Trade-off between photovoltaic systems installation and agricultural practices on arable lands: An environmental and socio-economic impact analysis for Italy

(1)

Contents lists available atScienceDirect

Land

Use

Policy

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

Trade-off

between

photovoltaic

systems

installation

and

agricultural

practices

on

arable

lands:

An

environmental

and

socio-economic

impact

analysis

for

Italy

S. Sacchelli

a,b,∗

_,

_G.

_Garegnani

b

_,

_F.

_Geri

c

_,

_G.

_Grilli

b,c

_,

_A.

_Paletto

d

_,

_P.

_Zambelli

b

_,

_M.

_Ciolli

c

_,

D. Vettorato

b

a_Department_of_{Agricultural,}_Food_and_Forest_Systems_Management,_University_of_Florence,_P.le_delle_Cascine_18,₅₀₁₄₄_Florence,_Italy b_European_Academy_of_Bolzano,_Via_G._Di_Vittorio_16,₃₉₁₀₀_Bolzano,_Italy

c_Department_of_Civil,_{Environmental}_and_Mechanical_Engineering,_University_of_Trento,_via_Mesiano_77,₃₈₁₂₃_Trento,_Italy

d_Council_for_Agricultural_Research_and_Economics_–_Forest_Monitoring_and_Planning_Research_Unit_(CRA-MPF),_P.za_Nicolini_6,_38123,_Villazzano,_Trento, Italy

a

r

t

i

c

l

e

i

n

f

o

Articlehistory:

Received13October2015

Receivedinrevisedform1April2016 Accepted24April2016 Keywords: Solarenergy Spatialanalysis Open-sourcemodel Sustainabilityconstraint Cropproduction Trade-offevaluation

a

b

s

t

r

a

c

t

Thepaperintroducesanddiscussesanopen-sourcespatial-basedmodel(calledr.green.solar)ableto quantifytheenergyproductionfromsolarphotovoltaic(PV)ground-mountedpanels.Socio-economic andenvironmentalimpactscanbeevaluatedbythemodel.Themodelstartsfromthetheoreticalquantity ofsolarPVpotentialenergyandestimatesareductionoftotalamountofenergybasedonlegal,technical, recommendedandeconomicconstraints.Modeloutputswereusedforatrade-offanalysisbetween energyproductionandtraditionalcropsforfood/feedcultivationonnotirrigatedarableland.Themodel wastestedatregionallevelforaMediterraneancontext(Italy).Theresultsconﬁrmthattheeconomic proﬁtabilityofPVsystemsfollowsanorth-southgradient,butthemainimpactsarerelatedtolocal peculiarities–suchasthedisposalofnotirrigatedarablelandandthepresenceofconstraints,inparticular thelandscapemaintenance,themorphologicalvariablesandthespecializationindex–andcropyields. ©2016ElsevierLtd.Allrightsreserved.

1. Introduction

Inordertocopewithnegativeeffectsofclimatechange, sev-eralpoliticalmeasuresandactionshavebeenappliedworldwide inrecent years.Normativerules havebeenparticularlyfocused onthereductionofcarbondioxideemissionsandsubstitutionof fossilfuelswithrenewableenergy(RE)sources.Inthissense,the EuropeanCommissionreleasedtheDirective2009/28/EConthe promotionoftheuseofenergyfromrenewablesources.This Direc-tive–alsoknownas20-20-20strategy –reportsonmandatory nationaltargetsandmeasuresfortheuseofenergyfromrenewable sources,highlightingatthesametimetheneedofnationalREaction plans.Despitetodateseveralenvironmentalandsocio-economic beneﬁtshavebeenrecognizedtoRE,intherecent scientiﬁc lit-eratureagrowinginterestisgiventotheevaluationofpotential negativeimpactsaswellasintegratedanalysis(seee.g.,Valodka

∗ Correspondingauthorat:DepartmentofAgricultural,FoodandForestSystems Management,UniversityofFlorence,P.ledelleCascine18,50144Florence,Italy.

E-mailaddress:sandro.sacchelli@uniﬁ.it(S.Sacchelli).

and Valodkien ˙e, 2015; Bilgili et al., 2016).Taking intoaccount

the Directive 2009/28/EC, sustainability criteria for RE produc-tionarestrictlydefinedonlyforbiofuelsandbioliquids.However, alsotheotherREsources(i.e.geothermal,hydropower,windand solar power) can affect a specific production and/or consump-tionareasinecological,socialandeconomicterms.Particularly, theseREsourcescanhavesignificantimpactsoncertain Ecosys-temServices(ESs).Tocopewithriskofnegativeimpacts,anumber of studiesand models havebeen carried out,payingparticular attentiontobiomass/biofuelsproduction(seee.g.Verkerk etal.,

2011;DominikandRainer,2014;UphamandSmith,2014),wind

power(Kouloumpisetal.,2013;Yuanetal.,2015), hydropower

(Daini,2000;Chenetal.,2015)andsolarenergy(Kaygusuz,2009;

WandererandHerle,2015).

Oneoftheﬁrststudiesfocusedonassessmentofthe poten-tial impacts of solar energy was developed by Neff (1981). In thatwork,theauthorpointedoutsomeimportantrelationships betweentheimplementationofphotovoltaic(PV)technologyand theconsequencesonpublicoccupationalsafetyandhealth.A par-ticularemphasiswasgiventotheindirecteffectsonlabormarket aswellastoenvironmentalconsequences.Inthissense,landuse, http://dx.doi.org/10.1016/j.landusepol.2016.04.024

(2)

Nomenclature

_Theo Conversion efﬁciency related to the Carnot efﬁ-ciencylimit(%)

SEN Totalsolarenergy(kWh/m2year−1pereachkWpof

installedpower)

SEN Totalsolarenergy(kWh/m2year−1pereachkWpof

installedpower)

THEN Theoreticalenergy(MWh/pixelyear−1)

nsres North-southresolutionofrastermap(m) ewres East-westresolutionofrastermap(m) LEEN Legalenergy(MWh/pixelyear−1)

AL Pixelclassiﬁedasnotirrigated arablelands(code 2111ofIVth_level_corine_land_cover)

LC Pixelclassiﬁedasareaswithlandscapeconstraint NA Pixelincludedinprotectedareas

TEEN Technicalenergy(MWh/pixelyear−1)

k ActualnetavailablesurfaceforPVplantsinstallation (%)

PVplantefﬁciency(%) sl Slope(%)

alt Altitude(masl) m Municipality

r Region

NIALm Municipalsurfaceofnotirrigatedarableland(ha)

NIALr Recommendedenergy(MWh/pixelyear−1)

FR Pixelclassifiedashighfloodrisk LR Pixelclassifiedashighlandsliderisk ER Pixelclassifiedashighearthquakerisk

REV RevenuesfromPVenergyselling(D/pixelyear−1) p MarketpriceofPVenergy(D/MWh)

inc Additional optional incentives for PV energy (D/MWh)

