Friend or Foe: Exotic Flora and Ecosystem Function

2010·3.5 MB·English

by  Melanie, Gibson Birtchnell

ResearchReports
GosperCandVivian-SmithG(2006)Selectingreplacements Rodriguez LF (2006) Can invasive species facilitate native
forinvasiveplantstosupportfrugivoresinhighlymodified species?Evidenceofhow,whenandwhytheseimpactsoc-
sites:AcasestudyfocusingonLantanacamara.Ecological cur.BiologicalInvasions8,927-939.
managementandRestoration7,197-203. SchmidtB,RenowdenCandQuinD(2009)SouthernBrown
GriffinG,StaffordS,MortonG,AllanGandMastersK(1989) BandicootStrategicManagementPlanfortheformerKoo
StatusandimplicationsoftheinvasionofTamarisk(Tamariz WeeRupSwampArea.(EcologyAustralia:Fairfield)
aphylla) ontheFinke River,NorthernTerritory,Australia. Scott Mills G, DunningJB and Bates JM (1989) Effects of
JournalofEnvironmentalManagement29,297-315. urbanization on breeding bird community structure in
LawrieS(2001)Findinganimalsamongsttheweeds:anAu- southwesterndeserthabitats.TheCondor91,416-428.
dcietedsiynsgtsemo.ftIhne2t0h0i1r,djGoeiontgrCaopnhfyer-enAceSpoaftitahleONdeywsseZye.alParno-d ShEafraontdhSPt,uCalretveJr(l2y0J0,5D)udCloenytrTo,lTaoyfloTramJ,aVrainxRiinptehreCW,eWseteekrsn
GeopraphicalSocietyandtheInstituteofAustralian Geog- UnitedStates: Implicationsforwatersalvage,wildlifeuse,
raphers. pp. 191-195. (eds Holland P, Stephenson F and and riparian restoration. EnvironmentalManagement 35,
WearingA)(BrebnerPrint:Hamilton) 231-246.
Loyn R(1986) The20minutesearch-asimplemethodfor SoggeMK,SferraSJandPaxtonEH(2008)Tamarixashabitat
countingforestbirds.Corella1,59-60. forbirds:Implicationsforriparianrestorationinthesouth-
LoynRandFrenchK(1991)Birdsandenvironmentalweeds westernUnWitedStates.RestorationEcology16,146-154.
in south eastern Australia. Plant Protection Quarterly 6, Sutherland (1996) Ecological census techniques, a hand-
137-148 book.(CambridgeUniversityPress:PortMelbourne)
NeDiloanfrWu,giCvaotrtoeursalblirCd,sKaassniostwsrkaiinfJoraensdtsMuccKceesnsniaonSin(2w0e0e6d) WhNiotne-uJGn,ifAonrtmosbiMrJd,asFsietmzbsliamgoenssiJnAurabnadnPeanlvmierronGmeCnt(s2:00t5h)e
dominatedoldfieldregrowthofsubtropicalAustralia?Bio- influenceofstreetscapevegetation. Landscapeand Urban
logicalConservation129,393-407. Planning71,123-135.
NiasR(1986)Nestsitecharacteristicsandreproductivesuc- YeatesLandBarmutaL(1999)Theeffectsofwillowandeu-
cessintheSuperbFairy-wren.Emu86, 139-144. calypt leaveson feedingpreferences andgrowthofsome
ParkesD,NewellGandChealD(2003)Assessingthequality AustralianaquaticInvertebrates.AustralianJournalofAp-
ofnative vegetation: The habitat hectares approach. Eco- pliedEcology24,593-598.
logicalManagementandRestoration4,S29-S38.
PeterJ(2000) BirdsandBoxthorn. The VictorianNaturalist
11,63-66. Received24May2010;accepted17June2010
Randall J (1996) Weedcontrolforthepreservation ofbio-
logicaldiversity. WeedTechnology10,370-383.
Friend or foe: exotic flora and ecosystem function
Melanie Birtchnell1 and Maria Gibson
PlantEcologyResearchUnit,SchoolofLifeandEnvironmentalSciences
DeakinUniversity,221BurwoodHighway,Burwood,Victoria,3125
'[email protected]
Abstract
Exotic flora, particularly weeds, are renowned for out-competing and displacing native flora, consequently
affecting native faunaandpollinatorrelationships. Nonetheless, itstands to reason thatweeds mustprovide
somecompensatoryecologicalvalue.Thisstudyassessedwhetherweedsarefriendorfoetoecosystemfunc-
tionbyconsideringthequalityandquantityofpollen offeredbywidespreadweedsinAustralianecosystems.
