Effects of Afforestation on Plant Diversity and Soil Quality in Semiarid SE Spain
Abstract
INIA, grant number RTA2005‐00009‐C02‐01
Full text
Forests2021,12,1730.https://doi.org/10.3390/f12121730www.mdpi.com/journal/forests Article EffectsofAfforestationonPlantDiversityandSoilQualityin SemiaridSESpain CarmenSegura 1 ,MaríaN.Jiménez 2 ,EmiliaFernández‐Ondoño 3 andFranciscoB.Navarro 4, * 1 DepartmentofSustainableAgricultureSciences,RothamstedResearch,NorthWyke, Okehampton,DevonEX202SB,UK;carmen.segura‐[email protected] 2 DepartmentofBotany,FacultyofPharmacy,CampusdeCartuja,UniversityofGranada, 18071Granada,Spain;[email protected] 3 DepartmentofSoilScienceandAgriculturalChemistry,FacultyofScience,UniversityofGranada, C/SeveroOchoa,s/n,18071Granada,Spain;[email protected] 4 AreaofAgricultureandEnvironment,InstituteofAgriculturalResearchandTrainingofAndalusia (IFAPA),GovernmentofAndalusia,CaminodePurchils/n,18004Granada,Spain *Correspondence:[email protected] Abstract:Farmlandafforestationhasbeenpromotedinrecentdecadesandisoneofthemainstrat‐ egiesincludedintheUNDecadeonEcosystemRestorationtorecoverdegradedareas.However, theimpactsofafforestationonplantdiversityandsoilqualityindicatorsarestillnotwell‐under‐ stoodinsemiaridenvironments.Inthisstudy,weassessedtherelationshipsbetweenplantdiversity indicators(abundance,totalrichness,richnessbyfunctionalgroups,andShannondiversity)anda largenumberofvariablesin48afforestationsitesinsoutheastSpain.Weconsideredassociateden‐ vironmentalfactors,suchasgeographical,climaticoredaphicvariables,age,andland‐usehistory. Wecomparedplantdiversityandsoilpropertiesfollowingland‐usechangefromcerealcropping toafforestation,whichisoneofthemostcommonland‐usechangesinMediterraneanareas.Plant diversityinafforestedsiteswasfoundtobedependentonpreviouslanduse,theproximityofnat‐ uralvegetation,severalsoilproperties(texture,pH,andtotalnitrogen),andplantationage.Affor‐ estedsoilsshowedhigherplantdiversityandanimprovementinedaphicparametersrelatedto multifunctionalityinsemiaridecosystems(i.e.,soilorganiccarbon,nitrogen,andpotassium)than arablecroppedsoils. Keywords:activerestoration;functionalgroups;plantrichness;soilproperties;land‐usechange; Mediterraneanregion;UNDecadeonEcosystemRestoration 1.Introduction Inrecentdecades,variousglobalandregionalinitiativeshavepromotedactiveenvi‐ ronmentalrestorationasoneofthemainworldwidenature‐basedsolutionstoprevent soilerosion,recoverdegradedlands,andlimittheeffectsofclimatechange[1].Inthis sense,theareathathasbeenreforested(plantingtreesinpreviouslyforestedlandscapes) andafforested(plantingtreesinlandscapesthatwerenotpreviouslyforested)hasin‐ creasedbyover123millionhasince1990,andcurrentlyrepresent7%oftheworld’sforest area[2].Scanteffortshavebeendevotedtomonitoringthesuccessoftreeplantingin achievingobjectivesandassessingthesuccessofrestoration[3].TheUnitedNationsGen‐ eralAssemblydeclared2021–2030astheDecadeonEcosystemRestoration,withanaim ofplantingbillionsoftreesthroughrestorationinitiativesaroundtheworld[3].Thisstrat‐ egyisderivedfromtheSustainableDevelopmentGoalsAgenda(i.e.,SustainableDevel‐ opmentGoal15.3),addressingbothclimatechangemitigationandbiodiversityandeco‐ systemservices[4]. Citation:Segura,C.;Jiménez,M.N.; Fernández‐Ondoño,E.;Navarro, F.B.EffectsofAfforestationonPlant DiversityandSoilQualityin SemiaridSESpain.Forests2021,12, 1730.https://doi.org/10.3390/ f12121730 AcademicEditor:BartoszAdamczyk Received:13October2021 Accepted:6December2021 Published:8December2021 Publisher’sNote:MDPIstaysneu‐ tralwithregardtojurisdictional claimsinpublishedmapsandinstitu‐ tionalaffiliations. Copyright:©2021bytheauthors.Li‐ censeeMDPI,Basel,Switzerland. Thisarticleisanopenaccessarticle distributedunderthetermsandcon‐ ditionsoftheCreativeCommonsAt‐ tribution(CCBY)license(https://cre‐ ativecommons.org/licenses/by/4.0/).
