Global Ecol Biogeogr. 2019;1–12. wileyonlinelibrary.com/journal/geb  | 1© 2019 John Wiley & Sons Ltd

Received:29May2018  |  Revised:16November2018  |  Accepted:25November2018DOI:10.1111/geb.12875

R E S E A R C H P A P E R

Large birds travel farther in homogeneous environments

Marlee A. Tucker1,2  | Olga Alexandrou3 | Richard O. Bierregaard Jr.4 |  Keith L. Bildstein5 | Katrin Böhning‐Gaese1,2 | Chloe Bracis1  | John N. Brzorad6 |  Evan R. Buechley7,8 | David Cabot9 | Justin M. Calabrese10,11 |  Carlos Carrapato12 | Andre Chiaradia13,14  | Lisa C. Davenport15,16,17 |  Sarah C. Davidson18,19  | Mark Desholm20 | Christopher R. DeSorbo21  |  Robert Domenech22 | Peter Enggist23 | William F. Fagan11 | Nina Farwig24 |  Wolfgang Fiedler19,25 | Christen H. Fleming10,11 | Alastair Franke26,27 |  John M. Fryxell28 | Clara García‐Ripollés29,30 | David Grémillet31,32 |  Larry R. Griffin33 | Roi Harel34  | Adam Kane35  | Roland Kays36,37 |  Erik Kleyheeg19,38  | Anne E. Lacy39 | Scott LaPoint19,40  | Rubén Limiñana41  |  Pascual López‐López42  | Alan D. Maccarone43 | Ugo Mellone41 |  Elizabeth K. Mojica44,45  | Ran Nathan34 | Scott H. Newman46 |  Michael J. Noonan10,11 | Steffen Oppel47  | Mark Prostor26 | Eileen C. Rees48  |  Yan Ropert‐Coudert49 | Sascha Rösner24  | Nir Sapir50 | Dana Schabo24 |  Matthias Schmidt51 | Holger Schulz23,52 | Mitra Shariati53 | Adam Shreading22 |  João Paulo Silva54,55,56 | Henrik Skov57 | Orr Spiegel58  | John Y. Takekawa59,60 |  Claire S. Teitelbaum61  | Mariëlle L. van Toor19 | Vicente Urios41 |  Javier Vidal‐Mateo41 | Qiang Wang62 | Bryan D. Watts44 | Martin Wikelski19,25 |  Kerri Wolter63 | Ramūnas Žydelis64 | Thomas Mueller1,2

1SenckenbergBiodiversityandClimateResearchCentre,SenckenbergGesellschaftfürNaturforschung,Frankfurt(Main),Germany2DepartmentofBiologicalSciences,GoetheUniversity,Frankfurt(Main),Germany3SocietyfortheProtectionofPrespa,Prespa,Greece4BiologyDepartment,UniversityofNorthCarolinaatCharlotte,Charlotte,NorthCarolina5HawkMountainSanctuary,AcopianCenterforConservationLearning,Orwigsburg,Pennsylvania6ReeseInstituteforConservationofNaturalResources,Lenoir‐RhyneUniversity,Hickory,NorthCarolina7HawkWatchInternational,SaltLakeCity,Utah8BiodiversityandConservationEcologyLaboratory,UniversityofUtah,SaltLakeCity,Utah9SchoolofBiological,EarthandEnvironmentalScience,UniversityCollegeCork,Cork,Ireland10SmithsonianConservationBiologyInstitute,NationalZoologicalPark,FrontRoyal,Virginia11DepartmentofBiology,UniversityofMaryland,CollegePark,Maryland12InstitutodaConservaçãodaNaturezaedasFlorestas,ParqueNaturaldoValedoGuadiana,CentroPolivalentedeDivulgaçãodaCasadoLanternim,Mértola,Portugal13PhillipIslandNatureParks,Victoria,Australia14SchoolofBiologicalSciences,MonashUniversity,Clayton,Australia

2  |     TUCKER ET al.

15FloridaMuseumofNaturalHistory,Gainesville,Florida16DepartmentofBiology,UniversityofFlorida,Gainesville,Florida17CollegeofScienceandEngineering,JamesCookUniversity,Cairns,Australia18DepartmentofCivil,EnvironmentalandGeodeticEngineering,TheOhioStateUniversity,Columbus,Ohio19DepartmentofMigrationandImmuno‐Ecology,MaxPlanckInstituteforOrnithology,Radolfzell,Germany20BirdLifeDenmark,Copenhagen,Denmark21BiodiversityResearchInstitute,Portland,Maine22RaptorViewResearchInstitute,Missoula,Montana23StorchSchweiz,Kreuzlingen,Switzerland24ConservationEcology,FacultyofBiology,Philipps‐UniversityMarburg,Marburg,Germany25DepartmentofBiology,UniversityofKonstanz,Konstanz,Germany26ArcticRaptorsProject,Nunavut,Canada27DepartmentofBiologicalScience,UniversityofAlberta,Edmonton,Canada28DepartmentofIntegrativeBiology,UniversityofGuelph,Guelph,Canada29VertebratesZoologyResearchGroup,UniversityofAlicante,Alicante,Spain30EnvironmentScienceandSolutionsSL,Valencia,Spain31Centred’EcologieFonctionnelleetEvolutive,UMR5175,CNRS–UniversitédeMontpellier–UniversitéPaul‐ValéryMontpellier–EPHE,Montpellier,France32PercyFitzPatrickInstituteofAfricanOrnithology,DST‐NRFCentreofExcellence,UniversityofCapeTown,Rondebosch,SouthAfrica33Wildfowl&WetlandsTrust,CaerlaverockWetlandCentre,Caerlaverock,UK34MovementEcologyLaboratory,DepartmentofEcology,EvolutionandBehavior,AlexanderSilbermanInstituteofLifeSciences,TheHebrewUniversityofJerusalem,Jerusalem,Israel35SchoolofBiologyandEnvironmentalScienceandEarthInstitute,UniversityCollegeDublin,Dublin,Ireland36NorthCarolinaMuseumofNaturalSciences,Raleigh,NorthCarolina37DepartmentofForestryandEnvironmentalResources,NorthCarolinaStateUniversity,Raleigh,NorthCarolina38Ecology&BiodiversityGroup,InstituteofEnvironmentalBiology,UtrechtUniversity,Utrecht,TheNetherlands39InternationalCraneFoundation,Baraboo,Wisconsin40Lamont‐DohertyEarthObservatory,ColumbiaUniversity,Palisades,NewYork41DepartamentodeDidácticaGeneralyDidácticasEspecíficas,UniversityofAlicante,Alicante,Spain42UniversityofValencia,CavanillesInstituteofBiodiversityandEvolutionaryBiology,TerrestrialVertebratesGroup,Valencia,Spain43BiologyDepartment,FriendsUniversity,Wichita,Kansas44CenterforConservationBiology,CollegeofWilliamandMary,VirginiaCommonwealthUniversity,Williamsburg,Virginia45EDMInternational,Inc,FortCollins,Colorado46FoodandAgricultureOrganizationoftheUnitedNations,RegionalOfficeforAfrica,Accra,Ghana47RSPBCentreforConservationScience,RoyalSocietyfortheProtectionofBirds,Cambridge,UK48Wildfowl&WetlandsTrust,Gloucester,UK49Centred’EtudesBiologiquesdeChizé,UMR7372CNRSUniversitédeLaRochelle,Villiers‐en‐Bois,France50DepartmentofEvolutionaryandEnvironmentalBiology,InstituteofEvolution,UniversityofHaifa,Haifa,Israel51BirdLifeÖsterreich,Vienna,Austria52SchulzWildlifeConsulting,Bergenhusen,Germany53FacultyofGeo‐InformationScienceandEarthObservation(ITC),UniversityofTwente,Enschede,TheNetherlands54CIBIO/InBIOAssociateLaboratory,UniversidadedoPorto,Vairão,Portugal55CentreforAppliedEcology‘Prof.BaetaNeves’/InBIOAssociateLaboratory,InstitutoSuperiordeAgronomia,UniversidadedeLisboa,Lisbon,Portugal56CentreforEcology,EvolutionandEnvironmentalChanges,FaculdadedeCiênciasdaUniversidadedeLisboa,Lisbon,Portugal57EcologyandEnvironmentDepartment,DHI,Hørsholm,Denmark58SchoolofZoology,FacultyofLifeSciences,TelAvivUniversity,TelAviv,Israel59U.S.GeologicalSurvey,WesternEcologicalResearchCenter,Vallejo,California60SuisunResourceConservationDistrict,SuisunCity,California61OdumSchoolofEcology,UniversityofGeorgia,Athens,Georgia62KeyLaboratoryofWetlandEcologyandEnvironment,NortheastInstituteofGeographyandAgroecology,ChineseAcademyofSciences,Changchun,People’sRepublicofChina63VulProNPC,Skeerpoort,SouthAfrica64OrnitelaUAB,Vilnius,Lithuania

     |  3TUCKER ET al.

