Content uploaded by Zach Cravens
Author content
All content in this area was uploaded by Zach Cravens on Oct 19, 2021
Content may be subject to copyright.
J Appl Ecol. 2017;1–9. wileyonlinelibrary.com/journal/jpe
|
1
© 2017 The Authors. Journal of Applied Ecology
© 2017 British Ecological Society
Received:12July2017
|
Accepted:24October2017
DOI:10.1111/1365-2664.13036
RESEARCH ARTICLE
Illuminating prey selection in an insectivorous bat community
exposed to artificial light at night
Zachary M. Cravens1 | Veronica A. Brown2 | Timothy J. Divoll3 | Justin G. Boyles1
1CooperativeWildlifeResearch
Laboratory,DepartmentofZoology,Southern
IllinoisUniversity,Carbondale,IL,USA
2UniversityofTennesseeGenomicsCore
Facility,Knoxville,TN,USA
3CenterforBatResearch,Outreach,and
Conservation,IndianaStateUniversity,Terre
Haute,IN,USA
Correspondence
ZacharyM.Cravens
Email:zcravens@siu.edu
Funding information
MissouriDepartmentofConservation
HandlingEditor:MatthewStruebig
Abstract
1. Lightpollutionhasbeenincreasingaroundtheglobeandthreatenstodisturbnatu-
ralrhythmsofwildlifespecies.Artificiallightimpactsthebehaviourofinsectivo-
rousbatsinnumerousways,includingforagingbehaviour,whichmayinturnlead
toalteredpreyselection.
2. Inamanipulativefieldexperiment,wecollectedfaecalsamplesfromsixspeciesof
insectivorousbatsinnaturallydarkandartificiallylitconditions,andidentifiedprey
items using molecular methods to investigate effects of light pollution on prey
selection.
3. Proportionaldifferencesinidentifiedpreywerenotconsistentandappearedtobe
speciesspecific.Redbats,little brown bats and grey bats exhibited expected in-
creasesinmothsatlitsites.Beetle-specialistbigbrownbatshadasizeableincrease
in beetle consumption around lights, while tri-coloured bats and evening bats
showed little change in moth consumption between experimental conditions.
Dietary overlap was high between experimental conditions within each species,
anddietarybreadthonlychangedsignificantlybetweenexperimentalconditionsin
onespecies,thelittlebrownbat.
4. Policy implications.Ourresults,buildingonothers,demonstratethatbat–insectin-
teractionsmaybemorenuancedthanthecommonassertionthatmothconsump-
tion increases around lights. They highlight the need for a greater mechanistic
understandingofbat–light interactions to predict whichspecieswill be most af-
fectedbylightpollution.Givendifferencesinbatandinsectcommunities,weadvo-
catebiologists,landstewardsandcivilplannersworkcollaborativelytodetermine
lighting solutions that minimize changes in foraging behaviour of species in the
localbatcommunity.Sucheffortsmayallowstakeholderstomoreeffectivelycraft
management strategies to minimize unnatural shifts in prey selection caused by
artificiallights.
KEYWORDS
allotonicfrequencyhypothesis,artificiallight,bat–insectinteractions,bats,dietaryoverlap,
earedmoth,faecalDNA,LED,Lepidoptera,lightpollution
2
|
Journal of Applied Ecology
CRAVENS Et Al.
1 | INTRODUCTION
The biological world is orderedaround the natural rhythm of alter-
natingnightand day.Asareliablesignalovergeologictime, mostor-
ganismshaveevolvedinrelationto temporalcycles oflightanddark
periods (Gaston, Bennie, Davies, & Hopkins, 2013). However, fast-
paced urbanization beginning in the 20th centuryhas led to a dra-
matic increase in artificial light at night (ALAN; Holker, Moss, etal.,
2010). Global light pollution is increasing and has nearly doubled
overthe past25years(Holker,Moss,etal., 2010). Currently,almost
90%of Europeand halfthe United Statesexperiences light-polluted
skies (Falchi etal.,2016), but those levels have remained relatively
constantoverthelastseveraldecades(E. L.Koenetal.unpubl.data).
Conversely,developing regionswith above-averagespeciesrichness
haveexperiencedrecentincreasesinlightpollutionextentcompared
to areas with low to moderate richness (E. L. Koen etal. unpubl.
data).This trendwill likelycontinueas the majority ofurbangrowth
isexpectedto occurnearcurrentlyprotected land (i.e. dark refugia;
Güneralp&Seto,2013).Encroachmentofartificiallightintoremaining
darkareaswillincreasinglythreatenbiodiversityas30%ofvertebrates
and >60% of invertebratesare nocturnal and therefore likely to be
stronglyimpactedbyALAN(Holker,Wolter,Perkin,&Tockner,2010).
Most bats have evolved unique behavioural and morphological
adaptations (e.g. echolocation) to navigate in the absence of light
(Neuweiler,1990;terHofstede&Ratcliffe,2016).Avoidanceofliten-
vironmentsislikelyasignificantultimatecauseofnocturnalityinbats
becauseitreducessusceptibilitytopredationbyvisualhunters,suchas
diurnalbirdsofprey(Rydell&Speakman,1995;Speakman,2001;Voigt
&Lewanzik,2011).This selectivepressureis strongenough thatbats
generallyemergefromroostsjustaftersunset(Duverge,Jones,Rydell,
&Ransome,2000),despiteapulseofinsectactivityjustpriortosunset
(Rydell,Entwistle, & Racey, 1996). Therefore,bats seem to prioritize
darkerconditionsoverahigherenergeticpay-offundernaturalcondi-
tions,andtheglobalpervasivenessofALANmayaffectthistrade-off.
Artificiallightatnightimpactsbatspeciesinnumerousways,often
leading to roost abandonment, spatial avoidanceand delayed emer-
gence (reviewed in Rowse, Lewanzik, Stone, Harris,& Jones, 2016;
Stone,Harris, &Jones,2015). Impacts on bat foragingbehaviour are
lessclearanddependontaxon-specifictraitsandenvironmentalcon-
ditions.Forexample,clutter-adaptedbatsgenerallyavoidlitconditions,
whetherinaconsistentlyliturbanorsemi-urbanenvironmentorinan
experimentallylit environment (Lacoeuilhe, Machon, Julien, LeBocq,
& Kerbiriou, 2014; Schoeman, 2016; Stone,Jones, & Harris, 2009).
