the Habitats grid: The on spatial dimension does matter for red-listing Journal for Nature Conservation

(1)

ContentslistsavailableatScienceDirect

Journal

for

Nature

Conservation

jou rn a l h o m e p a g e :w w w . e l s e v i e r . d e / j n c

Habitats

on

the

grid:

The

spatial

dimension

does

matter

for

red-listing

Daniela

Gigante

a,∗

_,

_Bruno

_Foggi

b

_,

_Roberto

_Venanzoni

a

_,

_Daniele

_Viciani

b

_,

_Gabriella

_Buffa

c

a_Department_of_Chemistry,_Biology_and_{Biotechnology,}_University_of_Perugia,_Borgo_XX_giugno,_74,_I-06121_Perugia,_Italy b_Department_of_Biology,_University_of_Florence,_Via_G._La_Pira_4,_I-50121_Florence,_Italy

c_Department_of_{Environmental}_Science,_Informatics_and_Statistics,_University_Ca’_Foscari_of_Venice,_Via_Torino_155,_I-30172_Venezia_Mestre,_Italy

a

r

t

i

c

l

e

i

n

f

o

Articlehistory:

Received19December2015 Receivedinrevisedform2March2016 Accepted29March2016

Keywords: Areaofoccupancy Emergentproperties Patternofspatialoccupancy Plantcommunity Spatialscale Threatassessment

a

b

s

t

r

a

c

t

BesidesspeciesRedLists,recently,avarietyofframeworkshavebeenproposedforassessinghigher levelsofbiologicalorganisation,i.e.ecosystems,habitats,plantcommunities.Mostoftheseprotocols referto‘plantspeciesassemblages’or‘vegetationtypes’asproxiesforecosystemsorhabitats.Indeed,the habitatconceptbasedonplantcommunitieshasacquiredacentralroleasakeyapproachforbiodiversity conservationabovethespecieslevel.Plantcommunities,likeeverycomplexbiologicalsystem,hold scale-dependent‘emergent’propertieswhichvaryasafunctionofthescaleofobservation.Withreferenceto red-listing,thesescale-dependentpropertieshavefar-reachingconsequencesforbothidentificationand classification,aswellasforrepresentationandevaluation,andbecomeparticularlychallengingwhen dealingwithcriteriaregardingdeclineindistributionorrestricteddistribution.Therecentdiscussion onthered-listingprotocolshasevidencedseveralaspectsthatclaimspecialeffortsforasuitableuse. Inthepresentpaper,startingwiththeanalysisofsomerecentlyproposedprotocolsforthered-listing ofhabitatsandecosystems,wediscussandtestsome‘emergent’propertiesofspeciesassemblages, providingcuesforreflection.Basedonavarietyoftheoreticalmodelsandscientificoutcomesinliterature fromthelastdecades,wetheorisethatplantcommunitiesownsomeintrinsic,ecologicallybasedand scale-dependentspatialfeatures,whichgiverisetodifferenttypesofpatternofspatialoccupancy.We discussamodelwhere,innaturalconditions,thepossiblepatternsofspatialoccupancyarereferredto 3basictypes:areal,linearandpoint.Thisapproachishereproposedasatooltodiscriminateamong differentbroadcategoriesofplantcommunity-basedhabitattypesandoptimisetheirassessmentin thered-listingprocess.Startingfromahomogeneousdataset,theproposedcasestudiesprovethatthe choiceofthescaleaffectsthecomprehensionofthehabitats’occurrence,withasubstantialrelapseon theestimatesoftheirdistributionsize.Inparticular,habitatswithlinearandpointdistribution,often naturallysmallinsizeanddispersed,aremoresusceptibletobiasedevaluationoftheiractualdistribution andconsequentlyoftheirthreatstatus.Theintrinsicspatialattributesofplantcommunitiesshouldnot beneglectedinared-listingprocessandclaimfora‘habitat-tailored’approach.Theuseofdifferent grid-cellsizesandthresholdsforthethreemainpatternsofspatialoccupancyhereproposed,mightcertainly avoidinaccuratestatements.

1. Introduction

Theconservationofbiodiversityisawidelyacknowledged

tar-get for humankind (Cafaro & Primack,2014), as shown by an

increasingarrayofregional,nationalandinternationalagreements

and frameworksfor hindering biodiversity loss(see, e.g.CITES,

1973;EuropeanCommission,2011;UnitedNations,1976,1992).

Themostchallengingaspectstemsfromthecomplexityof

biodi-versity,thatinitselfcomprisesbothmultiplelevelsoforganisation

∗ Correspondingauthor.

E-mailaddress:daniela.gigante@unipg.it(D.Gigante).

(e.g.genes,species,communitiesandecosystems)andtheir

inter-actingrelationshipsatalltheintegrationlevels(Allen&Starr,1982;

Margules&Pressey,2000;Noss,1990).

Inthelastdecades,theInternationalUnionforConservationof

Nature(IUCN)hasledthedevelopmentofquantitativecriteriafor

thecreationofredlistsofthreatenedspeciesthatallowfor

trans-parent, objective and repeatable risk assessments (IUCN, 2012,

2013;Maceetal.,2008).Byrankingspeciesatriskofextinction,

theIUCNRedListsprovideaglobalindicationonthestateofone

levelofbiodiversityandmakegovernmentsandsocietyawareof

thetrendsinextinctionrisk(Baillie,Hilton-Taylor,&Stuart,2004;

Butchartetal.,2004;McCarthy,Thompson,&Garnett,2008).

http://dx.doi.org/10.1016/j.jnc.2016.03.007

(2)

DespitethecuttingedgeimportanceofspeciesRedLists,the

realisationthatanapproachfocusedexclusivelyonthespecieslevel

isunﬁttoconserveallcomponentsofbiodiversityledthe

scien-tiﬁccommunity,conservationprofessionalsandinstitutionstobe

increasinglyconcernedwithbiodiversityassessments,addressing

higherlevelsofbiological organisation(Izco,2015;Keith,2009;

Keithet al.,2013,2015; Kontula&Raunio,2009; IUCN,2015a;

Nicholson,Keith,&Wilcove,2009;Rodríguezetal.,2011,2012,

2015).Ecologicalcommunitiesmaymoreefﬁcientlyrepresentthe

biological diversity as a whole, compared to the species-level

approach,whichoftenlacksdirectinformationabout

fundamen-talabioticcomponents,thusmissingboththetargetsofprotecting

ecologicalpatternsandprocesses,andensuringthepersistenceof

ecosystemfunctionsandstructure(Balmfordetal.,2002;Cowling

etal.,2004;MillenniumEcosystemAssessment,2005;Noss,1996; SecretariatoftheConventiononBiologicalDiversity,2010).

Fur-thermore,theconservationofcommunitiesorecosystemscanalso

actasasurrogateforthespecies,particularlyforthosespeciesyet

undescribedorpoorlyknown(Cowling&Heijnis,2001;Nicholson

etal.,2009),thusprovidingapreciousservicewhenconsidering

that,despitestrenuousefforts,onlylessthan5%oftheestimated

numberofdescribedspecies(lessthan7%whenconsideringonly

plants)hasbeenevaluatedforinclusionintheIUCNRedListby

2015(IUCN,2015b).Theassessmentofcommunitiesorecosystems

alsoallowstoincorporatefurtherinformation,suchastheroleof

speciesrichness/diversity,offeringprecioustoolsbothforspecies

andhabitat’sprioritization(see,e.g.,Bergetal.,2014;Lindenmayer

etal.,2008;Pärteletal.,2005).

Early lists of endangered plant

communi-ties/habitats/ecosystems have been compiled since the 1980s,

focusingeitheronspeciﬁcecosystemtypesoronnational

bound-aries(seee.g.,Moravecetal.,1983;Schulte&Wolf-Straub,1986).

