Upload
others
View
2
Download
0
Embed Size (px)
Citation preview
THE REGENERATION NICHE OF TREES AT THE ALPINE TREELINE
-
CONSTRAINTS OF MICROCLIMATE AND THE ALPINE GRASSLAND
VEGETATION ON GERMINATION AND SEEDLING ESTABLISHMENT
Die von der Fakultät für Mathematik und Naturwissenschaften
der Carl von Ossietzky Universität Oldenburg
zur Erlangung des Grades und Titels eines
Doktors der Biologie (Dr. rer. nat.)
angenommene Dissertation von
Hannah Loranger, geboren am 26.11.1985 in Duisburg
Erstgutachter: Prof. Dr. Gerhard Zotz
Zweitgutachter: Prof. Dr. Martin Diekmann
Tag der Disputation: 30.05.2016
v
CONTENTS
Chapter 1 – General Introduction ....................................................................................... 7
Defining the term “treeline”............................................................................................... 8
Historical retrospect and recent interest .......................................................................... 9
Causes of treeline formation ............................................................................................ 11
The regeneration niche of treeline trees ......................................................................... 13
Thesis outline...................................................................................................................... 15
Chapter 2 – Impacts of soil microclimate on early establishment of trees at the
alpine treeline: idiosyncratic responses and the importance of soil moisture ......... 19
Abstract ............................................................................................................................... 19
Introduction ........................................................................................................................ 21
Methods ............................................................................................................................... 25
Results ................................................................................................................................. 32
Discussion ........................................................................................................................... 39
Consistency of limiting factors during early establishment ................................... 39
Temperature, moisture, and their interactions driving early establishment
success ............................................................................................................................. 40
Species-specific responses ............................................................................................ 42
Implications for local treeline patterns and dynamics ............................................. 46
Conclusions ........................................................................................................................ 48
Acknowledgements ........................................................................................................... 48
Appendix ............................................................................................................................ 50
Chapter 3 – Competitor or facilitator? The role of grassland vegetation for
germination and seedling performance of tree species at the alpine treeline ......... 53
Abstract ............................................................................................................................... 53
Introduction ........................................................................................................................ 56
Materials and Methods ..................................................................................................... 59
Results ................................................................................................................................. 66
Discussion ........................................................................................................................... 74
Competition dominates the interaction between tree seedlings and neighbouring
vegetation ....................................................................................................................... 75
vi
Seasonal shifts from competition to facilitation depend on the leaf functional
type .................................................................................................................................. 79
Acknowledgements ........................................................................................................... 82
Appendix ............................................................................................................................ 83
Chapter 4 – A cool experimental approach to explain elevational treelines, but can
it explain them? ..................................................................................................................... 87
Abstract ............................................................................................................................... 87
Introduction ........................................................................................................................ 89
Confounding temperature conditions ............................................................................ 91
Confounding moisture conditions .................................................................................. 93
Alternative experimental setups ..................................................................................... 95
Alternative treeline-forming mechanisms in Nothofagus and other genera .............. 99
Conclusion ........................................................................................................................ 102
Appendix .......................................................................................................................... 103
Chapter 5 – Synthesis ........................................................................................................ 105
Local and intrinsic factors driving the regeneration response .................................. 106
Outlook and future research needs ............................................................................... 110
References ............................................................................................................................ 114
Summary .............................................................................................................................. 132
Zusammenfassung ............................................................................................................. 134
Danksagung ......................................................................................................................... 137
Lebenslauf ........................................................................................................................... 139
Authors’ contributions ...................................................................................................... 141
Allgemeine Erklärung ....................................................................................................... 143
7
CHAPTER 1
GENERAL INTRODUCTION
Life on earth is incredibly diverse, but none of the millions of species already
catalogued or still to discover is evenly distributed over the available space. The
arising fundamental questions of “where?”, “how?” and “why?” encompass the
central interest of ecology: to describe and understand the interactions that
determine the distribution and abundance of organisms (Krebs 1972). In this context,
a special relevance can be attributed to distributional boundaries, which offer unique
opportunities to study limiting factors and their interactions due to increasingly
stressful conditions for the focal organisms. Alpine treelines are among the most
conspicuous vegetation limits in existence, and mark not only the distribution
boundary of single tree species, but of the entire life-form “tree”, due to an
increasingly unfavourable heat balance (Körner 2012). They are at the same time
important ecological boundaries, since site conditions such as topoclimate and soil
Chapter 1
8
properties change significantly over the transition from closed forest to alpine tundra
(Holtmeier 2009). These characteristics designate alpine treelines as particularly
interesting study systems, which, although having been linked to a common
isotherm of 5 – 7 °C mean growing season temperature on a global scale (Körner and
Paulsen 2004), still lack comprehensive mechanistic explanations for locally varying
patterns and elevational positions. This thesis aims at contributing to a more
complete understanding of alpine treeline patterns and dynamics by focussing on the
early establishment of treeline trees with regard to abiotic as well as biotic potentially
limiting factors.
DEFINING THE TERM “TREELINE”
There is no general consensus reached yet regarding the definition and application of
the term “treeline”, clearly distinguishing it from similar boundary concepts such as
“timberline” or “forest line” (Körner 1998). For this reason I adopted the concise and
suitable definition established by Bader (2007), which uses treeline as an abbreviation
for treeline ecotone, as it is rather a transition zone than a sharply delimited line (Fig.
1.1). Treeline is then defined as “the existent transition between continuous upper
montane forest and continuous alpine vegetation”. Therein, the term “existent” refers
to the actual presence of a treeline, irrespective if it is at the climatic limit or
depressed by anthropogenic influences. The term “continuous” emphasizes the
“clear dominance of either vegetation type”.
General Introduction
9
Fig. 1.1 Relatively undisturbed treeline of the Rif Blanc Forest, Galibier massif,
French Alps at approximately 2300 m a.s.l. adjoining the subalpine pastures of the
historically deforested Lautaret Pass. The dominant tree species are Larix decidua and
Pinus uncinata; land-use impacts are limited to occasional sheep-grazing. A. From a
distant view the limit of tree cover appears to follow a relatively sharp line. B. A
close-up of the same treeline reveals its transitional nature, with reduced tree size
and increasing patchiness of tree cover with increasing elevation.
HISTORICAL RETROSPECT AND RECENT INTEREST
The elevation of tree cover on high mountains was essentially modified by human
activities as early as the Bronze Age (Wick and Tinner 1997), however, it took a long
time in the course of history until a scientific interest in this noticeable vegetation
boundary arose. The first descriptive records of elevational vegetation belts date to
observations of the renaissance polymath Leonardo da Vinci (1452-1519) in the
Italian Alps, followed by those of Conrad Gessner (1515-1565) in the Swiss Alps. The
latter was presumably the first to mention an elevational limit of tree growth as
related to climatic factors, namely temperature and growing season length
(Holtmeier 1965). In the following centuries, remarks on alpine treelines were rather
accidental in connection with geological or botanical studies, and only about 200
A B
Chapter 1
10
years ago this vegetation boundary itself was moved more into the focus of scientific
inquisitiveness (Marek 1910). An impressive prelude was given by the famous Latin
America – expedition of Alexander von Humboldt (1769-1859), from which he laid
the foundations of biogeography by relating elevation and plant distribution on
mountain slopes to isotherms, using treelines as an important reference point (von
Humboldt and Bonpland 1805). The initiation of actual treeline research started with
systematic measurements of treeline elevations (e.g. Wahlenberg 1813) and, as a
consequence of previous observations, the study of different thermal parameters as
potentially overarching explanation (reviewed in Holtmeier 2009). The results of this
early research were synthesized in the extensive works of Brockmann-Jerosch (1919)
and Däniker (1923), the former considering the complex climate character of treelines
on a global scale and the latter developing a thorough theory about tree growth
limitations at high elevations due to heat deficiency that is still in the centre of
debates today. Däniker was also the first to apply ecological research methods, which
were extended in the 1930s with a stronger focus on tree physiology by several
pioneers of experimental ecology (e.g. Michaelis 1934; Steiner 1935; Pisek and
Cartellieri 1939). The heavy destruction caused by avalanches in the European Alps
in the 1950s then gave rise to a new perspective on alpine treelines with regard to
social and economic value of ecosystem services, stimulating more studies on the
potential limit of protective mountain forests and motivating the establishment of
specific research stations (Obergurgl 1953, Austria; Stillberg/Davos 1959,
Switzerland; Holtmeier 2009). This increase of local or regional, often species-specific
knowledge coincided with an increasingly broader approach of treeline research,
including large-scale bioclimatological works (Daubenmire 1954; Hermes 1955) and
General Introduction
11
worldwide comparisons of alpine treelines (Troll 1973; Holtmeier 1974; Wardle 1974),
highlighting the global nature of the treeline phenomenon. From these currents of
treeline research emerged two distinct approaches with differing focuses of interest
that dominate the field today: while the global approach aims at determining the
general factors causing the formation of treelines, the local or landscape-approach
seeks to understand why treelines differ among locations (Malanson et al. 2011).
The rapidly increasing number of publications in both lines of research in
recent years is the consequence of a new global challenge, climate change. Global
temperatures have risen by about 0.7 °C over the last century and climate models
predict a further increase of 1-5 °C for the century to come (IPCC 2013). As a
thermally limited distribution boundary treelines are expected to move upslope or
poleward (Grace 2002), and indeed, there is evidence for advancing tree cover to
higher elevations and latitudes as well as increased radial growth in treeline trees
(Rolland et al. 1998; Kullman 2007; Shiyatov et al. 2007; Vittoz et al. 2008). However,
responses are not uniform, since locally stable or even receding treelines have been
reported as well (Harsch et al. 2009). In this regard, the recent interest in the treeline
topic reflects the urgent need of a more mechanistic understanding of limiting factors
at different spatial scales, which is a prerequisite for reliable predictions of treeline
shifts and the associated consequences for sensitive mountain ecosystems.
CAUSES OF TREELINE FORMATION
On a global scale, the elevational limit of trees is ultimately set by heat deficiency.
While this relationship was observed early in the history of treeline research, it was
especially the establishment of the growth-limitation hypothesis postulated by
Chapter 1
12
Körner (1998), which connected correlations of isotherms and global treeline
positions with a likely mechanism. This hypothesis comprises that the life-form tree
is characterized by an upright stem with high-reaching branches, which are tightly
coupled to the atmosphere and its temperature fluctuations. The resulting low tissue
temperatures under harsh conditions limit cell division and thus growth,
distinguishing trees from low-stature alpine vegetation with a much more favourable
heat balance. This concept is supported to the disadvantage of competing
explanations such as the carbon-limitation hypothesis (Stevens and Fox 1991) by
strong evidence: i) at the lower temperature threshold of growth of approximately 5
°C photosynthesis rates are still positive (Grace 2002; Rossi et al. 2008), ii) this lower
limit of growth corresponds well to the isotherm of 5 – 7 °C associated to positions of
treelines worldwide (Körner and Paulsen 2004), which can differ however
substantially in more stress-related factors such as seasonality, snow cover or solar
irradiance (e.g. comparing temperate and tropical alpine treelines) and iii) many
studies report an increase of mobile carbon stores with increasing elevation,
suggesting that while the assimilation of photosynthates is not restricted at treeline,
their utilisation for growth might be (e.g. Hoch and Körner 2003, 2012; Shi et al. 2008;
Fajardo et al. 2012).
However, limitations of carbon gain as well as a set of other hypothesis
regarding the causes of treeline formation including climatic stress, nutrient
deficiency and disturbances (summarized in e.g. Tranquillini 1979; Körner 1998;
Wieser and Tausz 2007) may well play an important role at a local scale. As i)
treelines can vary locally up to several hundred metres in their elevational positions,
ii) patterns range from abrupt treelines over scattered tree islands to diffuse forms
General Introduction
13
and iii) treelines do not all respond uniformly to a warming climate, locally varying
environmental factors unrelated to temperature must be responsible (Holtmeier and
Broll 2005; Harsch and Bader 2011; Case and Duncan 2014). Among the abiotic
conditions especially water availability seems to play a critical role, as moisture
deficits can be positively linked to climate warming and have the potential to
override positive treeline responses to increasing temperatures (e.g. Barber et al.
2000; Daniels and Veblen 2004). Further site-specific complexity is added by biotic
factors, because interactions with other organisms can be beneficial (Mattes 1994;
Akhalkatsi et al. 2006) or detrimental (Moir et al. 1999; Cairns and Moen 2004) for
treeline trees. These abiotic and biotic site-specific conditions, but also intrinsic
factors such as tree species ecology (Körner and Paulsen 2004) and life-stage
dependent requirements (Barbeito et al. 2012) may interact with temperature to
determine local treeline patterns and dynamics. Thus, studying these
interdependencies is essential for a more complete understanding of the mechanisms
governing alpine treelines, thereby exploring what prevents some treelines from
reaching their climatic limit and facilitating predictions of future dynamics.
THE REGENERATION NICHE OF TREELINE TREES
The regeneration niche of a species can be defined as “the expression of the
requirements for a high chance of success in the replacement of one mature
individual by another” (Grubb 1977). For treelines to remain stable or to advance
even to higher elevations, such a replacement of trees within and above this ecotone
is a necessary prerequisite. However, the regeneration niche of trees can be expected
to become constricted under the increasingly harsh conditions and the substantially
Chapter 1
14
different ecosystem properties in the alpine tundra (Fig. 1.2A). While seed
availability and successful dispersal are important precursors, germination and
seedling establishment are the critical steps that determine the compatibility between
the environment and the respective regeneration niche, and are widely recognized as
major life-history bottlenecks for treeline trees (Stevens and Fox 1991; Smith et al.
2003). Germination depends on environmental cues such as a specific temperature or
moisture range and is the first, critical life-stage transition (Baskin and Baskin 2001).
Subsequently, young seedlings are in the most vulnerable life-stage of a tree due to
their small size and the resulting low water and reserve storage capacity (Cui and
Smith 1991; Wang and Zwiazek 1999; Johnson et al. 2011). Furthermore, a different
physiology compared to adult trees (e.g. Tegischer et al. 2002) might contribute to a
varying or higher sensitivity to environmental stressors (Germino et al. 2002;
Barbeito et al. 2012; Moyes et al. 2013). These characteristics are reflected in the
natural distribution of seedlings in the treeline ecotone, which often shows a
decreasing density along the elevational gradient (Cuevas 2000) and a spatial
association with shelter elements (Germino and Smith 1999; Smith et al. 2003; Batllori
and Camarero 2009). In this context, other plants such as those forming alpine
grassland vegetation might play an ambiguous role for tree regeneration (Fig. 1.2B),
as they can not only provide shelter by improving microclimatic conditions (Smith et
al. 2003; Maher et al. 2005), but also act as strong competitors for resources (Nambiar
1990; Moir et al. 1999). Also the specific types of treelines, especially abrupt forms,
appear tightly linked to regeneration patterns (Harsch and Bader 2011), and might
arise from a lack of safe sites (sensu Harper 1977) in the alpine tundra (Tranquillini
1979). Consequently, understanding the limitations of tree recruitment at their
General Introduction
15
distribution boundary is a key component in the discussion about treeline causality
and predictions.
Fig. 1.2 A. Young Sorbus aucuparia - seedling experiencing stressful high-light
conditions on a sunny morning following a frost night. B. Freshly germinated Larix
decidua – seedling surrounded by dense herbaceous alpine vegetation, which might
act as strong competitor or as facilitator providing shelter.
THESIS OUTLINE
The aim of this thesis is to deepen the knowledge of which environmental factors,
considering abiotic and biotic interactions, constrict the regeneration niche of treeline
trees, and how. To include the potentially most important drivers of local treeline
patterns (i.e. site-specific environmental factors, life-stage-specific responses, tree
species ecology), I investigated the effect of abiotic and biotic interactions with
respect to i) the two critical, successive first life-stages germination and early
seedling establishment and ii) general or species-specific responses by comparing
five important European treeline tree species. Hereafter I will give a brief outline of
the specific focus of the following chapters 2 to 5.
Chapter 2 and 3 present results obtained in common garden field experiments
conducted in the experimental garden of the alpine research station Joseph Fourier in
the French Alps near the local treeline (Lautaret Pass, 2100 m a.s.l., Fig. 1.3). The
A B
Chapter 1
16
possibility of combining the experimental manipulation of the microclimate as well
as biotic interactions with fine-scale measurements of two critical early life-stages in
multiple species produced a unique dataset, addressing for the first time in unity this
complexity of interdependent factors and their potential to shape local treelines.
In Chapter 2 we tested the impact of microclimate – more specifically the
effects of temperature, water availability and the interaction of both – on germination
and first-year seedling survival of the focal species. Besides the well-established
importance of temperature, water availability was shown to affect treelines in general
(González de Andrés et al. 2015) and their regeneration in particular (Moyes et al.
2013). As both climate variables can be strongly interdependent, their interaction
might be more important for treeline responses than their absolute values (Ohse et al.
2012).
Chapter 3 deals with the ambiguous role of alpine grassland vegetation for
tree establishment at and above the alpine treeline, since both positive (Germino et
al. 2002; Maher et al. 2005) and negative (Moir et al. 1999; Dullinger et al. 2003)
interactions have been documented before. Here, we focussed on stress- (survival) as
well as resource-related responses (biomass, mobile carbon reserves) to establish the
relative importance of facilitative and competitive effects.
Additionally, we planned a field experiment to address the effect of the
changing microclimatic conditions for the growing tree seedling as it emerges from
the buffered vegetation cover and becomes increasingly coupled to the atmosphere,
since this important transition might account for life-stage-specific susceptibilities to
environmental stressors (Barbeito et al. 2012) and growth limitations with increasing
General Introduction
17
size (Körner 1998). To test the effect of seedling height within and above the alpine
grassland vegetation independently from ontogenetic influences and tree size, we
conceived planting tubes that were vertically adjustable on a metal pole, in which
seedlings of the same age class could be grown at different heights within and above
the vegetation layer. However, preliminary temperature measurements in the
planting tubes in the ground and suspended at 1 m height showed that in spite of
insulation, ventilation layers and reflective shields, soil temperatures were not
comparable and would thus bias the seedling response to the treatments.
Nevertheless, this testing provided valuable experience and data, which we applied
in Chapter 4 to contest a study (Fajardo and Piper 2014) using a similar design
without controlling neither seedling root-zone temperature nor soil moisture. In this
chapter, the consequently questionable interpretation of results in Fajardo and
Piper’s study is put in context with the potential impact of fluctuating temperatures
and moisture deficits, and alternative experimental setups for this after all elegant
approach are explored.
Finally, Chapter 5 synthesises the results and conclusions of the previous
chapters and gives an outlook to future research.
Chapter 1
18
Fig. 1.3 View from the Chaillol Mountain on the Lautaret Pass (2050 m a.s.l.),
showing the Alpine Research Station Joseph Fourier of the University Grenoble with
the associated Lautaret Alpine Botanical Garden (white arrow) and the experimental
garden (black arrow, 2100 m a.s.l.). On the left in the background, relatively
undisturbed treelines principally formed by Larix decidua and Pinus uncinata can be
seen at approximately the same elevation.
19
CHAPTER 2
IMPACTS OF SOIL MICROCLIMATE ON EARLY ESTABLISHMENT OF
TREES AT THE ALPINE TREELINE: IDIOSYNCRATIC RESPONSES AND
THE IMPORTANCE OF SOIL MOISTURE
Hannah Loranger, Gerhard Zotz, Maaike Y. Bader
Submitted
ABSTRACT
On a global scale, temperature is the main determinant of arctic and alpine treeline
position. However on a local scale, treeline form and position vary considerably due
to other climatic factors, tree species ecology and life-stage-dependent responses. For
treelines to advance poleward or uphill, the first steps are germination and seedling
establishment. These earliest life stages may be major bottlenecks for treeline tree
populations and will depend differently on climatic conditions than adult trees. We
Chapter 2
20
investigated the effect of soil temperature and moisture on germination and early
seedling survival in a field experiment in the French Alps near the local treeline (2100
m a.s.l.) using passive temperature manipulations and two watering regimes. Five
European treeline tree species were studied: Larix decidua, Picea abies, Pinus cembra,
Pinus uncinata and Sorbus aucuparia. In addition, we monitored the germination
response of three of these species to low temperatures under controlled conditions in
growth chambers. The early establishment of these trees at the alpine treeline was
limited either by temperature or by moisture, the sensitivity to one factor often
depending on the intensity of the other. The results showed that the relative
importance of the two factors and the direction of the effects are highly species-
specific, while both factors tend to have consistent effects on both germination and
early seedling survival within each species. We show that temperature and water
availability are both important contributors to establishment patterns of treeline trees
and hence to species-specific forms and positions of alpine treelines. The observed
idiosyncratic species responses highlight the need for studies including several
species and life-stages to create predictive power concerning future treeline
dynamics.
Keywords: alpine treelines, climate change, early seedling survival, germination,
temperature-moisture-interactions, time-to-event-analysis
Effects of Microclimate
21
INTRODUCTION
Treelines are conspicuous transition zones between two very different vegetation
types. There is a growing concern about how global climate change may affect these
systems, and as a consequence much attention has been drawn to both alpine and
arctic ecotones in recent years. Treelines could represent a distinct indicator of
climate warming since temperature is recognized as the main determinant of treeline
position on a global scale, roughly following a common isotherm of 5-7 °C mean
growing season temperature (Körner and Paulsen 2004). Many studies show recent
advances of treelines poleward and to higher elevations, as well as increasing radial
growth of the trees forming these ecotones (Rolland et al. 1998; Kullman 2007;
Shiyatov et al. 2007; Qi et al. 2015). However, stable or receding treelines have been
found (Harsch et al. 2009), and treeline position may vary considerably at a local
scale (Holtmeier and Broll 2005; Case and Duncan 2014). Such local variations can be
due to locally varying environmental conditions unrelated to temperature such as
precipitation (Holtmeier and Broll 2005), tree species ecology (Körner and Paulsen
2004) and life-stage dependent environmental dependencies (Barbeito et al. 2012;
Greenwood et al. 2015).
These abiotic and biotic factors may also interact with temperature to
determine the form and dynamics of a treeline at a specific site. For example, the
consequences of moisture deficits – which can be positively linked to climate
warming – have been shown to override positive temperature responses with respect
to growth (Barber et al. 2000; González de Andrés et al. 2015) and regeneration
(Barton 1993; Daniels and Veblen 2004; Moyes et al. 2013). In such cases, treeline
Chapter 2
22
shifts may depend more on the interactions of temperature and water availability
than on their absolute values (Ohse et al. 2012). As a result, it is commonly observed
that tree cover is slow or unable to expand to its ultimate thermal boundary
(Holtmeier 2009). The underlying mechanisms remain however difficult to
disentangle and there is an urgent need for quantitative assessments of the specific
environmental conditions and associated mechanisms preventing the establishment
of different tree species beyond current treelines.
Treelines represent distribution boundaries for an entire life-form – the tree.
Consequently, ecosystems above the treeline differ fundamentally from those below,
e.g. in regard to soils and microclimates (Sullivan and Sveinbjörnsson 2010; Thébault
et al. 2014). This presents particular challenges for a successful tree regeneration and
establishment in the treeline ecotone and beyond, as required for an upward
distributional shift. Previous studies have shown that traits essential for regeneration
such as the number of seed-bearing fruits or the number of viable seeds often
decrease with increasing elevation, thereby reducing the probability of seedling
establishment especially above treeline (Cuevas 2000; Kroiss and HilleRisLambers
2015). While seed production and dispersal are unequivocal prerequisites for tree
regeneration, subsequent germination and seedling establishment have also been
widely recognized as potential life-history bottleneck of treeline tree populations
(Stevens and Fox 1991; Germino et al. 2002; Smith et al. 2003; Johnson et al. 2011).
Germination represents the earliest, critical life-stage transition and should thus be
subject to strong natural selection (Baskin and Baskin 2001). Furthermore, the
conditions during germination also influence the phenotypic expression of post-
germination traits, thereby affecting later seedling performance (Donohue et al.
Effects of Microclimate
23
2010). Once successfully germinated, the germinant enters the most vulnerable life-
stage of a tree, characterized by the highest mortality of the whole life-cycle (Cui and
Smith 1991; Johnson et al. 2011). Most studies investigating environmental
dependencies of both early life-stages find that favourable conditions are concordant
(i.e. the same conditions are favourable for both, seed and seedling), though others
report conflicting requirements (reviewed in Schupp 1995). Hence, it remains unclear
to what extent the effects of environmental conditions on regeneration success are
life-stage specific.
The natural seedling distribution in treeline ecotones, a result of limitations to
both early life-stages, is often found to be related to stress-reducing site features such
as reduced sky exposure or shelter from strong winds (Germino and Smith 1999;
Smith et al. 2003; Batllori and Camarero 2009). Furthermore, seedling density often
decreases with elevation (Cuevas 2000). Both observations are in line with the view
that the lack of safe sites (sensu Harper 1977) and the harsh climatic conditions in the
alpine zone might restrict the regeneration of treeline trees (Tranquillini 1979). Most
research focusing on the earliest stages of tree regeneration at treeline sites used
germinating seeds principally to study subsequent young seedling survival and
physiology (Germino and Smith 1999; Germino et al. 2002; Moyes et al. 2013), or
lumped germination and subsequent seedling survival due to long observation
intervals (Zurbriggen et al. 2013). Others explicitly including germination responses
at finer temporal scales mainly used elevation gradients to study recruitment
responses, without actively manipulating microclimate (Ferrar et al. 1988; Castanha
et al. 2012). As the process of germination differs in genetic regulation and
environmental sensitivity from survival mechanisms in emerged seedlings and may
Chapter 2
24
thus be evolutionarily decoupled, seedling emergence needs to be monitored
frequently following individual seeds (and seedlings). Moreover, potentially
complex interactions of microclimatic variables and responses of early tree
establishment ask for experimental manipulations of more than one limiting factor.
