Adaptation to climate change and Finnish Forests

October 22, 2007 The House of Sciences (Tieteiden talo), Kirkkokatu 6, Helsinki, Finland FOREST SCIENCE DAY arranged by The Finnish Society of Forest Science Adaptation to climate change and Finnish Forests 1

Mauri Timonen: presentation & poster abstracts: Ilmaston muutokset puulustojen valossa...3 The Changing Climate of Finland...5 Climate from the 7641-year unbroken Scots pine tree-ring chronology for Finnish Lapland...6 Climate patterns in Northern Fennoscandia during the Last Millennium...8 2

Metsätieteen päivä 23.10.2007 Esitelmän abstrakti Esitelmä: http://lustiag.pp.fi/mtp_231007.pdf Mauri Timonen ILMASTON MUUTOKSET PUULUSTOJEN VALOSSA Lapin männyn kasvu on viileänä kesänä heikompaa kuin lämpimänä kesänä. Tähän itsestään selvyydeltä tuntuvaan väittämään sisältyy kokonainen tieteenala, dendrokronologia (puulustotiede). Se, että juuri kesän lämpöolot vaikuttavat Lapin männyn kasvuun, perustuu resurssien niukkuuteen: lämpötila on männyn kasvun minimitekijä. Kuivuus on Lapin humidisen ilmaston vuoksi vain harvoin puun kasvun minimitekijä, joten kosteuden vaihtelut eivät näy kasvussa. Etelä-Suomessa sekä lämpötila että kosteus vaikuttavat puiden kasvuun, mutta eri tavoin puulajista riippuen. Pintajuurinen eteläsuomalainen kuusi reagoi herkästi kuivuuteen, mutta syvemmälle juurensa ulottava mänty paljon tunnottomampi sille. Kesien vaihtelut näkyvät kapeina ja leveinä vuosirenkaina useimmissa saman ilmastoalueen puissa. Koska kesät vuodet eivät ole veljeksiä keskenään, muodostuu eri levyisistä lustoista viivakoodin kaltainen sormenjälki, mitä ominaisuutta käytetään hyväksi puun kasvuajankohdan määrittämisessä. Lapin elävistä metsänrajamännyistä, keloista ja niiden makrofossiileista on koottu prof. Matti Erosen johdolla 7640 vuoden pituinen vuodentarkka vuosilustosarja. Se on Euraasian pisin ja maailman toiseksi havupuulustosarja Kalifornian vihnemännyn (8702 v) jälkeen. Ruotsalaisilla on 7413 vuoden pituinen metsänrajamännyn lustosarjansa, venäläisillä 7358-vuotinen lehtikuusisarjansa ja saksalaisilla peräti 10479-vuotinen tammisarjansa. Lapin metsänrajamännyn pitkää lustosarjaa pidetään pituutensa, vuodentarkkuutensa ja hyvän ilmastovasteensa perusteella yhtenä parhaista ilmastomuutostutkimuksen ns. proksilähteistä eli ilmastosta likimääräisesti kertovina muuttujina. Ilmastovasteella tarkoitetaan puun kasvuparametrin ja ilmastomuuttujan välistä tilastollisesti merkitsevää yhteyttä. Lapin männyllä se on esimerkiksi vuosiluston leveyden ja kesä-heinäkuun välinen riippuvuus. Lustonleveys ei suinkaan ole ainoa ilmastosta kertova proksi, esimerkiksi puuaineksen tiheys ja lustoihin varastoitunut ilman 13C ja 12C -hiiliisotooppien suhde heijastavat kesäkauden ilmaston erityispiirteitä. Metsänrajamännyn vuosilustoissa on myös globaalimuutosten jälkiä, mikä voidaan todeta mm. pohjoisen pallonpuoliskon suurilmaston vaihteluita kuvaavan NAO-indeksin ja lustoleveyden välisenä yhteytenä. Tuhannen vuoden aikaperspektiivissä tarkasteltuna viime vuosisadan kasvun vaihtelut eivät poikkea olennaisesti aiempien vuosisatojen vaihteluista. Lapin männyn kasvu, joka kuvaa varsin tarkasti myös Lapin ilmaston vaihteluita, on vaihdellut viimeisten 100 vuoden aikana heikkokasvuisesta alusta (1900-1913) hyväkasvuiseen mutta trendiltään alenevaan vaiheeseen (1915-1960), sen jälkeen vaihtelevaan, mutta selvästi keskitasoa heikompaan 3

