j. Field Ornithol., 63(4):455-461
EFFECT
OF DEVELOPMENT
COMPOSITION
OF TWO
WETLANDS
IN STATEN
CHRISTINA
ON BIRD
SPECIES
URBAN
FORESTED
ISLAND,
NEW YORK
DOWD
Division of Fish and Wildlife
New YorkStateDepartmentof EnvironmentalConservation
47-40
27 Street
LongIsland City, New York llOII USA
Abstract.--Urban developmentoftenresultsin removal,fragmentationand degradationof
woodlandhabitat.A 2-yr breedingbird censuswasconducted
at two urban, forestedwetlands
in Staten Island, Richmond County, New York, to evaluatethe influenceof development
on bird speciescomposition.
Bird speciescomposition
at the RichmondCreek site,a habitat
fragmentsurroundedby development,was dominatedby non-forest,urban-residentand
human-attractedspecies.The Latourette Park site, locatedwithin a woodedparcel in the
StatenIsland Greenbelt,supportedsignificantlymore forest-interiorand human-intolerant
species.
The two sitesdid not differ significantlyin the numberof migrantor residentbird
species.
The studyindicatesthat forestislandsin New York City canprovidenestinghabitat
for area-sensitivebird species,but developmentthat encroachesupon or degradesthese
habitatspromotesthe urbanizationof the forestbird community.
EFECTO DEL DESARROLLO URBANO EN LA COMPOSICI•)N
ESPECIES
DE
AVES
DE
DOS
ANEGADOS
DE
FORESTADOS
URBANOS EN STATEN ISLAND, NEW YORK
Sinopsis.--E1desarrollourbanoa menudoresultaen la remoci6n,fragmentaci6ny degradaci6n de habitats forestados.Pot dosaftosse condujeroncensosde avesen dos anegados
forestados
urbanosdel Condadode Richmonden StatenIsland, New York, para evaluar la
influenciadel desarrolloen la composici6n
de especies
de aves.La composici6n
de avesen
un habitat fragmentado,rodeadopot desarrollourbano en Richmond,rue dominadopot
avestipicasde lugaresno forestados,
residentes
de fireasurbanasy avesatraldaspot actividadesde humanos.Una 1ocalidaddel Parque Latourette,localizadoen una parcelaforestadade StatenIsland, alberg6un nfimerosignificativamente
mayor de avesusualesdel
interior de bosquesy de pocatoleranciahacia el humano.Las doslocalidadesestudiadas
no difierensignificativamente
en lo referentea especies
residentes
o migratorias.E1 estudio
indicaque las "islasde bosques"en la ciudadde New York puedenproveerde habitatde
anidamientopara especiesde avessensitivas,pero el desarrolloque usurpa estasfireaso
degradael habitat promuevela urbanizaci6nde la comunidadde avesde bosques.
Numerousstudieshaveshownthat forestbird populationsin suburban
and developingareas of easternNorth America have exhibited declines
in neotropical-migrantand forest-interiorspecies.These declinesresult,
in part, from the lossof forestarea aswell asfragmentationand isolation
of remainingforests(Ambueland Temple 1982, 1983; Askinsand Philbrick 1987; MacClintock et al. 1977; Whitcomb 1977; Wilcove 1988).
Lossofforesthabitatis particularlyacutein urbanareaswherethe percent
of forestareais alreadymuchreducedandwhereremainingforesthabitats
are fragmentedby residentialand commercialdevelopments.
As urban
sprawl increases,forestedareas such as parks, greenbeltsand wetlands
becomeincreasinglyimportant as habitat islandsfor forestbird species
(Adamsand Dove 1989, Tilghman 1987). These natural areasare also
455
456]
C. Dowd
j. Field
Ornithol.
Autumn
1992
subjectto increaseddisturbancefrom human activitiesand encroaching
development.
