Early arrival at breeding grounds: Causes, costs and a ... · ROTIC E T AL.Journal of Animal Ecolog y | 1629 studied the eastern flyway (Figure 1). White stork exhibits substan-tial,

J Anim Ecol. 2018;87:1627–1638. wileyonlinelibrary.com/journal/jane  | 1627© 2018 The Authors. Journal of Animal Ecology © 2018 British Ecological Society

Received:30April2018  |  Accepted:6August2018DOI: 10.1111/1365-2656.12898

R E S E A R C H A R T I C L E

Early arrival at breeding grounds: Causes, costs and a trade- off with overwintering latitude

Shay Rotics1  | Michael Kaatz2 | Sondra Turjeman1 | Damaris Zurell3 Martin Wikelski4,5 | Nir Sapir6 | Ute Eggers7 | Wolfgang Fiedler4,5 | Florian Jeltsch8 |  Ran Nathan1

1MovementEcologyLaboratory,DepartmentofEcology,EvolutionandBehaviour,AlexanderSilbermanInstituteofLifeSciences,TheHebrewUniversityofJerusalem,Jerusalem,Israel;2VogelschutzwarteStorchenhofLoburge.V.,Loburg,Germany;3GeographyDepartment,Humboldt-UniversitätzuBerlin,Berlin,Germany;4DepartmentofMigrationandImmuno-Ecology,Max-Planck-InstituteforOrnithology,Radolfzell,Germany;5DepartmentofBiology,UniversityofKonstanz,Konstanz,Germany;6TheAnimalFlightLaboratory,DepartmentofEvolutionaryandEnvironmentalBiology,UniversityofHaifa,Haifa,Israel;7DepartmentofPlantEcologyandConservationBiology,InstituteforBiochemistryandBiology,UniversityofPotsdam,Potsdam,Germanyand8Berlin-BrandenburgInstituteofAdvancedBiodiversityResearch(BBIB),Berlin,Germany

CorrespondenceShayRoticsEmail:[email protected]

Funding informationDFG,Grant/AwardNumber:NA846/1-1andWI3576/1-1;MinervaCenterforMovementEcology;MinistryofScienceandTechnology

HandlingEditor:JasonChapman

Abstract1. Earlyarrivalatbreedinggroundsisofprimeimportanceformigratingbirdsasitisknowntoenhancebreedingsuccess.Adults,malesandhigherqualityindividualstypicallyarriveearlier,andacrossyears,earlyarrivalhasbeenlinkedtowarmerspringtemperatures.However,themechanismsandpotentialcostsofearlyarrivalarenotwellunderstood.

2. Todeepentheunderstandingofarrivaldatedifferencesbetweenindividualsandyears,westudiedtheminlightoftheprecedingspringmigrationbehaviourandatmosphericconditionsen route.

3. GPSandbodyacceleration(ACC)datawereobtainedfor35adultwhitestorks(Ciconia ciconia)overfiveyears(2012–2016).ACCrecordsweretranslatedtoen-ergy expenditure estimates (overall dynamic body acceleration;ODBA) and tobehaviouralmodes,andGPSfixeswerecoupledwithenvironmentalparameters.

4. Attheinterindividuallevel(withinyears),earlyarrivalwasattributedprimarilytodepartingearlierformigrationandfrommorenorthernwinteringsites(closertobreedinggrounds),ratherthantomigrationspeed.Infact,early-departingbirdsflewslower,experiencedweakerthermalupliftsandexpendedmoreenergydur-ingflight,butstillarrivedearlier,emphasizingthecostandthesignificanceofearlydeparture. Individuals thatwintered furthersoutharrived laterat thebreedinggroundsbutdidnotproducefewerfledglings,presumablyduetopositivecarry-overeffectsofadvantageouswinteringconditions(increasedprecipitation,veg-etationproductivityanddaylighttime).Therefore,earlyarrivalincreasedbreedingsuccessonlyaftercontrollingforwinteringlatitude.Malesarrivedslightlyaheadoffemales.Betweenyears,latearrivalwaslinkedtocoldertemperaturesen route throughtwodifferentmechanisms:strongerheadwindscausingslowermigrationandlowerthermalupliftsresultinginlongerstopovers.

1628  |    Journal of Animal Ecology ROTICS eT al.

1  | INTRODUC TION

Arrivaltimeofmigratorybirdstotheirbreedinggroundsisofhighimportance due to its negative correlation with breeding success(Newton,2008;Smith&Moore,2005).Birds thatarriverelativelyearly benefit from acquiring higher quality territories, nesting lo-cationsandmates(Gunnarssonetal.,2006;Janiszewski,Minias,&Wojciechowski,2013;Møller,1994;Newton,2008;Smith&Moore,2005). Furthermore, earlier arriving parents produce relativelyearly-hatching offspring (Smith &Moore, 2005; Vergara, Aguirre,&Fernandez-Cruz, 2007)which in turn attainhigherpostfledgingsurvivalrates(Lok,Veldhoen,Overdijk,Tinbergen,&Piersma,2017;Monros, Belda, & Barba, 2002; Verboven & Visser, 1998). Arrivaltimeisalsoofprimeimportanceinrelationtoglobalwarming,whichtriggers advances in springphenologyandanearlierpeak in foodduring breeding (Menzel etal., 2006). Correspondingly, advancesin arrival timewere recorded inmanymigrating species (Gordo&Sanz,2006;Huppop&Huppop,2003;Usui,Butchart,&Phillimore,2017),butbirdpopulationsthatdonotadjusttheirtimingcansuf-ferfromdetrimentaltrophicmismatch(Both,Bouwhuis,Lessells,&Visser,2006;Sainoetal.,2011).Considering theprofoundeffectsofarrivaltimeonindividualfitnessandpopulationdynamics,under-standingitsunderlyingcausesofvariationisofmajorimportanceinbirdecology.

Inlinewiththat,alargevolumeofresearchtargetedthevariationinarrivaltimebetweenindividualsandyears.Interindividualvaria-tionwasmostlystudiedbydescribingwhichindividualsarrivefirst;theseweretypicallytheadults(Dittmann&Becker,2003;Newton,2008;Sergioetal.,2014),males(Cadahiaetal.,2017;Ouwehand&Both, 2017) and higher quality individuals (Blums,Nichols, Hines,Lindberg,&Mednis, 2005;Dittmann&Becker, 2003;Matyjasiak,2013). However, which spring migration properties underlie indi-vidualdifferencesinarrivaltimewasrarelystudied(butseeLemkeetal., 2013; Ouwehand & Both, 2017); that is, do early-arrivingbirdsflyfaster,stopless,departearlierformigrationorfromcloserwinteringsites?Recentbiotelemetry-basedstudieshighlightedthesignificant role of departure date rather than migration progress(speedandstopovers)indeterminingarrivaltime(Lemkeetal.,2013;

Ouwehand&Both,2017;Sergioetal.,2014).However,otherstudiesalsopointedout the importanceof speed (McKinnon,Macdonald,Gilchrist,&Love,2016;Schmaljohannetal.,2016),andtherewerecontradictingfindingsregardingthelinkagebetweenwinteringsitedistance and arrival time (Gunnarsson etal., 2006; Kentie etal.,2017; Lok etal., 2017). Thus, current findings are ambiguous andlimitedtoafewspecies,andthenatureoftherelationshipsbetweenthemigratorypropertiesandarrivaldateisnotclear.Furthermore,thecostsofearlyarrivalaremuchlessacknowledgedthanitsben-efits.Afewstudieshavepointedoutthedrawbacksoffacingharshenvironmental conditionsuponarriving (too)early in thebreedinggrounds(Newton,2008),butsimilareffectsthatcanpotentiallyactonearlymigrantsen routewereoverlooked.Thesequestionshigh-lighttheneedtostudytheinterplaybetweenthereturnmigrationproperties and arrival time to uncover basic aspects of the birds’migratory“race”tothebreedinggrounds.

At the interannual level, arrival timevariationhasbeenshownto correlate with several environmental factors (Gordo, 2007),wherethemainfindingspointedoutthathightemperaturesalongthemigration routeand in thebreedinggroundspromotedearlierarrival (Cadahia etal., 2017; Gordo, Tryjanowski, Kosicki, & Fulin,2013;Huppop&Huppop,2003;Marra,Francis,Mulvihill,&Moore,2005; Vaitkuviene,Dagys, Bartkeviciene,&Romanovskaja, 2015).However, there is a lack of direct evidence connecting these en-vironmental factors with migratory journeys of individual birds,which is needed to develop amoremechanistic understanding oftheir effects (Gordo, 2007). This information gap originated fromthe difficulty to obtain high-resolution tracking data of migratingbirds,thoughinrecentyears,thishasbecomemorefeasiblewiththeongoingadvancesinbiotelemetrytechnology.