RPV RevenuespresentvalueforPVplants(D/pixel) CPV CostspresentvalueforPVplants(D/pixel) NPVPV NetpresentvalueforPVplants(D/pixel)

r Discountrate(%)

d Yearlydecayofperformanceofphotovoltaic mod-ules(%)

lc LifecycleforPVplants(years) P InstalledPVpower(MW/pixel)

u Unit cost for ﬁxed ground-mounted PV panels installation(D/MW)

iC Purchase and installation cost for PV plants

(D/pixel)

gC Cost for PV plants connection to electric grid

(D/pixel)

RAL Cost for rent of not irrigated arable land (D/ha

year−1)

rC Surfacerentcost(D/pixelyear−1)

mC MaintenancecostforPVplants(D/pixelyear−1)

cC CleaningcostforPVplants(D/pixelyear−1)

aC AdministrativeandconsultanciescostsforPVplants

(D/pixelyear−1)

sC InsurancecostforPVplants(D/pixelyear−1)

dC DecommissioningcostforPVplants(D/pixel)

x Speciﬁccrop

NR Netrevenuesforcrop(D/hayear−1) GAP Grossagriculturalproduction(D/hayear−1) C Costforcropproduction(D/hayear−1) GAP Grossagriculturalproduction(D/hayear−1) C Costforcropproduction(D/hayear−1) NPVX Netpresentvalueforcrops(D/hayear−1)

rot Rotationperiodforcrop(years)

thermalandclimaticeffectsandemissionswereidentiﬁedas rel-evantissuestobeevaluated.Abalanceinpositive andnegative impacts ofsolarPV energywasdeﬁnedin SwapnilDubeyet al.

(2013),byacategorizationofconsequencesindifferentclasses:(i)

landuseandlandscape,(ii)infrastructure,(iii)political,(iv)energy market,(v)industry,R&D,educationand(vi)public&marketing. Moreinsightsaboutlarge-scalePVplants weregiveninPhillips

(2013).TheauthordepictedhowthePVsystemscanbeconducive

toachievingahighlevelofsustainability,comparedtotraditional energysourcesforbothconstructionandoperationphases. Detri-mentaleffectcouldberevealedforfewwildlifespecies(i.e.forflight hazards).Neutralimpactsweredefinedforotherfeaturessuchas visualaesthetics,landoccupationorhabitatfragmentation.In addi-tion,unknowneffectswerehighlightedbytheauthor,inparticular relatedtosoilandwaterimpactaswellastolocalclimatic vari-ation(changeinsurfacealbedoandothersurfaceenergyflows). LifeCycleAssessment(LCA)approach–includingdisposal,and/or recyclingphaseofpanels–isanotherappliedmethodologyforPV impactappraisal(seee.g.FthenakisandChulKim,2009;Turconi

etal.,2013;Dubeyetal.,2013).Arecentapproachdealswiththe

analysisofPVimpactonESsfollowingtheclassiﬁcationproposed bytheMillenniumEcosystemAssessment(Hastiketal.,2015).

Aliteraturereviewaboutterritorialandlandscapeimpactsfor solarpowerplantswasimplementedbyChiabrandoetal.(2009), with a real application for ground-mountedPV. Among differ-entpotentialnegativeeffectstheauthorsintroducedanin-depth assessmentofglareriskduetopanels.ZanonandVerones(2013)

stressed the risk of PV conﬂicts on the use of fertile areas or theimpactoftechnicalequipmentonthelandscape.Public per-ception of PV systems was investigated by Tsantopoulos et al.

(2014)inGreecewithresultingenvironmentally-friendly,

sustain-able and socially acceptableopinions for this REfrom citizens.

Heras-Saizarbitoriaetal.(2011)investigatedthepublicacceptance

ofPVsolarenergyinSpainthroughtheroleplayedbythemedia. However,asshowninBrudermannetal.(2013),althoughsome decision makers–suchasfarmers–usuallyhave ratherstrong eco-attitudesandethicalconsiderationsaboutPVsystems imple-mentation,thesefactorsdonotseemtobegoodpredictorswith respecttotheadoptionofPVtechnology.

Anawkwardproblemconcerningground-mountedPVplants is oftendepictedinlandusecompetitionwithcropproduction. Some studiesshowedthe importanceof site characteristicsfor trade-offanalysis:forexample,soilfertilityortypeofagricultural land(arableland,marginallandetc.)wereconsideredwith differ-entdegreesofsuitabilityforPVenergyproduction/cropcultivation

(Nonhebel,2005;Sliz-Szkliniarz,2013;CalvertandMabee,2015).

APVenergyvs.foodtrade-offwasanalyzedinNonhebel(2005)

stressingtheyieldimportanceofdifferentlocations.The evalua-tionofground-basedPVapplicationsrelatedtolandqualitywere carried out in a GIS-basedmodel of Sliz-Szkliniarz (2013).In a studybyCalvertandMabee(2015)marketparameters–energy densityaswellaspotentialelectricityproduction–werechosen as key elementstoestablisha trade-offanalysis betweensolar energyandenergycropscultivationonmarginallandinOntario (Canada).Optimizationtechniquessuchastheagrivoltaicsystem, implementedbymeansofLandEquivalentRatios,wereappliedto combineinasameareaPVplantsandagricultureproductionin ordertomaximizetotalenergyefﬁciency(forbothsolarpaneland crops)(Duprazetal.,2011).

Asoutlinedintheliterature,aconsistentnumberofscientific worksconcerningpotentialconflictbetweenPVplantsand agri-culturalproductionwasdepicted.Nevertheless,theexaminationof theabovementionedstudiesdenotesthepresenceofafewflexible andupdatableDecisionSupportSystems(DSS)suitablefor analy-sisatdifferentscale,indifferentcontextsandwithdiverseinput datasetavailabletodecisionmakers.

(3)

Fig.1. Studyarea.

In this framework, the paper continues ourprevious works

(Garegnani et al., 2015a) whose objective was to implement

and test a new Geographic Information System (GIS) based model named r.green (http://www.recharge-green.eu/approach/

). This model is carried out with a modular and multistep procedure that enables the quantiﬁcation of energy from the-oretical to economic. Reduction of energy availability can be takeninto accountthrough theevaluation of potentialimpacts on ESs. To date the r.green.biomassfor (Zambelli et al., 2012;

Sacchelli et al., 2013; http://grass.osgeo.org/grass70/manuals/

addons/r.green.biomassfor.html) and r.green.hydro (Garegnani

et al., 2015b; http://grass.osgeo.org/grass70/manuals/addons/r.

green.hydro.html)sub-modelsareavailableasadd-onsfor

Quan-tumGISandGRASSGISsoftware;theseDSSsarefocusedonforest biomassforenergyproductionandhydropoweranalysis, respec-tively.Specifically,theaimofthisworkwastodevelopandapply ther.green.solarsub-modelfocusedonthequantificationof sustain-ableenergyfromfixedground-mountedphotovoltaic(PV)panels, inordertomaketheDSSfreelyavailableinadd-onrepositoryof QuantumGISandGRASSGISsoftwares.Ther.green.solaroutputs wereusedtodevelopatrade-offanalysisbetweentraditional agri-culturalproductionandimplementedPVplantsonarablelandby theintegrationofspatialanalysisandeconomicindexes.

