challenges in the conservation of wide‐ranging nomadic

J Appl Ecol 20191ndash11 wileyonlinelibrarycomjournaljpe emsp|emsp1copy 2019 The Authors Journal of Applied Ecology copy 2019 British Ecological Society

Received12May2018emsp |emsp Accepted3February2019DOI1011111365-266413380

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

Challenges in the conservation of wide- ranging nomadic species

Dejid Nandintsetseg12 emsp| Chloe Bracis123emsp| Kirk A Olson45emsp| Katrin Boumlhning-Gaese12emsp| Justin M Calabrese56 emsp| Buyanaa Chimeddorj7emsp| William F Fagan68emsp| Christen H Fleming56emsp| Michael Heiner9emsp| Petra Kaczensky1011emsp| Peter Leimgruber5emsp| Dalannast Munkhnast12emsp| Theresa Stratmann12emsp| Thomas Mueller125

1SenckenbergBiodiversityandClimateResearchCentreSenckenbergGesellschaftfuumlrNaturforschungFrankfurt(Main)Germany2DepartmentofBiologicalSciencesGoetheUniversityFrankfurtFrankfurt(Main)Germany3IfremerChannelandNorthSeaFisheriesResearchUnitBoulogne-sur-MerFrance4WildlifeConservationSocietyMongoliaCountryProgramUlaanbaatarMongolia5SmithsonianConservationBiologyInstituteNationalZoologicalParkFrontRoyalVirginia6DepartmentofBiologyUniversityofMarylandCollegeParkMaryland7WorldWideFundforNatureMongoliaProgramUlaanbaatarMongolia 8SESYNCUniversityofMarylandAnnapolisMaryland9TheNatureConservancyFortCollinsColorado10NorwegianInstituteforNatureResearchTrondheimNorway11ResearchInstituteofWildlifeEcologyUniversityofVeterinaryMedicineViennaAustriaand12BatResearchCentreofMongoliaUlaanbaatarMongolia

CorrespondenceDejidNandintsetsegEmaildnandintsetseggmailcom

Funding informationBundesministeriumfuumlrBildungundForschungGrantAwardNumber01LC1710Aand01LC1820ARobertBoschFoundation

HandlingEditorRichardFuller

Abstract1 Conservationofnomadicspeciespresentssignificantconservationchallengesbe-causeofunpredictabilityintheirmovementsandspaceuseLong-termstudiesonnomadicspeciesofferinginsightsintothevariabilityinspaceusewithinandbe-tweenyearsarelargelymissingbutarenecessarytodevelopeffectiveconserva-tionstrategies

2 We examined the temporal variability in space-use of Mongolian gazelle a nomadicspeciesWetracked22individualsfor1ndash3yearswithGPSandusedtheresultingmovementpatternstoevaluateconservationstrategiesassociatedwiththeir nomadic movements in the intact open plain grasslands of MongoliaIndividualsexhibitedahighdegreeofvariabilityinspaceusewithinandbetweenyearsoftenusingdifferentwinteringareasindifferentyearsTheindividualrangesizevariedasmuchasthreefoldbetweenyearswithanestimatedaverageannualindividualrangesizeof~19000km2andalifetimerangeof~100000km2

3 Comparing simulated and empiricalGPS trajectories for theMongolian gazelleshowedthattheyavoideddisturbedareas(egoilfields)anddidnotpreferpro-tectedareasImportantlynosingleprotectedareaintheregionwaslargeenoughtocovertheannualrangeofanyofthetrackedgazelle

4 Becauseoftheirwide-rangingmovementsthepresenceoflinearinfrastructureandtheresultingbarriereffectsareaparticularconcernWefoundthatfencesalongthenationalborderwereabsolutebarriersaffectingmovementsofabout80of all tracked individualsWhen gazelle encounter the border fence theymoved amedian distance of 11km along fences suggesting frequent crossing optionsareneededtomakebarrierspermeable

2emsp |emsp emspenspJournal of Applied Ecology NANDINTSETSEG ET Al

1emsp |emspINTRODUC TION

Integrating animal movements into conservation management iscriticaltoconservationsuccess(AllenampSingh2016Mcgowanetal2017)Conservationchallengesaremagnifiedforhighlymobilespe-cies that have large ranges and high spatiotemporal variability inspaceuseSpatiotemporalvariabilityand itspotential implicationsfor conservation management remain understudied in large partduetoa lackofsuitabledatasets (KaysCrofoot JetzampWikelski2015RungeMartinPossinghamWillisampFuller2014)

ConsiderationofmovementstrategiessuchasrangeresidencyandmigrationiscrucialtoprotectingwildlifeForexampleconser-vationof long-distancemigrantswithpredictablemigrationroutescan be facilitated through corridors that maintain connectivitybetween seasonal habitats (Sawyer Kauffman Nielson amp Horne2009)Inothercasesmanagementeffortsfocusonprotectingareasthatareregularlyusedduringkeypartsofthespeciesrsquolifecycle(egbreedingandwinteringareasGeldmannetal2013)

Whilemigratoryandrangeresidentspeciesarewellstudiedfewstudies address nomadic species and even fewer consider conser-vationplanning for them (Rungeetal2014)Nomadic species likeThomsonsgazelleEudorcas thomsoniiintheSerengetiPlainsorkhulanEquus hemionus in theGobiDesert are typically found in resource-pooraridenvironmentswithdynamicresources(McNaughton1976NandintsetsegKaczenskyGanbaatarLeimgruberampMueller2016)Thekeycharacteristicsofnomadicmovementsarenon-seasonalspa-tialvariabilityandtemporalunpredictability in interannualspaceuse(Jonzeacutenamp Knudsen 2011) However few studies on nomadic spe-cies examine the predictability of space use bymultiple individualsmonitoredovermorethan1yearwhich iskey informationfor theirlong-termsurvivalStudiesontheuseofprotectedareas(PAs)ortheeffectsofdisturbanceonnomadicspeciesarelikewisescarce

HerewestudytheMongoliangazelleProcapra gutturosa(here-aftergazelle)aprominentexampleofanomadicungulate(Muelleretal2008Olsonetal2010)andoneofthemostnumerousgazelle

speciesgloballyTheyarenativetotheopensteppeofMongoliaandtheadjacentareasofRussiaandChina(Mallon2008)Over95oftheglobalpopulationofgazelleoccursinMongoliassteppeBasedona2002estimatedpopulationsizeof1milliongazelle(Olsonetal2010) thespecies is listedasLeastConcern in the IUCNRedList(Mallon2008)Althoughthepopulationtrendisassessedasstablein the IUCNRedList therehasbeenno regional scalepopulationcensussince2002andthusthepopulationtrendisunknownThepopulationdoesundergosubstantialfluctuationsinabundanceduetooverhuntingdiseaseoutbreaksandextremeweathereventsandgazelleareconsideredasendangeredinMongoliasRedList (Clarketal2006)

Gazelle undertake long-distance movements driven by highinter- and intraannual variability in resource availability in thesteppe (Mueller etal 2008) Observed group sizes range from asingleindividualtomorethan200000individuals(OlsonMuellerBolortsetseg etal 2009) An individual gazelle can roam over32000km2in1year(Olsonetal2010)Thespeciesrsquocoredistribu-tionrangeineasternMongoliais200000km2plusmn3100km2(Flemingetal2014)Apartfromattentiongiventoannualrangesizesrela-tively little isknownabout the lifetimespace requirementsor thepredictabilityofareasusedbyindividualgazelleparticularlyatcriti-calstagesoftheirlifecycle(egcalvingandwinteringperiods)

