Svensk Kärnbränslehantering ABSwedish Nuclear Fueland Waste Management CoBox 5864SE-102 40 Stockholm Sweden Tel 08-459 84 00 +46 8 459 84 00Fax 08-661 57 19 +46 8 661 57 19

R-06-121

Modelling carbon and water flows in terrestrial ecosystems in the boreal zone – examples from Oskarshamn

Louise Karlberg, Stockholm Environment Institute (SEI)

David Gustafsson, Per-Erik Jansson

Royal Institute of Technology KTH, Department of Land

and Water Resources Engineering

December 2006

ISSN 1402-3091

SKB Rapport R-06-121

Keywords: Mean residence time, Carbon sequestration, Forest ecosystems, Grasslands, Multiple criteria of acceptance.

This report concerns a study which was conducted for SKB. The conclusions and viewpoints presented in the report are those of the authors and do not necessarily coincide with those of the client.

A pdf version of this document can be downloaded from www.skb.se

Modelling carbon and water flows in terrestrial ecosystems in the boreal zone – examples from Oskarshamn

Louise Karlberg, Stockholm Environment Institute (SEI)

David Gustafsson, Per-Erik Jansson

Royal Institute of Technology KTH, Department of Land

and Water Resources Engineering

December 2006

�

Contents

1 Introduction 5

2 Modeldescription 7

3 Carbonturnoverinfourhypotheticalterrestrialecosystems 9�.1 Ecosystemdescription 9�.2 Parameterisationandmodelapplication 9�.� Results 10�.4 Discussion 1�

4 CarbonandwaterbudgetsatfoursiteslocatedwithintheOskarshamnstudy-area 15

4.1 Ecosystemdescription 154.2 Parameterisation 154.� Results 164.4 Discussion 18

5 Conclusion 21

Acknowledgement 2�

References 25

Appendix 29

5

1 Introduction

Carbonturnovertimescalescanbeusedasanapproximationoftheaccumulationoftraceelements,suchasradionuclidesoriginatingfromunderground,nuclearwasterepositories.Thus,dependingoncarbonandwaterturnoverrates,differentecosystemsmightbemoreorlesssuitableforthelocationofsuchrepositories.Althoughbothcarbonbudgetsandcarbonturnovertimeshavebeenestablishedforseveralecosystemsandatseverallocations,ithasbeenshownthattheoutcomeishighlydependentonclimateandthecorrectestimationofsoilrespiration;bothfactorsofwhicharehighlysite-specific/Medlynetal.2005/.Itisthereforeimportanttomakeananalysisofwaterandcarbonbudgetsandfluxesforeachtentativelocationforadeeprepository.Byusingmodellingtoolsitispossibletoincorporatelargeamountsofempiricaldataintheestimationsofwaterandcarbonturnover.Furthermore,byconductingatypeofsensitivityanalysisinthemodelsimulations,anapproximationofpotentialvariabilityintheestimatescanbecalculated.

TheSwedishNuclearFuelandWasteManagementCo(SKB)arecurrentlyinvestigatingtwosites(ForsmarkandOskarshamn/SKB2005ab/)aspossiblelocationsforadeeprepositoryofradioactivewaste.Thisstudyfocusesoncarbonandwaterflowsinterrestrialecosystemscommonlyoccurringatbothsites.Insectionthreeofthereport,wepresentmodelparameterisa-tionsforfourhypotheticalterrestrial,borealecosystems.Carbonturnoverinthesesystemsaresimulatedusingamethodbasedonmultiple-criteriaofacceptance.Secondly,insectionfour,weapplytheseparameterisations,withsomemodificationsforsite-specificempiricaldata,onfourtentativesitesforundergroundnuclearwasterepositoriesatOskarshamn.Site-specificdataonsoilrespirationisusedasacomparisontosimulatedvalues.

Sectiontwoandthreehave,withminormodifications,beenpublishedinAmbio/Karlbergetal.2006a/.

7

2 Modeldescription

TheCoupModelisaphysically-based,ecosystemsmodellingpackage/JanssonandMoon2001/thatcanbeusedtodesignaconceptualmodelforaspecificecosystem/JanssonandKarlberg2004/.Themodeldescribestheinteractionbetweenbiogeochemicalandhydrologicalprocessesinasoil-plant-atmospheresystem.Fluxesofwater,heat,andmatterarecalculatedforalayeredsoilprofileandoneorseveralvegetationlayersabovewithtimeseriesofmeteorologicaldataasthedrivingforce.

Theabioticpartofthemodelisbasedontwocoupledpartialdifferentialequationsforthewaterandheatflowsinthesoil:theRichard’sequation(water)andtheFourierlawofdiffusion(heat),respectively/JanssonandHalldin1979/.Surfaceboundaryconditions,suchasevapotranspira-tion,soilsurfacetemperature,andsnowmeltareestimatedfromenergybalancecalculationswherenetradiationisbalancedbyturbulentfluxesofsensibleandlatentheat,andsurfaceheatflow/AlvenäsandJansson1997,Gustafssonetal.2004/.Wateruptakefromthesoilisbasedonasoil-plant-atmosphere-continuumapproach,consideringthefluxofwaterfromthesoilthroughtheplantasaresponsetothedemandofwaterfromtheatmosphere,i.ePenman-Monteithequation/Penman195�,Monteith1965,Lindroth1985/.Snowaccumulationandmeltaredescribed,aswellasthepartitioningbetweeninfiltrationtothesoilorsurfacerunoffattheuppermostsoilboundary.

Thebioticpartofthemodelsimulatesplantgrowth,aswellascarbonandnitrogenturnoverinthesoil/Johnssonetal.1987,Eckerstenetal.1998/.Biomassispartitionedintoseveralabovegroundandbelowgroundpoolsofcarbonandnitrogen.Grossproductionofcarbon(GPP),drivenbysolarradiation/Monteith1977/andregulatedbyleafnitrogencontent,wateruptake,andairtemperature,isallocatedtodifferentcompartmentsoftheplant;leaves,stem,coarseroots,andfineroots,accordingtopre-specifiedpatterns.Eachcompartmentisassumedtohaveapotentialcarbontonitrogenratio,whichsubsequentlygivesrisetoanitrogendemand.Plantrespirationispartitionedongrowthandmaintenancerespirationfromallplantcompartments/Karlbergetal.2006b/.Dailylitterfalliscalculatedasfractionsofabove-groundandbelow-groundpartsoftheplantenteringthesoilorganicpools.Twopoolsofdifferentturnoverratewereusedtorepresentthesoilorganicmaterial,calledlitterandhumus.Themostimportantinputstothebioticpartarethuscharacteristicsgoverningtheplantlife-cyclesuchasallocationpatterns,plantassimilationandrespiration,nutrientuptakebyplants,externalnitrogeninputstothesoil,andfinallydecompositionandredistributionofdifferentdecompositionproductsinthesoilprofile.

ThemostcentralcomponentforinteractionbetweenthebioticandtheabioticpartsoftheCoupModelistheleafareaindex(LAI),whichgovernsbothinterceptionofradiationandofprecipitation.Boththelossesofwaterandcarbonfromthesoilbyeithertranspirationorrespiration,arestronglyrelatedtotemperatureandmoistureofthesoil.

9

3 Carbonturnoverinfourhypotheticalterrestrialecosystems

3.1 EcosystemdescriptionFourterrestrialecosystemswereselectedtorepresentthehypotheticalsystemsincludedinthestudy(Table�-1).Thesesystemswereselectedbothbecausetheyarelikelytodifferintermsofcarbonturnovertimes,andalsobecausetogethertheyarecommonlyoccurringinsouthernandcentralSweden.Thefirstecosystem,asemi-naturalgrassland,ischaracterisedbythelackofatreelayer,andafieldlayerconsistingofamixtureofgrassesandherbsgrowingonclay.Aforestdominatedbyalder(Alnus glutinosa)withahighgroundwatertablewaschosentorepresentthesecondecosystem.Thisdeciduoustreehassymbioticnitrogen-fixatingbacteriainitsrootnodules.Duetotheamplesupplyofnitrogen,itonlyretainsasmallfractionofitsnutrientsbeforesheddingitsleavesintheautumn.Thefieldlayerwascharacterisedbynitrophilic,lushgrassesandherbsgrowingonawetorganogenicsoiltype.Apineforest(Pinus sylvestris),growingonathinlayeroftillwithafieldlayerdominatedbycowberry,waschosenasthethirdecosysteminthestudy.Anotherconiferousforestontillwasselectedtorepresentthefourthecosystem;inthiscaseNorwayspruce(Picea abies).Inthisforest,thefieldlayerconsistedmainlyofblueberryandsomebroad-leavedgrasses.Lastly,amanagedforestsimilartotheNorwayspruceecosystemincompositionwasalsoincludedinthestudyasacomparisontothenaturalecosystem.ManagementwasassumedtofollowthegeneralpracticeforsouthernSwedenasrecommendedbytheSwedishForestAgency/SwedishForestAgency2005/.Thus,themanagedspruceforestwasclearedafter15years,resultinginaremovalof60%ofthetreebiomass,whichwasleftintheforestaslitter.Inaddition,theforestwasthinnedafter40andafter80years,affecting25%ofthetreebiomass.Whereasleaves,coarserootsandfinerootsremainatthesitetoformlitter,80%ofthestembiomass(ofthetreesaffectedbythinning)isremoved.Thefieldlayerwasassumedtobeunaffectedbytheseoperations.

