Population structure of the damselfly Calopteryx japonica Selys in an isolated small habitat in a cool temperate zone of Japan (Zygoptera: Calopterygidae)

12 Pages·1998·2.7 MB·English

by  Watanabe, M

Odonatologica27(2):213-224 June 1. 1998
Populationstructure ofthedamselfly
Calopteryxjaponica Selys inanisolatedsmall
habitatinacool temperatezoneofJapan
(Zygoptera: Calopterygidae)
M.Watanabe¹M.Taguchi² andN.Ohsawa³
1 DepartmentofBiology, Faculty ofEducation,Mie University,Tsu514-8507,Japan
e-mail:[email protected]
2 Yaei-HigashiHighSchool, Sagamihara229-1133, Japan
3 Kumon Kokusai Junior-Senior High School,Totsuka,Yokohama244, Japan
Received May 14, 1997/Revised andAcceptedJuly 1, 1997
Thepopulationstructure was studiedinasmall streaminacooltemperatezoneof
Japanin 1989and 1990, usingamark-release-recapturemethod. Theestimateddaily
numberofmaleswas500(1989)and 150(1990),whilethatoffemales was450(1989)
and 100(1990). Theoperationalsexratioineach yearwasprobablyunity. Thedaily
estimate numberofimmigrantsin each yearwasca. 10%ofthepopulation,and the
dailysurvival ratewasmorethan 80%. Therefore,thepopulationsinbothyears were
considered tobeclosed. The distributionofeach individual perchingwassurveyed.
Theperching siteofthe damselflydependedupon the sunlit areaonthe bank ofthe
stream. Accordingly,theyconcentrated onthe west bankduringthemorning,and the
east bank duringthe afternoon. However,alongthe bank theperchingsites ofmales
wasshowntobe aregulardistribution,due toterritorial behaviour. Thereweremany
small insects thatcould be prey forthe damselfliesalongthe streamand theedgeof
thepaddyfieldsnearby. Dipteraninsects weredominantpotentialprey inthishabi-
tat. Since the habitatofthe damselflyis surrounded bypaddy fieldsand lakes, the
maintenance ofthepopulationprobablydependsupon the abundance ofsubstratefor
oviposition andthelarval habitat.
INTRODUCTION
TheodonatefaunaonrunningwaterinthecooltemperatezoneofcentralHonshu,
Japan ischaracterizedbyspecies foundby smallriversinsomeplains andinsome
alpine landscapes. The formerhabitats are mainly used forrice paddy fields. In
214 M.Watanabe, M.Taguchi& N.Ohsawa
Oh-machiCity, Nagano Prefecture,thereareseveralisolatedplains surroundedby
mountainsorhills, andwenoticedthatfourspecies ofCalopterygidae, Calopteryx
japonica, C. atrata, C. corneliaandMnaispruinosa costalisare abundantthere.
Thehabitatofthedamselfliesconsists mainly ofslow-flowing streamswithmuddy
bottoms and such streams service the paddy fields. In thesummer, the streams
provide a perfect habitatfordamselflies, particularly C. japonica and C. atrata.
Thereisunpolluted water,abundantaquatic vegetation and someshady areaspro-
ducedby bankvegetation. ISHIDA(1976) pointed outthat Calopteryxpopulations
oftenattainhigh densities.
population is restricted to open habitats and after emer-
gence,itstays bythewater andshows territorialbehaviourwhensexually mature
(MIYAKAWA,1983). Immatureadultsdonot leavetheiremergencesitesandas
suchare atypical (CORBET, 1962).Therefore,eachC. japonicapopulation seems
tobearelatively closedpopulation ifsuitablehabitatsare isolated.Theimmature
periodisnormallytheprincipal dispersal,buttheabsenceofdispersalinC.japonica
means that the population structure ofthis species couldbe intensively studied.
However,littleattentionhasbeenpaid tothepopulation ecology ofC. japonicain
Japan, though SUZUKI& TAMAISHI (1981) studied thepopulation parameters
ofthe closely related C. Cornelia.The reproductive behaviourof Calopteryx
damselflieshasbeendescribedby many investigators (e.g. HIGASHI & UEDA,
1982; MIYAKAWA, 1982; WAAGE, 1984; HOOPER, 1995; PLAISTOW &
SIVA-JOTHY, 1996).
