Grassland management for meadow birds in the Netherlands is unfavourable to pollinators

(1)

BasicandAppliedEcology43(2020)52–63

Grassland

management

for

meadow

birds

in

the

Netherlands

is

unfavourable

to

pollinators

M.F.

(Marco)

Tanis

a

,

Leon

Marshall

a,b

,

J.C.

(Koos)

Biesmeijer

a,c

,

Lisette

van

Kolfschoten

a,d,∗

a_Naturalis_Biodiversity_Center,_P.O._Box_9517,₂₃₀₀_RA_Leiden,_The_Netherlands

b_Agroecology_and_Pollination_Group,_Landscape_Ecology_&_Plant_Production_Systems_(LEPPS/EIB),_Université LibredeBruxelles(ULB),AvenueFranklinRoosevelt50,1050Brussels,Belgium

c_Institute_of_{Environmental}_Sciences_(CML),_Leiden_University,_Einsteinweg_2,₂₃₃₃_CC_Leiden,_The_Netherlands d_Department_of_Plant_Ecology_and_Evolution,_Evolutionary_Biology_Centre,_Uppsala_University,_Norbyvägen_18D, UppsalaSE-75236,Sweden

Received6March2019;accepted6December2019 Availableonline16December2019

Abstract

Agriculturalintensificationandlossofsemi-naturalgrasslandhavecontributedtobiodiversitydecline,includingpollinator species,inpasturesaroundthe world.To reversethedecline,agri-environmentalschemeshavebeenimplemented,varying widelyineffectiveness.Inaddition,manycountries,includingtheNetherlands,haveestablishednaturereservesinwhich semi-naturalgrasslandsarerestoredandareoftenmanagedforspecificgroupsofspecies,e.g.meadowbirdsorplants.Theeffects ofsuch measuresoninsectbiodiversityarenotwellknownbutrecentreportsonthe dramaticdeclineof insectbiomassin naturereserveshaveputevenmoreattentiontotheimpactoflanduseandmanagementonbiodiversity.Thisstudycompares pollinator abundanceandspecies richnessin threecommon semi-naturalgrassland management types inthe Netherlands: (1)haymeadows,(2)herb-richgrasslandsand(3)meadowbirdgrasslands.Pollinatorabundanceandspeciesrichnesswere assessedinelevenstudyareas,eachwithallthreemanagementtypespresent.Standardizedtransects,insectsamplingwithina standard20mintimeframeandplot-basedflowersurveyswereusedinspringandsummertoassesstherelationshipsbetween managementregime,floralabundanceanddiversityandpollinatorcommunities.Theresultsshowthatmeadowbirdgrasslands havelowerpollinatorabundanceanddiversityandalessuniquepollinatorassemblagethanbothothertypes.Moreover,flower abundancehasapositiveeffectonpollinatorabundanceandflowerdiversityhasapositiveeffectonpollinatorspeciesrichness. Theseresultsindicatethatmeadow-birdgrasslandsareacomparativelyunfavourablehabitatforbees,hoverfliesandbutterflies, whichmaybeexplainedbyalackofflowersaswellasunsuitablemowingpractices.Measuresbenefittingbothinsectivorous birdsandflower-visitinginsects,suchasrotationalmowing,couldremediatethisimbalance.

Keywords: Agriculturalintensification;Insectdecline;Meadowbirdconservation;Mowing;Pollinators;Semi-naturalgrasslands

∗_{Corresponding}_author.

E-mailaddresses:marcotanis@hotmail.com(M.F.Tanis),leon.marshall@naturalis.nl(L.Marshall),koos.biesmeijer@naturalis.nl(J.C.Biesmeijer), lisette.vankolfschoten@ebc.uu.se(L.vanKolfschoten).

https://doi.org/10.1016/j.baae.2019.12.002

(2)

M.F.Tanisetal./BasicandAppliedEcology43(2020)52–63 53

Introduction

Overthepastdecades,agricultureinEuropehasrapidly intensified to increase productivity, causing widespread lossesinfarmlandbiodiversity(Matson,Parton,Power, & Swift,1997;Krebs,Wilson,Bradbury,&Siriwardena,1999; Benton,Vickery,&Wilson,2003).Intensificationhasbeen linkedtodramatic declinesinthe populationsof farmland birds(Chamberlain,Fuller, Bunce,Duckworth,&Shrubb, 2000;Donald,Green,&Heath,2001;Benton,Bryant,Cole, &Crick, 2002) andinsects (Benton et al., 2002),as well as homogenizationof communities (Gossner etal., 2016). Intensive agriculture is most likely responsible for insect declines in nature reserves embedded within the agricul-turalmatrix,witha75%declineininsectbiomassoverthree decadesinGermannaturereserves,(Hallmannetal.,2017) as a prime example. Agricultural intensification is there-foreconsideredtobethemaindriverofbiodiversitydecline inEurope (Bentonetal.,2003;Tscharntke,Klein,Kruess, Steffan-Dewenter,&Thies,2005;Pottsetal.,2016).

Pollinators are one example of a declining group of insects(Thomasetal.,2004;Biesmeijeretal.,2006;Kosior et al., 2007; Goulson, Lye, & Darvill, 2008; Potts et al., 2010,2016;VanSwaayetal.,2015).Pollinators,in particu-larbees,performimportantecosystemservices.Atleast75% offoodcropspeciesdependtosomedegreeonanimal polli-nation(Kleinetal.,2007),whilelargeanddiversepollinator communitieshavebeenshowntoimprovefruitset(Garibaldi etal.,2013;Winfreeetal.,2018).Moreover,around88%of allwildplantspeciesdependonanimalpollination(Ollerton, Winfree, & Tarrant,2011), making pollinators crucial for ecosystemfunctioning.