Forests2021,12,17302of17 Thisnewemphasisontreeplantingisoccurringwithinacontextofintensedebate regardingthesuitabilityofadoptingpassiveoractiverestorationstrategies.Therelative effectivenessofactiverestorationstrategiesrelativetopassiverecoveryofformeragricul‐ turallandforecosystemfunctionsisnotwell‐understood[5].Ithasbeenproposedthat theactiveafforestationofdegradedfieldsoragriculturalareasshouldonlybeadopted whenecosystemrestorationmaybeimpossiblepassively[6–8].Forsomeauthors,affor‐ estationisoftentheonlyviableapproachtopreventsoilerosionandrestoreecosystem functions,andotherspromoteactiveinterventionasanopportunityforclimatechange mitigationthatcouldhavepositiveeffectsonbiodiversity,particularlywhennativespe‐ ciesareplanted[9–11]. Theinfluenceofafforestationonplantdiversityandsoilqualityisdependenton land‐usehistory,whethertreespeciesareplanted,thetimeelapsedsincerestorationbe‐ gan,andlandscapecontext,amongstothersenvironmentalvariables[6,12–15].Studies conductedunderdryconditionsthatincorporateafforestationsuccessconsideringboth plantdiversityandsoilqualityrecoveryarescarce,eventhoughsemiaridandsubhumid areasaccountfor25%oftheterrestrialEarth,andahugeagriculturalarea(cultivatedor abandoned)hasbeenafforestedinrecentdecades.Theseregionsareconsideredextremely vulnerabletolanddegradationanddesertification[2,10,16].Previouslanduseisakey driverofplantdiversityandsoilqualitychangesfollowingafforestation[5,17].However, somestudiesconductedinsemiaridthatincludedbothplantdiversityandsoilhavefo‐ cusedongrasslandsandnaturalvegetationinsteadoffarmlandafforestation[18,19]. Otherstudiesthatreportedpositivelinksbetweenplantdiversityandedaphicproperties, suchassoilorganiccarbon(SOC),totalnitrogen(TN),phosphorus(P),andpotassium(K), havebeenconductedinsecondaryforests,abandonedlands,andaromaticspeciesculti‐ vation[20–22]. Studiesthatreportsoilqualityimprovementfollowingafforestationofarableland didnotconsiderplantdiversity[23,24].Abandonmentofagriculturallandpresentsop‐ portunityfortherecoveryofsoilpropertiesrelatedtofertility,andanincreaseinbiodi‐ versitywithlittleneedforactiverestorationandatlessexpense[25].Understanding whichedaphicandotherenvironmentalvariablesexplainplantdiversityundersemiarid conditions,particularlyinafforestationprojects,couldsupportanydecision‐makingpro‐ cesses.However,despitesoilbeingrecognizedasamaindriverofplantdiversity,the extenttowhichedaphicparameterscontrolplantdiversityisstillnotwell‐known[26]. Theuncertaintiesrelatedtotheeffectsofafforestationonplantdiversityandsoil haveimportantimplicationsforEuropeanUnionCommonAgriculturalPolicysubsidies [27–29].EuropeanafforestationeffortshavebeenincludedintheCommonAgricultural Policysince1992,mainlyinMediterraneancountriessuchasSpain,Portugal,andItaly [30].Underthisframework,theFarmlandAfforestationSchemeaimedtoreduceproduct surpluses,promoteearlyretirementopportunitiesforfarmers,increaseforestresources, andprovidesoilandbiodiversityprotectionbyofferingfarmersannualincentivepay‐ mentsfortheconversionoffarmlandtoafforestation[31].Morethan730,000haoffarm‐ landhasbeenafforestedinSpain,approximately50%ofwhichwaspreviouslydevoted toarablecrops[32].Thelackofplanningandmeasurableobjectiveshasresultedinaffor‐ estedsitesspreadhaphazardly,notfollowinganytechnicalorterritorialcriteria.Thishas complicatedthewide‐scalemonitoringofafforestationimpactsonbiodiversityandsoil quality,especiallywhenbaselineassessmenthasnotbeenestablished[33,34]. Withananticipatedexpansionofafforestedlandbytheendofthe21stcentury,itis importanttoensurenotonlytheaccuratemanagementofexistingafforestation,butalso thecarefulplanningofnewlyplantedforeststomaximizethebenefitsforbiodiversityand sustainability[1,3,35].Anaccurateassessmentofsoilchangefollowingafforestation shouldalsobeincludedasaprincipalcomponentofrestorationmonitoring[36,37],even thoughmanyfactorsareinvolvedandrelationshipsbetweenplantdiversityandsoilmay
Forests2021,12,17303of17 notbeevident[26,38,39].Identifyingenvironmentalfactorsthatindicateafforestationsuc‐ cessaswellasquantifyingchangesinplantdiversityandsoilqualityindicatorsfollowing plantingwillhelpimproverestorationsuccess. Inthisstudy,weaimedto(i)elucidatewhichenvironmentalvariablescanexplain plantdiversity,and(ii)compareplantdiversityandedaphicpropertiesinpairedcereal cropsandafforestationsites.Wefocusedonplantspeciesrichness,plantdiversity,and soilqualityindicatorsinatypicalsemiaridMediterraneanarea.Speciesrichnessanddi‐ versityareamongtheprincipalmetricschosentoassessthesuccessofrestorationinitia‐ tivesfromabiodiversitypointofview:theyaredriversandpredictorsofecosystemmul‐ tifunctionality(understoodastheabilityofecosystemstoprovideandmaintainmultiple functionsandservicessimultaneously),forinstance,bybufferingtheeffectsofclimate changeanddesertificationinsemiaridclimates,andbyinfluencingnutrientcyclingand carbonstorage[40,41].Somerecommendationsforevaluatingplantdiversityrestoration haveencouragedtheassessmentoffunctionalgroups[36,38].Identificationofendemism levelsandconservationstatusiscrucialtotheevaluatingoftheactiverestorationeffects onbiodiversity[12],especiallyinbiodiversityhotspotssuchastheMediterraneanregion. Wehypothesizedthat(1)previouslanduse,location,climate,andedaphicproper‐ ties,suchastexture,SOC,macronutrients,andparametersrelatedtosoilwatercontent, canexplainplantdiversityinMediterraneansemiaridafforestation;and(2)afforestation improvesplantdiversityandsoilqualityindicators,comparedtocerealcroppinginsem‐ iaridenvironments.Weexploredenvironmentalvariablesthatcouldexplainplantdiver‐ sityinafforestedfarmlandsusinggeneralizedlinearmodels,quantifyingandcomparing plantdiversityandsoilqualityindicatorsinpairedafforestationandcerealcropsites.To solvethelackofbaselinesamplingbeforeafforesting,weselectedcerealcropsascontrol sitesbecausehalfoftheafforestedareainthestudyregionwasarableland[6,27,32]. 2.MaterialsandMethods 2.1.StudyArea ThestudyarealiesinthenortheastoftheProvinceofGranada(southeastIberian Peninsula)withintheconfinesoftheBaeticmountainsystem(Figure1).Itextendsover 5220km2andincludestheadministrativeregionsofGuadix,Baza,andHuéscar.Within thisarea,atotalof241farmswereafforestedbetween1993and2006.Thesizeofthefarms variedbetween1.15and97.73ha(averageof26.85ha±1.36SE).Themostcommontree specieswerePinushalepensisMill.plantedinsingleormixedstandswithQuercusilexL. subsp.ballota(Desf.)Samp.Thesewereplantedatlowdensities(300–500stemha−1)using thelinearsubsoilingtechnique.Nosilviculturaltreatmentsorgrazingwereadoptedin theafforestedfarmlandsovertime.Weselectedatotalof48oftheseafforestedfarmsfor thisstudyaccordingtocriteriaofrepresentativenessofthethreeadministrativeregions, accessibility,andeconomiccostrelativetosamplingeffortandsoilanalysis(Figure1). TheregionhasaMediterraneanmacroclimatewithinmeso‐,supra‐,andoromediter‐ raneanthermotypesandsemiarid,dry,andsubhumidombrotypes[42].Themostfre‐ quentsoilsareLeptosols(LithicLeptosolsandRendzicLeptosols),Calcisols(PetricCal‐ cisolsandHaplicCalcisols),Cambisols(EutricCambisolsandCalcaricCambisols),and Regosols(LepticRegosols,CalcaricRegosolsandEutricRegosols).Therearealsosmall areasofCalcaricPhaeozems[43].