To our knowledge, no study has ever addressed both of these aspects in a field
experiment on regeneration limitations at treelines.
To summarize, any attempt to understand current treeline patterns and
positions mechanistically and to predict their future dynamics requires investigating
local treelines with regard to microclimate-, species- and life-stage-specific responses.
In this study, within a single field experiment, we assessed the germination and early
seedling-establishment responses of five important European treeline tree species to
the variation of two important microclimatic variables, temperature and moisture.
Accordingly, we asked the following research questions: (a) Do responses of treeline
trees to microclimatic conditions vary with life stage, i.e. during germination and
early seedling establishment? (b) Do temperature and water availability interact to
determine germination and early seedling survival? (c) Do different treeline tree
species show consistent responses to temperature and moisture conditions? In
addition to the field experiment, we monitored the germination response to low
temperatures under controlled conditions in growth chambers for three of the five
study species. This allowed us to assess temperature responses along a defined
gradient and at a finer temporal scale, complementing the results from the more
complex field study.
Effects of Microclimate
25
METHODS
Effect of soil moisture and temperature under field conditions
Study site and species
A common garden germination experiment was set up in the experimental garden of
the alpine research station Joseph Fourier in the French Alps near the local treeline
(Lautaret Pass, 2100 m a.s.l., 45°02’N, 6°24’E). The site is situated in a climatic
transition zone between the wet outer Alps and the dry inner Alps (Ozenda 1988),
with 11°C as the mean temperature of the warmest month (July) and an average
annual precipitation of 1230 mm (Choler et al. 2001). The study species comprise four
important treeline-forming conifers of the European Alps: Larix decidua, Picea abies,
Pinus cembra, Pinus uncinata as well as the deciduous angiosperm Sorbus aucuparia,
which also occurs up to treeline elevation (Brändli 1998). Seeds of subalpine origin
from the inner Alps were obtained from a commercial seed producer (Herzog Baum,
Samen und Pflanzen GmbH, Gmunden, Austria) and a forestry office (Kantonaler
Forstgarten Rodels, Rodels, Switzerland), except for seeds of S. aucuparia, which were
available only from colline origin in Hungary (Table 2.1). Information on seed
germinability – either provided by the supplier or determined from standard
germination trials – was used to adjust the seed quantity sown per plot (Table 2.1).
Relatively large seed quantities were sown to account for a potentially lower
germination success under field conditions, allowing a reliable estimation of
germination proportions and ensuring a sufficient number of seedlings to monitor
subsequent survival. Due to time constraints, the seed quantity had to be reduced in
the third experimental block.
Chapter 2
26
Table 2.1. Seed characteristics and seed quantities used in the germination field
experiment
Species Elevation of
seed source
(m a.s.l.)
Germinability (%) Seed quantity (#)
Larix decidua 1800 – 2000 33 % a 240 (120)
Picea abies 1100 – 1400 63 % b 120 (60)
Pinus cembra 1300 - 2850 89 % b 30
Pinus uncinata 2100 78 % a 120 (60)
Sorbus aucuparia 400 – 1400 86 % b 120 (60)
a Germinability of seed lot determined by own standard germination trial (winter 2012/2013)
b Germinability of seed lot provided by seed supplier
Seeds originated always from the inner Alps, except for Sorbus aucuparia, which was only available
from Hungary. Numbers in brackets indicate reduced seed quantity sown in the third experimental
block. For P. cembra seed quantity was always limited to 30 seeds per row due to the large seed size of
1-1.5 cm.
Experimental design
Fifteen experimental plots (70 x 30 cm) were arranged in three blocks to account for
spatial heterogeneity, with approximately 5 m distance between the centers of two
adjacent blocks and 20 cm distance between plot edges. All blocks were enclosed by a
60 cm–high wire-mesh fence as protection against rodents. The vegetation cover on
the plot surface was removed and plots were excavated to a depth of 15 cm to
remove rocks and large roots from the soil. The soil of plots from the same block was
then mixed and returned to the plots. This procedure was done both to create a
homogenous growth substrate within blocks and to remove the effects of biotic
interactions such as competition or facilitation by neighboring vegetation, allowing
to focus on abiotic factors. In October 2013, seeds were sown in one row of 60 cm
Effects of Microclimate
27
length per species, allowing 3 cm spacing between rows and 5 cm plot margin. Rows
were randomly assigned to one of the five species. Seeds were sown in 2 cm deep
grooves, distributing seeds evenly with the fingertips and closing up the soil. Seeds
of P. cembra were limited to 30 seeds per row and placed individually due to their
large size.
In spring of 2014, two watering regimes and two types of installations for
passive temperature manipulation, open-top chambers (OTCs; passive warming) and
shade roofs (passive cooling), were set up to create a gradient of soil temperature and
soil moisture across all experimental plots (Fig.2.1A). OTCs were conceived as
hexagons (Marion et al. 1997; r = 80 cm, h = 30 cm) with 3 mm thick acrylic glass
panels transmitting 92 % of solar radiation, including UV. Shade roofs consisted of a
plot-sized wooden frame covered with a shade net, providing 70 % shade on the plot
surface but allowing rain water to pass. The roofs were supported by four 30 cm high
metal poles at the plot corners with 20 cm shade net curtains on each side to prevent
the penetration of low-angle sunshine. Control and warming treatments were
crossed with a watering treatment, with watered plots receiving 3 mm irrigation on
days without rainfall throughout the study period (in total adding up to 35 % of the
May-September precipitation in 2014). Since cooling through shading was already
expected to decrease evapotranspiration and thus increase soil moisture, this
treatment was not included in the additional-watering regime. All five microclimate
treatments (control (C), watered control (C+W), passive warming (OTC), watered
passive warming (OTC+W), passive cooling (Sh)) were replicated in each of the three
experimental blocks. Treatments were initiated directly after snowmelt in a block-
wise manner due to a highly heterogeneous snow cover, with about four weeks
Chapter 2
28
between the start in the first block (mid-April) and the last block (mid-May). The
study period covered the complete growing season of 2014 from snowmelt to early
September.
Fig. 2.1 A. Experimental set-up to manipulate soil temperature and soil moisture by
using shade roofs (on the left, passive cooling) and open-top chambers (on the right,
passive warming) crossed with a watering treatment. B. Individual monitoring of
emerging seedlings by using colored pins (yellow: Larix decidua, green: Pinus cembra,
white: Sorbus aucuparia, red: Pinus uncinata, blue: Picea abies, black: dead seedling).
Microclimate
The soil moisture content (%) was measured monthly with a hand-held sensor
inserted 15 cm in each plot center (TRIME-PICO64, IMKO Micromodultechnik,
Ettlingen, Germany), while the soil temperature was measured at 5 cm depth with
external sensors of permanently installed temperature loggers (Hobo ProV2, Onset
Corp, Bourne, MA, USA). Since comparative measurements in the same
microclimatic treatments in 2013 had shown that there is no significant temperature
difference between watered plots and their control (C vs. W: p = 0.43; OTC vs.
OTC+W: p = 0.76; two-sample t-test, n = 5), soil temperature was only recorded in
the three temperature-relevant treatments C(+W), OTC(+W) and Sh. In each block, a
temperature logger was assigned by chance to either a watered plot or its control and
then inserted in the plot center, recording data in 30 min intervals.
A B
Effects of Microclimate
29
Integrated variables were calculated to obtain a quantitative gradient of both
soil moisture and soil temperature across all plots. Mean soil moisture content (soil
MC; %) was calculated as the seasonal average of four monthly measurements for all
plots, giving a soil MC gradient with 15 observation points. Soil heat accumulation
relevant for germination and seedling survival was expressed in growing degree
days with a base temperature (Tb) of 2 °C, which is the lower temperature limit for
germination of at least two of the study species (Løken 1959; Barclay and Crawford
1984). In the plots where temperature was recorded, the number of soil growing
degree days (soil GDD; #) was calculated by summing up the positive differences
between temperature recordings and the base temperature over the whole study
period and dividing the results by the measurement interval fraction of a day (30 min
/ 24 h = 48, Eq. 2.1), giving a soil GDD gradient with nine observation points.
/
( )
. 2.148
i b
i b
i T T
T T
Eq GDD
for temperature recordings (Ti) higher than the base
temperature (Tb).
Seedling survey
Emerging seedlings were recorded weekly and each marked with a colored pin to
allow the assessment of individual survival (Fig. 2.1B). Since hypogeous germination
could not be monitored directly, seedling emergence was used as a proxy for total
germination success by calculating the proportion of sown seeds that emerged as
seedlings (including subsequently dead individuals). Seedling survival was
calculated as the proportion of emerged seedlings that survived until the end of the
growing season.
Chapter 2
30
Effect of temperature under controlled conditions
Complementing the field experiment, a standard germination trial investigating the
effect of low temperatures on germination was performed for three of the five
studied species under controlled conditions in growth chambers (Economic Delux
Snijders Scientific, Thermotec, Weilburg, Germany) in winter 2014 / 2015. Batches of
25 seeds of P. uncinata and P. abies and 60 of L. decidua were placed on moist paper
tissue in sealable plastic boxes (volume = 280 ml) with six replicates for every
temperature treatment. Four low-temperature treatments comprised constant
regimes of 16 °C, 12 °C, 8 °C and 4 °C and all treatments included a 12 h / 12 h light-
dark-cycle. As control treatment we used the settings 20 / 15 °C 12h / 12h,
previously identified as optimal for the same seed lots (germination percentages: P.
uncinata: 79 %, P. abies: 63 %, L. decidua: 33 %). Germination boxes were rotated on
their tray every other day to assure homogeneous temperature exposure.
Successful germination was recorded for every single seed in two- to three-
day intervals as 1 cm growth of the radicle. Germinated seedlings were removed.
The temperature treatments were discontinued when a species showed no further
germination for two weeks. The remaining seeds of the two warmest treatments
(control, 16 °C) were all non-viable, their soft texture and liquid discharge indicating
decay of the embryo, unambiguously indicating maximum germination. Seeds in the
three cooler treatments (12 °C, 8 °C, 4 °C) still showed a very slow increase in
germination after eleven weeks so that remaining, healthy-looking seeds were
transferred to the control temperature for another three weeks to test their viability.
Effects of Microclimate
31
Statistical analysis
All analyses were performed using R 3.2.1 (R Core Team 2015). Germination and
survival in the field experiment at the end of the growing season 2014 were
expressed as a two-column vector of counts of successes and failures per species per
plots and analysed with binomial generalized linear models (GLM), including block,
soil MC and soil GDD and the interaction soil MC : soil GDD as explanatory
variables. Using these continuous gradients as explanatory variables instead of the
treatments allowed differentiating the relative effects of soil temperature and
moisture as well as detecting potential interactions between them. In cases of
overdispersion, the standard errors were corrected by using a quasi-GLM model
(Zuur et al. 2009). Non-significant terms were removed from the full models in a
backwards stepwise approach. To facilitate interpretation, significant interactions
between soil MC and GDD are shown graphically by plotting the predicted values
from the model along the whole range of one of these two variables (on the x-axis)
and for three fixed values of the other variable: low (25 % quartile), intermediate
(median) and high (75 % quartile). The variable chosen to represent the x-axis was for
each final model the one with the lower p-value. Note that the resulting curves are
predictions from the models so that they do not directly relate to specific data points;
sections of variables which were not measured are extrapolations of the models.
Since temperature and moisture extremes were also linked to reduced light
intensities by the shading roofs, we evaluated the potentially confounding effect of
light by performing an additional analysis excluding the three shaded plots.
Germination data of the growth chamber experiment were analyzed using
time-to-event analysis (McNair et al. 2012) by first assessing the random variation
Chapter 2
32
among replicates with a Cox proportional-hazards model including a frailty term.
There was no evidence of variability in frailty levels for any of the three species in
any temperature treatment so that data from the six replicates could be pooled. Non-
parametric time-to-event analysis was then used to compare temperature treatment
differences in the germination pattern of each species with a log-rank test using the
survdiff – function in R (survival library; Therneau 2015). Results give test statistics
and significance levels of group (temperature treatment) pairwise comparisons for
each species. The p-values were Holm-adjusted to account for family-wise error rates
in multiple comparisons. Treatment differences of survivor functions were
graphically displayed by showing the inverse Life-table estimates of survivor
functions with point-wise 95 % confidence intervals computed with the R-function
lifetab (KMsurv library; Klein & Moeschberger 2012).
RESULTS
Field experiment: soil moisture and temperature effects on early regeneration
Soil microclimate
The soil moisture content (soil MC) gradient ranged from 20 % to 35.5 %, with the
cooling treatment being the wettest (mean = 33.4 % ± 1.9 SD, n = 3) and the non-
watered control the driest (mean = 25.8 % ± 5.3 SD, n = 3) (Fig. 2.2), with consistent
relative differences between treatments. As expected, heat accumulation of the soil
was highest in the warming treatment (mean = 1313 GDD ± 101 SD, n = 3) and lowest
in the cooling treatment (mean = 769 GDD ± 46 SD, n = 3). The complete gradient
over the nine measured plots ranged from 733 to 1421 GDD (Fig. 2.2).
Effects of Microclimate
33
Fig. 2.2 Microclimatic conditions in the field experiment with manipulations of soil
moisture (watering, shading) and soil temperature (passive warming and cooling
treatments). Shown are the integrated microclimate variables mean soil moisture
content (soil MC; %) and total number of soil growing degree days > 2 °C (soil GDD;
#) per treatment (A, C; n = 3) and the respective gradient of soil moisture and soil
temperature over all 15 experimental plots (B, D; sorted by y-axis value, so order
differs for soil MC and soil GDD). Treatment abbreviations indicate: C = Control, W
= watered control, OTC = passive warming (open top chamber), OTC+W = watered
passive warming, Sh = passive cooling (shading roof).
Seedling emergence and survival
Maximum seedling emergence (%) at the end of the growing season under field
conditions was invariably lower than germination under optimum conditions in a
standard germination trial (L. decidua: 25 %, P. abies: 40 %, P. cembra: 63 %, P.
uncinata: 59 %, S. aucuparia: 23 %; see Appendix Fig. S2.1). Overall, seedling survival
at the end of the growing season exceeded an average of 50 % for all species, but
differed considerably among species (L. decidua: 53 %, P. abies: 65 %, P. cembra: 94 %,
P. uncinata: 72 %, S. aucuparia: 67 %; see Appendix Fig. S2.1).
The responses of seedling emergence and first-season survival in the five
study species to soil moisture and soil temperature were highly idiosyncratic. While
So
ilM
C (
%)
So
ilG
DD
(#)
Treatments Experimental plots
A B
DC
Chapter 2
34
higher soil moisture had a positive effect on both stages of early establishment in L.
decidua, it had a generally negative effect on P. cembra (Fig. 2.3, Table 2.2). Similarly,
higher soil temperature generally positively affected seedling emergence in P.
uncinata while having a negative effect on both stages of early establishment in S.
aucuparia (Fig. 2.3, Table 2.2). Within species, however, the principally affecting
climate variable and the direction of its effect were generally consistent for seedling
emergence and survival. There were significant interactions between soil
temperature and moisture in i) the emergence of P. cembra, P. uncinata and S.
aucuparia, as well as ii) the survival of P. abies (Fig. 2.3, Table 2.2): the negative effect
of high soil moisture was reduced (Fig. 2.3B, E) or even inversed (Fig. 2.3D, H) as
temperature increased. Conversely, the negative effects of high temperature were
reduced (Fig. 2.3B, E) or reversed (Fig. 2.3D, H) as the soil moisture content
increased. Finally, a block effect in seedling emergence of S. aucuparia indicated that
emergence was significantly higher in the block with earlier snowmelt (Fig. 2.3E).
An additional analysis excluding the shaded plots showed that effects found
when including all plots were generally maintained even on this shortened
temperature and moisture gradient, with one exception: the significant negative
temperature effects on emergence and survival in S. aucuparia disappeared (see
Appendix Table S2.1).
Effects of Microclimate
35
Fig. 2.3 Seedling emergence as proportion of sown seeds (A-E) and survival as proportion of emerged seedlings (F-J) for the five study
species in response to soil moisture (soil MC; %), soil temperature (soil GDD, #) or the interaction of both. Shown are binomial GLM
for significant responses, non-significant responses are displayed as open circles for observed values. Significant interactions are
shown using fixed values of soil GDD (soil MC) plotted along the complete range of soil MC (soil GDD) for P. cembra (P. uncinata, P.
abies, S. aucuparia), line types indicating: dotted = low intensity, dashed = intermediate intensity, solid = high intensity. Note that the
resulting curves are predictions from the models so that they do not directly relate to specific data points. The significant block effect
in seedling emergence of S. aucuparia is displayed by varying hues of grey: black = early snowmelt date (block 1), medium grey =
intermediate snowmelt date (block 2), light grey = late snowmelt date (block 3).
Em
erg
en
ce
Su
rviv
al
Larix decidua Pinus cembra Picea abies Pinus uncinata Sorbus aucuparia
Soil moisture content (%) Soil growing degree days (#)
A B C D E
F
G H I J
Chapter 2
36
Table 2.2 Summary of binomial GLM testing the effect of soil microclimate (soil
moisture content, soil growing degree days) on the proportions of germination and
subsequent survival at the end of the growing season 2014 of five treeline tree species
grown in a field experiment in the French Alps at 2100 m.
Rows give complete models with χ2- or F-values (for GLM and quasi-GLM, respectively) for
germination and survival data at the end of the growing season 2014 with the explanatory variables
block, soil moisture content (Soil MC), soil growing degree days > 2 °C (Soil GDD) and their
interaction (Soil MC : Soil GDD) in the order tested in the model; non-significant variables (given in
parenthesis) were removed from the models based on G- or F-tests, respectively, in a stepwise process
with superscripts indicating the order in which they were removed. The minimum adequate model is
given in bold and significance levels are indicated as: . p<0.1,* p<0.05, ** p<0.01, *** p<0.001. Arrows
indicate whether partial slopes are positive or negative.
Growth chamber experiment: germination response to low temperatures
Germination of the three species was significantly reduced by decreasing
temperatures, but with species-specific differences. In L. decidua the results of the
survivor functions differed significantly mainly between the three highest
temperature treatments, showing a 10 % decrease of germination probability per
treatment (Fig. 2.4A, Table 2.3). Contrasting, in P. abies, the results of the survivor
functions principally differed at the lower end of the temperature gradient (Fig. 2.4B,
Table 2.3), where differences mainly arose from an increasing delay in the onset of
Block Soil MC Soil GDD Soil MC : Soil
GDD
Larix decidua
Germination (F2,9 < 0.01)1 F1,13 = 8.18 * ↑ (F1,12 = 0.31)3 (F1,11 = 0.25)2
Survival (F2,9 = 0.96)1 F1,13 = 7.25 * ↑ (F1,12 = 1.28)3 (F1,11 = 0.27)2
Pinus cembra
Germination (χ2 2 = 4.49)1 χ2
1 = 5.89 * ↓ χ2 1 < 0.01 χ2
1 = 4.91 * ↑
Survival (χ2 2 = 3.01)2 χ2
1 = 10.83 *** ↓ ( χ2 1 = 1.14)3 ( χ2
1 = 0.52)1
Sorbus aucuparia
Germination χ2 2 = 8.1 * χ2
1 = 1.32 χ2 1 = 36.02 *** ↓ χ2
1 = 6.99 ** ↑
Survival (χ2 2 = 1.38)1 (χ2
1 = 0.06)3 χ2 1 = 8.85 ** ↓ ( χ2
1 = 0.36)2
Pinus uncinata
Germination (F2,9 = 0.1)1 F1,11 = 6.84 * ↑ F1,11 = 21.62 *** ↑ F1,11 = 5.11 * ↑
Survival (F2,10 = 0.5)2 (F1,13 = 1.86)4 (F1,1 2= 2.46)3 (F1,9=0.08)1
Picea abies
Germination (F2,12 = 1)4 (F1,10 = 0.53)2 (F1,11 = 1.03)3 (F1,9= 0.08)1
Survival (χ2 2 = 0.85)1 χ2
1= 0.16 χ2 1= 0.55 χ2
1 = 5.75 * ↑
Effects of Microclimate
37
germination. Only for the 4 °C treatment the probability of having germinated was
significantly lower (~ 30 % lower) at the end of the experiment than in all other
treatments, though its slope was still positive, potentially indicating a further
increase a longer time span (Fig. 2.4B). In P. uncinata, the most important decrease (~
30 %) in the probability of having germinated occurred at intermediate temperatures,
as was shown by highly significantly different results of survivor functions between
the 12 °C and the 16 °C treatments, while at the high and low end of the gradient,
results were statistically indistinguishable (Fig. 2.4C, Table 2.3).
Viability tests for seeds that did not germinate after eleven weeks in the lower
temperature treatments (12 °C, 8 °C, 4 °C) showed that seed viability was generally
not reduced. In almost all cases, a similar germination success as in the control
treatment was achieved after three additional weeks under control conditions (Fig.
2.4). Only the seeds of L. decidua coming from the 12 °C treatment showed a
substantial (~ 10 %) reduction in their germination success (Fig. 2.4A).
Chapter 2
38
Fig. 2.4 Inverse Life-table estimates of the survivor functions of germination data
from a growth chamber experiment representing the probability of having
germinated at four low temperature treatments and a control (22 / 15 °C, 12h / 12h)
over time, including seeds of L. decidua (A), P. abies (B) and P. uncinata (C). Seeds
were pooled over replicates (n = 6) giving a total of 360 (L. decidua) or 150 (P. abies, P.
uncinata) seeds, respectively. Grey lines represent point-wise 95% confidence
intervals for each treatment. Maximum germination in the control and 16°C-
treatments was achieved after 42 days, with all remaining seeds being non-viable.
After eleven weeks, remaining healthy-looking seeds in the three lower temperature
treatments (12 °C, 8 °C and 4 °C) were transferred to control conditions to test their
viability. Respective treatment symbols at day 100 give mean ± SD of final
germination success after three additional weeks of control conditions.
A Larix decidua
B Picea abies
C Pinus uncinataPro
bab
ilit
y o
fh
avin
gg
erm
ina
ted
Time (days)
Effects of Microclimate
39
Table 2.3 Summary of log-rank tests comparing the survivor functions of seeds of
three treeline tree species germinated under controlled conditions in growth
chambers in four permanently low temperature treatments and a control (15 / 22 °C
12h /12h).
Data were pooled over replicates giving a total of 360 (150) seeds for L. decidua (P. abies, P. uncinata),
respectively. Rows show results of group (temperature treatment) pairwise comparisons for each
tested species presenting χ2-values and Holm-adjusted p-values; significant differences are given in
bold, marginally significant differences are given in italics; significance levels are indicated as: .
p<0.1,* p<0.05, ** p<0.01, *** p<0.001.
DISCUSSION
Our results show that the early establishment of the focal treeline tree species is
affected by temperature and water availability in a very idiosyncratic manner.
However, the importance of both climate factors and the direction of their effect on
germination and survival tended to be consistent over both stages of early
establishment within each species. Interactions of both climate variables indicated
that the sensitivity to one factor often depends on the intensity of the other.
Consistency of limiting factors during early establishment
The consistent effect of microclimate over the life-stage transition from germination
to first-year seedling survival (Fig. 2.3) is in accordance with previous studies (Ferrar
Treatment Larix decidua Picea abies Pinus uncinata
comparisons χ2 1 p χ2 1 p χ2 1 p
4 °C vs. 8 °C 0.1 1 61.9 <0.001 *** 7.4 0.17
4 °C vs. 12 °C 15.5 0.08 . 71.8 <0.001 *** 20.8 0.01 *
4 °C vs. 16 °C 0.8 1 89.2 <0.001 *** 125.0 <0.001 ***
4 °C vs. Control 32.1 <0.01 ** 64.3 <0.001 *** 195.0 <0.001 ***
8 °C vs. 12 °C 11.4 0.15 11.3 0.15 5.2 0.3
8 °C vs. 16 °C 0.14 1 28.9 <0.01 ** 100.0 <0.001 ***
8 °C vs. Control 38.8 <0.001 *** 11.2 0.15 177.0 <0.001 ***
12 °C vs. 16 °C 21.0 <0.001 *** 14.0 0.16 71.7 <0.001 ***
12 °C vs. Control 80.5 <0.001 *** 3.6 0.42 147.0 <0.001 ***
16 °C vs. Control 22.4 0.01 * 1.1 1 17.3 0.07 .
Chapter 2
40
et al. 1988; Castanha et al. 2012). This is of particular importance since limitations of a
populations’ distribution are primarily imposed during these most critical life-stages
(Grubb 1977; Harper 1977). In this context, a high level of consistency over two
critical early life-stages will reduce regeneration restrictions arising from seed-
seedling conflicts. On the other hand it should increase the impact of relatively stable
limiting environmental factors, which could be particularly restricting for
regeneration in the harsh conditions of a species’ distribution range edge. In contrast,
a variable factor such as irregular freezing events during the growing season can be
temporarily decoupled from a short, susceptible life-stage such as germination, but is
more likely to affect the longer subsequent stage of the young seedling (Shen et al.
2014). Hence, the degree of concordance or conflict in the environmental
requirements between seed and seedling can have a direct impact on the quantity
and the distribution of recruits (Schupp 1995).