A main driver of pollinator decline is land-use intensi-fication,in particular the loss of semi-natural, extensively managedandnectar-richgrasslands(Osborne,Williams,& Corbet, 1991; Goulson, Hanley, Darvill, Ellis, & Knight, 2005; Goulson, Nicholls, Botías, & Rotheray, 2015; Van Swaayetal.,2015).Thesegrasslandsareamainhabitatfor pollinatinginsects,providingfoodandresourcesfor repro-duction,such as nesting sitesfor bees andhost plantsfor butterflylarvae(Osborneetal.,1991;VanSwaayetal.,2015). Thereplacementofsemi-naturalhabitatbymonoculturesof intensivelymanagedgrasslandshasledtowidespreadhabitat lossandfragmentationforpollinators,contributingtotheir declines(Pottsetal.,2010).

Tocounterbiodiversitydeclineinagriculturallandscapes, agri-environmentschemes(AES)havebeenimplementedin manyEuropean countries.Theseschemessupport farmers who adapt the management of their land to increase bio-diversity, e.g.through extensification.Theeffectiveness of AES has been mixed (Kleijn et al., 2006; Batáry, Dicks, Kleijn,&Sutherland, 2015)andlocalincreasesinspecies diversityandabundance(Batáryetal., 2010)seem depen-denton the structure and management of the surrounding landscape(Kleijn,Rundlöf,Scheper,Smith,&Tscharntke,

2011;Batáryetal.,2015).Anotherapproachoftentakenfor thepreservationoffarmlandbiodiversityistoconvert agri-culturallandintonaturereserves.Thesesubsidizedreserves are typicallymanagedtopromotecertainaspects of biodi-versity,forinstancefocusedonincreasingbreedingsuccess of meadowbirds(Ausden&Hirons,2002).Inthe Nether-lands,thisapproachhasbeensuccessfulformeadowbirds. They show a positive trend in protected areas while they are declining on farmlands with specific AES targeted at increasingthenumberofmeadowbirds(vanEgmond&de Koeijer, 2006). However, the effect of protected areas on insectabundanceanddiversityisnotmonitoredand there-forelargelyunknown,exceptforbutterfliesandgrasshoppers (InterProvinciaalOverleg,2018:seeAppendixA).Giventhe importanceofsemi-naturalgrasslandsforinsects(Öckinger &Smith, 2007),aswellasthefunctionalroleofinsectsas pollinatorsofwildflowers(Ollertonetal.,2011)andfood sourceforbirds(Beintema,Thissen,Tensen,&Visser,1991), moreknowledgeontheeffectofthedifferentmanagement typesontheinsectcommunityisneeded.

Tofillpartofthisknowledgegap,weassessedabundance andspecies richnessof pollinatorsandfloweringplantsin threedifferentmanagement typesofprotectedareasinthe WesternpartoftheNetherlands:(1)haymeadows,(2)herb rich grasslands and(3) meadow bird grasslands, to deter-minetheimpactofsemi-naturalgrassland managementon pollinators.Specifically,weassessedtheeffectofgrassland managementonpollinatorspeciesrichness,abundanceand assemblagecomposition.Weexpectedhaymeadowstohave thehighestpollinatorspeciesrichnessandabundance,aswell asthemostdiversespeciescomposition.Thesegrasslandsare typicallynutrient-poor,floristicallyrichanddiverseandare mownlateintheyear(seeAppendixB).Meadowbird grass-lands,onthe otherhand,tendtobericherinnutrients,are oftenmown inJuneandwere thusexpectedtobe floristi-callypoorer,havefewerflowersandconsequentlythelowest pollinatorabundanceanddiversityofthethreemanagement types.

Material

and

methods

Study

areas

(3)

54 M.F.Tanisetal./BasicandAppliedEcology43(2020)52–63

Fig.1. Locationsof the studyareaswithin Het GroeneHartand Midden-Delflandregionsof the Netherlands.Grasslands ofall three managementtypeswerealsopresentinstudyareas1and8butarenotvisibleonthemapduetotheirabsenceinthemostrecentversion ofthespatialdataset(InterProvinciaalOverleg,2018),asthesewereeitherrelativelyrecentconversionsfromagricultureorunsubsidized grasslands.

highgroundwater tables,periodicalinundation, less inten-sive grazingandextendedmowingdates. Thereservesare owned byNGOsandthegovernment,whodecideonhow thereservesaremanaged.Farmerswholeasethelandfrom theseorganisationscarryoutthemanagement.Often,parcels, i.e.plotsoflanddelineatedbyditches,withinthesereserves aremanagedaccordingtooneofthreetypes:haymeadows, whicharebotanicallyrichmesotrophicgrasslandsusedfor hay;meadowbirdgrasslands,whicharegrasslands richin meadowbirdswheremowingisdelayeduntilafterthe breed-ing season; and herb-richgrasslands, which are eutrophic grasslands often rich in herbs and thistles which do not meetthecriteria forthe othertwo types(see AppendixA; Appendix. B; Inter Provinciaal Overleg, 2018).The three management typesoften occur inclose proximity toeach otherandareembeddedinintensiveagriculture.Thiscreates anopportunitytostudytheeffectsofthethreemanagement types on insect pollinators, as we account for the spatial dependencyofthestudyarea,whichistreatedasarandom effectintheanalyses.

Toselectthestudyareas,thefollowingwasdone:first,all parcelsmanagedashaymeadows,herb-richgrasslandsand

meadowbirdgrasslandsweremapped(Fig.1;AppendixC). Studyareaswhereallthreemanagementtypeswerepresent withina2-kmradiuswereselected.Highlyfragmentedareas, definedasareaswhereatleastoneofthethreemanagement typewasrepresentedonlybysingleparcelssurroundedby differentmanagement,wereexcludedtopreventedgeeffects. Thirteenpossiblestudyareasmetthesecriteria.Fromthese thirteenareas,elevenwerechosen,asweexcludedtwostudy areasfromaclusteroffouroverlappingstudyareas.Finally, withineachstudyarea,threerepresentativesites,oneforeach managementtype,werechosenrandomly(seeAppendixD).

