Large Impacts of Climatic Warming on Growth of Boreal Forests since 1960

La ge Impac s o Clima ic Wa ming on G ow h o Bo eal Fo es s since 1960 Pekka E. Kauppi 1,2 *, Maximilian Posch 3 , Pen i Pi inen 4 1Depa men o En i onmen al Sciences, Uni e si y o Helsinki, Helsinki, Finland, 2Finnish Fo es Resea ch Ins i u e (METLA), Van aa, Finland, 3Coo dina ion Cen e o E ec s (CCE), Na ional Ins i u e o Public Heal h and he En i onmen (RIVM), Bil ho en, The Ne he lands, 4Clima e Se ice Cen e, Finnish Me eo ological Ins i u e, Helsinki, Finland Abs ac Bo eal o es s a e sensi i e o clima ic wa ming, because low empe a u es hold back ecosys em p ocesses, such as he mobiliza ion o ni ogen in soils. A g eening o he bo eal landscape has been obse ed using emo e sensing, and he seasonal ampli ude o CO 2 in he no he n hemisphe e has inc eased, indica ing wa ming e ec s on ecosys em p oduc i i y. Howe e , ield obse a ions on esponses o ecosys em p oduc i i y ha e been lacking on a la ge sub-biome scale. He e we epo a signi ican inc ease in he annual g ow h o bo eal o es s in Finland in esponse o clima ic wa ming, especially since 1990. This inding is ob ained by linking me eo ological eco ds and o es in en o y da a on an a ea be ween 60uand 70uno he n la i ude. An addi ional inc ease in g ow h has occu ed in esponse o changes in o he d i e s, such as o es managemen , ni ogen deposi ion and/o CO 2 concen a ion. A simila wa ming impac can be expec ed in he en i e bo eal zone, whe e wa ming akes place. Gi en he la ge size o he bo eal biome – mo e han en million km 2 – impo an clima e eedbacks a e a s ake, such as he u u e ca bon balance, anspi a ion and albedo. Ci a ion: Kauppi PE, Posch M, Pi inen P (2014) La ge Impac s o Clima ic Wa ming on G ow h o Bo eal Fo es s since 1960. PLoS ONE 9(11): e111340. doi:10.1371/ jou nal.pone.0111340 Edi o : Oli ie Bouche , Uni e si e ´Pie e e Ma ie Cu ie, F ance Recei ed July 3, 2014; Accep ed Sep embe 23, 2014; Published No embe 10, 2014 Copy igh : ß2014 Kauppi e al. This is an open-access a icle dis ibu ed unde he e ms o he C ea i e Commons A ibu ion License, which pe mi s un es ic ed use, dis ibu ion, and ep oduc ion in any medium, p o ided he o iginal au ho and sou ce a e c edi ed. Da a A ailabili y: The au ho s con i m ha all da a unde lying he indings a e ully a ailable wi hou es ic ion. All ele an da a a e wi hin he pape o can be ound in he e e ences lis ed. Funding: The au ho s ha e no suppo o unding o epo . Compe ing In e es s: The au ho s ha e decla ed ha no compe ing in e es s exis . * Email: pekka.kauppi@helsinki. i In oduc ion Va ia ions o ambien empe a u e in bo eal o es s s ongly a ec he ecosys em p ocesses [1,2,3]. Inc eased g eenness and ampli ied seasonal changes o CO 2 ha e been obse ed [4,5], bu he imp o emen o p oduc i i y has no been co ela ed wi h wa ming obse a ions. I clima ic wa ming has impac s on o es g ow h, his may signi ican ly a ec he ca bon budge , anspi- a ion and o es albedo [6,7]. G owing deg ee-days (GDD) a e he annual sum o daily empe a u es abo e a p e-de ined h eshold. GDD combines he e ec s o a ou able empe a u e on he du a ion and in ensi y o g ow h. Two a icles published in he 1980 s p ojec ed an inc ease in he g ow h o bo eal o es s unde u u e clima e wa ming [8,9]. A eg ession was epo ed, showing how o es g ow h is posi i ely co ela ed wi h GDD anging om 700 deg ee-days and low inc emen in no he nmos bo eal o es s o 1,350 deg ee-days and high inc emen in sou he n bo eal o es s [8]. Mo eo e , he eg ession was used o explo e hypo he ical wa ming scena ios [9]. The