2. Methodology

Dueto thefact that Mediterranean areais one of themost promisingforsolarenergyavailabilityinEurope,Italywaschosen ascasestudywithafocusataregionallevel(Fig.1).

Theworkwasdevelopedinthreephases.Intheﬁrstphasethe r.green.solarmodelwasimplementedasbashscriptsableto quan-tifyelectricsolarenergyavailabilityclassiﬁedin:

• Theoretical; • Legal; • Technical;

Fig.2.Generalframeworkofthework.

• Recommended; • Economic.

Inthesecondphasetheeconomicproﬁtabilityofagricultural foodandfeedproductiononarablelandsforeachregion,was ana-lyzed.Eventually,performanceofPVplantaswellastrade-offand potentialconﬂictamongPVplantsandtraditionalagricultural prac-ticeswereestimatedaccordingtofollowingindicators,asbetter explainedinSection2.2:

• NetPresentValueforPVplants;

• NetPresentValueforagriculturalproduction; • InternalRateofReturnforPVplants;

• SafetyMarginofsolarelectricenergyprice;

• PotentialcroplossesincaseofPVpanelsinstallationonarable lands.

ThegeneralframeworkoftheworkisreportedinFig.2. As a matter of fact the Directive 2009/28/EC was adopted byItalyin2010,witha NationalRenewableEnergyActionPlan (NREAP).AccordingtotheNREAP,themainsupportmechanism for electricity production from PV plants wasthe feed-intariff mechanism(ContoEnergia).Thismechanismprovidesa sequen-tialreductionupto2013.Since2013,PVincentivesstoppedand onlytaxdeductionaswellasfacilitationforself-consumptionhave beenmaintained.However,despitethefewpossibilitiesto imple-mentground-mountedPVpanelsonarablelandnowadays,agreat numberofPVplantshavebeenrealized.Inotherterms,developed impactanalysisrepresentsaspatialevaluationsuitableforboth implementedPVsystemsandpotentialfutureapplication. 2.1. Implementationofr.green.solarmodel

TheGIS-basedtoolcomputesamultistepprocedureto quan-tify solar PV energy, taking into account the legal, technical, recommended and economic constraints. The ﬁrst step was datasetintegration.Themodelautomaticallyimportsthevariables

(Table1)andtransformsthemintoarastermap(incaseofshapeﬁle

(4)

Table1 Inputdataset.

Variable Variabletypology

Solarradiation Raster

DigitalTerrainModel(DTM) Raster

CorineLandCovermap Raster

Landscapeconstraints Shapeﬁle

Naturalprotectedareas Shapeﬁle

Floodrisk Shapeﬁle

Seismicrisk Shapeﬁle

Landsliderisk Shapeﬁle

Arablelandspecializationindex Shapeﬁle

Roads Shapeﬁle

Regionsboundary Shapeﬁle

tospatial-independentcoefﬁcientsappliedinthecasestudy,were reportedinAppendixA.

Theoretical energy derives from Photovoltaic Geographical Information System (PVGIS, www.re.jrc.ec.europa.eu/pvgis), a spatial-basedassessmentofsolarelectricenergyresourcefromPV systemsinEurope,Africa,andSouth-WestAsia(ˇSúrietal.,2007; Huldetal.,2012).Thosedataareobtainedfromtheapplication ofr.sun moduleofGRASSGIStool (seee.g.Nguyenand Pearce, 2010)andrepresentlong-termyearlyaverages,basedonsatellite dataretrievalforglobalirradiationonanoptimally-inclinedsurface (period1998–2011).Theoreticalenergywascomputedby trans-formationoforiginaldatafromPVGISintoequivalentenergyper i-thpixel(expressedinMWh/year)takingintoaccountthe phys-icallaws.Inourcasewesetasquaredpixelresolution equalto 100×100m(pixelsurfaceof1ha).Infact,inItaly 30%offarms haveanagriculturalsurfacesmallerthan1ha(ISTAT,2010).Field sizehasaninﬂuenceontheaccuracyoftheappliedmethodology

(BrownandPervez,2014).Therefore,tobalanceoutputdetailand

computationaltimeofthemodelaswellastoobtainanexhaustive representationofthefactorsanalyzedwithareasonableamount ofdataprocessing(Romanoetal.,2015),apixelresolutionof1ha waschosen.

Mostsolarcellsonthemarketarebasedonsiliconwafersand theuppertheoreticalwasstudiedbyShockleyandQueisser(1961). Anoptimalcellwithabandgapof1.3eVislimitedbytransmission lossesofphotonsto31%(310Wpm−2).

Ifwedonotconsiderthecurrenttechnology,accordingto ther-modynamiclaws,theconversionefﬁciencyisrelatedtotheCarnot efﬁciencylimit,whichisnearly95%(Green,2002).Noticethatthe Carnotlimitisonlyatheoreticallimitandcannotbebuiltinpractice withtechnologycurrentlyavailable.Thislimitisthenthe upper-mostvaluethatitcanbetheoreticallyreached(Eq.(1)).

THEN i=Theo×SEN i×nsres×ewres/1000 (1)

Legalenergywasdepictedastheamountoftheoreticalenergy availableonexploitablesurfacesfromanormativepointofview. Accordingtothesepremises,suitableareasforPVplants implemen-tation(inourcasenotirrigatedarableland)werehighlightedfrom Corine LandCoverMap (EuropeanEnvironment Agency,2010); a limit of 10% of total available surface was applied based on ItalianLegislativeDecree28/2011.Spatialconstraintswerethen deﬁnedtodepictinappropriateareas.Avisualaestheticlimitwas appliedbytheintroductionofthenationallandscapeconstraints

map (SITAP, 2015)aswellasavoidingtheinsertion ofPV

pan-els in natural protected areas including national, regional and provincial parks, national and provincial reserves,natural pro-tectedareasoflocalinterestaswellasNatura2000networksites

(NationalCartographicPortal,2015).Itisworthmentioninghow

additionalconstraintscouldbedeﬁnedinregionallawsand regula-tions(e.g.constraintsforPVsystemsimplementationaredepicted forProtectedDesignationsofOriginandProtectedGeographical

Indicationsterritories).However,duetothelackofuniformdata atregionallevelforItalyandtoimplementaprecautionary eval-uationofPVenergyimpactduetocropsubstitution,regionaland sub-regionallegalconstraintswerenotintroduced.Legalenergy wasdeﬁnedasinEq.(2).

LEEN i=THEN i

∀

i∈(AL&¬LC&¬NA) (2)

TechnicalenergytakesintoaccountactualPVplantsavailable surface,morphologicalcharacteristicsandsolarcellefﬁciency.In particular, shadoweffectand spacefor maneuverwere consid-eredbydepictionofasuitablepercentageontotalsurface(Calvert

andMabee,2015;Karavelietal.,2015).Upperlimitsforterrain

slope andaltitudeweredefinedtoavoidimproperareasforPV plantimplementation(Bedinetal.,2011).Technicalmanualsand researchoutputswerefinallyevaluatedtostressplantsefficiency

(seee.gBedinetal.,2011;MillerandLumby,2012)(Eq.(3)).

