DISS. ETH NO. 22114
TRAIT-SPECIFIC RESPONSES OF CONIFERS TO IRRIGATION
AND DROUGHT AT DIFFERENT TEMPORAL SCALES
A thesis submitted to attain the degree of
DOCTOR OF SCIENCES of ETH ZURICH
(Dr. sc. ETH Zurich)
presented by
LINDA M. FEICHTINGER
Dipl.-Biologist, University Potsdam
born on 31.07.1981
citizen of Germany
accepted on the recommendation of
Prof. Dr. Nina Buchmann, examiner
Dr. Andreas Rigling, co-examiner
Dr. Kerstin Treydte, co-examiner
2014
Table of contents
Summary..................................................................................................
1
Zusammenfassung....................................................................................
5
General introduction................................................................................
9
Chapter I
Growth adjustments of conifers to drought and to
century-long irrigation.............................................
Chapter II
25
Long-term growth and physiological adjustments of
mature Scots pine and European larch to water
availability...............................................................
Chapter III
59
Trait-specific responses of Scots pine to irrigation at
different temporal scales…………………………………….…
85
Synthesis…………………………………….…………………………………………….…………..
117
Acknowledgements……………………………………………………………………………....
127
Summary
1
Summary
Drought is one of the most important environmental constraints on forest productivity
and the key factor in climate-related forest dieback. Drought-induced effects on forest
ecosystems are likely to gain in importance as climate projections predict an increase in
the frequency and severity of future drought events. Predictions of future forest
development, however, remain difficult, as the mechanisms underlying the adjustment
of mature trees to long-term variations in water availability are not well understood.
Therefore, the main objectives of this dissertation were to investigate the adjustments
to (1) long-term differences in water availability, and (2) changes in the water regime
of radial tree growth, physiology and morphology of dry forest ecosystems in innerAlpine valleys in Switzerland (Valais) and in Italy (Vinschgau). Trees growing under
naturally dry conditions (control) were compared with trees that had been (1)
permanently irrigated since establishment, growing along open water channels
(chapters I-III), (2) undergone a sudden stop in irrigation, when the channels were
drained (chapters I & II), and (3) exposed to a two-year re-irrigation, when a channel
dad been re-established (chapter III).
In the first chapter, the long-term effects of irrigation and drought on tree growth were
investigated by comparing the basal area increment (BAI) between irrigated and nonirrigated (control) Scots pine (Pinus sylvestris L.) in Valais as well as European larch
(Larix decidua Mill.) and black pine (Pinus nigra Arnold) in Vinschgau for the period
from 1920 to 2009. Moving-average response functions were used to track climate and
drought responses of trees over time. Furthermore, the adjustments in radial growth
of Scots pine and larch to an abrupt stop in irrigation were analysed. Irrigation
2
Summary
promoted the radial growth of all tree species investigated compared to the control:
(1) directly through increased soil water availability, and (2) indirectly through
increased soil nutrients and humus contents in the irrigated plots. Irrigation led to a full
elimination of growth responses to climate for European larch and black pine, but not
for Scots pine, which might become more sensitive to drought with increasing tree size
in Valais. For the control trees, the response of the latewood increment to water
availability in July/August has decreased in recent decades for all species, but increased
in May for Scots pine only. The sudden irrigation stop caused a drop in radial growth to
a lower level for Scots pine or similar level for larch compared to the control for up to
ten years. However, both tree species were then able to adjust to the new conditions
and subsequently grew with similar (Scots pine) or even higher growth rates (larch)
than the control.
The second chapter focuses on the physiological responses to long-term irrigation and
drought as well as on the impacts of an irrigation stop on Scots pine and European
larch. Measurements of tree-rings and Δ13C in latewood were combined to assess radial
growth and the physiology of gas exchange of two Scots pine (Valais) and one
European larch stand (Vinschgau). Despite the higher radial growth of the irrigated
trees compared to the control trees at all sites, the irrigation led to a major increase in
Δ13C at only one of our sites (pine). In contrast, Δ13C of the irrigated larch trees
increased only slightly compared to the control, while we found no differences for
trees at the second pine site. This suggests that the adjustments of Scots pine and
European larch to the local water availability occurred mainly through modifications in
radial and height increment, whereas variations in physiology appear to be limited and
highly site-specific. In response to the irrigation stop, larch demonstrated a higher
sensitivity in radial growth and physiology than Scots pine. Surviving trees of Scots pine
Summary
3
and larch, however, were able to adjust to the new and drier conditions. Both species
showed similar radial growth rates and Δ13C values than control trees after about ten
to 15 years after the irrigation stopped.