Theaim ofthepresentstudy was toobtain quantitative informationontheadult
population structureof C.japonica. We used mark-recapture techniques to esti-
mate population sizeand longevity. The abundanceofprey insects was also as-
sessed. Sincethematingbehaviourwas observedduringtheexperiments, thepopu-
lationstructure will alsohelp inunderstanding the C.japonica mating system.
MATERIALS ANDMETHODS
PopulationsofC. japonicawerestudied in an isolated plainofOh-machi City. The habitat con-
sistedofasmall stream(230 m long, 10m wide), which wasrelativelystraightfrom north to south
becauseitserved asachannelbetween twolakes (LakeNakatsuna andLake Kizaki). Bothsides ofthe
streamweresurrounded bypaddyfields. Towardsthe south ofthe streamthe smallvalleywithnarrow
paddyfields continuestoLake Kizaki.
AdultsofC.japonicaflyoverthe streamfromlateJunetomidAugust.Theaquaticvegetationwas
abundant and available forovipositionand larval life in 1989 but decreased in 1990. A halfofthe
aquaticvegetationwhich providedlarval habitats wascut offin the springof1990.
Adultcensuseswerecarriedoutalongbothsidesofthe streamon mostlycleardaysfor 14daysand
7 days,inlateJulytomidAugustof 1989and 1990,respectively, duringwhichthematingactivityof
C.japonicawasconsidered tobehigh.Ittookabout2hourstopatrolonesideofthestream.Toclarify
thedistribution ofperching sites,visual censuswasalso donefrom0500to 1800 onthe 8th August
1989. Plantsonthebanks wereoccasionallytrimmedby farmers,andthe vegetationwasabout0.5-1
m talland the commonheightoftheperching site of C.japonicawas usuallyless than 1 m. In
Populationstructure inCalopteryxjaponica 215
addition, werecorded thenumber ofaggressiveflightscarriedoutagainstothermales, thenumberof
copulationsand the numberoffemales ovipositing throughoutthe 8thAugust 1989.
Adultswerecapturedwithanetandmapped.Eachwas givenanindividualnumberthatwasmarked
withawhite felt-tippedpen ontheundersurface oftheleft hindwing.Everydamselflycapturedwas
released immediatelyatthe samecapture site afterrecordingthe date, site number,sexand age.The
markingwasconsidered tohave minoreffectsontheir flightactivities (andtheprobabilityofmating)
becausemost damselfliesbegan tofly normallysoon aftermarkingandthen toperch. Damselflies
woundedduringmarkingweretreatedasdeadindividuals in thecalculations. The age ofthedamselfly
wasestimated intotwoclasses (immatureand matureadults),mainlyby wingandbodyconditions,as
inthecaseoftheotherdamselflyspecies(e.g.WATANABE, 1991).Inmostpopulations,therecapture
probabilitiesofthe twosexesweredifferent(e.g.WATANABE&HIGASHI, 1989). The sexesofC.
japonicawereprocessedseparately throughJolly’sanalysis (JOLLY, 1965).
On8th August 1989, small insects thatwereregarded asprey fordamselflies inthehabitat were
collected around noon.We used4survey lines (50min each):inside the emergent plants,abovethe
emergent plants (1.5m above watersurface), abovethe bank plants(0.5m above the ground)and
abovetherice plants(heightofrice plantwas0.5 m)next tothebank. Thisdivisionisrelatedtothe
real compositionofthe habitat with differentheightsfromthe ground.AplateofA4-size (0.062m2)
witha sticky surface wasprepared towhich small insects wouldadhere.Wemoved theplatealong
each survey line at 5m/s,that is,wecould collect small insects thatexistedin thespace of3.1 m3
(=0.062m2x50 m)/10s.Samplingforeach survey linewasrepeatedfor 5times.Each insect adhered
totheplatewas examined under amicroscope(x40)anditsbody lengthwasmeasured.