TEEN i=LEEN i×k×

∀

i∈(sl≤20&alt≤800) (3)

Inadditiontothelegalandtechnicallimits,otherconstraints couldbeintroduced,toreducepotentialenvironmentaland socio-economicimpactsduetoPVplantsinstallation.Infact,themodel hasthepossibilitytoinsertlimitsthatcanbeincludedasoptional maps,suggestedbydecisionmakers.Duetonational characteris-tics,weintroducedahazardconstraintonarablelandspotentially subjecttoﬂood,landslideaswellasearthquakes.Theseareaswere consideredunsuitableterritoriesbecauseofpossiblereductionof technicalenergyanddamageofPVsystems(NationalCartographic

Portal,2015).Moreover,aSpecializationIndex(SI)relatedtoarable

landswasdeﬁnedasanindicatoroftheweightthatthislandcover reachesat municipalitylevel,withrespecttoa generalcontext

(Andinietal.,2013).Inotherwords,SIisaparticularvaluethat

isusefulfordetectingimportantlocaldistrictsforaparticular pro-ductionintheagricultural,industrialorservicessector,aswellas forterritorialcharacteristics.IftheSIhasavaluegreaterthanor equalto2itmeansthatintheexaminedareathereisahigh spe-cializationfortheconsideredparameter(Fagarazzietal.,2009).In ourcasestudytheSI_≥2indicatedahighimportanceof munici-palnotirrigatedarablelandswithrespecttoregionalcontextand, asconsequence,agriculturallandsthatshouldnotbeusedforPV energyproduction(Eq.(4)).

SIi=

NIALm/AASm

NIALr/AASr

∀

i∈(m&r) (4)

NIALandAASdataderivedfromVIth_ISTAT_Agricultural_National

Census(http://dati-censimentoagricoltura.istat.it/).

Thus, recommendedenergytakenintoaccountalltheabove mentionedconstraints(Eq.(5)):

REEN i=TEEN i

∀

i∈(¬FR&¬LR&¬ER&¬SI≥2) (5)

Thefinalsub-modelofr.green.solarcomputestheeconomic dis-posalofenergy.Inthefirststep,revenuesfromenergysellingcan bequantifiedas(Eq.(6)):

REV_i₌REEN i×(p+inc) (6)

Actualisedvalueofrevenuescanbecomputedas(Eq.(7)): RPVi=REVi×

(1+r+d)lc−1

(r+d)×(1+r+d)lc (7)

Implementationas wellasoperating and maintenance costs (O&M)ofPVplantsinclude(Bedinetal.,2011):(i)purchaseand installation,(ii)connectiontoelectricgrid,(iii)surfacerenting,(iv) maintenance,(v)cleaning,(vi)administrativeandconsultancies, (vii)insurance,(viii)decommissioningcosts.

Purchaseandinstallationcostsarebasedontheinstalledpower:

(5)

Table2

Costsforconnectiontotheelectricgrid.

Distance(m) Cost(D) D≤200 186 200<D≤700 279 700<D≤1200 836 D>1200 1950 Table3

Quantiﬁcationofsurfacerent,maintenance,cleaning,administrative,consultancies

andinsurancecosts.

Typeofannualcost Cost(D)

Surfacerent rC i=RAL×nsres×ewres∀i∈AL(9)

Maintenance mC=iC i×0.01(10) Cleaning cC i=if(iC i×0.001< 1000)then1000else(iC i×0.001)(11) Administrativeand consultancies aC i=if(iC i×0.001< 3000) then3000else(iC i×0.001)(12) Insurance sC i=if(iC i×0.0015< 2000)then2000else(iC i×0.0015)(13)

Costsfortheconnectiontothegridaredifferentiatedaccording tothedistance.Intheabsenceofanationaldatasetongeographic distributionofgrid,aﬁrstapproximationconsideredthedistance fromi-thpixeltoroads(NationalCartographicPortal,2015).That costsvaryaccordingtoTable2.

Surfacerentcosts foreach landuseare basedondatafrom NationalInstituteofAgriculturalEconomics(INEA,2014).Theother annualcosts–pointfrom(iv)to(vii)–werecomputedasa per-centageoftheinstallationcosts(Table3)(Bedinetal.,2011):

Attheendofitslifecycle,theplantmustbedismantled,andthe decommissioningcostsmustbetakenintoaccount(Eq.(14)).

dC i=iC i×115 (14)

Actualisedcostscanbeexpressedas(Eq.(15)): CPVi=iC i+gC i+(rC i+mC i+cC i+aC i+sC i) ×(1+r) lc −1 r×(1+r)lc + dC i (1+r)lc (15)

Eventually,theNetPresentValuecanbecomputed(Eq.(16)):

NPVPV i=RPVi−CPVi (16)

2.2. Trade-offanalysis

TheanalysisofcompetitionbetweenPVpanelsandcropsfor food/feedwasbasedontheselectionofsuitableplantationsforeach Italianregion.ThefocuswasondatafromINEA(2013)thattake intoaccounteconomicanalysisfortheproductionofcrops gen-erallycultivatedonnon-irrigatedarablelands(cerealsandgrain leguminous,industrialcrops,foragecrops1_)._For_each_production, theannualnetrevenueswerecomputedas(Eq.(17)):

NRx,r=GAPx,r−Cx,r (17)

Then,thenetpresentvalueforcropcultivationwascalculated basedona4yearscroprotationperiod,onatotalinvestmentlength equaltothePVpanelslifecycle.Inordertodevelopa

precaution-1_The_examined_crops_are:_oat,_chickpea,_spelt,_broad_bean,_durum_wheat,_wheat, buckwheat,lentil,whitelupin,millet,barley,gardenpea,rye,aromaticand ofﬁci-nalherbs,rape-seed,sunﬂower,lavender,alfalfa,Perennialrye-grass,Frenchgrass, Spanishesparcet,Egyptianclover,Crimsonclover,Whiteclover,Redclover, com-monvetch.

aryevaluationforPVdeployment,themoreconvenientcrop(from economicpointofview)waschosenforeachregion(Eq.(18)): NPVX r=MAX(NRx,r)×

(1+r)lc−1 r×(1+r)lc −

MAX(NRx,r)

(1+r)y (18)

wherey∈(rot·n,...,lc)withn=(1,2,...,lc/rot).

OnceNPVPVandNPVXwerecompared,twoeconomicindexes

forPVenergyproductionwerequantified:internalrateofreturn (IRR)andsafetymargin(SM).Thefirstgivesanideaofthe invest-ment’sprofitability.Ingeneral,theIRRcorrespondstothediscount ratethatmakestheNPVequalto0(Eq.(19)).Thelatterrepresents thepotentialdecreaseofcurrentenergypricethatmaintaina con-venienceinrenewableenergyplantsimplementationinrespectof cropscultivation(Eq.(20)):

IRRi=r|NPVPV i=0 (19)

SMi=p|NPVPV i,r=NPVX,r (20)

Theﬁnalevaluationsconsidered:(i)ananalysisbasedona per-centageofeconomicsurfacethatcanbehypotheticallyusedforPV energyproduction.Forthatareas,itwascomputedtheamountof potentialdeclineofcropsduetoPVplantsimplementation;(ii)a sensitivityanalysisbasedondiscountratevariationfor computa-tionofPVplants’economicefﬁciency.

