ORIGINALARTICLE
Roleofexotictreestandson thecurrent distributionandsocial behaviourof Swainson’shawk,Buteoswainsoniinthe ArgentinePampas
Jose´Herna´nSarasola*andJuanJose´Negro
DepartmentofAppliedBiology,Estacio´n Biolo´gicadeDon˜ana,ConsejoSuperiorde InvestigacionesCient´ıficas(CSIC),Avda.de Mar´ıaLuisas/n,Pabello´ndelPeru´,Sevilla, Spain
*Correspondence:Jose´Herna´nSarasola, DepartmentofAppliedBiology,Estacio´n Biolo´gicadeDon˜ana,CSIC,Avda.deMar´ıa Luisas/n,Pabello´ndelPeru´,41013Sevilla, Spain.
E-mail:
ABSTRACT
AimTheArgentinePampaswasextensivelyandabruptlyalteredbyEuropean colonization.Between 1880and1885theindigenoushuman inhabitants were completelydisplaced,andnativegrasslands werereplacedbyexoticpastureplants and crops. One of the most important ecologicalchanges in the Pampas landscape,theintroductionoftallexotictreespecies,hasreceivedlittleattention, anditseffectonwildlifehasneverbeenassessed.Wehavemadeanintensive surveyofhabitatuseofSwainson’shawk,Buteoswainsoniinitsmostimportant non-breedingquarters,thePampasofArgentina,aimingtocharacterizethesites usedbyhawksforcommunalroosting.
LocationPampasgrasslands,Argentina.
MethodsWesurveyed30,000kmofroadsbycarduringtheaustralsummer from2001to2004,coveringthemainnon-breedingareaoccupiedbySwainson’s hawks. Theirroostsiteswerelocatedbydirectobservationofbirdsroostingin treestandsclose totheroads,bysurveyingpotentialroostsitesaround places wheregroupsofpre-andpost-roostinghawks wererecorded,andbygathering informationfromlocalfarmers.
ResultsSwainson’shawksexclusivelyused stands ofexotictree speciesfor roosting.Eucalyptusviminaliswaspresentinall34roostssurveyed,andin59%of themitwastheonlyspeciespresent.Theremainingexotictreespecieswere the Siberianelm,Ulmuspumila,pines(Pinusspp.)and cypress(Cupressusspp.). Flocksizesatroostsiteswereunusuallyhighforaraptor, withanaverageof
658individuals(range8–5000hawks,n¼27flocks).
Main conclusionsTheintroduction ofexotictreesmayhave resultedinthe expansion ofthe suitablehabitat for Swainson’shawks,permitting a recent colonizationoftheArgentine Pampas.Treestandsmayhavealsochangedthe communalroostingbehaviourofthisraptor, byvirtueoftheirprovidingnew structural elementsin aregion that almost completelylackedtreesprior to Europeanoccupation.
Keywords
Anthropogeniclandscape,ArgentinePampas,Buteoswainsoni,exotictree,roost site,socialbehaviour,Swainson’shawk.
INTRODUCTION
Forcommunallyroostingraptors,theavailabilityofsuitable roostsitesisconsidereda limitingfactorintheirnon-breeding grounds(Newton,1979).Among Neotropicalmigratorybirds,
oneoftheraptorspeciesthatwouldbepotentiallyaffectedby roost availabilityisSwainson’shawk,Buteoswainsoni.This hawkbreedsinNorthAmerica andmigratestosouthernSouth America during the boreal winter (England et al., 1997), performing a migratory trip ofc.10,000km (Fuller etal.,
1998). Althoughtheirprimaryaustralrangeislocatedinthe PampasofArgentina,Swainson’shawkshavebeenreported irregularlyinsmallnumbersinBrazil,Colombia,CostaRica, Panama,andMexico (Englandetal.,1997).
Onthebreedinggrounds,Swainson’shawksaresolitaryand territorial, withpairsoccupyinghomerangesthat leadtoa spacingof1.5–2.5kmmean distancebetweenneighbouring nests(Englandetal.,1997). Duringmigrationandwintering, thehawksbecomegregarious,huntinginflocksandroosting communally. Theshiftinspatialdistributionbetweenbreeding and non-breeding grounds islinkedto seasonalchangesin feedinghabits. The breeding-seasondiet ofthe Swainson’s hawk is similar to that of other temperate-zone buteos (Englandetal.,1997).Thenon-breedingseasondietisatypical and almost exclusivelyinsectivorous (White etal., 1989; Jaramillo,1993).InthePampas,flocksofupto12,000birds havebeenreported following grasshopper(Orthoptera) out- breaks(Goldstein,1997).
