Citation
Effects of wind energy generation and white-nose syndrome on the viability of the Indiana bat

Material Information

Title:
Effects of wind energy generation and white-nose syndrome on the viability of the Indiana bat
Series Title:
PeerJ
Creator:
Erickson, Richard A.
Thogmartin, Wayne E.
Diffendorfer, Jay E.
Russell, Robin E.
Szymanski, Jennifer A.
Publication Date:
Language:
English
Physical Description:
1 online resource

Subjects

Subjects / Keywords:
Conservation biology ( lcsh )
Ecology ( lcsh )
Zoology ( lcsh )
Genre:
serial ( sobekcm )

Notes

Abstract:
Wind energy generation holds the potential to adversely affect wildlife populations. Species-wide effects are difficult to study and few, if any, studies examine effects of wind energy generation on any species across its entire range. One species that may be affected by wind energy generation is the endangered Indiana bat (Myotis sodalis), which is found in the eastern and midwestern United States. In addition to mortality from wind energy generation, the species also faces range-wide threats from the emerging infectious fungal disease, white-nose syndrome (WNS). White-nose syndrome, caused by Pseudogymnoascus destructans, disturbs hibernating bats leading to high levels of mortality. We used a spatially explicit full-annual-cycle model to investigate how wind turbine mortality and WNS may singly and then together affect population dynamics of this species. In the simulation, wind turbine mortality impacted the metapopulation dynamics of the species by causing extirpation of some of the smaller winter colonies. In general, effects of wind turbines were localized and focused on specific spatial subpopulations. Conversely, WNS had a depressive effect on the species across its range. Wind turbine mortality interacted with WNS and together these stressors had a larger impact than would be expected from either alone, principally because these stressors together act to reduce species abundance across the spectrum of population sizes. Our findings illustrate the importance of not only prioritizing the protection of large winter colonies as is currently done, but also of protecting metapopulation dynamics and migratory connectivity.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
K26-05267 ( USFLDC DOI )
k26.5267 ( USFLDC handle )

USFLDC Membership

Aggregations:
Karst Information Portal

Postcard Information

Format:
Serial

Downloads

This item is only available as the following downloads:


Full Text

PAGE 1

Submitted 11August2016 Accepted 24November2016 Published 22December2016 Correspondingauthor RichardA.Erickson, rerickson@usgs.gov Academiceditor NigelYoccoz AdditionalInformationand Declarationscanbefoundon page13 DOI 10.7717/peerj.2830 Distributedunder CreativeCommonsPublic DomainDedication OPENACCESS Effectsofwindenergygenerationand white-nosesyndromeontheviabilityof theIndianabat RichardA.Erickson 1 ,WayneE.Thogmartin 1 ,JayE.Diffendorfer 2 RobinE.Russell 3 andJenniferA.Szymanski 4 1 UpperMidwestEnvironmentalSciencesCenter,UnitedStatesGeologicalSurvey,LaCrosse, WI,UnitedStates 2 GeosciencesandEnvironmentalChangeScienceCenter,UnitedStatesGeologicalSurvey,Denver, CO,UnitedStates 3 NationalWildlifeHealthCenter,UnitedStatesGeologicalSurvey,Madison,WI,UnitedStates 4 DivisionofEndangeredSpecies,UnitedStatesFishandWildlifeService,Onalaska,WI,UnitedStates ABSTRACT Windenergygenerationholdsthepotentialtoadverselyaffectwildlifepopulations. Species-wideeffectsaredifficulttostudyandfew,ifany,studiesexamineeffectsof windenergygenerationonanyspeciesacrossitsentirerange.Onespeciesthatmaybe affectedbywindenergygenerationistheendangeredIndianabat Myotissodalis ,which isfoundintheeasternandmidwesternUnitedStates.Inadditiontomortalityfrom windenergygeneration,thespeciesalsofacesrange-widethreatsfromtheemerging infectiousfungaldisease,white-nosesyndromeWNS.White-nosesyndrome,caused by Pseudogymnoascusdestructans ,disturbshibernatingbatsleadingtohighlevelsof mortality.Weusedaspatiallyexplicitfull-annual-cyclemodeltoinvestigatehowwind turbinemortalityandWNSmaysinglyandthentogetheraffectpopulationdynamics ofthisspecies.Inthesimulation,windturbinemortalityimpactedthemetapopulation dynamicsofthespeciesbycausingextirpationofsomeofthesmallerwintercolonies. Ingeneral,effectsofwindturbineswerelocalizedandfocusedonspecificspatial subpopulations.Conversely,WNShadadepressiveeffectonthespeciesacrossitsrange. WindturbinemortalityinteractedwithWNSandtogetherthesestressorshadalarger impactthanwouldbeexpectedfromeitheralone,principallybecausethesestressors togetheracttoreducespeciesabundanceacrossthespectrumofpopulationsizes.Our findingsillustratetheimportanceofnotonlyprioritizingtheprotectionoflargewinter coloniesasiscurrentlydone,butalsoofprotectingmetapopulationdynamicsand migratoryconnectivity. Subjects ConservationBiology,Ecology,EnvironmentalSciences,Zoology Keywords Endangeredspeciesassessment,Full-annual-cycle,Migratoryconnectivity,Wind turbinemortality,White-nosesyndrome,Populationassessment,Indianabat, Myotissodalis INTRODUCTION Windenergygenerationholdspotentialasanalternativeenergysourcetofossilfuels butalsoposesnewthreatstowildlife Kuvleskyetal.,2007 .Inadditiontothelossof habitatassociatedwithwindturbineplacement,collisionswithwindturbinesmaycause mortalityduringmigration Kunzetal.,2007 ; NationalResearchCouncil:Committeeon Howtocitethisarticle Ericksonetal.,Effectsofwindenergygenerationandwhite-nosesyndromeontheviabilityoftheIndiana bat. PeerJ4:e2830;DOI10.7717/peerj.2830

PAGE 2

Figure1Mapofinputdata. Mapofinputdata,speciesoccurrencemap,andIndianaBatspeciesrange. ``USFWSsummerobservations''arefromtheUSFishandWildlifeService.AllUSFishandWildlifeServicedataarefromtheendangeredspeciesprogramandexactlocationsareconfidential.``BISONsummer observations''arefromtheBiodiversityInformationServingOurNationdatabaseBISON;http://bison. usgs.ornl.gov.``Clarketal.observations''arecapturedatafrom Clark,Bowles&Clark ``Hibernacula''datarefersthewinterhibernaculadatafromtheUSFishandWildlifeService.``Windturbinedata ''comesfrom Diffendorferetal. .Thewhite-to-bluecolorgradientdepictslowtohighsuitabilityfromtheoccurrencemodel.Thegridboundaryistheoutlineofthegridcellsusedfortheoccurrencemodel. theStatusofPollinatorsinNorthAmerica,2007 ; Arnettetal.,2008 .Onespeciespossibly facingthreatsfromwindenergyistheIndianabat Myotissodalis ,anendangeredspecies foundinthemidwesternandeasternUnitedStatesFig.1.TheIndianabatmigrates seasonallybetweenmaternitycoloniesandhibernaculacavesandmineswherethespecies overwinters,exposingthespeciestodifferentialseasonalrisktowindenergy Pruitt& TeWinkel,2007 ; Piorkowskietal.,2012 Full-annual-cycleFACmodelsincludemortalityandreproductionforallseasons ofaspecieslifecycleanddifferfromtraditionalmodelsthatlumpallseasonstogether Hostetler,Sillett&Marra,2015 Taylor&Norris developedanavianFACmodel thathasbeenappliedtoMexicanfree-tailedbats Wiederholtetal.,2013 andatheoretical modelof Myotis spp. Ericksonetal.,2014 .Traditionally,mostmigrationmodelsfocus onsummerandwinterhabitatusee.g.,breedingandnon-breedingsitesformigratory birdsormaternityandhibernatingsitesformigratorycavebatsratherthanmigratory pathways Taylor&Norris,2010 .Modelingmigratorypathways,however,iscriticalto Ericksonetal., PeerJ ,DOI10.7717/peerj.2830 2/19