Field

sampling

(4)

M.F.Tanisetal./BasicandAppliedEcology43(2020)52–63 55 Ateachsite,transectsof200mweresetouttomeasure

pol-linatorabundance.Thesetransectsconsistedoffourparallel sectionsof 50mrunning lengthwiseat 50–100mdistance fromthefieldedge.Distance betweenthesections andthe outerditcheswerekeptequal,dependentonthewidthofthe parcel,withthesectionsbeingwiderapartwithincreasing width.Theminimumdistancebetweensectionswas5mand themaximum20m(seeAppendixE).Forthisreason,parcels narrowerthan25mandwiderthan80mwereexcluded.

Along thesetransects,everybee, butterfly, hoverfly and soldierflyeasilyidentifiabletogenusorspecieslevelinthe field(seeAppendixF)andseeninavirtual5-mcagearound theobserverwasidentifiedandcounted(thismethodisknown as‘Pollardwalk’andfollowsVanSwaayetal.,2018).After walkingthetransect,atimedsearchoftwentyminuteswas conductedwithintheareaoftheparcelenclosedbetween50 and100mfrom the edge toobtain an estimateof species richness.Duringthese20min,asmanyspeciesaspossible wereidentifiedorcollectedwhennotdirectlyidentifiableto species.Toensuresufficientactivity,samplingofpollinators wasonlydonebetween10a.m.and5:30p.m.Furthermore, weatherconditions wererecorded andalwaysmet the fol-lowingcriteria:norain,windspeedbelow6Bftandeithera temperaturebetween13and17◦Cwithlessthan50%cloud cover,oratemperatureabove17◦C(VanSwaayetal.,2018). Collectedspecimens wereidentified usingguides (bees: Falk & Lewington, 2017; hoverflies:Van Veen, 2004 and Schulten,2018;soldierflies:Reemer,2014)andonlinephoto albums (Falk, 2018). All identifications were checked by JohnSmitof EIS-NL (EuropeanInvertebrateSurvey– the Netherlands).

Besides recording pollinators, all flowering plantswere identifiedtospeciesin1-m2plotssetoutevery10malongthe transect.Allinflorescenceswerecountedperspecies(Batáry etal.,2010; AppendixG)toestimate flowerdiversityand abundance.Flowerandpollinatorsurveyswerealwaysdone onthesameday.Thethreegrasslandsiteswithinastudyarea werenormallysampledonthesameday,butincasesofbad weather,uptofivedaysapart.

Subsequently,pollinatorspeciesrichnesswasdetermined by combining the species lists of the transect and the timedsearchpersiteandremovingdoubleobservations(see AppendixH).Pollinatorandflowerabundancewere deter-minedasthetotalnumberofobservationsalongthetransects. Lastly, flower diversity was calculated with the Shannon diversityindex(Shannon,1948).

Data

analysis

Threegeneralizedlinearmixedeffects models(GLMM) wereconstructed,oneforspeciesrichness,andtwofor abun-dance.Forspeciesrichnessresponse wepooledallspecies recordsacrossallsamplingdatespersite.Forspecies abun-dancewe analysed springand summersamples separately becauseabundanceisstronglyinfluencedbyseason.

Forpollinatorspeciesrichness,adatasetwasconstructed containingthenumberofspeciesfoundperparcel,the man-agement type of each parcel and the averages of flower abundance, flower diversity, cloud cover, wind speed and temperature over the two sampling rounds as explanatory variables.Wetestedforcollinearitybetweenvariablesusing variance inflationfactors(VIFs)andremovedallvariables withVIFs>1.4(seeAppendixI).Flowerabundancewasnot includedinthemodelsofpollinatorspeciesrichness,dueto itscorrelationwithmanagementtype.Wefittedapoisson dis-tributedmodelwithmanagementtype,flowerdiversity,cloud cover,windspeedandtemperatureandallpossibletwo-way interactionsasexplanatoryvariables.

Forpollinatorabundancespringandsummermodelswere fitted separately and included management type, flower abundance, cloud cover, wind speed and temperature as explanatory variables. We again tested for collinearity between variables using variance inflation factors (VIFs) and removed all variables with VIFs >1.4 (see Appendix I).Flowerdiversitywasexcluded,asitwascorrelatedwith flower abundance. Both Poissondistributedmodels of the count datawereoverdispersed, thereforewe usednegative binomialmodels.Allpossibletwo-wayinteractionsbetween theexplanatoryvariableswereagainincluded.

Study area wasincluded as arandom effectand notas afixedeffectinallthreemodels toaccount forthe spatial dependencybetweenparcels inthesameareaandbecause the inherent variationbetween the eleven study areas was notofinterest.Thefinalmodelsforbothpollinatorspecies richnessandabundancewereselectedbased onthelowest AIC (Akaike, 1987) using dredge inthe MuMIn package in R statistics (version 1.15.6; Barton, 2016). All models withadifferenceinAICoflessthan2arepresentedinthe results,sincemodelswithinthisrangehavesimilarstrengthin explainingthevariationobserved.Additionally,wetestedfor differencesinflowerabundancepermanagementtypewith aKruskal-Wallistestcombinedwithapairwisecomparison testusingWilcoxonranksum.

Finally, to visualize the community composition across types, Venn diagrams were constructed. These show how manypollinator speciesare uniquetocertainmanagement typesandhow manyspeciesoverlapbetweenthe manage-menttypes.Non-metricmultidimensionalscaling(NMDS) wasusedtovisualisethesimilarityinpollinatorspecies com-positionbetweenthedifferentmanagementtypes.Lastly,we calculated a beta-diversity value to see if any differences betweensitesweresignificant.

