Content uploaded by Martí March Salas
Author content
All content in this area was uploaded by Martí March Salas on Jan 08, 2018
Content may be subject to copyright.
Appl Veg Sci. 2018;1–9. wileyonlinelibrary.com/journal/avsc
|
1
Applied Vegetation Science
© 2018 International Association
for Vegetation Science
Received:18August2017
|
Accepted:14November2017
DOI: 10.1111/avsc.12355
RESEARCH ARTICLE
An innovative vegetation survey design in Mediterranean cliffs
shows evidence of higher tolerance of specialized rock plants
to rock climbing activity
Martí March-Salas1 | Miguel Moreno-Moya2 | Gemma Palomar3,4 | Pablo Tejero-Ibarra5 |
Emily Haeuser6 | Luis R. Pertierra7
1DepartmentofBiodiversityandEvolutionary
Biology,MuseoNacionaldeCiencias
Naturales(MNCN-CSIC),Madrid,Spain
2DepartmentofAgroforestryEcosystems,
PolytechnicUniversityofValencia,Valencia,
Spain
3ResearchUnitofBiodiversity(UO-CSIC-PA),
Mieres,Spain
4DepartmentofBiologyofOrganismsand
Systems,UniversityofOviedo,Oviedo,Spain
5DepartmentofBiodiversityand
Restoration,InstitutoPirenaicodeEcología
(IPE–CSIC),Jaca,Huesca,Spain
6Ecology,DepartmentofBiology,Universityof
Konstanz,Konstanz,Germany
7AreaofBiodiversityandConservation,
DepartmentofBiologyandGeology,ESCET,
UniversityKingJuanCarlos,Móstoles,Madrid,
Spain
Correspondence
MartiMarch-Salas,DepartmentofBiodiversity
andEvolutionaryBiology,MuseoNacional
deCienciasNaturales(MNCN-CSIC),Madrid,
Spain.
Email:martimarchsalas@gmail.com
Funding information
UniversitatPolitècnicadeValència
Co-ordinatingEditor:LauchlanFraser
Abstract
Questions:Isrockclimbingpressure,togetherwithmicrotopographicconditions,dis-
turbingcliffplantcoverandcomposition?Whataretheclimbingimpactsonrockspe-
cialist and non-specialist species? Can a case-control approach, not previously
implemented in cliff environments, offer additional value for actual and long-term
ecologicalresearch?
Location:Chulilla,Levantecoast,Spain.
Methods:We surveyed in situninerockclimbingroutesinorderto examine differ-
encesinplantspeciesrichnessandvegetationcoverbetweenunclimbedandclimbed
transects.Toevaluatetheeffectofrockclimbingonvegetation,weimplementeda
case-control methodology using the two zones immediately adjacent to common
climbingroutesascontrolpoints(i.e.unclimbedtransects).Threequadratsof3m×3m
wereestablished at different cliff heights.Allidentifiedspecieswerecategorizedas
eitherspecializedrockspeciesornon-specializedrockspeciesbasedontheirhabitat
preferencesfromliterature.Non-specializedrockspecieswerefurtherdifferentiated
as either moderately associated with rocky environments or strict generalists. The
rockclimbingimpactoneachgroupofspecieswasanalysedusingLMM.
Results:OurresultsprovideevidenceoftheeffectsofrockclimbingonaMediterranean
cliff,whichhasreceivedlittleattentionsofar. Significantly fewer generalist species
were present on climbed compared to unclimbed transects, while specialized and
moderatelyspecialized rock specieswerenotsignificantlyaffectedby rock climbing
intensity.Furthermore, while rock-specificandmoderatelyspecializedspeciescould
copewithmicrositeheterogeneity,areaswithfewercrackshadsignificantlynegative
effectsongeneralistspecies.
Conclusions:Moderaterockclimbingactivityoncliffenvironmentsmightnotreduce
the presence of specialized rock-dwelling species; however, this activity inherently
impactsthebiodiversityofcliffecosystemsdueto itslargeeffectongeneralistspe-
cies.Werecommendthatfutureconservationstudiesaccountforthedegreeofspe-
cies dependence on rocky habitats to better understand rock-climbing impacts in
these singular ecosystems. According to our experience, the implementation of an
2
|
Applied Vegetation Science MARCH- SALAS et AL.
1 | INTRODUCTION
Despitealackofknowledgeonrockclimbingimpacts,recentstudies
havereported disturbances to cliff vegetation fromthis increasingly
popular activity (Clark& Hessl, 2015; Lorite etal., 2017). However,
such disturbances to cliff plant communities remainto be assessed
withspecificconsiderationofthedifferentrockadaptationstrategies
ofspecies, togetherwiththeeffects ofmicrotopographicconditions
of the rock surface. In this regard, the review ofHolzschuh (2016)
pointedoutthe relevanceofthesephysicalandbiologicalfactorsfor
adequatecharacterizationoftheaffectedplantcommunitiesandthe
consequentassessmentofenvironmentalimpacts.Yetmostprevious
studieshavesomeselectionbiasaccordingtoHolzschuh(2016),and
somoresystematicapproachesaredesirabletosupportcurrentfind-
ings. Furthermore, specifically regarding biological factors, such as
adaptationofrockplant lifestrategies,theseeffects haveonlybeen
examined once and that sutdy focused on an alpine environment
(Müller,Rusterholz,&Baur,2004).Toincreaseknowledgeofthiscom-
munity, we examine all these effectsin a Mediterranean cliff plant
communityaffectedbyrockclimbing,inaregionlittlestudiedbutalso
recentlyshowntobedisturbed(Loriteetal.,2017).