The third compares the adjustments in radial growth and morphological needle and
shoot traits of irrigated trees of mature Scots pine with those of control trees growing
under naturally dry conditions at three sites in Valais. The trees growing along two
channels had been irrigated since germination (>70 years), whereas those along
another previously drained channel had been irrigated only from 2010 to 2012, when
the channel was re-established, and could thus be used to quantify the short-term
effects of re-irrigation. Linear mixed models revealed that needle and shoot lengths as
well as early- and latewood basal area increments (BAIs) were most responsive to
short-term and long-term irrigation. However, the magnitude of the response to the
short-term irrigation exceeded that of the long-term irrigation. An extreme drought
during the first half of 2011 led to an immediate decrease in the needle length, needle
width, and early- and latewood BAIs of the control trees, whereas the shoot length and
needle numbers of control trees reacted with a 1-year delay to the extreme drought, as
the shoots were responding to water availability of previous year's summer. Such
negative responses to dry climatic conditions were even found in irrigated trees at one
of our sites, which might be linked to tree growth becoming more sensitive to drought
with increasing tree height and leaf area.
The results of this dissertation demonstrated that the adjustments of different conifers
to local water availability were achieved through the modification in the tree size
(radial and height growth) as well as in the leaf area, whereas variations in the tree
physiology appeared to be limited and highly site specific on the longer time scales.
Furthermore, the long-term responses of trees to changes in water availability are
4
Summary
likely to be different in magnitude from short-term responses. Therefore, integrated
approaches and measurements of multiple traits are required in order to understand
the effects of drought on tree performance and to uncover non-linear processes on
longer time scales.
Zusammenfassung
5
Zusammenfassung
Extreme Trockenperioden stellen einige der wichtigsten Einschränkungen der
Produktivität der Wälder dar und werden in Verbindung mit klimabedingtem
Waldsterben gebracht. Trockenheitsbedingte Einflüsse auf Waldökosysteme werden
sehr wahrscheinlich an Bedeutung gewinnen, da Klimaprognosen eine Zunahme in der
Häufigkeit und im Ausmass von Trockenperioden vorhersagen. Wie sich Wälder in
Zukunft entwickeln werden, bleibt jedoch weiterhin schwierig zu beurteilen, da die
zugrundeliegenden Mechanismen der langfristigen Anpassung adulter Bäume
weitgehend unklar sind. Die folgende Dissertation untersucht diesbezüglich das
Anpassungsvermögen verschiedener Koniferen an (1) langfristige Unterschiede in der
Wasserverfügbarkeit sowie (2) Veränderungen im Wasserhaushalt und dessen
Auswirkungen auf das radiale Wachstum, die Physiologie und die Morphologie von
Bäumen auf Trockenstandorten in inner-Alpinen Trockentälern der Schweiz (Wallis)
und Italiens (Vinschgau). Bäume auf natürlich trockenen Standorten (Kontrolle) wurden
mit solchen verglichen, welche (1) seit Etablierung von historischen Kanälen bewässert
wurden (Kapitel I-III), (2) einem abrupten Bewässerungsabbruch ausgesetzt waren
(Kapitel I & II), und (3) für einen Zeitraum von zwei Jahren wiederbewässert wurden
(Kapitel III).
Im ersten Kapitel werden die Langzeit-Auswirkungen von Bewässerung und
Trockenheit auf das Baumwachstum untersucht. Hierfür wurde der Zuwachs der
Basalfläche von bewässerten und unbewässerten Bäumen (Kontrollen) der Wald-Föhre
(Pinus sylvestris L.), der Europäischen Lärche (Larix decidua Mill.) und der SchwarzFöhre (Pinus nigra Arnold) über den Zeitraum von 1920 bis 2009 verglichen. Um die
Klima- und Trockenheitsreaktionen der Bäume und deren Veränderungen über die Zeit
6
Summary
zu untersuchen, wurde eine gleitende Mittelwerts-Analyse durchgeführt. Des Weiteren
wurde das Anpassungsvermögen von Wald-Föhre und Lärche untersucht, welche einer
abrupten Veränderung des Wasserhaushaltes ausgesetzt wurden. Die Bewässerung
führte zu einer Zunahme des radialen Wachstums bei allen untersuchten Baumarten
im Vergleich zur Kontrolle. Dies war (1) direkt auf die erhöhte Wasserverfügbarkeit
zurückzuführen, und (2) indirekt auf eine erhöhte Nährstoffverfügbarkeit und einen
erhöhten Humusgehalt der bewässerten Böden. Im Gegensatz zur Lärche und SchwarzFöhre, wurde das Wachstum der Wald-Föhren unabhängig von der Bewässerung von
trockenen Klimabedingungen negativ beeinflusst. Dies könnte mit einer erhöhten
Klimasensitivität des Radialwachstums mit zunehmender Baumgrösse erklärt werden.
