URN:NBN:fi:tsv-oa5019 Contrasting short-term performance of mountain birch (Betula pubescens ssp. czerepanovii) treeline along a latitudinal continentality-maritimity gradient in the southern Swedish Scandes LISA ÖBERG AND LEIF KULLMAN Öberg, Lisa & Leif Kullman (2012). Contrasting short-term performance of mountain birch (Betula pubescens ssp. czerepanovii) treeline along a latitudinal continentality-maritimity gradient in the southern Swedish Scandes. Fennia 190: 1, pp. 19–40. ISSN 1798-5617. Positional treeline shift is a fundamental aspect and indicator of high-mountain vegetation response to climate change. This study analyses treeline performance during the period 2005/2007−2010/2011 in the Swedish Scandes. Focus is on mountain birch (Betula pubescens ssp. czerepanovii) along a regional climatic maritimity-continentality gradient. Treeline upshift by 3.0 yr-1 in the maritime part differed significantly from retreat by 0.4 m yr-1 in the continental part of the transect. This discrepancy is discussed in terms of differential warming-induced snow cover phenology patterns and their influence on soil moisture conditions. In the continental area, earlier and more complete melting of prior relatively rare late-lying snow patches, even high above the treeline, has progressed to a state when melt water irrigation ceases. As a consequence, soil drought sets back the vigor of existing birches and precludes sexual regeneration and upslope advance of the treeline. In the maritime area, extensive and deep snow packs still exist above the treeline and constrain its position, although some release is taking place in the current warm climate. Thereby, the birch treeline expands upslope as the alpine snow patches shrink, but continue to provide sufficient melt water throughout the summer. Treeline rise appears to have been based primarily on seed regeneration over the past few decades. This is a novelty, since prior (1915−2007) treeline advance was accomplished mainly by in situ shifts in growth form of relict krummholz birches, in some cases millennial-old, prevailing above the treeline. By the snow phenology mechanism, birch can benefit from climate warming in the maritime region, which contrasts with the situation in the continental region. This discrepancy should be accounted for in projective models. In a hypothetical case of sustained warming, the subalpine birch forest belt may expand less extensively than often assumed, although ad- vance may continue for some time in snow rich maritime areas. Keywords: Swedish Scandes, treeline, revisitation study, climate warming, re- generation, Betula pubescens ssp. czerepanovii Lisa Öberg, Department of Natural Sciences, Engineering and Mathematics, Mid Sweden University, SE 85170 Sundsvall, Sweden. E-mail: lisa.oberg@miun.se Leif Kullman, Department of Ecology and Environmental Science, Umeå Uni- versity, SE 901 87 Umeå Sweden. E-mail: leif.kullman@emg.umu.se Introduction Performance of cold-marginal treelines (in a broad sense) is commonly seen as a key element in the context of climate-driven transformation of the high-mountain landscape (Payette et al. 2001; Holtmeier 2003; Kullman 2010a, 2012; Harsch & Bader 2011; Malanson et al. 2011; Öberg & Kull- man 2011a). It is generally accepted that treelines are ultimately constrained and structured by heat deficiency, although the finer ecological and eco- physiological mechanisms are still not fully under- stood (Grace et al. 2002; Hoch & Körner 2003; Holtmeier 2003; Lloyd 2005; Körner 2012). Some 20 FENNIA 190: 1 (2012)Lisa Öberg and Leif Kullman theoretical doubt and discussion have been raised as to the responsiveness of treelines to climate change (e.g. Noble 1993; Kupfer & Cairns 1996; Körner 1999). Nevertheless, a wealth of empirical data world-wide have evidenced increased high- mountain tree establishment and upward treeline shifts in concert with post-Little Ice Age climate warming, let be with intra-regional variations of different magnitudes and rates (Kullman 1979; Juntunen et al. 2002; Esper & Schweingruber 2004; Lloyd 2005; Caccianiga & Payette 2006; Kapralov et al. 2006; Tape et al. 2006; Danby & Hik 2007; Shiyatov et al. 2007; Batllori & Gutièr- rez 2008; Devi et al. 2008; Harsch et al. 2009; Kharuk et al. 2009; Kullman & Öberg 2009; Elliott & Kipfmueller 2011; Leonelli et al. 2011; Öberg & Kullman 2011a; Kirdyanov et al. 2012; Singh et al. 2012). Paleoecolocical data further support the contention of substantial treeline displacements in covariance with climate change throughout the Holocene (Tinner & Kaltenrieder 2005; Kullman & Kjällgren 2006; MacDonald et al. 2008; Paus 2010; Öberg & Kullman 2011a, 2011b). To some extent, contrasting results and opinions with re- spect to treeline mobility may relate to different treeline definitions and to the fact that some stud- ies are based on just a few sites and therefore pos- sibly biased with respect to the omnipresent topoclimatic continuum (cf. Kullman & Öberg 2009). At the local scale, the basic control of treeline performance (elevation, structure and species composition) is more complex and spatially vari- able than just a plain function of ambient tempera- ture. For example, the treeline position is locally modulated by strong winds, snow cover duration, effective soil moisture, geomorphology, soil depth, human impacts, herbivory, pathogens and histori- cal legacy, all interacting in non-linear feedback systems (Walsh et al. 1994; Holtmeier & Broll 2005; Lloyd 2005; Gehrig-Fasel et al. 2007; Kull- man & Öberg 2009; Aune et al. 2011; Leonelli et al. 2011). In particular, the role of wind seems to be much underrated (Seppälä 2004; Gamache & Payette 2005; Holtmeier & Broll 2010; Kullman 2010a). Strong wind exposure and associated snow cover redistribution may frequently con- strain the treeline at a lower elevation than would be the case in less exposed mountain environ- ments (Kullman & Öberg 2009; Elliott & Kipfmuel- ler 2010). The current concern with the alleged anthropo- genic forcing of the world’s climate and associated ecological impacts has commonly focused on treeline change as a sensitive indicator (early warning) of pending broadscale transformation of alpine and arctic ecosystems and landscapes (Neilson 1993; Holtmeier 2003; Graumlich et al. 2005; Nagy 2006; Payette 2007; Kullman 2010a, 2010b; Malanson et al. 2011). For the case of predictive and dynamic model- ling of the forest-alpine tundra interface and sub- alpine/alpine vegetation (cf. Kullman 2012), em- pirical data (“experiments by nature”) on treeline performance along macroclimatic and topoclimat- ic gradients are mandatory (e.g. Fagre et al. 2003; Körner 2005; Holtmeier & Broll 2007; Batllori & Gutièrrez 2008). This contention was particularly stressed by a recent regional and observational multi-site study, focusing on positional treeline change by different tree species over the past cen- tury, 1915−2007, in the southern Swedish Scan- des (Kullman & Öberg 2009; Kullman 2010a). Treeline rise was recorded on 95% of all revisited localities, although with site-specific extent and a strong relationship to local topography and associ- ated wind, snow cover and soil moisture condi- tions. The maximum magnitude was about 200 m in elevation, much the same for all three investi- gated species, viz. Betula pubescens ssp. czerepanovii (mountain birch), Picea abies (Nor- way spruce) and Pinus sylvestris (Scots pine), al- though with differential rates over shorter sub-pe- riods. This figure is more or less what should be predicted from a lapse rate of 0.6 °C per 100 m altitude (Laaksonen 1976) and for a treeline sys- tem in equilibrium with recorded summer tem- perature warming (1.4 °C) during the observation period. Relative to the interval 1915−1975 and in contrast to Picea and Pinus, treeline rise of Betula slowed down during the sub-period 1975−2007 (Kullman & Öberg 2009). Retarded treeline rise of birch after the mid- 1970s was particularly discernible in the most continental (southernmost) part of the investigated region, where it also manifested as declining vigor of many treeline markers, i.e. individual treeline trees. As trees have reached increasingly higher elevations, they have faced wind stress to such a degree that further upshift is largely impaired. In addition, soil drought, as a consequence of earlier seasonal snow melt, was suggested as a proximate cause for flattened treeline advance (Kullman & Öberg 2009). This is within a region where a rela- tively thin snow cover and scarcity of late-lying snow patches have “always” provided marginal FENNIA 190: 1 (2012) 21Contrasting short-term performance of mountain birch … conditions for the comfort of mountain birch, which displays a close spatial association with the most snow rich parts of the terrain (Samuelsson 1917; Kullman 2010a, 2012). Mountain birch has its ecological optimum in cool, maritime climates and depends more on soil humidity than Picea and Pinus (Hämet-Ahti & Ahti 1969; Wielgolaski 1975; Kullman 1981, 1986). During the warm and dry early Holocene, the treeline ecotone in the whole study area was mainly composed of the relatively drought-resistant Pinus sylvestris (Kullman 1995; Öberg & Kullman 2011b). The relatively