Mediterraneancliffenvironmentsarecharacterizedbyharshcon-
ditions,such asextremedryness, high erosion,intenseradiationand
poorlydevelopedsoils.Despitetheselimitingconditions,cliffsprovide
ahighlydiversemosaicofmicrohabitatswhereawiderangeofplant
species are able to grow, often including endemic species (Larson,
Matthes,&Kelly,2000). Cliff ecosystems sustain plant speciesspe-
cialized for dwelling within rock balconies, including chamaephytes
thatcolonizethemostrestrictiverockymicrohabitats,aswellassome
generalistspeciesthatarenotspecificallyadaptedtocliffsbutareable
toestablishinrupicoloushabitatsundercertaincircumstances(Davis,
1951;Escudero, 1996). However,the life histories ofthese species
areremarkablydifferent;while specialistrockplantshavedeveloped
stress-tolerant traits to survive adverse abiotic factors (De Micco
&Aronne,2012;Surina & Martinčič, 2014), less adapted plants are
moresensitivetostressfulenvironmentalconditionsand,hence,may
be more vulnerable to external impacts. Mediterraneancliff micro-
habitatsand their occupants contributesubstantiallyto biodiversity
intemperateregions, highlightingtheconservationvalue ofthecliff
habitats.
Althoughmostclifffacesremainunaffectedbydirecthumandis-
turbancesduetotheirinaccessibility,theincreaseinhumanactivities
like mountaineering, especially rock climbing, may endanger these
habitatsthroughtheircumulativepressures.Becauseofthepotential
conflictswithconservationofbiodiversity,restrictionsonrockclimb-
ingactivityarethereforeunderconsiderationatmanysites.Primarily,
conservationpolicies have focused on theprotectionofbird repro-
ductivecycles.However,concernsregardingthepotentialdecreasein
plantspeciesrichnessincliffcommunitiesduetoclimbingimpactson
rocksurfaceshavealsobeenraised(e.g.Loriteetal.,2017;McMillan
&Larson, 2003; Rusterholz,Müller,& Baur,2004;Thiel & Spribille,
2007).Studiesevaluatingtheeffectsofrockclimbinghaveprincipally
focusedon impactstothestructureand compositionof overallcliff
vegetation(Mülleretal.,2004;Nuzzo,1996;Vogler&Reisch,2011),
andharmfulconsequencesofrockclimbingonendemicspecieshave
been noted (López-Pujol, Álvarez, Bosch, Simon, & Blanché, 2006).
Nevertheless,rock-climbingimpactsoncliffvegetationstillneedsfur-
therlong-termresearch.Conservationplansareoftenfocusedonthe
effectsofrockclimbingoncataloguedendemic species,disregarding
plantcommunity structure and habitatdisturbances, partly because
dedicatedresearchisstillcurrentlyscarce.Forinstance,itremainsrel-
ativelyunknownwhetherplantlifehistorydrivescapacitytowithstand
rock climbing pressures (but see Müller etal., 2004). Furthermore,
inEurope,to our knowledge,thereareonlya few examples ofrock
climbingimpactsoncliffvegetationinSwitzerland(e.g.Müller,2006;
Mülleretal.,2004),Spain(Loriteetal.,2017)andGermany(e.g.Thiel
&Spribille,2007).
RelevantstudiesintheMediterraneanregionarelargelyabsent(for
anotableexception,seeLoriteetal.,2017),inspiteofitshighlybiodi-
versecliffs(Groombridge&Jenkins,2002),althoughsomegreylitera-
turemayexistfromlocalstudieswithpurelypracticalfoci(e.g.internal
assessments and reports within natural areas under management).
Moreover, scientific study of rockclimbing effects poses additional
challengesduetolimitedaccessibility,and this mayalso offer some
explanationastothelowamountofdedicatedliterature(Holzschuh,
2016).Accountingforclimbing effectsovertime would alsoprovide
deeperknowledgeofthissystem, especially given the projected in-
creaseinclimbingpopularity(Cordell,2012),butthishasreceivedlittle
consideration.Disturbance effects on habitats often occurgradually
overalongtemporalscale,therefore,long-termmonitoringisapref-
erableassessmentapproach(Knappetal.,2012;Lugoetal.,2012),yet
suchstudieshavenotbeenconductedincliffenvironments.
Current effortsto study trampling impacts in areas with limited
monitoringcapabilitiestendtofocusonspatialcomponentsfollowing
acase-controlmethodologicalapproach(seeTejedoetal.,2012,2016),
provingavaluableframeworkforreplicationincliffenvironments.To
adjacentcase-controlsurveydesignformonitoringcliffvegetationcanhelp improve
andunifymethodologyforsuchstudies,asthisisstillanunderdevelopedfield.
KEYWORDS
climbingimpact,generalistsversusspecialists,limestone,long-termconservation,
microtopography,plantcommunities,rockadaptationstrategy,speciesrichness,vegetation
cover
|
3
Applied Vegetation Science
MARCH- SALAS et AL.
accountforspatialcomponents, itisnecessarytosamplealongrock
climbingroutesaswellasatnearbycontrolpoints.Byreplicatingthis
designacrossdifferentrockclimbingroutesanddifferentpointsinthe
route,it ispossibletoexamine the ecologicalimpactsofrockclimb-
ing.Somekeyenvironmentalfactorsareknowntoinfluencecliffveg-
etationstructure,suchashabitatfeatures(DoCarmo,DeCampos,&
Jacobi,2016;Hespanholetal.,2011).ResultsfromKuntzandLarson
(2006b)revealedastronginfluenceofmicrositeheterogeneityoncliff-
facevegetationcommunity structure. Inaddition, Kuntz and Larson
(2006b)foundthatonundisturbedcliffs,vascularplant,bryophyteand
lichenrichnessandabundancearecontrolledbylocalphysicalfactors
onthecliffface.Therefore,as demanded by Holzschuh (2016), it is
necessary to consider these physicalfactors (e.g. microtopography)
whenexaminingtheeffectsofrockclimbingonvegetation(e.g.Clark
&Hessl, 2015;Kuntz& Larson,2006a);however,moststudies have
neglected this keyaspect. Furthermore, it is also likely that species
withdifferinglifeformswill responddifferentlyto cliffstructureand
disturbancefromrockclimbingactivity,withdifferentlevelsofvulner-
ability.Revealingwhichfactorsdetermineactualcliff-facecomposition
andwhichplantscancopewiththepotentialimpactsofrockclimbing
willbeparticularlyhelpfulforpreservingbiodiversityinthesesingular
ecosystems,byofferingvaluable guidance forpracticalconservation
programmes.