AmosaicofPAswasestablishedinparttosafeguardremaininggazelle populations in easternMongoliawhich are threatened bypoachingcompetitionwithlivestockandhabitatlossandfragmenta-tion(OlsonampFuller2017Figure1)TheseunfencedPAsoftenhaveahabitatcharacteristicssimilar to their immediatevicinityFencesalongsidetheTrans-MongolianRailwayandMongoliasnationalbor-derareprovenbarriersforgazellefurtherfragmentingtheirhabitatoutsidethespeciesrsquocorerange (Itoetal2008)Furthermore therapidlygrowingextractiveindustryinMongoliahasledtoproposalsto constructnew railwaysMongoliahas a statepolicyon railwaytransportation (Parliament resolutionno32 in2010)whichplanstheconstructionof56835kmofnewrailways(GansukhMingamp

5 Synthesis and applicationsWe show that for nomadic specieswhose space use variesgreatlyacrossyearsmultiyearmovementdataareessentialforsoundcon-servationplanningWeemphasizethatplace-basedapproachesaloneareinsuffi-cient to conserve wide-ranging nomadic species and that linear infrastructureincludingfenceshighwaysandrailroadsisofparticularconcernBecausenomadicspecieslackdefinedmovementcorridorsweadvocateintegratedlanduseplan-ningthatprioritizespermeabilityacrosstheentirelandscapeandfacilitateslong-distancemovementsWesuggestthatconservationstrategiesfornomadicspeciesinaridandsemi-aridregionsbereconsideredbasedonmultiyearconnectivityas-sessmentsatthelandscapescale

K E Y W O R D S

animalmovementcrossingstructurelandusenomadicpermeabilityprotectedareaungulatewide-rangingspecies

emspensp emsp | emsp3Journal of Applied EcologyNANDINTSETSEG ET Al

Ali2018)androadcorridorsacrossthecorerangesofseveralungu-latesincludinggazelle(Figure1)Theseemergingdevelopmentswillincreasehabitatfragmentationandlikelyresultinadditionalmove-mentbarrierstogazellemovements(Batsaikhanetal2014)

Hereweusedmultiple-yearGPSmovementdataof gazellein the Eastern Steppe to examine spatiotemporal variability inspace use relative to conservation measures for gazelle Firstwe measured the variability in space use among individuals inthesameyearandwithinindividualsbetweenyearsSecondweevaluatedcurrentconservationmeasuresforgazellebyexamin-ingtheuseofthePAsandbyassessingtheeffectsofanthropo-genicdisturbanceongazellemovementsLastlyweassessedtheeffectoflinearinfrastructureongazellemovementsandproviderecommendations for the future conservationof gazellewhichcanbe relevant for otherwide-rangingnomadic species in aridandsemi-aridregions

2emsp |emspMATERIAL S AND METHODS

21emsp|emspStudy area

Ourstudywasconducted in theEasternSteppeofMongoliaoneofthelargesttemperategrasslandsintheworld(Figure1CarbuttHenwoodampGilfedder2017)ThesteppefeaturesbroadplainsandrollinghillsdominatedbygrassesPrecipitation(~200mmannually)canbehighlyvariableacrossspaceandtimeresultinginhighvaria-bilityinvegetationproductivity(VandandorjGantsetsegampBoldgiv2015)Landusepolicyfavourscommunalusebytraditionallivestockpastoralistsalthoughotherlanduseoccurs(egagriculturesettle-mentsmining)

22emsp|emspEmpirical movement data

Weuseddatafrom22gazelle(12collaredinOctober2014Sample110inSeptember2015Sample2)whichweremonitoredfor1ndash3yearsprovidingatotalof12166dailyGPSpositions(Figure1detailsinTableS1andananimationofthemovementsinAnimationS1)Althoughourdatasetis limitedto22individuals itcurrentlyconstitutesthelargestlong-termdatasetofgazelle(Nandintsetsegetal2019)

23emsp|emspSimulated movement model

Weusedsimulatedrandommovementpathsasanullmodeltoassessthreeobjectives(a)thepredictabilityofspaceusebetweenyears(b)theuseofPAsand(c)theeffectofdisturbanceongazellespaceuseWegeneratedcorrelatedrandomwalksbydrawingsteplengthsandturningangles fromtheempiricaldistributionsofeachGPStrackedgazelleWedrewdailystepsforeachindividualstrackingperiodWesimulatedgapsintheserandompathsbyremovingpositionsfromthefullsimulatedtrajectoriescorrespondingtothegapsintheempiricalmovementpathsThisresultedinsimulatedpathsthathadthesamenumberofpositions as theempirical pathsWebounded simulatedpaths by the Mongolian border and the Trans-Mongolian RailwayfencesthatrestrictgazellemovementsWesimulated1000replicatepathsforeachindividualForfurtherdetailsseeDescriptionS1

24emsp|emspRange estimation and temporal variability in space use

Toexaminevariabilityinannualspaceuseofgazelleamongindividu-alsandacrossyearsweusedminimumconvexpolygonsasametric

F IGURE 1emspMongoliangazelleGPSlocationswithexistingandproposeddisturbancesandprotectedareasinthecorerangeofMongoliangazelleinMongolia


of annual spaceuseper individual (95MCPmaptool r package Calenge2017)WeusedMCPbecauseitcanbecomparedtoprevi-ous studies anddatadensities across individualswere similarWecalculatedannualrangesforeachgazelleresultinginatotalof40annualrangesNotallgazellesurvivedthefull3yearstherewere2213and5 individualswith12and3yearsof trackingdatare-spectively(TableS1)

We used Autocorrelated Kernel Density Estimation (AKDEctmm041rpackage)anovelhomerangeestimatorthatallowstheuseofautocorrelatedmovementdata toreliablyestimate lifetimeranges as described in Fleming etal (2015) Lifetime range refersto predictions of range use beyond the tracking period assumingthatthemovementbehaviourstaysconstant(CalabreseFlemingampGurarie2016Flemingetal2015)Wecouldnotestimatethelife-timerangesusingAKDEfor10individuals(thosetrackedforasingleyear) because the semi-variance (ie the average square displace-ment vs time-lag) didnot approach an asymptotewith increasingtime-lags that is they likelyhadnotbeen tracked longenough toallowforaccurateestimates

In addition we examined the predictability of an individualsspaceuseWespecificallywereinterestedinhowthepredictabilityinspaceusevariedbetweenwinteringandcalvingperiodsWeeval-uated this variability for12 femalegazelleby calculating thepair-wisedistancesbetweencentroidsofasequenceof14-dayintervalsthroughouttheyearthuscomparingthesametimeofyearbetweenyearsforeachgazellewithatleast2yearsofdataWeusedapairedt test tocompareaveragepairwisedistancesbetweencalvingandwinteringperiodsforeachindividualDuetosynchronizedandshortbirthingandnursingperiods(OlsonFullerSchallerLhagvasurenampOdonkhuu2005)wedefinedthecalvingperiodasthetimespan-ning25Juneuntil23JulyTomatchthenumberofdaysforcalvingandwinteringperiodswealsoselectedthree14-dayintervalsinthemiddleofwinteranddefinedthewinteringperiodas8Januaryto5FebruaryTocomparethewinteringandcalvingmeanpairwisedis-tancesof the taggedgazelle to thenullmodelwealsocalculatedthepairwisedistanceinthesamewayforeachsimulatedindividualtrajectoryWethenestimatedthep-valuewitharandomizationtesttodetermineifthemeanpairwisedistancesoftaggedgazelleduringwinteringandcalvingperiodsweresignificantlydifferent thanthenulldistributionsbasedonthesimulatedpaths(FigureS3)