UsingtheHoneybeeApismelliferaasacasestudy,andinformationderivedfromhighlyexperiencedcommer-
cialapiarists,wedetermined that32exotic plants are importantpollensources. Mostspecies offered high to
veryhighqualitypollen. Pollen qualityvariedtemporally, spatiallyandinfraspecifically. Fifteen specieswere
consideredmorebeneficialtoA.melliferathanothers;onlysevenspecieswereconsideredlessbeneficial.Thus,
exoticfloracontributepollen resourcesthatarevaluabletomaintainecosystemfunction,particularlyattimes
whenfloweringnativespeciesarefew. (TheVictorianNaturalist127(4)2010,124-136)
Keywords:exoticflora,weeds,ecosystemfunction,pollen, pollinators
Introduction
Exotic flora - globally - have a bad reputation. ferewithnativespeciesfurtherbyaffectingpolli-
Their roll-call of maladies often includes out- natorrelationships,whichcanimpactgreatlyon
competing and displacing native flora (e.g. Vi- theecologyandevolutionofnativefloralspecies
tousek et al. 1987; Meiners et al. 2001; Levine (Ashman etal 2004). We refer to those species
et al 2003) consequently affecting native fauna mostproficientatsuchmaladiesasweeds.Ubiq-
(Vitousek etal 1987). Exotic species mayinter- uitousasweedsare,itstands to reason thatthey
124 TheVictorian Naturalist
ResearchReports
must provide some level ofecosystem service. this paper also consideredwhetherApis mellif-
Certainly, there are reports of the habitat val- era displayed preferences for particular pollen
ues offered by weeds (e.g. Donald et al. 2001), sources. Astheseaspects are likelytoaffectna-
although in many cases (including throughout tive invertebrate (and,possibly,vertebrate) pol-
Australia),exoticfloraprovidehabitatforexotic linators,theinformationobtainedisofgreatim-
fauna(Vitouseketal 1987)! portanceindeterminingthe ecologicalvalue of
Managing healthy ecosystems demands an exoticpollenflora(thatis,floratargetedforpol-
understanding of which species provide high len). This was part ofa larger study (Birtchnell
quality floral products and how temporal fac- 2008) which investigatedthe flowering ecology
tors affect floral resources, for example, pollen of south-east Australian melliferous (honey-
quality and availability. Pollen quality usually producing) flora and used observational data
is determined bynutritional levels (Keller etal from highly experienced, commercial apiarists
2005) andanumberofstudieshaveinvestigated to provide insight into otherwise difficult and
pollenqualityintermsofnutrition,byanalysing time-consumingecologicalexaminations.
protein (Roulston et al 2000; Somerville 1999;
Methodology
Somerville andNicol2006; Hanleyet al 2008),
Theapiarists
amino acid (Roulston and Cane 2000; Somer-
villeandNicol2006),lipid(ToddandBretherick Sixty-six apiarists were contacted and inter-
1942;Youssefefal 1978;Dayetal 1990;Singhet viewedforthebroaderstudyintofloralecology.
al 1999;ManningandHarvey2002; Somerville These66apiariststhenweresentquestionnaires
2005) and mineral composition (Herbert and relatingtopollenqualityandA. melliferanutri-
Miller-Ihli 1987). These are important works tion. The questionnaire consisted of20 closed
andopen-endedquestionspertainingtopollen
andprovideknowledgeofthenutritionalbreak-
down of pollens, but argument remains as to flora.Onlyresultsrelatingtotheexoticfloraare
presentedhere.
whichnutritionalcomponentsshouldbeusedto
Apiarists resided in the Australian states and
deTtheermihenaelptohlloefnqtuhaelihtyig(hKleylleorrgetanailse2d0,05)s.ocial territories of (southern) Queensland, New
South Wales, the Australian Capital Territory
Honeybee (‘bee) Apis mellifera L. and hive
(ACT), SouthAustralia, VictoriaandTasmania
health could be used to reflect pollen quality.
Firstly,A. melliferaisfarmedintensivelyinhives duringtheirbeekeepingyears (Fig. 1);however,
the migratory nature ofAustralian beekeeping
and observed closelyby apiarists (beekeepers).
often necessitates shifting hives interstate. This
Secondly, ifpollen quality is poor, A. mellifera
and hive health will deteriorate. If protein is rangedefinesthe‘studyarea (Fig. 1).
i(nKslueffiincsicehntm,idbtroaonddKreoanrdiongsd1e9c7r6e)asaensdmlaonrgkeevdiltyy atEeadcchomampeiarrciisatlilnyvfoolrveadmiinnithmisumstoufd3y0hyaedarsopaenrd-
is decreased (Sakagami and Fukuda 1968). Fi- Tmhainsaegnesduraemdianpiiamriusmtsohfa3d50ahnivienstiamtaatneyaonndeltoinmge-.
nally,apiaristshaveobservedvariationinpollen
‘quality’andfactorswhichmaycausesuchvari- termunderstandingofpollenqualityanditsvari-
ation over extended periods oftime. These ob- ation,hivemanagementandA.melliferanutrition.
