Urban effects on native avifauna review

LandscapeandUrbanPlanning74(2006)46–69

Urbaneffectsonnativeavifauna:areview

JamesonF.Chace

a,∗

,JohnJ.Walsh

b,c
aDepartmentofBiology,VillanovaUniversity,Villanova,PA19085-1699,USA
DepartmentofUrbanDesignandPlanning,UniversityofWashington,Seattle,WA98195,USA
DepartmentofPhysicalPlanning,Alterra,P.O.Box47,6700AAWageningen,TheNetherlands
Received5December2003;accepted31August2004

Availableonline24November2004Abstract
Theeffectofurbanizationcanbeimmense,yetourunderstandingisrudimentary.Here,wecompilethemostrecentinformation
on urban impacts on avian populations and communities. Compared to other vertebrates, birds are easily monitored by skilled
observers and provide a mechanism to explore urban effects and responses to different urban designs. Taxonomically, bird
communities in distinctly different habitats are most different in the least disturbed sites and the most similar in the most
urbanized sites. Urbanization tends to select for omnivorous, granivorous, and cavity nesting species. Increased urbanization
typicallyleadstoanincreaseinavianbiomassbutareductioninrichness.Unlikemostpasserines,raptorsmayhavehomeranges
that extend beyond the urban boundary and therefore do not need to meet all their ecological requirements within urban areas.
Urban habitats are often of superior quality to raptors because there they are often free from persecution and have an adequate
food supply. The processes that underlie the patterns of population and community level responses need more attention, but
severalareasofhavebeenidentiedasbeingimportant.Birdsrespondtovegetationcompositionandstructure,andurbanareas
thatretainnativevegetativecharacteristicsretainmorenativespeciesthanthosethatdonot.Avianfecundityinurbanareasisa
reection of species-specic adaptability to urban resources, and to levels of nest predation and nest parasitism. Additionally,
non-consumptivehumanactivitiesthatincreasewithurbanizationarerecognizedashavingnegativeimpactsonavianpopulations
and communities. Avian survivorship in urban areas is inuenced by risk of collision with man-made objects, changes in the
predator assemblage, food supply, and disease. Missing are thorough investigations in the regions of highest human population
growth, e.g. Southeast Asia. Additionally, there is a paucity of information from regions of high avian diversity, e.g. tropical
forests.Clearly,localknowledgeandstudyisrequiredbeforeimplementationofmanagementpoliciestoreduceurbanimpacts
on bird communities. Hopefully, such policies will include long-term monitoring. Demographic parameters of fecundity and
survivorship need to be examined in conjunction with measures of community diversity and density across the urban gradient
to better understand the quality of different urban habitats, and the variation of quality among spatial patterns of urbanization
withinthenativehabitatmatrix.
©2004ElsevierB.V.Allrightsreserved.Keywords:Avianecology;Birds;Landscapeecology;Urbanplanning;Urbanization

∗
Correspondingauthor.Tel.:+16105197534;fax:+16105197863.
E-mailaddress:jameson.chace@villanova.edu(J.F.Chace).
0169-2046/$20.00©2004ElsevierB.V.Allrightsreserved.

doi:10.1016/j.landurbplan.2004.08.007
J.F.Chace,J.J.Walsh/LandscapeandUrbanPlanning74(2006)46–6947

1. Introduction
Urbanizationcanbedenedasconcentratedhuman
presenceinresidentialandindustrialsettingsandtheir
associatedaffects(CringanandHorak,1989;Marzluff,
1997),andforthepurposesofecologicalstudiesurban
centers have been quantied as containing more than
2500 people (Dumouchel, 1975). The urban extent of
mostmetropolitanareasisexpandingintoadjacentru-
rallandscapes(AligandHealy,1987;WorldResources
Institute, 1994; UN, 1997). With the projected global
increaseofurbanization,landcoverconversionsforur-
ban use will only increase altering ecosystem patterns
andprocesses(Grimmetal.,2000).
The factors determining which species can coex-
ist with human settlement include: (1) the presence
and patch size of remnant (native) vegetation (Emlen,
1974; Gavareski, 1976; Rosenberg et al., 1987; Mills
et al., 1989; Catterall et al., 1991); (2) competition
with exotic species that have a longer history of hu-
man cohabitation (Major et al., 1996); (3) non-native
predators (Churcher and Lawton, 1987; Paton, 1990);
(4)thestructureandoristicattributesofplantedveg-
etation (Tweit and Tweit, 1986; Green et al., 1989);
(5) supplementary feeding by humans (Recher, 1972;
Brittingham,1990;Majoretal.,1996);and(6)residual
pesticides(Majoretal.,1996).
Thenumberofstudiesthatdescribeavianresponses
to urbanization is immense and growing (Marzluff et
al.,2001).Fortheurbanplanner,weattempttosumma-
rizeintoonepaperthepatternsofavianpopulationand
communityresponsetotheurbanenvironmentbyma-
jor habitat types. Where possible we illustrate major
points with specic species (scientic names in Ap-
pendix A) to assist the urban planner and others in-
volvedintheplanningprocesstoidentifylocalspecies
responsestolandscapechanges.Next,wereviewstud-
ies that have illuminated some of the major processes
that have contributed to the observed patterns of pop-
ulation and community change with urbanization. Fi-
nally, we outline future avian research needs that will
specicallyaidurbanplanningdecisions.
2. Patternsofurbanimpactsbymatrixhabitat