WestudiedthespringmigrationofadultwhitestorksequippedwithadvancedGPS–bodyacceleration(ACC)transmittersaimingtounderstandvariationinarrivaltimebetweenindividualsandyears.Whitestorksareiconic,long-distance,migrants,whichmostlybreedinEurasiaandmigrate tosub-SaharanAfrica.Theymigrateduringdaylightusingsoaring-glidingflight,utilizingthermalupliftstomini-mizetravelcosts(Leshem&YomTov,1996;Roticsetal.,2016)alongtwocentralflyways,eastandwestoftheMediterranean;here,we

5. Thisstudyshowedthatdistinctmigratorypropertiesunderliearrivaltimevaria-tionwithinandbetweenyears.Ithighlighted(a)anoverlookedcostofearlyarrivalinducedbyunfavourableatmosphericconditionsduringmigration, (b)an impor-tantfitnesstrade-offinstorksbetweenarrivaldateandwinteringhabitatqualityand(c)mechanisticexplanationsforthenegativetemperature–arrivaldatecorre-lationinsoaringbirds.Suchunderstandingofarrivaltimecanfacilitateforecastingmigratingspeciesresponsestoclimatechanges.

K E Y W O R D S

arrivaldate,birdmigration,breedingsuccess,carry-overeffects,Ciconia ciconia,climatechange,whitestork

     |  1629Journal of Animal EcologyROTICS eT al.

studiedtheeasternflyway(Figure1).Whitestorkexhibitssubstan-tial,andyetunexplained,individualvariationinwinteringsitesalongthe eastern flyway (Figure1; Berthold, Kaatz, & Querner, 2004),whichmayaffecttheirsubsequentarrivaltimetobreedinggrounds.Uponarrival,storksdisplayhighfidelitytoformernestsiteandmate(Barbraud,Barbraud&Barbraud,1999),butvitalclashesovernestsare fairly common (pers.obs.;Wuczynski,2005).As inother spe-cies,earlierarrivalwaslinkedtoenhancedbreedingsuccessinwhitestorks(Fulin,Jerzak,Sparks,&Tryjanowski,2009;Janiszewskietal.,2013;Kosicki,Sparks,&Tryjanowski,2004).

Adult storks from a breeding population in Saxony-Anhalt,Germany, were fitted with solar transmitters that recorded high-resolutionGPSandbodyacceleration (ACC)data.GPS fixeswerecoupled with environmental parameters (Dodge etal., 2013), andACCrecordswereusedtoapproximateactivity-relatedenergyex-penditure(bycalculatingODBA;Wilsonetal.,2006)andtodeducebehaviouralmodes(Roticsetal.,2016).Overall,weused90springmigrationtracksconsistingofca.400,000GPS-ACCrecords,from35adult storksacross fiveyears (2012–2016).Thesemultifaceteddataallowedustostudythecauses,costs,trade-offsandimplica-tionsofinterindividualandinterannualvariationinarrivaltime.

Attheinterindividuallevel,wefirstexaminedwhichofthefol-lowingspringmigrationpropertiesprimarilycharacterizedtheearly-arrivingindividuals:(a)migrationspeed(dailydisplacement),(b)totalstopovers duration, (c) departure time and (d) departure location(latitude of lastwintering site). As discussed above, therewas nostrongbasisfor informedpredictionsregardingtherelative impor-tanceofthesepropertiesforarrivaltime,butseveralrecentstudiesputforwardthesignificanceofdeparturetime(Lemkeetal.,2013;Ouwehand&Both,2017;Sergioetal.,2014).Wefurtherexploredthe relationships between themigration properties targeting twopotentialmigratorytrade-offs:(a)Alongwiththewell-documentedbreedingenhancementbyearlyarrival,whichwasre-assessedhere,we examined a potential cost in the formof highmigratory flighteffort (flightODBA)whilemigratingearlierdue to less favourableatmospheric conditions. (b) We investigated whether individualsthatwintered further southdepartedearlierormigrated faster tocompensate for the longer journey, or alternatively arrived laterandhad lowerbreedingsuccess (as inspoonbills,Loketal.,2017).Infact,someofthestudiedindividualswinteredthousandsofkilo-metres furthersouth than thecommonlyusedwintering region inthe Sahel (Figure1), andwe aimed to understand the drivers andconsequences of this by comparing environmental conditions andbehaviouratwinteringsitesandsubsequentbreedingsuccess.Thelast interindividual analyses were to examine whether protandry(male-firstarrival)exists inwhitestorks;previousstudieshavere-portedcontradictoryfindingsinthisregard(Barbraud&Barbraud,1999;Vergaraetal.,2007).

Inourinterannualanalyses,weaimedtogobeyondthewidelyreported correlation between high-temperature and early arrival(Gordo, 2007; Usui etal., 2017), to examine the effects of atmo-spheric factors en route (wind and thermal uplift) on migratoryparameters(speed,stopovers,departuretime)thatmayunderliean-nualdifferencesinarrivaltime.

2  | MATERIAL S AND METHODS

2.1 | Study site and tracking data

Wetrapped62adultstorksinthestateofSaxony-Anhalt,Germany,and fitted them with solar GPS-ACC transmitters (e-obs GmbH;Munich,Germany) thatweighed55g includingharness, ca. 2%oftheaveragestork’sweight(seeRoticsetal.,2016).Birdsexwasde-terminedbymolecularmethods(SupportingInformationAppendixS1). The transmitters recorded GPS fixes every 5min when solarconditionsweregood(95%ofthetime)orevery20min,otherwise.EveryfiveminutesanACCburstof3.8swasrecordedat10.54Hzforthethreeperpendicularaxes.ODBA—avalidproxyforactivity-related energy expenditure (Wilson etal., 2006)—was calculatedforeveryACCburst (seeRoticsetal.,2016fordetails).DatawerestoredonboardandweredownloadedviaaVHFradiolinkuponlo-cating thestork (Roticsetal.,2016).Outof62 tagged individuals,weuseddata from35birds.For21birds,datawerenotavailable

F IGURE  1 Whitestorkspringmigrationtracks.Dotsandsquaresmarkspringmigrationdeparturelocations(i.e.winteringsite):reddotsfortheSahelandpalebluesquaresformoresouthernlocations.Sixty-oneofthe90departurelocationswereintheSahel(overlappingreddots).Athirdofthetrackswererandomlyexcludedfromthefiguretoreducethevisualload[Colourfigurecanbeviewedatwileyonlinelibrary.com]

1630  |    Journal of Animal Ecology ROTICS eT al.

duetobirdsnotbeingfound intheyearaftertagging (n=15), tagmalfunctions(n=5)oruser-relatederrors(n=1).Sixindividualsthatmigrated through theWestern European flyway and wintered inSpainwereexcludedfromtheanalysisastheirmigrationandwinter-ingweresubstantiallydifferentfromallothersthattooktheeasternflywayandwinteredinAfrica.

Nestswere identified based on the tracking data and verifiedwith field observations. The number of fledglings was monitoredbygroundobservations anddrone-based filming flights (ca. every3weeks).Fledglingnumbercouldnotbeobtainedforthreenestingevents,whichwereexcludedfromthebreedingsuccessanalysis.

2.2 | Environmental data

EachGPSfixwasannotatedwithenvironmentaldataofwind,ther-mal uplift velocity, ambient temperature, precipitation and NDVIusing the Env-DATA track annotation tool of MoveBank (Dodgeetal.,2013;seeSupportingInformationAppendixS2fordetails).

2.3 | Data analysis

Arrival time to the breeding area was defined as the date of ap-proachingwithin20kmofthenest. Itwasthesameasthearrivaldatetothenestitselfin85%ofthecases,buttheformerwaspre-ferredasitdisregardedthetimethebirdspentsearchingforanestafter arriving in the nesting area (results were consistent acrossmethods).Dateswereanalysedasdayofyear(DOY):serialdaynum-berfromJanuary1.

