Agricultural abandonment in Mediterranean reclaimed peaty soils: Long-term effects on soil chemical properties, arbuscular mycorrhizas and CO2 flux

Agricultural

abandonment

in

Mediterranean

reclaimed

peaty

soils:

long-term

effects

on

soil

chemical

properties,

arbuscular

mycorrhizas

and

CO

2

flux

Elisa

Pellegrino

a,1,

_*

_,

_Simona

_Bosco

a,1

_,

_Valentina

_Ciccolini

a,1

_,

_Chiara

_Pistocchi

a,b

_,

Tiziana

Sabbatini

a

,

Nicola

Silvestri

c

,

Enrico

Bonari

a

a_{InstituteofLifeSciences,ScuolaSuperioreSant}0_{Anna,P.zaMartiridellaLibertà33,56127Pisa,Italy} b_{GroupofPlantNutrition,USYSETHZurich,Eschikon33Lindau,Switzerland}

c

DepartmentofAgriculture,FoodandEnvironment,UniversityofPisa,ViadelBorghetto80,56124Pisa,Italy

ARTICLE INFO Articlehistory: Received31March2014

Receivedinrevisedform6September2014 Accepted9September2014

Availableonlinexxx Keywords: Peatsoil Soilquality Landusechange

Arbuscularmycorrhizalfungi(AMF) SoilCO2flux

Soilrespirationpartitioning

ABSTRACT

In the last century, most peatlands werereclaimed for agricultural purposes,which led to peat

degradationandtoseveresubsidence,andthustoowetconditionsforcrops.Insomeareasthishas

therefore led to wide agricultural abandonment. However, studies on the effect of agricultural

abandonment as a potential restoration tool are lacking. In this study, the effectivenessand the

restorationpotentialofagriculturalabandonmentinreducingpeatdegradationandinimprovingsoil

microbialbiodiversitywereevaluated.Themainchemicalparameters,arbuscularmycorrhizal(AM)

fungaldiversityandsoilrespirationpartitioningwereusedtoassessthelong-termeffectof15yearsof

agricultural abandonment (Aband) in a Mediterranean reclaimed peatland. An intensive maize

cultivation (Cult) in thesame areawas used as acomparison. Multivariate analysesshowed that

15yearsofagriculturalabandonment:didnotaffectthemainsoilchemicalparameters,exceptfor

NH4+whichwaslowerintheAbandthanintheCult;increasedAMfungalrootcolonizationandthe

diversityintermsofnumberoffamiliesofAMfungiretrievedinroots,butdecreasedsoilAMfungal

richness;reducedtotalsoilrespirationanditsautotrophiccomponentbut increasedrespirationby

heterotrophs;determinedalowerfluctuationofsoilCO2fluxresponsetoairtemperaturethantheCult.

Thus, although somesoil quality parametersweresignificantlyimproved,15 years of agricultural

abandonment may not leadto an effective restoration. Consequently, alternative and sustainable

solutionsfortheprotectionandpreservationofMediterraneanpeatlandsneedtobedeveloped.

ã2014ElsevierB.V.Allrightsreserved.

1.Introduction

Wetlands are species-richhabitats performing valuable

eco-systemservicessuchasfloodprotection,waterquality

enhance-mentandcarbon(C)sequestration.Theirprotectionisofficiallya priorityfor168nationsthathaveratifiedtheRamsarConvention (VerhoevenandSetter,2010).Thesehabitatscoverabout6%ofthe

globalsurfaceareaandabout60%arerepresentedbypeatlands,

whichplayanimportantroleintheglobalCcycleasalong-term

sink. In Europe peatlands cover about 20% of the land area.

AlthoughmostarelocatedinnorthernEurope,somesmallsitesare alsosituatedintheMediterraneanarea(Montanarellaetal.,2006).

Inthelast200years,mostpeatlandshavebeenreclaimedin

Europeforagriculturalpurposesbecauseoftheirnaturalfertility, whichhasthusdestroyedtheiroriginalcharacterandecological

functions(Verhoevenand Setter, 2010).The increasein aerobic

conditionsduetothereclamationandagriculturalpracticesledto significantincreasesinsoilorganicmatter(SOM)oxidation.Major consequencesincludedthehighreleaseofcarbondioxide(CO2),

nutrientlossestowaterbodies,biodiversitylossesanddegradation ofthepeatwithadrasticdecreaseinsoilqualitytogetherwitha severesubsidencewhichledinsomeareastoconditionsthatare toowetforcrops.

Soilqualitycanbeevaluatedusingchemical,biochemicaland biologicalindicators(DoranandParkin, 1996).Withinthechemical andbiochemicalparametersofsoil,SOMandsoilrespirationare

themainandmostsuitableindicatorstomeasureCstorageand

degradation.Asregardssoilrespiration,CO2fluxisconsideredasa

proxyfor SOM decompositionand itspartitioning betweenthe

*Correspondingauthor.Tel.:+39050883181;fax:+39050883526. E-mailaddress:e.pellegrino@sssup.it(E.Pellegrino).

1

Theseauthorscontributedequallytothiswork.

http://dx.doi.org/10.1016/j.agee.2014.09.004

0167-8809/ã2014ElsevierB.V.Allrightsreserved.

ContentslistsavailableatScienceDirect

Agriculture,

Ecosystems

and

Environment

microbialandtherootcomponentisrequiredinordertoidentify CO2sources andtocalculatetheircontributiontothetotalflux

(Kuzyakov, 2006).Inaddition,in ordertoassesssoilhealthand functioning,anevaluationofthediversityofbeneficialmicrobes suchasarbuscularmycorrhizal(AM)fungiisvaluablesincethese

fungiareinvolvedinplantgrowthandnutrientuptake

enhance-mentandinimprovingsoilstructure(SmithandRead,2008).

Severalstudieshaveinvestigatedthesoilqualitydegradationof

reclaimedpeatlandsintensivelyused for agriculture.Numerous

strategieshavebeenproposedfortheirrestoration,suchas the rewettingwithorwithouttheintroductionofplantspecies,anda shifttoextensive-grazingsystemsortolessintensiveagriculture use(PfadenhauerandGrootjans,1999;ZeitzandVelty,2002).In

theagricultural peatlandswhere none of thesestrategies have

beenadopted,thesubsidenceandconsequentinabilitytosupport

cropgrowth haveledtothemassiveabandonmentoftheareas

(JoostenandClarke,2002).

However,studiesarelackingontheeffectoftheagricultural

abandonment in reclaimed peatlands, and in particular on the

mostvulnerableMediterraneanareaswhere climaticconditions

favorarapidmineralizationoftheSOM.Therefore,the effective-nessandtherestorationpotentialoftheagriculturalabandonment inreducingthepeatdegradationandinimprovingsoilmicrobial biodiversityneedtobeevaluated.

Weassessed thelong-termeffect of15 yearsof agricultural

abandonmentonthesoil qualityof a Mediterranean reclaimed

peatlandlocatedintheMassaciuccoliLakebasin(Tuscany,Italy).

Tothisend,chemicalparameters,AMfungalmoleculardiversity

andcommunitystructure,andinsitusoilrespirationpartitioning

wereused.Theagriculturalabandonmentwascomparedwithan

intensivelycultivatedpeatysoilrepresentedbyacontinuousmaize (ZeamaysL.)croppingsystem,whichisthemainsysteminthearea (Silvestrietal.,2012).

2.Materialsandmethods

2.1.Fieldsite

The experimental site is situated in the south of the

Massaciuccoli Lake basin (43490N–10₁₉0_{E) within the natural}

parkof Migliarino-SanRossore-Massaciuccoli (Pisa, Italy).Since 1930,mostofthebasinhasbeendrainedbyacomplexnetworkof artificialcanals, ditches and pumping stations. This hasforced

water from reclaimed areas into the lake, causing a severe

subsidencerangingfrom3to4cmyear1,causedbycompaction andpeatoxidation(Pistocchietal.,2012).Thesoilwasclassifiedas HistosolaccordingtotheUSDAsystem(SoilSurveyStaff,1975)and

asRheicHistosolaccordingtotheFAOsystem (IUSS,2006).The

thicknessoftheorganichorizonisabout3m(from2to4m)witha

SOMcontentof29.2%(minimum20.1%–maximum55.4%

(Walk-ley-Black)),aclaycontentof25%,a pHof4.9, abulkdensityof 0.5gcm3 inthe 0–30cm depthlayer. Therefore,these organic soilscanbedefinedalsoaspeatandpeatysoil(IPCC,2006).

During the year the water tableis maintainedby pumping

stations at a quite stable level, ranging from 0.40 to 0.60m

(Pistocchi et al., 2012). Therefore, the upper soil layer is only

occasionally subjected to water saturation. The climate is

Mediterranean (Csa) according to the Köppen–Geiger climate

classificationmap(Kotteketal.,2006).Summersaredryandhot, rainfallismainlyconcentratedinautumnandspring(meanannual rainfallca.945mmyear1)andmeanmonthlyairtemperatureat 2mrangingfrom7CinFebruaryto30CinAugust(withamean

of 14.8Cyear1).Average monthly maximum,mean and

mini-mum temperatures and rainfalls over the period 1990–2012,

recordedataweatherstationlocatedintheMassaciuccolibasin

are shown in Fig. S1.The site was cultivatedwith maize until

15yearsagowhen agriculturalpracticeswereabandonedinan

areaofabout15haandthenaturalsuccessionvegetationwereleft todevelop.Thesurroundingarea(about15ha)isstillcultivated

withmaize.

2.2.Experiment1:effectofagriculturalabandonmentonchemical parameters,roottraitsandAMfungi

Theaimofthisexperimentwastoevaluatethelong-termeffect

oftheabandonmentofagriculturalpracticescomparedtomaize

cultivationintermsofchemicalsoilparameters,roottraits,suchas

the root biomass and AM fungal colonization, and AM fungal

diversityandcommunitystructure.

2.2.1.Experimentalset-up

Theexperimentwasacompletelyrandomizeddesignwithland

use as treatment and three replicates (n=3; field replicates of 0.7ha)(Fig.S2).Wehadsixcases(fieldreplicates),threeofthemof

one land use and another three of the other land use. At the

beginning, before the experiment started, fields looked all the

sameandwehadnumberedthem1–6andletarandomnumber

generatortopickthethreefieldsthatreceivedsametreatment.The firstlandusetypewasanagriculturalpeatysoilleftabandonedfor 15years(Aband).Fieldreplicateswerelefttodevelopunderthe naturalsuccessionvegetation.Afloristicsurveyshowedthatthe

most commonspecies wereAbutilontheophrasti L., Amaranthus

retroflexusL.,ArctiumlappaL.,Artemisiasp.,Atriplexsp.,Bidenssp., Biphorasp.,Calystegiasp.,DaturastramoniumL.,Echinochloa crus-galli L., Galium sp., Humulus lupulus L., Linaria sp., Phragmites australis L, Typha latifolia L., Lythrum salicaria L., Phytolacca

americana L., Rumex crispus L., Silene alba L. and Xanthium sp.