PAGE 3

understandingtheeffectsofmortalityfromwindenergygenerationonmigratingwildlife. Hereweapplythetheoreticalmodeldevelopedby Ericksonetal. totheentirerange oftheIndianabatsothatwemayassesshowcurrentwindenergydevelopmentmayaffect thespecies. TheIndianabatwasoneofthefirstspecieslistedundertheEndangeredSpeciesActof 1973.ThisactwaspassedwiththegoalofprotectingthenaturalheritageoftheUnited StatesofAmericaandallowsforplantsandanimalstobelistedaseitherendangeredor threatened OfficeoftheFederalRegister,1973 .Althoughtheoriginallistingdidnotspecify areason,theconsensusamongbatexpertswasthathumandisturbanceofhibernating batscausedpopulationdeclines,promptingthelisting Pruitt&TeWinkel,2007 ; Officeof theFederalRegister,1967 .Besideswindturbines Langwigetal.,2012 ; Arnett&Baerwald, 2013 ,thespeciesalsofacesthreatsfromwhite-nosesyndrome Thogmartinetal.,2013 andhibernaculumvandalism Crimminsetal.,2014 ,aswellasbroadthreatsfromclimate change,habitatloss,andlandusechange Pruitt&TeWinkel,2007 ; Loeb&Winters, 2012 ; Weber&Sparks,2013 .Populationsofthespeciesappearedtoberecoveringprior tothearrivalofwhite-nosesyndromeWNS Thogmartinetal.,2012a ,butdeclined asthediseasespread Turner,Reeder&Coleman,2011 .Recentresearchsuggeststhese declinesmaynotbeassevereasinitiallyfeared,butconcernsremainforspeciesexistence Thogmartinetal.,2013 ; Powersetal.,2015 White-nosesyndromeaffectscavebatssuchastheIndianabatduringhibernation andmaycauseupto100%mortality,resultinginextirpationsoflocalpopulations Turner,Reeder&Coleman,2011 ; Fricketal.,2015 Pseudogymnoascusdestructans ,the fungalcausativeagentofWNS,appearstoopportunisticallyinfectbatsduringhibernation i.e.,infectsbatswhentheirimmunesystemsarelessactiveduringhibernationinduced torpor Langwigetal.,2015 .Afterinfection,WNSinitiatesaphysiologicalcascadeof disturbancesthatoftenleadstothedeathofbats Willisetal.,2011 ; Cryanetal.,2013 ; Warneckeetal.,2013 ; Verantetal.,2014 .PriortothearrivalofWNS,nodemographic populationmodelse.g.,matrixpopulationmodelsincontrasttostatisticalpopulation modelsexistedforanybatspecies Hallam&Federico,2009 ,andpost-WNSarrival, modelingeffortshavelargelyignoredspatialconnectionsbetweenpopulationse.g., Thogmartinetal.,2012a ; Fricketal.,2010 .Conversely,thespatialmodeldevelopedby Ericksonetal. didnotconsiderWNSandalsodidnotmodeltheobservedspatial arrangementofpopulations. BothwindenergydevelopmentandWNSarespatiallyexplicitthreatstotheIndiana bat.WindturbinesprimarilyaffectIndianabatsalongtheirmigrationroutesof7 documentedkillshavebeenduringthefallorspring;http://www.fws.gov/midwest/wind/ wildlifeimpacts/inbafatalities.html#Table1,whileWNSaffectsthesurvivalofindividuals overwinteringincavesandmines.Theeffectsofthesetwostressorshavenotbeenjointly studiedfortheIndianabat,orforanybatspeciesonarange-widescale AmericanWind WildlifeInstitute,2014 .Herein,weexaminethepopulation-leveleffectsofwindenergyon theIndianabatacrossitsentirerange.Wealsostudytheinteractionbetweenwindenergy andWNStounderstandwhetherthemagnitudeofmortalityfromwindmaybesufficient toprecluderecoveryorincreaseriskofextirpation. Ericksonetal., PeerJ ,DOI10.7717/peerj.2830 3/19

PAGE 4

METHODS WeusedaFACmodeltoexplorepotentialimpactsofwindenergydevelopmentonthe Indianabat.Ourmodelincludeddatafrommultiplesources,includinghabitatandwind turbinedataFig.1.WedescribedourmodelusingtheOverview,DesignConcepts,and Detailsprotocol Grimmetal.,2006 ; Grimmetal.,2010 aspartofourtransparentand comprehensivemodel``evaludation''TRACEdocumentation Schmolkeetal.,2010 ; Augusiak,VandenBrink&Grimm,2014 ; Grimmetal.,2014 SupplementalInformation3. WealsoincludeourcodeasSupplementalInformation4andourdatahavebeenpublished toaUSGSwebpage Erickson,2016 .Withintheremainderofthissection,weprovidean overviewofourmodelingapproachanddescriptionofthedatausedwithinthisapproach. Thecoreofourpopulationmodelisaseriesofdifferenceequationspreviouslydescribed in Ericksonetal. andlistedinourTRACEdocumentation.Themodelkeepstrack ofgroupsoffemaleIndianabatsusingapathwaybetweenahibernaculumandamaternity colony. Ericksonetal. formulatedthemodeltoincludedensityatbothmaternity sitesandhibernaculafollowing Taylor&Norris .Wemodifiedthemodeltoonly includedensityatthematernitycolonies,whichaffectedabaselinesurvivalrate.Wemade thismodificationbecausehibernaculaareunlikelytobelimitingtheIndianbatpopulation sizes.Specifically,thetotalIndianabatpopulationappearstobeatleastoneorderof magnitudelowerthanpre-Europeansettlementsizesandthenumberofhibernaculahas remainingrelativelystableorincreasingthroughtimeasbatscolonizeoldmines Pruitt& TeWinkel,2007 Webasedourlifehistoryparametersuponpreviousmodels Thogmartinetal.,2012b ; Erickson,Thogmartin&Szymanski,2014 andselectedparametervaluessothattheannual populationgrowthratewas1.02withoutthedensityeffect.Thisvalueisconcordantwith pre-WNSgrowingIndianabatpopulations Thogmartinetal.,2013 ; Thogmartinetal., 2012a ; Thogmartinetal.,2012b OurmodellandscapecoveredmuchoftheeasternUnitedStatesFig.1.Thelandscape wasdividedintoapproximately33,0006500-hagridcellsbecausethisresolutionis consideredtobeequivalenttothehomerangeareaofanIndianabatmaternitycolony bytheUSFWSJSzymanski,pers.obs..Furthermore,allhibernaculaincloseproximity <10kmareconsideredoneunitformanagementbytheUSFWS Pruitt&TeWinkel, 2007 .Thecenterofeachgridcellcontaininghibernaculawasconnectedtothecenter ofallothergridcellstocreatemigratorypathways.Weassumedamaximummigration distanceof500km,becausemostdocumentedIndianabatmigrationroutesappeartobe ashorterdistancethanthis Gardner&Cook,2002 ; Winhold&Kurta,2006 .Weexcluded all``empty''hibernaculai.e.,thosehistoricallyoccupied,butnowemptyfromourmodel. Weonlyusedthehighest20%ofsummermaternitygridcellsbasedupontheprobabilityof Indianabatoccurrence,whichisdescribedinthenextparagraphFig.1.Thisleftuswith approximately50,000possiblepathwaysbetweenhibernaculaandhighqualitymaternity gridcells.Batswereplacedonthemodellandscapeatarandomsubsetofhibernacula duringmodelinitialization.Inanygivenrun,approximately5,000ofthe50,000pathways wereoccupied. Ericksonetal., PeerJ ,DOI10.7717/peerj.2830 4/19