Thisislikelybecauselight-intolerantspeciesmayassociateapredatory
riskwithlitenvironments(Jones &Rydell, 1994).Conversely,numer-
ous species have been observed feeding at artificial lights (Acharya
&Fenton,1999; Clare, Fraser,Braid, Brock Fenton,& Hebert,2009;
Rydell,1992;Svensson&Rydell,1998).Artificial lightinterfereswith
insectnavigationalcues,causingattractiontoandunusuallyhighden-
sitiesaround lights (vanLangevelde, Ettema,Donners,WallisDeVries,
&Groenendijk,2011).Higherdensitiesalone maymakeaerialinsects
more vulnerable to predation from bats, but in some prey species,
changesinbehaviouraroundlightsmayalsoplayanimportantrole.For
example,artificiallightappearstointerferewithhighlyevolvedmecha-
nismsearedmothsusetodetectbatecholocationandavoidpredation
(Acharya&Fenton,1999;Svensson&Rydell,1998;Wakefield,Stone,
Jones,&Harris,2015).Observationsofbatsforagingatlightsareusu-
allyinurbanorsemi-urbanareas(except,seeMinnaar,Boyles,Minnaar,
Sole,& McKechnie,2015),wherestreetlightsarea consistentpart of
thenocturnalenvironment.Fromthesestudies,apatternhasemerged
thatconsumptionofmoths,specificallyearedmoths,increasesatlights
(Belwood & Fullard, 1984; Hickey & Fenton, 1990; Minnaar etal.,
2015;Svensson&Rydell,1998).However,theuniversalityofthispat-
ternisunclear,bothwithinandacrossbatcommunities.
Weevaluatedeffectsoflightpollutiononpreyselectionofbatsat
acommunitylevel.Thebatcommunityinthestudyareaisrepresented
byspecieswithdifferentwingmorphologies,foraginghabitsanddiets,
soifthegeneralpatternofincreasedmothconsumptionaroundlights
isfoundinall membersofthiscommunity,thepatternis likelytobe
robust.To testthispattern,wemanipulatednaturallydarkareaswith
a short-term artificial light treatment. We collected faecal samples
frombatscapturedinbothlit andunlit environmentsandused next-
generationsequencing ofinsectDNAextracted fromfaecalsamples
tomeasuredifferencesinfrequencyofinsectpreybetweenunlitand
lit conditions. We predicted bat consumption of moths (including
earedmoths) toincreaseand consumption ofbeetlestodecrease in
artificiallighttreatmentsrelativetonaturallydarkareas.
2 | MATERIALS AND METHODS
2.1 | Study site
Ourstudy was conducted in a 15-county region of western–south-
western Missouri, USA, during summer (May–August) 2016. The
eastern half of the study area is within the Ozark Highlands physi-
ographicregion, whichisa heavilyforestedlandscape dominatedby
oak-hickory forests. To the west, the land transitions to the Osage
Plains, a region historically dominated by prairie but now heavily
convertedtoagriculturewithlimitedforestandwoodlands(Raeker,
Fleming,Morris,Moser,&Trieman,2010).
2.2 | Experimental design
We erected temporary lights along naturally dark forest roads or
streamsonpubliclandsandhadtwoexperimentalconditions:unlit
(control) and lit (light pollution treatment). Distance between lit
and unlit sites was at least 2km to minimize overlap in foraging
rangesbyindividualbats,butsiteswerechosenwithsimilarhabitat
andlandscape features.At litsites,we used50-WLED (Shenzhen
Lepower Opto Electronics Co., China) producing 4,200 lumens at
5,500K.Lightswereelevated3mfromthegroundonametalpole
and powered by a 12-V lead acid battery. We used LED lighting
asitisbecoming morecommon inoutdoorlightingapplicationsas
olderstyles,suchasmercuryvapour,arebeingphasedout.Wenet-
tedeachsurveylocationforthreenightsandranlightsforallthree
nightsfrom21.00to05.00hr.Onthefirsttwonights,wecaptured
|
3
Journal of Applied Ecology
CRAVENS Et Al.
batsat a nearbyunlitsiteas a control andonthe third night cap-
turedbats atlit sites(Minnaaretal., 2015).Delaying captureatlit
sitesuntilthethirdnightallowedbatstobecomeaccustomedtothe
litcondition, aswellas provide timeforthem to choosetoforage
inthenewlylitenvironment.We makenoassumptionthatallbats
capturedatlit sites will necessarily be foraging around the lights;
moreover,weexpect some species may be less prone to foraging
atlights than others and thereforeshow less pronounced dietary
shifts.Nets were placed in flyways within 25mof the light in an
appropriatenettinglocation.
Wenettedalongforestedroadsorstreamsat20locationsthrough-
outthesummer.Weheldbatsinclothbagsfor30–45min, storedall
deposited faecal pellets in 1.5-ml microcentrifuge tubes with silica
beadsandassignedauniquesampleIDtoallowrandomsubsampling,
whennecessary,formolecularanalysis.Sampleswerekeptfrozenafter
thefieldseasonat−20°Cfor4monthsbeforeprocessingforDNA.
2.3 | Molecular analysis
We extracted DNA from one to three pellets of guano from each
individualbatusingPowerSoil®DNAIsolationKit(MoBioLaboratories
Carlsbad,CA)followingmanufacturer’sspecifications,withtheminor
modificationofincreasingthefirst4°Cstepfrom30mintoovernight.
Wediscarded samples withinsufficientfaecal matter (<1fullpellet).
Redbat(Lasiurus borealis)samplesweretoonumeroussowesubsam-
pledbyrandomly selectinglit andunlit pairsfrom thesamesite.The
analytical methodology follows T. J. Divoll etal. (unpubl. data), and
wehaveincludedadetaileddescriptionoftheworkflowinDataS1.
2.4 | Data analysis
Sequences were analysed using the QIIME (www.qiime.org) platform
(Caporasoetal.,2010)andtheworkflowoutlinedinT.J.Divolletal.
(unpubl. data) (https://github.com/tdivoll/bat-diet-metabarcoding)
withoneadditionalstepto only keep sequences within 10 bp of our
targetamplicon.Forwardandreversereadswerejoined,andprimerse-
quenceswereclipped. Wefiltered outsequences smallerthan147bp
orgreater than 167bp. Sequences were clustered intomolecularop-
erationaltaxonomic units(MOTUs)using the SWARMmethodwith a
resolutionof2(Mahé,Rognes,Quince,deVargas,&Dunthorn, 2014).
ToaccountforpotentialOTUinflation,weexcludedMOTUsthatwere
not present at least 10 times in at least one sample. We performed
filteringusingacustom Pythonscriptemployingthe“pandas”package
(McKinney,2010).WeconductedfurtherfilteringofremainingMOTUs
byconsideringwithin-sampleMOTU occurrences<10aspotentialse-
quencingerrors and removing them.Weextracted representative se-
quences from each MOTU cluster, based on abundance, to compare
againstareferencedatabase(T.J.Divolletal.unpubl.methods).