Results

Pollinator

species

richness

(5)

56 M.F . T anis et al. / Basic and Applied Ecology 43 (2020) 52–63

Table1. ThebestfourGLMMmodels(AIC<2)explainingspeciesrichness.“Type”referstowhichgrasslandtypesweretested.“HRB”=herb-richgrasslands,“HM”=haymeadows and“MBG”=meadowbirdgrasslands.Theestimates(exp=exponential)refertotherateratiosofapoissondistributedmodel.The95%confidenceintervalofthelograteratiosare presentedforeachmodelcovariate.

Model1 Model2 Model3 Model4

Explanatoryvariables Est. Est.(exp) 2.5% 97.5% Est. Est.(exp) 2.5% 97.5% Est. Est.(exp) 2.5% 97.5% Est. Est.(exp) 2.5% 97.5%

(6)

Fig.2. Modelcoefficientestimateswith95%confidenceintervalsshowinginfluenceofgrasslandmanagementon(A)pollinatorspecies richness,(B)pollinatorabundanceinspring,and(C)summerpermanagementtype.AllmodelsrepresentthebestmodelaschosenbyAIC (seeTables1–3).Asterisksrepresentpairwisecomparisonwherethe95%confidenceintervalofthemeandifferencedoesnotincludezero. PhotoofHelophiluspendulus,commonhoverflyonthesegrasslands(photobyauthor).

speciesrichnesswasloweronmeadowbirdgrasslandsthan onothertypes, andsimilaronhaymeadowsandherbrich grasslands(Fig.2A;Table1).In addition,speciesrichness increasedwithhigherflowerdiversity(inoneofthebest mod-els),lowertemperatures(inthreeoutoffourmodels),andless cloudcover(inthethirdbestmodel)(Table1;AppendixJ). Notwo-wayinteractionswereincludedinthebestmodels.

Pollinator

abundance

Spring

Thesinglebestmodelincludedmanagementtype,flower abundanceandcloud cover(Table2).Again,management typehadastatisticallycleareffectonpollinatorabundance (Table2):abundancewasloweronmeadowbirdgrasslands thanonbothothertypes,andwasslightlyhigheronherb-rich grasslandsthanonhaymeadows(Fig.2B;Table2). More-over,flowerabundanceandcloudcoverhadapositiveeffect onpollinatorabundance(Table2;AppendixK).

Summer

Allthreetopmodels explainingpollinatorabundancein summer(Table3)includedastatisticallycleareffectof man-agementtype:meadowbirdgrasslandshadlowerpollinator abundancethanbothhaymeadowandherb-richgrasslands

Fig.3. Meanflowerabundanceofmajorforageplantsperparcel acrossmanagementtypes.Inbrackets:totalnumberofplantspecies found per management type. Ranunculus species were pooled. Flower species with less than 30 flower heads per parcel were includedinthecategoryother(seeAppendixL).

(7)

Table2. Thebest GLMM-Model (AIC) explainingspring pollinator abundance.“Type”refers to which grasslandtypes were tested. “HRB”=herb-richgrasslands,“HM”=haymeadowsand“MBG”=meadowbirdgrasslands.Theestimates(exp=exponential)refertothe rateratiosofanegativebinomialdistributedmodel.The95%confidenceintervalofthelograteratiosarepresentedforeachmodelcovariate.

Model1

Explanatoryvariables Estimate Estimate.(exp) 2.5% 97.5%

Type HRG-HM 0.63 1.88 −0.10 1.37 MBG-HM −0.68 0.50 −1.48 0.11 MBG-HRG −1.32 0.27 −2.06 −0.58 Flowerabundance 1.40 4.05 0.31 2.48 Cloudcover 1.23 3.41 0.36 2.10 AIC 236.8

ascloudcover,whichwasincludedinonemodel.No two-wayinteractionswereincludedinthebestmodelsforboth summerandspring.

Pollinator

species

composition

Atotalof1688pollinatorswererecordedorcollectedin thisresearch.Of these, 447 werebees, 1146 hoverflies or soldierflies,and95butterflies.Intotal,75pollinatorspecies werefound,including43hoverflyspecies,14beespecies,12 butterflyspecies,5soldierflyspeciesandoneburnetmoth species (seeAppendix M).Almost halfof the species (36 spp.)werefoundatleastonceinallmanagementtypes.Ofthe remainingspecies,mostwereeitheruniquetohaymeadows (9 spp.), herbrich grasslands(11 spp.), or both(11spp.), whilemeadowbirdgrasslandshadonlythreeuniquespecies (Fig.4A).

TheNMDS(stress<0.1)outputshowsthatoverallthe pol-linatorcommunitieswerehighlysimilaracrossthedifferent types.Thiswasconfirmedbyabeta-diversityanalysis,which showedthecommunitiestobesimilar(p=0.336).Basedon visualinspectionoftheoutput(Fig.4B),thespecies commu-nityfoundonhaymeadowspossessedthemostspecies-rich community,whichissimilartothatonherb-richgrasslands. Meadowbirdgrasslandcommunitiesharbourasubsetofthe othertwotypes(Fig.4B).

Flower

abundance

&

composition

Haymeadowshadthehighestflowerabundance,followed byherb-richgrasslandsandmeadowbirdgrasslands(Fig.3). The difference between the three types was significant (KruskalWallistest:p<0.05),aswasthedifferencebetween haymeadowsandmeadowbirdgrasslands(p=0.01),butnot betweenhaymeadowsandherb-richgrasslands(p=0.133)or meadowbirdgrasslandsandherb-richgrasslands(p=0.197). Themostabundantnectarplantsonallthreetypesof grass-lands were members of the genus Ranunculus, consisting of R. repens and R. acris on meadow bird and herb-rich grasslands andof R.ﬂammulaand R.acris onhay mead-ows(seeAppendixL).Rhinanthus minorwascommon on

haymeadows,whileCirsiumpalustrewascommonon herb-rich grasslands.Meadow birdgrasslands hadnoparticular commonspeciesapartfromRanunculusrepens(Fig.3).