servations have beenmade over decades (often Recruitmentofapiaristswhofulfilledtheselection
generations) andtheirlivelihoodhas depended criteriawasundertakenusingtwo methods: first,
onaccurateassessmentsofsuchvariation.Thus, the gatekeeper’ approach (Berg 1999), whereby
apiarists representan alternativebutimportant contact details for 11 apiarists were provided by
source of long-term observational data. Their thebeekeeperwhoinitiallysuggestedtheresearch
understanding ofpollen qualityandquantityis concept;second,the‘snowball’technique(Gilbert
1993; Robson 1993), whereby each respondent
critical to their livelihood. Their observations,
therefore, are likelyto besignificantlymoreac- wasaskedto providedetails ofotherexperienced
curateandcontinuousthananyothersources. apiarists.Thesetechniquescommonlyareusedin
This studyaimedto assesswhetherweeds are socialresearch (e.g.Mesquitaetal 2001;Momar-
friend orfoe to ecosystem functioningbycon- tinetal 2002; PoczwardowskiandConroy2002)
sideringthepollen qualityand quantityoffered andemployexistinginterpersonalnetworkswith-
by common, widespread weeds to Australian in closedcommunities and, in this case, a closed
ecosystems. UsingApis melliferaasacasestudy, industry, to encourage participation in research
(McLean and Campbell 2003). To ensure addi-
Vol 127 (4) 2010 125
Research Reports
tionalrespondentswererenownedfortheirexper- 1). Few apiarists provided comments relating
tise,eachpotentialparticipantwasrecommended to short-term infraspecific variation in pollen
independentlybyat least two otherparticipating quality, but those who did considered tem-
apiarists. Participants were recruited and inter- perature and rainfall to be most significant in
viewedinaccordancewithDeakinUniversityeth- determining pollen quality. Hot temperatures
icsrequirements(Permit:EC92-2003). were considered detrimental but rainfall was
Results beneficial. Budding/floweringintensityandsoil
Thirty apiarists (45%) returned completed type also were believed to affectpollen quality,
forms. This response rate was very good; nor- but it is unknown whether these factors were
mally, a25%to35% responseis consideredad- detrimentalorbeneficial topollenquality.
equate(Somervilleand Nicholson 2005). How- Approximately half of respondents (13/28)
ever, not all apiarists completed all questions. observed long-term variation in pollen qual-
Thus, thesample size n’ displayed inparenthe- ity in exotic species such as H. radicata. One
sesafterresults reflectsthe numberofrespons- respondent commented that all ground flora
es to thatquestion, rather than the number of showed long-term variation; another two stat-
/a3p0i)a.risFtosrwehxoamprleet,urtnheed ntohteatqiuoensti[o12n/n1a7i]rein(di.ie-. Ledontgh-attealrlmspveacriieastivoanriinedaldlesppeencdieisngatosnocmleimsaittee.s
cates that 17 apiarists answered the particular wasobservedbytenapiarists (10/29) - particu-
questionand,ofthese 17responses, 12apiarists larlythe Victorian/South Australian mallee re-
providedthesamefinding. gionandMaryboroughinCentralVictoria.
Fourteenfactorsthatinfluencedratingofpol-
Pollenquality lenqualitywereidentified (Fig. 3) andmostre-
Apiarists provided assessments ofpollen qual- spondents (21/30) used more than one factor.
ity for 32 exotic species, which belonged to 14 Pollencolour, sizeandvolumewereconsidered
families(Table 1).Thefamiliesmostcommonly important most often (Fig. 3) with taste and
represented were the Asteraceae, Fabaceae (six texturebeingless frequentlycited (itshouldbe
specieseach)andRosaceae(fivespecies) (Table remembered that thenumber ofcitations does
1). Pollen qualitywas assessed using a scale of not make one factor more or less important:
oneto five,whereonereflectedverylowpollen one factor mayhave agreater influence in one
qualityandfivereflectedveryhighpollenqual- site than another and particular factors may
ity. Whilst many species used for pollen were be easier to discern than others). Apis mellif-
native (64%) (Birtchnell2008),apiaristsidenti- era health and longevity, colony health, brood
fiedtheexoticflorapresentedinTable 1 asvital healthandbroodlayoutbythequeenalsowere
in providing adequate pollen to maintain A. mentioned as important indicators of pollen
mellifera and colonyhealth. qualitybutwere citedbyless than fiveapiarists
Thevastmajorityofexoticfloraofferedmid-to- (Fig. 3). Similarly, soil type, rainfall/soil mois-
high quality pollen (Table 1). Most species that ture and seasonal variation were cited by less
hadanaverageratingofbetweenfourandfivehad than five respondents (Fig. 3). For example, in
a narrow range ofratings (usually between four deep sand (Fig. 3), H. radicata and Medicago
and five) (Table 1). Most species were assigned a sativa were cited as producing copious pollen
rangeofratingsbyeachapiaristwhousedthemas of high quality which maintains vigorous A.
apollensource:forexample, Trifoliumrepenswas melliferabroodhealth.
usedbyonlyoneapiarist,yetwasassignedarange Pollenquantityandavailability
of3-5(Table 1)thushighlightingthatvariationin Most apiarists found pollen was available
pollenqualitywithinaspeciescanoccur.