WethencomparedtherepresentativesetofsequencestotheCOI
databaseinBOLD(Ratnasingham&Hebert,2007)usingthepackage
“bold”(Chamberlain, 2017)inr (R CoreTeam,2016).Weconsidered
onlythefirst40 recordsforeach representativeMOTUandthenfil-
tered recordswith <98% similarity and country of origin outside of
UnitedStatesand Canada.The entireoutput foreachrepresentative
was then separated into two groups:high quality with at least one
match(≥99.36% similarity)andlow qualitywith allmatches(>98.0%
but <99.36% similarity). We made taxonomic identifications based
onthese groupings, and in all cases where there was disagreement,
identificationwasmadeatthenexthighestleveloftaxonomy.Inthe
high-quality group, matches <99.36% did notchange the identifica-
tion,regardlessoftaxonomicdivergence,andinthelow-qualitygroup,
variation in per cent match was not considered for identification,
onlydisagreement.Asanexample,agivenMOTUhas a100%match
fromtheBoldpackage outputforthemothAristotelia rubidella(fam-
ily:Gelechiidae),and a 98.92% match for the moth Hillia iris (family:
Noctuidae).Because the second match is less than 99.36% (a single
basepairdifferenceassuming157bp)andthefirstmatchis≥99.36%,
weidentified thepreyitemasA. rubidella. IftheH. iris had ≥99.36%
matchthen, because there wasdisagreementat the family level,we
wouldhaveidentified the item only as Lepidoptera.UniqueMOTUs
assignedto the same taxonomywere collapsedintoa single MOTU,
representingonebatpreyitem.Thismayleadtocertainordersbeing
over or under split due to differences in genetic variation (Brown,
Chain, Crease, MacIsaac, & Cristescu, 2015); however, this should
notbias ourresults whenmeasuringwithin-specieschange between
experimentalconditions.
2.5 | Statistical analysis
Wecalculatedpercent frequency of occurrence of insect prey or-
ders (number of samples containing an order divided by the total
occurrencesof allorders) foreachbat speciesin bothexperimental
conditions.WithinorderLepidoptera,wealsocalculatedpercentfre-
quencyofoccurrenceofearedmothsforeachbatspeciesasfollows:
We defined families Sphingidae, Noctuidae, Notodontidae,
Geometridae and Pyralidae as eared moths, as they are known to
havetympanateorgansused forpredatoravoidance(terHofstede &
Ratcliffe,2016).Wewereunabletoquantifyabundanceofpreyitems
givenvariationininsectDNAdegradationasitpassesthroughabats
intestinaltractanddifferencesinPCRamplification.Forallotheranal-
yses,weusedthe collapsedsetofuniqueMOTUassumedto bebat
preyspecies.
We used the EcoSimR 0.1.0 package (Gotelli, Hart, & Ellison,
2015)in r to determinedietaryoverlap amongthesixbat species
and to assess effects of artificial light. Nullmodels were used to
determinewhether extent of niche overlapwaslowerthan would
beexpectedbychance.WeusedPianka’s(1973)measureofniche
overlap and generated 1,000 bootstraprandomizations of MOTU
diet composition using the “ra3” algorithm. We conducted this
analysisincludingall MOTUs(all-preyanalysis)aswellas excluding
preyonlyeatenbyasingleindividual(common-preyanalysis;asper
Brownetal., 2014;Clare,Symondson, Broders,etal., 2014; Clare,
No. of samples with eared moths
No. of eared moth occurrences in dataset
4
|
Journal of Applied Ecology
CRAVENS Et Al.
Symondson,&Fenton,2014).Weusedthe iNEXTpackage (Hsieh,
Ma,&Chao,2016)inrtodetermineextentofdietaryspecialization
anddiversityusingthefirstthreeHillnumbers(oreffectivenumber
ofspecies):q=0(speciesrichness),q=1(exponentialofShannon’s
entropyindex)andq=2(inverseofSimpson’sconcentrationindex)
aswell as the chao2 asymptotic estimatorforthose numbers. Hill
numbershave been increasinglyusedforbiodiversityanalysisand
arepreferredoverotherdiversityindicesgiventheyareintuitiveand
statisticallyrobust(Chaoetal.,2014).
3 | RESULTS
Wecaptured453batsfromsixspecies(bigbrownbats[Eptesicus fus-
cus];red bats;grey bats [Myotis grisescens];little brownbats[Myotis
lucifugus]; evening bats [Nycticeius humeralis]; and tri-coloured bats
[Perimyotis subflavus]) across both experimental conditions (n = 297
duringunlitandn=151duringlit) spanning61nights(n=42during
unlitand n=19 during lit).Lightdidnot appear to attractnewspe-
ciesaswecapturedmostoftheexpectedspeciesbasedonregional
species distributions, at both lit and unlit sites. We analysed DNA
from 188 faecal samples from the six species (Table 3) and recov-
ered71,992,648sequencing reads. After performing bioinformatics
processing, these reads were clustered and filtered down to 3,078
MOTUs. Using representative sequences of the 3,078 MOTUs, we
identified1,129(36.7%)withmatchingsequencesintheBOLDda-
tabase,belonging to 15 insectorders.After collapsing MOTUs with
thesametaxonomy,wewereleftwith487uniqueMOTUsorunique
preyitems.
Ingeneral,Lepidoptera,ColeopteraandDipterawerethemostcom-
monly identified orders and their combinedproportion was relatively
constant(range~69% to~83%)foreachbat speciesinboth treatment
groups. Specifically, Coleoptera were the most commonly identified
preyfor big brown and evening bats, and Lepidopterawere the most
common preyfor red and little brown bats in both treatmentgroups.
Dipterawerethemostcommon preyidentifiedin the diet of greyand
tri-coloured batsat unlit sites, but the most common-prey items atlit
siteswereLepidopteraforgreybatsandColeopterafortri-colouredbats.
Basedonorder-leveltaxonomyofprey,greybatsweretheonly
species with a significant shift between treatments (χ2=10.11,
p =.02), although significance is lost after a Bonferronicorrection
(Figure1).Further, this maybe relatedto our smallertotalsample
size for this species. Little overallvariation in prey selection was
detectedin anyotherspecies (p >.15).Analysisofdietaryoverlap
valuestellasimilarstory(resultsofall-preyandcommon-preyanal-
yseswere similar; therefore, all-preyvaluesarereported).Overlap
betweenlit and dark treatment groups exceeded0.6, thevalue at
whichdietsaregenerallyconsideredtorepresentbiologicalsimilar-
ity(Pianka&Pianka,1976),forallspecies(seeTable1).Withinaspe-
cies,redbatshadthehighestdegreeofoverlapbetweenlitandunlit
conditions(Ojk0.906,p <.001).Theresultsweregenerallylesscon-
clusivewhenwelimited ourcomparisonofoverlapvaluesbetween
treatment groups toprey items identified as Lepidoptera, but red
batsstillhadasignificantdegreeofoverlap(Ojk0.9059,p <.001).In
general,valuesfordietaryoverlapbetweenspeciespairswerelower
thanthosefoundwithinspeciesbetweentreatmentgroups(Table2).