The most abundant plant species were Calystegia sp. (17.5%),

P. australis (15%), A. lappa (13.8%) and B. tectorum (13.8%). No fertilizersorotheragriculturalpracticeswereapplied,exceptfor anannualvegetationcuttingattheendofthewinterseason.

Thesecondlandusetypewasanintensivelycultivatedpeaty

soil representedby continuous maize (Cult). Each year in late

springfieldreplicateswereploughed(30–35cm)andharrowed,as

themainandsecondarytillage,respectively.Maizewassownat

thebeginningofJuneatarateof75,000seedsha1with7517cm

rowspacing and harvested in late September. Fertilizationwas

applied at sowing and at mechanical weeding with rates of

32kgha1N,96kgha1P,96kgha1Kand138kgha1N,

respec-tively. Chemicaland mechanical post-emergenceweedcontrols

wereapplied.Thefifteenyearaveragemaizeyieldwas6.4tha1. 2.2.2.Sampling

InJuly2011onecombinedsoilsample,resultingfrompooling sevensoilcores,wascollectedfromeachfieldreplicate(0–30cm depth)inordertocoverchemicalandAMfungalspatialvariability.

SamplingwascarriedoutonlyonceinJuly,sincemid-summeris

the best choice because sampling should not be close to soil

treatments and becausethe variability in chemical parameters

changes slightly duringtheyear(Pellegrinoet al.,2011).These facts,alongwiththefactthatAMfungiconsistentlymaintainthe

same patterns of variability in differently managed systems,

althoughwithseasonalchanges,weretakenintoaccountwhen

choosingJulyasasinglesamplingtime(Vandenkoornhuyseetal., 2002;Oehletal.,2010;DiBeneetal.,2013).

Soilsamplesusedforthechemicalparameterdeterminations

were oven dried at 60C and then sieved at 2mm, while for

genomicDNAextractionrootswerecarefullypluckedwithforceps. Soil DNAextractswerestored at4C. Asregardstheroottrait

determinations, seventurfs wereextracted(20cm depth) from

each fieldreplicateand thencombined.In thelaboratory,roots

driedat60Cforrootdryweight(DW)measurements,whereasfor

AM fungal root colonization assessment and genomic DNA

extraction,rootsubsamplesweretakenandstoredat4C. 2.2.3.Soilchemicalanalyses

Soilsampleswereanalyzedfor:pH;electricalconductivity,EC; exchangeablepotassium,Kexch; total nitrogen,Ntot;ammonium,

NH4+;nitrates,NO3;soilorganicmatter,SOM;totalphosphorus,

Ptot;availablephosphorus,Pavailandorganicphosphorus,Porg.Soil

pHandECweremeasuredindeionizedwater(1:2.5and1:2,w/v,

respectively).Kexchwasdeterminedbyatomicabsorption.Ptotand

Pavail were determined by colorimetry using perchloric acid

digestionandanextractionwithsodiumbicarbonate,respectively

(OlsenandSommers,1982).PorgwasevaluatedusingtheMetha

extraction(HenceandAnderson, 1962).Ntotwasdeterminedbythe

macroKjeldahldigestionprocedure,whileNO3andNH4+were

determinedbytheKjeldahlmethodafterKCl2Nextractionand,in thecaseofNO3,afterreductionwithDevarda’salloy.SOMwas

measured using the modified Walkley-Black wet combustion

method(NelsonandSommers,1982).SoilC/Nratiowascalculated bydividingSOC((SOM/1.7)10)bytotalN.

2.2.4.RootdeterminationandAMfungalrootcolonization

Fromthecombinedturfsofeachfieldreplicate,soilsubsamples

(meansoilDWca.400g)wereusedtodeterminerootDW.Roots

weremanuallycollectedwithforcepsandwashedbywet-sieving

anddecantingdowntoamesh sizeof250

m

m. Afterremoving

organicdebris,allliveanddeadrootfragmentswereoven-dried

andweighedtodeterminerootDW.RootDWpergramofsoilwas

calculated.

AM fungalrootcolonizationwas assessedundera

stereomi-croscope(OlympusSZX 9,Olympus Optics, Tokyo,Japan), after

clearingandstainingwithlactic acidinsteadofphenol(Phillips

and Hayman, 1970), following the gridline intersect method

(McGonigleetal.,1990).Therootsweremountedonmicroscope

slidesandexaminedatmagnificationsof125–500,andverifiedata magnificationof1250.

2.2.5.AMfungaldiversity:extractionofgenomicDNA,PCR amplification,cloningandsequencing

SoilDNAwasextractedfrom0.5gofsoilusingthePowerSoil1 MoBiokit (MoBioLaboratoriesInc.,NY,USA)(n=6),whileroot

DNAwas extracted from 100-mg fresh root samplesusing the

DNeasy1PlantMiniKit(n=6)(Qiagen,Germantown,MD,USA).

DNA quality was checked on a ND-1000 spectrophotometer

(NanoDrop Technology, Wilmington, DE). The direct extraction

oftheDNAfromsoilwasperformedtopreventthehostpreference

effect as previously described by Davison et al. (2012). PCR

amplification was performed using the primer pair NS31 and

AM1targetingthesmallsubunitribosomalRNA(SSUrRNA)region

(Simon etal.,1992; Helgasonetal.,1998).Althoughlonger and higherdiscriminating regionsareavailable (Krügeret al.,2012;

Pellegrinoetal., 2012), the NS31/AM1SSUregionwastargeted

becausemostGlomeromycota diversitydataareobtainedusing

thisregion, whichprovides alargercomparativeDNAsequence

data-set.

PCRwasperformedusingthetemperatureprofiledescribedby

Helgasonetal.(1998).NoPCRinhibitionproblemswereregistered.

PCRampliconsweregeneratedfrom10ng

m

L1genomicDNAin

volumes of 20

m

L with 0.5U of GoTaq1 hot start polymerase

(PromegaCorporation,Madison,WA,USA),0.2

m

Mofeachprimer

(NS31/AM1), 0.2mM of each dNTP, 1.25mM of MgCl2 and 1x

reactionbuffer,usingtheS1000ThermalCyclerTM(BIORAD,USA). Beforeligation,thequantityandqualityofthePCRamplicons

were checked by a spectrophotometer (NanoDrop Technology,

Wilmington,DE). ThePCRampliconswerethenligatedintothe

pGem1-TEasyvector(PromegaCorporation,Madison,WA,USA)

andusedtotransformXL10-Gold1UltracompetentEscherichiacoli cells (Stratagene1, LaJolla, CA, USA). At least 25 recombinant clonesperampliconlibrary(n=12)werescreenedforthec.

550-bp-longNS31/AM1fragmentonagarosegels.

Atotalof270PCRproductsobtainedfromclones(ameanof

23perlibrary)weresequencedusingtheNS31/AM1primersinan

ABI Prism1 3730XL automated sequencer (Applied Biosystem,

Foster City, CA, USA) at the High-Throughput Genomics Unit

(Seattle,WA,USA).

2.3.Experiment2:effectofagriculturalabandonmentonsoilCO2flux

Theaimofthisexperimentwastoevaluatethelong-termeffect

ofagriculturalabandonmentcomparedwithmaizecultivationon

totalsoilrespirationfluxandautotrophicandheterotrophicCO2

fluxcomponents.

2.3.1.Experimentalset-upandsampling

Theexperimentaldesignwasasdescribedaboveforexperiment

1. Six experimental blocks (one for each field replicate) were

establishedforsoilrespirationmonitoring(Fig.S3a,b).Sixblocks

were considered as an adequate number due to the texture

homogeneity of the mineral component of the soil (Table S1).

Blockswerelocatedabout300mapart.Ineachfieldplot,theblock

consisted of two 20-cm diameter open-ended PVC collars: a

surface collar (7.0-cm deep collar inserted 1cm into the soil)

pressedfirmlyontotheshallowsurfacelayerwithoutcuttingany roots(Fig.S3a);a25-cmdeepcollarinserted20cmintothesoil (Fig.S3b).Thecollardepthwasevaluatedasbeingappropriateto exclude90%offinerootsfromthesoilvolume(Heinemeyeretal.,

2007).Plants insidethe collarswereremoved,leaving theroot

systems intact. The surface and the deep collars provided a

measureofthetotalsoilrespiration(Rs)andoftherespirationby

heterotrophs (Rh,soil microorganisms and mesofauna),

respec-tively(Heinemeyeretal.,2007,2011).Thesemeasurementswere

usedtocalculatethecontributionoftherootcomponentdefinedas respirationbyautotrophs(Ra=Rs Rh).

SoilCO2fluxwasmeasuredusinganon-steady-state

through-flow chamber equipped with a portable infrared gas analyzer

(IRGA) (Licor LI-820) connected to a steel chamber with a

headspace volume of 6186cm3 _{(chamber B, West Systems Srl,}

Pontedera,Italy).To guaranteea tightseal withthecollars, the chamberhad arubberring thatfitsintothecollarlip.TheCO2

concentrationwasmeasuredpersecond,whiletheincreaseinthe

headspace (ppm/s) was checked for linearity for a period of

Table1

Soilchemicalparameters(0–30cmdepth)ofabandonedagriculturalpeatysoils (Aband)andofamaizecultivation(Cult).

Chemicalparameters Aband Cult pH* 5.30.4** 4.60.1 EC(mScm1) 0.90.1 1.80.9 Kexch(mgkg1) 560.084.6 397.051.9 Ntot(gkg1) 11.81.2 13.02.2 NO3-(mgkg1) 59.018.6 42.315.7 NH4+(mgkg1) 56.34.5a 148.09.6b SOM(%) 28.34.6 25.74.1 C/N 13.71.0 11.50.3 Ptot(mg kg1) 2846.7229.2 2709.0329.3 Pavail(mgkg1) 76.313.6 70.37.9 Porg(mgkg1) 2054.7198.1 2153.3363.4 * _{EC:electricalconductivity;K}

exch:exchangeablepotassium;Ntot:totalnitrogen;

NO3:nitrate;NH4+:ammonium;SOM:soilorganicmatter;C/N:carbon/nitrogen

ratio;Ptot:totalphosphorus;Pavail:availablephosphorus;Porg:organicphosphorus. **

ValuesaremeansSEofthreefieldreplicatesforeachtreatment.Valuesinthe samerowfollowedbydifferentlettersarestatisticallydifferentbetweenlanduses accordingtoANCOVA(P<0.05).