1  | INTRODUC TION

Animalmovementplaysan importantrole inshapingawiderangeofecologicalphenomena,fromspeciessurvivaltoecosystemfunc‐tioning and patterns of biodiversity (Nathan et al., 2008; Viana,Santamaría,& Figuerola, 2016). As animalsmove across the land‐scape, they interactwith individualsof thesameordifferent spe‐cies(e.g.,predator–preyinteractions),carryoutecologicalfunctions(e.g., seed dispersal) and mediate processes (e.g., disease dynam‐icsandgeneflow)(Bauer&Hoye,2014).Thesearchforresourcesis one underlying driver of animal movements (La Sorte, Fink,

Hochachka,DeLong,&Kelling,2014;López‐López,García‐Ripollés,&Urios,2014),whereresourcescanbefood,water,cover,suitablebreeding habitat and access tomates. The link between resourceabundance andmovement has been found in animal home‐rangepatterns,wherehome‐rangesize,ortheareausedbyananimaltoreproduceandsurvive,decreaseswithincreasingdensityoffoodre‐sources (Koubaetal.,2017).Thespatialarrangementofresourcesand theproximityofhabitatscontainingvital resources (i.e., land‐scape complementarity) are also important factors affecting ani‐malmovements (López‐Lópezetal.,2014;Monsarratetal.,2013).For example, changes in resource distributions can lead to shifts

CorrespondenceMarleeA.Tucker,SenckenbergBiodiversityandClimateResearchCentre,SenckenbergGesellschaftfürNaturforschung,Senckenberganlage25,60325Frankfurt(Main),Germany.Email:tucker.marlee@gmail.com

Funding informationNationalTrustforScotland;PenguinFoundation;TheU.S.DepartmentofEnergy,Grant/AwardNumber:DE‐EE0005362;AustralianResearchCouncil;NASA'sArcticBorealVulnerabilityExperiment(ABoVE),Grant/AwardNumber:NNX15AV92A;NetherlandsOrganizationforScientificResearch,Grant/AwardNumber:VIDI864.10.006;BCC;NSFAward,Grant/AwardNumber:ABI‐1458748;U.K.DepartmentforEnergyandClimateChange;‘JuandelaCierva‐Incorporación’postdoctoralgrant;IrishResearchCouncil,Grant/AwardNumber:GOIPD/2015/81;DECC;GoetheInternationalPostdoctoralProgramme,PeopleProgramme(MarieCurieActions)oftheEuropeanUnion'sSeventhFrameworkProgrammeFP7/2007‐2013/underREAgrantagreementno[291776];GermanAerospaceCenterAward,Grant/AwardNumber:50JR1601;ScottishNaturalHeritage;SolwayCoastAONBSustainableDevelopmentFund;COWRIELtd.;HeritageLotteryFund;RobertBoschStiftung;NSFDivisionofBiologicalInfrastructureAward,Grant/AwardNumber:1564380;SpanishMinistryofEconomyandCompetitiveness,Grant/AwardNumber:IJCI‐2014‐19190;Energinet.dk;NASAAward,Grant/AwardNumber:NNX15AV92A;MAVAFoundation;FundaçãoparaaCiênciaeTecnologia,Grant/AwardNumber:SFRH/BPD/118635/2016;NationalKeyR&DProgramofChina,Grant/AwardNumber:2016YFC0500406;GreenFundoftheGreekMinistryofEnvironment

Editor:VéroniqueBoucher‐Lalonde

AbstractAim:Animalmovementisanimportantdeterminantofindividualsurvival,popula‐tiondynamicsandecosystemstructureandfunction.Nonetheless,itisstillunclearhowlocalmovementsarerelatedtoresourceavailabilityandthespatialarrange‐mentofresources.Usingresidentbirdspeciesandmigratorybirdspeciesoutsidethemigratoryperiod,weexaminedhowthedistributionofresourcesaffectsthemovementpatternsofbothlargeterrestrialbirds(e.g.,raptors,bustardsandhorn‐bills)andwaterbirds(e.g.,cranes,storks,ducks,geeseandflamingos).Location:Global.Time period:2003–2015.Major taxa studied:Birds.Methods:WecompiledGPStrackingdatafor386individualsacross36birdspe‐cies.Wecalculatedthestraight‐linedistancebetweenGPSlocationsofeachindi‐vidualatthe1‐hrand10‐daytime‐scales.Foreachindividualandtime‐scale,wecalculatedthemedianand0.95quantileofdisplacement.Weusedlinearmixed‐ef‐fectsmodelstoexaminetheeffectofthespatialarrangementofresources,meas‐ured as enhanced vegetation index homogeneity, on avian movements, whileaccountingformeanresourceavailability,bodymass,diet,flighttype,migratorystatusandtaxonomyandspatialautocorrelation.Results:Wefoundasignificanteffectofresourcespatialarrangementatthe1‐hrand 10‐day time‐scales. On average, individual movements were seven timeslongerinenvironmentswithhomogeneouslydistributedresourcescomparedwithareasoflowresourcehomogeneity.Contrarytopreviouswork,wefoundnosig‐nificanteffectofresourceavailability,diet, flighttype,migratorystatusorbodymassonthenon‐migratorymovementsofbirds.Main conclusions:Wesuggestthatlongermovementsinhomogeneousenviron‐mentsmightreflecttheneedfordifferenthabitattypesassociatedwithforagingandreproduction.Thishighlightsthe importanceof landscapecomplementarity,wherehabitatpatcheswithinalandscapeincludearangeofdifferent,yetcomple‐mentary resources.Ashabitathomogenization increases, itmight forcebirds totravelincreasinglylongerdistancestomeettheirdiverseneeds.

K E Y W O R D S

enhancedvegetationindex,landscapecomplementation,movementecology,productivity,spatialbehaviour,terrestrialbirds,waterbirds

4  |     TUCKER ET al.

betweenmovementstrategies(e.g.,rangeresidentversusnomadic;Reluga&Shaw,2015)andaffectthesearchbehavioursofindividu‐alswhileforaging,includingsteplengthandpathtortuosity,depend‐ing on howheterogeneously distributed the resource patches are(Smith,1974;Spiegel,Leu,Bull,&Sih,2017).

Examining the link between avianmovement and resources isimportantnotonly forbuilding abetter understandingof theun‐derlyingdriversofanimalmovement(Kleyheegetal.,2017;Nathanetal.,2008),butalsoforunderstandinghowgloballandscapemod‐ificationwill impactbirdmovementpatterns (Gilbert et al., 2016).Previousresearchonthelinkbetweenbirdmovementandresourceshas largely focused upon single populations andmigratorymove‐ments(Koubaetal.,2017;Thorupetal.,2017),withlessattentiononhownon‐migratorymovementsareimpactedbyresourcesacrossmultiplespecies.Here,weaimtoexaminehowthedistributionofre‐sourcesaffectsnon‐migratorymovementpatternsatthewithin‐day(1hr)andwithin‐season(10day)scalesacross36avianspeciesandfivecontinents.

Wepredictedshortermovementswhenresourcesarehetero‐geneouslydistributed (i.e., lowhomogeneity),becauseheteroge‐neousareasprovideadiverserangeofhabitats(includingdiverseresources)within a smaller area (Da Silveira,Niebuhr,Muylaert,Ribeiro,&Pizo,2016).Thismeansthatindividualsdonotneedtotravellongdistancestofulfilcomplementaryresourceneeds(e.g.,foragingversusreproduction).Wealsoexpectedastrongereffectofenhancedvegetationindex(EVI)homogeneityatthe1‐hrscale(i.e.,asteeperslope),becausehourlymovementsarelesslikelytoincludelongerinter‐patchmovementsfoundatthe10‐dayscale.Therefore, changes to the landscape (e.g., homogenization) thatresult inresourcesbeingfartherapartwouldresult inbirdscov‐eringlongerdistancesmorefrequentlytofindtheresourcestheyneed.