RESULTS
C. japonica passes through one generation per year. As in many calopterygid
species, there is sharp behavioural difference between males and females.
Fig. 1. Changesin estimateddailynumber ofmaleCalopteryxjaponicain 1989 and 1990withthose
forimmature(squares) and maturemales(triangles),calculatedbyJolly’smethod. Thebars repre-
sent standard deviations.
216 M.Watanabe, M.Taguchi& N.Ohsawa
Long-term observations (unpublished) show that males incessantly stay and de-
fendperching sites againstrivals,whilethefemalesflyaboutthestream withinter-
mittentfeeding, perchingandovipositing. Notonly didthiscause arelativelysmall
proportion ofthefemalestobemarked inthestudy stream, buttheprobability of
recapturing amarked femalewas lowerthanthat ofrecapturing amarkedmale.
In 1990,therewas ashortenedemergenceperiod duetolittleprecipitation inthe
rainy season.Someaquatic vegetation witheredduetolackofwater,thenahalfof
the remaining aquatic vegetation hasbeencut andremoved by fanners. In addi-
tion, 1990was marked by a severe mid-summerdrought withextraordinary high
temperatures. Consequently, the drought may have reduced the adult numbers
both by interfering with larvalgrowthandby accelerating theageing ofadults.
Adultswere foundin thewholehabitat, butwere concentratedonboth sidesof
thestream. These areas supplied adultperching andoviposition sites. Sincemost
adults perched onthe plants except during territorialflights, the mark-recapture
work was donealongthebanksofthestream.Theadultmalesdefendedterritories
and a few ofthem were alsoobserved engaged in sneaking behaviouron some
days. When sneaking, males did not defendterritoriesbut attempted to steal fe-
malesfromothermales’territories.
Figure 1 showsestimatedpopulation densitiesofmaleswiththeir standardde-
viations (SD). In 1989,therewas a relatively constant numberof about200
immatureand300 mature males,respectively. The estimatedpopulation size in
1990was muchsmallerwithroughly 50immatureand 100mature males,respec-
tively.
Fig. 2.Changes in estimated dailynumber offemale Calopteryxjaponicain 1989 and 1990with
those forimmature (squares)andmaturefemales (triangles), calculated by Jolly’smethod. The bars
represent standard deviations.
Population structure in Calopteryxjaponica 217
TableI
Estimated daily immigrants,dailysurvival rates and expected longevities- ofCalopteryxjaponica
(±SE)
Numberof Expectedlongevity
immigrants Survivalrate (days)
1989 <J immatures 45±12 (5) 0.80±0.07 ( 8) 5.0
matures 8±17 (6) 082±0.09 (10) 5.6
5 immatures 19±20 (6) 0.82±0.09 ( 7) 5.6
matures 12±29 (7) 0.93±0.02 (10) 14.3
1990 S immatures l± 4 (4) 0.63±0.11 ( 5) 2.7
matures 6± 1 (4) 0.87±0.06 ( 5) 7.7
5 immatures - (0) 1.00 ( 1) -
matures 7± 5 (4) 0.87±0.05 ( 5) 7.7
*: Expected longevity (L)wascalculated by thefollowingequationusingextimated dailysurvival
rate(S):L=l/(1-S).
():Numberofdaysextimated.
In 1989,theestimatednumbersoffemaleswerealsostablethroughoutthestudy,
withabout 100 immatureand350 mature females,respectively (Fig.2). Thus,in
1989,theoperational sex ratiowas closeto unity.However,in 1990,wecollected
a fewimmature females, mostof which maturedduringthe experimental period
andnoneofwhichwererecaptured asimmatureadults.Theestimateddailynumber
offemaleswas about 100.
TableI shows the estimatednumberofdaily immigrants. In 1989,daily immi-
grants ofimmatureadults were about 1/5 ofthe estimated daily population for
eachsex. Afewnewadultsemerged inthesurveyperiod, thoughthesurveyperiod
was lateintheemergenceseasonofC. japonica.Therefore, mostimmigrantsmay
not betmeimmigrants fromotherhabitats,butnewadultsemerged inthehabitat.