Das
Spätholz-Wachstum
der
Kontroll-Bäume
wurde
zu
Beginn
der
Untersuchungsperiode u.a. von der Wasserverfügbarkeit im Juli/August beeinflusst,
wohingegen in den letzten Jahrzehnten vor allem das Klima im Juni das Wachstum aller
Baumarten beeinträchtigte. Für das Wachstum der Wald-Föhre hatte zudem die
Wasserverfügbarkeit
im
Mai
an
Bedeutung
zugenommen.
Der
abrupte
Bewässerungstopp führte zu einer starken Wachstumsreduzierung von Wald-Föhre
und Lärche im Vergleich zur Kontrolle. Nach circa zehn Jahren konnten sich jedoch
diejenigen Individuen beider Arten, welche den Bewässerungsabbruch überlebten, an
die neuen Bedingungen anpassen und zeigten ähnliche Wachstumsraten wie die
Kontroll-Bäume.
Das zweite Kapitel konzentriert sich auf die Analyse der physiologischen Reaktionen
der Wald-Föhre und der Europäischen Lärche auf Langzeit-Bewässerung und
Trockenheit sowie eine kurzfristige Veränderung in der Wasserverfügbarkeit. Hierfür
wurden das jährliche Wachstum und das Δ13C des Spätholzes von bewässerten und
Kontroll-Bäumen ermittelt. Trotz des erhöhten Radialwachstums aller bewässerten
Summary
7
Bäume im Vergleich zur Kontrolle führte die Bewässerung lediglich zu einer Erhöhung
der Δ13C Werte an einem Standort (Wald-Föhre). Im Gegensatz dazu wiesen die
bewässerten Lärchen bzw. Wald-Föhren des zweiten Standortes eine geringe bzw.
keine signifikante Erhöhung der Δ13C Werte im Vergleich zu den Kontroll-Bäumen auf.
Dies lässt vermuten, dass sich sowohl Föhre als auch Lärche langfristig durch
Veränderungen des radialen Wachstums und der Baumgrösse an unterschiedliche
Wasserverfügbarkeit anpassen und Anpassungen in der Baumphysiologie scheinbar
limitiert und stark standortsabhängig sind. Diejenigen Föhren und Lärchen, welche den
Bewässerungsabbruch
überdauerten,
konnten
sich
jedoch
an
die
neuen
Wachstumsbedingungen anpassen. Beide Arten zeigten jährliche Zuwachsraten und
Δ13C Messwerte ähnlich denen der Kontroll-Bäume nach ca. zehn bis 15 Jahren.
Das dritte Kapitel untersucht die morphologische Anpassungsfähigkeit von Nadel- und
Triebparametern der Wald-Föhre auf Bewässerung und Trockenheit an drei Standorten
im Wallis. An zwei Standorten wurden Bäume untersucht welche bereits seit der
Keimung bewässert wurden (>70 Jahren), wohingegen am dritten Standort die Bäume
lediglich über eine Periode von zwei Jahren von 2010 bis 2012 bewässert wurden.
Unabhängig von der Dauer der Bewässerung, reagierten Nadel- und Trieblängen sowie
Früh- und Spätholzwachstum der Föhre am stärksten im Vergleich zur Kontrolle; die
Nadelbreite, die spezifische Blattfläche und die Nadel-Langlebigkeit wurden hingegen
entweder nicht oder nur kaum beeinflusst. Das Ausmass der Reaktionen auf die
zweijährige Bewässerung war jedoch um einiges grösser als diejenige auf die
langfristige Bewässerung. Darüber hinaus führte eine extreme Dürre in der ersten
Hälfte des Jahres 2011 zu einer sofortigen Reduktion der Nadellänge und -breite sowie
des Frühholz- und Spätholzwachstums der Kontroll-Bäume. Im Gegensatz dazu
reagierten die Trieblänge und die Nadelmenge der Kontroll-Bäume mit einem Jahr
8
Summary
Verzögerung auf die extreme Trockenheit. Dies stimmt damit überein, dass das
Wachstum der Triebe vor allem durch das Klima im Sommer des Vorjahres beeinflusst
wird. Ähnlich der Kontrolle reagierten auch die bewässerten Bäume an einem der
Standorte auf die Trockenheit in 2011. Letzteres lässt sich vermutlich auf eine erhöhte
Sensibilität gegenüber Trockenheit mit zunehmender Baumgrösse und Blattfläche
erklären.
