Current Biology 19, 1415–1419, August 25, 2009 ª2009 Elsevier Ltd All rights reserved DOI 10.1016/j.cub.2009.06.

052

Report
Noise Pollution Changes Avian
Communities and Species Interactions

Clinton D. Francis,1,* Catherine P. Ortega,2 To date, noise has been associated with declining bird densi-
and Alexander Cruz1 ties [10–14], prompting conservation concerns that many
1Department of Ecology and Evolutionary Biology, University species may be excluded from otherwise suitable habitat
of Colorado at Boulder, 334 UCB, Boulder, CO 80309, USA as a result of ecological sensitivities or intolerance to noise
2San Juan Institute of Natural and Cultural Resources, Fort [1–4]. Additionally, individuals that settle in noisy habitats
Lewis College, 1000 Rim Drive, Durango, CO 81301, USA may have reduced reproductive success because noise inter-
feres with detection of approaching predators [1, 3]. Despite
previous links between noise and bird declines, evidence
Summary demonstrating a direct negative influence of noise on birds
has been equivocal because previous efforts have employed
Humans have drastically changed much of the world’s methods with insufficient controls over other stimuli associ-
acoustic background with anthropogenic sounds that are ated with noise, such as the physical alteration of habitat,
markedly different in pitch and amplitude than sounds in community location at habitat edges versus interior habitat,
most natural habitats [1–4]. This novel acoustic background or visual disturbance presented by moving traffic or equipment
may be detrimental for many species, particularly birds [1]. [1, 3, 15]. These uncontrolled variables could also explain
We evaluated conservation concerns that noise limits bird observed bird declines. Additionally, these previous studies
distributions and reduces nesting success via a natural have not accounted for the negative influence of noise on the
experiment to isolate the effects of noise from confounding observer’s ability to detect birds [5]. We tested conservation
stimuli and to control for the effect of noise on observer concerns that noise results in declines in bird densities,
detection biases [5]. We show that noise alone reduces nest- community species richness, and reproductive success via
ing species richness and leads to different avian communi- a unique study design that controlled for the effects of stimuli
ties. Contrary to expectations, noise indirectly facilitates often associated with noisy habitats and detection problems
reproductive success of individuals nesting in noisy areas caused by noise. We show how noise, in the absence of other
as a result of the disruption of predator-prey interactions. influential stimuli, can have either a negative or an indirect posi-
The higher reproductive success for birds within noisy habi- tive effect on birds as a result of altered species interactions
tats may be a previously unrecognized factor contributing to produced by species-specific responses to noise (Figure 1A).
the success of urban-adapted species and the loss of birds
less tolerant of noise. Additionally, our findings suggest Nesting Community Richness and Community
that noise can have cascading consequences for communi- Composition
ties through altered species interactions. Given that noise We located and monitored nests for three breeding seasons at
pollution is becoming ubiquitous throughout much of the our study sites among the scattered natural-gas extraction
world, knowledge of species-specific responses to noise infrastructure within pinyon (Pinus edulis)-juniper (Juniperus
and the cumulative effects of these novel acoustics may osteosperma) woodlands of northwestern New Mexico. Our
be crucial to understanding and managing human-altered design provided a natural experiment that permitted isolation
landscapes. of noise as a single experimental stimulus. Treatment sites
included woodland habitat adjacent to natural gas wells with
Results and Discussion noise-producing compressors, which aid in extraction and
transportation of gas through pipelines and run 24 hours
Nearly anyone who has been near a busy roadway, an airport, a day, 365 days a year, aside from periodic maintenance and
or industrial equipment can attest to the intensity of sounds during our two-hour nest searching efforts and surveys. Wood-
produced by human activities. Many of these anthropogenic land habitats adjacent to natural gas wells that lacked noise-
sounds can be physically harmful or distracting to humans or generating compressors were used as control sites (see
wildlife and are considered noise pollution (hereafter referred Figure S1 available online). We measured noise amplitudes at
to as noise). Noise, characterized by high amplitudes and nests and throughout sites to characterize differences in the
low spectral frequencies, is typical to habitats in and around acoustic background between treatment and control sites
human-altered landscapes [1–4, 6–8]. These acoustics have (Figures S2 and S3). Given that noise can reduce avian detection
emerged swiftly on a global scale; therefore, noise presents probabilities [5] and may hamper researcher ability to locate
an evolutionarily novel source of acoustic interference for nests, we turned off all compressors (n = 9) during nest search-
many species and a potentially significant force influencing ing efforts for the first two years of the study and for half of all
the ecology and evolution of many animals [1]. Because of their treatment sites (n = 5 turned off; n = 5 left on) in the third year.