type
Total breeding bird density is often higher in urban
areas than in the surrounding native habitat (Walcott,1974; Gavareski, 1976; Lancaster and Rees, 1979;
BeissingerandOsborne,1982).However,speciesrich-
ness is usually lower in urban areas, where the avian
community is dominated by a few, often introduced,
species (Gavareski, 1976; Lancaster and Rees, 1979;
Beissinger and Osborne, 1982; Cam et al., 2000). Ur-
banizationfavorsafewspeciesbutselectsagainstmost
such that the avian community composition of urban
environments differs dramatically from local natural
environments (Beissinger and Osborne, 1982; Rosen-
berg et al., 1987; Mills et al., 1989; Jokim

¨aki and
Suhonen, 1993; O’Connell et al., 2000). Urbaniza-
tion tends to favor granivores, aerial insectivores, and
ground foraging insectivores (Emlen, 1974; Allen and
O’Conner, 2000), and residents over migrants (Allen
and O’Conner, 2000; Kluza et al., 2000; Poague et
al.,2000).Instudieswheretheurbangradientismore
nely divided, the peak of avian diversity is found in
areasofmoderatelevelsofdisturbance,ofteninsubur-
ban areas or at the urban/wildland interface (Jokim¨aki
andSuhonen,1993;Blair,1999).Pre-developmentbird
speciesdropoutgraduallyfromthecommunityassites
becomemoreurban(Blair,2001).

2.1. Desertscrub
The avian community composition of the desert
scrubsurroundingTucson,Arizonamarkedlychanges
within city limits (Emlen, 1974). Urbanization favors
seed eaters, ground foraging insectivores, water de-
pendent and crevice nesting species of desert scrub
ecosystems (e.g. white-winged dove, Inca dove, and
northern mockingbird), as well as exotics (e.g. house
sparrow and European starling). Insectivores, cavity
nesting species, and species with Type A territories
(i.e.non-overlappingterritorieswherethehomerange
is approximately the same size and the defended ter-
ritory space) disappeared from the urbanized desert
scrubcommunity.Urbanandwilddensitiesoftheash-
throated ycatcher and curve-billed thrasher did not
differsignicantly,seeminglyunaffectedbyurbaniza-
tion(Emlen,1974).
Mills et al. (1989)found a similar response among
Tucson’s avifauna. Density of exotic (house sparrow,
rockdove,Europeanstarling)andurban-adaptednative
species (mallard, Inca dove, American robin, bronzed
cowbird, and great-tailed grackle) increased signi-
cantlywithurbanization.Othernativespeciesexhibited
48J.F.Chace,J.J.Walsh/LandscapeandUrbanPlanning74(2006)46–69
no signicant change in density across the urban gra-
dient.
Both studies found a decrease in diversity with ur-
banization(Emlen,1974;Millsetal.,1989);however,
Millsetal.’s(1989)densityandbiomassestimatesare
mostrevealingofurbanimpacts.Theurbancommunity
averaged 1230 individuals per 100 acres with an avi-
faunal biomass of 53,208g, whereas the desert scrub
community averaged a density of 47 individuals with
2052gofbiomassoverthesamearea.