eco-physiological mechanisms o his co ela ion a e no ully unde s ood. Howe e , p ocess-based ecosys em models sugges ha an accele a ing ni ogen cycle is an impo an , hough concei ably no he only mechanism [3]. And indeed, a conside able wa ming has occu ed in no he n la i udes since he 1980 s [10]. In Canada, a posi i e and signi ican g ow h esponse o wa ming has been epo ed, based on in en o y measu emen s on 1,267 si es, combining obse a- ional da a wi h models, which ela e s and g ow h o s and age [11]. He e we analyse a la ge amoun o da a om Finland, collec ed and compiled by he Finnish Me eo ological Ins i u e (FMI) and he Finnish Fo es Resea ch Ins i u e (METLA), and apply he empi ical me hodology used ea lie [8]. All bo eal o es s g owing on a land a ea o 302,000 km 2 in Finland we e co e ed in he moni o ing sys em. Fo es da a om Finland a e e y well documen ed and ha e been made a ailable in e e ence [12]. We desc ibe he s udy egion, de i e spa ially ep esen a i e clima e ends in e ms o G owing Deg ee Days (GDD), es ima e a eg ession be ween o es g ow h and GDD om ecen da a d awing on he geog aphic co- a ia ion o he wo a iables, and inally compa e he esul s wi h he indings and p ojec ions om he 1980 s. Ma e ials and Me hods The me hod o his esea ch is based on he empi ical inding ha spa ial and empo al a ia ions o o es g ow h (annual inc emen ) co ela e wi h a ia ions o GDD [8,9,13]. The s udy a ea is loca ed in Finland, s e ching om he sou he n o he no he n limi o he bo eal biome and is di ided in 15 egions (Fig. 1). The Ah enanmaa egion (A ˚land Islands) is excluded as i has a empe a e cha ac e and he o es ed a ea is small (abou 1,000 km 2 ). Region 1 su ounding Helsinki is a he sou he n edge o he bo eal zone, whe e ag icul u e and se lemen s a e widesp ead bu , ne e heless, 54% o he land is co e ed by PLOS ONE | www.plosone.o g 1 No embe 2014 | Volume 9 | Issue 11 | e111340 o es s. Region 15 be ween 68uand 70uN is pa ly loca ed beyond he A c ic imbe line and is he only one whe e o es land does no p edomina e as he main land co e ype. The dis ance be ween he mos no he n and sou he n egions is abou 900 km. Du ing 1961–2013 he FMI has eco ded daily mean empe - a u e a 120–210 s a ions ac oss Finland. S a ion da a on he daily mean empe a u e and he beginning and end o he g owing season ha e been in e pola ed on o a 10 km 610 km g id using a k iging me hod [14,15]. GDD o each g id poin and each yea has been calcula ed using a h eshold empe a u e o +5uC. F om hese es ima es we compu ed he spa ial a e age GDD o each egion and each yea . Linea eg essions we e i ed o hese ime se ies. The 11 h Na ional Fo es In en o y (NFI) o Finland was ca ied ou in 2009–2012. The sample plo s o his in en o y p o ide a igo ous ne wo k o measu emen s, which was dis ibu ed o e all egions. Fo es g ow h (o annual inc emen ) e e s o he accumula ion o woody issue on ee s ems du ing one g owing season. A na ional (and No dic) con en ion o he concep ‘ o es g ow h’ was applied, which does no ake in o accoun he con ibu ion o he ees los in na u al mo ali y. Howe e , he g ow h o ecen ly ha es ed ees is included in o he g ow h es ima e. The concep is simila o ‘g oss g ow h’ as e e ed o in [11]. The me hods o measu ing o es inc emen a e well es ed, because he e has been a pe sis en economic incen i e o ob ain accu a e in o ma ion on his na u al esou ce [16]. The numbe o sample plo s measu ed in he 11 h NFI a ied be ween egions om 567 o 4,377 (Table 1). While he o al numbe o plo s was la ge (39,276), an e en mo e impo an cha ac e is ic o he o es