Assessing Community Annoyance of Helicopter Noise (2017)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

5 1.1 Introduction The literature review performed by the research team initially identifies prior design and analysis approaches used for research on community response to aircraft noise. Review of these prior design and anaylsis approaches then leads to a discussion of hypotheses that merit consider- ation in field studies.1 The review then identifies annoyance as the primary noise effect of concern and distinguishes between the direct annoyance of airborne noise and the indirect annoyance of secondary emissions (vibration and rattling sounds) that may be induced by helicopter acoustic emissions. A recent increase in concern with helicopter noise complaints is then discussed. The next topics addressed are the potential influences of nonacoustic factors in community response to helicopters and the usefulness of laboratory and field findings about helicopter annoyance. The review concludes with a summary of prior findings. 1.2 Understanding of Helicopter Noise Versus Fixed-Wing Aircraft Noise Community reaction to helicopter noise has been less studied and less well understood than community reaction to fixed-wing aircraft noise for a variety of reasons. Most obviously, exposure to helicopter noise remains a more geographically limited problem than exposure to fixed-wing aircraft noise, and affects far fewer people. For example, out of a total of 232,567 active aircraft in the domestic U.S. fleet of commercial and general aviation aircraft, only 11,245 are helicopters (FAA 2011). Despite the smaller numbers of people affected by exposure to helicopter noise than by exposure to noise from fixed-wing aircraft, helicopter noise can nonetheless be distinctive and highly annoying. As described in Appendix A, noise emissions of helicopters are more complex, variable, and unpredictable than those of fixed-wing aircraft. (The appendix provides a brief tutorial on the sources and characteristics of helicopter noise in various flight regimes.) Helicopter noise emissions vary not only with flight regime, orientation with respect to the flight path, and speed, but also with manner of operation. A fixed-wing aircraft flyover characteristically produces a simple and familiar “haystack” temporal pattern. Fixed-wing aircraft noise increases more or less monotonically as an aircraft flies toward an observer, reaches a peak at about the time that the aircraft is directly overhead, and then monotonically decreases as it flies away from the observer. In areas within a few miles of runway ends, high-speed, fixed-wing aircraft usually follow pre- dictable paths and distribute their noise emissions symmetrically with respect to the flight path. In contrast, the spatial distribution of helicopter noise is more complex than that of fixed- wing aircraft because of source directivity, dependence of emissions on flight regime, and the C H A P T E R 1 Literature Review

6 Assessing Community Annoyance of Helicopter Noise operational flexibility of rotary-wing flight. High-speed impulsive (HSI) helicopter noise is concentrated in the plane of the rotor disk and in the direction of forward flight. Blade-vortex interaction (BVI) noise (“blade slap”) is also impulsive sounding and is concentrated forward and downward, along the helicopter’s flight path. Broadband emissions of rotary-wing aircraft are typically greater on the side of the aircraft with the counter-torque rotor. Helicopters may approach and depart a landing pad at low speeds, and to and from more than one direction. The flexibility of rotary-wing flight also means that the time pattern of helicopter noise intrusions is less predictable than that of fixed-wing aircraft. Helicopters typically operate at lower altitudes than fixed-wing aircraft and can orbit a location on the ground or hover in place for prolonged periods. These flight characteristics can render individual helicopter operations more audible, for longer periods of time, than fixed-wing aircraft overflights in urban ambient noise environ- ments. Further, the low-frequency noise emissions of helicopters can excite more indoor rattle and vibration in residences than fixed-wing aircraft in flight at greater altitudes. For all of these reasons, helicopter noise is often thought to be more annoying on a per-event basis than fixed-wing aircraft noise of comparable sound level. It is also commonly believed that the repetitive impulsive nature of helicopter noise is its most annoying characteristic. Neither of these interpretations is necessarily correct, nor the complete story. In particular, it remains unclear whether the supposed “excess” annoyance of helicopter noise (vis-à-vis that of fixed- wing aircraft noise) is acoustic or nonacoustic in origin. 