3.2 ParameterisationandmodelapplicationAnumberofparametervaluescharacterisingtheecosystemswerederivedfromtheliterature;eithervaluesfromfieldmeasurementsorfrommodellingstudies(TableA1).Theseparametersdescribed,forexample,carbonallocationintheplant,maintenanceandgrowthrespiration,nitrogendemandandplantlitterfall,andwerecalledprimaryparameters.Someofthosewereplantorsoilspecific,whereastherestwereofamoregeneralnatureandwerethereforeassumedtobethesameforallsystems.Theparameterisationofthefieldlayerinthealderforestwasusedtorepresentthegrasslandvegetation.Foreachecosystemtherewerealsoanumberofparametersthatcouldnoteasilybedeterminedfromtheliterature,socalledsecondary

Table3-1. Descriptionofthemaincharacteristicsoftheecosystemsincludedinthestudy.

Name Treelayer Fieldlayer Soiltype Management

Grassland none Grass and herbs Clay NoneAlder Alder (Alnus glutinosa) Grass and herbs Peat NonePine Scots pine (Pinus sylvestris) Cowberry (Vaccineum vitis-idea) Till (thin)/

bedrockNone

Spruce Norway spruce (Picea abies) Blueberry (Vaccineum myrtillis) Till NoneSpruce,managed

Norway spruce (Picea abies) Blueberry (Vaccineum myrtillis) Till Clearing, thinning and harvest

10

parameters(TableA2).Instead,arangeofvalueswasspecifiedforeachofthem,assumingthatthe“true”valuewouldliesomewherewithinthatrange.Randomnumbersweregeneratedforeachsecondaryparameterwithintherespectiverangeofvalues.

Onehundredsimulationswererunforeachecosystemwiththeprimaryparametersandusingthepre-generatedrandomnumbersassecondaryparameters.Simulationswerebasedonaone-yeardata-setcontaininghourlyclimaticdatafromameteorologicalstationatthenorthernpartoftheÖlandisland(57°N’22’’1.�,17°E’5’’4�.4)1981/Larsson-McCannetal.2002/,scaledtoberepresentativefortheSKBstudyareainÄspö,Oskarhamn.Thisdatawasrecycledandusedasdrivingdatafor100-yearsimulations.Thereasonforchoosingsuchalongsimulationperiodwastoinsurethatsoilandplantcarbonremainedstableinthenaturalecosystems.Anotherreasonwasthat100yearsistheapproximaterotationperiodforaspruceproductionforestinsouthernSweden.Consequently,theinitialcarboncontentsintheplantandsoilwasparameterisedaccordingtoaveragecarbonplantandsoillevelsinnaturalecosystems,exceptforinthemanagedforestecosystem,wheretheinitialplantcarboncontentwaschosentorepresentanewlyplantedtree.Simulationswerecomparedtopre-specifiedcriteriaofaccept-anceincludingbothsitespecificandgenericdata(TableA�–A5).AnerrorfunctionadoptedfromBarrett/Barrett2002/wasusedtofindthebestparameterssetswithrespecttoallcriteriaofacceptance:

( ) ( )∑=

+−−=M

iiiii poooswE

1minmax

2

wheresiisasimulatedvalue,oi,omax,andomin,arethepre-specifiedoptimum,maximum,andminimumvaluesallowedforaspecificcriteria,respectively,piisapenaltyfactorthatis1ifsiisoutsidethemaximumandminimumrange(otherwise0),andwiisaweightingfactortakingvaluesbetween0and1,whichwasusedtoweighttheimportanceofdifferentcriteria.Thetensimulationswiththelowesterrorfunctionwereusedintheanalysistocalculateaverageandstandarddeviationofthesecondaryparameters,andofthesimulatedcarbonbudgets(TableA2).

3.3 ResultsCarbonstoragevariedbyafactortwobetweenecosystems(Figure�-1).Thealderecosystemhadthehighesttotalcarboncontent(25,000gCm–2)ofallsystems,closelyfollowedbythenaturalspruceforest(18,000gCm–2).Grassland,pineandmanagedspruceallhadaboutthesametotalcarbonstorage(around1�,000gCm–2).Lookinginsteadatthedistributionofcarbonwithinthesystems,similaritiesexistbetweenthegrasslandandthealdersystems,wherethemajorityofthecarboninthesystemwaslocatedinthesoilbiomass(97%and7�%respectively).Onthecontrary,intheecosystemsdominatedbyconiferoustrees,theplantbiomassconsistedofabouthalfofthetotalcarbonstorage.Acomparisonbetweenthenaturalandthemanagedspruceecosystemsshowsthatthesoilwasrathersimilarinthetwosystems,whereastheplantbiomasswaslowerintheproductionsystem.Thevariabilityintheestimationsofcarbonstoragecanbeassessedbystudyingtheratiobetweenthestandarddeviationandthemean,fordifferentpartsofthesystems.Forthesoilinallsystemsthisfigurewasratherlow(7–8%),whilethetreebiomassestimationshadalargervariability(12–26%forallsystems)(Figure�-1).Thelargestvariabilitywasfoundintheestimatesofthefieldlayerbiomass(24–75%forallsystems).

Totalfluxesofcarboninandoutoftheecosystemsalsovariedbyafactortwobetweeneco-systems(Figure�-1).Thealderandthenaturalspruceforesthadthehighestcarbonfluxes(around1,600gCm–2yr–1),followedbypine(1,000gCm–2yr–1)andlastlythegrassland(600gCm–2yr–1).Sincetheproductionsystemisnotinsteady-state,theinfluxofcarbonexceededtheeffluxbyabout100gCm–2yr–1.Theuncertaintiesintheestimationsofthefluxesweregenerallygreatercomparedwiththestorages,butshowedasimilarpattern.Thus,theratio

11

Figure 3-1. Carbon budgets for the hypothetical ecosystems. a) grassland b) alder c) pine d) spruce and e) spruce managed forest. Carbon storages (gC m–2) in bold and carbon fluxes (gC m–2 yr–2) in italics, including standard deviations.

E.

D.

B.

370 ± 90

300 ± 60

600 ± 130

11900 ± 990

0.3 ± 0.05

280 ± 60

320 ± 90

290 ± 170

510 ± 140

210 ± 90

920 ± 700

630 ± 240

18300 ± 1200520 ± 120

9.3 ± 2.5

440 ± 140

6400 ± 740

610 ± 540

390 ± 230

560 ± 350

400 ± 310

40 ± 20

230 ± 70

180 ± 90

280 ± 150

5600 ± 1200

7900 ± 530

440 ± 320

0.3 ± 0.02

720 ± 250

430 ± 140

300 ± 50

770 ± 260

650 ± 150

900 ± 220

250 ± 60

9400 ± 1500

8600 ± 620

50 ± 20

0.1 ± 0.04

280 ± 50

80 ± 60

120 ± 130

590 ± 140940 ± 210

250 ± 60

5300 ± 1400

8600 ± 620

30 ± 20

0.1 ± 0.04

100 ± 110

Carbon storage (gC m-2)

Carbon efflux (gC m-2 yr-1)

Carbon influx (gC m-2 yr-1)

Carbon internal system flux (gC m-2 yr-1)

A.

C.

12

betweenthestandarddeviationandthemeanforsoilheterotrophicrespirationvariedbetween17–�0%,whilethecorrespondingfigureswere22–59%and22–110%forthetreeandthefieldlayer(excludinggrassland)respectively(Figure�-1).Thegrasslandandthenaturalspruceforestgenerallyhadlowervariabilityintheestimationoffluxescomparedtotherest.

Meanresidencetime(MRT),thatistheaveragetimeanassimilatedcarbonmoleculestaysinacertainpartofsystembeforebeingdischargedorrespired,wascalculatedfortheaboveandbelowgroundpartsofthetreesandthefieldlayers,aswellasforthelitterandhumusfractionsofthesoil(Figure�-2).ThelowestMRTwasfoundtobeinthefieldlayer(approximatelyoneyear).Incomparison,treeshadanMRToftenyears,whichthusissignificantlyhigher.However,iftheturnovertimeintreesiscalculatedforaboveandbelowgroundpartsseparately,thepictureisaltered(Figure�-2).AbovegroundMRTfortreeswas25yearsonaverage,whilethecorrespondingfigureforthebelowgroundtreecomponentswaslessthantwoyears.TherewasalsoalargedifferencebetweentheMRTforlitterandhumus(Figure�-2).WhiletheformerhadanMRToflessthantwoyears,theMRTinhumuswasfoundtobeapproximately150years.Onaverage,totalMRTfortheentireecosystemwasaround15years.

VariabilityintheestimationsofMRTwasgreatestforthefieldlayer,inparticularinthepine,alderandproductionecosystems,whileitwaslowestforthesoilcomponents.DespitethisvariationintheMRTestimations,somedifferencesbetweentheecosystemscouldbeidentified(Figure�-2).Forinstance,theMRTofthebelowgroundtreebiomasswasmorethantwotimeshigherintheconiferousecosystems(1.7years)comparedtothealder(0.7years).Furthermore,theMRTofthefieldlayervegetationwasabout2.5timeshigherforthemanagedspruceforestcomparedtothenaturalspruceecosystem,bothabovegroundandbelowground.MRTofthefieldlayervegetationalsodifferedbetweenthenaturalpineandspruceecosystems,wheretheformerhadbothhighermeanvaluesandstandardvariations.Finally,MRTforthehumuspoolsvariedbetween1�5yearsforthenaturalspruceforestto165yearsforthealderforest,whereastheMRTforthelitterpoolvariedverylittlebetweentheecosystems(rangingfrom1.8yearsto2.�years).

Toassesstheuncertaintiesinthesimulatedcarbonbudgetsanddifferencesbetweentheeco-systems,itisimportanttorecognisethenumberandtypesofcriteriamet.Whenaggregatedintogroups,thepercentageofcriteriafulfilledinthefinalsimulationsshowedasimilarpatternwithintheecosystem,withtwoexceptions:alderandgrassland(Table�-2).Thealdersystemhadalowernumberoffulfilledcriteria,especiallywithregardstothetreelayerandlitterfall.

