Quantifying water use by temperate deciduous forests in South Korea : roles of species diversity, canopy structure, and complex terrain

Quan i ying wa e use by empe a e deciduous o es s in Sou h Ko ea: oles o species di e si y, canopy s uc u e, and complex e ain Disse a ion zu E langung des Dok o wü de (D . e . na .) de Fakul ä ü Biologie, Chemie und Geowissenscha en de Uni e si ä Bay eu h on Eunyoung Jung aus Deajeon, Süd-Ko ea Bay eu h, Juni 2013 ! i Abs ac Abou se en y pe cen o Sou h Ko ea is co e ed wi h o es s, mos o which a e ound in he moun ain egions since moun ains ecei e mo e ain all and a e di icul e ains no sui able o ag icul u e. Because moun ains a e impo an wa e sou ces o ci ies and human popula ion downs eam, pe o ming wa e balance o o es ca chmen s has become a esea ch p io i y. The ongoing shi om coni e ous o species- ich deciduous o es s due o a changing go e nmen policy and he an icipa ed changes in u u e clima e, associa ed wi h inc easing amoun o ain all and empe a u e will also impac o es wa e use, calling o an u gen need o unde s and how o es s, in hei cu en s a us, use wa e . The knowledge is i al o p edic ing wa e equi emen s o he u u e o es . The wa m-deciduous empe a e o es s ound in Sou h Ko ea, howe e , ha e a high di e si y o ee species, ha e mul i-laye ed canopies and a e mos ly loca ed on ugged moun ainous e ains, which make i di icul o quan i y o es wa e use, a basic equi emen o ca chmen wa e budge ing. The main objec i es o his s udy we e o: (1) iden i y he oles o species di e si y in ee and o es wa e use, (2) examine he impac o canopy s uc u e on o es anspi a ion, and (3) e alua e he in luence o e ain on o es wa e use. Si e-speci ic s udies we e ca ied ou in h ee di e en na u al deciduous o es s, namely, Gyebang (GB), Gwangneung (GN) and Haean (HA) o es si es, ep esen ing he gene al s uc u e o S. Ko ean o es s. GB si e is known o i s high species di e si y, GN si e is an old o es g ow h a climax, wi h clea ly de ined unde s o y and o e s o y canopy laye s while he HA si e was loca ed wi h in a ca chmen , wi h s ong ele a ion changes wi hin sho ho izon al dis ances, ising om 400 o 1,000 m a.s.l., and in di e en aspec s. Fou loca ions wi h a ying ele a ions and aspec s we e chosen in he HA si e. T ee wa e use (TWU) and canopy anspi a ion (EC) we e es ima ed om sap lux densi y measu ed wi h he mal dissipa ion p obes. Unde s o y anspi a ion (EU) was measu ed using s em hea balance while ecosys em e apo anspi a ion (Eeco) was de e mined using eddy co a iance echnique. Ai empe a u es (Ta), p ecipi a ion, sola adia ion, apo p essu e de ici (VPD), wind speed we e measu ed om wea he s a ions and soil wa e con en was measu ed om equency domain e lec ome y (FDR) senso s a he espec i e s udy si es. Vege a ion su eys, including diame e a b eas heigh (DBH), ee densi y, species composi ion, sapwood a ea (AS), and lea a ea index we e pe o med in all he si es. Canopy conduc ance (GC) and s oma al sensi i i y o VPD we e assessed based on anspi a ion and mic oclima e measu ed a each si e. A unc ional allome ic ela ionship was es ablished be ween AS and DBH, and also be ween TWU and DBH o all he s udy si es; i s o single species and hen combining all he species ei he in a single si e o in all he si es. I espec i e o ee species, AS and maximum TWU we e signi ican ly co ela ed wi h DBH in a powe unc ion o AS (R2 = 0.77, P <0.0001) and bo h in powe (R2 = 0.63, P <0.0001) and sigmoid unc ions (R2 = 0.66, P <0.0001) o TWU, o he co-occu ing species as well as ac oss he si es, ! ii sugges ing ha DBH can be a good p edic o o s and AS and maximum TWU, based on he es ablished allome ic unc ions. Ea ly bud b eak and de elopmen o he unde s o y compa ed o he o e s o y canopy esul ed in an ea lie onse o o es anspi a ion, wi h EU con ibu ing 22% and 14% be ween Ap il and May o he o al o es anspi a ion. This high con ibu ion was a o ed by high adia ion and VPD in he unde s o y, since he o e s o y was s ill unde eloped and open. Despi e diminishing VPD and ligh condi ions in he unde s o y be ween June and Augus , he unde s o y con inued o anspi e a subs an ial amoun o wa e , con ibu ing 10% o he o al anspi a ion. The seasonal pa e ns o bo h EO and EU we e synch onized o canopy de elopmen , while VPD and adia ion de e mined daily ends. EO and EU accoun ed o 80% o Eeco in sp ing bu only 60% du ing he monsoon pe iod due o lowe ed adia ion inpu , VPD, and plan a ea index (PAI). Thus, Eeco is la gely in luenced by anspi a ion a e and i s seasonal a ia ion and also canopy s uc u e. Ea ly sa u a ion o EC a ela i ely low VPD and also a apid dec ease in GC wi h inc easing VPD we e obse ed in he o es s and loca ed a he highes ele a ion s udied (950 m) in he HA si e, compa ed o he GN and he o he o es s ands in HA. These di e ences in anspi a ion a es and s oma al esponse can be explained by g ea e s oma al sensi i i y o VPD o 0.83 ound a he 950 m si e compa ed o 0.63–0.66 in he o he s udy si es. Howe e , he main con olling ac o o he change in s oma al sensi i i y a he 950 m s and is unce ain. Al hough maximum daily EC we e co ela ed wi h AS o he o es s ands a di e en si es (R2 = 0.78, P <0.01), annual EC declined wi h inc easing ele a ion, i.e., 176 >175 >110 >90 mm yea −1 a 340 >450 >650 >950 m, espec i ely. Decline in o al EC was due o he decline in annual Ta, day ime VPD, and leng h o g owing season a highe ele a ions. The GB si e, which was loca ed a 960 m ele a ion, howe e , did no display a same esponse pa e n as hose obse ed a he 950 m si e. I is likely because hese si es we e unde di e en en i onmen al condi ions, i.e., GB si e is exposed o highe Ta and highe humidi y, and is shel e ed (lowe wind speeds). These obse a ions emphasize he complexi y associa ed wi h es ima ion o anspi a ion in ugged e ains, since gene al p inciples do no always apply and he spa ial pa e ns o o es anspi a ion a e complex. Complexi y a ising om mul iple ee species composi ion when es ima ing o es wa e use can be educed by applying unc ional allome ic ela ionship linking ee size and wa e use. Fo es canopy s uc u e and physical loca ion should be aken in o accoun since hey in luence he way o es s use wa e esou ces by al e ing mic oclima e and plan physiology. Based on ou indings, es ima ion o o es wa e use on ugged e ains equi e epea ed measu emen s a ela i ely small spa ial scales since he d i ing ac o s change apidly o e e y na ow e ical dis ances. ! iii Zusammen assung In Südko ea is ca. 70% de Fläche on Wald bedeck , welche sich haup sächlich übe gebi gige und du ch hohe Niede schläge gekennzeichne e, landwi scha lich nich nu zba e Gebie e e s eck . Diese Gebi gs egionen s ellen eine wich ige Wasse essou ce ü die s äd ische Be ölke ung da , sodass die Wasse bilanzie ung on bewalde en Einzugsgebie en zu einem p imä en Fo schungsgegens and gewo den is . Die gegenwä ige Ve lage ung on Nadelwälde n hin zu a en eichen Laubwälde n als Folge eine sich e ände nden S a egie de südko eanischen Regie ung, zu Anpassung an die sich e ände nden klima ischen Bedingungen, wie s eigende Niede schläge und Tempe a u en, üh zu einem e ände en Wasse e b auch de Waldbes ände. Dahe soll e de S a us des gegenwä igen Wasse e b auchs de Bes ände umgehend un e such we den, um Vo he sagen übe den zukün igen Ve b auch e en zu können. Die im wa m-gemäßig en Klima in Südko ea e b ei e en Wälde sind du ch eine hohe A endi e si ä und einen ielschich igen Au bau in ih e S uk u gekennzeichne . Du ch ih e Lage im sch o en, ze klü e en Gelände, ges al e sich die quan i a i e E assung des Wasse e b auchs diese Wälde umso schwie ige . Seine E assung is jedoch ü eine Wasse bilanzie ung in den Waldbes änden eine Vo ausse zung. Das übe geo dne e Ziel diese S udie is (i) zu iden i izie en, welche Rolle die A endi e si ä in Bezug au den Wasse e b auch sowohl einzelne Baumindi iduen als auch des gesam en Waldbes andes spiel , (ii) den Ein luss de K onendachs uk u au die T anspi a ion des Waldbes andes zu un e suchen und (iii) den Ein luss des Geländes au den Wasse e b auch des Waldes zu e aluie en. Fü die Un e suchungen diese S udie wu den d ei, ü Südko ea ep äsen a i e, na ü liche Laubwälde in Gyebang (GB), Gwangneung (GN) und Haean (HA) ausgesuch . Das GB-Waldgebie zeichne sich besonde s du ch eine hohe A en iel al aus. Das GN-Waldgebie bes eh aus eine al en Klimax- Waldgesellscha und läss sich s uk u ell in eine Un e holzschich und eine Baumschich gliede n. Haean (HA) zeichne sich hingegen du ch einen s a ken Höheng adien (400 m bis 1000 m ü. NN) übe ku ze Dis anzen und du ch eine Exposi ion in alle Himmel ich ungen aus, sodass in HA insgesam ie S ando e mi un e schiedlichen Höhenlagen und Exposi ionen ü die Un e suchungen ausgewähl wu den. De Wasse e b auch on Baumindi iduen (TWU) und die K onendach anspi a ion (EC) wu den mi els de Sa lussme hode, welche die Sa lussdich e miss , un e such . Die Un e holz anspi a ion (EU) wu de mi de „s em hea balance“ Me hode (SHB) gemessen. Am S ando GN wu de zu E assung de Ökosys eme apo anspi a ion (Eeco) die Eddy-Ko a ianz-Me hode benu z . Die Ins alla ion on We e s a ionen dien e de E assung on Lu empe a u (Ta), Niede schlag, Sola s ahlung, Wasse damp sä igungsde izi (VPD) und de Windgeschwindigkei . Zusä zlich wu de FDR-Sonden ins allie , um den Bodenwasse gehal zu messen. Zu Un e suchung de Vege a ion wu de de S ammdu chmesse in B us höhe (DBH), die Bes andesdich e, die Splin holz läche (AS), de ! i Bla lächenindex und die A enzusammense zung an allen S ando en e ass . Basie end au de T anspi a ion und dem Mik oklima wu de die K onendachlei ähigkei (GC) und die s oma ä e Emp indlichkei bezüglich des VPD un e such . Nich nu die Be ücksich igung einzelne A en, sonde n auch die Einbeziehung alle A en an einem S ando sowie alle S ando e e gab eine unk ionale allome ische Beziehung, sowohl zwischen As und DBH, als auch zwischen TWU und DBH. Unabhängig on de jeweiligen Bauma zeig en die Analysen einen signi ikan en Zusammenhang zwischen As und BHD, ausged ück in eine Po enz unk ion (P <0.0001). Auch de maximale TWU ko elie e signi ikan mi dem DBH im Sinne on Po enz- und sigmoidalen Funk ionen (P <0.0001), sowohl ü einzelne A en als auch s ando übe g ei end. Au g und diese E gebnisse is de DBH un e Be ücksich igung de einge üh en allome ischen Funk ionen ein gu e Schä zg öße ü die Bes andessplin holz läche As und dem maximalen Wasse e b auch (TWU). De im Ve gleich zu Baumschich ühe Bla aus ieb im Un e holz üh e zu einem e höh en An eil de Un e holz anspi a ion on 22% im Ap il und 17% im Mai an de Gesam anspi a ion des Waldbes andes. Diese hohe An eil wu de du ch die hohe Eins ahlung und das hohe Wasse damp sä igungsde izi (VPD) im Un e holz au g und de lich en Baumschich begüns ig . Alle dings ug das Un e holz auch on Juni bis Augus mi einem be äch lichen An eil on 10% zu Gesam anspi a ion bei, obwohl die sola e Eins ahlung und das VPD du ch die en wickel e Baumschich ge inge wa en. De saisonale Ve lau on EO und EU e lie synch on zu En wicklung des K onendaches, wäh end hingegen de Tagesgang du ch das Mik oklima, die sola e Eins ahlung und das VPD ges eue wu de. Im F ühling ugen EO und EU mi einem An eil on 80% zu gesam en Eeco bei. Wäh end de Monsunpe iode e inge e sich diese An eil jedoch au g und eine ge inge en Eins ahlung, einem ge inge em VPD und einem minimie en P lanzen lächenindex (PAI) au 60%. Die Eeco wi d dahe s a k du ch den saisonalen Ve lau de T anspi a ions a e als auch du ch die K onendachs uk u beein luss . Im Ve gleich zu GN, GB, und den nied ige en S ando en im HA-Einzugsgebie konn e eine ühe Sä igung on Ec bei ela i ge ingem VPD und auch eine ela i s a ke Abnahme on GC bei s eigendem VPD am höchs en S ando in 950 m ü.NN in HA beobach e we den. Die un e schiedlichen T anspi a ions a en und die s oma ä e Reak ion können du ch eine g öße e s oma ä e Emp indlichkei in Bezug au ein VPD on 0.82 e klä we den. Im Ve gleich dazu wiesen die ande en S ando e nu ein VPD zwischen 0.63−0.66 au . Dennoch sind die e u sachenden Fak o en ü eine Ve ände ung de s oma ä en Emp indlichkei in g öße en Höhenlagen mi gewisse Unsiche hei beha e . Obwohl die maximale ägliche EC mi AS (R²=0.78, P <0.01) de Waldbes ände an e schiedenen S ando en ko elie e, konn e eine Abnahme de jäh lichen Ec, mi 176 > 175 > 110 > 90 mm Jah −1, bezüglich de Höhenlage, 340 > 450 > 650 > 950 m ü.NN beobach e we den. Eine Abnahme de gesam en EC läß sich au die Abnahme de Jah esdu chschni s empe a u , des ageszei lichen VPD und de Länge de Vege a ionspe iode in g öße en Höhenlagen zu ück üh en. Die E gebnisse des S ando es GB, welche sich in ähnliche Höhenlage au 960 ! m be and, zeig en jedoch im Ve gleich zum HA-S ando in 950 m ü.NN un e schiedliche Reak ionsmus e , welche au die geschü z e Lage mi höhe en Tempe a u en, höhe e Lu euch igkei und ge inge en Windgeschwindigkei en zu ückzu üh en sind. Diese Beobach ungen un e s eichen die Schwie igkei en bei de Schä zung de Bes andes anspi a ion im ze klü e en, s eilem Gelände, da au g und de komplexen äumlichen Mus e de Waldbes and anspi a ion nich imme allgemeingül ige P inzipien abzulei en sind. Du ch die mul iple Zusammense zung de Bauma en in einem Bes and ges al e sich die Schä zung des Wasse e b auchs als schwie ig, dennoch können diese Schwie igkei en du ch die Anwendung unk ionale allome ische Beziehungen zwischen dem Baumum ang und dem Wasse e b auch eduzie we den. Insbesonde e soll e die K onendachss uk u und die geog a ische Lage aus eichend be ücksich ig we den, da diese Fak o en den Wasse e b auch de Waldbes ände du ch die Ve ände ung des Mik oklimas und de P lanzenphysiologie beein lussen. In Anbe ach de E gebnisse soll e die Schä zung des Wasse e b auchs de Bes ände anhand wiede hol e Messungen au ela i klein äumige Skala e olgen, da sich die eibenden Fak o en schnell und au ela i ku ze Dis anz ände n können. ! i Acknowledgemen s Fi s o all, I would like o hank PD D . Dennis O ieno, Depa men o Plan Ecology, Uni e si y o Bay eu h, o his guidance on how o be a scien is du ing my Ph. D. He is no only he bes eache in my whole li e bu also he bes iend who I was able o discuss wi h abou any hing. I am e y hono ed o be he i s Ph. D. g adua e unde him in Ge many. I also hank P o . D . John Tenhunen, Depa men o Plan Ecology, Uni e si y o Bay eu h, o gi ing me a lo o oppo uni ies o lea n and o see mo e in he wo ld. I admi e his en husiasm on he esea ch, posi i e a i ude in li e, and wise leade ship. I would like o hank D . Hyojung Kwon, O egon S a e Uni e si y, o he encou agemen and hough ul commen s on analyzing he da a, w i ing he manusc ip s, and also ha ing a meaning ul li e. I lea ned a lo om he h ough uncoun able mee ings and discussions while she s ayed in Bay eu h. I am g a e ul o Ma ga e e Wa inge o he excellen suppo on p epa a ion o he ield wo k in Ko ea and A ica. I belie e ha I could no accomplish all he ield wo k wi hou he help in sol ing unexpec ed echnical p oblems. I also hank Fiede ike Ro he, Bä bel Heindl-Tenhunen, and Sand a Thomas o hei emendous ca es o me o ha e a pleasan s ay in Bay eu h. I hank all my colleagues o he TERRECO and KiLi p ojec s, who ga e me inspi a ion o he possible u u e esea ch and enjoyable li e in Bay eu h, Haean, and Nkweseko. Special hanks o S e e Linde o aking ca e o me in e e y possible ways, o Bo a Lee o jus being he e o me, o Bumsuk Seo o s imula ing me o s udy ou subjec in mo e dep h, Saem Lee o lis ening o me a en i ely all he ime, Ma ianne o he help on Ge man summa y, Emily o English co ec ions, Thomas o Ge man co ec ions, and Sina o encou aging me and eaching me p omising spi i o he . I would like o hank all my iends who I spen un o ge able ime oge he in Bay eu h: Family Gä di z, Family Jeoung, Family Made , Family O o, Family Pa k Hoseon, F ank, Hee a, Yangmin, and Yoolim. Finally, I would like o exp ess my special hanks o my husband Jae-Woo and my pa en s o hei endless suppo , us , pa ience, and lo e. ! ii Table o con en s Abs ac ……………………………………………………………………………………………………..i Zusammen assung………………………………………………………………………………………...iii Acknowledgemen s……………………………………………………………………………………….. i Table o con en s…………………………………………………………………………………………. ii Lis o igu es………………………………………………………………………………………………x Lis o ables……………………………………………………………………………………………..xiii Lis o abb e ia ions and symbols………………………………………………………………………..x 1 De ailed summa y.………………………………………………………………………………1 1.1 Gene al in oduc ion and li e a u e e iew.…………………………………………………………..1 1.1.1 Tempe a e deciduous o es s: dis ibu ion and s uc u e.…………………………………………….…1 1.1.2 Fo es s in Sou h Ko ea…………………………………………………………………………………..2 1.1.2.1 His o y o o es s in Sou h Ko ea………………………………………………………..2 1.1.2.2 Cu en s a us o o es s in Sou h Ko ea………………………………………………....2 1.1.2.3 The u u e o o es s in Sou h Ko ea…………………………………………………….3 1.1.3 Regula ion o wa e use by o es ecosys ems…………………………………………………………..4 1.1.4 Es ima ion o wa e use by o es s in Sou h Ko ea………………………………………………….…..5 1.1.5 S a emen o esea ch challenges………………………………………………………………………..6 1.1.6 Objec i es o he esea ch……………………………………………………………………………….7 1.2 Gene al ma e ials and me hods……………………………………………………………………...10 1.2.1 Desc ip ion o s udy si es……………………………………………………………………...10 1.2.2 Me hods………………………………………………………………………………………..15 1.2.2.1 Sap low measu emen s………………………………………………………………...15 1.2.2.2 Canopy conduc ance……………………………………………………………………18 1.2.2.3 Biome ic measu emen s……………………………………………………………….19 1.2.2.4 Mic ome eo ological measu emen s……………………………………………………20 1.2.2.5 Eddy co a iance lux measu emen s…………………………………………….……..20 1.2.2.6 Wa e use e iciency……………………………………………………………………20 1.3 Gene al esul s and discussions……………………………………………………………………...21 1.3.1 T ee and o es wa e use in di e se species composi ion ………………………………………….….21 1.3.2 Impac o he o es s uc u e on o es wa e use……………………………………………………...25 1.3.3 Fo es wa e use in he complex e ain………………………………………………………………..26 1.4 Gene al conclusions…………………………………………………………………………………30 1.5 Lis o manusc ip s and speci ica ion o con ibu ions……………………………………………...31 1.6 Re e ences…………………………………………………………………………………………...32 2 Up-scaling o s and anspi a ion o an Asian empe a e mixed-deciduous o es om single ee sap low measu emen s……………………………………………………………41 Abs ac …………………………………………………………………………………………………...41 2.1 In oduc ion………………………………………………………………………………………….42 2.2 Ma e ials and me hods………………………………………………………………………………43 2.2.1 S udy si e…………………………………………………………………………………………….…43 2.2.2 Vege a ion.…………………………………… ……………………………………………………….43 2.2.3 Mic ome eo ology……………………………………………………………………………………...44 2.2.4 T ee allome ics………………………………………………………………………………………...45 ! xi Table 4.4 Annually a e aged mic oclima es (sola adia ion (RS), ai empe a u e (Ta), mean day ime apo p essu e de ici (VPD), annual ain all and wind speed) and soil wa e con en (θ) du ing g owing season o he s udy si es in Haean ca chmen , Sou h Ko ea in 2010. ± a e s anda d de ia ion (SD). Table 4.5 Maximum lea a ea index (LAI), leng h o he g owing season, o al canopy anspi a ion (EC, mm) du ing he g owing season and mean ing wid h o Que cus mongolica a each s udy si e. ± a e s anda d de ia ion (SD). Table 4.6 Canopy conduc ance (GC, mm s−1) on clea days in June, i s espec i e apo p essu e de ici (VPD, kPa) o day ime ( om 8:00h o 18:00h) and GC e (GC a VPD = 1 kPa) a each si e. ± a e s anda d de ia ion (SD). ! x Lis o abb e ia ions and Symbols Abb e ia ion/Symbol De ini ion Uni AIC Akaike’s in o ma ion c i e ion - AS sapwood a ea [m2] BA basal a ea [m2] cp speci ic hea a cons an p essu e [J kg−1 K−1] DBH diame e a b eas heigh [cm] EC canopy anspi a ion [mm h−1, mm d−1] Eeco ecosys em e apo anspi a ion [mm h−1, mm d−1] Emax maximum s and anspi a ion [mm h−1, mm d−1] EO o e s o y anspi a ion [mm h−1, mm d−1] EU unde s o y anspi a ion [mm h−1, mm d−1] Fd sap lux densi y [g m−2 s−1] GA ae odynamic conduc ance [m s−1] gb bounda y laye conduc ance [m s−1] GC canopy conduc ance [mm s−1] GC e canopy conduc ance a VPD = 1 kPa [mm s−1] g u bulen conduc ance [m s−1] GV gas conduc ance o wa e apo [m3 kPa kg−1 K−1] KoFlux Ko ean Flux g oup - LAI lea a ea index [-] LAImax maximum lea a ea index [-] LAIU unde s o y lea a ea index [-] MLT modi ied lookup able - PAI plan a ea index [-] PAR pho osyn he ic ac i e adia ion [mol m−2 s−1] PPFD pho osyn he ic pho on lux densi y [mol m−2 s−1] ! x i RMSE Roo -mean-squa e-e o [-] RN ne adia ion [W m−2] RS sola adia ion [W m−2] SHB s em hea balance - SLA speci ic lea a ea [cm2 g−1] SWAT Soil and Wa e Assessmen Tool - Ta ai empe a u e [°C] TDP he mal dissipa ion p obes - Tk ai empe a u e in kel in [K] U wind speed abo e he ege a ion laye [m s−1] VPD apo p essu e de ici [kPa] WUE wa e use e iciency - z canopy heigh [m] z0 oughness leng h [m] α a enua ion coe icien o wind speed inside he canopy [-] Δ change o sa u a ion wa e apo p essu e wi h empe a u e [Pa K−1] γ psychome ic cons an [Pa K−1] θ soil wa e con en [%] κ on Ka man cons an [-] λ la en hea o apo iza ion o wa e [J kg−1] ρ densi y o d y ai [kg m−3] ρw densi y o wa e [kg m−3] ! Chap e 1 – De ailed summa y ! 1 Chap e 1 De ailed summa y 1.1 Gene al in oduc ion and li e a u e e iew 1.1.1 Tempe a e deciduous o es s: dis ibu ion and s uc u e Tempe a e deciduous o es s a e widely dis ibu ed ac oss he globe, co e ing an a ea o app oxima ely 7.8 million km2 wo ldwide (Allaby 2006). They occu in Eas e n No h Ame ica, wes e n and cen al Eu ope, eas e n Asia, Nea Eas , and in pa s o Sou h Ame ica such as Pa agonia and Chile. These o es s occu in ela i ely wa m- empe a e mois clima es, wi h an a e age empe a u e o he coldes mon h anging be ween −18 and −3°C and he wa mes mon h anging be ween 18 and 30°C (Röh ig 1991a; Allaby 2006). The mean annual p ecipi a ion anges om 750 o 1,500 mm, and is alloca ed e enly h oughou he season o mos o he egions excep o eas e n Asia, which expe iences se e e ain s o ms du ing summe (Röh ig 1991a; Allaby 2006). Compa ed o he o es s in Eu ope and No h Ame ica, he empe a e deciduous o es s o eas e n Asia a e wo o h ee imes highe in plan species di e si y (La ham and Rickle s 1993; Qian and Rickle s 1999). His o ically, his subs an ial di e ence in plan di e si y appea s o esul om g ea e physiog aphic he e ogenei y in Asia, which allowed o allopa ic specia ion in esponse o clima e and sea le el luc ua ions a e empe a e o es zones became disjunc in he la e Te ia y (Qian and Rickle s 2000). Mo eo e , epea ed glacia ion du ing he Pleis ocene was mo e ex eme in Eu ope and No h Ame ica han in eas e n Asia (Qian and Rickle s 2000). In eas e n No h Ame ica, mo e uni o m clima e and simple geog aphy ha e no os e ed e olu iona y condi ions among he same amilies as in eas e n Asia (Qian and Rickle s 2000). In he case o Eu ope, e en lowe di e si y is ound in o es ege a ion due o he es ic ions on e ugia o mig a ing species ha we e imposed by he Alps ba ie and he loca ion o he Medi e anean Sea du ing Pleis ocene glacia ions (Ellenbe g 1978), esul ing in la ge numbe s o ex inc ions. Gene ally, he empe a e deciduous o es s o eas e n Asia can be classi ied in o wo main g oups, namely cool- and wa m- empe a e deciduous o es s (Ki a 1991). The cool- empe a e deciduous o es s mainly inhabi wes e n Japan, while he wa m- empe a e deciduous o es s occu in o he pa s o Japan, China and Ko ean Peninsula (Nakashizuka and Iida 1995). Cha ac e is ically, he cool- empe a e deciduous o es s a e domina ed by Fagus c ena a, which accoun s o mo e han 80% o he o al o es basal a ea (Nakashizuka 1987). Wa m- empe a e deciduous o es s, on he o he hand, a e species- ich, o example, Chap e 1 – De ailed summa y ! 