high rate of treeline rise, as re- corded predominantely in maritime areas of the southern Scandes (Kullman & Öberg 2009), may be understandable in terms of a snow cover that over quite large areas of mountain slopes supplies sufficient melt water for sustained vitality of exist- ing and upcoming treeline birches. In many parts of this landscape, a surplus of late-melting snow still appears to control the treeline, which re- sponds with uphill advance as the constraint im- posed by late-lying snow eventually weakens when the summer climate warms and earlier snow melt takes place (Fig. 1). Divergent treeline patterns and responses be- tween continental and maritime climates are pos- sibly coupled to soil moisture conditions, the amount of snow and the timing of its seasonal melting (cf. Grace 1997; Gansert 2004; Payette 2007; Aune et al. 2011). This is an aspect of gen- eral relevance for the generation of realistic land- scape-scale models for an anticipated warmer fu- ture. In fact, snow cover phenology is suggested to be more essential for high-mountain plant life than direct effects of ambient air temperature (cf. Holt- meier & Broll 2005; Körner 2005; Schmidt et al. 2006; Kullman 2007a, 2007b; Wipf et al. 2009). Increasing soil drought and moisture stress in re- sponse to warming-induced earlier snow melt in the summer is frequently discussed in connection with treeline performance in different parts of the world (e.g. Barber et al. 2000; Lloyd & Fastie 2002; Daniels & Veblen 2004; Millar et al. 2004; Fig. 1. Late-lying snow patches and associated meltwater are essential for the existence of subalpine birch forest, but also constrain the elevational position of the treeline. Mt. Storsnasen, 10 July 2011. 22 FENNIA 190: 1 (2012)Lisa Öberg and Leif Kullman Gamache & Payette 2005; Wilmking et al. 2005; Holtmeier & Broll 2007; Shrestha et al. 2007; Batl- lori & Gutièrrez 2008; Green & Pickering 2009). Betula pubescens ssp. czerepanovii possesses eminent capabilities both for seed-based and veg- etative reproduction. The latter option relies on fast-growing basal sprouts, which provides a steady turn-over of stems as the old ones eventu- ally die from senescence or physical damage (Kal- lio & Mäkinen 1978; Kullman 1981). It has been hypothesized that by this mechanism individual Betula specimens may, under certain circumstanc- es, survive almost eternally, fluctuating between krummholz and arborescent modes as climate fa- vourability shifts (Kallio & Mäkinen 1978; Jonsson 2004; Kullman 2010a). Treeline rise by mountain birch in the Swedish Scandes during the first half of the 20th century was accomplished predomi- nantly by a swift phenotypic height-growth re- sponse of a pool of ancient krummholz specimens existing above the former treeline, hypothetically as relicts from a period with warmer climate, prior to the neoglacial cooling phase, initiated more than 3500 years ago (Kullman 2003; Öberg & Kullman 2011a). In very recent time, however, in- creasing seed viability and resultant reproduction has been discerned for Betula in moderately moist leeside slopes, where the depth and duration of the snow cover have diminished in favour of birch growth and establishment (Kullman 2007b; Kull- man & Öberg 2009). Presumably, a common shift to sexual regeneration at the treeline is imminent in maritime climates (cf. Holtmeier & Broll 2007). The present study focuses on both these principal modes of birch treeline regeneration and their rel- ative roles in connection with recent climate warming along a regional continentality-maritimi- ty gradient. Based on the background outlined above, we hypothesize (1) that summer warming and associ- ated snow phenology have recently forced con- trasting responses of the Betula treeline between regions with continental and maritime climatic conditions, respectively and (2) that recent treeline regeneration is predominantly seed-based in the former and vegetative in the latter region. We test- ed these assumptions by a revisitation study using accurate positional baseline data, retrieved over a relatively short observation period. Thereby, the possible operation (or not) of distinct climatic events of importance for population dynamics may be observed (cf. Holtmeier & Broll 2005; Danby & Hik 2007). In addition, any confounding processes, e.g. herbivory, human impact or other non-climatic disturbances, can be better under- stood by a short-time approach than revisitation studies with longer intervals. Study area The study comprises treeline sites distributed along a regional transect, extending 250 km north- ward from the southernmost mountains with an alpine treeline in the Swedish Scandes, 61° 05’ to 63° 25’N (Fig. 2). The mountains reach 900−1800 m a.s.l. (highest in the north) and the valley floors are at 600−800 m a.s.l. The regional climate dis- plays a clear spatial discontinuity between weakly maritime in the north to a more continental char- acter in the south. Accordingly, the transect can be divided into two distinct, sharply separated parts, with respect to climatic maritimity-continentality, calculated as the sum of the mean temperature dif- ference between July and June and between day and night during June (Raab & Vedin 1995). The southern, continental part (C) is characterized by a value of 30−40 and the northern, maritime part (M) by 25−30. Other indices provide a similar view of a radical change of climate character at virtually the same point along the transect (e.g. Tuhkanen 1980). Southwards, annual precipita- tion decreases by about 300 mm and the propor- tion of snow decreases by 10%. In the south, the number of days with snow cover and the maxi- mum snow depth is relatively low. Moreover, the seasonal snow melt is earlier and windiness be- comes less pronounced towards the south (Raab & Vedin 1995). The difference in macroclimatic character ex- erts a discernible influence on the treeline posi- tion, which is more than 300 m higher in the very south compared to the northern extremity (Kjäll- gren & Kullman 1998). In addition, the floristic composition of the subalpine birch forests just be- low the treeline clearly reflects contrasts in climate character between north and south (Hämet-Ahti 1963). The focal point of the study, i.e. the birch treeline, is by convention narrowly defined as the maximum elevation (m a.s.l.), at a given site, of trees with a minimum height of 2 m (cf. Miehe & Miehe 2000; Batllori & Gutièrrez 2008). This strict definition is practical for the purpose of intercom- parisons over time and space. The chosen critical stem height implies that treeline markers are not FENNIA 190: 1 (2012) 23Contrasting short-term performance of mountain birch … entirely covered by snow and thereby the treeline is coupled to the climate of the free atmosphere during the entire year. Since the treeline is usually formed by isolated trees in a matrix of alpine tun- dra, the impact of intraspecific competition and shelter is minimized and therefore the relation to climate and climate change is optimized (cf. Kör- ner 2007; Kullman 2010a). It may be argued that the treeline, based on the uppermost individual birch within each transect, is sensitive to disturbance without relevance for the treeline ecotone as a whole. In the present case, however, this caveat is largely reduced by the use of a large number of sampling localities. Moreover, since the study is short-term and in- volves relocation and repeat photography of indi- viduals, any such putative events are under con- trol. Thus, we are quite confident that treeline change under these premises provide a reliable view of the dynamic situation prevailing in the treeline ecotone in general. Any kind of “forest limit” is impractical for the current purposes, as the transition zone (ecotone) between closed forest and alpine tundra is spa- tially variable and unique in structure for each in- dividual mountain slope, which also impacts the responsiveness to climate change and variability (Kullman 2010a, 2012; Harsch & Bader 2011). Obviously, the closed forest has self-stabilizing properties, which further invalidates its use for ecological monitoring (cf. Körner 1999; Dullinger et al. 2004; Kullman 2010a). The uppermost treeline in the study area is formed by Betula pubescens ssp. czerepanovii, which usually forms a subalpine belt above the coniferous (boreal) forest, where Picea abies and Pinus sylvestris alternate as dominants. As a rule, the treeline of Betula extends about 50 and 100 m above the treelines of Picea and Pinus, respective- ly (Kjällgren & Kullman 2002). With increasing continentality at the geographi- cal scale, the width of the subalpine birch belt as well as the vertical distance between the birch treeline and the treelines of conifers decline. In the south, the birch treeline attains its highest levels in the Swedish part of the Scandes, 1140 m a.s.l., while values around 800 m a.s.l. prevail in the north. The subalpine birch forest has been utilized for extensive grazing by semi-domestic reindeer (on- going) and livestock (until the 1940s). However, the treeline, as defined below, is virtually pristine in character (Kilander 1965; Kullman 1979; Kjäll- gren & Kullman 1998). One reason may be that treeline markers of mountain birch are often situ- ated high above the closed forest, in steep and boulder-rich terrain, not easily accessible to man and grazing animals (cf. Sundqvist et al. 2008). Moreover, mountain birch has a strong regenera- tion capacity, also at the sapling stage, which re- duces its sensitivity to physical disturbance, e.g. herbivory (Kallio & Mäkinen 1978; Kullman 2010a). Further support for treeline naturalness is given by the fact that regional treeline dynamics during the past century are much the same for Betula, Picea and Pinus, despite their widely differ- ent regeneration strategies and varied utilization by herbivores and humans (Kullman & Öberg 2009). With respect to human impact, Swedish treelines differ from those in most other parts of Europe (cf. Gehrig-Fasel et al. 2007; Rössler et al. 2008; Potthoff 2009), as recent treeline dynamics is merely a consequence of natural forces (cf. Nagy 2006). Fig. 2. Location map showing the extent of the study transect and the individual treeline sites. 