Departure date was identified with a backward–forward al-gorithm; starting fromapoint atwhich thebirdwasundoubtedlymigrating (crossing 17.5°N northward), we searched backwardsuntil reachinga stationaryphaseof five consecutivedaysofdailydisplacement<50km.Fromthispoint,wesearchedforwardforthefirst three consecutive days of (a)more than 50km displacementeachday, (b)more than150kmtotaldisplacementand (c)generalnorthwarddirection(azimuth>320°and<110°).Departuredaywasdefinedasthefirstofthesethreedays.Themethodwasextensivelyvalidatedbyvisualexaminationofthetracks.Accordingly,departurelocationwas the last stationary locationbeforedeparturedate. Itwas included in analyseseitherby its ºN latitude (negativevaluessouthoftheequator)orcategorizedintotwoclasses:Sahel-(latitude>9°N)andmoresouthern-winteringlocations(Figure1).

Migrationspeedwastheaveragedailydisplacementinprogressdays(>30kmdisplacement),whereasstopoverdurationwasthetotalnumberofstationarydays(<30km).Migratoryflightenergyexpendi-turewasdeducedfromthemeanODBAduringflight(speed>5m/s).Windwasexaminedduringflightandunlessspecifiedotherwisealsothermaluplift(seeSupportingInformationAppendixS2fordetails).

Therewasveryhighvariationinthestorkmigrationtracksdueto different departure (wintering) locations (Figure1). Given ourresearchobjectives,weaimedtocomparesimilarmigrationtracksamong individuals toavoid theprominent,potentiallymasking,ef-fectsofmigratingindifferentgeographicalregions(Chevallieretal.,

2010;Klaassen,Strandberg,Hake,&Alerstam,2008),asforexam-ple,amigrationjourneystartingfromSouthAfricaandfromSudan(Figure1).Therefore,themigrationpropertiesofspeed,stopovers,flight cost and en route environmental conditions were examinedwithinaspatialwindowbetweenlatitudes20°and51.5°N,inwhichallbirdsdisplayedsimilartracks(Figure1).Migrationdepartureloca-tionanddatewerecalculatedirrespectivelyofthisspatialwindow.

We also explored the birds’ late wintering period of the twomonths prior to migration and compared Sahel-wintering andsouthern-wintering conditions: NDVI, precipitation, diurnal tem-perature(7:00–16:00GMT)anddaylightlength(calculatedinMatlabwiththesuncyclefunction,Pawlowicz,2009),andthewinteringbe-haviour: daily distance in stationary days, and relative time spentforaging.Thelatterwastheratioofwalkingandpeckingrecordsdi-videdbytotalrecords,basedonACCdataclassifiedintobehaviouralmodeswithsupervisedmachinelearning(seeRoticsetal.,2016fordetails).Werepeated thiscomparison fora fixedwinteringperiodofDec-Jan,dismissingthelinkto,andpotentialeffectsof,migrationonsetandtheresultswererobust(notreported).

Springmigrationsofindividualsthatdidnotattempttobreedinthatyear(i.e.didnothaveanest;eightcasesoriginatingfromsevenbirds)wereextremeoutliersintheirphenology,reachingthebreed-ingareas23±3daysaftertheaveragearrivaltimeofnestingbirds,and were thus excluded from the analyses except for when por-trayingtheirdifferences (Non-nesting storkssectionoftheresults).Furthermore,ourbasic,underlyingworkingassumptionwasthatthestorkshadamotivetoarriveearlyforbreedingpropose,whichcan-notbeascertainedinthenon-nestingcases.

2.4 | Statistical notes

Thespringmigrationdeparturetime,departurelatitude,speedandstopover lengthhavealtogether straightforwardeffectsonarrivaldatewhenincludedinasinglemodel(SupportingInformationTableS1)derivingfromthebasicspeed–time–distancekinematicrelation.Wewereinterestedindeterminingwhichofthesemigratoryprop-ertiesprimarilyexplained individualvariation inarrival timewithinyears, that iswhich propertiesmainly characterized early-arrivingindividuals. For this, the effect of eachmigratory property on ar-rivaltimewasexaminedseparatelywithalinearmixedmodel(LMM;year and individualasrandomfactors)andthemodels’likelihoodandmarginalR2were compared.MarginalR2was calculated followingNakagawaandSchielzeth(2013)usingtherpackagemumin(Barton,2016).

Generalizedlinearmixedmodels(GLMMs)witherrordistributionaccordingtothedependentvariable(normaldistributionwastestedwith Lilliefors test) and LMMswere usedwith year and individual asrandomfactorsin(almost)allstatisticalanalyses.Theexceptionswereanalyseswheretheannualeffectswereofexplicitinterestandthus the yearwas a fixed factor (e.g. annual differences in arrivaltimereportedinFigure4).Tofurtherexaminedifferencesbetweenyears, post hoc tests (Tukey) were conducted using the lsmeans() R function (Lenth, 2016). Individual consistency indeparture time

     |  1631Journal of Animal EcologyROTICS eT al.

andlocationwasexaminedbycalculatingrepeatabilityacrossyears(Intraclasscorrelation)usingtherptr rpackage(Stoffel,Nakagawa,&Schielzeth,2017).Multicollinearitywastested,verifyingthatallpre-dictorsinmultipleregressionshadavarianceinflationfactor(VIF)<3(Zuur,Ieno,&Elphick,2010).

3  | RESULTS

3.1 | Individual variation in arrival time—effect of migration departure date

Spring migration departure time and location (latitude), speed andstopoverlengthallhadstatisticallysignificanteffectsonarrivaltimewhenexaminedtogether(SupportingInformationTableS1).Examiningthem separately revealed that departure time and latitudewere ofmajor importance inexplaining individualvariationwithinyears,butnotmigrationspeedandstopoversduration(Figure2).Thus,theearly-arriving individualswerethosethatdepartedearlierandfrommorenorthernwintering locations, but not necessarily progressed faster.In fact, early departure date was associatedwith slowermigrationspeed (Figure3a; Supporting Information Appendix S3) and longerstopovers(GLMMwithPoissonerrordistribution;β=−0.032±0.006,t80=−4.76, p<0.001), emphasizing that early-departing individualsarrivedearlieratbreedinggroundsdespitetheirslowermigrationpro-gress.Correspondingly,individualsthatdepartedearlierexperiencedlowerthermalupliftduringmigration(Figure3b)andexhibitedhigherflightenergyexpenditure (flightODBA;Figure3c).Additionally,ear-liermigrantshadslightlylessdaytimeen routewhichmightbelinked

to reduced thermals and slower migration (Supporting InformationAppendix S3). Individual departure date was rather consistent be-tweenyears(repeatabilityr=0.51±0.11,p<0.001)andarrivaldate(repeatabilityr=0.49±0.12,p<0.001).

3.2 | Wintering location and breeding success

Birds thatwintered further south than theSaheldidnotcompen-satefortheirlongermigrationdistancebydepartingearlier;onthecontrary,theydepartedlater(Table1)andarrivedlateratthebreed-inggrounds(Figure2b).Yet,theirlatearrivaldidnotresultinlowerbreeding success (see below), possibly because of better environ-mental conditionsexperienced in thesouthern-winteringgrounds,reflected by higher NDVI and precipitation, milder temperatures(Table1)andlongerdaylighttimecomparedtotheSahel-winteringregion(Table1).Correspondingly,duringstationarywinteringdays,southern-winteringbirdsmovedhalf the amountofdailydistanceand spent relatively more time foraging compared to the Sahel-winteringindividuals(Table1).Individualwinteringhabits(Sahelvs.Southern) were moderately repeatable (r=0.45±0.17, p<0.001)indicatingsignificantbutnotabsoluteconsistency(sixoutofthe26birds thatwintered in theSahel, alsowintered southof theSahelin other years, see Supporting InformationAppendix S4 formoredetails).Wintering site selection was not affected by year or sex(GLMMwithbinomialerrordistribution;year:F4,76=1.06,p = 0.38; sex:F1,76=0.004,p=0.95).

Winteringlocationandarrivaltimecounteractedeachotherinaf-fectingbreedingsuccess.Southernwinteringandearlyarrivalincreased

F IGURE  2 Therelationshipsbetweenarrivaltime(dayofyear,DOY)andspringmigrationproperties:(a)departuretime,(b)departurelocation,(c)migrationspeedand(d)stopoverlengthbasedon82migrationsfrom34individuals.Eachplot’sregressionlineandR2—whichisamarginalR2(Nakagawa&Schielzeth,2013)—arebasedonaseparatelinearmixedmodeldetailedinSupportingInformationTableS2[Colourfigurecanbeviewedatwileyonlinelibrary.com]

1632  |    Journal of Animal Ecology ROTICS eT al.

fledglings number (Table2), but because southernwinteringwas as-sociatedwithlatearrival (Figure2b),thesefactorscounteractedeachotherwhenexaminedseparately(i.e.withoutcontrollingforeachother;

Supporting Information Table S5). Thus, wintering latitude affectedbreedingsuccessonlyaftercontrollingforarrivaltimeandviceversa.