Figure 3-2. Mean residence time of carbon in different parts of the ecosystems; tree layer and field layer are separated on above ground and below ground components, and the soil organic carbon is separated on the fast “litter” pool and the slow “humus” pool according to the simulations. AG = above ground, BG = below ground.

0.1

1

10

100

1000

Grassland

MR

T (y

ears

)

Tree AG Tree BG Field AG Field BG Litter Humus

ProductionSprucePineAlder

1�

Incontrast,thegrasslandsimulationsperformedbetteroverallthantheaverage.Generally,criteriarelatedtocarbonstoragesinthetreelayerandinthesoil,aswellasnitrogenprocesses,werefulfilledtoahigherextentcomparedtocriteriarelatedtorespiration,litterfall,andleafareaindex,whichhadapoorerover-allresultinallecosystems.

3.4 DiscussionDuetothewetconditionsinthealderecosystem,decompositionofcarbonisslow,resultinginlargecarbonstorageinthesoil.Thehightotalcarboncontentinthenaturalspruceecosystemcouldinsteadbeattributedtothehighcarboncontentintrees.Fluxesofcarbontoandfromtheecosystemswerealsogreatestforthespruceandalderecosystems.Inthealderecosystem,thehighproductivityisduetotheabilityofthetreetofixatenitrogen.Becauseofnitrogen-richlitterfallfromthetree,nitrogenlevelsinthesoilarehigh,whichisreflectedinlowC/Nratios.Thefieldlayerbenefitsfromthesehighnitrogenlevels,causingahighproductivityinthefieldlayeraswell.Onthecontrary,thetreesandthefieldlayercompetestronglyfornitrogenintheconiferousecosystems.Therefore,inthespruceecosystemitisinsteadhighleafareaindicesthatgenerateahighproductivity.Meancarboncontentinthetreeoveritsonehundredyearlifespaninthemanagedforestislowcomparedwithtreesinthenaturalspruceforest.Thisisduetothefactthatthemeanvalueforthemanagedforestalsoincludesperiodfromwhenthetreewasplanteduntilitreachedmaturity.Furthermore,thefieldlayerintheproductionforestislowercomparedwiththenaturalforest.Mostlikely,thisisaneffectofahightreeleafareaindexintheproductionsystem,resultinginlightstressforthefieldlayer.

Becausethesizeofthestoragepoolsinrelationtothefluxeswereofthesameorderofmagnitudeinallsystems,theMRTsconsequentlyturnedouttobeverysimilar.AlthoughtheMRTforthewholeecosystemwasaround15years,MRTvarieddrasticallydependingontherouteofcarbonthroughtheecosystem.Notsurprisingly,thehighestturnoverrateswerefoundinthefieldlayer.Thefieldlayerturnoverrateswereslightlyhigherinthenaturalspruceforestcomparedtothemanagedspruceandthenaturalpine,mainlyasaresultoftherelativelylowfieldlayerproductionintheselatterecosystems.Generally,abovegroundMRTwashigherthanbelowgroundMRTfortreesinallecosystems,duetoaslowturnoverofcarboninthestems.AnotherinterestingaspectonturnoverinthedifferenttreeswasthelowerMRTofthebelowgroundbiomassinalder,whichprobablywasareflectionofthehigherrootactivityinthenitrogenfixatingsystem.Carbonthatdoesnotleavetheplantsthroughrespirationcontinuesthroughthesystemtothesoilwhereitformslitter.Litterisquicklydecomposedresultinginanotherrespiratorylossofcarbonfromthesystem.Theremainingmaterialformshumus,whichhasanestimatedturnovertimeofmorethanonehundredyears.ThevariationofsimulatedMRTbetweenecosystemsforthehumuspoolscanbeexplainedbydifferentsoiltemperatureandsoilmoistureregimes(notshown)incombinationwiththeparametervaluesforhumus

Table3-2. Averagepercentageoffulfilledcriteriainthebestsimulationsselectedthroughthemultiplecriteriaparameterisationmethod.The30–45criteriaformodelacceptancehavebeenaggregatedintogroupsofsimilarnature.

Carbonstorages Carbonflows MiscellaneousSoil Tree Field NPP2 Respiration Litterfall LAI3 Nprocesses Waterflows

Grassland 95 n.a. 78 75 100 20 50 100 n.a.Alder 100 67 46 43 55 8 70 66 n.a.Pine 100 93 66 48 40 40 50 54 n.a.Spruce 100 90 74 47 50 28 45 96 n.a.Sprucem1 100 80 66 50 25 42 60 81 58Mean 99 83 66 53 54 28 55 79 58

1Managed forest. 2Net primary production. 3Leaf area index.

14

decomposition(TableA2).Inconclusion,theturnovertimeforanatomofcarbonandassociatedtraceelementssuchasradionuclidesenteringtheecosystemcanvarybetweenonetomorethanonehundredyears.Ifthefateofradionuclidesisofinterest,itisthusofimportancetoknownotonlyhowmuchisassimilatedintothebiomass,butalsohowthetraceelementsareallocatedtodifferentpartsofthesysteminordertopredicttheturnovertimeofthetraceelements.Tomakeitevenmorecomplex,theallocationanddecompositionpatternsmaydifferbetweentraceelementsandcarbon.However,thepresentsimulationofthecarbonturnoverpatternsinthefiveselectedecosystemsmaybeastartingpointforamoreelaborateanalysisoftraceelementturnover.

Themajoruncertaintiesinsimulatedcarbonbudgetsaswellasinthemodelparameterisations,wererelatedtocarbonfluxes,i.e.photosynthesis,respirationandlitterfall.Thiswasevidentfromthelargestandarddeviationinthesimulatedfluxes,andintherelativelylownumberofcriteriametforrespirationandnetprimaryproductioncomparedtotheothergroupsofcriteria.Nonetheless,webelievethatanacceptableleveloffulfilledcriteriawasachievedforallecosystemsinordertodrawconclusionsonthevariationofbehaviourwithinandbetweenthesesystems.Carbonbudgetsforborealecosystemsarecharacterizedbyasmallnetexchangeasaresultoflargeinflowsandoutflows.Forsuchecosystems,itisofhighimportancetohavewell-definedcriteria,notonlywithregardstotherelationshipbetweentheinflowsandoutflowsofcarbon,butalsoontheabsolutelevels,toavoidover-parameterisations.Theestimatedturnovertimesforthewoodypartofthetrees(approximately20–�0years),maybetoohighcomparedtootherstudies/e.g.Barrett2002/,whichindicatesthatrespirationaswellasassimilationofcarbonmayhavebeenoverestimated.Ontheotherhand,theturnovertimeofabout150yearsintheslowlydecomposingsoilorganicmatter(humus)iswellwithintherangeofreportedvalues/e.g.Schulzeetal.2005/.Netecosystemcarbonfluxdataweremissinginthisstudy,whichcouldhaveimprovedtheparameterisationoftheplantandsoilrespiration.Ontheotherhand,theresultingcarbonbudgetsforthespruceecosystemweresimilartothosepresentedbyforinstanceMedlynandco-workers/Medlynetal.2005/,whoreportedatotalsystemrespira-tionofabout900–1,600gCm–2yr–1comparingspruceandpineforestsinSweden,UK,andFrance.Similarly,thetotalsystemrespirationfromanalderforestinGermanywasestimatedtobeabout1,800gCm–2yr–1,whichisjustslightlyhigherthanthosereportedinthisstudy/Kutchetal.2001b/.

Thisstudydemonstratedamethodforhowtoidentifycrucialecosystemsbehaviourusingadetailedprocessorientedmodel.Themodelwassuccessfullycombinedwithclimaticdatafromonesiteandanumberofdifferentdatasourcestoprovideconsistentdescriptionsofcarbonfluxesfordifferentterrestrialecosystems.Obviously,thecriteriaformodelacceptancecouldbedesignedinmanywaysandthebasicfunctionaldifferencesbetweentheassumedparametersettingscouldbediscussed.Thecriteriaweretosomeextentsubjectivelychosenandcouldnoteasilybefulfilledbycomparisonwithindependentdata.However,theapproachwasoperationalandtransparentwhensynthesizingknowledgefrommanydifferentsourcesandinvestigations.Obviousproblemsarethattheperiodofsimulationisverylongandthattheecosystemsaretoalargeextentonlysimplifiedrepresentationsofpossiblerealecosystems.Thismeansthataconventionalvalidationandcalibrationofmodelparametersisnotarealisticalternative.Thedescriptionofcarbonturnoverinthedifferentecosystemsprovidedaframeforadualunderstandingofthemodeldesign:firstoftheconsequencesoftheinteractionbetweenprocessesdescribedinthemodel,andsecondlyoftheabilityofthemodeltoestimateimportantcharacteristicssuchasturnovertimeindifferentcomponentsoftheecosystems.Furthermore,theparameterisationderivedfromthisstudycoulddirectlybeusedfordifferenttypesofsimulationexperiments,suchastheimpactofdifferentlandusepracticesonthetransportandretentionofdifferentradionuclides.Inordertounderstandtraceelementturnoverinterrestrialecosystemsitisimportanttorecognisenotonlythedilutionandallocationinthebiomassfollowingcarbonassimilationbytheplantandturnoverinthesoil,butalsoplantwateruptakefromthesoilwithitsassociatedtraceelementuptake.Investigationswerethelinkbetweentraceelementturnoverandfluxesofbothwaterandbiomassinthesoil-plantsystemiscrucialcouldbenefitfromthetypeofmodeldescriptionspresentedinthisstudy.

15

4 CarbonandwaterbudgetsatfoursiteslocatedwithintheOskarshamnstudy-area

4.1 EcosystemdescriptionFoursiteslocatedwithintheOskarshamnstudy-areawereincludedinthispartofthestudy:asemi-naturalgrasslandonafinesandwithahighsoilorganicmattercontent,analderforestonpeat,amanagedspruceforestonpeatandfinallyapineforestontill(Table4-1).Allecosystemsaresimilartothecorrespondinghypotheticalsystemsdescribedintheprevioussectionintermsofthevegetationcover,butdiffersinsoilnitrogenandcarboncomposition.