jaksoon (1961-1995) ja viime aikoina (1997-2007) jälleen hyväkasvuiseen trendiltään nouseva jaksoon. Talvikauden sääolot vaikuttavat pohjoisen havumetsävyöhykkeen puiden kasvuun. Keskeisenä säätelevänä tekijänä on routa, jonka poistuminen maasta riippuu talvikauden sääsuhteista. Mitä pidempään kuluu aikaa kasvukauden alkamisen +5 o C kynnyslämpötilan ylittämiseen, sitä lyhyemmäksi jää kasvukausi esimerkiksi männyllä, joka lopettaa aina kasvunsa samoihin aikoihin heinäkuun alkupuoliskolla. Pitkät vuosilustosarjat ovat pidentäneet metsäntutkijoiden aikaperspektiiviä sadoista vuosista tuhansiin vuosiin. Se on osaltaan auttanut ymmärtämään aiempaa syvällisemmin kasvun vaihteluita ja ilmaston muutoksia. Tiedämme mm., että nykyistä lämpimämmät ja kylmemmät vaiheet ovat vuorotelleet viimeisten kahdeksan vuosituhannen aikana. Nykyistä lämpimämpien ilmastojaksojen puolesta puhuvat nykyistä mäntymetsänrajaa jopa 200 m ylempänä tai liki 100 km pohjoisempana sijaitsevien tunturijärvien pohjamudista löytyneet männyn tukkipuukokoiset megafossiilit. Männyn ja kuusen metsänrajat ovat liikkuneet heilurimaisesti ilmaston vaihteluiden tahdissa. Tuhatvuotisessa aikaperspektiivissä tarkasteltuna 1900-luvun liki puolivuosisatainen suotuisa jakso (1915 1960) kuuluu vuosituhannen lämpimimpiin. Sen seurauksena syntyneet, nykyisin 45 90 vuotiset männiköt, muodostavat maantieteellisesti pohjoisimman (polaarisen) ja tuntureilla ylimmän (alpiinisen) metsänrajan. Mikäli 1990-luvun lopulla alkanut ilmaston suotuisampi vaihe jatkuu vielä 10-20 vuotta, saamme ylimmälle metsänrajalle uuden, 2000-luvun alun puusukupolven. Metsänrajapuut sen aikanaan kertovat, kehittyikö nykyisestä suotuisasta ilmastovaiheesta uusi lämpökausi vai viilenikö sittenkin? 4

Metsätieteen päivä 23.10.2007 Posteri 1: http://lustiag.pp.fi/onko_ilmasto_muuttumassa13.pdf Timonen, M. 1, Mielikäinen, K. 2 & Helama, S. 3 1 Finnish Forest Research Institute, Rovaniemi Research Unit, PL 16, FI-96301 ROVANIEMI, FINLAND 2 Finnish Forest Research Institute, Vantaa Research Unit, PL 18, FI-01301 VANTAA, FINLAND 3 Department of Geology, University of Helsinki, Gustaf Hällstöminkatu 2 (P.O. Box 64), 00014 HELSINKI, FINLAND mauri.timonen@metla.fi THE CHANGING CLIMATE OF FINLAND Annual temperatures are generally rising in the northern hemisphere, but the trend is not that clear in Finland. First, there was an ascending of temperatures up to the mid-1930s, after which a descending trend up to the mid-1980s occurred. Thereafter, an ascending trend up has prevailed to the present. The rise of warming trends seems, however, to be less significant as moving towards the north. The temperature record of Sodankylä climate station (Northern Finland, starting from 1901), has proved to be a useful general climatic indicator at the Finnish timberline regions. This record shows that global warming cannot so far been recognized as drastically elevating temperatures or as increased tree growth. According to the Sodankylä record, the last seven summers were, however, warmer than the100-yr average of the record. Tree-ring indices from northern Finland show similar results. This may indicate a trend-like warming, or just the beginning of a warmer short-period cycle. Another interesting issue is the rise of monthly minimum temperatures almost to the level of the 100-yr average of the record. As a conclusion, the Finnish pine timberline continues its advancing to the north and to higher altitudes. Steady climate pattern in Finland is a result of the geographical location of Finland just beside the Scandinavian mountains, the atmospheric circulation (the prevailing positive NAO) and the warmth of Gulf Stream. This climatic setting seems to result in mean temperatures close to long-term averages, mild winters, cool summers and smaller in yearto-year variation. If the easterly winds dominate in Finland, the summers are warmer and winters colder. Tree-rings are an ultimate proxy tool for providing climate information from locations where no climate stations exist. Tree-rings work fairly well also for interpreting past climate changes. Metla, in cooperation with the universities of Helsinki and Joensuu, completed a 7520-yr tree-ring chronology of timberline Scots pine in 1999. This record (updated length 7641 years in 2007) is used as a basis for our studies of past climate variations. Considering the causes of the currently trend-like warming, it is useful to look how climate varied during and after the warm and moist Atlantic period (about 7 500 to 5 000 years ago). At that time, pine timberlines reached their maximum appearance in Finnish Lapland, growing 100-200 m higher and even 80 km northern than today. Summer mean temperatures were accordingly 2 to 4 degrees warmer than today. Obviously, the warmer summer climate of Atlantic period did not seem cause any crucial consequences, e.g. the melting of glaciers did not cause a dramatic global climate cooling, comparable to 8.2k event. 5

