Suction duration and numbers of Araneae and Auchenorrhyncha collected in grassland when using an open-ended cylinder to define the sample area

2018·0.19 MB·

by  Andrew Cherill

©EntomologicaFennica.7September2018
Suction duration and numbers of Araneae
and Auchenorrhyncha collected in grassland
when using an open-ended cylinder to define the sample area
AndrewCherrill,RichardBurkmar,HenryQuenu&EmanZentane
Cherrill,A.,Burkmar,R.,Quenu,H.&Zentane,E.2018:Suctiondurationand
numbersofAraneaeandAuchenorrhynchacollectedingrasslandwhenusingan
open-endedcylindertodefinethesamplearea.—Entomol.Fennica29:105–
111.
Suctionsamplersaretypicallyusedwithtouch-downsamples,wherethenozzle
isheldstationaryontheground,orwithinanopen-endedcylinder,wherethenoz-
zleismovedthroughtheenclosedvegetation.Withtouch-downsampling,pub-
lishedestimatessuggestthatvirtuallyallindividualsandspeciesofgrasslandin-
vertebrate are captured within 3 s. Effects of suction duration on catch when
samplingwithinacylinderhaverarelybeeninvestigated.Here,itwasfoundthat
extrapolationfromstudiesoftouch-downsampling,basedontherelativeareasof
theG-vacnozzleandthecylinder,canbeusedtoguideminimumsamplingdura-
tionsforAuchenorrhyncha(leafhoppersandplanthoppers)whenusingaG-vac.
Additionaltime,however,maybenecessarywhensamplingAraneae(spiders),
probablybecausemovementoftheG-vacnozzlewithinthecylinderresultsin
disturbanceofthevegetationsuchthatmorespecimensaremadeavailabletobe
capturedthanwithtouch-downsampling.
A.Cherrill,CentreforIntegratedPestManagement,DepartmentofCropand
EnvironmentSciences,HarperAdamsUniversity,Shropshire,TF108NB,United
Kingdom;E-mail:[email protected]
R. Burkmar, Field Studies Council, Preston Montford, Shropshire, SY4 1HW,
UnitedKingdom
H.Quenu,DepartmentofAgriculture,PurpanSchoolofEngineering,Toulouse
Cedex3,France
E.Zentane,PlantProtectionDepartment,CollegeofAgriculture,Universityof
Tripoli,Libya.
Received26July2017,accepted23November2017
1.Introduction morpha and Fulgoromorpha) (Southwood &
Henderson 2000, Hill et al. 2005). G-vacs are
Modifiedgardenblow-vacs,commonlyreferred popularbecausetheyarerelativelyinexpensive,
toasG-vacs,areusedwidelytosamplegrassland andbecausetheycapturegrasslandinvertebrates
invertebrates and particularly Araneae (spiders) whichliveclosetotheground.Suchspeciesare
and Auchenorrhyncha (leafhoppers and plant- often under-represented in sweep-net samples
hoppers of the Hemipteran suborders Cicado- (Samu & Sarospataki 1995, Stewart & Wright
106 Cherrilletal. (cid:127) ENTOMOL.FENNICAVol.29
1995,Stewart2002).G-vacscanalsobeusedto 2.Methods
obtaincrudedensityestimateswheresamplingis
conductedwithinanenclosedarea–typicallyde- 2.1.Studysites
fined by an open-ended cylinder many times
wider thantheG-vacnozzle(Samuetal. 1997, Threegrassland sites with flatterrain and inter-
Hilletal.2005).Thetimerequiredtocaptureall, nallyhomogenousvegetationwereselectedtoin-
or most invertebrates, when using a suction de- clude a range of agricultural management re-
vicewithinanopen-endedsamplingcylinderhas, gimes.Site1wasanungrazedgrassland,mownat
however,beeninvestigatedrarely(Henderson& intervalsfrommid-Julyonwards(altitude60m,
Whitaker1977,Törmälä1982). latitude52°46’N,longitude2°25’W).Site2was
Incontrast,Brooketal.(2008)estimatedthat anunmownpasturegrazedbydairycattleandsit-
90% of Coleoptera, Araneae and Auchenor- uated within 500 mof Site 1. Site 3 was unim-
rhynchaspeciesarecapturedinthefirst3sofsuc- proved sheep-grazed grassy heath (altitude 420
tionwhenthenozzleofaVortisdeviceisheldsta- m,latitude52°31’N,longitude2°53’W).Vegeta-
tionaryonthegroundatasingleunenclosedpoint tion at Sites 1 and 2 was dominated by sown
(commonly known as a “touch-down” sample). grasses and was representative of mesotrophic
Moreover,Belletal.(2000)foundnoincreasein grasslandtypeMG7aLoliumperenne–Trifolium
numbersofAraneaecapturedbeyondthefirst1s repensleywithintheNationalVegetationClassi-
suctionofatouch-downsamplewhenusingaG- fication(NVC)(Rodwell1992).Site3wasdomi-
vac. natedbysemi-naturalvegetationcharacteristicof
When using a cylinder, the time needed to poor acidic soils and was most similar to NVC
capture a representative sample of invertebrates type U1 Festuca ovina – Agrostis capillaris –
willbeatleasttheminimumtimerequiredfora Rumexacetosella.