2.2. Closedcanopyforests
Temperate:Beissinger and Osborne (1982)found
the avian community response of biomass and den-
sity to urbanization in Oxford, Ohio similar to what
Emlen (1974)reported in Arizona. Forests contained
higher species richness but lower density (1020–1253
individualsper40ha;4455–5736gbiomass)compared
to residential areas (1320–1667 individuals per 40ha;
biomass 12,637–15,445g). Urbanization favors seed
eaters,ominvores,andgroundforagers,whileselecting
againsthighcanopyandfoliageforagers,insectivores,
bark gleaners and drillers. Specically, rock doves,
mourningdoves,chimneyswifts,Americanrobins,Eu-
ropean starlings, house sparrows, and common grack-
les responded positively to urbanization; ycatchers,
red-eyed vireos, cerulean warblers, and most wood-
peckers (except northern ickers and downy wood-
peckers) responded negatively.
The effects of fragmentation of the eastern de-
ciduous forest by agriculture and suburban develop-
ment on bird populations has received considerable
attention (e.g.Askins et al., 1990; Hagan and John-
ston, 1992; Friesen et al., 1995; Kluza et al., 2000).
Forests fragmented by urbanization into small forest
tracts(4–25ha)resultedinthedeclineofforestinterior
speciessuchastheeasternwood-pewee,woodthrush,
scarlettanager,rose-breastedgrosbeak,andBaltimore
oriole, while great-crested ycatchers and red-eyed
vireos were the least affected under those conditions
(Friesen et al., 1995). Likewise, lightly treed parks
often support suburban bird communities rather than
species of forest interior or forest-dependent species
(Hudsonetal.,1997).Forestsfragmentedbyruralres-
idential development (0.01–6.7house/ha) in western
Massachusettshadfewermigratoryandforest-interior
species (e.g. veery, wood thrush and ovenbird;Kluzaetal.,2000).Alongwoodedrights-of-wayinthegrass-
land matrix of Nebraska, total relative abundance did
notdifferbetweenruralandurbanforests,whilespecies
richnesswashigherintheurbanareasexceptduringmi-
grationwhenmigratorybirdsappearedtopreferentially
choose the rural woodland habitat over urban (Poague
etal.,2000).
Marzluff’s (1997)review of urbanization impacts
on southwestern ponderosa pine forest communities
suggestssimilartrends.Hummingbirds,corvids,swal-
lows, blackbirds, European starlings, nches, Ameri-
can robins, and house sparrows respond positively to
urbanization. Flycatchers, tanagers, vireos, warblers,
thrushes, and bluebirds respond negatively to urban-
izationinponderosapineforests.
A few studies have measured temporal responses
of bird populations and communities to urban-
ization (Walcott, 1974; Aldrich and Cofn, 1979;
Nowakowski,1996;Fitzgeraldetal.,1999).Suchstud-
ies provide direct inferences to changes associated
withurbandevelopment.InCambridge,Massachusetts
Walcott (1974)compared birds censused on two plots
between1860and1964.Naturalhabitatdeclinedfrom
95to30%,andfrom50to15%onthetwosites,respec-
tively. Bird fauna changed from 26 species of mostly
migratory breeders and a few transients, to 9 species,
largely permanent residents. In Fairfax County, Vir-
ginia,theavifaunalcommunitywasexaminedbetween
1942and1979(AldrichandCofn,1979).Thenumber
of species increased as the deciduous forest landscape
of 1942 became dominated by suburban residential
in 1979 (23 species versus 29 species), and the den-
sityofbirdsincreasedby140%(195territoriesversus
277territories).Mostinterestingwasthedramaticshift
in community composition among the ten most abun-
dantspecies,especiallythosethatwereextirpatedafter
1942 and those that colonized by 1979 (Table 3). In
theexpandingurbancenterofOlsztyn,Poland,sixrare
species disappeared, ve new species appeared (most
signicantlytherockdove),11specieshaveincreased
signicantly, and 15 species have decreased signi-
cantly between 1968 and 1993 (Nowakowski, 1996).
Overthecourseof5yearsofurbanizationinMissouri,
Fitzgerald et al. (1999)found that the blue-gray gnat-
catcher,indigobunting,acadianycatcher,black-and-
white warbler, and the northern parula have declined
signicantly(thelatterthreespeciesareforestinterior
specialists).Comparingbirdpopulationsinresidential
J.F.Chace,J.J.Walsh/LandscapeandUrbanPlanning74(2006)46–6949
communitiesofdifferentages(35,20,and10yearsold)
in Alberta, Canada,Edgar and Kershaw (1994)found
that density, richness, and evenness increased with ur-
ban community age. The only species common to all
three areas was the introduced house sparrow which
wasthemostabundancespeciesinallthreeurbancom-
munities.
Tropical: In Panama, 11 habitats were sur-
veyed across an anthropogenic disturbance gradi-
ent, including residential areas (Petit et al., 1999).
Nearctic–Neotropicalmigratoryspeciesweremostnu-
merous in residential, lowland forest fragments and
shade-coffeeplantations.Higherportionsoffrugivores
and nectarivores characterize residential avian com-
munities compared to native forest areas. In Jakarta,
Indrawan and Wirakusumah (1995)found avian rich-
ness had decreased by over 50% in 40 years (199

speciesto
A higher abundance of human commensal species
(e.g. house swift and common myna) are associ-
ated with increases in built areas, while parkland and
rainforest-associated bird species (e.g. greater green
and short-tailed babbler) have increased abundance
withnativeandmanagedvegetationinthetropicalcity
of Singapore (Sodhietal.,1999).

2.3. Grasslands
In Saskatoon, Saskatchewan species richness was
not signicantly different between rural and urban
sites, however abundance was signicantly higher in
the urban sites (Sodhi, 1992). Most of the urban bird
fauna was composed of non-grassland and introduced
species, with the house sparrow being the most abun-
danturbanspecies.Thehornedlarkandwesternmead-
owlarkdominatedtheruralsites,whiletheclay-colored
sparrowwastheonlygrasslandspeciesrelativelycom-
moninurbansites.
GrasslandbirdsinColoradorespondtoedgesandto
theextentofurbanization(Bocketal.,2001).Bockand
hiscolleaguesfoundgrasslandnestingbirdabundance
twice as great on interior plots as on edge plots. Ves-
per sparrow, savannah sparrow, grasshopper sparrow,
bobolink, and western meadowlark were most sensi-
tive to urbanization. American robin, European star-
ling,commongrackle,housench,andhousesparrow
were ve times more abundant on grassland/suburban
edges than on grassland interior plots. Reduced abun-dance of grassland species (vesper sparrow, grasshop-
per sparrow, and horned lark) along the urban edge is
duetothelossofpreferredgrasslandcover-typesatthe
urban-wildlandinterface(Haireetal.,2000).
Engle et al. (1999)examined the impact of low-
density rural sprawl on bird communities of the Great
PlainsnearTulsa,Oklahoma.Humandevelopmentfa-
vored the barn swallow, dickcissel and grasshopper