4.2 ParameterisationMeasureddataonsoilcarbonandnitrogencontentateachsitewasusedtoparameterisetheinitialconditionsofthelitterandhumuspoolsinthesite-specificsimulations.Totalsoilrespiration(autotrophicandheterotrophic)atthegrassland,spruceandpineforestsiteswasmeasuredmonthlyfromMarch2004toMarch2005(Tagessonunpublisheddata).Thesemeasurementswereusedtocalibratethesimulationsbyadjustingthegroundwaterlevel,sincethisvariablewasunknown.Duetodifferentexposuretooxygen,achangeingroundwaterlevelpredominantlyaffectsheterotrophicrespirationandhasalargeimpactonsoilsrichinorganicmattersuchasthoseatthestudy-site.

Tocreatereasonableinitialconditionsatthebeginningofthesoilrespirationmeasurementperiodin2004,simulationswerestartedfrom199�andendedinJuly2005.ClimaticvariablesweremeasuredatÄspömeteorologicalstation,about�0kmfromthestudyarea/SKB2005c/.Hourlydataonprecipitation,airtemperature,globalradiation,windspeedandrelativehumiditywasavailablefrom200�-09-09to2005-07-07.Fromthisdata-setthevariablesfrom2004werereplicatedseveraltimestocreatealongertime-seriesrangingfrom199�-01-01to200�-09-09.Thesetwodata-setwerethencombinedtoformacontinuousseriesfrom199�-01-01to2005-07-07.

Inordertogetanestimateofthevarianceinthedifferentpartsofthecarbonbudgets,therelativevariance(i.e.thestandarddeviationdividedbythemean)estimatedforthehypotheticalsystemswasmultipliedwiththemeanvaluesfromthesite-specificsimulations.Carbonturn-overwasnotcalculatedforthesite-specificecosystemssincenoneofthemwerefoundtobeinsteady-state.Sincethedepthoftheorganiclayerwasnotknown,onlytheorganiccontentintheuppermostmeterofthesoilisshownintheresults.

Table4-1. Descriptionofthemaincharacteristicsoftheecosystemsincludedinthestudy.

Name Treelayer Fieldlayer Soiltype

Grassland none Grass and herbs ClayAlder Alder (Alnus glutinosa) Grass and herbs PeatPine Scots pine (Pinus sylvestris) Cowberry (Vaccineum vitis-idea) Till (thin)/bedrockSpruce,managed Norway spruce (Picea abies) Blueberry (Vaccineum myrtillis) Peat

16

4.3 ResultsSoilrespirationwasgenerallywelldescribedinthemodelsimulations(Figure4-1).Forthegrassland,themaindeviationbetweensimulatedandmeasuredvaluesoccursinAugustandSeptember(Figure4-1a).Duringthesemonthssimulatedvaluesfarexceedmeasurements,causingtheannualsimulatedmeantobehigherthanthemeasured(�.45and2.89gCm–2day–1respectively).Nonetheless,theagreementbetweensimulatedandmeasuredvalueswasstillquitehigh(r2=0.72).Simulatedsoilrespirationinthepineforestwasoverestimatedduringthewintermonthswhereasduringsummer,itwassometimesunderestimated(Figure4-1b).Thiscausedthesimulatedannualmeansoilrespirationtobesomewhatlarge(�.87gCm–2day–1,comparedwiththemeasured�.17gCm–2day–1),andtheoverallagreementbetweensimulationsandmeasurementswaslowerthanforthegrassland(r2=0.60).Thebestcorrelationwasfoundforthespruceforest,inwhichsimulatedsoilrespirationagreedwellwithmeasurementsthroughouttheentireyear(r2=0.85)(Figure4-1c).Meanannualsoilrespirationinthesimula-tionwas2.72gCm–2day–1comparedwith2.55gCm–2day–1measuredintheforest.

Carbonbudgetswereestimatedforeachecosystembasedontheresultsfromthesimulations(Figure4-2).Comparedwiththehypotheticalecosystems,thesite-specificsystemshadlargerfluxesofcarbontoandfromthesystem.Duetothehighorganiccontentinthesoils,thevastmajorityofcarbonislocatedbelowgroundinallecosystems.Noneoftheecosystemsareinsteady-stateintermsofcarbonstorage(Figure4-2).Forthegrassland,theannualaveragenetecosystemexchange1isabout–600gCyr–1,whileitisclosetozerointhespruceforest(–80gCyr–1)andispositiveinthealderandpineforests(2�0gCyr–1and200gCyr–1respectively).Soilorganiccontentdecreasesannuallyinthegrasslandandthespruceforest,whileitincreasesinthealderforestandremainsstableinthepineforest.Inallecosystems,thecarboncontentinthevegetationlayereitherincreasesorremainsratherstable.

1 Anegativesignindicatesalossofcarbonfromthesystem.

Figure 4-1. Simulated (crosses) and measured (open circles) soil respiration from March 2004 to March 2005. Hourly mean values measured during daytime.

17

Annualwaterflowswereestimatedinthesimulations(Figure4-�).Theseestimatesshowthatmostofthewaterenteringtheecosystemsasprecipitationleavesassubsurfacedrainageorsurfacerunoff,exceptinthealderforest.Thisisparticularlytrueforthegrasslandandthepineforest,inwhichevapotranspirationisabout�0%ofincomingprecipitation.Thecorrespondingfigureisaround45%forthespruceforest.Intheconiferousforestssurfacerunoffexceedsdrainage,whiletheoppositeholdstrueforthegrasslandandalderforest.Despiteaverylowinterceptionevaporation,totalevapotranspirationinthealderforestisabout50%.Averyhighsoilevaporationconstitutesthemajorityofthatfigure.

Figure 4-2. Carbon budgets for the site-specific ecosystems located within the Oskarshamn study area. a) grassland b) alder c) pine and d) spruce managed forest. Carbon storages (gC m–2) in bold and carbon fluxes (gC m–2 yr–2) in italics, including standard deviations estimated from Figure 3-1.

D.

B.

Carbon storage (gC m-2)

Carbon efflux (gC m-2 yr-1)

Carbon influx (gC m-2 yr-1)

Carbon internal system flux (gC m-2 yr-1)

A.

C.

870 ± 210

1470 ± 290

1830 ± 400

36700 ± 3050

820 ± 180

960 ± 270

8.2 ± 1.4

300 ± 70

340 ± 300

160 ± 90

250 ± 70

280 ± 120

520 ± 400

920 ± 350

49000 ± 3210

16 ± 4.3

570 ± 180

6850 ± 790

690 ± 760

130 ± 90

180 ± 140

930 ± 170

850 ± 920

1080 ± 260

1770 ± 400

410 ± 100

9080 ± 2400

86600 ± 6240

3.2 ± 1.3

830 ± 490

1390 ± 870

890 ± 690

130 ± 70

550 ± 170

420 ± 210

630 ± 340

7850 ± 1650

20000 ± 1340

1070 ± 780

0.2 ± 0.02

18

4.4 DiscussionSimulatedsoilrespirationwasfittedtomeasurementsbyadjustingthegroundwaterlevel,whichpredominantlyaffectedheterotrophicrespiration.Bydoingso,itwaspossibletoaffecttheannualmeanrateofsoilrespiration,butnotthedynamicbehaviourofsimulatedsoilrespirationoverthewholeyear.Despitethis,thesimulateddevelopmentofsoilrespirationovertheyearshowedahighcorrelationwithmeasurements,asindicatedbythehighr2-values.Hence,themodelseemscapableofaccuratelydescribingsoilrespirationprocesseswithintheyear.Thelowestcorrelationwasfoundinthepineforest,whichisprobablyduetodifficultiesrelatedtoestimatingsoilwaterfluctuationsandsurfacerunoffinapatchyterrain.

Comparedwiththehypotheticalecosystems,thesite-specificsystemsdifferedpredominantlyinsoilcomposition.AllsoilsfromtheOskarshamnstudy-areawereveryrichinorganicmatter,andsoilrespirationwashighatallmeasuredsites.Theestimatesfromthesimulationindicatesthatasmuchas80%oftotalsoilrespirationisheterotrophicrespirationinthegrassland.Forthespruceandpineforeststhecorrespondingfigureswere45%and�5%respectively.Arapidlydecreasingcarboncontentinthesoilatthegrasslandsite,indicatesthatthearearecentlyhasbeendrained,therebyexposingnon-decomposedcarbontoaerobicconditionsleadingtohighratesofheterotrophicrespiration.Asimilarsituationseemstoprevailatthespruceforestsite,althoughatamuchlowerrate.Thissiteisstillratherwet,resultinginamuchlowerdecreaseinsoilcarboncontentcomparedtothegrassland.Atbothsites,thesoilswillcontinuetoloosecarbonuntilmostcarbonabovethegroundwaterlevelhasbeendecomposed.Iffurtherdrainageisimposed,morenon-decomposedcarbonwillbeexposeddependingonthedepthofthepeatlayerofthesoil.Anothereffectofthisprocessisanaturalfertilisationofthesoil.Ascarbondioxideistransferredtotheatmosphere,nitrogenisleftbehindinthesoil.Thiswasseeninthesimulationsbothatthegrasslandandthespruceforestsitesasahighrateofphotosynthesis.Thesoilatthepineforestsiteseemsstableintermsofcarboncontent.Comparedwiththehypotheticalforest,thissoilisricherinnitrogencontent,whichcausesplantgrowthtobesignificantlyhigher.Thehighgrowthratealsocontributestohighsoilrespirationrates.Duetoverywetsoilconditions,thesoilinthealderecosystemistheonlysoilthatincreasesincarboncontentovertime;however,sinceneithersoilrespirationnorgroundwaterlevelweremeasuredatthissite,thisshouldonlybeconsideredasaroughestimate.