Discussion

The type of management applied on protected, semi-naturalgrasslandsintheNetherlandsaffectsbothpollinator species richness and abundance as well as their floral resources. Meadow bird grasslands have fewer pollinator species,whichoccurinlowerabundance,andfewerunique species than either herb-rich grasslands or hay meadows. These results indicate that meadow bird grasslands are a comparativelypoorhabitatforpollinators.

Impacts

of

mowing

and

grazing

on

pollinators

Mowing (or grazing) is necessary to maintain semi-naturalgrasslands,thatotherwisedevelopintoforests,which explainswhyallthreemanagementregimesincluderegular mowingorgrazing.However,mowingcanhaveanegative effectonarthropods(Morris,2000),boththroughdirect mor-tality(Humbert,Ghazoul,&Walter,2009),andremovalof forageandreproductionhabitat(Cizek,Zamecnik, Tropek, Kocarek,&Konvicka,2012).Partialmowingmaymitigate these negative effects, as it leaves refuge areas (de Vries &Knotters, 2000;Humbert etal.,2009; Buri,Arlettaz,& Humbert,2013)andimprovesgrasslandheterogeneity,which is advantageous to insects (Cizek et al., 2012) and birds (Verhulst, Kleijn, Loonen, Berendse, & Smit, 2011). Low intensitygrazingalsocreatesfavourablehabitatforpollinator species(Kruess&Tscharntke,2002).

(8)

M.F . T anis et al. / Basic and Applied Ecology 43 (2020) 52–63 59

Table3. ThebestthreeGLMMmodels(AIC<2)explainingsummerpollinatorabundance.“Type”referstowhichgrasslandtypesweretested.“HRB”=herb-richgrasslands,“HM”=hay meadowsand“MBG”=meadowbirdgrasslands.Theestimates(exp=exponential)refertotherateratiosofanegativebinomialdistributedmodel.The95%confidenceintervalofthelog rateratiosarepresentedforeachmodelcovariate.

Model1 Model2 Model3

Explanatoryvariables Est. Est.(exp) 2.5% 97.5% Est. Est.(exp) 2.5% 97.5% Est. Est.(exp) 2.5% 97.5%

(9)

Fig.4. Pollinatorcommunitystructureindifferentgrasslandmanagementtypes.(A)Venndiagramoftheoverlapinpollinatorspecies.(B) Non-metricmultidimensionalscalingoutputforpollinatorspeciescommunity.

continuous forage in close proximity to their nests (Potts etal.,2010).Managementofherb-richgrasslandsandhay meadows,ontheotherhand,involvespartialmowingor graz-ing(see AppendixB),whichleaves partof the vegetation standingandthussupportspollinatorpopulations through-outtheyear.Thisdifferenceinmowingpracticeisthemost likelyexplanationforthedifferencesinpollinatorsbetween meadowbirdgrasslandsandtheothertwotypes.

Lack

of

ﬂoral

resources

in

meadow

bird

grasslands

limits

pollinators

Flowerabundanceand diversityare majorfactors shap-ingpollinatorcommunities(Biesmeijeretal.,2006;Scheper etal.,2014).Inourstudy,flowerabundancehadapositive effectonpollinatorabundanceandflowerdiversityon pol-linatorspeciesrichness.Ofallthethreemanagementtypes, meadowbird grasslandshavethelowestflowerabundance

andrichness,possiblyexplainingwhymeadowbird grass-landsarepoorinpollinators.

(10)

polli-M.F.Tanisetal./BasicandAppliedEcology43(2020)52–63 61 natorabundanceinhaymeadowsandherb-richgrasslands,

despitehaymeadowshavingonaveragemoreflowers.

A

landscape

suitable

for

pollinators?

Anotherresultofthisresearchistheapparentlackofbee species on all three types of grasslands. This is counter-intuitive, as pollinator diversitycan be very high inopen (semi-)natural grasslands (Öckinger & Smith, 2007) and manybeespeciesusedtothriveinthesesameregions(Peeters etal., 2012). However, the sampledgrasslands tend tobe partofopen,monotonouslandscapes,becauseoftheoverall intensificationandhomogenizationofgrasslandscombined withthe focus of nature management mostly on meadow birds,which favour breeding inopen landscapes to avoid predators. They lack the small landscape elements, trees, buildings and dykes, (see Appendix N)that bees need as nesting sites andrefuges, as the meadows themselves are oftentoo wetfor nesting.Wild bees neednesting sites in closeproximitytotheirfeedingsites,duetotheirrelatively small forage range (Zurbuchen et al., 2010). Notably, the onlyspecieswhichwasfoundregularly,Macropiseuropaea, isabletonestinwetsoil(Peetersetal.,2012).The combi-nationoflarge-scalemowingandlandscapehomogenization hasledtolowandintermittentforageresources,low nest-ingavailabilitywithaseverelyimpoverishedbeecommunity as aresult (only14 of 51possible beespecies; Appendix O).

Theapparentlackofbeespeciesisinstarkcontrastwith thehoverflyspeciesfound(43of66possiblespecies).The reasonmaybethatallflyspeciesarehighlymobileandthat manyhaveaquaticlarvae,andcanreproduceintheditches andwetlands inthe area. The same goes for 8 of the 12 butterflyspeciesfound,whicharemigratoryspeciesunlikely tobeboundtothegrasslands.Incontrast,specieslikelytobe affectedbymowing,suchasstationarybutterflies(4species, outof8possiblespecies)andphytophagoushoverflyspecies (3species)wererarelyfound.