reliance on acoustic communication, birds have been viewed Contrary to previous reports of reduced densities of birds as
as especially vulnerable to the novel acoustics of noise [1–4]. a result of road noise [10–13, 16], we found no difference in
Specifically, noise may disrupt acoustic communication [1–4, community nest density between treatment and control sites
6], interfere with detection of warning signals [1, 3], and elevate (t = 20.38, df = 52, p > 0.70; Figure 1B). Despite no difference
stress levels [1, 9]. in nest density, we observed 21 species nesting at treatment
sites and 32 species nesting at control sites. Rarefaction and
nesting species richness estimates from EstimateS species
*Correspondence: [email protected] richness estimation software (http://purl.oclc.org/estimates)
Current Biology Vol 19 No 16
1416

Figure 1. Influence of Noise on Nesting Commu-

nity, Nest Density, and Nesting Richness at Treat-
ment and Control Sites
(A) Interaction web showing the pathway by
which noise negatively influences species rich-
ness of the breeding community but indirectly
facilitates avian reproduction because fewer
nests fail as a result of predation. The plus or
minus signs refer to the direction of the effect
for each interaction. Figure and table numbers
indicate which figures or tables present data sup-
porting each pathway step. Noise negatively
influences the nesting communities’ species
richness and also a major nest predator, the
western scrub-jay (Aphelocoma californica).
Because the scrub-jay has reduced occupancy
rates in noisy areas, fewer nests fail as a result
of predation. These changes brought about by
noise result in an indirect positive effect on nest
success (dashed line).
(B) There was no difference in mean nest density
between treatment and control sites. Results are
shown as mean 6 standard error of the mean
(SEM).
(C) Rarefaction and richness estimate curves
reflect the observed difference in nesting species
richness, with higher estimates of nesting
species richness at control sites (black lines
and solid symbols) than at treatment sites (gray lines and open symbols). Richness estimates are denoted as follows: solid lines = rarefaction (Sobs), dia-
monds = first-order jackknife, squares = second-order jackknife, circles = Chao 1, triangles = bootstrap. (See http://purl.oclc.org/estimates for explanations
and calculations of estimators.)

supported the observed richness difference between the two Besides the presence of noise, there were minor, albeit insig-
site types (Figure 1C). This difference was reflected by species nificant, differences between site types in terms of number of
that were unique to one of the two site types: nests of 14 pinyon trees and amount of bare ground (Table S1). These
species were found only at control sites, yet nests of 3 species two variables had no effect on nesting patterns in terms of
were found only at treatment sites. nest density or density of nesting species (see Figure S4). There
In addition to the difference in richness between treatment were no additional differences in habitat features between
and control sites, we detected clear differences in the compo- treatment and control sites that would explain selection for
sition of the nesting communities at each site type. Analysis of treatment or control sites in each species’ nest placement
similarity (ANOSIM) indicated that the nesting species compo- (Table S1).
sition at treatment and control sites was significantly dissimilar We also detected avoidance of noise in terms of nest place-
(R = 0.19, p < 0.001). This difference can be attributed to the ment within treatment sites. Gray flycatchers (Empidonax
difference in nesting species richness between noisy treatment wrightii), gray vireos (Vireo vicinior), black-throated gray
and control sites, but also to the presence of indicator species warblers (Dendroica nigrescens), and spotted towhees (Pipilo
for each site type [17]. The black-chinned hummingbird (Archi- maculatus) all nested significantly farther away from the well
lochus alexandri) and house finch (Carpodacus mexicanus) pad at treatment sites than at control sites, suggesting avoid-
were strongly associated with treatment sites (Dufrêne-Legen- ance of noise generated at treatment-site well pads (Table 1).
dre indicator species analysis: black-chinned hummingbird Nests parasitized by the brown-headed cowbird (Molothrus
indicator value (IV) = 0.66, p = 0.001; house finch IV = 0.49, ater) were also significantly farther away from the well pad at
p = 0.001). This strong association was reflected by the relative treatment sites than at control sites (Table 1). No species nested
abundance of each of these species at treatment sites: closer to the well pad at treatment sites than at control sites.