in en o y sys em is ha he plo s we e loca ed using a well- es ed s a is ical algo i hm [16]. The e o e, he sample ep esen s all o es s in each egion. Fo es g ow h is di ec ly measu ed om sample ees obse ing he wid h o he i e mos ecen comple e ee ings and he s em elonga ion o he leade shoo du ing a i e- yea pe iod p eceding he obse a ion. As he sample ees o g ow h measu emen s a e s a is ically chosen on each plo , he esul s ep esen he g ow h o all ees g owing on o es land in each egion. A linea eg ession was i ed o he da a o desc ibing he dependence o g ow h on GDD. The da a e e o he yea 2008, which coincides wi h he a e age ime o issue o ma ion o he e ospec i e g ow h obse a ions as aken in 11 h NFI du ing 2009–2012. The app oach o his s udy is s ic ly empi ical and is based on high-quali y obse a ions on daily empe a u e and o es g ow h, which in Finland a e unique wi hin he bo eal biome. Measu e- men s om he pe iod 1961 o 2012 ( o 2013 o GDD) we e chosen, hus ob aining a su icien ly long pe iod o ime o end analyses. Ea lie eco ds and epo s we e a ailable e e ing o he pe iod 1931–1960 and we e used as a e e ence. The ea lie wo k [8,9] add essed he same a ea in Finland and applied g ow h obse a ions o Finland’s 3 d NFI, which was conduc ed in 1951–1953. No ing hese sampling yea s and he ime lag o e ospec i e moni o ing, he analysis e e ed o he o es inc emen du ing he pe iod 1946–1952. In he ea lie pape s [8,9] GDD da a o he clima e no mal pe iod 1931–1960 we e applied. The me hod was imp o ed in his new analysis as ollows. The k iging me hod, which was no ye a ailable o [8], p o ides a be e spa ial ep esen a ion o he empe a u e da a. As he o es ed a ea a ies be ween egions, he g ow h obse a ions in eg ession analysis we e weighed by he a ea o o es land (see Table 1). In de e mining he g ow h-GDD eg ession, he iming o he empe a u e and g ow h da a was be e ma ched ( he obse a ions e e ing o 2006–2011). Figu e 1. The bo eal zone as p esen ed in [9] and he 15 egions o his s udy loca ed wi hin he bo eal biome in Finland. doi:10.1371/jou nal.pone.0111340.g001 Wa ming in Bo eal Fo es s PLOS ONE | www.plosone.o g 2 No embe 2014 | Volume 9 | Issue 11 | e111340 Resul s Wa ming ends Clima ic wa ming was quan i ied om he GDD eg essions, compa ing hei es ima es o he g owing seasons o 1961 and 2008 (Fig. 2). The la e yea coincided wi h he o es g ow h as measu ed in he 11 h NFI. Wa ming was clea ly obse ed as he GDD was highe o 2008 han o 1961 in all egions. The GDD o 2008 (and 1961) in each egion e e o he es ima es as ob ained om he ime se ies eg ession, no GDD as measu ed in ha pa icula yea , no he i e-yea mo ing a e age (see Fig. 2). The wa ming ends we e s a is ically signi ican (p.95% in all egions). In absolu e e ms, he GDD wa ming was la ges in he sou he nmos egion 1 (+233 deg ee days) and smalles in he no he nmos egion 15 (+124 dd). Howe e , in line wi h p ojec ions [9], he la ges ela i e change was in he no he nmos egion 15 (+23.8%); while he smalles ela i e change occu ed in he sou h-eas e n egion 5 (+15.2%). Compa ing he mo ing a e ages wi h he eg essions i appea s ha he wa ming in all egions was pa icula ly signi ican since 1990. In all egions, he i e-yea a e age o he la es obse ed yea 2011 is highe han he eg ession es ima e o 2011 and o 2008 (Fig. 2). Impac o wa ming on g ow h The obse a ions o his s udy can be compa ed wi h he ea lie esul s [8]. Rela ing he mos ecen g ow h obse a ions o he 11 h NFI o he a e age GDD o 2006–11, a new eg ession was ob ained wi h almos he same slope as ea lie (Fig. 3). Di e ences be ween