throughout that species’ flowering period
Pollen quality varied within a species (infra-
(24/30). Three found thatpollen was available
specific variation) during a single flowering
either throughout the flowering periodoronly
event (21/29), on a seasonal basis (in longer
during part of the flowering period. Twenty
f(l15o/w2e4r)in(gFisgp.ec2)i.esE)ch(2i3u/m27p)laanntdagfirnoemumsi,tBertaosssiictae laepinaarviastislafboleunddurvianrgiaftlioowneirnintghe(2q0u/a2n9t)i.tTywoofpaopli--
scpupm.,oSfafliicxinsaplpe.,anHdyptohcishtoleers,isinrapdairctaitcau,larT,arwaexrae- dareipstesncdeodmmoennttheedsptehcaitesv,arainadblweeproelelxehniqbuiatnetdibtyy
cited as having variable pollen quality (Table some flora but not others. Yet another stated
126 TheVictorian Naturalist
Research Reports
that less pollen was available at either end of vourite pollens/pollen sources (28/29). Eight-
the season. Twenty-six apiarists noted that A. eenexotic specieswerecitedasbeingfavourite
melliferabegancollectingpollen fromasource pollen sources (Table 2). Arctotheca calendula
as soon as flowering commenced (26/30) and pollen was cited most often (6/29) as being
16believedA. melliferacontinuedtodosountil favoured by A. mellifera followed by Brassica
,
floweringinthatspecies ceased (16/30). napus (5/29), E. plantagineum (5/29) and Tri-
The period of pollen collection usually was folium spp. (4/29) (Table 2). All other species
longer than the period of nectar collection werecitedlessthanfourtimeseach.Whilstthe
(15/27). Five apiarists considered that nec- numberofcitationsforeachspeciesmayreflect
tar and pollen collection periods were similar therelativevalue and‘favour’displayedforflo-
(5/27), four thatnectar collection occurred for ralsourcesbyA. mellifera, italsomaybeindic-
alongerperiod(4/27) andthreethattheperiod ative ofthe relativeabundance ofeach species.
of nectar and pollen collection varied (3/27). Four apiarists commented that Apis mellifera
Sometimes, pollen collection commenced at preferred pollen sourced from species such as
theonsetofflowering(2/30)whilenectaryields Trifolium spp., E. plantagineum, A. calendula,
commencedsubsequently.Alternatively,pollen Pyrusspp.andT.officinaleonthebasisthatthey
and nectar collection could slowly diminish hadhigherprotein contentandresulted gener-
together but sometimes both would cut out allyinlarge,healthycolonies.Indeed,theseflo-
overnight (1/30), orpollenwouldremainto be ralspecieswereidentified asproducers ofhigh
collectedwhen no nectar was available or vice qualitypollen(Table 1)andassourcesfavoured
versa. One respondentnotedthatnectar secre- byA. mellifera (Table2),indicatingthathigher
tion stopped once pollination had occurred. proteincontentmaybethekeytolinkingqual-
Three respondents commented that nectar ity, preferenceandA. mellifera health.
availabilitywas controlled bytemperature and ThenotionthatA. mellifera have pollen pref-
rainfall. erences was supported furtherbya number of
particular observations. For example, the na-
Pollenpreferences
tive and ubiquitous Eucalyptus camaldulensis
Nearly all respondents believed A. mellifera hadabundantnectarandpollenyetA. mellifera
display pollen preferences and, thus, have ‘fa- visited thenearby T. officinale. Variationinthe
J
'Ti-,
4 %
A
/
>r
NORTH *
(
Queensland
V$} AuMldliJ y
Brisbane^
South
Australia
A NewSouth
Wa,M
/-
JSydmty
“t . , • Canberraj
Victoria AustralianCapitalTerritory
Melbourne
’*olalas/fofitonai.hiw*ir|iapiansl*
jinvolvedinthisMLtayrowdon Tasmania
Studyare* ^^lobart
|
Fig. 1. Map ofAustralia showingstates/territories in which apiarists resided and extentofmigratoryrange
coveredbythisstudy(studyarea).AdaptedfromAustralianGovernment.