24 FENNIA 190: 1 (2012)Lisa Öberg and Leif Kullman More detailed data concerning the differentia- tion of climate, geology, plant cover and human impact along the concerned transect are given by Kullman & Öberg (2009). The Holocene vegeta- tion history, with particular respect to treeline and forest vegetation is quite well-known (Segerström & von Stedingk 2003; Bergman et al. 2005; Kull- man & Kjällgren 2006; Öberg & Kullman 2011a, 2011b). Materials and methods The present study draws on a multi-site regional sample of treeline positions gathered in ~500 m wide belt transects by the early-20th century (Smith 1920) and with resurveys carried out in the mid-1970s (Kullman 1979) and 2005−2007 (Kull- man & Öberg 2009). Here we focus specifically on elevational change of the Betula treeline over the period 2005/2007– 2010/2011, for simplicity henceforth cited as 2007−2011, using a random subsample of spatially very precise treeline records from the years 2005−2007 (Kullman & Öberg 2009) as a benchmark. All sites were visited with four years in between. A total of 44 sites were fair- ly evenly distributed among two areas (C and M), representing different climate types, with respect to thermic continentality/maritimity (see above). By the survey in 2005−2007, the treeline posi- tions were accurately defined by the use of a GPS navigator (Garmin 60 CS) repeatedly calibrated against topographical maps. Consistently, the de- viations were less than 5 m. Reported data are rounded off to the nearest 5 m. In addition, the treeline markers were photographed, which fur- ther helped to relocate the exact positions for the individual birch trees, which constituted the treeline 2005−2007. Centred on the treeline mark- ers as starting points, a ~500 m wide belt transect, trending upslope, was systematically scrutinized for the accurate positions of any new birch indi- viduals fulfilling the treeline criteria of an upright stem, at least 2 m tall. In those cases, where stasis or upslope displacement could not be document- ed, the belt transect was extended downslope until the first individual birch complying with the treeline definition was found. The regeneration mode behind recorded treeline rise was assessed from the presence or not of birch specimens (irrespective of size) by the mid-1970s (Kullman 1979) at the same elevation as the new treeline, 2010/2011. Absence was tak- en as evidence of treeline rise accomplished by seed-based establishment of new individuals dur- ing the past 30 years or so. Repeat photography was carried out from identi- cal vantage points in order to document growth and vitality changes of treeline markers (if any) and to provide further detail to the analysis of treeline performance over the study period. Some treeline sites were visited at several occasions during the four-year period, also during winter and spring, in order to assess snow cover conditions, weather in- juries and other disturbances, e.g. herbivory. The option of height growth increment of an- cient krummholz birches growing above the treeline, as a potential mechanism of treeline shift (Kullman 1979, 2010a), is tested by radiocarbon- dating of wood remnants preserved in the soil be- neath the canopy of a living and representative tree-sized birch, which grew as krummholz prior to the early 20th century and attained tree-sized after the early 20th century (Kullman 1979, 1993). The concerned birch (Mt. Getryggen, 905 m a.s.l., area M) is 3.7 m high (2010), with three living stems of similar stature and one dead stem, 1.9 m high. Maximum winter snow depth under the can- opy is 0.8 and 0.3 m at the wind and lee side, re- spectively. By the early 20th century, this individu- al birch grew 75 m above the contemporary treeline (Kullman 1979). Radiocarbon-dating of unearthed subfossil Betula wood, physically con- nected to below-ground parts of the living birch, was conducted by Beta Analytic Inc., Miami (USA). The radiocarbon ages are expressed as cal- ibrated years before present (cal. yr BP), which re- fer to the intercept of the radiocarbon age with the calibration curve (for further details, see Öberg & Kullman 2011b). Browsing by semi-domestic reindeer (Rangifer tarandus L) and other herbivores is sometimes thought to influence the potential for birch treelines to respond positively to climate warming (e.g. Olofsson et al. 2009; Aune et al. 2011). This motivated us to assess each treeline marker for signs of herbivory. The nomenclature of vascular plant taxa is ac- cording to Mossberg & Stenberg (2003). Climate evolution over the past century Climate evolution over the past century is repre- sented by the official weather stations Storlien/ FENNIA 190: 1 (2012) 25Contrasting short-term performance of mountain birch … Visjö valen (595 m a.s.l.) and Särna (435 m a.s.l.), in area M and C, respectively. These stations have long and fairly homogeneous records and display similar trends in covariance with the entire region (cf. Alexandersson 2006; SMHI 2012). Through- out, summer and winter refer to the mean temper- atures of June−August and December−February, respectively. For the period 1915−2011, summer temperatures rose by 1.1−1.4 °C and winter tem- peratures by 0.9−1.4 °C. For both periods, the highest figures were recorded at Storlien/Visjöva- len. During the four-year period in particular focus here, summer temperatures were 1.0 and 0.8 °C warmer than the average for 1915−2011, for Stor- lien/Visjövalen and Särna, respectively. Corre- sponding figures for the winter are –0.1 and –1.3 °C, respectively. Precipitation in the region has increased by 5−10% over the same period of time, and most significantly by 10−20% during the summers over the past two decades. Concurrently, winter precipitation remained fairly stable (Alex- andersson 2006; SMHI 2012). The days with snow cover have decreased (Moberg et al. 2005). Growing season soil temperatures are suggested as a general constraint of treeline position (e.g. Körner & Paulsen 2004). In order to account for this aspect, we analyzed root-zone temperature records annually (1985−2001) at a treeline site, under the canopy of a birch copse in the northern part of the study area (Mt. Storsnasen, 850 m a.s.l.), where the maximum snow depth is ~0.7 m. A view of this site in summer and winter is given in Figure 3. Temperatures were recorded by resist- ance termistors (TO-03R), manufactured by T. Johnsson Inc. Umeå, Sweden. The sensors were installed at a depth of 30 cm in the upper mineral soil, right below the organic raw humus layer. At this depth, short-term temperature variations are significantly damped (cf. Körner 2007), as evi- denced by daily measurements over some weeks during all seasons. Readings have been carried out 2−3 times per month, which should provide an in- tegrated view of seasonal trends in soil tempera- ture, comparable between different years (cf. Har- ris 2001). Based on these measurements, we present an annual index of the summer tempera- ture favourability. This refers to the highest reading for the July−August period, which always con- tained the highest annual record. Soil temperature measurements 1985−2011 are displayed in Figure 4. A strong rising linear trend (+3.4 °C) of growing season soil temperature is evident for the period 1985−2011. Proxy support for a changing mountain climate in the study area, with direct relevance for birch performance, is provided by continuous glacier and snow patch recession (cf. Kullman 2004b, 2007a, 2007b; Öberg & Kullman 2011a). In addi- tion, a photographic landscape-scale time series (2008−2011) of annual birch leaf unfoldning and landscape-scale snow melt in the northern part of the transect is compared with a photo of the same view captured by Dr. Harry Smith on June 11 in 1914. At this date, the birches were totally devoid of leaves and snow covered most parts of the sur- veyed landscape (Fig. 5). During the four years Fig. 3. Summer and winter views of the site for soil temperature records. Mt. Storsnasen, 29 July 2010 and 3 March 2011. 