Manycommoncriteriawereusedfortheevaluation(e.g.rarity,

range,speciescomposition,naturalness,humanpressure,aesthetic

oreducationalvalue), buttheassessment wasmostlybasedon

expertknowledge,e.g.long-termﬁeldexperience(Blab,Riecken,

&Ssymank,1995;Noss,LaRoe,&Scott,1995;Paal,1998)and,in

somecases,keyconceptswerenotunderpinnedbysound

theoret-icalbackgrounds.Morerecently,avarietyofframeworks,founded

onrelevantecological theories,have beenproposedfor

assess-ingthe threat status of plantcommunities/habitats/ecosystems

(Bergetal.,2014;Biserkov,2011;Essl,Egger,&Ellmauer,2002;

Lindgaard&Henriksen,2011;Walkeretal.,2006),insomecases

promptedbygovernmentagencies.Nevertheless,differentlyfrom

species,asyetthereisnoacknowledgedinternational

methodol-ogyonwhichtobasethehabitatred-listing.InEurope,Rodwell,

Janssen,Gubbay,andSchaminée(2013)recentlystartedaproject

aimedatdevelopingaRedListoftheEuropeanhabitattypes,while

theIUCNcouncil(CEM-IUCN&Provita,2012)formallyendorsed

theprotocolproposedbyKeithetal.(2013).

Besides a common origin from the protocol developed for

species(IUCN,2013;Maceetal.,2008),thevariousmethodologies

adopted for plant communities/habitats/ecosystems red-listing

shareseveralcharacteristics:(a)althoughsomeprotocolsdeﬁne

assessmentunitsoflargedimensions,suchas‘ecosystems’(Keith

etal.,2013)or‘Land Environments’ (Walkeretal., 2006), most

ofthemreferto‘plantspeciesassemblages’or‘vegetationtypes’,

usedasproxiesforecosystemsorhabitats;(b)asarule,the

assess-mentisbasedonquantitativecriteriaandonlyfewprotocolsrely

onqualitativeones,i.e.onthebestexpertisesupportedby

spe-ciﬁcparadigms(e.g.Biserkov,2011);(c)alltheprotocolsinclude

declineandrestrictedsizeinspatialdistributionaskeycriteria,

andalmost allidentifyquantitativethresholds,oftenanalogous

yetless severethanthose adoptedforspecies;(d)theyarestill

ratherlackinginsuitabletoolstoincorporatemeasuresof

‘ecologi-calfunction’,i.e.thecapacityofcommunitiestosupporttheirwhole

diversityofspeciesandtosustaintheirfunctionalrolesin

land-scapes(Nicholsonetal.,2009),withsomeremarkableexceptions

givingspecialimportancetospeciesdiversityortothepresenceof

threatenedtaxaintheevaluationofhabitatquality(e.g.Andreas

&Lichvar,1995;Bacchetta,Farris,&Pontecorvo,2012;Bergetal., 2014;Gauthier,Debussche,&Thompson,2010).

1.1. Emergentproperties:the“patternofspatialoccupancy”

Plantcommunities,likeeverycomplexbiologicalsystem,hold

aggregateor‘emergent’properties,whichcausethewholetobe

morethanthemeresumofitsparts(Bissonette,1997;Halley&

Winkler,2008;vanderMaarel&Franklin,2013).Innatural

con-ditions,propertiesofplantcommunitiessuchascompositionand

structurearisefromtheinteractionofbothcoarse-andﬁne-scale

ﬁlters(Dale,1999;Lortieetal.,2004).Onaﬁnescale,vegetation

patternsareruledbyspeciessizeandgrowthpattern,aswellas

bytheinteractionsamongplantindividuals.Onalargerscale,they

areinﬂuencedbyphysicalandgeomorphologicfeatures(i.e.

val-leys,ridges,slopes,waterbodies)thatcreatespatialandecological

heterogeneity(Dale,1999;Greig-Smith,1979;Palmer,1988).

By interacting withspatially distributed environmental

gra-dients,organisms,communitiesandecologicalsystemsarethus

arrayedinspacetoformdistinctpatternsorconﬁgurations,i.e.

‘spe-ciﬁcarrangementofspatialelements’(Turner,Gardner,&O’Neill,

2001),thatexhibitacertainamountofpredictability(Dale,1999).

Thus,althoughtheboundariesbetweendifferentplant

commu-nitiesareinherentlymoreuncertainthanisthecaseforspecies

(Nicholsonetal.,2009),therecognitionanddelimitationofstands

ofvegetationintheﬁeldcanbebasedoninternalcharacteristics,

e.g.structural,physiognomicandﬂoristicuniformity,andexternal

ones,e.g.discontinuitywiththesurroundingvegetation(vander

Maarel&Franklin,2013).

Onthisground,itcanbeassumedthateachplantcommunity

ownsintrinsicspatialfeatures,ecologicallyfounded,whichaffect

itsspatialdistributioninnaturalconditions.Inparticular,the

envi-ronmentalheterogeneityaccountsforanintrinsicpropertyofeach

plantcommunitythatwecall‘patternofspatialoccupancy’(PSO).

InaccordancewithDale(1999),werefertospatialpatternas

‘nonrandomnessinspatialarrangement,whichthenpermits

pre-diction’.Theabioticenvironmentisspatiallystructured,resulting

inpatchypatternsorgradients.Innaturalconditions,whenthe

plantspeciesand assemblagesarenotsubjectedtohuman

con-straints,plantcommunitiescan displayseveral,yet predictable,

PSOs.Dale(1999) proposedtheconceptsof‘point pattern’ and

‘patternonanenvironmentalgradient’withreferencetothetype

of representation of plant individuals and communities

distri-bution.Withdifferentalthoughcloselyrelatedaims, theFrench

phytosociological-synusialschooldevelopedageometricapproach

to plant communities, introducing the shape (‘forme spatiale,

linéairecontinue,linéairediscontinue,ponctuelle’)asanessential

propertyofvegetationpatchesinlandscapeanalysis(Géhu,1974,

1991;Julve,1986).

Asageneralmodel,weassumethatthePSOmostlytendsto

displaythreemainpatterns:areal(i.e.withanextended

distribu-tion;e.g.broadleavedtemperateforests,naturalandsemi-natural

grassland formations), linear (i.e. with a distribution in strips,

wherelengthismuchgreaterthanwidth;e.g.riparianand

water-dependant formations,coastal plantcommunities)or point(i.e.

witha naturallyscatteredspatialdistribution,e.g.vegetationof

temporaryponds).

Byreﬂectingtheecologicaldrivingforces,these3modelsofPSO

arerepresentativeofnaturalconditionsandassuchtheyshouldbe

consideredasanintrinsicfeatureofeachplantcommunity(namely

theydonotgiveaccountofartiﬁcial,human-induceddistribution,

(3)

Propertiesofplantcommunities,suchascompositionand

struc-tureaswellastheirspatialpattern,areinherentlyscale-dependent

concepts(Gray, Phan,&Long, 2010; Turner,O’Neill,Gardner,&

Milne,1989;Wiens,1989,1990).Scaledependencecanbedeﬁned

asthedegreetowhichecologicalphenomenavaryasafunction

ofgrainandextent,thetwo maincomponentsofscale(Palmer

&White,1994).Sincedifferentpatternsmayemergeatdiffering

scalesofinvestigation,itiscrucialthatresolution(i.e.grainsize)

andextentbedeﬁnedtoaccuratelyrepresenttheecological

pro-cessortheobjectunderstudy.Differently,thedetectedpatterns

willhavelittlesigniﬁcanceandwillleadtoinaccurateconclusions.

Inared-listingprocedure,thisissuebecomesparticularly

chal-lengingwhendealingwithcriteriaregardingdecliningorrestricted

distribution,i.e.traitsbasedonspatialaspectsoftheunitunder

assessment.Forinstance,oneoftheworldwideacceptedmetrics

usedtoevaluatethereductioningeographicdistributionofany

assessmentunitisthe‘Areaofoccupancy’(AOO),correspondingto

theareaofsuitablehabitatactuallyoccupiedbytheassessment

unit,usually calculatedbycrossing theoccurrenceswithagrid

andsummingtheareas(orcountingthenumber)oftheoccupied

cells(IUCN,2012,2013;CEM-IUCN&Provita,2012).Assuch,AOO

dependsonthecell’ssizeandisafunctionofthescaleatwhichitis

measured.Consequently,assessmentsaresubjectedtosigniﬁcant

distortionifthedistributionisassessedatdifferentspatialscales.

Giventhegeneralavailabilityofspatialdata,criteriabasedon

dis-tributionwilllikely bethemostfrequentlyused(Boitani,Mace,

&Rondinini,2015).Inthissense,thedevelopmentofconsistent

methodsrequiresacarefulinvestigationoftheimpactofthescale

ontheassessmentoutcomes.