Thegeneral aim ofthisstudy istodeterminetheeffects ofrock
climbing on cliff plant communities, while considering microtopo-
graphicfeaturesandspecieslifestrategies(i.e.rockspecialistorgen-
eralistspecies),throughacase-controlstudyapproachnotpreviously
implementedinthesehabitats.Forthispurpose,wefirst analysethe
rockclimbingeffectsonplantspeciesrichnessandvegetationcoverin
Chulilla(Spain), a highbiodiversityarea oftheMediterraneanregion
subjectedtorockclimbing.Second,wetestifrockclimbingaffectsthe
compositionofspecificspeciesofthesite,andwhethertheeffectvar-
iesbetweencliffspecialistandgeneralistspecies.Third,wesummarize
theimplicationsofourresultsforconservationandmanagementplans
FIGURE1 StudyarealocatedinChulilla,Levantecoast(Spain),wherewesampledninerockclimbingroutesincludedinfoursectors
(i.e.(a)Barranco,(b)PeñetaII,(c)PeñetaI,(d)Fantasía)
4
|
Applied Vegetation Science MARCH- SALAS et AL.
insimilarrockyhabitats.Wediscussthepotentialbenefitsofoursur-
veydesign,notpreviouslyimplementedintheserockyenvironments,
asareferencestandardizedvegetationsurveyapproachthatcouldbe
usedasastartingpointforlong-termecologicalresearch(LTER)stud-
iesincliffssubjectedtohumanrockclimbingpressures.
2 | METHODS
2.1 | Study site
We surveyed in situ cliff formations along the village of Chulilla,
Valencia, in the southwest Iberian system mountains of Spain
(39.6513494°N,0.8500479°W),whicharesubjectedtofrequentrock
climbingactivity.TheseoutcropsinChulillaconsistofmarlsandlime-
stones from the Superior and Middle Jurassic periods (Santisteban,
2012).Thedominantvegetation communityonthe cliffsisaJasonio
glutinosae–Teucrietum thymifolii association (Agulló, Alonso, Juan,
Villar,&Crespo,2010)includingsomeendemicssuchasSarcocapnos
enneaphylla(L.)DC.andTeucrium thymifoliumSchreb.
Rockclimbingactivityin Chulilla started inthe 1980s; in recent
years,ithasbecomeoneofthemostpopularrecreationalrockclimb-
ingareasintheMediterraneanregion(RoldánGutierrez,2013).Most
routes are used forsport rock climbing, which requires drilling into
therocktofixpermanentsafetyequipment.Currently,over900rock-
climbingroutesaredistributedacross30cliffsinChulillaataltitudes
from250to636ma.s.l.,withamaximumcliffheightof50m.
2.2 | Field survey design and data collection
Nine rock-climbing routes, distributed along four outcrops (hence-
forthreferredtoassectors)ofChulillawerechosenfortheirfeasibility
toconductvegetationsurveysinsitu:‘Barranco’,‘PeñetaI’,‘PeñetaII’
and‘Fantasía’(see Figure1 map, Table S1). Field surveys were con-
ductedfrommid-MaytoearlyJul2014,coincidingwithfloweringsea-
sonofthevascularplants.Wesurveyedroutesvaryinginorientation
rangingfromeastto southwest (and delineated them with standard
cardinal and intercardinal directions) to optimize biodiversity cap-
ture.Asanindirectmeasureof rockclimbingintensity,weidentified
beginner, intermediate and experienced level rock climbing routes
(5.4–5.11 on the rock climbing difficulty scale; Table S1), assuming
anegativecorrelationbetweendifficultyandrockclimbingintensity.
Toexaminedifferencesincliffsbetweenunclimbedandclimbed
areas,wedesigned acase-controlsurveywith3m×3m quadrats,
each composed of three1m×3m transects (i.e. plots): a central
transectalongacommonlyused climbingroute(i.e. climbedtran-
sect)andthetwoadjacentzonesnotrequiredforrockclimbing(i.e.
unclimbedtransects) used as controlpoints.Theproximityofthe
controlpointstotheclimbedtransectprecludesthepossibilitythat
differencesinotherphysicalfeaturescouldactasdriversofdiffer-
ences between climbed and unclimbed transects.To examine the
entirecliff-faceplantdistribution(Kuntz&Larson,2006a;Figure2),
weestablishedthreequadratsatdifferentaltitudesalongtherock
climbing route: top, middle and bottom (Figure2). The bottom
quadratwasestablished 1–2mabovetheground,withthemiddle
andtop quadrats placed above at proportionaldistancesso as to
span the cliff height. Forclimbing route denoted ‘a1’ it was only
feasibletoestablishthebottomquadratbecausethecliffwasonly
7-mhigh.Intotal,wesampled25quadratscontaining75transects,
including 50 unclimbed and 25 climbed.
We identified all species in each transect using Castroviejo
(1986–2012) and measured their respective percentage cover. To
assessvegetationcoverforeachspecies,we measuredandsummed
thesquareareaofallindividuals ofa species, and used thattocal-
culate the proportion of the total transect area (3m2) covered by
thatspecies. FollowingtheMülleretal.(2004)classificationand the
Castroviejo(1986–2012)index,specieswerecategorizedas:(S)spe-
cialized rock-dwelling species:allspecieswithstrictassociationtorocky
habitats;(M)moderately associated to rocky habitats species:speciesas-
sociated with other habitats but able to live recurrently in rocky habi-
tats;and(G)generalist species:ubiquitousspecieswithnoassociation
torockyhabitats,andveryrestrictedwithinrockyhabitats.Wecon-
sideredthetwolastgroupstogetherasnon-specializedrockspecies,
asnoneofthesespecieshavedirectassociationwithrockyhabitats.
Physical variables of each rock-climbing route (i.e. cliff height,
aspect and average slope)were also recorded. Total surface ofthe
quadratwasdividedincracks/crevices(hereinafterreferredascracks),
ledges,smoothrocksurfaceandoverhangpercentagetoaccountfor
microtopographvariations.
FIGURE2 Sampledesign.Wesampledthreequadrats(3m×3m)
inthebottom,middleandtopsectionsofthecliffface.Eachquadrat
wasdividedintothreeplots(i.e.transects):acentralclimbedtransect
(C)andtwoadjacentunclimbedtransects(U)
|
5
Applied Vegetation Science
MARCH- SALAS et AL.
