Page 1 of 18
European Journal of Applied Sciences – Vol. 10, No. 3
Publication Date: June 25, 2022
DOI:10.14738/aivp.103.12369. Kullman, L. (2022). Forest-Limit (Betula pubescens ssp. czerepanovii) Performance in the Context of Gentle Modern Climate
Warming. European Journal of Applied Sciences, 10(3). 168-185.
Services for Science and Education – United Kingdom
Forest-Limit (Betula pubescens ssp. czerepanovii) Performance in
the Context of Gentle Modern Climate Warming
Leif Kullman
Department of Ecology and Environmental Science
Umeå University, 901 87 Umeå
ABSTRACT
In the context of post-Little Ice Age climate change, regional forest-limit (strictly
defined) adjustment was studied over the period 1915 to 1975. The study
comprised nearly 200 sites, in the form of systematically distributed belt transects.
Forest-limit positions (m a.s.l.) were reconstructed for the years around 1915 and
compared to the situation about 1975. During that interval, summer temperatures
(June-August) increased by 1.2 °C. Concurrently, elevational upshifts were
ubiquitous, site specific and rather modest; c. 17 altitudinal meter rise over a period
of about 60 years. This is far less than expected for the climate change alone and
indicates a pronounced disequilibrium situation. In comparison, the treeline in the
same area, i.e. the uppermost > 2 m tall birches, advanced by average c. 30 m. Thus,
it appears that the treeline is a more sensitive biomonitor than the forest-limit.
Therefore, the treeline should be primarily focused in the context of environmental
monitoring. Given that the current relatively warm climate phase continues, the
subalpine birch forest belt may eventually recede and give way to a subalpine pine
belt. The obtained modest forest-limit advancement is so small that flourishing
model simulations of extensive birch forest expansion over most of the current
alpine tundra, appear as large and unfounded exaggerations. The alpine tundra of
the Scandes, with its characteristic biodiversity, is likely to prevail for a foreseeable
future.
INTRODUCTION
The upper distribution of biota in high-mountain regions is ultimately constrained by thermal
deficiency (Tranquillini 1979; Körner & Paulsen 2004; Holtmeier & Broll 2005; Nagy 2006). In
that context, arboreal vegetation plays a pivotal role, as its dynamics influences the composition
and structure of lower vegetation strata, surface albedo and carbon sequestration (Theurillat
& Guisan 2011; Camarero & Gutiérrez 2004; Greenwood et al. 2016). It is well known that, on
most time-scales, upper ”treelines” respond sensitively to climate change and variability
(Kullman 1998, 2021a; Fagre et al. 2003; Kullman & Öberg 2009; Hofgaard et al. 2009; Liang et
al. 2011; Holtmeier & Broll 2011). However, less is known about couplings between climate
change and the upper boundary of closed forest, henceforth termed ”forest-limit” (Donato
2013). This is an important matter, since projections, drawing on proposed anthropogenic
future warming, claim extensive and pending elevational forest-limit expansion at the cost of
alpine tundra over the present century. Upshifts by 200-600 altitudinal meters are simulated
and are assumed to change the general appearance of the mountain region (Kellomäki et al.
1997; Moen et al. 2004; ACIA 2005; Kaplan & New 2006). The credibility of such projections is
dubious and based on immature climate and ecological science, since few observational studies
Page 2 of 18
169
Kullman, L. (2022). Forest-Limit (Betula pubescens ssp. czerepanovii) Performance in the Context of Gentle Modern Climate Warming. European
Journal of Applied Sciences, 10(3). 168-185.
URL: http://dx.doi.org/10.14738/aivp.103.12369
have specifically focused on modern elevational shifts of the forest-limit, strictly defined.
Accordingly, more conservative and seemingly realistic inferences are launched by other
researchers (e.g. Rössler et al. 2008; Willis & MacDonald 2011; Hofgaard et al. 2013; Bandekar
& Odland 2017; Kullman 2019). The sparsity of relevant studies in this field may relate to the
difficulty of finding a stringent forest-limit definition, which is practically applicable in the field,
particularly in retrospect.
It is of utmost importance for dynamic landscape comprehension to uphold the distinction
between the treeline and the forest-limit, as they constitute opposite ends of a vegetation
continuum, the so called treeline ecotone (cf. Treter 1984). Possibly, they respond to climate
change and variability with different time lags (Hermes 1955; Körner 2007; Bekker & Malanson
2009; Kharuk et al. 2010; Rannow 2013; Kullman 2021c; Bryn & Potthoff 2022). The present
study focuses specifically on temporal dynamics of the forest-limit.