1.3 Noise Effects of Concern 1.3.1 Annoyance The Federal Interagency Committee on Noise (FICON) (1992) considers annoyance an attitude (that is, a covert mental process) as its preferred general indication of adverse air- craft noise impacts. In this context, annoyance is gauged by the self-reporting of opinions in community-wide social surveys, in response to questions such as “While you’ve been at home over the last (day/week/year), have you been not at all, slightly, moderately, very, or extremely annoyed by aircraft noise?” Schultz (1978) and his successors have produced several quantita- tive dosage-response relationships to predict the prevalence of a consequential degree of aircraft noise-induced annoyance attributable to cumulative noise exposure. Nearly all of the field studies from which such relationships have been inferred have dealt with annoyance produced by fixed- rather than rotary-wing aircraft operations. Most dosage-response relationships attempt to predict the prevalence of aircraft noise-induced annoyance in communities from a single independent variable—cumulative noise exposure—as estimated either by direct measurement or by noise modeling. Such relationships account for less than half of the variance in the association between noise exposure and annoyance. Only in recent years has a practical, quantitative method emerged for incorporating an additional vari- able into predictions of annoyance prevalence rates. As described in Appendix C, the second predictor variable is the sum total of community-specific, nonacoustic influences on annoyance.2 Even if it is assumed that the annoyance of exposure to noise produced by helicopters is best understood in entirely acoustic terms, a further question remains: is that annoyance produced solely by the airborne acoustic energy that helicopters produce or by secondary emissions (rattling noises and vibration) induced by helicopter noise in residences. 1.3.2 Direct Annoyance of Airborne Noise Created by Helicopters Figure 1-1 compares three dosage-response relationships between cumulative aircraft noise exposure and the prevalence of aircraft noise-induced annoyance in average communities.

Literature Review 7 The solid black line, the community tolerance level (CTL) relationship, is the one recom- mended in the 2016 revision of International Standards Organization (ISO) Standard 1996-1.3 (Appendix C provides additional detail about the methods described in the latest revision of the ISO Standard.) If helicopter noise is more annoying, decibel-for-decibel, than fixed-wing aircraft noise, the CTL curve seen in Figure 1-1 (developed for fixed-wing aircraft) will be shifted toward the left side of the graph. Figure 1-2 illustrates a family of dosage-response relationships corresponding to increases in the annoyance of helicopter noise exposure by amounts ranging from 3 to 10 dB. For example, if helicopter noise proves to be 3 dB more annoying than fixed-wing aircraft noise, analyses of survey data may be expected to produce a dosage-response relationship similar to the dashed curve to the left of the one seen in Figure 1-1. Note that the curves in Figure 1-2 differ both in positions on the abscissa, and in their slopes, for reasons discussed in Appendix C. The shapes of the curves are identical no matter where they are horizontally. However, the horizontal position affects the slope of a given curve at a particular dose (i.e., DNL value), and hence the rate at which annoyance grows with increasing dose at that level. 1.3.3 Annoyance Due to Secondary Emissions The primary structural resonance in conventional wood frame construction for single-family detached dwellings is typically in the 10–25 Hz frequency region, the same frequency region as the fundamental (one per revolution) frequency of the main rotor system of many helicopters. This means that helicopter operations can easily induce noticeable vibration in homes near helipads and flight paths. Even modest levels of structural vibration, which might escape direct notice, can cause lightweight or suspended architectural elements (windows, doors, bric-a-brac on shelves, pictures on walls, crockery in cupboards, HVAC ducts, and other household paraphernalia) to Figure 1-1. Comparison of revised ISO Standard 1996-1 dosage-response curves with earlier FICON curve.