Figure 4-3. Participation of incoming precipitation for the site-specific ecosystems located within the Oskarshamn study area.

Soil evaporation Transpiration

Interception Surface runoff

Drainage

Grassland

Pine

Alder

Spruce

19

ThesimulatedcarbonbudgetscanalsobecomparedtodescriptivemodelsoncarbonfromthreeofthedifferentecosystemsinOskarshamn,butwithslightlydifferentsoilconditions/Löfgrenetal.2006/.Duetotherapidcombustionofsoilcarboninthegrasslandsimulationsandtheconcurrentfertilisationofthesoil,NPP,biomasscarboncontentandconsequentlyalsolitterfallbecomesverylarge,comparedwiththedescriptivemodel.Itispossiblethatthecurrentmodeldesignandparameterisationtendstocauseanexaggeratedimpactofsoilnitrogenonplantgrowth.Heterotrophicsoilrespirationwasalsolargerinthesimulationscomparedwiththedescriptivemodel,whichmaypartlybeexplainedbyaslightoverestimationoftotalsoilrespirationinthesimulations.Despitethis,simulatedheterotrophicrespirationisstilllargerthaninthedescriptivemodel.However,ifheterotrophicrespirationwastobeloweredinthesimulations,rootrespirationhadtobeincreasedsubstantiallyinordertocorrelatetotalsoilrespirationwithmeasurements.Thus,itispossiblethatsoilheterotrophicrespirationisslightlyunderestimatedinthedescriptivemodel.Inthepineforestecosystemsimulation,therelativelylownitrogenstresscausesbiomass,NPPandlitterfalltobehighcomparedwiththedescriptivemodelbyLöfgrenandco-workers/Löfgrenetal.2006/.Similartothegrassland,soilheterotrophicrespirationishighinrelationtothevaluesgiveninthedescriptivemodel.Finally,thesimulatedspruceecosystemwascomparedtoacorrespondingsystemdescribedbyLöfgrenandco-workers/Löfgrenetal.2006/.WhilebothNPPandlitterfallweresimilarinthesimulationsandthedescriptivemodel,soilheterotrophicrespirationwasagainhigherinthesimulations.Dissolvedorganiccarbonwasofthesameorderofmagnitudeinallecosystemsinboththesimulationsandthedescriptivemodels.Toconclude,thebiomasscarboncontentandtheassociatedvariablesNPPandlitterfall,aredependentonthenutrientstatusofthesoil,whichseemedtodifferbetweenthesimulationsandthedescriptivemodels.Heterotrophicrespirationishighinthesimulationscomparedwiththedescriptivemodels,whichmightbeareflectionofthedifficultyinseparatingheterotrophicrespirationfromtotalsoilrespiration.Hence,itseemsthatbothmodelsdescribethecarbonbudgetsoftheecosystemsinasimilarway.

Inthegrasslandecosystem,lowevapotranspirationiscausedbylowtranspirationsincethereisonlyonetranspiringvegetationlayer.Alotofwaterinfiltratesinthesoilandleavesthesystemassubsurfacedrainage.Thealderforestisinsteadcharacterisedbyverywetconditionsinlow-landterrain.Thiscausesahighsoilevaporationandaratherlowsurfacerunoff.Interceptionevaporationwasprobablyunderestimatedinthemodelsimulations,butontheotherhand,transpirationmighthavebeenexaggeratedsinceitwasnotaffectedbythenegativeimpactsfromprolongedwaterlogging.Surfacewaterconditionsweredifficulttoportrayinthepineforestsimulations,sincetheterrainisheterogeneous,consistingofathinsoilcoverinterspacedwithbarerock.Theseconditionscausealargesurfacerunoffofwaterwhiledrainageisclosetozeroduetonearlyimpermeablebedrock.Bothtreesandfieldlayerregularlyfacewaterstressconditions,leadingtorelativelylowtranspirationlevels.Inthespruceforest,wetsoilmoistureconditionswereratherfavourablefortranspiration,thusgivingrisetofairlyhightranspirationrates.Thehighgroundwatertablealsocausedhighratesofsurfacerunoffwhiledrainageremainedrelativelylow.

21

5 Conclusion

Carbonbudgetsandmeanresidencetimeswereestimatedinfourhypotheticalecosystems.Thegreatestuncertaintiesintheestimationslieinthecalculationoffluxestoandfromthefieldlayer.Aparameterisationmethodbasedonmultiplecriteria,synthesisingawiderangeofempiricalknowledgeonecosystembehaviour,provedtobeusefulbothintheestimationofunknownparameters,todemonstratemodelsensitivity,andtoidentifyprocesseswhereourcurrentknowledgeislimited.Theparameterisationsderivedfromthestudyofthehypotheticalsystemswereusedtoestimatesite-specificcarbonandwaterbudgetsforfourecosystemslocatedwithintheOskarshamnstudy-area.Measuredsoilrespirationwasusedtocalibratethesimulations.Ananalysisofthesimulatedcarbonfluxesindicatedthattwooftheecosystems,namelythegrasslandandthespruceforest,werenetsourcesofcarbondioxide,whilethealderandthepineforestwerenetsinksofCO2.Intheformercase,thiswasinterpretedasaresultofrecentdrainageoftheorganogenicsoilsandtheconcurrentincreaseindecomposition.TheresultsfromthestudyconformedratherwellwithresultsfromapreviousstudyoncarbonbudgetsfromtheOskarshamnstudyarea.

2�

Acknowledgement

Anumberofpersonshavecontributedtothisreport.MagnusSvensson,KTH,hasbeenhelpfulwithprovidingusefulliteraturereferences,andhassharedhisknowledgeinecosystemsmodellingofborealforests.AnnikaNordin,TryggvePersson,BengtOlssonandTomEricsson,SLUandUmeåPlantScienceCentre,kindlyprovideddatafortheparameterisationoftheecosystems.AndersLöfgren,TobiasLindborgandUlrikKautsky,SKB,havealsobeenhelpfulwithprovidingsite-specificdatafromOskarshamnforthesimulations.AndersLindrothandTorbernTagesson,LundUniversity,sharedtheirsoilrespirationmeasurementsforthesecondpartofthestudy.Finally,ClareBradshaw,StockholmUniversity,commentedonpartsofapreviousversionofthemanuscript.SKBprovidedthefundingforthisstudy.

25

References

AlvenäsG,JanssonP-E,1997.Modelforevaporation,moistureandtemperatureofbaresoil:calibrationandsensitivityanalysis.AgrForestMeteorol88,47–56.

BarrettDJ,2002.SteadystateturnovertimeofcarbonintheAustralianterrestrialbiosphere.GlobalBiogeochemCy16(4)art.1108,21pp.

BerggrenD,BergkvistB,JohanssonM-B,LangvallO,MajdiH,MelkerudP-A,NilssonÅ,WeslienP,OlssonM,2004.AdescriptionofLUSTRA’scommonfieldsites.DepartmentofForestSoils,SwedishUniversityofAgricultureSciences,Uppsala,Report87.41pp.

BettsAK,BallJH,1997.Albedoovertheborealforest,JGeophysRes102,28901–28909.

BlombäckK,1998.Carbonandnitrogenincatchcropsystems–modellingofseasonalandlong-termdynamicsinplantandsoil.Ph.D.Thesis.ActaUniversitatisAgriculturaeSueciae,Agraria1�4.SwedishUniversityofAgriculturalSciences,Uppsala.�1pp.

BosselH,1996.TREEDYN�forestsimulationmodel.EcolModel90,187–227.

CanhamCD,PaceML,PapaikMJ,PrimackAGB,RoyKM,MarangerRJ,CurranRP,SpadaDM,2004.Aspatiallyexplicitwatershed-scaleanalysisofdissolvedorganiccarboninAdirondacklakes.EcolAppl14(�),8�9–854.

ChapinFS,MatsonP,MooneyHA,2002.Principlesofterrestrialecosystemecology.Springer-Verlag,NewYork.4�6pp.

Charles-EdwardsDA,DoleyD,RimmingtonGM,1986.Modellingplantgrowthanddevelopment.AcademicPress,Australia,2�5pp.

ChungHH,BarnesRL,1977.PhotosynthateallocationinPinus taeda.I.Substraterequire-mentsforsynthesisofshootbiomass.CanJForestRes7,106–111.

DillyO,BachH-J,BuscotF,EschenbachC,KutchWL,MiddelhoffU,PritschK,MunchJC,2000.CharacteristicsandenergeticstrategiesoftherhizosphereinecosystemsoftheBornhövedLakedistrict.ApplSoilEcol15,201–210.

EckerstenH,JanssonP-E,JohnssonH,1998.SOILNmodel,ver.9.2,User’smanual.DivisionofHydrotechnics,Communications98:6,DepartmentofSoilSciences,SwedishAgriculturalUniversity,Uppsala.11�pp.

EricssonT,1994.Nutrientdynamicsandrequirementsofforestcrops.NewZealJForSci24(2/�),1��–168.

EricssonT,1995.Growthandshoot:rootratioofseedlingsinrelationtonutrientavailability.PlantSoil168(1),205–214.

FieldC,MooneyHA,1986.Thephotosynthesis-nitrogenrelationshipinwildplants.In:Ontheeconomyofplantformandfunction.Givnish,T.J.(ed.),CambridgeUniversityPress,Cambridge.pp25–55.

Flower-EllisJGK,1985.LitterfallinanageseriesofScotspinestands:Summaryofresultsfortheperiod197�–198�.Institutionenförekologiochmiljövård.Sverigeslantbruksuniversitet.Rapport19,75–94.

FosterTE,BrooksJR,2005.Functionalgroupsbasedonleafphysiology:aretheyspatiallyandtemporallyrobust?Oecologica144,��7–�52.