25emsp|emspAssessing use of PAs and effects of disturbance on space use

WeevaluatedthegazellersquouseofPAsbycomparingtheproportionofdailyGPSpositions insidePAs foreachgazelle (ie the ratioofthenumberofgazelledailyGPSpositionsinsidethePAstothetotalnumberofdailyGPSpositions) tothatof thesimulatedpathsWethenusedarandomizationtesttoassesswhetherthemedianpro-portionofPAuseacrossalltaggedgazellewassignificantlygreaterthanthedistributionofmedianPAusebythesimulatedindividualsunderthehypothesisthattaggedgazelleusePAsmorethanthenullmodel

Tomeasure effects of human-induced disturbances on gazellemovementsweusedacumulativedisturbance index layer for theEastern Steppe (Heiner etal 2016) This layerwas created usingfive anthropogenic factors including herder household locationsagriculturaluseexistingminingareaspopulationcentresand lin-earinfrastructuresuchasroadsandrailwaysThedisturbanceindexrangesfrom0 (no disturbance) to1 (high disturbance)Foreachga-zelleweextracted thedisturbance indexpixel valuesateachpo-sition and used the median across locations to characterize thedisturbance experienced by gazelleWe applied the same proce-dure to the1000 simulatedgazellepaths for each individualWethencompared the rangeofmediansof thedisturbance index forthesimulatedpathstothemedianofthedisturbance indexofthetaggedgazelleUsingarandomizationtestweassessedwhetherthemediandisturbance indexacrossallgazellewassignificantly lowerthanthatforsimulatedpathsunderthehypothesisthattaggedga-zelleusedisturbedareaslessthanthesimulationmodelwhichhasnoavoidancebehaviour

26emsp|emspEffects of linear features on gazelle movements

Toexplorehowgazellereacttolinearbarriersweexaminedtheef-fectsoftheMongolianborderfenceongazellemovementsWhengazelleapproach theborder fence they frequently travelalong itpresumablytryingtocrossWecalculatedthetraveldistancesalongtheborder fence for eachencounterwith theborderWeconsid-eredpositionswithin5kmofthebordertobefenceencountersandtreatedallsubsequentpositionswithin10kmoftheborderfenceaspartofthesameencountertravellingalongtheborderbeforegivingup (Figure5b)Weselected5kmas the thresholdbasedonvisuallineofsightdistancestothehorizonunderperfectclarityandflat-ness (Martiacutenez-Garciacutea CalabreseMuellerOlsonamp Loacutepez 2013)Wefeelthisisreasonablebecausegazellemaybeabletouseothernon-visual cues (eg acoustic and social cues) and given our dailysamplingitispossiblethatthegazellewerecloserthan5kmtotheborderfencebetweentwosamplingeventsWeusedalargergiving-upthresholdof10kmtoavoidbreakingasingleencounterintomul-tipleencounterssimplybecauseonedatapointwasslightlyfartherawayTodefinethetraveldistanceswefirstprojectedeachgazelleposition inanencounterevent to theclosestpointon theborderForeachfenceencounterwethencalculatedthedistancealongtheborderencompassingalltheprojectedpointsontheborderWealsocalculatedthenumberofborderencountersandnumberofdaysga-zellestayedinproximitytotheborder

3emsp |emspRESULTS

31emsp|emspSpatiotemporal variability in space use of gazelle

Themeanannual rangeofasinglegazellewas19346km2 (n=40annual ranges Figure2) but this varied greatly among individuals


(SD=9265km2)Thelargestannualrange(53422km2)ofasinglegazellewasmorethaneighttimeslargerthanthesmallest(6431km2Figure2)Withinthesameyearandthesameregiontherangesizeamong individuals varied up to six times Considerable variabilityin space use also occurred by the same individuals among yearsAnnualrangesizevarieduptothreetimesforthesameindividualacrossyears(egfrom17890to53422km2foroneorfrom12696to37447km2foranotherindividualTableS1)

Theaveragerangesizeacrossallgazellevariedlessamongyearsforthe12femalesthatwereallcaughtinthesamelocationandthesameyear (Sample1) range sizesvaried from14934km2 (n=12SD=5502km2)inyear1to23556km2(n=8SD=14113km2)inyear2and17602(n=5SD=5805km2)inyearthree(Sample1inFigure2)Similarlyforthe10individualsofSample2theaveragerangewas21500km2(n=10SD=7854)inyear2and20637km2 (n=5SD=7854km2)inyear3(Sample2inFigure2)

LifetimerangesofindividualgazelleestimatedwithAKDEaver-aged100800km2 (n=12SD=45356km2) ranging from38100to 167841km2 The average range crossing time was 6months(Figure2TableS2)TheaveragelifetimerangewassixtimeslargerthanthetotalsizeofthePAs(15000km2Figure2)

The predictability of space use across years based on the 14-dayintervalpairwisedistancesshowedthemeanpairwisedistancewas134kmindicatingthatingeneralgazelledidnotvisitthesameplacesacrossyears(Figure3)Themeanpairwisedistancebetweencalvingareas indifferentyearswas91km(n=12SD=49)whichwas significantly lower than distances for the wintering periods(p=0002df=11)Incontrastduringwinteringperiodsthemeanpairwisedistancewas176km(n=12SD=91)indicatingthatindi-vidualgazellesoughtandutilizedwinteringareasthatwerefartherapartindifferentyearsthanthedistancesbetweenareasoccupied

duringthecalvingperiod(Figure3)Theshorterpairwisedistancesduringcalvingtimeindicatethatindividualgazelleinhabitedareasinrelativeproximitytothosewhichtheyhadused inpreviousyearsWhencomparingmeanpairwisedistanceduringcalvingtothenullmodelwefoundthreeindividualsthathadsignificantlysmallerdis-tancesthanthenullmodel indicating individual-levelvariability inspatialpredictabilityduringcalvingperiods(TableS3)Duringwin-terspairwisedistancesforallindividualswerenotsignificantlydif-ferentthanthenullmodel(TableS3)

32emsp|emspEffects of protected and disturbed areas on gazelle space use

GazellepassedthroughsevenPAsbut theydidnotusethePAsmorethanexpectedbychance(Figure4aAnimationS1)Theme-dian of the proportion of positions inside PAs of tagged versussimulatedgazellewassimilarandwedidnotdetectasignificantdifference(Figure4aAnimationS1)Howeverwefoundasignifi-cantdifferencebetweenthemediandisturbanceindexfortaggedand simulated gazelle paths indicating that gazelle avoided dis-turbed areas (eg population centres extractive industry sitesFigure4b)

33emsp|emspEffects of border fences on gazelle movements

About80(17of22)ofthetaggedgazelleencounteredtheborderfenceatleastonceduringthestudyperiod(foratotalof39fenceencounters)eventhoughtheoriginaltagginglocationswerefarfromtheborderfence(upto100kmFigure1)MovementbehavioursofgazellewereextremelyvariablewhenapproachingtheborderfenceOnaveragegazellemovedalongtheborderfencefor10daysbut

F IGURE 2emspAnnualrange(MCP)andlifetimerangeestimations(AKDE)ofindividualMongoliangazellecomparedtothesizesofprotectedareasintheeasternMongoliaonalogscaleNotallindividualssurvivedtheentirestudyperiodandthesamplesizedecreaseswithstudyyearsforbothsamplesTheboxplotofannualrangesindicatewithin-yearvariabilityandcomparisonamongboxplotsindicatesbetween-yearvariabilityofrangesThetrianglerepresentsthetotalareaofallsevenprotectedareas


somegazellemovedalongtheborderforaslongas59daysandstillothersturnedbackwithinadayThedistancegazellemovedalongtheborderfencerangedfromafewhundredmetersto80kmwithamediandistanceof11km(Figure5a)