2.3 | Statistical analysis
We used the Shapiro-Wilk normality test and Bartlett test for ho-
mogeneityofvariancesto checknormalityand homoscedasticityfor
speciesrichness and percentage plantcover.Allvariableswere nor-
mal except M species richness and vegetation cover. We applied a
^0.7transformationtoMrockyhabitat species richness and a ^0.2
transformationtovegetationcovertoobtainnormaldistributionsfor
subsequentanalyses(seebelow).Therewerenoheteroscedasticdata.
Using the lme4package(Bates,Mächler,Bolker,&Walker,2015)
inR(v3.3.1;RFoundationforStatisticalComputing,Vienna,Austria),
weperformedLMMsandsimplelinearregressionstoassessthecom-
bined effects of rock climbing impact and microtopographic condi-
tionsonspeciesrichnessandtotalpercentageplantcoveratthestudy
sites.Becausethepercentageplantcovervalueswereallconsistently
low,wewereabletoanalysebothresponsevariablesinthisway.To
simplify analyses and avoidover-fitting, we first performed individ-
uallinearregressions,testingspeciesrichnessandcoverresponsesto
eachmicrotopographicvariable. Since the percentageof cracks was
theonlymicrotopographicvariablefoundtobesignificant,thisfeature
alonewasincludedinthesubsequent LMMsassessing rockclimbing
impacts.Inthese LMMs, percentageofcracks,quadratposition(i.e.
bottom,middle and top), rockclimbingtransects(i.e.climbedorun-
climbed)andinteractionswereusedasfixedfactors,whilerockclimb-
ingroutenested within sector was used as random factor.Forboth
speciesrichness andcover,wetested ourthreespeciesgroupsboth
jointlyandseparately.Themostparsimoniousmodelwasdetermined
usingstep-wise backward elimination. Furthermore,we constructed
linearmodelstoevaluatetheeffectoftheslopeandaltitudeofeach
rock-climbingroute.Wealsoseparatelyanalysedtheeffectofthedif-
ficultyoftherouteanditsaspect.Weexaminedthe main effectsof
allthe models usingChi-square tests. When main effects were sig-
nificantin factorscontainingmorethantwolevels,weapplied post-
hoccontrasttestsinR(lsmeanspackage;Lenth,2016)usingtheTukey
adjustment.
3 | RESULTS
We recorded a total of 34 vascular plant species (Table S2). All of
themhaveaMediterraneannativedistribution,exceptCeterach offici-
narumWilld.andOpuntia maxima Mill.,whichare originallyfromthe
Holarctic and Neartic, respectively. Among the Mediterranean spe-
cies,seven(i.e.21.87%)wereidentifiedasIbero-Levantineendemics,
athirdofthetotalspecies(i.e.29.41%)wererockspecialistwhilethe
rest(i.e. 70.59%)werenon-specialistrockspecies. Withinthelatter,
ca25%werespecieshavingmoderateassociationwithrockyhabitats
andca.75% were generalist species (Table S2).Meanplantspecies
richness on the 75 cliff-face transects was 2.68±0.18 species/m2
(mean±SE)andmeanvegetationcoverwas1.45±0.33%perarea.
3.1 | Physical condition effects
Overallspecies richnessrosesignificantly(
χ2
1
=7.38,p = .007) asthe
percentageofcracksintheclifffaceincreased.Wedidnotfindanyef-
fectofthecracksonvegetationcover(
χ2
1
=0.87,p = .351).However,
TABLE1 ResultsofLMMtestingtheeffectsofquadrat(i.e.top,middle,bottom),climbingeffect(i.e.climbed,unclimbed)andpercentageof
crackseffectsonspeciesoverallandwithingroups,differentiatedbyassociationwithrockyhabitats
Species Factor
Species richness Vegetation cover
df χ2p- value Sig. R2 (%) df χ2p- value Sig. R2 (%)
Overallsp. Quadrat 2 10.76 .005 ** 29.80 2 10.80 .006 ** 12.04
Climbingeffect 1 2.17 .140 1 3.67 .055
%cracks 1 7.38 .007 ** 1 0.87 .351
Specializedspp. Quadrat 2 1.52 .469 16.82 2 0.07 .967 6.65
Climbingeffect 10.04 .843 1 0.08 .773
%cracks 10.40 .517 1 1.96 .161
Notspecialized
spp.
Quadrat 2 16.78 <.001 *** 34.93 2 20.00 <.001 *** 25.85
Climbingeffect 14.08 .043 * 1 2.30 .130
%cracks 1 18.76 <.001 *** 1 6.77 .009 **
Moderately
associatedspp.
Quadrat 2 6.79 .033 * 11.65 2 9.08 .011 * 15.46
Climbingeffect 1 0.38 .538 1 0.06 .812
%ofcracks 1 1.92 .166 1 0.48 .490
Strictgeneralist
spp.
Quadrat 2 12.82 .002 ** 45.61 2 18.98 <.001 *** 29.40
Climbingeffect 14.12 .042 * 1 3.96 .047 *
%cracks 120.43 <.001 *** 1 3.92 .003 **
Allinteractionsbetweenthefactorsweresignificantinallmodels(all:
χ2
1
≤1.66,p ≥ .20),thereforetheseresultswerenotincludedinthetable.Marginal
R2values(in%)arereportedforallreducedmodels.Significancesareindicatedwithasterix(*.05>p≥.01;**0.01>p≥.001;***p<.001).
6
|
Applied Vegetation Science MARCH- SALAS et AL.
byexaminingeachspecies groupseparately,wefounda positiveef-
fectof the percentage ofcracksforgeneralists(Table1).There was
no effect of crack percentage on specialized rock species richness
(
χ2
1
=0.40,p = .517)orvegetationcover(
χ2
1
=1.96,p = .16).Similarly,
speciesmoderatelyassociatedwith rocky habitat were not affected
bycrackpercentage.However,therichnessofnon-specialized spe-
cies increased as the percentage of cracks increased (
χ2
1
=18.76,
p < .001),asdidtheircover(
χ2
1
=6.77,p = .009;Table1).