The pollinator fauna on all three types of grasslands consists almost completely of common species; only two recorded species could be considered of conservation priority: Bombus jonellus (Reemer, 2018) and Neoascia geniculata (Reemer et al., 2009). This indicates that the species of conservation interest have disappeared com-pletely,eventhoughtheycouldstillbefoundintheseareas a few decades ago (Peeters et al., 2012; Reemer et al., 2009).

A

place

for

pollinators

in

(meadow

bird)

grassland

reserves

in

the

Netherlands?

TheGrasslandnaturereservesmayfailtoprovideessential habitatrequirementsforbeesandanumberofhoverflyand butterflyspecies,withmeadowbirdgrasslandsbeing partic-ularlypoorhabitatfor bees.Currently aboutathirdof the

naturereservesintheGroeneHartandMidden-Delflandare managedasmeadowbirdgrassland(AppB.).Thisfocuson afewspeciesofmeadowbirdshasclearlybeendetrimental topollinatinginsects,which,undercurrentregimes,mustbe conservedelsewhere.Thismay,however,bedetrimentalto themeadowbirdsthemselves,asavariedpollinator commu-nitycouldalsobenefitthem.Manyfloweringplantsdepend oninsectsfortheirpollinationandseedset(Ollertonetal., 2011),andriskpollinationfailureifpollinatorsarelacking (Wilcock&Neiland,2002).Therefore,avariedand abun-dantcommunityofpollinatorsisimportantformaintaining flower-richgrasslands,whicharerichininsects, an impor-tant foodsource formeadow birds(Beintemaetal., 1991; Vickeryetal.,2001).

Ourresultsalsoemphasizetherelativeimportanceofhay meadowsandherb-richgrasslands.Herb-richgrasslandsare oftenseenasanintermediatestagetowardsanothertypeof ‘nature’.Nonetheless,pollinatorsaremostabundantinthese grasslands, likely due to the presence of Cirsium thistles. Haymeadows,ontheotherhand,harbourdiverseplant com-munitiesonwhichthemostdiversepollinatorcommunities are maintained,underliningtheir importancefor pollinator conservation.

In summary, ourfindings highlight theneed for amore balanced approach for the conservation of wet grasslands of the Netherlands andtheir fauna,as theyindicate that a singlefocusonmeadowbirdsisunfavourableforbees, but-terfliesandhoverflies.Animprovementcouldbeto:include alargerproportionofhaymeadowsinthelandscape, imple-ment rotationalmowingandincrease heterogeneity within andbetweenparcels.Thiscouldleadtoafriendlierlandscape forbothbirdsandpollinators,whichwouldbebroadly sup-portedbysocietalpartnersintheNetherlands(seeDeltaplan BiodiversityRecovery,2018).

Acknowledgements

WethankMaartenvan‘tZelfdeforthemap,NiekeKnoben forgeneralassistance,VincentKalkmanforhelpingdevelop theideaandJohnSmitforspecimenidentification.Thanks also to NinoukVermeer, Sietse Kleinjan, ReneGarskamp (Zuid-Hollands Landschap), Martijn van Schie, Guus van Oostwaard and Gerard Haak (Natuurmonumenten), Cor Noorman (Groenzoom), Nico de Bruyn and Casper Zuy-derduin (Staatsbosbeheer) for access tothe sitesandtheir assistance, andtoLucasAlferink, Dimmyvan Ruitenand HansvanderKooijforfieldassistance.

Appendix

A-P

Supplementary

data

(11)

References

Akaike,H.(1987).FactoranalysisandAIC.InSelectedpapersof HirotuguAkaike.pp.371–386.NewYork,NY:Springer. Ausden,M.,&Hirons,G.J.M.(2002).Grasslandnaturereserves

forbreedingwadingbirdsinEnglandandtheimplicationsfor the ESA agri-environment scheme. Biological Conservation, 106(2),279–291.

Barton,K.(2016).MuMIn:Multi-modelinference.Rpackage ver-sion1.15.6.

Batáry,P.,Baldi,A.,Sarospataki,M.,Kohler,F.,Verhulst,J.,Knop, E.,...&Kleijn,D.(2010).Effectofconservationmanagement onbees andinsect-pollinatedgrasslandplantcommunities in threeEuropeancountries.Agriculture,Ecosystems& Environ-ment,136(1–2),35–39.

Batáry,P.,Dicks,L.V.,Kleijn,D.,&Sutherland,W.J.(2015).The roleofagri-environmentschemesinconservationand environ-mentalmanagement.ConservationBiology,29(4),1006–1016. Benton, T.G., Bryant, D.M.,Cole,L., & Crick,H. Q.(2002). Linkingagriculturalpracticetoinsectandbirdpopulations:A historicalstudyoverthreedecades.JournalofAppliedEcology, 39(4),673–687.

Benton,T.G.,Vickery,J.A.,& Wilson,J.D.(2003).Farmland biodiversity:ishabitatheterogeneitythekey?TrendsinEcology &Evolution,18(4),182–188.

Beintema,A.J.,Thissen,J.B.,Tensen,D.,&Visser,G.H.(1991). Feedingecologyofcharadriiformchicksinagricultural grass-land.InBreedingecologyofmeadowbirds(Charadriiformes); Implicationsforconservationandmanagement.,85.

Biesmeijer, J. C., Roberts, S. P., Reemer, M., Ohlemüller, R., Edwards, M., Peeters, T., ... & Settele, J. (2006). Parallel declinesinpollinatorsandinsect-pollinatedplantsinBritainand theNetherlands.Science,313(5785),351–354.