2–3min,andcalculatedandrecordedinthefieldbya palmtop

computerconnected via Bluetooth. CO2 flux was calculated by

performinglinearregressionsontheloggedCO2data(R2>0.95),

whichwerecorrectedforatmosphericpressureandair

tempera-ture. An internal fan maintained the homogeneity of the air

mixture within the chamber during the measurements. Soil

moistureand temperaturewererecordedateach measurement

nexttoeachcollarbyaprobe(DecagonDevisesECH2O-TE/EC-TM)

insertedintoasoildepthof5cm.

In 2012 monitoring was undertaken between 14 May and

13Augustwithameasurement frequencyof onceortwiceper

week(atotalof21measurementsperblock).Thesamplingperiod waschosenbytakingintoconsiderationthatinordertoevaluate theimpactofalandusechangeonsoilbiochemicalparameters,it isnecessarytosampleatamuchlaterdatefromsoiltreatments

(Picci and Nannipieri, 2002). Daily mean temperatures and

rainfallsduringthesamplingperiod(from MaytoAugust2012)

wererecorded(Fig.3a).Measurementsweretakenbetween8a.m.

and12a.m.,becausemid-dayvaluesofCO2fluxareassumedtobe

representativeofadailyaverageflux(Davidsonetal.,1998;Luo

andZhou,2006).Samplingstartedtendaysaftercollarinsertion,

whichisasufficienttimelapsetoallowforconsiderablerootdeath

andtoexcludetheRacomponentfromdeepcollar.

2.3.2.Modelingthesoilrespiration

ThecumulatedRsandRh(from14Mayto13August2012)were

calculatedusingtheexponentialrelationshipbetweensoilCO2flux

and air temperature. The Van’t Hoff empirical exponential

equation (Q10),a simplified version of theRothamsted Carbon

Model (RothC), with temperature as an independent variable

(ColemanandJenkinson, 1999)andtheLloydandTaylor(LT)(Lloyd

and Taylor, 1994) models were used in order to assess the

sensitivityofsoilrespirationtoairtemperature.Thethreemodels werefittedonthemeasuredsoilCO2fluxofRsandRhforbothland

uses,andwerethencorrelatedwithairtemperature.

TheperformanceofthethreemodelsofCO2fluxresponsetoair

temperaturewasevaluatedusingtheAkaikeinformationcriterion

(AIC); the root mean squared error (RMSE) and the adjusted

R-squarevalue (Rsd. ad). The best statistical fit was chosen to

calculatethecumulativeflux. 2.4.Dataandsequenceanalysis

AnupdatedAMfungalreferencedatasetof59NS31/AM1public

sequences (ca. 550bp) was created using only morphotype

sequences,includingthemajorityoftheAMfungalspecieslisted

inSchüßlerandWalker(2010)phylotaxonomicclassification(this

alignmentisinanopen-accessdatabasehttps://sites.google.com/

site/restomedpeatland/microbiology).Thereference datasetand

theircorrespondencewiththeclosest(similarityhigherthan99%) virtualtaxon(VT)afterblastsearchagainsttheMaarjAMdatabase (Öpiketal.,2010)isshowninFig.S4andTableS2.TheAMfungal referencedatasetwasusedforthefurtheralignmentofthe

newly-generated sequences using Bioedit (Hall, 1999), after having

checked the quality of their electropherograms by Vector NTI

Advance10(Invitrogen,USA)andaffiliationtotheGlomeromycota usingabasiclocalalignmentsearchtool(BLAST)search(Altschul etal.,1997).Analignmentofatotalof195sequences(21fromthe referencedataset, 13fromNCBIafterblastsearch(similarityhigher than99%),160newlygeneratedsequencesandtheCorallochytrium

limacisporumsequenceL42528astheoutgroup)wastrimmedto

thesamelength(ca.490bp)andmanuallyrefined.Phylogenetic

trees were inferredby the neighbor-joining(NJ) analysisusing

MEGAversion5.1(Tamuraetal.,2011http://www.megasoftware.

net)and theKimura2-parametermodel(Kimura,1980).Branch

support values correspond to 1000 bootstrap replicates. The

phylograms were drawn by MEGA 5.1 and edited by Adobe

IllustratorCS6.

The phylogramwas usedtoassignthe newly-generatedAM

fungal sequences to molecular operational taxonomic units

(MOTUs)onthebasisofabootstrapvalueof75.Inaddition,the

correspondence ofeach MOTU withtheclosestVT(Öpiket al.,

2010)wascomputed.AMfungalMOTUrichnessandtheShannon

index (H0) werecalculated usingPrimer v6(Clarke and Gorley,

2006http://www.primer-e.com).ThesuitabilityoftheAMfungal

communitysamplingwasverifiedbyColemanrarefactioncurves

(Coleman, 1981)inEstimateSversion9.1(Colwell,2013http://purl.

oclc.org/estimates) using individual-based rarefaction curves

(GotelliandColwell,2001)withclones/sequences consideredas

unitsofreplication.

Allnewly-generated sequencesweresubmitted tothe EMBL

nucleotide sequence database(http://www.ebi.ac.uk/embl/) and

areavailableunderaccessionnumbersHG425705–HG425864.

Tocompensatethescarceinterspersionofthefieldreplicates generatedbyourrandomizationandtoprovethattheeffectsthat wedetectedwereunlikelytohavebeencausedbyapurelyspatial

phenomenon, we accounted for plotpositions byapplying the

analysis of covariance (ANCOVA). Land use was used as fixed

factorandthespatialcoordinatesoftheplots(latitude/longitude) ascovariables.Therefore,wefactoredoutthedistance variable

that may affect the studied dependent variables, allowing a

propercomparison ofthetreatments.Spatial coordinateswere

first transformed into radiants and thenstandardized. All the

dependent variables were log- or arcsin-transformed when

necessarytofulfill theassumptionfor theANCOVA. Whenthe

assumptions for theANCOVAwerenot fulfilled,even after the

appropriatetransformation,datawereanalyzedusingtheMann–

Whitney non-parametric test or the Kruskal–Wallis H

non-parametrictest,followedbytheMann–Whitney test. Allthese

analyseswereperformedinSPSSversion21.0(SPSSInc.,Chicago, IL,USA).

Constrainedordinationanalyses(partialredundancyanalysis, pRDA)wereusedtoinvestigatetheinfluenceofthedifferentland

uses (usedas explanatoryvariables) onthechemical,rootand

CO2 parameters and on the AM fungal relative abundances of

MOTUs (used asresponsevariables) accountingfor thespatial

coordinates of the plots (used as covariables). In partial

con-strained analysis, information that can be explained by the

covariablesis_firstextractedandthentheexplanatoryvariables

areusedto“explain”theresidualvariation.RDAswereusedto

assesstheinfluenceofdifferentmatrices(rootsandsoil)onthe

explanatory variables. RDAlinear method (Lepš and Šmilauer,

2003)wasusedsincethegradientofthedetrended

correspon-dence analysis was lower than four. All data were

log-trans-formed, centered and standardized by the response variables.

MonteCarlopermutationtestswereperformedusing499random

permutations(unrestrictedpermutation)in ordertodetermine

thestatisticalsignificanceoftherelationsbetweenlanduseand

responsevariables.RDAsandpRDAswereperformedbyCanoco

for Windowsv.4.5(terBraakandŠmilauer,2002).Thebiplots

weredrawnbyCanoDrawforWindows.

Table2

Rootdryweightandarbuscularmycorrhizal(AM)fungalrootcolonizationofthe naturalsuccessionvegetationandofthemaizeoccurringinabandonedagricultural peatysoils(Aband)andinamaizecultivation(Cult),respectively.

Parameters Aband Cult Rootdryweight(mgg1soil) 1.10.5*

0.60.2 AMfungalrootcolonization(%) 33.22.0b 22.21.7a

*

ValuesaremeansSEofthreefieldreplicatesforeachtreatment.Valuesinthe samerowfollowedbydifferentlettersarestatisticallydifferentbetweenlanduses accordingtoANCOVA(P<0.05).

Fig.1.Neighbor-Joining(NJ)treeofarbuscularmycorrhizalfungal(AMF)sequencesderivedfromrootsandsoil(shownastriangleandcircle,respectively)ofagricultural abandonedpeatysoils(Aband)andofamaizecultivation(Cult)(green/openandred/filledsymbols,respectively).Theanalysisisbasedonpartialnuclearsmallsubunit ribosomalRNAgenesequences(SSU550bp;NS31/AM1fragment),andthetreeisrootedwithareferencesequenceofCorallochytriumlimacisporum(L42528).AMF sequenceswereclassifiedin11moleculartaxonomicunits(MOTUs)affiliatedto:Funneliformissp.(8),Rhizophagusspp.(4,5),Sclerocystissp.(3),Scutellosporasp.(10)andto additionaltaxa,suchasGlomusspp.(1,2,6,7,9)oranunculturedGlomeromycota(11).Piesindicatetheproportionsofsequencesintothetwolanduses(Aband,green;Cult, red)andmatrixes(roots,lightcolor;soil,darkcolor).Thetreeisdrawntoscale,withbranchlengthsinthesameunitsasthoseoftheevolutionarydistancesusedtoinferthe phylogenetictree.TheevolutionarydistanceswerecomputedusingtheKimura2-parametermethodandareintheunitsofthenumberofbasesubstitutionspersite. Bootstrappingisbasedon1000replicates.Theanalysisinvolved195nucleotidesequences.EvolutionaryanalyseswereconductedinMEGA5.Sequencesobtainedinthe presentstudyareshownbysymbolsandtheiraccessionnumbersareshowninFig.S3.ThecorrespondencebetweenMOTUsandtheclosestvirtualtaxa(VT)afterblastsearch againsttheMaarjAMdatabase(Öpiketal.,2010)isshowninFig.S3andTable3.