he esul s we e as ollows: (i) The new egion 15 in he e y no h widened he ange o lowe alues; (ii) The wa ming impac mo ed he da a poin s o he igh ; and (iii) O he d i e s a ec ing g ow h we e e lec ed in he ele a ed posi ion o he new eg ession abo e he old one. Rep oducing success ully he g ow h- o-GDD eg ession and ob aining an imp o ed i lends new suppo o using GDD in analyses o spa ial and in e -annual a ia ions o o es g ow h in he bo eal zone. The new da a we e empo ally be e ma ched and spa ially mo e ep esen a i e han hose epo ed ea lie [8], and in he new analysis he g ow h es ima es we e weigh ed by he a ea o o es land. These imp o emen s o he me hod and he inclusion o egion 15 con ibu ed o he be e i ( 2 = 0.94). GDD abo e 1,350 deg ee-days we e common in many egions in 2006–11, unlike in he mid-20 h cen u y (Figs. 2 and 3). Howe e , wa ming was no he only d i e a ec ing g ow h. Fo es managemen , a g adual change in he species composi ion and age s uc u e o o es s, CO 2 e iliza ion, imp o ing wa e -use e iciency, ni ogen deposi ion and o he (unknown) ac o s can ha e played a ole. In he ea lie p ojec ions a maximum g ow h o 6 m 3 ha –1 y –1 was pos ula ed, expec ing d ough limi a ions o come in o play [9]. This u ned ou o be inco ec , as g ow h abo e 6 m 3 ha –1 y –1 was now obse ed in he majo i y o egions, e en hough he a ea-weigh ed coun y a e age did no exceed 5m 3 ha –1 y –1 . We canno p edic whe he d ough limi a ions will come in o play in he u u e, especially as u u e clima e may igge changes in p ecipi a ion [17]. Wha would ha e been he g ow h o hese bo eal o es s had he wa ming no occu ed? This ques ion was add essed in he ollowing way: Fo e e y egion we calcula ed he GDD o he yea s 1961 and 2008 om he egional eg essions gi en in Fig. 2. Inse ing he pai s o GDD alues in o he new GDD-g ow h eg ession (Fig. 3), we compu ed he wa ming impac on g ow h pe hec a e in e e y egion as he di e ence o he wo ob ained g ow h alues. This di e ence was mul iplied by he o es a ea o each egion (see Table 1), esul ing in es ima es o egional g ow h due o wa ming; and he di e ence o he o e all g ow h was assigned o o he ac o s (Table 2). Table 1. Land co e and o es sample size by egions. Region Land a ea Fo es y land Fo es land Numbe o measu ed o es plo s in 2009–2012 1000 km 2 1 6.7 4.3 3.6 918 2 7.2 5.1 4.7 1066 3 16.9 10.9 9.8 2513 4 14.3 9.6 9.3 2380 5 10.5 7.7 7.4 1879 6 12.6 9.7 9.2 2300 7 14.3 12.6 12.2 3095 8 19 14.3 12.6 3034 9 16.7 14.4 13.8 3258 10 16.8 14 13.5 3197 11 17.8 15.8 14.6 3430 12 21.5 20.4 17 3264 13 35.5 31.3 24.8 4377 14 64.5 62.5 42.2 3998 15 28.1 27.9 7.8 567 To al 302.4 260.5 202.5 39276 ‘Fo es y land’ is a land use concep e e ing o lands wi h no o he p io i y assigned excep o es y. ‘Fo es land’ is a land co e concep e e ing o lands, which a e es ima ed o p oduce a leas 1 m 3 o wood pe hec a e and yea as a long- e m a e age. Da a a e om [12] and om METLA (A. Ihalainen, pe s. comm.). While he loca ion o he s udy a ea is he same as in [8] he geog aphical bo de s o egions sligh ly di e and, in pa icula , egion 15 in he e y no h was no epo ed in [8]. doi:10.1371/jou nal.pone.0111340. 