PAGE 5

Habitatoccurrenceofmaternitysitesforeachgridcellwasmodeledwithalogistic regression.Webuiltuponpreviousworktoidentifycovariatestoconsider Loeb& Winters,2012 ; Weber&Sparks,2013 ; Farmer,Cade&Stauffer,2002 ; Milleretal.,2002 ; Yates&Muzika,2006 .Weinitiallycomparedseveralmodelsthatincludeddifferent combinationsofmeanmonthlytemperature,differentmeanmonthlyprecipitations,land cover,meanelevation,andmaximumslope.WeusedtheWatanabeAkaikeinformation criterionWAICforourmodelselection Watanabe,2010 becausethismethodisfully Bayesianandconsidersparameterdistributionsunlikeothermodelselectionapproaches Gelmanetal.,2013 .WeusedStanversion2.4,http://www.mc-stan.org,asimplemented throughRStan,tofitourmodelsandcalculatetheWAICvalues Hoffman&Gelman,2014 Ourfinalmodelincludedcropcover,deciduousforestcover,andMayprecipitation.The completeparametervaluesforthismodelaredescribedinourTRACEdocumentation SupplementalInformation3. WeusedtheWNS-spreadMapfrom12March2015https://www.whitenosesyndrome. org/resources/maptomodelspreadofthediseasethroughtime.WemodeledWNS thedisease,ratherthan Pseudogymnoascusdestructans thefungusbecausethisiswhat theNorthAmericanWNSresponsegrouptracks.Weassumedthatanyhibernaculum withoutWNSwouldhaveWNSby2016becauseWNShasspreadacrosstheentirerange oftheIndianabat.Weadaptedthewhite-nosesyndromemodelusedin Thogmartinet al.b tobeacontinuoustimefunctionratherthanapiecewisediscretefunction. WeusedalogisticfunctiontodescribebatsurvivalfromWNSthroughtime Bolker, 2008 .Themodeldependeduponthebaselinewintersurvivalrateandthearrivalyearof white-nosesyndrome,andalsoincludedaslopetermandinterceptterm.Theintercept termisoffsetbythearrivalyeartermtoaccountforWNSarrivingduringdifferent yearsandweincludedthreedifferentintercepttermstoaccountdifferentWNSsurvival scenarios.ParameterwerebaseduponourparameterizationasdescribedintheTRACE DocumentationSupplementalInformation3. WemodeledtheeffectofwindturbinesontheIndianabatbydecreasingsurvivalbased uponthenumberofturbinesfoundalongamigratorypathwayorinamaternitysiteor hibernaculumgridcell.Thenumberofturbinespresentdecreasedthebaselinesurvival. Duetouncertaintyaboutthenumberofbatskilledbyturbines Arnett&Baerwald,2013 weusedthreemortalityscenarios:Alowmortalityscenariowith1of1,000batsflying byaturbinekilled,amediummortalityscenariowithoneoutof100batsflyingbya turbinekilled,andahighmortalityscenariowithoneoutof10batsflyingbyaturbine killed.Thesethreescenarioswerechosentoboundthespectrumofpossibleresponsesand accommodateuncertaintyinIndianabatmortalityatturbines,aswellasincludea``safety factor''consideration SuterII,2006 .Thisuncertaintyinmortalityexistsbecause,todate, only7IndianabatshavebeenreportedkilledatwindenergyfacilitiestotheUSFWSFig.2; http://www.fws.gov/midwest/wind/wildlifeimpacts/inbafatalities.html.Themediumand highscenariosareunlikelytobeanaverageor``expected''mortalityrateforallmigratory pathways,butrepresentworstcasecollisionsrisksforbatsflyingthroughturbines.The actualcollisionriskforanyspecificwindturbineandanIndianabatlikelydependsupon manylocalenvironmentalconditionsthatchangetemporallye.g.,winddirectionduring Ericksonetal., PeerJ ,DOI10.7717/peerj.2830 5/19

PAGE 6

Figure2MapofknownIndianabatfatalities. CountieswithknownIndianabatfatalitiesatwildfacilities.ThefatalitiesmappedarethoseknowntotheUSFishandWildlifeServiceasofApril2015.Thefigure isfrom``IndianaBatFatalitiesatWindEnergyFacilities''byLoriPruittandJenniferOkajima,USFishand WildlifeService,IndianaFieldOfficehttp://www.fws.gov/midwest/wind/wildlifeimpacts/inbafatalities. html.ThefigurewascreatedbyUSGovernmentemployeesduringtheirofficialdutiesandisthereforein thepublicdomain. thefewminutesthatanIndianabatisflyingthroughaturbinefarm.Usingastochastic collisionriskwouldpossiblyimprovemodelrealism,butonlyifwehadameaningful distributionfromwhichtodraw. Turbinelocationdatawerefrom Diffendorferetal. .Winterandsummermortality fromwindturbinesonlyconsideredthemortalityfromthecellcontainingthecolony becauseofthespeciessmallhomerangeduringnon-migratoryseasons Pruitt&TeWinkel, 2007 .Nohibernaculacellshadturbinespresentwithinthem.Eachmigratorypathwaywas bufferedoneachsideby1-km,2-km,10-km,and20-km.Thisbufferdistanceaccountedfor uncertaintyintheIndianabatmigrationroute.Wefocusedonthe2-kmbufferpathway -kmwidebecauseUSFWSexpertsconsiderthistobethemostreasonablescenario JSzymanski,pers.obs.,2013. Themodelwasprogrammedin R RCoreTeam,2014 usingthe data.table package Dowleetal.,2014 .Weparallelizedourcodeforthestochasticrunsusingthe doSNOW package RevolutionAnalytics&Weston,2014 .OurcodeisincludedasSupplemental Information4. Ericksonetal., PeerJ ,DOI10.7717/peerj.2830 6/19

PAGE 7

Figure3TotalpopulationoffemaleIndianabatspredictedbythemodel. Thefigureisfacetedonthe x -axisbyWNSmortalityscenarios.Thefigureisfacetedonthe y -axisbywindturbineexposurescenarios. `` Migratoryonly ''referstobatsonlybeingkilledalongthemigrationpathwayswhereas`` Both ''allowsthe batstobekilledatboththesummerandwinterhabitatsaswellasalongthemigratorypathway.Weonly showtheresultsfromincludingturbinesfoundwithina2-kmbufferofthemigratorypathway. RESULTS WNShadthelargestimpactonthemodeledpopulationdynamicsoftheIndianabat Fig.3.ThehighestWNSmortalityscenariocauseda 95%decline,whichcausedextreme imperilmentforthespecies.ThemediumWNSmortalityscenariodecreasedthepopulation sizeby 80%whereastheWNSmortalityscenariosreducedthetotalpopulationsizeby 50%.UncertaintyexistedaboutwhereIndianabatsliveandmigrateonthelandscape, whichaffectedtheirmortalityfromwindturbineswithinthemodelandledtomodel uncertaintyi.e.,aprobabilisticoutputseenintheNoWNSscenarios.However,the inclusionofWNSoverwhelmedthisspatialuncertaintyandalmostcompletelyreduced themodel'suncertaintyi.e.,therangeoftheresultingprobabilitydistributiondeclinedto nearzero. IncludingWNSaspartofthesimulationsappeared,atfirstglance,toovershadowthe effectsofwindturbinemortality.Windturbinemortalityaffectedthesysteminanuanced andsubtlemanner.InthescenarioswithoutWNS,thelowestwindturbinemortalityrate causedadeclineoflessthan1%intotalpopulationsize,themediummortalityratecaused a3%decline,andthehighmortalityratecauseda6%decline.Thesedifferenceswere Ericksonetal., PeerJ ,DOI10.7717/peerj.2830 7/19