4emsp |emspDISCUSSION

41emsp|emspVariability of space use in nomadic movements

Variability of individual behaviour including individual variationin space use is a key factor in ecology and evolution that shouldreceive greater attention in conservation (Merrick amp Koprowski2017)Thisisparticularlyimportantfornomadicspeciesthatdisplaymovementbehaviourswith large individualdifferenceswithinandbetweenyears

Ourstudyhighlightstheimportanceoflong-termmonitoringofspaceusefornomadicspeciesandemphasizesthreeaspectsofthe

spatiotemporal dynamics of space use in nomadism (a) individualdifferencesinspaceusewithinandbetweenyears(b)lackoffidelitytoparticularareasforkeystagesinthelifecycle(egcalvingruttingwintering)and(c)largelifetimeranges

Firstwith regard to variability in spaceuse among individualsin the same year we found that an individual gazelle occupied alargeareainasingleyearonaverage~19000km2mdashaboutthesizeofKrugerNationalParkinSouthAfricaSomegazellehaduptosixtimeslargerrangesthanothersindicatingindividualbehaviouraldif-ferencesinspaceuse(Figure2)Moreoverwefoundgazelleexhibitsubstantialbehaviouralflexibilityovertimewithrangesvaryingbyafactorofthreefrom1yeartothenextSecondthisvariabilitywasalsoprevalentinthelackofsitefidelitytospecificwinteringareasandtheaveragedistancebetweenwinteringareasindifferentyearswas176km(SD=91km)Theareasusedbygazelleduringthecalv-ingperiodontheotherhandshowed lessvariabilitybut thedis-tancebetweencalvinggroundsindifferentyearswasrelativelylarge

F IGURE 3emspPredictabilityofspaceusebetweenyearsEachboxplotrepresentsthedistributionofpairwisedistancesbetweentwo14-daymeanlocationsofthesameindividualindifferentyearsAreasusedbyMongoliangazelleduringwinteringperiodsweremorevariablethancalving periods

F IGURE 4emsp (a)AcomparisonofthemedianproportionofprotectedareausebytaggedMongoliangazelle(blackcircle)withthemediansof1000replicatesimulations(boxplot)showedthatMongoliangazelleusedprotectedareassimilarlytorandomchance(p=015) (b)AcomparisonofthemediandisturbanceindexforthetaggedMongoliangazelle(blackcircle)withmediansfrom1000replicatesimulationsshowedastrongavoidanceofdisturbedareas(p=0003)Weestimatedthep-valueusingarandomizationtest


(91plusmn49km)suggestinglittleevidencefortheexistenceoflsquocalvinggroundsrsquo or seasonal ranges referred to in the literature (Gunn ampMiller1986ItoTsugeetal2013Leimgruberetal2001Olsonetal2010)

Thebetween-yearvariabilityinspaceuseislikelydrivenbyun-predictable changes in resource availability across the landscapeashasbeenalsoshownforothernomadicspecies inaridenviron-ments(JonzeacutenampKnudsen2011RoshierDoerrampDoerr2008)Insummer gazellemovements are driven by the patchy distributionofhigh-quality vegetationdue to rainfall variability (Mueller etal20082011)Inwintergazellemovementsarelikelydrivenbyacom-binationofhigherforageavailabilityandshallowsnowdepthwhichareunpredictableinspaceandtime(ItoTsugeetal2013LuoLiuLiuJiangampHalbrook2014)explainingthelackoffidelitytowin-teringareasovertimeSearchingoutlowersnowdepthsforeasiermovementandbetteraccesstoforageisawidespreadbehaviourofungulatesinwinter(AvgarMosserBrownampFryxell2013GilbertHundertmarkPersonLindbergampBoyce2017NicholsonArthurHorneGartonampDelVecchio2016)

Lastlywe found that individual gazelle have extremely largeestimated lifetime ranges Although the average annual rangewas 19346km2 (MCP) the estimated average lifetime range fora single gazellewas100800km2 (AKDEFigure2)which is halfof the population core range of gazelle (~200000km2 Flemingetal 2014) and is four times larger than the area covered by~12million wildebeestConochaetus taurinus during their annualmigration through the SerengetindashMara ecosystem (25000km2)(Thirgood etal 2004) The average lifetime range for gazelle isaboutthesamesizeasthetotalareausedby54caribousRangifer tarandus grantiover4years in theCanadianNorthwestTerritory(84543km2Nicholsonetal2016)Howeverwenotethattheseestimatesofotherungulatesindifferentsystemswerenotcalcu-latedwithAKDEandthatactuallifetimerangesofgazellemaybesomewhat smaller thanwe estimated here because AKDE doesnottakeintoaccountbarriers

42emsp|emspConservation strategies for nomadic species

Nomadicgazellesindividualvariabilityinspaceusebetweenyearstheirlargelifetimeareaneedsandespeciallytheirlackoffidelitytowinter-ingandcalvingareasallhighlighttheimportanceof landscapeperme-abilityAllthegazellewetrackedhadlifetimerangeslargerthananyofthePAs(Figure2)andthePAswereonlysporadicallyusedbygazelle(Figure4aAnimationS1)indicatingthatPAsarenotaneffectivecon-servationmeasureforthespeciesTheindividuallifetimerangeofover100000km2suggeststhatthescaleofconservationmanagementmustgoconsiderablybeyondthescaleofexistingPAs(upto6000km2)AtpresenthoweverPAsarecurrentlytheonlyconservationmeasureforthegazelleandcover~8ofthegazellerangeinMongolia

Studies on highlymobile species have established that spatiallystatic PAs are not themost effective conservationmeasure (Rungeetal 2014Thirgoodetal 2004) Forwide-ranging specieswherePAsalonearenotsufficient for theirconservationanumberofdy-namic conservation concepts have been suggested (a) mobile PAs(b)PAnetworks (c)biodiversityoffsetsand(d) landscape-levelman-agementMobilePAsaimtotemporarilyprotectareaswhereanimalsareknowntoaggregate(TaillonFesta-bianchetampCocircteacute2012)TheseareasmayshiftalongpredictablechangesofsuitablehabitatsthroughtheyearLikewisePAnetworksaimtoconservecriticalareasalongmovementcorridorsorspatiallypredictablecorerefugeslikebreed-ingandwinteringareas(RoshierRobertsonampKingsford2002SinghampMilner-Gulland2011)BothmobilePAsandPAnetworksdependuponlocationsthatareknowntobeimportantatsomepointthrough-out the year In contrast biodiversity offsets are location-basedapproacheswiththeunderlyingideathatdetrimentallandscapemodi-ficationscanbeoffsetbyconservationmeasuresindifferentspatiallydistinctareasof impact (BullSuttleSinghampMilner-Gulland2013GordonBullWilcoxampMaron2015)

With nomadicmovements however an entire region is inter-connectedoverthecourseofseveralyearsThuswhileanyoftheaforementioned approaches can be important components of a