3. Results

Table4showspotentialavailablesurfacesforPV

implementa-tionandenergyfromlegal,technical,recommendedandeconomic viewpoint. Asmatter of fact these variables assume a relevant importanceforterritorialplanning;theoreticalenergyis syntheti-callyreportedinFig.4.AhighpotentialforPVenergyproduction isrelatedtoSicily,EmiliaRomagna,Lombardy,Veneto,Apuliaand Sardinia.Thisdenotesthedisposalofalargeamountofnotirrigated arablelands(seelegalenergy).

Areductionofbothenergyandsurfacedisposalisevidentincase ofintroductionoftechnicalconstraints.Theinclusionoftechnical and recommendedconstraintsconsiderablyreduced theenergy potentialof someregionsin northernItaly. Specially, Trentino-South Tyroland Liguria highlighta reduction of recommended energyupto97.2% and79.3%in comparisonwithlegalenergy, respectively.Higherdecreaseofrecommendedenergyinsouthern regionsisfoundforthefollowingregions(Fig.3):Calabria(60.9%), Abruzzo(50.4%),Campania(48.3%),Sardinia(33.7%),Molise(24.6%) andBasilicata(24.5%).Intheseregionsthemajorlimitsarerelated torecommendedconstraints,inparticulartheearthquakeriskand theSI(Fig.4).Someofthecentralandnorthernregions–suchas EmiliaRomagna,Veneto, Lombardyand Piedmont–seemtobe favoritebythelowamountofsurfacewithmorphological (tech-nical) constraints,i.e. slope and altitude (see Figs. 3and 4).In thesecasesthereductionofenergyavailabilityfromlegalto rec-ommendedrangesfrom0.3%to1.2%.Inthoseregionsitdependson thelowweightofrecommendedconstraintsinrespecttothelegal andtechnicalones(inparticular,asexpressedbyFig.4,aconsistent overlapbetweenthefewareawithrecommendedconstraintsand legal/technicallimitsishighlighted).Recommendedandeconomic energiesshowthesameresults.Thisisduetothefactthatneither currentdiscountrateexceedsIRRnorpriceofenergyislowerthan safetymargin.Therefore,theeconomicproﬁtabilityofPVplantsis alwaysguaranteed.

NPVPV wascomputedastheaveragevalueofpixelwith

eco-nomicproﬁtabilityforPVplants;theanalysiswascarriedoutby meansofzonalstatisticoperationsforeachregion(Fig.5a).NPVX

derivesfromEq.(18)(Fig.5b).

Resultsdenoteanorth-south gradientofconvenienceforPV plants.TheaverageNPVPVrangesfrom169,798D/haof

(6)

Trentino-Table4

Energypotential(GWh/year×10−3₎_and_available_surface_(ha)_per_region.

Region Legal Technical Recommended Economic

Energy Surface Energy Surface Energy Surface Energy Surface

Piedmont 716.5 41,081 107.32 41,022 106.23 40,604 106.23 40,604 AostaValley 0.0 0 0.00 0 0.00 0 0.00 0 Lombardy 1382.1 80,724 206.70 80,455 206.69 80,451 206.69 80,451 Trentino-SouthTyrol 5.2 358 0.08 36 0.02 10 0.02 10 Veneto 1,227.6 71,915 183.98 71,850 183.14 71,525 183.14 71,525 Friuli-VeneziaGiulia 308.0 18,647 46.15 18,624 44.11 17,757 44.11 17,757 Liguria 8.9 536 1.15 459 0.27 111 0.27 111 EmiliaRomagna 1480.9 86,463 221.14 86,066 220.89 85,967 220.89 85,967 Tuscany 897.3 49,778 133.09 49,212 123.52 45,736 123.52 45,736 Umbria 404.7 22,570 58.27 21,636 53.66 19,938 53.66 19,938 Marche 647.8 37,674 91.07 35,279 85.53 33,140 85.53 33,140 Lazio 847.9 44,839 123.97 43,620 119.39 41,952 119.39 41,952 Abruzzo 146.0 8216 19.94 7471 10.94 4078 10.94 4078 Molise 211.4 11,735 28.44 10,508 24.16 8850 24.16 8850 Campania 377.5 20,509 53.72 19,412 30.14 10,611 30.14 10,611 Apulia 1203.7 63,685 179.31 63,228 155.66 54,739 155.66 54,739 Basilicata 555.8 30,057 75.06 26,954 63.76 22,701 63.76 22,701 Calabria 352.2 18,299 45.41 15,584 20.96 7158 20.96 7158 Sicily 1588.5 77,542 219.19 71,133 210.31 68,232 210.31 68,232 Sardinia 1023.7 51,418 150.55 50,382 100.79 34,079 100.79 34,079 Total 13,385.6 736,047 1945 712,929 1760 647,637 1760 647,637

Fig.3.ReductionofPVsurfacefromlegaltoeconomicparametersforeachItalianregion(ha).

SouthTyrolto287,282_D/haofSicilytakingintoaccounta20-years PVsystemslifecycleandadiscountrateof3%(Fig.5a).

AsimilartrendisdenotedforbothaverageIRRandSM(Fig.6). IRR varies from 31% (Trentino-South Tyrol) to 49% (Sicily).SM rangesfrom54D/MWhofLiguriato69D/MWhofSicily.Agreat proﬁtabilityofPVinvestmentsisdenotedbybothindexes.

ThisaspectwasalsoconﬁrmedbytheanalysisofFig.5aand b,inwhichthedifferencebetweenNPVPV andNPVX reachesan

orderofmagnitude(rangefromNPVPV,Umbria=10×NPVX,Umbriato

NPVPV,Sicily=48×NPVX,Sicily).

Inthisframework,itisinterestingtoevaluatethepotentialdrop incropproductionduetoPVplantsimplementation.Three scenar-ioswerecarriedoutassuming5%,10%and15%ofeconomicsurface useandrealdataconcerningthecropyield(INEA,2013).Results arereportedinTable5.

Basingonyieldofcropthat maximizeNPVX foreach region,

resultsshowhowpotentialagriculturallossesdo notfollowPV economicconvenience.Asmatterof fact,relevantdecreasingof cropproductionaredepictedforregionwithacombinationofhigh cropyieldaswellasavailabilityofnotirrigatedarablelands(i.e. EmiliaRomagna,Veneto,LombardyandPiedmont).

Aﬁnalremarkregardsthepotentialvariabilityoftechnicalas wellaseconomicparametersandtheirimpactonPVplants

prof-itability.Availabletechnologysuggestshowastrongincreasein plantsefficiencycannotbeforecastedatshort-mediumterm.On theotherhand,isdemonstratedthatoneofthemostsignificant variableforeconomicefficiencyisdiscountrate.Giventhispremise asensitivityanalysisforNPVPV computation,basedon

modiﬁca-tionofdiscountrate,wasdeveloped(Cozzietal.,2014).Resultsare expressesbyTable6.