Yet for two species, P. abies and P. uncinata, only one of the two studied life-
stages showed a significant response (Fig. 2.3). This might indicate a change in their
susceptibility to the two microclimatic factors during early establishment, which is
supported by a capability of germinating under a large range of conditions for P.
abies (Løken 1959 and Fig. 2.4B) and a relatively resistant seedling stage in P. uncinata
(Batllori et al. 2010).
Temperature, moisture, and their interactions driving early establishment success
The results from our growth chamber experiment, where decreasing temperatures
invariably reduced germination (Fig. 2.4), are in line with previous studies
demonstrating the importance of temperature for treelines in general (Rolland et al.
Effects of Microclimate
41
1998; Körner and Paulsen 2004) and for early regeneration stages in particular
(Germino and Smith 1999; Smith et al. 2003). However, germination success of L.
decidua and P. abies was still considerable at low temperatures (Fig. 2.4A-B). This
concords with the results of our field experiment, where seedling emergence of
neither species was affected by temperature. Inversely, P. uncinata was particularly
temperature-sensitive in the growth chambers and showed a positive temperature
response of seedling emergence in the field, confirming the consistency between both
experiments.
Our field experiment further revealed that almost all seedling emergence and
survival responses were sensitive to water availability, though these responses often
showed an interaction with temperature. In L. decidua moisture was even the only
significant variable, implying that depending on the species, temperature may play
rather a subordinate role in limiting regeneration. Our findings thus add to a
growing body of evidence that other factors than temperature alone, e.g. water
availability, determine seedling distributions at alpine treelines (Ferrar et al. 1988;
Sullivan and Sveinbjörnsson 2010; Greenwood et al. 2015; Kroiss and
HilleRisLambers 2015; Moyes et al. 2015). While germination often requires relatively
high moisture conditions as environmental cue and to set the necessary physiological
processes in motion (Baskin and Baskin 2001), young seedlings depend on it for a
longer period of time due to their generally shallow, simple rooting system and their
large, transpiring surface area relative to the low water storing capacity (Johnson et
al. 2011). Both early life-stages are thus much more affected by water shortage than
established trees, potentially creating a bottleneck for regeneration in treeline and
Chapter 2
42
alpine tundra ecotones, which often exhibit highly variable water holding capacities
(Holtmeier and Broll 2005).
An important result of this study is that the effects of both temperature and
moisture availability on early establishment cannot be decoupled from one another.
Especially the combination of opposite extremes, e.g. high temperatures at low soil
moisture or low temperatures at high soil moisture had limiting effects on both early
life-stages of the study species. Both combinations have previously been shown to
restrict tree development, either by temperature-induced moisture stress (Barber et
al. 2000; Lloyd and Bunn 2007) or cold soil conditions and insufficient aeration
limiting root zone activity (LeBarron 1945; Islam and Macdonald 2004).
Species-specific responses
The observed early-establishment responses to the abiotic environment of the five
tree species were highly idiosyncratic. The temperature response of germination in
the growth chambers revealed a specific pattern for each species (Fig. 2.4, Table 2.3),
possibly indicating an adaption to different ranges of germination temperatures.
These tendencies were confirmed in the field experiment, which further showed
contrasting responses to soil moisture and temperature among all studied species.
Consequently, explaining and understanding observed patterns in the regeneration
limitations of local treeline tree populations requires the consideration of tree species
identity (Wardle 1985; Ball et al. 1991; Sullivan and Sveinbjörnsson 2010; Dufour-
Tremblay et al. 2012) and a detailed connection to the ecology of each individual
species. Hence, in the following we present a tailored, species by species
interpretation of the results.
Effects of Microclimate
43
The results of Larix decidua match the ecological features of a typical subalpine,
high-elevation tree species with a high tolerance to cold conditions (see Rameau et al.
1993; Brändli 1998). This was reflected in the relatively high success of germination
down to 4 °C in the growth chambers and the absence of a temperature response for
both early life-stages in the field. Soil moisture, on the other hand, positively affected
seedling emergence and survival (Fig. 2.3A, F), which can be related to the increased
water demand and low water use efficiency of the deciduous life-form compared to
evergreen conifers (Matyssek 1986). This feature seems to be already inherent to the
earliest stages of regeneration, even though first-year seedlings are not deciduous
yet.
In contrast, soil moisture had a negative effect on both early life-stages of
Pinus cembra. This negative effect was for seedling survival but was reduced by
increasing temperatures for seedling emergence. While this appears surprising at
first, it can be explained by a combination of limiting biotic factors and the life-
history strategy of this species. First, P. cembra is the highest-occurring tree species in
Europe and mostly occurs on steep sloping terrain where moisture limitations are
most severe (Brändli 1998). This is not only due to its higher tolerance to the harsh
subalpine conditions, but also caused by its low competitive capacities in relation to
other high-elevation tree species (Ulber et al. 2004). Second, seed dispersal relies on
the nutcracker (Nucifraga caryocatactes), a bird hiding seeds specifically in shallow
caches in open, wind-exposed (Kajimoto et al. 1998) and early snow-free (Mattes
1994) sites. Both of these aspects suggest that P. cembra is well adapted to rather dry
regeneration sites, as corroborated by impressively deep tap roots already present in
small seedlings (Hättenschwiler & Körner 1995, up to 20 cm in 1-yr old plants, i.e.
Chapter 2
44
nearly 4-fold the aboveground plant height, personal observation). Such a rooting
system might, however, be disadvantageous as soil moisture increases since deeper
roots aggravate problems associated with insufficient aeration and cool soil
temperatures (Scott et al. 1987). Furthermore, seedlings are highly vulnerable to
snow fungi promoted by prolonged snow cover (Senn 1999) and in a previous
germination trial (data not shown) we observed high seed mortality due to fouling.
Both findings indicate a general susceptibility of early life-stages to pathogens under
high moisture conditions.
The contrasting responses of early establishment in the other two conifers,
Picea abies and Pinus uncinata, can be directly related to their respective distribution
range. In growth chambers and in the field, the germination response of P. abies was
not or only weakly affected by colder conditions (Figs 2.3C, 2.4B), which is supported
by previous studies reporting germination responses temperatures as low as 2 °C
(Løken 1959). Seedling survival, however, responded to an interaction of
temperature and moisture, with increasing soil moisture compensating a negative
effect of high temperatures (Fig. 2.3H). These findings are in line with the ecological
requirements of this boreal-subalpine tree species, tolerating a wide amplitude of
environmental conditions except drought stress, which is reflected in its absence
from the south side or continental ranges of the European Alps (Rameau et al. 1993;
Brändli 1998).
Pinus uncinata, on the other hand, is a heliophile subalpine tree species with a
southern distribution (Pyrenees, southern European Alps, Rameau et al. 1993) and
accordingly its germination was strongly limited by colder temperature (Fig. 2.4C)
Effects of Microclimate
45
and responded positively to higher temperature under sufficient soil moisture
conditions (Fig. 2.3D). Seedling survival was not affected by the manipulated
microclimatic gradients, which is in line with the high tolerance of P. uncinata to
drought and a relatively robust seedling stage (Rameau et al. 1993; Batllori and
Camarero 2009).
Finally, the only broad-leaved and distributionally ubiquitous species Sorbus
aucuparia displayed the counterintuitive response of both early establishment stages
being negatively affected by increasing temperatures. This effect may partly be
explained by a limitation of our study design, in which the plots with shading roofs
(passive cooling) had the coolest temperatures but also an important change in light
conditions. In an additional analysis removing these plots, we showed that the
negative temperature effects on the performance of S. aucuparia disappeared
(Appendix Table S2.1; importantly, removing these plots did not change the general
effects in the conifer species), suggesting that those negative effects were actually an
artefact caused by an increased performance under shaded conditions. This is
supported by the literature, reporting evidence for shade-tolerant seedlings in S.
aucuparia (Raspé et al. 2000; Zywiec and Ledwoń 2008). However, in the case of
seedling emergence, a significant block effect (Table 2.2) indicated a true temperature
effect, i.e. higher emergence with earlier snowmelt, which means colder
temperatures during germination. Furthermore, there is a trend towards higher
seedling emergence in the low-temperature plots even when considering only the
reduced gradient (C and C+W, see Appendix Fig. S2.1). And finally, a true
temperature response for seedling emergence is supported by the germination ability
of S. aucuparia at temperatures as low as 2 °C (Barclay and Crawford 1984) and the
Chapter 2
46
previously found relationship of increasing temperatures reducing germination in an
alpine soil seed bank (Hoyle et al. 2013). Sorbus aucuparia, which possesses traits of
both pioneer and climax species (Zywiec et al. 2013), might benefit from increased
germination at low temperatures in two ways: First, low temperatures could act as an
additional germination cue to increase germination under conditions suitable for
seedling establishment (i.e. shade) and second, it might favour an early germination
time to avoid competing with faster growing species.
Implications for local treeline patterns and dynamics
The regeneration responses found in our study may offer an explanation for
observed patterns and dynamics of treeline tree populations, although such
observations are surprisingly difficult to find in the current literature. For example,
the re-invasion of abandoned subalpine pastures by trees was shown to be restricted
to colluvial soils alongside forest edges for L. dedicua, while being concentrated in
convex relief forms for P. cembra (Didier 2001). According to our results, this may
well be due to the respective early-establishment soil-moisture requirements of these
species (Fig. 2.3A-B, F-G). As another example, water shortage appears not, at least
not yet, to be an issue for recent dynamics of European alpine treelines, since even at
the southern treelines of P. uncinata seedling recruitment is common under current
climatic conditions (Batllori et al. 2010). This species, however, is also known to be
particularly drought resistant (Rameau et al. 1993), while, in line with our results,
boreal spruce forests have already been shown to suffer increasingly from
temperature-induced drought stress (Barber et al. 2000). Accordingly, we found an
important interaction between temperature and soil moisture for seedling survival of
P. abies (Fig. 2.3H). Hence, depending on local changes in precipitation, growth and
Effects of Microclimate
47
recruitment of high-elevation populations of this species could become restricted by
a warming climate even though they were, until recently, positively affected by it
(Bolli et al. 2007) and thereby affect the responsiveness of species-specific treelines to
increasing temperatures. However, note that a direct relation to local treeline features
will remain difficult, because many treelines are subject to intense anthropogenic
influences and are currently not at their climatic limit (Wick and Tinner 1997). Land
use can thus be a primary driver of their spatial pattern and recent dynamics, in
particular in the European Alps (Didier 2001; Bolli et al. 2007; Vittoz et al. 2008).
Therefore, in addition to climatic factors, land-use history needs to be taken into
account in observational studies of treeline dynamics.
Our results can also be linked to the important contribution that species-
specific requirements of the earliest life-stages exert on the shape and dynamic of a
local treeline (Harsch and Bader 2011). For example, if young seedlings require shade
or shelter – as did S. aucuparia in our study – they will be most successful near
existing trees and treeline tree populations will tend to occur in clustered spatial
patterns (Smith et al. 2003). Conversely, species requiring increased temperature or
light conditions – such as P. cembra and P. uncinata according to our results (Fig. 2.3)
– may perform better in open microsites and their treeline populations may develop
a scattered distribution (Holtmeier 2009). Consequently, abrupt treelines, if not
caused by disturbances, are primarily explained by high seedling mortality beyond
the forest edge and less so by growing season temperatures, which makes them less
responsive to the current climate change. In diffuse treelines in contrast, growth is
more and more limited by temperature with increasing elevation or latitude, and
Chapter 2
48
accordingly most treeline advances can be expected in this treeline type (Harsch and
Bader 2011).
CONCLUSIONS
Recruitment as a population bottleneck plays a crucial role in the discussion about
driving forces and future dynamics of treeline ecotones. To our knowledge this is the
first study to link the two earliest stages of tree establishment in a multi-species
approach experimentally manipulating two potentially limiting microclimatic
variables. We show that responses are highly idiosyncratic, but generally consistent
over both life-stages within each species, which increases the impact of limiting
climate variables in a relatively stable environment. Furthermore, interactions of
temperature and moisture highlight the complex interplay of microclimatic factors
influencing the regeneration success and confirm the importance of other factors than
temperature, such as water availability, for the understanding of treeline dynamics.
Our study contributes to the understanding of species-specific requirements and
limitations of the vulnerable stages of early establishment, which can be used to
explain current treeline patterns and predict future responses in the context of their
local climatic conditions.
ACKNOWLEDGEMENTS
We are very grateful to Serge Aubert, former director of the Lautaret Alpine
Botanical Garden and the affiliated research station Joseph Fourier, for enabling this
study by allowing us to use the experimental space and facilities of the station, and
thank the gardener team for support with any technical problem. We also thank
Effects of Microclimate
49
several field assistants, especially Carla Sardemann and Mathilde Vicente, for their
help in installing, maintaining and monitoring the experiment. The study was
funded by the German Research Foundation (DFG, BA 3843/5-1&2).
Chapter 2
50
APPENDIX
Fig. S2.1 Proportions of seedling emergence and subsequent survival (mean ± SD; n = 3) at the end of the first growing season for five
treeline tree species grown in a field experiment with manipulations of soil moisture and soil temperature in the French Alps at 2100
m a.s.l.. Treatment abbreviations indicate: C = Control, W = watered control, OTC = passive warming (open top chamber), OTC+W =
watered passive warming, Sh = passive cooling (shading roof). Different letters stand for, where present, significant differences. Note
that there are no error bars for emergence of Pinus cembra, OTC (C), because the number of seedlings was the same in all three
replicates.
Em
erg
en
ce
Su
rviv
al
Treatments
Larix decidua Pinus cembraPicea abies Pinus uncinata Sorbus aucuparia
A B C D E
F G
H
I J
a
b
abab
ab
aa
b
ab
ab
a
bb
abab
bb b
ab a
Effects of Microclimate
51
Table S2.1 Summary of binomial GLM testing the effect of soil microclimate (soil
moisture content, soil growing degree days) for reduced gradients (excluding plots
with shading roofs) on the proportions of seedling emergence and subsequent
survival at the end of the growing season 2014 of five treeline tree species grown in a
field experiment in the French Alps at 2100 m.
Rows give complete models with χ2- or F-values (for GLM and quasi-GLM, respectively) for
germination and survival data at the end of the growing season 2014 with the explanatory variables
block, soil moisture content (Soil MC), soil growing degree days (> 2 °C, Soil GDD) and their
interaction (Soil MC : Soil GDD) in the order tested in the model as indicated by superscripts; non-
significant variables (given in parenthesis) were removed from the models based on G- or F-tests,
respectively, in a stepwise process with superscripts indicating the order in which they were removed.
The minimum adequate model is given in bold and significance levels are indicated as: . p<0.1,*
p<0.05, ** p<0.01, *** p<0.001. Arrows indicate whether partial slopes are positive or negative.
Block Soil MC GDD Soil MC : GDD
Larix decidua
Germination (F2,6 = 0.01)1 F1,10 = 11.2 * ↑ (F1,9 = 0.26)3 (F1,8 = 0.42)2
Survival (F2,6 = 0.32)1 F1,10 = 9.46 * ↑ (F1,9 < 0.01)3 (F1,8 = 0.13)2
Pinus cembra
Germination χ2 2 = 6.75 * (χ2 1 = 2.65)3 (χ2 1 = 0.47)2 (χ2 1 = 2.66)1
Survival (χ2 2 = 0.42)2 χ2 1 = 6.47 * ↓ ( χ2 1 = 1.95)3 ( χ2 1 = 0.02)1
Sorbus aucuparia
Germination (F2,6 = 0.99)1 (F1,9 = 0.67)3 (F1,10 = 1.19)4 (F1,8 = 0.96)2
Survival (χ2 2 = 0.95)3 ( χ2 1 = 0.12)2 (χ2 1 = 0.23)4 ( χ2 1 = 0.1)1
Pinus uncinata
Germination (F2,6 = 0.58)1 F1,9 = 4.36 . ↑ F1,9 = 7.32 * ↑ (F1,9= 1.11)2
Survival (F2,7 = 1.29)2 F1,9 = 7.39 * ↑ F1,9 = 8.28 * ↑ (F1,6=0.03)1
Picea abies
Germination (F2,9 = 1.43)4 (F1,7 = 0.01)2 (F1,8 = 3.23)3 (F1,6 = 0.06)1
Survival (χ2 2 = 4.45)4 (χ2 1= 0)2 (χ2 1= 0.5)3 (χ2 1= 0.26)1
53
CHAPTER 3
COMPETITOR OR FACILITATOR? THE ROLE OF GRASSLAND
VEGETATION FOR GERMINATION AND SEEDLING PERFORMANCE OF
TREE SPECIES AT THE ALPINE TREELINE
Hannah Loranger, Gerhard Zotz, Maaike Y. Bader
Submitted
ABSTRACT
Alpine treelines constitute conspicuous vegetation boundaries in mountain
ecosystems that are expected to move upslope with a warming climate. However,
treeline responses are inconsistent and remain difficult to predict since many factors
unrelated to temperature, such as biotic interactions, can influence them on a local
scale. Especially during early regeneration stages, tree seedlings can be strongly
Chapter 3
54
influenced by alpine herbaceous vegetation through both competition and
facilitation. We aimed to understand the relative importance of these two types of
interactions, in dependence of vegetation structure, for treeline regeneration
dynamics. We studied the effect of herbaceous alpine vegetation on germination and
first-year seedling performance in a field experiment in the French Alps (2100 m
a.s.l.) with five important European treeline tree species: Larix decidua, Picea abies,
Pinus cembra, Pinus uncinata and Sorbus aucuparia. We focussed on how reserve
storage and allocation, studied via measurements of non-structural carbohydrates
and seedling biomass, are affected by varying vegetation cover and how this
interaction changes seasonally during the first year. The results show the dominance
of negative vegetation impacts, including general competition effects as well as tree-
species-specific susceptibilities to combinations of competition and indirect
vegetation effects via microclimate or pathogens. However, evergreen tree seedlings
appear to benefit from protection by the senescent herbs in autumn, leading to
increased carbohydrate reserves at the end of the winter. Thus, the interaction with
herbaceous vegetation switches seasonally from competition to facilitation. It is
currently unclear whether this effect promotes long-term net facilitation for tree
seedlings or if competitive interactions with the herbaceous vegetation prevail,
highlighting the need of long-term studies evaluating the impact of biotic
interactions at alpine treelines. Since early regeneration determines whether treelines
remain stable or move upslope, our findings contribute to the understanding of
observed treeline patterns, dynamics and their local variation in the context of site-,
life-stage and species-specific processes. This has important implications for the
development of predictive models of treeline dynamics, in which these ‘local’ aspects
Effects of Herbaceous Vegetation
55
should be incorporated in addition to more global drivers like changes in
temperature.
Keywords: biotic interactions, early establishment, Larix decidua, non-structural
carbohydrates, Picea abies, Pinus uncinata, Pinus cembra, Sorbus aucuparia
Chapter 3
56
INTRODUCTION
The fast, ongoing changes in global climate have already caused important shifts in
species distributions, which are predicted to continue in the future (Parmesan and
Yohe 2003; Chen et al. 2011). Treeline ecotones form the most conspicuous vegetation
boundary in alpine and arctic ecosystems and are particularly interesting in the light
of distributional shifts, since an entire life-form – the tree – reaches a thermal limit
(Tranquillini 1979; Körner 1998). This global pattern led to the prediction of an
upslope, or in arctic systems poleward, migration of tree cover with a warming
climate (Grace 2002). However, observations of local treeline positions and recent
dynamics show contrasting patterns (Holtmeier and Broll 2005; Harsch et al. 2009;
Case and Duncan 2014). While climatic factors are important in driving species
responses to climate changes, they interact with non-climatic factors in a close and
complex manner (Sutherst et al. 2007; Brown and Vellend 2014). Disregarding non-
climatic factors when considering distributional shifts, in particular species-specific
responses and species interactions, can therefore lead to unreliable predictions (Davis
et al. 1998).
Biotic interactions belong to such non-climatic factors with the potential to
limit or extend a species’ ecological and geographic range (Wisz et al. 2013 and
references therein). They can thus be an important factor causing variation in local
treeline responses, as shown by the impact of soil biota (Hasselquist et al. 2005),
animals (via herbivory and disturbances, Cairns and Moen 2004) and neighbouring
plants (e.g. Hobbie et al. 1999; Akhalkatsi et al. 2006; Grau et al. 2012) on the
distribution of trees at their elevational or latitudinal range edge. Plant-plant
Effects of Herbaceous Vegetation
57
interactions might be particularly important for the successful establishment of
seedlings, with neighbouring vegetation either causing competition (Venn et al. 2009)
or facilitating seedling growth (Sullivan and Sveinbjörnsson 2010). Under stressful
conditions the importance of positive (facilitative) interactions tends to increase
relative to negative (competitive) interactions (the ‘stress gradient hypothesis’, SGH,
Bertness and Callaway 1994). While several studies provide support for the SGH in
alpine plant communities (Choler et al. 2001; Callaway et al. 2002; He et al. 2013 and
references therein), it has recently been argued that species-interactions can switch
back to competition at the extreme ends of a stress gradient (Michalet et al. 2014). For
the alpine treeline, the interaction between existing alpine vegetation and
establishing tree seedlings is likely a balance between positive and negative effects,
depending on the alpine vegetation type, the local climatic conditions, and the
tolerance limits of the tree species involved.
Herbaceous vegetation may facilitate tree germination and seedling
performance at treeline in many ways, e.g. via increased canopy and soil
temperatures through passive solar warming and soil insulation (Körner and Paulsen
2004), or via decreased exposure to high solar radiation through shading (Germino et
al. 2002; Maher et al. 2005; Bader et al. 2007), reducing the risks of desiccation and
photoinhibition (Ball et al. 1991; Germino and Smith 1999). Furthermore, it can
reduce desiccation stress and mechanical damage produced by wind and by frost
heave, with increased seedling survival compared to open patches (Noble and
Alexander 1977; Carlsson and Callaghan 1991; Ryser 1993). On the other hand, for
seedlings reaching above the herbaceous canopy or growing in open patches
surrounded by vegetation, temperature extremes may be exacerbated compared to
Chapter 3
58
bare ground due to radiative warming and cooling of the vegetation (Ball et al. 1997;
Germino et al. 2002). Neighbouring grasses and herbs can also be strong competitors
for resources such as light, water and nutrients (Nambiar 1990; Moir et al. 1999).
These contrasting effects of herbaceous vegetation appear to strongly depend on
vegetation height and density.
Many of the potential effects of alpine vegetation on seedling performance are
related to the seedling’s carbon balance. Adult trees at treeline do not appear to be
limited by carbon (e.g. Hoch and Körner 2003; Shi et al. 2008; Fajardo et al. 2012;
Molina-Montenegro et al. 2012). Seedlings, however, differ substantially from adult
trees in terms of physiology (Day et al. 2001; Tegischer et al. 2002) and the
microclimate they experience (Körner 1998). More specifically, treeline trees in their
early establishment may be subjected to greater carbon limitations, since they have
only small amounts of productive and storage tissues, resulting in a risk of carbon
starvation (Wang and Zwiazek 1999; Li et al. 2002). Furthermore, seedlings are
covered under snow well into spring at most non-tropical treelines, resulting in a
much shorter growing season for reserve accumulation and tissue maturation
(Bansal and Germino 2009). Finally, young seedlings may be more susceptible to
cold-induced photoinhibition (Ball et al. 1991; Germino and Smith 1999; Bader et al.
2007), limiting photosynthetic efficiency and lowering photosynthesis when
sustained (Close et al. 2000). All these limitations can be mitigated or aggravated by
the potentially positive or negative impacts that alpine vegetation exerts on young
tree seedlings’ resource acquisition (Germino et al. 2002; Maher and Germino 2006).
Effects of Herbaceous Vegetation
59
The aim of our experimental study was to assess the balance of positive and
negative effects of alpine herbaceous vegetation on early seedling establishment in
different treeline tree species. Our research questions were: (a) How do varying
intensities of herbaceous vegetation cover influence the microclimate experienced by
treeline tree seedlings? (b) How do seedlings of five common treeline tree species
respond to varying vegetation cover? c) How is the carbon balance during early
establishment affected by adjacent herbaceous vegetation, and how do these effects
differ between seasons and species?
MATERIALS AND METHODS
Study site and species
The study site was located in the experimental garden of the Alpine Research Station
Joseph Fourier in the French Alps near the local treeline (Lautaret Pass, 2100m a.s.l.,
45°02’N, 6°24’E). The site is situated in a climatic transition zone of the wet outer
Alps to the dry inner Alps (Ozenda 1988), with 11°C as the mean temperature of the
warmest month (July) and an average annual precipitation of 1230 mm (Choler et al.
2001). In winter permanent snow cover typically lasts 4-5 months with a moderate
depth of 2-3 m (Franck Delbart, Alpine Research Station Joseph Fourier, personal
communication). Due to past transformation of natural forests to pastures there is no
natural treeline at the Lautaret pass and the vegetation is dominated by species-rich
Festuca paniculata – meadows (Quétier et al. 2007).