Vol 127 (4) 2010 127
Research Reports
PhRKIiioeaBngsbyohal-rcetopaePo1agiflenmalaaPme(coic5nell)e;ilaaqeeeGuesnae(l1((rqi)2nut;)o-ay;.)lPG.iIostenapySrep--acoeimfcIPeainoseecdaxseioacntaceaieeaercaeea(esc1(p)ph2(;er)1c.ef)Ri;saueemtsCniutl-oyeenRdiu-antiCanspuccacdearuleaeeresnbtco(hie1fetn)sa;d1eci-Bse5nr)ag:a(e1-Ao(sr1rBt)der;feal-rOsecsnAotissafctaeat-crheveaOaeicnerreaya(ga3e)lvr;eao(r6cAw)aes;gpapeoFela-lrb(eaA1tn)s-;ipnqShFguaoaalbdloiafe-tclypeSaoaacalleenliaedc(ena6c5)(eq?1aua)Rae;loviB(wset1ory)>r-y.;
Average Range Numberof Numberof
pollen of respondents respondents
quality rating citing who
Species Family rating species consider
(n=30) pollen
quality
variable
CentaureasolstitialisL. Ast 5 5 2 o
UlexeuropaeusL. Fab 5 5 2 0
PEyrroudsiucmomcmicuuntiasriLu.m (L.)L’Her. GReosr 55 55 11 oo
Citrussinensis(L.)Osbeck,Citrusspp. Rut 5 5 1 o
Trifoliumspp. Fab 4.9 4to5 9 o
Prunusdulcis(Mill.) D.A.Webb Ros 4.8 4to5 5 o
Brassicarapa Bra 4.6 4to5 6 o
ViciafabaL. m Fab 4.5 4to5 2 o
ATsripfhooldieulmufsrfaigsitfuelroususL. AFsapb 44..54 44ttoo55 51 01
EChcohnidurmilpllaajnutnacgeianeL.umL. BAsotr 4.41 24ttoo55 134 o1
Romulearosea(Ewart) M.P.deVos Iri 4 4 2 o
MalusdomesticaBorkh. Ros 4 4 1 o
CTruicfuomliiusmmryerpieoncsaLr.pusNaudin CFuabe 44 3t4o5 11 01
Arctothecacalendula(L.)Levyns Ast 3.7 3to5 19 2
Rubusfruticosusspp.agg. Ros 3.7 3to4 3 0
MedicagosativaL. Fab 3.7 3to4 1 0
Brassica rapassp.sylvestris(L.)Janch. Bra 3.3 1 to5 6 0
BrassicanapusL. Bra 3.2 1 to5 15 0
Thistles(Undet.spp.) Ast 3.2 1to5 6 2
Taraxacum officinaleWeber Ast 3.2 1 to5 5 0
OenotherabiennisL. Ona 3 2to4 2 0
HeliotropiumamplexicauleVahl Bor 3 3 1 0
Salixspp. Sal 2.5 1 to4 2 0
HypochoerisradicataL. Ast 1.9 1 to4 7 0
PinusradiataD.Don Pin 1 1 3 o
Prunussp. (Cherry) Ros 1 1 1 0
LoliumperenneL. Poa 1 1 1 0
Zea maysL. Poa 1 1 1 0
preferred pollen source could occur within a considered less beneficial were believed to de-
single day, according to most apiarists (20/30). crease only A. mellifera health and longevity
Thisprincipallywas attributed to weather con- (Table 4). Four species were included in both
draiitnifoanlsl,w(9i/n3d0)an(dFitg.he4o)ccsuurcrhencaseotfemsptoerrmast.ure, sliasttis:vaA,.scuaglgeensdtuilnag,sBp.atniaapluasn,dT/.oorfftiecminpaolreaalnvdarMi.-
Pollen effectsonA. melliferahealth ation affects pollen nutrition. Actotheca calen-
dulawascitedmostoftenasbeingmorebenefi-
Fifteen exotic pollen sources were considered
more beneficial to A. mellifera than others cial (8/30), followed byE.plantagineum (7/30)
(Table 3). Seven species were cited only once
(Table 3) but only seven were considered less
bmeonerfeicbieanlef(iTcaiball’ep4)o.llSepnecwieersectohnosuigdhetretdotsopeyciiefli-d l(oTawbelreq3u)a,listuyggpeolslteinn,ghtahvaettahmeoserespreecsitersicptreodvdiidse-
tribution or are targeted byfewer apiarists due
callyincreaseA. mellifera health,broodhealth,
to extended intervalsbetweenflowering events
longevity and hive health (Table 3). Species orpoornectarproduction.
128 TheVictorianNaturalist
Research Reports
Brassica napus and H. radicata were cited better. This was attributed to multiple sources
mostoftenasbeingproducersoflessbeneficial resulting in stronger A. mellifera individuals
pollen (3/30 and 2/30 respectively) (Table 4). and colonies, individuals with stamina, in-
Ofthe fourspecies included in both the ‘more creasedlongevityandimprovedbrood rearing.
beneficial’ and ‘less beneficial’ lists, the major- Itwasbelievedthatmultiplepollensourcespro-
ity were cited most often as ‘more beneficial’ videda morebalanceddietandhigherprotein.
(Tables 3 and4).Forexample,A. calendulawas Three respondents, however, stated that some
cited eight times as producing morebeneficial singlesourcesofpollenmatchedthebenefitsof
pollen (Table 3) and only once as being less multiplesources, forexample, Trifoliumrepens,
beneficial(Table4). T.fragiferumandE.plantagineum.
Apiarists observed that healthy A. mellifera Most apiarists stated thatworker bees collect
were stronger, had greater stamina and were pollen frommultiplesources (27/30). Thepro-
fatter. A healthybrood had greater vigour and portion collected from each source varied and
filled a larger expanse ofthe frame (the com- depended on:
ponent ofthebee hive inwhich the queenlays • diversityofavailablepollen;
eggs and where young bees are reared). Pol- • quantityofpollenavailable;
len produced byeight exotic floral specieswas • pollencomposition;
of sufficiently high quality to improve brood • distancebetweenapiaryandpollen source;
health; seven species increased A. mellifera • tasteandrequirementoftheindividualhive;
health, but only few species increased A. mel- • timeofday;
lifera longevity (three species) and hive health • season;
(onespecies) (Table3). Interms ofdetrimental • weather (specifically temperature, rainfall,
impacts, onlydecreasedA. mellifera healthand wind);
longevity resulted from less beneficial pollen • duration offlowering;
(threeandonespeciesrespectively) (Table4). •theprevalence/densityofinsects,presumably
There was little difference in reports by apia- competingforfloralresources.