Forests2021,12,17304of17 Figure1.Studyareaandsitesincluded.Purplecirclesindicatethelocationofafforestedsitessampledin2006(N=43). Greenpinsindicatethelocationofpairedsitessampledin2006–2007toperformthediversitycomparisonbetweenaffor‐ estation(N=6)andcerealcrops(N=6).Blackcrossesindicatethelocationofpairedsites,afforestation(N=18),andcereal crops(N=18)wheresoilpropertiesweremeasuredin2011. 2.2.DataCollection Fromthe48farmlandsafforestation,weselected43sitesforthestudyofexplanatory environmentalvariablesforplantdiversity(Table1andFigure1;Supplementarymaterial TableS1).Withineachsite,asampleplot(20×20m)wasidentified,atleast50maway fromtheedgetoavoidanyedgeeffects.Foreachplot,five20mlonglineartransectswere selected,separatedfromeachotherby4m,andalignedperpendiculartofurrowsmade duringplanting. Thedataofvascularplants(includingafforestedspecies)wererecordedwherethe tipofapointertouchedthevegetationperpendicularlyevery100cmalongtransects,ina mannersimilartothepointquadrattechnique(cf.[44]).Weestimatedspeciesabundance (numberofindividualsofeachspeciesperplot)andrelativecover(%)oftheaerialparts ofeachspecies.Baresoil(%)wasalsoestimated.Allspecieswereidentifiedandclassified intofunctionalgroups:annuals(includingannualorbiennialplants),perennialforbs,per‐ ennialgrasses,andwoodyspecies(includingdwarfscrubsandshrubs).Conservationsta‐ tuswascheckedtoquantifyendemismandtoidentifyendangeredspecies,accordingto Cabezudoetal.[45]. Plantdiversitywasconsideredintermsofspeciesrichnessandspeciesdiversity.Spe‐ ciesrichnessisreportedasthetotalnumberofspeciesperplot(Total_R)andasrichness byfunctionalgroup(Annuals_R,Forbs_R,Grasses_R,andWoody_R).Speciesdiversity wascalculatedastheShannon–Wienerindex(H’): H’=−Σp i lnp i ,(1) wherep i istheabundanceofiexpressedasaproportionofthetotalspeciesabundancein eachafforestedplot. Differencesinplantdiversitybothincludingandexcludingafforestedspecieswere examined(Total_HandH_noaff,respectively).
Forests2021,12,17305of17 Table1.Environmentaldatacollectedforeachfarmlandafforestation(N=43).*Nominalvariables. VariableTypeVariableUnitMean(SD)MinimumMaximum GeographicElevationabovesealevel(Elevation)m1237.0(304.2)881.01939.0 Distancetocrops(D_crops)m188.5(308.0)0.01488.0 Distancetonaturalvegetation(D_n_veg)m61.9(133.7)0.0531.5 Distancetoshrubpatches(D_shrub)m1241.0(1022.0)90.54675.0 DistancetoQuercuswoodland(D_Qil)m1727.0(2219.0)0.09834.0 ClimaticAnnualPrecipitation(Pp)mm394.0(65.2)311.0547.6 MeanTemperature(T)°C12.68(0.98)10.9014.10 EdaphicAvailablewater(AW)%6.82(2.90)2.2212.74 Soilmoistureatfieldcapacity(33kPa) (SM33)%16.24(5.78)5.0928.58 Soilmoistureofthewiltingpoint(1500 kPa)(SM1500)%9.42(3.81)2.8719.02 Watersaturation(WS)%38.39(7.59)27.6356.42 Gravel(>2mm)%37.24(18.02)0.0073.42 Sand(2–0.05mm)%45.28(16.75)3.2375.65 Coarsesilt(0.05–0.02mm)(CSilt)%10.27(4.35)3.0724.67 Finesilt(0.02–0.002mm)(FSilt)%22.15(9.66)8.3757.25 Clay(<0.002mm)%22.30(8.99)6.1238.47 Cationexchangecapacity(CEC)Cmol(+) kg−111.28(6.13)3.2725.39 Electricalconductivity(EC)mS cm−10.89(0.40)0.512.21 Exchangeablesodium(Na)Cmol(+) kg−10.06(0.07)00.29 Exchangeablemagnesium(Mg)Cmol(+) kg−11.86(1.28)0.305.98 Exchangeablepotassium(K)Cmol(+) kg−10.36(0.27)0.061.28 Solublecalcium(s_Ca)mgL−13.71(2.41)0.679.81 Solublemagnesium(s_Mg)mgL−10.65(0.47)0.162.59 Solublepotassium(s_K)mgL−10.40(0.51)0.023.05 Solublesodium(s_Na)mgL−11.61(0.98)0.134.78 CaCO3(CaCO3)%24.36(23.24)0.1073.78 pH7.90(0.55)6.728.83 Soilorganiccarbon(SOC)%1.24(0.82)0.303.19 Totalnitrogen(TN)%0.11(0.05)0.050.24 NO2−(NO2)mgL−13.29(8.65)0.0041.60 NO3−(NO3)mgL−111.10(44.41)0.00233.40 SO42−(SO4)mgL−146.60(64.58)6.81301.50 Cl−(Cl)mgL−130.25(22.01)10.53117.2 GeologicalLithology* LanduseAgeofafforestation years8.4(2.1)5.012.0 Previouslanduse(cerealcroporold‐ field)* Plantcover(Cover)%67.49(13.15)40.0091.00 CoverofPinushalepensis(CovPha)%5.93(6.44)0.0032.00 CoverofQuercusilexsubsp.ballota (CovQil)%0.26(0.62)0.003.00