3.Results

3.1.Experiment1:effectofagriculturalabandonmentonsoilchemical parameters,rootsandAMfungi

3.1.1.Soilchemicalparameters,rootdryweightandAMfungalroot colonization

Fifteen years of agricultural abandonment (Aband) did not

significantlymodifysoilquality,asmeasuredbyitsmainchemical parameters,comparedwithmaizecultivation(Cult),exceptforsoil

NH4+ concentrationwhich was around three-fold lower in the

Abandthan in theCult (Table 1).However,it is noteworthyto

highlight anincreasing trend in the Abandsoil organic matter

content,which increasedby9%incomparisonwiththeCult,as

wellasotherchemicalparameters,suchasthePtotandPavail,which

increasedby8%and5%,respectively.

Asregardsroots, DWvalues werenot significantly different

between land uses, whereas AM fungal root colonization was

significantlyhigher(30%)intheAbandthanintheCult(Table2). 3.1.2.AMfungaldiversity

ThePCRprimerpairNS31/AM1,whichtargetsthe30_endofthe

SSUrRNAgene(550bp),was usedtoamplifyandscreen the

clonelibrariesobtainedfromthe12crudeDNAextractsoffield

rootandsoilsamples.A totalof366clones werescreenedand

270 showedtheexpected AMfungal bandlength.132positive

cloneswereobtainedfromtheAband(63and69fromroot(R)and

soil samples (S), respectively) and 138 positive clones were

obtainedfromtheCult(65and73fromRandS,respectively).After

BLASTchecking,about40%ofthesequenceswereexcludeddueto

sequencingerrorsorPCRprimerunspecificity.

The obtainedsequences weregroupedinto 11 different AM

fungalmolecularoperationaltaxonomicunits(MOTUs)(atotalof

sevenandeightintheAbandandtheCult,andsixandeightin

rootsandsoil,respectively),whichwerephylogeneticallyaf_filiated

toFunneliformissp.(Fun1_AMASS),fivedifferentGlomusspp.(from

Glo1_AMASS to Glo5_AMASS), two Rhizophagus sp.

(Rhizo1_A-MASS and Rhizo2_AMASS), Sclerocystis sp. (Sclero1_AMASS),

Scutellosporasp.(Scut1_AMASS)and anuncultured

Glomeromy-cota (Uncult1_AMASS) (Fig. 1; Fig. S5). The correspondence

between our MOTUs and the closest (similarity higher than

99%)VTafterblastsearchagainsttheMaarjAMdatabaseisshown

in Table 3. Three out of eleven MOTUs, although three were

doubletons (Glo5_AMASS, Rhizo1_AMASS and Scut1_AMASS),

wereconsideredforfurtheranalyses.

The rarefaction curves showing the relation between the

number of sequences and the number of observed AM fungal

MOTUsretrievedfromrootsandsoiloftheAbandandtheCult

(Fig.S6)demonstratedthatthesamplingeffortwassufficientas

theaccumulationcurvesreachedtheasymptote.

AsshowninthepiechartsinFig.1,threeAMfungalMOTUs

wereexclusivelyretrievedintheAband:Fun1_AMASSinthesoil

and theothers (Scut1_AMASS and Uncult1_AMASS) within the

roots.FourAMfungalMOTUswereonlyfoundinthesoiloftheCult

(Glo2_AMASS,Glo5_AMASS,Rhizo1_AMASSandSclero1_AMASS).

By contrast, Rhizo2_AMASS, Glo1_AMASS, Glo3_AMASS and

Glo4_AMASSshowedaubiquitousbehavior.

TheMOTUrichnessandShannonbiodiversityindex(H0)were

calculated toevaluate theAM fungal community richness and

diversityinthetwodifferentland-uses,consideringinthisway,

not only the number,but alsothe relative proportionsof taxa

(Table4).WeobservedasignificantlylowerAMfungalrichnessin thesoiloftheAbandthanintheCult.Interestingly,inthemaize cultivation,thesoilshowedhighervaluesofbothindexesinterms

of the roots.Although a differentialtrend was found and both

diversityindexeswerethree-foldhigherintheAbandthaninthe Cult,15yearsof agricultural abandonmentdidnot significantly

affectAMfungalcommunityrichnessandevennessintheroots

(Table4).

pRDAsshowedthatlandusechangesignificantlyaffectedonly

the soil AM fungal composition and structure, and that the

differentmatrixes havedifferentAM fungalassemblages.pRDA

revealedthatlanduseexplained87.2%(IandIIaxes)ofthewhole variance(Fig.2a),andthatitseffectontheAMfungalcommunity wassignificant(P=0.036).Ashighlightedbythearrows,

Glo1_A-MASS and Fun1_AMASS showed a preferential presence in the

Aband,Glo2_AMASSandSclero1_AMASSintheCult(Fig.2a).

As regards the Cult and the Aband, RDAs showed that the

differentmatrixesexplained89.7%and89.2%(IandIIaxes)ofthe wholevariance,respectively(Fig.2b,c),andthattheireffectonthe AMfungalcommunitywassignificant(P=0.002).Asshownbythe

arrows, in theCult Glo2_AMASS, Sclero1_AMASS, Glo1_AMASS,

Glo4_AMASS,Glo5_AMASSandRhizo1_AMASSaresoil

preferen-tial,whileGlo3_AMASSandRhizo2_AMASSarerootpreferential

(Fig.2b).AlsointheAband,thearrowsshowedthatRhizo2_AMASS

and Glo3_AMASSarepreferentialof therootsandthat

Glo1_A-MASSofthesoil(Fig.2c).Interestingly,thisRDAbiplothighlighted

thatScut1_AMASS/Uncult1_AMASSandFun1_AMASS,exclusively

presentintheAband,werepreferentialoftherootsandofthesoil, respectively.

3.2.Experiment2:effectofagriculturalabandonmentonsoilCO2flux

3.2.1.SoilCO2fluxmeasurements

Both air and soil temperatures and soil moisture varied

considerablyduringthemonitoringperiod(Fig.3a,b).FromMay

to August, the mean air and soil temperatures ranged from

12.0C to 26.6C (average value over the period 21.5C) and

from 16.1C to 30.2C, respectively; soil moisture (% v:v)

decreasedinbothlandusesfrom27.5%to2.2%andfrom30.8%

to15.8% for theAband and theCult,respectively(Fig.3b). Rs,

Table3

Listofthemoleculartaxonomicunit(MOTU)ofthesequencesretrivedinthisstudy andtheircorrespondencewiththeclosestvirtualtaxon(VT)afterblastsearch againsttheMaarjAMdatabase(Öpiketal.,2010).TheaccessionnumberofMaarjAM typesequencesisalsoshown.

MOTU Code* _{VT Accessionnumber}

Fun1_AMASS 8 VT065/067 Y17635 Glo1_AMASS 1 VT309 FJ194511 Glo2_AMASS 2 VT309 FJ194511 Glo3_AMASS 6 VT093 EU332715 Glo4_AMASS 7 VT219 AM849279 Glo5_AMASS 9 VT419 EU340305 Rhizo1_AMASS 4 VT090 Y17648 Rhizo2_AMASS 5 VT113/114 AJ418876 Sclero1_AMASS 3 VT310 FJ194510 Scut1_AMASS 10 VT049 AF074340 Uncult1_AMASS 11 VT242/151 AB015052

* _{CodenumberusedinFig.1.}

Table4

Arbuscularmycorrhizal(AM)fungalmolecularoperationaltaxonomicunit(MOTU) richnessandShannon(H0_{)indexwithinthenativeplantspeciesrootsandthesoilof}

abandonedagriculturalpeatysoils(Aband)andwithinthemaizerootsandthesoil ofamaizecultivation(Cult).

Roots Soil

Parameter Aband Cult Aband Cult MOTUrichness 4.31.7*

1.70.3A 3.30.3a 6.00.6bB Shannonindex(H0_{) 1.10.5 0.40.2A 1.00.1 1.50.2B} *

ValuesaremeansSEofthreefieldreplicatesforeachtreatment.Valuesinthe samerowfollowedbydifferentsmalllettersarestatisticallydifferentbetweenland uses,accordingtoANCOVA(P<0.05).

RhandRasteadilyincreasedfollowingthetrendofairandsoil

temperatures (Fig. 3c–e). Rs ranged from 19.97 to 54.95g

CO2m2day1 and from 9.74 to 63.35g CO2m2day1 in the

Aband and in the Cult, respectively, while Rh from 17.34 to

35.43gCO2m2day1andfrom6.79to29.12gCO2m2day1in

the Abandand theCult,respectively.

TheAbandshowed significantlyhigherRs valuesin thefirst

periodofthemonitoring(fromthemiddletotheendofMay)than

Fig.2.PartialRedundancyAnalysis(pRDA)biplotbasedonsoilrelativeabundancesofarbuscularmycorrhizalfungal(AMF)molecularoperationaltaxonomicunits(MOTUs) (Funneliformissp.,Fun1_AMASS;Glomusspp.,fromGlo1_AMASStoGlo5_AMASS;Rhizophagusspp.,Rhizo1_AMASSandRhizo2_AMASS;Sclerocystissp.,Sclero1_AMASS; Scutellosporasp.,Scut1_AMASS;unculturedGlomeromycota,Uncult1_AMASS)usedasresponsevariable,andlanduses(agriculturalabandonedpeatysoils,Aband;amaize cultivation,Cult)usedasenvironmentalvariables.Plotcoordinates(latitude/longitude)wereusedascovariables(a).RDAbiplotsoftheAMFMOTUs,usedasresponse variables,andthematrixes,soilandroots,usedasenvironmentalvariablesintheCult(b)andtheAband(c).pRDAbiplotbasedonsoilchemicalparameters(pH;electrical conductivity,EC;exchangeablepotassium,Kexch;totalnitrogen,Ntot;ammonium;nitrates;soilorganicmatter,SOM;carbon/nitrogenratio,C/N;totalphosphorus,Ptot; availablephosphorus, Pavail;organicphosphorus, Porg);CO2 flux(totalsoilrespiration, Rs;respirationby heterotrophs,Rh;respiration byautotrophs,Ra); root

measurements(totalrootdryweight,RDW;AMFrootcolonization,AMFc);AMFdiversity(AMFMOTUrichnessofrootsandsoil,MOTUrandMOTUs,respectively);and Shannonindex(H’)ofrootsandsoil,H0_rootsandH0_{soil,respectively),usedasresponsevariables;landuses,AbandandCult,usedasenvironmentalvariables.Plotcoordinates}

(latitude/longitude)wereusedascovariables(d).The1stand2ndaxesaccountedfor87.2%,89.7%,89.2%and74.7%ofthetotalvarianceexplainedbyallcanonicalaxesfora,b, candd,respectively.TheMonteCarlopermutationaltestsshowedthatAMFassemblageswerestatisticallydifferentbetweenthesoiloftheAbandandoftheCultand betweensoilandrootsofbothCultandAbandsystems(P=0.036,a;P=0.002,bandc).Landuseswerealsostatisticallydifferentconsideringsoilchemicalparameters,CO2

theCult,whereas, later,weobservedanopposite trend.TheRa rangedfrom2.63to27.82gCO2m2day1andfrom2.33to38.55g

CO2m2day1intheAbandandtheCult,respectively.Rainthe

Cult was significantly higher than in the Aband in the second

halvesofJuneandofJuly,whileRsonlyinJuly(Fig.3c,e). 3.2.2.ModelselectionofsoilCO2fluxresponsetoairtemperature

TocalculatethecumulatedvaluesoftheRsandRhofthesoil CO2respiration,weselectedthebestofthemostcommonlyused

models: Q10, LT and simplified RothC. These models were

compared onthebasis ofAIC, RSMEandRsq. ad(Table 5).The

LTmodelshowedthebestfitforRsandRhintheAbandandthe

Cult.TheLTmodelminimizedtheAICandtheRSMEvalues,while

thesimplifiedRothCmodelmaximizedtheRsq.advalues(Table5).