F IGURE 5emspTheeffectsoftheMongolianborderfenceonMongoliangazellemovements(a)ThedistributionofMongoliangazelletraveldistancesalongtheborderfence(b)IllustrationoftwoencountersofasingleMongoliangazellewiththeborderfenceandtheirtraveldistancesForencounter1(inblue)theMongoliangazelleleftthefencewithinadayandthetraveldistancealongtheborderwas~20kmForencounter2(inorange)theMongoliangazellewasintheproximityoftheborderfor~20daysandthetraveldistancewas~50km


conservation strategy for nomadic species landscape-levelman-agementthatconsidersallpartsofthelandscapeiskeytoensuringpermeabilityacrosstheentire landscape(KremenampMerenlender2018PoianiRichterAndersonampRichter2000)Maintainingper-meability throughout the entire landscape is critical for nomadicspeciestocopewiththepatchyandephemeraldistributionofre-sourcesandtoavoidadverselyaffectedareasduringextremecon-ditionsForexampleduringourstudyperiodaregionaldroughtinthesummerof2015wasfollowedbyanextremelycoldandsnowyseverewinter (Raoetal2015)ourmovementdata showed thatsomegazelleescapedthemostseverewinterconditionsbymovinglongdistancestothenortheastcrossingthefrozenKherlenRiver(Figure S2 Animation S1) Evidence from other ungulate speciesduring severe winter conditions shows that if large-scale escapemovements are not possible dramatic population declines canresult (Kaczenskyetal2011)Thereforeakeyquestionforcon-servation ishowpermeability across theentire landscape canbemaintained especially in the faceof infrastructuredevelopments(Ascensatildeoetal2018)

A critical issue for landscape permeability is linear infrastruc-ture that can prevent nomadic species from accessing unpredict-able and ephemeral resources across an entire landscape Forgazelle and other wide-ranging ungulates around the world thenegative effect of linear infrastructure such as fences highwaysandrailroads iswidelyobserved inbarriereffects thatcutoffen-tire areas of the landscapewhere animalsmight have tomove toescape harsh conditions or access migration routes and seasonalranges (Ito Lhagvasuren etal 2013Wingard Zahler VicturineBayasgalanampBayarbaatar2014XiaYangLiWuampFeng2007)Inourcasethefencealongthenationalbordercannotbecrossedbygazelle (Figure1AnimationS1)andthemovementsof80ofthe22trackedindividualswereaffectedbytheborderfenceWhenapproachingtheborder fencegazellemovementswereextremelyvariablewithregardtofindingacrossingsomegazellemovedex-tended periods along the border fence while others immediatelygaveupandmovedawayInadditionseveralpreviousstudieshaveshownthatfencingcauseshighmortalityingazelleandotherwide-ranginglargemammalsthroughoutEurasia (Itoetal2008Linnelletal 2016 Olson Mueller Leimgruber etal 2009) Migratoryungulatesthatfacehabitatfragmentationandbarriereffectsoftenexhibit significantpopulationdeclinesorhaveperishedaltogether(Harris ThirgoodHopcraft CromsigtampBerger 2009WilcoveampWikelski2008)

Identifying suitable design spacing and locations for crossingstructuresandmovementcorridorsalongthemigrationroutesareamitigationmeasureforminimizingthelandscape-scaleimpactsoflin-earbarriersonmigratoryungulates(Bastille-RousseauWittemyerDouglas-Hamilton ampWall 2018 Sawyer Lebeau amp Hart 2012)Thisconservationmitigationreliesonareasrepeatedlyusedbymi-gratoryungulateswhichshowstrongfidelitytoroutesandseasonalrangesIncontrastnomadicspeciesaredifficulttomanagebecausetheir key areas and seasonal ranges are not clearly defined andtheydonotexhibit repeateduseof same locations Inwide-open

ecosystems in arid environments such as theEasternSteppe andtheKazakhSteppewheregazelleandsaigaantelopeSaiga tatarica tataricaoccur respectively identifyingcriticalcorridorsandcross-ingsischallengingbecausenomadicpopulationsrequiresuchlargeexpansesofhabitat

43emsp|emspConservation of the Eastern Steppe

TheMongoliangovernmentproposed56835kmofnew railwaysandroadcorridorsacrossthecorerangesofseveralungulates in-cluding gazelle (Figure1 Animation S1 Batsaikhan etal 2014Gansukh etal 2018) Any development projects are requiredto conduct environmental impact assessments in Mongolia (Lawof Mongolia on Environmental Impact Assessments 2011) andMongoliahasapprovedthewildlifecrossingstandardforroadandrailroads (Mongolian National Standard 2015) which states thatthelocationsofcrossingstructuresmustbeselectedbasedonsci-entific knowledgeonanimalmovements and theirmovement cor-ridorsWhilethesemitigationstandardsandguidelinesexistthereisaclearlackofstrategyforimplementationandrecommendationsbasedonscientificknowledge

Weshowthatgazelleavoidpopulationcentresareaswithahighdensityofroadsoilextractionfieldsandlarge-scaleinten-siveagriculture(Figure3AnimationS1)InadditiondisturbancessuchastheproposedrailwayintheEasternSteppewillfragmentthesteppe(Figure1AnimationS1)andwillbecomeanimperme-ablebarriertogazellemovementsiffencedAvoidingorminimiz-inganylandscape-scaleimpactsfrominfrastructuredevelopmenton the permeability of the steppe should be a development planningpriority

Animpedimenttothatgoalisthelackofaregion-widecompre-hensiveland-useplanTheEasternSteppeisunderthestewardshipofmultipleownersandissubjecttoavarietyofmanagementprac-ticesandregulationsCurrentlydifferentgovernmentagencies(egMinistryofFoodandAgricultureMinistryofMineral andEnergyResourcesMinistryofInfrastructureandDevelopmentMinistryofEnvironmentandTourism)aswellasprivatesectors(egextractiveindustriestransportationcompanies)thatarededicatedtolandde-velopment often act without considering landscape permeabilityandtheconservationofwide-ranginganimals

WeemphasizethattheEasternStepperemainsoneofthelarg-estandleastfragmentedtemperategrasslandsintheworldandastrongholdofthelargestremainingpopulationofopenplainsungu-latesworld-wideandthattheirlarge-scalenomadicmovementsarerecognizedbytheConventionofMigratorySpeciesTheecologicalintegrityof the steppe canbepreservedwhere gazelle continueto benefit fromunrestricted access throughout the landscape bylimitinginfrastructureexpansionThiscouldbeachievedbydesig-natingtheregionsofthesteppecurrentlycategorizedasldquopasturerdquoand ldquomanagementrdquo via traditional land use practices as an IUCNcategoryVprotectedlandscapewhereconservationobjectivesaresetacrosslargeareasandthemanagementiscarriedoutbyarangeof actors (Dudley 2008) Such a designation could help preserve


gazelleandotherendangeredspeciesaswellasthenomadicpas-toralculture

Wherelinearinfrastructurecannotbereroutedtoavoidconflictwerecommendthatbecauseofthelackoffidelityingazellemove-mentscrossingoptionsshouldbeveryfrequentsimilartothehighdensityofcrossingoptionsinothersuccessfulmitigationmeasuresformigratoryungulates(SeidlerGreenampBeckmann2018)Onav-erage an individual gazellemoved11kmalong theborder fencesbeforegivingupitscrossingattemptThisdistancemightbeafirstminimumestimateonthenecessaryfrequencyofpotentialcrossingoptionsalonglinearbarriers

5emsp |emspCONCLUSIONS

The nomadic movements of ungulates remain largely unknownMultiple-year monitoring data required to examine ungulatesrsquomovementcharacteristicsinwide-openaridenvironmentsandtoidentifytheconservationmeasuresneededrarelyexistIncompleteknowledgeaboutanimalmovementscanresultininaccuratecon-servationassessmentsandineffectivemanagementactions(AllenampSingh2016Rungeetal2014)Inadditionmovementstudiestodaterarelyexploretheroleofthesemovementsinshapingpop-ulationabundanceUltimatelymovementdataneedtobecoupledwithrobustpopulationcensusestounderstandhowdemographicprocessesarelinkedtospaceuseWeencourageintegratedlandusemanagementpoliciesatthelandscapescalethataccountforlandscapepermeabilityfornomadicspecieswhereverpossibleInparticularweadvocatethatmultiyearmovementdataisessentialformakingconnectivityassessmentsinaridandsemi-aridregionswherewide-rangingnomadicspeciesoccur