Forests2021,12,17306of17 Theedaphiccharacteristicsandgeographic,climatic,geological,andlandusevaria‐ blesforeachafforestedsitearesummarizedinTable1.Compositesoilsampleswerecol‐ lectedfromthecenterandcornersofeachquadrattoadepthof0–10cmandmixedto obtainonerepresentativesampleperplot.Soilsampleswereair‐driedandsieved(<2 mm).Gravels(>2mm)wereweighedandstoredseparately.Soiltexturewasanalyzedby theRobinsonpipettemethod[46].Availablewatercontentwascalculatedbythediffer‐ encebetweenmoisturecontentatfieldcapacityextractedinapressureplateat−33kPa andmoistureatplantwiltingpoint,measuredat−1500kPa[47].Exchangeablebases (Mg2+,K+,andNa+)wereextractedwith1NNH4OAc,andcation‐exchangecapacitywas determinedbysaturationinsodiumbywashingthesoilsampleswithalcoholandextract‐ ingthesodiumadsorbedwith1NNH4OAc[46].pHwasmeasuredinasoilsuspension indistilledwater(1:2.5).SoilorganiccarbonwasdeterminedusingtheWalkleyandBlack method[48],modifiedbyTyurin[49].TheKjeldahlmethodwasusedtocalculatetotalN [50].CaCO3equivalentsweredeterminedusingthemanometricmethod[51].Asaturated extractwaspreparedfromeachsampletodetermineitselectricalconductivity[52].Solu‐ blecalcium,magnesium,potassium,andsodiuminthesaturatedextractsweredeter‐ minedbyatomic‐absorptionspectrometry.Availablephosphorus(onlymeasuredin2011) wasdeterminedusingOlsen’smethod[53].NO2−,NO3−,SO42−,andCl‐inthesaturated extractsweredeterminedbyhigh‐precisionliquidchromatography(HPLC). GeographicvariableswereobtainedfromtheLandUseandVegetationCoverMap ofAndalusia(scale1:25,000)[54].Climatedatawerederivedfrominformationprovided bytheNationalMeteorologicalInstituteofSpainusingtheinterpolationmethodpro‐ posedbySánchez‐Palomaresetal.[55].Lithologicaldatawereobtainedfromgeological mapspublishedbytheSpanishInstituteforGeologyandMining[56].Lithologywasdi‐ videdintothefollowingcategories:(1)carbonaterocks,includinglimestones,dolomites, marbles,limestonecrusts,andconglomerates;(2)siliceousrocks,includingmicaschists, phyllites,andquartzites;(3)siltswithclay;and(4)sands. Datarelatingtoplantationage,land‐useandmanagementwereprovidedbythe MinistryofAgricultureandFisheriesofAndalusia.RelativecoversofP.halepensisandQ. ilexsubsp.ballotameasuredwithinthesampledafforestedareaswereincludedasland‐ usevariables. Apairedsitecomparisonapproachwasfollowedtocompareplantdiversityandsoil characteristicsbetweenafforestedandcerealcropsitesbecausebaselinesamplingbefore afforestingwasunavailable.Atotalofsixpairedsiteswithsimilarecologicalcharacteris‐ ticswerechosentocompareplantdiversity(Figure1).Additionalsoilsamplingwascar‐ riedoutin18pairedsitestocomparesoilparametersattwosoildepths(0–5and5–10cm) betweenbothlanduses(Figure1).Adjacentcerealcropsconsideredinthesepairedcom‐ parisonsweresampledthesameasafforestedplots. 2.3.StatisticalAnalyses Apreliminaryexplorationofdifferencesbetweenthepreviouslanduses(afforesta‐ tionplantedincroplandsversusold‐fields,N=43)wasperformedbytwo‐samplet‐tests andtwo‐sampleWilcoxontest(non‐parametrictest(W))forallthestudyvariables. Generalizedlinearmodels(GLMs)wereusedtoanalyzeplantdiversityindicatorsin theafforestedsites(N=43)includinggeographic,climatic,edaphic,geological,andland‐ usedataasexplanatoryvariablesofdiversity.Tostart,correlationanalysiswasperformed onquantitativeexplanatoryvariables,andhighlycorrelatedvariables(r>|0.75|)were excludedfromsubsequentmodellingsteps.APoissonerrordistributionwasusedto modelAbundance,Total_R,andrichnessbyfunctionalgroup(Annuals_R,Forbs_R, Grasses_R,andWoody_R).AGaussianerrordistributionwasusedtomodelTotal_H. Collinearityandoverdispersionwerecheckedformodelvalidation.Varianceinflation factors(VIFs)wereusedtoidentifycollinearitybetweenvariables.Colinearvariables wereremovedfromthemodels.Homoscedasticityandnormalitywereevaluatedbyplot‐ tingresidualsandpredictedvalues.Thebestmodelswereselectedaccordingtothelower
Forests2021,12,17307of17 Akaikeinformationcriterion(AIC).Theproportionofvarianceexplainedbythemodel wascalculatedasapseudoR2=[(nulldeviance—residualdeviance)/nulldeviance]×100. Initialmodelselectionwasperformedfollowingastepwiseprocedure.Inordertofitthe mostparsimoniousmodel,anynonsignificantfactorswereremovedfromthevariables includedintheinitialmodelwhentheAICwascomparable.Finally,theeffectofeach significantexplanatoryvariableontheresponsevariablewasexploredbyexaminingfac‐ torcoefficientsandplottingadjustedpredictionsoftheresponsevariableusingthemost parsimoniousfittedmodel.Comparisonsbetweenafforestedandadjacentcerealplots wereconductedbytwo‐samplet‐tests.AllstatisticalanalyseswereperformedusingR software,andggplot2,corrplot,car,MASS,andggeffectspackages[57–62]. 