26 FENNIA 190: 1 (2012)Lisa Öberg and Leif Kullman Fig. 5. Phenological view of the east-facing slope of Mt. Lillulvåfjället. The initial image (top) was captured by H. Smith, 11 June 1914. The following sequence (from left to right) represents the situation at virtually the same date of the years 2008, 2009, 2010 and 2011. Fig. 4. Maximum summer (July-August) root-zone tem- peratures obtained annually at the treeline (Fig. 3). FENNIA 190: 1 (2012) 27Contrasting short-term performance of mountain birch … here concerned, snow melt has been well under- way and birch leafing entirely completed at the same date (± 2 days). Overall the subalpine/alpine landscape in the Scandes has become drier since the 1930s or so (Smith 1957; Komárková & Wielgolaski 1999; Kullman 2007a, 2007b, 2010b). Results Treeline change 2007 to 2011 was assessed at 44 sites – 24 in area M and 20 in area C. Of the total sample, 21 sites (48%) displayed treeline rise by 5 m or more (max. 45 m), 18 sites were indifferent (41%), while 5 sites (11%) showed retreat by 5 m or more (max 35 m). A total of 18 out of 21sites (86%) with treeline advance was confined to area M. The investigated sites represent all aspects, al- though with a preponderance for south and north. Detailed statistics concerning the magnitude of treeline change within area M and C, respectively, is given in Table 1. Figure 6 accounts for the extent of treeline change in relation to latitudinal position within the geographical transect. Clustering of sites with upslope treeline displacement in the north (area M) is evident, while stability and re- traction prevail in the south (area C). It is visually striking that the sharp transition between the mari- time and the continental parts of the transect (see above) almost exactly coincides with these funda- mentally different treeline performances. The magnitude of treeline change is significant- ly larger in area M than in area C (t-test, p=0.0007). Area C displays a negative average shift. Of those sites representing treeline descent, 4 out of 5 were in area C. The annual rates for the whole transect and the two sub-areas (M and C) are 1.4, 3.0 and –0.4 m/year, respectively (Table 2). The etiology behind treeline retraction was a complex and time transgressive affair, involving gradual foliage loss, dieback of stems and their fi- nal breakage (Fig. 7). The main pattern was a re- duction in maximum stem height to below 2 m. When the dominant stem had died, short sprouts always existed at the trunk base. In no case was treeline retreat associated with death of individual treeline markers. In some cases, observations in late summer evidenced dead and drying leaves, indicative of severe drought during the summer. Since stem and twig dieback occurred above as well as below the maximum snow depth position, it appears unlikely that frost desiccation could be responsible for this process. At the localities with recorded treeline stasis, the treeline markers did not increase perceivably in height and they fre- quently displayed reduced vitality of the same kind as described above (Fig. 8). All sites (area M and C) with treeline stability, retraction, or discernible loss of vigor are in par- ticularly wind-exposed landscape sections, virtu- ally without late-lying snow patches located above the treeline (Fig. 9). The only case of lowered Fig. 6. Treeline change at all individual sites, plotted against the latitudinal position along the entire maritimity-continental- ity transect, according to the geographical positional system RT90. The hatched line separates the maritime (M) and the continental areas (C). 28 FENNIA 190: 1 (2012)Lisa Öberg and Leif Kullman Table 1. Treeline positions (m a.s.l.) 2007, 2011and change (m) over the period 2007-2011. Geographical coordinates are given as degrees lat. long. Localities above and below the horizontal line belong to area M and C, respectively. FENNIA 190: 1 (2012) 29Contrasting short-term performance of mountain birch … Table 2. Absolute change and annual rate of change, separately for area M, C and the entire transect. Fig. 7. Treeline retraction over the four-year study period, exemplified at two sites. Left. Almost complete dieback of the main stem, although with live basal sprouts. Mt. Städjan, 6 June 2011. Right. Slight reduction in height of the tallest stem implies that this old-growth birch does not fulfill the 2 m treeline definition. Mt. Sonfjället (Gråsidan), 1 August 2011. Fig. 8. Typically, treeline markers at sites with a stable treeline dis- played reduced vitality in terms of foliage loss and drying twigs and stems. Left. Mt. Sonfjället (Korpflyet), 3 July 2011. Right. Mt. Brattriet, 12 August 2011. 30 FENNIA 190: 1 (2012)Lisa Öberg and Leif Kullman treeline in area M was due to geomorphic instabil- ity in the form of a minor landslide. In no case could reindeer browsing, which is quite strong along the entire transect, account for treeline low- ering or substantial physiognomic recession. Predominant treeline rise and infilling during the present study period are largely confined to habitats where establishment and growth were previously constrained by too much late melting snow. This discrepancy in snow cover characteris- tics and phenology between area M and C, par- ticularly during late spring and early summer, has been documented by surveys at each locality (Kullman 1979, updated). The typical character of this course of change is highlighted from Mt. Getryggen (area M), one of the most intensively studied mountains in this respect (Fig. 10, 11). Notably, the majority of birches representing re- cent treeline rise are quite low and slender speci- mens, 2−2.2 m high and with basal trunk diame- ters of 3−8 cm (see Fig. 11E). Treeline rise has been accomplished by saplings which were recorded as near-tree sized by 2005−2007, when they grew slightly above the contemporary treeline. Annual height growth during the study period was in the range 15−25 cm, which resulted in height incre- ment by 0.5−1 m during the four year period, thereby passing the critical limit of 2 m. Widely scattered saplings, 0.5−1.5 m high, occur within a zone 0−25 m above the new and most recent treeline. The study of the clonal age of a treeline birch, which transformed from krummholz to tree-size during the past century, clearly sustains its ancient age and establishment much earlier than the mod- ern warming phase (Fig. 12). Coring of the stem at the root collar revealed that this specimen existed, presumably as krummholz, in the late-18th cen- Fig. 9. Characteristic site with stable treeline, which is positioned right at the rim of closed birch forest. The terrain higher upslope is strongly windswept and virtually no late-lying snow patches accumulate here. Mt. Hårdeggen, 8 August 2009. FENNIA 190: 1 (2012) 31Contrasting short-term performance of mountain birch … Fig. 10. In the maritime area, birch is frequently invading hollows and lee slopes, where previously too much of late-melting snow prevented establishment and growth. Mt. Getryggen, 8 August 2011. tury. Two pieces of decaying wood were excavat- ed and radiocarbon-dated underneath the stools of living and dead stems. One sample yielded a mod- ern age within the range AD 1490−1950 (Beta- 230891) and another dated 4770 cal. year BP (Beta-264397). Most of the new treeline markers, which ap- peared during the study period, seemed to origi- nate from seed that had germinated in recent time, apparently during the past 2−3 decades. In some cases, however, they were recorded as low-grow- ing (< 2 m tall) shrubs by surveys of the same transects carried out in the mid-1970s (Kullman 1979). Multiple stems and stools of decaying stems bases further indicate a pre-recent origin. Thus, with respect to the mode of reproduction and mechanism of treeline rise, the present phase of birch treeline advance (mainly in area M) relies both on seed reproduction and accelerated height growth of old-growth krummholz specimens, i.e. a kind of phenotypic plasticity (Kullman 1979, 2010a; Öberg & Kullman 2009). Overall, the im- portance of the latter mechanism seems to have diminished substantially relative to the situation prevailing at the time of the survey carried out in the mid-1970s (Kullman 1979). Signs of reindeer browsing on twigs and branch- es were found on 58 and 51% of the treeline mark- ers in area M and C, respectively. In most cases, only the tips of annual shoots were browsed and in no case was reindeer impact obviously responsi- ble for treeline retreat during the study period. Characteristically, treeline markers in area M grow in spots with predominant Vaccinium myrtil- lus and some species typical of moderate snow beds, e.g. Alchemilla alpina and Viola biflora. In area C, treeline birches are confined to more mon- ospecific and xerophilic Vaccinium myrtills-heaths. 