ACKNOWLEDG EMENTS

WearegratefultofieldassistantsaswellasWorldWideFundforNatureMongolia (WWF) for gazelle capture permissions All ani-malswerecaptured followingstandardprotocolsapprovedby theMinistry of Environment and Green Development in Mongolia(licenses 65621 54275) The project was funded by RobertBosch Foundation and theGerman FederalMinistry of EducationandResearch (BMBF 01LC1710A and01LC1820A) JMC CHFandWFFweresupportedbyNSFABI1458748WethankDavidRoshierandoneanonymousreviewerforhelpfulcomments

AUTHORSrsquo CONTRIBUTIONS

DNdevelopedtheideasandmethodologywithcontributionfromTM CB and KAO DN TM KAO and DM collected themovement data DN performed the analysis with contributionsfromTMCBJMCandCHFMHdevelopedthedisturbanceindex layer TS completed the animation of gazelle movementsDNledthewritingofthemanuscriptAllauthorsrevisedthemanu-scriptandhelpedwiththeinterpretationofresults

DATA ACCE SSIBILIT Y

The data used in this study are available in the Dryad DigitalRepository httpsdoiorg105061dryad23nt63c (Nandintsetsegetal2019)

ORCID

Dejid Nandintsetseg httpsorcidorg0000-0002-5376-3808

Justin MCalabrese httpsorcidorg0000-0003-0575-6408

Thomas Mueller httpsorcidorg0000-0001-9305-7716

R E FE R E N C E S

AllenAMampSinghNJ(2016)Linkingmovementecologywithwild-lifemanagementandconservationFrontiers in Ecology and Evolution3(January)1ndash13httpsdoiorg103389fevo201500155

AscensatildeoFFahrigLClevengerAPCorlettRT JaegerJAGLauranceFampPereiraHM (2018)EnvironmentalchallengesforthebeltandroadinitiativeNature Sustainability1(May)1ndash4httpsdoiorg101038s41893-018-0059-3

AvgarTMosserABrownGSampFryxellJM(2013)Environmentaland individual drivers of animalmovement patterns across awidegeographical gradient Journal of Animal Ecology 82(1) 96ndash106httpsdoiorg101111j1365-2656201202035x

Bastille-Rousseau G Wittemyer G Douglas-Hamilton I ampWall J(2018) Optimizing the positioning of wildlife crossing structuresusing GPS telemetry Journal of Applied Ecology (January) 1ndash9httpsdoiorg1011111365-266413117

Batsaikhan N Buuveibaatar B Chimed B Enkhtuya O GalbrakhDGanbaatarOhellipWhittenT(2014)Conservingtheworldsfin-estgrasslandamidstambitiousnationaldevelopmentConservation Biology001ndash4httpsdoiorg101111cobi12297

Bull J W Suttle K B Singh N J amp Milner-Gulland E J (2013)Conservationwhennothing stands stillMoving targetsandbiodi-versityoffsetsFrontiers in Ecology and the Environment11203ndash210httpsdoiorg101890120020

CalabreseJMFlemingCHampGurarieE(2016)CtmmAnRpackageforanalyzinganimalrelocationdataasacontinuous-timestochasticprocess Methods in Ecology and Evolution7(9)1124ndash1132httpsdoiorg1011112041-210X12559

CalengeC(2017)PackagelsquoadehabitatHRrsquoPackage ldquoAdehabitatHRrdquoCarbutt C HenwoodW D amp Gilfedder L A (2017) Global plight

of native temperate grasslands Going going gone Biodiversity and Conservation 26(12) 2911ndash2932 httpsdoiorg101007s10531-017-1398-5

ClarkELMunkhbatJDulamtserenSBaillieJEMBatsaikhanNKingSRBStubbeM(2006)SummaryconservationactionplansforMongolianmammalsRegional Red List Series21ndash101

DudleyN(2008)GuidelinesforApplyingProtectedAreaManagementCategories

FlemingCHCalabreseJMMuellerTOlsonKALeimgruberPampFaganWF (2014)Non-Markovianmaximumlikelihoodestima-tionofautocorrelatedmovementprocessesMethods in Ecology and Evolution5462ndash472httpsdoiorg1011112041-210X12176

FlemingCHFaganWFMuellerTOlsonKALeimgruberPampCalabreseJM(2015)Rigoroushome-rangeestimationwithmove-mentdataAnewautocorrelatedkernel-densityestimatorEcology96(5)1182ndash1188httpsdoiorg10189014-20101

GansukhUMingXampAliSA(2018)Analysisofthecurrentsitua-tionofMongolianrailwayanditsfuturedevelopment International


Business Research 11(5) 119ndash128 httpsdoiorg105539ibrv11n5p119

GeldmannJBarnesMCoadLCraigieIDHockingsMampBurgessND (2013) Effectivenessof terrestrial protected areas in reduc-inghabitatlossandpopulationdeclinesBiological Conservation161230ndash238httpsdoiorg101016jbiocon201302018

GilbertSLHundertmarkKJPersonDKLindbergMSampBoyceMS (2017)Behavioralplasticity inavariableenvironmentSnowdepth and habitat interactions drive deer movement in winterJournal of Mammalogy 98(1) 246ndash259 httpsdoiorg101093jmammalgyw167

GordonABullJWWilcoxCampMaronM(2015)Perverseincen-tivesriskunderminingbiodiversityoffsetpoliciesJournal of Applied Ecology52532ndash537httpsdoiorg1011111365-266412398

GunnAampMillerFL(1986)Traditionalbehaviourandfidelitytocar-iboucalvinggroundsbybarren-groundcaribouRangifer6(2)151httpsdoiorg107557262640

HarrisGThirgoodSHopcraftJGCCromsigtJPGMampBergerJ(2009)Globaldeclineinaggregatedmigrationsoflargeterrestrialmammals Endangered Species Research7(May) 55ndash76 httpsdoiorg103354esr00173

HeinerMMcRaeBLkhagvasurenBBatsaikhanNGalbadrakhDTsogtsaikhanBhellipKieseckerJ(2016)Modelinghabitatconnectiv-ity of a nomadicmigrant facing rapid infrastructure developmentKhulanhabitatconnectivityintheSoutheastGobiRegionMongoliaCapacitybuildingforMongolianMinistryofEnvironmentandGreenDevelopment in relation to biodiversity and conservation in thesouthernGobiDesertFinalReport1ndash29

Ito T Y Lhagvasuren B Tsunekawa A Shinoda M Takatsuki SBuuveibaatarBampChimeddorjB(2013)Fragmentationofthehab-itat ofwildungulatesby anthropogenicbarriers inMongoliaPLoS ONE8(2)1ndash9httpsdoiorg101371journalpone0056995

ItoTYOkadaABuuveibaatarBLhagvasurenBTakatsukiSampTsunekawaA(2008)One-sidedbarrierimpactofaninternationalrailroadonMongoliangazellesJournal of Wildlife Management72(4)940ndash943httpsdoiorg1021932007-188

ItoTYTsugeMLhagvasurenBBuuveibaatarBChimeddorjBTakatsuki S hellip Shinoda M (2013) Effects of interannual varia-tions in environmental conditions on seasonal range selection byMongoliangazellesJournal of Arid Environments9161ndash68httpsdoiorg101016jjaridenv201212008

Jonzeacuten N amp Knudsen E (2011) Uncertainty and predictability ThenichesofmigrantsandnomadsInEJMilner-GullandJMFryxellampARESinclair (Eds)Animal migration A synthesis (pp91ndash109)OxfordUKOxfordUniversityPress