This study focusedon breeding bird populationsin two forestedwetlands in Staten Island, Richmond County, New York, New York. The
studysiteswere chosento include a relatively undisturbedarea in a large
city park and a site surroundedby residentialdevelopment.The purpose
of the studywas to documentand compareavian use of thesetwo areas
and to identify the extentto which adjacentdevelopmentmay affectthe
speciescompositionof breedingbirds. The resultsare discussed
in terms
of Staten Island biogeography.
METHODS
Study areas.--Bird specieswere monitored in two 7-ha plots. The
Latourette Park plot is located within the Staten Island Greenbelt, a
1000-ha natural area managedby the New York City Department of
Parksand Recreation.The studyarea is in a deciduousswampdominated
by red maple, Acer rubrum, spicebush, Lindera benzoin,jewelweed,Impatiens capensisand skunk cabbage,Symplocarpus
foetidus.A multibranchedstream flows through the wetland. The study area is a rectangular plot 200 m wide and 350 m long.The plot is boundedby wetland
to the north, upland woodsof oak, Quercusspp. and red maple to the
eastand by a public golf courseto the west and south.The plot boundary
varies between 5 and 50 m from the forest/golf courseedge. Approximately 1 ha of upland forest is included within the plot. The nearest
residential developmentis located400 m to the north.
The Richmond Creek plot is locatedapproximately 1200 rn southof
the Latourette
Park
site and is connected to the Greenbelt
via a wooded
stream corridor. The study area is deciduousswamp dominatedby red
maple, black willow, Salix nigra, spicebush,arrowwood, Viburnumdentatum and jewelweed. The wetland is borderedalong its length by residential developmentand is dissectedat severallocationsby pavedroads
5-6 m in width. The roads divide the study area into severalrectangles
varying from 100 to 200 m wide and totalling 750 m long. Water impoundedby roadconstruction
hasdevelopedinto a small (0.4 ha) emergent
marsh of cattail, Typha latifolia and arrow arum, Peltandra virginica.
Severalhousesand yards are includedin the plot. Destructionof wetland
vegetation has occurred from human activities including construction,
filling, power line maintenance,dumpingand dischargeof septicwastes.
Invasivespeciessuchas Phragmitescommunis,Japaneseknotweed,Polygonurncuspidatum,
Japanesehoneysuckle,
Lonicerajaponica,poisonivy,
Rhus radicans,multiflora rose, Rosa multiflora and greenbrier, Smilax
rotundifoliadominate the disturbed areas.
These two sites were chosenfor comparisonbecauseof their similar
geographicallocation, similar vegetativecharacteristicsand the common
origin of their streams.I have assumedfor the purposesof comparison
that, prior to the existingdevelopmentand alteration of the wetland, the
Richmond
Creek
site was similar
to the Latourette
Park
site and that
Vol.63,No.4
UrbanWetland
Breeding
Birds
[457
this comparisonis essentiallya beforeand after development
scenario(see
Emlen 1974).
Censusmethods.--Bird speciesnestingin eachplot were identifiedusing
the spotmappingmethod(Wakeley 1987). Censuseswere conductedin
1989 and 1990 between8 May and 29 June and between0510 and 0845
hours.The number of territorial malesor breedingpairs was determined
for all species.Each specieswas classifiedaccordingto three variables
(residentstatus,habitat preferenceand tolerancefor associationwith
humans). Each variable has three or four categoriesdefinedas follows.
(1) Status:permanentresident,neotropicalmigrant or short-distance
migrant. (2) Habitat: forestinterior, forest edgeor non-forest.(3) Human
asociation:(a) Urban: Speciescommonlyfound in urban, residentialdevelopmentsand associated
with human residences
for foodand nestsites.
Urban-residentialareasare definedas city blocks,with individualhomes
havingyardswith grass,shrubberyand streettrees.(b) Attracted:Species
that inhabit wooded, urban, residential areas and are attracted to these
areasof humanresidencefor food,nestmaterials,etc.(c) Tolerant: Species
nestingin suitablehabitat in the vicinity of human residencebut not
attractedto it becauseof the presenceof people. (d) Intolerant: Species,
usually forest-interior or area-sensitive,infrequently found nesting in
associationwith urban developments.