In thiswork,we focused on data‐rich, large species, includingterrestrialbirds(e.g.,raptors,hornbillsandbustards)andwaterbirds(e.g.,ducks,geese,storks,cranesandflamingos).WeusedtheEVI,whichmeasuresvegetationproductivity,asasatellite‐derivedproxyfor resources.Satellite‐basedvegetation indiceshavebeenshowntobegoodproxiesforavarietyofresourcesandhavebeenusedtopredictbirddiversitypatterns(Tuanmu&Jetz,2015)andmovement(LaSorteetal.,2014).Asameasureforthespatialarrangementofresources,weusedarecentlypublishedmetricofEVIhomogeneitythat estimates the similarity of EVI between adjacent 1‐kmpixels(Tuanmu&Jetz,2015).Withthismeasure,anylandscapeandhabitat(e.g., grasslands, forestsor agricultural lands) is consideredhomo‐geneousiftherearenochangesorfewchangesofhabitattypeatthe1‐kmscale.

In addition to thedistributionof resources,we includedothercovariates that affect avian movements, including mean resourceavailability, bodymass, diet, flight type andmigratory status.Wepredictedshorter1‐hrand10‐daymovementswhenfoodresourcesare in high abundance (i.e., high EVI), because animals can ful‐fil their requirements (e.g., foodand shelter)withina smaller area(Gilbert et al., 2016). Allometric scaling relationships have shown

that animals of greater body size usually fly farther owing to en‐ergyefficiency, increasedflightspeedsand increasedresourcere‐quirements(Alerstam,Rosén,Bäckman,Ericson,&Hellgren,2007).In addition,differences in theabundanceanddistributionof foodresourcesacrossdifferentdietcategoriesshouldtranslateintodif‐ferentmovementpatternsacrosscarnivores,herbivoresandomni‐vores(Alerstametal.,2007;Tamburello,Côté,&Dulvy,2015).Wecontrolled for thesedifferencesby includingdietasacovariate inour analysis. Finally, there are different energetic costs and flightspeedsassociatedwithflappingversussoaringflight.Flappingflightisfaster,butsoaringflightismoreenergeticallyefficient,whichgen‐erallyleadstolongerflightdistances(Hedenstrom,1993;Watanabe,2016).Forthisreason,weincludedflighttypeinouranalyses,withtheexpectation thatsoaringbirdswould fly longerdistancesovershortandlongtimeperiods.Wealsoincludedmigratorystatus(i.e.,migratoryornon‐migratory)asacovariateinourmodelstoaccountforanypotentialdifferencesinmovementdistancesacrossthetwostrategies(Alerstametal.,2007).

2  | METHODS

2.1 | Data

We compiled GPS tracking data for 36 terrestrial and freshwaterbirdspeciesbetween2003and2015,spanning4,638,594locationsacross386 individualsandfivecontinents (Figure1).ThemajorityofthedatawereobtainedfromMovebank(https://www.movebank.org/)andtheMovebankDataRepository(https://www.datareposi‐tory.movebank.org/)orweredirectlycontributedbyco‐authors(seeSupportingInformationAppendixS1).

2.2 | Movement metric

Ourmovementmetricwasdisplacement,which isthestraight‐linedistancebetweentwolocations.Wechosetoexamineavianmove‐ments at the 1‐hr and 10‐day scales because they enabled us toexamine short‐ (i.e.,within‐day) and long‐term (i.e.,within‐season)movementsandmaximizedthecontrastbetweenscaleswhilepre‐servingsufficientsamplesizesatthespeciesandindividuallevels.Tostandardizethesamplingfrequencyamongstudies,wesubsampledlocationdatasothatintervalsbetweenconsecutivelocationswereeither1hror10days.Westartedthesubsamplingalgorithmfromthefirst locationofeachindividual,andthesubsamplingprecisionwasset to the inter‐location interval±4% (e.g., for the1‐hrscale,resultingininter‐locationintervalsvaryingbetween57and62min).Thereweresomeindividualsthatdidnothavedataforboththe1‐hrand10‐dayscalesowingtothedifferenttrackingregimesofthedata,wheresomeindividualshaddataevery15min,whereasoth‐ershadonlyonelocationperday.Thisresultedinsomeindividualsnothaving the fine‐scaledata for the1‐hrscaleanalysis.Someofthe individualswere trackedwith tags thatwere switchedoff forsetperiodsoftheday(e.g.,nights)toreducebatteryuse.Toavoidany bias in the sampling at the 1‐hr time‐scale, we included only

     |  5TUCKER ET al.

locationsthatoccurredbetween06:00a.m.and6:00p.m.localtime,becausethisenabledustoincludemovementsbetweenthefeedingareaandtheroost,whileavoidingtheroostingperiodwhenbirdsarelikelytobemoresettled,particularlyinthewintermonths.Thisalsomeant thatwe includedonly birdswith diurnalmovement behav‐iours.Toexcludemigratoryperiods,we includedonlyspeciesthatwere non‐migratory (all seasons) ormigratory species outside themigratoryperiod(i.e.,summerandwintermovementsonly).Summerandwintercategorieswerebasedonmonthand latitude.SummerincludedJune,JulyandAugust(NorthernHemisphere;latitude>0)orDecember,JanuaryandFebruary(SouthernHemisphere;latitude<0).Winter includedDecember, January andFebruary (NorthernHemisphere)orJune,JulyandAugust (SouthernHemisphere).Wecategorized species as non‐migratory (n=27) or migratory (n=9)based on Eyres, Böhning‐Gaese, and Fritz (2017), who broadlycategorizedthemovementbehaviourof10,443birdspeciesasdi‐rectionalmigrant (seasonalmovementswith a specific geographi‐cal direction), dispersive migrant (seasonal movements without aspecificgeographicaldirection),nomadic(irregularmovements,notseasonal or with geographical direction) and resident (sedentarymovements).Wedefinedspeciesasmigratoryiftheywereclassifiedas ‘directionalmigratory’ or ‘dispersivemigratory’ and non‐migra‐toryifclassifiedas‘resident’.Toensurethatwedidnotincludethebeginningorendofmigrationduringthesummerorwinterforeachindividual,wecalculatedthecentroidofthedensestclusterofpointsfor each season.Clusterswere identifiedbasedon kernel densityestimation,whereaclusterisdefinedbythelocalmaximumoftheestimated density function (see Supporting InformationAppendixS2forRpackagedetails).Wethencalculatedacirclecentredontheclustercentroidwitharadiusequaltothemaximaldisplacementdis‐tancecalculated for that individual and time‐scale,withaminimalradiussizeof30kmforspecieswithveryshortmaximaldisplace‐ments.Weincludedonly locationsthatoccurredwithinthiscircle,andwedidthisforeachseasonseparatelytoavoidtracksthatexitedandre‐enteredthecircle(seeSupportingInformationAppendixS3foragraphicalrepresentationofthismethodology).

For the remaining 1‐hr and10‐day displacement data,we cal‐culated the geodesic distance between the subsampled locations.Weremovedoutliersbasedonmaximalmovementspeeds(>23m/s;Alerstam et al., 2007) and removed any stationary locations (i.e.,displacements <10m, based on averageGPS error).We removedstationarylocationsbecausewewantedtofocusonperiodswhenindividualsweremovingratherthanduringstationaryperiods,suchas roostingornesting.Wethencalculatedtworesponsevariablesforeachindividual:themediandisplacementdistanceandthe0.95quantiledisplacementdistance(i.e.,long‐distancemovements).Welog10‐transformed the displacement values tomeet the normalityassumptionofthedistributionofresidualsfromthelinearmixed‐ef‐fectsmodels.