Plants were distributed differently along the cliff face in terms
of both species richness (
χ2
2
=10.76, p = .005) and vegetationcover
(
χ2
2
=10.80, p = .006; Figure3). Average richness and cover of vas-
cularplantswassignificantlylowerin bottomquadrats(species rich-
ness=2.04±0.29, vegetation cover=0.53±0.16%) than in the
middle (3.04±0.30, 1.92±0.72%) and top quadrats (3.04±0.28,
2.02±0.70%). Again, the effect differed between rock-specialized
and non-rock-specialized species (Figure3). While both richness
(
χ2
2
=1.52, p = .469) and cover(
χ2
2
= 0.07, p = .967) of rock specialist
speciesdid not differ across the differentquadrats ofthe cliff face,
non-specializedrockspeciesrichness (Figure4;
χ2
2
=16.78,p < .001)
andcover(Figure4;
χ2
2
=20.00,p < .001) weresignificantlyhigherin
themiddleandtopquadratsthaninthebottomone(Table1,Figure4).
Thiseffectwaspresentinbothnon-specializedrockspeciessubgroups
(Table1).Theeffectofquadratlocationalsodifferedacrosslifeforms.
Finally,we did not find significant influenceof the average slope of
theroute,cliffheight,thedifficultyoftheroute,northeaspectonthe
vegetationcoverandrichnessonthesampledcliffs(TableS3).
3.2 | Rock climbing effect
Wedid not find evidence thatrock-climbing activity affectedover-
allspeciesrichness(
χ2
1
=2.17,p = .140)andonlyamarginaleffecton
vegetationcover(
χ2
1
=3.67,p = .055)inouranalysesacrossallspecies
sampled.However,weagainfoundthattherockclimbingeffect dif-
feredbetweenspeciesgroups. Rock-climbing activity didnotaffect
rockspecialistspecies(richness:
χ2
1
=0.04,p = .843;cover
χ2
1
=0.08,
p = .773) but non-specialist species richness was significantly lower
inclimbedzones(richness,
χ2
1
=4.08,p = .043;covernotsignificant:
χ2
1
=2.30,p = .130).Amongnon-specialists,generalistspeciesrichness
wassignificantlyaffectedbyrock-climbingactivity(
χ2
1
=4.12,p = .042),
withlowerspeciesrichnessintheclimbedzones,whilemoderaterock
specialistspecies were notsignificantlyaffected(
χ2
1
=0.38,p = .538;
Figure5).The samepatternwasfound invegetationcover(general-
ists,
χ2
1
=3.96,p = .047; moderate rockspecialist,
χ2
1
=0.06,p = .812;
Figure5).Interactionsbetweenrockclimbingeffectandpercentage
ofcrackswerenotsignificantinspeciesrichnessorvegetationcover
inanyspeciesgroup(all
χ2
1
≤1.66,p ≥ .20).
4 | DISCUSSION
Althoughcliff environmentecosystemsare consideredrelativelyun-
affectedby humans overthelastcenturies, increasing rockclimbing
activity is altering vegetation present in this habitat (e.g. Adams &
Zaniewski,2012;Farris,1995;Loriteetal.,2017;McMillan&Larson,
2003;Thiel & Spribille, 2007).Our results support theexistence of
negativeeffectsofrock climbingpressuresoncliffvegetation. More
specifically,ourresultsconfirmedthefindingsofMülleretal.(2004),
where rock climbing negatively impacted species diversity and af-
fectedspeciescomposition(Figure S1;TableS4).However,contrary
FIGURE3 Left:Speciesrichnessperquadrat;Right:Covertureof
vegetation(%)perquadrat.Means±SEareshownforeachquadrat
2.0
2.5
3.0
1
2
Vegetation cover (%
)
Bottom
Quadrat height
Middle Top
Species richness
Bottom
Quadrat height
Middle Top FIGURE4 Top:Cliffheight(i.e.quadrat)effectonspecies
richnessoneachgroupofspeciesaccordingtotheirrockassociation;
Bottom:Cliffheight(i.e.quadrat)effectonthepercentageof
vegetationcoveroneachgroupofspeciesaccordingtotheirrock
association.Means±SEareshownforeachspeciesgroup
|
7
Applied Vegetation Science
MARCH- SALAS et AL.
to Müller etal. (2004), we found these effects were mostly limited
togeneralistspecies, whilerockspecialistspecieswerenotaffected.
Toourknowledge,ourcase-controlmethodologicalapproachhasnot
previously been implemented in these environments. Furthermore,
thestudywasconductedinaMediterraneancliffenvironment,oneof
thelessstudiedecosystemsinthistopic.
The higher tolerance capacities of rock specialist plants to
climbing-relateddisturbancemayberelatedtotheiradaptationstothe
extremeandlimitingconditionsofcliffecosystems(García,Espadaler,
&Olesen,2012).DeMiccoandAronne(2012)foundthatspeciesen-
demictotheseecosystemspossessadaptivestructuraltraitsanddis-
playspecificbehaviourstohelpcopewiththepressuresimposedby
extremecliffenvironments.Typically,rockspecialistplantsareableto
colonizebarerock(Escudero,1996),establishinginmicrositeswhere
rootssystemscan find anchorage (Matthes-Sears& Larson, 1995).
This could help them to resist and withstand rock-climbing distur-
bance.Additionally,somerupicolousplantshavemechanismstotake
upwatermoreefficiently,includingfromdriersubstrates(Gouvra&
Grammatikopoulos,2007).Although the majorityofthese traits are
relatedtotheregulationofattachmenttothesubstrate,wateruptake
andstorage,someadvantageousreproductivestrategiescanalsobe
found.Forinstance,S. enneaphyllausesantsas pollinatorsand seed
dispersers(Davis,1951;Garcíaetal.,2012).Overall,suchstrategies
provide advantages to rock specialist plants, potentially affording
themhighertolerancetohumanperturbations.
Tofreeupcracksforhand-andfootholds,climbersoftenremove
someplants present along the rock-climbingroute.Thiscan reduce
the substrate presentwithin these cracks, and alter the seed bank.