sparrowatlowhumandensities(

−2

)and
the American robin, common grackle, European star-
ling,housesparrow,andpurplemartinathigherdensi-

ties(>12peoplekm

−2

).Mostnotably,forestandedge
species(e.g.tuftedtitmouse,Bewick’swren,Kentucky
warbler, summer tanager, chipping sparrow), were re-
placedbyspeciesassociatedwithhumandevelopment,
especiallyinthehigherdensityruralareas.
2.4. Australianeucalyptusforest,bushlands,and

subtropicalrainforest
Sewell and Catterall (1998)surveyed bird commu-
nities across an urban gradient from large patches of
bushland to urban sites without native vegetation in
Brisbane, Queensland. They detected a decrease in
diversity and an increase in abundance with increas-
ingurbanization,patternbroadlyconsistentwithother
studies (Emlen, 1974; Rosenberg et al., 1987; Mills
et al., 1989). However, suburban sites had the highest
abundanceanddiversity,especiallyamonglow-density
housingthatretainedeucalyptuscanopystructure,and
amongtheexotic-plantedsuburbs.However,thesesites
had a low abundance of eucalyptus forest birds, there-
fore suburban planting does not reverse the effects
of deforestation of eucalyptus forest, but promotes
a distinctive suburban bird community. Within urban
environments, bird communities show little seasonal
variationascomparedtowoodlandoreucalyptusforest
bird communities in Queensland, Australia (Catterall
etal.,1998).
Inwetsclerophyllandsubtropicalrainforest,Wood
(1996)comparedbirdcommunitiesina5.4hadegraded
public reserve with an adjacent 55-year old residen-
tialareaofWollongong,NewSouthWales(population
200,000).Inallseasons,thereservehadhigherspecies
richness but lower total abundance. There were 13 re-
servespecialists(e.g.white-browedscrubwren,eastern
yellow robin, brown gerygone, and eastern whipbird,
ofwhichthelattertwoarethemostvulnerabletolocal
50J.F.Chace,J.J.Walsh/LandscapeandUrbanPlanning74(2006)46–69
extinction), 5 exotic, suburb specialists (e.g. common
myna, red-whiskered bulbul, European starling, and
house sparrow), and 17 generalists. Specialist species
were similar in Wollongong and in a similar study in
Brisbane (Catterall et al., 1989). Wood estimates that
15 species have been extripated with suburban devel-
opment.

2.5. Coastalsage-scrub
Large coastal sage-scrub reserves in urban areas
may ameliorate urban impacts on the native commu-
nity assemblage (Crooks and Soul´e, 1999).Sauvajot
and Buechner (1993)found no relationship of diver-
sity and abundance associated with degrees of urban
exposure in chaparral habitats of the Santa Monica
Mountains; however, the study lacked an appropriate
control site. The control site was a 5000ha state park
butwasnotindependentoftheurbanarea.Ratherthan
an independent control, it served as a source popula-
tionforsomeotherwiseurban-sensitivespecieslocated
on the urban treatment plots. Distance from urban ar-
easisanimportantaspect(Munyenyembeetal.,1989;
Catteralletal.,1989,1991;Bolgeretal.,1997).Many
coastalsage-scrubspeciesoccupiedintacthabitateven
ifimmediatelyadjacenttourbanareas.However,some
of these coastal sage-scrub species respond to habitat
patchiness: sage sparrow, California thrasher and Cal-
ifornia towhee were less common at edges, whereas
the northern mockingbird and European starling were
morecommonatedges(Kristanetal.,2003).
Steep undeveloped slopes in southern California
coastal sage scrub serves as temporary habitat within
urban matrix. Scrub-specialist disappearance in this
landscapeisrelatedtofragmentareaandfragmentage
(i.e. time since isolation;Crooks and Soul´e, 1999).
Within this urbanized landscapeBolger et al. (1997)
found that species distributions were likely inuenced
by landscape-scale vegetation patterns and by the ag-
gregateamountofurbanizedareas.Theurban-edgeaf-
fected species, that exhibited a signicant reduction
inabundancewithin200–500mofdevelopmentedge,
were sage, rufous-crowned, lark, and black-chinned
sparrows.Edgeenhancedspecies,withelevatedabun-
dance within 1000m from an edge, included the
house nch, northern mockingbird, lesser goldnch,
and Anna’s hummingbird. Interestingly, within one of
largestexpansesofsagescrubwithinoneofthelargesturbanizing landscapes, Los Angeles,Cooper (2002)
foundthatseveralsensitivespecies,includingCalifor-
niagnatcatcher,wereconrmedorsuspectedofbreed-
inginpatchessmallerthan100haduringthestudy,and
oftenattheurbaninterface.