3.Results 3.1.EnvironmentalCharacteristicsandPlantSpeciesinAfforestedSites Thirty‐nineenvironmentalvariableswererecordedforeachofthe43afforestedsites (Table1;TableS1).Atotalof252plantspecieswereidentified(TableS2,taxonomyac‐ cordingtoBlancaetal.[63]).Onlyeightofthesehadbeenartificiallyintroduced:Pinus halepensis,Quercusilexsubsp.ballota,PinuspineaL.,PrunusmahalebL.,Juniperusphoenicea L.,PinusnigraJ.F.Arnold,PrunusaviumL.,andQuercusfagineaLam.Theremainingnative speciesweredividedas53.57%annuals,14.29%forbs,6.35%perennialgrasses,and 25.79%woodyspecies.Ofthesespecies,atotalof52(20.63%)belongedtoarestrictedarea ofdistribution(TableS2):28wereendemictotheIberianpeninsulaandNorthAfrica (11.11%),15wereendemictotheIberianpeninsula(5.95%),and13wereendemictothe southandsoutheastregionsoftheIberianpeninsula(5.15%).Weidentifiedonlyoneen‐ dangeredspecies(0.40%)thatappearsintheAndalucianRedListofthreatenedspecies (Centaureapulvinata(Blanca)Blanca).Themeandiversityestimatedpersamplewhenin‐ troducedtreespecieswereincluded(Total_H)was2.54±0.40(SD),and2.49±0.44when theyweredisregarded(H_noaff). 3.2.EffectofthePreviousLandUseonPlantDiversityandEnvironmentalVariables Annual_R,CovPhal,D_n_veg,Mg,andAgeweresignificantlyhigherinafforested sitesthaninpairedcroppedsites(t=4.2305,df=41,p‐value<0.001;Wilcoxontest,W= 319.5,p‐value<0.05;W=305.5,p‐value<0.05;W=320,p‐value<00.05;andW=351,p‐ value<0.01,respectively).Grasses_R,Woody_R,andD_cropswerehigherinafforested old‐fields(W=74,p‐value<0.001;W=33,p‐value<0.0001;andW=120,p‐value<0.01, respectively).NosignificantdifferenceswerefoundbetweenpreviouslanduseforAbun‐ dance,Total_R,Total_H,andForbs_R,andtheremainingenvironmentalvariables(Sup‐ plementaryMaterial,FiguresS1–S4). 3.3.EffectsofEnvironmentalVariablesandSoilPropertiesonPlantDiversity ThequantitativeparametersincludedasexplanatoryvariablesinGLMswerese‐ lectedtakingintoaccountacorrelationcoefficientlowerthan|0.75|(FigureS5).Addi‐ tionally,previouslanduseandlithologywereincludedinthemodelsascategoricalex‐ planatoryvariables.Allmodelresultsandpredictedresponsesgraphsareavailablein SupplementaryMaterial. ForAbundance,themostparsimoniousmodelrevealedasignificantassociationonly withCovPhaandCover(TableS3andFigureS6). TheselectedmodelforTotal_Rshowedsignificantnegativerelationshipswith D_n_veg,D_shrub,s_K,CSilt,SOC,Age,andOld‐fieldfactor(Table2;Figure2;TableS4). Themodelrevealedevidenceofpositiveeffectsofs_Mg,TN,pH,andCoveronTotal_R (Table2andFigure2).
Forests2021,12,17308of17 Table2.ResultsforTotal_Rbestandmostparsimoniousmodel(pseudoR2=75.92%;AIC=258.3). EstimateSEzValuePr(>|z|) Intercept1.390000.734101.8940.058 D_n_veg−0.000960.00037−2.5660.010 D_shrub−0.000080.00004−2.0910.037 s_Mg0.263800.092702.8460.004 s_K−0.306200.09801−3.1240.002 CSilt−0.029480.00944−3.1210.002 SOC−0.212000.07265−2.9190.004 pH0.227200.086422.6290.009 TN3.186001.179002.7030.007 Age−0.066330.02660−2.4940.013 PreviouslanduseOld‐field−0.351900.09792−3.593<0.001 Cover0.015080.002895.219<0.001 D_n_veg=distancetonaturalvegetation(m);D_shrub=distancetoshrubpatches(m);s_Mg= solubleMg(mgL−1);s_K=solubleK(mgL−1);CSilt=coarsesiltfraction(0.05–0.02mm,%);SOC= soilorganiccarbon(%);TN=soiltotalnitrogen(%);Age=afforestationage(years);Cover=plant cover(%);Previouslanduse=landusepriortoafforestation(old‐fieldorcerealcrops). Figure2.RelationshipsbetweenTotal_Randexplanatoryandresponsevariablesidentifiedbythebestfitmodel.Black linesarethefittedlinescalculatedusingtheggpredictfunctioninR,providingthepredictedvaluesonthescaleofthe response.Theshadedarearepresentsthe95%confidenceinterval.Pointsrepresentmeasuredvalues(N=43).(a)D_n_veg =distancetonaturalvegetation(m);(b)D_shrub=distancetoshrubpatches(m);(c)s_Mg=solubleMg(mgL−1);(d)s_K =solubleK(mgL−1);(e)CSilt=coarsesiltfraction(0.05–0.02mm,%);(f)SOC=soilorganiccarbon(%);(g)TN=soiltotal nitrogen(%);(h)pH;(i)Age=afforestationage(years);(j)Cover=plantcover(%);(k)landusepriortoafforestation. Forfunctionalgroups,Annuals_Rwassignificantlynegativelyassociatedwith D_Qil,K,CSilt,NO3,SandandSilt+claylithology,Old‐fieldpreviouslanduse,andAge (TableS5andFigureS7).PositiveeffectswerefoundforMg,AW,EC,pH,Siliceousrocks, CovPhal,andCover.