001 Wa ming in Bo eal Fo es s PLOS ONE | www.plosone.o g 3 No embe 2014 | Volume 9 | Issue 11 | e111340 A g ow h o 103.9 million m 3 y 21 was obse ed in he bo eal o es s o Finland e e ing o app oxima ely 2008 as desc ibed in he 11 h NFI (Table 1, based on [12], excluding Ah enanmaa (A ˚land)). This can be compa ed o he obse ed g ow h o 49.5 million m 3 y 21 as measu e in he 4 h NFI in 1960–63 [18]. Wa ming-induced inc emen is es ima ed a 28.4 million m 3 y 21 (Table 2). An addi ional change o +26 million m 3 y 21 would hence be he esponse o o he d i e s, such as imp o ed o es managemen , CO 2 e iliza ion and o he (unknown) mechanisms. Gi en unce ain ies, we es ima e ha he g ow h o bo eal o es s o Finland would ha e been abou 75 million m 3 in 2008, had he g owing season empe a u es emained a he le el o he 1960’s. Discussion and Conclusions Mo e han hal o he signi ican g ow h imp o emen in he bo eal o es s o Finland has been a esponse o clima ic wa ming. Such a la ge change in p oduc i i y has g ea ly a ec ed he ecological and economic pe o mance o hese o es s. The long in e al o abou 60 yea s and he la ge geog aphic scale om he no he n o he sou he n edge o he bo eal zone gi e con idence Figu e 2. Annual G owing Deg ee Days (GDD) 1961–2013 o he egions (blue ci cles). Also shown a e he linea eg ession lines (black) and he 5-yea mo ing a e ages om 1963 o 2011 ( ed lines). doi:10.1371/jou nal.pone.0111340.g002 Wa ming in Bo eal Fo es s PLOS ONE | www.plosone.o g 4 No embe 2014 | Volume 9 | Issue 11 | e111340 o ob aining conclusi e esul s. The eg ession analyses we e applied as hey a e simple and anspa en and enable compa ison wi h ea lie esul s [8,9]. The analysis in his pape add essed he s a is ical co ela ion be ween o es g ow h and g owing deg ee days. The e could be causal mechanisms a ec ing o es g ow h, which a y in ime and space and a e un ela ed o empe a u e and, ne e heless, igge simila spa ial pa e ns as hose depic ed in Table 2. Howe e , i has been well documen ed ha he in e -annual a ia ion o g ow h in he bo eal zone co ela es wi h changes in empe a u e. Mo eo e , he slope o he eg ession is ep oduced using wo independen da a se s (see Fig. 3). The e o e, we conclude ha he spa ial and empo al a ia ion o g owing season empe a u e is he main causal ac o a ec ing a ia ions o o es g ow h in his egion. New esea ch is needed o elabo a e on he ecological and eco- physiological mechanisms by which inc easing empe a u es s imula e o es g ow h. An empi ical analysis like his canno cha ac e ize he mechanisms o such la ge impac s. I is possible ha he a ious d i e s ein o ce one ano he based on syne gis ic mechanisms, hus c ea ing an agg ega ed wa ming impac . Analogous a ibu ion p oblems a e me in plan sciences in assessing he ela i e e ec o plan geno ype and o he en i onmen in a ec ing plan pheno ype and yield [19]. Fin- land-speci ic land managemen , ni ogen deposi ion, o dis u - bance his o y may ha e a ec ed he wa ming esponse. Howe e , wa ming has been widely obse ed [10], CO 2 concen a ion is spa ially in a ian and simila wa ming esponses ha e been epo ed ecen ly om Canada [20] and in Russia [21]. Indi ec obse a ions ein o ce his iew [4,5]. The e o e, a signi ican wa ming impac on p oduc i i y has likely occu ed mo e b oadly in o es ecosys ems o he bo eal biome. The g ow h concep ha we applied is simila o ‘‘g oss g ow h’’ as desc ibed in [11]. The ac ion o g oss g ow h ha decomposes in he o es has been ela i ely small in Finland (na ional o al ,5 million m 3 annually). Fo es i es ha e been i ually non-exis en ; and ees, which ha e been damaged by s o ms o insec s, a e o en ha es ed o indus ial o ene gy use. The g ow h concep in he Finnish NFI has no changed since he 1960’s and, he e o e, he ime se ies a e di ec ly compa able. Ecological and economic asse s a e a s ake in Finnish o es y. Fo example, a majo expansion plan o o es indus ies in Finland was ecen ly announced, hough no ye con i med, d awing on he ample imbe esou ces. The new plan