ThemajorityofSwainson’s hawkscongregateduring the australsummerinthePampas,aregionthathas been highly modifiedbyhumansduringthelastcentury.Inthisregionthe naturalgrasslands havebeenconvertedtoarablelands (Soriano, 1992;Viglizzoetal., 1997), but have also been alteredstructurallyby theintroduction ofexotictreespecies (Zalbaetal.,2000;ZalbaVillamil,2002)(Fig.1).Although habitatuseby Swainson’shawksinthisareahasrecentlybeen assessedat a landscape scale (Canavelli etal., 2003), no information existsregardingspecific characteristicsofroost sites usedbyhawksontheirnon-breedinggroundsin Argentina.Severalauthors (reviewedin White etal.,1989; but seealsoWoodbridgeetal.,1995;Goldsteinetal.,1996,
2000;Canavellietal., 2003), however,suggestthe use of introducedEucalyptusasroostsites.
Tall-tree standsarenovelinthePampasandtheir introduction hasadded structural complexityto the rather two-dimensional grasslandsof this region. Typically,every ranch house in the Pampas issurrounded byatree stand
locallycalled ‘monte’. Properties usually comprise several hundred hectares,whereasthemontescoverlessthan 10ha (Fig.1). Inthiswork,weassessthecharacteristicsofroostsites usedbySwainson’shawksthroughanintensivesurveyofthe regionconsideredtobethemainnon-breeding groundsfor thespeciesintheAmericas.Ouraimwastodeterminetheuse ofgrovesofexotictreespeciesascommunal roostsbythis species.Basedonourresults,wediscusstheimplicationsthat thestandsofexotictreemayhave inthecurrentdistribution andcommunalroostingbehaviourofthespeciesinArgentina, aswellaspossiblepastscenariosfortherelationshipbetween hawksandthePampaslandscapepriortoEuropeancoloniza- tion.
MATERIALSANDMETHODSLocatingroostsites
Between2001and2004wesurveyed30,000kmofroadsbycar incentralArgentina,includingLaPampa,Co´rdoba,Buenos AiresandSantaFeprovinces(Fig.2).Thisregion,commonly called theArgentinePampas,comprisesc. 58millionhaof moreorless fertilelandsuitableforcattlerearingandcrop production. Theclimateishumid to subhumid temperate, withannualrainfallrangingfrom600mminthesouth-westto
1100mminthenorth-east(Bukartetal.,1999).
RoostsitesusedbySwainson’shawkswerelocatedduring roadsurveysinthreeways:(1)by directobservationofhawks roostinginstandsclose toroads,(2)bysurveyingpotential roostsiteswherepre-andpost-roostinghawkswereobserved, mainlyperchedontheground and fencepostsintheearly morning and at sunset, and (3) bygathering information providedbylocalfarmers.Forsituations(1)and(2)wewere abletoestimatethenumberofhawksusingtheroostbydirect counting.Wheninformationabouttheuseof aroostsite was providedbylandowners,we corroboratedthisfindingby looking for indirect evidenceof occupancy by Swainson’s
Figure1 Aerialviewofthesouthern ArgentinePampasnearBalcarce,Buenos Airesprovince.Croplandhasreplacednative grasslandinmostofthearea,while human settlementsandtheestablishmentofthe
‘estancias’havebeenaccompaniedbythe introduction ofsmallstandsofexotictree speciesfortimber,shadeandasshelterbelts (PhotobyR.C.Leo´n,reprintedfromSoriano,
1992).
–32°
SantaFe province
Usingmappingsoftware,wecalculatedashapeindexforthe roostsitesasfollows:
Córdoba / EntreRiprovince / os / S¼L—W;
L
province
whereL isthelengthofthemajoraxiscrossingthecentreof
–34° / theroost(roostlength),andWis thelengthoftheminoraxis
alsocrossingthecentreoftheroost(roostwidth).Thisindex
rangesfrom0,whenlength andwidthoftheroostareequal,to
BuenosAires / 1,forthemaximumdifferencebetweenthesetwomeasures.
–36° / province / Wecorrelatedtheestimatednumberofhawksusingtheroost
againstroostsize,shapeindex,linealedgeandroostorienta-
tionusingnonparametric Spearman’srankcorrelationproce-
LaPampa province / dure(Zar,1996).