Further,evenqualitativeshiftsinconsumptionofthetwomost im-
portantpreyitems, Coleopteraand Lepidoptera,variedacrossspe-
cies(Figure1).Therewasalsonoindicationoftheexpectedincrease
ineared-mothconsumptionaroundlights,andtheonlyspecieswith
asignificantshiftinmothsidentifiedasearedmoths,bigbrownbats
(p <.007),consumedfewerearedmothsinthelittreatment.
Diversityestimates showed that dietary breadth did not change
substantiallybetween experimentalconditionsformostspecies,and
FIGURE1 TheproportionofMOTUsidentifiedinthedietofsixspeciesofinsectivorousbatsunderexperimentallylitandnaturallydark
conditions
Proportion of MOTUs by Order
Coleoptera
Lepidoptera
Diptera
Other
.6
.4
.2
Big brown bat
Lit Unlit Lit Unlit
Lit Unlit Lit Unlit Lit Unlit Lit Unlit
0
Red bat Grey bat Little brown bat Evening bat Tri-coloured bat
|
5
Journal of Applied Ecology
CRAVENS Et Al.
noclearpattern existsin thedirection ofchange(Table3). Onlylittle
brownandtri-colouredbatshadnooverlapinthe95%confidencein-
tervalsfordiversityaccumulationcurvesofthefirstHillnumber(q=0)
using the Chao2 estimation of incidence-based richness estimation
(Figure2).Thelittlebrownbatistheonlyspeciesthathadsignificantly
higherestimateddietarydiversityinlitconditionsinallthreediversity
measures.Ingeneral,redbatshadthebroadestdietarydiversity,while
bigbrownbatshadthenarrowest,andthispatternheldforeachHill
numberwhetherobservedorestimated(Table3).
4 | DISCUSSION
Wedetermineddietofsixspeciesofinsectivorousbatstoexaminethe
impactofALAN,atthecommunitylevel,onpreyselection.Contraryto
expectations,nospeciesinthiscommunity showeda significantshift
indiet as seen in another study using a similar experimental design.
Further,evenignoringstatisticalsignificance,ourdatadonotsupport
aconsistenttrendinshiftsindietarynichebetweennaturallydarkand
experimentally lit conditions that would suggest an existing pattern
weare missing due tolowpower.Proportional differences in identi-
fiedpreyappeartobespecies-related.Redbats,littlebrownbatsand
greybats followedtheexpected pattern atlit sites withhighermoth
andlowerbeetleconsumptionfrequencies.Bigbrownbatsarebeetle
specialists,andtherewasasubstantialincreaseintheproportionof
beetlesidentifiedunder lit conditions. Evening bats and tri-coloured
batsshowednochangeinmothorbeetleproportionsunderdarkand
litconditions.Therewasahighdegreeofdietaryoverlapforallspecies
betweentheexperimentalconditions(Ojk > 0.719forallspecies).This
maybeabiologicalresult,suggestingthateitherbatsdidnotchooseto
forageinanartificiallylitconditionorthatbatsdidnotselectdifferent
preyinthepresenceoflight.Alternatively,thismaybeamethodologi-
callimitation as we areunableto determine true abundanceofeach
preyitemwithinanindividualbat,sotheamountof aparticular prey
itemmay change without a change intheproportionof unique prey
itemsidentifiedinouranalysis.Additionally,dietarybreadthwassimi-
larbetweenlitandunlitsites,exceptforlittle brownandtri-coloured
bats.Therewasahighdegreeofoverlapinthe95%confidenceinter-
valsbetweentreatmentgroupsintheinterpolationandextrapolation
curvesofdietarybreadthfortheotherfourspecies(Figure2).Overall,
diversityandbreadthestimates suggestbatswere notfeeding selec-
tivelyonadistinctpreygroupinthepresenceoflight.
Pairwise comparisons between species, within each treatment
group,providefurtherevidence forspecies-specific changes indiet,as
opposedtoanoverallpatterncommontoallspecies(Table2).Forexam-
ple,thedegreeofoverlapbetweenbigbrownbatsandredbatswasless
atlitsites(Ojk =0.345)thanunlitsites(Ojk =0.536),suggestingincreased
dietarydifferentiationinthepresenceoflight.Similarly,littlebrownbats
andgreybatsexhibitedthegreatestdietaryoverlapwithbigbrownbats
atunlitsites,andredbatsatlitsitesbecauseofincreasedconsumptionof
Lepidoptera.Finally,eveningbatshadahighdegreeofoverlapwithbig
brownbatsatunlitsites(Ojk =0.705),whichistobeexpectedasevening
All MOTUs Common- prey analysis
Observed
mean
p (observed
≥expected)
Observed
mean
p (observed
≥expected)
Alltreatmentsandspp. 0.70446 <.001 0.70968 <.001
Littreatmentallspp. 0.60420 <.001 0.61796 <.001
Controlallspp. 0.66239 <.001 0.66866 <.001
Lit/unlittreatment
Bigbrownbat 0.85907 <.001 0.86922 <.001
Redbat 0.90643 <.001 0.91358 <.001
Greybat 0.71912 <.001 0.74098 <.001
Littlebrownbat 0.81780 <.001 0.83112 <.001
Eveningbat 0.82670 <.001 0.83715 <.001
Tri-colouredbat 0.74422 <.001 0.75555 <.001
Lepidopteraalltreatmentsandspp. 0.64481 <.001 0.66121 <.001
Lepidopteralitallspp. 0.45916 <.001 0.48400 <.001
Lepidopteracontrolallspp. 0.52352 <.001 0.54295 <.001
Lepidopteralit/unlit
Bigbrownbat 0.48110 .250 0.50903 .190
Redbat 0.90586 <.001 0.91503 <.001
Greybat 0.56171 .570 0.60760 .470
Brownbat 0.59638 <.001 0.62704 .004
Eveningbat 0.56600 .083 0.57387 .099
Tri-colouredbat 0.45748 .766 0.49490 .713
TABLE1 Dietoverlapbetweenthesix
speciesofinsectivorousbatsevaluatedin
thisstudy.Observedmeanvaluesbelow
0.6aregenerallyacceptedtorepresent
biologicallysignificantresourcepartitioning
6
|
Journal of Applied Ecology
CRAVENS Et Al.
batstypically preferColeoptera (Feldhamer,Whitaker,Krejca,& Taylor,
1995;Whitaker,1972).However,thedegreeofoverlapdecreasesinthe
presenceoflight(Ojk =0.544),becauseeveningbatswerenotexploiting
higherconcentrationsofbeetlesatlitsitesaswerebigbrownbats.