rists as to whether one or multiple sources of Ifhigh qualitypollen was available, A. mellifera
pollen were sufficient to maintain hive health collectedmainlyfromthatsource,collectingonly
andto buildhive populations; however, all but asmallvolumeofpollenfromothersources.
three stated that multiple pollen sources were
Variationinpollenquality
Fig.2. Infraspecificvariationofpollenqualityinexoticandnativeflora
Vol 127 (4) 2010 129
Research Reports
Fig.3.Factorsinfluencingratingofpollenquality.
sights.Certainly,ourabilitytostudy
thecontributionofpollensources is
facilitated by the use ofA. mellifera
as it is asocialinsectand, so,brings
pollen back to the hive where pol-
len‘traps’canbeinstalledtoremove
pollen from returning bees, allow-
ing analysis ofpollen quantity and
quality. Regrettably, analysis ofpol-
len collected by non-social inverte-
brates and vertebrates, generally, is
notso simple.
Apismelliferakeep onlyasmallre-
serve ofpollen within a hive at any
one time (Pernal and Currie 2001)
making A. mellifera and, conse-
quently, their colony susceptible to
short-termenvironmentalvariations
Fig.4.Factorsaffectingdailypreferenceforapollensource
in pollen quality and availability. In
Discussion pollen-dependent species which are solitary
More than one-third of the important pollen (and, so, also keep only small reserves ofpol-
sources identified by apiarists were exotic spe- len),thisalsowouldbepertinentandhighlights
cies. Whilst this could be influenced by the theimportanceofadiversefloralcommunity,to
availabilityofapiaries (siteswherebeehivesare ensure temporal availability offloral resources
parked’) on oradjacent to exotic flora (for ex- such as pollen. Exotic plant species contribute
ample, crops), it certainly reflects the extent to to both floral diversity and pollen availability
which A. mellifera is dependent on exotic pol- and, so, could be a vital resource for solitary
len sources for pollen requirements. Whether pollen-dependentspecieswhen few nativespe-
this dependency on exotic pollen accurately cies (orcrops) areflowering.
reflectstheextenttowhich nativeinvertebrates Pollen sourced from exotic flora is avital re-
(and vertebrates) are dependent on exotic source for A. mellifera, and presumably also
pollen sources requires further investigation; is critical for countless pollen-dependent ver-
nonetheless, this study has given valuable in- tebrates and invertebrates (e.g. Churchill and
130 TheVictorian Naturalist
Research Reports
apiarists’observationsofpollensources,pollen
Table2.Pollensources‘favoured’byApismellifera
qualityandquantity,andtheeffectsofdifferent
Species Numberof pollen onA. mellifera health show the impor-
citations(n=29) tant contribution exotic floramake to the pol-
Arctothecacalendula 6 lenresourcebase.
Brassicanapus 6 Apiarists identifiedthatthebulkofharvested
Echiumplantagineum 5
pollengenerallycamefromonlyafewplantspe-
Trifoliumspp. 4
Prunusdulcis 3 cies. This was expected as it also was observed
Pyrussp. 3 inotherstudies (e.g. Synge 1947; Shawer 1987;
Salixspp. 2 Cortopassi-Laurino and Ramalho 1988). The
Taraxacumofficinale 2 three exotic species used most frequentlywere
Allintroducedgroundflora 1 A. calendula B.napusandE.plantagineum.Kel-
BBrraassssiiccaatrraappaassp.sylvestris 11 ler et al. (20,05) reviewed about 25 studies and
Citrussinensis 1 found B. napus ranked as one ofthe five top
Hypochoerisradicata 1 pollen sources globally. This is not surprising
Malussp. 1 considering the abundance ofthis crop world-
?Mostgroundflora 1 wide. It is likely thatpollen from floral species
Trifoliumfragiferum 1
Ulexeuropaeus 1 identifiedinthisstudyalsoprovidescriticalnu-
Vidafaba 1 trition to a range ofnative vertebrates and in-
?denotesspecieswithunknowngeographicorigin vferrotmebrhaitgehs,quaaslitthyespeolslpeenc.ieTshuasl,soitwoisullidkebleynetfhiatt
Christensen 1970; Turner 1984; van Tets and the exotic species identified here are now im-
Hulbert 1999; Pestell and Petit 2007). The ex- portant components ofAustralian ecosystems
tent to which species utilise pollen and the intermsoftheirroleinmaintainingpopulation
range ofpollen sources consumed, however, is dynamics.
incrediblyvariable. Forexample, Roulston and Apiarists assess pollen quality on a holistic
Cane(2000)conductedacomparisonofthedi- basis.Theyobservewhetherdetrimentaleffects
gestionofpollenindifferentanimalsandfound occur, either immediately or with time, and
thatthepercentageofpollengrainsemptiedby so their assessment ofpollen quality may not
passingthoughthe digestivetractrangedfrom matchthose determinedbychemical analyses.