2 he dominan genus Que cus occu as 66 di e en species (Röh ig 1991b). This high species di e si y likely esul ed om he wa m and humid summe condi ions associa ed wi h high adia ion and mois u e inpu s om he monsoon (Röh ig 1991a). Dominan gene a include Que cus, Ca pinus, Ulmus, and Tilia (Velichko and Spasskaya 2002). Thus, he wa m- empe a e deciduous o es s o Ko ea and o he pa s o Asia ep esen an impo an ecosys em ype wi h mul i-laye ed physiognomy (Kim 2002), whe e bo h he o e s o y and unde s o y a e well de eloped and display di e se species composi ions. As a esul , la ge di e ences a e expec ed in e ms o wa e use be ween he cool-and wa m- empe a e deciduous o es s. 1.1.2 Fo es s in Sou h Ko ea 1.1.2.1 His o y o o es s in Sou h Ko ea Sou h Ko ea is a moun ainous coun y, wi h 70% o i s land co e ed by moun ains wi h ele a ions o up o 2,000 m a.s.l. sepa a ed by deep and na ow alleys. Two hi ds o he land a ea is, he e o e, di icul e ain ha is no sui able o ag icul u e and is, ins ead, unde o es co e . Mos s eams o igina e om he high ele a ion o es s, and supply mos o he wa e equi emen s o he popula ion downslope. Du ing he colonial pe iod o 1910 o 1945 and he Ko ean Wa o 1950 o 1953, o es s we e excessi ely de as a ed, which led o equen loods and landslides. In o de o ehabili a e o es s, he go e nmen ini ia ed a na ion-wide, la ge-scale e o es a ion p og am om 1973, wi h a goal o es o ing 1 million ha o o es s wi hin a sho pe iod by plan ing as g owing ee species, mos ly coni e s (Pinus ko aiensis, Abies holophylla and La ix lep olepis) and es ic ing he bu ning o o es s o c ea e land o cul i a ion. Th ough his ac ion, abou 730 housand ha o deg aded land was es o ed h ough an ex ensi e plan a ion o abou 10 billion ees on an a ea o o e 350 housand ha (Lee e al. 1997; Ko ea Fo es Se ice 2009). 1.1.2.2 Cu en s a us o o es s in Sou h Ko ea In he ecen pas , o es co e age in Sou h Ko ea has been declining a an ala ming a e o 40 housand ha pe yea (Ko ea Fo es Se ice 2006), due o con e sion in o ag icul u al land, u baniza ion and expansion o he manu ac u ing indus y (Youn e al. 2009). The p opo ional co e age o coni e ous and deciduous o es s has changed as well. Coni e ous o es s ha e declined om 55% o 42% (600 housand ha), while he a ea co e ed wi h deciduous o es s has inc eased om 17% o 40% (540 housand ha), be ween 1972 and 2008 (Ko ea Fo es Se ice 2009). Al hough o al o es a ea has declined, he g owing s ock o ees has inc eased 11 imes du ing he las 40 yea s (i.e., g owing s ock was 70 million m3 in 1972 and inc eased o 800 million m3 in 2010), as a esul o na u al e-g ow h and imp o ed o es managemen p ac ices (Ko ea Fo es Se ice 2011). A subs an ial p opo ion o he o es s is a ea ly and mid succession s ages. Chap e 1 – De ailed summa y ! 3 Acco ding o he Ko ea Fo es Se ice epo o 2009, o es s unde 10 yea s and o e 51 yea s o age co e 7% and 2% o he o al o es a ea, espec i ely. Mos (close o 70%) o he o es a ea is occupied by ees aged be ween 20 and 40 yea s. Oaks occupy 75% o he a ea unde na u al deciduous o es s, wi h Q. mongolica being one o he dominan species g owing om 100 m o 1,800 m a.s.l., bu mos ly abundan a a ound 700 m a.s.l. (Chung and Lee 1965). 1.1.2.3 The u u e o o es s in Sou h Ko ea The clima e o Sou h Ko ea has expe ienced a g adual wa ming du ing he 20 h cen u y (Oh e al. 2004; Na ional Ins i u e o En i onmen al Resea ch 2011). In he las cen u y alone, he a e age ai empe a u e inc eased by abou 1.5°C, which is wice as high compa ed o he global wa ming p ojec ions. This has been a ibu ed o apid indus ializa ion (Oh e al. 2004; Kwon 2005). This empe a u e inc ease is con ibu ing o he cu en shi s in plan species dis ibu ion anges. Fo example, bamboos (Phyllos achys) ha e shi ed om 35°N o 36°N in he con inen al egion and 38°N in he eas e n coas al egion du ing he las 200 yea s (Gong 2001; Oh e al. 2004). Also, he ee line o Ko ean i (Abies ko eana), which g ows only a high ele a ions, has been con inuously mo ing upwa ds and he species is now h ea ened wi h high mo ali y a e o 20–50% in i s na u al ange (Lim e al. 2008). Based on he high- esolu ion clima e simula ions o 2021 o 2050, unde he B2 scena io (IPCC 2000), wa ming in he ange o 1–4 °C is expec ed in he no he n pa o S. Ko ea du ing he cold season (Im e al. 2008). P ecipi a ion will also be egionally a iable, wi h inc eased summe ains in he no h, bu a decline in he sou he n egions (Im e al. 2008). Based on hese clima e p ojec ions, deciduous and mixed o es s could inc ease by 60% and 10%, espec i ely, while coni e ous o es co e could dec ease by 10% by he yea 2080 (Shin e al. 2012). The p ojec ed changes in clima e and o es s uc u e s imula e in e es s in assessing he ongoing changes and how hey will in luence o es wa e budge in he sho - and long- e m since mos o he coun y’s wa e equi emen is me by wa e om o es ed moun ains. Al eady, hyd ological simula ions wi h he SWAT (Soil and Wa e Assessmen Tool) model, using clima e scena ios o ca. 4°C and 20–35% p ecipi a ion inc ease, espec i ely, show ha an e apo anspi a ion (Eeco) inc ease o 15–20% is expec ed be ween 2000 and 2080 (Pa k e al. 2011). Mos o es s a e loca ed in moun ains and hey a e pe cei ed as wa e ese oi s o ag icul u al lands and popula ions downs eam. His o ically, ule s o he ancien Ko ean kingdoms ga e a high p io i y o o es p o ec ion since hey conside ed he wa e egula ion unc ion o o es s as a undamen al se ice o ag icul u e (Youn e al. 2009), a concep ha is widesp ead among Sou h Ko eans o da e. Fo es soils pe o m as ese oi s o wa e om p ecipi a ion and losing wa e by uno and Eeco. Any ege a ion changes s ongly in luence Eeco, which necessa ily a ec s uno , because Eeco is one o he la ge componen s o he o es hyd ologic budge . Fo example, coni e o es s p oduce less uno han deciduous o es s due o hei highe a es and longe season o Eeco (Swank Chap e 1 – De ailed summa y ! 4 and Douglass 1974). Unde s anding how o es ca chmen s s o e ain wa e and also he egula ion o wa e elease om o es s, ei he as i e discha ge o h ough Eeco is, he e o e, c i ical o he managemen o na u al wa e esou ces (Chapin e al. 2011). 1.1.3 Regula ion o wa e use by o es ecosys ems In mos o es ecosys ems, canopy anspi a ion is de e mined by he p e ailing mic oclima e, soil mois u e s a us and he plan cha ac e is ics (Kö ne 1994; Schulze e al. 2005). A g adien in apo p essu e be ween he in e cellula spaces and he su ounding ai ou side he lea su ace de e mines he a e o wa e ans e om he lea in o he a mosphe e and consequen ly, anspi a ion a e: as long as he s oma a emain open. Unde a o able soil mois u e condi ions, ligh in ensi y con ols s oma al opening and s oma al conduc ance inc eases wi h highe ligh in ensi ies (Schulze e al. 2005). On a daily basis, du ing ample soil mois u e a ailabili y, he e o e, canopy anspi a ion inc eases exponen ially wi h inc easing apo p essu e de ici (VPD), as long as he p e ailing pho osyn he ic pho on lux densi y (PPFD) is high enough o allow o ull s oma al opening (G anie and B éda 1996; O en and Pa aki 2001; Ewe s e al. 2002). A highe VPD inc eases in anspi a ion a e, howe e , egula ed by he s oma a such ha he species-maximum capaci y o hyd aulic conduc i i y is no su passed and ca i a ion does no occu . Fo example, sa u a ion o daily s and anspi a ion o Eu opean beech o es s ands occu ed a mean daily VPD o 2.5–3.0 kPa, while o a sp uce s and, anspi a ion sa u a ed a VPD o 2.0–2.5 kPa (Kös ne 2001). Simila ly, mo e p onounced s oma al closu e in esponse o inc easing VPD (>2.5 kPa) in Eu opean beech (F. syl a ica) han in sessile oak (Q. pe aea) has been epo ed, e ealing la ge sensi i i y o beech o VPD changes (A anda e al. 2000). S oma al closu e occu s o p e en he de elopmen o dange ously low wa e po en ials, which can cause ca i a ion, and o p o ec he conduc ing essels (Jones and Su he land 1991). Du ing a ully de eloped canopy s age in deciduous o es s, soil wa e a ailabili y egula es he po en ial anspi a ion a es, while PPFD and VPD con ol diu nal pa e ns o anspi a ion wa e loss (Kö ne 1994). When soil mois u e is limi ed, lea wa e po en ial declines. Unde such condi ions, s oma al closu e will occu in o de o con ol anspi a ion wa e loss (Cocha d e al. 1996). In Eu opean and No h Ame ican empe a e o es s, se e al s udies ound ha he c i ical alue o ela i e ex ac able wa e om he soil was abou 0.4, which is calcula ed as he a io o ex ac able wa e (a ailable soil wa e – minimum soil wa e ) and maximum ex ac able wa e (soil wa e con en a ield capaci y – minimum soil wa e ), a which soil wa e con en begins o limi maximum anspi a ion (Black 1979; G anie 1987; G anie e al. 1999; Wilson and Baldocchi 2000). While en i onmen al d i e s con ol o es s and anspi a ion in a ela i ely sho pe iod o ime, o es s uc u e, i.e., numbe and size o he ees, age and species composi ion, egula es s and anspi a ion o e Chap e 1 – De ailed summa y ! 5 longe ime scales. T anspi a ion om monocul u al o es s ands is likely o be dependen on he o al sapwood a ea (AS) o he s and o lea a ea index (LAI) (Ewe s e al. 2002; Wullschlege e al. 2001). Fo example, Zimme mann e al. (2000) ound ha canopy anspi a ion was co ela ed wi h s and AS which was de e mined by s and densi y and ee AS in pine o es monocul u e s ands, wi h di e se ages anging om 28 o 383 yea s o age. Kös ne (2001) showed an inc easing pa e n o he maximum s and anspi a ion wi h LAI o i e Eu opean beech o es s, ega dless o he s and age. In mixed o es s ands, howe e , di e ences in species composi ion may modi y a es o anspi a ion, since species di e in wa e - esou ce acquisi ion, xylem ana omy, and phenology. Fo example, ee species wi h deep oo ing sys ems like Q. alba show highe a es o anspi a ion as he soil d ies, compa ed o shallow oo ed ones such as Ace ub um (O en and Pa aki 2001; Bo a d e al. 2005). O en e al. (1999) showed ha he s oma al conduc ance o ing-po ous species was less sensi i e o a ia ions in ligh and VPD han di use-po ous species. This hen esul ed in lowe mean canopy conduc ance and lowe maximum canopy anspi a ion o he o es s ands composed o a high p opo ion o ing-po ous species han di use-po ous species (O en and Pa aki 2001). The iming o lea lushing and senescence also a y among deciduous species (Vi asse e al. 2009), which can ha e an impac on annual canopy anspi a ion, since i de e mines he pe iod o ac i e lea anspi a ion. Complexi y in e ain, such as along moun ain slopes and alleys o exposi ion inc eases he complexi y in pa e ns o ee anspi a ion and he quan i ies o wa e used by o es s ands. A highe ele a ions anspi a ion a es a e likely o dec ease because o lowe ai empe a u es (Ta), highe humidi y (highe ain all equencies and amoun s) and lowe VPD (Kubo a e al. 2005; Kö ne e al. 2007; Kumagai e al. 2008; McDowell e al. 2008; Ma yssek e al. 2009). Soil cha ac e is ics also change wi h ele a ion, wi h ela i ely shallow soils ound a highe ele a ions compa ed o down slope due o inc eased uno and e osion a highe ele a ions and deposi ion a lowe ele a ions (Hi obe e al. 1998; Ta eno e al. 2004; T omp- an Mee eld and McDonnell 2006). Sou h- acing aspec s ecei e highe sola adia ion, which esul s in wa me and d ie condi ions on sou h han no h- acing aspec s (Van de Wa e e al. 2002). Spa ial a iabili y in mic oclima e and soil p ope ies a di e en ele a ions in complex e ains a e, he e o e, likely o gene a e he e ogeneous anspi a ion a es de ined by he complex in e ac ions o s and s uc u e, edaphic and he p e ailing mic oclima ic condi ions abo e he o es s ands. 1.1.4 Es ima ion o wa e use by o es s in Sou h Ko ea Using di e en app oaches, a numbe o s udies ha e a emp ed o asce ain wa e budge s o isola ed o es s in S. Ko ea. Kim and Woo (1988) and Lee e al. (1989) epo ed ha di ec in e cep ion wa e loss by he o es canopy in he plan ed coni e ous o es s was 15–20% highe compa ed o na u al deciduous Chap e 1 – De ailed summa y ! 6 o es s by measu ing h ough all and s em low unde he canopy. In compa ison, Eeco was highe in coni e ous o es s han deciduous o es s g owing oge he a simila ele a ions, based on he calcula ions using he Tho n hwai e me hod (Kim 1987). Since 2001, he Ko ean Flux g oup (KoFlux) has been assessing Eeco in di e en o es ypes using he eddy co a iance echnique in an a emp o ob ain an o e all wa e budge o he S. Ko ean o es s (Kim e al. 2006; Kang e al. 2009; Kang e al. 2012). Kang e al. (2009) showed a cha ac e is ic seasonali y, wi h mid-season dep essions in Eeco ha a e associa ed wi h he educed amoun o a ailable ene gy du ing he monsoon season. The applica ion o he eddy co a iance echnique in he es ima ion o o es wa e use in S. Ko ea is, howe e , challenging and ques ions a e aised ega ding da a accu acy, since mos o he o es s a e loca ed in moun ainous landscapes ha a e highly he e ogeneous and complex. One o he main assump ions o he eddy echnique is a la and homogeneous e ch o oo p in measu emen si es (Baldocchi e al. 1988), which is no me in mos o hese o es s ands. Fo es anspi a ion can be measu ed by sap low echniques a ela i ely high empo al scales (Wullschlege e al. 1998). Since hey apply a single ee le el, he echniques a e no limi ed by e ain complexi y and ee species di e si y (Wilson e al. 2001; Kumagai e al. 2008). Fo es s and anspi a ion can be de e mined by summing up he alues o anspi a ion by e e y single ee in he s and mul iplied by i s espec i e AS. A easonable scaling p ocess om ee o s and le el equi es accu a e es ima es o wa e use om a limi ed numbe o ep esen a i e ees wi hin he s and (Kumagai e al. 2008). The sap low measu emen s a e, he e o e, sui able o es ima ion o wa e use by o es s ands, he analysis o species e ec s on o es wa e use and o pa i ioning Eeco in o anspi a ion and e apo a ion (Wilson e al. 2001; Fo d e al. 2007). In S. Ko ea, sap low measu emen s ha e been used o es ima e ee anspi a ion o majo species such as Que cus mongolica and La ix lep olepis (Han and Kim 1993; Han and Kim 1996), bu no o es ima e o es s and anspi a ion so a . Compa ed o Eu ope and No h Ame ica whe e signi ican esea ch in he empe a e o es s has been conduc ed, knowledge on he s uc u e and unc ion o he empe a e o es s in Asia is s ill lagging behind, bo h a egional and local scales. Thus, mo e s udies a e needed in o de o ill he gaps in knowledge, which will allow o a mo e in o med and sus ainable o es managemen . 1.1.5 S a emen o esea ch challenges Na u al egene a ion o o es s in Sou h Ko ea has a o ed he expansion o he species- ich deciduous o es s o e coni e s, which we e massi ely plan ed in he p e ious es o a ion p og ams. This shi in o es composi ion is likely o change he hyd ology o mos S. Ko ean o es s in a way ha is no ye well unde s ood. Clima e p ojec ions o S. Ko ea (Im e al. 2008) also show u u e changes in ain all pa e ns and amoun s and inc eases in Ta, which a e likely o signi ican ly impac wa e use by o es s. These o es s Chap e 1 – De ailed summa y ! 13 Table 1.2 S and s uc u e o he six na u al deciduous o es si es in Gyebangsan (GB), Gwangneung (GN), and Haean (HA). LAI, BA, AS and DBH indica e lea a ea index, basal a ea, sapwood a ea and diame e a b eas heigh (DBH). T ees wi h diame e a b eas heigh (DBH) ≥5 cm and ees wi h 2 cm ≤ DBH <5 cm a e selec ed o he o e s o y (O/S) and he unde s o y (U/S), espec i ely. S and densi y and DBH-based da a a e in e pola ed om he in en o y conduc ed in 2008 o GB and GN, and in 2010 o HA. GB GN HA 450N 650N 650S 950N S and age [y s] ca. 50 ca. 200 ca. 30 ca. 30 ca. 30 ca. 20 LAI No measu ed 4.3 4.9 5.3 6.3 5.7 T ee densi y [ ees ha-1] O/S 1,025 63 1,252 1,640 2,523 4,350 U/S 1,730 1,035 5,165 960 3,969 15,050 BA [m2 ha-1] O/S 24.0 38.5 21.8 17.5 20.5 22.4 U/S No measu ed 0.9 2.3 0.8 2.5 1.0 S and AS [m2 ha-1] O/S 5.7 16.5 12.5 13 15.5 10.5 A e age DBH [cm] O/S 13.3 25.7 10.8 10.8 9.4 6.2 Max DBH [cm] O/S 40.7 64.0 28.4 33.5 23.5 9.9 Max ee heigh [m] 15 20 10 12 10 5 Species composi ion (BA co e [%]) O/S Tilia amu ensis (31) Ulmus da idiana (14) Que cus mongolica (12) Ace mono (8) Ace pseudo- sieboldianum (5) Maackia amu ensis (4) Co nus con o e sa (4) Que cus se a a (71) Ca pinus laxi lo a (22) Ca pinus co da a (5) Que cus mongolica (24) Alnus sibi ica (18) Que cus aliena (15) Que cus se a a (15) Ulmus lacinia a (10) Que cus den a e (8) Tilia mandshu ica (6) Qu cus den a e (65) Be ula da u ica (19) Que cus mongolica (14) Que cus mongolica (50) Tilia mandshu ica (25) Que cus den a e (14) F axinus hynchophylla (4) Que cus se a a (4) Que cus mongolica (72) F axinus hynchophylla (13) Euonymus hamil onianus (9) U/S No measu ed Euonymus oxyphyllus Cel is jessoensis So bus alni olia S y ax obassia Euonymus ala us Rhododend on yedoense Rhododend on schlippenbachii Qu cus den a e S ephanand a incisa Que cus mongolica Ace pseudo- sieboldianum Euonymus ala us Ace pseudo- sieboldianum S aphylea bumalda Chap e 1 – De ailed summa y ! 14 Figu e 1.2 Clima ic cha s ollowing Wal e and Lie h (1967) o Gyebangsan (GB), Gwangneung (GN), and Haean (HA) o es si es. Chap e 1 – De ailed summa y ! 15 1.2.2 Me hods 1.2.2.1 Sap low measu emen s Sap low echniques we e used o measu e anspi a ion o indi idual ees. Fo ees wi h s em diame e la ge han 5 cm, he mal dissipa ion p obes (TDP; G anie 1987) we e applied (Figu e 1.3a). The TDP is gene ally accep ed as a eliable me hod o es ima ing ee anspi a ion and has been widely used in p e ious s udies o es ima e o es wa e use in a ious ecosys ems (Ba bou e al. 2005; G anie and B éda 1996; Ma yssek e al. 2009; O en and Pa aki 2001; Wullschlege e al. 2001; Zeppel e al. 2006). Fo he unde s o y ees wi h s em/b anch diame e s anging om 9 o 13 mm, s em hea balance (SHB) echnique (Saku a ani 1981; 1984 and imp o ed by Weibel and de Vos 1994) was employed (Figu e 1.3b). The SHB me hod has been success ully employed o es ima e anspi a ion o whole saplings (Lei e al. 2010; Weibel and de Vos 1994) and b anches o la ge ees (O ieno e al. 2007). The TDP was applied in all he s udy si es, while he SHB was used only in he GN si e. Sample ees we e selec ed acco ding o species and ee size dis ibu ion in he s udy plo s. Cha ac e is ics o he sample ees o sap low measu emen s such as ee species, DBH, ee heigh , and sapwood a ea, a e summa ized in Table 2.1 (chap e 2), Table 3.1 (chap e 3), and Table 4.2 (chap e 4). Sap low measu emen s we e ca ied ou in June and July 2008 and epea ed o e he same pe iod in 2009 a he GB si e (chap e 2), Ap il o Sep embe in 2008 a he GN si e (chap e 3), and om May o Oc obe in 2010 a he HA si es (chap e 4). Chap e 1 – De ailed summa y ! 16 Figu e 1.3 Sap low me hods: (a) A schema ic ep esen a ion o he mal dissipa ion p obe (TDP, G anie 1987) ins alled on o a ee. The uppe p obe was hea ed wi h a cons an cu en powe supply while he lowe one ( e e ence) was no hea ed. (b) A schema ic ep esen a ion o s em hea balance (SHB, Saku a ani 1981; 1984), showing a angemen o he he mocouples a ound he ee s em. A, B and C ep esen he espec i e empe a u e di e ences eco ded a he logge . Q low, Q , Q and Pin ep esen con ec i e hea loss by he sap low, e ical hea conduc ion, adial hea conduc ion and hea ing powe , espec i ely. (c) Ins alla ion o TDP in he ees wi h DBH >5 cm and (d) ins alla ion o SHB in he unde s o y ees wi h s em/b anch diame e s 9−13 mm. Chap e 1 – De ailed summa y ! 17 A modi ied Ja is-S ewa model as de ined by Whi ley e al. (2008) was used o es ima e canopy anspi a ion (EC) o he pe iod when da a gaps occu ed due o powe ailu e. Whi ley e al. (2008; 2009) exp essed E in he same way as GC desc ibed by Ja is (1976) and S ewa (1988). EC=ECmax ⋅ 1(RS)⋅ 2(VPD)⋅ 3( θ )⋅ 4(LAI) (1) The unc ions o sola adia ion (RS), apo p essu e de ici (VPD), soil wa e con en (θ), and lea a ea index (LAI) a e a se o scaling e ms educing a maximum EC (ECmax) in esponse o changes in each one o he a iables. Daily EC was de e mined by he unc ions using he op imal es ima es o pa ame e s. A unc ion desc ibing a adia ion esponse was, 1(RS)=RS 1000 ! " #$ % &⋅1000 +k1 RS+k1 ! " #$ % & (2) whe e, k1 is an empi ical coe icien desc ibing he cu a u e o he ela ionship. A unc ion esponse o EC o VPD was, 2(VPD)=k2⋅VPD⋅exp(−k3⋅VPD) (3) whe e, k2 and k3 a e he pa ame e s desc ibing he a e o changes in VPD. A unc ion desc ibing soil mois u e esponse was exp essed as h ee-phase ela ionships, 3( θ )= ! " # $ # 0, θ < θ w θ − θ w θ c− θ w , θ w< θ < θ c 1, θ > θ c (4) whe e, θw is he wil ing poin and θc is he ield capaci y. A unc ion o LAI esponse o EC was, 4(LAI)=LAI LAImax (5) whe e, LAImax is maximum LAI. Model pa ame e iza ion was pe o med using measu ed da a on daily basis ia nonlinea leas squa es analysis using R (R de elopmen Co e Team, 2009). To a oid e o s o di ision by ze o and condi ions o we canopy, day ime da a be ween 8 and 18h was used o he calcula ions, and he da a o he ainy days was excluded. Roo -mean-squa e-e o (RMSE) and an ag eemen index de eloped by Willmo (1981) we e used o e alua e he ag eemen be ween he p edic ed EC and he obse ed EC (see Table 4.3 in chap e 4). The bes es ima ions o pa ame e s (k1, k2, and k3) a e shown in Table 4.3 as well. To al gap- illed pe iods o each si e we e 27, 54, 68, and 8 days o 450N, 650N, 650S, and 950N, espec i ely. P edic ed EC was only used o calcula e annual EC o each si e, bu i was no used o calcula ions o canopy conduc ance as well as o analyses o he ela ionship be ween EC and con olling ac o s. Chap e 1 – De ailed summa y ! 