singletouch-downsamplemultipliedbytheratio Ateachsite,agrid12mby20mwasmarked
of the area of the cylinder to that of the nozzle. outandsub-dividedintofifteen4mby4mgrid-
Longer may be required, however, because squares.Vegetationheightwasquantifiedusinga
movementofthenozzlewithinthecylindermay singledrop-platemeasurement(Cherrill&Rush-
enhancecaptureofspecimensthatwouldother- ton1993)atthecentreofeachgrid-square(x±
wisebehiddenbeneaththevegetation.Thepres- SD:Site1,12.3±2.2cm;Site2,15.1±3.4cm;
entstudyaddressesthisproblembyinvestigating Site3,3.7±1.0cm)(n=15ateachsite).
the numbers of individuals and species of
AraneaeandAuchenorrhynchacapturedwithin-
2.2.Suctionsampling
creasing sample duration when using a G-vac
within an open-ended cylinder at each of three TheG-vacsuctionsamplerwasamodifiedMc-
grasslands.Inthisstudy,theG-vacnozzlehadan CullochGBV345gardenblower/vacuum(Stew-
areaof0.01m2,whiletheareaofthecylinderwas art&Wright1995,Stewart2002).Theendofthe
0.18m2. collectingpipewassawnoffperpendiculartoits
Ifitisassumedthataperiodof3sisappropri- lengthtogiveanozzlewithacrosssectionalarea
ate for a touch-down sample, then most speci- of0.01m2.Anylon1mmmeshbagwasheldin
mens would be expected to have been captured placeattheendofthepipebyarubberband.Esti-
fromthecylinderinjustunder60s(i.e.18multi- mated air velocity within the nozzle was20.3
pliedby3stogiveanequivalentsamplinginten- ms–1(Zentaneetal.2016).
sityintermsofsecondsperm2).Totestthishypo- Suctionsamplesateachsiteweretakenona
thesisthreesuccessivesamplesof30sweretaken singledayonly(15th,17thand21stJuly2014for
fromwithinthecylindersuchthattheaccumula- Sites1,2and3respectively).Minimumairtem-
tion of specimens and species could be investi- peraturesduringsampling,measuredinshadeat1
gatedafter30s,60sand90ssampling. mabovegroundusingadrybulbmercuryther-
mometer,atthethreesiteswere30°C,26°Cand
24°Crespectively.Vegetationandleaflitterwas
drytothetouch.
ENTOMOL.FENNICAVol.29 (cid:127) G-vacsuctionduration 107
Table1.TotalnumbersofindividualsandspeciesofAraneaecollectedafter90ssuctionsampling.