The study species comprise the four important treeline-forming conifers Larix
decidua L., Picea abies (L.) Karst, Pinus cembra L., Pinus uncinata Ram. as well as the
broadleaved Sorbus aucuparia L., which also occurs up to treeline elevation in the
Chapter 3
60
Alps (Brändli 1998, personal observation). Seeds of subalpine origin were obtained
from a commercial seed producer (Herzog Baum, Samen und Pflanzen GmbH,
Gmunden, Austria) and a forestry office (Kantonaler Forstgarten Rodels, Rodels,
Switzerland) providing seeds from the inner Alps, except for seeds of Sorbus
aucuparia, which were available only from colline origin in Hungary. Seeds were
germinated beforehand in plastic trays with a homogeneous mixture of commercial
propagation substrate and sand under cool greenhouse conditions (10 – 15 °C) in
March 2013 in Oldenburg, Germany. After approximately one month, seedlings were
transported to the study site and kept indoors at the same temperature regime as in
the nursery due to harsh winter weather until the end of May 2013. Prior to
transplanting into the experiment in early June 2013, seedlings were subjected to an
acclimatization period of three weeks in lightly shaded nursery beds (roofs with
shading cloths providing 40 % shade and allowing penetration of direct sunlight in
the evening) of the Lautaret Alpine Botanical Garden.
Experimental design
Twelve experimental plots (70 x 30 cm) were arranged in three blocks to account for
spatial heterogeneity, with four treatments per block. There was approximately 5 m
distance between the centers of two adjacent blocks and 20 cm distance between
plots. All blocks were surrounded by a 60 cm high wire-mesh fence as protection
against rodents. The treatments consisted of three levels of vegetation cover (Fig. 3.1):
full vegetation cover (FV), intermediate vegetation cover (IV) and bare ground (BG).
Additionally, a shaded bare ground treatment (Sh) was included to study the effect
of light reduction without further biotic interactions between tree seedlings and
herbaceous vegetation. In the full vegetation treatment, the locally occurring dense,
Effects of Herbaceous Vegetation
61
intact alpine grassland vegetation was left unmanipulated. These plots had a
vegetation height at peak biomass in mid-July of approximately 25 cm and a PAR
(photosynthetic active radiation) reduction of 80 % compared to bare ground
(measured at seedling height in the plot centres with a PAR-sensor, Microstation,
Onset Corp, Bourne, MA, USA). For the intermediate vegetation treatment, the
herbaceous vegetation was reduced by regular cutting to a height of 7 cm, which was
approximately seedling height. In this treatment, tree seedlings were largely relieved
of light competition (PAR reduction of 10 % compared to BG) but still subject to
belowground competition and a microclimate essentially modified by surrounding
vegetation. For the bare ground treatments with and without shading (BG and Sh),
plots were excavated to a depth of 15 cm to remove all above- and belowground
plant material. The soil of both vegetation-free plots of a block was then mixed and
returned, creating a homogenous growth substrate within blocks. The shaded
treatments were covered with shading roofs, i.e. plot-sized wooden frames covered
with shade cloth creating light conditions similar to the full vegetation treatment (70
% PAR reduction compared to BG) while allowing rain water to pass. The roofs were
supported by four 30 cm high metal poles at the plot corners and featured 20 cm
shade-cloth curtains on each side to prevent the penetration of low-angle sunshine.
Tree seedlings were planted into the treatment plots in 60-cm rows, with one row per
species and ten seedlings per row. This allowed for 6 cm spacing within and 5 cm
between rows. After planting, seedlings were watered daily for two weeks to reduce
planting stress and associated mortality. Seedling survival was recorded monthly
throughout the growing season and once again in the following spring after
snowmelt.
Chapter 3
62
Fig. 3.1 Experimental set-up including three levels of vegetation cover: A. Full
vegetation cover (no manipulation of the initial alpine grassland vegetation after
seeding /planting of the seedlings). B. Intermediate vegetation cover (reduction of
the vegetation by regular cutting to a height of approximately 7 cm). C. Bare ground
(removal of all above- and belowground plant material). Note the protective shield of
the humidity sensor (10 cm height, furthest left white circle), which is freely visible in
the intermediate vegetation treatment and almost completely overgrown in the full
vegetation treatment (lower left side). Pictures were taken at peak biomass in mid-
July 2014.
This seedling experiment was accompanied by a seed germination
experiment. Seeds of the same five tree species, originating from the same seed
sources, were sown in the four treatments (FV, IV, BG and Sh) in autumn 2013 as
described in Loranger et al. (2016, submitted). Starting after snowmelt in spring 2014,
emerging seedlings were recorded weekly and marked with a colored pin to allow
monitoring each individual. Since germination of soil-covered seeds could not be
monitored directly, seedling emergence was used as a proxy. Germination rates were
defined as the total proportion of emerged seedlings from sown seeds (including
subsequently dead individuals, at the end of the growing season in September 2014).
Microclimate
The microclimate within treatments was monitored by measuring air and soil
temperature with permanently installed temperature loggers (Hobo ProV2, Onset
Corp, Bourne, MA, USA) featuring two external sensors at 5 cm belowground and,
A B C
Effects of Herbaceous Vegetation
63
protected by a sunshield, at 10 cm aboveground, over the complete growing season
from June to September. Relative air humidity (%) was measured using loggers
(HumiLog “rugged”, Driesen & Kern GmbH, Bad Bramstedt, Germany) with an
external humidity sensor protected by a rain/sun shield. All sensors were placed in a
central position in the plot and data were recorded in 30-minute intervals. Due to
limited numbers of HumiLog loggers, air humidity was measured only in three
treatments out of four – i.e. FV, IV, and BG – and in two treatments simultaneously,
comparing IV and BG treatments, respectively, with FV in two subsequent
measurement periods of six days. Furthermore, soil moisture content (%) was
measured monthly throughout the growing season with a hand-held sensor inserted
15 cm deep in each plot center (TRIME-PICO64, IMKO Micromodultechnik,
Ettlingen, Germany).
Air and soil temperatures were recorded in 2013 and 2014, but due to logistical
problems air humidity and soil-moisture contents were recorded only in 2014 and
were therefore used only to determine relative differences between treatments. Air
and soil temperatures showed the same treatment effects for 2013 and 2014. Annual
cumulative precipitation measured by the nearest weather station were very similar
in both years, with 970 mm and 932 mm, respectively (Besse en Oisan Weather
Station Météo France, 1525 m a.s.l., 45°04’N, 6°10’E). Microclimate data are therefore
presented for 2014 only.
Biomass sampling and carbohydrate analysis
Complete seedlings were harvested to determine seedling biomass and carbohydrate
concentrations. Sampling took place twice, at the end of the first growing season
Chapter 3
64
(September 11 – 13, 2013, one block per day) and in the following spring three-to-six
days after snowmelt (May 19 – 22, 2014). For the deciduous species S. aucuparia this
timing was before autumn senescence of leaves and after the opening of leaf buds in
spring. The evergreen conifer species had only first-year needles in both sampling
periods, since their leaf buds were still closed during the spring sampling. This
includes also the generally deciduous conifer L. decidua, whose seedlings do not shed
their needles during the first years. In each sampling, one seedling per species per
plot was randomly selected, carefully excavated and immediately stored on ice.
Samples were brought to the field laboratory within 2 hours, where roots were
washed thoroughly and samples were then microwaved for 90 s at 600 W to stop
enzymatic reactions (Popp et al. 1996). Samples were then dried to constant weight at
70 °C for 48 h and stored on silica gel in sealed containers until further analysis.
Additionally, the initial seedling biomass before planting was determined for five
randomly selected seedlings of each species (L. decidua: 14.8 ± 3.9 mg, P. abies: 22.5 ±
4.6 mg, P. cembra: 170.6 ± 47.7 mg, P. uncinata: 11.6 ± 3.2 mg, S. aucuparia: 24.4 ± 5.5
mg, mean ± SD, n = 5).
Non-structural carbohydrates (NSC) are defined here as the sum of soluble,
low-molecular-weight sugars (glucose, sucrose, fructose and maltose) and starch. The
chemical analysis of NSC took place in the Functional Ecology laboratory of the
University of Oldenburg, where biomass samples were separated into leaf, stem and
root biomass and weighed to ± 0.01 mg. The separated seedling part samples were
than ground to a fine powder and soluble sugars were extracted by heating
approximately 5 mg of this powder mixed with distilled water for 30 min at 80 °C.
After three consecutive centrifugation and dilution steps to wash out soluble sugars
Effects of Herbaceous Vegetation
65
from the sample residue, the combined supernatant was used for measuring soluble
sugar concentration via High Performance Liquid Chromatography (HPLC, ICS-
3000, Dionex Corporation, Sunnyvale, California). For starch, each residue from the
first extraction was mixed with α-amylase, amyloglucosidase, distilled water and
acetate buffer and heated for 2 h at 30 °C to break down starch into glucose. Again,
after three consecutive centrifugation and dilution steps, glucose concentration
originating from starch was measured using HPLC. Concentrations (mg g-1 dry
weight) of each sugar were determined via comparison with standards. It is
important to note that results of NSC-analyses were shown to differ importantly
between laboratories and applied methods (Quentin et al. 2015), so that no cross-
comparisons of absolute values are possible without cross-calibration. Results
originating from the same laboratory and the same method, however, should be
comparable.
Statistical analysis
All analyses were performed using R 3.2.2 (R Core Team 2015). Treatment effects on
plot microclimate were evaluated by analysing seasonal averages of daily minimum,
maximum and mean air and soil temperatures as well as seasonal averages of soil
moisture content with single-factor analysis of variance (ANOVA). For relative air
humidity, which was obtained for only two treatments simultaneously, replicates
were in time (i.e. daily minimum, maximum and mean per day of measurement) and
differences were analysed with paired t-tests for each experimental block separately.
Survival of transplanted seedlings over summer (September 2013) and winter
(May 2014), and total seasonal germination of sown seeds (September 2014) were
Chapter 3
66
expressed as two-column-vectors of successes and failures for each species and
analysed with binomial generalized linear models (GLM), including block and
treatment as explanatory variables. In cases of overdispersion, the standard errors
were corrected by using a quasi-GLM model (Zuur et al. 2009).
At the level of whole seedlings, differences in biomass and carbohydrate
reserves (total NSC concentrations) between treatments and seasons were tested
using two-way factorial ANOVA or analysis of covariance (ANCOVA) for the latter,
controlling for size-effects by including seedling biomass as covariable. On the level
of separated seedling parts, differences in total soluble sugars, starch and NSC were
analysed with an ANCOVA including treatment, part and season as independent
factors with their respective two-way interactions, and biomass as covariable. Where
necessary, data were transformed to meet model assumptions and non-significant
terms were removed from the full models in a backwards stepwise approach. All
statistically significant results (p < 0.05) were evaluated using Tukey honestly
significant difference criteria for pairwise comparisons with Bonferroni corrections
for multiple comparisons. For each species, response variables were analysed in a
separate model, and block was included as a factor in all models to account for the
spatial structure of the experiment.
RESULTS
Microclimate
The four treatments applied in this study (BG, IV, FV, Sh) significantly affected the
microclimate experienced by the germinating and transplanted seedlings. The soil
moisture content in the bare ground treatment was consistently and significantly
Effects of Herbaceous Vegetation
67
lower than in the other treatments (26 ± 5 % vs. 33 ± 3 %; means ± SD; F3,30 = 5.74, p <
0.01). Relative air humidity (rh) was generally lower in the bare ground and the
intermediate vegetation compared to the full vegetation treatment (daily minimum
rh in BG vs. FV: 45 ± 2 vs. 61 ± 8 % and IV vs. FV: 46 ± 3 vs. 56 ± 3 %, means ± SD, n =
3; see Appendix Table S3.1).
Mean and maximum daily air temperatures were significantly lower in the
shading and full vegetation treatment and higher in the intermediate vegetation
treatment compared to bare ground (Fig. 3.2B-C). The minimum daily air
temperature, however, varied much less; the only significant effect being that
shading was colder than full vegetation (Fig. 3.2A). Overall the intermediate
vegetation treatment had the largest air temperature fluctuations (Fig. 3.2). The mean
daily soil temperature was lowest in the shading treatment and highest in the bare
ground and intermediate vegetation treatments (Fig. 3.2C). Thus, soils under
artificial shading were generally coldest, while the bare ground treatment
experienced both the warmest and the second-coldest temperatures and had the
largest soil temperature fluctuations. The soil under the intermediate vegetation
treatment was significantly warmer than the full vegetation treatment (Fig. 3.2), but
both treatments provided a buffering effect leading to relatively low maximum and
high minimum soil temperatures, i.e. small daily soil-temperature fluctuations (Fig.
3.2A-B).
Chapter 3
68
Fig. 3.2 Seasonal means (± SD, n = 3 plots) of daily minimum (A), maximum (B) and
mean (C) temperatures measured in four treatments with manipulations of
vegetation cover and microclimate in a seedling transplant field experiment at 2100
m a.s.l. in the French Alps over the whole growing season (June 6 to September 4,
2014). Open circles indicate air, filled circles soil temperatures. Different letters
indicate significant differences, with upper case letters for air temperatures and
lower case italic letters for soil temperatures. BG = Bare ground (no vegetation
cover), IV = Intermediate vegetation cover, FV = Full vegetation cover, Sh = Shading
(no vegetation cover).
Seedling performance and total reserve accumulation
The only treatment significantly affecting seedling survival was the full vegetation
cover, decreasing survival in three of the five species (L. decidua: F3,19 = 5.3, p<0.01; P.
uncinata: F3,18 = 24.3, p<0.001; S. aucuparia: F3,18 = 6.9, p<0.01). While in both conifers
all dead seedlings were at least partly covered by mould fungi, suggesting a causal
relationship, there was no visible cause of death in S. aucuparia. Moreover, survival
was significantly reduced in L. decidua and P. abies after the winter in all treatments
(F1,19 = 5.2, p<0.05) (F1,20 = 13.5, p<0.01). Pinus cembra was not affected by any
treatment and there was no additional winter mortality (Fig. 3.3).
Treatments
Tem
pe
ratu
re(°
C)
B
a
bc
d
AB AB A
A
a ab
c
AB
C C
CB
a
b
c c
A
B
CC
Effects of Herbaceous Vegetation
69
Fig. 3.3 Means (± SD, n = 3) of survival percentage (first row), total seedling biomass
(second row) and total non-structural carbohydrates (NSC) concentrations per
seedling (third row). Seedlings were grown in a field experiment with manipulations
of vegetation cover and microclimate in the French Alps at 2100 m a.s.l. and samples
were taken in two subsequent seasons: at the end of the first summer (September
2013, black bars) and at the end of the first winter (May 2014, grey bars). Seedling
biomass did not differ between seasons and the data were therefore pooled over
seasons (no extra growth over the winter). Different letters stand for significant
treatment differences, significant seasonal differences or interactions between
treatments and season are represented by the symbols S* and I*, respectively. BG =
Bare ground (no vegetation cover), IV = Intermediate vegetation cover, FV = Full
vegetation cover, Sh = Shading (no vegetation cover).
Seedling biomass was lower in full and intermediate vegetation compared to
bare ground and shading for L. decidua, P. uncinata, S. aucuparia and P. abies, whereas
in the latter the effect of intermediate vegetation was less pronounced (Table 3.1, Fig.
3.3). Pinus cembra was again exceptional: all vegetation- or roof-covered treatments
resulted in lower seedling biomass compared to the bare ground (Table 3.1, Fig. 3.3).
Bio
mas
s(m
g)N
SC c
on
cen
trat
ion
s(m
g g-1
dw
)Su
rviv
al(%
)
Treatments
S*
aa
a
b
a
a
a
b
I*
aa
b
b
a
c
ab
bc
a
b bb
aa
b
b
a
c
ab
bc
abab
aba a a
b
S*Larix decidua Picea abies Pinus cembra Pinus uncinata Sorbus aucuparia
a a ab
S*
BG BG BG BG BG
Chapter 3
70
Table 3.1. Summary of ANOVA/ANCOVA results testing the effect of vegetation
cover (comprising bare ground, bare ground with shading, intermediate and full
vegetation cover as treatments) on the total biomass and total NSC concentrations of
seedlings of five treeline tree species grown in a field experiment in the French Alps
at 2100 m. The two sampling seasons were the end of the first growing season and
the end of the first winter in 2013/2014.
Total Biomass Total NSC
F P F p
Larix decidua
Treatment 9.05 <0.001 9.04 <0.001
Season 0.37 0.55 1.30 0.27
Treatment:Season 0.63 0.61 1.65 0.22
Picea abies
Treatment 8.73 <0.001 0.34 0.80
Season 1.03 0.32 0.01 0.90
Treatment:Season 0.35 0.79 1.53 0.24
Pinus cembra
Treatment 10.22 <0.001 0.33 0.80
Season <0.01 0.95 0.15 0.70
Treatment:Season 0.46 0.71 6.02 <0.01
Pinus uncinata
Treatment 15.51 <0.001 7.83 0.001
Season 5.11 0.04 0.49 0.49
Treatment:Season 1.28 0.32 2.96 0.07
Sorbus aucuparia
Treatment 8.59 <0.001 0.89 0.47
Season 0.52 0.48 7.14 0.01
Treatment:Season 0.13 0.94 0.42 0.74
Shown are the main effects treatment and season as well as their interaction for both response
variables total biomass and total NSC concentrations of each species reporting F- and p-values.
Significant effects are given in bold and italics.
The total non-structural carbohydrate (NSC) concentration per seedling was
lower in full vegetation compared to all other treatments in L. decidua and P. uncinata.
In P. cembra, an interaction between treatment and season indicated that this was the
case only at the end of the summer, while seedling NSC concentration in full
vegetation increased over the winter half-year in contrast to the other treatments
Effects of Herbaceous Vegetation
71
(Table 3.1, Fig. 3.3). A similar pattern was found in the other three conifers (Fig. 3.3).
In the only broadleaved species, S. aucuparia, total NSC concentrations were not
affected by treatments, but decreased significantly over the winter half-year, while P.
abies was neither significantly affected by treatment nor by season (Table 3.1, Fig.
3.3).
None of the different treatments affected seedling emergence, except for a
positive effect of shade roofs on S. aucuparia (F3,8 = 4.6, p<0.05, data not shown).
Organ-dependent reserve accumulation
The patterns of reserve accumulation in seedling parts as a function of treatment and
season is consistent with that of entire seedlings. For the four conifers, the main
effects indicated that i) seedlings in the full vegetation treatment had generally lower
carbohydrate concentrations compared to the other treatments, ii) reserve
accumulation (especially starch) was highest in the roots, and iii) starch reserves
decreased over winter (Fig. 3.4, Appendix Table S3.2). Carbohydrate concentrations
within treatments and seedling parts often differed between seasons, but patterns
were plant-part and species specific, which is shown by many highly significant
interactions (Fig. 3.4, Appendix Table S3.2).
In the following paragraphs we present more specific results. Frequently,
trends are similar, but not always significant for all NSC components and species.
Hence, for conciseness, the NSC component and the species with significant effects
are displayed in parentheses.
Chapter 3
72
Fig. 3.4 Mean concentrations of non-structural carbohydrates (NSC) as a sum of soluble carbohydrates (black bars) and starch (white
bars) in three plant parts (leaves, stems and roots) of seedlings of five treeline tree species. Seedlings were grown in a field experiment
with manipulations of vegetation cover and microclimate in the French Alps at 2100 m a.s.l. and samples were taken in two
subsequent seasons: at the end of the first growing season (September 2013) and at the end of the first winter (May 2014). BG = Bare
ground (no vegetation cover), IV = Intermediate vegetation cover, FV = Full vegetation cover, Sh = Shading (no vegetation cover).
Shown are means of three seedlings ± SD for total NSC only. For L. decidua in May 2014, FV, only one seedling could be sampled.
Leave
sSte
mR
oo
ts
Treatments
Larix decidua Picea abies Pinus cembra Pinus uncinata Sorbus aucuparia
Sep May Sep Sep Sep SepMay May May May
NSC
co
nce
ntr
atio
ns
(mg
g-1d
w)
BG BG BG BG BG BG BG BG BG BG
H2
Effects of Herbaceous Vegetation
73
Carbohydrate concentrations among the conifers were significantly lower in
full vegetation (L. decidua and P. uncinata: all three NSC components, P. cembra:
starch; Fig. 3.4, Appendix Table S3.2). This effect was strongly linked with an
interaction of treatment and season, showing reduced carbohydrate concentrations in
full vegetation only at the end of the summer but not anymore after the winter half-
year (P. uncinata: soluble sugars, P. abies: starch, by trend also in L. decidua: soluble
sugars). In P. cembra the seasonality in treatment effects was even stronger, with
increased NSC concentrations after the winter in full vegetation, while in the other
treatments concentrations decreased during winter as in the other species, ultimately
yielding significantly higher NSC concentrations in the full vegetation treatment than
in bare ground in spring (Fig. 3.4, Appendix Table S3.2). An additional treatment
effect was observed in P. uncinata, in which NSC concentrations were significantly
higher in intermediate vegetation (and by trend in the shading treatment) than on
bare ground (Fig. 3.4, Appendix Table S3.2).
The interactions between seedling part and season reveal that carbohydrate
reserves in the conifers decreased over the winter half-year in leaves (L. decidua:
soluble sugars) or both leaves and stem (P. cembra and P. uncinata: starch), while root
reserves remained relatively stable. In P. abies carbohydrate concentrations of
seedlings parts did not differ with the season (Fig.3.4, Appendix Table S3.2).
Variation in carbohydrate concentrations in seedlings of the broadleaved S.
aucuparia generally differed from that in the conifers, with one exception: Reserve
accumulation (especially starch) was highest in stem and roots, in contrast to the
evergreen species but similar to the deciduous L. decidua (Fig. 3.4, Appendix Table
Chapter 3
74
S3.2). In S. aucuparia, the only treatment effect showed an interaction with seedling
parts, with a significantly lower soluble sugar concentration in leaves from the
intermediate vegetation treatment compared to leaves from the other three
treatments (Fig. 3.4, Appendix Table S3.2). While starch and NSC concentrations
decreased over the winter as in the conifers, soluble sugars were significantly higher
in spring (Fig. 3.4, Appendix Table S3.2).
DISCUSSION
The influence of neighbouring herbaceous vegetation on early establishment of
treeline trees revolves around the importance of two opposed interactions –
competition and facilitation. While evidence of both interaction types has been
previously documented at different treeline sites (Moir et al. 1999; Germino et al.
2002; Dullinger et al. 2003; Maher et al. 2005), our study suggests that negative
impacts dominate. We found general competition effects of herbaceous vegetation on
tree seedlings as well as species-specific susceptibilities to combinations of
competition and indirect vegetation effects via microclimate or pathogens.
Competition is not evident across all seasons, however, and seasonally the outcome
may even switch to facilitation: in autumn / winter, evergreen species may benefit
from the protection by the remains of high vegetation to allow increased carbon
gains than in sites with low or no vegetation cover. The net effect in terms of growth
and survival is still generally negative for the first year, but this autumn window to
improve the carbon balance may be of utmost importance for the interaction between
trees and the alpine vegetation in determining the long-term tree establishment
success.
Effects of Herbaceous Vegetation
75
Competition dominates the interaction between tree seedlings and neighbouring
vegetation
Seedlings of the five focal tree species generally produced less biomass in plots with
neighbouring herbaceous vegetation (FV and IV) compared to plots without it (BG
and Sh; Fig. 3.3). Compared with the initial biomass, this means no biomass gain
with vegetation versus an approximate doubling without it. The only common
characteristic within these two pairs of treatments being the presence or absence of
neighbouring vegetation, competition rather than microclimate arguably causes this
result. This pattern further points towards belowground rather than aboveground
competition for light in determining growth, because in most species biomass in the
shading treatment – reducing light availability without neighbouring vegetation –
was similar to the bare ground treatment.
The observed dominance of negative vegetation effects is in line with
numerous studies in forestry on the detrimental effect of grass cover on tree
seedlings (e.g. Sims and Mueller-Dombois 1968; Nambiar 1990; Ellis and Pennington
1992), with the general observation that belowground competition for water and
nutrients prevails over competition for light in nutrient-poor environments (Coomes
and Grubb 2000; Strand et al. 2006; Bloor et al. 2008; Axelsson et al. 2014). Given i)
functional differences such as dense root mats of grassy vegetation versus shallow,
simple root system of young seedlings (Casper and Jackson 1997) and ii) nitrogen
being particularly limiting in cold ecosystems such as treelines (Thébault et al. 2014),
competition for soil resources may outweigh the potentially facilitative effects of
neighbouring herbaceous vegetation as documented at different treeline sites (Moir
et al. 1999; Dullinger et al. 2003). The contrast of our results with reports of
Chapter 3
76
exclusively positive interactions between treeline tree seedlings and grassland
vegetation (Germino et al. 2002; Maher et al. 2005) are probably related to differences
in the structure of the alpine plant communities in the respective study areas and /
or to different susceptibilities of the studied tree species. This is supported by the
findings of Bansal et al. (2011) using a gradient of herb cover, along which they
detected evidence for above- and belowground competition in treeline tree seedlings,
in line with our results, primarily under full herb cover, while survival of their two
study species was affected contrastingly by intermediate herb cover. Furthermore,
the type of response parameter measured might lead to contrasting results, since a
focus on stress effects (e.g. photoinhibition, mortality) may over-emphasize the
positive aspects of the interaction without revealing resource competition. Finally,
locally differing resource-related abiotic conditions (soil types, precipitation) as well
as alpine plant species composition can add further variation. For example, the
presence of legumes is known to enhance soil nitrogen (Li et al. 2015; Bowman et al.
1996), which could result in reduced competition for soil resources in favour of
facilitative effects.
In addition to these general patterns, there were several responses reflecting
species-specific negative impacts of neighbouring vegetation and / or microclimate.