Forests2021,12,17309of17 SignificantpositiveassociationswerefoundforpHandCoveronForbs_R;however, Kseemedtoexertanegativeeffect(TableS6andFigureS8). ForGrasses_R,theselectedmodelshowedapositiverelationshipwithAW,CSilt, andOld‐fieldpreviouslanduse(TableS7;FigureS9).Evidenceofanegativeeffectof D_shrubandCaCO3onGrasses_Rwasfound. TheselectedmodelrevealedevidenceofthenegativeeffectsofGravelandCaCO3on Woody_R(TableS8;FigureS10).Onotherhand,apositiveeffectwasfoundonWoody_R byOld‐fieldasthepreviouslanduse. ThemostparsimoniousmodelforTotal_Hincluded12variables,notallassociated withsignificanteffects(Table3;TableS9).D_n_veg,D_shrub,s_K,CaCO3,andPrevi‐ ous_land_use(Old‐fieldfactor)hadanegativeeffectonTotal_H,whereass_Mg,pH, CovQil,andCovershowedapositiveeffect(Figure3). Table3.InfluenceoffactorsidentifiedbythemostparsimoniousmodelperplantShannonIndex diversity(Total_H)(pseudoR2=74.71%;AIC=11.24). FactorEstimateSEtValuePr(>|t|) (Intercept)−0.922200.91590−1.0070.322 D_n_veg−0.000980.00039−2.5310.017 D_shrub−0.000180.00004−4.285<0.001 D_Qil−0.000050.00003−1.9690.058 s_Mg0.503300.135303.7200.001 s_K−0.260500.11160−2.3350.026 CSilt−0.017160.01152−1.4900.147 CaCO3−0.006720.00261−2.5760.015 pH0.476300.115504.123<0.001 Age−0.062440.03068−2.0360.051 Old‐fieldpreviouslanduse−0.359100.10910−3.2920.003 CovQil0.136400.064512.1150.043 Cover0.013120.004053.2410.003 D_n_veg=distancetonaturalvegetation(m);D_shrub=distancetoshrubpatches(m);D_Qil= distancetoQuercuswoodland(m);s_Mg=solubleMg(mgL−1);s_K=solubleK(mgL−1);CSilt= coarsesiltfraction(0.05–0.02mm,%);CaCO3(%);pH;Age=afforestationage(years);CovQil= Quercusilexcover(%);Cover=plantcover(%)
Forests2021,12,173016of17 32. Vadell,E.;DeMiguel,S.;FernándezCenteno,G.;Robla,E.;Lerner,M.;PemánGarcía,J.Afforestationofagriculturallands: Balanceofaforestpolicyinstrumentforland‐usechange.Cuad.Soc.EspañolaCienciasFor.2019,45,1–20. https://doi.org/10.31167/csecfv0i45.19497. 33. Madsen,L.M.TheDanishafforestationprogrammeandspatialplanning:Newchallenges.Landsc.UrbanPlan.2002,58,241– 254.https://doi.org/10.1016/S0169‐2046(01)00224‐9. 34. Molina,C.M.TheRestorationofForestLandscapes.InTheConservationofCulturalLandscapes;Agnoletti,M.,Ed.;CABInternational: Wallingford,UK,2006;pp.195–208. 35. Horák,J.;Brestovanská,T.;Mladenović,S.;Kout,J.;Bogusch,P.;Halda,J.P.;Zasadil,P.Greendesert?:Biodiversitypatternsin forestplantations.For.Ecol.Manag.2019,433,343–348.https://doi.org/10.1016/j.foreco.2018.11.019. 36. Nunes,A.;Oliveira,G.;Mexia,T.;Valdecantos,A.;Zucca,C.;Costantini,E.A.C.;Abraham,E.M.;Kyriazopoulos,A.P.;Salah, A.;Prasse,R.;etal.EcologicalrestorationacrosstheMediterraneanBasinasviewedbypractitioners.Sci.TotalEnviron.2016, 566–567,722–732.https://doi.org/10.1016/j.scitotenv.2016.05.136. 37. Costantini,E.A.C.;Branquinho,C.;Nunes,A.;Schwilch,G.;Stavi,I.;Valdecantos,A.;Zucca,C.Soilindicatorstoassessthe effectivenessofrestorationstrategiesindrylandecosystems.SolidEarth2016,7,397–414.https://doi.org/10.5194/se‐7‐397‐2016. 38. Pausas,J.G.;Carreras,J.Theeffectofbedrocktype,temperatureandmoistureonspeciesrichnessofPyreneanScotspine(Pinus sylvestrisL.)forests.Vegetatio1995,116,85–92.https://doi.org/10.1007/BF00045281. 39. Gleixner,G.;Kramer,C.;Hahn,V.;Sachse,D.Theeffectofbiodiversityoncarbonstorageinsoils.InForestDiversityandFunction: TemperateandBorealSystems;Scherer‐Lorenzen,M.,Korner,C.,Schulze,E.D.,Eds.;Springer:Berlin,Germany,2005;pp.165– 183. 