2.6. Oakwoodlands
Along the pre-development to urban gradient,
species richness and abundance were found to be
highest in areas of intermediate disturbance (e.g. golf
courses and low density residential), while the pre-
development species dropped out of the community
with increasing urbanization (Blair, 1996). The avian
communitywasdividedintothreemajorclassesbased
ontheirresponsetourbanization:urbanavoiders,sub-
urbanadaptable,andurbanexploiters(Blair,1996).Of
the pre-development community, the Western scrub-
jay, Anna’s hummingbird, and the mourning dove
persisted across the urban gradient. Urban avoiders
included the dark-eyed junco, blue-gray gnatcatcher,
ash-throated ycatcher, Steller’s jay, wrentit, west-
ern wood-pewee, Hutton’s vireo. Urban exploiters in-
cluded the white-throated swift, rock dove and house
sparrow; 30 other species were classied as urban
adaptable(Blair,1996).
Scott (1993)found that initial development (i.e.
during the time of physical vegetation removal and
construction) caused the loss of a number of species
in southern California (including northern mocking-
bird, lazuli bunting, blue grosbeak, Costa’s humming-
bird, ash-throated ycatcher, California gnatcatcher,
phainopepla,Hutton’svireo,orange-crownedwarbler,
and Bullock’s oriole), but the increase in two species
(black-chinnedhummingbirdandCaliforniathrasher).
Urbanization in the oak woodlands of California
tends alter plant community composition and struc-
ture (Blair, 1996), as seen elsewhere (Beissinger and
Osborne, 1982; Rudnicky and McDonnell, 1989). In-
creased urbanization in the suburbs may actually in-
crease structural diversity (Emlen, 1974; Beissinger
and Osborne, 1982) leading to the higher abundances
and richness found byBlair (1996).However,inthe
most urban environments avian richness decreased
to six species, including the three urban exploiters,
speciesthatareadaptabletothereducedoristiccom-
positionandstructure(Blair,1996).Moststudieshave
examined avian community composition in urban and
J.F.Chace,J.J.Walsh/LandscapeandUrbanPlanning74(2006)46–6951
non-urban areas, whereasBlair’s (1996)work reveals
changes that occur along the urban gradient. These
results can be projected to predict trends in native
communities as an area becomes urbanized over time.
Blair’s(1996)resultsfromtheoakwoodlandsofCali-
forniaareconsistentwiththeintermediatedisturbance
hypothesis (Connell, 1978). They are also consistent
withMcDonnell et al.’s (1993)suggestion that biotic
factors are more limiting to avian populations at the
rural end, while physical factors are more limiting at
theurbanendoftheurbanizationgradient.

2.7. Generalizations
Taxonomically, the bird communities in distinctly
differenthabitatssuchasCaliforniacoastalsage-scrub
andOhioeasterndeciduousforestaremostdifferentin
theleastdisturbedsitesandthemostsimilarinthemost
urbanized sites (Blair, 2001). Urbanization selects for
omnivorous, granivorous, and cavity nesting species
(Emlen, 1974; Lancaster and Rees, 1979; Beissinger
andOsborne,1982;Rosenbergetal.,1987;Millsetal.,
1989; Allen and O’Conner, 2000; Kluza et al., 2000).
Increased urbanization leads to an increase in avian
biomass but a reduction in richness, a pattern largely
consistent across forests, desert scrub, and grassland
habitats. Few studies have examined the response of
species composition and abundance as development
progresses.
The avian community changes with urban devel-
opment (Scott, 1993). As time from development in-
creasesbirdcommunitiesbecomeatoncemoredistinct
from the native community (Wood, 1996; but also see
EdgarandKershaw,1994)butmorehomogenouswith
other urban areas (Blair, 2001). This is one area needs
to be explored in more detail. Studies with a greater
coverageoftimesincedevelopmentandoveragreater
range of plant communities are needed. Missing are
more thorough investigations in the regions of highest
humanpopulationgrowth,SoutheastAsia,andthereis
apaucityofinformationfromborealforestandtropical
forestbiomes.

3. Urbanimpactsonraptors
Studies of passerine responses to urbanization are
oftendevoidofimportantreproductiveinformationthat

Table1
TenmostabundantspeciesatLakeBarcroft,FairfaxCounty,Virginia
1942and1979(datafromAldrichandCofn,1979)