Theseactionswouldhaveastrongerinfluenceonthegeneralistspe-
ciespresentthat requirea richer substratewith more resources for
establishment. Our results confirmedthe importance of microtopo-
graphicconditionsasimportantdriversofdiversitypatterns,inagree-
mentwithotherstudies(DoCarmo &Jacobi,2016;Kuntz&Larson,
2005, 2006a,b; Lundholm & Larson, 2003). While specialized rock
specieswereable tosurvivewithout ahighnumberofcracks,non-
specializedrockspeciesweregreatlyaffectediftheamountofthese
microtopographicfeaturesdecreased.However,slopeandaspectdid
notsignificantlyinfluencevegetationcompositioninourstudy.
Thelowcoverandrichnessofgeneralistspeciesinthelowestparts
ofthecliffface(i.e.bottom;Mülleretal.,2004)comparedwithhigher
parts(i.e.middleandtop)couldstillberelatedtoclimber-drivendistur-
bance.Inthe bottompart ofthecliff,generalistspecies richnessand
vegetation cover were similar in the climbed and unclimbed transects.
Nevertheless,therecould be multiple reasonsforthis result. When
climbersstart to ascend a rockclimbingroute,theyusuallyevaluate
routeoptionsandwillclearsubstrateanddustaccumulatedonseveral
potentialholdssuitableforvegetationestablishment.Inaddition,the
bottompartofthecliffisoftenusedforbouldering(i.e.rockclimbing
withoutropesorharnesses).Bothactivities can result inthefurther
reductionofsubstrateinrockcracks,whichmayleadtoadecreasein
generalistspecies,asfoundinourstudy.Herbivorycouldalsoexplain
thereducedpresence of species at the bottom of a cliff face;while
specialized rock species have some traits and defence systems to
reduceherbivory(Aronne,DeMicco,&Barbi,2010),generalistplants
may be more vulnerable.
Ourfindings showedthatthe degree ofspecies dependence on
rockyhabitatplaysanimportantroleintheirresistancetorockclimb-
ingpressures.Species with less specializationtocliffconditions(i.e.
lessrockadaptationstrategies)weremorevulnerabletorockclimbing
effects.OurvegetationsurveyshowedthatthecliffsofChulillamostly
havenon-specializedrockspecies (>70%ofspeciesfound),in accor-
dancewith Mülleretal. (2004).Therefore,theloss ofthese species
willdrasticallyreducespecies diversityinthesesingular ecosystems,
as well as change the composition and structureof the plant com-
munity.Theassociateddeclineinvegetationcovermayalsoendanger
othercharacteristicsofthesehabitatsbypromotingerosiveprocesses.
Thisresearchadditionallyrepresentsastepforwardinecological
researchon cliffs, andparticularlyareaswith rockclimbingactivity.
Inourcase-controldesign, we examined the impacts of rockclimb-
ing using a spatial approach, but also provided a starting pointfor
FIGURE5 Top:Rockclimbingeffectonspeciesrichnessoneach
groupofspeciesaccordingtotheirrockassociation;Bottom:Rock
climbingeffectonthepercentageofvegetationcoveroneachgroup
ofspeciesaccordingtotheirrockassociation.Means±SEareshown
foreachspeciesgroup
8
|
Applied Vegetation Science MARCH- SALAS et AL.
monitoringseries,which might allowstudyofthe long-term effects
ofrockclimbingonplantcommunitieslivingoncliffs.Periodicallyre-
peatingthe surveyinsituonlyrequiresa detaileddescriptionofthe
placementof surveyunitsalongeachrock-climbingroute.As far as
weknow,ourcharacterizationofthestudyareainChulillarepresents
thefirstattempttoinstigatelong-termresearchmonitoringofrock
climbingeffects.Ourcase-controldesignincludesacentralsurveying
transectlocatedwheretheintensityofrockclimbingismaximal,and
twocontiguouslateraltransectsascontrolpoints.Still,notallclimbers
followexactlythesamelaneacrossarock-climbingroute.Therefore,
special attention must be paid to the placement ofthe in situ sur-
veytransectsin orderto optimizethecontrolfunctionofthelateral
transects.Followingourexperienceanddata,wesuggestsurveying
transects5-mwide×3-mhigh(insteadof3m×3m)mightincrease
the potential ofthe analysis. This would calls for higher separation
betweenthecontrolsandclimbedtransect,whichcouldremovenoise
inthedataacquisitionfromcasualascentdeviations,ascontrolpoints
cannot be considered completelyundisturbed. In addition, we also
recommendseparateanalysesforthelowerpartsoftherock-climbing
routes,astheseappeartobemorecomplexareaswithotherecologi-
calinteractionsapparentlymaskingthedirecteffectsofrockclimbing.
Inconclusion,despite the strong influence of physical and envi-
ronmentalfactors,increasingpopularityofrockclimbingcouldbean
importantdeterminantincliff-facefloralstructure,since italtersthe
normaldevelopmentofplantsinsuchenvironments.Werecommend
thatfuturemanagementandconservationplansand studiesaccount
not only forendemic species, which are often more protected, but
alsoforgeneralistspecies.This,togetherwithstandardizationofthe
surveydesignbasedon ourproposal,wouldhelpeffortstopreserve
localbiodiversityandtofullyunderstandthepotentialimpactsofrock
climbingactivityoncliffhabitats.
ACKNOWLEDGEMENTS
We thank the Polytechnic University of Valencia and Botanical
GardenoftheUniversityofValenciaforinstitutionalsupport.Weare
gratefultoOlgaMayoralandJaimeGüemesforfruitfuldiscussionson
thesurveymethodsand experimentaldesign,aswelltheir latersug-
gestions.Wealsothanktherefereesandeditorsfortheirsuggestions.
AUTHOR CONTRIBUTIONS
MMS and MMM designed the study. MMM designed the survey
methodology and implemented the study, collecting all the data.
MMSandGPcarriedoutthestatisticalanalysis.MMS,MMM,EHand
LRPgeneratedthefiguresofthemanuscript.MMSandMMMwrote
thefirst draftofthemanuscript. MMS, MMM,GP,PTI,EH andLRP
revisedthemanuscript.
REFERENCES
Adams,M. D.,&Zaniewski,K. (2012). Effectsofrecreationalrockclimb-
ing and environmental variation on a sandstone cliff-face lichen
community. Botany- Botanique,90,253–259. https://doi.org/10.1139/
b11-109
Agulló,J. C., Alonso, M. Á., Juan,A.,Villar, J. L., & Crespo, M. B. (2010).