0% (a Honeyeater - bird) to 98.2% (A. mellif- For example, pollen from some native species
era) andvaried depending on the floralsource were considered excellent quality by apiarists
ofthepollen. In anotherstudy,batspeciesthat (that is, they were rated > 4) (Birtchnell and
regularlyconsumedpollenextractedpollency- Gibson2006)butprotein analyses suggestpol-
toplasm (so, digested pollen) more efficiently lenfrom the samesourcesare ofaverage qual-
than those bat species which did not consume ity (Somerville and Nicol 2006). This contrast
pollen regularly(HerreraandMartinezdelRio in reports ofpollen quality may highlight the
1998).DigestionofpollenbyA. melliferadiffers underestimation of quality typical of chemi-
dependingon the floral source (and the age of cal analyses ofpollen collected byA. mellifera
thebee), withpollen from some speciespossi- (Roulston and Cane 2000). Generally, chemi-
blybeingdigestedby osmoticshock’ (Kroonet cal composition ofpollen is reported on a per
al. 1974).Incontrast,pollenfromothersources weightbasis, thus reportedvalues probably do
such as T. officinale and M. sativa both iden- notaccountforthe addedweightofthe nectar
,
tifiedin this study as important and beneficial orhoneysugarswhich are added to thepollen
pollen sources, were digested slowlybydegra- by A. mellifera prior to transport. Therefore,
dation (Peng et al 1985; 1986). Thus, in order it is likely that the concentration ofchemical
todeterminethepotentialnutritionalvalue(i.e. constituents in the pollen itselfwill be greatly
effects on organism health and, so, ecosystem underestimated in chemical analyses owing to
health) it would be necessary to know the bi- the(highlyvariable)contributionmadebynec-
ologyofthe species consuming the pollen and tarorhoneyto the pollen mass (Roulston and
the nature ofthe pollen itself. Thus, the actual Cane2000). Furthermore,spatialandtemporal
contribution of pollen to broader ecosystem influencesonharvestedpollenmayaccountfor
health is difficult to quantify. Nonetheless, discrepancies between reported values. There-
Vol 127 (4) 2010 131
Research Reports
Table3.Impactof morebeneficial pollenonApismellifera.Speciesinboldwereconsideredbothmore
beneficial(Table3)andlessbeneficial(Table4).
Species Increased Increased Increased Increased Numberof
A. mellifera A. mellifera A. mellifera A. mellifera respondents
health brood health longevity hivehealth
Arctothecacalendula X X
8
Echiumplantagineum X X X
7
Brassicanapus X
6
Trifoliumspp. X X X 6
Brassicarapa X X X
4
Prunusdulcis X X 3
Taraxacumofficinale X 2
Ulexeuropaeus
2
Asphodelusfistulosus X
Chondrillajuncea X 1
?Daisies(Asteraceae) X 1
1
Medicagosativa
Pyrussp. X 1
1
Rubusfruticosusspp.agg.
1
Salixspp.
1
?denotesspecieswithunknowngeographicorigin
Table4. Impactof‘lessbeneficial’pollenonApismellifera.Speciesinboldwereconsideredbothmorebenefi-
cial(Table3)andlessbeneficial (Table4)
Species Decreased Decreased Decreased Decreased Numberof
A. mellifera A. mellifera A. mellifera A. mellifera respondents
health brood health longevity hive health
Brassicanapus X X
3
Hypochoerisradicata X 2
Arctothecacalendula
1
Heliotropiumamplexicaule
Medicagosativa X 1
1
Oenotherabiennis
1
Taraxacumofficinale
1
fore, theway an apiarist assessespollen quality Nicol 2006). A numberoffrequently usedpol-
maybemore accurate than results determined lensources,however,showedhighproteincon-
bychemical analyses, as apiarists’ observations tents, e.g. E. plantagineum and Trifolium spp.
are notnecessarilyaffected bythe relative ratio (Somervilleand Nicol2006).
ofnectar to pollen, but rather bythe impact of It is well known that pollen quality varies
the pollen pellets on A. mellifera health, lon- amongst plantspecies (Somerville2000; Keller
gevity and other biological parameters, and etal. 2005). Thisstudyhas identified that apia-
are based on observations of pollen over an rists observed great variation in pollen quality
extended period rather than asingle (orshort- within a species, with just under halfthe spe-
term) collectionevent. cies cited being given different quality ratings
The contrast in values of pollen quality re- (Table 1).Thenumberofrespondentswhocon-
ported in this study and those reported else- sider any one species to have a variable pollen
where also mayresultbecause apiaristsconsid- quality, however, is extremely small (Table 1).