3.3 | Non- nesting storks

Non-nestingstorksarrivedlateratthebreedinggroundscomparedto nesting storks (LMM; β=23±3.3days, t88=7.1, p<0.001,SupportingInformationFigureS1).Thus,includingthenon-nestingstorks in theabovebreeding success analysis increased thenega-tiveeffectof late arrival (GLMMwithPoissondistribution; arrivaltime:β =−0.039±0.009, t84=−4.12,p<0.001;wintering latitude:β =−0.023±0.008days, t84=−2.87, p=0.005). Compared tonesting individuals, non-nesting birds departed later for springmigration (LMM; β=13.6±4.0days, t88=3.41, p<0.001), tooklonger stopovers (GLMM with Poisson distribution; non-nesting:12.25±3.43days, nesting: 7.88±0.58, t88=4.85, p<0.001) andmigrated slower (LMM; β =−15.44±6.17km/days, t88=2.50,p=0.014), but no differences were found in wintering sites se-lection (Southern vs. Sahel; GLMM with binomial distribution;β=0.27±1.03,t88=0.26,p=0.80).

3.4 | Sex differences

Males arrived at the breeding area five days earlier than females(LMM;β=−5.11±2.61days,t80=−1.96,p=0.05).However,thedif-ferencebetweensexeswasnotclear-cut;selectingrandomlyamaleanda female from the sameyearyieldedmale-first arrivalonly in68%ofthecases(basedon10,000randomselections).Similarly,insevencasesinwhicharrivaltimesofbothpair-mateswereavailable(originatingfromthreetaggedpairsacrossmultipleyears),themalearrivedonaverage6.5daysaheadofitsfemale,butonlyinfiveofthecaseswashethefirst.Therewerenosex-relateddifferencesintheothermigrationproperties(departuretimeandlocation,speed,stopovers,fightODBA;GLMMs,N.S.).

3.5 | Interannual variation

Thereweredetectabledifferencesinarrivaltimebetweenthestudyyears (LMM; F4,52.8=7.7, p<0.001) with birds arriving earlier in

F IGURE  3 Effectsofdeparturedayon(a)migrationspeed(dailydisplacement),(b)thermalupliftexperiencedenrouteand(c)flightenergyexpenditure(82tracksfrom34individuals).Alleffectsaresignificantatp<0.001(LMMs;seeSupportingInformationTableS3forstatisticaldetails)andtheirmarginalR2sare0.18,0.35and0.22,respectively[Colourfigurecanbeviewedatwileyonlinelibrary.com]

ParameterSahel (mean ± SE)

Southern (mean ± SE) t80 p

NDVI 0.24±0.01 0.47±0.02 11.59 <0.001

Precipitation(mm/day) 0.007±0.006 7.13±0.80 13.77 <0.001

Daytimetemperature(°C)

30.48±0.19 26.26±0.60 −8.67 <0.001

Daylight(hours) 11.47±0.01 12.28±0.10 11.75 <0.001

Dailydistance(km) 69.06±3.10 35.59±3.78 −6.28 <0.001

Relativeforagingtime 0.21±0.01 0.30±0.01 5.98 <0.001

Departureday(DOY) 49.8±1.23 56.8±1.82 2.29 0.02

Note.DOY:dayofyear;seeSupportingInformationTableS4forrandomeffectvariances.

TABLE  1 ComparisonsofenvironmentalconditionsandbehaviourduringwinteringbetweenstorksintheSahelbeltandinmoresouthern-winteringsites(Figure1).EachlinedetailsparameterdifferencesthatwereexaminedusingaLMMwithwinteringsite(Sahel/Southern)asafixedfactor,year and individualasrandomfactorsandtheparameterasthedependentvariable.Dailydistanceandrelativeforagingtimewerecalculatedduringstationarywinteringdays

     |  1633Journal of Animal EcologyROTICS eT al.

2014comparedto2012,2013 and 2015,butnotcomparedto2016 (Figure4a). Correspondingly, temperatures during migration werewarmer in 2014thanin2012, 2013 and 2015(Figure4b).Twodiffer-entmigratorypatternsunderlietherelativelyearlyarrival in2014: sloweraveragemigrationspeedin2012 and 2015thatcouldbere-lated to strongerheadwindsduring flight in theseyears (Figure4)and longerstopovers in2013 and 2015 thatcouldbeexplained inpart by lower thermal uplift conditions in 2013 (during flight andwhile on the ground, see Supporting Information Appendix S2;Figure4). Correspondingly, migration speed was negatively asso-ciated with headwind velocity (LMM; β=−9.59±2.80km/days,

t80=3.42,p<0.001),andstopoverdurationwasnegativelyaffectedbythermaluplift(GLMMwithPoissondistribution;β=−1.47±0.31,t80=−4.72,p<0.001).Migrationdeparturetimeanddeparturelo-cationdidnotdifferbetweenyears (GLMMs;N.S).Year2016 pre-sented an intermediate arrival time value between year2014 and theothers(2012, 2013 and 2015),notdifferingfromanyofthestudyyears.

4  | DISCUSSION

Our study aimed to illuminate individual differences in arrivaltimeby investigating thespringmigrationproperties.We foundthat departing for springmigration early and frommore north-ernlocationsunderlietheearlyarrivalofindividualswithinyears(Figure2);however,theycameatthecostofmigratingandwin-tering in less favourableconditions.Migrationspeedandstopo-verlengthmediatedbetween-yeardifferencesinarrivaltimethatwereassociatedwithannualvariation inatmosphericconditionsen route (Figure4), but had smaller impacts on individual differ-enceswithinyears.

Storksthatdepartedearlierforspringmigrationarrivedearlieratthebreedinggrounds,despitemigratingslowerandtakingmorestopovers.Thisemphasizestheimportanceofmigrationdeparturetime,inlinewithpreviousstudiesinmigratingbirdsalongdifferentflyways(Lemkeetal.,2013;Ouwehand&Both,2017;Sergioetal.,2014).Thisrelationshipalsoexplainswhytheprevailingeffectsofmigrationdeparturetimeandlocationmaskedtheeffectofmigra-tionspeedonindividualarrivaltimewithinyears,whenexaminedseparately. A very similar association between early departureand slowermigrationwas described in black kites (Sergio etal.,2014),anditwassuggestedthatearly-departingbirds(whichweregenerally older) mainly minimized energy expenditure and thustravelledmoreslowly,whereaslateonesminimizedmigrationtimeandtravelled“inahurry”toadvancetheirarrival.Inourcase,wesuggestadifferentexplanationasearliermigrantswereexposedtolessfavourableconditionsofweakerthermalupliften route,anessentialtransportresourceforsoaringbirds(Hedenstrom,1993;Sapir,Wikelski,McCue,Pinshow,&Nathan,2010).This resultedinhigher flight costs (flightODBA),most likelydue tousing rel-ativelymorecostlyflappingvs.glidingflight (Roticsetal.,2016).Thus,migratingearlierwasmorestrenuousintermsofflighteffortwhichcanalsoexplaintheslowermigrationprogress.Morenota-bly,thesefindingspresentexplicitmigration-relatedcostsofearlyarrivalatbreedinggroundsthathavenotbeenacknowledgedthusfar,namelyenergyandtimecosts inducedbyunfavourablecon-ditionsen route.Thisaddsup toharshenvironmental conditionsatthebreedingareasthemselvesuponearlyarrival(Møller,1994;Newton,2008).

Earlymigratingbirdsneedtobesufficientlyfittocopewiththesecosts,andcorrespondingly,birdphysicalconditionhasbeenfoundtoaffectdeparturetiming(Cooper,Sherry,&Marra,2015).Ourfind-ingsshowedthatmigrationdeparturetimewasaratherconsistent

TABLE  2 Arrivaltimeandwinteringlatitudeeffectsonbreedingsuccess(fledglingsnumber)basedonaGLMMwithPoissonerrordistributionandyear and individualasrandomfactors.Lowerwinteringlatitudesmeanmoresouthernlocations

Parameter β SE T76 p

Arrivaltime(days)

−0.017 0.008 −2.14 0.03

Winteringlatitude(°N)

−0.021 0.010 −2.07 0.04

Note.Yearvariance:0.09;individualvariance:<0.001.