2.3 | Environment and life history data

WeannotatedeachGPSlocationwiththemeanEVIacross2001–2012andEVIhomogeneityacross2001–2005usingpubliclyavail‐ableglobaldatasetswith1‐kmresolution(SupportingInformationAppendixS4:Hengl,Kilibarda,Carvalho‐Ribeiro,&Reuter,2015;Tuanmu&Jetz,2015).ThemeanEVIdatawerecalculatedusingmonthly MODIS EVI time‐series data (MOD13A3; Hengl et al.,2015), and the EVI homogneity data were calculated using the16‐dayMODISEVItime‐seriesdata(MOD13Q1;Tuanmu&Jetz,2015).TheEVIisamodifiedversionofthenormalizeddifferencevegetationindex(NDVI),whichisdesignedtodealwithstructuralvariations in high‐biomass regions and is able to decouple thecanopybackgroundsignalfromatmosphericinfluences(Hueteetal.,2002).ThismeansthatEVIismoresensitivetodifferencesinheavily vegetated areas (i.e., when vegetation is dense, EVI candifferentiate between different vegetation types) owing to thecorrectionforatmospherichazeandthelandsurfacebeneaththevegetation.TheEVIhomogeneitymetricwasoriginallydevelopedforexamininghowbirdspeciesrichnesswasrelatedtohabitathet‐erogeneity(seeTuanmu&Jetz,2015)andthusprovidedanidealandtesteddatasettoexaminehowhabitatheterogeneityimpacts

F I G U R E 1  Globalpatternsofenhancedvegetationindex(EVI)homogeneityspanningfromlow(darkblue)tohigh(yellow) Note.Thepinkcirclesrepresenttheaveragelongitudeandlatitudepositionforeachofthe386individualsacross36speciesincludedinthestudy −150 −100 −50 0 50 100 150

−60

−20

020

4060

80

Longitude

Latit

ude

0.20.40.60.81.0

EVI H

omog

enei

ty

6  |     TUCKER ET al.

avianmovements. The EVI homogeneity is a proxy for the spa‐tialdistributionofvegetationproductivityandreflectsfine‐grainland‐coverheterogeneity.ItiscalculatedbasedonthesimilarityofEVIvalueswithinasetneighbourhood(foradditionaldetails,seeTuanmu&Jetz,2015).TheEVIandEVIhomogeneitydataareter‐restrial‐basedmeasures,wherecellsthatincludedwaterweresetas‘NA’andwaterwas,therefore,excludedfromouranalyses.WeassumethatEVIcapturestheresourcesusedbywaterbirdsbasedonpreviouswork(Henry,Ament,&Cumming,2016),althoughwenote thatwaterbodies are also an important resource thatwerenotincludedinouranalyses.ToexaminetheaverageEVIandEVIhomogeneityexperiencedbyeachindividual,wecalculatedmeanvalues for each individual using the annotated EVI and EVI ho‐mogeneityvalues.Wealsoincludedspecies‐leveltraits,includingbodymassfromtheEltonTraits1.0database(Wilmanetal.,2014),diet[carnivore(n=20),herbivore(n=14)oromnivore(n=2)]andflighttype[soaringandflapping(n=18)orflappingonly(n=18)].Inthecaseofflighttype,soaringspeciesareabletousebothflap‐ping and soaring flight. Bodymass values ranged from600g to9.5kgandwerelog10‐transformedbeforeanalyses.

Lastly,toattempttoaccountfortheEVIandEVIhomogeneityvaluesexperiencedbyindividualswhileflying,wealsoranthemod‐els using theweightedmean values of EVI andEVI homogeneity.Weighted mean values were calculated along each displacementsegment (i.e., a straight‐line distance between two sequential lo‐cations),whereweightswerebasedon theproportionof theseg‐mentthatoccurredineachpixel.Forthefinalanalysis,weaveragedtheseweightedaverageEVIandEVIhomogeneityvaluesforeachindividual.

2.4 | Analyses

Ourfinaldatabase(seeSupportingInformationAppendicesS5andS6)includedindividualmedianand0.95quantiledisplacementval‐uesfor1‐hrand10‐daydisplacements,theassociatedmeanvaluesforbodymass,EVIandEVIhomogeneity,anddiet,flighttypeandmigratorystatuscategories.Weincludedonlyindividualsthathadtrackingdataforaminimumof1weekofhourlylocationsor60daysof10‐daylocations.Weranfourlinearmixed‐effectsmodels:twoforeachtime‐scale,onewith themedianandtheotherwith the0.95 quantile displacement distances as the dependent variable,andbodymass,EVI,EVIhomogeneity,flighttypeanddietasthefixedeffects.Weincludedanestedrandomeffecttoaccountfortaxonomy(i.e.,order/family/genus/species).Giventhatthetrack‐ing data are spatially autocorrelated, we accounted for this cor‐relationintheregressionmodelsusingaGaussianfunctionbasedonthedistancesbetweenthemeanlongitudeandlatitudeofeachindividual.Foreachmodel,wecheckedtheresidualsfornormality(i.e.,Q–Qplots).Weexaminedthecollinearityamongvariablesandfound that all correlation coefficients among the predictor vari‐ableswere |r|≤ .53,which isbelowthecommoncut‐offvalueof0.7 (Dormannetal.,2013).Wealsocheckedformulticollinearityusingvarianceinflationfactors(VIFs)andfoundthatallVIFswere

<2.0,whichisbelowthecommonlyacceptedcut‐offvalueof4.0(Zuur,Ieno,&Elphick,2010).Weexaminedthegoodness‐of‐fitforeachmodelusingthemarginalR2(varianceexplainedbythefixedeffects)andconditionalR2 (varianceexplainedbybothfixedandrandom factors) values for eachmodel (Nakagawa& Schielzeth,2013).Wecalculatedthemodelpredictionsusingthemeanvalueofthecontinuouspredictors (e.g.,massandEVI)andvaryingthecovariate of interest (e.g., EVI homogeneity).We chose tomakepredictions for carnivorous soaringmigrants because this is thepredominant combination inourdata.We tested fordifferencesbetween the slope estimates for EVI homogeneity for the 1‐hrmodels,the10‐daymodelsandbetweenthe1‐hrand10‐daymod‐els.We did this using the difference between EVI homogeneitycoefficientestimatesandtheassociatedconfidenceintervalscal‐culated via error propagation basedonClark (2007: see chapter5.6.2andappendixD.5.3).TheEVIhomogeneityslopeestimatesweredeemednot significantwhen the95%confidence intervalsoverlappedzero.AllanalyseswereperformedinRv.3.4.3(RCoreTeam,2017),anddetailson theRpackagesused in theanalysescanbefoundintheSupportingInformation(AppendixS2).

3  | RESULTS

Wefoundasignificantpositiverelationshipbetweendisplacementand EVI homogeneity at both the 1‐hr and 10‐day time‐scales(Table1;Figures2and3).TheresultsweresimilarfortheweightedmeanEVIandEVIhomogeneityanalyses(SupportingInformationAppendixS7).Onaverage,displacementswereuptoseventimeslongerinareaswithhighEVIhomogeneity(Figure2),suchasde‐sert regions (the maximal EVI homogeneity value was .85). Forexample,modelpredictionsfor1‐hrmediandisplacementsforcar‐nivoroussoaringindividualswere1.02km(±SE1.63km,range=0.62–1.65km,n=168) in areas of high EVI homogeneity versus0.14km(±SE1.47km,range=0.10–0.21km,n=168)inareasoflowEVIhomogeneity(Figure2a).The1‐hrlong‐distancedisplace‐ments for carnivorous soaring individuals were 10.20km (±SE 1.57km,range=6.48–16.07km,n = 168)inareasofhighEVIho‐mogeneity versus2.40km (±SE 1.45km, range=1.66–3.48km,n=168)inareasoflowEVIhomogeneity(Figure2a).

Therewas no significant difference between the slope coeffi‐cients of the 1‐hr and 10‐day displacements for both themedianand long‐distance models (Supporting Information Appendix S8).Contrarytoourpredictions,theseresultssuggestthatmovementsatbothtime‐scaleswereequallysensitivetodecreasinghomogeneity.

Ourmodelsexplained52–71%ofthevariationinaviandisplace‐mentsatthe1‐hrand10‐daytime‐scaleswhenaccountingforbothrandom and fixed effects, and 10–38% of the variationwhen ac‐countingforthefixedeffectsalone(i.e.,bodymass,meanEVI,EVIhomogeneity,diet,flighttypeandmigratorystatus;Table1).WedidnotfindanysignificanteffectsofmeanEVI,bodymass,diet,flighttypeormigratorystatusonmedianorlong‐distancedisplacementsateithertime‐scale(Table1;Figure3).

     |  7TUCKER ET al.