GoverdeM,ArnoneJA,ErhardtA,2002.Species-specificreactionstoelevatedCO2andnutrientavailabilityinfourgrassspecies.BasicandAppliedEcology�,221–227.

26

GustafssonD,StähliM,JanssonP-E,2001.Thesurfaceenergybalanceofasnowcover:comparingmeasurementstotwodifferentsimulationmodels.TheorApplClimatol70,81–96.

GustafssonD,LewanL,JanssonP-E,2004.Modelingwaterandheatbalanceoftheboreallandscape–comparisonofforestandarablelandinScandinavia.JApplMeteorol4�(11),1750–1767.

HelmisaariH-S,MakkonenK,KellomäkiS,ValtonenE,MälkönenE,2002.Below-andabove-groundbiomass,productionandnitrogenuseinScotspinestandsineasternFinland.ForestEcolManag165,�17–�26.

HoffmannF,1995.FAGUS,amodelforgrowthanddevelopmentofbeech.EcolModel8�,�27–�48.

JanssonP-E,HalldinS,1979.Modelfortheannualwaterandenergyflowinalayeredsoil.In:Comparisonofforestandenergyexchangemodels.Halldin,S.(ed.).SocietyforEcologicalModelling,Copenhagen.pp145–16�.

JanssonP-E,MoonD,2001.Acoupledmodelofwater,heatandmasstransferusingobjectorientationtoimproveflexibilityandfunctionality.EnvironModellSoftw16(1),�7–46.

JanssonP-E,KarlbergL,2004.Coupledheatandmasstransfermodelforsoil-plant-atmospheresystems.RoyalInstituteofTechnology,DeptofCivilandEnvironmentalEngineering.,Stockholm,Sweden,4�5pp.

JohanssonT,2000.Biomassequationsfordeterminingfractionsofcommonandgreyaldersgrowingonabandonedfarmlandandsomepracticalimplications.BiomassBioenerg18,147–159.

JohnsonDW,LindbergSE,1992.Atmosphericdepositionandforestnutrientcycling.EcologicalStudies91,Springer-Verlag,NewYork.707pp.

JohnssonH,BergströmL,JanssonP-E,PaustianK,1987.Simulatednitrogendynamicsandlossesinalayeredagriculturalsoil.AgriculturalEcosystemsandEnvironment18,���–�56.

KarlbergL,GustafssonD,JanssonP-E,2006a.Modelingcarbonturnoverinfiveterrestrialecosystemsintheborealzoneusingmultiplecriteriaofacceptance.Ambio�5(8),448–458.

KarlbergL,Ben-GalA,JanssonP-E,ShaniU,2006b.Modellingtranspirationandgrowthinsalinity-stressedtomatounderdifferentclimaticconditions.EcolModel190(1–2),15–40.

KinersonR,RalstoneCW,WellsCG,1977.Carboncyclinginaloblollypineplantation.Oecologia29,1–10.

KutchWL,StaackA,WötzelJ,MiddelhoffU,KappenL,2001a.Fieldmeasurementsofrootrespirationandtotalsoilrespirationinanalderforest.NewPhytol150,157–168.

KutchWL,EschenbachC,DillyO,MiddelhoffU,SteinbornW,VanselowR,WeisheitK,WötzelJ,KappenL,2001b.ThecarboncycleofcontrastinglandscapeelementsoftheBornhövedLakedistrict.In:EcosystemapproachestolandscapemanagementinCentralEurope.Tenhunen,J.D.etal.(eds.),Springer-Verlag,BerlinHeidelberg.pp75–95.

LarcherW,2003.Physiologicalplantecology:ecophysiologyandstressphysiologyoffunctionalgroups.4ed.Springer-Verlag,Berlin.52�pp.

Larsson-McCannS,KarlssonA,NordM,SjögrenJ,JohanssonL,IvarssonM,KindellS,2002.Meteorological,hydrologicalandoceanographicalinformationanddataforthesiteinvestigationprograminthecommunityofOskarshamn.SKBTR-02-0�,SvenskKärnbränslehanteringAB.

LindborgT,2005.Descriptionofsurfacesystems.PreliminarysitedescriptionSimpevarpsubarea–Version1.2.R-05-01.SvenskKärnbränslehanteringAB.27�pp.

LinderS,TroengE,1981.Theseasonalvariationinstemandcoarserootrespirationina20-yearoldScotspine(Pinus sylvestrisL.).MitteilungenderForstlichenBundes-Versuchs-anstalt(Wein)142,125–1�9.

27

LindrothA,1985.Canopyconductanceofconiferousforestsrelatedtoclimate.WaterResourRes21,297–�04.

LöfgrenA,2005.Oskarshamn/Forsmarksiteinvestigation.Estimationofbiomassandnetprimaryproductioninfieldandgroundlayer,andbiomassinlitterlayerofdifferentvegetationtypesinForsmarkandOskarshamn.SKBP-05-80,SvenskKärnbränslehanteringAB.�7pp.

LöfgrenA,MilianderS,TruvéJ,LindborgT,2006.CarbonbudgetsforcatchmentsacrossamanagedlandscapemosaicinsoutheastSweden.Ambio�5(8),pp459–468.

MarklundL-G,1988.Biomassafunktionerförtall,granochbjörkiSverige.SLU,Inst.förskogstaxering,Rapport45,Umeå.7�pp(InSwedish).

MedlynBE,BerbigierP,ClementR,GrelleA,LoustauD,LinderS,WingateL,JarvisPG,SigurdssonBD,McMurtrieRE,2005.Carbonbalanceofconiferousclimates:Model-basedanalysis.AgrForestMeteorol1�1,97–124.

MiddletonEM,ChanSS,RusinRJ,MitchellSK,1997.OpticalpropertiesofblackspruceandjackpineneedlesatBOREASsitesatSaskatchewan,Canada.CanadianJournalofRemoteSensing2�,108–119.

MonteithJL,1965.Evaporationandtheatmosphere.In:Thestateandmovementofwaterinlivingorganisms.19thSymp.Fogg,G.E.(ed.).Soc.Exp.Biol.,TheCompanyofbiologists,Cambridge.pp205–2�4.

MonteithJL,1977.ClimateandtheefficiencyofcropproductioninBritain.PhilosTRoySocB281,277–294.

NilssonL-O,ÖstergrenM,WiklundK,2001.Hurpåverkasträdenovanmark?In:Skogaby-försöket–effekteravlångvarigkväve-ochsvaveltillförseltillettskogsekosystem.Persson,T.andNilsson,L.-O.(eds.).Naturvårdsverket,Rapport517�,Stockholm.51–74(InSwedish).

NordinA,NäsholmT,1997.Nitrogenstorageformsinnineborealunderstoreyspecies.Oecologica110,487–492.

NordinA,NäsholmT,EricsonL,1998.EffectsofsimulatedNdepositiononunderstoryvegetationofaborealconiferousforest.FunctEcol12,691–699.

NäsholmT,EkbladA,NordinA,GieslerR,HögbergM,HögbergP,1998.Borealforestplantstakeuporganicnitrogen.Nature�92,914–916.

OkeTR,1987.Boundarylayerclimates.Methuen,London,pp12and1�8–1�9.

OlssonM,2000.Bidrarmarkentillväxthuseffekten–ellerkandenhjälpaossattlösaproblemen?KungligaSkogs-ochLantbruksakademinstidsskrift1�9(1�),41–50(InSwedish).

PengYY,DangQ-L,2003.Effectsofsoiltemperatureonbiomassproductionandallocationinseedlingsoffourborealtreespecies.ForestEcolManag180,1–9.

PenmanHL,1953.Thephysicalbasisofirrigationcontrol.In:Reportofthe1�thinternationalhorticulturalcongress1952.Synge,P.M.(ed.).Roy.HorticulturalSoc.,London,vol:II.pp91�–924.

PenningdeVriesFWT,1975.Thecostofmaintenanceprocessinplantcells.AnnalsofBotany�9,77–92.

PenningdeVriesFWT,vanLaarHH,1982.Simulationofplantgrowthandcropproduction.Simulationmonographs,Wageningen,theNetherlands.�07pp.

PerssonT,NilssonL-O,2001.Skogabyförsöket–effekteravlångvarigkväve-ochsvaveltillförseltillettskogsekosystem.Naturvårdsverket,Rapport517�,Stockholm.220pp(InSwedish).

RaichJW,TufekciogluA,2000.Vegetationandsoilrespiration:Correlationsandcontrols.Biogeochemistry48,71–90.

28

Reich PB, Walters MB, Ellsworth DS, Vose JM, Volin JC, Gresham C, Bowman WD, 1998. Relationships of leaf dark respiration to leaf nitrogen, specific leaf area and leaf life-span: a test across biomes and functional groups. Oecologia 114(4), 471–482.

Schulze E-D, 2000. Carbon and nitrogen cycling in European forest ecosystems, Ecological Studies 142, Springer: Berlin, Heidelberg, New York, 500 pp.

Schulze E-D, Beck E, Müller-Hohenstein K, 2005. Plant ecology. Springer-Verlag, Berlin. 702 pp.

Sellers PJ, Hall FG, Kelly RD, Black A, Baldocchi D, Berry J, Margolis H, Ryan M et al. 1997. BOREAS in 1997: Scientific results, experimental overview and future directions. J Geophys Res 102, 28731 – 28770.

SKB, 2005a. Description of surface systems. Preliminary site description. Forsmark area – version 1.2. SKB R-05-03, Svensk Kärnbränslehantering AB.

SKB, 2005b. Description of surface systems. Preliminary site description. Simpevarp sub area – version 1.2. SKB R-05-01, Svensk Kärnbränslehantering AB.

SKB, 2005c. Description of climate, surface hydrology, and near-surface hydrogeology. Simpevarp 1.2. SKB R-05-06, Svensk Kärnbränslehantering AB.