32 FENNIA 190: 1 (2012)Lisa Öberg and Leif Kullman Discussion As outlined below, we consider both of the ini- tially launched hypotheses to be validated by the obtained results. Current treeline performance dis- plays a temporally tight and ultimate relationship with the character of the prevailing thermal cli- mate and appears to be independent of herbivory, e.g. reindeer grazing. This result contrasts with generalizing hypotheses that reindeer grazing may have a substantial inhibitory effect on warming- induced treeline upshifts (Cairns & Moen 2004; Olofsson et al. 2009; Van Bogaert et al. 2011). A similar conclusion has also been reached for treeline areas in northernmost Sweden (Hedenås et al. 2011). However, this circumstance does not Fig. 11. Upper. The temporal course of treeline change along a local elevational transect on the south-facing slope of Mt. Getryggen, according to observations at different points of time. A. 830 m a.s.l. (Smith 1920), B. 905 m a.s.l. (Kilander 1955), C. 930 m a.s.l. (Kullman 1979), D. 930 (Kullman & Öberg 2009), E. This study. Lower. Present-day character of the trees marking past and present treeline positions (A-E). 30 April 2011. FENNIA 190: 1 (2012) 33Contrasting short-term performance of mountain birch … rule out the possibility that in a reindeer-free sys- tem the abundance of birches would be higher than is currently the case. On the other hand, ob- servations in ungrazed mountain areas do not sup- port this option (Kullman 2004a, 2005). Despite virtually similar magnitudes of recent climate warming, short-term positional treeline re- sponses differ significantly north and south of a distinct breakpoint, separating a maritime climate in the north (area M) from more continental condi- tions in the south (area C). Treeline rise is relatively most prevalent and extensive in the former area, while stability and retreat characterize the latter. This discrepancy indicates that treeline dynamics by mountain birch is only indirectly mediated by the course of ambient temperature change and more directly by some variable associated with the maritimity/continentality continuum, most likely snow cover phenology and related soil moisture conditions. This assumption is based on tenden- cies gleaned in more extensive studies over longer periods of time (Kullman 1979; Kullman & Öberg 2009). In the south (area C), the treeline is generally positioned relatively close to the mountain peaks, which potentially may hamper its ability to reach the potential limit (see discussion by Odland 2010). However, this circumstance can hardly ac- count for the differential performance along the studied geographical transect. This contention is based on the fact that both treeline position and vitality declined in the south, which needs a tem- poral change of growth preconditions. Moreover, it is well documented that most of the mountains in the continental area have supported tree and forest growth virtually up to the highest peaks dur- ing periods with more favourable climates during earlier parts of the Holocene (Öberg & Kullman 2011b; Kullman 2012) When contemplating the role of effective soil moisture in alpine/subalpine regions it is impor- tant to make a distinction between moisture pro- vided by liquid precipitation and by the gradual melting of late-lying snow patches. To a large ex- tent, the seasonal water balance of high-moun- tain ecosystems is a function of snow cover char- acteristics, in particular its spring/summer phe- nology (e.g. Peterson 1998; Beniston 2003; Hall & Fagre 2003). The fundamental role of melt wa- ter in alpine/subalpine plant ecology is widely recognized (e.g. Smith 1957; Holway & Ward 1963; Philipp 1978; Kullman 1986; Keller & Körner 2003; Smith et al. 2003; Shrestha et al. 2007; Millar et al. 2012). Drying-out of the top- soil and associated nutrient shortage represents a recurrent hazard during the summer over those parts of the Scandes where there is no continuous supply of melt-water (Østbye 1975; Wielgolaski & Kärenlampi 1975). Sustained snow melting throughout the summer is vital for establishment and growth of mountain birch (Kullman 1986; At- kinson 1992; Sveinbjörnsson et al. 1992). Thus, projections of future evolution of treeline vegeta- tion based solely on wet precipitation models Fig. 12. Left. Ancient mountain birch which grew as a prostrate krummholz-individual, 75 m above the treeline of the early- 20th century. Mid. Coring at the ground surface level yielded 221 year rings. Right. Wood remnants in the soil underneath the stem base were radiocarbon-dated and indicate that this individual specimen existed here almost 5000 years ago. 34 FENNIA 190: 1 (2012)Lisa Öberg and Leif Kullman therefore risk to be misleading (e.g. Young et al. 2011). The character of birch decline in the continental area suggests that soil drought and desiccation have recently reached critical levels, when pre- conditions for sustainable growth and seed regen- eration have gradually over the past century be- come decidedly sub-optimal. An expression of this situation is provided by an age structure study of mountain birch at a typical site in the treeline eco- tone (area C), which displays a pronounced “geri- atric” age frequency pattern, with a preponder- ance of low-vitality ancient trees, representing peak regeneration in the mid 19th century. This sharply contrasts with insignificant sexual regen- eration over the warmer and drier 20th century and particularly during recent decades (Kullman 2012). The coupling between the present state of treeline performance, earlier and more complete snow melt and thereby drier soils, is supported by landscape-scale studies, which document disap- pearance of snow beds and associated plant com- munities in the continental area (Kullman 2004a, 2005), despite increased summer precipitation. Obviously, under these circumstances, the drought-intolerant mountain birch (Vaartaja 1955) and its treeline are no longer able to benefit from recent climate warming. Due to gradually en- hanced moisture stress, trees and treelines in this area are not responding to rising temperatures with upslope advances as they did during earlier warming phases of the 20th century (Kullman & Öberg 2009). This kind of shift in response pattern, implying that moisture stress rather than heat defi- ciency is limiting for treeline rise, is reported also from continental Alaska and Sierra Nevada (e.g. Taubes 1995; Lloyd 1997; Driscoll et al. 2005; Millar et al. 2012). In contrast to retarded birch treeline advance, more drought-resistant ever- green conifers, pine in particular, are currently more expansive. At some localities in the conti- nental area, pine is even “leap frogging” over the sparse and disintegrating birch belt (Kullman 2004a, 2010a, 2010b, 2012). This currently emerging situation has a paleoecological analogy. During the early Holocene, when temperatures were about 3 °C higher than present, no birch belt existed and pine formed the transition between forest and alpine tundra (Nesje et al. 1991; Kull- man 1995; Barnett et al. 2001; Kullman & Kjäll- gren 2006; Öberg & Kullman 2011b). The relatively more frequent treeline rise in the maritime region is consistent with air and soil tem- perature evolution, birch phenology series and glacier/snow pack performance. Upshifts, in con- trast to stability and retraction, have a clear spatial relation to sites influenced by a relatively deep and persistent snow cover, as assessed by field sur- veys at each locality during spring and early sum- mer (Kullman 1979, updated). This is a setting that in its most extreme form does not support estab- lishment and growth of mountain birch (Björk & Molau 2007). In accord with the current warm phase this constraint is becoming gradually re- leased and new ground with a suitable snow cover and sufficient soil moisture for tree growth by birch is exposed at the margin of receding snow patches (Fig. 1). It appears that most new treeline markers have originated from seed establishment of new indi- viduals over the past 30 years or so. In contrast, during earlier parts of the 20th century, treeline upshifts were mainly accomplished by transforma- tion of krummholz birches to more erect and arbo- rescent forms as climate warming initiated a surge of fast-growing stems (Kullman 1979, 2010a). These originated from a pool of relictual shrub birches, some of which have prevailed above the treeline for several past millennia (Fig. 12), charac- terized by neoglacial cooling (Kullman 2003, 2010a). Gradually during the past century, this pool seems to have become “depleted” as most individuals have turned into tree form and a shift from predominant vegetative to sexual reproduc- tion has taken place quite recently in snow rich and maritime areas. This contention is supported also by a virtually new phenomenon, implying that, within snow rich areas, birch saplings (20−30 years old) are quite commonly established up to 300−700 m above the treeline (Kullman 2004b, 2007a, 2007b; Sundqvist et al. 2008; Öberg & Kullman 2011a). This indicates that treeline vege- tation is currently in a strongly progressive phase in these environments. An analogous reproduc- tional shift may account for extensive upslope ad- vances of many vascular plant species in the al- pine zone of the maritime area, i.e. a kind of “ther- mophilization/xerophilization” of the alpine plant cover in general (Kullman 2007a, 2007b, 2010b). The annual rate of treeline advance for the en- tire area (M + C), i.e. 1.4 m year-1, is double the figure assessed for the period 1915−2007, which contained both warm and cool phases (Kullman & Öberg 2009), indicating that treeline change is currently performing in distributional equilibrium with the prevailing climate. Similar rates during FENNIA 190: 1 (2012) 35Contrasting short-term performance of mountain birch … the past 20 years are reported for the Ural Mon- tains (Russia) (Kharuk et al. 2009). In this context, it should be stressed that the short-term tempera- ture equilibrium, empirically demonstrated in this study, invalidates generalizations that, at the scale of years to decades, the birch treeline is out of equilibrium with climate (e.g. Staland et al. 2010). Given sufficient soil moisture, there is a tight cou- pling between climate warming and treeline re- sponses, as described and discussed from widely different regions (Weissberg & Baker 1995; Luck- man & Kavanagh 2000; Daniels & Veblen 2004; Elliott 2011; Singh et al. 2012). Since the air temperature rise during the study period has been quite modest, in contrast to more consistent soil temperature increase, there is rea- son to stress the contribution of high soil tempera- tures for the present phase of birch treeline rise in particular and for treeline life in general (cf. Karls- son & Nordell 1996; Körner & Paulsen 2004; Gehrig-Fasel et al. 2008; MacDonald et al. 2008). A main conclusion from the present study is that one and the same climatic fluctuation does not everywhere lead to the same pattern and magni- tude of birch treeline change (cf. Kullman & Öberg 2009; Virtanen et al. 2010; Elliott & Kipfmueller 2011). In the present context, snow cover phenol- ogy appears to have a pivotal role for differential treeline dynamics. This circumstance has essential implications for projections of treeline perform- ance in an allegedly warmer future. It now appears that such modelling needs to account for divergent responses along the ubiquitous continentality- maritimity gradients in the Scandes. Existing mod- els (e.g. Moen et al. 2004; Young et al. 2011) do not make this distinction, which results in quite unrealistic outputs, projecting a threat to the per- sistence of an extensive treeless alpine zone in Scandinavia. In a hypothetical case of continued warming, mountain birch is likely to advance sub- stantially only in the most maritime and snow rich districts of the Scandes, where sufficient soil mois- ture would still be available (cf. Öberg & Kullman 2011a). In this context, it is important to stress also that recent treeline rise is manifested by small and sparsely distributed trees. We can see no indica- tion whatsoever for a pending landscape-scale transformation from alpine tundra to closed birch forest, as young seedlings and sapling are only sparsely spread above the treeline. However, in- filling of snow glades in the upper birch forest belt by seeded birches occurs commonly in the mari- time area (Kullman 2007b; Kullman & Öberg 2009), which implies some densification of the treeline ecotone. In a long-term perspective, puta- tive warming and soil drought may reach critically high levels also in the maritime region, which would cause disintegration and extirpation of the birch belt and its replacement with the more drought-resistant Pinus sylvestris (Kullman & Kjäll- gren 2006; Kullman 2010a). An analogous projec- tion, based on recent vegetational trajectories, has been made for Betula pubescens ssp. czerepanow- ii in the Ural Mountains of Russia (Kapralov et al. 2006). Conclusions • During the period 2007/2008–2010/2011, treeline advance and stability/retreat prevailed in areas with maritime and continental climate, respectively. • Current treeline performance occurred in a con- text of historically high summer and winter air temperatures and raised root zone tempera- tures. • Treeline rise was proximately conditioned by re- lease from snow cover constraint on birch es- tablishment and growth. • Treeline stability and retreat related to soil drought as a consequence of more or less com- plete melt-out of most snow patches already during the early summer. • Prior treeline rise (1915−2007), predominantly accomplished by increased height growth of relict krummholz birches, has recently given way to seed-based establishment of new trees above the former treeline (2005−2007). • The discrepant response pattern between treelines in maritime versus continental cli- mates, despite a similar course of climate change, should be accounted for in projective vegetation models for high mountain regions. • In case of continued warming, the subalpine birch forest is unlikely to expand on a broad front over the alpine tundra since the continen- tal type of snow phenology conditions and as- sociated birch responses will eventually come to prevail over much of the Scandinavian moun- tain world. 36 FENNIA 190: 1 (2012)Lisa Öberg and Leif Kullman REFERENCES Alexandersson H 2006. Klimat i förändring. En jäm- förelse av temperatur och nederbörd 1991−2005 med 1961−1990. SMHI Faktablad 29, 1–8. Aune S, Hofgaard A & Söderström L 2011. Contrast- ing climate- and land-use-driven tree encroach- ment pattern of subarctic tundra in northern Nor- way and the Kola Peninsula. Canadian Journal of Forest Research 41: 3, 437–449. Atkinson MD 1992. Betula pendula Roth (Betula ver- rucosa Ehrh.) and B. pubescens Ehrh. Biological Flora of the British Isles 175. Journal of Ecology 80: 4, 837–870. Barber PV, Juday GP & Finney B 2000. Reduced growth of Alaska white spruce in the twentieth century from temperature-induced drought stress. Nature 405, 668–673. Barnett C, Dumayne-Peaty L & Matthews JA 2001. Holocene climatic change and tree-line response in Leirdalen, central Jotunheimen, south central Norway. Review of Palaeobotany and Palynology 117: 1, 119–137. Batllori E & Gutièrrez E 2008. Regional treeline dy- namics in response to global change in the Pyr- enees. Journal of Ecology 96: 6, 1275–1288. Beniston M 2003. Climatic change in mountain re- gions: a review of possible impacts. Climatic Change 59: 1−2, 5–31. Bergman J, Hammarlund D, Hannon G, Barnekow L & Wohlfarth B 2005. Deglacial vegetation succes- sion and Holocene tree-limit dynamics in the Scandes Mountains, west-central Sweden: strati- graphic data compared to megafossil evidence. Review of Palaeobotany and Palynology 134: 3−4, 129–151. Björk RG & Molau U 2007. Ecology of alpine snowbeds and the impact of global change. Arc- tic, Antarctic, and Alpine Research 39: 1, 34–43. Caccianiga M & Payette S 2006. Recent advance of white spruce (Picea glauca) in the coastal tundra of the eastern shore of Hudson Bay (Quebec, Canada). Journal of Biogeography 33: 12, 2120– 2135. Cairns DM & Moen J 2004. Herbivory influences tree lines. Journal of Ecology 92: 6, 1019–1024. Danby RK & Hik DS 2007. Variability, contingency, and rapid change in recent subarctic alpine treeline dynamics. Journal of Ecology 95: 2, 352– 363. Daniels LD & Veblen TT 2004. Spatiotemporal influ- ences of climate on altitudinal treelines in north- ern Patagonia. Ecology 85: 5, 1284−1296. Devi N, Hagedorn F, Moiseev P, Bugmann H, Shiya- tov S, Mazepa V & Rigling A 2008. Expanding for- ests and changing growth forms of Siberian larch at the Polar Ural treeline during the 20th century. Global Change Biology 14: 7, 1581–1591. Driscoll WW, Wiles GC, D’Arrigo RD & Wilmking M 2005. Divergent tree growth response to recent climatic warming, Lake Clark National Park and Preserve. Geophysical Research Letters 32, L20703. Dullinger S, Dirnböck T & Grabherr G 2004. Model- ling climate change-driven treeline shifts: relative effects of temperature increase, dispersal and in- visibility. Journal of Ecology 92: 2, 241–252. Elliott GP 2011. Influences of 20th century warming at the upper tree line contingent on local-scale in- teractions: evidence from a latitudinal gradient in the Rocky Mountains, USA. Global Ecology and Biogeography 20: 1, 46–57. Elliott GP & Kipfmueller KF 2010. Multi-scale influ- ences of slope aspects and spatial pattern on eco- tonal dynamics at upper treeline in the Southern Rocky Mountains, USA. Arctic, Antarctic, and Al- pine Research 42: 1, 45–56. Elliott GP & Kipfmueller KF 2011. Multiscale influ- ences of climate on upper treeline dynamics in the southern Rocky Mountains, USA: evidence of intraregional variability and and bioclimatic thresholds in response to twentieth-century warm- ing. Annals of the Association of American Geog- raphers 101: 6, 1181–1203. Esper J & Schweingruber FH 2004. Large-scale tree- line changes recorded in Siberia. Geophysical Re- search Letters 31, 1–5. Fagre DB, Peterson DL & Hessl AE 2003. Taking the pulse of mountains: ecosystem responses to cli- matic variability. Climatic Change 59: 1, 263– 282. Gamache I & Payette S 2005. Latitudinal response of subarctic tree lines to recent climate change in Eastern Canada. Journal of Biogeography 32: 5, 849–862. Gansert D 2004. Treelines of the Japanese Alps − al- titudinal distribution and species composition un- der contrasting winter climates. Flora 199: 2, 143−156. Gehrig-Fasel J, Guisan A & Zimmermann NE 2007. Tree line shifts in the Swiss Alps. Climate change and land abandonment. Journal of Vegetation Sci- ence 18: 4, 571–582. Gehrig-Fasel J, Guisan A & Zimmermann NE 2008. Evaluating thermal treeline indicators based on air and soil temperature using an air-to-soil tempera- ture transfer model. Ecological Modeling 213: 3−4, 345−355. Grace J 1997. The oceanic tree-line and the limit for tree growth in