4  | DISCUSSION

WehaveshownthatEVIhomogeneityisakeyfactorassociatedwithavianmovements,wheremovementswereonaverageseventimeslonger in areas of high EVI homogeneity (e.g., deserts) comparedwith areas of low EVI homogeneity (e.g., mixed broadleaved andneedle‐leavedforests).Theincreaseindisplacementwithincreasinghomogeneityislikelytobeareflectionofthedifferenthabitattypes(including microhabitat heterogeneity) required for survival (e.g.,

foodresourcesortreecoverforpredatoravoidance)andreproduc‐tion (e.g., nesting sites). Somebird species (e.g., upland sandpiper,Bartramia longicauda)havelargerhome‐rangesizesinhomogeneousenvironments,suchaspasturesorgrasslands,becausethestructureof thesehabitatsdoesnotmeetallof thebiological requirementsof the bird, meaning that they increase their ranging behaviouruntiltheirrequirementsaremet (Sandercocketal.,2015;Stanton,Kesler,&Thompson,2014).Therefore,landscapecomplementation,where a single landscape includes habitat patches with different

TA B L E 1  Modelcoefficients,R2,p‐valuesandsamplesizesoflinearmixed‐effectsmodelspredictingthemedianand0.95quantileofindividualdisplacementsfor1‐and10‐daytime‐scales

1 hr 10 days

Median 0.95 quantile Median 0.95 quantile

Estimate (SE) p Estimate (SE) p Estimate (SE) p Estimate (SE) p

Mass 0.385(0.265) .283 0.175(0.174) .419 0.155(0.237) .532 −0.427(0.264) .145

EVI −0.58(0.436) .185 −0.053(0.328) .872 −0.225(0.409) .582 0.795(0.484) .102

EVI_Homogeneity 1.198 (0.323) < .001 0.881 (0.23) < .001 2.427 (0.311) < .001 2.292 (0.434) < .001

Diet(H) 0.088(0.33) .807 −0.065(0.272) .827 0.056(0.302) .857 0.017(0.403) .968

Diet(O) 0.129(0.56) .833 −0.654(0.395) .196 −0.359(0.459) .456 −0.908(0.553) .139

FlightT_Soar 0.469(0.32) .281 0.195(0.224) .476 0.123(0.315) .723 −0.202(0.419) .663

MigStatus_NM 0.231(0.148) .259 0.213(0.099) .164 0.252(0.195) .232 0.082(0.206) .699

R2 marginal .376 .360 .261 .102

R2conditional .696 .706 .518 .566

Species 19 35

Individuals 168 356

Note.Predictorvariables includedfixedeffectsforbodymass (Mass),enhancedvegetation index (EVI),EVIhomogeneity (EVI_Homogeneity),diet(H=herbivoreandO=omnivorecoefficients),flighttype(FlightT;soaringcoefficientvaluesshownhere)andmigratorystatus(MigStatus_NM;non‐migratorycoefficientvaluesshownhere).Themodelalsoincludedanestedrandomeffectaccountingforthetaxonomy,andaGaussianspatialauto‐correlationstructure.Boldvaluesindicatesignificanceatp<.05.

F I G U R E 2  Avian(a)1‐hrand(b)10‐daymedian(0.5quantile;yellow)andlong‐distance(0.95quantile;purple)displacementswithincreasingenhancedvegetationindex(EVI)homogeneity Note.Plotsincluderegressionlinesfromthelinearmixed‐effectsmodelsand95%confidenceintervals.AnEVIhomogeneityvalueofzeroindicatesareasoflowhomogeneity,andvaluesof0.8representareasofhighhomogeneityatalocalscale

(a) (b)

8  |     TUCKER ET al.

butcomplementaryresourceswithincloseproximity,islikelytobeanimportantfeatureforshapingavianmovements(Mueller,Selva,Pugacewicz,&Prins, 2009). The linkbetweenmovement andEVIhomogeneitymightalsosuggestthatitisimportanttomaintainland‐scapecomplementarity inhuman‐modifiedareasthathaveshiftedfromheterogeneous tohomogeneous landscapes (e.g., croplands),whichmightreducethedistancescoveredbyindividualbirdsand,inturn,thepotentialnegativeeffectsoftheselongertraveldistances(e.g.,increasedenergeticcosts).

Interestingly, we did not find a significant effect of themeanabundanceof resources, contrary toourpredictions andpreviousresearchonsinglepopulationsofbirds(Dodgeetal.,2014;Stantonetal.,2014).Thisdifferencecouldbeattributabletopreviousstud‐iesfocusingonlong‐distancemovements,suchasmigration,ornotincluding the effect of spatial arrangement of resources, or both.WecanruleoutthepossibilityofspatialarrangementofresourcesmaskingtheeffectofEVI,becauseweranourmodelsexcludingEVIhomogeneityandstillfoundnosignificanteffectofEVI(SupportingInformation Appendix S9). Although vegetation indices, such asEVI, have been shown to underlie bird behaviour (La Sorte, Fink,Hochachka,DeLong,&Kelling,2013)anddiversitypatterns(Tuanmu&Jetz,2015),itmayalsobethecasethatmeanEVIisnotthebestproxyofresourcesusedbybirds,particularlyonasmallscale(e.g.,dailymovements).Itisassumedthatvegetationindicesprovidein‐formationacrossseveraldietcategories;however, theymightper‐formpoorly fornon‐herbivore species, specifically those that relyonscavenging.Wealsoranourmodelswithaninteractiontermbe‐tweenmeanEVIanddiettotestfordifferencesintheresponsetoEVIacrossdietcategories (Supporting InformationAppendixS10).Theinteractiontermwassignificantonlyforthelong‐distance10‐day displacements, suggesting thatwewere unable to detect dif‐ferencesbetweendietcategoriesforhourlymovementsusingEVI

ata1‐kmresolution.Wealsonotethatwedidnotaccountfortheseasonalvariation in resourceavailability,whichmay impactavianmovements. Our study focused on terrestrial resources that arelikely to capture some of the resources used by waterbirds (e.g.,crops),butfuturestudiesshouldinvestigatetheroleofaquaticre‐sources onwaterbirdmovements.Overall, productivitymeasures,suchasEVI,arecurrentlythebestproxyforfoodresourcesavail‐able,andourresultsindicatethatEVIhomogeneityisapotentiallyusefulproxyofthespatialarrangementofresourcesandhasanim‐portantroleinshapingavianforagingmovements.

Alsocontrarytoourpredictions,wedidnotfindasignificanteffectofbodymassondisplacements.Thelackofrelationshipbe‐tweendisplacementandmasscouldalsobearesultofthelimitedrangeofbodymass included inourdatabase, spanning600g to9.5kg,andthelowsamplesizeofsmallbirdsincludedinourstudy.This isbecauseof the limitedavailabilityofhigh‐resolutiondatafor terrestrialbirds<250g,owing to theweightofcurrentGPStracking technologiesandthe limitedbattery life forsmallerde‐vices(López‐López,2016).Basedonallometricrelationships,birdswithsmallerbodymasses(e.g.,<600g)shouldtravelshorterdis‐tancesanduseasmallerareabasedonreducedresourcerequire‐ments,energyefficiencyandflightspeedsincomparisontolargerspecies(Alerstametal.,2007).Astrackingtechnologiesimprove,itwillbecomepossibletotracksmallerspeciesandthenre‐exam‐inethisrelationshipacrossabroaderrangeofavianbodymass.