Specieswere classifiedby a combinationof personalobservations
and
referenceto other studies(Ambuel and Temple 1982, Askins and Philbrick 1987, Galli et al. 1976, Robbins et al. 1989, Whitcomb et al. 1981).
Specieswere assignedto the human association
categoryaccordingto their
highesttolerancefor human presence.The number of speciesin each
categorywas tallied for both sites.The speciescomposition(or number
of speciesin each category)was comparedbetweensitesusing a G test
for goodness
of fit (Sokal and Rohlf 1981) for eachof the threevariables.
RESULTS
Resultsof the 2-yr censusare summarizedin Table 1, which liststhe
species,classificationand locationfor all breedingbirds. Eight of the 30
breedingspecieswere found nestingat both sites.
Consideringboth yearstogether,19 breedingspecieswere detectedat
the LatourettePark site(Table 1). In bothyears,Gray Catbird,Dumetella
carolinensis
was the mostnumerousbreedingspeciesfollowedby House
Wren, Troglodytes
aedon,Tufted Titmouse, Parusbicolorand Red-eyed
Vireo, Vireoolivaceus.
A majority (95%) of the 19 specieswere associated
with either the forestinterior or the forestedge.Five species(27%) were
neotropicalmigrants. Nine species(47%) were consideredintolerant of
human presence.
A total of 19 breedingspecieswas observedat the Richmond Creek
sitealso.In bothyears,Red-wingedBlackbird,Agelaius
phoeniceus,
Gray
Catbird and EuropeanStarling, Sturnusvulgariswere the mostnumerous
breedingspecies.Only 58% of the specieswere associated
with either the
forest interior or the forest edge. The remaining 42% were non-forest
458]
C. Dowd
j. Field
Ornithol.
Autumn
1992
Vol.63,No.4
UrbanWetland
Breeding
Birds
[459
species.Two species(10%) were neotropicalmigrants and two species
(10%) were classifiedas intolerant of human presence.
Although the two studysiteshavethe samenumber of breedingspecies,
the speciescomposition
differssignificantlywith respectto the habitat (G
= 8.79, df = 2, P = 0.012) and human association(G = 13.36, df = 3,
P -- 0.004) variables.Overall, the Latourette Park site supporteda more
typical forestcommunitywhich was significantlylesstolerant of human
presencethan the Richmond Creek bird community. The Richmond
Creek site supportedmore non-forestspeciesand more speciesattracted
to or tolerant of human presence.The sitesdid not differ significantly
with respectto the statusvariable (G = 1.80, df = 2, P = 0.05), probably
becausepermanent residentscomprised42% of the speciesnesting at
Latourette Park and, even though Richmond Creek supportedonly two
neotropicalmigrants, 42% of all speciesnestingthere were migrants.
DISCUSSION
In a studyof StatenIslandbreedingbirds,Siebenheller(1981) reported
the extirpation of severalforestspeciesfrom the island including Blackand-white Warbler, M•otilta varia, American Redstart, Setophagaruticilla and Ovenbird, Sierusaurocapillus.Habitat lossand developmentwere
cited as probablecausesfor the lossof thesespeciesas breedingbirds.
Though they regularly occuron the island during migration, apparently
habitat of sufficientsizeor quality for establishmentof breedingterritories
doesnot exist within the remaining woodlands.Other species,such as
Wood Thrush, Hylocichlamustelir•aand Red-eyedVireo, have experienceddeclines.Sincethe publicationof that report, residentialdevelopment on Staten Island has proceededapace.
The immediate impact of developmenton Staten Island is the permanent
removal
of forest habitats
and the substitution
of residential
and
edgehabitats (see also Aldrich and Coffin 1980). Developmentalso reduceshabitat value through incidentalimpactssuchas erosion,sedimentation, pollution and depositionof debriswhich degradeadjacentwetland
and foresthabitats.At the RichmondCreek site,developmenthasresulted
in substantialfragmentationof the streamcorridor by pavedroads.The
resultingmix of fragmentedforest,disturbedareasand landscapedyards
probably doesnot provide the quality of breeding habitat required by
many forest species.For example, the Wood Thrush territory at Richmond Creek consistedof fragmentsof deciduousswamp and disturbed
woods. Parts of the territory were separatedby paved roads. Nesting
success
in sucha fragmented,urban territory is not likely for this areasensitivespecies(Andrle and Carroll 1988, Robbins et al. 1989).