18 1.2.2.2 Canopy conduc ance Biological egula ion o anspi a ion occu s a he s oma a and is measu ed by s oma al conduc ance. Canopy conduc ance (GC) was es ima ed om es ima ed s and anspi a ion de i ed om sap low measu emen s and mic oclima e (Ta and VPD) acco ding o he Penman−Mon ei h equa ion (G anie e al. 1996; Kös ne 1992; O en e al. 1998), GC=( ρ w⋅G ⋅Tk)⋅EC VPD (6) whe e, ρw is densi y o wa e , G is gas conduc ance o wa e apo , Tk is ai empe a u e in kel in, and EC is canopy anspi a ion. This simpli ica ion o he Penman−Mon ei h equa ion is based on he assump ion ha ee canopies a e well coupled o he a mosphe e, when lea es a e exposed o su icien ly high wind speeds, which esul s in la ge ae odynamic conduc ance (GA) han GC. Thus, VPD can be used as an app oxima ion o he o al d i ing o ce o anspi a ion. The assump ion o s ong coupling was es ed in all he s udy si es by compa ing GA and GC. The esul s we e in ag eemen wi h he assump ion, i.e., GC is close o 1% o GA, o GB and HA si es, bu no o GN, pa icula ly he unde s o y laye . The e o e, his simpli ied equa ion was used o he s udies in GB (chap e 2) and HA (chap e 4), whe eas a sepa a e equa ion a anged om he Penman−Mon ei h equa ion was used o GN si e (chap e 3) (Mon ei h 1965; He bs e al. 2008). GC= λ ⋅E⋅ γ ⋅GA Δ ⋅ RN+ ρ ⋅cp⋅VPD⋅GA− λ ⋅E⋅(Δ+ γ ) (7) whe e, λ is he la en hea o apo iza ion o wa e , γ is he psychome ic cons an , Δ is he change o sa ua a ion wa e apo p essu e wi h empe a u e, RN is ne adia ion, ρ is he densi y o d y ai , and cp is he speci ic hea o ai a cons an p essu e. Using his equa ion, canopy conduc ance o he o e s o y and he unde s o y we e calcula ed sepa a ely. GA o he o e s o y and he unde s o y we e also es ima ed using he bounda y laye conduc ance (gb) and he u bulen conduc ance (g ) acco ding o Kös ne e al. (1992), Mangnani e al. (1998), and Ta eishi e al. (2010) as, GA −1=gb −1+g −1 (8) gb=b⋅U⋅[1−exp(− α / 2)] dm⋅ α (9) g = κ 2⋅U LAI ⋅{ln[(z−d) / z0]}2 (10) whe e, b is he p opo ionali y coe icien , U is he wind speed abo e he ege a ion laye s, α is he a enua ion coe icien o wind speed inside he canopy, dm is he cha ac e is ic dimension calcula ed as he squa e oo o a lea a ea, κ is he on Ka man cons an , LAI is he lea a ea index o he canopy, z is he canopy heigh , d is he ze o-plane displacemen , and z0 is he oughness leng h. To calcula e GC based on hese wo models, da a om he measu emen s be ween 11h and 17h (day ime) we e used. Chap e 1 – De ailed summa y ! 19 To assess s oma al sensi i i y o VPD, a modi ied Lohamma ’s unc ion was applied as, GC(VPD)=GC e −m⋅ln(VPD) (11) whe e, GC e is he canopy conduc ance a VPD = 1 kPa and –m is he sensi i i y o GC esponse o VPD (O en e al. 1999). This analysis only ook in o accoun VPD la ge han 0.6 kPa o keep he e o s in GC es ima es unde 10% (Ewe s and O en 2000). GC e was also used o compa e he capaci y o wa e use esponse o VPD among di e en ee species o o es s ands (O en e al. 1999). 1.2.2.3 Biome ic measu emen s To examine seasonal changes o he o es co e and quan i y he maximum lea a ea o he o es s, we measu ed plan a ea index (PAI) and maximum lea a ea index (LAI) a GN and HA si es. PAI was measu ed e e y mon h du ing he ege a i e pe iod using a plan canopy analyze (LAI-2000, LI-COR Inc., Lincoln, USA) unde di use ligh condi ions a ixed 6−12 sampling poin s. The maximum LAI o he o es si es in HA was es ima ed om lea li e collec ed wi h 1 m high, 0.5 m × 0.5 m li e aps andomly placed a i e poin s abo e he o es loo . The li e was ans e ed o he labo a o y, so ed acco ding o species, d ied o 48 hou s a 75°C and weighed. Speci ic lea a ea (SLA) was de e mined om he a io o single lea a ea and d y mass o each species. To al lea a ea o each species om each si e was compu ed om o al lea d y weigh o e he season mul iplied by SLA and di ided by he a ea o he li e ap. The maximum unde s o y LAI (LAIU) o GN si e was measu ed by collec ing lea samples o all he unde s o y ees in h ee plo s wi h 2 m × 2 m size in ea ly July when LAIU was a he peak. The o al a ea o he lea samples was measu ed using a lea a ea me e (LI-3100, LI-COR Inc.). We pe o med ege a ion su eys o all ee s ems la ge han 5 cm in DBH in a 40 m × 50 m plo a he GB si e, in a 30 m × 30 m plo a he GN, and in 25 m × 25 m plo s a he HA si es. Based on he su ey, mean DBH, basal a ea (BA), ee densi y, and species composi ion we e de e mined o each plo (see Table 1.2). To calcula e sapwood a ea (AS) o he sample ees, we measu ed ba k and sapwood dep hs om he ex ac ed co es aken om he s ems a sap low senso heigh a he end o he measu emen s using an inc emen co e . Sapwood was iden i ied by dying he co e samples using b omoc esol g een (Sigma Chemicals, Ge many) (Bu ows 1980). AS was calcula ed om he measu ed DBH and dep hs o ba k and sapwood. Based on he calcula ed AS and measu ed DBH, an allome ic unc ion was es ablished (Ve essy e al. 1995, Meinze e al. 2005) as, AS= α DBH β (12) whe e, α is a cons an and β is he allome ic scaling exponen . Allome ic ela ionships we e made o each one o he s udied species as well as o all he species oge he . Since he di e ences in he es ima ions o s and AS ( o al AS o all ees in he plo ) om sepa a e eg essions and om a combined species eg ession Chap e 1 – De ailed summa y ! 20 we e less han 10% a all si es, we chose he gene al eg ession o u he analyses ela ed o AS. Fo example, he alues o s and AS de e mined by he gene al eg ession we e used o he es ima ions o canopy anspi a ion o each si e. 1.2.2.4 Mic ome eo ological measu emen s Ai empe a u e (Ta), p ecipi a ion, ne adia ion (RN) o sola adia ion (RS), pho osyn he ic ac i e adia ion (PAR), ela i e humidi y o wa e apo densi y, and wind speed (U) we e measu ed wi h a 20 m owe in GB, wi h a 40 m owe in GN, and a 2 m abo e he g ound in he open space nex o he o es si es in HA. Mic oclima es below he canopy o Ta, PAR, and humidi y we e measu ed a 5 m in GB, a 4 m (bu a 2 m o PAR) in GN, and a 2 m in HA si es. VPD was de i ed om ai empe a u e and ela i e humidi y (Mu ay 1967). In he GN si e, U in he unde s o y a a heigh o 4 m was addi ionally measu ed. Soil wa e con en (θ) and empe a u e we e measu ed a 30 cm dep h in all he si es. Da a we e ead e e y 30 seconds, a e aged, and s o ed e e y 30 minu es using da a logge s. Soil wa e e en ion was de e mined om measu ed θ and soil cha ac e is ics (i.e., soil ex u e and bulk densi y). 1.2.2.5 Eddy co a iance lux measu emen s A he GN si e, e apo anspi a ion (Eeco) was measu ed wi h an eddy co a iance sys em ins alled on a 40 m high owe in GN si e o he same pe iod as sap low measu emen s (chap e 3). Da a quali y was con olled using he s anda dized KoFlux p o ocol including plana i o a ion, Webb-Pea man-Leuning co ec ion, spike de ec ion, and gap illing (Hong e al. 2009). Da a gaps we e illed using a modi ied lookup able ollowing he me hod p oposed by FLUXNET (Reichs ein e al. 2005) and modi ied by Kang e al. (2012). 1.2.2.6 Wa e use e iciency F om each si e, i e sun- and i e shade-lea es each om i e Q. mongolica canopy ees we e collec ed on 24 June 2010, du ing mid season. The samples we e o en-d ied a 75°C o 48 hou s and hen ball-milled be o e subjec ed o 13C/12C iso opic a io analysis a BayCEER – Labo a o y o Iso ope Biogeochemis y, Ge many. Analyses we e conduc ed wi h an elemen al analyze NA 1108 (CE Ins umen s, Milan, I aly) coupled o an iso ope a io mass spec ome e del a S (Finnigan MAT, B emen, Ge many) ia an open spli in e ace ConFlo III (Finnigan MAT, B emen , Ge many) as desc ibed by Bida ondo e al. (2004). S anda d CO2 gas was calib a ed wi h espec o in e na ional s anda d (CO2 in Pee Dee Belemni e) by use o he e e ence subs ance NBS 16 o 20 o ca bon iso opic a io p o ided by he in e na ional A omic Ene gy Chap e 1 – De ailed summa y ! 21 Agency IAEA, Vienna, Aus ia. The 13C/12C iso opic a ios, deno ed as del a alues we e calcula ed acco ding o he equa ion δ 13C=Rsample Rs d −1 " # $% & '×1000 (13) whe e δ13C is he iso ope a io o ca bon in del a uni s ela i e o he PDB s anda d. Rsample and Rs d a e he 13C/12C o he samples and he PDB s anda d, espec i ely. δ13C was used as an index o seasonally in eg a ed wa e use e iciency (WUE) (Tieszman and A che 1990). 1.3 Gene al esul s and discussions 1.3.1 T ee and o es wa e use in di e se species composi ion An allome ic equa ion was used o ind a gene al ela ionship be ween DBH and AS o di e en species co-occu ing in each s udy si e. A signi ican ela ionship, AS = αDBHβ, linking AS o DBH, was es ablished (see Figu e 2.5 in chap e 2, Figu e 3.1 in chap e 3, and Figu e 4.2 in chap e 4). These eg ession models we e applied o calcula e AS o all he ees in each s udy plo . To examine he applicabili y o his gene al ela ionship ac oss he si es, we combined he da a o 13 deciduous ee species g owing in di e en s udy si es. We ound ha all he species, ega dless o hei loca ion, signi ican ly (R2 = 0.77, P <0.0001) i ed in o a single powe cu e (Figu e 1.4a). This indica es ha he gene al eg ession model es ablished in his s udy can be applicable o simila o es s in S. Ko ea, ou side ou s udy si es loca ed in o he places. To educe complexi y in es ima ion o anspi a ion in mixed deciduous o es s, a simila analysis ela ing TWU and DBH was pe o med. Maximum daily TWU was signi ican ly ela ed o DBH in a powe unc ion o he species co-occu ing a each s udy si e (P <0.001 o all si es). To check i his ela ionship is applicable ac oss he s udied si es, he 12 ee species measu ed in he h ee di e en s udy si es we e combined in he analysis. A signi ican (R2 = 0.77, P <0.0001) single powe unc ion was es ablished o he combined da a (Figu e 1.4b), al hough he e was a endency o highe TWU o he ees g owing in he HA si e, which can be explained by g ea e alues o maximum Fd, compa ed o he o he s g owing in GB and GN si es (Figu e 1.5). This esul was consis en wi h he p e ious s udies conduc ed in empe a e o es in No h Ame ica (Wullschlege e al. 2001), opical ain o es (Meinze e al. 2005), and open woodland in Aus alia (Zeppel and Eamus 2008). Ou indings suppo he hypo hesis ha di e en ee species g owing oge he in common loca ions ha e con e ging unc ion in wa e use de e mined by ee sizes (Kalla akal e al. 2013). As long as soil wa e is no limi ing, ansien changes in ee anspi a ion a e due o he p e ailing mic oclima e, while he po en ial maximum anspi a ion is de e mined by he xylem anspo capaci y (O en and Pa aki 2001). This implies ha he same powe scaling pa ame e s, de e mining AS and Chap e 1 – De ailed summa y ! 22 TWU o single ees by he DBH sizes, can be applied o di e en ee species g owing in deciduous o es s in S. Ko ea as long as ees a e exposed o simila en i onmen s. Thus, ou indings can be applied o es ima e maximum daily TWU and s and anspi a ion using a simple empi ical allome y model and DBH o he a ge si es, i espec i e o he species. The powe eg essions and he s a is ics o hese analyses a e shown in Table 1.3 and 1.4. To iden i y he mos sui able unc ion explaining he ela ionship be ween TWU and DBH, he same da a se was i ed o he h ee-pa ame e sigmoid unc ion (Meinze e al. 2005) (Figu e 1.4b). We ound simila goodness o i be ween a sigmoid and powe unc ion (see AIC alues in Table 1.4). Meinze e al. (2005) ound supe io i ing om a sigmoid unc ion han powe unc ion o angiospe m ees, which was suppo ed by he e idences ha anspi a ion and pho osyn hesis a e limi ed as ees g ow abo e a h eshold size (McDowell e al. 2002, Niineme s 2002). Based on ou s udy, howe e , bo h powe and sigmoid unc ions can be used o he es ima ion o TWU om DBH. Unlike AS and TWU, he e was no signi ican ela ionship be ween maximum Fd and DBH (Figu e 1.5). This is consis en wi h he s udy conduc ed by Phillips e al. (1999). Meinze e al. (2001), howe e , ob ained a nega i e co ela ion, while O en e al. (1998) ound posi i e co ela ion be ween maximum Fd and DBH sugges ing ha DBH can be a good p edic o o Fd among species om di e se loca ions. Se e al s udies ha e also conside ed wood densi y as an al e na i e p edic o no only o Fd bu also o hyd aulic conduc i i y and s oma al conduc ance (Bucci e al. 2004; O’G ady e al. 2009; Kalla ackal e al. 2013). This is an al e na i e app oach ha we ecommend in u u e s udies, since wood densi y combines bo h he ee age and g ow h condi ions. Chap e 1 – De ailed summa y ! 29 Figu e 1.7 Cumula i e canopy anspi a ion (EC) om Gwangneung (GN) and Haean (HA) si es. Figu e 1.8 Rela ionships be ween ele a ion and (a) annual mean Ta (b) annual day ime mean VPD, (c) leng h o g owing season, and (d) o al canopy anspi a ion (EC) o he Gwangneug (GN) and Haean (HA) si es. Chap e 1 – De ailed summa y ! 30 1.4 Gene al conclusions This s udy in es iga ed he challenges imposed by mul iple ee species, mul iple canopy laye ing and complex e ain when a emp ing o de e mine wa e use by wa m- empe a e o es species in S. Ko ea. Ou main indings and sugges ions a e summa ized below:  DBH was co ela ed wi h AS (in a powe unc ion) and maximum TWU (bo h in powe and sigmoid unc ions) no only o he ee species co-occu ing in a single o es s and bu also g owing in di e en ly aged and s uc u ed deciduous o es s.  This unc ional ela ionship be ween DBH and TWU p o ides a ela i ely simple bu accu a e app oach o he p edic ion o wa e use by ees and o es s ands in mixed deciduous o es s, hus educing he complexi y a ising om mul iple ee species.  Tempo al di e ences in he o e s o y and unde s o y de elopmen s, wi h ea lie bud b eak and lea expansion o he unde s o y ees, al e ed he mic oclima e in he unde s o y and hence he a e o EU be o e and a e he o e s o y canopy ma u i y.  The seasonal pa e n o bo h EO and EU a e egula ed by canopy de elopmen and mic oclima e, p ima ily VPD and PAR, which de e mine hei daily end.  EU signi ican ly con ibu ed o o al o es anspi a ion du ing he whole g owing season, wi h he highes con ibu ions in Ap il and May, which esul ed in ela i ely high o al anspi a ion ea ly in he season.  Since Eeco is s ongly in luenced by EO and EU, bo h he o e s o y and unde s o y should be conside ed when es ima ing Eeco o o es s consis ing o a well-de eloped unde s o y laye .  High s oma al sensi i i y o VPD o he o es loca ed a highe ele a ion esul ed in ea ly sa u a ion and midday dep ession o EC a ela i ely low day ime VPD and possibly high WUE, al hough he main d i e o he shi in s oma al sensi i i y is s ill unclea .  To al EC du ing he g owing season dec eased wi h inc easing ele a ion, co esponding o he dec ease in Ta, day ime VPD, and leng h o g owing season a he highe ele a ion. These a iables should be ca e ully aken in o accoun o he es ima ion o o es wa e use in moun ainous egions in complex e ains. Ou s udy add essed he challenges in ol ed in es ima ing wa e use by he wa m- empe a e deciduous o es s in S. Ko ea, comp ised o di e se ee species, mul i-laye ed canopy, and ugged moun ainous e ains. We obse ed ha he complexi y a ising om mul iple ee species can be sol ed using a unc ional ela ionship be ween DBH and TWU. We also ound ha o o es s wi h a mul i-laye ed canopy, bo h EO and EU should be aken in o conside a ion when es ima ing Eeco. Finally, in complex e ains, ele a ion alone Chap e 1 – De ailed summa y ! 31 does no de e mine o es wa e use, bu wo ks in andem wi h mic oclima e and plan g ow h cha ac e is ics, ac o s ha need o be de e mined a ela i ely small spa ial scales since hey change apidly. 1.5 Lis o manusc ip s and speci ica ion o con ibu ions This disse a ion includes h ee manusc ip s. The i s manusc ip is published in Plan Ecology and he second manusc ip is published in Jou nal o Plan Resea ch. The hi d manusc ip is submi ed o Plan and Soil and is cu en ly ‘unde e iew’. Speci ic con ibu ions by he co-au ho s o each manusc ip a e lis ed below. Manusc ip 1 (Chap e 2) Au ho s EY Jung, D O ieno, B Lee, JH Lim, SK Kang, MWT Schmid , J Tenhunen Ti le Up-scaling o s and anspi a ion o an Asian empe a e mixed-deciduous o es om single ee sap low measu emen s S a us Published in Plan Ecology (2011) 212:383-395 Con ibu ions EY Jung: concep s, ield wo k, discussion and p esen a ion o esul s, manusc ip p epa a ion, co esponding au ho D O ieno: concep s, ield wo k, discussion o esul s, manusc ip edi ing B Lee: ield wo k JH Lim: ield wo k, logis ics in Ko ea SK Kang: logis ics in Ko ea MWT Schmid : discussion o esul s J Tenhunen: concep s, manusc ip edi ing Manusc ip 2 (Chap e 3) Au ho s EY Jung, D O ieno, H Kwon, B Lee, JH Lim, J Kim, J Tenhunen Ti le Wa e use by a wa m- empe a e deciduous o es unde he in luence o he Asian monsoon: con ibu ions o he o e s o y and unde s o y o o es wa e use S a us Published in he Jou nal o Plan Resea ch (2013) DOU 10.1007/s10265-013-0563-5 Con ibu ions EY Jung: concep s, ield wo k, p esen a ion and discussion o esul s, manusc ip p epa a ion, co esponding au ho D O ieno: concep s, ield wo k, discussion o esul s, manusc ip edi ing H Kwon: ield wo k, discussion o esul s, manusc ip edi ing B Lee: ield wo k Chap e 1 – De ailed summa y ! 32 JH Lim: ield wo k, logis ics in Ko ea J Kim: logis ics in Ko ea J Tenhunen: concep s, manusc ip edi ing Manusc ip 3 (Chap e 4) Au ho s EY Jung, D O ieno, H Kwon, S Be ge , M Haue , J Tenhunen Ti le In luence o ele a ion on canopy anspi a ion o empe a e deciduous o es s in a complex moun ainous e ain o Sou h Ko ea S a us Submi ed o Plan and Soil; cu en s a us ‘in e ision’ Con ibu ions EY Jung: concep s, ield wo k, p esen a ion and discussion o esul s, manusc ip p epa a ion, co esponding au ho D O ieno: concep s, ield wo k, discussion o esul s, manusc ip edi ing H Kwon: discussion o esul s, manusc ip edi ing S Be ge and M Haue : ield wo k J Tenhunen: concep s, logis ics in Ko ea 1.6 Re e ences Allaby M (2006) Biomes o he Ea h: Tempe a e Fo es s. New Yo k, NY: Chelsea House A anda I, Gil L, Pa dos JA (2000) Wa e ela ions and gas exchange in Fagus syl a ica L. and Que cus pe aea (Ma uschka) Liebl. In a mixed s and a hei sou he n limi o dis ibu ion in Eu ope. T ees 14:344–352 Baldocchi DD, Hicks BB, Meye s TP (1988) Measu ing biosphe e-a mosphe e exchanges o biologically ela ed gases wi h mic ome eo ological me hods. Ecology 69:1331–1340 Baldocchi DD, Vogel C (1996) A compa a i e s udy o wa e apo , ene gy and CO2 lux densi ies abo e and below a empe a e b oadlea and a bo eal pine o es . T ee Physiol 16:5–16 Baldocchi DD, Wilson KB, Gu L (2002) How he en i onmen , canopy s uc u e and canopy physical unc ioning in luence ca bon, wa e and ene gy luxes o a empe a e b oad-lea ed deciduous o es – an assessmen wi h he biophysical model CANOAK. T ee Physiol 22:1065–1077 Ba bie S, Gosselin F, Balandie P (2008) In luence o ee species on unde s o y ege a ion di e si y and mechanisms in ol ed – A c i ical e iew o empe a e and bo eal o es s. Fo Ecol Manage 254:1–15 Ba bou MM, Hun JE, Walc o AS, Roge s GND, McSe eny TM, Whi ehead D (2005) Componen s o ecosys em e apo a ion in a empe a e coni e ous ain o es , wi h canopy anspi a ion scaled using sapwood densi y. New Phy ol 165:549–558 Chap e 1 – De ailed summa y ! 33 Be bigie P, Diawa a A, Lous au D (1991) A mic oclima ic s udy o he e ec o d ough on e apo anspi a ion in a ma i ime pine s and and i s unde s o y. Ann Sci Fo 48:157–177 Bida ondo MI, Bu gha d B, Gebaue G, B uns TD, Read DJ (2004) Changing pa ne s in he da k: iso opic and molecula e idence o ec omyco hizal liaisons be ween o es o chids and ees. P oc R Soc Lond B 271:1799−1806 Black TA (1979) E apo anspi a ion om Douglas i s ands exposed o soil wa e de ici s. Wa e Resou Res 15:164–170 Bo a d BD, Cu is PS, Vogel CS, Su H-B, Schimid HP (2005) En i onmen al con ols on sap low in a no he n ha dwood o es . T ee Physiol 25:31–38 Bucci SJ, Golds ein G, Meinze FC, Scholz FG, F anco AC, Bus aman e M (2004) Func ional con e gence in hyd aulic a chi ec u e and wa e ela ions o opical sa anna ees: om lea o whole plan . T ee Physiol 24:891−899 Bu ows LE (1980) Di e en ia ing sapwood, hea wood and pa hological wood in li e moun ain beech. New Zealand Fo es Se ice, Fo es Resea ch Ins i u e, P o ec ion Fo es y Repo 172 Campbell-Clause M (1998) S oma al esponse o g ape ines o wind. Aus J Exp Ag ic 38:77−82 Canham DC, Finzi AC, Pacala SW, Bu bank DH (1994) Causes and consequences o esou ce he e ogenei y in o es s: in e speci ic a ia ion in ligh ansmission by canopy ees. Can J Fo Res 24:337−349 Chapin FS, Ma son PA, Vi ousek PM (2011) P inciples o Te es ial Ecosys em Ecology. Second edi ion. Sp inge , New Yo k Chung TH, Lee WC (1965) A s udy o he Ko ean woody plan zone and a o able egion o he g ow h and p ope species. Thesis Collec ion o Sungkyunkwan Uni e si y 10:329−435 Cocha d H, B éda N, G anie A (1996) Whole ee hyd aulic conduc ance and wa e loss egula ion in Que cus du ing d ough : e idence o s oma al con ol o embolism? Ann Sci Fo 53:197−206 Co dell S, Golds ein G, Muelle -Dombois D, Webb D, Vi ousek PM (1998) Physiological and mo phological a ia ion in Me oside os polymo pha, a dominan Hawaiian ee species, along an al i udinal g adien : he ole o pheno ypic plas ici y. Oecologia 113:188-196 Di ma C, Elling W (2006) Phenological phases o common beech (Fagus syl a ica L.) and hei dependence on egion and al i ude in Sou he n Ge many. Eu J Fo Res 125:181−188 Ehle inge JR (1993) Ca bon and wa e ela ions in dese plan s: an iso opic pe spec i e. In Ehle inge JR, Hall AE, Fa quha GD (Eds.), S able iso opes and plan ca bon-wa e ela ions, Academic P ess, San Diego, Cali o nia, USA Ellenbe g H (1978) Zu Popula ionsökologie des Rehes (Cap eolus cap eolus L., Ce idae) in Mi eleu opa. Spixiana 2, München Ewe s BE, O en R (2000) Analyses o assump ions and e o s in he calcula ion o s oma al conduc ance om sap lux measu emen s. T ee Physiol 20:579−589 Chap e 1 – De ailed summa y ! 34 Ewe s BE, Mackay DS, Gowe s ST, Ahl DE, Bu ows SN, Saman a SS (2002) T ee species e ec s on s and anspi a ion in no he n Wisconsin. Wa e Resou ces Res 38:1103 Fo d GR, Hubba d RM, Koeppel BD, Vose JM (2007) A compa ison o sap lux-based e apo anspi a ion es ima es wi h ca chmen -scale wa e balance. Ag ic Fo Me eo ol 145:176−185 Gong WS (2001) Tempo al and spa ial changes in he dis ibu ion o Phyllos achys habi a ion (in Ko ean). Ko Geog aph Soc 36:444−457 G anie A (1987) E alua ion o anspi a ion in a Douglas- i s and by means o sap low measu emen s. T ee Physiol 3:309−320 G anie A, B éda N (1996) Modelling canopy conduc ance and s and anspi a ion o an oak o es om sap low measu emen s. Ann Sci Fo 53:537−546 G anie A, Huc R, Ba igah ST (1996) T anspi a ion o na u al ain o es and i s dependence on clima ic ac o s. Ag ic Fo Me eo ol 78:19−29 G anie A, B éda N, Bi on P, Ville e S (1999) A lumped wa e balance model o e alua e du a ion and in ensi y o d ough cons ain s in o es s ands. Ecol Mod 116:269−283 Han SS, Kim SH (1993) Ecophysiological in e p e a ions on he wa e ela ions pa ame e s o ees (VII) − Measu emen o wa e low by he hea pulse me hod in a La ix lep olepis s and. J Ko Fo Soc 82: 152–165 Han SS, Kim SH (1996) Ecophysiological in e p e a ions on he wa e ela ions pa ame e s o ees (IX) − Measu emen o he anspi a ion a e by he hea pulse me hod in a Que cus mongolica s and. J Ko Fo Soc 85: 288–299 Ha on TJ, Rup ech J, Geo ge RJ (2003) P eclea ing hyd ology o he Wes e n Aus alia whea bel : Ta ge o he u u e. Plan Soil 257:341–356 He bs M, Rosie PTW, Mo ec o MD, Gowing DJ (2008) Compa a i e measu emen s o anspi a ion and canopy conduc ance in wo mixed deciduous woodlands di e ing in s uc u e and species composi ion. T ee Physiol 28:959−970 Hi obe H, Naoko T, Iwa subo G (1998) Spa ial a iabili y o soil ni ogen ans o ma ion pa e ns along a o es slope in a C yp ome ia japonica D. Don plan a ion. Eu J Soil Biol 34:123−131 Hoch G, Kö ne C (2005) G ow h, demog aphy and ca bon ela ions o Polylepis ees a he wo ld’s highes ee line. Func Ecol 19:941−951 Hong J, Kwon H, Lim JH, Byun YH, Lee J, Kim J (2009) S anda diza ion o KoFlux eddy-co a iance da a p ocessing. Ko J Ag ic Fo Me eo ol 2:12−26 Hubick KT, Fa quha GD, Sho e R (1986) Co ela ion be ween wa e -use e iciency and ca bon iso ope disc imina ion in di e se peanu (A achis) ge mplasm. Aus alian J Plan Physiol 13:803−816 Chap e 1 – De ailed summa y ! 