We found little evidence of increased consumption of eared
moths under artificially lit treatments; in fact, eared-moth propor-
tionsdecreased(althoughnot significantly) at lit sites for most spe-
cies.Conversely,Capeserotinebats(Neoromicia capensis)significantly
increasedeared-mothconsumptionatexperimentallymanipulatedlit
sitesinSouthAfrica(Minnaaretal.,2015).Nearlyeverymothspecies
(92.9%)wasidentifiedas anearedmoth atthatstudysite,while the
proportionofearedmothsinthecommunitywestudiedislikelycon-
siderably lower(Dodd, Lacki, & Rieske, 2008). Our results mayalso
beanartefactofouruseoftheBold Systemsdatabase asnumerous
potential eared moths had multiple family-level identifications and
were thus only identifiedto the ordinal level. This may be because
thesemothspeciesare notyetinthe BoldSystemsdatabaseorthat
wehadsequenceddegradedDNA.
Based on our currentwork and that of others, we propose four
responsesintermsofpreyselectionbybatsaroundlight.First,known
specialists may take advantage of artificial light-induced phototaxis
(van Langevelde etal., 2011) to increaseprey consumption of their
preferredprey.Inthebatcommunitywestudied,twodietaryspecial-
ists(big brown and red bats) consumed proportionally more of their
preferredprey at lit sites. Big brown bats,with their powerful jaws,
preferbeetles(Agosta,2002;Clare,Symondson,etal.,2014),whilered
batsprefersofter-bodiedLepidoptera(Acharya&Fenton,1999;Clare
etal.,2009).Second,somegeneralistspeciesmayshowdietaryshifts
TABLE2 Pairwisecomparisonofdietoverlapbetweensixspeciesofinsectivorousbatsinlitandnaturallyunlitexperimentaltreatments.
Valuesbelowthediagonalaretheobservedmean,andnumbersabovethediagonalarethecorrespondingpvalues
Big brown bat Red bat Little brown bat Grey bat Evening bat Tri-coloured bat
Lit
Bigbrownbat 0.037 0.001 0.073 0.022 0.001
Redbat 0.34471 0.001 0.001 0.001 0.001
Littlebrownbat 0.57593 0.65952 0.001 0.001 0.001
Greybat 0.46833 0.68382 0.77017 0.002 0.001
Eveningbat 0.54435 0.43585 0.63214 0.59877 0.001
Tri-colouredbat 0.64095 0.54714 0.76412 0.69536 0.70188
Unlit
Bigbrownbat 0.001 0.001 0.001 0.001 0.002
Redbat 0.53586 0.001 0.001 0.001 0.001
Littlebrownbat 0.76932 0.6034 0.001 0.001 0.001
Greybat 0.64075 0.51874 0.81686 0.001 0.001
Eveningbat 0.70514 0.58402 0.76158 0.74654 0.001
Tri-colouredbat 0.54008 0.57741 0.67231 0.69197 0.77191
TABLE3 Diversityestimatesbetweenexperimentallylitandnaturallydarkconditionsinsixspeciesofinsectivorousbats
Bat species
n (samples
analysed)
Treatment
group Richness
q = 1, Shannon diversity
effective no. of MOTUs
q = 2, Simpson diversity
effective no. of MOTUs
Obs. Est. Obs. Est.
Bigbrownbat 7Lit 66 54.40 85.67 44.50 59.98
14 Unlit 100 69.19 97.66 52.36 59.34
Redbat 35 Lit 200 111.89 164.19 68.26 74.86
39 Unlit 213 119.98 157.49 74.33 80.70
Greybat 7Lit 107 89.76 148.73 74.57 107.04
9Unlit 108 81.69 134.30 61.59 77.21
Littlebrownbat 9Lit 119 89.51 176.47 65.47 83.73
29 Unlit 150 87.57 114.84 57.70 62.41
Eveningbat 6 Lit 73 60.29 101.71 49.63 67.52
16 Unlit 120 81.45 105.00 58.58 66.18
Tri-colouredbat 6 Lit 91 75.31 146.47 60.63 86.83
11 Unlit 95 72.87 95.91 55.64 66.89
|
7
Journal of Applied Ecology
CRAVENS Et Al.
toincludegreaterconsumptionofmothsaroundlights.Inourcommu-
nity, twogeneralist species (grey bat and little brown bat) exhibited
sucha pattern ofincreased mothconsumptionand decreasedbeetle
consumptionunder artificial light. Greybats had a 64.9% increase in
Lepidopterapreyatlitsites,thehighestwithin-orderpercentincrease.
Third,somespeciesmayshownoshiftinpreyselectionaroundlights.
Twospecies(eveningbatandtri-colouredbats) exhibitedlittle differ-
ence in the proportion ofbeetles and moths under the two experi-
mentalconditions. Forthesespecies,thelack of dietary changemay
berelatedto their morphology.Tri-colouredbatsareclutter adapted
(Menzeletal.,2005), andwhile eveningbatsarenotcompletelyclut-
teradapted,theyareweakfliers(Norberg&Rayner,1987);therefore,
thesebatsmaybeavoidingthelightsto avoidpredation.Fourth,spe-
cialist species may decrease consumption of their preferredprey in
favourof moths around lights. No species in our study showedthis
response,butithasbeennotedinCapeserotinebats,abeetlespecial-
istthatincreasesconsumptionofmothsaroundlights(Minnaaretal.,
2015).
Artificiallightingatnighthasvariedeffectsonbatspeciesandthe
mechanismgoverningbehaviouralresponsestolightisunclear.Ingen-
eral,species with morphological adaptationsthatfavourfasterflight
inrelativelyunclutteredhabitatsareconsideredlight-tolerantspecies
(Rowseetal.,2016).Thesespeciesoftenfeedonpositivelyphototac-
tic prey aroundtemporally stable light sources, such as streetlights
(Schoeman, 2016). Conversely, slower flying species with greater
manoeuvrabilityto forageinand around cluttered habitats are con-
sideredlight-intolerant(Rowseetal., 2016).Thesespeciesareoften
foundinlowerdensitiesinartificiallylitenvironmentsandmayactively
avoidartificiallight,although presumablylight-intolerantMyotisspe-
cieshavebeenrecordednearsingle,experimentallightsetupsindes-
ertenvironments(Bell, 1980;Fenton& Morris,1976). Itmaybethat
light-intolerantspeciesinnon-desert regionsarenot avoiding lights,
butrathertheopenhabitatinwhichstreetlightsarefound.Evenlight-
tolerantspecies seem to preferstreetlightsinrural areas overurban
landscapes(Geggie&Fenton,1985).