1942 1979

Red-eyedVireo

NorthernCardinal

Ovenbird

NorthernMockingbird

WoodThrush SongSparrow

ScarletTanager

BlueJay

HoodedWabler

EuropeanStarling

AcadianFlycatcher

GrayCatbird

EasternWood-pewee AmericanRobin

Red-belliedWoodpecker HouseSparrow

BlueJay MourningDove

NorthernCardinal CarolinaChickadee
Extirpated.
Newtosite.
is more descriptive of habitat quality than measures
of abundance (Van Horne, 1983). Studies of raptors
can be illuminating in this regard, as demographic pa-
rameters have often been measured on urban nesting
owls, hawks, falcons, and eagles. However, an impor-
tant caveat is that unlike most passerines raptors may
havehomerangesthatextendbeyondtheurbanbound-
aryandthereforedonotneedtomeetalltheirecological
requirementswithinurbanareas.
Urban habitats are often of superior quality to rap-
tors(CringanandHorak,1989)becausetheretheyare
often free from persecution that is more typical of ru-
ralregions.Freedomfrompersecutionandanadequate
foodsupplymayallowraptorstoinhabitotherwiseun-
suitable nesting sites (Newton, 1986). Small raptors
(e.g. eastern screech owl, tawny owl, Scops-owl, and
collared Scops-owl, and sparrowhawks) often inhabit
greenbeltsandparks(Clarketal.,1984;Newton,1986).
Great horned owls are common in metropolitan areas,
such as Seattle, Washington (Lambert, 1981). Raptors
with large home range requirements are more likely
to be negatively affected by habitat reduction or frag-
mentationthattypiestheurbanenvironmentthanare
smaller raptors (Newton, 1979; Phillips et al., 1984;
Scott,1985)(Table1).
Species that meet their food requirements within
the urban setting can exhibit positive population re-
sponses. Many falcon species respond well to urban
environments because of the large biomass of small
birds (Dietrich and Ellenberg, 1981; Horak, 1986;
Newton, 1986; James et al., 1987). The peregrine fal-
con,lesserkestrel,andAmericankestrelrespondwell,
52J.F.Chace,J.J.Walsh/LandscapeandUrbanPlanning74(2006)46–69

Table2

Summaryofurbanimpactsonraptors

Species Datatype

Notes Region Reference

ABC
PeregrineFalcon X 34–58%ofregionalpop.nestin

urbanarea.60U.S.CitiesCadeetal.(1996)
Threemostcommonpreyitems:

rockdove,northernicker,blue

jay

Populationsizenow>historical

abundance
PrairieFalcon X Scarceonurbanplots Boulder,COBerryetal.(1998)

LesserKestrel X Predationinurban

Lowerpreydeliveryrates,and

lowersuccess,inurban
JapaneseLesserSparrowhawk X Feedsprimarily(90%)onsmall

birdsinsuburbanTokyoJapanUeta(1992)
AmericanKestrel X Notsensitivetourbanization Boulder,COBerryetal.(1998)
Sparrowhawk X Success:village>rural>wild TheNetherlandsDiermen(1996)

Villagehawksbreedearlierand

laymoreeggs

MississippiKite X Urbanpopulationsvery

successfulMidwestUSParker(1996)
Red-shoulderedHawk X X 32%nestinurbanareas(N

=170)s.CABloom and McCrary

(1996)

Successfullynestinnon-native

trees

X Avoidnestingnearroadsand

buildingsBednarzand

Dinsmore(1981)
X >50%nestsinexotictrees CARottenborn(2000)

Successinexotictrees>native

trees

X Signicantlyavoidedsuburban

habitatNJBosakowskiand

Smith(1997)
Red-tailedHawk X Notsensitivetourbanization Boulder,COBerryetal.(1998)

X Signicantlyassociatedwith

suburbanhabitatNJBosakowskiand

Smith(1997)
Swainson’sHawk X Notsensitivetourbanization Boulder,COBerryetal.(1998)
FerruginousHawk X Scarceonurbanplots Boulder,COBerryetal.(1998)
Rough-leggedHawk X Scarceonurbanplots Boulder,COBerryetal.(1998)
Cooper’sHawk X Highestknownbreedingdensity
insuburbsWIRoseneldetal.

(1996)

Highestknownreproductive

successinsuburbs(clutch

size=4.2,nestlings=4)WI

X Urban>wild:clutchsize AZBoalandMannan

(1999)

Urban

X Smallerhomerangesinurban

areasTucson,AZMannanandBoal

(2000)
NorthernGoshawk X Avoidssuburbanhabitat NJBosakowskiand

Smith(1997)
J.F.Chace,J.J.Walsh/LandscapeandUrbanPlanning74(2006)46–6953

Table2(Continued)

Species Datatype

Notes Region Reference

ABC

X Winterrangeconcentratednear

urbanareasFinlandTornbergandColpaert

(2001)
BaldEagle X Scarceonurbanplots Boulder,COBerryetal.(1998)

Osprey X Absenceofsuitablenesttreeshas

leadto>75%useofman-made
structures(powerlines)GermanyMeyburgetal.(1996)

Man-madenestsuccess(n=366)

>natural(n=258)
EasternScreechOwl X Citypopulationmoredense,
moreproductive,andmorestableTXGehlbach(1996)

Avoidssuburbanhabitat NJBosakowskiand

Smith(1997)
BarredOwl X Pref.matureforestoverareasof
humanhabitationNJBosakowskietal.