Buri, P.,Arlettaz,R.,& Humbert, J.Y. (2013). Delaying mow-ingandleavinguncutrefugesboostsorthopteransinextensively managedmeadows:Evidencedrawnfromfield-scale experimen-tation.Agriculture,Ecosystems&Environment,181,22–30. Chamberlain,D.E.,Fuller,R.J.,Bunce,R.G.H.,Duckworth,J.

C.,&Shrubb,M.(2000).Changesintheabundanceoffarmland birdsinrelationtothetimingofagriculturalintensificationin EnglandandWales.JournalofAppliedEcology,37(5),771–788. Cizek,O.,Zamecnik,J.,Tropek,R.,Kocarek,P.,&Konvicka,M. (2012).Diversificationofmowingregimeincreasesarthropods diversityinspecies-poorculturalhaymeadows.JournalofInsect Conservation,16(2),215–226.

Deltaplan Biodiversity Recovery, 2018. https://www. samenvoorbiodiversiteit.nl.

deVries,M.W.,&Knotters,J.C.(2000).Effectenvangefaseerd maaibeheeropdeongewerveldefaunavangraslanden.De Lev-endeNatuur,101(2),37–41.

Donald,P.F.,Green,R.E.,&Heath, M.F.(2001).Agricultural intensificationandthecollapseofEurope’sfarmlandbird popu-lations.ProceedingsoftheRoyalSocietyofLondonB:Biological Sciences,268(1462),25–29.

Falk,S.,&Lewington,R.(2017).VeldgidsBijenvoorNederland enVlaanderen.Utrecht/Antwerpen:KosmosUitgevers. Falk, S. (24-10-2018). retrieved on 24-10-2018, from Flickr:

https://www.flickr.com/photos/63075200@N07/albums/with/ 72157629248800794.

Garibaldi,L.A.,Steffan-Dewenter,I.,Winfree,R.,Aizen,M.A., Bommarco,R.,Cunningham,S.A.,...&Bartomeus,I.(2013). Wildpollinatorsenhancefruitsetofcropsregardlessofhoney beeabundance.Science,339(6127),1608–1611.

Gossner,M.M.,Lewinsohn,T.M.,Kahl,T.,Grassein,F.,Boch, S.,Prati,D.,...&Arndt,H.(2016).Land-useintensification causesmultitrophichomogenizationofgrasslandcommunities. Nature,540(7632),266.

Goulson,D.,Hanley,M.E.,Darvill,B.,Ellis,J.S.,&Knight,M.E. (2005).Causesofrarityinbumblebees.BiologicalConservation, 122(1),1–8.

Goulson,D.,Lye,G.C.,&Darvill,B.(2008).Declineand con-servationof bumblebees.AnnualReviewof Entomology,53, 191–208.

Goulson,D.,Nicholls,E.,Botías,C.,&Rotheray,E.L.(2015).Bee declinesdrivenbycombinedstressfromparasites, pesticides, andlackofflowers.Science,347(6229),1255957

Hallmann,C.A.,Sorg,M.,Jongejans,E.,Siepel,H.,Hofland,N., Schwan,H.,...&Goulson,D.(2017).Morethan75percent declineover27yearsintotalflyinginsectbiomassinprotected areas.PLoSOne,12(10),e0185809.

Humbert,J.Y.,Ghazoul,J.,&Walter,T.(2009).Meadow harvest-ing techniquesand their impactson field fauna.Agriculture, Ecosystems&Environment,130(1–2),1–8.

Inter Provinciaal Overleg. (2018). Informatiemodel natu-urbeheer (IMNAB): Index Natuur en Landschap. https://www.bij12.nl/onderwerpen/natuur-en-landschap

Kleijn,D.,Baquero,R.A.,Clough,Y.,Diaz,M.,DeEsteban,J., Fernández,F.,...&Knop,E.(2006).Mixedbiodiversity ben-efits ofagri-environmentschemesinfiveEuropeancountries. EcologyLetters,9(3),243–254.

Kleijn,D.,Rundlöf,M.,Scheper,J.,Smith,H.G.,&Tscharntke, T.(2011).Doesconservationonfarmlandcontributetohalting thebiodiversitydecline?TrendsinEcology&Evolution,26(9), 474–481.

Klein, A.M.,Vaissiere,B.E.,Cane,J.H.,Steffan-Dewenter,I., Cunningham,S.A.,Kremen,C.,...&Tscharntke,T.(2007). Importance of pollinators in changing landscapes for world crops.ProceedingsoftheRoyalSocietyofLondonB:Biological Sciences,274(1608),303–313.

Kosior,A.,Celary, W.,Olejniczak,P.,Fijał,J.,Król,W.,Solarz, W.,...&Płonka,P.(2007).Thedeclineofthebumblebeesand cuckoobees(Hymenoptera:Apidae:Bombini)ofWesternand CentralEurope.Oryx,41(1),79–88.

Krebs,J.R.,Wilson,J.D.,Bradbury,R.B.,&Siriwardena,G.M. (1999).Thesecondsilentspring?Nature,400(6745),611. Kruess,A.,&Tscharntke,T.(2002).Grazingintensityandthe

diver-sityofgrasshoppers,butterflies,andtrap-nestingbeesandwasps. ConservationBiology,16(6),1570–1580.

Matson,P.A.,Parton,W.J.,Power,A.G.,&Swift,M.J.(1997). Agriculturalintensificationandecosystemproperties.Science, 277(5325),504–509.

Morris,M.G.(2000).Theeffectsofstructureanditsdynamicson theecologyandconservationofarthropodsinBritishgrasslands. BiologicalConservation,95(2),129–142.

Öckinger,E.,&Smith,H.G.(2007).Semi-naturalgrasslandsas populationsourcesforpollinatinginsectsinagricultural land-scapes.JournalofAppliedEcology,44(1),50–59.