36 (92%) of 39 black-chinned hummingbird nests and 29
(94%) of 31 house finch nests were at treatment sites. These
species also accounted for a large proportion of the treat- Table 1. Species Nesting Significantly Farther from the Plot Origin
ment-site nesting community. Black-chinned hummingbird at Treatment Sites Than at Control Sites
nests accounted for 17% and house finch nests accounted Species Treatment Control t p
for 14% of all nests at treatment sites. In contrast, nests of
Gray flycatcher (n = 67) 261.6 6 16.3 206.4 6 14.7 2.51 0.015
these species were uncommon at control sites, each repre-
Gray vireo (n = 14) 275.3 6 14.5 187.2 6 33.8 2.40 0.043
senting fewer than 3% of all control-site nests. The mourning Black-throated gray 285.7 6 9.8 188.5 6 24.8 3.66 0.006
dove (Zenaida macroura) and black-headed grosbeak (Pheuc- warbler (n = 11)
ticus melanocephalus) were strongly associated with control Spotted towhee (n = 33) 267.7 6 25.3 157.4 6 22.5 3.26 0.003
sites (mourning dove IV = 0.41, p = 0.001; black-headed gros- Cowbird-parasitized 297.5 6 22.1 171.7 6 30.3 3.36 0.003
beak IV = 0.19, p = 0.025). Twenty-two (97%) of 23 mourning nests (n = 21)
Nests of all species (n = 400) 221.6 6 7.6 200.5 6 7.7 1.96 0.052
dove nests and all black-headed grosbeak nests (n = 5) were
located at control sites. Mourning dove nests represented Results are presented as mean nest distance (m) from origin 6 SEM. t values
12% and black-headed grosbeak nests represented 3% of were determined by two-tailed Welch two-sample t test. Nests of all species
were pooled, uncorrected for the number of nests for each species.
the control-site nesting community.
Avian Community Consequences of Noise Pollution
1417

Our results confirm the conservation concern that noise competition pressure or additional cues representative of
negatively affects breeding bird communities through a predation risk. Further research is needed to identify mecha-
reduction in nesting species richness, but the decline in rich- nisms responsible for settlement in noise areas and the poten-
ness was not reflected by a reduction in nest density of the tial tradeoffs associated with living in noisy conditions, such as
breeding community as a whole. Rather, we documented declines in feeding rates [24, 25].
a change in the composition of the community, with species-
specific responses to the noise disturbance that ranged from Influence of Noise on Nest Success
positive to negative but were predominantly negative. To determine whether noise negatively influences nest
Although the negative influence of noise on birds has been success, we monitored all nests until they fledged or failed.
implicated in a number of studies [10–14, 16, 18, 19], findings Nest predation was the major cause of nest failure (76% of
have primarily been restricted to studies using surveys of all failures) throughout the study area, followed by abandon-
individuals [10–14, 16, 18], often under conditions in which ment (13%) and brown-headed cowbird brood parasitism
evidence for the effect of noise on birds is weak because of (9%). In terms of apparent (observed) nest success, and
effects of uncontrolled confounding stimuli or potential detec- counter to expectations, 13% of nests with known outcomes
tion errors (but see [14] for analytical methods for dealing with at treatment sites (n = 205) failed to predation, and 32% of
differences in detection probability). Because we were able to nests with known fates (n = 174) were depredated at control
control for the effects of other influential stimuli and detection sites (c21 = 12.1, p < 0.001). We further estimated nest success
biases, our results provide especially strong evidence that in terms of daily nest survival (DNS), calculated via the logistic-
noise alone reduces habitat quality for numerous species. exposure method [26], and used likelihood-ratio tests to
Yet two species were much more common at noisy treatment assess model performance. For the nesting community as
sites than at control sites, prompting a need to identify what- a whole, the DNS model with the inclusion of a site-type
ever mechanism is causing the different responses among covariate was significantly better than a constant DNS model
species. (likelihood-ratio test, c21 = 18.3, p < 0.001). DNS was higher
In general, species-specific responses to noise remain at treatment sites (0.989, 95% confidence interval [CI]:
poorly described, and the mechanisms responsible for these 0.981–0.994) than at control sites (0.974, 95% CI: 0.969–0.980;
responses are largely unknown [1–4, 6]. Species’ avoidance btreatment = 0.85 6 0.20 standard error [SE]), reflecting the
of noisy habitat may be a result of ecological intolerances of difference in predation. Assuming a 23-day nest cycle, the pre-
noise or species’ inability to effectively communicate through dicted nest success from these DNS estimates was 22%
the din of human activities [1–4, 6, 18, 19]. Vocal frequency higher at treatment sites than at control sites (Figure 2A). Inclu-
characteristics of indicator species suggest that the latter sion of those habitat features that differed slightly between
may have occurred at our sites. The two control-site indicator treatment and control sites (amount of bare ground and
species have vocalizations characterized by low frequencies: number of pinyon trees) did not improve DNS model perfor-
mourning dove vocalizations have an emphasized frequency mance over the model with the site-type covariate. (For likeli-
(i.e., the frequency at which the vocalization has the highest hood-ratio test results, see Supplemental Data.)