FIGURE 4 Annualdifferencesin(a)springmigrationpatternsand(b)environmentalconditionsen route.Underneatheachmigratorypatternin(a)isapotentiallyrelatedenvironmentalfactorin(b).Zscores(standarddeviationsfromtheoverallmean)aredisplayedtounifythevariablesononeY-axis.*denotessignificantdifferencescompared to year 2014(p<0.001;Tukey’sposthoctestfollowingaGLMMwithindividualasarandomfactor).Legenddisplays:year(samplesize).Thedashedlineacrosstheheadwindbarsin(b)markszerowindspeed(rawvalue),belowwhichheadwindisnegative(i.e.positivetailwind)[Colourfigurecanbeviewedatwileyonlinelibrary.com]

1634  |    Journal of Animal Ecology ROTICS eT al.

featureoftheindividual,inaccordancewithpreviousfindingsfromdifferent flyways (Tottrup etal., 2012; Yamamoto etal., 2014),likely related to sex (discussed below), age and individual quality(Dittmann&Becker,2003;Matyjasiak,2013;Newton,2008;Sergioetal., 2014).Thus, presumably the fittest storks startedmigratingandarrivedearlieratthebreedinggrounds.Alongsideitscosts,ourdata reconfirmed the renowned association of early arrival withenhancedbreedingsuccess,aftercontrollingforwinteringlocation(latitude).

Storksthatwinteredatmoresouthernlatitudesarrivedlateratthebreedinggrounds.Obviously,theytookalongerjourney,butnotably,theyalsodeparted later.Thishints that thedeparturedecisionwasirrespectiveof thewinteringdistance,butpossibly reliedonanen-dogenouscircannualclock(Gwinner,1996)tunedwithexternalsignalsleadingtoadelayeddepartureatmoresouthern-winteringlatitudes.Such potential signals could be day-length (Kumar etal., 2010) andtemperature(Sokolov&Tsvey,2016)thathavedifferenttrendsduringspringinthedifferentwinteringsites.Weproposethatthevariationindeparturetimewasrelatedtoindividualqualitywithinwinteringsitesandtorelevantenvironmentalcuesbetweensites.Insomecontrasttoourfindings,birdsinbetterwinteringhabitatswerereportedtoad-vancetheirdeparturetime(Paxton&Moore,2015),stressingthattheinternalandexternalfactorsregulatingmigrationonsetvarybetweenstudysystemsandshouldbefurtherinvestigated.

Even though they arrived late to thebreeding grounds, storksfrom southern-wintering sites did not suffer from lower breedingsuccess. In fact, wintering at southern latitudes enhanced breed-ingsuccessaftercontrollingforarrivaldate.Aprobablecausewasthe better environmental conditions at more southern-winteringsites, involving milder temperatures and increased day-length,precipitation and NDVI. The latter was positively correlated withinsect abundance (Schlaich etal., 2016), a prime food resource ofwhite storks (Cheriak, Barbraud,Doumandji, & Bouguessa, 2014).Correspondingly, southern-wintering storks moved half the dailydistance compared to Sahel-wintering ones during winter, pre-sumably due to more abundant resources. Similarly, Montagu’sHarriers(Circus pygargus)winteringintheSahelexhibitedanegativeNDVI-dailydistancecorrelation(Schlaichetal.,2016).Additionally,southern-winteringstorksspentrelativelymoretimeforaging,prob-ably due to reducedmovement time and extended daylight time.Theseenhancementsmayexplaintheremarkablelengtheningofmi-grationdistanceexhibitedbysomeoftheindividuals,winteringupto4,500kmfarthersouththanthecentralwinteringregion(Figure1).Furthermore, our findings add support to the elusive evidence ofcarry-overeffects(Loketal.,2017;Norris,Marra,Kyser,Sherry,&Ratcliffe,2004),inwhichwinteringconditionshavedelayedimpactslateron,duringbreeding.

Therewasanapparenttrade-offbetweenarrivaltimeandwin-teringlatitude.

Early-arrivingbirdswinteredandmigratedunderlessfavourableenvironmental conditions but enjoyed themultiple advantages ofearly arrival (discussed in the Introduction). Contrarily, southern-winteringbirdsexperiencedbetterenvironmentalconditionsduring

winteringandduringtheirdelayedmigration,arrivinglateratbreed-inggroundsbutprobably lessexhausted.Thesetwofactorsactedinoppositedirections,balancingeachother inrespecttobreedingsuccess,suchthatearlyarrivalaswellasmoresouthernwinteringenhancedfledglingnumber,butonlyaftercontrollingforeachoth-er’s effect. At the individual level, storks rather consistently usedone of the two strategies. Frompopulation and evolutionary per-spectives,divergentwinteringstrategiesreflectmigratoryplasticitythatenhancesthestorksabilitytoadjusttoglobalchanges(Gordo&Sanz,2006)andtocopewithhighlyvariableenvironmentalcondi-tions,likerainfallintheSahel(Nevoux,Barbraud,&Barbraud,2008).More broadly, birdmigration is a flexible phenomenon (Alerstam,Hedenström,&Åkesson,2003;Newton,2008),andmanyspeciesdisplaysignificantvariationinmigrationflywaysbetweenandwithinpopulations(e.g.Barbraud,Barbraud,&Barbraud,1999;Shamoun-Baranes,Burant,Loon,Bouten,&Camphuysen,2017;Weimerskirchetal.,2017).Furtherresearchonthedifferencesandtrade-offsbe-tweenmigration strategieswould advance our knowledge on theevolutionandconservationofbirdmigration.

Eventhoughearlyarrivalandsouthern-winteringbalancedeachotherinaffectingfledglingnumbers,thereisstillanadvantageforearly arrival in the form of having earlier-hatching offspring thattypicallyhavehighersurvival(Loketal.,2017;Monrosetal.,2002;Verboven & Visser, 1998), as was also seen in our study popula-tion(S.Rotics,unpublisheddata).Similarly,spoonbills (Platalea leu-corodia) thatwintered farther away arrived later at their breedinggroundsandstilldidnotraiselessoffspring,buthadloweroffspringrecruitmentrates(Loketal.,2017).Thiscouldexplainwhymostofourstorkswintered in theSahel,preferringearlyarrivaloverwin-teringhabitatquality.Theremightalsobesurvival implicationsforthedifferentmigratorystrategies (e.g.Lok,Overdijk,Tinbergen,&Piersma,2011;Roticsetal.,2017).Betterwinteringandmigratingconditionsmaypromotesurvivalofsouthern-winteringindividuals,orontheotherhand,reducedmigrationriskmaybenefitonesthatwintered closer to breeding grounds. Such implications, however,couldnotbeevaluatedhereduetothenatureofthedata;onlytracksofreturningadultswereavailable(seeMethods)anddisappearancecouldnotbeconfidentlyassignedtomortalityvs.largebreeding-sitedisplacement.

Theabovewintering-arrival time trade-off canexplain thede-creaseintheeffectofarrivaldateonstorkbreedingsuccessinSpain(Gordoetal.,2013)andthelackofdifferencesinfledglingnumberbetween sedentary and migratory storks in France (Massemin-Challetetal.,2006),assumingthattheearly-arriving/sedentarybirdswinteredinlower-qualityhabitats.However,ourresultscontradictpreviousstorkstudiesthatfoundaneffectofarrivaltimeonbreed-ing successwithout controlling forwintering location (Fulin etal.,2009; Janiszewski etal., 2013;Kosicki etal., 2004).Onepotentialexplanationisthatwinteringsitevariationwaslowerintheseearlierstudiesascomparedtoourstudypopulation.Thiscouldarisefrominherentpopulationdifferencesorfromtemporalchanges—arecenttrendofincreaseinstorkswinteringvariation(Martín,Onrubia,delaCruz,&Ferrer,2016).Alternatively,much larger samplesizes in

     |  1635Journal of Animal EcologyROTICS eT al.

previous studies (n>1,000 inFulin etal., 2009; Janiszewski etal.,2013)alloweddetectionoftheeffectofarrivaltimeindependently.Hypothetically,arrivaltimemighthavelessprominenteffectsinthelong-termmonogamouswhite stork, since the individual doesnotneedtofindanewpairmate,butjusttoarrive“ontime”torejoinitsformerpartner.

In eight cases in our study, adult storks that returned to thebreedinggroundsdidnotnest.Theirarrivaldateswereexception-allylateastheydepartedforspringmigrationtwoweekslater,mi-gratedslowerandtooklongerstopovers,comparedtonestingbirds.Thesepatternsmayimplythatthedecisionnottonestwasmadeinadvanceandwasthenmanifestedbymorerelaxedspringmigrationbehaviour.Alternatively, tardymigration and failure to nestmightbothbetheoutcomesofinferiorphysicalconditions.