Aimofthepresentpaperistoshowthatthespatialpatternof

plantcommunities,andinparticulartheirPSO,mightplayarolein

ared-listingprocedurewhenapplyingquantitativecriteria,

specif-icallythoserelatingtospatialoccurrence.Inparticular,theuseof

the‘Areaofoccupancy’(AOO)mightbringtoinconsistent

assess-mentsacrosshabitattypeswithdifferentPSO.Ourﬁrsthypothesis

isthat,asadirectconsequenceofthespatialdistributionof

ecolog-ical,physiognomicandstructuralvariables,thePSOsigniﬁcantly

affectsthehabitat’sassessment.Oursecondhypothesisisthatthe

useofaﬁxedgrid-cellsizeforhabitatswithdifferentPSOisnot

appropriateandleadstobiasedassessment,evenmoredistorted

forlinearandpointPSO.

2. Materialsandmethods

ToshowhowthePSOmightaffecttheAOOcalculationand,

con-sequently,theapplicationofthered-listingcriteria,wetookinto

accountfourcasestudies.

Asassessmentunitsweconsideredthehabitatsasdeﬁnedbythe

European‘Habitat’Directive92/43/EEC(HD;seee.g.Evans&Arvela,

2011)andlistedinitsAnnexI.ThedescriptionofHDHabitats(from

hereoncapitalised,toavoidanymisunderstandingwithother

con-ceptsofhabitat),isfoundedontherelatedplantcommunities.They

areusuallyidentiﬁed,andevenmapped,basedona

phytosociologi-calapproach(Biondi,2011;Braun-Blanquet,1964;Dengler,Chytr ´y,

&Ewald,2008;Westhoff&vanderMaarel,1978),andaremostly

describedatthelevelofasyntaxonomicalliance(Biondietal.,2012;

Bunceetal.,2013;Evans,2006,2010;Rodwelletal.,2002).Being

deﬁnedonthegroundoftheirplantcommunities,HDHabitatscan

beassumedtoownthesameintrinsic,ecologicallybased,spatial

features.

Wetookintoaccount126AnnexIHabitatsoccurringinItaly

(Biondietal.,2009).WeusedthedatareportedbytheItalian

Min-istryoftheEnvironmentandProtectionofLandandSeaforthe

3rdreportex-Art.17(Genovesietal.,2014;MATTM,2013).More

detaileddata,derivedfromtheHabitatmapsincludedinNatura

2000SitesManagementPlansfromtheRegionsUmbria,Veneto

andTuscany(unpublisheddata),havebeenusedforasubsetof41

habitats.

Totestourhypotheses,AnnexIHabitatsweregroupedonthe

groundoftheirPSO.TheadoptedcriteriaforHabitatattribution

tooneofthethreePSO(point,linearorareal)tookintoaccount,

asmaindrivers,theirintrinsicecologicfeatures,suchasthe

pres-enceofastrongecologicalgradient,andtheenvironmentalspatial

heterogeneity.

WeconsideredaslineartheriparianandcoastalHabitats,plus

allthoseHabitatswhich,forintrinsicreasons,whenrepresented

onamapappearintheshapeofastretch(e.g.rockywalls,steep

slopes withchasmophytic vegetation). We considered as point

thoseHabitatswhich,forintrinsicreasons,whenrepresentedon

amapappearintheshapeofspots,e.g.,thetemporaryponds,the

inlandsaltmeadows,theshallowstandingoligotrophicwaters,all

characterisedbyarestricteddistributionduetotheirecological

peculiarity.TheHabitatswhosedistributiondidnotfallinthe

for-mertwotypeshavebeenreferredtothearealPSO.Thecomposition

ofthegroupsisreportedinTable1;forHabitatdescriptionssee

Biondietal.(2009,2012).

Totest ifdifferentHabitats,groupedaccordingtotheirPSO,

exhibitsigniﬁcantdifferencesintheaveragespatialextent,ﬁrstly

weconsideredatheoreticalcasestudy(a),representedbythree

HabitatswiththesameactualsurfacebutdifferenttypesofPSO,

andtheirinterrelationwitha10×10km2_grid,_on_whose_base_the

AOOhasbeencalculated.

Then,inthecasestudy(b),weanalysedtheaveragetotalsurface

perNatura2000siteoccupiedbyeachoftheconsidered126Annex

IHabitats,mappedandmeasuredallovertheItalianNatura2000

network(2585sites).

Incase study(c)dataontheactualsurface(km2₎_have _been

relatedtothecalculatedAOO(basedona10×10km2_grid),_with

theaimtocheckifthe3PSO-basedgroupsofHabitatsevidenced

any differentialresponse. In order to normalise data, we

com-paredtheratiobetweenthetotalgrid-basedAOOsurfaceandthe

totalactualsurface(km2₎_for_each_Habitat._This_metrics_gives_also

accountonhowlargeisthegapbetweenthetwoconsideredvalues.

Eventually,inthecasestudy(d),weappliedtheAOOcalculation

processtoarestrictednumber (41)of Habitatspresent inItaly,

forwhichdetailedcartographicdatawereavailable,withtheaim

todetectiftheAOOcalculationisaffectedbythePSOandbythe

adoptedgridcell-size.The41Habitattypesusedforcasestudy(d)

areincludedinthecompletelist(Table1),withtheindicationofthe

territorialscaletheyhavebeenconsideredat(national,regionalor

local).ForeachconsideredHabitattype,thedetailedcartographic

dataallowedtocalculatetheAOOsurfaceatfourdifferentgrid-cell

sizes(10×10km2_,₂_×₂_km2_,₁_×₁_km2_and_0.5_×_0.5_km2_).

Forallthecasestudies,thestatisticalsigniﬁcanceofthe

differ-encesamongtheconsideredgroupsofdatahasbeenanalysedby

applyingthenonparametricKruskal-WallisTest.Thecorrelation

analyseshavebeenbasedonthenon-parametricSpearman’srank

coefﬁcient.ThesoftwareAnalystSoftStatPlus:macv2009hasbeen

usedforstatisticalanalyses.Thecartographicdatahavebeen

pro-cessedbyusingtheOpenSourceGeographicInformationSystem

QuantumGISQGIS2.0.1(QGISProject,2014).

3. Results

3.1. Casestudy(a):asimpleexerciseof‘form’

The three theoretical Habitats sharing the same surface

(2092.5ha)anddifferinginthetypeofPSO(areal,linearandpoint),

arespatiallyrepresentedinFig.1.Theymightcorrespond,e.g.,toa

(4)

Fig.1. Exempliﬁcativerepresentationofthreetheoreticalhabitatswiththesameactualsurface(2,092.5ha)butdifferenttypesofPSO,andtheirinterrelationwitha 10×10km2_grid;_the_respective_calculated_values_of_the_AOO_surfaces_are_reported_(map_processed_by_using_QGIS_2.0.1).

gallery(92A0)andaMediterraneantemporarypondwithIsoëtes

hystrix(3170),respectively.ThecalculatedAOOsurface,basedona

10×10km2_grid,_resulted_respectively_in_400,₉₀₀_and₁₅₀₀_km2_.

3.2. Casestudy(b):rangesofsizeofHabitats:scaleandgrain

playarole

Resultsofthecomparisonamongtheaveragetotalsurfaceof

AnnexIHabitatsperNatura2000site,groupedaccordingtotheir

PSO,areshowninFig.2.ThemeanHabitatsurfacepersiteshoweda

decreasingtrendfromareal(311.8ha),tolinear(79.5ha),topoint

Habitats(28.0ha),resultinginstatisticallysigniﬁcantdifferences

(p<0.001)amongthethreePSOtypes,onawide(national)scale.

ThisindicatesthatthethreePSO-basedgroupsof Habitattypes

showdistinctranges ofsize andask forown peculiar scalesof

representation.

3.3. Casestudy(c):atheoreticalexerciseofformappliedtoactual

distributiondata

Resultsoftheanalysisontherelationbetweenthecalculated

AOOandtheactualsurfaceofeachHabitattypeareshowninFig.3.