ThespectralcompositionoftheLEDsused inthisexperimentmay
furtherexplainthelackofaconsistentresponseinourexperiment.LEDs
donotinducephototaxistothesamedegreeasotherlightsources,es-
peciallymercuryvapour(Eisenbeis&Eick,2011;Huemer,Kühtreiber,&
Tarmann, 2010),likelybecause LEDsdonotproducelightinthelower
UVspectrum (Stone etal., 2015). Other formsoflight which lack UV
light,suchashighpressuresodium, also attract fewer insects (Rydell,
2005).Lightsourceswithlowerinsectabundancehavesignificantlyless
batactivity (Blake,Hutson, Racey,Rydell, &Speakman,1994);in fact,
batactivitycanchangebyasmuchas anorderofmagnitudedepend-
ingon lighting technology(Rydell,1992). Interestingly,light-intolerant
batsdonotappearasaverseto LEDsas othertechnologies (Lewanzik
&Voigt,2017).LoweraversionmayberelatedtoUVasevidencesug-
gestslight-intolerantbatsareavoiding UVlightspecifically (Gorresen,
Cryan,Dalton,Wolf,&Bonaccorso,2015).The lackofUVin LEDlight
maychangetheperceptionbythesebatsleadingtodecreasedaversion.
Therefore,LEDlightingmayhavelessofanegativeimpact,atleastwith
respecttoforaging,forbatsandtheirinsectprey.
Numerousstudieshavereportedbatsfeedingatartificiallights(Hickey
& Fenton, 1990; Minnaar etal., 2015; Rydell, 1992; Schoeman, 2016),
and some of these studies havecompared differences in diet with unlit
sitesto determine adietaryshift (Hickey& Fenton,1990; Minnaaretal.,
2015).A patternhasemergedthat bats generally consumemoremoths,
andmoreearedmothsspecifically,underartificiallight.Inparticular,much
oftheworkon effectsofartificial lightonforagingbatsinNorthAmerica
has focused on hoary,red and Hawaiian hoary bats (Acharya & Fenton,
1992,1999; Belwood &Fullard,1984;Fullard,2001;Hickey,Acharya,&
Pennington,1996;Hickey& Fenton,1990;Jacobs,1999).Hoaryandred
bats are generallyconsidered moth specialists; therefore, an increase in
mothconsumptionaroundlightsmaybeexpected(andissupportedbyour
results).Thelackofconsistentdietarychangeinourstudymayberelated
tothebroaderrangeofspeciessampledandsuggestscautioninassuming
auniversalresponsein dietaryshiftsaroundlights forallspecies.Theoft-
citedpattern, which isquicklybecoming a paradigm,thatALAN leads to
increasesinmothconsumptionininsectivorousbatsmaynotbethe case
FIGURE2 Interpolation(rarefaction)andextrapolationofdietaryspeciesrichnessforeachexperimentalconditioninsixspeciesof
insectivorousbatsusingtheChao2estimationforincidence-basedsampledata.Richnessisextrapolatedtotwicethesamplesizeand
bootstrapped500times
Species diversity
Interpolation Extrapolation (95% CI)
Unlit (actual sample size)
Lit
Number of sampling units
Big brown bat
Grey bat
Evening bat
Red bat
Tri-coloured bat
Little brown bat
8
|
Journal of Applied Ecology
CRAVENS Et Al.
forall species. Our results underscorethe need forabettermechanistic
understandingofinteractiveeffectsoflightsonbatsandtheirinsectprey
topredictwhichbatspecieswillbemoststronglyaffectedbylightsandto
craftmanagementplanstolimitnegativeeffectsoflightsonforagingbats.
ACKNOWLEDGEMENTS
We thank the Missouri Department of Conservation (MDC) for fi-
nancialsupportandwithinMDC,KellyRezac,Tony ElliottandShelly
Colatskie for logistical support. Jenna Holub provided assistance in
thefield. Gary McCracken provided laboratory space and materials.
WealsothankThomas Lilley and one anonymous reviewer for pro-
vidingcommentsonthemanuscript.Allexperimentsreportedherein
wereapproved bytheAnimal Care andUse Committee ofSouthern
IllinoisUniversity(15-044).
AUTHORS’ CONTRIBUTIONS
Z.M.C. and J.G.B. conceived the idea and experimental methodol-
ogy;Z.M.C.collectedthedata;andZ.M.C.,T.J.D.andV.A.B.analysed
thedata.Allauthorscontributedcriticallytothedraftsandgavefinal
approvalforpublication.J.G.B.oversawtheproject.
DATA ACCESSIBILITY
Underlyingdata are available in the Dryad DigitalRepository https://
doi.org/10.5061/dryad.gm6kk(Cravens,Brown,Divoll,&Boyles,2017).
ORCID
Zachary M. Cravens http://orcid.org/0000-0002-6680-738X
REFERENCES
Acharya,L., & Fenton,M. B. (1992). Echolocationbehaviourofvespertil-
ionid bats (Lasiurus cinereus and Lasiurus borealis) attackingairborne
targets including arctiid moths. Canadian Journal of Zoology, 70,
1292–1298.
Acharya,L., & Fenton, M.B.(1999).Batattacks and moth defensive be-
haviouraroundstreetlights.Canadian Journal of Zoology,77,27–33.
Agosta,S.J.(2002).Habitatuse,dietandroostselectionbythebigbrown
bat(Eptesicus fuscus)inNorthAmerica:Acaseforconservinganabun-
dantspecies.Mammal Review,32,179–198.
Bell,G. P.(1980). Habitatuseand response topatchesof preyby desert
insectivorousbats.Canadian Journal of Zoology,58,1876–1883.
Belwood,J.J.,&Fullard,J.H.(1984).Echolocationandforagingbehaviour
intheHawaiianhoarybat, Lasiurus cinereus semotus. Canadian Journal
of Zoology,62,2113–2120.
Blake,D.,Hutson,A. M.,Racey,P.A.,Rydell,J., &Speakman,J.R. (1994).
Useof lamplit roadsbyforaging bats in SouthernEngland.Journal of
Zoology,234,453–462.
Brown,D.S.,Burger,R.,Cole,N.,Vencatasamy,D.,Clare,E.L.,Montazam,
A.,& Symondson, W.O. C. (2014). Dietarycompetition between the
alien Asian musk shrew (Suncus murinus) and a re-introduced pop-
ulation of Telfair’s skink (Leiolopisma telfairii). Molecular Ecology, 23,
3695–3705.