Male storks arrived at the breeding grounds on average fivedaysaheadoffemales,similartoalargenumberaspeciesinwhichprotandrywas described (Cadahia etal., 2017; Saino etal., 2010).However,thesexdifferences instorkarrivalwerenotverystrong(p=0.05), and thedividewasnotabsolute (sometimes the femalearrivedfirst).Webelievethatthesensitivetrackingdataallowedusto identify themhere and the use of less precise nest-monitoringmethodsmayaccountfortheambiguityregardingstorkprotandryinpreviousstudies(Barbraud&Barbraud,1999;Tortosa&Redondo,1992;Vergaraetal.,2007).Thus,assumingthefirststorkarrivingatthenestisthemale,asinGordoetal.(2013),wouldbecorrectinca.70%ofthecasesaccordingtoourdata.

Interannual differences in arrival time reconfirmed the asso-ciation between early arrival and higher temperatures en route (Gordoetal., 2013;Huppop&Huppop,2003;Marraetal., 2005;Vaitkuviene etal., 2015). Linking stork migration tracks with at-mospheric factors uncovered two nonexclusive mechanisms thatwereresponsibleforthisassociation:(1) lowertemperatureswereaccompaniedbynorth-to-southwinds (thatpossiblybrought coldweather along the migration range) which presented strongerheadwinds for the northward-migrating storks resulting in slowermigration speed. (2) Lower thermal upliften route in colder yearswas linked with longer stopover time, probably due to elevatedflighteffort(Roticsetal.,2016).Thus,delayedarrivalincolderyearsresultedfromincreasedheadwindsand/orreducedthermaluplift.Correspondingly,windsareknowntoaffectflightspeed(Shamoun-Baranes etal., 2003;Vansteelant etal., 2015) and strong thermalupliftreducesflighteffort(Chevallieretal.,2010;Hareletal.,2016;Sapiretal.,2010)andstopoverduration(Duerretal.,2015;Nourani&Yamaguchi,2017).However,asfarasweknow,thisstudyisthefirsttolinktheseatmosphericfactorsexplicitlywithinterannualdif-ferencesinarrivaltime.

In summary, distinct factors mediated interindividual (withinyear) and interannual variation in arrival time, similar to previousfindings inblack-tailedgodwits (Gunnarssonetal.,2006).Wesug-gestthatmigrationdeparturetimingand locationwerechieflyde-termined by intrinsic factors and therefore explained arrival timevariationwithinyears,whereasmigrationspeedandstopoverweremorerelatedtoenvironmentalfactors,accountingforthevariation

between years. At the individual level, our study emphasized theimportanceofspringmigrationdeparturetimeandpointedoutin-creasedflighteffortforearlymigrants,aswellasapotentialtrade-offbetweenearlyarrivalandwinteringhabitatquality.Consideringthesignificanceofspringdeparturetiming,moreknowledgeontheendogenousandexternalmechanismsthatmodulatebirddeparturedecisionsisrequired.This,togetherwiththeinsightsonarrivaldatereportedhere,canfacilitateforecastingmigratingbirds’responsesunderclimatechangescenarios.

ACKNOWLEDG EMENTS

We thankH.G.Benecke,T. Schaffer, andW.Sender andhis crewintheDrömlingNatureParkfortheiressentialhelpinthefield;W.Heidrich and F. Kuemmeth from e-obs GmbH for their dedicatedtechnical support.We acknowledge the generous funding of DIPgrants(DFG)NA846/1-1andWI3576/1-1toRN,FJ,andMW.Thisstudy was also supported by the Minerva Center for MovementEcologygrantedtoR.N.S.R.wassupportedbyadoctoralbirdstudyscholarshipoftheMinistryofScienceandTechnology,Israel.

AUTHOR’ S CONTRIBUTIONS

S.R.,R.N.,F.J.andM.W.conceivedthe idea.S.R.andM.K.carriedout the fieldworkwith the help of S.F., U.E.,M.W. andD.Z. S.R.wrotethefirstdraftandallauthorscontributedtotherevisions.

DATA ACCE SSIBILIT Y

TheGPS-ACCdataareavailableinMovebankDataRepositorywithdoi:https://doi.org/10.5441/001/1.v8d24552(Roticsetal.,2018).

ORCID

Shay Rotics http://orcid.org/0000-0002-3858-1811

Damaris Zurell http://orcid.org/0000-0002-4628-3558

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

Barbraud,C.,&Barbraud, J.C. (1999). Is thereageassortativematingintheEuropeanwhitestork?Waterbirds: The International Journal of Waterbird Biology,22,478–481.https://doi.org/10.2307/1522129

Barbraud,C.,Barbraud,J.-C.,&Barbraud,M.(1999).Populationdynam-ics of theWhite StorkCiconia ciconia inwestern France. Ibis,141,469–479.

Barton, K. (2016). MuMIn: Multi-Model Inference. R package version1.15.6.

Berthold,P.,Kaatz,M.,&Querner,U.(2004).Long-termsatellitetrackingofwhitestork(Ciconia ciconia)migration:Constancyversusvariabil-ity. Journal of Ornithology, 145, 356–359. https://doi.org/10.1007/s10336-004-0049-2

Blums,P.,Nichols,J.D.,Hines,J.E.,Lindberg,M.S.,&Mednis,A.(2005).Individual quality, survival variation and patterns of phenotypic

1636  |    Journal of Animal Ecology ROTICS eT al.

selectiononbodyconditionandtimingofnestinginbirds.Oecologia,143,365–376.https://doi.org/10.1007/s00442-004-1794-x

Both,C.,Bouwhuis, S., Lessells,C.M.,&Visser,M.E. (2006).Climatechange and population declines in a long-distancemigratory bird.Nature,441,81–83.https://doi.org/10.1038/nature04539

Cadahia,L.,Labra,A.,Knudsen,E.,Nilsson,A.,Lampe,H.M.,Slagsvold,T., & Stenseth, N. C. (2017). Advancement of spring arrival in along-term study of a passerine bird: Sex, age and environmen-tal effects. Oecologia, 184, 917–929. https://doi.org/10.1007/s00442-017-3922-4

Cheriak, L., Barbraud, C.,Doumandji, S., & Bouguessa, S. (2014).Dietvariability in the White Stork Ciconia ciconia in eastern Algeria.Ostrich,85, 201–204. https://doi.org/10.2989/00306525.2014.971451

Chevallier, D., Handrich, Y., Georges, J. Y., Baillon, F., Brossault, P.,Aurouet,A.,…Massemin,S.(2010).Influenceofweatherconditionsontheflightofmigratingblackstorks.Proceedings of the Royal Society B- Biological Sciences, 277, 2755–2764. https://doi.org/10.1098/rspb.2010.0422

Cooper,N.W.,Sherry,T.W.,&Marra,P.P.(2015).Experimentalreduc-tionofwinterfooddecreasesbodyconditionanddelaysmigrationina long-distancemigratorybird.Ecology,96,1933–1942.https://doi.org/10.1890/14-1365.1

Dittmann,T.,&Becker,P.H.(2003).Sex,age,experienceandconditionasfactorsaffectingarrivaldateinprospectingcommonterns,Sterna hirundo. Animal Behaviour, 65, 981–986. https://doi.org/10.1006/anbe.2003.2128

Dodge, S., Bohrer, G., Weinzierl, R., Davidson, S., Kays, R., Douglas,D., … Wikelski, M. (2013). The environmental-data automatedtrack annotation (Env-DATA) system: Linking animal trackswith environmental data. Movement Ecology, 1, 3. https://doi.org/10.1186/2051-3933-1-3

Duerr,A.E.,Miller,T.A.,Lanzone,M.,Brandes,D.,Cooper,J.,O’Malley,K., … Katzner, T. (2015). Flight response of slope-soaring birds toseasonalvariationinthermalgeneration.Functional Ecology,29,779–790.https://doi.org/10.1111/1365-2435.12381

Fulin,M.,Jerzak,L.,Sparks,T.H.,&Tryjanowski,P.(2009).Relationshipbetween arrival date, hatching date and breeding success of thewhitestork(Ciconia ciconia)inSlovakia.Biologia,64,361–364.