35 Im E-S, Ahn J-B, Kwon W-T, Gio gi F (2008) Mul i-decadal scena io simula ion o e Ko ea using a on- way double-nes ed egional clima e model sys em. Pa 2: u u e clima e p ojec ion (2021−2050). Clim Dym 30:239−254 I o A, Saigusa N, Mu ayama S, Yamamo o S (2005) Modeling o g oss and ne ca bon dioxide exchange o e a cool- empe a e deciduous b oad-lea ed o es in Japan: Analysis o seasonal and in e annual change. Ag ic Fo Me eo ol 134:122−134 IPCC (2000) Special epo on emissions scena ios. A Special epo o wo king g oup III o he in e go e nmen al panel on clima e change (Nakiceno ic N and lead au ho s). Camb idge Uni e si y P ess, Camb idge Ja is PG (1976) The in e p e a ion o he a ia ions in lea wa e po en ial and s oma al conduc ance ound in canopies in he ield. Phil T ans Roy Soc Lond B 273:593−610 Jones HG, Su he land RA (1991) S oma al con ol o xylem embolism. Plan Cell En i on 14: 607−612 Kalla ackal J, O ieno DO, Reineking B, Jung E, Schmid MWT, G anie A, Tenhunen JD (2013) Func ional con e gence in wa e use o ees om di e en geog aphical egions: a me a-analysis. T ees 27: 787−799 Kang M, Pa k S, Kwon H, Choi HT, Choi YJ, Kim J (2009) E apo anspi a ion om a deciduous o es in a complex e ain and a he e ogeneous a mland unde monsoon clima e. Asia-Paci ic J A mos Sci 45:175−191 Kang M, Kwon H, Cheon JH, Kim J (2012) On es ima ing we canopy e apo a ion om deciduous and coni e ous o es s om Asian monsoon clima e. J Hyd ome eo l 13:950−965 Kellihe FM, Leuning R, Schulze ED (1993) E apo a ion and canopy cha ac e is ic o coni e ous o es s and g asslands. Oecologia 95:153−163 Kim JS (1987) Fo es e ec s on he lood discha ges and he es ima ion o he e apo anspi a ion in he small wa e shed. Ko ea Fo es Resea ch Ins i u e Repo s, Vol.35 Kim JH (2002) Communi y ecological iew o he na u al deciduous o es in Ko ea. In: Ecology o Ko ea, Bumwoo, Seoul Kim KH, Woo BM (1988) S udy on ain all in e cep ion loss om canopy in o es (I). Jou nal o Ko ean Fo es y Socie y, Vol.77, No.3 Kim J, Lee D, Hong J, Kang S, Kim S-J, Moon S-K, Lim J-H, Son Y, Lee J, Kim S, Woo N, Kim K, Lee B, Lee B-L, Kim S (2006) Hyd oKo ea and Ca boKo ea: c oss-scale s udies o ecohyd ology and biogeochemis y in a he e ogeneous and complex o es ca chmen o Ko ea. Ecol Res 21:881−889 Ki a T (1991) Fo es ecosys ems o eas and sou heas Asia in a global pe spec i e. Ecological Resea ch 6:185−200 Ko ea Fo es Se ice (2006) S a is ical Yea book o Fo es y. Seoul, KFS Chap e 1 – De ailed summa y ! 36 Ko ea Fo es Se ice (2009) Na ional Repo on Sus ainable Fo es Managemen in Ko ea, 2009. Daejeon, KFS Ko ea Fo es Se ice (2011) 2010 Basic S a is ics o Fo es . Daejeon, KFS Kö ne C, Fa quha GD, Roksandik Z (1988) A global su ey o ca bon iso ope disc imina ion in plan s om high al i ude. Oecologia 74:623−632 Kö ne C, Fa quha GD, Wong SC (1991) Ca bon iso ope disc imina ion by plan s ollows la i udinal and al i udinal ends. Oecologia 88:30−40 Kö ne C (1994) Lea di usi e conduc ances in he majo ege a ion ypes o he globe. In: Schulse E-D, Caldwell MM (eds.) Ecophysiology o Pho osyn hesis. Sp inge -Ve lag, Heidelbe g Kö ne C (2003) Alpine Plan Li e (2nd ed.), Sp inge -Ve lag, Be lin Heidelbe g New Yo k Kö ne C (2007) The use o ‘al i ude’ in ecological esea ch. T ends Ecol E ol 22:569–574 Kös ne BMM, Schulze E-D, Kellihe FM, Hollinge DY, Bye s JN, Hun JE, McSe eny TM, Mese h R, Wei PL (1992) T anspi a ion and canopy conduc ance in a p is ine b oad-lea ed o es o No ho agus: an analysis o xylem sap low and eddy co ela ion measu emen s. Oecologia 91:350−359 Kös ne B (2001) E apo a ion and anspi a ion om o es s in Cen al Eu ope – ele ance o pa ch-le el s udies o spa ial scaling. Me eo ol A mos Phys 76:69−82 Kubo a M, Tenhunen J, Zimme mann R, Schmid M, Adiku S, Kakuba i Y (2005) In luences o en i onmen al ac o s on he adial p o ile o sap lux densi y in Fagus c ena a g owing a di e en ele a ions in he Naeba Moun ains, Japan. T ee Physiol 25:545−556 Kumagai T, Ta eishi M, Shimizu T, O suki K (2008) T anspi a ion and canopy conduc ance a wo slope posi ions in a Japanese ceda o es wa e shed. Ag ic Fo Me eo ol 148:1444−1455 Kwon W-T (2005) Cu en s a us and pe spec i es o clima e change sciences (in Ko ean wi h English abs ac ). J Ko Me eo ol Soc 41:325−336 La ham RE, Rickle s RE (1993) Global pa e ns o ee species ichness in mois o es s: ene gy-di e si y heo y does no accoun o a ia ion in species ichness. Oikos 67:325−333 Lausche F (1976) Wel wei e Typen de Höhennabhängigkei des Niede schlags. We e Leben 28:80−90 Lee JH, Kim TH, Lee WK, Choi K, Lee CY, Joo JS (1989) A s udy on egula ing discha ges by o es . Ko ea Fo es Resea ch Ins i u e Repo s, Vol.38, Seoul Lee YH, Pa k TS, Youn YC, Ryu TK, Lim YJ, Choi JC, Byun WH, Kim EG (1997) His o y o Ko ean Fo es y Policy. Ko ea Fo es Se ice, Seoul Lee IK, Lim JH, Kim C, Kim YK (2006) Nu ien dynamics in decomposing lea li e and li e p oduc ion a he long- e m ecological esea ch si e I M . Gyebangsan (in Ko ean wi h English abs ac ). J Ecol Field Biol 29:585−591 Chap e 1 – De ailed summa y ! 37 Lei H, Zhi-Shan Z, Xin-Rong L (2010) Sap low o A emisia o dosica and he in luence o en i onmen al ac o s in a e ege a ed dese a ea: Tengge Dese , China. Hyd ol P ocess 24:1248−1253 Lim JH, Chun JH, Woo SY, Kim YK (2008) Inc eased declines o Ko ean i o es caused by clima e change in Moun ain Halla, Ko ea. In: O al P esen a ion A : In e na ional Con e ence “Adap a ion o Fo es s and Fo es Managemen o Changing Clima e wi h Emphasis on Fo es Heal h: A Re iew o Science, Policies, and P ac ices”, Umeá, Sweden, FAO/IUFRO, 25–28 Augus 2008 Magnani F, Leona di S, Togne i R, G ace J, Bo ghe i M (1998) Modeling he su ace conduc ance o a b oad-lea canopy: e ec s o pa ial decoupling om he a mosphe e. Plan Cell En i on 21:867−879 Ma yssek R, Wiese G, Pa zne K, Blaschke H, Häbe le K-H (2009) T anspi a ion o o es ees and s ands a di e en al i ude: consis encies a he han con as s? Eu J Fo Res 128−579596 McDowell NG, Phillips N, Lunch C, Bond BJ, Ryan MG (2002) An in es iga ion o hyd aulic limi a ion and compensa ion in la ge, old Douglas- i ees. T ee Physiol 22:763−774 McDowell NG, Whi e S, Pockman WT (2008) T anspi a ion and s oma al conduc ance ac oss a s eep clima e g adien in he sou he n Rocky Moun ains. Ecohyd ol 1:193−204 Meinze FC, Golds ein G, And ade JL (2001) Regula ion o wa e lux h ough opical o es canopy ees: do uni e sal ules apply? T ee Physiol 21:19−26 Meinze FC, Bond BJ, Wa en JM, Wood u DR (2005) Does wa e anspo scale uni e sally wi h ee size? Func Ecol 19:558−565 Miyajima Y, Takahashi K (2007) Changes wi h al i ude o he s and s uc u e o empe a e o es s on Moun No iku a, cen al Japan. J Fo Res 12:187−192 Mon ei h JL (1965) E apo a ion and en i onmen . Symp Soc Exp Biol XIX:205−234 Mon ei h JL, Unswo h MH (1990) P inciples o en i onmen al physics. Edwa d A nold, London Mu aoka H, Koizumi H (2005) Pho osyn he ic and s uc u al cha ac e is ics o canopy and sh ub ees in a cool- empe a e deciduous b oadlea ed o es : Implica ion o he ecosys em ca bon gain. Ag ic Fo Me eo ol 134:39−59 Mu ay FW (1967) On he compu a ion o sa u a ion apo p essu e. J Appl Me eo ol 6:203−204 Nakashizuka T (1987) Regene a ion dynamics o beech o es s in Japan. Vege a io 69:169−175 Nakashizuka T, Iida S (1995) Composi ion, dynamics and dis u bance egime o empe a e deciduous o es s in Monsoon Asia. Vege a io 121:23−30 Nasaha a KN, Mu aoka H, Nagai S, Mikami H (2008) Ve ical in eg a ion o lea a ea index in Japanese deciduous b oad-lea ed o es . Ag ic Fo Me eo ol 148:1136−1146 Na ional Ins i u e o En i onen al Resea ch (2011) Ko ea Clima e Change Valua ion 2010, Incheon, NIER Niineme s U (2002) S oma al conduc ance alone does no explain he decline in olia pho osyn he ic a es wi h inc easing ee age and size in Picea abies and Pinus syl es is. T ee Physiol 22:515−535 Chap e 1 – De ailed summa y ! 38 O’G ady AP, Cook PG, Eamus D, Duguid A, Wischusen JDH, Fass T, Wo ldege D (2009) Con e gence o ee wa e use wi hin an a id-zone woodland. Oecologia 160:643−655 Oh J-H, Kim T, Kim M-K, Lee S-H, Min S-K, Kwon W-T (2004) Regional clima e simula ion o Ko ea using dynamic downscaling and s a is ical adjus men . J Me eo ol Soc Jpn 82:1629−1643 O en R, Phillips N, Ka ul G, Ewe s BE, Pa aki DE (1998) Scaling xylem sap lux and soil wa e balance and calcula ing a iance: a me hod o pa i ioning wa e lux in o es s. Ann Sci Fo 55:191−216 O en R, Phillips N, Ewe s BE, Pa aki DE, Megonigal JP (1999) Sap- lux-scaled anspi a ion esponses o ligh , apo p essu e de ici , and lea a ea educ ion in a looded Taxodium dis ichum o es . T ee Physiol 19:337−347 O en R, Pa aki DE (2001) T anspi a ion in esponse o a ia ion in mic oclima e and soil mois u e in sou heas e n deciduous o es s. Oecologia 127:549−559 O ieno DO, Schmid MWT, Ku z-Besson C, Lobo Do Vale R, Pe ei a JS, Tenhunen JD (2007) Regula ion o anspi a ional wa e loss in Que cus sube ees in a Medi e anean- ype ecosys em. T ee Physiol 27:1179−1187 Pa k J-Y, Pa k M-J, Ahn S-R, Pa k G-A, Yi J-E, Kim G-S, Sini asan R, Kim SJ (2011) Assessmen o u u e clima e change impac s on wa e quan i y and quali y o a moun ainous dam wa e shed using SWAT. T ans ASABE 54:1725−1737 Phillips N, O en R, Zimme mann R, W igh SJ (1999) Tempo al pa e ns o wa e lux in ees and lianas in a Panamanian mois o es . T ee 14:116−123 Qian H, Rickle s RE (1999) A compa ison o he axonomic ichness o ascula plan s in China and he Uni ed S a es. Am Na 154:160−181 Qian H, Rickle s RE (2000) La ge-scale p ocesses and he Asian bias in species di e si y o empe a e plan s. Na u e 407:180−182 R De elopmen Co e Team (2009) R: A Language and En i onmen o S a is ical Compu ing. Vienna, Aus ia: R Founda ion o S a is ical Compu ing Reichs ein M e al. (2005) On he sepa a ion o ne ecosys em exchange in o assimila ion and ecosys em espi a ion: e ies and imp o ed algo i hm. Glob Change Biol 11:1424−1439 Richa dson AD, Bailey AS, Denny EG, Ma in CW, O’Kee e J (2006) Phenology o a no he n ha dwood o es canopy. Glob Change Bio 12:1174−1188 Röh ig E (1991a) Clima ic condi ions. pp. 7−16. In: Röh ig E and Ul ich B (eds.) Tempe a e Deciduous Fo es s. Else ie , Ams e dam Röh ig E (1991b) Flo al composi ion and i s e olu iona y de elopmen . pp. 17−33. In: Röh ig E and Ul ich B (eds.) Tempe a e Deciduous Fo es s. Else ie , Ams e dam Saku a ani T (1981) A hea balance me hod o measu ing wa e lux in he s em o in ac plan s. J Ag ic Me eo ol 37:9–17 Chap e 2 – Up-scaling o s and anspi a ion o an Asian empe a e mixed-deciduous o es om single ee sap low measu emen s ! 45 (UA-002-08, Onse , USA) a 1 m (below he canopy) and a 20 m heigh (abo e he canopy), pho osyn he ic ac i e adia ion (PAR) (LI-190, LI-COR, USA) a 20 m, ai empe a u e and humidi y (HMP35C, Cambell Scien i ic Inc., USA) a 5, 10, and 20 m heigh below, wi hin and abo e he canopy, and soil wa e con en and empe a u e (5TE, Decagon De ices, USA) a −5, −15 and −30 cm. These pa ame e s we e measu ed con inuously du ing he expe imen al pe iod. Da a we e a e aged and logged e e y 30 min, ei he wi h logge s buil in o he senso s (ligh in ensi y) o a cen al da a logge (DL2e, Del a-T De ices, UK). Vapo p essu e de ici (VPD) was calcula ed om ai empe a u e and ela i e humidi y. P ecipi a ion da a we e ob ained om an au oma ed wea he s a ion buil by he Ko ea me eo ological adminis a ion loca ed 5.2 km away om ou s udy si e. 2.2.4 T ee allome ics The sample ees we e selec ed acco ding o species and ee size dis ibu ion in he plo (Figu e 2.2). T. amu ensis was he dominan species in he plo , occupying 31.5% o he o al basal a ea. U. da idiana, Q. mongolica, A. mono, and C. con o e sa we e co-dominan and occupied 13.6, 10.7, 8.1, and 4.4% o he o al basal a ea o he plo , espec i ely. Thus, hese dominan and co-dominan species co e ed almos 70% o he o al basal a ea o he s udy plo . We chose i e ees each om Q. mongolica, T. amu ensis, and U. da idiana and h ee ees each om C. con o e sa and A. mono, espec i ely, o he sap low measu emen s. DBH anges o 13.2−38.2 cm we e conside ed (Table 2.1). Figu e 2.2 (a) Pe cen age o basal a ea o sample ee species in he s udy plo . The basal a ea (BA) o he i e measu ed species occupied 70% o all ees in he plo . (b) S udy si e wi h speci ic loca ion and ela i e size o ees. To al plo size was 2,000 m2. Chap e 2 – Up-scaling o s and anspi a ion o an Asian empe a e mixed-deciduous o es om single ee sap low measu emen s ! 46 Table 2.1 S udied sample ees and espec i e diame e o b eas heigh (DBH), ee heigh , sapwood dep h, canopy a ea, mean sap lux densi y (Fd), and maximum Fd a M . Gyebangsan, June and July in 2008 and 2009. DBH was sapwood dep h was es ima ed by empi ical eg ession models o DBH (see Figu e 2.5). P ojec ed canopy a ea was measu ed in la e all, 2008 and sap lux densi y (Fd) was measu ed du ing June 2008 and 2009 a T ees wi h a sapwood dep h la ge han 20 mm had wo senso s in di e en dep h Species Sample ees DBH (cm) T ee heigh (m) Sapwood dep h (cm) Canopy a ea (m2) Mean Fd (g m−2 s−1) Max Fd (g m−2 s−1) Q. mongolica Q1a 27.6 14 2.9 16.0 21.9 ± 3.4 27.7 Q2a 28.4 14 3.1 37.8 15.9 ± 3.9 23.8 Q3 20.3 13 2.0 19.8 21.8 ± 3.9 30.8 Q4 13.3 12 1.4 9.1 4.3 ± 2.6 11.7 Q5a 38.2 15 4.9 46.5 24.4 ± 4.0 34.6 T. amu ensis T1 29.2 17 1.5 18.8 38.9 ± 9.3 54.5 T2 18.9 14 0.8 9.0 7.7 ± 2.8 13.5 T3 26.8 15 1.3 14.7 38.7 ± 8.9 53.2 T4 13.2 12 0.6 16.8 21.4 ± 7.8 33.9 T5 17.8 13 0.8 18.5 18.6 ± 7.7 32.7 U. da idiana U1 23.1 15 1.6 11.2 22.9 ± 5.4 32.0 U2 23.4 14 1.7 37.2 14.3 ± 2.4 18.3 U3 28.6 16 2.2 22.7 23.3 ± 3.9 30.4 U4 26.1 15 1.9 37.1 21.3 ± 4.9 30.1 U5 18.7 15 1.3 18.5 28.9 ± 4.1 35.4 C. con o e sa C1 25.3 15 1.2 62.5 34.0 ± 7.6 48.8 C2 22.3 15 1.0 41.9 33.4 ± 7.0 49.3 C3 17.6 15 0.7 28.8 15.7 ± 5.7 27.9 A. mono A1 15.0 13 0.3 13.3 15.7 ± 5.7 27.0 A2 22.0 14 0.5 38.7 30.7 ± 12.4 50.2 A3 13.8 11 0.3 12.1 31.9 ± 13.9 57.8 Chap e 2 – Up-scaling o s and anspi a ion o an Asian empe a e mixed-deciduous o es om single ee sap low measu emen s ! 47 To es ima e he sapwood a ea (AS) o he sample ees, an inc emen bo e was used o ex ac co es o sapwood a he senso ins alla ion heigh (abou 1.3 m heigh ) on same species, bu di e en ees om hose ins alled wi h he sap low senso s. Sapwood dep h was de e mined isually on hose co es since sapwood and hea wood we e clea ly dis inguishable. AS was de e mined om sapwood dep h and ee DBH based on he equa ion (Ve essy e al. 1995; Meinze e al. 2005): AS= α ⋅DBH β (1) whe e α is a cons an and β is he allome ic scaling exponen , and bo h species-speci ic coe icien s. Coe icien s o he eg ession models o each measu ed species, numbe o samples and R2 a e p o ided in he legend o Figu e 2.5. The g ound-p ojec ed c own a ea (Acp, m2) o sample ees was measu ed in eigh ho izon al di ec ions using a compass, c own mi o , and measu ing ape. The oc agonal a ea was calcula ed as he sum o eigh iangles (Schmid 2007). These esul s we e used o compu e canopy conduc ance (GC, mm s−1). 2.2.5 T ee sap low Sap lux densi y (Fd) was measu ed in he ee s ems o i e ees pe species using he he mal dissipa ion me hod (G anie 1987) du ing June and July 2008 and epea ed du ing he same pe iod in 2009. This pe iod was chosen as i was conside ed he mos ac i e pe iod in he con ex o plan wa e use, jus be o e he onse o he Monsoon ains. All senso ins alla ions we e made on he no h- acing side o he ees o a oid exposu e o he sun and minimize di ec sho -wa e adia ion (Wilson e al. 2001; Wullschlege e al. 2001). In addi ion, he senso s we e co e ed wi h a adia ion shield (S y o oam shee s wi h aluminium oil) o u he minimize he di ec he mal load. Powe o hea ing he senso s was p o ided by lead-acid ba e ies ha we e echa ged wi h sola panels ia a cha ge con olle . Each senso consis ed o a pai o 2 mm diame e p obes e ically aligned ca. 15 cm apa . Each p obe included a 0.2 mm diame e coppe - cons an an he mocouple. The wo he mocouples we e joined a he cons an an leads, so ha he ol age measu ed ac oss he coppe leads p o ided he empe a u e di e ence be ween he hea ed uppe p obe and he lowe e e ence. Hea ing ac oss he en i e leng h o he 20 mm uppe p obe was achie ed wi h a cons an cu en o 120 mA supplied o a cons an an hea ing wi e, esul ing in a hea ing powe o 200 mW (G anie 1987). Senso s we e placed in he ou e 20 mm o he sapwood (annulus 1, 0−20 mm adial sapwood dep h). In cases whe e he ee unk was la ge wi h a sapwood adius g ea e han 20 mm (Table 2.1), a second senso was implan ed 20 o 40 mm in o he sapwood. Senso s we e spaced 10–15 cm ci cum e en ially, away om he i s senso pai , on he same side o he s em o a oid azimu h di e ences. Tempe a u e di e ences we e measu ed e e y 5 min and a 30-min mean alue was logged (DL2e wi h LAC-1 in single ended mode, Chap e 2 – Up-scaling o s and anspi a ion o an Asian empe a e mixed-deciduous o es om single ee sap low measu emen s ! 48 Del a-T De ices, England). Sap lux densi y (Fd, g m−2 s−1) o each senso was calcula ed om ΔT in acco dance wi h G anie (1987), assuming ze o Fd (i.e., ΔTmax) a nigh and VPD nea ze o: Fd=119 ⋅K1.231 (2) whe e, K=(ΔTmax − ΔT) ΔT (3) T ee wa e use (TWU, kg h−1) was ob ained by mul iplying Fd by sapwood c oss-sec ional a ea (AS, m2). TWU =(Fdi ⋅ASi i=1 n ∑) (4) whe e, Fdi is sap lux densi y o he annulus i (g m−2 s−1) and ASi is sapwood a ea o he annulus i (m2). This ook in o accoun he second annulus ing, in case a second senso was ins alled in o he ee. Fo example, i = 1 was annulus ing 0−20 mm sapwood dep h, i = 2 was annulus ing 20−40 mm sapwood dep h. Canopy anspi a ion (EC, mm pe day) was compu ed by summing he con ibu ions om all he ees in he s udy plo : EC=TWUj j=1 n ∑×Aplo −1 (5) whe e, TWUj is ee wa e use o ee j (kg h−1) and Aplo is plo a ea (m2). TWU o he ees on which senso s we e no ins alled was es ima ed om he ela ionship be ween Fd and he compu ed AS o each species (Eq. 1). 2.2.6 Es ima ion o canopy conduc ance Canopy conduc ance was calcula ed om he sap low measu emen s o s and/canopy anspi a ion, in ela ion o clima e a iables: hal hou ly a e aged ai empe a u e, VPD, and canopy anspi a ion as desc ibed by Kös ne e al. (1992): GC=( ρ w⋅G ⋅Tk)⋅EC VPD (6) whe e, ρw is densi y o wa e (998 kg m−3) and G is gas cons an o wa e apo (0.462 m3 kPa kg−1 K−1), Tk is ai empe a u e (K), and EC is canopy anspi a ion (mm s−1) (Schmid 2007). To es ima e GC based on his model, da a om measu emen s be ween 10 and 15 h, when hal hou ly a es o EC we e highes , we e used. This model assumed ha ee canopies we e well coupled o he a mosphe e, so ha ae odynamic conduc ance (GA) was la ge han GC (Kös ne e al. 1992; Phillips and O en 1998). Chap e 2 – Up-scaling o s and anspi a ion o an Asian empe a e mixed-deciduous o es om single ee sap low measu emen s ! 49 2.2.7 S a is ical analyses Fd and en i onmen al a iables we e eco ded as hal hou ly a e aged alues. These a iables, including TWU and GC es ima ed om Fd we e con e ed in o daily a e ages. Da a a e p esen ed as mean ± s anda d de ia ion (SD). Fd, TWU, and GC we e compa ed be ween yea s and also among ee species using one-way ANOVA. Whe e di e ences we e ound among species, a pos -hoc K uskal-Wallis es was ca ied ou . No mali y o samples was es ablished by es ing he esiduals ob ained om he ANOVA. Measu ed and es ima ed EC by he ela ionship be ween TWU and DBH we e compa ed wi h - es . Reg ession analysis was es ed wi h Pea son co ela ion es . All s a is ical analyses including eg ession models we e based on a 0.05 signi icance le el and pe o med wi h R e sion 2.6.2 (R De elopmen Co e Team, 2008). 2.3 Resul s 2.3.1 Mic ome eo ological and soil mois u e measu emen s Daily mean ai empe a u es o e he measu emen pe iod o June and July we e abou 19.3°C in 2008 and 18.3°C in 2009. A e aged daily VPD we e 0.34 kPa in 2008 and 0.28 kPa in 2009, while he summed daily PAR we e 32.6 mol m−2 d−1 in 2008 and 25.9 mol m−2 d−1 in 2009 espec i ely (Figu e 2.3a). The o al amoun o p ecipi a ion eco ded du ing he measu emen pe iod was 89 mm in 2008, and 178.5 mm in 2009. Mean soil wa e con en (θ) wi hin he 30cm soil p o ile was 0.24 ± 0.04 m3 m−3 in 2008 and 0.21 ± 0.05 m3 m−3 in 2009 (Figu e 2.3a). Be o e he onse o ou expe imen s 2009 was compa a i ely d ie han 2008, as demons a ed by lowe θ a he beginning o measu emen s. A ains o m e en on June 3, 2009 amoun ing o 73 mm, howe e , signi ican ly aised θ ( om 0.11 o 0.29 m3 m−3), and θ he ea e was compa able o 2008. 2.3.2 T anspi a ion a e and canopy conduc ance Mean maximum Fd o he 21 ees measu ed was 247.5 ± 93.1 kg m−2 h−1 in 2008 and 271.5 ± 97.1 kg m−2 h−1 in 2009 (Table 2.2). The e was no signi ican (F = 0.88, P = 0.35) di e ence in Fd be ween he wo yea s. And also, mean maximum TWU was 21.0 ± 21.8 kg d−1 in 2008 and 32.9 ± 22.0 kg d−1 in 2009 (Table 2.2). A compa ison o maximum TWU om di e en yea s showed same esul s (F = 1.00, P = 0.33). Mean daily Fd o Q. mongolica, T. amu ensis, U. da idiana, C. con o e sa, and A. mono we e 40.9 ± 17.8 kg m−2 h−1 (n = 5), 49.5 ± 26.1 kg m−2 h−1 (n = 5), 49.2 ± 8.4 kg m−2 h−1 (n = 5), 59.5 ± 24.5 kg m−2 h−1 (n = 3) and 55.4 ± 17.8 kg m−2 h−1 (n = 3). TWU and GC a e aged o e he measu emen pe iod a e shown in Table 2.3. The mean daily TWU anged om 1.2 kg d−1 o A. mono wi h DBH o 15.0 cm o 70.1 kg d−1 o Q. Chap e 2 – Up-scaling o s and anspi a ion o an Asian empe a e mixed-deciduous o es om single ee sap low measu emen s ! 50 mongolica wi h DBH o 38.2 cm. And mean GC amoun ed om 0.7 mm s−1 o Q. mongolica wi h DBH o 13.3 cm o 16.1 mm s−1 o T. amu ensis wi h 29.2 cm. Mean maximum GC o he s and was 5.6 ± 4.8 mm s−1. The a e aged EC was 0.64 ± 0.26 mm d−1 in 2008 and 0.70 ± 0.30 mm d−1 in 2009. The maximum EC occu ed a ound day 177 in 2009 (June 26, 0.97 mm d−1, Figu e 2.4), coinciding wi h he highes daily o al PAR and VPD. The e we e no signi ican (F = 0.31, P = 0.73) di e ences in daily anspi a ion among Q. mongolica, T. amu ensis, and U. da idiana. The pe cen age mean con ibu ion o he h ee species was abou 30% each, while C. con o e sa and A. mono each accoun ed o abou 4% o he o al anspi a ion. The e was no signi ican in luence o species on Fd (P = 0.82), TWU (P = 0.19) and GC (P = 0.23). Figu e 2.3 (a) Daily mean apo p essu e de ici (VPD, kPa) and daily amoun s o pho osyn he ic ac i e adia ion (PAR, mol m−2 d−1), (b) ain all (mm d−1) and soil wa e con en (θ, m3 m−3) eco ded a he s udy si e du ing June 2008 and 2009 when sap low measu emen s we e conduc ed. Chap e 2 – Up-scaling o s and anspi a ion o an Asian empe a e mixed-deciduous o es om single ee sap low measu emen s ! 