8 Assessing Community Annoyance of Helicopter Noise rattle audibly. Such rattling noises can be annoying in their own right, whether or not accompanied by noticeable vibration, or by audible helicopter noise. Figure 1-3, adapted from Fidell et al. (2002a), shows a relationship between the prevalence of annoyance due to aircraft noise-induced rattle and a single-event measure of low-frequency noise level. The measure, known as low frequency sound level (LFSL), is the sum of the sound exposure levels in the six one-third octave bands between 25 and 80 Hz. 35 60 85 110 Day Night Average Noise Level (DNL), dB Grand Average Fixed Wing Aircraft CTL -3 dB -6 dB -10 dB Decreasing Tolerance 0 10 20 30 40 50 60 70 80 90 100 Pe rc en t H ig hl y An no ye d Increasing Tolerance Figure 1-2. Family of hypothetical dosage-response curves for differing levels of community sensitivity. Figure 1-3. Relationship between LFSL and the prevalence of high annoyance with rattle. (Note: the % high annoyance due to rattle at MSP appears at 87.5 dB on the graph.)

Literature Review 9 1.3.4 Complaints In July of 2013, the Washington, D.C., Court of Appeals found that helicopter noise could adversely affect a residential population at an A-weighted cumulative noise level more than 20 dB lower than FAA’s customary criterion of “significant” noise impact (Ldn = 65 dB). The court ruled in Helicopter Association International, Inc. v. Federal Aviation Administration, Case No. 12-1335 (C.A. D.C., Jul. 12, 2013) that the FAA was justified in mandating compulsory com- pliance with an offshore flight route for helicopters,4 even when the noise created by helicopter operations did not exceed Ldn = 45 dB at affected residences. The ruling seems to rely solely on a high number of noise complaints rather than any specific acoustic measure. Complaints, a behav- ior, are not the same quantity as annoyance, an attitude. A recent study has made some progress in suggesting a potential relation between the behavior and the attitude (Fidell et al. 2012). Note that the referenced study made a clear distinction between numbers of complaints, number of complainers and segregating complainers by numbers of complaints. Except for the most prolific complainers, a common pattern was observed leading to the conclusion that tracking the number of non-prolific complainers may provide an indication of community attitudes about noise. This is a topic about which more, and very possibly quite productive, research could be done. The court’s ruling implies an A-weighted difference on the order of 20 dB between the annoy- ance of helicopter and fixed-wing aircraft noise. Conventional analyses, such as those identified by ISO 1996 and discussed in Appendix C, however, “penalize” helicopter noise by less than 10 dB in an attempt to equalize predictions of the annoyance of rotary- and fixed-wing noise. The order of magnitude difference between the findings of the Court of Appeals and current (acoustically driven) noise impact evaluation methods suggests that metrics sensitive to acoustic factors alone may not be fully capable of predicting community response to helicopter noise. 1.4 Noise Metrics Useful for Quantifying Helicopter Noise Two frequency weighting networks and families of noise metrics are commonly employed in the U.S. to express sound levels of both fixed- and rotary-wing aircraft. For aircraft noise certification purposes, the FAA has required frequency weighting, called the tone-corrected per- ceived noise level, abbreviated PNL(T), developed in the 1950s. For predicting and assessing environmental impacts of aircraft noise exposure, the FAA endorses the A-weighting network, developed in the 1930s.5 Each metric supports a family of single-event and cumulative exposure metrics to deal with exposure that varies from instantaneous through annual time frames.6 Concern about noise metrics appropriate for predicting the annoyance of exposure to rotary-wing aircraft noise has peaked several times since the 1950s. As discussed in Appen- dix B, a 1982 literature review by Molino (1982) compares the findings of 34 earlier analyses of the annoyance of helicopter noise, the earliest of which date to the 1960s (cf. Crosse et al. 1960, Niese 1961, Robinson et al. 1961, and Pearsons 1967). The findings of these early studies are neither consistent nor definitive. These and other studies (e.g., Powell, 1981) do not fully support Molino’s conclusion that there is “no need to measure helicopter noise any differently from other aircraft noise.” The common belief that rotary-wing aircraft noise causes more annoyance on a decibel-for- decibel basis than fixed-wing aircraft noise has led to the practice of imposing decibel-denominated “penalties” on A-weighted (but not PNL-weighted) measures of helicopter noise for purposes of assessing environmental impacts of helicopter noise. This may be an expedient way of accom- modating the supposed excess annoyance of helicopter noise, but is not necessarily the most systematic or defensible way.