Theproportion ofthe dailyimmigrationofmature adultstotheestimatedpopula-
tionwereabout1/30. Mostimmigrantsalsoconsistedoftheadultsthatmaturedin
thishabitat,becausethey alreadyhadtheindividualnumbermarkedattheimma-
ture stage.Therewas fewtrue immigrantsfromotherhabitats.Therefore,thepopu-
lationwas isolated.Thesimilarpopulation trendwas alsoobserved in 1990.
Table1 alsoshows theestimateddailysurvivalrate (S) withtherough averageof
the expected residence rate (L), by the equation: L=l/(l-S). The expected resi-
dencetimewas less than5 days forimmatureadults, suggesting thatittakesabout
5 days tosexually matureafteremergence.Matureadultsmaysurvive morethan 1
week. Consequently, theadultlongevity ofC. japonica wasmorethan 2 weeks.
Mostmale C. japonica perched low, near the water surface on the emergent
plants, whichwereopenandregarded asa lookoutpoint. Then, maturemalesoc-
cupied theperching sitetoestablish theterritory.Theperching sites closertothe
water surface were defendedmore aggressively than the others. Most females
218 M.Watanabe,M.Taguehi&N. Ohsawa
perched ontheemergentvegetation ofthesemi-shadedsite.
In themorning, only thewestbank ofthe stream receives direct sun lightbe-
causeof easternhigh mountains.About70%ofmalesandabout85% offemales
perched there irrespective theirages. Aroundnoon,bothbanksof thestream were
exposed to direct sunlight and adults perched oneach sideofthestream. In the
afternoon,theeast bankofthestreamreceived directsunlight duetowestern high
mountains, andabout65%ofmalesandabout85%offemalesperched there.Both
sexes preferred torest indirect sunlight.
The stream was split into 10m sections to facilitaterecording the locationof
individuals.Then,thenumberofsectionswas 46.Theemergentplants forperch-
Fig. 3. Diurnal changesinthe distributionparameter (mean crowding) ofperchingsites for adult
Calopteryxjaponicaonthe eastbank (circle)andwest bank (square)ofthe stream. The dottedline
shows therandom distribution (m*/m=l).
Populationstructure in Calopteryxjaponica 219
ing andovipositing weredistributedevenlyamongeachsection. Figure3presents
thediurnalchange inthem*-mindex (IWAO, 1968)for bothsidesof the stream.
Thedistributionofperching siteforbothsexes wasroughly evenlythroughout the
day, though thepopulation size ineachsampling timevariedduetothemovement
ofsunmentionedabove.Inparticular, distributionsofmature individualsalong the
bank was random inreproductive hours (0900-1500)during whichmalesshowed
territorialbehaviourexpellingtheirrivals fromtheirperching sites.
Thediurnalactivity ofeach sex was countedas thenumberofflightsperhour
andthe numberofmating flightsper hourformales(Fig. 4). Aftersunrise, resting
adultsonthegrassbegantobeactivebuttheflightactivitywas weak.Thepurpose
oftheflightactivityuntil0700wasregarded aspreying.After0800,males showed
cruisingflights, searchedtheirperching site, andthendefended.Thepeak activity
occurredaroundnoon, during whichtandem,copulatory andguarding behaviours
werefrequently observed.Oviposition bycompletely submerged femalesoccurred
in living plant material.Take-overs or displacement fromtheperching site were
alsoobserved. Intheevening, malesexclusively preyed onsmallinsects around
theirperching siteandthenrested.Flightactivity offemaleswasrestrictedtoprey-
ing onsmall insects andfor searching around theresting site, so that their peak
Fig.4.Diurnalchangesinflightactivity ofCalopteryxjaponica.Trianglesandcirclesshowthe number
offlightsperhourformalesand females,respectively.The barsrepresent standard errors. The shaded
areameansthe numberofmatingbehaviours permaleper hour,whichconsists oftandem,copulating
and guardinganovipositingfemale.
220 M.Watanabe, M.Taguchi& N.Ohsawa
activity was morning and
evening, though reproductive
activitieswereobservedaround
noon.