Scotland. Botanical Journal of Scot- land 49: 2, 223–236. Grace J, Berninger F & Nagy L 2002. Impacts of cli- mate change on the tree line. Annals of Botany 90, 537–544. Graumlich LJ, Waggoner LA & Bunn AG 2005. De- tecting global change at alpine treeline: coupling paleoecology with contemporary studies. In Hu- ber UM, Bugmann HKM & Reasoner MA (eds). Global change and mountain regions, 501−508. Springer, Dordrecht. Green K & Pickering CM 2009. The decline of snow patches in the Snowy Mountains of Australia: im- FENNIA 190: 1 (2012) 37Contrasting short-term performance of mountain birch … portance of climate warming, variable snow, and wind. Arctic, Antarctic, and Alpine Research 41: 2, 212–218. Hall MHP & Fagre DB 2003. Modeled climate-in- duced glacier change in Glacier National Park, 1850−2100. BioScience 53: 2, 131–140. Hämet-Ahti L 1963. Zonation of the mountain birch forests in northernmost Fennoscandia. Annales Botanici Societatis Zoologicæ Botanicæ Fennicæ “Vanamo” 34, 1–127. Hämet-Ahti L & Ahti T 1969. The homologies of the Fennoscandian Mountain and coastal birch for- ests in Europe and North America. Vegetatio 19: 1−6, 208–219. Harris SA 2001. Twenty years of data on climate-per- mafrost-active layer variations at the lower limit of alpine permafrost, Marmot Basin, Jasper National Park. Geografiska Annaler 83A: 1−2, 1–14. Harsch MA, Hulme PE, McGlone MS & Duncan RP 2009. Are treelines advancing? A global meta- analysis of treeline response to climate warming. Ecology Letters 12: 10, 1040–1049. Harsch MA & Bader MY 2011. Treeline form – a po- tential key to understanding treeline dynamics. Global Ecology and Biogeography 20: 4, 582– 596. Hedenås H, Olsson H, Jonasson C, Bergstedt J, Dahl- berg U & Callaghan T 2011. Changes in tree growth, biomass and vegetation over a 13-year period in the Swedish sub-Arctic. Ambio 40: 6, 672–682. Hoch G & Körner C 2003. The carbon charging of pines at the climatic treeline: a global compari- son. Oecologia 135: 1, 10–21. Holtmeier F-K 2003. Mountain timberlines: ecology, patchiness and dynamics. Kluwer, Dordrecht. Holtmeier F-K & Broll G 2005. Sensitivity and re- sponse of northern hemisphere altitudinal and po- lar treelines to environmental change at land- scape and local scales. Global Ecology and Bio- geography 14: 5, 395–410. Holtmeier F-K & Broll G 2007. Treeline advance – driving processes and adverse factors. Landscape Online 1, 1–33. Holtmeier F-K & Broll G 2010. Wind as an ecological agent at treeline in North America, the Alps and the European Subarctic. Physical Geography 31: 3, 203–233. Holway JG & Ward RT 1963. Snow and meltwater effects in an area of Colorado Alpine. The Ameri- can Midland Naturalist 69, 189–197. Jonsson TH 2004. Stature of subarctic birch in rela- tion to growth rate, life span and tree form. Annals of Botany 94: 5, 753–762. Juntunen V, Neuvonen S, Norokorpi Y & Tasanen T 2002. Potential for timberline advance in north-- ern Finland, as revealed by monitoring during 1983−99. Arctic 55: 4, 348–361. Kallio P & Mäkinen Y 1978. Vascular flora of Inari Lapland. 4. Betulaceae. Reports from the Kevo Subarctic Research Station 14, 38–63. Kapralov DS, Shiyatov S, Moiseev P & Fomin V 2006. Changes in composition, structure, and altitudinal distribution of low forest at the upper limit of their growth in the Northern Ural Mountains. Russian Journal of Ecology 37: 6, 367–372. Karlsson PS & Nordell O 1996. Effects of soil tem- perature on the nitrogen economy and growth of mountain birch seedlings near its present low temperature distribution limit. Ecoscience 3: 2, 183–189. Keller F & Körner C 2003. The role of photoperiodism in alpine plant development. Arctic, Antarctic and Alpine Research 35: 3, 361–368. Kharuk VI, Ranson KJ, Im ST & Dvinskaya ML 2009. Response of Pinus sibirica and Larix sibirica to cli- mate change in southern Siberia alpine forest- tundra. Scandinavian Journal of Forest Research 24: 2, 130–139. Kilander S 1955. Kärlväxternas övre gränser på fjäll i sydvästra Jämtland samt angränsande delar av Härjedalen och Norge. Acta Phytogeographica Suecica 35, 1–198. Kilander S 1965. Alpine zonation in the southern part of the Swedish Scandes. Acta Phytogeographica Suecica 50, 78–84. Kirdyanov AV, Hagedorn F, Knorre A, Fedotova EV, Vaganov EA, Naurzbaev MM, Moiseev PA & Ri- gling A 2012. 20th century tree-line advance and vegetation changes along an altitudinal transect in the Putorana Mountains, northern Siberia. Boreas 41: 1, 56–67. Kjällgren L & Kullman L 1998. Spatial patterns and straucture of the mountain birch tree-limit in the southern Swedish Scandes – a regional perspec- tive. Geografiska Annaler 80A: 1, 1–16. Kjällgren L & Kullman L 2002. Geographical patterns of tree-limits on Norway spruce and Scots pine in the southern Swedish Scandes. Norwegian Jour- nal of Geography 56: 4, 237–245. Komárková V & Wielgolaski FE 1999. Stress and dis- turbance in a cold region ecosystem. In Walker R (ed). Ecosystems of the world, 39–122. Elsewier, Amsterdam. Körner C 1999. Alpine Plant Life. Functional Plant Ecology and High Mountain Ecosystems. Spring- er, Berlin. Körner C 2005. The green cover of mountains in a changing environment. In Huber UM, Bugmann HKM & Reasoner MA (eds). Global change and mountain regions, 367–375. Springer, Dor- drecht. Körner C 2007. Climatic treelines: conventions, glo- bal patterns, causes. Erdkunde 61: 4, 316–324. Körner C 2012. Alpine Treelines. Springer, Basel. Körner C & Paulsen J 2004. A world-wide study of high altitude treeline temperatures. Journal of Bio- geography 31: 5, 713–732. Kullman L 1979. Change and stability in the altitude of the birch tree-limit in the southern Swedish Scandes. Acta Phytogeographica Suecica 65, 1–121. 38 FENNIA 190: 1 (2012)Lisa Öberg and Leif Kullman Kullman L 1981. Some aspects of the ecology of the Scandinavian subalpine birch forest belt. Wahlen- bergia 7, 99–112. Kullman L 1986. Demography of Betula pubescens ssp. tortuosa sown in contrasting habitats close to the birch tree-limit in Central Sweden. Vegetatio 65: 1, 13–20. Kullman L 1993. Tree limit dynamics of Betula pubes- cens ssp. tortuosa in relation to climate variability: evidence from central Sweden. Journal of Vegeta- tion Science 4: 6, 765–772. Kullman L 1995. Holocene tree-limit and climate his- tory from the Scandes Mountains, Sweden. Ecol- ogy 76: 8, 2490–2502. Kullman L 2003. Recent reversal of Neoglacial cli- mate cooling trend in the Swedish Scandes as evi- denced by mountain birch treelimit rise. Global and Planetary Change 36: 1, 77–88. Kullman L 2004a. Tree-limit landscape evolution at the southern fringe of the Swedish Scandes (Dala- na province) – Holocene and 20th century per- spectives. Fennia 182: 2, 73–94. Kullman L 2004b. Early Holocene appearance of mountain birch (Betula pubescens ssp. tortuosa) at unprecedented high elevations in the Swedish Scandes: megafossil evidence exposed by recent snow and ice recession. Arctic, Antarctic, and Al- pine Research 36: 2, 172–180. Kullman L 2005. Gamla och nya träd på Fulufjället – vegetationshistoria på hög nivå. Svensk Botanisk Tidskrift 99: 6, 315–329. Kullman L 2007a. Long-term geobotanical observa- tions of climate change impacts in the Scandes of West-Central Sweden. Nordic Journal of Botany 24: 4, 445–467. Kullman L 2007b. Modern climate change and shift- ing ecological states of the subalpine/alpine land- scape in the Swedish Scandes. Geo-Öko 28: 3−4, 187–221. Kullman L 2010a. One century of treeline change and stability – experiences from the Swedish Scandes. Landscape Online 17, 1–31. Kullman L 2010b. A richer, greener and smaller al- pine world: review and projection of warming- induced plant cover change in the Swedish Scan- des. Ambio 39: 2, 159–169. Kullman L 2012. The alpine treeline ecotone in the southernmost Swedish Scandes – dynamism on different scales. In Myster R (ed). Ecotones be- tween forest and grassland, 271−298. Springer, New York. Kullman L & Kjällgren L 2006. Holocene treeline ev- olution in the Swedish Scandes: recent treeline rise and climate change in a long-term perspec- tive. Boreas 35: 1, 159–168. Kullman L & Öberg L 2009. Post-Little Ice Age tree line rise and climate warming in the Swedish Scandes: a landscape ecological perspective. Journal of Ecology 97: 3, 415–429. Kupfer JA & Cairns DM 1996. The suitability of mon- tane ecotones as indicators of global climatic change. Progress in Physical Geography 20: 3, 253–272. Laaksonen K 1976. The dependence of mean air tem- perature upon altitude and latitude in Fennoscan- dia (1921−1950). Annales Academiæ Scientiarum Fennicæ: AIII: 119, 5–19. Leonelli G, Pelfini M, Morra di Cella U & Garavaglia V 2011. Climate warming and the recent treeline shift in the European Alps: the role of geomorpho- logical factors in high-altitude sites. Ambio 40: 3, 264–273. Lloyd AH 1997. Response of tree-line populations of foxtail pine (Pinus balfouriana) to climate varia- tion over the last 1000 years. Canadian Journal of Forest Research 27: 6, 936−942. Lloyd AH 2005. Ecological histories from Alaskan tree lines provide insight into future change. Ecol- ogy 86: 7, 1687–1695. Lloyd AH & Fastie C 2002. Spatial and temporal vari- ability in the growth and climate response of treeline trees in Alaska. Climatic