10 Assessing Community Annoyance of Helicopter Noise The tactic of assigning penalties treats the assumed excess of annoyance of helicopter noise as a simple problem of measurement, while ignoring the underlying causes of the supposed excess annoyance. Since the evidence supporting the assumption of excess annoyance is not definitive, the issue may not simply be one of physical measurement, however. The supposed excess could be attributable to operational factors (the characteristic shorter slant ranges and relatively longer duration of helicopter operations vis-à-vis fixed-wing aircraft operations) rather than inher- ent differences in noise-induced annoyance. The supposed excess could also be attributable to entirely nonacoustic factors. Although a good deal has been learned since Molino’s 1982 review about the mechanisms that generate rotary-wing aircraft noise in different flight regimes, it is only recently that systematic means have become available to focus more closely on potential nonacoustic factors that influence annoyance judgments (Appendix C provides greater detail about these means). To the extent that excess annoyance of helicopter noise is attributable to the annoyance of rattle and vibration (to which A-weighted noise metrics are insensitive), A-weighted noise met- rics are unlikely to adequately predict the overall annoyance of helicopter overflights of residen- tial populations, if the helicopter noise has strong low-frequency components as is the case for heavy military aircraft. 1.5 Nonacoustic Contributions to Community Reaction to Helicopter Noise FAA (2004) summarized many operational, situational, and other nonacoustic factors that contribute to adverse community response to helicopter noise. These include low flight alti- tudes; long hover durations; times, numbers, and frequencies of operations; fear of crashes; and attitudes of misfeasance and malfeasance. Most of these factors similarly affect the annoyance of fixed-wing aircraft, but to lesser degrees. Perceptions of the necessity for flight operations can differ greatly for a range of rotary-wing missions. The necessity of medical evacuation, search and rescue, law enforcement, firefighting, and some heavy lift construction missions is widely acknowledged. The necessity for other rotary-wing flight operations is less apparent. For example, large fixed-wing aircraft are self-evidently the most efficient mode of public trans- portation for regularly scheduled, long-haul carriage of hundreds of passengers per flight. As such, their necessity is generally taken for granted. In contrast, short-haul private transportation of individuals by helicopter is widely viewed as a luxurious choice (or “a rich man’s toy,” in the words of FAA’s 2004 Report to Congress) rather than a practical necessity. Similarly, the limited ground visibility from fixed-wing airplanes and high flight speeds and altitudes pose little threat to domestic privacy. Helicopters hovering over residences are a different matter. Few would consider long duration hovering to permit paparazzi to photograph private events to be truly necessary. Likewise, fixed-wing aircraft in the vicinity of airports necessarily approach and depart run- ways on flight paths corresponding to runway alignments. The motivation and necessity for non- emergency (e.g., air tour), small rotorcraft operations are not as apparent. Given their flexibility of flight, why must helicopters approach a particular house so closely on their way to and from landing pads? Why must multiple news gathering helicopters orbit the same traffic accident? 1.6 Laboratory Versus Field Studies of Helicopter Annoyance Studies of the annoyance of rotary-wing aircraft noise have been conducted under both labo- ratory and field conditions. Laboratory studies offer greater precision of control over listening conditions than field studies, but lack the residential context of field studies. It is also difficult to