It was difficult to collect
small insects in amongst the
emerging plantsusingtheplate
with glue due to complex
leavesandtwigs. However,we
collected 18.4 individuals/m3
onthepaddyfields(Fig. 5).The
paddyfieldsalong thebankhad
the largest community among
the sample lines (paddy fields
- bank: x-9.46, 0.1>P>0.05;
paddy fields - emergent
plants:x2=18.72,P<0.005).The
body length of most insects
collected was 1-2 mm, which
occupied 54% oftotal insects
collectedon thebank. Diptera,
Hymenoptera and Hemiptera
weremajorinsects inthishabi-
tat. Although some ofthespe-
cies caught were classified as
pest insects ofrice, they were Fig.5,Communitiesofsmallinsects collectedontheplate
regarded asprey insects forC. in relation totheir body size. The meannumberofinsects
adhered tosuch plateareshown inparenthesis.
japonica.
DISCUSSION
Inmanystudieson odonatepopulations, malecapturesoutnumberfemalecap-
tures(e.g. WATANABE&HIGASHI, 1989),becausemalesarepresentatthebreed-
ing site.Inthespecies adopting territorialtactics, particularly, mature malesreturn
to water to establishterritoriesbut mature femalestendto visitonly to oviposit.
Then, the probability ofrecapturefor territorialmales is higher than thatforfe-
males. Therefore,in odonate species with territorialbehaviourvariance on esti-
matedmale numbers shouldbe lowerthan thaton femalenumbers. However, in
thepresentstudy, aconsiderablenumberoffemaleswerere-captured. Fieldobser-
vations suggestthatfemalestendto stay by waterthroughout the day. HIGASHI
(1976)pointed outthat themaidenflightsofC. corneliadidnotdisperse immature
adultsfaraway fromthewater fromwhichthey emerged, though C. maculatais
Populationstructure in Calopteryxjaponica 221
oftenconsideredaforest damselfly(e.g. TAYLOR & MERRIAM, 1996).
Theproportionofestimatednumberofimmigrantsto theestimatedpopulation
showedthat C.japonica has high sitefidelity. Theoperational sex ratio was con-
sideredas unity.Inmanyodonate species sincefemalesvisitby waterfor repro-
duction,theprolonged residenceby femalesinthehabitatmay increasetheprob-
ability ofmating throughout theirlife span. Although manyobservations thatfe-
males can refuse to copulate havebeen reported, malesand femalesof odonate
species are polyandrous. Therefore, femalesvisiting in the reproductive area are
regarded as being receptive for mating. Consequently, the operational sex ratio
shouldincrease.HIGASHI&UEDA(1982)reported thatthesexratioofC. cornelia
was approximately 1:1. FORSYTH & MONTGOMERIE (1987) also found an
equal population phenology between the sexes of C. maculata in Canada. In
contrast, thenumberoffemalesof M.pruinosa ona mountainstreamwas lower
thanthatofmales,becausethefemalesleftthestreamafterovipositing (HIGASHI,
1976). Anequal sex ratiointhereproductive area wasalsoreportedforAnisoptera
suchasSympetrumpedemontanum, whichinhabitspaddyfieldswithoutanyflights
away fromthefields(TAGUCHI &WATANABE, 1985).
HIGASHI& UEDA(1982) statedthat teneralsofC. Corneliahadatendency to
move upstream as is thecase with many aquatic insects. CORDERO (1989) also
observed asignificantly highertendency tofly upstreaminimmaturethaninma-
ture malesofC. haemorrhoidalis.However,inthepresent study, LakeNakatsuna
is located upstreamofthe habitat.Therefore, littlemovement offemalesout of
theirnatalarea isreflectedintheobserved stabilitiesinpopulation structure ofC.
japonica. We did not detect any difference in daily survival between sexes.
MIYAKAWA (1982) reported that themean longevityofC. japonica in Saitama
Pref.was thesamebetweensexes.