PAGE 8

reducedwithinthelowWNSmortalityscenariosanddisappearedwithinthemediumand highWNSmortalityscenarios.ThisdifferencebetweenWNSscenariosoccurredbecause WNSkilledbatsthatwouldhaveotherwisebeenkilledbywindturbinestrikes.Wealso foundthatturbine-causedmortalityatcolonieswasnegligiblecomparedtomortality alongthemigratorypathways.Presumably,thisisbecausetherewaslittleoverlapbetween modeledsummermaternitycoloniesandwindturbinesandnooverlapbetweenknown hibernaculumcellsandwindturbines. DespitekillingfewerindividualsthanWNS,windturbinesaffectedthemetapopulation dynamicsoftheIndianabatmorethanWNSforallscenariosotherthanthehigh-WNS mortalityscenarioFig.3.WithoutWNS,thelow-andmedium-windmortalityscenarios decreasedthenumberofmigrationpathwaysby6%whereasthehighWNSmortality scenariocausedalmostallofthepathwaystogoextinct<99%. ThelossofmigratorypathwayscorrespondedtothelossofmaternitycoloniesFig.4 andwintercoloniesFig.S1.Windturbinescausedthelossofmaternitycoloniesprimarily intwoclusters:oneinnorthernIllinoisandIndianathewesternclusterandthesecond intheAppalachiansofWestVirginiaandPennsylvaniatheeasterncluster.Thewestern clustercorrespondedtoanareawithmoderateabundancesofIndianabatmaternity coloniesandhighabundancesofwindturbines.Theeasternclustercorrespondedtoan areawithhighabundancesofIndianabatcoloniesandmoderateabundancesofwind turbines.AsWNSmortalityratesincreased,thelossofmaternitycoloniesshiftedsouth andwest.Thiscorrespondedtoashiftfromareaswithwindturbinestoanareawherethe majorityofthewintercoloniesforthespecieshavebeenfound.Itisalsoworthnotingthat someofthelowwindturbinemortalityscenarioshavefewerandmoredisperseddeaths thanotherscenarios.Thespatialdensityplotsreflectedtheuncertainty,spatialvariability, andnumberofmortalitiesinsimulations Wickham,2009 .Plotsfromscenarioswith fewermortalitiesandgreaterspatialvariabilitywouldhavelargershadedregionsthanplots fromscenarioswithmoremortalitiesandlessspatialvariabilitybecausethefirstscenarios havemoreuncertaintyintheconfidenceregions. Theextirpationofwintercoloniesfollowedadifferentpatternthanthelossofmaternity coloniesFig.S1.Windturbinemortalitiesledtothelossofthreeclustersofwinter colonies.OneclusterwasintheAppalachianregionsofPennsylvaniaandWestVirginia andwassimilartotheeasternclusterofmaternitycolonieslost,andwaslocatednearahigh densityofwindturbines.AsecondclusteroccurredmostlyinwesternKentucky,whereas athirdclusteroccurredmostlyinsouthernMissouri.Thesetwoclusterswerenotlocated nearanywindenergygenerationfacilities.AsmodeledWNSmortalityratesincreased,the locationsofwintercolonieslosttendedtobecomemoreevenlydistributedacrosstherange ofwintercolonies. Althoughweonlycompared4differentwindturbinemortalitylevels,interpolationto differentlevelsofmortalitymaybepossibleFig.S2.Ourlowestlevelofwindturbine mortalityhadverylittleeffectonthefinaltotalpopulationsize.Adeclineinpopulation sizeoccurredasmortalityfromwindincreasedbetweenthelowandmediumwindturbine mortalityscenarios.Asmallerdeclineoccurredbetweenthemediumandhighwindturbine mortalityscenarios.Thissmallerdeclinesuggestsalevelingoffofmortalitye.g.,apointof Ericksonetal., PeerJ ,DOI10.7717/peerj.2830 8/19

PAGE 9

Figure4Mapofmaternitycolonieslostunderdifferentexposurescenarios. Thefigureisfacetedon the x -axisbydifferentWNSmortalityscenarios.Thefigureisfacetedonthe y -axisbydifferentwindturbinemortalityrates.Weonlyshowtheresultsfromincludingturbinesfoundwithina2-kmbufferofthe migratorypathway.Wealsodidnotplotthescenariosthatonlyincludedtakeoccurringalongmigratory pathways.Theshadingistherelativedensityofcolonieslost.Thedensityissubplotspecificandonlyqualitativecomparisonsshouldbemadeacrosssubplots.Furthermore,theareaandshadingofthedensity variesacrossplotsbecauseofthetheshadingalgorithmusedby ggplot2 .Thisplottingprogramshrinksthe densityasthenumberofpointsincreasesandthevariabilityamongpointsdecreases. diminishingreturn,suchthatfurtherincreasesinthemortalityratefromwindturbines, astheyarecurrentlyconfiguredacrosstheUnitedStates,wouldhavelittleadditionaleffect becausewindturbinesremovedallgroupsaffectedbyenergygeneration.Thus,thebats aredepopulatedfromtheturbineareaswhenmortalityishigh. DISCUSSION ThecurrentjuxtapositionofwindenergyfacilitieswithintherangeoftheIndianabatmay leadtoameaningfulimpactonthepopulationdynamicsofthespecies,dependinguponthe magnitudeofriskfromcollisionfacedbybatsinmigration.Althoughwindenergymayhave someeffectonthesimulatedtotalpopulationsizeFig.3,theeffectsofwindturbinesonthe metapopulationdynamicsand,specifically,onmigrationalconnectivityoftheIndianabat arelikelymoreimportantowingtothereductioninnumberofmigratorypathwayswithin ourmodelFig.S1.Atthesimulatedratesofmortalityfromturbines,windenergyfacilities holdthepotentialtoextirpatesmallerover-winteringpopulations Barclay&Harder,2003 ; Jones,Purvis&Gittleman,2003 .Survivalofthesesmallersub-populationsislikelycritical forthespeciestosurviveWNSbecausesmallerwintercoloniesappearlessatriskfrom WNS Thogmartinetal.,2012b ; Wilderetal.,2011 .Thisfindingalsohighlightsimportant differencesincompensatoryandadditivemortality.Atthepopulation-level,windturbine developmentandwhite-nosesyndromeappeartobecompensatorysourcesofmortality i.e.,ifwindturbinesdidnotkillIndianabats,white-nosesyndromewouldkillthem Ericksonetal., PeerJ ,DOI10.7717/peerj.2830 9/19

PAGE 10

anyways.However,whenexaminingthetwostressorstogetheratthemeta-population level,thetwostressorsquitelikelyareadditive.Windturbinemortalitywouldbemore likelytoextirpatesmallhibernaculawhereaswhite-nosesyndromewouldbemorelikely toextirpatelargehibernacula.Ourfindingalsoraisesconcernsaboutwindturbinesand WNSproducingasynergisticeffectonthepopulationdynamicsofthespecies,whereeach stressorhasamuchgreaterimpactwhenconsideredjointlythanwouldbeexpectedfrom thatstressoractingalone. CurrentUSFWSmanagementoftheIndianabatfocusesonprotectinglargewinter coloniesbecausemostoftheindividualbatsuseafewcaves Pruitt&TeWinkel,2007 ; Thogmartin&McKann,2014 .Additionally,currentmodelsusedbytheUSFWSfor issuingincidentaltakepermitsignorethespatialstructureofthepopulation Thogmartin etal.,2012b ; Erickson,Thogmartin&Szymanski,2014 .TheUSFWSmaybenefitfrom explicitlyconsideringmetapopulationdynamicsasWNSkillsagrowingportionof thepopulationandwindenergyproductionincreases.Specifically,placingadditional emphasisonprotectingsmallwintercoloniesmaybeprudent Thogmartin&McKann, 2014 .Additionally,amorecompletemodelforWNSmighthelpguideconservation effortsbecausedifferentriskfactorsappeartoaffectsurvival Boyles&Willis,2009 ; Flory etal.,2012 ; Wilderetal.,2011 .Empiricallyquantifyingandunderstandingtheseeffects willbecriticaltounderstandingthedynamicsofthespeciesandthediseaseaffectingit Thogmartinetal.,2013 SimilartoWNS,apaucityofdataexistsformodelinghowwindturbinesaffectIndiana batsurvival.Thisdeficitofdatacreatedsomeofthegreatestuncertaintyinourmodel becauseourwindturbinemortalityscenariosvariedbyordersofmagnitude.Asof2015,the USFWShasonlyreceivedreportsofsevenIndianabatsbeingkilledatwindturbinefacilities Fig.2;http://www.fws.gov/midwest/wind/wildlifeimpacts/inbafatalities.html#Table1. EstimatingthenumberofIndianabatskilledbywindturbinesisdifficultduetoalackof standardizedprotocolsforsamplingwindturbinesforallspecies Huso,2011 ; Huso,2013 hamperingmeta-analysisacrossstudysites Loss,Will&Marra,2013 ; Beston,Diffendorfer &Loss,2015 .Additionally,theIndianabatisdifficulttofindbecauseitisasmallspecies thatdecomposesquicklyafterdeathandisdifficulttocorrectlyidentify Arnettetal., 2011 .Further,thenostandardizedreportingframeworkexistsforwindenergymortality withintheUnitedStates.Abetterunderstandingoftheconditionsunderwhichturbines killIndianabatswouldnotonlyallowabetterunderstandingofthespeciespopulation dynamics,butalsoallowforpossibleprotectivemeasurestobetaken Arnettetal.,2011 Thelackofdataonwindturbinecollisionrisklimitspopulation-levelassessments forallspecies,notjusttheIndianabat Loss,Will&Marra,2013 ; Beston,Diffendorfer& Loss,2015 .Todate,fewstudiese.g., Carreteetal.,2009 ; Schaub,2012 haveexamined range-wideeffectsofwindturbinesonaspecificspecies;oursisthefirsttolookat multiplestressorsatthepopulation-level.Ourfindingsillustratehowmortalityfromwind turbinesinteractswithotherstressors.Modeledwindturbinesstronglyaffectedandoften extirpatedsmallsub-populationswhereasmodeledWNScausedafairlyuniformdecline acrosstheentirerange.Thisresultalsodemonstratestheneedforgreaterunderstandingof compensatorymortalitywhenexaminingincidentaltake McGowanetal.,2011 .Although Ericksonetal., PeerJ ,DOI10.7717/peerj.2830 10/19