would s a ope a ing in 2017 and would be loca ed in egion 6. I con i med, he new plan is he la ges single in es men p ojec in o es indus y o da e in Finland. The plan would p ocess abou 4 Figu e 3. Reg essions ela ing o es g ow h o G owing Deg ee Days. The new eg ession ( ed) e e ing o 2006–11 is based on da a om he 15 egions shown in Figu e 1. The black do on he ed line shows he a ea-weigh ed a e age o all 15 egions. The old eg ession (blue) as published in [8] was based on 19 da a poin s as eco ded in he mid-20 h cen u y. doi:10.1371/jou nal.pone.0111340.g003 Table 2. Measu ed g ow h by egions, he wa ming impac , and he g ow h un ela ed o wa ming. Region Mesu ed g ow h Wa ming impac G ow h un ela ed o wa ming Million m 3 y 21 1 2.6 0.7 1.9 2 2.9 0.8 2,1 3 6.6 1.7 4.9 4 7.2 1.6 5.6 5 5.3 1.1 4.2 6 6.2 1.4 4.8 7 8.9 2.0 6.8 8 6.4 1.9 4.5 9 8.8 2.1 6.8 10 9.4 2.0 7.4 11 8.8 2.1 6.7 12 7.2 2.1 5.1 13 10.7 3.5 7.2 14 11.7 4.9 6.8 15 1.5 0.8 0.7 To al 103.9 28.4 75.6 All esul s e e o he yea 2008, wi h es ima es o 1961 used as e e ence. doi:10.1371/jou nal.pone.0111340. 002 Wa ming in Bo eal Fo es s PLOS ONE | www.plosone.o g 5 No embe 2014 | Volume 9 | Issue 11 | e111340 million m 3 o ound wood annually. Timbe sho age would es ic such ini ia i es in absence o he wa ming impac s. Mo e b oadly, he implica ions o clima ic wa ming a e immensely di e se. The inc emen o woody biomass eeds in o o es ca bon sinks [22]. Resea ch should add ess he clima e eedbacks and assess how hyd ology o o es albedo will change as o es g ow h and s anding biomass g adually inc ease in he ci cumpola bo eal o es s [23–26]. Pape s in he 1980 s sugges ed ha bo eal o es s a e sensi i e o clima ic wa ming, should a wa ming occu [8,9]. This s udy epo s ha a wa ming has occu ed by 2013 and bo eal o es s ha e esponded wi h accele a ing g ow h. Al hough he da a co e only Finland, simila esponses ha e been epo ed based on esea ch in o he pa s o no he n hemisphe e o es s [21]. I is impo an o con inue and expand he moni o ing p og ams o bo h clima ic changes and ecosys em esponses, and o u he imp o e hei scien i ic quali y. Acknowledgmen s The au ho s hank An i Ihalainen, Pekka No¨jd, Pauline S enbe g, Heikki Tuomen i a and wo anonymous e iewe s o aluable commen s, and he Finnish Fo es Resea ch Ins i u e (METLA) and Finnish Me eo olog- ical Ins i u e(FMI) o collec ing he big da a, p ocessing hem and making hem publicly a ailable. The au ho s hank Uni e si y o Helsinki, METLA, FMI, and RIVM o hei esea ch in as uc u e. Au ho Con ibu ions Analyzed he da a: PK MP. Con ibu ed o he w i ing o he manusc ip : PK MP PP. Ini ia ed he esea ch: PK. P ocessed he me eo ological obse a ions: PP. Re e ences 1. Goulden ML, Wo sy SC, Ha den JW, T umbo e SE, C ill PM, e al. (1998) Sensi i i y o bo eal o es ca bon balance o soil haw. Science 279: 214–217. 2. Hy o¨nen R, A ˚g en GI, Linde S, Pe sson T, Co u o MF, e al. (2007) The likely impac o ele a ed [CO2], ni ogen deposi ion, inc eased empe a u e and managemen on ca bon seques a ion in empe a e and bo eal o es ecosys ems: a li e a u e e iew. New Phy ologis 173: 463–480. 3. Melillo JM, McGui e AD, Kickligh e DW, Moo e III B, Vo osma y CJ, e al. (1993) Global clima e change and e es ial ne p ima y p oduc ion. Na u e 363: 234–240. 4. Myneni RB, Keeling CD, Tucke CJ, As a G, Nemani RR (1997) Inc eased plan g ow h in he no he n high la i udes om 1981 o 1991. Na u e 386: 698– 702. 5. G a en HD, Keeling RF, Pipe SC, Pa a PK, S ephens BB, e al. (2013) Enhanced seasonal exchange o CO 2 by no he n ecosys ems since 1960. Science 341: 1085–1089. 