Inboththemorning andtheevening whentheambienttemperaturedecreasedto
20°C, adultsofC. japonica preferred thesunlitarea forperching, suggesting that
they showedthermoregulatory behaviour,basking. Sincetheambienttemperature
aroundnooninthisstudyareaexceededover30°C,thethresholdofbody tempera-
ture tobecome active mightberelatively high. WATANABE(1991) reported that
sexually mature malesofMnaispruinosa changed thewatercontent oftheirbod-
ies, whereas mature females did not but they stayed inside the forests to avoid
directsunlight. Inthepresentstudy,adultsofC.japonicadonot haveshady forests
available in whichto escape the high radiation. They still perched on the sunlit
area.Thus,itis possible thatacertainthermoregulation system haveoperatedun-
derdirectsunlight,openhabitatconditions.Therearesomereports thatfemalesof
Calopteryx spp. submerge to oviposit (e.g. CORBET, 1962). We frequently ob-
served them bathing inthe stream, which means that the body temperaturemust
decrease. Suchbehaviourcouldmakeitpossible forC.japonicatostay inanopen
area.
WATANABE& MATSUNAMI(1990) showedthatthedistributionofthemale
222 M.Watanabe, M.Taguchi& N.Ohsawa
emerald damselfly, Lestes sponsa, was random in the forestfloorbecause they
established territories at the sunflecks randomly distributedin the forest. Since
maleC. japonica alsoestablishedterritoriesexpellingtheirrivals(e.g.AIDA,1974),
the distributionoftheperching sites in thebanks ofthe stream mustberegular.
Abundanceofbank vegetation along thestreamalso supported theregular distri-
butionoftheperching sites. Although we didnot surveythedistributionofperch-
ing sites in 1990,cutting offtheaquatic vegetation andemergentplants couldbe
caused the distributionpattern ofthe damselflies to be more clumped. COR-
DOBA-AGUILAR (1994) pointed out that more aggressive males of Hetaerina
cruentata were moreabundantinsubstrates closertothe water.Thedistributionof
thefemaleH.macropusreflectswherethey fedratherthanwherethey bred(EBER-
HARD, 1986).
PIERCEetal.(1985)havemadequantitative connectionsbetweenaquatic veg-
etationand predation risk ofodonate larvafrom fish. CONVEY (1988) reported
thatthemajoradvantage ofperch possession toCoenagrion puella larvae is inthe
reductionofpredation. CORBET etal.(1960) statedthat©donatelarvaeareeaten
by anumberoffreshwaterpredators including thelargerwaterbeetles, somebugs,
newtsandfish.Inthepresentstudy, suchpredatorshavebeenidentified:forexam-
plewe foundalotofdamselfliesinsidethestomachofblack bass (unpublished).
On theotherhand,eggsof C. splendens placed in fast-flowing water developed
significantly fasterand showedsignificantly lowermortalitythan theeggsplaced
in slow-flowing water, due to growth ofencrusting algae (SIVA-JOTHY et al.,
1995).Inthestudy area, however,theharvestofaquatic vegetation might causea
destructionofthe C. japonica population. It is possible, however, that adultre-
sources might notbealimiting factor, whichwouldexplain therelatively consist-
ent large numbersof perching andresting sitesby water.
CORBET (1962)statedthatadult odonatesprey onsmallinsects irrespectiveof
theirtaxon. HIGASHI(1973) reported thatC. corneliaquickly responded toevery
smallinsectpassing through neartheperching site.Inthepresentstudy, mostsmall
insects thatadhered tothecollecting plate mightbesuitableprey for damselflies.
Adultsfeed onsmall insects, mainly Diptera andHymenoptera, and usually stay
on the emergent plants or on grasses along the stream bank. Prey insects were
abundantin thishabitat.Dispersion and behaviourofdamselfliesofCalopteryx
species weremodifiedby thesizeandavailability ofoviposition resources (MEEK
& HERMAN, 1991). Therefore, thequalityandthe quantityofaquatic vegetation
forlarvalhabitatsmaybemoreimportanttomaintainthepopulation ofC.japonica
thanthoseofterrestrialvegetation foradults.
ACKNOWLEDGEMENTS
This work has beensupportedprimarily by agrant toM. WATANABEfromThe Foundationfor
Earth Environment. We are extremely grateful toMessrs.Y. FUKUI,T. HIGASHI, AANDO, Y.