–38° / RESULTS
Wefoundatotalof34roostsitesusedbySwainson’shawks(3,18
km / and 13 in 2001–2002,2002–2003and 2003–2004austral
050100 / summers,respectively)inBuenosAires(n¼14),LaPampa
–40° / (n¼15)and Co´rdoba provinces(n¼5)(Fig.2).Twenty-
–64° / –62° / –60° / –58°–56° / sevenoftheseroosts werefound bydirect observation of
Figure2 ArgentinePampas(shadedarea)accordingtoViglizzo etal.(1997)andZalbaVillamil(2002).Theareasurveyedfor thisworkincludesthePampasregioncomprisingBuenosAires,La Pampa,Co´rdobaandSantaFe province,excludingEntreR´ıos. BlackpointsarethelocationsoftheSwainson’shawkroostsites foundduring2001to2004australsummers.
hawks(e.g.moultedfeathersandregurgitationpellets onthe ground).Wefollowedthesameproceduretoconfirmtheuse ofaroostwhenhawksgatherednearapotentialroostsite.
Roost-sitecharacteristics
Foreachroostsitewerecordedthepercentageofevery tree species inthegroveandthetotalareaoccupiedbythetree stand. Wedid thisbywalkingthe perimeter ofeachroost with a GlobalPositioning Systemand estimating the area usingOziExplorer3.95mapping software.Asplanted trees withinastandarealltypicallyofthesameage,themeantree heightwasobtainedfromfiverandomlyselectedtreesusing a Suunto clinometer. The trees were selectedat random points along the long axisof the roost (roost length, see below).Whentheroostwascomposedofmorethanonetree species,werandomlyselectedindividualsfromthemaintree speciesin the roost. The percentage of canopy coverwas estimated using a convexsphericaldensiometer (Lemmon,
hawks,eitherleavingorarrivingattheroostorperchingnearby,
whiletheremainingsevenwerereportedbylocalfarmers.The meansizeofroostingflocks(±SD)was658±1318birds,with valuesranging from8toanestimate 5000hawks.
Allroostsiteswerecomposedofexotictreespecies.Fifty- ninepercentoftheroostsitesweremono-specificgroves of Eucalyptusviminalis(Labill.1806),theremainingbeingmixed groves composedofvariablepercentagesofE.viminalis, Siberianelm(UlmuspumilaLinneaus1753),pine(Pinusspp.) and cypress(Cupressusspp.) (Table1).Eucalyptusviminalis wastheonlytreespeciespresentinallroostsitessampled.The density of trees and high canopy cover (Table2) also evidencedtheartificialnatureoftheroostsites.Thenumber ofSwainson’shawksusingaroostwasnegativelycorrelated withthelinearedgeofthegroves(rs ¼)0.44, t25 ¼)2.46, P¼0.02) but was not correlated either with the size (rs ¼0.20, t25 ¼1.03, P¼0.30), shape (rs ¼)0.30, t25 ¼)1.61, P¼0.11)orwiththeorientation oftheroosts (rs ¼0.09, t25 ¼0.45, P¼0.65),wheres denotestheuseof Spearmanrankcorrelationinsteadof Pearsoncorrelation coefficient.Theelongateshapeoftheroosts(Table2)reflected oneofthemainpurposesoftreeplantinginthearea,namely tocreatewindbreaks.
Table 1 Treespeciescomposition(expressedasthepercentageof eachspecies)in34treegrovesusedbySwainson’shawks for roostingintheArgentinePampas.Thenumberofroostsisinthe
bodyofthetable
1956) at five randomly located points inside the roost
selectedin the samewayasfor tree-height measurements. Finally,the total tree density was calculated by counting everytreeinthestandanddividingthisnumberbythearea occupiedbytheroost.Wealso estimatedthelineardistances fromthecentreoftheroosttothenearesthuman building andpavedroad.
1–30%31–60%60–99%100%
Eucalyptusviminalis23920
Ulmuspumila5110
Cupressusspp.6010
Pinusspp.5100
Table 2 RoostcharacteristicsofSwainson’shawksinthe
ArgentinePampas(n¼34groves)
MeanSDMinMax
Area(ha)5.75.90.426.1
Shapeindex0.760.250.170.98
Treeheight(m)20.34.210.829.0
Plantdensity(treeha)1)288.2227.786.41111.1
Canopycover(%)87.912.934.3100.0
Distancetonearestroad(m)570.7591.80.02030.0
Distancetonearestbuilding(m) 445.6486.17.01700.0
DISCUSSION
AllgrovesusedbySwainson’shawksforroostingweresmall treeplantations ofexotictree species.Themajority ofthe roostswerelocatedingrovesaroundfarmhousesandstables. Thissuggeststhat thecloseproximityofhumans and their pets,suchascatsand dogs,did not deterthe hawksfrom roostingininhabitedgroves.