40. Maestre,F.T.;Quero,J.L.;Gotelli,N.J.;Escudero,A.;Ochoa,V.;Delgado‐Baquerizo,M.;García‐Gómez,M.;Bowker,M.A.; Soliveres,S.;Escolar,C.;etal.Plantspeciesrichnessandecosystemmultifunctionalityinglobaldrylands.Science2012,335,214– 219.https://doi.org/10.1126/science.1215442. 41. Lucas‐Borja,M.E.;Delgado‐Baquerizo,M.Plantdiversityandsoilstoichiometryregulatesthechangesinmultifunctionality duringpinetemperateforestsecondarysuccession.Sci.TotalEnviron.2019,697,134204. https://doi.org/10.1016/j.scitotenv.2019.134204. 42. Rivas‐Martinez,S.;Loidi,J.BioclimatologyoftheIberianPeninsula.ItineraGeobot.1999,13,41–47. 43. IUSSWorkingGroupWRB.WorldReferenceBaseforSoilResources1998;ReportNum.84;IUSSWorkingGroupWRB:Roma, Italy,1998. 44. Chalmers,N.;Parker,P.TheOUProjectGuide:FieldworkandStatisticsforEcologicalProjects;TheDorsetPress:Dorchester,UK, 1989. 45. Cabezudo,B.;Talavera,S.;Blanca,G.;Salazar,C.;Cueto,M.;Valdés,B.;Hernández‐Bermejo,J.E.;Herrera,C.M.;Hiraldo,C.R.; Navas,D.ListaRojadelaFloraVascularAmenazadadeAndalucía;ConsejeríadeMedioAmbiente,JuntadeAndalucía:Sevilla, Spain,2005. 46. SoilConservationService.SoilSurveyLaboratoryMethodsandProceduresforCollectingSoilsSamples;SoilConservationService: Washington,DC,USA,1972. 47. Cassel,D.K.;Nielsen,D.R.Fieldscapacityandavailablewatercapacity.InMethodsofSoilAnalysis.Part.1:Physicaland MineralogicalMethods;Klute,A.,Ed.;AmericanSocietyofAgronomy:Madison,WI,USA,1986;pp.901–926. 48. Walkley,A.;Black,I.A.AnexaminationofDegtjareffmethodfordeterminingsoilorganicmatterandaproposedmodification ofthecromictitrationmethod.SoilSci.1934,34,29–38. 49. Tyurin,I.V.Analyticalprocedureforacomparaturestudyofsoilhumus.Tr.Pochr.Inst.Dokuchaeva1951,33,5–21. 50. Bremner,J.M.Nitrogenavailabilityindexes.InMethodsofSoilAnalysis.Part.2.ChemicalandMicrobiologicalProperties;Black, C.A.,Evans,D.D.,Esminger,T.E.,Clark,F.E.,Eds.;AmericanSocietyofAgronomy:Madison,WI,USA,1965;pp.1324–1345. 51. Williams,D.E.Arapidmanometricmethodforthedeterminationofcarbonateinsoils.SoilSci.Soc.Am.Proc.1948,13,27–129. 52. USSalinityLaboratory.DiagnosisandImprovementofSalineandAlkaliSoils;USDepartmentofAgriculture:Washington,DC, USA,1954. 53. Olsen,S.R.;Sommers,L.E.Phosphorus.InMethodsofSoilAnalysis:ChemicalandMicrobiologicalProperties;Page,D.L.,Ed.; AmericanSocietyofAgronomyandSoilScienceSocietyofAmerica:Madison,WI,USA,1982;pp.403–430. 54. ConsejeríadeMedio.AmbienteMapadigitaldeUsosyCoberturasVegetalesdelSuelodeAndalucíaaEscala1:25.000;Juntade Andalucía:Sevilla,Spain,2003. 55. Sánchez‐Palomares,O.;SánchezSerrano,F.;Carretero‐Carrero,M.P.ModelosyCartografíadeEstimacionesClimáticas TermopluviométricasParalaEspañaPeninsular;INIA:Madrid,Spain,1999. 56. IGMEMapaGeológicodeEspaña.Escala1:50.000;InstitutoGeológicoyMinerodeEspaña:Madrid,Spain,1980. 57. Venables,W.N.;Ripley,B.D.ModernAppliedStatisticswithS;Springer:NewYork,NY,USA,2002;ISBN0387‐95457‐0. 58. Wickham,H.ggplot2:ElegantGraphicsforDataAnalysis;Springer:NewYork,NY,USA,2009;pp.27–41 59. Wei,T.;Simko,V.RPackage“Corrplot”:VisualizationofaCorrelationMatrix(Version0.84);2017.Availableonline: https://github.com/taiyun/corrplot(accesedon7December2021). 60. Lüdecke,D.ggeffects:TidyDataFramesofMarginalEffectsfromRegressionModels.J.OpenSourceSoftw.2018,3,772. https://doi.org/10.21105/joss.00772. 61. Fox,J.;Weisberg,S.AnRCompaniontoAppliedRegression,3rded.;SagePublications:ThousandOaks,CA,USA,2019.