BothHabitatswithapointandalinearPSOwereheavilyshifted

infavourofinﬂatedvaluesofthegrid-basedsurface(respectively:

R=0.952,p<0.001;R=0.759,p<0.001).Conversely,Habitatswith

anarealPSOshowedacleartendencytocorrelatewith

proportion-allylowervaluesoftheAOOsurface(R=0.807,p<0.001).

3.4. Casestudy(d):effectsofgridsizeandpatternofspatial

occupancyonAOO

TheeffectofAOObasedonaﬁxedgrid-cellsizeforHabitats

withdifferentPSOisshowninFig.4.Theyaxisshowstheratio

AOO/actualsurface,foreachPSO-basedgroupofHabitats

(aver-agevaluespergrouparerepresented);thexaxisshowsfourseries

ofdata,referringtothefourgrid-cellsizesusedfortheAOO

cal-culation(10×10km2_,₂_×₂_km2_,₁_×₁_km2_and_0.5_×_0.5_km2_)._For

pointHabitats(and,atalesserextent,forlinearHabitats)thegap

betweenAOOand actualsurfacewasalwaysremarkablyhigher

thanforarealhabitats.Thisgaplargelyincreasedwhen shifting

fromasmall-toalarger-sizedgrid.Overall,alargegrid-cellsize

producedhighAOOestimates,whileasmallersizeproducedAOO

surfacesthat betterapproximatedtherealsurfaces(Spearman’s

correlation,respectively:R=0.952,p<0.001;R=0.759,p<0.001;

R=0.807,p<0.001). Furthermore,ifwe considerthetrendlines,

whichrepresenthowtheratioAOO/actualsurfaceincreaseswith

theprogressiveenlargementofthegrid-cellsize,afurtherinsight

becomesevident:theinﬂationissigniﬁcantlymorepronounced

whenconsideringHabitatswithapointorlinearPSOcomparedto

anarealPSO,showingdifferencesofseveralordersofmagnitude

(5)

Fig.2.AverageAnnexIHabitat(H)surface(ha)perNatura2000SiteinItaly,withHabitatsgroupedaccordingtotheirPSO:point,linearorareal;elaborationbasedonoriginal datafrom2585SCIand/orSPAfromItaly;thestatisticalsigniﬁcanceofthedifferencesamonggroupsisreported,basedonKruskal-WallisTest(DataSource:MATTM,2013).

Fig.3.RelationsbetweentheAOOsurface(basedona10×10km2_grid)_and_the_total_actual_surface_(km2₎_for_126Habitats_occurring_in_Italy;_symbols_and_trendlines_are

(6)

Fig.4.RatiobetweentheAOOsurface(calculatedwithprogressivelylargergrid-cellsize:0.5×0.5,1×1,2×2,10×10km2₎_and_the_actual_measured_surface,_for₄₁_Italian

AnnexIHabitattypesforwhichexhaustiveanalyticdatawereavailableatdifferentscales.Habitatsaregroupedaccordingtothe3modelsofPSO;bars:standarderror;the statisticalsigniﬁcanceofthedifferencesamonggroupsisreported,basedonKruskal-WallisTest:*p<0.05;**p<0.01;***p<0.001(Datasources:Natura2000Management Plans;MATTM,2013).

4. Discussion:lessonslearnedandperspectives

Theresults of thefourcase studieshelp pointing out some

issuesthatmightbeworththinkinganddiscussing.Asmain

out-come,resultshighlighttheprominentroleofscaleandgrainwhen

quantifyingthedistributionrangeofa Habitatandofanyplant

community-basedhabitattype.

Theimportanceofthescalehasbeenwidelyrecognisedasit

concernsalltypesofecologicaldataandisafundamentalaspect

oftheenvironmentalheterogeneity,whoseinterpretationdepends

onthelevelofobservationestablishedwhenstudyinganecological

system(Battisti&Fanelli,2015;Levin,1992;O’Neilletal.,1991,

1996).

Intheory,anyscalecanbeusedtostudyecologicalproblems,

andnospeciﬁcspatialscalemaybeuniversallyapplied;

never-theless,thescaleatwhichenvironmentalpatternsarequantiﬁed

inﬂuencestheresult(Brandt,1998; Turneret al.,2001; Wiens,

1989; Wu,2004)andinappropriatescalesmayfailindetecting

patterns(Li&Wu,2004;Kirchheimeretal.,2015).With

increas-inggrainsize(anddecreasingscale),unitswhichareintrinsically

rareornaturallyoccurasverysmallstandsmaybecomeless

rep-resentedorevendisappear(Turneretal.,2001).Thisisgenerally

truewhendistributionmapsbasedonpolygonsareused.Butitis

trueaswellwhenadoptingtheAOOasaproxyfordistribution.

EstimatesofAOOappeartobehighlysensitivetospatialgrain,as

alsosuggestedbyotherAuthors(seee.g.,Hartley&Kunin,2003;

Nicholsonetal.,2009).Thesurface(andthenumberofcells)

cor-respondingtotheAOOobviouslydependsonthecellsizeandisa

functionofthescaleatwhichitismeasured.

The here discussed case studies show that the AOO

scale-dependenceisalsostronglyaffectedbyanemergentpropertyof

plantcommunities:thePSO,aninherentlyscale-dependent

con-cept. With reference to red-listing, scale-dependent properties

havefar-reachingconsequencesforbothidentiﬁcationand

classi-ﬁcation,aswellasforrepresentationandevaluation.Inconsistency

intheclassiﬁcationandmappingscaleofecologicalcommunities

mayleadtoinaccuratestatementsandaffecttheoutcomesofthe

assessment(Nicholsonetal.,2009).

Thecasestudy(a)showedthedramaticeffectofthePSOonthe

AOOcalculation:thePSOaffectsthenumberofunitsofoccurrence,

regardlessofthegrid-cellsize.Weusedagridwith10×10km2

cellsize,asrecommendedbyKeithetal.(2013),butsimilarresults

wouldbeobtainedwithothertheoreticalhabitatsandgridswith

largerorsmallercells.Althoughsharingthesametotaldistribution

surface,theAOOofthe3theoreticalHabitatsdrasticallyvaried.

TheinappropriatenessofAOObecomesparticularlyapparentfor

thoseHabitats(andrelatedplantcommunities)whosenaturalPSO

istheresultofstrongenvironmentaldrivingforces,which

deter-minepeculiargeometriesintheirspaceoccupancy.Itmightbethe

case,forinstance,ofthevegetationcolonisingtheland

immedi-atelyadjacenttoariver,thesocalled‘riparianzone’,adeﬁnition

thataccountsforitsmoreorlessnarrowandlineardistribution

(Cressie,1993;Diggle,1983;Malanson,1993;Upton&Fingleton, 1985,1989).Similarly,theintrinsicfragmentationofapointHabitat

typewillalwaysresultinaremarkableinﬂationofitsAOO,which

inFig.3isoneorderofmagnitudehigherthanthearealone,dueto

itsscattereddistribution.

ThethreemodelsofPSOadoptedinouranalysisshow,with

asimpleapproach,thattheintrinsicspatialpatternsofplant

com-munitiesshouldnotbeneglectedwhenassessingtheirdistribution

size.Starting from a homogeneousdata set,theproposed case

studiesprovedthatthechoiceoftheresolutionscaleaffectsthe

comprehensionofthepatternsofspaceoccupancyandoccurrence

oftheHabitatsandthecorrelationsbetweenHabitatsand

envi-ronment(Greig-Smith,1983;Juhász-Nagy&Podani,1983;Reed,

Peet,Palmer,&White, 1993),witha substantialrelapse onthe

measurementoftheirAOO.

Casestudy(c)demonstratedthattheuseofacommongridfor

theAOOcalculationaffectstheoutcomesinadifferentialwayfor

the3PSOtypes,generatinginconsistentresults.Habitatswitha

pointorlinearPSOarethosewithamajorrisktobeover-estimated

whenrelyingonAOO,whileforthosewithanarealPSOtheAOO

(7)

Table1

AttributionoftheItalianAnnexIHabitatstothe3PSO.

ListoftheAnnexIHabitatsoccurringinItaly,withtherelatedcodes(datasources: MATTM,2013),consideredinthedataanalyses,groupedonthegroundoftheir patternofspatialoccupancy(PSO);thelettersinbrackets‘N’,‘R’and‘P’indicatethe Habitattypesselectedforcasestudy(d)andreporttheterritorialrangethedata referto(respectively:national,regional,provincial).