KaczenskyPGanbataarOAltansukhN EnkhsaikhanN StaufferCampWalzerC (2011) Thedangerof having all your eggs inonebasket-wintercrashofthere-introducedPrzewalskishorsesintheMongolianGobiPLoS ONE6(12)1ndash8httpsdoiorg101371jour-nalpone0028057

KaysRCrofootMCJetzWampWikelskiM(2015)Terrestrialanimaltrackingasaneyeon lifeandplanetScience348(6240)aaa2478httpsdoiorg101126scienceaaa2478

Kremen C ampMerenlender A (2018) Landscapes thatwork for bio-diversity and people Science 362(6412) eaau6020 httpsdoiorg101126scienceaau6020

LawofMongoliaonenvironmentalimpactassessments(2011)LeimgruberPMcSheaWJBrookesCJBolor-ErdeneLWemmerCamp

LarsonC(2001)Spatialpatternsinrelativeprimaryproductivityandga-zellemigrationintheEasternSteppesofMongoliaBiological Conservation102(2)205ndash212httpsdoiorg101016S0006-3207(01)00041-6

Linnell J D C Trouwborst A Boitani L Kaczensky P Huber DReljic S hellip Breitenmoser U (2016) Border security fencing andwildlife The end of the transboundary paradigm in Eurasia PLoS Biology14(6)1ndash13httpsdoiorg101371journalpbio1002483

LuoZLiuBLiuSJiangZampHalbrookRS (2014) InfluencesofhumanandlivestockdensityonwinterhabitatselectionofMongoliangazelle(Procapra gutturosa)Zoological Science31(1)20ndash30httpsdoiorg102108zsj3120

MallonD (2008)Procapra gutturosa The IUCN Red List of Threatened SpeciesdoieT18232A7858611

Martiacutenez-GarciacuteaRCalabreseJMMuellerTOlsonKAampLoacutepezC(2013)OptimizingthesearchforresourcesbysharinginformationMongoliangazellesasacasestudyPhysical Review Letters110(24)1ndash5httpsdoiorg101103PhysRevLett110248106

Mcgowan J Beger M Lewison R L Harcourt R Campbell HPriestM hellip Possingham H P (2017) Integrating research usinganimal-borne telemetry with the needs of conservation man-agement Journal of Applied Ecology 54 423ndash429 httpsdoiorg1011111365-266412755

McNaughtonSJ(1976)SerengetimigratorywildebeestFacilitationofenergy flowbygrazingScience (New York NY)191(4222) 92ndash94httpsdoiorg101126science191422292

MerrickM J amp Koprowski J L (2017) Shouldwe consider individ-ual behavior differences in applied wildlife conservation stud-ies Biological Conservation 209 34ndash44 httpsdoiorg101016 jbiocon201701021

MuellerTOlsonKADresslerGLeimgruberPFullerTKNicolsonChellipFaganWF (2011)Howlandscapedynamics link individual-topopulation-levelmovementpatternsAmultispeciescomparisonofungulaterelocationdataGlobal Ecology and Biogeography20(5)683ndash694httpsdoiorg101111j1466-8238201000638x

MuellerTOlsonKA FullerTK SchallerGBMurrayMGampLeimgruberP(2008)InsearchofforagePredictingdynamichabi-tatsofMongoliangazellesusingsatellite-basedestimatesofvegeta-tionproductivityJournal of Applied Ecology45(2)649ndash658httpsdoiorg101111j1365-2664200701371x

NandintsetsegDBracisCOlsonKABoumlhning-GaeseKCalabreseJM Chimeddorj B hellipMueller T (2019)Data from Challengesin conservation of wide-ranging nomadic species Dryad Digital Repositoryhttpsdoiorg105061dryad23nt63c

Nandintsetseg D Kaczensky P Ganbaatar O Leimgruber P ampMueller T (2016) Spatiotemporal habitat dynamics of ungulatesinunpredictableenvironmentsThekhulan (Equus hemionus) in theMongolianGobidesertasacasestudyBiological Conservation204313ndash321httpsdoiorg101016jbiocon201610021

NicholsonKLArthurSMHorneJSGartonEOampDelVecchioP A (2016) Modeling caribou movements Seasonal ranges andmigration routesof the central arctic herdPLoS ONE11(4) 1ndash20httpsdoiorg101371journalpone0150333

OlsonKAampFullerTK(2017)WildlifehuntingineasternMongoliaEconomicanddemographicfactorsinfluencinghuntingbehaviorofherding householdsMongolian Journal of Biological Sciences15(1)37ndash46httpsdoiorg1022353mjbs20171505

OlsonKAFullerTKMuellerTMurrayMGNicolsonCOdonkhuuDhellipSchallerGB(2010)AnnualmovementsofMongoliangazellesNomads in the eastern Steppe Journal of Arid Environments74(11)1435ndash1442httpsdoiorg101016jjaridenv201005022

OlsonKAFullerTKSchallerGBLhagvasurenBampOdonkhuuD(2005)Reproductionneonatalweightsandfirst-yearsurvivalofMongolian gazelles (Procapra gutturosa) Journal of Zoology 265(3)227ndash233httpsdoiorg101017S0952836904006284

OlsonKAMuellerTBolortsetsegSLeimgruberPFaganWFampFullerTK (2009)Amega-herdofmorethan200000MongoliangazellesProcapra gutturosaAconsequenceofhabitatqualityOryx43(01)149httpsdoiorg101017S0030605307002293

Olson K A Mueller T Leimgruber P Nicolson C Fuller T KBolortsetsegShellipFaganWF(2009)Fencesimpedelong-distanceMongolian gazelle (Procapra gutturosa) movements in Drought-strickenlandscapesMongolian Journal of Biological Sciences745ndash50


Poiani K A Richter B D Anderson M G amp Richter H E (2000)Biodiversity conservation at multiple scales Functional siteslandscapes and networks BioScience 50 133ndash146 httpsdoiorg1016410006-3568(2000)050[0133BCAMSF]23CO2

RaoMPDaviNKDArrigoRDSkeesJNachinBLelandChellipByambasurenO(2015)Dzudsdroughtsandlivestockmortalityin Mongolia Environmental Research Letters10074012httpsdoiorg1010881748-9326107074012

RoshierDADoerrVAJampDoerrED (2008)Animalmovementindynamic landscapes Interactionbetweenbehavioural strategiesand resourcedistributionsOecologia156(2)465ndash477httpsdoiorg101007s00442-008-0987-0

RoshierDARobertsonAIampKingsfordRT(2002)ResponsesofwaterbirdstofloodinginanaridregionofAustraliaandimplicationsforconservationBiological Conservation106399ndash411

RungeCAMartinTGPossinghamHPWillisSGampFullerRA (2014) Conserving mobile species Frontiers in Ecology and the Environment12(7)395ndash402httpsdoiorg101890130237

Sawyer H Kauffman M J Nielson R M amp Horne J S (2009)Identifyingandprioritizingungulatemigrationroutesforlandscape-levelconservationEcological Applications19(8)2016ndash2025httpsdoiorg10189008-20341

SawyerH LebeauCampHart T (2012)Mitigating roadway impactstomigratory mule deermdashA case study with underpasses and con-tinuous fencingWildlife Society Bulletin 36 492ndash498 httpsdoiorg101002wsb166

SeidlerRGGreenDSampBeckmannJP (2018)Highwayscross-ingstructuresandriskBehaviorsofGreaterYellowstonepronghornelucidateefficacyofroadmitigationGlobal Ecology and Conservation15e00416httpsdoiorg101016jgecco2018e00416

Singh N J ampMilner-Gulland E J (2011) Conserving a moving tar-get Planning protection for a migratory species as its distribu-tion changes Journal of Applied Ecology 48 35ndash46 httpsdoiorg101111j1365-2664201001905x