Skogsdata, 2003. Aktuella uppgifter om de svenska skogarna från Riksskogstaxeringen. SLU, Inst. för skoglig resurshushållning och geomatik. 114 pp (In Swedish).

Strand M, Lundmark T, Söderbergh I, Mellander P-E, 2002. Impacts of seasonal air and soil temperatures on photosynthesis in Scots pine trees. Tree Physiol 22, 839–847.

Swedish Environmental Protection Agency, 2005. Sweden’s National Inventory Report 2005 – Submitted under the United Nations Framework Convention on Climate Change. Report 5451, Swedish Environmental Protection Agency, Stockholm, Sweden, pp 216. Available at: http://www.naturvardsverket.se/bokhandeln.

Swedish Forest Agency, 2005. Grundbok för skogsbrukare. Skogsstyrelsens förlag, Jönköping, Sweden. 190 p (In Swedish).

Thornley JHM, Cannell MGR, 1992. Nitrogen relations in a forest plantation – soil organic matter ecosystem model. Annals of Botany 70, 137–151.

Tsialtas JT, Pritsa TS, Veresoglou DS, 2004. Leaf physiological traits and their importance for species success in a Mediterranean grassland. Photosynthetica 42(3), 371–376.

Uri V, Tullus H, Lõhmus K, 2002. Biomass production and nutrient accumulation in short-rotation grey alder (Alnus incana (L.) Moench) plantation on abandoned agricultural land. Forest Ecol Manag 161, 169–179.

Waring RH, Running SW, 1998. Forest ecosystems. Analysis at multiple scales. Second edition. Academic Press, California, USA. 370 pp.

Wikström F, Ericsson T, 1995. Allocation of mass in trees subject to nitrogen and magnesium limitation. Tree Physiol 15, 339–344.

Personal communication and unpublished dataNordin A, Associate Professor, Umeå Plant Science Centre, Sweden.

Olsson B, Researcher, Dept. of Ecology and Environmental Research, SLU, Sweden.

Svensson M, PhD student, Dept. of Land and Water Resources Engineering, KTH, Sweden.

Tagesson T, PhD student, Dept. of Physical Geography and Ecosystems Analysis, Lund University, Sweden.

29

Appendix

TableA1. Differencesbetweenplanttypes,primaryparameters.

Property Spruce Pine Alder Blueberry Cowberry Grass/herb Unit Ref.

Photosynthesisandresp.Radiation use efficiency1 8 8 8 8 8 8 gDwMJ–1 /Charles-Edwards et al. 1986/Optimum temp. interval 10–25 15–30 15–25 10–20 10–20 10–20 ºC /Hoffmann 1995, Strand et al. 2002, Larcher 2003,

Peng and Dang 2003/Min/max temp –4/40 –5/40 –2/40 –2/40 –2/40 –2/40 ºC /Hoffmann 1995, Larcher 2003, Peng and Dang 2003/Optimum C/N ratio, leaf 24 24 11 24 24 11 – /Wikström and Ericsson 1995/Threshold C/N ratio, leaf 77 77 77 cal. cal. cal. – /Field and Mooney 1986, Wikström and Ericsson 1995/Growth resp.2 0.21 0.21 0.21 0.21 0.21 0.21 gC gC–1 /Chung and Barnes 1977, Linder and Troeng 1981,

Penning de Vries and van Laar 1982/Maint. resp. stem3 0.00015 0.00015 cal. cal. cal. cal. day–1 /Kinerson et al. 1977, Linder and Troeng 1981,

Bossel 1996/Maint. resp. fine roots 0.006 0.006 cal. cal. cal. cal. day–1 /Linder and Troeng 1981, Bossel 1996/AllocationRoot mass alloc. coef.4 0.00004 0.00004 n.a. n.a. n.a. n.a. – /Helmisaari et al. 2002/Frac. stem to coarse root 0.15 0.15 0.12 0.8 0.75 0 – /Nordin et al. 1998, Johansson 2000, Helmisaari

et al. 2002, Uri et al. 2002, Larcher 2003/Litterfall(oldbiomass)Leaf (during dormancy)5 cal. 0.055 0.1 0.1 0.1 0.1 day–1 AssumedStem 0.000018 0.000018 0.000018 0.00091 0.00091 0.1 day–1 AssumedCoarse roots 0.000018 0.000018 0.000018 0.00091 0.00091 0.00091 day–1 AssumedFine root 0.0027 0.0027 0.0027 0.0027 0.0027 0.0027 day–1 Assumed

�0

Property Spruce Pine Alder Blueberry Cowberry Grass/herb Unit Ref.

MinC/Nratioplants6

Leaf 22 22 11 23 30 15 gC gN–1 /Ericsson 1994, Nilsson et al. 2001, Goverde et al. 2002/ (Nordin pers. comm.)

Stem and coarse roots 830 830 265 37 40 15 gC gN–1 /Ericsson 1994, Goverde et al. 2002/ (Nordin pers. comm., Olsson pers. comm.)

Fine root 40 40 18 64 69 40 gC gN–1 /Ericsson 1994, Persson and Nilsson 2001, Nordin and Näsholm 1997/ (Nordin pers. comm.)

PhysicalpropertiesLeaf mass/unit leaf area 100 80 35 35 65 85 gC m–2 /Bossel 1996, Middleton et al. 1997, Reich et al. 1998,

Waring and Running 1998, Johansson 2000, Tsialtas et al. 2004, Foster and Brooks 2005/

Max height 30 20 16 0.3 0.2 0.8 m (Tagesson unpublished data)Root lowest depth 0.7 0.4 0.45 0.3 0.3 0.55 m /Lindborg 2005/Plant albedo 8 8 15 10 10 25 % /Oke 1987, Betts and Ball 1997, Gustafsson et al. 2001/Max. surface coverage 0.9 0.84 0.97 1 1 1 m2 m–2 (Tagesson unpublished data)

cal. = calibrated; n.a. = not available. 1Photosynthesis calculated as a linear function of absorbed radiation multiplied by response functions for leaf temperature and leaf C/N ratio.

2Growth respiration is deducted from gross photosynthesis before allocation to different storage organs of the plant and calculation on maintenance respiration. 3Same for coarse roots. 4Factor describing an increase in the allocation to roots with tree biomass. 5All plant types loose their leaves in the autumn at the onset of dormancy except spruce that looses its needles constantly over the season. 6Minimum C/N ratio measured in fertilisation experiments.

�1

TableA2. Calibratedparameters,differentecosystems.

Property Grassland Alder Pine Spruce Sprucemanaged Unit

PlantThreshold C/N ratio leaf, Field1 144 ± 44 116 ± 19 123 ± 39 115 ± 33 123 ± 33 gC gN–1

Leaf allocation coef., Tree n.a. 0.25 ± 0.06 0.24 ± 0.05 0.21 ± 0.06 0.18 ± 0.05 –Root allocation coef., Tree n.a. 0.17 ± 0.07 0.47 ± 0.05 0.47 ± 0.06 0.50 ± 0.05 –Leaf allocation coef., Field 0.63 ± 0.12 0.55 ± 0.09 0.19 ± 0.08 0.21 ± 0.09 0.23 ± 0.06 –Root allocation coef., Field 0.36 ± 0.04 0.45 ± 0.11 0.13 ± 0.05 0.10 ± 0.05 0.10 ± 0.04 –Mobile allocation coef, Tree2 n.a. 0.09 ± 0.06 0.16 ± 0.09 0.28 ± 0.12 0.32 ± 0.13 gC gC–1

Mobile allocation coef, Field2 0.40 ± 0.21 0.96 ± 0.13 0.68 ± 0.32 0.48 ± 0.21 0.56 ± 0.22 gC gC–1

Maint. resp. leaf, Tree n.a. 0.0051 ± 0.00093 0.0030 ± 0.0016 0.0038 ± 0.0012 0.0028 ± 0.0014 day–1

Maint. resp. stem, Tree3 n.a. 0.00032 ± 0.00012 n.a. n.a. n.a. day–1

Maint. resp. fine roots, Tree n.a. 0.0058 ± 0.0027 n.a. n.a. n.a. day–1

Maint. resp. leaf, Field4 0.010 ± 0.0034 0.015 ± 0.0088 0.017 ± 0.011 0.025 ± 0.0049 0.0010 ± 0.0070 day–1

Maint. resp. stem, Field3, 4 0.0051 ± 0.0017 0.0076 ± 0.0044 0.0087 ± 0.0056 0.012 ± 0.0025 0.0049 ± 0.0035 day–1

Maint. resp fine roots, Field4 0.0034 ± 0.0011 0.0051 ± 0.0029 0.0058 ± 0.0037 0.0083 ± 0.0017 0.0033 ± 0.0023 day–1

Leaf litterfall rate, Tree5 n.a. n.a. n.a. 0.00049 ± 0.00025 0.00044 ± 0.00019 day–1

N fixation, Tree6 n.a. 0.34 ± 0.23 n.a. n.a. n.a. gN gN–1

SoilC/N ratio microbes7 14 ± 2.3 12 ± 3.5 20 ± 2.9 19 ± 1.8 19 ± 2.9 gC gN–1

Decomposition rate humus 15.3 ± 3.4 20.7 ± 3.1 18.6 ± 3.9 18.6 ± 3.6 16.1 ± 3.2 day–1·10–5

Organic uptake rate litter8 18.5 ± 5.7 13.5 ± 6.0 16.8 ± 6.2 20.4 ± 3.4 17.1 ± 5.5 day–1·10–5

Organic uptake rate humus9 18.5 ± 5.7 13.5 ± 6.0 16.8 ± 6.2 20.4 ± 3.4 17.1 ± 5.5 day–1·10–7

1Parameters allowed to vary according to literature values /Field and Mooney 1986, Thornley and Cannell 1992, Wikström and Ericsson 1995/. 2Plant retention of carbon at leaf abscission in relation to original carbon content in leaf. Fixed relationship between tree and plant according to Ericsson /Ericsson 1994/. 3Same for coarse roots. 4Maintenance respiration in the field layer is distributed between the different plant organs as: leaf = 3; stem = 1.5; root = 1 according to Penning de Vries /Penning de Vries 1975/. 5Constant over the whole season for Norway spruce, only after the onset of dormancy for pine and alder. 6Fraction of total N plant demand. 7The C/N ratio of microbes affects the mineralisation rate of nitrogen. 8The organic uptake of nitrogen represents the uptake of amino acids by mycorrhiza to the plant. 9Assumed to be a hundredth of the organic uptake from litter.