Metsätieteen päivä 23.10.2007 Posteri 2: http://lustiag.pp.fi/7638-pisyfi3b.pdf Mielikäinen, K. 1, Timonen, M. 2, & Helama, S. 3 1 Finnish Forest Research Institute, Vantaa Research Unit, PL 18, FI-01301 VANTAA, FINLAND 2 Finnish Forest Research Institute, Rovaniemi Research Unit, PL 16, FI-96301 ROVANIEMI, FINLAND 3 Department of Geology, University of Helsinki, Gustaf Hällstöminkatu 2 (P.O. Box 64), 00014 HELSINKI, FINLAND kari.mielikainen@metla.fi CLIMATE FROM THE 7641-YEAR UNBROKEN SCOTS PINE TREE- RING CHRONOLOGY FOR FINNISH LAPLAND Tree rings are of great importance as high-resolution proxies of past climate. A long well established dendrochronological record from a climatologically sensitive area can provide valuable indications of the Holocene climatic variability. Here we report a 7640-yr (length in 2006) continuous Scots pine tree-ring record from the treeline area of Northern Fennoscandia. Scots pine (Pinus sylvestris, L.) immigrated to northern Finnish Lapland by 9.5-9 ka calbp and spread in favourable climatic conditions to a larger area than that occupied by pine forests today. The time of the maximum extent was between 7 and 4.5 ka calbp. A large number of subfossil pine trunks and stumps have been preserved in small lakes in Lapland in the present treeline area and also beyond it. An earlier work in Lapland resulted in several dozens of radiocarbon dates for subfossil pine wood. The dated sample discs could be used to tie the initial floating chronologies to the radiocarbon timescale. The master curve was completed in 1999 by professor Matti Eronen s research group in the ADVANCE-10K project (http://www.cru.uea.ac.uk/advance10k/ ). The present pine tree-ring chronology extends from the present time to 5633 B.C. Treering samples of Scots pine (Pinus sylvestris L.) were collected from living trees, dead standing logs, old buildings, and subfossil wood from small lakes, selected dataset containing at present 1081 tree-ring series in all. The latter archive is the major source of samples. The area is situated between 68 and 70 N, 20 and 30 E, located in the northern part of the boreal forest belt in Fennoscandia, between the Swedish Scandes and the Kola Peninsula. The long pine tree-ring curve indicates interannual variations of June-July temperatures in northern Fennoscandia. The annual resolution has made it possible to reconstruct a highfrequency record of temperature variability over thousands of years. It gives also a potential to study many of past climatic and environmental variables. The high-resolution proxy record covering a major part of the Holocene can be used now for many kind of analyses, measurements and correlations with other proxies. However, it is rather difficult to extract information about low-frequency climatic variations from this data. The Finnish supra-long pine chronology is so far not long enough to overlap the specific cooling period observed in northern hemisphere about 8200 years ago. It seems there was a drastic drop of 1-5 o C in annual mean temperatures lasting for a few hundred years. The event is connected to a slowdown of the Gulf stream, which was probably caused by the cold 6

pent up waters in Hudson Bay releasing into the North Atlantic. It is possible that similar flip flop event might take place as resulted by global warming. Our target is now try to find over 8000 year pine megafossils and analyse from tree-rings, how Finnish climate changed in those days. In theory we can conclude the following: if the westerlies prevailed during the cool Gulf stream, the summers were cooler than normal, thus resulting reduced growth in trees. In the case of the easterlies prevailing, continental weather caused warm, even hot summers. This should be seen as increased growth in trees. 7