HabitatswithanarealPSO

•Openseaandtidalareas:1110,1120,1130,1140,1150,1160,1170 •Temperateheathandscrub:4030,4060,4070,4080,4090 •Sub-Mediterraneanandtemperatescrub:5110,5130 •Mediterraneanarborescentmatorral:5210,5230

•Thermo-Mediterraneanandpre-steppebrush(except5320because relatedtothegradientofthelittoralsystems):5330

•Phrygana:5410,5420,5430

•Naturalgrasslands:6110,6130,6150,6170

•Semi-naturaldrygrasslandsandscrublandfacies:6210,6220,6230, 6240,62A0

•Sclerophillousgrazedforests(dehesas):6310

•Moliniameadowsoncalcareous,peatyorclayey-siltladensoils(Molinion caeruleae):6410

•Mesophilegrasslands:6510,6520

•Sphagnumacidbogs:7110(N,R),7120,7140(P),7150 •Alkalinefens:7230(R,P)

•AlpinepioneerformationsofCaricionbicoloris-atrofuscae:7240 •Scree:8110,8120,8130

•SiliceousrockwithpioneervegetationoftheSedo-Scleranthionorofthe Sedoalbi-Veroniciondillenii:8230

•Limestonepavements:8240

•Fieldsoflavaandnaturalexcavations:8320 •Permanentglaciers:8340

•ForestsoftemperateEurope(except91B0,91E0,91F0becausedepending onthepresenceofwatergradients):9110(R),9120,9130,9140,9150, 9160,9180(R),9190,91AA,91D0,91H0,91K0,91L0(R),91M0(R) •Mediterraneandeciduousforests(except92A0,92C0,92D0because dependingonthepresenceofwatergradients):9210(R),9220,9250,9260 (R)

•Mediterraneansclerophyllousforests:9320,9330,9340(N),9350,9380 •Temperatemountainousconiferousforests:9410(R),9420,9430 •MediterraneanandMacaronesianmountainousconiferousforests:9510, 9530,9540,9560,9580,95A0

HabitatswithalinearPSO

•Seacliffsandshingleorstonybeaches:1210,1240 •Salicorniaandotherannualscolonisingmudandsand:1310 •Spartinaswards(Spartinionmaritimae):1320

•Mediterraneanandthermo-Atlanticsaltmarshesandsaltmeadows: 1410,1420,1430

•Saltandgypsuminlandsteppes:1510

•SeadunesoftheAtlantic,NorthSeaandBalticcoasts:2110,2120(R), 2130(R),2160

•SeadunesoftheMediterraneancoast:2210,2230,2240,2250,2260, 2270

•Hardoligo-mesotrophicwaterswithbenthicvegetationofCharaspp.: 3140

•NaturaleutrophiclakeswithMagnopotamionorHydrocharition–type vegetation:3150

•Runningwater:3220,3230,3240(R),3250,3260,3270,3280(R),3290 •LowformationsofEuphorbiaclosetocliffs:5320

•Semi-naturaltall-herbhumidmeadows(except6410):6420,6430(R) •Calcareousfens(except7230,7240whichhaveanarealdistribution): 7210(R),7220(P)

•Calcareousrockyslopeswithchasmophyticvegetation:8210(R) •Siliceousrockyslopeswithchasmophyticvegetation:8220(R) •ForestsoftemperateEuropedependingonwaterorimpluvia:91B0, 91E0(R),91F0(R)

•Mediterraneandeciduousforestsdependingonwaterorimpluvia:92A0 (R),92C0,92D0(R)

HabitatswithapointPSO •Inlandsaltmeadows:1340(N)

•InlandduneswithopenCorynephorusandAgrostisgrasslands:2330(N) •Standingwater(except3140,3150):3110,3120,3130,3160,3170(R) •Submergedorpartiallysubmergedseacaves:8330

evidencedincasestudy(d),theappropriatenessofagrid-cellsize fortheAOOcalculationisdeeplycontingentontheHabitatPSO, andHabitattypeswithapointorlinearPSOwillalwaysrisktobe overestimatedwhentheirAOOiscalculatedbasedonacommon grid-cellsize.

Theissuehasalreadybeenfacedwithreferencetospecies red-listing.Manyauthorsagreedinrecognisingthatdifferentspatial scalesofobservationcanrevealdifferentthresholdpatternsfor differentspecies(Forman,1995;Huggett,2005;MacNallyetal.,

2002).TheriskofinconsistenciesandbiascausedbytheAOO

esti-mationat differentscaleswasalready emphasised byIUCN for

species,withreferencetospecialisedtaxawithapatchy

distribu-tion,evidencingtheneedforscalecorrectionfactors(IUCN,2012,

2013).IUCNevenapproachedtheproblemofspecieswith‘linear’

habitats,eventuallysuggestingtoneglectthispeculiarecological

distribution,sinceinsuchcasesthehabitatactualsurfacemightbe

sonarrowthatitshouldnearlyalwaysbeassignedtoathreatened

status,basedonthe‘B’criterion(IUCN,2013).

Nevertheless,thereisageneralagreementontherecognition

ofintrinsic(speciﬁc)needsofdifferentspecieswithreferenceto

pressuresandextinctionrisk(Johnson,2013).

Dykstra (2004) reported that there are no generalresponse

modelsinspeciesdecline,makingtheidentiﬁcationofgenerally

validthresholdsparticularlyhard.Indeed,ageneralisationishard

toﬁndforsinglespecies,duetotheenormousvarietyofecological

requirementsandtraits,includingsizeandspecialisation.Similarly,

itispracticallyimpossibletoadoptspeciﬁcrulesforeachsingle

Habitat.However,someintrinsicspatialpropertiesofeachplant

community-basedhabitattype(e.g.thehereproposedPSO)might

representagoodmodelandofferamajoradvantageinthatthey

allowtotakeintoaccount,atleastinpart,someprominent

commu-nityattributesandtomodulatenotonlymetricsbutalsothresholds

andcriteriaofassessment.Additionally,itisperhapsworthto

con-siderthat,althoughAOOisoneofthemostusedmetricstoassess

theconservationstatusofahabitat,itshouldnotbeusedalone,but

shouldbeonlyoneinasetofproxies.

IUCN(2012)recommendstouseascalethatis‘appropriateto

relevantbiologicalaspects’oftheassessmentunit.Inthislight,the

proposalbyKeithetal.(2013)tocalculateAOOonthebaseofa

standardisedspatialgrain,recommendinga10×10km2_grid_and

ﬁxingcommonquantitativethresholdsforalltheassesseditems,

appearsratherdifﬁculttounderstand,ifnotevenlackingaclear

theoreticalbasis,assuggestedbyBoitanietal.(2015).

Ourﬁndingshighlighttheneedofacertainadaptabilityinthe

red-listcriteriawhenassessingecologicalcommunities,with

par-ticular referencetotheirspatialdimension, in accordance with

Nicholsonetal.(2009)whoarguedthatcommunitiesshouldbe

listedatthebroadestscaleatwhich theymeetcriteriafor

list-ing.Habitat types witha linearor a pointPSO, naturallysmall

in size,scatteredorwitha stretcheddistribution,need a

ﬁner-scaleapproachtobeaccuratelyidentiﬁedandcorrectlyassessed,

asalreadyemphasisedforspecies(seee.g.,Jiménez-Alfaro,Draper,

&Nogués-Bravo,2012).

Although it is beyond thescope of this study, the reported

casestudiesconcurtoconﬁrmthatplantcommunitiesare

suit-ableoperationalobjectsinthered-listingprocess.Despite their

greatcomplexity,plantcommunitiesandplantcommunity-based

habitattypesareorganizedintheformofspatiallyrecognisable

patches.Theyappeartobeunequivocalassessmentunits,better

deﬁnedcomparedtotheconceptof‘ecosystem’usedbyKeithetal.

(2013),asalreadyindicatedbyBoitanietal.(2015).Thereisahuge

convergencebetweenplantcommunities,HDHabitatsandother

ofﬁciallyacknowledgedvegetationclassiﬁcationsystems,suchas

EUNIS(Davies,Moss,&Hill,2004)orCORINEbiotopes(Commission

oftheEuropeanCommunities,1991),asclearlyexempliﬁedbythe

(8)

aremanyeffortstoﬁndaninteroperabilitybetweenexisting

classi-ﬁcationsofhabitatsandplantcommunities(see,e.g.,Rodwelletal.,

2002;Schaminéeetal.,2012,2013a,2013b).