amplitude) near 527 Hz [20], and black-headed grosbeak notes To more thoroughly examine the relationship between noise
range from 1.5 to 4.0 kHz [21]. These frequency ranges overlap and nest predation, we estimated daily nest predation (DNP)
with most anthropogenic noise (<2.0 kHz) and are within the by excluding all nests that failed for reasons other than preda-
frequency range of noise produced at our treatment sites tion and used only those nests that were successful or depre-
(<5 kHz; Figure S2). Acoustic masking likely limits these dated. In this context, estimates of DNP were inverse
species to control sites where their vocalizations can be heard. measures of DNS. As expected, the DNP model including
In contrast, treatment-site indicator species have vocalization nest placement at treatment or control sites was an improve-
frequencies that may escape the masking effects of noise or ment over a constant DNP model (likelihood-ratio test, c21 =
are capable of adjusting vocal signals in response to noise. 27.0, p < 0.001). DNP was much higher at control sites than
Black-chinned hummingbird vocalizations span 1.5–12.0 kHz at treatment sites (Figure 2B). In other words, probability of
but generally have the most energy above 5 kHz [22], and not being depredated was higher at treatment sites than at
house finches are known to sing with higher minimum frequen- control sites (DNS btreatment = 1.14 6 0.24 SE). Given that the
cies in response to urban noise [23]. Signal adjustments may composition of the breeding community differed at treatment
not permit house finches to escape masking effects of noise and control sites, species-specific differences in DNP could
entirely but might shift signals to higher frequencies at which potentially explain differences in nesting success between
compressor noise has less acoustic energy (Figure S2). site types; therefore, we also used measured noise amplitudes
Signal plasticity or use of frequencies above those domi- at each nest to predict DNP for three species common to
nated by noise may facilitate black-chinned hummingbirds treatment and control sites and for the entire community.
and house finches in their ability to inhabit noisy areas, yet DNP models including amplitude were significantly better
these attributes do not explain each species’ preference for than constant DNP models (likelihood-ratio tests, community:
treatment sites over control sites in their nest site selection. c21 = 30.6, p < 0.001; gray flycatcher: c21 = 7.0, p = 0.004;
To our knowledge, this is the first evidence demonstrating spotted towhee: c21 = 3.5, p = 0.04; chipping sparrow [Spizella
that some species select for noisy habitats over quiet habitats, passerina]: c21 = 4.2, p = 0.02). Increases in noise amplitude
and this finding provides an intriguing focus for future resulted in lower DNP (i.e., the probability of a nest escaping
research. These species may use noise as a settlement cue predation increased) for all three species and for the pooled
in habitat selection; however, in light of our evidence docu- nesting community (Figure 2C). These results suggest that
menting an altered community structure plus higher nest higher nest success at treatment sites can be attributed to
success and lower levels of predation in noisy areas (see noise rather than to different rates of nest success among
below), the possibility exists that these species are responding species and that the difference results from reduced nest
indirectly to noise via factors such as lower interspecific predation with increased noise amplitudes.
Current Biology Vol 19 No 16
1418

Figure 2. Influence of Noise on Nest Success,

Nest Predation, and a Major Nest Predator, the
Western Scrub-Jay
(A) Predicted nest success from daily nest
survival (DNS) estimates was higher at treatment
sites than at control sites for the entire nesting
community. Error bars denote standard error (SE).