Grovesizedidnotseemtobealimitingfactorforthenumber ofhawksroostingin aplantation,atleastforthe roostsizes and hawkflockswerecorded.Basedonthespatialdistributionof pellets andmoultedfeathersfoundonthegroundofthelargest roosts,weobservedthatSwainson’shawkswerenothomogen- eouslydistributedwithinthetreestandandselectedperipheral treesforperching.Thispatternwas mostevidentwherelarge grovessupportedfewer birdsthanwouldbepredictedbasedon areaoredge lengthofthehabitat.Therelationshipbetween habitatedgelengthandbirdnumberwas significantbut negativelycorrelated.Thesedatasuggesttwopossibleexplana- tions fortherelationship betweenbird abundance and tree availabilityatrootsthatarenotmutuallyexclusive.First,the hawksfavouredgesoverinteriorsofwoodlotsbecausetheyare notwelladaptedtoflyinginconfinedareas(delHoyoetal.,
1994).Second,whateverbirdsarefeedinginanareaaredrawnto theroostsite,anditishawkdensityperunit area,ratherthantree density,thatdrivesthisrelationship.Aspatialanalysisofthe spatialdensityofwoodlotsrelative tothedensityofhawksis needed toresolvethispatternfully.
TheArgentinePampaspriortoEuropeancolonization
Asinotherlowlandareasworldwide,intheArgentinePampas theareaofcultivatedlandsincreasedgreatly inthetwentieth century at the expense of natural areas, most of them grasslandsand pseudo-steppes (Viglizzoetal.,1997).Until theearly nineteenthcentury,treeswerevirtuallyabsentfrom theregion, andshrubsformedonlyaminorcomponentofthe vegetation (Darwin, [1898] 1983), to the point that the shortageoffirewoodwasamajorinconveniencetotravellersin thePampas(ZalbaVillamil, 2002).Theonlynativetrees presentintheinteriorofthePampasbefore European settlements were the ‘ombu´’ (Phytolaca dioica Linneaus
1762),whichgrewscatteredthrough thegrasslandsandthe
‘saucecolorado’(Salixhumboldtiana Willdenow1805),the
latterofwhichwas restrictedtoriparianareas.Theestablish- ment of the ‘estancias’(the local name for ranches) was accompaniedbyafforestationusingforeignspeciesfortimber, shade,and asshelterbelts(Hudson, [1918]1963;Zalba& Villamil,2002).
Exotic treestandsandthedistributionofSwainson’s hawks
Theecologyand distribution oftheSwainson’shawkinits australgroundsbeforethearrivalofEuropeancolonizers are unknown. Thefirstreferencesfoundintheliteraturetothe speciesin this region came from pioneer naturalists and ornithologistswhosurveyed theareaatthebeginningofthe twentieth century. Someof theseauthors alreadyreported flocksroostinginEucalyptustrees(Ambrosetti,1919;Delius,
1953;Olrog, 1967).None of theseauthors mentioned the adaptiveimplications ofthe useofthesesmall,man-made forestsbySwainson’shawks.Thelackofprior observations, andtherapidacceptanceofthetreesbythehawksdidnot allow these initial observers to see this as a process of behavioural adaptation.Giventhestrongrelationshipobserved between exotic tree species and roost sites selected for Swainson’shawks,severalkeyquestionsareraised.Ifthefirst exotictreestandswereplantedinthePampasc.125yearsago, whateffectshavethesesmallandman-madewoodlandsonthe ecologyofSwainson’shawksduringthenon-breedingseason? Swainson’shawksareknowntobehighlyvagile,raisingthe possibilityofrapidrangeexpansion.Goldsteinetal.(2000), forexample,found that fourout of22radio-taggedhawks wereusinganarea located500kmnorthofthesitewherethey hadbeencaptured2monthsearlierinArgentina.Thesedata suggestthat Swainson’shawkscould respond to improved resourceavailabilitybyenlargingtheireffectivenon-breeding areainafewyears.TheArgentinePampasmayhavebeen marginalareasinthedistributionofSwainson’shawksinthe past,withmostofthehawkpopulation inhabitingsavannas andsteppeswithatleastsomearborescent growth,forexample theChaco,andtheecotoneregionsamongPampasgrasslands andthexerophiticforestoftheEspinalbiometothenorth.If this were the case, the Pampas could have been rapidly
colonizedbythehawksfollowingexotictreeintroduction.