First, P. cembra was the only species to show a lower biomass production in all
vegetation- or roof-covered plots compared to bare ground, which may be due to the
moister conditions in these treatments. A negative impact of soil moisture on early
establishment of this species has been shown previously (Loranger et al. 2016,
submitted), which agrees with its distribution at rather dry sites (e.g. steep slopes;
convex, exposed topographies; Brändli 1998; Didier 2001). Nevertheless, survival of
Effects of Herbaceous Vegetation
77
this species was high in all treatments and NSC concentrations were reduced only in
full vegetation cover at the end of the summer, indicating a low sensitivity to the
presence or absence of vegetation cover. This response might be enabled by the large
and rich seeds of this species, generally characterized as competition-intolerant,
providing the seedlings with long lasting reserves at least during the first growing
season (Ulber et al. 2004). Also P. abies appeared relatively insensitive to varying
vegetation cover, since we found no significant treatment effect for survival or NSC
concentrations. This may explain this species’ capacity to invade abandoned
subalpine mountain pastures (Bolli et al. 2007), where herb cover tends to be well
developed.
The most sensitive species were L. decidua and P. uncinata, in which full
vegetation cover resulted in strong reductions in survival, growth and total NSC
concentrations (Fig. 3.3). These two species are characterized as heliophiles,
regenerating on bare or sparsely vegetated ground (Rameau et al. 1993; Didier 2001;
Batllori and Camarero 2009), which matches their generally negative responses to full
vegetation. Interestingly, however, while growth was also reduced in intermediate
vegetation, there was no apparent limitation for NSC accumulation in this treatment.
This indicates that growth was limited through belowground competition, but
photosynthesis was not, due to sufficient light availability. Similarly interesting, the
shading treatment affected neither survival, growth nor NSC concentrations of both
these species (L. decidua and P. uncinata; Fig. 3.3). A possible explanation is that 30 %
light availability in the shading treatment (compared to 20 % under full vegetation)
may be above the necessary threshold to obtain a positive carbon balance, and thus
to support growth and survival. However, this remains difficult to confirm without
Chapter 3
78
an independent light availability gradient. Doubtlessly, these two species can tolerate
relatively shady conditions as long as belowground competition is absent. Since
seedling survival was negatively affected only by the full vegetation treatment, it can
be concluded that both below- and aboveground competition are needed to decrease
the seedling survival of these species. Notably, seedlings of these species (L. decidua
and P. uncinata) dying in full vegetation were always at least partly covered by
mould fungi. These seedlings probably experienced the highest intensities of
competition (i.e. above- and belowground). This can result in morphologically and
physiologically weakened individuals, resulting in a high susceptibility to pathogen
infections (Seiwa 1998), which is a primary cause of tree seedling mortality
(Yamazaki et al. 2009). Furthermore, pathogens might have been promoted in full
vegetation due to the humid and cool microclimate and close contact to neighbouring
plants.
In S. aucuparia, growth was also strongly reduced under both vegetation
treatments, but NSC concentrations were not. Seedlings of S. aucuparia are known to
be shade-tolerant (Raspé et al. 2000). Accordingly, carbon storage appeared to be
actively maintained at the expense of growth, a strategy associated with shade-
tolerance (Kobe 1997). Similarly to the previous two species, survival of S. aucuparia
was negatively affected by the combination of above- and belowground competition.
While in this case there was no visible cause of seedling mortality, the results are also
in accordance with the species’ regeneration niche in forest understorey (Zywiec and
Ledwoń 2008). There, light availability is low, but dense herbaceous vegetation is
generally absent, preventing the limiting combination of intense belowground
Effects of Herbaceous Vegetation
79
competition and shaded conditions experienced by seedlings in our full vegetation
treatment.
The preceding discussion highlights how competition (here belowground) can
alter the range of abiotic conditions that a species can tolerate, i.e. we show that the
fundamental niche (versus realized niche; see Hutchinson 1957) of L. decidua, P.
uncinata and S. aucuparia spans over a larger range of light conditions than expected
from their observed distribution. Indeed, L. decidua and P. uncinata – considered
heliophile species – can establish well in shaded conditions without belowground
competition (shading treatment), while S. aucuparia – considered shade-tolerant and
regenerating mainly in the understorey – does well in strong light conditions but not
in shaded conditions with competition. This strongly suggests that competition
causes their restricted observed realized niche.
Seasonal shifts from competition to facilitation depend on the leaf functional type
The negative impact of full vegetation cover on carbohydrate accumulation in
evergreen conifers was highly seasonal. Carbohydrate concentrations in this
treatment were lowest at the end of the summer, but after the winter half-year this
negative effect had disappeared in evergreen seedlings (P. abies, P. uncinata), or even
became positive in P. cembra (Figs 3.3-3.4). Hence, while reserves generally decreased
during winter, they increased for evergreen seedlings in the full vegetation treatment,
possibly due to a release from competition in this period. During summer, there was
intense competition by high and dense vegetation as indicated by the low biomass
increment of seedlings. At the end of the summer, at the time of the first sampling,
the aboveground parts of grasses and herbs started to die back. Since a permanent
Chapter 3
80
snow cover rarely develops at this site before December, seedlings with evergreen
leaves had a considerable period to be photosynthetically active. However,
transitional seasons such as autumn are particularly stressful, as frequent
combinations of cold temperatures and high irradiation can cause photoinhibition
(Germino and Smith 1999). Seedlings in the full vegetation treatment could now
benefit from partial shade in the remaining dead vegetation matrix and buffered
temperature conditions (Germino et al. 2002) and maintain or even increase their
carbohydrate reserves, while photosynthetic activity in the other treatments was
limited. To our knowledge, such a seasonal switch from competition to facilitation
has never been shown before for treeline tree species (but see Kikvidze et al. 2006 and
Venn et al. 2009 for examples in subalpine, herbaceous vegetation). These findings
demonstrate the dynamic potential of biotic interactions, even over short periods of
time, and their important influence on the early establishment of trees at their
elevational limit.
Unsurprisingly, the deciduous seedlings of S. aucuparia could not benefit from
the release of competition after the retreat of the herbaceous vegetation.
Characteristic for deciduous trees, they showed a large build-up of starch reserves in
stem and roots at the end of the summer, which then decreased in all treatments until
the end of the winter, probably most strongly after resuming activity in spring, i.e. at
bud break, shortly before our sampling (Kozlowski 1992). Seedlings of the deciduous
L. decidua are evergreen during their first years and show an intermediate response.
In this regard, the tendency of a seasonal vegetation effect indicates a certain benefit
from the extended productive season. This might contribute to advantages of the
Effects of Herbaceous Vegetation
81
ontogenetic switch of leaf types in this species as a strategy to alleviate competition
for small seedlings.
Our results show that herbaceous vegetation, which is dominant at and above
many treeline sites, exerts an important, mostly limiting impact on young tree
seedlings. This interaction can change depending on season and leaf type of the
seedlings, with a switch to facilitation leading to increased carbohydrate reserves in
the following spring. For most species this increase compensated for the lower NSC
concentrations due to competition in the previous summer, while for P. cembra it
even resulted in a net increase. It is, however, currently unclear whether these
increased reserves can sustain seedling growth and survival in dense herbaceous
vegetation over longer periods of time and whether the growth disadvantage from
the first year can be compensated.
In contrast to the clear responses in seedling performance, the germination
response was not affected by vegetation cover, showing that the abiotic conditions
remain within the range required by this earliest life-stage. Nevertheless, our seeds
were sown directly in the soil, whereas free-falling seeds will often be prevented
from reaching the soil by dense vegetation cover, undoubtedly reducing germination
success.
Although early seedling establishment is only the first step to successful
regeneration, our results suggest that tree establishment at and above alpine treelines
with dense herbaceous vegetation might be facilitated by disturbances creating sites
without vegetation cover. Indeed, such a relationship of seedling establishment with
exposed mineral soil caused by geomorphological processes or animal activities has
Chapter 3
82
been documented before (Malanson et al. 2009 and references therein) and might be
particularly important for species sensitive to dense vegetation cover, such as L.
decidua and P. uncinata. At the same time, without ground disturbances dense
vegetation cover may impede regeneration and explain treelines that are stable or
respond only slowly to warming temperatures. While these findings are an
important contribution towards an explanation of local treeline patterns and
dynamics, long-term studies including at least two growing seasons are urgently
needed to clarify if the observed seasonal facilitation results in a longer-term net
facilitation. Such studies should also include different life-stages and tree species to
take into account the dynamic potential of plant-plant interactions at a distribution
boundary.
ACKNOWLEDGEMENTS
We are very grateful to Serge Aubert, former director of the Lautaret Alpine
Botanical Garden and the affiliated research station Joseph Fourier, for enabling this
study by allowing us to use the experimental space and facilities of the station, and
thank the gardener team for support with any technical problem. We also thank
several field assistants, especially Jasmin Baruck, Gesa Pries, Verena Schenk and Eric
Thurm, for their help in installing, maintaining and monitoring the experiment. The
study was funded by the German Research Foundation (DFG, BA 3843/5-1&2).
Effects of Herbaceous Vegetation
83
APPENDIX
Table S3.1. Summary of paired t-tests comparing microclimatic conditions between
treatments. Compared were daily minimum, maximum and mean relative air
humidity (rh) in treatments with three different levels of vegetation cover measured
in a seedling transplant experiment in the French Alps at 2100 m.
Treatment
comparisons Block 1 Block 2 Block 3
FV vs. IV
rh min d = 12.1, p < 0.001 d = 12.5, p < 0.001 d = 4.1, p = 0.19
rh max d = 0.4, p = 0.06 d = 1.0, p = 0.06 d = - 0.5, p = 0.52
rh mean d = 6.1, p < 0.001 d = 9.6, p < 0.01 d = - 2.7, p = 0.09
FV vs. BG
rh min d = 20.1, p < 0.001 d = 23.0, p < 0.001 d = 7.0, p = 0.09
rh max d = 0.6, p = 0.08 d = -0.2, p = 0.36 d = 0.1, p = 0.51
rh mean d = 9.1, p < 0.001 d = 9.2, p < 0.001 d = 4.1, p < 0.001
Rows show results of treatment pairwise comparisons of minimum, maximum and mean values
giving the mean of the differences (d, %) and the p-value with significant effects given in bold. BG =
Bare ground (no vegetation cover), IV = Intermediate vegetation cover, FV = Full vegetation cover.
Chapter 3
84
Table S3.2. Summary of ANCOVA models testing the differences in non-structural carbohydrates (NSC) and of its two components
(soluble carbohydrates and starch) between treatments, plant parts and seasons. Seedlings of five treeline tree species were sampled at
the end of the first growing season and the end of the first winter 2013/2014 in a field experiment in the French Alps at 2100 m.
Soluble Carbohydrates Starch NSC
F p Effect F p Effect F p Effect
Larix decidua
Treatment 6.02 0.001 BGa, IVa, FVb, Sha 6.64 <0.001 BGab, IVa, FVb, Sha 15.37 <0.001 BGa, IVa, FVb, Sha
Part 2.13 0.13 … 13.14 <0.001 Lb, Sa, Ra 4.1 0.02 Lb, Sab, Ra
Season 1.18 0.28 … 9.23 <0.01 13a, 14b 7.33 <0.001 13a, 14b
Treat:Part 0.83 0.55 … 1.98 0.09 … 1.05 0.41 …
Treat:Season 1.63 0.19 … 1.88 0.15 … 3.28 0.03 …
Part:Season 11.35 <0.001 L13>{R13, S13,
L14}
R14>L14
0.17 0.85 … 8.38 <0.001 L14<{R14, S14, L13}
Picea abies
Treatment 1.68 0.18 … 1.48 0.23 … 1.67 0.18 …
Part 2.36 0.10 … 10.69 <0.001 Lb, Sb, Ra 1.99 0.15 …
Season 0.05 0.83 … 0.05 0.82 … <0.01 0.95 …
Treat:Part 0.73 0.63 … 0.41 0.87 … 0.49 0.81 …
Treat:Season 0.43 0.74 … 6.33 <0.001 Sh13>{FV13, Sh14} 2.34 0.08 …
Part:Season 0.09 0.91 … 2.25 0.11 … 0.72 0.49 …
Pinus cembra
Treatment 1.96 0.13 … 10.17 <0.001 BGa, IVa, FVb, Sha 2.03 0.12 …
Part 0.67 0.52 … 43.13 <0.001 Lb, Sb, Ra 10.70 <0.001 Lb, Sb, Ra
Season 0.57 0.45 … 10.31 <0.01 13a, 14b 0.31 0.58 …
Treat:Part 0.89 0.51 … 2.01 0.08 … 1.16 0.34 …
Treat:Season 5.59 <0.01 FV14>{BG14,
FV13}
6.26 <0.001 BG14< BG13 10.00 <0.001 BG14<FV14
FV13<{ BG13, IV13,
FV14}
Part:Season 1.74 0.19 … 10.98 <0.001 [L, S]13>[L, S]14 0.08 0.93 …
Effects of Herbaceous Vegetation
85
Pinus uncinata
Treatment 8.26 <0.001 BGab, IVa, FVb, Sha 97.28 <0.001 BGa, IVa, FVb, Sha 18.75 <0.001 BGb, IVa, FVc, Shab
Part 11.03 <0.001 La, Sb, Ra 28.57 <0.001 Lb, Sb, Ra 15.08 <0.001 La, Sb, Ra
Season 0.75 0.39 … 27.05 <0.001 13a, 14b 2.71 0.11 …
Treat:Part 1.36 0.25 … 1.68 0.15 … 1.00 0.43 …
Treat:Season 5.28 <0.01 FV13<(BG, Sh,
IV)13
5.92 <0.01 BG13>BG14;
Sh13>Sh14
6.94 <0.001 BG14<{ IV14, BG13}
Part:Season 1.60 0.21 … 33.19 <0.001 L13>L14; S13>S14
R14>(L14, S14)
7.75 0.001 R14>{L14, S14}
Sorbus aucuparia
Treatment 2.71 0.05 … 2.64 0.06 … 0.53 0.66 …
Part 0.77 0.47 … 16.68 <0.001 Lb, Sa, Ra 6.96 <0.01 Lb, Sa, Ra
Season 7.14 0.01 13b, 14a 102.4 <0.001 13a, 14b 31.10 <0.001 13a, 14b
Treat:Part 2.88 0.02 IV-L<{FV-L, Sh-L} 0.45 0.84 … 1.25 0.30 …
Treat:Season 1.25 0.30 … 0.69 0.56 … 0.80 0.50 …
Part:Season 36.58 <0.001 L14<{ L13, S14,
R14}
R13<{R14, L13}
S13<{S14, L13}
5.42 <0.01 L13>L14
S13>{L13, S14}
R13>{L13, R14}
2.17 0.12 …
Shown are the main effects treatment, seedling part and season as well as their two- and three-way-interactions for the three response variables soluble
carbohydrates, starch and total NSC concentrations of each species reporting F- and p-values. Significant effects (p<0.05) are given in bold and italics, marginally
significant effects (p<0.1) only in italics. Notes - Treatment: BG = Bare ground (no vegetation cover), IV = intermediate vegetation, FV = full vegetation, Sh =
Shading (no vegetation cover). Part: L = leaves, S = stem, R = roots. Season: 13 = September 2013, 14 = May 2014. Significant differences between treatments,
parts, or seasons are represented with different letters in superscript, with the first letter in alphabetical order representing higher carbohydrate concentrations (a
> b).
87
CHAPTER 4
A COOL EXPERIMENTAL APPROACH TO EXPLAIN ELEVATIONAL
TREELINES, BUT CAN IT EXPLAIN THEM?
Maaike Y. Bader, Hannah Loranger, Gerhard Zotz
Published in American Journal of Botany
ABSTRACT
At alpine treeline, trees give way to low-stature alpine vegetation. The main reason
may be that tree canopies warm up less in the sun and experience lower average
temperatures than alpine vegetation. Low growth temperatures limit tissue
formation more than carbon gain, but whether this mechanism universally
determines potential treeline elevations is the subject of debate. To study low-
temperature limitation in two contrasting treeline tree species, Fajardo and Piper (
Chapter 4
88
American Journal of Botany 101: 788–795) grew potted seedlings at ground level or
suspended at tree-canopy height (2 m), introducing a promising experimental
method for studying the effects of alpine-vegetation and tree-canopy microclimates
on tree growth. On the basis of this experiment, the authors concluded that lower
temperatures at 2 m caused carbon limitation in one of the species and that treeline-
forming mechanisms may thus be taxon-dependent. Here we contest that this
important conclusion can be drawn based on the presented experiment, because of
confounding effects of extreme root-zone temperature fluctuations and potential
drought conditions. To interpret the results of this elegant experiment without
logistically challenging technical modifications and to better understand how low
temperature leads to treeline formation, studies on effects of fluctuating vs. stable
temperatures are badly needed. Other treeline research priorities are interactions
between temperature and other climatic factors and differences in microclimate
between tree canopies with contrasting morphology and physiology. In spite of our
criticism of this particular study, we agree that the development of a universal
treeline theory should include continuing explorations of taxon-specific treeline-
forming mechanisms.
Key words: alpine treeline; carbon balance; ecophysiology; Fuscospora; growth
limitation; methodology; micrometeorology; Nothofagus; Pinus; timberline
Methodological Approaches
89
INTRODUCTION
In a recent paper in the American Journal of Botany , Fajardo and Piper (2014)
presented an experimental approach that compares tree seedling performance in
conditions typical for low-stature alpine vegetation and conditions representing tree
canopy conditions at treeline. The rationale of this approach was that treelines are
assumed to form because tree canopies cannot warm up as much as low-stature
alpine vegetation can, due to a stronger coupling to atmospheric conditions (Körner
1998). The lower temperatures experienced by a tree canopy do not allow growth,
either because carbon assimilation is insufficient (the source-limitation or carbon-
limitation hypothesis), or because growth processes (e.g., cell division or
lignification) are directly impaired (the sink-limitation or growth-limitation
hypothesis). Although most results of recent research, in particular on elevational
patterns of non-structural carbohydrate contents in trees (summarized by Hoch and
Körner 2012) and on low-temperature limits to wood formation (e.g. Rossi et al.
2008), support the growth limitation hypothesis, some findings seem to point at
carbon limitation (Wiley and Helliker 2012; Dawes et al. 2013; though see Palacio et
al. 2014). This issue is thus not generally resolved.
The new approach used by Fajardo and Piper (2014) aimed to study low-
temperature limitation in trees experimentally by placing individuals of similar size
and developmental stage in temperature regimes usually experienced by individuals
of very different sizes, thereby avoiding any confusion of microclimatic effects with
ontogenetic and size effects. Even though tree seedlings were used, this approach
was not aimed at studying seedling performance as such but used seedlings as a
Chapter 4
90
small-version model of a taller tree to study responses of tree growth to
microclimatic conditions.
In the experiment, one of the two species tested, Nothofagus pumilio , showed a
decrease in growth and non-structural carbohydrate (NSC) reserves when suspended
at 2 m above the ground, where mean temperatures were lower than at ground level.
The other species, Pinus contorta, showed no response. The authors concluded that
their study is the first unambiguous test of the mechanism behind growth limitations
at treeline elevation, with evidence for differences between treeline-forming taxa.
Such an important finding deserves thorough scrutiny. Here, we argue that due to
their particular experimental method, the observed carbon limitation may be due to
other factors than low mean temperature and does not represent strong and
unambiguous evidence for a treeline-specific phenomenon.
However, we do agree with the authors that potential taxon-specific, growth-
limiting mechanisms at treeline should be seriously considered when further
developing alpine-treeline theory. First, because we should continue to question
whether there is really one mechanism explaining this low-temperature life-form
boundary or whether equally valid representatives of this life-form are constrained
differently. And second, because only by systematically exploring the variation
beyond a universal treeline-forming mechanism can we hope to understand and
predict treeline elevations in real landscapes.
Methodological Approaches
91
CONFOUNDING TEMPERATURE CONDITIONS
It is an excellent idea to test differences in plant performance under low-vegetation-
and tree-canopy-temperature regimes independent of ontogeny and size. Growing
seedlings at ground level and at canopy level (but outside an actual canopy) to mimic
the different levels of atmospheric coupling is a highly promising approach.
However, there is one fundamental problem: soil temperatures in suspended pots
fluctuate very strongly. They fluctuate much more strongly than soil temperatures at
ground level and also more strongly than air temperatures around the suspended
pots ( but less than air temperatures at ground level, see Fig. 2 of Fajardo and Piper,
2014). In the experiment of Fajardo and Piper (2014), suspended seedlings thus
experienced similar temperature fluctuations in roots (3–10 ° C) and shoots (3–8 ° C),
whereas ground level seedlings experienced low root-zone fluctuations (6–8 ° C) and
high shoot-zone fluctuations (2–14 ° C).
Such strong temperature fluctuations in suspended pots clearly do not mimic
root-zone conditions below a tree canopy, which are very stable because of the large
soil volume and the shade provided by the canopy (Körner and Paulsen 2004). In the
experiment of Fajardo and Piper (2014), the average temperature in the suspended
pots was about 1 K lower than in ground-level pots. In that sense, the suspended
pots did mimic one aspect of soil conditions under a tree canopy, which are nearly
always cooler during the growing season than under nearby alpine vegetation
(Bendix and Rafiqpoor 2001; Bader et al. 2007; Körner 2012). The crucial question
here is, however, what do these mean temperatures mean physiologically? Biological
rates respond nonlinearly to temperature, so that at a given mean temperature,
Chapter 4
92
fluctuating temperatures should lead to different mean biological rates than constant
temperatures. On the other hand, mean growing season temperatures are the most
consistent thermal parameter at treelines worldwide (Körner and Paulsen 2004;
Cieraad et al. 2014; Paulsen and Körner 2014) and experimental results from Hoch
and Körner (2009) showed similar growth rates in two conifer species at constant and
variable temperatures. Although these observations still await a physiological
explanation, they suggest that mean temperature really has a biological meaning. The
lower mean root-zone temperature in the suspended pots could thus rightfully be
expected to slow seedling growth. However, as long as it cannot be excluded that the
strong temperature fluctuations contributed to the observed differences in seedling
performance, the observed effect cannot be unequivocally attributed to the different
mean temperatures.
How likely is it that these fluctuations were really a problem? The answer
depends on how temperature differences between root and shoot and on how
temperature fluctuations in general affect seedling physiology. Although these seem
two very basic biological questions, there is surprisingly little information to guide
an answer (Pregitzer et al. 2000). Opposite effects of root and shoot temperature on
plant nutrient status (Weih and Karlsson 2001) and biomass allocation patterns
(Larigauderie et al. 1991) suggest that temperature differences between roots and
shoots, aside from their absolute temperatures, can affect whole-plant performance.
However, we did not find experiments focusing explicitly on this question.
Temperature-fluctuation-effects on plant growth are hardly studied either. In an
experiment addressing this for trees at low mean temperatures, conifer seedlings
grew similarly well under constant or variable (ca. 6 K amplitude around 6 ° C or 12 °
Methodological Approaches
93
C) daily and seasonal temperature regimes, with only slight positive effects of
fluctuations for Larix decidua but not for Pinus mugo (Hoch and Körner 2009). In
contrast, in the only other experiment with trees that we could find, temperature
variability (5 or 10 K amplitude around 23 ° C, with a 5-d fluctuation) clearly affected
growth and root to shoot ratios in poplar ( Populus deltoids × nigra ) cuttings, with
positive effects of fluctuations at the intermediate but not at the greater amplitude
(Cerasoli et al. 2014). Obviously, this limited and ambiguous evidence does not allow
generalizations. Additional experiments addressing temperature-fluctuation effects
on tree growth, addressing differences between species or functional types, the mean
temperature and the amplitude of the fluctuations, are clearly desirable. Apart from
identifying potential artifacts in experiments like that of Fajardo and Piper (2014),
such studies could greatly contribute to understanding the physiological meaning of
different temperature parameters for treeline formation.
CONFOUNDING MOISTURE CONDITIONS
There is a second concern: how severe was drought stress in these suspended plots,
and did this stress affect the results? Seedlings were only watered during the first
month of the growing season. Even though precipitation during the growing season
was ca. 500 mm at this Patagonian site (Fajardo and Piper 2014), this statistic does not
preclude that rainless periods were frequent and that the soil in the pots dried out
repeatedly during these periods, especially in the suspended pots. If P. contorta is less
sensitive to drought than N. pumilio , this sensitivity could explain the species-
specific responses. The documented carbon limitation of N. pumilio (reduced growth
Chapter 4
94
associated with low NSC contents) could thus be caused by water stress and not by
low temperature.
The authors argued that water shortage should not have affected N. pumilio
seedlings, based on results of earlier experiments where seedlings did not respond to
watering in ambient and warmed conditions (Piper et al. 2013). However, not
requiring extra watering in full-soil conditions does not imply drought tolerance in
suspended pots. Playing devil’s advocate, one could even argue that this earlier
experiment did show a trend, though not significant, to higher growth rates under
watering (Piper et al. 2013). Similar negative effects of drought (as a result of
experimental warming) have been observed at treeline in New Zealand for seedlings
of Fuscospora cliffortioides ( previously Nothofagus solandri var. cliffortioides, Heenan
and Smissen 2013) (Melanie Harsch, University of Washington, Seattle, personal
communication) and in North America for recently germinated Pinus flexilis
seedlings (Moyes et al. 2013). In the contested experiment by Fajardo and Piper
(2014), the higher mortality (significant when combining both species) in the
suspended pots also suggests a stress factor confounding or aggravating a potential
temperature effect.
To conclude, the study by Fajardo and Piper (2014) does not, because it could
not, provide unambiguous support for low temperature- induced carbon limitation
in N. pumilio. Because adult trees of the same species show no decrease in NSC with
elevation and thus do not appear carbon limited (Fajardo et al. 2011), the
fundamentally different growth-limitation for Nothofagus compared with other
treeline trees remains to be shown.