Site1 Site2 Site3
Theridiidae
EnoplognathalatimanaHippa&Oksala,1982 4 0 0
Linyphiidae
Dicymbiumnigrum(Blackwall,1834) 1 1 0
Entelecaraflavipes(Blackwall,1834) 0 0 1
Dismodicusbifrons(Blackwall,1841) 0 0 0
Oedothoraxfuscus(Blackwall,1834) 7 6 4
Oedothoraxretusus(Westring,1851) 12 2 0
Gongylidiellumvivum(O.P.-Cambridge,1875) 2 0 0
Micrargusherbigradus(Blackwall,1854) 0 0 1
Micrargussubaequalis(Westring,1851) 1 0 0
SavigniafrontataBlackwall,1833 5 6 1
Millerianainerrans(O.P.-Cambridge,1885) 1 0 0
Erigonedentipalpis(Wider,1834) 2 17 3
ErigoneatraBlackwall,1833 9 44 3
Agynetadecora(O.P.-Cambridge,1871) 2 0 0
Meionetarurestris(C.L.Koch,1836) 0 2 0
Bathyphantesgracilis(Blackwall,1841) 18 53 0
Tenuiphantestenuis(Blackwall,1852) 13 42 0
Microlinyphiapusilla(Sundevall,1830) 3 0 0
Thomisidae
Xysticuscristatus(Clerck,1757) 1 0 0
Immatures
Unidentifiedfamilies 162 182 7
Linyphiidae 100 193 7
Tetragnathidae 6 0 0
Lycosidae 29 0 0
Thomisidae 1 0 1
Totalnumberofindividuals 379 548 28
Numberofspecies 15 9 6
G-vac samples were taken within an open- sand,althoughnotquantified,itispossiblethata
endedplasticcylinderwithacross-sectionalarea smallnumberofactivefliersmay haveescaped
of0.18m2,heightof60cminheightandweight fromthecylinderwhilethiswasbeingdone.
of5kg(sufficienttoformasealwiththeground
surface).Thecylinderwasplacedatthecentreof
2.3.Treatmentofinvertebratesamples
a grid-square. The G-vac was then used to take
threesuccessive30ssub-samplesbeforethecyl- Materialfromthethree30ssub-samplesineach
inderwasmovedtothenextgrid-square.Themo- grid-square were counted separately. Adult
torwasrunonfull-throttlewhilstthenozzlewas AraneaeandAuchenorrhynchawereidentifiedto
withinthecylinder,butwasidledwhilstthemesh species,whileimmaturespecimenswereidenti-
bag was replaced between 30 s sub-samples. fied to family. Nomenclature follows Merrett et
Within each 30 s sub-sample, the G-vac nozzle al.(2014)andWilsonetal.(2015).
wassweptoverthesurfaceofthevegetationfor5
s before the nozzle was repeatedly lowered and
2.4.Statisticalanalysis
raisedfromthegroundsurfaceinaseriesofover-
lapping “touch-downs” covering the enclosed Repeated measures Generalised Linear Model
areafortheremaining25s.Changingthemesh with negative binomial error distribution was
bagbetweensub-samplestookapproximately15 usedwithinIBMSPSSver24totestforaneffect
108 Cherrilletal. (cid:127) ENTOMOL.FENNICAVol.29
Table2.TotalnumbersofindividualsandspeciesofAuchenorrhynchacollectedafter90ssuctionsampling.
Site1 Site2 Site3
Cicadellidae
Aphrodesbicinctus(Schrank,1776) 3 0 0
Anoscopusalbifrons(Linnaeus,1758) 2 0 0
Anoscopusserratulae(Fabricius,1775) 3 0 0
Anoscopusunidentifiedfemales 2 0 0
Deltocephaluspulicaris(Fallen,1806) 0 0 19
Arocephaluspunctum(Flor,1861) 0 0 2
Arthaldeuspascuellus(Fallen,1826) 60 2 0
Psammotettixnodosus(Ribaut,1925) 0 0 2
Euscelisincisus(Kirshbaum,1858) 37 0 0
Cicadulapersimilis(Edwards,1920) 1 0 0
Zyginidiascutellaris(Herrich-Schaeffer,1838) 22 0 0
Delphacidae
Javesellaobscurella(Boheman,1847) 0 4 0
Javesellapellucida(Boheman,1847) 202 30 0
Javesellaunidentifiedfemales 0 23 0
Immatures
Cicadellidae 116 3 3
Delphacidae 292 23 0
Totalnumberofindividuals 740 85 26
Numberofspecies 8 3 3
of the independent variable “sample duration” 3.1.Numbersofindividuals
(30,60and90s)onthedependentvariablewhich
was the accumulated number of individuals (or
Sample duration had a significant effect on the
species).TheanalysiswasconductedforAraneae
cumulativenumberofAraneaecapturedatSite1
and Auchenorrhyncha within each site sepa-
(¤2=14.56, d.f. =2, P=0.01) andSite2(¤2=
rately.Ineachanalysis,thebetween-subjectfac-
70.54,d.f.=2,P=<0.001),butnotSite3(¤2=
torwasgrid-squareandthewithin-subjectfactor
3.21, d.f. = 2, P= 0.20). At both Sites 1 and 2,
was sample duration. Detection of a significant
numbersincreasedwitheachadditional30ssub-
effect of sample duration, based on Walds Chi-
sample.Incontrast,atSite3veryfewadditional
squareTest(withcriticalPof0.05),wasfollowed
Araneaewerecapturedinthesecondandthird30
by Bonferroni-corrected pairwise comparisons
ssub-samples(Fig.1a).