Lastly,wedidnotfindanysignificantdifferencesbetweensoar‐ing/flapping flightand flapping‐only flight. It ispossible that flightstrategyhasasmallerimpactonforagingmovementscomparedwithmigratorymovements,wherethetrade‐offbetweenflightdistanceandenergeticcostsisgreater(Hedenstrom,1993;Watanabe,2016).Alternatively,itcouldbethatflightbehaviours,suchasthermalsoar‐ing,were not captured at the temporal resolution of the tracking

F I G U R E 3  Modelcoefficients(±95%confidenceintervals)oflinearmixed‐effectsmodelspredictingaviandisplacementsusing:(a)bodymass;(b)meanenhancedvegetationindex(EVI);(c)EVIhomogeneity;(d)flighttype(soaring);and(e)migratorystatus(non‐migratory) Note.Modelswererunforthemedian(yellow)andlong‐distance(0.95quantile;purple)displacementsofeachindividual,calculatedacrossdifferenttime‐scales.Whentheerrorbarscrossthehorizontalline,theeffectisnotsignificant.SeeTable1fordetails

(a)

(d) (e)

(b) (c)

Time-scale Time-scale Time-scale

Time-scale Time-scale

     |  9TUCKER ET al.

datausedandourrestrictiontoexaminingtwo‐dimensionalmove‐ments(i.e.,onlylongitudeandlatitude).Thismeansthatindividualsthatusethermalsoaringcouldbecoveringlongerdistancesthatwearenotable todetectwithourcurrentanalysis (Tamburelloetal.,2015).Owing to thedisproportionate increase in flight costswithbodymassforflappingflyers,flappingflightismorecommoninsmallspecies(Hedenstrom,1993),andwiththeinclusionofthesespecieswe might see more divergent displacement behaviours betweentheseflyingstrategies.Itisalsopossiblethatthesizeofthesmallerbirdsinourdatasetthatarecharacterizedasactivefliers(i.e.,flap‐pingflight)usethisstrategyforonlyshortperiodsbecausetheyarestill too largetomaintain this flightstrategyenergetically for longperiods, thuspreventingus fromdetectinganydifferencesamongstrategiesinouranalysis.

Therandomeffect (i.e., taxonomy)explaineda largeportionofthe variance in avian movements (c.40–50%). Previous work hasexaminedspecies‐leveldifferencesinmovementpatterns,includingdifferences inhome‐rangesize (Haskell,Ritchie,&Olff,2002)andmigrationdistances/strategies (Alerstam,Hedenström,&Åkesson,2003;LaSorteetal.,2013)basedonspecies‐leveltraits(e.g.,bodysize and diet). Someof the variation among individualswithin thesame species is probably attributable to sex, because males andfemales have different movement patterns during brood rearing(Hernández‐Pliego, Rodríguez, & Bustamante, 2017). In addition,feather moult (i.e., feathers being shed and regrown) may impactavian movements, including periods of flightlessness (e.g., cranesand waterfowl post‐breeding) and reduced aerodynamic perfor‐mance of the wings (e.g., Falco peregrinus; Flint &Meixell, 2017).Variation inmoultpatternsandtheirconsequencesforbirdmove‐ment between species, populations and individualswere not con‐sideredhereowingtolackofdetailedmoultdatawhenmovementwas recorded.Reproduction isanothervitalpartofan individual’slifehistoryandofteninvolvesashiftinmovementpatternsowingtothedistributionofmates,lekkingsitesoravailabilityofnestingsitesorfoodresources(Cecere,Gaibani,&Imperio,2014;Rösner,Brandl,Segelbacher,Lorenc,&Müller,2014).Otherenvironmentalvariables,suchaswindspeedanddirection,werenotincludedinouranalyses,butmightalsoaccountforsomeoftheunexplainedvarianceofourmodels(Harel,Horvitz,&Nathan,2016;Melloneetal.,2015).

Anotherpotentialfactoraccountingforthewithin‐speciesvari‐ation in avian displacements is related animal personality, whereindividuals with different personalities are likely to differ in theirmovementstrategies(Patrick,Pinaud,&Weimerskirch,2017;Spiegeletal.,2017).Forexample,movementpatternsareexpectedtodifferaccordingtotheboldnessofindividuals,becausebolderindividualsmaydemonstratemoreexploratorymovementsandusemoreriskyenvironments(Spiegeletal.,2017).Thiscouldalsoberelatedtoageandexperience,becauseindividualswithmoreexperiencemightbelesslikelytoinhabitriskyenvironmentsandmightalreadyhaveiden‐tifiedwhere the reliable foodpatches are, contributing further tointraspecificvariation(López‐Lópezetal.,2014).

Acaveatofouranalysis is theassumption thatourcalculationof the EVI and EVI homogeneity values based on endpoints of

displacementsrepresentthemeanresourcesorresourcehomoge‐neity experiencedby the individualwhilemoving. In this context,without high‐resolution data collected over long durations, it willbedifficulttodiscernexactlywhattheindividualexperiencedoverextendedperiods.Nevertheless,our resultsclearlydemonstratearelationshipbetweenresourcesandavianmovements,becausewefoundsimilarresultsusingmodelsbasedonlyontheendcoordinatesof displacement segments and models using the weighted meanalong the entire straight‐line displacement segments (SupportingInformation Appendix S7). As higher‐resolution tracking data be‐comemore common, future studies canbegin to discern foragingbehavioursfrommovementtracksandexamineforagingpatternsinresponsetoresourcesatamacroecologicalscale.

5  | CONCLUSION

In conclusion, our study is the first to examine the relationshipbetween the distribution of resources and non‐migratory avianmovement patterns acrossmultiple species and regions.Wehavedemonstrated the importance of resource spatial distribution onshapingmovements,highlightingthepossibleeffectsof landscapehomogenization,whereindividualsmayneedtoflyfarthertomeettheirecologicalrequirements. It ispossiblethatcontinuinghabitathomogenization (e.g., intensification of agriculture) in landscapeswith a naturally high diversity of habitats will have negative im‐pacts on the abundance and diversity of birds (Jerrentrup et al.,2017)owingtothe lossofcomplementaryhabitats.Thismight, inturn, result in greatermovement requirements and higher energyexpenditure.

ACKNOWLEDG MENTS

The authors are grateful for financial and logistical support fromthe Robert Bosch Foundation, Goethe International PostdoctoralProgramme,PeopleProgramme(MarieCurieActions)oftheEuropeanUnion’sSeventhFrameworkProgrammeFP7/2007‐2013/underREAgrantagreementno[291776],‘JuandelaCierva—Incorporación’post‐doctoral grant, Spanish Ministry of Economy and Competitiveness(IJCI‐2014‐19190), 3M Gives, Cowrie Ltd, U.K. Department forEnergyandClimateChange,EnvironmentalImpactAssessmentstudyforKriegersFlakoffshorewind farm fundedbyEnerginet.dk,HawkMountainSanctuary,IrishResearchCouncilGOIPD/2015/81,MAVAFoundation, Green Fund of the Greek Ministry of Environment,MinervaCenterforMovementEcology,NASAAwardNNX15AV92A,NSF Division of Biological Infrastructure Award 1564380, GermanAerospace Center Award 50JR1601, Netherlands Organization forScientific Research grant number VIDI 864.10.006, NSF AwardABI‐1458748, Penguin Foundation, Australian Research Council,SolwayCoastAONBSustainableDevelopmentFund,ScottishNaturalHeritage, BBC, National Trust for Scotland, Heritage Lottery Fund,DECC, U.K. Department for Energy and Climate Change, NASA’sArcticBorealVulnerabilityExperiment(ABoVE)(#NNX15AV92A),the

10  |     TUCKER ET al.

NationalKeyR&DProgramofChina(2016YFC0500406),TheBaileyWildlifeFoundation,TheNatureConservancy,BiodiversityResearchInstitute, the Bluestone Foundation and The U.S. Department ofEnergy (DE‐EE0005362),U.S.ArmyAberdeenProvingGround,TheCenterforConservationBiology,ArcticNet,GovernmentofNunavut,NunavutWildlifeManagementBoard,CanadianCircumpolarInstitute,The MPG Ranch andWhooping Crane Eastern Partnership (www.bringbackthecranes.org). J.P.S. was supported by a postdoctoralscholarship fromFCT (FundaçãoparaaCiênciae tecnologia;SFRH/BPD/118635/2016).ThankyoutoMitchWeegman,EdBurrell,JohnSkilling,ArthurThirlwellandCarlMitchell forassistancewithcatch‐inganddatadownloadfortheGreenlandwhite‐frontedgeese,RichardHesketh for the Svalbard barnacle geese and AlynWalsh, MauriceCassidy,RichardHeskethfortheGreenlandbarnaclegeese.Movebankis hosted by the Max Planck Institute for Ornithology, and theMovebankDataRepositoryishostedbytheUniversityofKonstanz.