The LTmodelwas thus selectedfor modelingthesoil CO2 flux

responsetoairtemperature.

The plots of the model fitting are shown in Fig. S7. Clear

relationships betweenRsandRh fluxand airtemperaturewere

observedinbothlanduses.IntheAbandsigni_ficantrelationships

frommoderatetoweakwereobserved(RsRsq.ad=0.5;RhRsq.

ad=0.3;P<0.001)(Table5;Fig.S7b,d,respectively),whileinthe Cult strong andsignificantrelationships wererevealed(RsRsq. ad=0.7;Rh Rsq.ad=0.8;P<0.001)(Table5; Fig.S7a,c, respec-tively).

3.2.3.ImpactofagriculturalabandonmentontheresilienceofCO2flux

andoncumulatedsoilCO2flux

ThecoefficientsofvariationoftheRsandRhcomponentswere signi_ficantlylowerintheAbandthanintheCult,thusshowinga

higher fluctuation of soil CO2 flux response toair temperature

(Table6).

ThecurvesoftheRs,RhandRacumulatedCO2fluxesareshown

inFig.4.IntheAband,RsandRacumulatedCO2flux(399976

gm2period1 and 141964gm2 period1, respectively) was significantlylowerthanintheCult(436043gm2period1and

231651gm2 period1, respectively) (Fig. 4). Rs and Ra

cumulated soilCO2 fluxvariations,calculatedas((Aband–Cult)/

C- u- l- t))- - 1- 0-0,

Fig.3.Dailymaximum,meanandminimumtemperatures(_{C)andtotalrainfall}

(mm)overthemonitoringcampaign(a).Meansoiltemperature(C)(shownas circle)andmoisture(%;v:v)(shownassquare)ofagriculturalabandonedpeatysoils (opensymbols; Aband)andamaizecultivation(filled symbols;Cult)(b).Soil temperatureandmoistureweremeasuredoutsidethesixblocks(seeFig.S1)used duringtheCO2fluxmonitoringcampaign.TwocomponentsofthesoilCO2fluxes,

soilrespiration(Rs;shownastriangle)(c),respirationbyheterotrophs(Rh;shown asdiamond)(d)weremeasuredinthetwodifferentlandusesAbandvsCult(open andfilledsymbols,respectively).Respirationbyautotrophs(Ra;shownascircle) wascalculatedasdifferencebetweenRsandRh.Themonitoringcampaignranges fromthe14thofMaytothe13thofAugust2012withoneortwosoilCO2flux

measurementsperweek(n=21).ValuesaremeansSEofthreereplicateplotsfor eachlanduse.ForeachsamplingdateandsoilCO2fluxcomponent,statistically

significantdifferencesbetweenlandusesareshownbydifferentlettersaccording toANCOVA(P<0.05).

Table5

ComparisonofthreemodelsofCO2fluxresponsetoairtemperatureusingdifferent

modelselectioncriteria:theAkaikeInformationCriterion(AIC),theRootMean SquaredError(RMSE)andtheAdjustedR-squarevalue(Rsd.ad).Themeasurements (n=21)wererecordedfromMaytoAugust2012inabandonedagriculturalpeaty soils(Aband)andinamaizecultivation(Cult).

Criteria Q10*

LT RothC Rs Aband Cult Aband Cult Aband Cult AIC 317.9** 472.9 317.2 469.9 365.3 531 RMSE 9.1 9.8 9 9.6 16.2 15.9 Rsq.ad 0.5 0.6 0.5 0.7 0.7 0.8 Rh AIC 296.6 331.7 296.6 326.4 334.5 384.8 RMSE 7.1 3.2 7.1 3.1 11.3 5 Rsq.ad 0.3 0.8 0.3 0.8 0.7 0.8 *

Q10:Van’tHoff’sQ10model;LT:LloydandTaylormodel;RothC:simplified Rothamstedcarbonmodel.

were8%and39%,respectively.Incontrast,theRhcomponent

showedanoppositebehaviorwithsignificantlyhighervaluesin

theAband(2580_80gm2period1),thanintheCult(2045_8 gm2period1).IntheCult,theRhandRasoilCO2fluxpartitioning

was47%and53%,respectively,whereasintheAbanditwas65%

and35%,respectively.

3.3.Mainpatternsofsoilchemicalparameters,CO2fluxandAMfungal

diversityasaffectedbyagriculturalabandonment

pRDAshowedthatlanduseexplained74.7%(IandIIaxes)ofthe wholevariance,andthatitseffectonsoilqualityparameterswas significant(P=0.05),asshownbythebiplotarrowsrepresenting theAbandandtheCult(Fig.2d).Thebiplotclearlysuggestedthat

themostdiscriminantvariablesbetweenthetwolanduseswere

NH4+,soilAMFMOTUrichness,RaandRs,whichshowedlower

valuesintheAbandcomparedtotheCult,andRhandAMfungal

colonization,whichshowedhighervaluesintheAbandthanthe

Cult.

4.Discussion

Inthisworkweassessedforthefirsttimethelong-termeffect

onsoil qualityofagriculturalabandonmentin aMediterranean

reclaimed peaty soil. Agricultural abandonment (Aband) was

comparedtoanintensivelycultivatedpeatysoilrepresentedby

continuous maize (Cult). Multivariate analyses showed that

15yearsofagriculturalabandonment:(i)didnotaffectthemain soilchemicalparameters,exceptforNH4+whichwaslowerinthe

AbandthanintheCult;(ii)increasedAMfungalrootcolonization

androotAMfungaltaxonomicdiversity,butdecreasedsoilMOTU

richness;(iii)reducedsoiland rootrespirationtogetherwithan

increase in respiration by heterotrophs; (iv) increased soil

resiliencetoairtemperatureintermsofCO2fluxresponse.

4.1.Effectofagriculturalabandonmentonsoilchemicalparameters,

rootsandAMfungi(Experiment1)

4.1.1.Soilchemicalparameters

Tothebestofourknowledge,ourworkisthefirstonetoreport ontheimpactoftheabandonmentoftheagriculturalpracticesin organicdrainedsoil.However,ourresultsconfirmedpreviousdata

detected in mineral abandoned soils which, compared to

conventionalfarming,showednochangesinthemainchemical

aspects of soil quality due to fallow, land abandonment or

grassland reestablishment(Liebig et al.,2004;Pellegrino et al., 2011;Belletal.,2012).

On the otherhand, our findings are in contrast with other

studiesreportingnegativechangesduetolanduseintensification incomparisonwithagriculturalabandonmentorgrassland(Celik,

2005; Marriott et al., 2010; Gajic, 2013).These inconsistencies

couldbeattributedtothesamplingdepthandtothetimeelapsed sinceland-useconversion(Liebigetal.,2004;Jinboetal.,2007; Raiesi,2012;Gajic,2013).Infactdifferenceswereobservedonlyat

a 0–15cm depth and already six years after abandonment.

Interestingly,inlinewithourresults, Ewingetal. (2012)found nosignificantchangesinorganicsoilsoftotalorganiccarbonafter

15yearsofcropproductioninreclaimedwetlandsincomparison

withnaturalwetlandsoils.

ThefactthattheonlydifferenceobservedwasinthesoilNH4+

concentrationcouldbeexplainedbythehightemporalvariability ofsuchachemicalcompound(Violante,2000)orbythecloseness ofthesamplingtotheNfertilizationintheCult.

Concerning the C/N ratio, our data are in the value range

reportedforhighly-decomposedcultivatedpeatysoils(Berglund

etal.,2010).

4.1.2.RootdryweightandAMfungalrootcolonization

Theroot dryweight wasintherange of thoseobtainedfor

uncultivated natural fallow and maize (0.04–1.8mgg1)

(Kothari et al., 1990; Mekonnen et al., 1997). Despite the

well-known highly mycotrophic status of maize (Gavito and

Varela,1993), inour study theAM fungal rootcolonizationof

maizewaslowerthanthenaturalplantspeciesoccurringinthe

Aband. This cannot be explained by the plant species

composition in the Aband, since most of the plant species is

known to be a non-mycorrhizal status (Harley and Harley,

1987).Thereforethedifferenceis verylikelyduetothetillage,

fertilization, mechanical and chemical weeding carried out in

theCult(Helgasonetal.,1998;Vandenkoornhuyseetal.,2002;

Borriello et al., 2012). 4.1.3.AMfungaldiversity

AlthoughinthepasttheimportanceofAMFinwetlandswas

nottakenintoaccount,ourstudyontheAMfungaldiversityofa Mediterraneanreclaimedpeatysoilswasboostedbytheincreased

awareness of their occurrence and functionality in these key

ecosystems(Wolfeetal.,2007).

RootandsoilMOTUrichnessandH0fellintoasimilarrangeto previousworks (Helgasonetal.,1998; Danielletal.,2001;Hijri

etal., 2006;Borrielloet al.,2012), butweremostlylowerthan

thosereportedbyotherauthors(Jansaetal.,2002;Wolfeetal.,

2007;Oehletal.,2010).Theseinconsistenciesmightbeduetothe

pedo-climatic dissimilarities, but also to the differences in

detectionmethodologies.RootMOTUrichnessand H0 suggested

thattheAMfungaldiversitywasnotdepletedbytheintensi fica-tionofcultivation,inlinewithfindingsinfieldandmicrocosms (Danielletal.,2001;Johnsonetal.,2004;Hijrietal.,2006).The

Table6

Coefficientofvariationofthetotalsoilrespiration(Rs)andoftherespirationby heterotrophs(Rh)andbyautotrophs(Ra)oftwolanduses:abandonedagricultural peatysoils(Aband)vsamaizecultivation(Cult)overthemonitoringcampaign.