4.1. Conclusions

Withthepresentpaperweaimtoadvancesomeviewpointsthat

mightconsolidatethegroundsbutalsoopennewperspectivesfor

theincreasinginterestintheassessmentofhigher-than-species

biodiversitylevels.Facingtheriskofunreliableassessments,some

possiblesolutionsshouldgo inthefollowing directions:(i)the

actualdistributionsurfaceofaHabitatshouldalwaysbeseenas

theoptimalproxytotakeintoaccount;(ii)gridsshouldonlybe

limitedtocaseswherethelevelofknowledgeisreallylow;(iii)

whenusinggrids,thecellsize(or,possibly,thethresholdvalues

fordifferentthreatcategories)shouldbeselectedbasedonthePSO

oftheassessmentunit.

Thereissubstantialscopeforexploringthespatialdimensionof

Habitatsandfuturestudiesshouldimplementresearches

speciﬁ-callydesignedtoaddressthisissue.Itisindeedacentralpointfor

red-listassessmentsandhighlightswhatwebelievetobesome

novelfuturedirectionforthisﬁeld.

References

Allen,T.F.H.,&Starr,T.B.(1982).Hierarchy:perspectivesforecologicalcomplexity. Chicago,USA:UniversityofChicagoPress.

Andreas,B.K.,&Lichvar,R.W.(1995).Floristicindexforassessmentstandards:a casestudyfornorthernOhio.WetlandsResearchProgramTechnicalReport WRP-DE-8.ExperimentStation,Vicksburg,MA:U.S.ArmyCorpsofEngineers Waterways.

Bacchetta,G.,Farris,E.,&Pontecorvo,C.(2012).Anewmethodtosetconservation prioritiesinbiodiversityhotspots.PlantBiosystems,146(3),638–648. Baillie,J.E.M.,Hilton-Taylor,C.,&Stuart,S.N.(2004).IUCNredlistofthreatened

species.aglobalspeciesassessment.GlandandCambridge:IUCN.

Balmford,A.,Bruner,A.,Cooper,P.,Costanza,R.,Farber,S.,Green,R.E.,etal.(2002). Economicreasonsforconservingwildnature.Science,297,950–953. Battisti,C.,&Fanelli,G.(2015).Don’tthinklocal!Scaleinconservation,

parochialism:dogmaticbureaucracyandtheimplementingoftheEuropean Directives.JournalforNatureConservation,24,24–30.

Berg,C.,Abdank,A.,Isermann,M.,Jansen,F.,Timmermann,T.,&Dengler,J.(2014). RedListsandconservationprioritizationofplantcommunities—a

methodologicalframework.AppliedVegegetationScience,17(3),504–515. Biondi,E.(2011).Phytosociologytoday:methodologicalandconceptualevolution.

PlantBiosystems,145(1),19–29.

Biondi,E.,Blasi,C.,Burrascano,S.,Casavecchia,S.,Copiz,R.,DelVico,E.,etal. (2009).ManualeItalianodiinterpretazionedeglihabitatdellaDirettiva92/43/CEE. S.B.I.,M.A.T.T.M.,D.P.N..Accessed10.02.15.http://vnr.unipg.it/habitat Biondi,E.,Burrascano,S.,Casavecchia,S.,Copiz,R.,DelVico,E.,Galdenzi,D.,etal.

(2012).DiagnosisandsyntaxonomicinterpretationofAnnexIHabitats(Dir. 92/43/EEC)inItalyatthealliancelevel.PlantSociology,49(1),5–37. Biserkov,V.(Ed.).(2011).ReddatabookoftherepublicofBulgaria.Bulgarian

AcademyofSciences&MinistryofEnvironmentandWater.Accessed12.04.15. http://e-codb.bas.bg/rdb/en/vol3

Bissonette,J.A.(Ed.).(1997).Wildlifeandlandscapeecology:effectsofpatternand scale.NewYork:Springer-Verlag.

Blab,J.,Riecken,U.,&Ssymank,A.(1995).Proposalonacriteriasystemfora NationalRedDataBookofBiotopes.LandscapeEcology,10,41–50. Boitani,L.,Mace,G.M.,&Rondinini,C.(2015).Challengingthescientiﬁc

foundationsforanIUCNRedListofEcosystems.ConservationLetters,8(2), 125–131.

Brandt,J.(1998).Keyconceptsandinterdisciplinarityinlandscapeecology:a summing-upandoutlook.InJ.W.Dover,&R.G.H.Bunce(Eds.),Keyconceptsin landscapeecology(pp.421–434).Garstang,GreatBritain:IALE-UK.

Braun-Blanquet,J.(1964).Pﬂanzensoziologie.grundzügedervegetationskunde(3 ed.).Wien-NewYork.

Bunce,R.G.H.,Bogers,M.M.B.,Evans,D.,Halada,L.,Jongman,R.H.G.,Mucher,C. A.,etal.(2013).Thesigniﬁcanceofhabitatsasindicatorsofbiodiversityand theirlinkstospecies.EcologicalIndicators,33,19–25.

Butchart,S.H.M.,Stattersﬁeld,A.J.,Bennun,L.A.,Bennun,L.A.,Shutes,S.M., Akc¸akaya,H.R.,etal.(2004).Measuringglobaltrendsinthestatusof biodiversity:RedListIndicesforbirds.PLoSOneBiology,2,2294–2304. Cafaro,P.,&Primack,R.(2014).Speciesextinctionisagreatmoralwrong.

BiologicalConservation,170,1–2.

CEM-IUCN,&Provita.(2012).IUCNredlistofecosystems.thecommissionon ecosystemmanagement(CEM)ofinternationalunionforconservationofnature (IUCN)andprovita,caracas,Venezuela..Accessed21.09.12.http://www. iucnredlistofecosystems.org

CITES.(1973).Conventiononinternationaltradeinendangeredspeciesofwildfauna andﬂora.signedatwashington,D.C.,on3march1973.amendedatbonn,on22 june1979..Accessed16.06.15.http://www.cites.org/eng/disc/text.php CommissionoftheEuropeanCommunities(1991).HabitatsoftheEuropean

Community.CORINEbiotopesmanual.Luxembourg:Commissionofthe EuropeanCommunities.

Cowling,R.M.,Knight,A.T.,Faith,D.P.,Ferrier,S.,Lombard,A.T.,Driver,A.,etal. (2004).Natureconservationrequiresmorethanapassionforspecies. ConservationBiology,18,1674–1676.

Cowling,R.M.,&Heijnis,C.E.(2001).Theidentiﬁcationofbroadhabitatunitsas biodiversityentitiesforsystematicconservationplanningintheCapeFloristic Region.SouthAfricanJournalofBotany,67,15–38.

Cressie,N.A.C.(1993).Statisticsforspatialdata(reviseded.).NewYork:Wiley. Dale,M.R.T.(1999).Spatialpatternanalysisinplantecology.Cambridge:

Cambridge:UniversityPress.

Davies,C.E.,Moss,D.,&Hill,M.O.(2004).EUNIShabitatclassiﬁcationrevised2004.. Availableat.http://eunis.eea.europa.eu/upload/EUNIS2004report.pdf Dengler,J.,Chytr ´y,M.,&Ewald,J.(2008).Phytosociology.InS.E.Jørgensen,&B.D.

Fath(Eds.),Generalecology.encyclopediaofecology(Vol.4)(pp.2767–2779). Oxford:Elsevier.

Diggle,P.J.(1983).Statisticalanalysisofspatialpointpatterns.London:Academic Press.

Dykstra,P.D.(2004).ThresholdsinHabitatSupply:AReviewoftheLiterature. WildlifeReportNo.R-27,Victoria,BritishColumbia:BritishColumbiaMinistry ofSustainableResourceManagement.

Essl,F.,Egger,G.,&Ellmauer,T.(2002).Rotelistegefährdeterbiotoptypen österreichs.Wien:Konzept,UmweltbundesamtGmbH.