Wearenotaware ofanyotherexampleofaraptorialbird thatmayhaveincreaseditsdistribution asaresultofintro- duced plant species.Contrarily, introduced Eucalyptus tree forestshavebeenreportedtodecrease theprobabilityofthe presenceofnativecommon buzzardsButeo buteo(Linneaus
1758)insouthernSpain(BustamanteSeoane,2004).
Asforpossibleeffectsofthelargeaggregations ofSwainson’s hawksonotherspeciesofthePampas(seee.g.Terborghetal.,
2001), thelattershouldberememberedthatitisanartificial ecosysteminwhichtheoriginaldistribution andabundances ofotherspecies, whetherpossiblecompetitors orprey,have surelybeenaltered.It seems,however,that the Swainson’s hawk is the only bird speciesin the area specializingin grasshopperconsumption.
CONCLUSIONS
The current landscapeofthe ArgentinePampas comprises smallstandsofexotictreespeciesscatteredamongamatrixof crops and grasslands,most of which are ‘improved’.This landscapecontrastswiththenatural,almosttreelessgrasslands foundbythefirstEuropeancolonizers.Basedonhabitat-use studies,Swainson’shawksseemwelladaptedtotheagricul- tural settingthat resultsfrom the transformation ofnative grasslands (Canavellietal.,2003).However,akeyfeatureof thishabitatusemaybethepresence oftreesplantedtoform shelterbelts.Althoughchangesinroostingbehaviourcannot bediscounted,thepractice ofplantingEucalyptusappearsto haveenlargedtheavailablehabitatforSwainson’shawkinits australgrounds,allowing hawkstooccupyareaswherethey were probablyabsentinthepast.Currentpatternsof distribution,habitatuseandspatialaggregationofSwainson’s hawkshavecertainlybeeninfluencedbytheincorporation of exotictreestandsintheArgentinePampas.
ACKNOWLEDGEMENTS
We wishto thank MiguelSantillan,Maximiliano Galmes, Ramo´nSosaand AlejandroTravainifortheir valuablehelp in locatingand measuring roost sitesand conducting bird surveys,andtheSubsecretarı´adeEcologı´adelaProvinciade LaPampa for providing logisticalsupport to this project. We thank Rolando J. C. Leo´n for providing pictures, ElsevierCo. for authorizing photograph reproduction, and GeorgeTaylor for improvements to the Englishtext. We appreciatethecollaboration ofmanylocallandownerswho gave us permission to survey groves located on their ranches. This study received financial support from the JamesL.Baillie MemorialFund (Canada), theWildlife Conservation Society(USA),and UniversidadNacional de LaPampa(Argentina).Whiledoingthisresearch,J.H.S.was supportedby ascholarshipfromConsejoNacionalde InvestigacionesCient´ıficas yTe´cnicasde Argentina (CON- ICET).
REFERENCES
Ambrosetti,H.T.(1919)Notassobrealgunasrapaces.Hornero,
1,287–290.
Bukart, R., Ba´rbaro, N.O., Sanchez,R.O. Gomez, D.A. (1999) Eco-regionesde la Argentina.Administracio´n de ParquesNacionales,BuenosAires.
Bustamante,J. &Seoane,J. (2004)Predictingthedistribution offourspeciesofraptors(Aves:Accipitridae)insouthern Spain:statisticalmodelsworkbetter than existingmaps. Journal ofBiogeography,31,295–306.
Canavelli, S.B.,Bechard, M.J., Woodbridge, B., Kochert, M.N.,Maceda,J.J. &Zaccagnini,M.E.(2003)Habitatuse bySwainson’sHawkson their australwinteringgrounds incentralArgentina.JournalofRaptorResearch,37,125–
134.
Darwin,C.[1898](1983)ElviajedelBeagle.Guadarramaeds, Barcelona.
Delius, J.D.(1953)Algosobreellangostero.Hornero,10,80. England,A.S.,Bechard,M.J.Houston, C.S.(1997)Swain- son’sHawk(Buteoswainsoni).ThebirdsofNorthAmerica (ed.byA.Pooleand F.Gill),No.265.TheAcademyof Natural Sciences,Philadelphia,and The AmericanOrni-
thologistsUnion,Washington,DC.