51 Table 2.2 Maximum sap lux densi y (Fd) and maximum ee wa e use (TWU) a e aged o 21 measu ed ees om June 2008 and June 2009. June 2008 June 2009 F- alue P- alue Max Fd (kg m−2 h−1) 247.48 271.46 0.8834 0.3529 SD 93.13 97.14 Max TWU (kg d−1) 21.02 32.91 0.9971 0.3264 SD 21.77 22.02 Table 2.3 Mean ee wa e use (TWU, kg d−1) and canopy conduc ance (GC, mm s−1) o indi idual ee species. Species Sample ees Mean TWU (kg d−1) Mean GC (mm s−1) Q. mongolica Q1 27.8 ± 11.6 13.1 ± 4.3 Q2 20.1 ± 9.1 3.4 ± 0.9 Q3 14.2 ± 6.4 4.3 ± 1.7 Q4 1.3 ± 0.7 0.7 ± 0.2 Q5 70.1 ± 23.4 9.7 ± 4.1 T. amu ensis T1 20.4 ± 9.7 16.1 ± 4.3 T2 1.9 ± 0.9 2.0 ± 0.8 T3 17.2 ± 5.2 6.7 ± 2.2 T4 2.6 ± 1.3 2.1 ± 0.6 T5 3.7 ± 2.0 2.4 ± 0.6 U. da idiana U1 13.7 ± 5.7 13.4 ± 2.9 U2 9.2 ± 2.9 7.0 ± 1.2 U3 22.3 ± 7.7 9.2 ± 2.9 U4 17.3 ± 5.5 5.4 ± 2.3 U5 9.1 ± 3.7 2.9 ± 0.8 C. con o e sa C1 14.7 ± 5.4 4.2 ± 1.8 C2 10.8 ± 4.0 2.8 ± 0.9 C3 2.8 ± 1.3 1.5 ± 0.4 A. mono A1 1.2 ± 0.6 2.3 ± 0.6 A2 5.4 ± 3.1 2.6 ± 0.7 A3 2.0 ± 1.0 1.6 ± 0.5 Chap e 2 – Up-scaling o s and anspi a ion o an Asian empe a e mixed-deciduous o es om single ee sap low measu emen s ! 52 Figu e 2.4 Es ima ed canopy anspi a ion (EC, mm d−1) and anspi a ion o measu ed species, Que cus mongolica (Q.m.), Tilia amu ensis (T.a.), Ulmus da idiana (U.d.), Co nus con o e sa (C.c.), and Ace mono (A.m.). 2.3.3 Rela ionship be ween ee wa e use and ee size Pa ame e s de i ed om he allome ic equa ion (Eq. 1) ela ing AS and DBH o he i e di e en species a e shown in Figu e 2.5. The con e ged eg ession model in e ed om he i e di e en species showed a s ong ela ionship (n = 35, R2 = 0.81, P <0.001) be ween AS and DBH. This model was used o compu e AS o non-measu ed species o a i e a AS o he whole s udy plo . Q. mongolica (0.15 m2 ha-1), T. amu ensis (0.29 m2 ha-1), U. da idiana (0.16 m2 ha−1), C. con o e sa (0.03 m2 ha−1), and A. mono (0.06 m2 ha−1), accoun ed o 79.3% o o al AS which was 0.87 m2 ha−1 (s udy plo = 0.2 ha). Obse ed Fd, TWU and GC we e dependen on ee size, de e mined by DBH and AS. Fd and TWU had a s onge dependency on DBH han on AS. Fo example, a eg ession o Fd in indi idual ees agains AS did no show any ela ionship (n = 21, R2 = 0.03, P >0.44), bu a eg ession be ween Fd and DBH showed ha DBH could explain 21% o he obse ed a iabili y (n = 21, R2 = 0.21, P = 0.036) in Fd among he s udied ees (Figu e 2.6a). Mo eo e , he ela ionship be ween mean daily TWU and DBH was s onge (n = 21, R2 = 0.87, P <0.001) o all measu ed species (Figu e 2.6b). SA was also signi ican ly co ela ed wi h TWU, bu less han DBH (n = 21, R2 = 0.83, P <0.001). GC had signi ican ela ionship wi h bo h DBH and AS, and also like o he ela ions GC was co ela ed be e wi h DBH (n = 17, R2 = 0.63, P <0.001) (Figu e 2.6c) han AS (n = 17, R2 = 0.48, P = 0.002). EC es ima ed om measu ed TWU was compa ed wi h EC* es ima ed om modelled TWU based on he ela ionship o DBH and TWU (Figu e 2.6b). In his s udy, he empi ical model was TWU = Chap e 2 – Up-scaling o s and anspi a ion o an Asian empe a e mixed-deciduous o es om single ee sap low measu emen s ! 53 0.0002DBH3.4302 o mean daily TWU o e he measu emen pe iod (P <0.001). The e was a s ong ag eemen (n = 22, R2 = 0.94, P <0.0001) be ween he measu ed and modeled EC*. Figu e 2.5 Rela ionship be ween sapwood a ea (AS, cm2) de e mined om inc emen co e ex ac ed om mo e han i e samples pe species and hei espec i e diame e a b eas heigh (DBH) (n = 35, R2 = 0.81, P <0.001). The eg ession equa ion o Que cus mongolica was AS = 0.2067DBH2.1 (n = 9, R2 = 0.94), o Tilia amu ensis was AS = 0.1846DBH2.0 (n = 10, R2 = 0.85), o Ulmus da idiana was AS = 0.1745DBH2.1 (n = 3, R2 = 0.98), o Co nus con o e sa was AS = 0.0757DBH2.2 (n = 3, R2 = 0.98), o Ace mono was AS = 0.0215DBH2.4 (n = 3, R2 = 0.98), and o all he s udied species was AS = 0.053DBH2.4 (n = 35, R2 = 0.81) 2.3.4 Rela ionship be ween ee wa e use and clima e ac o s Bo h VPD (R2 = 0.78 and P <0.0001) and PAR (R2 = 0.91, P <0.0001) (Figu e 2.7) could explain mos o he daily luc ua ions in EC. EC inc eased wi h inc easing VPD, a aining a maximum a VPD = 0.5 kPa, bu la e d opped a highe VPD (Figu e 2.7a). On days when VPD was high, EC inc eased du ing mo ning hou s wi h inc easing PAR and eached a maximum a a ound midday, when PAR was >1,200 µmol m−2 s−1 (Figu e 2.7b). GC o he s and co esponding o he ligh condi ions mo e han 20 mol m−2 d−1 o PAR, was plo ed agains VPD (Figu e 2.7c). The selec ed da a showed a log-linea ela ionship be ween GC and VPD (R2 = 0.69, P <0.0001). This eg ession model ag eed wi h he simpli ied model o Lohamma e al. (1980): GC=b−c⋅(lnVPD) (7) whe e, b is GC a a e e ence VPD = 1 kPa. Chap e 2 – Up-scaling o s and anspi a ion o an Asian empe a e mixed-deciduous o es om single ee sap low measu emen s ! 54 Figu e 2.6 Rela ionship be ween DBH and (a) mean sap lux densi y (Fd, kg m−2 h−1) (n = 21, R2 = 0.21, P = 0.036), (b) mean ee wa e use (TWU, kg d−1) (n = 21, R2 = 0.87, P <0.001) and (c) canopy conduc ance (GC, mm s−1) signi ican (n = 17, R2 = 0.63, P <0.001), o all he measu ed ees. Chap e 2 – Up-scaling o s and anspi a ion o an Asian empe a e mixed-deciduous o es om single ee sap low measu emen s ! 61 Luis VC, Jiménez MS, Mo ales D, Kuce a J, Wiese G (2005) Canopy anspi a ion o a Cana y islands pine o es . Ag ic Fo Me eo ol 135:117–123 Meinze FC (2003) Fuc ional con e gence in plan esponses o he en i onmen . Oecologia 134:1–11 Meinze FC, Golds ein G, And ade JL (2001) Regula ion o wa e lux h ough opical o es canopy ees: Do uni e sal ules apply? T ee Physiol 21:19–26 Meinze FC, Bond BJ, Wa en JM, Wood u DR (2005) Does wa e anspo scale uni e sally wi h ee size? Func Ecol 19:558–565 O’B ien JJ, Obe baue SF, Cla k DB (2004) Whole ee xylem sap low esponses o mul iple en i onmen al a iables in a we opical o es . Plan Cell En i on 27:551–567 O’G ady AP, Cook PG, Eamus D, Duguid A, Wischusen JDH, Fass T, Wo ldege D (2009) Con e gence o ee wa e use wi hin an a id-zone woodland. Oecologia 160:643–655 O en R, Phillips N, Ka ul G, Ewe s B, Pa aki DE (1998) Scaling xylem sap lux and soil wa e lux in o es s. Ann Sci Fo 55:191–216 O en R, Pa aki DE (2001) T anspi a ion in esponse o a ia ion in mic oclima e and soil mois u e in sou heas e n deciduous o es s. Oecologia 127:549–559 Phillips N, O en R (1998) A compa ison o daily ep esen a ions o canopy conduc ance based on wo condi ional ime-a e aging me hods and he dependence o daily conduc ance on en i onmen al ac o s. Ann Sci Fo es 55:217–235 Phillips N, O en R, Zimme mann R, W igh SJ (1999) Tempo al pa e ns o wa e lux in ees and lianas in a Panamanian mois o es . T ees 14:116–123 R De elopmen Co e Team (2009) R: a language and en i onmen o s a is ical compu ing. R Founda ion o S a is ical Compu ing, Vienna, Aus ia, ISBN 3-900051-07-0. Accessed 17 No 2009 Schmid MWT (2007) Canopy anspi a ion o beech o es s in No he n Ba a ia−S uc u e and unc ion in pu e and mixed s ands wi h oak a colline and mon ane si es. Do o al disse a ion, Uni e si y o Bay eu h, Ge many, 239 pp. URL: h p://opus.ub.uni-bay eu h.de/ oll ex e/2008/428 Se an o S, Nikinmaa E, Riikonen A, Daley M, Pe ijohn C, Mikkelsen TN, Phillips N, Holb ook NM (2008) Linking xylem diame e a ia ions wi h sap low measu emen s. Plan Soil 305:77−90 Ve essy RA, Benyon RG, O’sulli an SK, G ibben PR (1995) Rela ionships be ween s em diame e , sapwood a ea, lea a ea and anspi a ion in a young moun ain ash o es . T ee Physiol 15:559−567 Wal e H, Lie h H (1967) Klimadiag amm-Wel a las. VEB Gus a Fische Ve lag, Jena Wessman CA, Abe JD, Pe e son DL, Melillo JM (1988) Remo e sensing o canopy chemis y and ni ogen cycling in empe a e o es ecosys ems. Na u e 335:154−156 Chap e 2 – Up-scaling o s and anspi a ion o an Asian empe a e mixed-deciduous o es om single ee sap low measu emen s ! 62 Wilson KB, Hanson PJ, Mulholland PJ, Baldocchi DD, Wullschlege SD (2001) A compa ison o me hods o de e mining o es e apo a ion and i s componen s: sap- low, soil wa e budge , eddy co a iance and ca chmen wa e balance. Ag ic Fo Me eo ol 106:153−168 Wullschlege SD, Hanson PJ, Todd DE (2001) T anspi a ion om a mul i-species deciduous o es as es ima ed by xylem sap low echniques. Fo Ecol Manage 143:205−213 Zeppel M, Eamus D (2008) Coo dina ion o lea a ea, sapwood a ea and canopy conduc ance leads o species con e gence o ee wa e use in a emnan e e g een woodland. Aus J Bo 56:97−108 Chap e 3 – Wa e use by a wa m- empe a e deciduous o es unde he in luence o he Asian monsoon: Con ibu ions o he o e s o y and unde s o y o o es wa e use ! 63 Chap e 3 Wa e use by a wa m- empe a e deciduous o es unde he in luence o he Asian monsoon: Con ibu ions o he o e s o y and unde s o y o o es wa e use Jou nal o Plan Resea ch (2013) DOI 10.1007/s10265-013-0563-5 Eun-Young Junga*, Dennis O ienoa, Hyojung Kwona,c, Bo a Leea, Jong-Hwan Limb, Joon Kimc, John Tenhunena aDepa men o Plan Ecology, Uni e si y o Bay eu h, D-95440 Bay eu h, Ge many bDepa men o Fo es Conse a ion, Ko ea Fo es Resea ch Ins i u e, 130-712 Seoul, Republic o Ko ea cDepa men o Landscape A chi ec u e and Ru al Sys ems Enginee ing, Seoul Na ional Uni e si y, 151-742 Seoul, Republic o Ko ea Abs ac Unde s anding he dynamics o anspi a ion o he o e s o y (EO) and unde s o y (EU) in o es s ands unde he in luence o he Asian monsoon is needed o imp o e es ima ion o o es wa e budge and iden i y key ac o s con olling o es wa e use unde clima e change. In his s udy, EO and EU o a empe a e deciduous o es s and loca ed in Sou h Ko ea we e measu ed du ing he g owing season o 2008 using sap low me hods. The objec i es o his s udy we e o (1) quan i y he o al anspi a ion o he o es s and, (2) de e mine hei ela i e con ibu ion o ecosys em e apo anspi a ion, and (3) iden i y ac o s con olling he anspi a ion o each laye . EO and EU we e 174 and 22 mm, espec i ely. To al anspi a ion accoun ed o 55% o he o al Eeco, e ealing he impo ance o unaccoun ed con ibu ions o Eeco (i.e., soil e apo a ion and we canopy e apo a ion). Du ing he monsoon pe iod, he e was a s ong educ ion in he o al anspi a ion, likely because o educ ions in pho osyn he ic ac i e adia ion (PAR), apo p essu e de ici (VPD) and plan a ea index (PAI). The a io o EU o EO declined du ing he same pe iod, indica ing an e ec o monsoon on he pa i ioning o Eeco in i s wo componen s. The seasonal pa e n o EO was synch onized wi h he o e s o y canopy de elopmen , while EU a ied in unc ion o he unde s o y canopy de elopmen as long as he o e s o y canopy emained open. Following he o e s o y canopy closu e, en i onmen al condi ions o he unde s o y con olled he a ia ions in EU. Chap e 3 – Wa e use by a wa m- empe a e deciduous o es unde he in luence o he Asian monsoon: Con ibu ions o he o e s o y and unde s o y o o es wa e use ! 64 Key-wo ds: Asian monsoon, Canopy conduc ance, T anspi a ion o o e s o y, Unde s o y, Wa m- empe a e deciduous o es 3.1 In oduc ion Tempe a e deciduous o es s in China and Ko ea composed o Que cus, Ca pinus, Ulmus, and Tilia species ha e de eloped oge he wi h he es ablishmen o wa m- empe a e summe condi ions, a s ong dec ease in win e empe a u es and an inc ease in monsoon-induced p ecipi a ions. These o es s ha e been compa ed wi h o he majo deciduous o es ypes o he egion and a e desc ibed as wa m- empe a e deciduous o es s (Nakashizuka and Iida 1995). In gene al, wo o h ee imes highe plan species di e si y is ound in he empe a e o es s o he Asian monsoon egion han in No h Ame ica (La ham and Rickle s 1993; Qian and Rickle s 1999). This di e ence in plan di e si y appea s o esul om a g ea e physiog aphic he e ogenei y in Asia, which allowed o an allopa ic specia ion in esponse o he sea le el luc ua ions a e empe a e o es zones became disjunc in he la e Te ia y (Qian and Rickle s 2000). Wa m and humid summe condi ions due o he monsoon also con ibu e o he g ea di e si y o plan species in he eas e n Asian o es s (Röh ig and Ul ich 1991). Thus, he wa m- empe a e deciduous o es s o Ko ea and o he pa s o Asia ep esen an impo an ecosys em ype wi h mul i-laye ed physiognomy (Kim 2002), whe e bo h he o e s o y and unde s o y a e well de eloped and display di e se species composi ions. As a esul , la ge di e ences a e expec ed in he pa i ioning o wa e use be ween he o e s o y and unde s o y, which may in luence o es ecosys em unc ions. In pa icula , hese di e ences can in luence o es ene gy lows, nu ien cycling and niche pa i ioning wi hin species (Jackson e al. 1995). The species composi ion o unde s o y ege a ion is in pa de e mined by canopy ee species and s uc u e which modi y mic oclima e, ligh a ailabili y, soil wa e con en , and soil nu ien s inpu s below he canopy (Canham e al. 1994; Augus o e al. 2003; Ba bie e al. 2008). Simul aneously, unde s o y ege a ion compe es wi h o e s o y ees o esou ces and may in luence ee g ow h (Riegel e al. 1992). In empe a e deciduous o es s, ligh a ailabili y a he unde s o y is high in ea ly sp ing be o e canopy closu e, bu dec eases in ea ly summe wi h he lea eme gence o o e s o y ees. Low adia ion and wind speed below he canopy c ea e an en i onmen o ela i ely low apo p essu e g adien , esul ing in low anspi a ion a es o he unde s o y (Landsbe g and Gowe 1997). Howe e , e en in ela i ely closed canopies, he e is enough ligh eaching he unde s o y o allow he physiological unc ioning o he ege a ion (Lie e s e al. 1999). Following p ecipi a ion e en s, as he o es canopy d ies, he a e o anspi a ion o he unde s o y is expec ed o inc ease wi h he inc easing apo p essu e g adien (Black and Kellihe 1989). Howe e , i will s ill emain below he canopy anspi a ion. Thus, al hough he unde s o y may be composed o plan species di e en om hose o he canopy, he anspi a ion o he o me may s ill be igh ly coupled o canopy p ocesses gi en ha he canopy can s ongly in luence he unde s o y Chap e 3 – Wa e use by a wa m- empe a e deciduous o es unde he in luence o he Asian monsoon: Con ibu ions o he o e s o y and unde s o y o o es wa e use ! 65 mic oclima e. T ends and pa i ioning o anspi a ion a bo h le els, he e o e, usually display s ong seasonali y acco ding o he dynamics o canopy co e and lea a ea index (Wullschlege e al. 2001). E apo anspi a ion (Eeco) in such o es s ands consis s o ou main componen s, namely, o e s o y anspi a ion (EO), unde s o y anspi a ion (EU), soil e apo a ion (Eg), and e apo a ion om we canopy su ace (Ew) (Ba bou e al. 2005). Quan i ying he ela i e con ibu ion o each componen o Eeco is essen ial in o de o de e mine hei impo ance and o iden i y key ac o s con olling o es wa e use. In u n, his allows o accu a e assessmen o wa e use by o es ecosys ems, p o ides empi ical da a o he pa ame e iza ion and calib a ion o alida ion o o es hyd ological models, which equi es all hese componen s, as well as in o ma ion o he sus ainable managemen o o es s (Ha on e al. 2003; Zeppel e al. 2006). To he bes o ou knowledge, he e ha e been no a emp s o pa i ion Eeco in o EO and EU in any wa m- empe a e o es s ands unde he in luence o he Asian monsoon clima e, al hough Eeco has been widely measu ed in se e al s udies including hose o he Asia Flux g oups (Kosugi e al. 2007; Shi e al. 2008; Tanaka e al. 2008; Kang e al. 2012). The Asian monsoon, which is cha ac e ized by s ong sou hwes su ace winds and hea y ains (Lau and Li 1984), leads o o es canopy dis u bance ha a ec s canopy s uc u e and unc ion, including canopy plan a ea index (PAI). The monsoon is also gene ally cha ac e ized by ho and humid condi ions, a educ ion in apo p essu e de ici (VPD) and low pho osyn he ic ac i e adia ion (PAR). All hese changes a e likely o in luence o es wa e use di e en ly han ha o he d y-summe deciduous empe a e o es s loca ed in Eu ope and No h Ame ica. Acco ding o he clima e simula ions o Sou h Ko ea (Im e al. 2008), he equency, in ensi y and du a ion o he Asian monsoon a e likely o inc ease in he nex 30 yea s. This will al e he s uc u e and unc ion o he wa m- empe a e deciduous o es s o Asia and how hey a e managed. The e is, he e o e, an u gen need o ill he gaps in he knowledge on he in luence o he Asian monsoon on o es wa e use and i s dynamics. In his s udy, we conduc ed anspi a ion measu emen s using sap low me hods a bo h he o e s o y and unde s o y o a wa m- empe a e deciduous o es s and loca ed in Pocheon-si, Gyeonggi-do, Sou h Ko ea. Ou objec i es we e o (1) quan i y EO and EU o e he g owing season, (2) de e mine hei ela i e con ibu ion o Eeco, and (3) iden i y con olling ac o s on he o al amoun and a es o EO and EU. The speci ic ques ions add essed we e: (1) How does o e s o y de elopmen in luences unde s o y mic oclima e? (2) How is he pa i ioning o Eeco in o EO and EU in luenced by o e s o y and unde s o y de elopmen ? and (3) How does he monsoon a ec on he o al amoun and a es o EO and EU? We hypo hesized ha he con ibu ion o bo h EO and EU o Eeco is s ongly dependan on mic oclima ic ac o s, mainly PAR and VPD ha con ol canopy anspi a ion and he s a us o he o e s o y and unde s o y canopy de elopmen . Chap e 3 – Wa e use by a wa m- empe a e deciduous o es unde he in luence o he Asian monsoon: Con ibu ions o he o e s o y and unde s o y o o es wa e use ! 66 3.2 Ma e ials and me hods 3.2.1 S udy si es The s udy was conduc ed in a wa m- empe a e deciduous o es s and o he Gwangneung Na ional A bo e um loca ed in he cen al pa o he Ko ean peninsula (37°45’25.37’’N, 127°9’11.62’’E) a an ele a ion o 340 m abo e sea le el. The si e is egis e ed as a KoFlux Supe si e (Kim e al. 2006) and long- e m ecological moni o ing s a ion (Oh e al. 2000). Mean annual p ecipi a ion o e he pas 25 yea s was 1,436 mm and was mainly concen a ed in he mon hs o la e June and la e July due o he in luence o he monsoon ainband. The Ko ean monsoon is cha ac e ized by pe sis en and in ense ain all and is known as Changma. In 2008, when he measu emen s we e made, Changma s a ed on June 17 and ended on July 26 (Figu e 3.2; Ko ea Me eo ological Adminis a ion 2011). Fo he las 30 yea s, mean annual empe a u e was 11.5°C and empe a u e anged o −11.5 and 30.0°C (Kim e al. 2006). Soil dep h anges om 0.4 o 0.8 m and he soil ex u e is p edominan ly sandy loam. The bed ock p ima ily consis s o g ani e gneiss and schis (MOST 1999). The si e is loca ed on a slope o 15°, acing sou hwes . The o es s and is a climax and is domina ed by 80- o 200-yea -old Que cus se a a Thunb. ex Mu ay and Ca pinus laxi lo a (Siebold & Zucc.) Blume a . laxi lo a o an a e age heigh o 18 m (Cho e al. 2007). The unde s o y is composed o a high di e si y o species o saplings and sh ubs and has an a e age heigh o 2 m (Lim e al. 2003). 3.2.2 Mic ome eo ological measu emen s Ai empe a u e (Ta), wa e apo densi y, pho osyn he ically ac i e adia ion (PAR), ne adia ion (RN), wind speed (U) and ain all we e measu ed a 40-m heigh owe . Below he canopy, Ta, wa e apo densi y, RN, and U we e measu ed a a heigh o 4 m. Ta and U a bo h heigh we e measu ed wi h a h ee- dimensional sonic anemome e (model: CSAT3, Campbell Scien i ic Inc., Logan, U ah, USA). PAR was measu ed by pho odiode senso s (model: BPW21, Os am Semiconduc o GmbH, Regensbu g, Ge many) a a heigh o 2 m a six andom loca ions in he plo o sap low measu emen s. Ligh ansmi ance was calcula ed as he a io o PAR a he 40-m heigh (PAR40) and PAR a he 2-m heigh (PAR2). Vapo p essu e de ici a he 40-m heigh (VPD40) and a he 4-m heigh (VPD4) we e de i ed om ai empe a u e and wa e apo densi y measu ed a he espec i e heigh s (Mu ay 1967). Soil wa e con en in eg a ed o e 0 o 30 cm dep h was measu ed using soil mois u e p obes (model: CS616, Campbell Scien i ic Inc.). The da a om he eddy co a iance sys em we e sampled a 10 Hz and he me eo ological da a we e moni o ed e e y 30 seconds, and hal -hou ly means o bo h da a we e calcula ed. The da a was s o ed on h ee ypes o da a logge s (model: CR-5000 and CR-3000, Campbell Scien i ic Inc. and DL2e, Del a-T De ices, Camb idge, UK). The measu emen s o me eo ological a iables a he o e s o y and unde s o y Chap e 3 – Wa e use by a wa m- empe a e deciduous o es unde he in luence o he Asian monsoon: Con ibu ions o he o e s o y and unde s o y o o es wa e use ! 67 we e used o analyze he impac s o en i onmen al condi ions on EO and EU and o calcula e canopy conduc ance (GC) o bo h laye s. 3.2.3 Biome ic measu emen s PAI, which includes unde s o y and o e s o y ees, was measu ed e e y h ee weeks h oughou he yea using a plan canopy analyze (model: LI-2000, LI-COR Inc., Lincoln, Neb aska, USA) unde di use ligh condi ions a 12 sampling poin s wi h 50 m x 50 m g id in e al o e he plo o sap low measu emen s (Kwon e al. 2010). In o de o measu e maximum unde s o y lea a ea index (LAIU), lea samples o all he saplings in h ee plo s measu ing 2 m × 2 m we e collec ed in ea ly July, and he a ea o he lea samples was measu ed wi h a lea a ea me e (model: LI-3100, LI-COR Inc.). LAIU eached i s peak in May and did no dec ease un il ea ly Sep embe . The e o e, we assumed ha he lea a ea measu emen conduc ed in July e lec ed he maximum alue o LAIU. In o de o measu e ba k and sapwood dep h o he sample ees o Q. se a a and C. laxi lo a, an inc emen co e was used o ex ac co es om he s ems a sap low senso ins alla ion heigh . Ba k and sapwood we e isually dis inguishable o bo h species. Sapwood a ea (AS) was calcula ed om he measu emen s o diame e a b eas heigh (DBH) and dep hs o ba k and sapwood. A non-linea eg ession was es ablished be ween AS and DBH o bo h species (Figu e 3.1a). The eg ession was used o es ima e o al AS o all ees o he s udy plo (As ) (Ve essy e al. 1995; Meinze e al. 2005). Figu e 3.1 Clima e diag am Rela ionships (a) be ween s em diame e a b eas heigh (DBH) and sapwood a ea (AS) in powe unc ion (AS = 1.664DBH1.483, n = 14, R2 = 0.90, P <0.001) and (b) be ween DBH and ee wa e use (TWU) in h ee-pa ame e sigmoidal unc ion (TWU = 97.07/(1 + e−[(DBH−49.85)/14.22]), n= 11, R2 = 0.97, P <0.001) o canopy ee species o Que cus se a a (Q.s.) and Ca pinus laxi lo a (C.l.) om he yea s 2007 and 2008. Chap e 3 – Wa e use by a wa m- empe a e deciduous o es unde he in luence o he Asian monsoon: Con ibu ions o he o e s o y and unde s o y o o es wa e use ! 