Brown,E. A., Chain, F.J.J., Crease,T.J.,MacIsaac, H. J., & Cristescu,M.
E.(2015).Divergencethresholdsanddivergentbiodiversityestimates:
Can metabarcoding reliably describe zooplankton communities?
Ecology and Evolution,5,2234–2251.
Caporaso, J. G., Kuczynski, J., Stombaugh, J., Bittinger, K., Bushman, F.
D.,Costello,E. K.,… Knight,R. (2010).QIIME allowsanalysis ofhigh-
throughputcommunitysequencingdata.Nature Methods,7,335–336.
Chamberlain, S. (2017). bold: Interface to BoldSystems ‘API’. R package
version0.4.0.https://cran.r-project.org/package=bold
Chao,A.,Gotelli,N.J.,Hsieh,T.C.,Sander,E.L.,Ma,K.H.,Colwell,R.K.,&
Ellison,A.M.(2014).RarefactionandextrapolationwithHillnumbers:
Aframeworkforsampling and estimationin speciesdiversity studies.
Ecological Monographs,84,45–67.
Clare,E.L., Fraser,E.E.,Braid,H. E.,BrockFenton,M., &Hebert,P.D.N.
(2009).Speciesonthe menuof ageneralist predator,theeastern red
bat(Lasiurus borealis):Usingamolecularapproachtodetectarthropod
prey. Molecular Ecology,18,2532–2542.
Clare,E. L., Symondson,W. O.C., Broders,H., Fabianek,F.,Fraser,E. E.,
MacKenzie, A., … Reimer, J. P. (2014). The diet of Myotis lucifugus
acrossCanada:Assessingforagingqualityanddietvariability.Molecular
Ecology,23,3618–3632.
Clare,E.L.,Symondson,W.O.C., &Fenton,M.B.(2014).Aninordinatefond-
nessforbeetles?variationinseasonaldietarypreferencesofnight-roosting
bigbrownbats(Eptesicus fuscus).Molecular Ecology,23,3633–3647.
CoreTeam, R.(2016).R: A language and environment for statistical comput-
ing.Vienna,Austria:RFoundationforStatisticalComputing.
Cravens,Z.M.,Brown,V.A.,Divoll,T.J.,&Boyles,J.G.(2017).Datafrom:
Illuminating prey selection in an insectivorous bat community, ex-
posed to artificial light at night. Dryad Digital Repository, https://doi.
org/10.5061/dryad.gm6kk
Dodd,L.E.,Lacki,M.J.,&Rieske,L.K.(2008).Variationinmothoccurrence
and implications for foraginghabitat of Ozark big-eared bats. Forest
Ecology and Management,255,3866–3872.
Duverge,P. L.,Jones, G., Rydell, J., & Ransome, R. D. (2000). Functional
significanceofemergencetiminginbats.Ecography,23,32–40.
Eisenbeis,G.,&Eick,K.(2011).StudiezurAnziehungnachtaktiverInsekten
an die Straßenbeleuchtung unterEinbeziehung von LEDs. Natur und
Landschaft,86,298–306.
Falchi,F.,Cinzano,P.,Duriscoe,D.,Kyba,C.C.M.,Elvidge,C.D.,Baugh,K.,
…Furgoni,R.(2016).Thenewworldatlasofartificialnightskybright-
ness.Science Advances,2,e1600377.
Feldhamer,G.A.,Whitaker,J.O., Krejca,J.K., &Taylor,S.J.(1995). Food
of the eveningbat (Nycticeius humeralis) and red bat (Lasiurus borea-
lis) from Southern Illinois.Transactions of the Illinois State Academy of
Science,88,139–143.
Fenton,M.B.,&Morris,G.K.(1976).Opportunisticfeedingbydesertbats
(MyotisSpp).Canadian Journal of Zoology,54,526–530.
Fullard,J.H.(2001).Auditorysensitivity ofHawaiianmoths(Lepidoptera:
Noctuidae) and selective predation by the Hawaiian hoary bat
(Chiroptera:Lasiurus cinereus semotus).Proceedings of the Royal Society
B- Biological Sciences,268,1375–1380.
Gaston,K.J.,Bennie,J.,Davies,T.W.,&Hopkins,J.(2013).Theecological
impactsofnighttimelightpollution:Amechanisticappraisal. Biological
Reviews,88,912–927.
Geggie,J.F.,&Fenton,M.B.(1985).AcomparisonofforagingbyEptesicus
fuscus(Chiroptera,Vespertilionidae)in urban and rural environments.
Canadian Journal of Zoology,63,263–267.
Gorresen, P. M., Cryan, P. M., Dalton, D. C., Wolf, S., & Bonaccorso,
F. J. (2015). Ultraviolet vision may be widespread in bats. Acta
Chiropterologica,17,193–198.
Gotelli,N.J.,Hart,E.M.,&Ellison,A.M.(2015).EcoSimR:Nullmodelanal-
ysisforecologicaldata.Rpackageversion0.1.0.
Güneralp, B., & Seto, K. C. (2013). Futures of global urban expan-
sion: Uncertainties and implications for biodiversity conservation.
Environmental Research Letters,8,014025.
|
9
Journal of Applied Ecology
CRAVENS Et Al.
Hickey,M.B.C.,Acharya,L.,&Pennington,S.(1996).Resourcepartitioning
bytwospeciesofvespertilionidbats(Lasiurus cinereus and Lasiurus bo-
realis)feedingaroundstreetlights.Journal of Mammalogy,77,325–334.
Hickey,M. B. C., & Fenton, M. B. (1990). Foragingby red bats (Lasiurus
borealis):Do intraspecificchases mean territoriality? Canadian Journal
of Zoology,68,2477–2482.
terHofstede,H.M.,&Ratcliffe,J.M.(2016).Evolutionaryescalation:The
bat-motharmsrace.Journal of Experimental Biology,219,1589–1602.
Holker,F., Moss,T., Griefahn, B., Kloas, W.,Voigt, C. C., Henckel, D.,…
Tockner, K.(2010). Thedark sideoflight:Atransdisciplinaryresearch
agendaforlightpollutionpolicy.Ecology and Society,15,1–11.
Holker,F.,Wolter,C.,Perkin,E.K.,&Tockner,K.(2010).Lightpollutionasa
biodiversitythreat.Trends in Ecology & Evolution,25,681–682.
Hsieh, T. C., Ma, K. H., & Chao, A. (2016). iNEXT: iNterpolation and
EXTrapolationforspeciesdiversity.Rpackageversion2.0.12.