betweenaccumulatednumberscapturedafter30,
60and90s. Sample duration had a significant effect on
AtSites1and2,allmeans(andStandardEr- the cumulative number of Auchenorrhyncha at
rors,SE)arebasedondatafrom15grid-squares, Site1(¤2=21.15,d.f.=2,P=<0.001).Thenum-
whileatSite3somesamplesweremislaidleaving berofindividualAuchenorrhynchaincreasedin
datafor10grid-squaresonly. thesecond30ssub-sampleonly,withfewadded
inthethird30s(Fig.1c).Sampledurationhadno
significant effect on cumulative numbers cap-
3.Results turedatSite2(¤2=2.07,d.f.=1,P=0.15)orSite
3(¤2=2.05,d.f.=1,P=0.11).Atthesesites,very
In total 955 specimens of Araneae and 851 few additional specimens were captured in the
Auchenorrhynchawerecapturedacrossthethree second30ssub-sample,andnonewerecaptured
sites. Nineteen species of Araneae and twelve in the third 30 s (with the consequence that the
speciesofAuchenorrhynchawerefound(Tables thirdsub-samplewasomittedfromtheseanaly-
1and2). ses)(Fig.1c).
ENTOMOL.FENNICAVol.29 (cid:127) G-vacsuctionduration 109
Fig.1.Mean(±SE)foraccumulativenumbersofindividualsandspeciesofAraneaeandAuchenorrhynchacap-
turedafter30s,60sand90s.Columnswithinasite,andbearingthesameletter,weresubjecttostatistical
analysisandwerefoundtobenotsignificantlydifferent(atP=0.05).Columnswithoutaletterhaveidentical
meanandSEtotheadjacentcolumn(s).Meansarebasedonn=15atSites1and2,andn=10atSite3.
3.2.Numbersofspecies 4.Discussion
Sample duration had a significant effect on the Several studies have investigated how numbers
cumulative number of species of Araneae cap- of species and specimens increase with suction
turedatSite1(¤2=11.82,d.f.=2,P=<0.01)and timeusingtouch-downsamples(Macleod etal.
Site2(¤2=27.48,d.f.=2,P=<0.001).Themean 1994,Belletal.2000,Brooketal.2008).Mac-
numberofAraneaespeciesincreasedinthethird leod et al. (1994) used a G-vac to repeatedly
30ssub-sampleatSite1,andinthesecond30s samplethesameunenclosedpointwithconsecu-
sub-sampleatSite2(Fig.1b).AtSite3,noaddi- tivetouch-downsuptoatotaldurationof160s.