(B) Daily nest predation (DNP, the inverse of DNS)
was calculated using only nests with known
outcomes that were successful or failed as
a result of predation (see Supplemental Data).
DNP was higher at control sites than at treatment
sites. Data are shown as DNP and SE.
(C) Increases in noise amplitude decreased DNP
for the nesting community and individual species
that nest at treatment and control sites (commu-
nity: DNS bdB = 0.092 6 0.02 SE; gray flycatcher:
DNS bdB = 0.06 6 0.04 SE; spotted towhee: DNS
bdB = 0.10 6 0.07 SE; chipping sparrow: DNS
bdB = 0.10 6 0.045 SE). Results are displayed as
DNP. Xs denote nesting community; diamonds
denote gray flycatcher; squares denote spotted
towhee; circles denote chipping sparrow.
(D) Baited artificial nests paired with motion-
triggered cameras were used to identify common
nest predators. Photograph shows a western
scrub-jay removing an egg from an artificial nest
at one of the study sites.
(E) The occupancy rate estimate for western
scrub-jays was significantly lower at treatment
sites than at control sites. Results are reported
as the proportion of point-count stations occupied
at treatment or control sites. Error bars denote SE.

Nest Predator Response to Noise areas, especially those that rely on acoustic cues to locate
To account for differences in nest predation between sites, we nests, may be less likely to locate nests because of the mask-
hypothesized that common nest predators were absent from ing effects of noise, which would also lead to increased nest
or less abundant at treatment sites than control sites. Using success of prey species with noise amplitude.
baited artificial nests paired with motion-triggered cameras, That noise changes patterns of nest predation has important
we identified the western scrub-jay (Aphelocoma californica) implications for additional species interactions in noisy land-
as the primary nest predator in our study area (Figure 2D). To scapes. For example, in our study area the scrub-jay is not
determine whether there was any evidence that noise influ- only a major nest predator, it is also a key mobile link for pinyon
ences occupancy rates of this important nest predator, we pine through dispersal of its seeds [28, 29]. Scrub-jay avoid-
used standard point-count surveys at treatment and control ance of noisy habitats may have negative consequences for
sites, with treatment-site compressors turned off. We esti- seedling recruitment that could result in decreased pinyon
mated scrub-jay occupancy with Presence occupancy pine densities in noisy areas, potentially affecting many organ-
modeling software (http://www.mbr-pwrc.usgs.gov/software/ isms and community dynamics that are dependent on pinyon
presence.html) and found the inclusion of a noise covariate pine [30, 31]. Knowledge of the full extent to which noise can
(noise-conditional model) to significantly improve occupancy trigger changes is urgently needed, given the rate at which
estimations over a uniform occupancy model (likelihood-ratio natural habitat is being transformed by human activities.
test, c21 = 17.3, p < 0.001). Scrub-jay occupancy rates deter- More insight on the cumulative consequences of noise pollu-
mined from the noise-conditional model were 32% higher at tion may be gained through studies that focus on species
control sites than at treatment sites (Figure 2E). This was the with important roles within communities across diverse habitat
pattern we expected to see and supports the pattern of lower types.
predation rates for nests at treatment sites.
Contrary to the concern that noise may negatively influence Conclusions
nest success [1, 3], our findings show that noise can have an The current study has important implications for both avian
indirect positive effect for individuals nesting in noisy areas. conservation and community ecology within human-altered
This result exemplifies the importance of examining the conse- landscapes. The change in the avian community is in line
quences of anthropogenic disturbance from a community- with earlier studies implicating the negative influence of noise
level perspective. In our study area, the decrease in nest on birds [8, 10–14]; however, we provide the first evidence of
predation was a result of the western scrub-jay’s avoidance this trend while simultaneously controlling for confounding
of noisy habitat. Scrub-jays’ intolerance of noisy habitat, stimuli and potential noise-caused detection biases. This is
much like the intolerance of the control-site indicator species, the strongest evidence to date that noise negatively influences
may be a result of acoustic masking of its vocalizations, which bird populations and communities, and acoustic masking may
include frequencies below 2 kHz [27]. Future research should be a dominant mechanism precluding many birds from
consider the possibility that nest predators present in noisy breeding in noisy habitats [1–4, 18, 19, 32, 33]. Because noise
Avian Community Consequences of Noise Pollution
1419

also indirectly facilitates reproductive success, species intol- 11. Kuitunen, M. (1998). Do highways influence density of land birds?
erant of noise may suffer from not only exclusion from noisy Environ. Manage. 22, 297–302.