Parameter Coefficientofvariation

Aband Cult CV_Rs(%) 23.3* a 32.0b CV_Rh(%) 19.1a 28.9 b CV_Ra(%) 39.3 36.7 *

Valuesinthesamerowfollowedbydifferentlettersarestatisticallydifferent betweenlandusesaccordingtotheMann–Whitneynonparametrictest(P<0.05).

Fig.4.CumulatedsoilCO2flux:soilrespiration,Rs(shownastriangle);respiration

byheterotrophs,Rh(shownasdiamond)andrespirationbyautotrophs,Ra(shown ascircle).Themeasurements(n=21)wererecordedfromMaytoAugust2012in agriculturalabandonedpeatysoils(Aband;opensymbols)andamaizecultivation (Cult;filledsymbols).Valuesaremeansofthreereplicateplotsforeachlanduse. Statisticallysignificantdifferencesbetweenlandusesareshownbydifferentletters accordingtotheANCOVA(P<0.05).

higherMOTUrichnessintheCultsoilthanintheAbandsoilmaybe

duetoweedmycotrophiccomposition(Poaceae,Solanaceaeand

Phytolaccaceae).

Glomeraceaewastheonlyfamilyretrievedinthesoil,whilea

memberofGigasporaceaewasalsodetectedwithintheroots.Our

dataareconsistentwithpreviousobservations:Glomeraceaeare

mainlyretrieved in agricultural soil and wetlands, while

Giga-sporaceaeare morefrequent in uncultivatedor woodlandsites

(Helgasonetal., 1998;Danielletal.,2001;Jansaetal.,2002;Wirsel,

2004;Wolfeetal.,2007).

Within Glomeraceae, the dominance of the Rhizo2_AMASS

within maize roots suggests that it can survivein agricultural

conditionsduetoitsabilitytocolonizetherootsfrommycelium

fragments(BiermannandLinderman,1983).Thepresencewithin

the roots of the natural plant species of a MOTU that is

phylogeneticallyaffiliatedtoScutellosporasp.islikelyduetothe lackofsoiltillagedisruptionoftheextraradicalhyphae,whichis essentialforthepropagationofGigasporaceae(Jasperetal.,1993).

Despite several authorshighlighting thedominance of

Funneli-formismosseae in arable land,grassland and alsowetlands, we

unexpectedlyretrieved Funneliformis sp. onlyin the soil of the Aband,anditwasnotthemostrepresentativespecies.

4.2.EffectofagriculturalabandonmentonsoilCO2flux(Experiment2)

4.2.1.SoilCO2fluxmeasurementsandcumulatedvalues

Differentiating the soil CO2 flux into its autotrophic and

heterotrophiccomponentsisimportantforinterpretingsoilCO2

sources(Baggs,2006;Kuzyakov,2006).Inourstudy,thesoilCO2

sourcesweresuccessfullymonitoredinabandonedandcultivated

peatysoilsusingtherootexclusionmethod.Inaddition,inorderto fillrespirationgaps andtoassess thetotalrespirationover the

monitoredperiod, theair temperaturewas usedasan

environ-mentalvariableforCO2modeling.Infact,althoughsomeauthors

havemodeled soil respirationusingenvironmental parameters,

suchassoilwatercontentandwatertabledepth(Almagroetal.,

2009;Correiaetal.,2012;Rowsonetal.,2013),mostauthorshave

found a highly significant relationship between soil or air

temperatureandsoilrespiration(LuoandZhou,2006;Richardson

etal.,2006;Subkeetal., 2006).Onthebasis ofthree different

criteria,wethusselectedtheLTmodelintermsoftheQ10andthe simplifiedRothC,inagreementalsowithDavidsonetal.(2006)and

LuoandZhou(2006).

WeusedthecumulatedvaluesoftheRs,extrapolatedusingthe LTmodel,toobtainthemeansofthedailyRsvalues.Thecalculated

dailyRs fluxes of the Abandand the Cult overthe monitoring

period(43.47and47.39gCO2m2day1,respectively)were

two-foldhigherthanreportedbyseveralauthorsasannualmeandaily values(from0.77to26.6gCO2m2day1)forfensandpeatlandsin

boreal and temperate areas (Silvola et al., 1996;

Kasimir-Klemedtsson et al., 1997; Danev_9ci9c et al., 2010; Heinemeyer

etal.,2011).Thislargefluxescouldbeexplainedbythefactthatwe

monitoredthespring–summerperiod,characterizedbyawarmer

averagetemperature(21.5_{C)thantheannualaverage}

tempera-ture(14.8C).Althoughthisestimationneedstobeconfirmedbya long-termmonitoring,astrongmineralizationrateispointedout inMediterraneanpeatysoilssubjectedtoagriculturalreclamation, implyingalargelossofSOCandsoildepth.

Inadditiontosuchfactors,thenotlimitingsoilwatercontent

couldhavedeterminedlargerfluxescomparedtothosecommonly

registeredin mineralsoilsinMediterraneanareas(from3.19to 25.09gCO2m2day1) (Almagro et al., 2009; Mancinelli et al.,

2010;Correiaetal.,2012).

Withregard tothe partitioning,the tworespiration

compo-nentsshowed anopposite performancesince theheterotrophic

componentwas predominantintheAbandand theautotrophic

componentintheCult.Anexplanationforthesignificantlyhigher dailyvaluesoftheRacomponentintheCultistobefindinthe boostofthefinerootsduringthegrowingseasonofthemaizecrop

(RochetteandFlanagan,1997;WerthandKuzyakov,2009).Such

effectresultsinaprogressiveincreaseofthefluxfromthesowing (firstpartofMay)tothecropdevelopment(midandlatesummer).

As suggestedby Subkeet al. (2006), we alsocompared our

resultsonsoilCO2partitioningusingtheRh/Rsratio.TheRh/Rs

ratiorangedfrom0.65 intheAband,withvalues similartothe

pristinepeatlandto0.47theCult,similartoothercroplandsvalues.

These ratios were similar to those reported in Mediterranean

croplands,wherethemeanratiowasaround0.50(Subkeetal.,

2006).Incontrast,ourratioswerelowerthanthosereportedfor

borealandtemperatepeatlands,rangingfrom0.72to0.97(Silvola

etal.,1996;WunderlichandBorken,2012).

Whenconsideringvaluescomingfromrootexclusionmethods

itisimportanttotakeinaccountthebiasthatthismethodimplies. Itisknownthatthemajorconcernassociatedwiththistechnique resultsinaincreaseinthedeadrootbiomassinthedeepcollar (comparedtotheshallowone)thatcontributetoRhandleadtoan

underestimation of Ra (Baggs, 2006; Subke et al., 2006;

Heinemeyer et al., 2007). Moreover, some authors highlighted

directcorrelationsbetweenRsandproductivityandproposedthat

inmoreproductivesystems,asthecaseofMediterraneanpeaty

soils compared to boreal peatlands, a greater amount of

assimilatedCisallocatedtoRa,thusreducingRh/Rsratio(Raich andTufekciogul,2000;Subkeetal.,2006).Finally,anotherfactorto

takeinconsiderationwhencomparingourratiovalues toother

ecosystemsisthatwemonitoredthespring-summerperiodwhen

theRacomponent,followingthetypicalannualphenologyofthe

vegetation,isatitsmaximum(Silvolaetal.,1996). 4.2.2.ResilienceoftheCO2fluxes

Inagreementwithourdata,comparingthefluctuationofthe

CO2fluxintheAbandcomparedtotheCult,Jacksonetal.(2003)

reportedalowerresilienceinthetilledcroppingsystemsthanin thenon-tilledones.Thiscouldbeexplainedbythehigherporosity, theinitiallowerbulkdensityandthelowerporeconnectivityof thelong-termtilledsoilscomparedtothenon-tilledsoils(Silgram

and Shepherd, 1999). Laliberté et al. (2010) also linked the

reductioninresiliencefollowinglanduseintensificationtoplant diversityandfunctionality.Togetherwiththis,ourmaizesystem

alsoshowedlowerplantdiversitythanin theabandonedpeaty

soil.

5.Conclusions

Our analysis of the chemical changes and of the peat

mineralization measured using respiration by heterotrophs

revealed15yearsofagriculturalabandonmentdoesnot necessari-lyleadtotheeffectiverestorationofaMediterraneanreclaimed peatlandintermsofthesoilquality.ThelargeCO2fluxthatwe

observed demonstrates the strong degradation rate of peat–

whethercultivatedorabandoned_–andtheimportanceofthis_flux

asadiffusesourceofCO2atthelandscapescale.Asaconsequence,

ithighlightsastrongmineralizationrateinMediterraneanpeaty soilssubjectedtoreclamation,implyinganimportantlossinSOC andinsoildepth.

However,thereweresomepositiveeffectsintheabandonment: anincreaseofthediversityintermsofnumberoffamiliesofAMF retrievedinroots,asmallreductionoftotalsoilCO2respirationand

alowerfluctuationofsoilCO2fluxresponsetoairtemperature.

Webelievethatourfindingscanbeusedtobetter

understand-ing on how best to protect and preserve the Mediterranean

peatlands.Theycanhelptodevelop alternativeandsustainable

hydrologicalconditions(e.g.,rewetting)asamajorfactortotrigger therestorationandtolimitthemineralizationratethuscontrolling thesubsidenceeffect.

Acknowledgements

Thisworkwassupportedbythe“ConsorziodiBonifica Versilia-Massaciuccoli”andfundedbythe“RegioneToscana”asapartof

the project “Restoration of a Mediterranean Drained Peatland”

(Restomedpeatland;

https://sites.google.com/site/restomedpeat-land/home).WewishtothankthetechnicalstaffoftheConsorzio

diBoni_ficaVersilia-Massaciuccoliformanagingtheexperimental site,andFabioTacciniandCindyCardenasforhelpinsoilandCO2

samplings.SincerestthankstoDr.GiorgioRagagliniandDr.Agata Klimkowskaforusefuldiscussions,toNikkiYeohandAdrianJohn WallworkforrevisingtheEnglishandtotheanonymousEditorfor constructivecriticisms.AspecialthanksgotoDr.PetrŠmilauerand Prof.MartiJ.Andersonforstatisticaladvice.