tionalspecieswerefoundinthesecondorthird30 CumulativenumbersofColeopteraandAraneae
ssub-samplesandhencethesedatawerenotsub- increasedfromthefirst5ssub-sampleto30sbut
jecttostatisticalanalysis(Fig.1b). not thereafter. The time after which no further
At Site 1, the number of Auchenorrhyncha benefitaccruedmayhavebeenoverestimatedbe-
speciesshowedasmallnon-significantincrease cause specimens were probably drawn in from
inthesecond30ssub-sample(¤2=3.16,d.f.=1, surroundingvegetationastheG-vacwasrepeat-
P=0.08),butnoadditionalspecieswerecaptured edly raised and lowered (Samu et al. 1997,
inthefinal30s(Fig.1d).Moreover,noadditional Cherrill 2015). This problem was overcome by
species were captured after the first 30 s sub- Bell et al. (2000) and Brook et al. (2008) who
sampleateitherSite2or3andhencethesedata sampled independent points for different dura-
werenotsubjecttostatisticalanalysis(Fig.1d). tions. These authors suggest that most Araneae
and Auchenorrhyncha are captured within the
firstfewsecondsofatouch-downsample.Bellet
al.(2000)foundnoincreaseinnumbersorspe-
ciesofspidersbetween1sand10ssuctionwitha
110 Cherrilletal. (cid:127) ENTOMOL.FENNICAVol.29
G-vac.Brooketal.(2008)estimatedthat90%of suctiondurationcanbeextrapolatedfromstudies
individuals and species of Coleoptera, Araneae using the alternative touch-down approach. A
andAuchenorrhynchaarecapturedwithin3sus- samplingintensityequivalentto3sperunitarea
ing a similar Vortis suction device. In a direct (whereunitareaisdefinedbythecross-sectional
comparison of the two devices, Zentane et al. areaofthenozzle)issufficienttocaptureallspe-
(2016)foundtheVortistobelessefficientthana cies and nearly all individuals of Auchenor-
G-vac,hencetheseestimatescanbeviewedasbe- rhyncha,butlongermaybeneededforAraneae,
ingbroadlycomparable. and particularly in taller grass. This is probably
In the present study, the G-vac nozzle was becausemovementofthenozzlewithinacylin-
usedtosweeptheupperlayersofthevegetation derresultsindisturbanceofthevegetationsuch
before being applied in a series of overlapping thatmoreAraneaearemadeavailabletobecap-
touch-downswithinacylinderofarea0.18m2.A turedthanwithtouch-downsampling.
30 s sub-sample therefore approximates to a
sampling intensity of approximately 1.7 s per Acknowledgements. We thank Chris Hickman for assis-
0.01m2(i.e.30sdividedby18).BasedonBellet tancewithfieldwork,NigelCane-Honeysettforassistance
with spider identification, Simon Jeffery for assistance
al. (2000) and Brook et al. (2008), touch-down
with producing the figure, and two anonymous referees
samplingwithinthecylinderfor60s(i.e.equiva-
whosecommentsgreatlyimprovedthemanuscript.DrE.
lent to approximately 3.4 s per 0.01 m2 touch- ZentanewassupportedbyasabbaticalfromtheUniversity
down) would be predicted to have captured ofTripoli.HenryQuenuwassupportedbytheinternship
nearly all specimens and species therein. Al- programmeatPurpanSchoolofEngineering,France.
though the approach to sampling within these
studies was not identical, the prediction agrees
References
with the results reported here for Auchenor-
rhyncha.Numbersofindividualsandspeciesdid
not increase after 60 s at any of the three sites Bell,J.,Wheater,C.P.,Henderson,R.&Cullen,R.2000:
Testingtheefficiencyofsuctionsamplers(G-vacs)on
(Figs.1cand1d).
spiders:theeffectofincreasingnozzlesizeandsuction
In contrast, numbers ofindividuals and spe-
time.—In:Toft,S.&Scharff,N.(eds.),European
ciesofAraneaeaccumulatedbeyond60ssamp- Arachnology: 285–290. Aarhus University Press,
ling, except at Site 3 where Araneae were rela- Denmark.358pp.
tivelyscarce(Figs.1aand1b).Overall,theresults Brook,A.J.,Woodcock,B.A.,Sinka,M.&Vanbergen,A.
J.2008:Experimentalverificationofsuctionsampler
suggestthatsamplingwithinthecylinderfor60s
captureefficiencyingrasslandsofdifferingvegetation
(crudely equivalent to a series of twenty 3 s
heightandstructure.—JournalofAppliedEcology
touch-downs) was sufficient for Auchenor- 45:1357–1363.
rhyncha. Longer may be needed, however, to Cherrill,A.J.2015:Suctionsamplingofgrasslandinverte-
reachthepointwherefurthersamplingcaptures bratesusingtheG-vac:Quantifyingandavoidingpe-
ripheralsuctioneffects.—EuropeanJournalofEnto-
no additional Araneae. The efficiency of samp-
mology112:520–524.
lingmayberelatedtothestructuralcomplexityof
Cherrill,A.J.&Rushton,S.P.1993:TheAuchenorrhyn-
thevegetation,becauseatthesitewiththeshort-
chaofanunimprovedmoorlandinnorthernEngland.