6. Foley JA, Ku zbach JE, Coe MT, Le is S (1994) Feedbacks be ween clima e and bo eal o es s du ing he Holocene epoch. Na u e 371: 52–54. 7. Bonan GB (2008) Fo es s and clima e change: o cings, eedbacks, and he clima e bene i s o o es s. Science 320: 1444–1449. 8. Kauppi P, Posch M (1985) Sensi i i y o bo eal o es s o possible clima ic wa ming. Clima ic Change 7: 45–54. 9. Kauppi P, Posch M (1988) A case s udy o he e ec s o CO 2 -induced clima ic wa ming on o es g ow h and he o es sec o : A. P oduc i i y eac ions o no he n bo eal o es s. In: Pa y ML, Ca e TR, Konijn N, edi o s. The Impac o Clima ic Va ia ions on Ag icul u e, Volume 1: Assessmen s in Cool Tempe a e and Cold Regions. Kluwe , Do d ech , Ne he lands. 183–195. 10. Ha man DL, Klein Tank AMG, Rus icucci M, Alexande LV, B oenniman B, e al. (2013) Obse a ions: a mosphe e and su ace. In: S ocke TF, Qin D, e al., edi o s. Clima e Change 2013: The Physical Science Basis, Camb idge Uni e si y P ess. 159–254. 11. Hembe RA, Ku z WA, Me sa an a JM, Black TA, Guy RD, e al. (2012). Accele a ing eg ow h o empe a e-ma i ime o es s due o en i onmen al change. Global Change Biology, 18: 2026–2040. 12. Yli alo E, edi o (2013) Finnish S a is ical Yea book o Fo es y. Finnish Fo es Resea ch Ins i u e. 13. Mikola P (1950) On a ia ions in ee g ow h and hei signi icance o g ow h s udies. Communica iones Ins i u i Fo es alis Fenniae 38: 1–131. 14. Hen onen H (1991) K iging in in e pola ing July mean empe a u es and p ecipi a ion sums. Repo 12, Dep . o S a is ics, Uni e si y o Jy a¨skyla¨, Jy a¨skyla¨, Finland. 41 pp. 15. Vena¨la¨inen A, Heikinheimo M (2002) Me eo ological da a o ag icul u al applica ions. Physics and Chemis y o he Ea h 27: 1045–1050. 16. Tomppo E, Heikkinen J, Hen onen HM, Ihalainen A, Ka ila M, e al. (2011) Designing and Conduc ing a Fo es In en o y - case: 9 h Na ional Fo es In en o y o Finland. Managing Fo es Ecosys ems Vol.22, Sp inge . 272 pp. 17. Bin anja R, Sel en FM (2014) Fu u e inc eases in A c ic p ecipi a ion linked o local e apo a ion and sea-ice e ea . Na u e 509: 479–482. 18. Huika i O, Seppa¨la¨ R, edi o s (1984) Yea book o o es s a is ics 1983. Folia Fo es alia 590: 76–81. 19. Finlay KW, Wilkinson GN (1963) The analysis o adap a ion in a plan -b eeding p og amme. C op and Pas u e Science 14: 742–754. 20. Peng Y, A o a VK, Ku z WA, Hembe RA, Hawkins BJ, e al. (2014) Clima e and a mosphe ic d i e s o his o ical e es ial ca bon up ake in he p o ince o B i ish Columbia, Canada. Biogeosciences 11: 635–649. 21. F os GV, Eps ein HE (2014) Tall sh ub and ee expansion in Sibe ian und a eco ones since he 1960 s. Global Change Biology 20: 1264–1277. 22. Pan Y, Bi dsey RA, Fang J, Hough on R, Kauppi PE, e al. (2011) A la ge and pe sis en ca bon sink in he wo ld’s o es s. Science 333: 988–993. 23. Swann AL, Fung IY, Le is S, Bonan GB, Doney SC (2010) Changes in A c ic ege a ion ampli y high-la i ude wa ming h ough he g eenhouse e ec . P oc Na l Acad Sci USA 107: 1295–1300. 24. Be s RA (2000) O se o he po en ial ca bon sink om bo eal o es a ion by dec eases in su ace albedo. Na u e 408: 187–190. 25. Luyssae S, Jamme M, S oy PC, Es el S, Pong a z J, e al. (2014) Land managemen and land-co e change ha e impac s o simila magni ude on su ace empe a u e. Na u e Clima e Change 4: 389–393. 26. Lukesˇ P, Rau iainen M, Manninen T, S enbe g P, Mo˜ us M (2014) Geog aphical g adien s in bo eal o es albedo and s uc u e in Finland. Remo e Sensing o En i onmen 152: 526–535. Wa ming in Bo eal Fo es s PLOS ONE | www.plosone.o g 6 No embe 2014 | Volume 9 | Issue 11 | e111340