At the sametime, this habitat mix is attractive to speciesthat tend to
be more tolerant of disturbanceto their nesting habitats. Red-winged
Blackbird,not typically found in deciduousswamps,was the mostabundant speciesat Richmond Creek. In 1990, four of the seventerritorial
malesusedthe 0.4-ha emergentmarsh.The otherthree usedsmall patches
460]
C. Dowd
J. Field Ornithol.
Autumn
1992
of Phragmitesgrowing in disturbedareas adjacentto road margins and
yards.Both the formationof the cattail marsh and the growth of Phragmitesare attributable to the developmentadjacentto the wetland.
Interestingly,Gray Catbird held the samenumber of territoriesat both
sites.Gray Catbirds apparentlycan toleratethe proximity of residential
developmentwhen suitably denseshrub habitat exists.At Richmond
Creek, Gray Catbird preferred areasof densevegetationwhere wetland
shrubsmixed with multiflora rose and greenbrier. In Latourette Park,
this speciespreferredareasof denseunderstoryand greenbrierin regenerating blowdownsor burned areas.
Edgeeffect.--Althoughthe two studysitesare the samesize,they have
different shapes.Due to its long, narrow shape,the RichmondCreek site
hasa higher ratio of edgeto forestinterior than the LatourettePark site.
Thus, the high proportion(42%) of non-forestspeciesat RichmondCreek
may be partly an edgeeffectdueto plot shape.This plot shape,however,
is dictatedby the development
alongthe streamcorridor.Fifty-five percent
of the 18 non-forestand forest-edgespeciesat Richmond Creek were
urban residentsor speciesattractedto human presence.At the Latourette
Park site,which is affectedby a golf courseedgeon two sides,only 27%
of the 15 non-forestand forest-edgespecieswere speciesattractedto
human presence.There were no urban speciesat the Latourette Park
site. The human-orientednature of the edgeat Richmond Creek attracts
such speciesas American Robin, Turdus migratorius,Northern Mockingbird,Mimus polyglottos
and EuropeanStarling, which are absentfrom
the Latourette Park plot.
Biogeography.--Theimportanceof forestsizefor the preservationand
managementof forestbird communitieshas been well documented(Ambuel and Temple 1982, 1983; Askins and Philbrick 1987; Galli et al.
1976; Tilghman 1987; Whitcomb 1977; Wilcove 1988). The StatenIsland
Greenbeltand contiguous,
woodedoutparcelsrepresentthe largestforest
area on Staten Island. Results from the Latourette Park breedingbird
censusindicate that this forest is large enoughto provide nestinghabitat
for somearea-sensitivespeciessuchas ScarletTanager, Piranga olivacea,
Wood Thrush, Veery, Catharusfuscescens
and Red-eyed Vireo. The
Greenbeltitself,however,is surroundedby developmentand represents
an isolatedforeston a true island.The nearestmainlandforestof any
sizeliesmorethan 50 km away in New York or New Jersey.The "island
effects"on forest birds are therefore compoundedand colonizationrates
from mainlandforestsare probablyvery low. The continuedexistenceof
the Veery and the ScarletTanager as breedingbirds on Staten Island is
consideredtenuous(Siebenheller1981). Further developmentadjacentto
the Greenbeltmay reducethe regionalforestarea to suchan extentthat
thesespeciesare also extirpated.
The findingsof this studyconfirmthat, evenin the New York City
metropolitanarea, large urban woodlandscan providenestinghabitat for
someof the neotropical-migrantspeciesthat havepopulationsin decline.