AUTHOR CONTRIBUTIONS

M.A.T. and T.M. conceived themanuscript,M.A.T. conducted theanalyses,andM.A.T.andT.M.wrotethefirstmanuscriptdraft.Co‐authorscontributeddatasetsandassistedwithwritingthefinalver‐sionofthemanuscript.

DATA ACCE SSIBILIT Y

Thedatausedinthisstudyareavailableatdatadryad.org(https://doi.org/10.5061/dryad.jc5616h).Mostoftheanimalmovementdataoriginatefromandarepublicallyavailableatwww.movebank.org.

ORCID

Marlee A. Tucker https://orcid.org/0000‐0001‐7535‐3431

Chloe Bracis https://orcid.org/0000‐0003‐4058‐7638

Andre Chiaradia https://orcid.org/0000‐0002‐6178‐4211

Sarah C. Davidson https://orcid.org/0000‐0002‐2766‐9201

Christopher R. DeSorbo https://orcid.org/0000‐0002‐2096‐3176

Roi Harel https://orcid.org/0000‐0002‐9733‐8643

Adam Kane https://orcid.org/0000‐0002‐2830‐5338

Erik Kleyheeg https://orcid.org/0000‐0001‐8026‐3887

Scott LaPoint https://orcid.org/0000‐0002‐5499‐6777

Rubén Limiñana https://orcid.org/0000‐0001‐8152‐3644

Pascual López‐López https://orcid.org/0000‐0001‐5269‐652X

Elizabeth K. Mojica https://orcid.org/0000‐0001‐6941‐4840

Steffen Oppel https://orcid.org/0000‐0002‐8220‐3789

Eileen C. Rees https://orcid.org/0000‐0002‐2247‐3269

Sascha Rösner https://orcid.org/0000‐0002‐6766‐1546

Orr Spiegel https://orcid.org/0000‐0001‐8941‐3175

Claire S. Teitelbaum https://orcid.org/0000‐0001‐5646‐3184

R E FE R E N C E S

Alerstam,T.,Hedenström,A.,&Åkesson,S.(2003).Long‐distancemigra‐tion:Evolutionanddeterminants.Oikos,103, 247–260.https://doi.org/10.1034/j.1600‐0706.2003.12559.x

Alerstam, T., Rosén,M., Bäckman, J., Ericson, P. G. P., & Hellgren, O.(2007). Flight speeds amongbird species:Allometric andphyloge‐neticeffects.PLoS Biology,5,e197.https://doi.org/10.1371/journal.pbio.0050197

Bauer,S.,&Hoye,B.J.(2014).Migratoryanimalscouplebiodiversityandecosystemfunctioningworldwide.Science,344,1242552–1242552.https://doi.org/10.1126/science.1242552

Cecere,J.G.,Gaibani,G.,&Imperio,S.(2014).Effectsofenvironmentalvariabilityandoffspringgrowthonthemovementecologyofbreed‐ing Scopoli’s shearwaterCalonectris diomedea. Current Zoology,60,622–630.https://doi.org/10.1093/czoolo/60.5.622

Clark, J. S. (2007).Models for ecological data. Princeton,NJ: PrincetonUniversityPress.

DaSilveira,N.S.,Niebuhr,B.B.S.,Muylaert,R.deL.,Ribeiro,M.C.,&Pizo,M.A.(2016).Effectsoflandcoveronthemovementoffrugiv‐orousbirdsinaheterogeneouslandscape.PLoS ONE,11,e0156688.https://doi.org/10.1371/journal.pone.0156688

Dodge,S.,Bohrer,G.,Bildstein,K.,Davidson,S.C.,Weinzierl,R.,Bechard,M.J.,…Wikelski,M. (2014).Environmentaldriversofvariability inthemovementecologyofturkeyvultures (Cathartes aura) inNorthand South America. Philosophical Transactions of the Royal Society B: Biological Sciences, 369, 20130195. https://doi.org/10.1098/rstb.2013.0195

Dormann, C. F., Elith, J., Bacher, S., Buchmann, C., Carl, G.,Carré, G., … Leitão, P. J. (2013). Collinearity: A review ofmethods to deal with it and a simulation study evaluat‐ing their performance. Ecography, 36, 27–46. https://doi.org/10.1111/j.1600‐0587.2012.07348.x

Eyres, A., Böhning‐Gaese, K., & Fritz, S. A. (2017). Quantification ofclimaticniches inbirds:Adding the temporaldimension. Journal of Avian Biology,48,1517–1531.https://doi.org/10.1111/jav.01308

Flint, P. L., & Meixell, B. W. (2017). Movements and habitat use ofwhite‐fronted geese (Anser albifrons frontalis) during the remigialmolt in Arctic Alaska, USA.Waterbirds, 40, 272–281. https://doi.org/10.1675/063.040.0308

Gilbert,N.I.,Correia,R.A.,Silva,J.P.,Pacheco,C.,Catry, I.,Atkinson,P.W.,…Franco,A.M.A.(2016).Arewhitestorksaddictedtojunkfood? Impacts of landfill use on the movement and behaviourof resident white storks (Ciconia ciconia) from a partially migra‐tory population.Movement Ecology, 4, 7. https://doi.org/10.1186/s40462‐016‐0070‐0

Harel, R.,Horvitz,N.,&Nathan,R. (2016).Adult vultures outperformjuvenilesinchallengingthermalsoaringconditions.Scientific Reports,6,27865.https://doi.org/10.1038/srep27865

Haskell, J. P., Ritchie, M. E., & Olff, H. (2002). Fractal geometry pre‐dicts varying body size scaling relationships for mammal and birdhome ranges. Ecology, 418, 527–530. https://doi.org/10.1038/nature00840

Hedenstrom,A. (1993).Migrationbysoaringorflappingflight inbirds:The relative importance of energy cost and speed. Philosophical Transactions of the Royal Society B: Biological Sciences,342,353–361.https://doi.org/10.1098/rstb.1993.0164

Hengl,T.,Kilibarda,M.,Carvalho‐Ribeiro,E.D.,&Reuter,H. I. (2015).Worldgrids—ApublicrepositoryandaWPSforglobalenvironmentallayers.WorldGrids.Retrieved fromhttp://worldgrids.org/doku.php/wiki:evmmod3(AccessedApril2017)

Henry,D.A.W.,Ament,J.M.,&Cumming,G.S. (2016).Exploringtheenvironmental drivers of waterfowl movement in arid landscapesusingfirst‐passagetimeanalysis.Movement Ecology,4,8.https://doi.org/10.1186/s40462‐016‐0073‐x

     |  11TUCKER ET al.

Hernández‐Pliego,J.,Rodríguez,C.,&Bustamante,J.(2017).Afewlongversus many short foraging trips: Different foraging strategies oflesserkestrelsexesduringbreeding.Movement Ecology,5,8.https://doi.org/10.1186/s40462‐017‐0100‐6

Huete,A.,Didan,K.,Miura,T.,Rodriguez,E.P.,Gao,X.,&Ferreira, L.G.(2002).OverviewoftheradiometricandbiophysicalperformanceoftheMODISvegetationindices.Remote Sensing of Environment,83,195–213.https://doi.org/10.1016/S0034‐4257(02)00096‐2

Jerrentrup, J. S., Dauber, J., Strohbach, M. W., Mecke, S., Mitschke,A.,Ludwig,J.,&Klimek,S. (2017). Impactofrecentchanges inag‐ricultural land use on farmland bird trends.Agriculture, Ecosystems & Environment, 239, 334–341. https://doi.org/10.1016/j.agee.2017.01.041

Kleyheeg,E.,vanDijk,J.G.B.,Tsopoglou‐Gkina,D.,Woud,T.Y.,Boonstra,D.K.,Nolet,B.A.,&Soons,M.B. (2017).Movementpatternsofakeystonewaterbird species are highly predictable from landscapeconfiguration. Movement Ecology, 5, 2. https://doi.org/10.1186/s40462‐016‐0092‐7

Kouba, M., Bartoš, L., Tomášek, V., Popelková, A., Šťastný, K., &Zárybnická,M. (2017).Home range sizeofTengmalm’sowlduringbreedinginCentralEuropeisdeterminedbypreyabundance.PLoS ONE,12,e0177314.https://doi.org/10.1371/journal.pone.0177314