Methodological Approaches
95
ALTERNATIVE EXPERIMENTAL SETUPS
Finding problems in an experiment is easy. More difficult and more useful is
proposing solutions. Understanding how low temperatures limit tree growth is not
only of fundamental biological interest but also affects model predictions of tree
growth in a warmer and CO 2 - richer future. Temperature effects can be studied in
isolation in fully controlled conditions in growing chambers (e.g. Hoch and Körner
2009). However, it is very difficult to mimic the typical treeline combination of high
radiation and low air temperature. For the question of carbon vs. growth limitation,
in particular, results from such experiments will be hard to translate to the real
world.
In the field, one option for studying temperature effects on treeline tree
growth is to warm existing tree canopies. However, this technically and logistically
very challenging manipulation is practicable only for small sections such as branches
(Lenz et al. 2013) or buds (Petit et al. 2011). Such sectional studies can yield important
information about local growth processes but cannot contrast carbon vs. growth
limitation directly, because these involve microclimatic effects on whole-tree carbon
gain and use.
Using seedlings as model tissue for tree performance is a sensible alternative.
Compared with adult trees, they allow for better replication and faster responses and
a distinction of microclimatic from ontogenetic and size effects. Suspended seedlings
experience tree canopy conditions, but the results may be confounded by soil
temperature fluctuations not seen in natural soil and by uncontrolled soil moisture.
The moisture issue is relatively easy to solve (disregarding logistic difficulties for
Chapter 4
96
maintaining an irrigation system at treeline). Whether temperature fluctuations are a
problem remains to be investigated, as discussed above. As long as this is unclear,
temperature needs to be controlled. Temperature fluctuations in suspended pots can
be reduced by using reflective material, insulation, and ventilation layers around the
pots. However, in trials related to a somewhat similar experiment, we were unable to
design pots that can be suspended in air while maintaining the stable temperature
conditions typical for full soil, let alone for full soil under a tree canopy (see
Appendix S4.1). We tried this in northern Germany (ca. 53 ° N) in early spring, with
relatively mild sunshine loads. At the alpine treeline in the middle of summer and at
lower latitudes, we would expect the problem to be worse. Active temperature
control using flowing water around the pots could work well from the physical point
of view, though from the practical point of view such a system would be very
challenging to install at most treeline sites.
Assuming that root-zone temperatures can be controlled, the question remains
which temperature regimes would allow an informative comparison between
canopy-level seedlings and ground-level seedlings. Seedlings can be used as models
to study tree growth in two distinct ways: as a model for tree tissue in general or as a
small-version model of a tall tree. These approaches require explicit assumptions
about temperature effects and require different temperature regimes. Using seedlings
as a model for tree tissue in general seems reasonable for some questions because
roots and shoots have similar temperature thresholds for growth (references in Rossi
et al. 2008) and both need to be warm enough for a plant to grow (Körner and Hoch
2006). For such a model, roots and shoots should ideally experience the same
temperature regimes. Therefore, root-zone temperatures in suspended and ground-
Methodological Approaches
97
level pots should be regulated to follow the respective air temperatures (and
irrigation adjusted accordingly).
The alternative, using seedlings as a small-version model for a tree, assumes
that whole-plant physiology is affected differently by root than by shoot temperature
(Larigauderie et al. 1991; Weih and Karlsson 2001) and that this is similar for
seedlings and adult trees. Such a small-version tree model is what Fajardo and Piper
(2014) had in mind for their experiment, which “aimed to mimic the low temperature
effects on meristematic shoot and root tissues of a taller tree and, eventually, the
temperature effects on the tree’s C balance as a whole”. For such a model,
temperatures in the pots should follow the respective root-zone temperatures in
alpine vegetation (ground-level pots) and below trees (suspended pots). Thus, only
suspended pots would need regulation (assuming the ground-level pots are buried
in the ground and temperatures there are naturally representative), and this
regulation could be based on measured soil temperatures under a nearby tree
canopy. Alternatively, if the question is focused on aboveground temperatures only,
root-zone temperatures could be similar in both treatments, i.e., could be regulated in
suspended pots to follow the temperature in the ground-level pots.
A common approach to manipulate temperature for seedlings at treeline is
warming, either passively by using transparent roofs (Germino and Smith 1999) or
open-top chambers (e.g. Danby and Hik 2007; Xu et al. 2012), or actively by using
infrared lamps (Moyes et al. 2013). All these methods are useful for studying thermal
constraints for seedlings and for mimicking future climate warming, keeping in mind
restrictions inherent to the different types of warming. For mimicking tree canopy
Chapter 4
98
conditions, however, seedlings need to be cooled, which is more challenging than
warming. Other than suspending seedlings in the air, options would be planting
under the tree canopy or with artificial cover (but there will be confounding effects of
shade), planting at higher elevations (but this option implies cooling both day and
night, in contrast to tree canopies, which are warmer at night than low vegetation), or
active cooling. Active cooling could be achieved via radiative cooling (e.g., using
peltier or other electric cooling elements near the plants) or via convective cooling
(using ventilators, supplying air from outside the soil–alpine vegetation boundary
layer). As an imitation of convective cooling (and at night: warming) of the tree
canopy, the ventilator option seems by far the better choice. This approach, on a
larger scale using wind machines, is commonly used in horticulture to prevent
radiation frosts (Perry 1998). As in the suspended-pot experiment presented by
Fajardo and Piper (2014) , such a setup would have to control for soil moisture
differences. Belowground temperatures could either be left to equilibrate with the air
(the easier option), or aboveground ventilation could be accompanied by
temperature regulation of the root zone to mimic soil conditions below a tree canopy,
e.g., using electric temperature elements as is sometimes used, though again usually
for heating only, in climate-change experiments (Melillo et al. 2002).
In all of these setups, apart from accounting for potential microclimatic
artifacts, the assumed role of the seedlings as models for tree growth should be made
explicit and validated before translating the results obtained with the seedling
models to adult trees.
Methodological Approaches
99
Comparing taxa regarding canopy–microclimate effects on tree growth is only
one side of the story, however. To understand fully the role of microclimate in
limiting tree growth, differences in canopy microclimates and tissue temperatures
due to taxon-specific leaf and crown morphologies, transpiration rates and albedo
also need to be considered (Leuzinger and Körner 2007). Data addressing such
differences for treeline trees are rare (Körner 2012), though they would be relatively
easy to obtain using data loggers. Such data would be another valuable contribution
toward understanding the mechanisms of treeline formation for different taxa and
functional tree types.
ALTERNATIVE TREELINE-FORMING MECHANISMS IN NOTHOFAGUS AND OTHER
GENERA
Nothofagus treelines have long been regarded as unusual, occurring at higher
temperatures than most northern-hemisphere treelines, which was explained by
genus-specific limitations (Körner and Paulsen 2004; Wardle 2008). However, several
recent studies suggest that mean temperatures in the growing season at these
treelines are actually quite comparable to those at other treelines (Mark et al. 2008;
Cieraad et al. 2014; Fajardo and Piper 2014). Another argument against genus-specific
limitations is presented in the study discussed here (Fajardo and Piper 2014), where
Pinus contorta seedlings did not outperform Nothofagus pumilio at 50 m above the
treeline. The authors argue that because of the lower mean temperatures in the
suspended pots, P. contorta should not outperform N. pumilio up to 330 m above the
treeline, although this argument disregards temperature extremes. However, in an
experiment in New Zealand started in the 1960s, Pinus contorta and other exotic
Chapter 4
100
conifers developed large stems up to 300 m above the local treeline, while saplings of
the native treeline species Fuscospora cliffortioides (Nothofagus solandri var.
cliffortioides), though surviving ca. 150 m above the treeline, have still not emerged
from the boxes that protected them as seedlings (Fig. 4.1). Clearly, Pinus is
outperforming this close relative of Nothofagus in this case (Wardle 1985). So the
question remains: do the Nothofagaceae form treelines for different reasons than
most other treeline tree families?
Fig. 4.1 A. Overview of the treeline tree-establishment experiment installed by Peter
Wardle in the 1960s in the Craigieburn Range, New Zealand (photos taken in March
2009 by M. Y. Bader). View down toward the treeline from the experimental garden
at 1450 m, ca. 150 m above the current treeline. Exotic tree species, many of which
have grown well at this elevation, have been cut down to prevent them from
spreading into the native ecosystem. Note the shrubby Fuscospora cliffortioides (≡
Nothofagus solandri var. cliffortioides), the local treeline tree species that appears to
have survived or established in the shelter of the now cut-down exotic trees. B.
Stump of Pinus contorta at 1450 m. The box is the remnant of an experimental 73%
shade treatment. The rosette plant in the box is the alpine species Aciphylla cf. aurea ,
Methodological Approaches
101
unrelated to the experiment. C. Shade box with barely grown F. cliffortioides
apparently unable to escape these sheltered conditions (at 1450 m) (experiment and
results described in Wardle, 1985 , 2008).
Most treelines composed of Nothofagaceae are very abrupt boundaries from
tall, closed forest to low alpine vegetation, suggesting that limitations to
establishment outside the forest rather than limited growth determine the position of
these treelines (Harsch and Bader 2011). Above the natural treeline, Fuscospora
cliffortioides seedlings depend on shade and/or frost protection, which also suggests
that tree establishment and treeline advance are not limited by low growing
temperatures alone but by the interaction with stressors like frost, excess radiation,
and wind (Wardle 1993, 2008). Such stressors cannot offer a universal explanation for
treeline formation, as they vary strongly among treelines and can occur at any
elevation (Körner 2012). In the absence of such stressors and in the case of resistant
species, treeline elevations can be controlled by low temperature limitations to
growth, either via source or sink limitation. In all other cases, regional peculiarities
and “taxon-specific” treelines emerge (Harsch and Bader 2011). As these are arguably
the rule rather than the exception (e.g. Piper et al. 2006; Wardle 2008; Holtmeier 2009;
Malanson et al. 2011), we embrace the recommendation of Fajardo and Piper (2014)
to keep developing a universal theory for treeline formation including variation
between tree taxa and regional climates. Experiments to this end should address low-
temperature limitations to growth as well as interactions with other climatic factors
in all tree life stages. Preferably, such experiments should include several members of
different functional tree types to allow generalizations beyond taxon-specificity.
Chapter 4
102
CONCLUSION
The presented experimental setup in Fajardo and Piper (2014) is a low-technology
approach for studying a fundamental question in functional plant biology: what
causes treelines? Though elegant, the experiment suffers from a potentially
disqualifying practical problem: the strong fluctuations in soil temperature (and most
probably soil moisture) in the suspended pots. As long as the effect of such
fluctuations is unknown, they cannot be ignored and should be experimentally
controlled. A promising alternative is to keep seedlings at ground level and ventilate
them to mimic the atmospheric coupling found in tree canopies. These solutions
require an infrastructure that is not usually available at treeline, though solar panels
and local water sources could allow these setups even in remote sites. Apart from
technical solutions, we discussed conceptual solutions: how can we interpret the data
given the temperature and moisture fluctuations? To do this, soil moisture data
would be needed as well as a much better understanding of the effects of
temperature fluctuations on tree growth. At this stage, an unambiguous
interpretation of the results of Fajardo and Piper (2014) seems impossible. Nothofagus
pumilio seedlings are carbon-limited before they are growth-limited at the
temperature and moisture conditions in the suspended pots, but it is unresolved
whether the limitations are really due to the lower mean temperature or (1) to the
strongly fluctuating root-zone temperature regime, or (2) to moisture stress.
Although their experiment did not allow unambiguous conclusions about the causes
of alpine treeline formation, it provides excellent food for thought on further
experiments toward this goal.
Methodological Approaches
103
APPENDIX
Fig. S4.1 A. Vertically adjustable planting-tube designed to test low-temperature
limitations of tree growth and vegetation microclimate on seedling performance by
growing same-aged tree seedlings at different heights within and above high alpine
grassland vegetation. An inner tube (3 cm planting core) is wrapped in 2 cm-thick
pipe insulation material and suspended in an outer reflective aluminium tube,
leaving a 1cm aeration layer through which air can circulate from the aeration holes
(top) via the open bottom. Note that the soil surface is covered with moss for
comparability of temperature measurements with the moss-covered ground. B.
Comparative temperature measurements between the soil in the planting tube at 100
cm height, the soil in the ground and the air at 100 cm height. Temperature sensors
were inserted 5 cm into the soil for soil temperature measurements and protected
from solar radiation by an aerated shield (not shown). Solid black line: soil
temperature in the ground, dotted black line: soil temperature in the planting tubes,
dotted grey line: air temperature.
Date
Tem
pe
ratu
re°C
A B
105
CHAPTER 5
SYNTHESIS
On the one hand, the studies presented in this thesis investigated experimentally
how abiotic and biotic environmental conditions affect the two earliest regeneration
stages of trees at the alpine treeline. Soil moisture was, in addition to temperature, a
critical factor for both germination and subsequent seedling establishment. The
interaction with neighbouring herbaceous vegetation affected only the seedling life-
stage and was principally negative, but could seasonally switch to facilitation for
evergreen tree species. While general patterns emerged for responses to both
microclimate and plant-plant-interactions, species-specific effects were more strongly
represented. On the other hand, we pointed out in a commentary the difficulty of
finding resourceful, novel approaches to test hypotheses and gain mechanistic
insights to explain a major vegetation boundary with an array of potentially limiting
and interacting factors, and explored the conceivable possibilities.
Chapter 5
106
LOCAL AND INTRINSIC FACTORS DRIVING THE REGENERATION RESPONSE
It is still under debate if an overarching explanation for treeline formation can be
found on a global scale, or if individual treelines, although ultimately limited by heat
deficiency, are each constrained by different sets and interactions of site- and taxon-
specific factors. The regeneration of treeline trees as a potential life-history bottleneck
plays an important role in this debate, and the presented studies show that early
establishment is importantly affected by local environmental factors such as
microclimate and biotic interactions as well as intrinsic factors such as life-stage
dependencies and tree species ecology.
In terms of environmental factors, we confirmed the well-established
importance of temperature for treelines in general (Rolland et al. 1998; Körner and
Paulsen 2004) and for early regeneration stages in particular (Germino and Smith
1999; Smith et al. 2003) by showing the expectedly negative impact of decreasing
temperatures on germination and early seedling establishment. We further found,
however, that temperature effects could not be decoupled from the effect of water
availability and that depending on the species, the importance of the latter
predominated (Chapter 2). While varying intensities of vegetation cover significantly
altered the microclimate regarding both temperature and water availability, the
results from this experiment suggest that direct negative and especially belowground
plant-plant interactions, i.e. competition, between tree seedlings and herbaceous
alpine vegetation prevail (Chapter 3). Both studies reveal a strong dependency of
early tree life-stages on soil resources such as water or nutrients, which can be linked
i) for germination to the requirement of high moisture conditions as external cue and
Synthesis
107
driver of physiological processes (Baskin and Baskin 2001) and ii) for seedlings to
their characteristically small size connected with only small amounts of productive
and storage tissue, simple rooting system and a consequently low resource-uptake or
storage capacity (Wang and Zwiazek 1999; Johnson et al. 2011).
The responses of the studied tree species to microclimatic manipulations, both
positive and negative, were generally consistent over the earliest life-stages of
germination and subsequent seedling establishment (Chapter 2). This concordant
pattern has important implications for the regeneration success: it reduces
restrictions caused by seed-seedling conflicts regarding the environmental factors
required for establishment, thereby mitigating the effect of the environmental filters
imposed on each life-stage and supporting that a safe site (sensu Harper 1977) for the
seed is also safe for the seedling (Schupp 1995). However, in a relatively stable
environment it can also be expected that limiting factors have accordingly a higher
impact on a regeneration process with concordant life-stage responses, whereas an
irregularly occurring stress factor may be temporarily decoupled from brief
susceptible stages such as germination (Shen et al. 2014). The fact that we did not
find concordant patterns over both early life-stages for interactions with the alpine
herbaceous vegetation, i.e. no effect on germination and important impacts on the
seedling stage (Chapter 3), can be related to two important conditions: i) due to the
reserves contained in the seed, the germination success is relatively independent
from external resources and thus much less susceptible to competition, which was
the dominant interaction affecting seedling performance, and ii) the abiotic
conditions, although modified by varying intensities of vegetation cover, were
apparently not altered beyond the range required by this earliest life-stage. However,
Chapter 5
108
since the presented studies were conducted over a relatively short period of time
with measurements taken primarily in the less stressful growing season and
germination was monitored from seeds sown directly in the ground, it remains
unclear how permanent versus variable environmental factors impact the overall
regeneration success and how vegetation cover affects the germination of free-falling
seeds.
The previous paragraphs show that general patterns emerged in both studies
investigating the impact of potential environmental limitations on early regeneration
responses, but nevertheless, species- or taxon-specific effects predominated. In terms
of abiotic environmental factors, the relative importance of the manipulated
microclimate variables temperature and moisture as well as the direction of effects
were highly idiosyncratic (Chapter 2). These patterns on the one hand fittingly
represent the characteristics of the species’ ecology: for example, the high importance
of moisture in comparison to temperature for both early establishment stages of L.
decidua, although as young seedling still evergreen, may reflect the low water use
efficiency of later deciduous life-stages (Matyssek 1986). On the other hand they may
relate specific limitations to the species’ geographical distribution, as germination in
P. uncinata, a species forming southern treelines such as in the Southern European
Alps and in the Pyrenees (Rameau et al. 1993; Batllori and Camarero 2009), was
particularly sensitive to low temperatures. In terms of interactions with the alpine
herbaceous vegetation, tree-species-specific negative responses were principally
caused by combinations of competition and indirect vegetation effects such as
microclimate and pathogens, which might decrease seedling performance through
physiological and morphological weakening followed by increased disease
Synthesis
109
susceptibility (Seiwa et al. 2008; Yamazaki et al. 2009). The conifer species however,
all evergreen in the early seedling stage, additionally showed the potential to
seasonally switch the interaction with high herbaceous vegetation from competition
to facilitation: between the dieback of the vegetation and the development of a
permanent snow cover they could benefit from an extended productive season in the
shelter of vegetation remains, thereby compensating for carbon losses due to
competition during the summer or even obtaining a positive carbon balance at the
end of the winter (Chapter 3). While previously shown for herbaceous alpine plant
communities (Kikvidze et al. 2006; Venn et al. 2009), such a seasonal switch in species
interactions has to our knowledge never been demonstrated for treeline trees before.
Thus, this taxon-specific response highlights the dynamic potential of biotic
interactions at a distribution boundary even over short periods of time. It remains
however still unclear if this seasonal benefit results in long-term net facilitation, or if
competitive effects of the herbaceous vegetation prevail.
The important and variable impact of the site-specific and intrinsic factors on
two critical stages of early establishment in our studies suggests a frequent restriction
of treeline tree regeneration before a temperature limit for growth is reached.
Germination and early seedling establishment are not sufficient, but indispensable
first steps towards the establishment of a mature tree. Although several subsequent
life-stages have to be completed before a tree contributes to the current tree line, or in
case of establishment in the alpine tundra, before a treeline has advanced upslope,
restrictions during these earliest stages can importantly influence the shape and
dynamic of the resulting treeline (Harsch and Bader 2011). Consequently, the
findings presented in this thesis may offer an explanation of observed treeline
Chapter 5
110
patterns, dynamics, and their local variation in the context of site-, life-stage and
species-specific factors. Moreover, they highlight the concurrently causative impact
of such “local” factors in addition to more global drivers such as temperature and the
resulting important implications for the development of predictive models.
OUTLOOK AND FUTURE RESEARCH NEEDS
Observational studies and simple experiments, i.e. comprising only one or few
variables of interest, are the essential foundation and a logical starting point in the
attempt to understand and then be able to predict a natural phenomenon.
Subsequently, the complexity and variability encountered in real landscapes
demands for more integrative approaches to gain a deeper mechanistic insight. In the
debate about treeline formation, this need is reflected in a recent increase of such
integrative studies on treeline tree regeneration as an important driver of treeline
formation. For example, tree-establishment responses are compared at different
temporal scales and life-stages (Bansal and Germino 2009), in multiple species
(Zurbriggen et al. 2013) or to more than one experimentally manipulated
microclimatic factor (Moyes et al. 2013) and combinations of climatic and non-
climatic factors (Grau et al. 2012).
In this spirit, the first two studies of this thesis (Chapter 2 and 3) were
conceived to combine several of those important aspects acting as abiotic and biotic
local factors on treelines in terms of tree regeneration. Thereby, we could present
new insights on the complexity of early establishment responses of treeline trees such
as the idiosyncratic effects of interacting microclimate variables or seasonal changes
of the interaction between tree seedlings and herbaceous alpine vegetation.
Synthesis
111
In the next step, it would be desirable to combine the experimental
manipulation of microclimatic factors and the structure of vegetation cover in a
single field experiment, since it can be expected that the response of both tree
seedlings and alpine vegetation to varying intensities of e.g. temperature and
moisture feeds back to their interaction (positive or negative) and its importance
(Blois et al. 2013). Establishing such an experiment at a site with more stressful
conditions than in our studies (e.g. more summer frost events at higher elevation or
stronger temperature fluctuations and solar radiation loads at tropical alpine
treelines) would allow for longer gradients of abiotic factors, thereby more likely
reaching the tolerance limits of more resistant species (such as P. abies in our studies)
and more appropriately testing hypotheses of plant-plant-interactions such as the
“stress-gradient-hypothesis” with regard to treeline tree seedlings (Bertness and
Callaway 1994). An additional and urgently needed feature to include is an
experimental duration of several years, starting with the initial establishment after
germination and employing a high temporal resolution of measurements. This
would allow to clarify the long-term impact of the seasonal facilitation effects of
alpine herbaceous vegetation we documented and to distinguish between
permanently and irregularly affecting environmental conditions. Finally, it is of
unchanged importance to include different life-stages and tree species to account for
varying treeline regeneration responses caused by such intrinsic factors.
While a project of these dimensions will arguably be difficult to implement
due to the required work load alone, few or no infrastructure and the difficult
accessibility of many treeline sites will further complicate an intended realization.
However, there is a great potential in collaborative initiatives including different
Chapter 5
112
research groups and institutions as known from long-term biodiversity experiments
such as the “Jena-Experiment” (Roscher et al. 2004), which offer important
advantages to large-scale experiments in terms of workforce, temporal stability and
financial aspects. Such an approach should therefore, ideally in the vicinity of an
associated alpine research station (e.g. Lautaret Pass, France; Obergurgl, Austria), be
considered also in the field of treeline research, especially with regard to the
increasing interest in this conspicuous vegetation boundary in the light of climate
change.
A first initiative in this direction was developed with the GTREE-network
(Global Treeline Range Expansion Experiment, Brown et al. 2013), an ongoing joint
effort of different treeline research-groups all over the world, in which we also
participate with a site at a natural treeline close to the Alpine Research Station Joseph
Fourier, Lautaret Pass, France. The aim of GTREE is to disentangle the impact of two
important precursors of treeline range expansion, namely seed availability and
suitability of substrate for establishment and survival, by implementing comparable
seeding experiments at already established treeline research sites worldwide. While
each participant is only responsible for a modest data collection at relatively low
costs, the aggregated, global dataset will allow a wide latitudinal and geographical
comparison of treeline tree regeneration responses. This approach not only allows to
investigate the generality of regeneration limitations on a global scale, but is
subsequently also intended to focus on long-term recruitment patterns and to
perform more region-specific analyses. Thus, studies of this scope could bring
unparalleled contributions for the understanding of the interdependencies that limit
treeline tree regeneration as an important driver of treeline formation.
Synthesis
113
Sometimes the difficulty of conceiving more integrative studies in order to test
hypotheses of treeline formation is not of organizational but of methodological
nature. A promising experimental design evaluated by us in a trial and used in a
study by Fajardo and Piper (2014) aimed at investigating the effect of atmospheric
coupling, i.e. tree-canopy temperature conditions, and vegetation microclimate on
tree growth by growing tree seedlings of different species in suspended or vertically
adjustable pots. The advantage of this approach is that low-temperature limitations
of tree growth ("growth limitation hypothesis", Körner 1998) can be studied in small,
easily replicable units of tree tissue independent from ontogenetic or plant-size
effects and at different heights within and above the alpine vegetation layer.
However, the occurrence of important confounding factors, i.e. high soil temperature
fluctuations and potential moisture deficits due to the comparatively small soil
volume (in spite of reflective outer material, insulation and ventilation layers), does
not allow an unbiased interpretation of results obtained from such a design (Chapter
4). Especially a lack of knowledge concerning the effect of variable versus constant
temperatures on plants in general and tree seedlings in particular conceals the
potential impact of such an artifact, thus urgently calling for further studies
investigating this question. While the presented experimental design may have been
proven unsuitable for the task at hand, alternatives such as the use of active cooling
should be pursued to study the important implications of a low-temperature growth
limit for treeline formation.
114
REFERENCES
Akhalkatsi M, Abdaladze O, Nakhutsrishvili G, Smith WK. 2006. Facilitation of
seedling microsites by Rhododendron caucaskum extends the Betula litwinowii Alpine
treeline, Caucasus Mountains, Republic of Georgia. Arctic Antarctic and Alpine
Research 38: 481–488.
Axelsson PE, Lundmark T, Högberg P, Nordin A. 2014. Belowground competition
directs spatial patterns of seedling growth in boreal pine forests in fennoscandia.
Forests 5: 2106–2121.
Bader MY. 2007. Tropical alpine treelines: how ecological processes control vegetation
patterning and dynamics. PhD Thesis, Wageningen University, Wageningen, The
Netherlands.