Gordo,O. (2007).Why are birdmigration dates shifting? A review ofweatherandclimateeffectsonavianmigratoryphenology.Climate Research,35,37–58.https://doi.org/10.3354/cr00713

Gordo,O.,&Sanz, J. J. (2006).Climate changeandbirdphenology:Along-termstudyintheIberianPeninsula.Global Change Biology,12,1993–2004.https://doi.org/10.1111/j.1365-2486.2006.01178.x

Gordo,O.,Tryjanowski,P.,Kosicki, J.Z.,&Fulin,M. (2013).Complexphenological changes and their consequences in the breed-ing success of a migratory bird, the white stork Ciconia cico-nia. Journal of Animal Ecology, 82, 1072–1085. https://doi.org/10.1111/1365-2656.12084

Gunnarsson,T.G.,Gill,J.A.,Atkinson,P.W.,Gelinaud,G.,Potts,P.M.,Croger,R.E.,…Sutherland,W.J.(2006).Population-scaledriversofindividualarrivaltimesinmigratorybirds.Journal of Animal Ecology,75,1119–1127.https://doi.org/10.1111/j.1365-2656.2006.01131.x

Gwinner,E.(1996).Circadianandcircannualprogrammesinavianmigra-tion.Journal of Experimental Biology,199,39–48.

Harel,R.,Duriez,O., Spiegel,O., Fluhr, J.,Horvitz,N.,Getz,W.M.,…Nathan, R. (2016). Decision-making by a soaring bird: Time, en-ergy and risk considerations at different spatio-temporal scales.Philosophical Transactions of the Royal Society B- Biological Sciences,371,2201050397.

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

Huppop,O.,&Huppop,K.(2003).NorthAtlanticOscillationandtimingofspringmigrationinbirds.Proceedings of the Royal Society B- Biological Sciences,270,233–240.https://doi.org/10.1098/rspb.2002.2236

Janiszewski,T.,Minias,P.,&Wojciechowski,Z.(2013).Reproductivecon-sequences of early arrival at breeding grounds in theWhite StorkCiconia ciconia. Bird Study,60,280–284.https://doi.org/10.1080/00063657.2013.778227

Kentie,R.,Marquez-Ferrando,R.,Figuerola,J.,Gangoso,L.,Hooijmeijer,J.,Loonstra,A.H.J.,…Piersma,T.(2017).DoeswinteringnorthorsouthoftheSaharacorrelatewithtimingandbreedingperformanceinblack-tailedgodwits?Ecology and Evolution,7,2812–2820.https://doi.org/10.1002/ece3.2879

Klaassen, R. H. G., Strandberg, R., Hake, M., & Alerstam, T. (2008).Flexibility in daily travel routines causes regional variation in birdmigrationspeed.Behavioral Ecology and Sociobiology,62,1427–1432.https://doi.org/10.1007/s00265-008-0572-x

Kosicki,J.,Sparks,T.,&Tryjanowski,P. (2004).Doesarrivaldate influ-ence autumn departure of theWhite Stork Ciconia ciconia?Ornis Fennica,81,91–95.

Kumar, V., Wingfield, J. C., Dawson, A., Ramenofsky, M., Rani, S., &Bartell,P.(2010).Biologicalclocksandregulationofseasonalrepro-ductionandmigrationinbirds.Physiological and Biochemical Zoology,83,827–835.https://doi.org/10.1086/652243

Lemke, H.W., Tarka,M., Klaassen, R. H. G., Akesson,M., Bensch, S.,Hasselquist,D.,&Hansson, B. (2013). Annual cycle andmigrationstrategies of a Trans-Saharan migratory songbird: A geolocatorstudyintheGreatReedWarbler.PLoS ONE,8,e79209.https://doi.org/10.1371/journal.pone.0079209

Lenth,R.V.(2016).Least-squaresmeans:TheRpackagelsmeans.Journal of Statistical Software,69,1–33.

Leshem,Y.,&YomTov,Y.(1996).Theuseofthermalsbysoaringmigrants.Ibis,138,667–674.https://doi.org/10.1111/j.1474-919X.1996.tb04768.x

Lok,T.,Overdijk,O.,Tinbergen,J.M.,&Piersma,T.(2011).Theparadoxofspoonbillmigration:Mostbirdstraveltowheresurvivalratesarelowest. Animal Behaviour, 82, 837–844. https://doi.org/10.1016/j.anbehav.2011.07.019

Lok,T.,Veldhoen,L.,Overdijk,O.,Tinbergen,J.M.,&Piersma,T.(2017).Anage-dependentfitnesscostofmigration?Oldtrans-Saharanmi-gratingspoonbillsbreedlaterthanthosestayinginEurope,andlatebreedershavelowerrecruitment.Journal of Animal Ecology,86,998–1009.https://doi.org/10.1111/1365-2656.12706

Marra, P. P., Francis, C. M., Mulvihill, R. S., & Moore, F. R. (2005).The influence of climate on the timing and rate of spring birdmi-gration. Oecologia, 142, 307–315. https://doi.org/10.1007/s00442-004-1725-x

Martín, B., Onrubia, A., de la Cruz, A., & Ferrer, M. (2016). Trendsof autumn counts at Iberian migration bottlenecks as a tool formonitoring continental populations of soaring birds in Europe.Biodiversity and Conservation,25,295–309.https://doi.org/10.1007/s10531-016-1047-4

Massemin-Challet, S., Gendner, J. P., Samtmann, S., Pichegru,L., Wulgue, A., & Le Maho, Y. (2006). The effect of migra-tion strategy and food availability on White Stork Ciconia ciconia breeding success. Ibis, 148, 503–508. https://doi.org/10.1111/j.1474-919X.2006.00550.x

Matyjasiak,P.(2013).Timingofarrivalfromspringmigrationisassociatedwith flight performance in themigratory barn swallow.Behavioral Ecology and Sociobiology, 67, 91–100. https://doi.org/10.1007/s00265-012-1429-x

McKinnon,E.A.,Macdonald,C.M.,Gilchrist,H.G.,&Love,O.P.(2016).Spring and fall migration phenology of an Arctic-breeding passer-ine. Journal of Ornithology, 157, 681–693. https://doi.org/10.1007/s10336-016-1333-7

Menzel, A., Sparks, T. H., Estrella, N., Koch, E., Aasa, A., Ahas, R., …Zust,A.(2006).Europeanphenologicalresponsetoclimatechange

     |  1637Journal of Animal EcologyROTICS eT al.

matchesthewarmingpattern.Global Change Biology,12,1969–1976.https://doi.org/10.1111/j.1365-2486.2006.01193.x

Møller, A. P. (1994). Phenotype-dependent arrival time and its conse-quencesinamigratorybird.Behavioral Ecology and Sociobiology,35,115–122.https://doi.org/10.1007/BF00171501

Monros, J. S., Belda, E. J., & Barba, E. (2002). Post-fledgingsurvival of individual great tits: The effect of hatchingdate and fledging mass. Oikos, 99, 481–488. https://doi.org/10.1034/j.1600-0706.2002.11909.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

Nevoux, M., Barbraud, J. C., & Barbraud, C. (2008). Nonlinear im-pact of climate on survival in a migratory white stork popu-lation. Journal of Animal Ecology, 77, 1143–1152. https://doi.org/10.1111/j.1365-2656.2008.01435.x

Newton,I.(2008).The migration ecology of birds,pp.399–456,617–637.London,UK:AcademicPress,Elsevier.

Norris,D.R.,Marra,P.P.,Kyser,T.K.,Sherry,T.W.,&Ratcliffe,L.M.(2004). Tropical winter habitat limits reproductive success onthe temperate breeding grounds in a migratory bird. Proceedings of the Royal Society B- Biological Sciences, 271, 59–64. https://doi.org/10.1098/rspb.2003.2569

Nourani,E.,&Yamaguchi,N.M.(2017).Theeffectsofatmosphericcurrentson the migratory behavior of soaring birds: A review.Ornithological Science,16,5–15.https://doi.org/10.2326/osj.16.5

Ouwehand, J., & Both, C. (2017). African departure rather than mi-gration speed determines variation in spring arrival in pied fly-catchers. Journal of Animal Ecology, 86, 88–97. https://doi.org/10.1111/1365-2656.12599

Pawlowicz, R. (2009). SUNCYCLE code adapted from: AIR_SEATOOLBOX (version2.0:8/9/99)basedonAppendixE in the1978edition of Almanac for Computers, Nautical Almanac Office, U.S.NavalObservatory.