PAGE 11

ourfindingsonlyexamineoneanthropogenicstressoraffectingthestabilityofonespecies, similartrendsareemergingworldwidewhereanthropogenicstressorsareaffectingthe stability,biodiversity,andproductivityofEarth'secosystems Dirzoetal.,2014 ; Harley, 2011 ; Hautieretal.,2015 TocapturethesalientlifehistoryaspectsoftheIndianabat,ourFACrequiredsignificant effortatparameterization.Thiseffortmaybepossibleforotherendangeredspeciessuchas thewhoopingcrane Grusamericana Nations,Howlin&Young,2013 ,butisnoteasily scalabletothehundredsofspecieskilledatwindturbines.Forspecieswhereitisnotpossible toconstructhigh-effort,high-inputmodels,probablythefirstandmostimportantquestion toaskwouldbe,``whatistheoverlapbetweenthespeciesrangeandwindturbines?''Asan exampleofsuchanassessment, Santosetal. appliedspatialdistributionmodelingto examinefourspeciesofbatsandwhatfactorsaffectedtheprobabilityofmortalityoccurring atagivenwindenergyproductionfacility.Similarly,workby Roscionietal. modeled theregionaleffectsofwindfarmsonbats,and Roscionietal. modeledtheeffectsof windfarmsonbatmigrationandpopulationconnectivity.Aspartofthespatialoverlap question,itisalsoimportanttonotonlyconsiderthe``where,''butalsothe``when.'' Obviously,aspecieswithnooverlapisnotdirectlyatrisk,butmightbeifwindenergy generationadverselyaffectsanimportantcompetitororpreyspecies. Theothermodelingeffortswedescribedaresimilarinthattheybroadlyseekto understandtheimpactsofwindenergydevelopmentonwildlife.Theseeffortsdiffered, however,ineithertheirscaleormodelingapproaches.Forexample Nations,Howlin& Young constructedanindividual-basedmodelforanextremelyrarespeciesthat wouldhavemoreofalocalizedriskofwindturbinemortality. Santosetal. used speciesdistributionmodelstoexaminespatialdistributionandriskusingdistribution modelingratherthanpopulationmodeling.Effortsby Roscionietal. ; Roscioniet al. weresimilartoourinthattheyexaminedspatialmigrationandnetworks. Specifically, Roscionietal. examinedthespatialconnectivityofabatspeciesinItaly andthepossibleeffectsofwindturbinedevelopmentonthespecies.However,ourapproach differsfrom Roscionietal. becausewefocusedonthepopulationdynamicsofthe species. Anotherimportantconsiderationisthespatialstructureofthepopulation:``Arethere distinctsubpopulationsoristhespecieswellconnectedacrossitsrange?''TheIndiana batformsdistinctsubpopulationsbecauseofitslifehistory,butotherspeciessuchas long-distancemigratorytreebatsorsomeavianspeciesmaynot.Thisspatialconnectivity isalsoimportantifonedecidestoconsiderthepossibilityofre-colonizationofextirpated populations.Thethirdimportantconsiderationthatemergesfromourresultswouldbe, ``Whataretheotherstressorsaffectingthepopulationandhowdotheyinteractwithwind energyproduction?'' Duetothelargenumberofparametersinanduncertaintywithinourmodel,additional researchdataandmodelimprovementcouldbeincorporatedtorefineourapproach. MoresummerfieldobservationsoftheIndianabatwouldbeespeciallybeneficial.A NorthAmericanBatMonitoringProgramisbeingdeveloped,butdoesnotcurrently haveextensivedataontheIndianabat Loebetal.,2015 .Thisdatawouldallowmore Ericksonetal., PeerJ ,DOI10.7717/peerj.2830 11/19

PAGE 12

certaintyinmodelingofsummerhabitat,whichwouldalsoallowforbetterunderstanding ofmigratoryroutes.Additionally, Paulietal. recentlymodeledhowdifferent landscapemanagementscenariosaffecttheIndianabatandmodelssuchastheirscould belinkedtopopulationmodelsifcomputationallimitsanddatalimitationscouldbe overcome.Usinganagent-basedmodelapproach,treatingmigratorygroupsasthe agent,oranindividualbased-modelingapproach,wouldalsoallowourmodeltocapture moresalientbehaviors Grimm&Railsback,2005 .Anotherimportantconsideration iscolonizationandre-colonizationmetapopulationdynamics.Indianabatshavebeen observedcolonizingabandonedmines Pruitt&TeWinkel,2007 ,whichwedidnot explicitlyconsider.Lastly,ourmodeldidnotconsiderdemographicstochasticityor demographicheterogeneity Melbourne&Hastings,2008 .Wedidnotincludethese componentsbecauseofcomputationallimits.Modelingthesewouldhaverequiredusing naturalnumbers,butweusedcontinuousnumbersbecauseitsimplifiedthemodeland decreasedcomputationtime.Additionally,demographicstochasticitywouldrequire theuseofeithercomputationallyintensemethodssuchasbinomialdistributionsor programming-intensemethodssuchasbranchingprocessmodels Caswell,2001 ; Erickson etal.,2015 CONCLUSION Undersomeofthemodeledlevelsofinfluence,windenergyproductionmaydeleteriously affectthepopulationdynamicsoftheIndianabat.Wefoundwindenergyproduction's effectswereprincipallyatthemetapopulation-levelandprimarilyaffectedsmallerwinter colonies.CombinedwithWNS,whichprincipallyaffectslargercolonies,managementmay needtoconsidermetapopulationdynamicsandfocusonprotectingsmallerIndianabat wintercoloniestoreduceriskofspeciesextinction.Ourfindingsalsoillustratethebroader importanceofconsideringFACsandmigratorynetworksratherthansimplyfocusingon localhabitatorhomogeneouslydistributedrange-widepopulations. ACKNOWLEDGEMENTS WethankVolkerGrimmforcommentsandhelpwiththeTRACEDocumentation.We thankLindseyHoffermanfromthePAGameCommissionforsharingtheWNSspread datawithus.WethankJoshTakacsattheGeosciencesandEnvironmentalChangeScience CenterforhelpwithHighPerformanceComputing,otherwise,oursimulationsmightstill berunning.WethankLoriPruitt,RobinNiver,andErikOlsonfromtheUSFWSforreading throughthemanuscriptandprovidingfeedback.WethankJackWaide,MarkGaikowski, RobinWhite,andtwoanonymousreviewersfortheirfeedbackonthismanuscript.Any useoftrade,product,orfirmnamesarefordescriptivepurposesonlyanddonotimply endorsementbytheUSGovernment.Theviewsexpressedinthisarticlearetheauthor's ownanddonotnecessarilyrepresenttheviewsoftheUSFishandWildlifeService.This workwasassistedthroughparticipationintheHabitatforMigratorySpeciesWorking GroupattheNationalInstituteforMathematicalandBiologicalSynthesis,sponsored Ericksonetal., PeerJ ,DOI10.7717/peerj.2830 12/19