68 In a 30 m x 30 m plo , he DBH o all s ems la ge han 5 cm in diame e (70 s ems in o al) was measu ed. Basal a ea was hen calcula ed om he measu ed DBH. The o al basal a ea o he plo was 38.5 m2 ha-1. The dominan ee species we e Q. se a a and C. laxi lo a, cons i u ing o 71% and 22% o he o al basal a ea, espec i ely. The a e age DBH o Q. se a a and C. laxi lo a we e 45.4 ± 13 and 33.8 ± 10 cm (he ea e s anda d de ia ion (SD) is indica ed as ±), espec i ely. The unde s o y was composed o saplings and sh ubs o di e en species, such as Euonymus oxyphyllus, Cel is jessoensis, S y ax obassia, Co nus kousa and So bus alni olia. The basal a ea o he unde s o y saplings was abou 7% o he o al basal a ea. 3.2.4 Measu emen o o e s o y and unde s o y anspi a ion O e s o y (EO) and unde s o y (EU) anspi a ion we e measu ed om Ap il 1 o Sep embe 27 in 2008, and o al anspi a ion (EO + EU) du ing his pe iod was conside ed as o al annual anspi a ion since lea senescence s a ed a he end o Sep embe and no e e g een species exis ed in he s udy a ea. EO was es ima ed om sap lux densi y (Fd, g m−2 s−1) measu ed using he mal dissipa ion p obes (TDP), which we e cons uc ed a he echnical labo a o y o he Depa men o Plan Ecology, Uni e si y o Bay eu h, based on he o iginal design by G anie (1987). The TDP consis ed o wo p obes (2 mm in diame e and 20 mm in leng h) aligned e ically 10 cm apa in o he sapwood. Each p obe con ained a coppe -cons an he mocouple and was connec ed in pa allel (G anie 1987). The empe a u e o he uppe p obe, cons an ly hea ed by a 0.2 W powe supply, was in luenced by he a e o con ec i e hea anspo away om he hea sou ce, wi h he e ically lowing sap in he xylem. The lowe p obe, which was an unhea ed e e ence, e lec ed he ambien sap empe a u e. The p obes we e inse ed in wo dep hs be ween 0 and 20 mm o he ou e ing and be ween 20 and 40 mm o he inne ing, a a heigh o 1.3 m abo e g ound, on he no h- acing side o minimize he e ec s o di ec sho wa e adia ion (Wilson e al. 2001; Wullschlege e al. 2001). A e ins alla ion, each p obe was co e ed wi h S y o oam shea hs and aluminum oil o a oid di ec he mal load om adia ion. Tempe a u e di e ences (ΔT) be ween he uppe and lowe e e ence p obes we e measu ed e e y 30 seconds and hal -hou ly means we e calcula ed. Da a we e eco ded on da a logge s (model DL2e, Del a-T De ices). Fd was calcula ed as a unc ion o ΔT acco ding o s anda d calib a ion o he TDPs (e.g., G anie , 1987), Fd=119 ⋅[(ΔTmax − ΔT)⋅ ΔT−1]1.231 (1) whe e ΔTmax is he maximum empe a u e di e ence be ween he wo p obes du ing a day when sap low was null. Fd a di e en sapwood dep hs o he same indi idual was used o calcula e he sapwood a ea weigh ed sap low densi y (Fd ) wi h he ollowing equa ion: Fd =(Fdo ⋅Aso +Fdi ⋅Asi )⋅(Aso +Asi )−1 (2) Chap e 3 – Wa e use by a wa m- empe a e deciduous o es unde he in luence o he Asian monsoon: Con ibu ions o he o e s o y and unde s o y o o es wa e use ! 69 whe e Fdo is he sap lux densi y o he ou e ing, Aso is he a ea o he ou e ing, Fdi is he sap lux densi y o he inne ing and Asi is he a ea o he inne ing. We used a weigh ed a e age Fd o es ima e o al o es anspi a ion because Fdi was lowe han Fdo in Q. se a a and C. laxi lo a by 0.2 and 0.8, espec i ely. Fd in he sapwood deepe han 40 mm sapwood dep h, i.e., ou side he hea ing p obe, was es ima ed by applying an empi ical unc ion desc ibed by Poya os e al. (2007) o Que cus spp. and Gebaue e al. (2008) o Ca pinus spp. EO (mm h−1) was calcula ed as: EO=Fd ⋅As ⋅Aplo −1 (3) whe e Fd is he mean Fd o he sample ees, As is he o al sapwood a ea o all ees in he plo , and Aplo is he plo a ea. In o de o de e mine ci cum e en ial di e ences in sap low, we ins alled mul iple sap low p obes, i.e., h ee pe species, wi hin he ou e 20 mm sapwood on he sou h- and no h- acing side o he s em om July o Oc obe . The wo se s o he p obes (on each ee) we e placed a ela i ely simila heigh s, bu ca e ully posi ioned apa so ha hea ing om one senso on he sou he n azimu h did no in e e e wi h he o he on he no h-side. Howe e , we obse ed no ci cum e en ial a ia ion in Fd be ween he sou h and no h sides o he o e s o y ee s ems (P <0.0001), hence we did no pe o m any u he co ec ions on EO. In o de o measu e EO, we selec ed h ee indi iduals o Q. se a a (dominan species) and h ee indi iduals o C. laxi lo a (sub-dominan species) wi h DBH be ween 24 and 59 cm and heigh s be ween 12 and 18 m. Fo he measu emen o EU, h ee indi iduals o E. oxyphyllus, one indi idual o C. jessoensis and one indi idual o S. alni olia saplings wi h s em diame e s anging om 1.7 o 3.0 cm (a 1 m abo e he g ound) and heigh s anging om 1.8 o 2.5 m we e selec ed (Table 3.1). The choice o a species was based on he heal h s a us and i s dominance wi hin he plo . The numbe o indi iduals used o he measu emen s was se e ely es ic ed by he Na ional A bo a um managemen policy since he Gwangneung o es is one o he only na u al o es s a climax whi hin Ko ea. These o es s ha e been placed unde s ic p o ec ion. Any des uc ion o ees o his o es is, he e o e, s ongly p ohibi ed. Howe e , he scaling o sap low measu emen s om plo -scale o s and-scale equi es ha a sample be ep esen a i e o he spa ial dis ibu ion o species and size classes wi hin a s and (Kös ne e al. 1998; O en e al. 1998; Kumagai 2005). The e o e, we used he esul s o p e ious measu emen s o compa ison. In he p e ious yea (2007), sap low measu emen s we e comduc ed on i e indi iduals o Q. se a a ees (dominan species), wi h DBH dis ibu ion wi hin he classes o 10−20, 20−30, 30−40, 40−50 and 50−60 cm which a e ep esen a i e o he DBH classes in his o es s and (unpublished da a). We examined he ela ionship be ween DBH and a es o sap low a e combining da a om he yea s 2007 and 2008. DBH explained 97% o he a ia ions in wa e use among ees in a sigmoidal unc ion ela ionship desc ibed by Meinze e al. (2005) (Figu e 3.1b), which ag ees wi h he indings o p e ious s udies in Chap e 3 – Wa e use by a wa m- empe a e deciduous o es unde he in luence o he Asian monsoon: Con ibu ions o he o e s o y and unde s o y o o es wa e use ! 70 simila o es ypes (Ve essy e al. 1995; Meinze 2003; Meinze e al. 2005; Jung e al. 2011). Thus, despi e he limi ed sample size used in subsequen analyses, we a e con inced ha he da a a e well ep esen a i e o he o es s and and ha he ype II e o is wi hin accep able limi s. Table 3.1 Cha ac e is ics o he sample ees o sap low measu emen s. DBH, AS, and AL indica e diame e a b eas heigh , sap wood a ea o he o e s o y ees, and lea a ea o he unde s o y ees, espec i ely. Sample ees T ee species T ee heigh [m] DBH [cm] AS/AL [m2] Q1 Que cus se a a 15 35.4 0.03 Q2 17 58.8 0.07 Q3 17 38.8 0.03 C1 Ca pinus laxi lo a 18 43.0 0.07 C2 16 33.0 0.04 C3 12 24.1 0.02 SHB1 Euonymus oxyphyllus 1.8 1.9 0.83 SHB2 2.0 2.1 0.98 SHB3 1.9 1.7 0.90 SHB4 Cel is jessoensis 2.5 2.0 1.08 SHB5 So bus alni olia 2.5 3.0 1.39 EU was es ima ed om sap low (F, g h−1), which was measu ed by he s em hea balance (SHB) echnique (Saku a ani 1981). This echnique has been success ully applied o accu a ely es ima e sap low o he baceous s ems (Bake and an Ba el 1987), saplings (Lei e al. 2010), and b anches o la ge ees (O ieno e al. 2007) wi h diame e s anging om 2 o 125 mm. The SHB senso s used o he measu emen s we e manu ac u ed a he elec onic wo kshop, Uni e si y o Bay eu h, acco ding o he o iginal design by Saku a ani (1981), imp o ed by Weibel and de Vos (1994) o s ems o b anches wi h diame e anging Chap e 3 – Wa e use by a wa m- empe a e deciduous o es unde he in luence o he Asian monsoon: Con ibu ions o he o e s o y and unde s o y o o es wa e use ! 77 Table 3.2 Mon hly mean o e s o y anspi a ion (EO), unde s o y anspi a ion (EU), e apo anspi a ion (Eeco) and mon hly accumula ed ain all du ing he g owing season o 2008. Numbe s in pa en hesis indica e s anda d de ia ion. Figu e 3.4 Rela ionship be ween mon hly cumula i e ain all and he a io o mon hly a e aged s and anspi a ion (= EO + EU) o mon hly a e aged ecosys em e apo anspi a ion (Eeco). The ela ionship was signi ican (R2 = 0.71, P <0.01, y = −0.131ln(x) + 1.3247). Mon h EO [mm d−1] EU [mm d−1] Eeco [mm d−1] EU/EO (EO + EU)/Eeco Rain all [mm mon−1] Ap 0.64 (± 0.4) 0.14 (± 0.1) 0.87 (± 0.3) 0.22 0.89 24.0 May 1.20 (± 0.4) 0.17 (± 0.1) 1.71 (± 0.7) 0.14 0.80 82.0 June 1.39 (± 0.4) 0.16 (± 0.1) 2.27 (± 1.1) 0.11 0.68 136.5 July 0.80 (± 0.5) 0.07 (± 0.1) 1.50 (± 0.8) 0.09 0.58 630.0 Aug 0.99 (± 0.3) 0.11 (± 0.1) 2.60 (± 1.0) 0.11 0.42 291.5 Sep 0.98 (± 0.4) 0.10 (± 0.1) 1.45 (± 0.9) 0.10 0.74 121.0 Chap e 3 – Wa e use by a wa m- empe a e deciduous o es unde he in luence o he Asian monsoon: Con ibu ions o he o e s o y and unde s o y o o es wa e use ! 78 Figu e 3.5 Rela ionships be ween anspi a ion and pho osyn he ic ac i e adia ion (PAR) and be ween anspi a ion and apo p essu e de ici (VPD). Numbe s o 40, 4, and 2 indica e 40-m, 4-m, and 2-m heigh s, espec i ely. No e ha PAR2 was mul iplied by a ac o 10 o he con enience o plo ing wi h PAR40 excep o Ap il. Closed and open ci cles indica e o e s o y and unde s o y, espec i ely. All he eg essions a e in polynomial unc ions. Chap e 3 – Wa e use by a wa m- empe a e deciduous o es unde he in luence o he Asian monsoon: Con ibu ions o he o e s o y and unde s o y o o es wa e use ! 79 3.3.5 Canopy conduc ance Mean day ime GC o he o e s o y was signi ican ly highe be o e Changma han a e ( - es , d = 42, = 5.02, P <0.0001). Thus, mean day ime GC o he o e s o y was 8.5 ± 4.5 mm s−1 be o e (May 24 o June 16) Changma compa ed o 4.5 ± 1.4 mm s−1 a e (Augus ) Changma (Table 3). Simila ly, mean day ime GC o he unde s o y dec eased om 2.6 ± 1.0 mm s−1 o 1.9 ± 1.2 mm s−1 a e Changma. The di e ence was, howe e , no s a is ically signi ican . Declining end o GC e was also ound a e Changma, showing GC e o he o e s o y a 12.8 mm s−1 be o e Changma compa ed o 11.9 mm s−1 a e Changma and GC e o he unde s o y a 2.2 mm s−1 and 1.0 mm s−1 (Table 3.3). Day ime GC o he o e s o y exponen ially and signi ican ly declined wi h inc easing day ime VPD, (R2 = 0.46, P = 0.0012 be o e Changma; R2 = 0.66, P <0.0001 a e Changma), while he ela ionship be ween GC o he unde s o y and VPD4 was no signi ican (Figu e 3.6). The a io o −m/GC e , which indica es he a ia ions o −m ( he sensi i i y o GC esponse o VPD; see Eq. 9) o GC e , o he o e s o y was 0.66 and 0.68 be o e and a e Changma, espec i ely, and ha o he unde s o y was 0.81 and 0.91. The obse ed – m/GC e a io o he o e s o y was close o he heo e ical a io o 0.6 (O en e al. 1999), while ha o he unde s o y was highe han he heo e ical a io (Table 3.3). Table 3.3 Mean alues o day ime apo p essu e de ici (VPD, kPa), daily sum o pho osyn he ic ac i e adia ion (PAR, mol m−2 d−1), canopy conduc ance (GC, mm s−1), canopy conduc ance a VPD = 1 kPa (GC e , mm s−1) and he a io o he sensi i i y o GC esponse o VPD o GC e (−m/GC e ) a he o e s o y and unde s o y o he pe iod be o e and a e Changma. Numbe s in pa an hesis indica e s anda d de ia ion. Be o e Changma (May 24−June 16) A e Changma (Augus 1−31) O e Unde O e Unde VPD 1.4 (± 0.5) 0.7 (± 0.4) 2.1 (± 0.5) 0.7 (± 0.3) PAR 42.4 (± 15.4) 2.3 (± 0.8) 39.9 (± 12.7) 2.7 (± 1.2) GC 8.5 (± 4.5) 2.6 (± 1.0) 4.5 (± 1.5) 1.9 (± 1.2) GC e 12.8 2.2 11.9 1.0 –m/GC e 0.66 0.68 0.81 0.91 Chap e 3 – Wa e use by a wa m- empe a e deciduous o es unde he in luence o he Asian monsoon: Con ibu ions o he o e s o y and unde s o y o o es wa e use ! 80 Figu e 3.6 Rela ionships be ween day ime canopy conduc ance (GC) and day ime apo p essu e de ici (VPD) o he o e s o y (closed) and he unde s o y (opened) be o e Changma (June 1−June 16) ( iangles) and a e Changma (Augus 1 o Augus 30) (ci cles). 3.4 Discussion 3.4.1 Regula ion o he o e s o y and unde s o y anspi a ion The seasonal pa e n o EO was synch onized o he seasonal de elopmen o lea a ea un il July, a he onse o he monsoon pe iod. Simila ly, pa e ns o EU ea ly in he season e lec ed he pa e n o LAIU de elopmen . Howe e , his ela ionship became dis o ed wi h inc easing de elopmen and closu e o he o e s o y canopy. A e ull o e s o y canopy de elopmen , EU was mainly in luenced by he p e ailing en i onmen al condi ions in he unde s o y. The unde s o y canopy had an ea lie bud b eak han he o e s o y canopy and he o me a ained i s maximum lea a ea ea lie han he la e . As a esul , maximum EU occu ed in la e Ap il, coinciding wi h he highes lea a ea, highes PAR and VPD in he unde s o y, a a ime when he o e s o y canopy was no ye ully de eloped. Maximum EO occu ed in ea ly June, coinciding wi h he peak o e s o y lea a ea. Di e ences in iming o he canopy de elopmen o he unde s o y and o e s o y ha e been epo ed o a simila o es ype in Japan (Nasaha a e al. 2008). In his empe a e o es s and, he unde s o y, he e o e, accoun s o mos o he o al lea a ea ea ly in he g owing season, since o e s o y lea de elopmen is delayed. These phenological di e ences be ween he wo laye s esul in o empo al shi s in unc ionali ies, in e ms o o es anspi a ion. Simila obse a ions ha e been Chap e 3 – Wa e use by a wa m- empe a e deciduous o es unde he in luence o he Asian monsoon: Con ibu ions o he o e s o y and unde s o y o o es wa e use ! 81 made by Unswo h e al. (2004), who epo ed shi s in he EO o EU a io du ing he g owing season due o he empo al di e en ia ion in phenology and canopy de elopmen o he dominan species o he wo laye s. Responses o EO and EU o he co esponding PAR and VPD we e simila h oughou he ege a i e pe iod, excep in Ap il when he lea es in bo h laye s we e in hei de eloping s ages. The beha io obse ed in Ap il is likely due o p ema u e lea es and less de eloped s oma a a his s age (Hiyama e al. 2005). Th oughou mos o he g owing season, excep in July, EO sa u a ed on he days when PAR eached 40 o 60 mol m−2 d−1, which a mos imes co esponded o a daily VPDmax o 1.8 o 2.0 kPa. EU, howe e , ne e eached sa u a ion likely due o he low ligh and VPD en i onmen in he unde s o y. Al hough PAR a he unde s o y was sligh ly highe in Augus and Sep embe , due o canopy opening (dec eased PAI) caused by he monsoon and yphoons, his change in PAR did no in luence he magni ude o EU since his pe iod was cha ac e ized by low VPD, low s oma al conduc ance and aging lea es. Acco ding o Kang e al. (2009b; 2010), decoupling coe icien (Ω), de ined as a deg ee o decoupling be ween ege a ion and he a mosphe e, was, on a e age, equal o 0.39 o he o e s o y and 0.15 o he unde s o y du ing he g owing season a Gwangneung. This was compa able wi h Ω o 0.35 o he o e s o y and 0.23 o he unde s o y in he empe a e deciduous oak o es in Tennessee whose o al con ibu ion o anspi a ion o Eeco was simila wi h ha o Gwangneung. Low Ω a bo h laye s demons a ed a g ea e dependence o EO and EU on VPD han PAR. Ano he no able obse a ion was he signi ican d op in he magni udes o EO and EU du ing and a e Changma. This was a ibu ed o he s ong decline in PAR and VPD (Figu e 3.2), he dec ease in PAI and he ad anced lea age. A dec ease in s oma al conduc ance wi h inc easing lea age has been epo ed in o he s udies (Field and Mooney 1983; Radoglou 1996; Rey and Ja is 1998). Consis en wi h his, a e Changma dec eased GC and GC e o bo h he o e s oy and unde s o y we e obse ed in ou s udy. Mo eo e , he a io o –m/GC e o bo h laye s was sligh ly inc eased a e Changma, which sugges s an inc eased s oma al sensi i i y o VPD due o aging lea es. G ea e –m/GC e o he unde s o y, compa ed o he o e s o y and he heo e ical a io o 0.6 (O en e al. 1999) migh be because o smalle VPD ange in he unde s o y (0.6–1.5 kPa) (He bs e al. 2008). He bs e al. (2008) epo ed ha he e is a endency o inc easing he alue o –m/GC e o low anges o VPD. These changes in GC and s oma al sensi i i y oge he wi h he al e ed en i onmen al ac o s a e Changma, may explain he decline in EO and EU a e Changma. 3.4.2 Pa i ioning o ecosys em wa e use The maximum a e o EO obse ed in his o es s and was 1.9 mm d−1 (Figu e 3.4). This alue is wi hin he ange o maximum EO a es epo ed o simila o es ypes loca ed wi hin he same la i udinal ange. Fo Chap e 3 – Wa e use by a wa m- empe a e deciduous o es unde he in luence o he Asian monsoon: Con ibu ions o he o e s o y and unde s o y o o es wa e use ! 82 example, he maximum EO epo ed o a empe a e deciduous o es s and in Duke, No h Ca olina, wi h PAI o 5.4, was 2.0 mm d−1 (O en and Pa aki 2001). Wullschlege e al. (2001) epo ed a alue o EO o 2.2 mm d−1 in a empe a e deciduous oak o es s and in Tennessee wi h a PAI o 6.2. On he o he hand, he a e age EU was 0.1 ± 0.1 mm d−1, which con ibu ed app oxima ely 13% o he o al s and anspi a ion amoun ing o 196 mm p.a. This p opo ion o EU o o al s and anspi a ion is simila o hose epo ed o o he empe a e o es s ands. EU o a simila empe a e oak o es s and accoun ed o 17% o o al s and anspi a ion (Wullschlege e al. 2001). Baldocchi and Vogel (1996) epo ed 5 o 25% con ibu ion o he unde s o y o he o al s and anspi a ion o a simila deciduous o es s and in No h Ame ica. Ou o al s and anspi a ion was, howe e , app oxima ely 30% lowe compa ed o alues measu ed in he empe a e deciduous o es s o No h Ame ica (i.e., Wullschlege e al. 2001). Despi e simila i ies in magni udes o he annual mean RN and Ta and annual p ecipi a ion be ween he wo si es, o al s and anspi a ion is lowe , in compa ison, a ou Asian s udy si e because o di e ences in dis ibu ion pa e ns o hese en i onmen al d i e s. In o al, EO and EU con ibu ed 47% and 8% o he annual Eeco, espec i ely. Consequen ly, he emaining 45% o Eeco o igina ed om Eg and Ew. P e ious s udies conduc ed a he same o es s and es ima ed 16% con ibu ion om Eg (Kang e al. 2009b) and 13~32% om Ew o annual Eeco (Kang e al. 2012). Kang e al. (2009b) epo ed ha Eg was high in Ma ch and No embe when ligh a ailabili y a he o es loo was high, bu was negligible du ing he g owing season when ligh a ailabili y a he o es loo and VPD we e low. E en hough Ew epo ed by Kang e al. (2012) was highly a iable, i s ela i ely high con ibu ion o Eeco e eals ha Ew is a majo componen o he pa i ioning o ecosys em wa e use, especially du ing he g owing season. The con ibu ion o s and anspi a ion (EO + EU) o Eeco a ied seasonally. Be ween Ap il and May, s and anspi a ion accoun ed o 80 o 89% o he o al wa e loss om he o es . This alue declined o 58% in July and 42% in Augus (Table 3.2). The decline in July was p ima ily due o he dec ease in EO in esponse o he declining PAR and VPD associa ed wi h Changma and PAI educ ion due o hea y ain all and s ong winds. Howe e , he educed con ibu ion o anspi a ion o Eeco in July and Augus may be due o he unaccoun ed amoun o Ew, which was ound o be highes in hese mon hs, co esponding wi h in ensi ied ain all (Kang e al. 2012). In Sep embe , he p opo ion inc eased o 74% as anspi a ion became he dominan con ibu o o Eeco. The pa i ioned componen s o Eeco in ou s udy indica e he impo ance o he unaccoun ed con ibu ions o Ew o Eeco o Gwangneung o es s and du ing he g owing season. Chap e 3 – Wa e use by a wa m- empe a e deciduous o es unde he in luence o he Asian monsoon: Con ibu ions o he o e s o y and unde s o y o o es wa e use ! 83 3.5 Conclusions The unde s o y con ibu es signi ican ly o he o al annual o es wa e budge . Since bud b eak in he unde s o y occu s almos one mon h ea lie han he o e s o y canopy, he unde s o y becomes he dominan sou ce o anspi a ion wa e loss ea ly in he season. I s dominance is, howe e , subdued wi h he de elopmen and closu e o he o e s o y canopy, e en hough i s ill accoun s o a signi ican p opo ion o he o al o es PAI mos o he yea . The seasonal pa e n and con ibu ion o EO o Eeco, howe e , a e s ongly synch onized o he o e s o y canopy de elopmen . Despi e he unde s o y comp ising o di e en plan species, i s ole in o al o es wa e budge seems o be go e ned by p ocesses o he o e s o y. Conside ing he daily maximum and mean a es o EO and EU and hei con ibu ions o Eeco, he wa m Asian empe a e deciduous o es is unc ionally (in e ms o wa e use) simila o he empe a e oak o es s in No h Ame ica. To al s and anspi a ion in he empe a e o es s in Asia, howe e , is lowe in compa ison, likely due o he dep ession in anspi a ion du ing he monsoon as a esul o decline in canopy PAI, PAR and VPD, sugges ing ha ecosys em p ocess (i.e., EO and EU) is decoupled om ain all inpu . Independen es ima es o EO and EU and hei pa i ioning o Eeco in his s udy can imp o e ou unde s anding o he seasonal dynamics o o es wa e use, wi h a holis ic assessmen and enhancemen o he pe o mance o o es hyd ology models, by highligh ing he need o inco po a ing sepa a e canopy de elopmen s age o he o e s o y and unde s o y, in he case, egions unde he in luence o monsoon clima es. 3.6 Acknowledgemen s This s udy was ca ied ou as pa o he In e na ional Resea ch T aining G oup TERRECO (GRK 1565/1) unded by he Deu sche Fo schungsgemeinscha (DFG) in coope a ion wi h he Uni e si y o Bay eu h, Ge many and he Ko ean Resea ch Founda ion (KRF) a Kangwon Na ional Uni e si y, Chuncheon, S. Ko ea and a g an (code: 1-8-3) om Sus ainable Wa e Resou ces Resea ch Cen e o 21s Cen u y F on ie Resea ch P og am. We would like o acknowledge he inpu om Ms. Ma ga e e Wa inge o Plan Ecology Depa men , Uni e si y o Bay eu h and s uden s o Seoul Na ional Uni e si y and Kangwon Na ional Uni e si y o hei suppo du ing ieldwo k. 3.7 Re e ences Augus o L, Dupouey JL, Range J (2003) E ec s o ee species on unde s o y ege a ion and en i onmen al condi ions in empe a e o es s. Ann Fo Sci 60:823−831 Chap e 3 – Wa e use by a wa m- empe a e deciduous o es unde he in luence o he Asian monsoon: Con ibu ions o he o e s o y and unde s o y o o es wa e use ! 84 Baldocchi DD, Vogel C (1996) A compa a i e s udy o wa e apo , ene gy and CO2 lux densi ies abo e and below a empe a e b oadlea and a bo eal pine o es , T ee Physiol 16:5−16 Ba bie S, Gosselin F, Balandie P (2008) In luence o ee species on unde s o y ege a ion di e si y and mechanisms in ol ed–A c i ical e iew o empe a e and bo eal o es s. Fo Ecol Manage 254:1−15 Ba bou MM, Hun JE, Walc o AS, Roge s GND, McSe eny TM, Whi ehead D (2005) Componen s o ecosys em e apo a ion in a empe a e coni e ous ain o es , wi h canopy anspi a ion scaled using sapwood densi y. New Phy ol 165:549−558 Black TA, Kellihe FM (1989) P ocesses con olling unde s o ey e apo anspi a ion. Philos. T ans R Soc London B 324:207−229 Campbell GS, No man JM (1998) An in oduc ion o en i onmen al biophysics. Sp inge , New Yo k Canham DC, Finzi AC, Pacala SW, Bu bank DH (1994) Causes and consequences o esou ce he e ogenei y in o es s: in e speci ic a ia ion in ligh ansmission by canopy ees. Can J Fo Res 24:337−349 Cho YC, Shin HC, Kim SS, Lee CS (2007) Dynamics and Conse a ion o he Gwangneung Na ional Fo es in Cen al Ko ea: A Na ional Model o Fo es Res o a ion. J Plan Biol 50:615−625 Ewe s BE, O en R (2000) Analyses o assump ions and e o s in he calcula ion o s oma al conduc ance om sap lux measu emen s. T ee Physiol 20:579−589 Field C, Mooney HA (1983) Lea age and seasonal e ec s on ligh , wa e and ni ogen use e iciency in a Cali o nia sh ub. Oecologia 56:348−355 G anie A (1987) E alua ion o anspi a ion in a Douglas- i s and by means o sap low measu emen s. T ee Physiol 3:309−320 G ime VL, Sinclai FL (1999) Sou ces o e o in s em hea balance sap low measu emen s. Ag ic Fo Me eo ol 94:103−121 Ham JM, Heilman JL, Lascano RJ (1990) De e mina ion o sol wa e e apo a ion and anspi a ion om ene gy balance and s em low measu emen s. Ag ic Fo Me eo ol 52:287−301 Ha on TJ, Rup ech J, Geo ge RJ (2003) P eclea ing hyd ology o he Wes e n Aus alia whea bel : Ta ge o he u u e. Plan Soil 257:341–356 He bs M, Rosie PTW, Mo ec o MD, Gowing DJ (2008) Compa a i e measu emen s o anspi a ion and canopy conduc ance in wo mixed deciduous woodland di e ing in s uc u e and species composi ion. T ee Physiol 28:959−970 Hiyama T, Kochi K, Kobayashi N, Si isampan S (2005) Seasonal a ia ion in s oma al conduc ance and physiological ac o s obse ed in a seconda y wa m- empe a e o es . Ecol Res 20:333−346 Hong J, Kwon H, Lim JH, Byun YH, Lee J, Kim J (2009) S anda diza ion o KoFlux eddy-co a iance da a p ocessing. Ko J Ag ic Fo Me eo ol 2:12−26 Chap e 3 – Wa e use by a wa m- empe a e deciduous o es unde he in luence o he Asian monsoon: Con ibu ions o he o e s o y and unde s o y o o es wa e use ! 