AppendixA.Supplementarydata

Supplementary data associated with this article can be

found, in the online version, at http://dx.doi.org/10.1016/j.

agee.2014.09.004. References

Almagro, M.,López, J., Querejeta, J.I., Martínez-Mena,M., 2009.Temperature dependenceofsoilCO2effluxisstronglymodulatedbyseasonalpatternsof

moistureavailability inaMediterraneanecosystem.SoilBiol.Biochem.41, 594–605.

Altschul,S.F.,Madden,T.L.,Schäffer,A.A.,Zhang,J.,Zhang,Z.,Miller,W.,Lipman,D.J., 1997.GappedBLASTandPSI-BLAST:anewgeneration ofprotein database searchprograms.NucleicAcidsRes.25,3389–3402.

Baggs,E.M., 2006.Partitioningthecomponentsofsoilrespiration:aresearch challenge.PlantSoil284,1–5.

Bell,L.W.,Sparling,B.,Tenuta,M.,Entz,M.H.,2012.Soilprofilecarbonandnutrient stocks under long-term conventional and organic crop and alfalfa-crop rotationsandre-establishedgrassland.Agric.Ecosyst.Environ.158,156–163.

Berglund,Ö.,Berglund,K.,Klemedtsson,L.,2010.Alysimeterstudyontheeffectof temperature on CO2 emission from cultivated peat soils. Geoderma 154,

211–218.

Biermann,B.,Linderman,R.G.,1983.Mycorrhizalroots:intraradicalvesiclesand extraradicalvesiclesasinoculum.NewPhytol.95,97–105.

Borriello,R.,Lumini,E.,Girlanda,M.,Bonfante,P.,Bianciotto,V.,2012.Effectsof differentmanagementpracticesonarbuscularmycorrhizalfungaldiversityin maizefieldsbyamolecularapproach.Biol.Fertil.Soils48,911–922.

Celik,I.,2005.Land-useeffectsonorganicmatterandphysicalpropertiesofsoilina southernMediterraneanhighlandofTurkey.SoilTill.Res.83,270–277.

Clarke,K.R.,Gorley,R.N.,2006.PRIMERv6:UserManual/Tutorial.PRIMER-ELtd., Plymouth.

Coleman,K.,Jenkinson,D.S.,1999.ROTHC-26.3.AModelfortheTurnoverofCarbon inSoils.Herts,RothamstedResearch,HarpendenHertfordshire,UK.

Coleman,B.D.,1981.Onrandomplacementandspecies–arearelations.Math.Biosci. 54,191–215.

Colwell,R.K.,2013.Estimates:StatisticalEstimationofSpeciesRichnessandShared SpeciesfromSamples.Version9andearlier.User’sGuideandapplication. Publishedat:http://purl.oclc.org/estimates.

Correia, A.C., Minunno, F., Caldeira, M.C., Banza, J., Mateus, J., Carneiro, M., Wingate,L.,Shvaleva,A.,Ramos,A.,Jongen,M.,Bugalho,M.N.,Nogueira,C., Lecomte,X.,Pereira,J.S.,2012.Soilwateravailabilitystronglymodulatessoil CO2effluxindifferentMediterraneanecosystems:modelcalibrationusingthe

Bayesianapproach.Agric.Ecosyst.Environ.161,88–100.

Danev9ci9c,T.,Mandic-Mulec,I.,Stres,B.,Stopar,D.,Hacin,J.,2010.EmissionsofCO2,

CH4 and N2Ofrom Southern Europeanpeatlands. Soil Biol.Biochem. 42,

1437–1446.

Daniell,T.J.,Husband,R.,Fitter,A.H.,Young,J.P.W.,2001.Moleculardiversityof arbuscularmycorrhizalfungicolonisingarablecrops.FEMSMicrob.Ecol.36, 203–209.

Davidson,E.A.,Belk,E.,Boone,R.D.,1998.Soilwatercontentandtemperatureas independentorconfoundedfactorscontrollingsoilrespirationinatemperate mixedhardwoodforest.GlobalChangeBiol.4,217–227.

Davidson,E.A.,Janssens,I.A.,Luo,Y.,2006.Onthevariability ofrespirationin terrestrialecosystems:movingbeyondQ10.GlobalChangeBiol.12,154–164.

Davison,J.,Öpik,M.,Zobel,M.,Vasar,M.,Metsis,M.,Moora,M.,2012.Communities ofarbuscularmycorrhizalfungidetectedinforestsoilarespatially heteroge-neousbutdonotvarythroughoutthegrowingseason.PloSOne7,e41938.

DiBene,C.,Pellegrino,E.,Debolini,M.,Silvestri,N.,Bonari,E.,2013.Short-and long-termeffectsofolivemillwastewaterlandspreadingonsoilchemicaland biologicalproperties.SoilBiol.Biochem56,21–30.

Doran,J.W.,Parkin,T.B.,1996.Quantitativeindicatorsofsoilquality:aminimum dataset.In:Doran,J.W.,Jones,A.J.(Eds.),MethodsforAssessingSoilQuality. SSSAInc.,Madison,WI,USA.

Ewing,J.M.,Vepraskas,M.J.,Broome,S.W.,White,J.G.,2012.Changesinwetlandsoil morphologicalandchemicalpropertiesafter15,20,and30yearsofagricultural production.Geoderma179,73–80.

Gajic,B.,2013.PhysicalpropertiesandorganicmatterofFluvisolsunderforest grassland,and100yearsofconventionaltillage.Geoderma200–201,114–119.

Gavito,M.E.,Varela,L.,1993.Seasonaldynamicsofmycorrhizalassociationsin maizefieldsunderlow-inputagriculture.Agric.Ecosyst.Environ.45,275–282.

Gotelli,N.J.,Colwell,R.K.,2001.EstimatingSpeciesRichness.FrontiersinMeasuring Biodiversity.OxfordUniversityPress,NewYork.

Hall,T.A.,1999.BioEdit:auser-friendlybiologicalsequencealignmenteditorand analysisprogramforWindows95/98/NT.Nucl.Acid.Symp.41,95–98.

Harley,J.L.,Harley,E.L.,1987.Acheck-listofmycorrhizaintheBritishflora–second AddendaandErrata.NewPhytol.115,699–711.

Heinemeyer,A.,Hartley,I.P.,Evans,S.P.,CarreiraDeLaFuente,J.A.,Ineson,P.,2007. ForestsoilCO2flux:uncoveringthecontributionandenvironmentalresponses

ofectomycorrhizas.GlobalChangeBiol.13,1786–1797.

Heinemeyer,A., DiBene, C.,Lloyd,A.R., Tortorella,D.,Baxter, R.,Huntley,B., Gelsomino,A.,Ineson,P.,2011.Soilrespiration:implicationsoftheplant-soil continuumandrespirationchambercollar-insertiondepthonmeasurement andmodellingofsoilCO2effluxratesinthreeecosystems.Eur.J.SoilSci.62,

82–94.

Helgason,T.,Daniell,T.J.,Husband,R.,Fitter,A.H.,Young,J.P.,1998.Ploughingupthe wood-wideweb?Nature394,431.

Hence,R.J.,Anderson,G.,1962.Acomparativestudyofmethodsofestimatingsoil organicphosphate.J.SoilSci.13,225–230.

Hijri,I.,Sýkorová,Z.,Oehl,F.,Ineichen,K.,Mäder,P.,Wiemken,A.,Redecker,D.,2006. Communitiesofarbuscularmycorrhizalfungiinarablesoilsarenotnecessarily lowindiversity.Mol.Ecol.15,2277–2289.

IPCC,2006.In:Eggleston,H.S.,Buendia,L.,Miwa,K.,Ngara,T.,Tanabe,K.(Eds.),IPCC Guidelinesfor NationalGreenhouse GasInventories,4. Intergovernmental PanelonClimateChange.

IUSSWorkingGroupWRB,2006.Worldreferencebaseforsoilresources.World.Soil ResourcesReportsNo.103.FAO,Rome.

Jackson, L.E., Calderon, F.J., Steenwerth,K.L., Scow, K.M., Rolston,D.E., 2003. Responsesofsoilmicrobialprocessesandcommunitystructuretotillageevents andimplicationsforsoilquality.Geoderma114,305–317.

Jansa,J.,Mozafar,A.,Anken,T.,Ruh,R.,Sanders,I.R.,Frossard,E.,2002.Diversityand structureof AMFcommunities asaffected bytillage in atemperate soil. Mycorrhiza12,225–234.

Jasper,D.A.,Abbott,L.K.,Robson,A.D.,1993.Thesurvivalofinfectivehyphaeof vesicular–arbuscular mycorrhizal fungi in dry soil and interaction with sporulation.NewPhytol.124,473–479.

Jinbo,Z.,Changchun,S.,Shenmin,W.,2007.Dynamicsofsoilorganiccarbonandits fractionsafterabandonmentofcultivatedwetlandsinnortheastChina.SoilTill. Res.96,350–360.

Johnson,D.,Vandenkoornhuyse,P.J.,Leake,J.R.,Gilbert,L.,Booth,R.E.,Grime,J.P., Young,J.P.W.,Read,D.J.,2004.Plantcommunitiesaffectarbuscularmycorrhizal fungaldiversityandcommunitycompositioningrasslandmicrocosms.New Phytol.161,503–515.

Joosten,H.,Clarke,D.,2002.WiseUseofMiresandPeatlands.Backgroundand Principles Including a Fraimwork for Decision-Making.International Mire ConservationGroupandInternationalPeatSociety,Saarijärvi,Finland.

Kasimir-Klemedtsson,Å.,Klemedtsson,L.,Berglund,K.,Martikainen,P.,Silvola,J., Oeema,O.,1997.Greenhousegasemissionsfromfarmedorganicsoils:areview. SoilUseManage.13,245–250.

Kimura,M.,1980.Asimplemethodforestimatingevolutionaryrates ofbase substitutionsthroughcomparativestudiesofnucleotidesequences.J.Mol.Evol. 16,111–120.

Kothari,S.K.,Marschnerf,H.,George,E.,1990.EffectofVAmycorrhizalfungiand rhizospheremicroorganismsonrootandshootmorphology:growthandwater relationsinmaize.NewPhytol.116,303–311.

Kottek,M.,Grieser,J.,Beck,C.,Rudolf,B.,Rubel,F.,2006.WorldmapoftheKöppen– Geigerclimateclassificationupdated.Meteorol.Z.15,259–263.