Bader MY, Van Geloof I, Rietkerk M. 2007. High solar radiation hinders tree
regeneration above the alpine treeline in northern Ecuador. Plant Ecology 191: 33–45.
Ball MC, Egerton JJG, Leuning R, Cunningham RB. 1997. Microclimate above grass
adversely affects spring growth of seedling snow gum (Eucalyptus pauciflora). Plant,
Cell and Environment 20: 155–166.
Ball MC, Hodges VS, Laughlin GP. 1991. Cold-induced photoinhibition limits
regeneration of snow gum at a tree-line. Functional Ecology 5: 663–668.
Bansal S, Germino MJ. 2009. Temporal variation of nonstructural carbohydrates in
montane conifers: similarities and differences among developmental stages, species
and environmental conditions. Tree physiology 29: 559–568.
Bansal S, Reinhardt K, Germino MJ. 2011. Linking carbon balance to establishment
patterns: Comparison of whitebark pine and Engelmann spruce seedlings along an
herb cover exposure gradient at treeline. Plant Ecology 212: 219–228.
Barbeito I, Dawes MA, Rixen C, Senn J, Bebi P. 2012. Factors driving mortality and
growth at treeline: a 30-year experiment of 92 000 conifers. Ecology 93: 389–401.
115
Barber VA, Juday GP, Finney BP. 2000. Reduced growth of Alaskan white spruce in
the twentieth century from temperature-induced drought stress. Nature 405: 668–673.
Barclay AM, Crawford RMM. 1984. Seedling emergence in the rowan (Sorbus
aucuparia) from an altitudinal gradient. Journal of Ecology 72: 627–636.
Barton AM. 1993. Factors controlling plant distributions: drought, competition, and
fire in montane pines in Arizona. Ecological Monographs 63: 367–397.
Baskin CC, Baskin JM. 2001. Seeds: Ecology, biogeography, and evolution of dormancy
and germination. San Diego: Academic Press.
Batllori E, Camarero JJ. 2009. Seedling recruitment, survival and facilitation in alpine
Pinus uncinata tree line ecotones. Implications and potential responses to climate
warming. Global Ecology and Biogeography 18: 460–472.
Batllori E, Camarero JJ, Gutiérrez E. 2010. Current regeneration patterns at the tree
line in the Pyrenees indicate similar recruitment processes irrespective of the past
disturbance regime. Journal of Biogeography 37: 1938–1950.
Bendix J, Rafiqpoor MD. 2001. Studies on the thermal conditions of soils at the
upper tree line in the Paramo of Papallacta (Estern Cordillera of Ecuador). Erdkunde
55: 257–276.
Bertness MD, Callaway R. 1994. Positive interactions in communities. Trends in
Ecology and Evolution 9: 187–191.
Blois JL, Zarnetske PL, Fitzpatrick MC, Finnegan S. 2013. Climate change and the
past, present, and future of biotic interactions. Science 341: 499–504.
Bloor JMG, Leadley PW, Barthes L. 2008. Responses of Fraxinus excelsior seedlings to
grass-induced above- and below-ground competition. Plant Ecology 194: 293–304.
Bolli JC, Rigling A, Bugmann H. 2007. The influence of changes in climate and land-
use on regeneration dynamics of Norway spruce at the treeline in the Swiss Alps.
Silva Fennica 41: 55–70.
Brändli U-B. 1998. Die häufigsten Waldbäume der Schweiz - Ergebnisse aus dem
116
Landesforstinventar 1983-85: Verbreitung, Standort und Haufigkeit von 30 Baumarten.
Birmensdorf, Switzerland: Eidgenössische Forschungsanstalt tu r Wald, Schnee und
Landschaft.
Brockmann-Jerosch H. 1919. Baumgrenze und Klimacharakter. Schweizerische
Naturforschende Gesellschaft. Pflanzengeographische Kommission. Beiträge zur
geobotanischen Landesaufnahme der Schweiz. Zürich: Rascher.
Brown CD, Johnstone JF, Mamet SD, Trant AJ. 2013. Global Treeline Range
Expansion Experiment field protocols. www.treelineresearch.com.
Brown CD, Vellend M. 2014. Non-climatic constraints on upper elevational plant
range expansion under climate change. Proceedings of the Royal Society-Biological
Sciences 281: 20141779.
Cairns DM, Moen J. 2004. Herbivory influences tree lines. Journal of Ecology 92: 1019–
1024.
Callaway RM, Brooker RW, Choler P, Kidvidze Z, Lortiek CJ, Michalet R, Paolini
L, Pugnaire FI, Newingham B, Aschehoug ET, Armasq C, Kikodze D, Cook BJ.
2002. Positive interactions among alpine plants increase with stress. Nature 417: 844–
848.
Carlsson BA, Callaghan T V. 1991. Positive plant interactions in tundra vegetation
and the importance of shelter. Journal of Ecology, 79: 973–983.
Case BS, Duncan RP. 2014. A novel framework for disentangling the scale-
dependent influences of abiotic factors on alpine treeline position. Ecography 37: 838–
851.
Casper BB, Jackson RB. 1997. Plant competition underground. Annual Review of
Ecology and Systematics 28: 545–570.
Castanha C, Torn MS, Germino MJ, Weibel B, Kueppers LM. 2012. Conifer
seedling recruitment across a gradient from forest to alpine tundra: effects of species,
provenance, and site. Plant Ecology & Diversity 6: 1–12.
117
Cerasoli S, Wertin T, McGuire MA, Rodrigues A, Aubrey DP, Pereira JS, Teskey
RO. 2014. Poplar saplings exposed to recurring temperature shifts of different
amplitude exhibit differences in leaf gas exchange and growth despite equal mean
temperature. AoB PLANTS 6.
Chen I-C, Hill JK, Ohlemüller R, Roy DB, Thomas CD. 2011. Rapid range shifts of
species associated with high levels of climate warming. Science 333: 1024–1026.
Choler P, Michalet R, Callaway RM. 2001. Facilitation and competition on gradients
in alpine plant communities. Ecology 82: 3295–3308.
Cieraad E, McGlone MS, Huntley B. 2014. Southern Hemisphere temperate tree
lines are not climatically depressed. Journal of Biogeography 41: 1456–1466.
Close DC, Beadle CL, Brown PH, Holz GK. 2000. Cold-induced photoinhibition
affects establishment of Eucalyptus nitens (Deane and Maiden) Maiden and Eucalyptus
globulus Labill. Trees - Structure and Function 15: 32–41.
Coomes DA, Grubb PJ. 2000. Impacts of root competition in forests and woodlands:
A theoretical framework and review of experiments. Ecological Monographs 70: 171–
207.
Cuevas JG. 2000. Tree recruitment at the Nothofagus pumilio alpine timberline in
Tierra del Fuego, Chile. Journal of Ecology 88: 840–855.
Cui M, Smith WK. 1991. Photosynthesis, water relations and mortality in Abies-
lasiocarpa seedlings during natural establishment. Tree Physiology 8: 37–46.
Danby RK, Hik DS. 2007. Responses of white spruce (Picea glauca) to experimental
warming at a subarctic alpine treeline. Global Change Biology 13: 437–451.
Daniels LD, Veblen TT. 2004. Spatiotemporal influences of climate on altitudinal
treeline in northern Patagonia. Ecology 85: 1284–1296.
Däniker A. 1923. Biologische Studien über Wald- und Baumgrenze, insbesondere
über die klimatischen Ursachen und deren Zusammenhänge. Vierteljahresschrift der
Naturforschenden Gesellschaft Zürich 68.
118
Daubenmire R. 1954. Alpine timberlines in the Americas and their interpretation.
Butler University Botanical Studies 11: 119–136.
Davis AJ, Jenkinson LS, Lawton JH, Shorrocks B, Wood S. 1998. Making mistakes
when predicting shifts in species range in response to global warming. Nature 391:
783–786.
Dawes MA, Hagedorn F, Handa IT, Streit K, Ekblad A, Rixen C, Körner C,
Hättenschwiler S. 2013. An alpine treeline in a carbon dioxide-rich world: Synthesis
of a nine-year free-air carbon dioxide enrichment study. Oecologia 171: 623–637.
Day ME, Greenwood MS, White AS. 2001. Age-related changes in foliar
morphology and physiology in red spruce and their influence on declining
photosynthetic rates and productivity with tree age. Tree physiology 21: 1195–1204.
Didier L. 2001. Invasion patterns of European larch and Swiss stone pine in
subalpine pastures in the French Alps. Forest Ecology and Management 145: 67–77.
Donohue K, Rubio de Casas R, Burghardt L, Kovach K, Willis CG. 2010.
Germination, postgermination adaptation, and species ecological ranges. Annual
Review of Ecology, Evolution, and Systematics 41: 293–319.
Dufour-Tremblay G, de Vriendt L, Lévesque E, Boudreau S. 2012. The importance
of ecological constraints on the control of multi-species treeline dynamics in Eastern
Nunavik, Québec. American Journal of Botany 99: 1638–1646.
Dullinger S, Dirnböck T, Grabherr G. 2003. Patterns of shrub invasion into high
mountain grasslands of the northern calcareous Alps, Austria. Arctic, Antarctic, and
Alpine Research 35: 434–441.
Ellis RC, Pennington PI. 1992. Factors affecting the growth of Eucalyptus delegatensis
seedlings in inhibitory forest and grassland soils. Plant and Soil 145: 93–105.
Fajardo A, Piper FI. 2014. An experimental approach to explain the Southern andes
elevational treeline. American Journal of Botany 101: 788–795.
Fajardo A, Piper FI, Cavieres LA. 2011. Distinguishing local from global climate
119
influences in the variation of carbon status with altitude in a tree line species. Global
Ecology and Biogeography 20: 307–318.
Fajardo A, Piper FI, Pfund L, Körner C, Hoch G. 2012. Variation of mobile carbon
reserves in trees at the alpine treeline ecotone is under environmental control. New
Phytologist 195: 794–802.
Ferrar P, Cochrane P, Slatyer R. 1988. Factors influencing germination and
establishment of Eucalyptus pauciflora near the alpine tree line. Tree Physiology 4: 27–
43.
Germino MJ, Smith WK. 1999. Sky exposure, crown architecture, and low-
temperature photoinhibition in conifer seedlings at alpine treeline. Plant Cell and
Environment 22: 407–415.
Germino MJ, Smith WK, Resor AC. 2002. Conifer seedling distribution and survival
in an alpine-treeline ecotone. Plant Ecology 162: 157–168.
González de Andrés E, Camarero JJ, Büntgen U. 2015. Complex climate constraints
of upper treeline formation in the Pyrenees. Trees: 941–952.
Grace J. 2002. Impacts of climate change on the tree line. Annals of Botany 90: 537–544.
Grau O, Ninot JM, Blanco-Moreno JM, van Logtestijn RSP, Cornelissen JHC,
Callaghan T V. 2012. Shrub-tree interactions and environmental changes drive
treeline dynamics in the Subarctic. Oikos 121: 1680–1690.
Greenwood S, Chen J-C, Chen C-T, Jump AS. 2015. Temperature and sheltering
determine patterns of seedling establishment in an advancing subtropical treeline.
Journal of Vegetation Science 26: 711–721.
Grubb PJ. 1977. The maintenance of species-richness in plant communities: the
importance of the regeneration niche. Biological Reviews 52: 107–145.
Harper JL. 1977. Population biology of plants. London; New York: Academic Press.
Harsch MA, Bader MY. 2011. Treeline form - a potential key to understanding
treeline dynamics. Global Ecology and Biogeography 20: 582–596.
120
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: 1040–
1049.
Hasselquist N, Germino MJ, McGonigle T, Smith WK. 2005. Variability of
Cenococcum colonization and its ecophysiological significance for young conifers at
alpine-treeline. New Phytologist 165: 867–873.
Hättenschwiler S, Körner C. 1995. Responses to recent climatewarming of Pinus-
sylvestris and Pinus-cembra within their montane transition zone in the Swiss Alps.
Journal of Vegetation Science 6: 357–368.
He Q, Bertness MD, Altieri AH. 2013. Global shifts towards positive species
interactions with increasing environmental stress. Ecology Letters 16: 695–706.
Heenan PB, Smissen RD. 2013. Revised circumscription of Nothofagus and
recognition of the segregate genera Fuscospora, Lophozonia, and Trisyngyne
(Nothofagaceae). Phytotaxa 146: 1–31.
Hermes K. 1955. Die Lage der oberen Waldgrenze in den Gebirgen der Erde und ihr
Abstand zur Schneegrenze. Kölner Geographische Arbeiten, Heft 5, Geographisches
Institut, Universität Köln.
Hobbie SE, Shevtsova A, Chapin FS. 1999. Plant responses to species removal and
experimental warming in Alaskan tussock tundra. Oikos 84: 417–434.
Hoch G, Körner C. 2003. The carbon charging of pines at the climatic treeline: A
global comparison. Oecologia 135: 10–21.
Hoch G, Körner C. 2009. Growth and carbon relations of tree line forming conifers at
constant vs. variable low temperatures. Journal of Ecology 97: 57–66.
Hoch G, Körner C. 2012. Global patterns of mobile carbon stores in trees at the high-
elevation tree line. Global Ecology and Biogeography 21: 861–871.
Holtmeier F-K. 1965. Die Waldgrenze im Oberengadin in ihrer physiognomischen und
ökologischen Differenzierung. Dissertation, Rheinische Friedrich-Wilhelms-Universität
121
Bonn, Deutschland.
Holtmeier F-K. 1974. Geoökologische Beobachtungen und Studien an der subarktischen und
alpinen Waldgrenze in vergleichender Sicht (nördliches Fennoskandien/Zentralalpen).
Erdwissenschaftliche Forschung, Bd. 8. Wiesbaden.
Holtmeier FK. 2009. Mountain timberlines - ecology, patchiness, and dynamics.
Dordrecht: Springer Netherlands.
Holtmeier FK, Broll G. 2005. Sensitivity and response of northern hemisphere
altitudinal and polar treelines to environmental change at landscape and local scales.
Global Ecology and Biogeography 14: 395–410.
Hoyle GL, Venn SE, Steadman KJ, Good RB, McAuliffe EJ, Williams ER, Nicotra
AB. 2013. Soil warming increases plant species richness but decreases germination
from the alpine soil seed bank. Global Change Biology 19: 1549–61.
von Humboldt A, Bonpland A. 1805. Essai sur la géographie des plantes. Paris: Chez
Levrault, Schoell et Compagnie.
Hutchinson GE. 1957. Concluding remarks. Cold Spring Harbor Symposia on
Quantitative Biology 22: 415–427.
IPCC. 2013. Summary for Policymakers. Climate Change 2013: The Physical Science
Basis. Contribution of Working Group I to the Fifth Assessment Report of the
Intergovernmental Panel on Climate Chan. Cambridge, New York: Cambridge
University Press, .
Islam MA, Macdonald SE. 2004. Ecophysiological adaptations of black spruce (Picea
mariana) and tamarack (Larix laricina) seedlings to flooding. Trees - Structure and
Function 18: 35–42.
Johnson DM, McCulloh KA, Reinhardt K. 2011. The earliest stages of tree growth:
development, physiology and impacts of microclimate. Size- and Age-Related Changes
in Tree Structure and Function. Dordrecht: Springer Netherlands, 65–87.
Kajimoto T, Onodera H, Ikeda S, Daimaru H, Seki T. 1998. Seedling establishment
122
of subalpine stone pine (Pinus pumila) by nutcracker (Nucifraga) seed dispersal on Mt.
Yumori northern Japan. Arctic and Alpine Research 30: 408–417.
Kikvidze Z, Khetsuriani L, Kikodze D, Callaway RM. 2006. Seasonal shifts in
competition and facilitation in subalpine plant communities of the central Caucasus.
Journal of Vegetation Science 17: 77–82.
Klein JP, Moeschberger ML modifications by JY. 2012. KMsurv: Data sets from
Klein and Moeschberger (1997), Survival Analysis. R package version 0.1-5. URL:
http://cran.r-project.org/package=KMsurv
Kobe RK. 1997. Carbohydrate allocation to storage as a basis of interspecific
variation in sapling survivorship and growth. Oikos 80: 226–233.
Körner C. 1998. A re-assessment of high elevation treeline positions and their
explanation. Oecologia 115: 445–459.
Körner C. 2012. Alpine treelines. Functional ecology of the global high elevation tree limits.
Basel: Springer.
Körner C, Hoch G. 2006. A test of treeline theory on a montane permafrost island.
Arctic Antarctic and Alpine Research 38: 113–119.
Körner C, Paulsen J. 2004. A world-wide study of high altitude treeline
temperatures. Journal of Biogeography 31: 713–732.
Kozlowski T. 1992. Carbohydrate sources and sinks in woody plants. Botanical
Review 58: 107–222.
Krebs C. 1972. Ecology. The experimental analysis of distribution and abundance. New
York: Harper & Row.
Kroiss SJ, HilleRisLambers J. 2015. Recruitment limitation of long-lived conifers:
implications for climate change responses. Ecology 96: 1286–1297.
Kullman L. 2007. Tree line population monitoring of Pinus sylvestris in the Swedish
Scandes, 1973-2005: implications for tree line theory and climate change ecology.
Journal of Ecology 95: 41–52.
123
Larigauderie A, Ellis BA, Mills JN, Kummerow J. 1991. The effect of root and shoot
temperatures on growth of Ceanothus greggii seedlings. Annals of Botany 67: 97–101.
LeBarron RK. 1945. Adjustment of black spruce root systems to increasing depth of
peat. Ecological Society of America 26: 309–311.
Lenz A, Hoch G, Körner C. 2013. Early season temperature controls cambial activity
and total tree ring width at the alpine treeline. Plant Ecology & Diversity 6: 365–375.
Leuzinger S, Körner C. 2007. Tree species diversity affects canopy leaf temperatures
in a mature temperate forest. Agricultural and Forest Meteorology 146: 29–37.
Li M, Hoch G, Körner C. 2002. Source/sink removal affects mobile carbohydrates in
Pinus cembra at the Swiss treeline. Trees - Structure and Function 16: 331–337.
Li Q, Song Y, Li G, Yu P, Wang P, Zhou D. 2015. Grass-legume mixtures impact soil
N, species recruitment, and productivity in temperate steppe grassland. Plant and Soil
394: 271–285.
Lloyd AH, Bunn AG. 2007. Responses of the circumpolar boreal forest to 20th
century climate variability. Environmental Research Letters 2: 045013 (13pp).
Løken A. 1959. [Germination experiments in refrigerator chamber]. Meddedels. f.
Vestlandets forstl. Forsøksstasjon 11: 1–19.
Maher EL, Germino MJ. 2006. Microsite differentiation among conifer species during
seedling establishment at alpine treeline. Ecoscience 13: 334–341.
Maher EL, Germino MJ, Hasselquist NJ. 2005. Interactive effects of tree and herb
cover on survivorship, physiology, and microclimate of conifer seedlings at the
alpine tree-line ecotone. Canadian Journal of Forest Research 35: 567–574.
Malanson GP, Brown DG, Butler DR, Cairns DM, Fagre DB, Walsh SJ. 2009.
Ecotone dynamics: invasibility of alpine tundra by tree species from the subalpine
forest. The changing alpine treeline — The example of Glacier National Park, MT, USA.
Developments in earth surface processes. Vol. 12. Amsterdam: Elsevier, 35–57.
Malanson GP, Resler LM, Bader MY, Holtmeier FK, Butler DR, Weiss DJ, Daniels
124
LD, Fagre DB. 2011. Mountain Treelines: a Roadmap for Research Orientation. Arctic
Antarctic and Alpine Research 43: 167–177.
Marek R. 1910. Waldgrenzstudien in den österreichischen Alpen. Petermanns
geographische Mitteilungen, Ergänzungsband 36, Heft 168. Gotha: Justus Perthes.
Marion G, Henry G, Freckman D, Johnstone J, Jones G, Jones M, Lévesque E,
Molau U, Molgaard P, Parsons A, Svoboda J, Virginia R. 1997. Open-top designs for
manipulating field temperature in high-latitude ecosystems. Global Change Biology 3:
20–32.
Mark AF, Porter S, Piggott JJ, Michel P, Maeglp T, Dickinson KJM. 2008.
Altitudinal patterns of vegetation, flora, life forms, and environments in the alpine
zone of the Fiord Ecological Region, New Zealand. New Zealand Journal of Botany 46:
205–237.
Mattes H. 1994. Coevolutional aspects of stone pines and nutcrackers. International
workshop on subalpine stone pines and their environment: the status of our knowledge. INT-
GTR-309. St. Moritz, Switzerland, September 5-11,1992. UT, USA: Intermountain
Research Station, 31–35.
Matyssek R. 1986. Carbon, water and nitrogen relations in evergreen and deciduous
conifers. Tree physiology 2: 177–187.
McNair JN, Sunkara A, Frobish D. 2012. How to analyse seed germination data
using statistical time-to-event analysis: non-parametric and semi-parametric
methods. Seed Science Research 22: 77–95.
Melillo JM, Steudler P a, Aber JD, Newkirk K, Lux H, Bowles FP, Catricala C,
Magill A, Ahrens T, Morrisseau S. 2002. Soil warming and carbon-cycle feedbacks
to the climate system. Science 298: 2173–2176.
Michaelis P. 1934. Ökologische Studien an der alpinen Baumgrenze V. Osmotischer
Wert und Wassergehalt während des Winters in verschiedenen Höhenlagen. Jahrbuch
für wissenschaftliche Botanik 80: 337–362.
Michalet R, Le Bagousse-Pinguet Y, Maalouf JP, Lortie CJ. 2014. Two alternatives to
125
the stress-gradient hypothesis at the edge of life: The collapse of facilitation and the
switch from facilitation to competition. Journal of Vegetation Science 25: 609–613.
Moir WH, Rochelle SG, Schoettle AW. 1999. Microscale patterns of tree
establishment near upper treeline, Snowy Range, Wyoming, U.S.A. Arctic and Alpine
Research 31: 379–388.
Molina-Montenegro MA, Gallardo-Cerda J, Flores TSM, Atala C. 2012. The trade-
off between cold resistance and growth determines the Nothofagus pumilio treeline.
Plant Ecology 213: 133–142.
Moyes AB, Castanha C, Germino MJ, Kueppers LM. 2013. Warming and the
dependence of limber pine (Pinus flexilis) establishment on summer soil moisture
within and above its current elevation range. Oecologia 171: 271–282.
Nambiar EKS. 1990. Interplay between nutrients, water, root growth and
productivity in young plantations. Forest Ecology and Management 30: 213–232.
Noble DL, Alexander RR. 1977. Environmental factors affecting natural regeneration
of Engelmann spruce in the central Rocky Mountains. Forest Science 23: 420–429.
Ohse B, Jansen F, Wilmking M. 2012. Do limiting factors at Alaskan treelines shift
with climatic regimes? Environmental Research Letters 7: 015505.
Ozenda P. 1988. Die Vegetation der Alpen im europäischen Gebirgsraum. Stuttgart; New
York: Gustav Fischer Verlag.
Palacio S, Hoch G, Sala A, Körner C, Millard P. 2014. Does carbon storage limit tree
growth? New Phytologist 201: 1096–1100.
Parmesan C, Yohe G. 2003. A globally coherent fingerprint of climate change
impacts across natural systems. Nature 421: 37–42.
Paulsen J, Körner C. 2014. A climate-based model to predict potential treeline
position around the globe. Alpine Botany: 1–12.
Perry KB. 1998. Basics of frost and freeze protection for horticultural crops.
HortTechnology 8: 10–15.
126
Petit G, Anfodillo T, Carraro V, Grani F, Carrer M. 2011. Hydraulic constraints limit
height growth in trees at high altitude. New Phytologist 189: 241–252.
Piper FI, Cavieres L a., Reyes-Díaz M, Corcuera LJ. 2006. Carbon sink limitation and
frost tolerance control performance of the tree Kageneckia angustifolia D. Don
(Rosaceae) at the treeline in central Chile. Plant Ecology 185: 29–39.
Piper FI, Fajardo A, Cavieres LA. 2013. Simulated warming does not impair seedling
survival and growth of Nothofagus pumilio in the southern Andes. Perspectives in Plant
Ecology, Evolution and Systematics 15: 97–105.
Pisek A, Cartellieri E. 1939. Zur Kenntnis des Wasserhaushaltes der Pflanzen, IV.
Bäume und Sträucher. Jahrbuch für wissenschaftliche Botanik 88: 22–68.
Popp M, Lied W, Meyer AJ, Richter A, Schiller P, Schwitte H. 1996. Sample
preservation for determination of organic compounds: microwave versus freeze-
drying. Journal of Experimental Botany, 47: 1469–1473.
Pregitzer KS, King JS, Burton AJ, Brown SE. 2000. Responses of tree fine roots to
temperature. New Phytologist 147: 105–115.
Qi Z, Liu H, Wu X, Hao Q. 2015. Climate-driven speedup of alpine treeline forest
growth in the Tianshan Mountains, Northwestern China. Global Change Biology 21:
816–826.
Quentin AG, Pinkard EA, Ryan MG, Tissue DT, Baggett LS, Adams HD, Maillard
P, Marchand J, Landhäusser SM, Lacointe A, Gibon Y, Anderegg WRL, Asao S,
Atkin OK, Bonhomme M, Claye C, Chow PS, Clément-Vidal A, Davies NW,
Dickman LT, Dumbur R, Ellsworth DS, Falk K, Galiano L, Grünzweig JM,
Hartmann H, Hoch G, Hood S, Jones JE, Koike T, Kuhlmann I, Lloret F, Maestro
M, Mansfield SD, Martínez-Vilalta J, Maucourt M, McDowell NG, Moing A,
Muller B, Nebauer SG, Niinemets Ü, Palacio S, Piper F, Raveh E, Richter A,
Rolland G, Rosas T, Saint Joanis B, Sala A, Smith RA, Sterck F, Stinziano JR,
Tobias M, Unda F, Watanabe M, Way DA, Weerasinghe LK, Wild B, Wiley E,
Woodruff DR. 2015. Non-structural carbohydrates in woody plants compared
among laboratories. Tree Physiology 35: 1146–1165.