Literature Review 11 accurately reproduce recorded or helicopter-like synthetic sounds under laboratory conditions while also preserving crest factor (ratio of peak value to average value of sound wave—important with impulsive noise), phase relationships (whether two sound waves are synchronized or shifted in time), low frequency, and other dynamics of rotorcraft noise emissions. On the other hand, while field studies provide the appropriate residential context for annoyance judgments, they lack the precision of control over acoustic conditions of laboratory studies. It follows that questions about potential nonacoustic influences on the “excess” annoyance of helicopter noise are not readily answered in laboratory studies and that questions about the detailed acoustic origin of excess annoyance are not readily answered in field settings. 1.7 Summary of Findings of Literature Review This literature review was conducted to identify pragmatically useful—that is, testable and relevant—hypotheses about the origins of annoyance with exposure to helicopter noise as a preliminary aid to the design of subsequent field research. The current review, as well as prior literature reviews such as those conducted by Molino (1982), Ollerhead (1985), and FAA (2004) document research undertaken in the last half-century to quantify and predict the individual and community annoyance of rotary-wing aircraft noise. Whether conducted under laboratory or field conditions, much of this research was intended, directly or indirectly, to inform decisions about aircraft noise regulatory policy. Understandably, the early research sought out low-hanging fruit: “magic bullet” noise metrics; non-systematic (ad hoc, regression-based) dosage-response relationships; evidence that demographic and socio- economic factors could account for non-trivial amounts of variance in a predictively useful manner, and so on. The reviewed literature provided little systematic, rigorous, or theory-based understanding of the annoyance of helicopter noise. Given what has been learned over the decades, some of the earlier exploratory research goals, hypotheses tested, study designs, and analysis approaches are not as relevant or appropriate today as they once may have been. For example, individual-level analyses intended to identify covariates that might arguably improve prediction of helicopter annoyance prevalence rates are now outdated. Individual differences such as demographic (sex, age, gender, nationality, etc.) account for relatively little variance in the relationship between noise exposure and annoyance, and are of little practical regulatory utility. Attitudinal differences (fear, suspected malfeasance, sense of necessity, etc.) as measured on a community-wide basis have significant effects on annoyance. Systematic means have recently become available for efficiently taking into consid- eration the net effects, rather than individual influences, of all of the nonacoustic factors that may affect the annoyance of helicopter noise exposure. The findings of individual studies on the annoyance of helicopter noise disagree about as often as they agree. The main point of agreement in the technical literature is that helicopter noise is much more variable and complex than fixed-wing aircraft noise. This variability and complex- ity make it more difficult to accurately and credibly model helicopter noise exposure (other than under idealized conditions7), particularly in the vicinity of helipads. It follows, in turn, that predictions of the prevalence of annoyance of exposure to helicopter noise are likely to be more uncertain than predictions of the annoyance of exposure to fixed-wing aircraft noise. A main point of disagreement is the degree to which main rotor impulsive noise controls the annoyance of helicopter noise. Many believe that impulsiveness “corrections” are appropriate for predicting the annoyance of exposure to helicopter noise; others believe that conventional A-weighted noise measurements suffice for predicting the annoyance of helicopter noise.