erquantityofpollenwhendeterminingquality. Arguably, the variation observed by apiarists
Thus, thefrequencyapollen resource was cited may exist because each has a different quality-
by apiarists does not necessarilyreflect its pol- benchmark; however, this is unlikely owing to
len quality. For example, A. calendula and B. theclose-knit, interdependentnatureofapicul-
napus were frequently cited yet protein analy- ture whereby apiarists often are reliant on the
ses showed the first to be ofpoor quality and accurate observations made and reported by
the latter of average quality (Somerville and anotherbeekeeper.Itisprobable,therefore,that
132 TheVictorian Naturalist
Research Reports
infraspecificvariation observedinpollenqual- itypollenatanytimeofflowering.Causesforthis
ityisspatial,thusinfluenced byenvironmental were unknown and are difficult to explain. Poor
factors specificto location. Indeed, halfthere- qualitypollenattheendoffloweringseemslogical,
spondents agreed that site variation in pollen assoilresourcescouldbedepletedbecauseofplant
quality occurred and was found to influence growth andcompetition, butatthe start offlow-
other aspects of flowering ecology (Birtchnell ering onewould expect resourcesto beadequate
2002; Birtchnell and Gibson 2006; Birtchnell ifnotbetterthan atlaterstagesofflowering.This
2008). Somerville (2000; 2005) and Manning couldbeduetochangedrainfallpatterns:apiarists
(2001) also showed that spatial variation in havereportedthatrainisnotfallingattheseason-
pollen quality could occur within a species. ally-appropriate times (Birtchnell 2008) - many
A number of factors other than spatial in- apiaristsnotedthatthishasaffectedfloweringpat-
fluence were identified as responsible for dif- ternsandnectarproduction(Birtchnell2008),and
ferences in pollen quality within a species. alsomayhaveaffectedpollenavailabilityandqual-
Temperature and rainfall were considered sig- ity.Researchintopotentialimplicationsofthisisa
nificant influences over pollen quality: pollen matterofurgency.
qualityreportedlydecreased during high tem- Forty-threepercentofbeekeepersbelievedthere
peratures and/orlowrainfall. Flowering inten- was long-term, infraspecific variation in pollen
sity(levelsofgeneralbudding)andsoiltypealso qualitybut50%ofbeekeepersdidnot.This,again,
were important. This has important implica- couldbeduetositespecificvariablesincludingthe
tionsforfaunalandinvertebratedynamics and natureofthepollenresourcesavailable,especially
highlights the importance of landscape-scale as the majorityofthose who had observed such
resource availability in ensuring ecosystem fit- variation noticedvariationwithinparticularspe-
ness. Assessing budding/flowering intensity cies such as H. radicata and between sites, par-
maybe agood indicator ofpollen quality as it ticularlytheVictorianmalleeandcentraldistricts,
easily is undertaken byapiarists, but quantita- whichtypicallyexperiencelowerrainfallthanelse-
tivedataisnecessarytodeterminewhetherthis where.Observedlong-termvariationinthesedry
isaperceivedoractuallinearrelationship.Simi- areas, therefore, could be exacerbated further by
larly,itwouldbeusefultocomparepollenqual- reduced annual rainfall levels predicted to occur
itywith chemicalanalyses ofvarious soil types morecommonlywithclimatechange.
toquantifywhethersoiltypeprovidesareliable Almostall apiarists involved in this studybe-
surrogate forpollen quality. TheWestern Aus- lievedA. melliferahadafavouritepollensource.
tralian Corymbia calophylla (Lindl.) K.D. Hill A number ofearlier studies demonstrated that
&L.A.S. Johnson (Myrtaceae)producedpollen A. mellifera colonies regulated pollen foraging
proteinwithhigheramino acidandlipidlevels in response to changing protein demands (e.g.
whenlocatedonheaviersoiltypescomparedto Dreller et al. 1999; Fewell and Bertram 1999).
thesamespeciesgrowinginsandycoastalsoils Colony foraging increased in proportion to
(Manning2001). decreased pollen storage (Lindauer 1952; van
Otherfactors affectingpollen qualityweretim- Laere and Martens 1971). Experimental ma-
ingwithin afloweringepisode andwithinasea- nipulation of stored pollen resulted in com-
son. Season couldbe a reflection ofthe different pensatoryresponsesintermsofthenumbersof
speciesinflower, afactwhichiswidelyacknowl- foragerssentfrom thehiveandsubsequentrate
edged in the literature (Keller etal. 2005), but it ofpollencollection(Camazine 1993;Dogterom
also could be a reflection ofweather conditions. and Winston 1999; Dreller et al. 1999; Fewell
Hot, dry conditions were cited earlier as reduc- and Bertram 1999). Pernal and Currie (2001)
ingpollenquality,sospeciesfloweringinsummer extend this notion further and document that
mightproducelesserqualitypollenpurelydueto changesinforagingatthecolonyleveloccurred
prevailingweatherconditions. Variation withina inresponsetodeficitsineitherquantityorqual-
single flowering episode also may be species de- ity ofpollen. This resulted from an increase in
pendent, asabout70% ofrespondentsstatedthis the proportion offoraging bees. The hive pro-
wascommon.Someplantspecieswereconsidered duced more foragers so an increased number
to have poor qualitypollen at eitherorboth the ofyoung, inexperiencedforagers weresent out
beginningoffloweringandthe endofflowering, in response to quality and quantity deficits of
whereas other species could produce poor qual- pollen. These tended to collect larger loads of
Vol 127 (4) 2010 133