The results also suggestthat developmentthat encroachesupon and
Vol.63,No.4
UrbanWetland
Breeding
Birds
[461
degradesforest habitat promotesthe proliferation of non-forest species
and the urbanizationof the forestbird community.
ACKNOWLEDGMENTS
This studywas conductedwith the cooperationof the Office of the Greenbelt,New York
City Department of Parks and Recreation. Numerous staff of the NYSDEC provided
comment and encouragement.In particular, I thank M. Sanderson,W. Richter and T.
Lloyd-Evansfor their assistance.
LITERATURE
CITED
ADAMS, L. W., AND L. E. DOVE. 1989. Wildlife reservesand corridors in the urban
environment.Nat. Inst. Urban Wildl., Columbia, Maryland. 91 pp.
ALDRICH,J. W., AND R. W. COFFIN. 1980. Breeding bird populaionsfrom forest to
suburbiaafter thirty-sevenyears.Am. Birds 34:3-7.
AMBUEL,J. B., AND S. A. TEMPLE 1982. Songbirdpopulationsin southernWisconsin
forests:1954 and 1979. J. Field Ornithol. 53:149-158.
•,
AND•.
1983. Area-dependentchangesin the bird communitiesand vegetationof southernWisconsinforests.Ecology64:1057-1068.
ANDRLE,R. F., AND J. R. CARROLL(Eds.) 1988. The atlas of breedingbirds in New
York State. Cornell Univ. Press,Ithaca, New York. 551 pp.
ASKINS,R. A., ANDM. J. PHILBRICK.1987. Effect of changesin regionalforestabundance
on the declineand recoveryof a forestbird community.Wilson Bull. 99:7-21.
EMLEN,J.T. 1974. An urban bird communityin Tucson,Arizona:derivation,structure,
regulation. Condor 76:184-197.
GALLI, A. E., C. F. LECK, AND R. r. r. FORMAN. 1976. Avian distribution patterns in
forestislandsof different sizesin central New Jersey.Auk 93:356-364.
MACCLINTOCK,L., R. F. WHITCOMB, AND B. L. WHITCOMB. 1977. Evidence for the
value of corridorsand minimization of isolationin preservationof bioticdiversity.Am.
Birds
31:6-12.
ROBBINS,
C. S., D. K. DAWSON,ANDB. m. DOWELL. 1989. Habitat area requirementsof
breedingforestbirds of the Middle Atlantic States.Wildlife Monog. 103:1-34.
SIEBENHELLER,
N. 1981. Breedingbirds of Staten Island, 1881-1981. Staten Island Inst.
Arts and Sci., Staten Island, New York. 48 pp.
SOKAL,R. R., ANDF. J. ROHLF. 1981. Biometry. W. H. Freeman, San Francisco,California. 859 pp.
TILGHMAN,N. G. 1987. Characteristicsof urban woodlandsaffecting breeding bird
diversityand abundance.LandscapeUrban Plann. 14:481-495.
WAKELEY,
J. S. 1987. Avian territory mapping:section6.3.4, U.S. Army Corpsof EngineersWildlife ResourcesManagement Manual, Tech. Rep. EL-87-7, U.S. Army
EngineerWaterwaysExperimentStation,Vicksburg,Mississippi.13 pp.
WHITCOMB,
R.F. 1977. Islandbiogeography
and "habitatislands"of easternforest.Am.
Birds
31:3-5.
•,
C. S. ROBBINS,J. F. LYNCH, B. L. WHITCOMB, M. K. KLIMKIEWICZ,AND D.
BYSTRAK.1981. Effectsof forestfragmentationon avifauna of the easterndeciduous
forest.Pp. 125-205, in R. L. Burgess,and D. M. Sharpe,eds.Forestislanddynamics
in man-dominatedlandscapes.
SpringerVerlag, New York, New York.
WILCOVE,D. S. 1988. Changesin the avifauna of the Great Smoky Mountains: 19471983.
Wilson
Bull.
100:256-271.
Received22 Aug. 1991; accepted20 Dec. 1991.