EuropeanCommission(2011).CommunicationfromthecommissiontotheEuropean parliament,thecouncil,theeconomicandsocialcommitteeandthecommitteeof theregions.Ourlifeinsurance,ournaturalcapital:anEUbiodiversitystrategyto 2020.Brussels:COM(2011)244ﬁnal.

EuropeanCommission.(2013).Interpretationmanualofeuropeanunion habitats—EUR28.Brussels:EuropeanCommission.

Evans,D.(2006).ThehabitatsoftheEuropeanunionhabitatsdirective.Biologyand Environment:ProceedingsRoyalIrishAcademy,B,106b(3),167–173.

Evans,D.(2010).InterpretingthehabitatsofAnnexI.Past,presentandfuture.Acta BotanicaGallica,157(4),677–686.

Evans,D.,&Arvela,M.(2011).AssessmentandreportingunderArticle17ofthe HabitatsDirective–ExplanatoryNotes&Guidelinesfortheperiod2007–2012– Finalversion.ETC-BD.

Forman,R.T.T.(1995).Landmosaics:theecologyoflandscapesandregions. Cambridge,UK:CambridgeUniversityPress.

Gauthier,P.,Debussche,M.,&Thompson,J.D.(2010).Regionalprioritysettingfor rarespeciesbasedonamethodcombiningthreecriteria.Biological

Conservation,143,1501–1509.

Géhu,J.M.(1974).Surl’emploidelaméthodephytosociologiquesigmatistedans l’analyse:ladéﬁnitionetlacartographiedespaysages.ComptesRendusde l’AcadémiedesSciences,279,1167–1170.

Géhu,J.M.(1991).L’analysesymphytosociologiqueetgéosymphytosociologique del’espace:Théorieetméthodologie.ColloquesPhytosociologique

‘PhytosociologieEtPaysages(1988)’,17,12–46.

Genovesi,P.,Angelini,P.,Bianchi,E.,Dupré,E.,Ercole,S.,Giacanelli,V.,etal.(2014). SpecieehabitatdiinteressecomunitarioinItalia:distribuzione,statodi conservazioneetrend.,dataavailableat.http://cdr.eionet.europa.eu/it/eu/ art17/envupyjhw

Gray,T.N.E.,Phan,C.,&Long,B.(2010).Modelingspeciesdistributionatmultiple spatialscales:gibbonhabitatpreferencesinafragmentedlandscape.Animal Conservation,13(3),324–332.

Greig-Smith,P.(1979).Patterninvegetation.JournalofEcology,67(3),755–779. Greig-Smith,P.(1983).Quantitativeplantecology.studiesinecology,9(3rded.).

UniversityofCalifornia:BlackwellScientiﬁcPublications.

Halley,J.D.,&Winkler,D.A.(2008).Classiﬁcationofemergenceanditsrelationto self-organization.Complexity,13(5),10–15.

Hartley,S.,&Kunin,W.(2003).Scaledependencyofrarity,extinctionrisk:and conservationpriority.BiologicalConservation,17,1559–1570.

Huggett,H.J.(2005).Theconceptandutilityof‘ecologicalthresholds’in biodiversityconservation.BiologicalConservation,124,301–310. IUCN(2012).IUCNRedListCategoriesandCriteria.Version3.1.2nded.Gland,

SwitzerlandandCambridge,UK.

IUCN.(2013).GuidelinesforusingtheIUCNredlistcategoriesandcriteria.version 10.1.preparedbythestandardsandpetitionssubcommittee.,availableat.http:// www.iucnredlist.org/documents/RedListGuidelines.pdf

IUCN.(2015a).InL.M.Bland,D.A.Keith,N.J.Murray,&J.P.Rodríguez(Eds.), GuidelinesfortheapplicationofIUCNredlistofecosystemscategoriesandcriteria, version1.0.Gland,Switzerland:IUCN,ix+93pp.

IUCN.(2015b).TheIUCNredlistofthreatenedspecies.version2015-4..Downloaded on1March2016.dataavailableat.http://www.iucnredlist.org/about/ summary-statistics

Izco,J.(2015).Riskofextinctionofplantcommunities:riskandassessment categories.PlantBiosysystems,149(3),589–602.

Jiménez-Alfaro,B.,Draper,D.,&Nogués-Bravo,D.(2012).Modelingthepotential areaofoccupancyatﬁneresolutionmayreduceuncertaintyinspeciesrange estimates.BiologicalConservation,147(1),190–196.

Johnson,C.J.(2013).Identifyingecologicalthresholdsforregulatinghuman activity:effectiveconservationorwishfulthinking?BiologicalConservation, 168,57–65.

(9)

Juhász-Nagy,P.,&Podani,J.(1983).Informationtheorymethodsforthestudyof spatialprocessesandsuccession.Vegetatio,51,129–140.

Julve,P.(1986).Problèmesconceptuelsdansladéﬁnitiondesunitésdeperception dupaysagevégétalenrapportaveclagéomorphologie.Colloques

Phytosociologiques‘VégétationEtGéomorphologie(1985),13,65–84. Keith,D.A.(2009).Theinterpretation:assessmentandconservationofecological

communities.EcologicalManagementandRestoration,10,3–15.

Keith,D.A.,Rodríguez,J.P.,Rodríguez-Clark,K.M.,Nicholson,E.,Aapala,K.,Alonso, A.,etal.(2013).ScientiﬁcfoundationsforanIUCNredlistofecosystems.Public LibraryofScience,5,62–111.

Keith,D.A.,Rodríguez,J.P.,Brooks,T.M.,Burgman,M.A.,Barrow,E.G.,Bland,L., etal.(2015).TheIUCNredlistofecosystems:motivations,challenges,and applications.ConservationLetters,8(3),214–226.

Kirchheimer,B.,Schinkel,C.C.F.,Dellinger,A.S.,Klatt,S.,Moser,D.,Winkler,M., etal.(2015).Amatterofscale:apparentnichedifferentiationofdiploidand tetraploidplantsmaydependonextentandgrainofanalysis.Journalof Biogeography,http://dx.doi.org/10.1111/jbi.12663

Kontula,T.,&Raunio,A.(2009).Newmethodandcriteriafornationalassessments ofthreatenedhabitattypes.BiodiversityandConservation,18,3861–3876. Levin,S.A.(1992).Theproblemofpatternandscaleinecology.Ecology,73,

1943–1967.

Li,H.,&Wu,J.(2004).Useandmisuseoflandscapeindices.LandscapeEcology,19, 389–399.

Lindenmayer,D.,Hobbs,R.J.,Montague-Drake,R.,Alexandra,J.,Bennett,A., Burgman,M.,etal.(2008).Achecklistforecologicalmanagementof landscapesforconservation.EcologyLetters,11,78–91.

Lindgaard,A.,&Henriksen,S.(Eds.).(2011).The2011norvegianredlistfor ecosystemsandhabitattypes.Trondheim:NorwegianBiodiversityInformation Centre.

Lortie,C.J.,Brooker,R.W.,Choler,R.,Kikvidze,Z.,Michalet,R.,Pugnaire,F.I.,etal. (2004).Rethinkingplantcommunitytheory.Oikos,107(2),433–438. MacNally,R.,Bennett,A.F.,Brown,G.W.,Lumsden,L.F.,Yen,A.,Hinkley,S.,etal.

(2002).Howwelldoecosystem-basedplanningunitsrepresentdifferent componentsofbiodiversity?EcologicalApplication,12,900–912. Mace,G.M.,Collar,N.J.,Gaston,K.J.,Hilton-Taylor,C.,Akc¸akaya,H.R.,

Leader-Williams,N.,etal.(2008).Quantiﬁcationofextinctionrisk:iUCN’s systemforclassifyingthreatenedspecies.ConservationBiology,22,1424–1442. Malanson,G.P.(1993).Riparianlandscapes.cambridgestudiesinecology.

Cambridge,UK:CambridgeUniversityPress.

Margules,C.R.,&Pressey,R.L.(2000).Systematicconservationplanning.Nature, 405,243–252.

MATTM.(2013).Bancadatinatura2000.Roma..Accessed21.10.14.http://ftp.dpn. minambiente.it/Natura2000/TrasmissioneCE2013

McCarthy,M.A.,Thompson,C.J.,&Garnett,S.T.(2008).Optimalinvestmentin conservationofspecies.JournalofAppliedEcology,45,1428–1435. MillenniumEcosystemAssessment.(2005).Ecosystemsandhumanwellbeing:

biodiversitysynthesis.Washington,DC:WorldResourcesInstitute. Moravec,J.,Balátová-Tuláˇcková,E.,Hadaˇc,E.,Hejn ´y,S.,Jeník,J.,Kolbek,J.,etal.