�2

TableA3. Criteriaofacceptance,differentecosystems.Stateandauxiliaryvariables.

Property Grassland Alder Pine Spruce Unit Ref.

Statevariables∆C Soil 0 (–100/100) 0 (–100/100) 0 (–100/100) 0 (–100/100) gC m–2 yr–1 Assumed∆C Tree n.a. 0 (–50/50) 0 (–50/50) 0 (–50/50) gC m–2 yr–1 Assumed∆C Field 0 (–20/20) 0 (–20/20) 0 (–20/20) 0 (–20/20) gC m–2 yr–1 AssumedC Soil 13,000 (11,000/15,000) 18,850 (14,000/23,000) 8,000 (7,000/16,000) 8,000 (7,000/16,000) gC m–2 /Blombäck 1998, Olsson 2000, Kutch et al. 2001b/C Tree n.a. 6,200 (4,000/9,000) 6,570 (3,000/7,000) 9,183 (9,000/9,400) gC m–2 /Johansson 2000, Helmisaari et al. 2002,

Skogsdata 2003/C Field 439 (250/630) 328 (5/660) 95 (60/130) 261 (90/420) gC m–2 /Löfgren 2005/∆N Ecosystem 0 (–5/5) 0 (–5/5) 0 (–5/5) 0 (–5/5) gN m–2 yr–1 AssumedC distribution, Tree:Leaf n.a. 2 (0/12) 4 (0/14) 4 (0/14) % of total /Johansson 2000, Helmisaari et al. 2002,

Uri et al. 2002/Stem n.a. 83 (73/93) 83 (73/93) 83 (73/93) % of total /Johansson 2000, Helmisaari et al. 2002,

Uri et al. 2002/Coarse roots n.a. 11 (1/21) 11 (1/21) 11 (1/21) % of total /Johansson 2000, Helmisaari et al. 2002,

Uri et al. 2002/Fine roots n.a. 4 (0/14) 2 (0/12) 2 (0/12) % of total /Johansson 2000, Helmisaari et al. 2002,

Uri et al. 2002/C distribution, Field:Leaf 55 (45/65) 55 (45/65) 15 (5/25) 20 (10/30) % of total /Nordin et al. 1998, Larcher 2003/Stem 10 (0/20) 10 (0/20) 20 (10/30) 15 (5/25) % of total /Nordin et al. 1998, Larcher 2003/Coarse roots 0 0 60 (50/70) 59 (49/69) % of total /Nordin et al. 1998, Larcher 2003/Roots 35 (25/45) 35 (25/45) 5 (0/15) 6 (0/16) % of total /Nordin et al. 1998, Larcher 2003/Auxiliary variablesC/N ratio soil 18 (15/25) 17 (15/23) 20 (15/25) 20 (15/25) gC gN–1 /Blombäck 1998, Dilly et al. 2000,

Kutch et al. 2001a, Berggren et al. 2004/N retention, Tree1 n.a. 0.25 (0/45) 0.15 (0.01/0.35) 0.4 (0.2/0.6) gN gN–1 /Ericsson 1994/N retention, Field2 0.7 (0.5/0.9) 0.7 (0.5/0.9) 0.7 (0.5/0.9) 0.7 (0.5/0.9) gN gN–1 /Ericsson 1995/LAI Tree n.a. 1.5 (0.5/8) 3.2 (2.2/4.2) 3.6 (2.6/4.6) m2 m–2 /Johansson 2000/ (Tagesson unpubl. data)LAI Field 3 (0.5/4) 2 (1/3) 1 (0.5/2) 1 (0.5/2) m2 m–2 AssumedMax C/N ratio, TreeStem n.a. n.a. n.a. 1,000 (0/1,500) gC gN–1 /Johnson and Lindberg 1992/Needles n.a. n.a. n.a. 90 (0/100) gC gN–1 /Johnson and Lindberg 1992/

1 Retention of nitrogen at leaf abscission.

��

TableA4. Criteriaofacceptance,differentecosystems.Flowvariables.

Property Grassland Alder Pine Spruce Unit Ref.

Organic of tot. N uptake 0.1 (0.01/0.8) 0.1 (0.01/0.8) 0.1 (0.01/0.8) 0.1 (0.01/0.8) gN gN– /Näsholm et al. 1998/N fixation n.a. 6 (4/8.5) n.a. n.a. gN m–2 yr–1 /Dilly et al. 2000/DOC leaching n.a. 12.5 (–25%/+25%) 3.5 (–25%/+25%) 3.5 (–25%/+25%) gC m–2 yr–1 /Canham et al. 2004/Plant resp./tot. resp. 0.5 (0.3/0.7) 0.7 (0.6/0.8) 0.5 (0.3/0.7) 0.5 (0.3/0.7) – /Waring and Running 1998, Kutch et al. 2001b/Heterotrophic/tot. soil resp. 0.70 (0.60/0.80) 0.35 (0.20/0.70) 0.25 (0.10/0.40) 0.25 (0.10/0.40) – /Dilly et al. 2000, Raich and Tufekcioglu 2000,

Kutch et al. 2001b/NPP/GPP 0.50 (0.40/0.60) 0.45 (0.35/0.60) 0.46 (0.25/0.60) 0.46 (0.25/0.60) – /Waring and Running 1998, Kutch et al. 2001b,

Chapin et al. 2002/NPP tree n.a. 950 (400/1,300) 670 (–10%/+10%) 700 (200/1,200) gC m–2 yr–1 /Helmisaari et al. 2002, Schultze et al. 2005/NPP tree above ground n.a. 400 (250/550) n.a. n.a. gC m–2 yr–1 /Kutch et al. 2001b, Uri et al. 2002/NPP field 222 (130/310) 40 (5/78) 25 (10/40) 12 (5/20) gC m–2 yr–1 /Löfgren 2005/Distr. NPP Tree:Leaf n.a. n.a. 14 (4/24%) 14 (4/24) % of NPP tree /Helmisaari et al. 2002/Stem n.a. n.a. 18 (8/28) 18 (8/28) % of NPP tree /Helmisaari et al. 2002/Coarse roots n.a. n.a. 8 (0/18) 8 (0/18) % of NPP tree /Helmisaari et al. 2002/Fine roots n.a. n.a. 60 (50/70) 60 (50/70) % of NPP tree /Helmisaari et al. 2002/Litterfall Tree:Leaf of above ground n.a. n.a. 0.70 (0.40/0.90) 0.70 (0.40/0.90) – /Waring and Running 1998/Above ground n.a. 129 (–10%/+10%) n.a. 70 (60/100) gC m–2 y–1 /Berggren et al. 2004/Leaf C n.a. n.a. 40 (–10%/10%) n.a. gC m–2 y–1 /Flower-Ellis 1985/Leaf N n.a. 6 (–10%/10%) n.a. n.a. gN m–2 y–1 /Uri et al. 2002/Below ground n.a. 135 (–10%/+10%) n.a. 445 (–10%/+10%) gC m–2 y–1 /Berggren et al. 2004/Litterfall Field:Above ground 72 (–25%/+25%) 41 (–25%/+25%) 8 (–25%/+25%) 4 (–25%/+25%) gC m–2 y–1 /Löfgren 2005/Below ground 150 (–25%/+25%) 84 (–25%/+25%) 17 (–25%/+25%) 8 (–25%/+25%) gC m–2 y–1 /Löfgren 2005/

�4

TableA5. Criteriaforacceptancespruce(managed).Onlycriteriadifferentfromspruce(natural).

Property Values Unit Ref.

C Tree f(age): 7 16 26 36 51 71 91 yrC Tree1 0.64 1.32 3.52 5.66 7.47 8.93 9.79 kgC m–2 /Marklund 1988, Skogsdata 2003, Swedish Environmental Protection Agency 2005/C Tree change f(age): 12 21 31 44 61 81 99 yr∆C Tree1 75 221 214 121 73 43 –30 gC m–2 yr–1 /Marklund 1988, Skogsdata 2003, Swedish Environmental Protection Agency 2005/

C dist. Tree: 15 years 35 years 100 years yrLeaf 9 9 4 % of total /Helmisaari et al. 2002/Stem 66 69 83 % of total /Helmisaari et al. 2002/Coarse roots 10 15 11 % of total /Helmisaari et al. 2002/Fine roots 15 7 2 % of total /Helmisaari et al. 2002/Distr. NPP Tree: 15 years 35 years 100 years yrLeaf 15 12 14 % of NPP tree /Helmisaari et al. 2002/Stem 39 26 18 % of NPP tree /Helmisaari et al. 2002/Coarse roots 3 3 8 % of NPP tree /Helmisaari et al. 2002/Fine roots 43 59 60 % of NPP tree /Helmisaari et al. 2002/Litterfall Tree: <30 years >30 years yrAbove ground 40 70 gC m–2 y–1 /Berggren et al. 2004/Below ground 118 445 gC m–2 y–1 /Berggren et al. 2004/

1Calculated from standing stock volume /Skogsdata 2003/ using expansion factors derived from Marklund /Marklund 1988/, according to the Swedish Environmental Protection Agency /Swedish Environmental Protection Agency 2005/.