Change 52: 4, 481–509. Luckman B & Kavanagh T 2000. Impact of climate fluctuations on mountain environments in the Ca- nadian Rockies. Ambio 29: 7, 371–380. MacDonald GM, Kremenetski KV & Beilman DW 2008. Climate change and the northern Russian treeline zone. Philosophical Transactions of the Royal Society B 363, 2285–2299. Malanson GP, Resler LM, Bader MY, Holtmeier F-K, Butler DR ,Weiss DJ, Daniels LD & Fagre DB 2011. Mountain treelines: a roadmap for research orientation. Arctic, Antarctic, and Alpine Re- search 43: 2, 167–177. Miehe G & Miehe S 2000. Comparative high moun- tain research on the treeline ecotone under hu- man impact. Erdkunde 54: 1, 34–50. Millar CI, Westfall RD, Delany DL, King JC & Graum- lich LJ 2004. Response of subalpine conifers in the Sierra Nevada, California, USA, to 20th-cen- tury warming and decadal climatic variability. Arctic, Antarctic, and Alpine Research 36: 2, 181– 200. Millar C, Westfall RD, Delany D, Bokach MJ, Flint AL & Flint LE 2012. Forest mortality in high-elevation whitebark pine (Pinus albicaulis) forests of eastern California, USA; influence of environmental con- text, bark beetles, climatic water deficit, and warming. Canadian Journal of Forest Research 42: 4, 749−765. Moberg A, Tuomenvirta AH & Nordli Ø 2005. Recent climatic trends. In Seppälä M (ed). The physical geography of Fennoscandia, 112–133. Oxford University Press, Oxford. Moen J, Aune K, Edenius L & Angerbjörn A 2004. Potential effects of climate change on treeline po- sition in the Swedish mountains. Ecology and So- ciety 9(1): 16, 1–11. Mossberg B & Stenberg L 2003. Den nya nordiska floran. Wahlström & Widstrand, Stockholm. FENNIA 190: 1 (2012) 39Contrasting short-term performance of mountain birch … Nagy L 2006. European high mountain (alpine) veg- etation and its suitability for indicating climate change impacts. Biology and Environment: Pro- ceedings of the Royal Irish Academy 106B: 3, 335–341. Neilson RP 1993. Transient ecotone response to cli- matic change: Some conceptual and modelling ap- proaches. Ecological Applications 3: 3, 489–498. Nesje A, Kvamme M & Løvlie R 1991. Holocene gla- cial and climate history of the Jostedalsbreen re- gion, western Norway; evidence from lake sedi- ments and terrestrial deposits. Quaternary Science Reviews 10: 6, 87–114. Noble IR 1993. A model of the responses of ecotones to climate change. Ecological Applications 3: 3, 396–403. Öberg L & Kullman L 2011a. Recent glacier recession – a new source of postglacial treeline and climate history in the Swedish Scandes. Landscape On- line 26, 1–38. Öberg L & Kullman L 2011b. Ancient subalpine clon- al spruces (Picea abies): sources of postglacial vegetation history in the Swedish Scandes. Arctic 64: 2, 183–196. Odland A 2010. Importance of mountain height and latitude for the altitudinal distribution limits of vascular plants in Scandinavia: are the mountains high enough? Fennia 188: 2, 149–162. Olofsson J, Oksanen L, Callaghan T, Hulme EP, Oksa- nen T & Suominen O 2009. Herbivores inhibit climate-driven shrub expansion on the tundra. Global Change Biology 15: 11, 2681–2693. Østbye F 1975. Hardangervidda, Norway. Ecological Bulletins 20, 225–264. Paus A 2010. Vegetation and environment of the Rødalen alpine area, Central Norway, with em- phasis on the early Holocene. Vegetation History and Archaeobotany 19: 1, 29–51. Payette S 2007. Contrasted dynamics of northern Labrador tree lines caused by climate change and migrational lag. Ecology 88: 3, 770–780. Payette S, Fortin M-J & Gamache I 2001. The subarc- tic forest-tundra: the structure of a biome in a changing climate. BioScience 51: 9, 709–718. Peterson DH 1998. Climate, limiting factors and en- vironmental change in high-altitude forests of Western North America. In Beniston M & Innes JL (eds). The Impacts of Climate Variability on For- ests, 191−208. Springer, Berlin. Philipp M 1978. Vegetation of a snow bed at God- hawn, west Greenland. Holarctic Ecology 1: 1, 45–53. Potthoff K 2009. Grazing history affects the tree-line ecotone: a case study from Hardanger, Western Norway. Fennia 187: 2, 81–98. Raab B & Vedin H 1995. Climate, lakes and rivers. In National Atlas of Sweden. Sveriges Nationalatlas Förlag, Stockholm. Rössler O, Bräuning A & Löffler J 2008. Dynamics and driving forces of treeline fluctuation and re- generation in central Norway during the past dec- ades. Erdkunde 62: 2, 117–128. Samuelsson G 1917. Studien über die Vegetation der Gebirgsgegenden von Dalarne. Nova Acta Regiae Societatis Scientiarum Upsaliensis Series 4: 4, 1–253. Segerström U & von Stedingk H 2003. Early- Holocene spruce, Picea abies (L.) Karst, in west central Sweden as revealed by pollen analysis. The Holocene 13: 6, 897–906. Schmidt NM, Battinger C & Forchhammer MC 2006. Reconstructing century-long snow regimes using estimates of High Arctic Salix arctica radial growth. Arctic, Antarctic, and Alpine Research 38: 2, 257–262. Seppälä M 2004. Wind as a geomorphic agent in cold climates. Cambridge University Press, Cam- bridge. Shiyatov SG, Terentev MM, Fomin VV & Zimmerman NE 2007. Altitudinal and horizontal shifts of the upper boundaries of open and closed forests in the polar Urals in the 20th century. Russian Jour- nal of Ecology 38: 4, 223–227. Shrestha BB, Ghimire B, Lekhak HD & Jha PK 2007. Regeneration of treeline birch (Betula utilis D. Don) forest in a Trans-Himalayan dry valley in Central Nepal. Mountain research and Develop- ment 27: 3, 259−267. Singh CP, Panigrahy S, Thapliyal A, Kimothi MM, Soni P & Parihar JS 2012. Monitoring the alpine treeline shift in parts of the Indian Himalayas us- ing remote sensing. Current Science 102: 10, 559−562. SMHI 2012. 19.02.2012. Smith H 1920. Vegetationen och dess utvecklingshis- toria i det centralsvenska högfjällsområdet. Almqvist & Wiksell, Uppsala. Smith H 1957. En botanisk undersökning av Neans dalgång. Svenska Vetenskapsakademiens Av- handlingar i Naturskyddsärenden 16, 1–21. Smith WK, Germino MJ, Hancock TE & Johnson DM 2003. Another perspective on altitudinal limits of alpine timberlines. Tree Physiology 23: 16, 1101– 1112. Staland H, Salomonsson J & Hörnberg G 2010. A thousand years of human impact in the northern Scandinavian mountain range. Long-lasting ef- fects on forest lines and vegetation. The Holocene 21: 3, 379–391. Sundqvist MK, Björk RG & Molau U 2008. Establish- ment of boreal forest species in alpine dwarf- shrub heath in subarctic Sweden. Plant Ecology & Diversity 1: 1, 67−75. Sveinbjörnsson B, Nordell O & Kauhanen H 1992. Nutrient relations of mountain birch growth at and below the elevational tree-line in Swedish Lap land. Functional Ecology 6, 213–220. 40 FENNIA 190: 1 (2012)Lisa Öberg and Leif Kullman Tape K, Sturm M & Racine C 2006. The evidence for shrub expansion in Northern Alaska and the Pan- Arctic. Global Change Biology 12: 4, 686–702. Taubes G 1995. Is a warmer climate wilting the for- ests of the North? Science 267: 5204, 1595. Tinner W & Kaltenrieder P 2005. Rapid responses of high-mountain vegetation to early Holocene envi- ronmental changes in the Swiss Alps. Journal of Ecology 93: 5, 936–947. Tuhkanen S 1980. Climatic parameters and indices in plant geography. Acta Phytogeographica Suecica 67, 1–105. Van Bogaert R, Haneca K, Hoogesteger J, Jonasson C, De Dapper M & Callaghan T 2011. A century of tree line changes in sub-arctic Sweden show local and regional variability and only a minor role of 20th century climate warming. Journal of Bioge- ography 38: 5, 907–921. Vaartaja O 1955. Factors causing mortality of tree seeds and succulent seedlings. Acta Forestalia Fennica 62: 3, 1−31. Virtanen R, Luoto M, Rämä T, Mikkola K, Hjort J, Grytnes J-A & Birks HJB 2010. Recent vegetation changes at the high-latitude tree line ecotone are controlled by geomorphological disturbance, pro- ductivity and diversity. Global Ecology and Bioge- ography 19: 6, 810–821. Walsh SJ, Butler DR, Allen TR & Malanson GP 1994. Influence of snow patterns and snow avalanches on the alpine treeline ecotone. Journal of Vegeta- tion Science 5: 5, 657–672. Weissberg PJ & Baker WL 1995. Spatial variation in tree regeneration in the forest-tundra ecotone, Rocky Mountain National Park, Colorado. Cana- dian Journal of Forest Research 25: 8, 1326–1339. Wielgolaski FE 1975. Functioning of Fennoscandian tundra ecosystems. In Wielgolaski FE (ed). Fenno- scandian tundra ecosystems, Part 2. Animals and systems analysis, 300–326. Springer, Berlin. Wielgolaski FE & Kärenlampi L 1975. Plant ecology of Fennoscandian tundra areas. In Wielgolaski FE (ed). Fennoscandian tundra ecosystems, Part 1. Plants and microorganisms, 94–102. Springer, Berlin. Wilmking M, D’Arrigo R, Jacoby GC & Juday GP. 2005. Increased temperature sensitivity and diver- gent growth trends in circumpolar boreal forest. Geophysical Research Letters 32, L15715. Wipf S, Stoeckli V & Bebi P 2009. Winter climate change in alpine tundra: plant responses to changes in snow depth and snowmelt timing. Cli- matic Change 94: 1−2, 105−121. Young AB, Cairns DM, Lafon CW, Moen J & Martin LE 2011. Dendroclimatic relationships and possi- ble implications for mountain birch and Scots pine at treeline in northern Sweden through the 21st century. Canadian Journal of Forest Research 41: 3, 450–459.