est grass (Site 3) no additional Araneae were —EcologicalEntomology18:95–103.
caughtafterthefirst30s(Figs.1aand1b).Suc- Henderson,I.&Whitaker,T.1977:Theefficiencyofanin-
tionsamplingisgenerallymoreefficientinshort sectsampleringrassland.—EcologicalEntomology
2:57–60.
vegetation (Henderson & Whitaker 1997, Hos-
Hill,D.,Fasham,M.,Tucker,G.,Shrewry,M.&Shaw,P.
sainetal.1999,Brooketal.2008,butseeSan-
2005: Handbook of biodiversity methods. — Cam-
ders&Entling2011)anditisalsoknownthatless bridgeUniversityPress,Cambridge.588pp.
structurally complex grasslands tend to support Horvath,R.,Magura,T.,Szinetar,C.&Tothmeresz,B.
fewer invertebrates (Langellotto et al. 2004, 2009:Spidersarenotlessdiverseinsmallandisolated
grasslands,butlessdiverseinovergrazedgrasslands:
Horvathetal.2009).
Afieldstudy(EastHungary,Nyirseg).—Agriculture,
Thepresentstudyprovidespartialsupportfor
EcosystemsandEnvironment130:16–22.
thehypothesisthatwhenusingaG-vacwithinan
Hossain,Z.,Gurr,G.M.&Wratten,S.D.1999:Capture
open-ended cylinder, the minimum necessary efficiencyofinsectnaturalenemiesfromtallandshort
ENTOMOL.FENNICAVol.29 (cid:127) G-vacsuctionduration 111
vegetation using vacuum sampling. — Annals of Sanders,D.&Entling,M.H.2011:Largevariationofsuc-
AppliedBiology135:463–467. tion sampling efficiency depending on arthropod
Langellotto,G.A.&Denno,R.F.2004:Responsesofin- groups,speciestraitsandhabitatproperties.—Ento-
vertebratenaturalenemiestocomplex-structuredha- mologiaExperimentalisetApplicata138:234–243.
bitats:ameta-analyticalsynthesis.—Oecologia139: Southwood,T.R.E.&Henderson,P.A.2000:Ecological
1–10. methods.3rdEd.—BlackwellScience,Oxford.575
Macleod,A.,Wratten,S.D.&Harwood,R.W.J.1994: pp.
Theefficiencyofanewlightweightsuctionsampler StewartA.J.A.&WrightA.F.1995:Anewinexpensive
forsamplingaphidsandtheirpredatorsinarableland. suction apparatus for sampling arthropods in
—AnnalsofAppliedBiology124:11–17. grasslands.—EcologicalEntomology20:98–102.
Merrett,P.,Russell-Smith,A.&Harvey,P.2014:Arevised Stewart,A.J.A.2002:TechniquesforsamplingAuche-
checklistofBritishspiders.—Arachnology16:134– norrhynchaingrasslands.—Denisia4:491–512.
144. Törmaälä,T.1982:Evaluationoffivemethodsofsamp-
Rodwell,J.S.1992:BritishPlantCommunities.Volume3. lingfieldlayerarthropods,particularlytheleafhopper
Grasslandsandmontanecommunities.—Cambridge community, in grassland. — Annales Entomologici
UniversityPress,Cambridge.552pp. Fennici48:1–16.
Samu,F.,Nemeth,J.&Kiss,B.1997:Assessmentofthe Wilson,M.,Stewart,A.,Biedermann,R.,Nickel,H.&Ni-
efficiencyofahand-heldsuctiondeviceforsampling edringhaus,R.2015:Theplanthoppersandleafhop-
spiders: improved density estimation or oversam- persofBritainandIreland.—Cicadina–Supplement
pling?—AnnalsofAppliedBiology130:371–378. 2:1–138.
Samu, F. & Sarospataki, M. 1995: Design and use of a Zentane,E.,Quenu,H.,Graham,R.I.&Cherrill,A.2016:
hand-held suction sampler, and its comparison with Suctionsamplersforgrasslandinvertebrates:compa-
sweepnetandpitfalltrapsampling.—FoliaEntomo- risonofnumberscaughtusingVortisandG-vacdevi-
logiaHungarica56:195–203. ces.—InsectConservationandDiversity9:470–474.