85 Im ES, Ahn JB, Kwon WT, Gio gi F (2008) Mul i-decadal scena io simula ion o e Ko ea using a on-way doubled-nes ed egional clima e model sys em. Pa 2: u u e clima e p ojec ion (2021-2050). Clim Dyn 30:239−254 Jackson PC, Ca elie J, Golds ein G, Meinze FC, Holb ook NM (1995) Pa i ioning o wa e esou ces among plan s o a lowland opical o es . Oecologia 101:197−203 Jung EY, O ieno D, Lee B, Lim JH, Kang SK, Schmid MWT, Tenhunen J (2011) Up-scaling o s and anspi a ion o an Asian empe a e mixed-deciduous o es om sing ee sap low measu emen s. Plan Ecol 212:383−395 Kang M, Pa k S, Kwon H, Choi HT, Choi YJ, Kim J (2009a) E apo anspi a ion om a deciduous o es in a complex e ain and a he e ogeneous a mland unde monsoon clima e. Asia-Paci ic J A mos Sci 45:175−191 Kang M, Kwon H, Lim JH, Kim J (2009b) Unde s o y e apo anspi a ion measu ed by Eddy-Co a iance in Gwangneung deciduous and coni e ous o es s. Ko J Ag ic Fo Me eo ol 11:233−246 Kang M, Kwon H, Cheon JH, Kim J (2012) On es ima ing we canopy e apo a ion om deciduous and coni e ous o es s om Asian monsoon clima e, J Hyd ome eo l 13:950−965 Kim J, Lee D, Hong JY (2006) Hyd oKo ea and Ca boKo ea: c oss-scale s udies o ecohyd ology and biogeochemis y in a he e ogeneous and complex o es ca chmen o Ko ea. Ecol Res 21:881−889 Kim JH (2002) Communi y ecological iew o he na u al deciduous o es in Ko ea. In: Ecology o Ko ea, Bumwoo, Seoul Ko ea Me eo ological Adminis a ion (2011) A whi e pape on Changma. Kosugi Y, Takanashi S, Tanaka H, Ohkubo S, Tani M, Yano M, Ka ayama T (2007) E apo anspi a ion o e a Japanese cyp ess o es . I. Eddy co a iance luxes and su ace conduc ance cha ac e is ics o 3 yea s. J Hyd ol 337:269−283 Kös ne BMM, Schulze ED, Kellihe FM, Hollinge DY, Bye s JN, Hun JE, McSe eny TM, Mese h R, Wei PL (1992) T anspi a ion and canopy conduc ance in a p is ine b oad-lea ed o es o No ho agus: an analysis o xylem sap low and eddy co ela ion measu emen s. Oecologia 91:350−359 Kös ne B, G anie A, Ce mák J (1998) Sap low measu emen s in o es s ands: me hods and unce ain ies. Ann Sci Fo 55:13−27 Kumagai T, Aoki S, Nagasawa H, Mabuchi T, Kubo a K, Inoue S, U sumi Y, O suki K (2005) E ec s o ee- o- ee and adial a ia ions on sap low es ima es o anspi a ion in Japanese ceda . Ag ic Fo Me eo ol 135:110−116 Kwon H, Kim J, Hong J, Lim JH (2010) In luence o he Asian monsoon on ne ecosys em ca bon exchange in wo majo ecosys ems in Ko ea. Biogeoscience 7:1493−1504 Chap e 3 – Wa e use by a wa m- empe a e deciduous o es unde he in luence o he Asian monsoon: Con ibu ions o he o e s o y and unde s o y o o es wa e use ! 86 Landsbe g JJ, Gowe ST (1997) Applica ions o Physiological Ecology o Fo es Managemen . Academic P ess, San Diego La ham RE, Rickle s RE (1993) Global pa e ns o ee species ichness in mois o es s: ene gy-di e si y heo y does no accoun o a ia ion in species ichness. Oikos 67:325−333 Lau K-M, Li M-T (1984) The monsoon o E. Asia and i s global associa ions−A su ey. Ame Me Soc 65:112−125 Lei H, Zhang ZS, Li XR (2010) Sap low o A emisia o dosica and he in luence o en i onmen al ac o s in a e ege a ed dese a ea: Tengge Dese , China. Hyd ol P oc 24:1248−1253 Lie e s VJ, Messie C, S ad KJ, Gend on F, Comeau PG (1999) P edic ing and managing ligh in he unde s o y o bo eal o es s. Can J Fo Res 29:796−811 Lim JH, Shin JH, Jin GJ, Chun JH, Oh JS (2003) Fo es s and s uc u e, si e cha ac e is ics and ca bon budge o he Kwangneung na u al o es in Ko ea. Ko J Ag ic Fo Me eo ol 5:101−109 Magnani F, Leona di S, Togne i R, G ace J, Bo ghe i M (1998) Modeling he su ace conduc ance o a b oad-lea canopy: e ec s o pa ial decoupling om he a mosphe e. Plan Cell En i on 21:867−879 Meinze FC (2003) Func ional con e gence in plan esponses o he en i onmen . Oecologia 134:1−11 Meinze FC, Bond BJ, Wa en JM, Wood u DR (2005) Does wa e anspo scale uni e sally wi h ee size? Func Ecol 19:558−565 Mon ei h JL (1965) E apo a ion and en i onmen . Symp Soc Exp Biol 19:205–234 MOST (1999) Geologic map o Seoul-Namchon epo . Minis y o Science and Technology, Ko ea Mu ay FW (1967) On he compu a ion o sa u a ion apo p essu e. J Appl Me eo 6:203−204 Nakashizuka T, Iida S (1995) Composi ion, dynamics and dis u bance egime o empe a e deciduous o es s in Monsoon Asia. Vege a io 121:23−30 Nasaha a KN, Mu aoka H, Nagai S, Mikami H (2008) Ve ical in eg a ion o lea a ea index in Japanese deciduous b oad-lea ed o es . Ag ic Fo Me eo ol 148:1136−1146 Oh JS, Shin JH, Lim JH (2000) Long- e m ecological esea ch p og amme in Ko ea Fo es Ins i u e. Ko J Ecol 23:131−134 O en R, Phillips N, Ka ul G, Ewe s B, Pa aki E (1998) Scaling xylem sap lux and soil wa e balance and calcula ing a iance: A me hod o pa i ioning wa e lux in o es s. Ann Sci Fo 55:191−216 O en R, Spe y JS, Ka ul GG, Pa aki DE, Ewe s BE, Phillips N, Schä e KVR (1999) Su ey and syn hesis o in a- and in e speci ic a ia ion in s oma al sensi i i y o apou p essu e de ici . Plan Cell En i on 22:1515−1526 O en R, Pa aki DE (2001) T anspi a ion in esponse o a ia ion in mic oclima e and soil mois u e in sou heas e n deciduous o es s. Oecologia 127:549−559 Chap e 4 – In luence o ele a ion on canopy anspi a ion o empe a e deciduous o es s in a complex moun ainous e ain o Sou h Ko ea ! 93 Table 4.1 Loca ions, sizes, geological ai s, soil cha ac e is ics and s uc u al cha ac e is ics o he s udy si es. T ees wi h diame e a b eas heigh (DBH) ≥1.0 cm we e conside ed o calcula ing basal a ea (BA), ee densi y, mean DBH, and s and sapwood a ea (AS). ± indica es s anda d de ia ion (SD). 450N 650N 650S 950N Coo dina es 128°7’50.09’’E 38°17’18.64’’N 128°8’26.07’’E 38°18’56.83’’N 128°8’27.13’’E 38°18’57.07’’N 128°6’0.86’’E 38°14’43.37’’N Ele a ion [m a.s.l.] 450 650 650 950 Plo a ea [m2] 575 750 325 200 Inclina ion [°] 20 23 15 21 Exposu e Sou heas Sou heas No hwes Sou heas Bed ock G ani e G ani e G ani e G ani ic gneiss Soil ex u e Loam Sandy-loam Sandy-loam Sandy (su ace) Loam Soil dep h [cm] 8–100 18–68 19–65 22–100 Soil N con en s [mg g-1] 1.6 ± 0.9 5.0 ± 2.3 5.0 ± 2.3 1.7 ± 0.7 S and age [y s] ca. 30 ca. 30 ca. 30 ca. 20 BA [m2 ha-1] 21.2 20.7 25.9 22.3 T ee densi y [T ee m-2] 0.4 0.2 0.5 1.4 Mean DBH [cm] 5.6 ± 5.8 9.4 ± 5.3 6.6 ± 4.4 4.2 ± 1.6 S and AS [m2 ha-1] 13.7 13.3 17 15.4 Canopy heigh [m] 10 12 10 5 Species composi ion (BA co e [%]) Que cus mongolica (24) Alnus sibi ica (18) Q. aliena (15) Q. se a a (15) Ulmus lacinia a (10) Q. den a e (8) Tilia mandshu ica (6) Q. den a e (65) Be ula da u ica (19) Q. mongolica (14) Q. mongolica (50) T. mandshu ica (25) Q. den a e (14) F axinus hynchophylla (4) Q. se a a (4) Q. mongolica (72) F. hynchophylla (13) Euonymus hamil onianus (9) Chap e 4 – In luence o ele a ion on canopy anspi a ion o empe a e deciduous o es s in a complex moun ainous e ain o Sou h Ko ea ! 94 4.2.2 Measu emen o abio ic ac o s 4.2.2.1 Mic ome eo ology Sola adia ion (RS), ai empe a u e (Ta), ain all, ela i e humidi y, and wind speed we e measu ed e e y 30 seconds, a e aged and logged e e y 30 minu es by au oma ic wea he s a ions (AWS; WS-GP1, Del a-T De ices, Camb idge, UK) ins alled a 2 m abo e he g ound. AWS a 450 m (450 AWS) and 950 m (950 AWS) we e ins alled in he open space nex o he espec i e. AWS a 650 m (650 AWS) was loca ed in an open space be ween 650N and 650S. Addi ional ai humidi y and empe a u e senso s (Funky Clima, ESYS GMBH, Be lin, Ge many) we e ins alled wi hin he c owns a each si e and hal -hou ly a e ages o i e- minu e da a eco ded. Vapo p essu e de ici (D) o each si e was de i ed om measu ed TA and ela i e humidi y (Mu ay 1967). 4.2.2.2 Soil wa e con en and soil wa e e en ion Soil wa e con en (θ) in 30 cm dep h was measu ed a each si e using soil mois u e senso s (5TE, Decagon De ices, Washing on, USA) and da a logged e e y 30 minu es (EM50 Da a logge , Decagon De ices) o e he pe iod o sap low measu emen s. Rela i e θ was de e mined a each si e as daily mean θ di ided by maximum alue o daily mean θ, allowing o a be e compa ison o soils wi h di e en ex u es. To analyze soil ex u e and bulk densi y, h ee soil samples we e collec ed om each s udy si e using a soil co e and a bulk densi y sample , espec i ely. A and B ho izons o he soils we e sepa a ely analyzed. Humus was elimina ed by H2O2. Sand, sil and clay con en s we e de e mined by we sie ing o sand and lase pa icle analyze (Mas e size S MAM5004, Mal e n Ins umen s, He enbe g, Ge many) o sil and clay in he Soil Physics labo a o y, Uni e si y o Bay eu h. Based on he bulk densi y and soil ex u e da a, soil hyd aulic pa ame e s o each si e, such as esidual wa e con en (θ ), sa u a ed wa e con en (θs) and empi ical shape pa ame e s (α and n) we e es ima ed using a compu e p og am RETC−Re en ion Cu e P og am (PC−P og ess, P ague, Czech Republic) and hen soil wa e e en ion cu es we e de e mined ollowing an Genuch en unc ion ( an Genuch en 1980, Schaap e al. 2001). F om hese e en ion cu es soil wa e e en ion le els, co esponding o measu ed daily mean θ, was ead ou and also wil ing poin (θw) and ield capaci y (θc) o each si e de e mined, and hen applied o gap− illing o canopy anspi a ion (EC; see he sec ion 4.2.4). Chap e 4 – In luence o ele a ion on canopy anspi a ion o empe a e deciduous o es s in a complex moun ainous e ain o Sou h Ko ea ! 95 4.2.3 Measu emen o bio ic ac o s 4.2.3.1 Biome ic da a Mon hly lea a ea index (LAI) a each si e was measu ed once a mon h wi h a plan canopy analyze (LAI- 2000, LI-COR Inc., Lincoln, USA) unde di use ligh condi ions a ixed 9 sampling poin s wi h 10 m × 10 m g id in e al in o de o de e mine he seasonal changes o LAI. The maximum lea a ea index (LAImax) was es ima ed om lea li e collec ed wi h 0.5 m × 0.5 m li e aps ( i e in each si e) andomly placed a ca. 1 m heigh abo e he o es loo in o de o calib a e he alues o LAI measu ed by LAI-2000. Lea li e was collec ed mon hly, ans e ed o he labo a o y and so ed acco ding o species. A sub-sample o lea li e was measu ed o lea a ea (LI-3100, LI-COR Inc.), d ied o 48 hou s a 75°C and weighed. Speci ic lea a ea was de e mined om he a io o lea a ea/d y mass (cm2 g−1). L o each species (AL) om each si e was compu ed om d y lea weigh mul iplied by speci ic lea a ea, summed o e he season and di ided by he a ea o he li e ap. In July 2010, a su ey o all s ems la ge han 2 cm in diame e a 1 m heigh was ca ied ou on 25 m x 25 m g ids. Based on his su ey, mean DBH, basal a ea (BA, m2 ha−1) and ee densi y (T ee m−2) we e calcula ed o each plo . Sapwood a ea (AS, m2) o sample ees was calcula ed om he sapwood dep hs a sap low senso heigh s, de e mined om ee co es ex ac ed a he end o he sap low measu emen s. Sapwood was iden i ied by dying he co e samples, using b omoc esol g een (Sigma Chemicals, Ge many) (Bu ows 1980). Da a om he sample ees we e used o build an allome ic unc ion om which AS o all he ees in he s and we e es ima ed (Ve essy e al. 1995, Meinze e al. 2005): € AS= α DBH β (1) whe e α is a cons an and β is he allome ic scaling exponen . Using his empi ical ela ionship o each si e, s and AS (m2 ha−1), i.e., o al AS o all ees in he plo (As ) pe g ound a ea (AG), was es ima ed on he basis o he s em su ey. Allome ic ela ionships o he s udied species (Que ucs spp., A. sibi ica, B. da u ica, and T. mandshu ica) a e shown in Figu e 4.1. Th ee espec i e eg essions in he o m o powe unc ion we e buil sepa a ely o Que cus spp. alone ( ing po ous; AS = 0.7642DBH1.8057, n = 31, R2 = 0.89, P <0.0001), o he es o he species wi hou Que cus spp. (di use po ous; As = 0.8838DBH1.8905, n = 13, R2 = 0.98, P <0.0001), and o all he species oge he (AS = 0.5974DBH1.9374, n = 44, R2 = 0.84, P <0.0001). To al s and AS es ima ed om sepa a e eg essions (Que cus spp. alone s. he es o he species) we e simila o sligh ly la ge han es ima es om a combined species eg ession (i.e., he di e ence was less han 1% a 450N and 650N, 5% a 650S, and 10% a 950N). We, he e o e, chose he gene al eg ession o u he analyses ela ed o AS. Chap e 4 – In luence o ele a ion on canopy anspi a ion o empe a e deciduous o es s in a complex moun ainous e ain o Sou h Ko ea ! 96 A e de e mina ion o o al AL and o al AS o each species om each si e, species-speci ic ALAS −1 was calcula ed o all s udied species om each si e. The ee co es ex ac ed o he de e mina ion o he AS we e b ough o he lab and ing wid hs we e isually measu ed using a s e eo mic oscope. Figu e 4.1 Rela ionship be ween s em diame e a b eas heigh (DBH, cm) and sapwood a ea (AS, cm2) o Que cus species (open symbols) and o he es o he s udied species (close symbols) om all s udy si es. Black solid line is he eg ession o Que cus spp.: AS = 0.7642DBH1.8057 (n = 31, R2 = 0.89, P <0.0001), g ay solid line is he eg ession o he es o he species: AS = 0.8838DBH1.8905 (n = 13, R2 = 0.98, P <0.0001), and b oken line is a eg ession o all species: AS = 0.5974DBH1.9374 (n = 44, R2 = 0.84, P <0.0001). Le e s o Q, As, Bd, Tm indica e Que cus spp., Alnus sibi ica, Be ula da u ica, Tilia mandshu ica, espec i ely. Chap e 4 – In luence o ele a ion on canopy anspi a ion o empe a e deciduous o es s in a complex moun ainous e ain o Sou h Ko ea ! 97 4.2.3.2 Sap lux densi y and anspi a ion To es ima e canopy anspi a ion (EC), dominan and sub-dominan ee species co e ing mo e han 80% o BA a each plo we e selec ed as sample ees o sap low measu emen s. Sample size pe species was decided based on he species dominance in each plo . T ees wi h la ge han 7 cm diame e a b eas heigh (DBH), we e chosen in o de o apply sap low p obes. De ailed in o ma ion o sample ees including species and numbe o indi iduals a e indica ed in Table 4.2. Sap lux densi y was moni o ed wi h 20 mm long he mal dissipa ion p obes hea ed a cons an powe supply. The p obes we e cons uc ed a he echnical labo a o y, Depa men o Plan Ecology, Uni e si y o Bay eu h, based on he o iginal design o G anie (1987). The p obes we e inse ed in o 0−20 mm o he sapwood in all he sample ees. Depending on he de e mined sapwood dep h (d), addi ional senso s we e ins alled deepe in o 20−40 mm o 40−60 mm o he sapwood in o de o co e he mos o he sapwood. Senso s we e ins alled a ca. 1.3 m heigh abo e he g ound and on he no h- acing side (azimu h) o he ees o minimize di ec sola hea ing (Wilson e al. 2001; Wullschlege e al. 2001). Fu he , each p obe was co e ed wi h S y o oam shea hs and aluminum oil o minimize di ec he mal load om he sun. The empe a u e di e ences (ΔT) be ween he hea ed and e e ence p obes aligned e ically 10 cm apa in he sapwood was eco ded, and by compa ing ΔT o he maximum empe a u e di e ence (ΔTmax) occu ing a p edawn when he e is no sap low, sap low densi y (Fd, g m−2 s−1) was calcula ed acco ding o G anie (1987): Fd=119 (ΔTmax − ΔT) ΔT # $ %& ' ( 1.231 (2) Na u al ∆T wi hou hea ing was negligible since all measu emen s in he p esen s udy we e ca ied ou in o es s ands wi h ela i ely closed canopies. The Clea wa e -co ec ion (Clea wa e e al. 1999) was applied o Js o ing-po ous Que cus wi h less han 20 mm sapwood dep h. Azimu h a ia ion in Js was no conside ed du ing EC es ima es, based on he indings o ou p e ious s udy (Jung e al. 2011), which showed no signi ican di e ences in Js in di e en azimu hs o he ee unk. T ee wa e use (TWU, kg h−1) o he indi idual ees was compu ed as: TWU =Fd As (3) Fd =(Fdi Asi ) ∑ Asi ∑ (4) whe e Fd is sapwood a ea weigh ed sap low densi y, Fdi is sap low densi y o annulus i, and Asi is sapwood a ea o annulus i. Canopy anspi a ion (EC, mm h−1) was calcula ed as: Chap e 4 – In luence o ele a ion on canopy anspi a ion o empe a e deciduous o es s in a complex moun ainous e ain o Sou h Ko ea ! 98 EC=Fd As AG −1 (5) whe e Fd is he mean Fd o he sample ees. Measu emen s we e conduc ed be ween May and Oc obe 2010, du ing he ege a ion pe iod. Fd da a we e sampled e e y 30 seconds, a e aged and eco ded e e y 30 minu es (DL2, Del a-T De ices, Camb idge, UK). The pe iod wi h ac i e anspi a ion was ega ded as he ege a i e/g ow h pe iod. 4.2.3.3 Canopy conduc ance Canopy conduc ance (GC, mm s−1) is es ima ed om anspi a ion pe uni lea a ea (EL, mm s−1) scaled om Fd (Mon ei h and Unswo h 1990) as: GC=KGEL VPD (6) whe e KG is he conduc ance coe icien as a unc ion o Ta (115.8 + 0.4236Ta, kPa m3 kg−1) accoun ing o empe a u e e ec s on he psych ome ic cons an , la en hea o apo iza ion, speci ic hea o ai a cons an p essu e and he densi y o ai (Phillips and O en 1998). EL (mm s−1) is anspi a ion pe uni lea a ea de e mined by: EL=Fd AS AL (7) This simpli ica ion o he Penman-Mon ei h equa ion is based on he assump ion ha o es s a e well coupled ae odynamically, when lea es a e exposed o su icien ly high wind speeds. Thus, VPD can be used as an app oxima ion o he o al d i ing o ce o anspi a ion. We es ed he assump ion o s ong coupling in all he s udied si es by compa ing ae odynamic conduc ance (GA) and GC o 8 days om 1–8 June. GC eached ca. 1% o GA, ag eeing wi h he assump ion o Eq. 6. To assess s oma al sensi i i y o VPD a each si e, a modi ied Lohamma ’s unc ion was applied. GC(VPD)=GC e −mlnVPD (8) whe e GC e is canopy conduc ance a VPD = 1 kPa and −m (i.e., –ΔGC/ΔlnVPD) is he sensi i i y o GC esponse o VPD (O en e al. 1999). S oma al sensi i i y analysis was limi ed o GC unde condi ions in which VPD ≥0.6 kPa in o de o minimize he unce ain ies o GC es ima es (Ewe s and O en 2000). Chap e 4 – In luence o ele a ion on canopy anspi a ion o empe a e deciduous o es s in a complex moun ainous e ain o Sou h Ko ea ! 99 Table 4.2 Species, numbe (n), anges o diame e a b eas heigh (DBH, cm), sapwood dep h (d, mm), maximum sap low densi y weigh ed by sapwood a ea o each ee (max Fd , g m−2 s−1) and maximum ee wa e use (max TWU, kg d−1) o all he sample ees wi h sap low senso s a each si e. Que cus mongolica Q. den a a Q. se a a Q. aliena Alnus sibi ica Be ula da u ica Tilia mandshu ica 450N n 5 3 1 1 3 - - DBH 14.7–25.3 13.6–16.3 12.0 26.0 11.5–25.5 - - d 22–40 23–33 20 36 40–70 max Fd 14.7–28.5 13.3–22.0 48.3 45.2 40.7–88.2 - - max TWU 8.6–40.9 6.6–8.5 13.3 34.5 18.1–59.1 - - 650N n 4 5 - - - 5 1 DBH 8.9–14.4 15.8–18.2 - - - 14.0–22.9 23.0 d 16–24 18–25 42–60 58 max Fd 15.2–21.8 11.2–15.7 - - - 33.1–64.2 36.1 max TWU 2.7–6.6 3.9–5.5 - - - 17.6–42.6 10.4 650S n 5 1 - - - - 4 DBH 9.9–17.8 14.3 - - - - 10.4–22.7 d 20–37 20 40–60 max Fd 19.4–26.5 13.6 - - - - 20.4–55.6 max TWU 2.8–9.1 3.3 - - - - 4.3–58.6 950N n 6 - - - - - - DBH 7.3–9.8 - - - - - - d 15–23 max Fd 15.7–41.2 - - - - - - max TWU 2.4–8.2 - - - - - - Chap e 4 – In luence o ele a ion on canopy anspi a ion o empe a e deciduous o es s in a complex moun ainous e ain o Sou h Ko ea ! 100 4.2.3.4 Gap- illing A modi ied Ja is-S ewa model as de ined by Whi ley e al. (2009) was used o es ima e EC o he pe iod when da a gap occu ed. Whi ley e al. (2008) and Whi ley e al. (2009) exp essed EC in he same way as GC, as de ined by Ja is (1976) and S ewa (1988). EC=ECmax 1(RS) 2(VPD) 3( θ ) 4(LAI) (9) The unc ions € i , which ake on alues be ween 0 and 1, a e a se o scaling e ms educing a maximum s and anspi a ion (ECmax, mm d−1) in esponse o changes in RS, VPD and θ. Daily es ima es o EC we e de e mined by he unc ions i using he op imal es ima es o pa ame e s. The unc ions o RS, VPD and θ we e aken om Whi ley e al. (2008) based on hose o S ewa (1988), W igh e al. (1995) and Ha is e al. (2004). A unc ion desc ibing a adia ion esponse is: € 1(RS)=RS 30 " # $ % & ' 30 +k1 RS+k1 " # $ % & ' (10) whe e k1 is an empi ical coe icien desc ibing he cu a u e o he ela ionship. I shows an asymp o ic unc ion sa u a ing a app oxima ely 30 MJ m−2 d−1. A unc ional esponse o EC o VPD was exp essed as: 2(VPD)=k2VPDexp(−k3VPD) (11) whe e k2 and k3 a e he pa ame e s desc ibing he a e o change a low and high VPD. This unc ion o VPD o Ec ollows Bol zmann dis ibu ion. A unc ion o soil mois u e esponse was desc ibed o be a h ee-phase ela ionship as: € 3( θ )= # $ % & % 0, θ < θ w θ − θ w θ c− θ w , θ w< θ < θ c 1, θ > θ c (12) whe e θw and θc a e wil ing poin and ield capaci y o each si e, espec i ely. A unc ion o LAI esponse o EC was desc ibed as: 4(LAI)=LAI LAImax (13) A model pa ame e iza ion was pe o med using measu ed da a on daily basis ia nonlinea leas squa es analysis. To a oid e o s o di ision by ze o and condi ions o we canopy, only he da a be ween 8:00h and 18:00h we e included, and he da a on he ainy days we e excluded. The da a om each s udy si e we e pa i ioned in o wo sepa a e se s o andom days in o de no o use same da a o bo h pa ame e iza ions and alida ion o he model. Roo -mean-squa e-e o (RMSE) and an ag eemen index (d), de eloped by Willmo (1984) we e used o e alua e he ag eemen be ween he p edic ed EC and he obse ed EC. The Chap e 4 – In luence o ele a ion on canopy anspi a ion o empe a e deciduous o es s in a complex moun ainous e ain o Sou h Ko ea ! 101 ideal model would gi e RMSE = 0 and d = 1. To al gap- illed pe iod o each si e was 27 days o 450N, 54 days o 650N, 68 days o 650S, and 8 days o 950N. The gap- illed da a we e only used o quan i ica ion o annual EC. Annual EC was es ima ed by summing up he measu ed and simula ed EC as Eq. 9–Eq. 13. Table 3 con ains he bes es ima es o pa ame e s (k1, k2, and k3) along wi h hei espec i e s anda d e o s. Pa ame e s o k2 and k3 we e ound o be s a is ically signi ican (P <0.001), bu k1 was no signi ican (P >0.1), wi h la ge s anda d e o s. RMSE was less han 0.2 and d was la ge han 0.9 o all s udy si es (Table 4.3). Table 4.3 Op imal es ima es o Ja is-S ewa model pa ame e s (k1, k2, k3) and s a is ical pa ame e s o e o assessmen such as oo -mean squa e-e o (RMSE) and index o ag eemen (d) o all si es. S anda d e o s a e gi en b acke s nex o each alue. 450N 650N 650S 950N k1 0.57 (1.63) 2.08 (3.81) 0.56 (1.24) 0.92 (3.92) k2 1.92 (0.21) 3.75 (0.65) 2.43 (0.24) 2.75 (0.45) k3 0.68 (0.07) 0.94 (0.12) 0.79 (0.08) 0.99 (0.14) RMSE 0.16 0.21 0.12 0.18 d 0.96 0.91 0.97 0.96 4.2.3.5 Wa e use e iciency and lea ni ogen con en s F om each si e, i e sun- and i e shade-lea es each om i e Q. mongolica canopy ees we e collec ed on 24 June 2010, du ing mid season. The samples we e o en-d ied a 75°C o 48 hou s and hen ball-milled be o e subjec ed o 13C/12C iso opic a io analysis a BayCEER – Labo a o y o Iso ope Biogeochemis y, Ge many. Analyses we e conduc ed wi h an elemen al analyze NA 1108 (CE Ins umen s, Milan, I aly) coupled o an iso ope a io mass spec ome e del a S (Finnigan MAT, B emen, Ge many) ia an open spli in e ace ConFlo III (Finnigan MAT, B emen , Ge many) as desc ibed by Bida ondo e al. (2004). S anda d CO2 gas was calib a ed wi h espec o in e na ional s anda d (CO2 in Pee Dee Belemni e) by use o he e e ence subs ance NBS 16 o 20 o ca bon iso opic a io p o ided by he in e na ional A omic Ene gy Chap e 4 – In luence o ele a ion on canopy anspi a ion o empe a e deciduous o es s in a complex moun ainous e ain o Sou h Ko ea ! 