(12)

Osborne,J.L.,Williams,I.H.,&Corbet,S.A.(1991).Bees, pollina-tionandhabitatchangeintheEuropeancommunity.BeeWorld, 72(3),99–116.

Peeters, T.M.,Nieuwenhuijsen, H., Smit,J., Vander Meer, F., Raemakers,I.P.,Heitmans,W.R.B.,...&Roos,M.(2012).

DeNederlandsebijen(Hymenoptera:ApidaeS.L.)Nederlandse Fauna11.NaturalisBiodiversityCenter&European Inverte-brateSurvey.

Potts,S.G.,Biesmeijer,J.C.,Kremen,C.,Neumann,P.,Schweiger, O.,&Kunin,W.E.(2010).Globalpollinatordeclines:Trends, impacts and drivers. Trends in Ecology & Evolution, 25(6), 345–353.

Potts, S.G., Imperatriz-Fonseca,V., Ngo,H. T., Aizen, M. A., Biesmeijer,J.C.,Breeze,T.D.,...&Vanbergen,A.J.(2016). Safeguardingpollinatorsandtheirvaluestohumanwell-being. Nature,540(7632),220.

Reemer,M.,Renema,W.,VanSteenis,W.,Zeegers,T.,Barendregt, A.,Smit,J.T.,...&VanderLeij,L.J.J.M.(2009).De Neder-landseZweefvliegen(Diptera:Syrphidae).NederlandseFauna 8.NationaalNatuurhistorischMuseumNaturalis& European InvertebrateSurvey.

Reemer,M.(2014).Veldtabelwapen-enbastvliegenvanNederland. Leiden:EISKenniscentrumInsecten.

Reemer,M.(2018).BasisrapportvoordeRodeLijstBijen.Leiden: EISKenniscentrumInsecten.

Scheper,J.,Reemer,M.,vanKats,R.,Ozinga,W.A.,vander Lin-den,G.T.,Schaminée,J.H.,...&Kleijn,D.(2014).Museum specimensreveallossofpollenhostplantsaskeyfactordriving wildbeedeclineinTheNetherlands.ProceedingsoftheNational AcademyofSciences,111(49),17552–17557.

Schulten, A. (2018). Zweefvliegen van Nederland en België. ’s Gravenland:Jeugdbondsuitgeverij.

Shannon,C.E.(1948).Amathematicaltheoryofcommunication. BellSystemTechnicalJournal,27(3),379–423.

Thomas,J.A.,Telfer,M.G.,Roy,D.B.,Preston,C.D.,Greenwood, J.J.D.,Asher,J.,...&Lawton,J.H.(2004).Comparativelosses ofBritishbutterflies,birds,andplantsandtheglobalextinction crisis.Science,303(5665),1879–1881.

Tscharntke, T., Klein, A. M., Kruess, A., Steffan-Dewenter, I., & Thies, C. (2005). Landscape perspectives on agricultural

intensificationandbiodiversity–ecosystemservicemanagement. EcologyLetters,8(8),857–874.

vanEgmond,P.M.,&deKoeijer,T.J.(2006).Weidevogelbeheer bijagrariersenterreinbeheerders.DeLevendeNatuur,107(3), 118–120.

VanSwaay,C.,VanStrien,A.,Aghababyan,K.,Astrom,S.,Botham, M., Brereton, T., ...& Feldmann, R. (2015). The european butterﬂyindicatorforgrasslandspecies:1990–2013.

VanSwaay, C. A.M.,Bos-Groenendijk,G. I.,van Deijk,J.R., vanGrunsven,R.H.A.,Kok,J.M.,Huskens,K.,...&Poot, M.(2018).Handleidinglandelijkemeetnettenvlinders,libellen en nachtvlinders. Rapport VS2018.011. De Vlinderstichting, Wageningen.

VanVeen,M.P.(2004).HoverﬂiesofNorthwestEurope: Identiﬁ-cationkeystotheSyrphidae.Utrecht:KNNVPublishing. Vickery,J.A.,Tallowin,J.R.,Feber,R.E.,Asteraki,E.J.,Atkinson,

P.W.,Fuller,R.J.,...&Brown,V.K.(2001).Themanagement oflowlandneutralgrasslandsinBritain:Effectsofagricultural practicesonbirdsandtheirfoodresources.JournalofApplied Ecology,38(3),647–664.

Vray,S.,Lecocq,T.,Roberts,S.P.,&Rasmont,P.(2017). Endan-gered by laws:Potential consequencesof regulationsagainst thistlesonbumblebeeconservation.AnnalesdelaSociété ento-mologiquedeFrance(NS),53(1),33–41.Taylor&Francis. Verhulst, J., Kleijn, D., Loonen, W., Berendse, F., & Smit, C.

(2011).Seasonaldistributionofmeadowbirdsinrelationto in-fieldheterogeneityandmanagement.Agriculture,Ecosystems& Environment,142(3–4),161–166.

Wilcock, C.,&Neiland,R. (2002).Pollinationfailurein plants: Whyithappensandwhenitmatters.TrendsinPlantScience, 7(6),270–277.

Winfree,R.,Reilly,J.R.,Bartomeus,I.,Cariveau,D.P.,Williams,N. M.,&Gibbs,J.(2018).Speciesturnoverpromotesthe impor-tance of beediversityforcroppollination atregionalscales. Science,359(6377),791–793.

Zurbuchen,A.,Landert,L.,Klaiber,J.,Müller,A.,Hein,S.,&Dorn, S.(2010).Maximumforagingrangesinsolitarybees:Onlyfew individualshavethecapabilitytocoverlongforagingdistances. BiologicalConservation,143(3),669–676.

Availableonlineatwww.sciencedirect.com