational variables on noise annoyance in residential areas. Journal of the Acoustical Society of America 93(5):2753–2763.
Finegold, L.S., and M.S. Finegold. 2002. Development of Exposure-Response Relationships Between Transportation Noise and Community Annoyance. Proceedings of the Japan Net-Symposium on “Annoyance, Stress, and Health Effects of Environmental Noise,” November–December. Available online at http://www.netsympo.com/2002/finegold/develop.pdf.
Finegold, L.S., C.S. Harris, and H.E. von Gierke. 1994. Community annoyance and sleep disturbance: Updated criteria for assessing the impacts of general transportation noise on people. Noise Control Engineering Journal 42:25–30.
Finegold, L.S., C. Oliva, and J. Lambert. 2008. Progress on Development of Noise Policies from 2003–2008. Proceedings of the 9th Congress on the Biological Effects of Noise: Noise as a Public Health Hazard. Foxwoods, Connecticut. Available online at http://www.bookmasters.com/marktplc/00726.htm.
Fletcher, H., and W.A. Munson. 1933. Loudness, its definition, measurement, and calculation. Journal of the Acoustical Society of America 5:82–108.
Flindell, I.H., and P.J.M. Stallen. 1999. Non-acoustical factors in environmental noise. Noise and Health 3:11–16.
Gjestland, T. 2008. Research on Community Response to Noise—in the Last Five Years. Proceedings of the 9th Congress on the Biological Effects of Noise; Noise as a Public Health Hazard, Connecticut. Available online at http://www.bookmasters.com/marktplc/00726.htm.
ISO (International Organization for Standardization). 1975. Acoustics—Methods for Calculating Loudness Level. Geneva, Switzerland: ISO. Available online at http://www.iso.org/iso/iso_catalogue/catalog_tc/catalogue_detail.htm?csnumber=4603.
ISO. 2003. Acoustics—Description, Measurement, and Assessment of Environmental Noise—Part 1: Basic Quantities and Assessment Procedures. Geneva, Switzerland: ISO. Available online at http://www.iso.org/iso/iso_catalogue/catalog_tc/catalogue_detail.htm?csnumber=28633.
Kryter, K.D. 1960. The measurement and meaning of perceived noise level. Noise Control 6(5):12–27.
Kryter, K.D. 2009. Acoustical model and theory for predicting effects of environmental noise on people. Journal of the Acoustical Society of America 125(6):3707–3721.
Kryter, K.D., and K.S. Pearsons. 1962. Judgment tests of the sound from piston, turbojet, and turbofan aircraft. Sound 1:24–31.
Kryter, K.D., and K.S. Pearsons. 1963. Some effects of spectral content and duration on perceived noise level. Journal of the Acoustical Society of America 35:866–883.
Miedema, H.M.E., and C.G. Oudshoorn. 2001. Annoyance from transportation noise: Relationships with exposure metrics DNL or DENL and their confidence intervals. Environmental Health Perspectives 109:409–416.
Miedema, H.M.E., and H. Vos. 1998. Exposure-response relationships for transportation noise. Journal of the Acoustical Society of America 104(6):3432–3445.
Miedema, H.M.E., and H. Vos. 1999. Demographic and attitudinal factors that modify annoyance from transportation noise. Journal of the Acoustical Society of America 105(6):336–344.
Schomer, P.D. 2002. On normalizing DNL to provide better correlation with response. Sound and Vibration 36(12):14–25.
Schomer, P.D. 2005. Criteria for assessment of noise annoyance. Noise Control Engineering Journal 53(4):132–144.
Schultz, T.J. 1978. Synthesis of social surveys on noise annoyance. Journal of the Acoustical Society of America 64:377–405.
Stevens, K.N., W.A. Rosenblith, and R.H. Bolt. 1955. A community’s reaction to noise: Can it be forecast? Noise Control 1:63–71.
Transport Canada. 2005. Aviation-Land Use in the Vicinity of Airports. Aircraft Noise, 8th ed. Available online at http://www.tc.gc.ca/CivilAviation/publications/TP1247/Part4/Part4-2.htm.
Von Gierke, H.E. 1975. Noise: how much is too much? Noise Control Engineering Journal 5(1):24–34.
WHO (World Health Organization). 2007. Night Noise Guidelines (NNGL) for Europe. Published by the World Health Organization in cooperation with the European Union. Available online at http://ec.europa.eu/health/ph_projects/2003/action3/docs/2003_08_frep_en.pdf.
Yano, T., T. Sato, and T. Morihara. 2007. Dose-Response Relationships for Road Traffic, Railway and Aircraft Noises in Kyushu and Hokkaido, Japan. Proceedings of INTER-NOISE 07, The 2007 International Congress and Exposition on Noise Control Engineering, Istanbul, Turkey, August 28–31. Available online at http://scitation.aip.org/journals/doc/INCEDL-home/cp/.