127
Quétier F, Lavorel S, Thuiller W, Davies I. 2007. Plant-trait-based modeling
assessment of ecosystem-service sensitivity to land-use change. Ecological Applications
17: 2377–2386.
R Core Team. 2015. R: A Language and Environment for Statistical Computing. R
version 3.2.2. URL: http://www.r-project.org/
Rameau J, Dumé G, Mansion D. 1993. Flore forestie re francaise: guide e cologique illustre.
2. Montagnes. Paris: Institut pour le Développement Forestier.
Raspé O, Findlay C, Jacquemart AL. 2000. Sorbus aucuparia L. Journal of Ecology 88:
910–930.
Rolland C, Petitcolas V, Michalet R. 1998. Changes in radial tree growth for Picea
abies, Larix decidua, Pinus cembra and Pinus uncinata near the alpine timberline since
1750. Trees 13: 40–53.
Roscher C, Schumacher J, Baade J. 2004. The role of biodiversity for element cycling
and trophic interactions: an experimental approach in a grassland community. Basic
and Applied Ecology 121: 107–121.
Rossi S, Deslauriers A, Griçar J, Seo JW, Rathgeber CBK, Anfodillo T, Morin H,
Levanic T, Oven P, Jalkanen R. 2008. Critical temperatures for xylogenesis in
conifers of cold climates. Global Ecology and Biogeography 17: 696–707.
Ryser P. 1993. Influences of neighbouring plants on seedling establishment in
limestone grassland. Journal of Vegetation Science 4: 195–202.
Schupp EW. 1995. Seed-seedling conflicts, habitat choice, and patterns of plant
recruitment. American Journal of Botany 82: 399–409.
Scott P, Bentley C, Fayle D, Hansell R. 1987. Crown forms and shoot elongation of
white spruce at the treeline, Churchill, Manitoba, Canada. Arctic and Alpine Research
19: 175–186.
Seiwa K. 1998. Advantages of early germination for growth and survival of seedlings
of Acer mono under different overstorey phenologies in deciduous broad-leaved
128
forests. Journal of Ecology 86: 219–228.
Seiwa K, Miwa Y, Sahashi N, Kanno H, Tomita M, Ueno N, Yamazaki M. 2008.
Pathogen attack and spatial patterns of juvenile mortality and growth in a temperate
tree, Prunus grayana. Canadian Journal of Forest Research 38: 2445–2454.
Senn J. 1999. Tree mortality caused by Gremmeniella abietina in a subalpine
afforestation in the central Alps and its relationship with duration of snow cover.
European Journal of Forest Pathology 29: 65–74.
Shen W, Zhang L, Liu X, Luo T. 2014. Seed-based treeline seedlings are vulnerable to
freezing events in the early growing season under a warmer climate: Evidence from a
reciprocal transplant experiment in the Sergyemla Mountains, southeast Tibet.
Agricultural and Forest Meteorology 187: 83–92.
Shi P, Körner C, Hoch G. 2008. A test of the growth-limitation theory for alpine tree
line formation in evergreen and deciduous taxa of the eastern Himalayas. Functional
Ecology 22: 213–220.
Shiyatov SG, Terent’ev MM, Fomin V V., Zimmermann NE. 2007. Altitudinal and
horizontal shifts of the upper boundaries of open and closed forests in the polar
Urals. Russian Journal of Ecology 38: 223–227.
Sims HP, Mueller-Dombois D. 1968. Effect of grass competition and depth to water
table on height growth of coniferous tree seedlings. Ecology 49: 597–603.
Smith WK, Germino MJ, Hancock TE, Johnson DM. 2003. Another perspective on
altitudinal limits of alpine timberlines. Tree Physiology 23: 1101–1112.
Steiner M. 1935. Winterliches Bioklima und Wasserhaushalt an der alpinen
Waldgrenze. Bioklimatische Beiblätter 2: 57–65.
Stevens GC, Fox JF. 1991. The causes of treeline. Annual Review of Ecology and
Systematics 22: 177–191.
Strand M, Lofvenius M, Bergsten U, Lundmark T, Rosvall O. 2006. Height growth
of planted conifer seedlings in relation to solar radiation and position in Scots pine
129
shelterwood. Forest Ecology and Management 224: 258–265.
Sullivan PF, Sveinbjörnsson B. 2010. Microtopographic control of treeline advance
in Noatak National Preserve, Northwest Alaska. Ecosystems 13: 275–285.
Sutherst RW, Maywald GF, Bourne AS. 2007. Including species interactions in risk
assessments for global change. Global Change Biology 13: 1843–1859.
Tegischer K, Tausz M, Wieser G, Grill D. 2002. Tree- and needle-age-dependent
variations in antioxidants and photoprotective pigments in Norway spruce needles
at the alpine timberline. Tree Physiology 22: 591–596.
Thébault A, Clément JC, Ibanez S, Roy J, Geremia RA, Pérez CA, Buttler A,
Estienne Y, Lavorel S. 2014. Nitrogen limitation and microbial diversity at the
treeline. Oikos 123: 729–740.
Therneau T. 2015. A Package for Survival Analysis in S. version 2.38. R package
version 2.38-1. URL: http://cran.r-project.org/package=survival
Tranquillini W. 1979. Physiological ecology of the alpine timberline. Berlin, Heidelberg:
Springer.
Troll C. 1973. The upper timberlines in different climatic zones. Arctic and Alpine
Research 5: 3–18.
Ulber M, Gugerli F, Bozic G. 2004. EUFORGEN Technical Guidelines for genetic
conservation and use for Swiss stone pine (Pinus cembra). Rome: International Plant
Genetic Resources Institute.
Venn SE, Morgan JW, Green PT. 2009. Do facilitative interactions with neighboring
plants assist the growth of seedlings at high altitudes in alpine Australia? Arctic,
Antarctic, and Alpine Research 41: 381–387.
Vittoz P, Rulence B, Largey T, Freléchoux F. 2008. Effects of climate and land-use
change on the establishment and growth of Cembran pine (Pinus cembra L.) over the
altitudinal treeline ecotone in the central Swiss Alps. Arctic, Antarctic, and Alpine
Research 40: 225–232.
130
Wahlenberg G. 1813. De vegetatione et climate in Helvetia septentrionali inter flumina
Rhenum et Arolam observatis et cum summi septentrionis comparatis tentamen. Zürich:
impensis Orell, Füssli et socc.
Wang Y, Zwiazek JJ. 1999. Effects of early spring photosynthesis on carbohydrate
content, bud flushing and root and shoot growth of Picea glauca bareroot seedlings.
Scandinavian Journal of Forest Research 14: 295–302.
Wardle P. 1974. Alpine timberlines. Arctic and alpine environments. London: Methuen,
371–402.
Wardle P. 1985. New Zealand timberlines. 1. Growth and survival of native and
introduced tree species in the Craigiebum Range, Canterbury. New Zealand Journal of
Botany 23: 219–234.
Wardle P. 1993. Causes of alpine timberline: A review of the hypotheses. Forest
development in cold climates. New York, USA: Plenum Press, 89–103.
Wardle P. 2008. New Zealand forest to alpine transitions in global context. Arctic
Antarctic and Alpine Research 40: 240–249.
Weih M, Karlsson PS. 2001. Growth response of Mountain birch to air and soil
temperature: Is increasing leaf-nitrogen content an acclimation to lower air
temperature? New Phytologist 150: 147–155.
Wick L, Tinner W. 1997. Vegetation changes and timberline fluctuations in the
central Alps as indicators of Holocene climatic oscillations. Arctic and Alpine Research
29: 445–458.
Wieser G, Tausz M. 2007. Trees at their upper limit - Treelife limitation at the alpine
timberline. Dordrecht: Springer Netherlands.
Wiley E, Helliker B. 2012. A re-evaluation of carbon storage in trees lends greater
support for carbon limitation to growth. New Phytologist 195: 285–289.
Wisz MS, Pottier J, Kissling WD, Pellissier L, Lenoir J, Damgaard CF, Dormann
CF, Forchhammer MC, Grytnes JA, Guisan A, Heikkinen RK, Høye TT, Kühn I,
131
Luoto M, Maiorano L, Nilsson MC, Normand S, Öckinger E, Schmidt NM,
Termansen M, Timmermann A, Wardle DA, Aastrup P, Svenning JC. 2013. The role
of biotic interactions in shaping distributions and realised assemblages of species:
implications for species distribution modelling. Biological Reviews 88: 15–30.
Xu Z, Hu T, Zhang Y. 2012. Effects of experimental warming on phenology, growth
and gas exchange of treeline birch (Betula utilis) saplings, Eastern Tibetan Plateau,
China. European Jounal of Forest Research 131: 811–819.
Yamazaki M, Iwamoto S, Seiwa K. 2009. Distance- and density-dependent seedling
mortality caused by several diseases in eight tree species co-occurring in a temperate
forest. Plant Ecology 201: 181–196.
Zurbriggen N, Hättenschwiler S, Frei ES, Hagedorn F, Bebi P. 2013. Performance of
germinating tree seedlings below and above treeline in the Swiss Alps. Plant Ecology
214: 385–396.
Zuur AF, Ieno EN, Walker NJ, Saveliev AA, Smith GM. 2009. Mixed Effects Models
and Extensions in Ecology with R. New York: Springer.
Zywiec M, Holeksa J, Wesolowska M, Szewczyk J, Zwijacz-Kozica T, Kapusta P.
2013. Sorbus aucuparia regeneration in a coarse-grained spruce forest - a landscape
scale. Journal of Vegetation Science 24: 735–743.
Zywiec M, Ledwoń M. 2008. Spatial and temporal patterns of rowan (Sorbus
aucuparia L.) regeneration in West Carpathian subalpine spruce forest. Plant Ecology
194: 283–291.
132
SUMMARY
Alpine treelines are important and conspicuous vegetation boundaries on high
mountains, which have caught the interest of naturalists and researchers for many
centuries. They are ultimately limited by heat deficiency and thus expected to
advance with the currently warming climate. However, treeline responses are locally
contrasting due to other climatic factors, tree species ecology and life-stage-
dependent responses. Since regeneration of trees is a prerequisite for treelines to
remain stable or to move upslope, especially the critical earliest life-stages of
germination and seedling establishment may present a major life-history bottleneck
for treeline tree populations. Studying the limitations of tree recruitment is therefore
an important step to gain a more mechanistic understanding of treeline formation
and to reliably predict their future dynamics.
The first two chapters of this thesis present studies investigating the effect of abiotic
and biotic environmental factors, respectively, on the two critical regeneration stages
of germination and early seedling establishment in five important European treeline
tree species. They show that local environmental factors in terms of microclimate and
biotic interactions exert an important impact on these earliest life-stages: i) early tree
establishment was either limited by temperature or moisture and often by
interactions of both, and ii) negative effects in interaction with herbaceous alpine
vegetation dominated, but evergreen tree species could benefit from a seasonal
switch to facilitation in autumn. While the responses to microclimate were generally
consistent over both early life-stages, only seedlings were affected by interactions
with herbaceous vegetation, and overall idiosyncratic patterns were predominant.
133
The important and variable impact of these site-specific and intrinsic factors on
treeline tree regeneration may further help to explain observed treeline patterns and
highlights the importance of their consideration for predictive models.
The third chapter presents a commentary contesting a recent study with a new,
promising approach of testing the low-temperature growth limitation of treeline
trees by growing tree seedlings in suspended pots. The criticism is based on the own
previous identification of considerable confounding factors in a similar experimental
design and the accordingly questionable interpretation of the study’s results is re-
evaluated with regard to the impact of high soil temperature fluctuations and
moisture deficiency. Additionally, species-specific treeline-forming mechanisms are
reviewed in relation to the study’s focal species, and alternatives for the after all
elegant method are explored.
The thesis concludes with an outlook in the potential of large-scale collaborative
projects to meet the need of more integrative studies that consider interacting
environmental factors, life-stage- and species-specific limitations as well as seasonal
and long-term responses. Such studies could bring unparalleled contributions for the
understanding of the interdependencies that limit treeline tree regeneration as an
important driver of treeline formation.
134
ZUSAMMENFASSUNG
Alpine Baumgrenzen sind wichtige und auffällige Vegetationsgrenzen hoher Berge,
die seit Jahrhunderten das Interesse von Naturforschern und Wissenschaftlern auf
sich ziehen. Sie sind schlussendlich durch einen Mangel an Wärme limitiert, was eine
Ausdehnung durch die aktuelle Klimaerwärmung erwarten lässt. Dennoch findet
man lokal stark variierende Muster und Dynamiken, welche durch andere
klimatische Faktoren und spezifische Bedürfnisse von Baumart oder Phase des
Lebenszyklus beeinflusst werden. Da die Regeneration von Bäumen die
Vorbedingung ist, damit Baumgrenzen stabil bleiben oder bergan voranschreiten,
könnten vor allem die kritischen Phasen der Keimung und Keimlingsetablierung
einen bedeutenden, lebensgeschichtlichen Engpass für Baumpopulationen der
Baumgrenze darstellen. Die Erforschung der Faktoren, die die Regeneration von
Bäumen limitieren ist daher ein wichtiger Schritt zu einem tieferen mechanistischen
Verständnis der Entstehung von Baumgrenzen und zu zuverlässigen Vorhersagen
ihrer zukünftigen Dynamik.
Die ersten beiden Kapitel dieser Dissertation präsentieren zwei Studien, welche
einmal die Wirkung abiotischer und einmal biotischer Umweltfaktoren auf die
beiden kritischen Regenerationsphasen Keimung und Keimlingsetablierung für fünf
wichtige europäische Baumarten der alpinen Baumgrenze untersuchen. Sie zeigen,
dass lokale Umweltfaktoren in Form von Mikroklima und biotischen Interaktionen
einen starken Einfluss auf die frühen Lebensphasen ausüben: i) frühe
Keimlingsetablierung war entweder durch Temperatur oder Feuchte, und häufig
135
durch Interaktionen beider Faktoren limitiert, und ii) negative Effekte durch die
Interaktion mit alpiner Graslandvegetation überwogen, aber immergrüne Baumarten
konnten im Herbst von einem saisonalen Wechsel zu einer förderlichen Beziehung
profitieren. Während die Reaktionen auf das Mikroklima generell in beiden frühen
Lebensphasen übereinstimmten, wurden nur Keimlinge von Interaktionen mit der
Graslandvegetation beeinträchtigt, und insgesamt überwogen artspezifische Effekte.
Der wichtige und variable Einfluss dieser ortsspezifischen und intrinsischen
Faktoren auf die Regeneration von Bäumen der Baumgrenze kann so im Weiteren
helfen, deren beobachtete Muster zu erklären und hebt hervor, wie wichtig ihre
Berücksichtigung in voraussagenden Modellen ist.
Das dritte Kapitel präsentiert einen Kommentar, der eine rezente Studie in Frage
stellt, in welcher das Anziehen von Keimlingen in hoch aufgehängten Behältnissen
als neuer, vielversprechender Ansatz zum Testen der Kältelimitierung des
Baumwachstums verwendet wird. Die Kritik basiert auf der eigenen vorherigen
Feststellung von beträchtlichen Störfaktoren in einem ähnlichen experimentellen
Aufbau, und die demzufolge fraglichen Interpretationen der Ergebnisse dieser Studie
werden unter der Berücksichtigung des Einflusses von starken
Bodentemperaturschwankungen und Wassermangel neu bewertet. Zusätzlich
werden artspezifische Mechanismen der Baumgrenzenbildung im Hinblick auf die
Zielarten der besagten Studie besprochen und Alternativen für die trotz allem
elegante Methode untersucht.
Die Dissertation schließt mit einem Ausblick auf das Potential von großangelegten,
gemeinschaftlichen Projekten um den Bedarf an integrativeren Studien zu decken,
136
welche interagierende Umweltfaktoren, art- und lebensphasenspezifische
Limitierungen sowie saisonale und Langzeitreaktionen berücksichtigen. Solche
Studien könnten einmalige Beiträge zum Verständnis der Wechselbeziehungen
leisten, welche die Regeneration von Bäumen als wichtigen Einflussfaktor der
Baumgrenzenbildung limitieren.
137
DANKSAGUNG
An dieser Stelle möchte ich mich zunächst bei meinen beiden Betreuern Maaike
Bader und Gerhard Zotz mit viel Nachdruck bedanken, die mir mit diesem
Promotionsprojekt eine sehr erfahrungsreiche und auch schöne Zeit in der
einmaligen Naturlandschaft der französischen Alpen ermöglicht haben, und deren
immer offene Türen / Ohren sowie unkomplizierte Email- / Skype-Kommunikation
ich sehr zu schätzen wusste!
Le meilleur concept ne peut pas aller bien loin sans un endroit à la hauteur de sa
mise en œuvre – cet endroit nous a été gracieusement fourni par la Station Alpine
Joseph Fourier de l’Université de Grenoble avec les zones expérimentales et le
Chalet-Labo au Jardin Alpin du Lautaret. Je remercie tout particulièrement l’ancien
directeur du Jardin Alpin, Serge Aubert, à qui le support de la recherche et la
réalisation de mes idées et de mes manips tenaient toujours très à cœur. Un grand
merci aussi à Karl Grigulis, Franck Delbart et Pascal Salze pour tout support
technique et administratif, et à toute l’équipe du Lautaret pour les bons moments
passés ensemble et les soirées mémorables dans le chalet des jardiniers.
Ein großer Dank gilt auch meinem Vorgänger Marc Müller, von dessen ersten
Erfahrungen ich bei der Planung meiner Experimente sehr profitieren konnte, und
allen helfenden Händen, ohne die ich dieses Projekt nicht hätte realisieren können!
Allen voran meine resoluten Feld-Hiwis Verena Schenk und Eric Thurm, Jasmin
Baruck und Gesa Pries, Carla Sardemann, Mathilde Vicente und Tizian Weichgrebe,
die jeweils über eine Feldsaison zusammen mit mir Wetter und Anstrengungen
138
getrotzt haben um die Experimente in Frankreich aufzubauen, zu pflegen und um
Messungen durchzuführen. Auch in Oldenburg hatte ich viel Unterstützung durch
die Hiwis Sven Wemken, Elif Gökpinar, Nawin Grabowsky, Nora Wissner, unsere
findigen TA’s Ingeborg Eden und Brigitte Rieger, sowie auf die eine oder andere Art
und Weise alle Mitglieder der AG’s Zotz und Albach. Vielen Dank an alle!
Schließlich hat meine Familie einen besonders großen Anteil an den Resultaten der
letzten Jahre:
Meine wunderbaren Eltern Horst und Hiltrud Kern, die schon früh mein Interesse an
der Natur gefördert haben und mich während meiner ganzen Entwicklung,
Ausbildung und auf allen neuen Wegen nach Kräften unterstützt haben, und mein
wunderbarer Bruder Boris, der mir immer wieder einen frischen und
unkonventionellen Blick auf die Dinge ermöglicht.
Et surtout mon merveilleux mari Jessy, qui était à mes côtés à travers toutes les
étapes de cette thèse, du terrain dans les hauteurs des Alpes jusque dans le labo à
Oldenburg, des moments les plus joyeux jusqu’au temps le plus difficile. Merci mon
amour d’avoir tenu ma corde de secours pendant cette escalade exigeante, j’ai
beaucoup hâte à nos aventures à venir!
139
LEBENSLAUF
PERSÖNLICHE DATEN
Name: Hannah Loranger
Geboren: 26.11.1985, Duisburg
Familienstand: verheiratet
Staatsangehörigkeit: Deutsch
WISSENSCHAFTLICHER WERDEGANG
6.2005 Abitur, Gymnasium Thomaeum, Kempen
10.2005 – 8.2007 Grundstudium Biologie an der Universität Osnabrück
9.2007 – 8.2008 Auslandsstudium an der Partneruniversität Université de
Sherbrooke, Québec, Kanada
9.2008 – 6.2011 Hauptstudium Biologie an der Universität Osnabrück mit dem
Abschluss Diplom, Prüfungsfächer: Botanik, Ökologie, Pflanzen-
physiologie, externe Diplomarbeit „Increasing invertebrate
herbivory along an experimental grassland plant diversity
gradient“ an der Universität Jena unter der Leitung von Prof. Dr.
Wolfgang W. Weisser und Jun. Prof. Dr. Till Eggers
7.2011 – 9.2011 Wissenschaftliche Hilfskraft am Institut für Ökologie, Universität
Jena
Seit 10.2011 Promotion an der Carl-von-Ossietzky Universität Oldenburg im
DFG-Projekt „The regeneration niche of trees at the alpine
treeline: climatic constraints on germination and seedling
establishment”
Leitung: Prof. Dr. Gerhard Zotz
140
PUBLIKATIONEN
Loranger J, Meyer ST, Shipley B, Kattge J, Loranger H, Roscher C & Weisser WW.
2012. Predicting invertebrate herbivory from plant traits: evidence from 51
grassland species in experimental monocultures. Ecology 93: 2674-2682.
Loranger J, Meyer ST, Shipley B, Kattge J, Loranger H, Roscher C, Wirth C & Weisser
WW. 2013. Predicting invertebrate herbivory from plant traits: Polycultures show
strong nonadditive effects. Ecology 94: 1499-1509.
Loranger H, Weisser WW, Ebeling A, Eggers T, De Luca E, Loranger J, Roscher C,
Meyer ST. 2014. Invertebrate herbivory increases along an experimental gradient
of grassland plant diversity. Oecologia 174: 183-193.
Bader M, Loranger H & Zotz G. 2014. A cool experimental approach to explain
elevational treelines. 2014. But can it explain them? American Journalof Botany
101 (9): 1403-1408.
KONFERENZBEITRÄGE
Loranger H, Zotz G & Bader 2014. Species-specific climate responses in tree
regeneration at the alpine treeline. Joint annual Meeting British Ecological
Society and Société Française d’Écologie, 9 – 12 December 2014, Lille, France
141
AUTHORS’ CONTRIBUTIONS
In the following section I present the contributions of all authors to the chapters 2 to
4 and their respective manuscripts or publications.
Chapter 2: Loranger H, Zotz G, Bader M Y. Impacts of soil microclimate on early
establishment of trees at the alpine treeline: idiosyncratic responses and the
importance of soil moisture. Resubmitted to AoB Plants
MB conceived the study with input from GZ, HL and MB planned the
experimental design, HL set up the experiment, collected the data, performed the
statistical analysis of the data and produced graphs and tables, HL wrote the initial
version of the manuscript with the support of MB and GZ, all authors contributed to
the revision of the manuscript.
Chapter 3: Loranger H, Zotz G, Bader M Y. Competitor or facilitator? The role of
grassland vegetation for germination and seedling performance of tree species at the
alpine treeline. Submitted to Functional Ecology
MB conceived the study with input from GZ, HL and MB planned the
experimental design, HL set up the experiment, collected the data, performed the
statistical analysis of the data and produced graphs and tables, HL wrote the initial
version of the manuscript with the support of MB and GZ.
Chapter 4: Bader M Y, Loranger H, Zotz, G. A cool experimental approach to explain
elevational treelines, but can it explain them? American Journal of Botany 101(9): 1–6
142
MB, HL and GZ planned the initial study, HL designed a prototype planting-
tube, collected and evaluated preliminary data, HL and MB decided to cancel the
experiment due to confounding factors arising from the specific design, MB wrote a
commentary contesting a study using a similar design without controlling for
confounding factors (Fajardo and Piper 2014, American Journal of Botany, 101: 788–795)
with the participation of GZ and HL.
Als Betreuer der Arbeit bestätige ich die Richtigkeit der Autorenbeiträge zu den
aufgeführten Kapiteln bzw. deren Manuskripten oder Veröffentlichungen.
…………………………………..
Prof. Dr. Gerhard Zotz
Hiermit bestätige ich die Richtigkeit der Autorenbeiträge zu den aufgeführten
Kapiteln bzw. deren Manuskripten oder Veröffentlichungen.
…………………………………..
Hannah Loranger
143
ALLGEMEINE ERKLÄRUNG
Ich füge folgende Erklärungen an gemäß § 11 Abs. 2 der Promotionsordnung (Stand:
13.01.2013) der Fakultät für Mathematik und Naturwissenschaften, Carl von
Ossietzky
Universität Oldenburg:
- Die Dissertation “The regeneration niche of trees at the alpine treeline -
Constraints of microclimate and the alpine grassland vegetation on
germination and seedling establishment” wurde von mir selbständig und nur
unter Verwendung der angegebenen Hilfsmittel verfasst.
- Ich strebe eine Promotion zur Doktorin der Biologie (Dr. rer. nat) an.
- Teile der Dissertation (Kapitel 4) wurden bereits veröffentlicht (siehe Author’s
contributions).
- Die Dissertation lag/liegt weder in ihrer Gesamtheit noch in Teilen einer
anderen wissenschaftlichen Hochschule zur Begutachtung in einem
Promotionsverfahren vor.
Hiermit bestätige ich die Richtigkeit der allgemeinen Erklärung.
…………………………………………
Hannah Loranger
I plant the first seedlings into the experiment, Lautaret June 2012