LaSorte,F.A.,Fink,D.,Hochachka,W.M.,DeLong,J.P.,&Kelling,S.(2013).Population‐levelscalingofavianmigrationspeedwithbodysize andmigration distance for powered fliers.Ecology,94, 1839–1847.https://doi.org/12‐1768.1

LaSorte,F.A.,Fink,D.,Hochachka,W.M.,DeLong,J.P.,&Kelling,S.(2014). Spring phenology of ecological productivity contributes tothe use of loopedmigration strategies by birds.Proceedings of the Royal Society B: Biological Sciences, 281, 20140984. https://doi.org/10.1098/rspb.2014.0984

López‐López,P.(2016).Individual‐basedtrackingsystemsinornithology:Welcome to theeraofbigdata.Ardeola,63, 103–136.https://doi.org/10.13157/arla.63.1.2016.rp5

López‐López,P.,García‐Ripollés,C.,&Urios,V.(2014).Foodpredictabil‐ity determines space use of endangered vultures: Implications formanagement of supplementary feeding.Ecological Applications,24,938–949.https://doi.org/10.1890/13‐2000.1

Mellone,U.,De La Puente, J., López‐López, P., Limiñana, R., Bermejo,A.,&Urios,V.(2015).Seasonaldifferencesinmigrationpatternsofasoaringbirdinrelationtoenvironmentalconditions:Amulti‐scaleapproach.Behavioral Ecology and Sociobiology,69,75–82.https://doi.org/10.1007/s00265‐014‐1818‐4

Monsarrat, S., Benhamou, S., Sarrazin, F., Bessa‐Gomes, C., Bouten,W.,&Duriez,O. (2013).Howpredictability of feedingpatches af‐fects home range and foraging habitat selection in avian socialscavengers?PLoS ONE,8, e53077.https://doi.org/10.1371/journal.pone.0053077

Mueller, T., Selva, N., Pugacewicz, E., & Prins, E. (2009). Scale‐sensi‐tive landscape complementation determines habitat suitabilityfor a territorial generalist. Ecography, 32, 345–353. https://doi.org/10.1111/j.1600‐0587.2008.05694.x

Nakagawa, S., & Schielzeth, H. (2013). A general and simple methodfor obtaining R2 from generalized linear mixed‐effects mod‐els. Methods in Ecology and Evolution, 4, 133–142. https://doi.org/10.1111/j.2041‐210x.2012.00261.x

Nathan,R.,Getz,W.M.,Revilla,E.,Holyoak,M.,Kadmon,R.,Saltz,D.,&Smouse,P.E.(2008).Amovementecologyparadigmforunifyingorganismalmovementresearch.Proceedings of the National Academy of Sciences of the USA,105,19052–19059.https://doi.org/10.1073/pnas.0800375105

Patrick,S.C.,Pinaud,D.,&Weimerskirch,H.(2017).Boldnesspredictsan individual’s position along an exploration‐exploitation forag‐ingtrade‐off.Journal of Animal Ecology,86,1257–1268.https://doi.org/10.1111/1365‐2656.12724

RCoreTeam. (2017).R: A language and environment for statistical com‐puting. Vienna, Austria: R Foundation for Statistical Computing.Retrievedfromhttps://www.R‐project.org/

Reluga, T. C., & Shaw, A. K. (2015). Resource distribution drives theadoption of migratory, partially migratory, or residential strat‐egies. Theoretical Ecology, 8, 437–447. https://doi.org/10.1007/s12080‐015‐0263‐y

Rösner, S., Brandl, R., Segelbacher, G., Lorenc, T., &Müller, J. (2014).Noninvasivegeneticsamplingallowsestimationofcapercaillienum‐bers and population structure in the Bohemian Forest. European Journal of Wildlife Research, 60, 789–801. https://doi.org/10.1007/s10344‐014‐0848‐6

Sandercock,B.K.,Alfaro‐Barrios,M.,Casey,A.E.,Johnson,T.N.,Mong,T.W.,Odom,K.J.,…Winder,V.L.(2015).Effectsofgrazingandpre‐scribedfireonresourceselectionandnestsurvivalofuplandsand‐pipersinanexperimentallandscape.Landscape Ecology,30,325–337.https://doi.org/10.1007/s10980‐014‐0133‐9

Smith, J. N. M. (1974). The food searching behaviour of twoEuropean thrushes. Behaviour, 48, 276–301. https://doi.org/10.1163/156853974X00363

Spiegel,O.,Leu,S.T.,Bull,C.M.,&Sih,A. (2017).What’syourmove?Movementasalinkbetweenpersonalityandspatialdynamicsinan‐imalpopulations.Ecology Letters,20,3–18.https://doi.org/10.1111/ele.12708

Stanton,R.A.,Kesler,D.C.,&Thompson,F.R.(2014).Resourceconfig‐uration andabundanceaffect spaceuseof a cooperativelybreed‐ing resident bird.The Auk,131, 407–420. https://doi.org/10.1642/AUK‐13‐186.1

Tamburello,N.,Côté,I.M.,&Dulvy,N.K.(2015).Energyandthescalingofanimalspaceuse.The American Naturalist,186,196–211.https://doi.org/10.1086/682070

Thorup,K.,Tøttrup,A.P.,Willemoes,M.,Klaassen,R.H.G.,Strandberg,R.,Vega,M.L.,…Rahbek,C. (2017).Resource trackingwithinandacrosscontinentsinlong‐distancebirdmigrants.Science Advances,3,e1601360.https://doi.org/10.1126/sciadv.1601360

Tuanmu,M.‐N.,&Jetz,W.(2015).Aglobal,remotesensing‐basedchar‐acterizationofterrestrialhabitatheterogeneityforbiodiversityandecosystem modelling. Global Ecology and Biogeography, 24, 1329–1339.https://doi.org/10.1111/geb.12365

Viana, D. S., Santamaría, L., & Figuerola, J. (2016).Migratory birds asglobaldispersalvectors.Trends in Ecology and Evolution,31,763–775.https://doi.org/10.1016/j.tree.2016.07.005

Watanabe, Y. Y. (2016). Flight mode affects allometry of migrationrangeinbirds.Ecology Letters,19,907–914.https://doi.org/10.1111/ele.12627

Wilman,H.,Belmaker,J.,Simpson,J.,delaRosa,C.,Rivadeneira,M.M.,&Jetz,W. (2014).EltonTraits1.0:Species‐level foragingattributesof the world’s birds and mammals. Ecology, 95, 2027. https://doi.org/10.1890/13‐1917.1

Zuur, A. F., Ieno, E. N., & Elphick, C. S. (2010). A protocolfor data exploration to avoid common statistical prob‐lems. Methods in Ecology and Evolution, 1, 3–14. https://doi.org/10.1111/j.2041‐210X.2009.00001.x

BIOSKETCHMarlee a. Tuckerisapostdoctoralresearcherwhoisinterestedinlarge‐scalepatternsinecologyandbiogeographyandinspe‐ciesvulnerabilitytochangingenvironments.Marlee’sresearchencompassesmacroecologicalquestionsrelatedtoallometricscaling,predator–preyinteractionsandanimalmovement.

12  |     TUCKER ET al.

SUPPORTING INFORMATION

Additional supporting information may be found online in theSupportingInformationsectionattheendofthearticle.Appendix S1Summaryofspeciesandnumberofindividualsperspe‐ciesincludedintheanalyses.Appendix S2DetailsofRpackagesusedintheanalysesincludingthespecificfunctionusedanditsroleintheanalyses.Appendix S3 Example of data selection process for migratoryspecies.Appendix S4Environmentaldataannotationsummary.Appendix S5Datausedintheanalyses.Appendix S6Datadistributionsofthedisplacementdatausedintheanalyses.

Appendix S7Results for themodels including theweightedmeanEVIandEVIhomogeneityvalues.Appendix S8ComparisonoftheEVIHomogeneityslopecoefficientestimates.Appendix S9ResultsformodelsexcludingEVIHomogeneity.Appendix S10 Resultsmodels including an EVIHomogeneity anddietinteractionterm.

How to cite this article:TuckerMA,AlexandrouO,BierregaardROJr.,etal.Largebirdstravelfartherinhomogeneousenvironments.Global Ecol Biogeogr. 2019;00:1–12. https://doi.org/10.1111/geb.12875