(1983).Rostlinnáspoleˇcenstva ˇCeskésocialistickérepublikyajejichohroˇzení.(Red listofplantcommunitiesoftheCzechRepublicandtheirendangerment.)1.ed. SeveroˇceskouPˇrír.,Suppl.1983(1),1–110.

Nicholson,E.,Keith,D.A.,&Wilcove,D.S.(2009).Assessingtheconservation statusofecologicalcommunities.ConservationBiology,23,259–274. Noss,R.F.(1990).Indicatorsformonitoringbiodiversity:ahierarchicalapproach.

BiologicalConservation,4,355–364.

Noss,R.F.(1996).Ecosystemsasconservationtargets.TrendsinEcologyand Evolution,11,351.

Noss,R.F.,LaRoe,E.T.,&Scott,J.M.(1995).EndangeredecosystemsoftheUnited States:apreliminaryassessmentoflossanddegradation.Washington,DC.U.S: GeologicalSurvey.

O’Neill,R.V.,Gardner,R.H.,Milne,B.T.,Turner,M.,&Jackson,B.(1991). Heterogeneityandspatialhierarchies.InJ.Kolasa,&S.T.A.Pickett(Eds.), Ecologicalheterogeneity(pp.85–96).NewYork:Springer-Verlag.

O’Neill,R.V.,Hunsaker,C.T.,Timmins,S.P.,Jackson,B.L.,Jones,K.B.,Ritters,K.H., etal.(1996).Scaleproblemsinreportinglandscapepatternattheregional scale.LandscapeEcology,11,169–180.

Paal,J.(1998).RareandthreatenedplantcommunitiesofEstonia.Biodiversityand Conservation,7,1027–1049.

Palmer,M.W.(1988).Fractalgeometry:atoolfordescribingspatialpatternsof plantcommunities.Vegetatio,75,91–102.

Palmer,M.W.,&White,P.S.(1994).Scaledependenceandthespecies-area relationship.AmericanNaturalist,144(5),717–740.

Pärtel,M.,Kalamees,R.,Reier,U.,Tuvi,E.-L.,Roosaluste,E.,Vellak,A.,etal.(2005). Groupingandprioritizationofvascularplantspeciesforconservation: combiningnaturalrarityandmanagementneed.BiologicalConservation,123, 271–278.

QGISProject.(2014).QGISuserguide.release2.0.january12,2014..Accessed 26.05.14.http://docs.qgis.org/2.0/pdf/QGIS-2.0-UserGuide-en.pdf Reed,R.A.,Peet,R.K.,Palmer,M.W.,&White,P.S.(1993).Scaledependenceof

vegetation-environmentcorrelations:acasestudyofaNorthCarolina piedmontwoodland.JournalofVegetationScience,4,329–340.

Rodríguez,J.P.,Keith,D.A.,Rodríguez-Clark,K.M.,Murray,N.J.,Nicholson,E., Regan,T.J.,etal.(2015).ApracticalguidetotheapplicationoftheIUCNRed ListofEcosystemscriteria.PhilosophicalTransactionsoftheRoyalSocietyB,370, 20140003.

Rodríguez,J.P.,Rodríguez-Clark,K.,Baillie,J.E.,Ash,N.,Benson,J.,Boucher,T.,etal. (2011).EstablishingIUCNredlistcriteriaforthreatenedecosystems. ConservationBiology,25,21–29.

Rodríguez,J.P.,Rodríguez-Clark,K.M.,Keith,D.A.,Barrow,E.G.,Benson,J., Nicholson,E.,etal.(2012).IUCNredlistofecosystems.S.A.P.I.EN.S5.2.Accessed 20.05.14.http://sapiens.revues.org.1286

Rodwell,J.,Janssen,J.,Gubbay,S.,&Schaminée,J.(2013).RedListAssessmentof EuropeanHabitatTypes—Afeasibilitystudy.ServiceContractNo.

070307/2012/624047/SER/B3,ReportfortheEuropeanCommission,DG Environment.

Rodwell,J.,Schaminée,J.H.J.,Mucina,L.,Pignatti,S.,Dring,J.,Moss,D.(2002).The DiversityofEuropeanVegetation.Anoverviewofphytosociologicalalliancesand theirrelationshipstoEUNISHabitats.Wageningen.EC—LNVReportnr. 2002/054.MinistryofAgricultureNatureManagementandFisheries,The NetherlandsandEuropeanEnvironmentalAgency.

Schaminée,J.H.J.,Chytry´ı,M.,Hennekens,S.M.,Mucina,L.,Rodwell,J.S.,&Tichy´ı,L. (2013b).DevelopmentofvegetationsyntaxacrosswalkstoEUNIShabitat classiﬁcationandrelateddatasets.Copenhagen:ReportfortheEuropean EnvironmentAgency.

Schaminée,J.H.J.,Chytry´ı,M.,Hennekens,S.M.,Jiménez-Alfaro,B.,Mucina,L., Rodwell,J.S.,etal.(2013a).ReviewofEUNISforesthabitatclassiﬁcation.(pp. 111).DraftReportEEA/NSV/13/005.

Schaminée,J.H.J.,Chytr ý,M.,Hennekens,S.M.,Mucina,L.,Rodwell,J.S.,&Tich ý,L. (2012).DevelopmentofvegetationsyntaxacrosswalkstoEUNIShabitat classificationandrelateddatasets.UnpublishedReportEEA/NSV/12/001. Wageningen,TheNetherlands:AlterraGreenWorld.

Schulte,G.,&Wolf-Straub,R.(1986).Vorläuﬁgerotelistedernordrhein-westfalen gefährdetebiotope:rotelistederinnordrhein-westfalengefährdetenpﬂanzen undtiere.SchriftenreiheDerLÖLF,4,19–27.

SecretariatoftheConventiononBiologicalDiversity(2010).GlobalBiodiversity Outlook3.Montréal.

Turner,M.G.,Gardner,R.H.,&O’Neill,R.V.(2001).LandscapeEcologyinTheoryand Practice:patternandprocess.NewYork,NY,USA:Springer-Verlag.

Turner,M.G.,O’Neill,R.V.,Gardner,R.H.,&Milne,B.T.(1989).Effectsofchanging spatialscaleontheanalysisoflandscapepattern.LandscapeEcology,3, 153–162.

UnitedNations(1976).ConventiononWetlandsofInternationalImportance especiallyasWaterfowlHabitat.ConcludedatRamsar,Iran,2/2/1971.UNTreaty SeriesN.14583,Vol.996,246–268.

UnitedNations(1992).ConventiononBiologicalDiversity.UnitedNations ConferenceonEnvironmentandDevelopment.UNTreatySeries.Vol.1760. Upton,G.J.G.,&Fingleton,B.(1985).Spatialdataanalysisbyexample.vol.I.point

patternandquantitativedata.NewYork:Wiley.

Upton,G.J.G.,&Fingleton,B.(1989).Spatialdataanalysisbyexample.vol.II. categoricalanddirectionaldata.NewYork:Wiley.

vanderMaarel,E.,&Franklin,J.(Eds.).(2013).Vegetationecology.JohnWiley& Sons.

Walker,S.,Price,R.,Rutledge,D.,Stephens,R.T.T.,&Lee,W.G.(2006).Recentloss ofindigenouscoverinNewZealandNewZealand.JournalofEcology,30(2), 169–177.

Westhoff,V.,&vanderMaarel,E.(1978).TheBraun–Blanquetapproach.InR.H. Whittaker(Ed.),Classiﬁcationofplantcommunities(2nded.,Vol.4,pp. 287–297).Junk,TheHague.

Wiens,J.A.(1989).Spatialscalinginecology.FunctionalEcology,3,385–397. Wiens,J.A.(1990).Ontheuseof‘grain’and‘grainsize’inecology.Functional

Ecology,4,720.

Wu,J.(2004).Effectsofchangingscaleonlandscapepatternanalysis:scaling relations.LandscapeEcology,19,125–138.