Krüger,M.,Krüger,C.,Walker,C.,Stockinger,H.,Schüssler,A.,2012.Phylogenetic referencedataforsystematicsandphylotaxonomyofarbuscularmycorrhizal fungifromphylumtospecieslevel.NewPhytol.193,970–984.

Kuzyakov,Y.,2006.SourcesofCO2effluxfromsoilandreviewofpartitioning

methods.SoilBiol.Biochem.38,425–448.

Laliberté,E.,Wells,J.A.,Declerck,F.,Metcalfe,D.J.,Catterall,C.P.,Queiroz,C.,Aubin, I.,Bonser,S.P.,Ding,Y.,Fraterrigo,J.M.,McNamara,S.,Morgan,J.W.,Merlos,D. S., Vesk, P.A., Mayfield, M.M., 2010. Land-use intensification reduces functionalredundancyandresponsediversityinplantcommunities.Ecol. Lett.13,76–86.

Lepš,J.,Šmilauer,P.,2003.MultivariateAnalysisofEcologicalDataUsingCANOCO. CambridgeUniversityPress,Cambridge.

Liebig,M.,Tanaka,D.,Wienhold,B.,2004.Tillageandcroppingeffectsonsoilquality indicatorsinthenortherngreatplains.SoilTill.Res.78,131–141.

Lloyd,L.,Taylor,J.,1994.Onthetempearturedependenceofsoilrespiration.Funct. Ecol.8,315–323.

Mancinelli,R.,Campiglia,E.,DiTizio,A.,Marinari,S.,2010.Soilcarbondioxide emissionandcarboncontentasaffectedbyconventionalandorganiccropping systemsinMediterraneanenvironment.Appl.SoilEcol.46,64–72.

Marriott,C.A.,Fisher,J.M., Hood,K.,Pakeman,R.J.,2010. Impactsofextensive grazingandabandonmentongrasslandsoilsandproductivity.Agric.Ecosyst. Environ.139,476–482.

McGonigle,T.P.,Miller,M.H.,Evans,D.G.,Fairchild,G.L.,Swan,J.A.,1990.Anew methodwhichgivesanobjectivemeasureofcolonizationofrootsbyvesicular– arbuscularmycorrhizalfungi.NewPhytol.115,495–501.

Mekonnen,K.,Buresh,R.J.,Jama,B.,1997.Rootandinorganicnitrogendistributions insesbaniafallow,naturalfallowandmaizefields.PlantSoil188,319–327.

Montanarella,L.,Jones,R.J.A.,Hiederer,R.,2006.Thedistributionofpeatlandin Europe.MiresPeat1,1–10.

Nelson,D.W.,Sommers,L.E.,1982.Totalcarbon,organiccarbonandorganicmatter, seconded.In:Page,R.H.,Keeney,D.R.(Eds.),MethodsofSoilAnalysis,Part2, Chemical and Microbiological Properties, Agronomy Monograph, vol. 9. AmericanSocietyofAgronomy,Madison,WI,pp.539–579.

Oehl,F.,Laczko,E., Bogenrieder,A., Stahr,K., Bösch,R.,vander Heijden,M., Sieverding, E., 2010. Soil type and land use intensity determine the composition of arbuscular mycorrhizal fungal communities. Soil Biol. Biochem.42,724–738.

Olsen,S.R.,Sommers,L.E.,1982.Phosphorus,seconded.In:Page,A.L.,Miller,R.H., Keeney,D.R.(Eds.),MethodsofSoilAnalysis,Part2,Chemicaland Microbio-logicalProperties,AgronomyMonograph,vol.9.AmericanSocietyofAgronomy, Madison,WI,pp.403–430.

Öpik,M.,Vanatoa,A.,Vanatoa,E.,Moora,M.,Davison,J.,Kalwij,J.M.,Reier,U.,Zobel, M., 2010. The online database MaarjAM reveals global and ecosystemic distributionpatternsinarbuscularmycorrhizalfungi(Glomeromycota).New Phytol.188,223–241.

Pellegrino,E.,DiBene,C.,Tozzini,C.,Bonari,E.,2011.Impactonsoilqualityofa 10-year-oldshort-rotation coppice poplar stand compared with intensive agriculturalanduncultivatedsystemsinaMediterraneanarea.Agric.Ecosyst. Environ.140,245–254.

Pellegrino,E.,Turrini,A.,Gamper,H.A.,Cafà,G.,Bonari,E.,Young,J.P.W.,Giovannetti, M.,2012.Establishment:persistenceandeffectivenessofarbuscular mycor-rhizalfungalinoculantsinthefieldrevealedusingmoleculargenetictracing andmeasurementofyieldcomponents.NewPhytol.194,810–822.

Pfadenhauer,J.,Grootjans,A., 1999.WetlandrestorationinCentralEurope:aimsand methods.Appl.Veg.Sci.2,95–106.

Phillips,J.M., Hayman,D.S.,1970.Improved proceduresforclearingrootsand staining parasitic and vesicular-arbuscular mycorrhizal fungi for rapid assessmentofinfection.Trans.Br.Mycol.Soc.55,158–161.

Picci,G.,Nannipieri,P.,2002.MetodidiAnalisiMicrobiologicadelSuolo.Franco Angeli,Roma,Italy.

Pistocchi,C.,Silvestri,N.,Rossetto,R.,Sabbatini,T.,Guidi,M.,Baneschi,I.,Bonari,E., Trevisan, D., 2012. A simple model to assess nitrogen and phosphorus contaminationin ungaugedsurfacedrainagenetworks:application tothe MassaciuccoliLakeCatchment,Italy.J.Environ.Qual.41,544–553.

Raich,J.W.,Tufekciogul,A.,2000.Vegetationandsoilrespiration:correlationsand controls.Biogeochemistry48,71–90.

Richardson,A.D.,Braswell,B.H.,Hollinger,D.Y.,Burman,P.,Davidson,E.A.,Evans,R. S.,Flanagan,L.B.,Munger,J.W.,Savage,K.,Urbanski,S.P.,Wofsy,S.C.,2006.

Comparingsimplerespirationmodelsforeddyfluxanddynamicchamberdata. Agric.For.Meteorol.141,219–234.

Rochette,P.,Flanagan,L.B.,1997.Quantifyingrhizosphererespirationinacorncrop underfieldconditions.SoilSci.Soc.Am.J.61,466–474.

Rowson,J.G.,Worrall,F.,Evans,M.G.,Dixon,S.D.,2013.Predictingsoilrespiration frompeatlands.Sci.TotalEnviron.442,397–404.

Schüßler,A.,Walker,C.,2010.TheGlomeromycota.Aspecieslistwithnewfamilies andnewgenera.TheRoyalBotanicGardenEdinburgh,TheRoyalBotanicGarden KewBotanischeStaatssammlungMunich.OregonStateUniversity.

Silgram, M., Shepherd,M.A.,1999.The effectsof cultivationonsoil nitrogen mineralization.Adv.Agron.65,267–311.

Silvestri,N.,Pistocchi,C.,Sabbatini,T.,Rossetto,R.,Bonari,E.,2012.Diachronic analysisoffarmers'strategieswithinaprotectedareaofcentralItaly.Ital.J. Agron.2,139–145.

Silvola,J.,Alm,J.,Ahlholm,U.,Nyhanen,H.,Martikainen,P.J.,1996.Thecontribution ofplantrootstoCO2fluxesfromorganicsoils.Biol.Fertil.Soils23,126–131.

Simon, L., Lalonde,M., Bruns,T.D.,1992. Specificamplification of18S fungal ribosomalgenesfromvesiculararbuscularendomycorrhizalfungicolonizing roots.Appl.Environ.Microbiol.58,291–295.

Smith, Read, D.J., 2008. Mycorrhizal Symbiosis, third ed. Academic Press, Amsterdam.

SoilSurveyStaff, 1975.Soiltaxonomy:abasicsystemofsoilclassificationformaking andinterpretingsoilsurveys.USDA-SCSAgric.Handb.436.U.S.Govt.Print. Office,Washington,DC.

Subke,J.-A.,Inglima,I.,FrancescaCotrufo,M.,2006.Trendsandmethodological impactsinsoilCO2effluxpartitioning:ametaanalyticalreview.GlobalChange

Biol.12,921–943.

Tamura,K.,Peterson,D.,Peterson,N.,Stecher,G.,NeiKumar,M.S.,2011.MEGA5: molecularevolutionarygeneticsanalysisusingmaximumlikelihood, evolu-tionary distance, and maximum parsimony methods.Mol. Biol. Evol. 28, 2731–2739.

terBraak,C.J.F.,Šmilauer,P.,2002.CANOCOReferenceManualandCanoDrawfor Wiindows User’s Guide: Software for Canonical Community Ordination (Version4.5).MicrocomputerPower,Ithaca.

Vandenkoornhuyse,P.,Husband,R.,Daniell,T.J.,Watson,I.J.,Duck,J.M.,Fitter,A.H., Young,J.P.W.,2002.Arbuscularmycorrhizalcommunitycompositionassociated withtwoplantspeciesinagrasslandecosystem.Mol.Ecol.11,1555–1564.

Verhoeven,J.T.,Setter,T.L.,2010.Agriculturaluseofwetlands:opportunitiesand limitations.Ann.Bot.London105,155–163.

Violante,P.,2000.MetodidiAnalisiChimicadelSuolo.FrancoAngeli,Milano.

Werth,M.,Kuzyakov,Y.,2009.Three-sourcepartitioningofCO2effluxfrommaize

fieldsoilby13Cnaturalabundance.J.PlantNutr.SoilSci.172,487–499.

Wirsel, S.G.R., 2004.Homogenousstandsof awetland grass harbourdiverse consortiaofarbuscularmycorrhizalfungi.FEMSMicrobiol.Ecol.48,129–138.

Wolfe,B.E.,Mummey,D.L.,Rillig,M.C.,Klironomos,J.N.,2007.Small-scalespatial heterogeneityofarbuscular mycorrhizalfungalabundanceandcommunity compositioninawetlandplantcommunity.Mycorrhiza17,175–183.

Wunderlich,S.,Borken,W.,2012.PartitioningofsoilCO2effluxinun-manipulated

andexperimentallyfloodedplotsofatemperatefen.Biogeosci.Discuss.9, 3477–3489.

Zeitz,J.,Velty,S.,2002.Soilpropertiesofdrainedandrewettedfensoils.J.Plant. Nutr.SoilSci.165,618–626.