12. Canaday, C., and Rivadeneyra, J. (2001). Initial effects of a petroleum
habitats that might be otherwise suitable but also higher rates
operation on Amazonian birds: Terrestrial insectivores retreat. Biodiver-
of nest predation relative to species inhabiting noisy areas. If sity and Conservation 10, 567–595.
this phenomenon is common to noisy environments, it may 13. Forman, R.T.T., Reineking, B., and Hersperger, A.M. (2002). Road traffic
help explain the high degree of success among urban-adapted and nearby grassland bird patterns in a suburbanizing landscape.
species and the homogenization of avian communities in Environ. Manage. 29, 782–800.
and around human-altered habitats [33, 34]. Perhaps more 14. Bayne, E.M., Habib, L., and Boutin, S. (2008). Impacts of chronic anthro-
pogenic noise from energy-sector activity on abundance of songbirds in
noteworthy, however, is that noise alone can disrupt species
the boreal forest. Conserv. Biol. 22, 1186–1193.
interactions, potentially influencing many organisms and 15. Dooling, R.J., and Popper, A.N. (2007). The effects of highway noise
processes indirectly. Noise pollution is becoming much more on birds (http://www.dot.ca.gov/hq/env/bio/files/caltrans_birds_10-7-
prevalent throughout much of the world. Knowledge of how 2007b.pdf).
species respond to this novel force, especially species with 16. Peris, S.J., and Pescador, M. (2004). Effects of traffic noise on passerine
critical links within the ecosystem, may be crucial to maintain- populations in Mediterranean wooded pastures. Appl. Acoustics 65,
357–366.
ing biodiversity and ecological processes in the growing
17. Dufrêne, M., and Legendre, P. (1997). Species assemblages and indi-
number of landscapes disrupted by our industrial clamor. cator species: The need for a flexible asymmetrical approach. Ecol.
Monogr. 67, 345–366.
Supplemental Data 18. Rheindt, F.E. (2003). The impact of roads on birds: Does song frequency
play a role in determining susceptibility to noise pollution? J. Ornithol.
Supplemental Data include Supplemental Experimental Procedures, one 144, 295–306.
table, and four figures and can be found with this article online at http:// 19. Habib, L., Bayne, E.M., and Boutin, S. (2007). Chronic industrial noise
www.cell.com/current-biology/supplemental/S0960-9822(09)01328-1. affects pairing success and age structure of ovenbirds (Seiurus auroca-
pilla). J. Appl. Ecol. 44, 176–184.
20. Tubaro, P.L., and Mahler, B. (1998). Acoustic frequencies and body
Acknowledgments
mass in New World doves. Condor 100, 54–61.
21. Ritchison, G. (1980). Singing behavior of the black-headed grosbeak,
We are grateful to our many research assistants for field support, especially
Pheucticus melanocephalus, with an emphasis on the function of
R. Kennedy and P. Nylander. We also thank anonymous reviewers for useful
singing in female. PhD thesis, Utah State University, Logan, UT.
suggestions and comments. This work was supported primarily by the
22. Rusch, K.M., Pytte, C.L., and Ficken, M.S. (1996). Organization of
United States Bureau of Land Management, ConocoPhillips, and Williams
agonistic vocalizations in black-chinned hummingbirds. Condor 98,
Energy, plus the Animal Behavior Society, Southwestern Association of
557–566.
Naturalists, a Beverly Sears Graduate Student Grant Award, Colorado Field
23. Fernández-Juricic, E., Poston, R., De Collibus, K., Morgan, T., Bastain,
Ornithologists, the Edna Bailey Sussman Foundation, the Mountain Studies
B., Martin, C., Jones, K., and Treminio, R. (2005). Microhabitat selection
Institute, an Ariel Appleton Research Fellowship with funding from the Wells
and singing behavior patterns of male house finches (Carpodacus
Fargo Foundation, and the University of Colorado Department of Ecology
mexicanus) in urban parks in a heavily urbanized landscape in the
and Evolutionary Biology, University of Colorado Graduate School.
western US. Urban Habitats 3, 49–69.