Huemer, P., Kühtreiber, H., & Tarmann, G. M. (2010). Anlockwirkung
moderner Leuchtmittel auf nachtaktive Insekten - Ergebnisse einer
Feldstudie in Tirol. Tiroler Landesumweltanwaltschaft & Tiroler
LandesmuseenBetriebsgesellschaftm.b.H.,Innsbruck,Austria.
Jacobs, D. S. (1999). The diet of the insectivorous Hawaiian hoary bat
(Lasiurus cinereus semotus)inanopenandaclutteredhabitat.Canadian
Journal of Zoology,77,1603–1608.
Jones,G.,&Rydell,J. (1994).Foragingstrategyandpredationriskasfac-
tors influencing emergence time in echolocating bats. Philosophical
Transactions of the Royal Society B- Biological Sciences,346,445–455.
Lacoeuilhe,A.,Machon,N.,Julien,J.-F.,LeBocq,A.,&Kerbiriou,C.(2014).
Theinfluenceoflowintensitiesoflightpollutiononbatcommunitiesin
asemi-naturalcontext.PLoS ONE,9,e103042.
Lewanzik,D.,& Voigt,C.C.(2017). Transitionfromconventionalto light-
emittingdiodestreetlightingchangesactivityofurbanbats.Journal of
Applied Ecology,54,264–271.
Mahé, F., Rognes, T., Quince, C., de Vargas, C., & Dunthorn, M. (2014).
Swarm:Robustandfastclusteringmethodforamplicon-basedstudies.
Peerj,2,e593.
McKinney,W.(2010).Datastructuresforstatistical computingin Python.
Proceedingsofthe9thPythoninScienceConference,pp.51–56.
Menzel,J.M.,Menzel,M.A.,Kilgo,J.C.,Ford,W.M.,Edwards,J.W.,&McCracken,
G. F.(2005). Effect of habitat and foraging height on bat activity in the
CoastalPlainofSouthCarolina.Journal of Wildlife Management,69,235–245.
Minnaar,C.,Boyles,J.G.,Minnaar,I.A.,Sole,C.L.,&McKechnie,A.E.(2015).
Stackingthe odds: Light pollution may shift the balance in an ancient
predator-preyarmsrace.Journal of Applied Ecology,52,522–531.
Neuweiler,G.(1990).Auditoryadaptationsforpreycaptureinecholocating
bats.Physiological Reviews,70,615–641.
Norberg,U.M.,&Rayner,J.M.V.(1987).Ecologicalmorphologyandflight
inbats(Mammalia,Chiroptera)–Wingadaptations,flightperformance,
foraging strategy and echolocation. Philosophical Transactions of the
Royal Society of London Series B- Biological Sciences,316,337–419.
Pianka,E.R.(1973).Thestructureoflizardcommunities.Annual Review of
Ecology and Systematics,4,53–74.
Pianka,E. R.,& Pianka,H. D.(1976). Comparativeecologyof twelvespe-
ciesofnocturnallizards(Gekkonidae)intheWesternAustraliandesert.
Copeia,1,125–142.
Raeker,G.,Fleming,J.,Morris,M.,Moser,K.,&Trieman,T.(2010).Missouri’s for-
est resource assessment and strategy: Seeking a sustainable future for Missouri’s
forest resources.JeffersonCity,MO:MissouriDepartmentofConservation.
Ratnasingham, S., & Hebert, P. D. N. (2007). BOLD: The barcode of life
datasystem.Molecular Ecology Notes,7,355–364.http://www.barcod-
inglife.org/
Rowse, E. G., Lewanzik, D., Stone, E. L., Harris, S., & Jones, G. (2016).
Darkmatters: The effects of artificial lighting on bats. In C. C.Voigt,
&T.Kingston(Eds.),Bats in the Anthropocene: Conservation of bats in a
changing world(pp.187–213).NewYork,NY:Springer.
Rydell, J. (1992). Exploitation of insects around streetlamps by bats in
Sweden.Functional Ecology,6,744–750.
Rydell,J. (2005). Bats and their insect prey at street-lights.InC. Rich, &
T. Longcore (Eds.), Ecological consequences of artificial night lighting
(pp.43–60).Washington,DC:IslandPress.
Rydell,J.,Entwistle,A.,&Racey,P.A.(1996).Timingof foragingflightsof
threespecies of bats in relation to insect activity and predationrisk.
Oikos,76,243–252.
Rydell,J.,&Speakman,J.R. (1995).Evolutionofnocturnalityin bats-po-
tentialcompetitors andpredatorsduring theirearly history.Biological
Journal of the Linnean Society,54,183–191.
Schoeman,M.C.(2016).Lightpollutionatstadiumsfavorsurbanexploiter
bats.Animal Conservation,19,120–130.
Speakman,J. R. (2001). The evolution of flight and echolocation in bats:
Anotherleapinthedark.Mammal Review,31,111–130.
Stone,E.L.,Harris,S.,&Jones,G.(2015).Impactsofartificiallightingonbats:
Areviewofchallengesandsolutions.Mammalian Biology,80,213–219.
Stone,E.L.,Jones,G.,&Harris,S.(2009).Streetlightingdisturbscommut-
ingbats.Current Biology,19,1123–1127.
Svensson,A.M., &Rydell,J.(1998). Mercuryvapourlampsinterferewith
thebatdefenceoftympanatemoths(Operophteraspp.;Geometridae).
Animal Behaviour,55,223–226.
Van Langevelde, F., Ettema, J. A., Donners, M.,WallisDeVries, M. F., &
Groenendijk,D.(2011).Effectofspectralcompositionofartificiallight
ontheattractionofmoths.Biological Conservation,144,2274–2281.
Voigt,C.C., & Lewanzik,D. (2011).Trappedinthedarkness ofthenight:
Thermalandenergeticconstraintsofdaylightflightinbats.Proceedings
of the Royal Society B- Biological Sciences,278,2311–2317.
Wakefield,A., Stone, E. L., Jones, G., & Harris, S. (2015). Light-emitting
diodestreet lightsreduce last-ditchevasivemanoeuvresbymothsto
batecholocationcalls.Royal Society Open Science,2,150291.
Whitaker,J.O.(1972).Foodhabits ofbatsfrom Indiana.Canadian Journal
of Zoology,50,877–883.
SUPPORTING INFORMATION
Additional Supporting Information may be found online in the
supportinginformationtabforthisarticle.
How to cite this article:CravensZM,BrownVA,DivollTJ,
BoylesJG.Illuminatingpreyselectioninaninsectivorousbat
communityexposedtoartificiallightatnight.J Appl Ecol.
2017;00:1–9. https://doi.org/10.1111/1365-2664.13036