Metsätieteen päivä 23.10.2007 Posteri 3: http://lustiag.pp.fi/holocene_trends1000_inqua.pdf Timonen, M. 1, Helama, S. 2, Holopainen, J. 2, Ogurtsov, M., G. 3, Eronen, M. 2, Lindholm, M. 1, Meriläinen, J. 4 & Mielikäinen, K. 5 1 Finnish Forest Research Institute, Rovaniemi Research Unit, PL 16, FI-96301 ROVANIEMI, FINLAND 2 Department of Geology, University of Helsinki, Gustaf Hällstöminkatu 2 (P.O. Box 64), 00014 HELSINKI, FINLAND 3 Cosmic Ray Laboratory, A.F. Ioffe Physico-Technical Institute, 194021, Polytechnicheskaya 26, St. Petersburg, Russia 4 SAIMA Centre for Environmental Sciences, University of Joensuu, P.O. Box 86, FI-57101 Savonlinna, Finland 5 Finnish Forest Research Institute, Vantaa Research Unit, PL 18, FI-01301 VANTAA, FINLAND mauri.timonen@metla.fi CLIMATE PATTERNS IN NORTHERN FENNOSCANDIA DURING THE LAST MILLENNIUM Introduction. Scots pine (Pinus sylvestris ) has proven to be a strong indicator of Northern Fennoscandian June-July temperatures in (Fig. 1). Due to good preservation of wood material and its accumulating into non-oxygen muddy bottoms of ice-cold lakes we have found enough samples to build a 7641-yr (length in 2007) continuous tree-ring chronology. We have also reconstructed a high-frequency record of temperature variability over thousands of years. We, however, concentrate in this context on the climate of the Last Millennium. Objectives: 1) Analysis of temperature variations and changes in Finnish climate during the Last Millennium. 2) Identification of natural temperature variation and changes and 3) Forecasts of natural temperature changes based on tree-rings. Data and Methods. Climate and tree-ring data was used in analyses. Older published data were needed for comparisons. Anomalous climate phases were quantified by four warmest and coolest 50-year periods, three warmest and coolest 100-year periods and the warmest a. The SAS JMP Statistical Discovery v6 software was used for spectral analysis and cycle modelling. Results. Warmest and coldest reconstructed 250-year periods occurred AD 931-1180 and AD 1601-1850 (Fig.5). These periods overlap with the Medieval Warm Period (MWP) and the Little Ice Age (LIA). Coldest and warmest of all reconstructed 100-year periods occurred AD 1587-1686 and AD 1895-1994, respectively. Approx. 60-year cycle is attributable to North Atlantic thermohaline circulation (THC) during the MWP but not during the LIA. The most significant cycles are 30-32, 37, 47-49, 81-85 and 95 years (Fig. 2). The strongest of these cycles is a period of 80-95 years. A tentative model was built for forecasting natural temperature changes for the rest of this century (Fig. 7). Conclusions. Detailed picture of temperature evolution shows that MWP was a long ameliorated interval with mean temperatures warmer than temperatures during the following centuries but not warmer than during the 20th century. Cooling of climate since the MWP until the termination of the LIA follows the hemispheric trend supposedly caused by orbital forcing and changes in Sun activity, amplification of volcanic signature years and hemispheric vegetation changes with amplifying mechanism from regional 8

forest-limit retreat. THC further appears to be an agent behind the initiation and continuation of MWP and the mid-lia transient warmth. Causes for warming/cooling are today more complicated because of human activity. Maybe some combination of different forcing factors prevails. Variations and changes in Finnish climate, as well as in growth of timberline pine, are closely connected to climate patterns in the Atlantic Ocean. The most significant cycles found in our supra-long pine chronology may thus indicate changes in the THC system. This connection, if the THC driven by cyclic forcings, gives a chance to build forecasts for the natural temperature variation component of climate. The last minimums occurred in 1610, 1710, 1795,1895 and 1970 and the last maximums in 1540, 1660, 1760, 1855 and 1925 (Figs 3, 6 and 7). Based on these cyclic data, a tentative forecast model suggests climate should be warmest some 2010-2020 and coolest some 2050-2060 (Fig.7). 9