Apreciaciones sintaxonómicas y nomenclaturales sobre las comuni-
dadesrupícolasdelaasociaciónJasonio glutinosae-Teucrietum thymifolii.
Lazaroa,31,155–163.https://doi.org/10.5209/rev_LAZA
Aronne,G.,DeMicco,V.,&Barbi,S.(2010).Hypocotyl features of Primula pa-
linuri Petagna (Primulaceae), an endemic and rare species of the Southern
Tyrrhenian Coast.InG.Giordani,V.Rossi&P.Viaroli(Eds.),Proceedings
of the Ecologia Emergenza Pianificazione, 18th Congresso Nazionale della
Societa Italiana di Ecologia (SItE ‘10),pp.113–119.
Bates,D.,Mächler,M., Bolker,B.M., &Walker,S.C. (2015).Fitting linear
mixed-effects models using lme4. Journal of Statistical Software, 67,
1–48.
Castroviejo,S.(1986–2012).Flora ibérica: Plantas vasculares de la Península
Ibérica e Islas Baleares.Madrid,Spain:C.S.I.C.RealJardínBotánico.
Clark, P., & Hessl, A. (2015). The effects of rock climbing on cliff-face
vegetation. Applied Vegetation Science, 18, 705–715. https://doi.
org/10.1111/avsc.12172
Cordell,H.K.(2012).Outdoor recreation trends and futures: a technical docu-
ment supporting the Forest Service 2010 RPA Assessment.
Davis,P.H.(1951).CliffvegetationintheEasternMediterranean.Journal of
Ecology,39,63–93.https://doi.org/10.2307/2256628
De Micco, V., & Aronne, G. (2012). Occurrence of morphological and
anatomical adaptive traits in young and adult plants of the rare
Mediterranean cliff species Primula palinuri Petagna. The Scientific
World Journal,2012,471814.https://doi.org/10.1100/2012/471814
DoCarmo,F.F.,DeCampos,I. C.,&Jacobi,C.M.(2016). Effectsoffine-
scalesurface heterogeneityon rock outcrop plant community struc-
ture. Journal of Vegetation Science,27,50–59.https://doi.org/10.1111/
jvs.12342
DoCarmo,F.F.,&Jacobi,C.M.(2016).Diversityandplanttrait–soilrela-
tionshipsamongrockoutcropsintheBrazilianAtlanticrainforest.Plant
and Soil,403,7–20.https://doi.org/10.1007/s11104-015-2735-7
Escudero, A. (1996). Community patterns on exposed cliffs in a
Mediterraneancalcareous mountain. Vegetatio,125, 99–110.https://
doi.org/10.1007/BF00045208
Farris, M. (1995). The Effects of Rock Climbing on the Cliff Flora of Three
Minnesota State Parks. Conservation Biology Research Grants Program,
Division of Ecological Services, Final report to the Minnesota department
of natural resources, US.
García, M. B., Espadaler, X., & Olesen, J. M. (2012). Extreme repro-
duction and survival of a true cliffhanger: The endangered plant
Borderea chouardii(Dioscoreaceae).PLoS ONE,7,e44657.https://doi.
org/10.1371/journal.pone.0044657
Gouvra,E.,&Grammatikopoulos,G.(2007).Diurnalandseasonaltrendsof
waterrelationsinfive co-occurring chasmophyticspecies.Flora,202,
237–248.https://doi.org/10.1016/j.flora.2006.06.005
Groombridge,B.,&Jenkins,M.D.(2002).World atlas of biodiversity: Earth’s
living resources in the 21st Century.Oakland,CA:UniversityofCalifornia
Press.
Hespanhol,H.,Séneca,A.,Figueira,R.,Sérgio,C.,Taylor,P.,Hespanhol,H.,
…Sérgio, C. (2011). Microhabitat effects on bryophytespeciesrich-
nessandcommunitydistributiononexposedrockoutcropsinPortugal.
Plant Ecology & Diversity,4,251–264.https://doi.org/10.1080/175508
74.2011.616546
Holzschuh, A. (2016). Does rock climbing threaten cliff biodiversity? A
critical review. Biological Conservation, 204, 153–162. https://doi.
org/10.1016/j.biocon.2016.10.010
Knapp,A.K.,Smith,M.D.,Hobbie,S.E.,Collins,S.L.,Fahey,T.J.,Hansen,
G.J.A.,…Webster,J.R.(2012).Past,present,andfuturerolesoflong-
term experiments in the LTER Network. BioScience, 62, 377–389.
https://doi.org/10.1525/bio.2012.62.4.9
Kuntz, K. L., & Larson, D.W. (2005). The relative influence of microhabi-
tat constraints and rock climbing disturbance to vegetation on Ontario’s
|
9
Applied Vegetation Science
MARCH- SALAS et AL.
Niagara Escarpment. In Proceedings of the Parks Reserve Forum of Ontario,
Guelph, Canada,pp.295–308.
Kuntz, K. L., & Larson, D. W. (2006a). Influences of microhabitat con-
straints and rock-climbing disturbance on cliff-face vegetation
communities. Conservation Biology, 20, 821–832. https://doi.
org/10.1111/j.1523-1739.2006.00367.x
Kuntz,K.L.,&Larson,D.W.(2006b).Microtopographiccontrolofvascular
plant,bryophyte andlichencommunitieson cliff faces.Plant Ecology,
185,239–253.https://doi.org/10.1007/s11258-006-9101-z
Larson,D.W.,Matthes,U.,& Kelly,P.E. (2000).Cliff ecology: Pattern and
process in cliff ecosystems.Cambridge,UK:CambridgeUniversityPress.
https://doi.org/10.1017/CBO9780511525582
Lenth,R.V.(2016).Least-squaresmeans:TheRpackage{lsmeans}.Journal
of Statistical Software,69,1–33.
López-Pujol,J.,Álvarez,N., Bosch, M.,Simon,J.,&Blanché,C. (2006).