(1987)
BurrowingOwl X Nestlingandedlingsuccessin

urban>ruralNMBotelhoand

Arrowood(1996)

Lowerpredationandlowernest

densitiesinurbanFLMillsapandBear

(2000)

Greaterfoodavailabilityinurban

TawnyOwl X Nestingsuccessin

urban=suburbanRomeRanazzietal.(2000)

X Behavioralswitchtosmallbird

preyinsuburbanenvironment

comparedtosmallmammalsin

naturalopenareasTomialojc(1970)
Great-hornedOwl X Associatedwithsuburbanhabitat NJBosakowskiand

Smith(1997)
A:Abundancedata;B:breedingdata;C:otherdatatype.
and have higher reproductive success in urban envi-
ronments (Table 2), although the prairie falcon ap-
pears to respond negatively to urbanization (Berry et
al., 1998). Sparrowhawks in The Netherlands have
greatersuccessinvillagesthaninruralareas,whichin
turn have greater success than those nesting in forests
(Diermen, 1996). Village sparrowhawks attract higher
quality mates and have larger clutch sizes (Diermen,
1996). Village hawks also show the least variety of
preyitemsselected,andthemostubiquitousneststruc-
ture selection (Diermen, 1996). Raptors that eat small
prey,includinginsects,cansuccessfullycolonizeurban
environments, such as the Mississippi kite, burrowing
owl, and screech owl. A number of small mammalian
human-commensals respond positively to urban envi-
ronments,andthereforeraptorsthatpreyonthemalso
respond positively, including the red-tailed hawk and
Swainson’s hawk. Whereas, those that prey on largermammalsoftendonotfareaswellintheurbanenviron-
ment, e.g. ferruginous hawk and rough-legged hawk.
Small accipiters respond positively to urban environ-
mentswheresmallbirdshaveclumpeddistributionsby
feeders;however,largeraccipitersoftendonotexhibit
asimilarresponse.Ospreyadapttourbanenvironments
when sh populations are high and persecution is low
(Spitzer et al., 1985). Vultures have increased world-
wide in response to early stages of urbanization, but
have decreased thereafter when waste supplies are re-
duced(Inigo,1986;Piperetal.,1986;Table3).
Theestablishmentofraptorsinurbanenvironments
is partially dependent on their ability to utilize new
or articial nesting substrates. Red-shouldered hawks
showavariedresponse;theyavoidsuburbanhabitatin
New Jersey, but nest successfully in non-native trees
in southern California (Table 2).Blue (1996)reports
thatosprey,red-tailedhawk,great-hornedowl,golden
54J.F.Chace,J.J.Walsh/LandscapeandUrbanPlanning74(2006)46–69

Table3
Raptoruseofman-madenestingsubstratesandtheimpactonreproductivesuccessandpopulationdynamics
Species Typeofstructure(n) Useandreproductivesuccess Region Reference
Red-tailedHawk Transmissiontowers(12) Articial100%(n=15)vs.natural

77%(n=84)WisconsinStoutetal.(1996)

Billboard(3)
Osprey Utilitytowers Useofutilitystructureshasincreased

withpopulationgrowthalongthe

WillametteRiver(OR)whileuseof

naturalstructureshasremained

constant,1976–1993production

equalOregonHennyandKaiser(1996)

HighuseoftowersinCanadaand

easternUSNewBrunswickStocek(1972)
Hydropoles(118) Increaseduseofarticialstructures

aspopulationincreasedsince1945
Transmissiontowers(14) GreatlakesBasinEwins(1996)

Communicationtowers(52)
Buildings(12) Reproductiveoutputfromarticial

sitesaveraged7.5%higherthan

naturalsites

Platforms(126)

Natural(579)
Powerlines Success:articialsites>naturalsites GermanyMeyburgetal.(1996)
Miscellaneous Absenceofsuitablenesttreeshas

leadto>75%useofman-made
structures(powerlines)GermanyMeyburgetal.(1996)

Success:man-madeneststructure

(n=366)>naturalstructure(n=258)
Utilitypoles Utilitypolenestsuccess=naturalsite

successOregonHennyandKaiser(1996)
FerruginousHawk Articialstructures Nestsuccess:articial>natural WyomingTigneretal.(1996)

Youngedged:articial>natural
Transmissiontowers Useoftransmissiontowers NorthDakotaGilmerandStewart(1983)
PeregrineFalcon Powerplants Highoccupancy UpperMidwestSeptonetal.(1996)
Buildings,bridges,towers Populationincreasesassociatedwith

articialsiteuseTemple(1988)

Bridges Successloweronbridgesthanat

naturalnestsitesSanFrancisco,

NewYorkCityBelletal.(1996)
AmericanKestrels Nestboxes Useinagriculturalareas IdahoBechardandBechard(1996)
Highusealonghighways IowaVarlandandLoughin(1993)
eagle,baldeagle,Americankestrel,Swainson’shawk,
eastern screech owl, Harris’ hawk, prairie falcon and
zone-tailed hawk use power line structures in the US.
Somespeciesaremoreexibleintheiruseofstructures
thanothers(Table2).Theuseofarticialstructureshas
been an important aspect of raptor population restora-
tion,andinsomecasesthepopulationstodayhaveex-
ceededhistoricnumbersbecauseofarticialstructures,
e.g. osprey (Table 2). Some species have even higher
reproductivesuccessonarticialnestsitesthannaturalsites, e.g. ferruginous hawks (Table 2). The impact of
urbanizationonmigrationroutesislikelytobeimpor-
tant,butthisaspecthasnotbeenfullyevaluated.
Behavioral response of raptors to humans is an im-
portantcomponentoftheirabilitytocolonizeurbanen-
vironments.Ferruginoushawkshomerangewasequal
between suburban and wild sites in Denver, Colorado,
while roosting time was lower for wild site hawks
(PlumptonandAndersen,1998).Red-tailedhawksha-
bituate to human intrusion with call and dive rates at
J.F.Chace,J.J.Walsh/LandscapeandUrbanPlanning74(2006)46–6955
nest sites lowest in areas of longest human contact
(Knight et al., 1989). Habituation to humans as young
mayleadtohighermortalityasadults.Cooper’shawk
nestlings that had frequent exposure to human intru-
sion(researchbiologistsbandingandweighingyoung)
weremorelikelytodiefromhuman-relatedcauses,es-
peciallyshooting(SnyderandSynder,1974).
4. Processesthatcontributetourbanimpacts