Table6highlightstheimportanceofdiscountrateforNPVPV

quantiﬁcationaswellashowitsvariationcanbringtorelevant instability of economic performance. Alsoin this case a north-southgradientisrevealedstressingahigherworseningofPVplants economicperformanceinnorthernregions,incaseofaugmented discountrate.

4. Discussionandconclusions

ThedevelopedmodelpermitsanevaluationofPVenergy avail-ability, basedonmodular andmultistep analysis.Starting from the total solar energy disposal and the theoretical availability, different constraintscanbe includedtoreduce theharvestable quantitiesfromlegal,technical,recommendedandeconomicpoint of view. The ﬂexible approach allows the consideration of

(7)

dif-Fig.4.Theoreticalandeconomicenergy;constraintsappliedforcomputationoflegal,technicalandrecommendedenergy.

ferent inputdatasetand themodiﬁcationof theconstraints,as wellasofvariablesfacilitatingapplicationsincontextswith differ-entcharacteristicsandnormativeprescriptions.Theraster-based methodconsents a multiscaleanalysis throughvarious level of

pixel aggregation(e.g. atmunicipal, regional or nationallevel). Potentialenvironmentaland socio-economicimpactsdue toPV plantsimplementationcanbeconsideredandreducedbythe def-initionofrelatedconstraints.

(8)

Fig.5.(a)AverageNetPresentValueforPVplants–NPVPV(kD/ha);(b)AverageNetPresentValueforcropproduction–NPVX(kD/ha).

Fig.6. SafetymarginandInternalRateofReturnforPVplants. Table5

ExampleofpotentialcroplossesincaseofPVpanelsinstallationonarablelands.

Region Surface(ha) Cropyield(t/hayear−1₎ _Potential_crop_losses_(t/year)

PVsurface(5%) PVsurface(10%) PVsurface(15%)

Piedmont 40,604 5.93 12,033 24,065 36,098 AostaValley 0 0.00 0 0 0 Lombardy 80,451 5.60 22,524 45,047 67,571 Trentino-SouthTyrol 10 0.00 0 0 0 Veneto 71,525 9.87 35,307 70,614 105,921 Friuli-VeneziaGiulia 17,757 9.40 8343 16,686 25,029 Liguria 111 9.68 54 107 161 EmiliaRomagna 85,967 9.75 41,927 83,853 125,780 Tuscany 45,736 2.09 4772 9545 14,317 Umbria 19,938 0.86 862 1724 2586 Marche 33,140 4.53 7503 15,005 22,508 Lazio 41,952 0.97 2042 4083 6125 Abruzzo 4078 4.11 838 1675 2513 Molise 8850 1.94 859 1718 2576 Campania 10,611 11.16 5923 11,846 17,770 Apulia 54,739 0.89 2439 4878 7317 Basilicata 22,701 6.90 7836 15,672 23,508 Calabria 7158 3.85 1378 2756 4134 Sicily 68,232 1.84 6262 12,524 18,786 Sardinia 34,079 2.95 5029 10,058 15,086

(9)

Table6

Sensitivityanalysisbasedondiscountrate.

NPVPV(D/ha) reductionofNPVPV(%)

Region r:1% r:2% r:3% r:4% r:5% “r”from1%to2% “r”from2%to3% “r”from3%to4% “r”from4%to5% “r”from1%to5%

Piedmont 267,892 239,796 220,517 193,593 174,543 −11.7% −8.7% −13.9% −10.9% −53.5% Lombardy 259,780 232,419 213,783 187,421 168,867 −11.8% −8.7% −14.1% −11.0% −53.8% Trentino-SouthTyrol 206,751 184,177 169,798 147,024 131,693 −12.3% −8.5% −15.5% −11.6% −57.0% Veneto 258,248 231,022 212,637 186,245 167,781 −11.8% −8.6% −14.2% −11.0% −53.9% Friuli-VeneziaGiulia 245,031 218,996 201,605 176,171 158,510 −11.9% −8.6% −14.4% −11.1% −54.6% Liguria 239,182 213,687 196,734 171,747 154,449 −11.9% −8.6% −14.5% −11.2% −54.9% EmiliaRomagna 259,753 232,387 213,783 187,381 168,823 −11.8% −8.7% −14.1% −11.0% −53.9% Tuscany 282,383 252,968 232,552 204,603 184,666 −11.6% −8.8% −13.7% −10.8% −52.9% Umbria 280,736 251,469 231,262 203,348 183,511 −11.6% −8.7% −13.7% −10.8% −53.0% Marche 261,703 234,158 215,359 188,860 170,183 −11.8% −8.7% −14.0% −11.0% −53.8% Lazio 307,320 275,641 253,470 223,569 202,108 −11.5% −8.7% −13.4% −10.6% −52.1% Abruzzo 279,626 250,477 230,116 202,550 182,792 −11.6% −8.8% −13.6% −10.8% −53.0% Molise 287,242 257,369 236,850 208,255 188,009 −11.6% −8.7% −13.7% −10.8% −52.8% Campania 306,588 275,001 252,610 223,078 201,679 −11.5% −8.9% −13.2% −10.6% −52.0% Apulia 306,989 275,346 253,183 223,332 201,895 −11.5% −8.8% −13.4% −10.6% −52.1% Basilicata 300,845 269,746 248,169 218,623 197,552 −11.5% −8.7% −13.5% −10.7% −52.3% Calabria 321,445 288,487 265,218 234,317 211,995 −11.4% −8.8% −13.2% −10.5% −51.6% Sicily 348,132 312,768 287,282 254,657 230,714 −11.3% −8.9% −12.8% −10.4% −50.9% Sardinia 326,621 293,204 269,516 238,283 215,652 −11.4% −8.8% −13.1% −10.5% −51.5%

Inthiswork,ther.green.solarmodelwasappliedtodefineenergy potentialfromground-mountedPVsystem,hypotheticallyinserted onnotirrigatedarableland.Infact,oneoftheaimsoftheresearch wastodepictatrade-offbetweenPVenergyandcropforfood/feed production.Infutureanalyses,anextensiontootherlandusecould beeasilycarriedoutwiththemodificationofafewinputdata.A potentialapplicationofthemodeltootherEuropeanandglobal regionsisalsofacilitatedbymodularcompositionandawide avail-abilityofinputdata.In thiscase,a particularattentionastobe paidto differentlegal constraintsthat canbein force inother Countriesaswellastodifferentinputeconomicvariables. Recom-mendedconstraintscouldalsobemodifiedbasingonparticipative approachesandfocusgroupinvolvinglocalexpertand stakehold-ers.

Althoughanhigherdisposalofsolarenergyperunitofsurface isshowninsouthernregionsofItaly,totalamountofPVenergyis stronglyinﬂuencedbytwomainparameters:(i)theavailabilityof notirrigatedarablelandand(ii)thepresenceofconstraints,related tothelandscapemaintenance,morphologicalvariables(slopeand altitude),theearthquakeriskandthespecializationindex.These features,linkedtocropyield,leadtoagreaterpotentialimpact–in termsofcropssubstitution–innorthernregionofItalyrespectto centralandsouthernones,unlessanorth-southincreasing gradi-entisshownforeconomicproﬁtability.Infact,averageNetPresent Value,InternalRateofReturnandSafetyMarginonelectricenergy pricestress a strong conveniencefor PV plants investmentsin regionsuchasSicily,SardiniaandCalabria.Withthesepremises, themodelcouldrepresentausefulDSSforpolicymakersandlocal stakeholders,toquantifyand communicatestrengthsaswellas weaknessesofPVplantsinaspatial-basedmanner.