Taillon J Festa-bianchetMampCocircteacute SD (2012) Shifting targets inthe Tundra Protection ofmigratory caribou calving groundsmustaccount for spatial changesover timeBiological Conservation147163ndash173httpsdoiorg101016jbiocon201112027

Mongolian National Standard Passages for wild ungulates along thehighwaysandrailwaysinsteppeandGobiareasMNS6515(2015)

Thirgood S Mosser A Tham S Hopcraft G Mwangomo EMlengeyaThellipBornerM(2004)Canparksprotectmigratoryun-gulatesThecaseoftheSerengetiwildebeestAnimal Conservation7(2)113ndash120httpsdoiorg101017S1367943004001404

Vandandorj S Gantsetseg B amp Boldgiv B (2015) Spatial and tem-poral variability in vegetation cover of Mongolia and its implica-tions Journal of Arid Land7(4) 450ndash461 httpsdoiorg101007s40333-015-0001-8

WilcoveDSampWikelskiM(2008)GoinggoinggoneIsanimalmi-grationdisappearingPLoS Biology6E188httpsdoiorg101371journalpbio0060188

Wingard J Zahler PVicturineRBayasgalanOampBayarbaatarB(2014) Guidelines for addressing the impact of linear infrastructure on large migratory mammals in Central AsiaUNEPCMSSecretariatWildlifeConservationSociety

XiaLYangQLiZWuYampFengZ(2007)TheeffectoftheQinghai-TibetrailwayonthemigrationofTibetanantelopePantholops hodg-soniiinHoh-xilNationalNatureReserveChinaOryx41(3)352ndash357httpsdoiorg101017S0030605307000116

SUPPORTING INFORMATION

Additional supporting information may be found online in theSupportingInformationsectionattheendofthearticle

How to cite this articleNandintsetsegDBracisCOlsonKA etalChallengesintheconservationofwide-rangingnomadicspecies J Appl Ecol 2019001ndash11 httpsdoiorg1011111365-266413380


1emsp |emspINTRODUC TION

Integrating animal movements into conservation management iscriticaltoconservationsuccess(AllenampSingh2016Mcgowanetal2017)Conservationchallengesaremagnifiedforhighlymobilespe-cies that have large ranges and high spatiotemporal variability inspaceuseSpatiotemporalvariabilityand itspotential implicationsfor conservation management remain understudied in large partduetoa lackofsuitabledatasets (KaysCrofoot JetzampWikelski2015RungeMartinPossinghamWillisampFuller2014)

ConsiderationofmovementstrategiessuchasrangeresidencyandmigrationiscrucialtoprotectingwildlifeForexampleconser-vationof long-distancemigrantswithpredictablemigrationroutescan be facilitated through corridors that maintain connectivitybetween seasonal habitats (Sawyer Kauffman Nielson amp Horne2009)Inothercasesmanagementeffortsfocusonprotectingareasthatareregularlyusedduringkeypartsofthespeciesrsquolifecycle(egbreedingandwinteringareasGeldmannetal2013)

Whilemigratoryandrangeresidentspeciesarewellstudiedfewstudies address nomadic species and even fewer consider conser-vationplanning for them (Rungeetal2014)Nomadic species likeThomsonsgazelleEudorcas thomsoniiintheSerengetiPlainsorkhulanEquus hemionus in theGobiDesert are typically found in resource-pooraridenvironmentswithdynamicresources(McNaughton1976NandintsetsegKaczenskyGanbaatarLeimgruberampMueller2016)Thekeycharacteristicsofnomadicmovementsarenon-seasonalspa-tialvariabilityandtemporalunpredictability in interannualspaceuse(Jonzeacutenamp Knudsen 2011) However few studies on nomadic spe-cies examine the predictability of space use bymultiple individualsmonitoredovermorethan1yearwhich iskey informationfor theirlong-termsurvivalStudiesontheuseofprotectedareas(PAs)ortheeffectsofdisturbanceonnomadicspeciesarelikewisescarce

HerewestudytheMongoliangazelleProcapra gutturosa(here-aftergazelle)aprominentexampleofanomadicungulate(Muelleretal2008Olsonetal2010)andoneofthemostnumerousgazelle

speciesgloballyTheyarenativetotheopensteppeofMongoliaandtheadjacentareasofRussiaandChina(Mallon2008)Over95oftheglobalpopulationofgazelleoccursinMongoliassteppeBasedona2002estimatedpopulationsizeof1milliongazelle(Olsonetal2010) thespecies is listedasLeastConcern in the IUCNRedList(Mallon2008)Althoughthepopulationtrendisassessedasstablein the IUCNRedList therehasbeenno regional scalepopulationcensussince2002andthusthepopulationtrendisunknownThepopulationdoesundergosubstantialfluctuationsinabundanceduetooverhuntingdiseaseoutbreaksandextremeweathereventsandgazelleareconsideredasendangeredinMongoliasRedList (Clarketal2006)

Gazelle undertake long-distance movements driven by highinter- and intraannual variability in resource availability in thesteppe (Mueller etal 2008) Observed group sizes range from asingleindividualtomorethan200000individuals(OlsonMuellerBolortsetseg etal 2009) An individual gazelle can roam over32000km2in1year(Olsonetal2010)Thespeciesrsquocoredistribu-tionrangeineasternMongoliais200000km2plusmn3100km2(Flemingetal2014)Apartfromattentiongiventoannualrangesizesrela-tively little isknownabout the lifetimespace requirementsor thepredictabilityofareasusedbyindividualgazelleparticularlyatcriti-calstagesoftheirlifecycle(egcalvingandwinteringperiods)

AmosaicofPAswasestablishedinparttosafeguardremaininggazelle populations in easternMongoliawhich are threatened bypoachingcompetitionwithlivestockandhabitatlossandfragmenta-tion(OlsonampFuller2017Figure1)TheseunfencedPAsoftenhaveahabitatcharacteristicssimilar to their immediatevicinityFencesalongsidetheTrans-MongolianRailwayandMongoliasnationalbor-derareprovenbarriersforgazellefurtherfragmentingtheirhabitatoutsidethespeciesrsquocorerange (Itoetal2008)Furthermore therapidlygrowingextractiveindustryinMongoliahasledtoproposalsto constructnew railwaysMongoliahas a statepolicyon railwaytransportation (Parliament resolutionno32 in2010)whichplanstheconstructionof56835kmofnewrailways(GansukhMingamp

5 Synthesis and applicationsWe show that for nomadic specieswhose space use variesgreatlyacrossyearsmultiyearmovementdataareessentialforsoundcon-servationplanningWeemphasizethatplace-basedapproachesaloneareinsuffi-cient to conserve wide-ranging nomadic species and that linear infrastructureincludingfenceshighwaysandrailroadsisofparticularconcernBecausenomadicspecieslackdefinedmovementcorridorsweadvocateintegratedlanduseplan-ningthatprioritizespermeabilityacrosstheentirelandscapeandfacilitateslong-distancemovementsWesuggestthatconservationstrategiesfornomadicspeciesinaridandsemi-aridregionsbereconsideredbasedonmultiyearconnectivityas-sessmentsatthelandscapescale

K E Y W O R D S

animalmovementcrossingstructurelandusenomadicpermeabilityprotectedareaungulatewide-rangingspecies























3emsp |emspRESULTS
















































ACKNOWLEDG EMENTS






ORCID




R E FE R E N C E S



























































































3emsp |emspRESULTS
















































ACKNOWLEDG EMENTS






ORCID




R E FE R E N C E S


















































































































ACKNOWLEDG EMENTS






ORCID




R E FE R E N C E S





































































challenges in the conservation of wide‐ranging nomadic

Documents