PAGE 13

bytheNationalScienceFoundationthroughNSFAward#DBI-1300426,withadditional supportfromTheUniversityofTennessee,Knoxville.ThisprojectwaspartoftheUS GeologicalSurvey'sWindEnergyImpactsAssessmentMethodologyWEIAMproject. ADDITIONALINFORMATIONANDDECLARATIONS Funding ThisworkwasassistedthroughparticipationintheHabitatforMigratorySpeciesWorking GroupattheNationalInstituteforMathematicalandBiologicalSynthesis,sponsoredbythe NationalScienceFoundationthroughNSFAward#DBI-1300426,withadditionalsupport fromTheUniversityofTennessee,Knoxville.ThisprojectwaspartoftheUSGeological Survey'sWindEnergyImpactsAssessmentMethodologyWEIAMproject.Thefunders hadnoroleinstudydesign,datacollectionandanalysis,decisiontopublish,orpreparation ofthemanuscript. GrantDisclosures Thefollowinggrantinformationwasdisclosedbytheauthors: NationalInstituteforMathematicalandBiologicalSynthesis. NationalScienceFoundation:NSFAward#DBI-1300426. TheUniversityofTennessee,Knoxville. CompetingInterests Theauthorsdeclaretherearenocompetinginterests. AuthorContributions RichardA.EricksonandWayneE.Thogmartinconceivedanddesignedtheexperiments, performedtheexperiments,analyzedthedata,wrotethepaper,preparedfiguresand/or tables,revieweddraftsofthepaper. JayE.DiffendorferandRobinE.Russellconceivedanddesignedtheexperiments, contributedreagents/materials/analysistools,wrotethepaper,revieweddraftsofthe paper. JenniferA.Szymanskiconceivedanddesignedtheexperiments,performedthe experiments,contributedreagents/materials/analysistools,wrotethepaper,reviewed draftsofthepaper,provideddataforsynthesis. DataAvailability Thefollowinginformationwassuppliedregardingdataavailability: USGS:http://dx.doi.org/10.5066/F75M63TN. SupplementalInformation Supplementalinformationforthisarticlecanbefoundonlineathttp://dx.doi.org/10.7717/ peerj.2830#supplemental-information. Ericksonetal., PeerJ ,DOI10.7717/peerj.2830 13/19

PAGE 14

REFERENCES AmericanWindWildlifeInstitute.2014. Windturbineinteractionswithwildlifeand theirhabitats:asummaryofresearchresultsandpriorityquestions.Technicalreport. AmericanWindWildlifeInstitute,Washington,D.C. ArnettE,BaerwaldE.2013. Batevolution,ecology,andconservation,chapterimpactsof windenergydevelopmentonbats:implicationsforconservation .NewYork:Springer. ArnettEB,BrownW,EricksonWP,FiedlerJK,HamiltonBL,HenryTH,JainA, JohnsonGD,KernsJ,KofordRR,NicholsonCP,O'ConnellTJ,PiorkowskiMD, TankersleyJRRD.2008. PatternsofbatfatalitiesatwindenergyfacilitiesinNorth America. TheJournalofWildlifeManagement 72 :61DOI10.2193/2007-221. ArnettEB,HusoMM,SchirmacherMR,HayesJP.2011. Alteringturbinespeedreduces batmortalityatwind-energyfacilities. FrontiersinEcologyandtheEnvironment 9 :209DOI10.1890/100103. AugusiakJ,VandenBrinkPJ,GrimmV.2014. Mergingvalidationandevaluationof ecologicalmodelsto`evaludation':areviewofterminologyandapracticalapproach. EcologicalModelling 280 :117DOI10.1016/j.ecolmodel.2013.11.009. BarclayR,HarderL.2003. Batecology,chapterlifehistoriesofbats:lifeintheslowlane Chicago:UniversityofChicagoPress. BestonJA,DiffendorferJE,LossS.2015. Insufficientsamplingtoidentifyspecies affectedbyturbinecollisions. TheJournalofWildlifeManagement 79 :513 DOI10.1002/jwmg.852. BolkerBM.2008. EcologicalmodelsanddatainR .Prinecton:PrincetonUniversityPress. BoylesJG,WillisCK.2009. Couldlocalizedwarmareasinsidecoldcavesreduce mortalityofhibernatingbatsaffectedbywhite-nosesyndrome? FrontiersinEcology andTheEnvironment 8 :92DOI10.1890/080187. CarreteM,Snchez-ZapataJA,BentezJR,LobnM,DonzarJA.2009. Largescale risk-assessmentofwind-farmsonpopulationviabilityofagloballyendangeredlonglivedraptor. BiologicalConservation 142 :2954 DOI10.1016/j.biocon.2009.07.027. CaswellH.2001. Matrixpopulationsmodels .Sunderland:SinauerAssociates. ClarkBK,BowlesJB,ClarkBS.1987. SummerstatusoftheendangeredIndianabatin Iowa. AmericanMidlandNaturalist 118 :32DOI10.2307/2425625. CrimminsSM,McKannPC,SzymanskiJA,ThogmartinWE.2014. Effectsofcavegating onpopulationtrendsatindividualhibernaculaoftheIndianabat Myotissodalis ActaChiropterologica 16 :129DOI10.3161/150811014X683345. CryanPM,MeteyerCU,BlehertDS,LorchJM,ReederDM,TurnerGG,WebbJ,Behr M,VerantM,RussellRE,CastleKT.2013. Electrolytedepletioninwhite-nose syndromebats. JournalofWildlifeDiseases 49 :398DOI10.7589/2012-04-121. DiffendorferJ,ComptonR,KramerL,AnconaZ,NortonD.2014. Onshoreindustrial windturbinelocationsfortheUnitedStatesthroughJuly,2013:dataseries817. Technicalreport.UnitedStatesGeologicalSurvey,Reston. Ericksonetal., PeerJ ,DOI10.7717/peerj.2830 14/19