Forests2021,12,173017of17 62. RCoreTeam.ALanguageandEnvironmentforStatisticalComputing;RCoreTeam:Vienna,Austria,2020.Availableonline: http://www.r‐project.org(accessedon29September2021). 63. Blanca,G.;Cabezudo,B.;Cueto,M.;Salazar,C.;MoralesTorres,C.FloraVasculardeAndalucíaOriental,Vols.4,2nded.;Juntade Andalucía:SevillaSpain,2011. 64. Kleijn,D.;Joenje,W.;LeCoeur,D.;Marshall,E.J.P.Similaritiesinvegetationdevelopmentofnewlyestablishedherbaceous stripsalongcontrastingEuropeanfieldboundaries.Agric.Ecosyst.Environ.1998,68,13–26.https://doi.org/10.1016/S0167‐ 8809(97)00098‐4. 65. SanzHerraiz,C.;LópezEstébanez,N.;MolinaHolgado,P.Influenciadelasrepoblacionesforestalesenlaevolucióndelas comunidadesvegetalesyorníticasdelaSierradelosFilabres(Almería).Ería2002,58,157–176. https://doi.org/10.17811/er.0.2002.157‐176. 66. Médail,F.;Quezél,P.LinkedreferencesareavailableonJSTORforthisarticle :BiodiversityHotspotsintheMediterranean Basin :SettingGlobalConservationPriorities.Conserv.Biol.1999,13,1510–1513. 67. Zhao,Y.;Li,M.;Deng,J.;Wang,B.Afforestationaffectssoilseedbanksbyalteringsoilpropertiesandunderstoryplantsonthe easternLoessPlateau,China.Ecol.Indic.2021,126,107670.https://doi.org/10.1016/j.ecolind.2021.107670. 68. Wulf,M.Plantspeciesrichnessofafforestationswithdifferentformeruseandhabitatcontinuity.For.Ecol.Manag.2004,195, 191–204. 69. Janssens,F.;Peeters,A.;Tallowin,J.R.B.;Bakker,J.P.;Bekker,R.M.;Fillat,F.Relationshipbetweensoilchemicalfactorsand grasslanddiversity.PlantSoil1998,202,69–78. 70. Liang,W.;Wei,X.Relationshipsbetweenecosystemsaboveandbelowgroundincludingforeststructure,herbdiversityand soilpropertiesinthemountainousareaofNorthernChina.Glob.Ecol.Conserv.2020,24,e01228. https://doi.org/10.1016/j.gecco.2020.e01228. 71. Hrivnák,R.;Slezák,M.;Jarcuška,B.;Jarolímek,I.;Kochjarová,J.Nativeandalienplantspeciesrichnessresponsetosoilnitrogen andphosphorusintemperatefloodplainandswampforests.Forests2015,6,3501–3513.https://doi.org/10.3390/f6103501. 72. Pausas,J.;Austin,M.Patternsofplantspeciesrichnessinrelationtodifferentenvironments:Anappraisal.J.Veg.Sci.2001,12, 153–166. 73. Zhang,Y.;Liu,T.;Guo,J.;Tan,Z.;Dong,W.;Wang,H.H.ChangesintheunderstorydiversityofsecondaryPinustabulaeformis forestsaretheresultofstanddensityandsoilproperties.Glob.Ecol.Conserv.2021,28,e01628. https://doi.org/10.1016/j.gecco.2021.e01628. 74. Basnou,C.;Vicente,P.;Espelta,J.M.;Pino,J.Ofnichedifferentiation,dispersalabilityandhistoricallegacies:Whatdrives woodycommunityassemblyinrecentMediterraneanforests?Oikos2016,125,107–116.https://doi.org/10.1111/oik.02534. 75. Abadie,J.;Avon,C.;Dupouey,J.‐L.;Lopez,J.‐M.;Tatoni,T.;Bergès,L.Landuselegaciesonforestunderstoryvegetationand soilsintheMediterraneanregion:Shouldweusehistoricalmapsorinsitulanduseremnants?For.Ecol.Manag.2018,427,17– 25.https://doi.org/10.1016/j.foreco.2018.05.050ï. 76. Amici,V.;Santi,E.;Filibeck,G.;Diekmann,M.;Geri,F.;Landi,S.;Scoppola,A.;Chiarucci,A.Influenceofsecondaryforest successiononplantdiversitypatternsinaMediterraneanlandscape.J.Biogeogr.2013,40,2335–2347. https://doi.org/10.1111/jbi.12182. 77. Cortina,J.;Amat,B.;Castillo,V.;Fuentes,D.;Maestre,F.T.;Padilla,F.M.;Rojo,L.Therestorationofvegetationcoverinthe semi‐aridIberiansoutheast.J.AridEnviron.2011,75,1377–1384.https://doi.org/10.1016/j.jaridenv.2011.08.003. 78. Sardans,J.;Peñuelas,J.Potassium:Aneglectednutrientinglobalchange.Glob.Ecol.Biogeogr.2015,24,261–275. https://doi.org/10.1111/geb.12259. 79. Liu,X.;Yang,T.;Wang,Q.;Huang,F.;Li,L.Dynamicsofsoilcarbonandnitrogenstocksafterafforestationinaridandsemi‐ aridregions:Ameta‐analysis.Sci.TotalEnviron.2018,618,1658–1664.https://doi.org/10.1016/j.scitotenv.2017.10.009. 80. Guo,L.B.;Gifford,R.M.Soilcarbonstocksandlandusechange:Ametaanalysis.Glob.Chang.Biol.2002,8,345–360. https://doi.org/10.1046/j.1354‐1013.2002.00486.x. 81. Cunningham,S.C.;Cavagnaro,T.R.;MacNally,R.;Paul,K.I.;Baker,P.J.;Beringer,J.;Thomson,J.R.;Thompson,R.M. Reforestationwithnativemixed‐speciesplantingsinatemperatecontinentalclimateeffectivelysequestersandstabilizescarbon withindecades.Glob.Chang.Biol.2015,21,1552–1566.https://doi.org/10.1111/gcb.12746. 82. Guo,Y.;Abdalla,M.;Espenberg,M.;Hastings,A.;Hallett,P.;Smith,P.Asystematicanalysisandreviewoftheimpactsof afforestationonsoilqualityindicatorsasmodifiedbyclimatezone,foresttypeandage.Sci.TotalEnviron.2021,757,143824. https://doi.org/10.1016/j.scitotenv.2020.143824. 83. Deng,Q.;McMahon,D.E.;Xiang,Y.;Yu,C.L.;Jackson,R.B.;Hui,D.Aglobalmeta‐analysisofsoilphosphorusdynamicsafter afforestation.NewPhytol.2017,213,181–192.https://doi.org/10.1111/nph.14119. 84. Yang,S.;Zhao,W.;Pereira,P.Determinationsofenvironmentalfactorsoninteractivesoilpropertiesacrossdifferentland‐use typesontheLoessPlateau,China.Sci.TotalEnviron.2020,738,140270.https://doi.org/10.1016/j.scitotenv.2020.140270.