Fuller,M.R.,Seegar,W.S.Schueck,L.S.(1998)Routesand travelratesofmigratingperegrinefalconsFalcoperegrinus and Swainson’shawks Buteoswainsoniin the Western Hemisphere.Journal ofAvianBiology,29,433–440.
Goldstein,M.I. (1997) Toxicologicalassessmentof a Neo- tropicalmigrantonitsnon-breedinggrounds:casestudyof the Swainson’s hawkin Argentina.MScThesis,Clemson University,Clemson,SC.
Goldstein,M.I.,Woodbridge,B.,Zaccagnini,M.E.,Canavelli, S.B.Lanusse,A.(1996)Anassessmentofmortality of Swainson’sHawks on wintering grounds in Argentina. JournalofRaptor Research,30,106–107.
Goldstein,M.I.,Bechard,M.J.,Parker,M.L.,Kochert,M.N.Lanusse,A.E.(2000)Abundance,behavior,andmortalityof Buteoswainsoninear SanFrancisco,Co´rdoba, Argentina. Hornero,15,117–121.
delHoyo,J.,Elliot,A.Sargatal,J.(1994)Handbookofthe birdsoftheworld.Vol.II:NewWorldvulturestoguineafowls. Lynxeditions,Barcelona.
Hudson, G.E.[1918](1963)Alla´lejosyhacetiempo.Peuser
Ed.,BuenosAires.
Jaramillo, A.P. (1993) Wintering Swainson’s Hawks in
Argentina: foodandage segregation. Condor,95,475–479. Lemmon,P.E.(1956)Asphericaldensiometerforestimating
forestoverstorydensity.ForestScience,2,314–320.
Newton, I.(1979)Populationecology ofraptors.T.A.D. Poyser,Vermillion,SD.
Olrog,C.C.(1967)Observacionessobreaves migratoriasdel hemisferionorte.Hornero,10,292–298.
Soriano,A.(1992)Rı´odelaPlatagrasslands.Ecosystemsofthe
World 8A. Natural grasslands:introduction and western hemisphere(ed.byR.T.Coupland), pp.367–407.Elsevier ScientificPublishingCo.,Amsterdam.
Terborgh,J.,Lopez,L.,Nun˜ez,P.,Rao,M.,Shahabuddin,G.,
Orihuela,G.,Riveros,M.,Ascanio,R.,Adler, G.H.,Lambert, T.D.Balbas,L.(2001)Ecologicalmeltdowninpredator- freeforestfragments.Science,294,1923–1926.
Viglizzo,E.F.,Roberto, Z.E.,Le´rtora, F.,LopezGay,E.
Bernardos,J. (1997)Climateandland-usechangeinfield- cropecosystemsofArgentina.AgriculturalEcosystemsand Environment,66,61–70.
White,C.M.,Boyce,D.A.& Straneck,R.(1989)Observations onButeoswainsoniinArgentina,1984,withcommentson food,habitatalterationandagriculturalchemicals. Raptors inthemodern world(ed.byB.U.MeyburgandR.D.Chan- cellor),pp.79–87.WorldWorkingGrouponBirdsofPrey andOwls,Berlin.
Woodbridge,B.,Finley,K.K. &Seager, S.T.(1995)Anin- vestigationoftheSwainson’sHawkinArgentina.Journalof Raptor Research,29,202–204.
Zalba,S.Villamil,C.B.(2002)Woodyplant invasionin relictualgrasslands.BiologicalInvasions,4,55–72.
Zalba,S.,Sonaglioni,M.I.,Compagnoni,C.A. &Belenguer,C.J. (2000)Usingahabitatmodeltoassess theriskofinvasionby anexoticplant.BiologicalConservation,93,203–208.
Zar,J.H.(1996)Biostatisticalanalysis,3rdedn.PrenticeHall, Princeton,NJ.
BIOSKETCHES
Jose´ Herna´nSarasolaisapost-graduateresearcheratDon˜anaBiologicalStation(CSIC).Hisresearchfocusesontheecologyand conservationofSwainson’shawksintheiraustralgroundsinArgentina.
JuanJose´ NegroisascientificresearcherattheDepartmentofAppliedBiologyoftheDon˜anaBiologicalStation(CSIC).Mostof hisresearchhasfocusedonbehaviouralecologyandconservationofraptorsandotherendangeredbirds.