PAGE 15

DirzoR,YoungHS,GalettiM,CeballosG,IsaacNJB,CollenB.2014. Defaunationin theAnthropocene. Science 345 :401DOI10.1126/science.1251817. DowleM,ShortT,LianoglouS,A.S.withcontributionsfromRSaporta,AntonyanE. 2014. data.table:extensionofdata.frame. Availableathttps://cran.r-project.org/web/ packages/data.table/index.html EricksonRA.2016. IndianaBatprojectdata. USGeologicalSurveydatarelease .Reston: USGeologicalSurvey.DOI10.5066/F75M63TN. EricksonRA,EagerEA,StantonJC,BestonJA,DiffendorferJE,ThogmartinWE.2015. Assessinglocalpopulationvulnerabilitywithbranchingprocessmodels:anapplicationtowindenergydevelopment. Ecosphere 6 :1DOI10.1890/ES15-00103.1. EricksonRA,RussellRE,DiffendorferJE,SzymanksiJA,ThogmartinWE.2014. A stage-structured,spatiallyexplicitmigrationmodelforcolonialspecieswithafocus on Myotis bats. LettersinBiomathematics 1 :1 DOI10.1080/23737867.2014.11414465. EricksonRA,ThogmartinWE,SzymanskiJA.2014. BatTool:anRpackagewithGUIfor assessingtheeffectofwhite-nosesyndromeandothertakeeventson Myotis spp.of bats. SourceCodeforBiologyandMedicine 9 :9DOI10.1186/1751-0473-9-9. FarmerAH,CadeBS,StaufferDF.2002. Evaluationofahabitatsuitabilityindex model.In:KurtaA,KennedyJ,eds. TheIndianabat:biologyandmanagementofan endangeredspecies .Austin:BatConservationInternational,172. FloryAR,KumarS,StohlgrenTJ,CryanPM.2012. Environmentalconditionsassociated withbatwhite-nosesyndromemortalityinthenorth-easternUnitedStates. Journal ofAppliedEcology 49 :680DOI10.1111/j.1365-2664.2012.02129.x. FrickWF,PollockJF,HicksAC,LangwigKE,ReynoldsDS,TurnerGG,Butchkoski CM,KunzTH.2010. Anemergingdiseasecausesregionalpopulationcollapseofa commonNorthAmericanbatspecies. Science 329 :679 DOI10.1126/science.1188594. FrickWF,PuechmailleSJ,HoytJR,NickelBA,LangwigKE,FosterJT,BarlowKE, BartonikaT,FellerD,HaarsmaA-J,HerzogC,HoracekI,VanderKooijJ, MulkensB,PetrovB,ReynoldsR,RodriguesL,StihlerCW,TurnerGG,Kilpatrick AM.2015. DiseasealtersmacroecologicalpatternsofNorthAmericanbats. Global EcologyandBiogeography 24 :741DOI10.1111/geb.12290. GardnerJ,CookE.2002. Seasonalandgeographicdistributionandquantificationof potentialsummerhabitat.In:KurtaA,KennedyJ,eds. TheIndianabat:biologyand managementofanendangeredspecies .Austin:BatConservationInternational,9. GelmanA,CarlinJB,SternHS,DunsonBD,VehtariA,RubinDB.2013. Bayesiandata analysis .BocaRaton:CRCpress. GrimmV,AugusiakJ,FocksA,FrankBM,GabsiF,JohnstonAS,LiuC,MartinBT, MeliM,RadchukV,ThorbekR,RailsbackSF.2014. Towardsbettermodelling anddecisionsupport:documentingmodeldevelopment,testing,andanalysisusing TRACE. EcologicalModelling 280 :129DOI10.1016/j.ecolmodel.2014.01.018. GrimmV,BergerU,BastiansenF,EliassenS,GinotV,GiskeJ,Goss-CustardJ,Grand T,HeinzSK,HuseG,HuthA,JepsenJU,JorgensenC,MooijWM,MullerB, Ericksonetal., PeerJ ,DOI10.7717/peerj.2830 15/19

PAGE 16

Pe'erG,PiouC,RailsbackSF,RobbinsAM,RobbinsMM,RossmanithE,Ruger N,StandE,SouissS,StillmanRA,VaboR,VisserU,DeangelisDL.2006. A standardprotocolfordescribingindividual-basedandagent-basedmodels. Ecological Modelling 198 :115DOI10.1016/j.ecolmodel.2006.04.023. GrimmV,BergerU,DeAngelisDL,PolhillJG,GiskeJ,RailsbackSF.2010. The ODDprotocol:areviewandfirstupdate. EcologicalModelling 221 :2760 DOI10.1016/j.ecolmodel.2010.08.019. GrimmV,RailsbackSF.2005. Individual-basedModelingandEcology .Princeton: Princeton. HallamTG,FedericoP.2009. Ecologicalandbehavioralmethodsforthestudyofbats, chapterapplicationofdynamicpopulationmodelstobats .Baltimore:JohnsHopkins UniversityPress,177. HarleyCD.2011. Climatechange,keystonepredation,andbiodiversityloss. Science 334 :1124DOI10.1126/science.1210199. HautierY,TilmanD,IsbellF,SeabloomEW,BorerET,ReichPB.2015. Anthropogenicenvironmentalchangesaffectecosystemstabilityviabiodiversity. Science 348 :336DOI10.1126/science.aaa1788. HoffmanMD,GelmanA.2014. TheNo-U-Turnsampler:adaptivelysettingpathlengths inHamiltonianMonteCarlo. JournalofMachineLearningResearch 15 :1351. HostetlerJA,SillettTS,MarraPP.2015. Full-annual-cyclepopulationmodelsfor migratorybirds. TheAuk 132 :433DOI10.1642/AUK-14-211.1. HusoM.2013. Estimatingdirectfatalityimpactsatwindfarms:howfarwe'vecome, wherewehaveyettogo.In: PNWWRMIX.Proceedingsofthewind-wildliferesearch meetingIX,Broomfield,CO,Nov.28,2012 .Washington,D.C.:AmericanWind WildlifeInstitute,175. HusoMM.2011. Anestimatorofwildlifefatalityfromobservedcarcasses. Environmetrics 22 :318DOI10.1002/env.1052. JonesKE,PurvisA,GittlemanJL.2003. Biologicalcorrelatesofextinctionriskinbats. TheAmericanNaturalist 161 :601DOI10.1086/368289. KunzTH,ArnettEB,EricksonWP,HoarAR,JohnsonGD,LarkinRP,StricklandMD, ThresherRW,TuttleMD.2007. Ecologicalimpactsofwindenergydevelopment onbats:questions,researchneeds,andhypotheses. FrontiersinEcologyandthe Environment 5 :315DOI10.1890/1540-9295[315:EIOWED]2.0.CO;2. KuvleskyWP,BrennanLA,MorrisonML,BoydstonKK,BallardBM,BryantFC.2007. Windenergydevelopmentandwildlifeconservation:challengesandopportunities. JournalofWildlifeManagement 71 :2487DOI10.2193/2007-248. LangwigKE,FrickWF,BriedJT,HicksAC,KunzTH,MarmKilpatrickA.2012. Socialitydensity-dependenceandmicroclimatesdeterminethepersistenceof populationssufferingfromanovelfungaldisease,white-nosesyndrome. Ecology Letters 15 :1050DOI10.1111/j.1461-0248.2012.01829.x. LangwigKE,FrickWF,ReynoldsR,PariseKL,DreesKP,HoytJR,ChengTL,Kunz TH,FosterJT,KilpatrickAM.2015. Hostandpathogenecologydrivetheseasonal Ericksonetal., PeerJ ,DOI10.7717/peerj.2830 16/19