Nevertheless,theapplication ofr.green.solarfor a localscale planningmustconsideradditionalanalysisanddata.Forexample, furtherconstraintsshouldbeevaluatedincaseofgeographic pecu-liarities(e.g.thepresenceofProtectedDesignationsofOriginand ProtectedGeographicalIndicationsterritoriesaswellasareaswith ahighspecializationindexforcropscultivatedonarableland).An in-depthanalysisofregulationcouldbealsodevelopedinorder toconsiderprovincialandregionalvariabilityaswellastemporal dynamicsofrulesandincentives.Additionalapplicationscouldgo beyondtheBooleanstructureofconstraintsbytheintroductionof weightedvalue,e.g.intheformofMultiCriteriaAnalysis.

Trade-off analysis can be improvedtaking into account the geographicalsuitabilityforeach cultivationand rotationamong differentcrops.Thisaspectwasheresimpliﬁedbythecomparison

ofPVenergyproductionwiththemoreproﬁtableregionalcrops, tocarryoutaprecautionaryanalysis.

Eventually,additionalfutureinsightscouldfocusonthe imple-mentationofdifferentscenariosandsub-models.Theevaluationof theeconomicperformancefordifferenttechnologies(e.g.ﬁxedvs single/dualaxistrackersground-mountedPVsystems)orthe quan-tiﬁcationofimpactonESs(e.g.avoidedCO2 emissioninrespect to fossil fuel and the impact onecological corridors) could be developedtoimprovethestudyandbetterhighlighttheexisting trade-offs.

Acknowledgments

Thisstudywasconductedintheframeoftherecharge.green project “Balancing Alpine Energy and Nature” (http://www.

recharge-green.eu),whichiscarriedoutwithintheAlpineSpace

Programme,andisco-ﬁnancedbytheEuropeanRegional Develop-mentFund.AuthorswishtoacknowledgeRecharge.Greenproject partners,fortheircontributiontotheresearch.

AppendixA.

seeTableA1.

TableA1

Quantiﬁcationofparametersusedinthecasestudy(spatial-independentvariables).

SymbolDescription Value

Theo ConversionefﬁciencyrelatedtotheCarnotefﬁciencylimit(%) 95

nsres North-southresolutionofrastermap(m) 100

ewres East-westresolutionofrastermap(m) 100

k ActualnetavailablesurfaceforPVplantsinstallation(%) 20

PVplantefﬁciency(%) 75

p PriceofPVenergy(D/MWh) 105.80

inc AdditionaloptionalincentivesforPVenergy(D/MWh) 0

r Dscountrate(%) 3

d Yearlydecayofperformanceofphotovoltaicmodules(%) 1

lc IifecycleforPVplants(years) 20

P InstalledPVpower(MW/pixel) 0.2

u Unitcostforﬁxedground-mountedPVpanelsinstallation

(D/kW)

2500

(10)

References

ˇSúri,M.,Huld,T.A.,Dunlop,E.D.,Ossenbrink,H.A.,2007.Potentialofsolar electricitygenerationintheEuropeanUnionmemberstatesandcandidate countries.Sol.Energy81,1295–1305.

Andini,M.,deBlasio,G.,Duranton,G.,Strange,W.C.,2013.Marshallianlabour marketpooling:evidencefromItaly.Reg.Sci.UrbanEcon.43(6),1008–1022.

Bedin,D.,Holland,E.,Chies,A.,Annunziata,V.,Virdis,A.,2011.PVSinBLOOM:

BusinessGuide–GroundPhotovoltaicInvestmentsonMarginalAreas.

IntelligentEnergyEuropeProgramme(accessed12.06.15)http://www.

pvsinbloom.eu/.

Bilgili,F.,Koc¸ak,E.,Bulut,Ü.,2016.Thedynamicimpactofrenewableenergy consumptiononCO2emissions:arevisitedEnvironmentalKuznetsCurve approach.Renew.Sust.Energ.Rev.54,838–845.

Brown,J.F.,Pervez,M.S.,2014.Mergingremotesensingdataandnational agriculturalstatisticstomodelchangeinirrigatedagriculture.Agr.Syst.127, 28–40.

Brudermann,T.,Reinsberger,K.,Orthofer,A.,Kislinger,M.,Posch,A.,2013.

Photovoltaicsinagriculture:acasestudyondecisionmakingoffarmers. EnergyPolicy61,96–103.

Calvert,K.,Mabee,W.,2015.Moresolarfarmsormorebioenergycrops?Mapping andassessingpotentialland-useconﬂictsamongrenewableenergy technologiesineasternOntario,Canada.Appl.Geogr.56,209–221.

Chen,S.,Chen,B.,Fath,B.D.,2015.Assessingthecumulativeenvironmentalimpact ofhydropowerconstructiononriversystemsbasedonenergynetworkmodel. Renew.Sust.Energ.Rev.42,78–92.

Chiabrando,R.,Fabrizio,E.,Garnero,G.,2009.Theterritorialandlandscapeimpacts ofphotovoltaicsystems:deﬁnitionofimpactsandassessmentoftheglarerisk. Renew.Sust.Energ.Rev.13,2441–2451.

Cozzi,M.,DiNapoli,F.,Viccaro,M.,Fagarazzi,C.,Romano,S.,2014.Orderedweight averagingmulticriteriaprocedureandcost-effectivenessanalysisforshort rotationforestryintheBasilicataRegion,Italy.Int.J.GlobalEnergy37, 282–303.

Daini,P.,2000.Environmentalimpactassessmentforhydroelectricpowerplants inTrentino(Italy)1990–1997:similarityandclusteringofstudies,sitesand projects.IAPA18(1),43–60.

Dominik,R.,Rainer,J.(Eds.),2014.Socio-EconomicImpactsofBioenergy Production.Springer.

Dubey,S.,Jadhav,N.Y.,Zakirova,B.,2013.Socio-economicandenvironmental impactsofsiliconbasedphotovoltaic(PV)technologies.EnergyProcedia33, 322–334.

Dupraz,C.,Marrou,H.,Talbot,G.,Dufour,L.,Nogier,A.,Ferard,Y.,2011.Combining solarphotovoltaicpanelsandfoodcropsforoptimisinglanduse:towardsnew agrivoltaicschemes.Renew.Energ.36,2725–2732.

EuropeanEnvironmentAgency,2010.CorineLandCover2006RasterData (accessed12.06.15) http://www.eea.europa.eu/data-and-maps/data/corine-land-cover-2006-raster.

Fagarazzi,C.,Sacchelli,S.,Chiaramonti,D.,Prussi,M.,Recchia,L.,2009.

Socio-economicandenvironmentalaspectrelatedtodevelopmentof agro-energeticchain:thecaseofTuscanyregion.In:Proceedingsof17th EuropeanBiomassConferenceandExhibition,Hamburg,pp.330–340.