With reference to Kullman (1979), the narrow and precise concept ”treeline” represents the
altitude (m a.s.l.) at any specified locality, where the uppermost birch, at least 2 m tall, grows at
a certain point of time. ”Forest-limit” refers to the highest point (m a.s.l.) at which birches of
tree height form closed stands at specific points of time. The configuration of the forest-limit is
highly variable between sites and may take forms as wedges, lobes, ”fingers” or clumps. For the
present analysis, the minimum area should be at least 5x5 m and contain 10 or more tree-sized
stems. Unfortunately, we have few direct observational data on forest-limit positions in the
early 20th century. Therefore, the past forest-limit has to be based on present-day visual
records and estimates of the upper limit of still living or dead stems which, based on learning
from extensive borings, could be judged to have been at least 2 m and alive in 1915 (Kullman
1979). Of course this approach implies some uncertainty, reasonably minimized by a large
number of study sites.
METHODS
The forest-limit dynamics 1915-1975 was reconstructed by the use of data from a regional
network of observations (1915) at 198 sites over a 8000 km2 area (Smith 1920), combined with
field observations in the mid-1970s and contemporary dendroecological analyses (Kullman
1979, updated). Records were obtained within elevational transects, running at right-angles to
the local slope contours. The altitudinal difference between the past and recent forest-limits,
defines the extent of forest-limit change over the period 1915-1975. It provides a measure of
the extent of afforestation of previously unforested alpine tundra.
Altitudinal measurements were carried out with a Paulin aneroid altimeter, with a reading
accuracy of±1 m, here rounded off to the nearest 5 m. Measurements were repeatedly
calibrated against modern topographical maps (1: 100 000).
Elevational forest-limit shift 1901-1975 was compared to treeline shift over the same period of
time (Kullman 1979). Significant differences were computed by one-way analysis of variance
(ANOVA), checked for normality.
Page 3 of 18
170
European Journal of Applied Sciences (EJAS) Vol. 10, Issue 3, June-2022
Services for Science and Education – United Kingdom
Study area
The study area comprises the provinces of Dalarna, Härjedalen and Jämtland in the southern
Swedish Scandes (Fig.1). The relief is alpine, with U-shaped valleys and the highest peaks
reaching 1000-1800 m a.s.l. The bedrock consists of amphibolites, quartzites, sparagmites and
calcareous tectonites. Podzolic soil and peat deposits cover the prevailing glacial deposits. The
regional climate ranges from weakly maritime in the northwest to a more continental character
in the eastern and southern parts. Data from Storlien/Visjövalen meteorological station (642 m
a.s.l.) display that mean temperatures for the months of January, July and the year, are -7.6, 10.7
and 1.1 oC, respectively, referring to the period 1961-1990.
Arboreal vegetation within the treeline ecotone is composed of Norway spruce (Picea abies L,
Karst,) and Scots pine (Pinus sylvestris), as alternating dominants. These sparse stands grade
upslope into a subalpine belt with predominant mountain birch (Betula pubescens ssp.
czerepanovii. Detailed accounts of the structure, composition and dynamics of the treeline
ecotone are given by different sources (Kullman 1981b, 2005; Carlsson et al. 1999; Jonsson
2004; Bandekar & Odland 2017, Kullman & Öberg 2018b). The treeline ecotone has been
impacted by human utilization for thousands of years; reindeer pastoralism, alternating with
livestock grazing and hay making by Nordic settlers (Kullman 1979; Josefsson et al. 2010;
Östlund et al. 2015).
Concerning impact of wild and semi-domestic reindeer, it has been claimed that grazing and
trampling constrain growth and reproduction of mountain birch in the treeline ecotone and
disguising the impact of climate warming (Cairns & Moen 2004; Olofsson et al. 2009; Van
Bogaert et al. 2011). This contention conflicts with opinions concerning the mountainscape in
general (Holmgren 1904; Kallio & Mäkinen 1978; Holmgren & Tjus 1996; Holtmeier 2003), as
well as age structure data (Kullman 2021b). In particular Tømmervik et al. (2009) found that
reindeer grazing reduces the lichen cover and may promote establishment and growth of
mountain birch. Accordingly, substantial regional treeline rise (birch and pine) during the past
100 years (Kullman & Öberg 2009) coincides with steadily rising reindeer herds (Kullman
2017b). Counter-intuitively, and with this background of empirical facts, it needs consideration
whether birch forest-limit progression and general landscape greening since early 20th century
is the combined effect of climate warming and increased grazing/trampling by free-roaming
reindeer.
In general, the treeline ecotone has been little affected by fire disturbance in the past (Kullman
1981a).
Impacts of past human land-use may display fundamental differences between of forest-limit
and treeline. It appears that treelines are less impacted in that respect, as a reflection of the
“Principle of Least Effort”. This implies that the use and of natural resources diminishes
towards the treeline, where the cost and effort of extraction exceeds the potential gain
(Blüthgen 1942; Shackleton & Prins 1992). Consequently, the treeline is generally in a more
natural stage and constitutes a relatively better and more sensitive bioindicator of climate
change than the forest-limit (Kullman 2008, 2015, 2017b; Körner 2007; Guo et al. 2005; Aitken
et al. 2008).