102 Agency IAEA, Vienna, Aus ia. The 13C/12C iso opic a ios, deno ed as del a alues we e calcula ed acco ding o he equa ion δ 13C=Rsample Rs d −1 " # $% & '×1000 (14) whe e δ13C is he iso ope a io o ca bon in del a uni s ela i e o he PDB s anda d. Rsample and Rs d a e he 13C/12C o he samples and he PDB s anda d, espec i ely. δ13C was used as an index o seasonally in eg a ed wa e use e iciency (WUE) (Tieszman and A che 1990). 4.2.3.6 S a is ical analysis S a is ical analyses including linea eg ession, ANOVA, and Tukey HSD, as well as model pa ame e iza ions we e conduc ed wi h R (R de elopmen Co e Team, 2010). Nonlinea cu e i s we e pe o med using Sigma Plo (Ve sion 11, SPSS, San Ra ael, CA). 4.3 Resul s 4.3.1 Ele a ion e ec s on mic oclima e Annual mean sola adia ion (RS), daily ai empe a u e (TA), ai humidi y and ain all di e ed among si es (Table 4). The annual mean RS was highe (ANOVA, P <0.0001) a 450 m > 950 m > 650 m. The annual mean TA and he mean annual day ime D we e highe a 450 m > 650 m > 950 m. Thus, TA and day ime D dec eased wi h inc easing ele a ion. Di e ences in TA and day ime D among he si es we e signi ican (ANOVA, P <0.001). Inc easing annual ain all was obse ed wi h inc easing ele a ion. In o he wo ds, 950m ecei ed highe amoun o ain all han lowe ele a ions, co esponding o highe ain all in ensi ies and equencies h ough he yea . Mean wind speed a 950 m was ca. 4.5 m s−1, which was abou wice as high as a 450 m (2.6 m s−1). Al hough absolu e alues o RS, Ta, VPD, and ain all we e di e en among si es, hei seasonal ends we e simila ac oss he si es (Figu e 4.2). RS inc eased du ing sp ing and eached i s maximum in June. A s ong decline in RS occu ed in July and Augus , du ing wi h he monsoon pe iod. A second peak in RS occu ed in Sep embe , a e he monsoon, bu his peak was lowe han he p e-monsoon alue. RS signi ican ly d opped du ing all and win e . The minimum and maximum Ta we e eco ded in Janua y and Augus , espec i ely. VPD inc eased s eadily du ing sp ing and eached i s maximum in June, bu declined o 20% o i s maximum be ween July and Augus , du ing he monsoon. A sligh inc ease in VPD occu ed in Chap e 4 – In luence o ele a ion on canopy anspi a ion o empe a e deciduous o es s in a complex moun ainous e ain o Sou h Ko ea ! 109 Figu e 4.5 Seasonal pa e ns o daily canopy anspi a ion (EC, mm d−1) a each s udy si e in 2010. Closed ci cles a e measu ed and open ci cles a e simula ed alues using Ja is-S ewa model (Eq. 11). The shaded a ea indica es he pe iod o he monsoon in Ko ea, 2010. Chap e 4 – In luence o ele a ion on canopy anspi a ion o empe a e deciduous o es s in a complex moun ainous e ain o Sou h Ko ea ! 110 Figu e 4.6 Rela ionship be ween apo p essu e de ici (VPD, kPa) and daily canopy anspi a ion (EC, mm d−1) a each si e om June o Sep embe 2010. Fi ed unc ions we e (a) EC = 1.70VPD(1−e(−2.21VPD)) o 450N, (b) EC = 1.30VPD(1−e(−2.28VPD)) o 650N, (c) EC = 1.48VPD(1−e(−1.58VPD)) o 650S, (d) EC = 1.48VPD(1−e(−5.71VPD)) o 950N. All he es ima ed pa ame e s o each eg ession we e s a is ically signi ican (P <0.0001). 4.3.4 Ele a ion e ec s on canopy conduc ance Canopy conduc ance (GC) is a measu e o he in ensi y o EC egula ion among o es s. Mean day ime GC on clea , sunny days be ween June (ma u e canopy) was simila a lowe si es, i.e., 450N, 650N, and 650S, which was abou 4 mm s−1 (TukeyHSD, Padj >0.8), while GC a he 950N was signi ican ly highe han he o he si es (ANOVA, F = 4.568, P = 0.003; Table 4.6). GC e , which deno es GC a VPD = 1 kPa, was 5.4 ± 0.8 mm s−1 a 450N, 3.6 ± 0.4 mm s−1 a 650N, 3.7 ± 0.5 mm s−1 a 650S, and 4.5 ± 1.5 mm s−1 a 950N. GC e di e ed signi ican ly (ANOVA, F = 24.52, P <0.0001) among he si es excep be ween 650N and 650S (TukeyHSD, Padj = 0.71). Thus, a he same ange o VPD, GC was highe a 450 > 950 > 650 m ele a ions. The esponse o GC o VPD unde sa u a ing RS (>400 W m−2) a each si e is plo ed in Figu e 4.7. The Chap e 4 – In luence o ele a ion on canopy anspi a ion o empe a e deciduous o es s in a complex moun ainous e ain o Sou h Ko ea ! 111 g adien o he exponen ial eg ession cu e ela ing GC o VPD was signi ican ly s eepe o he 950N compa ed o he o he si es. Figu e 4.7 Rela ionships be ween canopy conduc ance (GC, mm s−1) and apo p essu e de ici (VPD, kPa) unde sa u a ing global adia ion (RS >400 W m−2) a each s udy si e. All he es ima ed pa ame e s o each eg ession we e s a is ically signi ican (P <0.0001). Chap e 4 – In luence o ele a ion on canopy anspi a ion o empe a e deciduous o es s in a complex moun ainous e ain o Sou h Ko ea ! 112 Table 4.6 Canopy conduc ance (GC, mm s−1) on clea days in June, i s espec i e apo p essu e de ici (VPD, kPa) o day ime ( om 8:00h o 18:00h) and GC e (GC a VPD = 1 kPa) a each si e. ± a e s anda d de ia ion (SD). 450N 650N 650S 950N GC 4.4 ± 3.0 4.1 ± 4.1 4.3 ± 3.1 5.3 ± 6.2 VPD 1.4 ± 0.7 1.4 ± 0.8 1.3 ± 0.7 1.1 ± 0.5 GC e 5.4 ± 0.8 3.6 ± 0.4 3.7 ± 0.5 4.5 ± 1.5 In Figu e 4.8, he alues o sensi i i y o GC esponse o VPD (−m = –ΔGC/ΔlnVPD, see Eq. 8) we e plo ed agains GC e a each si e. The a e age o he empi ical slopes (GC e o –ΔGC/ΔlnVPD) o o all he sample ees a 450N, 650N, 650S, and 950N we e 0.64, 0.65, 0.67, and 0.84, espec i ely. The slope o he 950N si e was signi ican ly highe han hose a he o he si es (ANOVA, F = 24.33, P <0.0001; TukeyHSD, Padj <0.0001). Thus, he slopes o he 450N, 650N, and 650S we e close o he uni e sal slope o 0.6 (O en e al. 1999), while i was 25% highe a he 950N, which indica es highe s oma al sensi i i y o VPD a his si e/ele a ion. Figu e 4.8 The esponse o canopy conduc ance o apo p essu e de ici (–ΔGC/ΔlnVPD, see Eq. 10) plo ed agains he e e ence conduc ance (GC e ) o all s udied ees a each si e, i.e., Que cus spp. (Q), Alnus sibi ica (As), Be ula da u ica (Bd), and Tilia mandshu ica (Tm). The uni e sal a io o 0.6 sugges ed by O en e al. (1999) is indica ed by a b oken line. Chap e 4 – In luence o ele a ion on canopy anspi a ion o empe a e deciduous o es s in a complex moun ainous e ain o Sou h Ko ea ! 113 4.4 Discussion 4.4.1 In e ac ions among ele a ion, abio ic ac o s and ee g ow h pa ame e s A g adien along he ele a ion was ound o Ta, ain all, and VPD, while RS, wind speed and ela i e θ showed no endency. Low ele a ions a e likely o ha e highe Ta and less ain all (low humidi y), which esul in highe VPD (T omp- an Mee eld and McDonnell 2006). Kubo a e al. (2005) epo ed a dec easing VPD a highe ele a ions due o heigh -dependan dec ease in Ta. In ou s udy, he annual mean day ime VPD a he 450N was 0.5 kPa >0.4 kPa a 650N >0.2 kPa a 950N, co esponding o highe annual mean Ta and lowe annual ain all a lowe ele a ions. We obse ed signi ican and nega i e linea ela ionships be ween ele a ion and annual day ime mean D, annual mean Ta, and annual ain all (R2 = 0.98 o VPD, 0.99 o ele a ion, and Ta, and 0.95 o ain all; P <0.0001). The amoun o ain all, soil cha ac e is ics (mainly in il a ion a es), and soil dep hs in e ac o de e mine he amoun o wa e a ailable o plan s (Chapin e al. 2002). S udies in ugged moun ainous e ains epo con as ing pa e ns o soil mois u e a ailabili y along he ele a ional g adien . Fo example, Kumagai e al. (2008) obse ed ha he uppe slope soils we e cons an ly d ie han soils down slope in a Japanese ceda o es . On he o he hand, Kubo a e al. (2005) epo ed inc easing soil wa e con en wi h inc easing ele a ion as a esul o inc easing ain all a highe ele a ions. In ou s udy, despi e di e ences in soil dep hs and p ecipi a ion amoun s a he h ee di e en ele a ions, he e we e no signi ican di e ences in soil wa e e en ion among he si es. In mos cases, soil mois u e was close o ield capaci y a all si es (Figu e 4.1), as a esul o high ain all amoun s in Haean egion. The o es s s udied, he e o e, ne e expe ienced wa e s ess du ing he s udy pe iod and any di e ences in wa e use among he o es si es canno be a ibu ed o soil mois u e a ailabili y. A co ela ion be ween TWU and DBH (Figu e 4.3) and ha be ween AS and DBH (Figu e 4.2) among ee species and ac oss si es in di e en ele a ions p o ides a link be ween ee allome y and ee wa e use (Ve assy e al. 1995; Bucchi e al. 2004; Meinze e al. 2001; Meinze e al. 2005; Gebaue e al. 2008; Jung e al. 2011), demons a ing he de e minis ic ole o xylem wa e anspo on o e all o es wa e use as well as he dependence o bo h pa ame e s on ee size. We obse ed a nonlinea inc ease in AS wi h inc easing DBH, which was de ached om ele a ion since DBH sizes we e no de ined by ele a ion. Simila uni e sal unc ional ela ionships be ween DBH and AS ha e been epo ed o 24 co-occu ing canopy ee species in opical o es s (Meinze e al. 2001; Meinze e al. 2005) and o i e ee species g owing in an Asian empe a e o es (Jung e al. 2011). Howe e , Gebaue e al. (2008) epo ed conside ably highe AS (abou 80% o s em c oss-sec ional a ea) in di use po ous ees compa ed o 20% in ing po ous ees. In ou case, al hough ing and di use po ous ees we e dis inguishable acco ding o DBH:AS a ios, a uni e sal eg ession compa ing all he ees oge he was s a is ically signi ican (Figu e 4.1), sugges ing ha he Chap e 4 – In luence o ele a ion on canopy anspi a ion o empe a e deciduous o es s in a complex moun ainous e ain o Sou h Ko ea ! 114 species basically unc ion in a simila manne in e ms o wa e use (see Figu e 4.3). Ou esul s compa e a o ably wi h hose o Meinze e al. (2005), who compa ed 18 angiospe m species. Based on hese esul s, i was clea ha compa isons o wa e use and i s egula ion among he espec i e o es s ands a di e en ele a ions could be pe o med wi hou conside ing species composi ion. This was a depa u e om ou ini ial hypo hesis ha di e ences in species composi ion may mask di e ences a ising om ele a ion. In empe a e o es s, he ime du ing which ac i e lea anspi a ion occu s co esponds o he p oduc i e pe iod o he o es (Kö ne 2007) and has a s ong in luence on he o al wa e use by o es ecosys ems. The o al pe iod o anspi a ion dec eased wi h inc easing ele a ion and dec easing Ta, i.e., 4 days 100 m−1 and 6 days °C−1, which was compa able wi h he esul s om he simila ype o empe a e deciduous o es s in Eu ope and No h Ame ica (3–5 days 100 m−1 and 7–13 days °C−1) (Ro ze and Chmielewski 2001; Di e ma and Elling 2006; Richa dson e al. 2006; Vi asse e al. 2009), e en hough he ange o ele a ional g adien was ela i ely na ow (450 m o 950 m) in ou case. This g adien in du a ion o ac i e ee anspi a ion esul ed om di e ences in iming o lea onse and senescence a he di e en ele a ions, which is likely as a esul o Ta a ia ions among he si es. The ege a ion a highe ele a ions expe ienced delayed lea lush due o ex ended low win e empe a u es and an ea lie onse o lea senescence as a esul o apid cooling in au umn, a highe ele a ions. These empe a u e changes along he ele a ional g adien also in luence ee p oduc i i y and g ow h, as demons a ed by he ela ionship be ween annual ee ing wid h and mean annual empe a u e o he espec i e o es si es. This quali ies Ta as an impo an de e minan o ee g ow h and unc ioning on moun ain slopes. 4.4.2 Ele a ion e ec s on canopy anspi a ion and i s egula ion The annual EC declined om 175 mm yea −1 a 450N o 90 mm yea −1 a 950N. This end o o es canopy anspi a ion is consis en wi h p e ious indings o McDowell e al. (2008) who epo ed a decline in EC wi h inc easing ele a ion in o es s domina ed by di e en coni e ous species in he sou he n Rocky Moun ains. Ma yssek e al. (2009), howe e , ound no ela ionship be ween he magni ude o EC and ele a ion in mixed o coni e ous o es s ands a he collinea , moun ainous, and subalpine ele a ions. Bo h McDowell e al. (2008) and Ma yssek e al. (2009) compa ed EC om di e en o es ypes o o es s ands domina ed by di e en species, which makes i di icul o sepa a e he impac o ele a ion. Ou s udy add essed simila o es ypes domina ed by oaks. This made i possible o compa e EC om he di e en o es s ands and key ou ele a ion e ec s on o es wa e use. Al hough Kubo a e al. (2005) s udied ele a ion e ec s on JS o moun ainous beech o es s in Japan, o he bes o ou knowledge, no such s udy in annual EC ega ding o ele a ional g adien s has been conduc ed in simila o es ypes in Asia. Chap e 4 – In luence o ele a ion on canopy anspi a ion o empe a e deciduous o es s in a complex moun ainous e ain o Sou h Ko ea ! 115 We es ablished a unc ional ela ionship be ween EC and Ta as EC = 29.56Ta – 85.19 wi h R2 = 0.90 and P <0.0001 o he annual mean Ta and EC = 35.67Ta – 462.7 wi h R2 = 0.97 and P <0.0001 o he g owing season Ta. Based on hese eg essions, anspi a ion is ze o when he annual mean Ta and a g owing season’s mean Ta d op down o ~3°C and ~13°C, espec i ely. Kö ne (2003; 2007) sugges ed ha a s ong ela ionship be ween g owing season leng h and mean Ta con ibu es o he educ ion in annual EC a high ele a ions especially in humid, empe a e o es s whe e soil mois u e is no limi ing. P au sch e al. (2010) epo ed a s ong linea ela ionship be ween daily mean Fd and daily Ta, indica ing a empe a u e- dependence o wa e use in Eucalyp us egnans o es s in sou heas e n Aus alia. In ou s udy, howe e , we did no obse e any di ec empe a u e-dependence o EC on a daily scale, bu on an annual scale. On a daily basis, VPD was he dominan de e minan o anspi a ion (Figu e 4.6). Simila obse a ion was epo ed by Jung e al. (2011) o he deciduous o es in Ko ea. Ou daily Ta and VPD we e no co ela ed, likely due o he apidly changing humidi y a ou s udy si es. This may explain he lack o a ela ionship be ween Ta and EC on a daily basis, since Ta lags behind. On an annual basis, howe e , a signi ican linea ela ionship be ween Ta and VPD (R2 = 0.99, P <0.0001) was ound. Wa e anspi ed by o es s is de e mined by he s oma a. S oma al unc ioning is in luenced by VPD (Kellihe e al. 1997; Saugie e al. 1997), wind speeds (Campbell-Clause 1998; Schulze e al. 2005) and soil mois u e (Sala and Tenhunen 1996; Kellihe e al. 1997; Togne i e al. 2009). In ou s udy, VPD was he dominan de e minan o GC and also EC a all he si es. The simila i ies in he esponses o daily EC o VPD (Figu e 4.6) as well as GC o VPD (Figu e 4.7) among he 450N, 650N, and 650S sugges simila i ies in s oma al unc ioning among he si es. Di e ences in EC a 950N can be a ibu ed o di e ences in he esponse o GC o p e ailing VPD a he highes ele a ion (Figu e 4.7). A depa u e om his end obse ed o ees a he 950N, which was cha ac e ized by a conspicuous dep ession in EC be ween 10:00h and 13:00h (Figu e 4.4). An ea ly d op o EC a ela i ely low VPD (Figu e 4.6) and a s eepe slope o he cu e be ween VPD and GC (Figu e 4.7) we e likely due o high wind speeds, since his si e was mo e exposed. Campbell-Clause (1998) epo ed dec easing anspi a ion a es a wind speeds mo e han 4 m s−1. Mo eo e , signi ican di e ences in he esponse o GC o VPD occu ed among si es a di e en ele a ions wi h he 950N showing highe (0.83) s oma al sensi i i y o changes in VPD compa ed o he o he s udy si es (0.63–0.65) (Figu e 4.8). The g ea e sensi i i y caused he highe GC a 950N a ound VPD = 0.6–1.0 kPa, esul ing in highe Fd and hence highe EC (in 30min), bu he lowe GC a 950N a VPD >1.0 kPa, d i ing a d op in Fd and EC, consequen ly. Such adjus men s o local condi ions may ha e implica ions o he o e all o es s and wa e use. Fu he mo e, se e al s udies epo ed ha he in ensi y o exponen ially dec easing GC in ela ion o inc easing VPD migh a y among species (McNaugh on and Ja is 1991; O en e al. 1999; O en and Pa aki 2001; He bs e al. 2008), masking di e ences a ising om ele a ion. Fo example, highe s oma al Chap e 4 – In luence o ele a ion on canopy anspi a ion o empe a e deciduous o es s in a complex moun ainous e ain o Sou h Ko ea ! 116 sensi i i y o VPD has been epo ed o di use po ous (Ace ub um) compa ed o ing po ous species (Que cus alba) (O en and Pa aki 2001). A dissimila esul was ound by He bs e al. (2008) ha he o es wi h highe p opo ion o ing-po ous ees showed highe s oma al sensi i i y. In his s udy, we did no ind any signi ican di e ence in s oma al sensi i i y o changing VPD be ween di use and ing po ous ee species g owing oge he a he same ele a ion (Figu e 4.8). Ca bon iso ope composi ion has been used as an indi ec measu e o s oma al conduc ance and wa e use e iciency (WUE) o lea es (Hubick e al. 1986; Kö ne e al. 1988, 1991) and can be employed o desc ibe he a ia ions in plan wa e use s a egies (Ma shall and Zhang 1994; Sun e al. 1996; Oso io e al. 1998; Li e al. 2006). δ13C concen a ion in plan s p o ides an in eg a ed measu e o WUE du ing plan g ow h since he δ13C concen a ion o newly ixed ca bon inc eases unde condi ions o low in e nal CO2 concen a ion (Ehle inge 1993). In ou s udy, δ13C o Q. mongolica inc eased wi h ele a ion (Figu e 4.9), and he di e ence be ween 950N and he o he si es was signi ican (Padj <0.01, TukeyHSD). Highe δ13C a highe ele a ions sugges s ha WUE dec eased wi h inc easing ele a ion as obse ed by Kö ne e al. (1988; 1991), Spa ks and Ehle inge (1997), and Co dell e al. (1998). Figu e 4.9 S able ca bon iso ope (13C) composi ions in he lea es o Que cus mongolica dis ibu ed along an ele a ion g adien in he Haean ca chmen . E o ba s indica e s anda d de ia ion om he means o indi idual ees along his g adien . Chap e 4 – In luence o ele a ion on canopy anspi a ion o empe a e deciduous o es s in a complex moun ainous e ain o Sou h Ko ea ! 117 Al hough ou s udy conside ed a ela i ely na ow ele a ional g adien (i.e., 450−950 m), signi ican educ ion in o al EC occu ed. This was a ibu ed o changes in Ta, VPD, and he leng h o g owing season along he ele a ion, which in luence ee anspi a ion. We demons a ed ha he maximum daily wa e use o indi idual ees was uni e sally de ined by ee size ega dless o species and ele a ion. S em diame e is, he e o e, a good indica o o TWU i espec i e o species and ele a ion. Di e ences in species composi ion, he e o e, do no in luence wa e use by o es s ands in hese empe a e moun ain o es s. Di e ences in s oma al sensi i i y as obse ed along he ele a ional g adien , howe e , impac ed EC and should be conside ed when conduc ing o es wa e budge in he moun ain egions. 4.5 Acknowledgemen s This s udy was ca ied ou as pa o he In e na ional Resea ch T aining G oup, TERRain and ECOlogical He e ogenei y (TERRECO; GRK 1565/1) unded by he Deu sche Fo schungsgemeinscha (DFG) a Uni e si y o Bay eu h, Ge many and he Ko ean Resea ch Founda ion (KRF) a Kangwon Na ional Uni e si y, Chuncheon, Ko ea. The iso ope abundance analyses by he BayCEER – Labo a o y o Iso ope Biogeochemis y a e kindly acknowledged. 4.6 Re e ences Ba y RG (1981) Moun ain wea he and clima e. Me huen, London Benis on M (2003) Clima ic change in moun ain egions: a e iew o possible impac s. Clima ic Change 59: 5–31 Bida ondo MI, Bu gha d B, Gebaue G, B uns TD, Read DJ (2004) Changing pa ne s in he da k: iso opic and molecula e idence o ec omyco hizal liaisons be ween o es o chids and ees. P oc R Soc Lond B 271:1799−1806 Bucci SJ, Golds ein G, Meinze FC, Scholz FG, F anco AC, Bus aman e M (2004) Func ional con e gence in hyd aulic a chi ec u e and wa e ela ions o opical sa anna ees: om lea o whole plan . T ee Physiol 24:891−899 Bu ows LE (1980) Di e en ia ing sapwood, hea wood and pa hological wood in li e moun ain beech. New Zealand Fo es Se ice, Fo es Resea ch Ins i u e, P o ec ion Fo es y Repo 172 Campbell-Clause M (1998) S oma al esponse o g ape ines o wind. Aus J Exp Ag 38:77−82 Chapin FS III, Ma son PA, Mooney HA (2002) P inciples o Te es ial Ecosys em Ecology. Sp inge , New Yo k Clea wa e MJ, Meinze FC, And ade JL, Golds ein G, Holb ook NM (1999) Po en ial e o s in Chap e 4 – In luence o ele a ion on canopy anspi a ion o empe a e deciduous o es s in a complex moun ainous e ain o Sou h Ko ea ! 118 measu emen o nonuni o m sap low using hea dissipa ion p obes. T ee Physiol 19:681–687 Di ma C, Elling W (2006) Phenological phases o common beech (Fagus syl a ica L.) and hei dependence on egion and al i ude in Sou he n Ge many. Eu J Fo Res 125:181−188 Ehle inge JR (1993) Ca bon and wa e ela ions in dese plan s: an iso opic pe spec i e. In Ehle inge JR, Hall AE, Fa quha GD (Eds.), S able iso opes and plan ca bon-wa e ela ions, Academic P ess, San Diego, Cali o nia, USA Engelha d S, Ma yssek R, Huwe B (2009) Complexi y and in o ma ion p opaga ion in hyd ological ime se ies o moun ain o es ca chmen s. Eu J Fo Res 128:621−631 Ewe s BE, O en R (2000) Analyses o assump ions and e o s in he calcula ion o s oma al conduc ance om sap lux measu emen s. T ee Physiol 20:579−589 Gebaue T, Ho na V, Leuschne C (2008) Va iabili y in adial sap lux densi y pa e ns and sapwood a ea among se en co-occu ing empe a e b oad-lea ed ee species. T ee Physiol 28:1821−1830 G anie A (1987) E alua ion o anspi a ion in a Douglas- i s and by means o sap low measu emen s. T ee Physiol 3:309−320 G anie A, Bi on P, B éda N, Pon aille JY, Saugie B (1996) T anspi a ion o ees and o es s ands: sho and long- e m moni o ing using sap low me hods. Global Change Biol 2:265−274 Ha is PP, Hun ing o d C, Coxb PM, Gasha JHC, Malhi Y (2004) E ec o soil mois u e on canopy conduc ance o Amazonian ain o es . Ag ic Fo Me eo ol 122:215−227 Hi obe M, Tokuchi N, Iwa subo G (1998) Spa ial a iabili y o soil ni ogen ans o ma ion pa e ns along a o es slope in a C yp ome ia japonica D. Don plan a ion. Eu J Soil Biol 34:123−131 Hoch G, Kö ne C (2005) G ow h, demog aphy and ca bon ela ions o Polylepis ees a he wo ld’s highes ee line. Func Ecol 19:941−951 Hubick KT, Fa quha GD, Sho e R (1986) Co ela ion be ween wa e -use e iciency and ca bon iso ope disc imina ion in di e se peanu (A achis) ge mplasm. Aus alian J Plan Physiol 13:803−816 Ja is PG (1976) The in e p e a ion o he a ia ions in lea wa e po en ial and s oma al conduc ance ound in canopies in he ield. Phil T ans Roy Soc Lond B 273:593−610 Jung EY, O ieno D, Lee B, Lim JH, Kang SK, Schmid MWT, Tenhunen J (2011) Up-scaling o s and anspi a ion o an Asian empe a e mixed-deciduous o es om sing ee sap low measu emen s. Plan Ecol 212:383−395 Kellihe FM, Hollinge DY, Schulze ED, Vygodskaya NN, Beye s JN, Hun JE, McSe eny TM, Miluko a I, Soga che A, Va la gin A, Ziegle W, A ne h A, Baue G (1997) E apo a ion o an eas e n Sibe ian la ch o es . Ag ic Fo Me eo ol 85:135−147 Ko ea Fo es Se ice (2009) Na ional Repo on Sus ainable Fo es Managemen in Ko ea 2009