24. Quinn, J.L., Whittingham, M.J., Butler, S.J., and Cresswell, W. (2006).
Received: May 6, 2009 Noise, predation risk compensation and vigilance in the chaffinch
Revised: June 9, 2009 Fringilla coelebs. J. Avian Biol. 37, 601–608.
Accepted: June 19, 2009 25. Schaub, A., Ostwald, J., and Siemers, B.M. (2009). Foraging bats avoid
Published online: July 23, 2009 noise. J. Exp. Biol. 211, 3174–3180.
26. Shaffer, T.L. (2004). A unified approach to analyzing nest success. Auk
References 121, 526–540.
27. Curry, R.L., Peterson, A.T., and Langen, T.A. (2002). Western scrub-jay
1. Slabbekoorn, H., and Ripmeester, E.A. (2008). Birdsong and anthropo- (Aphelocoma californica). In The Birds of North America, No. 712, A.
genic noise: Implications and applications for conservation. Mol. Ecol. Poole and F. Gill, eds. (Philadelphia: The Birds of North America, Inc.).
17, 72–83. 28. Vander Wall, S.B., and Balda, R.P. (1981). Ecology and evolution of
2. Katti, M., and Warren, P.S. (2004). Tits, noise and urban bioacoustics. food-storage behavior in conifer-seed-caching corvids. Z. Tierpsychol.
Trends Ecol. Evol. 19, 109–110. 56, 217–242.
3. Warren, P.S., Katti, M., Ermann, M., and Brazel, A. (2006). Urban 29. Chambers, J.C., Vander Wall, S.B., and Schupp, E.W. (1999). Seed and
bioacoustics: It’s not just noise. Anim. Behav. 71, 491–502. seedling ecology of piñon and juniper species in the pygmy woodland of
4. Patricelli, G.L., and Blickley, J.L. (2006). Avian communication in urban western North America. Bot. Rev. 65, 1–38.
noise: Causes and consequences of vocal adjustment. Auk 123, 639– 30. Mueller, R.C., Scudder, C.M., Porter, M.E., Trotter, R.T., III, Gehring,
649. C.A., and Whitham, T.G. (2005). Differential tree mortality in response
5. Pacifici, K., Simons, T.R., and Pollock, K.H. (2008). Effects of vegetation to severe drought: Evidence for long-term vegetation shifts. J. Ecol.
and background noise on the detection process in auditory avian point- 93, 1085–1093.
count surveys. Auk 125, 600–607. 31. Brown, J.H., Whitham, T.G., Ernest, S.K.M., and Gehring, C.A. (2001).
6. Brumm, H., and Slabbekoorn, H. (2005). Acoustic communication in Complex species interactions and the dynamics of ecological systems:
noise. Adv. Stud. Behav. 35, 151–209. Long-term experiments. Science 293, 643–650.
32. Slabbekoorn, H., and Peet, M. (2003). Birds sing at a higher pitch in
7. Forman, R.T.T., and Alexander, L.E. (1998). Roads and their major
urban noise. Nature 424, 267.
ecological effects. Annu. Rev. Ecol. Syst. 29, 207–231.
33. Slabbekoorn, H., and den Boer-Visser, A. (2006). Cities change the
8. Reijnen, R., and Foppen, R. (2006). Impact of road traffic on breeding
songs of birds. Curr. Biol. 16, 2326–2331.
bird populations. In The Ecology of Transportation: Managing Mobility
34. Clergeau, P., Croci, S., Jokimäki, J., Kaisanlahti-Jokimäki, M.-J., and
for the Environment, J. Davenport and J.L. Davenport, eds. (Heidelberg:
Dinetti, M. (2006). Avifauna homogenization by urbanization: Analysis
Springer-Verlag), pp. 255–274.
at different European latitudes. Biol. Conserv. 127, 336–344.
9. Campo, J.L., Gil, M.G., and Davila, S.G. (2005). Effects of specific noise
and music stimuli on stress and fear levels of laying hens of several
breeds. Appl. Anim. Behav. Sci. 91, 75–84.
10. Reijnen, R., Foppen, R., Braak, C.T., and Thissen, J. (1995). Reduction
of density in relation to the proximity of main roads. J. Appl. Ecol. 32,
187–202.