Fthenakis,V.,ChulKim,H.,2009.Landuseandelectricitygeneration:alife-cycle analysis.Renew.Sust.Energ.Rev.13,1465–1474.

Garegnani,G.,Geri,F.,Zambelli,P.,Grilli,G.,Sacchelli,S.,Paletto,A.,Curetti,G., Ciolli,M.,Vettorato,D.,2015a.AnewopensourceDSSforassessmentand planningofrenewableenergy:r.green.In:ProceedingsofFOSS4GEurope, Como2015,14–17June2015,pp.39–50.

Garegnani,G.,Zambelli,P.,Curetti,G.,Grilli,G.,Biscaini,S.,Sacchelli,S.,Geri,F., Ciolli,M.,Vettorato,D.,2015b.Adecisionsupportsystemforhydropower productionintheGessoeVermenagnavalleys.In:E-proceedingsofthe36th IAHRWorldCongress,TheHague,theNetherlands.

Green,M.A.,2002.Thirdgenerationphotovoltaics:solarcellsfor2020andbeyond. PhysicaE14(1–2),65–70.

Hastik,R.,Basso,S.,Geitner,C.,Haida,C.,Poljanec,A.,Portaccio,A.,Vrˇsˇcaj,B., Walzer,C.,2015.Renewableenergiesandecosystemserviceimpacts.Renew. Sust.Energ.Rev.48,608–623.

Heras-Saizarbitoria,I.,Cilleruelob,E.,Zamanillo,I.,2011.Publicacceptanceof renewablesandthemedia:ananalysisoftheSpanishPVsolarexperience. Renew.Sust.Energ.Rev.15(9),4685–4696.

Huld,T.,Müller,R.,Gambardella,A.,2012.Anewsolarradiationdatabasefor estimatingPVperformanceinEuropeandAfrica.Sol.Energy86,1803–1815.

INEA,2013.AnalisiDeiRisultatiEconomiciAziendali–RICA-INEA(accessed 12.06.15)http://www.rica.inea.it/public/it/area.php.

INEA,2014.BancaDatiDeiValoriFondiari–INEA(accessed12.06.15) www.inea.it:8080/mercato-fondiario/banca-dati.

ISTAT,2010.SixthGeneralCensusofAgriculture2010(accessed12.06.15)http:// dati-censimentoagricoltura.istat.it/Index.aspx.

Karaveli,A.B.,Soytas,U.,Akinoglu,B.G.,2015.ComparisonoflargescalesolarPV (photovoltaic)andnuclearpowerplantinvestmentsinanemergingmarket. Energy84,656–665.

Kaygusuz,K.,2009.Environmentalimpactsofthesolarenergysystems.Energy SourcesPartA31(15),1376–1386.

Kouloumpis,V.,Liu,X.,Lees,E.,2013.Environmentalimpactsofrenewableenergy: gonewiththewind?renewableenergygovernance.Lect.NotesEnergy23, 203–215.

Miller,A.,Lumby,B.,2012.UtilityScaleSolarPowerPlants—AGuideFor DevelopersandInvestors.InternationalFinanceCorporation(IFC),Global EnvironmentFacility(GEF),AustrianMinistryofFinance(accessed12.06.15) www.ifc.org/hb-solarpowerplants.

NationalCartographicPortal,2015.ProgettoNatura(accessed12.06.15)http:// www.pcn.minambiente.it/GN/.

Neff,T.L.,1981.Thesocialcostofsolarenergy.In:AStudyofPhotovoltaicEnergy Systems.Pergamon,NewYork.

Nguyen,H.T.,Pearce,J.M.,2010.Estimatingpotentialphotovoltaicyieldwithr.sun andtheopensourceGeographicalResourcesAnalysisSupportSystem.Sol. Energy84,831–843.

Nonhebel,S.,2005.Renewableenergyandfoodsupply:willtherebeenoughland? Renew.Sust.Energ.Rev.9,191–201.

Phillips,J.,2013.Determiningthesustainabilityoflarge-scalephotovoltaicsolar powerplants.Renew.Sust.Energ.Rev.27,435–444.

Romano,G.,DalSasso,P.,TrisorioLiuzzi,G.,Gentile,F.,2015.Multi-criteria decisionanalysisforlandsuitabilitymappinginaruralareaofSouthernItaly. LandUsePolicy48,131–143.

SITAP,2015.LandscapeConstraintMapMinistryofCulturalHeritageandActivities andTourism(accessed12.06.15)http://www.sitap.beniculturali.it/.

Sacchelli,S.,Zambelli,P.,Zatelli,P.,Ciolli,M.,2013.Biomasfor—anopen-source holisticmodelfortheassessmentofsustainableforestbioenergy.iForest6, 285–293.

Shockley,W.,Queisser,H.J.,1961.Detailedbalancelimitofefﬁciencyofpn junctionsolarcells.J.Appl.Phys.32(3),510–519.

Sliz-Szkliniarz,V.B.,2013.Assessmentoftherenewableenergy-mixandlanduse trade-offataregionallevel:acasestudyfortheKujawsko–Pomorskie.Land UsePolicy35,257–270.

Tsantopoulos,G.,Arabatzis,G.,Tampakis,S.,2014.Publicattitudestowards photovoltaicdevelopments:casestudyfromGreece.Energ.Policy71,94–106.

Turconi,R.,Boldrin,A.,Astrup,T.,2013.Lifecycleassessment(LCA)ofelectricity generationtechnologies:overview,comparabilityandlimitations.Renew. Sust.Energ.Rev.28,555–565.

Upham,P.,Smith,B.,2014.UsingtheRapidImpactAssessmentMatrixto synthesizebiofuelandbioenergyimpactassessmentresults:theexampleof mediumscalebioenergyheatoptions.J.Clean.Prod.65,261–269.

Valodka,I.,Valodkien ˙e,G.,2015.Theimpactofrenewableenergyontheeconomy ofLithuania.Procedia213,123–128.

Verkerk,P.J.,Anttila,P.,Eggers,J.,Lindner,M.,Asikainen,A.,2011.Therealisable potentialsupplyofwoodbiomassfromforestsintheEuropeanUnion.Forest Ecol.Manag.261,2007–2015.

Wanderer,T.,Herle,S.,2015.Creatingaspatialmulti-criteriadecisionsupport systemforenergyrelatedintegratedenvironmentalimpactassessment. Environ.ImpactAssess.52,2–8.

Yuan,X.,Zuo,J.,Huisingh,D.,2015.Socialacceptanceofwindpower:acasestudy ofShandongProvince,China.J.Clean.Prod.92,168–178.

Zambelli,P.,Lora,C.,Spinelli,R.,Tattoni,C.,Vitti,A.,Zatelli,P.,Ciolli,M.,2012.AGIS decisionsupportsystemforregionalforestmanagementtoassessbiomass availabilityforrenewableenergyproduction.Environ.Modell.Softw.38, 203–213.

Zanon,B.,Verones,S.,2013.Climatechange,urbanenergyandplanningpractices: italianexperiencesofinnovationinlandmanagementtools.LandUsePolicy 32,343–355.