Paxton,K.L.,&Moore,F.R.(2015).Carry-overeffectsofwinterhabi-tatqualityonenroutetimingandconditionofamigratorypasser-ine during springmigration. Journal of Avian Biology,46, 495–506.https://doi.org/10.1111/jav.00614

Rotics, S., Kaatz,M., Feldman, S., Zurell,D.,Wikelski,M., Sapir,N.,…Nathan, R. (2018). Data from: Early arrival at breeding grounds:causes,costsandatrade-offwithoverwinteringlatitude.MovebankData Repository.https://doi.org/10.5441/001/1.v8d24552

Rotics,S.,Kaatz,M.,Resheff,Y.S.,Turjeman,S.F.,Zurell,D.,Sapir,N.,…Nathan,R.(2016).Thechallengesofthefirstmigration:Movementandbehaviourofjuvenilevs.adultwhitestorkswithinsightsregard-ingjuvenilemortality.Journal of Animal Ecology,85,938–947.https://doi.org/10.1111/1365-2656.12525

Rotics,S.,Turjeman,S.,Kaatz,M.,Resheff,Y.S.,Zurell,D.,Sapir,N.,…Nathan,R. (2017).Wintering inEurope insteadofAfricaenhancesjuvenile survival in a long-distancemigrant.Animal Behaviour,126,79–88.https://doi.org/10.1016/j.anbehav.2017.01.016

Saino, N., Ambrosini, R., Rubolini, D., vonHardenberg, J., Provenzale,A., Huppop, K., … Sokolov, L. (2011). Climate warming, ecologi-cal mismatch at arrival and population decline in migratory birds.Proceedings of the Royal Society B- Biological Sciences,278,835–842.https://doi.org/10.1098/rspb.2010.1778

Saino,N.,Rubolini,D.,Serra,L.,Caprioli,M.,Morganti,M.,Ambrosini,R.,&Spina,F. (2010).Sex-relatedvariation inmigrationphenologyinrelationtosexualdimorphism:Atestofcompetinghypothesesfortheevolutionofprotandry.Journal of Evolutionary Biology,23,2054–2065.https://doi.org/10.1111/j.1420-9101.2010.02068.x

Sapir,N.,Wikelski,M.,McCue,M.D.,Pinshow,B.,&Nathan,R.(2010).Flight modes in migrating european bee-eaters: Heart rate may

indicatelowmetabolicrateduringsoaringandgliding.PLoS ONE,5,e13956.https://doi.org/10.1371/journal.pone.0013956

Schlaich,A.E.,Klaassen,R.H.G.,Bouten,W.,Bretagnolle,V.,Koks,B., Villers, A., & Both, C. (2016). How individual Montagu’sHarriers cope with Moreau’s Paradox during the Sahelian win-ter. Journal of Animal Ecology, 85, 1491–1501. https://doi.org/10.1111/1365-2656.12583

Schmaljohann,H.,Meier,C.,Arlt,D.,Bairlein,F.,vanOosten,H.,Morbey,Y. E., … Eikenaar, C. (2016). Proximate causes of avian protandrydiffer between subspecies with contrasting migration challenges.Behavioral Ecology, 27, 321–331. https://doi.org/10.1093/beheco/arv160

Sergio, F., Tanferna, A., De Stephanis, R., Jimenez, L. L., Blas, J.,Tavecchia, G., … Hiraldo, F. (2014). Individual improvements andselectivemortalityshape lifelongmigratoryperformance.Nature,515,410–413.

Shamoun-Baranes, J.,Baharad,A.,Alpert,P.,Berthold,P.,Yom-Tov,Y.,Dvir, Y.,& Leshem,Y. (2003). The effect ofwind, season and lati-tudeonthemigrationspeedofwhitestorksCiconia ciconia,alongtheeasternmigrationroute.Journal of Avian Biology,34,97–104.https://doi.org/10.1034/j.1600-048X.2003.03079.x

Shamoun-Baranes,J.,Burant,J.B.,Loon,E.E.,Bouten,W.,&Camphuysen,C. J. (2017). Short distancemigrants travel as far as longdistancemigrants in lesser black-backed gulls Larus fuscus. Journal of Avian Biology,48,49–57.https://doi.org/10.1111/jav.01229

Smith,R.J.,&Moore,F.R.(2005).Arrivaltimingandseasonalreproduc-tive performance in a long-distancemigratory landbird.Behavioral Ecology and Sociobiology, 57, 231–239. https://doi.org/10.1007/s00265-004-0855-9

Sokolov,L.V.,&Tsvey,A.L.(2016).Mechanismscontrollingthetimingofspringmigrationinbirds.Biology Bulletin,43,1148–1160.https://doi.org/10.1134/S1062359016110145

Stoffel, M. A., Nakagawa, S., & Schielzeth, H. (2017). rptR:Repeatabilityestimationandvariancedecompositionbygeneral-izedlinearmixed-effectsmodels.Methods in Ecology and Evolution,8,1639–1644.

Tortosa,F. S.,&Redondo,T. (1992). Frequent copulationsdespite lowspermcompetition inwhitestorks (Ciconia ciconia).Behaviour,121,288–315.https://doi.org/10.1163/156853992X00408

Tottrup,A.P.,Klaassen,R.H.G.,Strandberg,R.,Thorup,K.,Kristensen,M.W., Jorgensen,P.S.,…Alerstam,T. (2012).Theannualcycleofa trans-equatorial Eurasian-African passerine migrant: Differentspatio-temporal strategies for autumn and spring migration.Proceedings of the Royal Society B- Biological Sciences,279,1008–1016.https://doi.org/10.1098/rspb.2011.1323

Usui,T.,Butchart,S.H.M.,&Phillimore,A.B. (2017).Temporal shiftsandtemperaturesensitivityofavianspringmigratoryphenology:Aphylogeneticmeta-analysis.Journal of Animal Ecology,86,250–261.https://doi.org/10.1111/1365-2656.12612

Vaitkuviene,D.,Dagys,M.,Bartkeviciene,G.,&Romanovskaja,D.(2015).TheeffectofweathervariablesontheWhiteStork(Ciconia ciconia)springmigrationphenology.Ornis Fennica,92,43–52.

Vansteelant,W.M.G.,Bouten,W.,Klaassen,R.H.G.,Koks,B.J.,Schlaich,A.E.,vanDiermen,J.,…Shamoun-Baranes,J.(2015).Regionalandseasonal flightspeedsofsoaringmigrantsandthe roleofweatherconditions at hourly and daily scales. Journal of Avian Biology, 46,25–39.https://doi.org/10.1111/jav.00457

Verboven,N.,&Visser,M.E.(1998).Seasonalvariationinlocalrecruit-mentofgreattits:Theimportanceofbeingearly.Oikos,81,511–524.https://doi.org/10.2307/3546771

Vergara,P.,Aguirre,J.I.,&Fernandez-Cruz,M.(2007).Arrivaldate,ageandbreedingsuccessinwhitestorkCiconia ciconia. Journal of Avian Biology,38,573–579.https://doi.org/10.1111/j.0908-8857.2007.03983.x

1638  |    Journal of Animal Ecology ROTICS eT al.

Weimerskirch,H.,Borsa,P.,Cruz,S.,deGrissac,S.,Gardes,L.,Lallemand,J.,…Prudor,A.(2017).Diversityofmigrationstrategiesamonggreatfrigatebirds populations. Journal of Avian Biology, 48, 103–113.https://doi.org/10.1111/jav.01330

Wilson,R.P.,White,C.R.,Quintana,F.,Halsey,L.G.,Liebsch,N.,Martin,G. R.,&Butler, P. J. (2006).Moving towards acceleration for esti-matesofactivity-specificmetabolicrate infree-livinganimals:Thecase of the cormorant. Journal of Animal Ecology, 75, 1081–1090.https://doi.org/10.1111/j.1365-2656.2006.01127.x

Wuczynski,A. (2005).TheturnoverofWhiteStorksCiconia ciconia on nestsduringspringmigration.Acta Ornithologica,40,83–85.https://doi.org/10.3161/068.040.0104

Yamamoto,T.,Takahashi,A.,Sato,K.,Oka,N.,Yamamoto,M.,&Trathan,P.N. (2014). Individualconsistencyinmigratorybehaviourofape-lagicseabird.Behaviour,151,683–701.https://doi.org/10.1163/1568539X-00003163

Zuur,A.F.,Ieno,E.N.,&Elphick,C.S.(2010).Aprotocolfordataexplo-rationtoavoidcommonstatisticalproblems.Methods in Ecology and Evolution,1,3–14.

SUPPORTING INFORMATION

Additional supporting information may be found online in theSupportingInformationsectionattheendofthearticle.

How to cite this article:RoticsS,KaatzM,TurjemanS,etal.Earlyarrivalatbreedinggrounds:Causes,costsandatrade-offwithoverwinteringlatitude.J Anim Ecol. 2018;87:1627–1638. https://doi.org/10.1111/1365-2656.12898