PAGE 17

dynamicsofafungaldisease,white-nosesyndrome. ProceedingsofTheRoyalSociety ofLondonB:BiologicalSciences 282 :20142335. LoebS,RodhouseTJ,EllisonLE,LausenCL,ReichardJD,IrvineKM,IngersollTE, ColemanJTH,ThogmartinWE,SauerJR,FrancisCM,BaylessML,StanleyTR, JohnsonDH.2015. AplanfortheNorthAmericanBatMonitoringProgram NABat.GeneralTechnicalReportSRS-208.Technicalreport.USDepartmentof AgricultureForestService,SouthernResearchStation,Asheville,NC. LoebSC,WintersEA.2012. Indianabatsummermaternitydistribution:effectsof currentandfutureclimates. EcologyandEvolution 3 :103DOI10.1002/ece3.440. LossSR,WillT,MarraPP.2013. EstimatesofbirdcollisionmortalityatwindfacilitiesinthecontiguousUnitedStates. BiologicalConservation 168 :201 DOI10.1016/j.biocon.2013.10.007. McGowanCP,RyanMR,RungeMC,MillspaughJJ,CochraneJF.2011. Theroleof demographiccompensationtheoryinincidentaltakeassessmentsforendangered species. BiologicalConservation 144 :730DOI10.1016/j.biocon.2010.10.020. MelbourneBA,HastingsA.2008. Extinctionriskdependsstronglyonfactorscontributingtostochasticity. Nature 454 :100DOI10.1038/nature06922. MillerN,DrobneyR,ClawsonR,CallahanE.2002. Summerhabitatinnorthern Missouri.In:KurtaA,KennedyJ,eds. TheIndianabat:biologyandmanagementof anendangeredspecies .Austin:BatConservationInternational,165. NationalResearchCouncil:CommitteeontheStatusofPollinatorsinNorthAmerica. 2007. StatusofpollinatorsinNorthAmerica .Washington,D.C.:NationalAcademies Press. NationsC,HowlinS,YoungDP.2013. Part2:collaborativelandscapeconservation approach:modelingpotentialimpactstomigratorywhoopingcranesfromwind powerdevelopment.In: PNWWRMIX.Proceedingsofthewind-wildliferesearch meetingIX.,Broomfield,CO,Nov.28,2012 .Washington,D.C.:AmericanWind WildlifeInstitute,175. OfficeoftheFederalRegister.1967. Notices. FederalRegister:TheDailyJournalofthe UnitedStates 32 :4001 Availableathttp://www.nmfs.noaa.gov/pr/pdfs/fr/fr324001.pdf OfficeoftheFederalRegister.1973. EndangeredSpeciesActof1973. FederalRegister: TheDailyJournaloftheUnitedStates 93 :884 Availableathttps://www.gpo. gov/fdsys/pkg/STATUTE-87/pdf/STATUTE-87-Pg884.pdf PauliB,ZollnerP,HaultonG,ShaoG,ShaoG.2015. Thesimulatedeffectsoftimber harvestonsuitablehabitatforIndianaandnorthernlong-earedbats. Ecosphere 6 :Article58DOI10.1890/ES14-00336.1. PiorkowskiMD,FarnsworthAJ,FryM,RohrbaughRW,FitzpatrickJW,Rosenberg KV.2012. Researchprioritiesforwindenergyandmigratorywildlife. TheJournalof WildlifeManagement 76 :451DOI10.1002/jwmg.327. PowersKE,ReynoldsRJ,OrndorffW,FordWM,HobsonCS.2015. Post-white-nose syndrometrendsinVirginiascavebats,2008. JournalofEcologyandThe NaturalEnvironment 7 :113DOI10.5897/JENE2015.0507. Ericksonetal., PeerJ ,DOI10.7717/peerj.2830 17/19

PAGE 18

PruittL,TeWinkelL.2007. Indianabat Myotissodalis draftrecoveryplan:first revision.Technicalreport.USFishandWildlifeService,FortSntelling,MN. RCoreTeam.2014. R:alanguageandenvironmentforstatisticalcomputing.Vienna:R FoundationforStatisticalComputing. Availableathttps://www.r-project.org/ RevolutionAnalytics,WestonS.2014. doSNOW:foreachparalleladaptorforthesnow package Availableathttp://CRAN.R-project.org/package=doSNOW RoscioniF,RebeloH,RussoD,CarranzaML,DifebbraroM,LoyA.2014. Amodelling approachtoinfertheeffectsofwindfarmsonlandscapeconnectivityforbats. LandscapeEcology 29 :891DOI10.1007/s10980-014-0030-2. RoscioniF,RussoD,DifebbraroM,FrateL,CarranzaM,LoyA.2013. Regionalscalemodellingofthecumulativeimpactofwindfarmsonbats. Biodiversityand Conservation 22 :1821DOI10.1007/s10531-013-0515-3. SantosH,RodriguesL,JonesG,RebeloH.2013. Usingspeciesdistributionmodelling topredictbatfatalityriskatwindfarms. BiologicalConservation 157 :178 DOI10.1016/j.biocon.2012.06.017. SchaubM.2012. Spatialdistributionofwindturbinesiscrucialforthesurvivalofred kitepopulations. BiologicalConservation 155 :111 DOI10.1016/j.biocon.2012.06.021. SchmolkeA,ThorbekP,DeAngelisDL,GrimmV.2010. Ecologicalmodelssupporting environmentaldecisionmaking:astrategyforthefuture. TrendsinEcology& Evolution 25 :479DOI10.1016/j.tree.2010.05.001. SuterIIGW.2006. Ecologicalriskassessment .BocaRaton:CRCpress. TaylorCM,NorrisDR.2010. Populationdynamicsinmigratorynetworks. Theoretical Ecology 3 :65DOI10.1007/s12080-009-0054-4. ThogmartinWE,KingRA,McKannPC,SzymanskiJA,PruittL.2012a. Populationlevelimpactofwhite-nosesyndromeontheendangeredIndianabat. Journalof Mammalogy 93 :1086DOI10.1644/11-MAMM-A-355.1. ThogmartinWE,KingRA,SzymanskiJA,PruittL.2012b. Spacetimemodelsfora panzooticinbats,withafocusontheendangeredIndianabat. JournalofWildlife Diseases 48 :876DOI10.7589/2011-06-176. ThogmartinWE,McKannPC.2014. Large-scaleclimatevariationmodifiesthewinter groupingbehaviorofendangeredIndianabats. JournalofMammalogy 95 :117 DOI10.1644/13-MAMM-A-098. ThogmartinWE,Sanders-ReedCA,SzymanskiJA,McKannPC,PruittL,KingRA, RungeMC,RussellRE.2013. White-nosesyndromeislikelytoextirpatetheendangeredIndianabatoveralargepartofitsrange. BiologicalConservation 160 :162 DOI10.1016/j.biocon.2013.01.010. TurnerGG,ReederD,ColemanJT.2011. Afive-yearassessmentofmortalityand geographicspreadofwhite-nosesyndromeinNorthAmericanbats,withalookat thefuture.Updateofwhite-nosesyndromeinbats. BatResearchNews 52 :13. VerantML,CarolMU,SpeakmanJR,CryanPM,LorchJM,BlehertDS.2014. Whitenosesyndromeinitiatesacascadeofphysiologicdisturbancesinthehibernatingbat host. BMCPhysiology 14 :10DOI10.1186/s12899-014-0010-4. Ericksonetal., PeerJ ,DOI10.7717/peerj.2830 18/19

PAGE 19

WarneckeL,TurnerJM,BollingerTK,MisraV,CryanPM,BlehertDS,Wibbelt G,WillisCK.2013. Pathophysiologyofwhite-nosesyndromeinbats:amechanisticmodellinkingwingdamagetomortality. BiologyLetters 9 :20130177 DOI10.1098/rsbl.2013.0177. WatanabeS.2010. AsymptoticequivalenceofBayescrossvalidationandwidelyapplicableinformationcriterioninsingularlearningtheory. TheJournalofMachine LearningResearch 11 :3571. WeberT,SparksD.2013. Summerhabitatidentificationofanendangeredbat Myotis sodalis ,acrossitseasternrangeoftheUSA. JournalofConservationPlanning 9 :53. WickhamH.2009. ggplot2:elegantgraphicsfordataanalysis .NewYork:Springer. WiederholtR,Lpez-HoffmanL,ClineJ,MedellnRA,CryanP,RussellA,McCracken G,DiffendorferJ,SemmensD.2013. Movingacrosstheborder:modelingmigratory batpopulations. Ecosphere 4 :Article114DOI10.1890/ES13.1. WilderAP,FrickWF,LangwigKE,KunzTH.2011. Riskfactorsassociatedwith mortalityfromwhite-nosesyndromeamonghibernatingbatcolonies. BiologyLetters 7 :950DOI10.1098/rsbl.2011.0355. WillisC,MenziesA,BoylesJ,WojciechowskiM.2011. Evaporativewaterlossisa plausibleexplanationformortalityofbatsfromwhite-nosesyndrome. Integrative andComparativeBiology 51 :364DOI10.1093/icb/icr076. WinholdL,KurtaA.2006. AspectsofmigrationbytheendangeredIndianabat, Myotis sodalis BatResearchNews 47 :1. YatesM,MuzikaR.2006. EffectofforeststructureandfragmentationonsiteoccupancyofbatspeciesinMissouriOzarkforests. JournalofWildlifeManagement 70 :1238DOI10.2193/0022-541X[1238:EOFSAF]2.0.CO;2. Ericksonetal., PeerJ ,DOI10.7717/peerj.2830 19/19