onbirdcommunities

4.1. Vegetationchanges
Thereisoftenastrongpositivecorrelationbetween
the volume and structure of native vegetation and na-
tive bird diversity and species richness (Emlen, 1974;
Millsetal.,1989).Likewise,non-nativespeciesdiver-
sity is correlated with exotic vegetation (Mills et al.,
1989).Emlen (1974)found that certain native desert
birds responded positively to urbanization in Tucson,
Arizona,acitythatmaintainsahighproportionofna-
tive vegetation. In support of Emlen’s assertion,Mills
et al. (1989)found that vegetation factors explained
a greater proportion of species diversity variance than
housing density. WhileGermaine et al. (1998)found
a strong positive correlation between housing density
and species richness for non-native species in Tucson,
nativebirdspeciesrichnessispositivelycorrelatedwith
the amount of native vegetation cover and negatively
correlatedwithdistancefromdesertwashes(Germaine
etal.,1998).
As the work in the Arizona desert attests, birds re-
spondtovegetationcompositionandstructure(Emlen,
1974;Millsetal.,1989;Germaineetal.,1998).Urban
environments are often savannah in nature (Dorney et
al.,1984),andusuallyhavegreaterfragmentedpatches
of native vegetation with more ground cover and less
canopy cover (Beissinger and Osborne, 1982). In ur-
banenvironments,vegetationismorefragmented,there
is less coverage at mid- and upper-canopy levels, and
there is more ground cover than in nearby wild sites.
Native plants are replaced by exotics, and few dead
standing trees remain. Therefore, it is not surprising
that in urbanized forest ecosystems avian communi-
tieslackcanopyforagingandbarkdrillingguilds,and
have an increase in ground gleaners (Beissinger and
Osborne,1982;Rosenbergetal.,1987).Theurbanen-vironment results in reduced diversity as total density
of a few species increases, often in response to food
provided by humans across the urban gradient. Such
urban environments favor cavity-nesting and ground
feedinggranivorousoromnivorousspecies(Lancaster
andRees,1979).
Native species typically drop out of the commu-
nity along the gradient from native to completely ur-
ban environments (Blair, 1996).Blair (1996)found
that in the oak woodlands of California total avian
richnesspeakedundermoderatelydisturbedconditions
(e.g. golf courses and low-density, detached, single-
unithomes),whichrunscountertotheresultsofmany
other studies (Graber and Graber, 1963; Tomialojc,
1970;Batten,1972;Emlen,1974;Guthrie,1974;Wal-
cott, 1974; Vale and Vale, 1976; Hohtola, 1978; De-
Graff and Wentworth, 1981; Jones, 1981; Beissinger
andOsborne,1982;Green,1984;Bezzel,1985;Rosen-
bergetal.,1987;Ruszczyketal.,1987).Thisisprimar-
ily due toBlair’s (1996)explicit design to explore the
avianresponsetothegradientofurbanizationwhilethe
previous work primarily examined discrete urban and
non-urban areas. This examination of the urban gra-
dient is illuminating of the processes that take place
during the urban transformation, and how avian com-
munities respond. Low levels of urban development
can increase the abundance and diversity of resources
availabletobirds.Whileurbanizationaltersplantcom-
munity composition, in some, especially arid, envi-
ronments, it may increase structural diversity (Emlen,
1974;RudnickyandMcDonnell,1989).Resourcesare
extremelydiminishedatthehighestlevelsofurbaniza-
tion in the commercial and business districts resulting
inthereductioninaviandiversityreportedinthestudies
previouslymentioned.Thisresultsfromtheremovalof
substantial areas of primary productivity, diminishing
vegetative cover, and compositional change to hardy
ornamental/exoticplantspecies(WhitneyandAdams,
1980);nativebirdsrespondpositivelywithnativeveg-
etation density, while non-native species respond pos-
itivelytoexoticplantbiomass(Millsetal.,1989).

4.2. Fragmentation
Forest area is key to predicting presence and abun-
dance of Nearctic–Neotropical migrants (Ambuel and
Temple, 1983; Askins et al., 1990). However, even
if forest size does not change with urbanization—the


Urban effects on native avifauna review

PDF file: 2006_Chace, Urban effects on native avifauna-review.pdf

doi:10.1016/j.landurbplan.2004.08.007