12 Assessing Community Annoyance of Helicopter Noise Table 1-1 summarizes the laboratory (controlled listening) and field (social survey) evidence for and against hypotheses about the origins of the supposed excess annoyance of helicopter noise. (Annotation is provided in Appendix B for only some of the cited sources.) The empty cells in Table 1-1 reflect the incomplete nature of understanding of the origins of annoyance with helicopter noise. Some of the implications of the findings of this literature review for the design of field studies include the following: • Neighborhood opinions about the annoyance of helicopter noise and fixed-wing aircraft noise exposure are likely to differ for nonacoustic reasons. Unless analytic means are employed to account for such community-specific differences, it may not be possible to reliably identify differences in opinions about fixed- and rotary-wing annoyance per se. HYPOTHESIS EVIDENCE OR ASSERTION CONSISTENT WITH HYPOTHESIS MARGINAL OR INCONCLUSIVE EVIDENCE OR ASSERTION EVIDENCE INCONSISTENT WITH HYPOTHESIS Decibel for decibel, rotary- wing aircraft noise is more annoying than fixed-wing aircraft noise No reliable, large-scale comparisons reported in peer-reviewed field studies More (2011); several other controlled-listening tests, which may not have controlled for confounding factors; tone- corrected effective perceived noise level [EPNL(T)] is a less consistent predictor of annoyance for rotary- than fixed-wing aircraft noise (Ollerhead 1982) Ollerhead, 1982 (2 dB average effect in effective perceived noise level, in direction opposite to predicted direction) Main rotor impulsive noise controls the annoyance of helicopter noise (and hence requires an impulsive noise “correction” to A-weighted measurements) Sternfeld and Doyle (1978); Man-Acoustics & Noise, Inc. (1976); Lawton (1976); Wright and Damongeot (1977); Galanter et al. (1977); Klump and Schmidt (1978) Fields and Powell (1987) (weak evidence at best); More (2011); Schomer and Wagner (1996); Magliozzi et al. (1975); Munch and King (1974) Patterson et al. 1977; Powell 1981; Ollerhead 1982—also ICAO, 1981 [no impulse correction needed for EPNL(T); effect of impulsiveness is confounded with level and duration]; Passchier-Vermeer, 1994; Ohshima and Yamada, 1993; Gjestland, 1994; Bisio et al., 1999 A-weighted noise measurements are inadequate for predicting the annoyance of rotary-wing aircraft noise Patterson et al. (1977); Schomer et al. (1991); Schomer and Neathammer (1987); Sternfeld et al. (1995); Edwards, (2002); Ollerhead (1982) More (2011) Molino, 1982 The annoyance of helicopter noise is strongly influenced by nonacoustic factors Leverton (2014); Ollerhead (1982); FAA (2004); Atkins et al. (1983) Situational and operational factors account for much of the annoyance of helicopter Ollerhead and Jones (1994); FAA (2004) Anecdotal evidence from noise popular press Cumulative noise metrics usefully predict the annoyance of exposure to helicopter noise Fields and Powell (1987) (“broad consistency”); Atkins et al. (1983) Secondary emissions (rattle) induced by helicopter noise strongly influence its annoyance Schomer and Neathammer (1987) The annoyance of helicopter noise is strongly influenced by its noticeability rather than its level per se Schomer and Wagner (1996) Annoyance is better predicted by time-integrated proximity to flight tracks than by acoustic measures Table 1-1. Evidence relevant to hypotheses about the annoyance of rotary-wing noise exposure.

Literature Review 13 • The flexibility of low-speed, rotary-wing flight lends itself to much more complex flight paths than those of fixed-wing aircraft. These complex flight paths cause the helicopter to accelerate/ decelerate along the flight path and can dramatically change blade vortex interaction (BVI) impul- sive noise level. The directivity of helicopter noise emissions further complicates noise exposure predictions based on flight tracks alone. Selecting sites with comprehensive flight track radar coverage and using sections of level flight rather than climbing and descending segments, the aircraft performance information will aid prediction, measurement, and interpretation of heli- copter noise exposure, minimizing the uncertainty of the dosage portion of the dosage-response analysis. In other words, differences of as little as 2 or 3 dB between the annoyance of rotary- and fixed-wing aircraft may be difficult to discern on the basis of social surveys undertaken in a limited number of communities. • Extensive efforts to confirm the utility of impulsive noise adjustments have yielded contradic- tory and inconclusive results. • Correlation analyses have shown that most of the noise metrics commonly used to quantify helicopter noise are so highly correlated with one another that no one metric differs mean- ingfully from others in its ability to predict the prevalence of annoyance of helicopter noise (Mestre et al. 2011). • Operational factors can also affect the annoyance of helicopter noise, but their effects may or may not be accounted for by integrated energy noise metrics. • Questions about potential nonacoustic influences on the “excess” annoyance of helicopter noise are not readily answered in laboratory studies, while questions about the detailed acous- tic origin of excess annoyance are not readily answered in field settings.