Allozyme variation and taxonomical implications of the en-
demic rocky plant Erodium rupestre (Geraniaceae). Biochemical
Systematics and Ecology, 34, 219–230. https://doi.org/10.1016/j.
bse.2005.11.002
Lorite,J.,Serrano,F.,Lorenzo,A.,Cañadas,E.M.,Ballesteros,M.,&Peñas,
J.(2017). Rock climbing alters plant species composition,cover,and
richnessin Mediterranean limestonecliffs.PLoS ONE,12, e0182414.
https://doi.org/10.1371/journal.pone.0182414
Lugo,A. E., Scatena,F.N., Waide, R. B., Greathouse, E. A., Pringle,C.
M., Willlg, M. R., … Thompson, J. (2012). Management implica-
tions and applications of long-term ecological research. In N.
Brokaw, T. Crowl & A. Lugo (Eds.), A Caribbean forest tapestry:
The multidimensional nature of disturbance and response (p. 305).
Oxford, UK: Oxford University Press. https://doi.org/10.1093/
acprof:osobl/9780195334692.001.0001
Lundholm, J. T., & Larson, D. W. (2003). Relationships between spa-
tial environmental heterogeneity and plant species diversity
on a limestone pavement. Ecography, 26, 715–722. https://doi.
org/10.1111/j.0906-7590.2003.03604.x
Matthes-Sears,U.,&Larson,D.W.(1995).Rootingcharacteristicsoftrees
inrock:AstudyofThuja occidentalisonclifffaces.International Journal
of Plant Sciences,156,679–686.https://doi.org/10.1086/297290
McMillan, M. A., & Larson, D. W. (2003). Effects of rock climbing on
the land snail community of the Niagara Escarpment in southern
Ontario, Canada. Conservation Biology, 17, 616–621. https://doi.
org/10.1046/j.1523-1739.2003.01362.x
Müller,S. (2006).Human impact on the vegetation of limestone cliffs in the
northern Swiss Jura mountains.Dissertation,UniversityofBasel,Basel,
Switzerland.
Müller,S.W.,Rusterholz,H.-P.,&Baur,B.(2004).Rockclimbingaltersthe
vegetation of limestone cliffs in the nothern SwissJura Mountains.
Applied Vegetation Science,7,35–40.
Nuzzo, V. A. (1996). Structureof cliff vegetation on exposed cliffs and
theeffect ofrock climbing.Canadian Journal of Botany,74,607–617.
https://doi.org/10.1139/b96-077
RoldánGutierrez, J. A. (2013). Chulilla climbing guidebook. Madrid,Spain:
Techrock.
Rusterholz,H.-P.,Müller,S. W., & Baur,B.(2004).Effectsofrock climb-
ing on plant communities on exposed limestone cliffs in the Swiss
Jura Mountains. Applied Vegetation Science, 7, 35–40. https://doi.
org/10.1111/j.1654-109X.2004.tb00593.x
Santisteban, C. (2012). Geolodia: El doble meandro encajado del río Turia.
Madrid,Spain:SociedadGeológicadeEspaña.
Surina,B.,&Martinčič,A.(2014).EcologyandnicheassemblyofCampanula
tommasiniana,anarrowendemic of Mt Ucka (Liburnian karst, north-
westernAdriatic).Acta Botanica Croatica,73,221–254.
Tejedo, P., Benayas, J., Cajiao, D., Albertos, B., Lara, F., Pertierra, L.
R., … Reck, G. K. (2016). Assessing environmental conditions of
Antarctic footpaths to support management decisions. Journal of
Environmental Management, 177, 320–330. https://doi.org/10.1016/
j.jenvman.2016.04.032
Tejedo,P.,Pertierra,L.R.,Benayas,J.,Convey,P.,Justel,A.,&Quesada,A.
(2012).TramplingonmaritimeAntarctica:Cansoil ecosystemsbeef-
fectivelyprotectedthroughexistingcodesofconduct?Polar Research,
31,1–13.https://doi.org/10.3402/polar.v31i0.10888
Thiel,H.,&Spribille,T.(2007).Lichensandbryophyteson shaded sand-
stone outcrops used for rock climbing in thevicinity of Göttingen
(southernLowerSaxony,Germany).Herzogia,20,159–177.
Vogler, F., & Reisch, C. (2011). Genetic variation on the rocks –
the impact of climbing on the population ecology of a typical
cliff plant. Journal of Applied Ecology, 48, 899–905. https://doi.
org/10.1111/j.1365-2664.2011.01992.x
SUPPORTING INFORMATION
Additional Supporting Information may be found online in the
supportinginformationtabforthisarticle.
Figure S1Percentagevegetationcoverofunclimbedandclimbedtran-
sects with specialized rock species, moderatelyassociated to rocky
habitatspeciesandgeneralistspecies
Table S1Differentrockclimbingroutessampled,with somephysical
parameters:difficulty,aspect(S:South;SE:Southeast;SW:Southwest;
E:East),slopeaverage(grades)andheightofroute(m).Fromeachcliff,
weincluded latitude and longitude. Fornumeric difficulties weused
theYosemiteDecimalSystem(YDS) fromtheUSA,andalso catego-
rizedifficultyleveloftheclimbingrouteas‘intermediate’or‘beginner’
(‘experienced’climbingrouteswerenotsampled)
Table S2Speciesfoundindifferentsampledcliffs.Wedividedspeciesby
degreeofdependenceontherockyhabitat(i.e.rockassociation:special-
izedrockspp. (S),moderatelyassociatedto rockyhabitatsspp.(M) and
generalistspp.(G)).Taxonomicgroupandfamilyofeachspeciesincluded.
Table S3Results of LMM testing the effects of: slope, cliff height,
climbingroutedifficultyandaspect.Theeffectsweretestedforoverall
speciesandfordifferentspeciesgroupsaccordingtotheirassociation
withrockyhabitats.Thesevariableswerenotsignificantinanymodel,
soallwereexcludedfromtheTable1
Table S4Presence(i.e.numberofindividualsperspecies)andvegetation
cover(%)ofeachspeciesinunclimbed(U)andclimbed(C)transects
How to cite this article:March-SalasM,Moreno-MoyaM,
PalomarG,Tejero-IbarraP,HaeuserE,PertierraLR.An
innovativevegetationsurveydesigninMediterraneancliffs
showsevidenceofhighertoleranceofspecializedrockplants
to rock climbing activity. Appl Veg Sci. 2018;00:1–9.
https://doi.org/10.1111/avsc.12355