Die Ergebnisse dieser Dissertation zeigen, dass die Anpassung verschiedener Koniferen
an die vorhandene Wasserverfügbarkeit vor allem mit Modifikationen des Radial- und
Grössenwachstums und der Blattfläche verbunden ist, und Anpassungen in der
Physiologie der Lärche und Wald-Föhre langfristig jedoch stark von den
Standortbedingungen abhängen und sehr variabel ausgeprägt sind. Darüber hinaus
konnte gezeigt werden, dass die untersuchten Bäume unterschiedlich auf kurz- und
langfristige Veränderungen in der Wasserverfügbarkeit reagieren. Dies ist vor allem auf
verzögerte Reaktionen und Anpassungen an Bewässerung und Trockenheit
zurückzuführen. Dementsprechend sollten sich Studien nicht auf die Analyse einzelner
Parameter beschränken und längere Untersuchungszeiträume in Betracht ziehen, um
nicht-lineare Prozesse zu berücksichtigen. Dies erst ermöglicht ein umfassendes
Verständnis der zugrundeliegenden Anpassungsmechanismen der Bäume an
Trockenheit.
General introduction
9
General introduction
Background
Climate change in Europe and the Central Alps
Water availability is crucial for ecosystem functioning, and alterations in its amount
and distribution will ultimately affect the productivity of most biomes on Earth (Zhao
and Running 2010). A decrease in water availability has frequently been reported to be
the key factor of forest decline and vegetation shifts in recent decades (Allen et al.
2010; Breshears 2005; Rigling et al. 2013). As forests cover approximately 30% of the
Earth’s total land surface, changes in ecosystem processes, goods and services will be
of increasing relevance (Anderegg et al. 2012) when the climate gets drier and warmer
(IPCC 2013).
In the mountainous regions of Central Europe, the rise in temperature has already
exceeded the global average (Ceppi et al. 2012), leading to a significant increase in
drought stress (Rebetez and Dobbertin 2004). The effects of reduced water availability
on forest ecosystems may not be consistent, with larger impacts on dry forests than on
mesic forests (Scherrer et al. 2011). Accordingly, high rates of drought-induced
mortality of Scots pine (Pinus sylvestris L.) were found in inner-Alpine dry valleys on
steep, south-facing slopes at lower altitudes and on shallow soils (Oberhuber et al.
2001; Rigling et al. 2013; Vacchiano et al. 2011). Reports concerning an increase in
mortality of other conifer species in the Alpine region, however, are far fewer.
10
General introduction
Vulnerability of coniferous species
The sub-boreal Scots pine has a wide spatial distribution, stretching from northern
Siberia to the Sierra Nevada (Figure 1) and is among the most economically important
tree species. At its southernmost range, Scots pine is restricted by high temperatures
and summer drought (Carlisle and Brown 1968) and is growing at its limit of drought
tolerance at dry sites in Valais (Dobbertin et al. 2005). It has a wide ecological range,
but due to its low competitiveness it is restricted to sites with extreme conditions, such
as those that are nutrient poor and dry (Carlisle and Brown 1968).
Figure 1. Distribution of Scots pine (blue), European larch (red) and
black pine (green) in Europe. Distribution maps from EUFROGEN 2009,
www.eufrogen.org, adapted by C. Bachofen (WSL).
Black pine (Pinus nigra Arnold) is mainly distributed along the Mediterranean basin
(Figure 1). Although small native populations exist in the Eastern Alps, it has been
extensively planted at dry sites throughout the Alps due to its low drought sensitivity
General introduction
11
and high growth rates (Lévesque et al. 2013; Martínez-Vilalta and Piñol 2002).
Furthermore, it has been considered as a substitute to more drought-prone conifers
such as Scots pine (Thiel et al. 2012). Despite its high drought tolerance, black pine has
suffered from negative growth trends at dry sites of lower elevations in the
Mediterranean since recent decades (Linares and Tiscar 2010; Martín-Benito et al.
2010).
European larch (Larix decidua Mill.) is distributed in the sub-continental climate of the
mountains of Central Europe, primarily in the Alpine region and the Carpathian
Mountains (Figure 1). Due to its low competiveness and high demand for light, it occurs
naturally in sub-Alpine to montane regions, but has been widely planted at lower
elevations (Englisch et al. 2011). Furthermore, larch was found to be one of the most
drought sensitive conifers at lower elevations in the European Alps (Eilmann and
Rigling 2012; Lévesque et al. 2013; Schuster and Oberhuber 2012). To estimate the
future development of Scots pine, European larch and black pine at low elevations in
inner-Alpine dry valleys their long-term adjustment to drought should be further
investigated.
Mechanisms of drought-induced forest mortality
Despite ongoing research, we still do not fully understand the underlying mechanisms
of drought-induced forest dieback. Two possible physiological mechanisms explaining
drought-mediated forest mortality are currently being discussed (McDowell et al.
2008): hydraulic failure and carbon starvation. The carbon starvation hypothesis
suggests that a decrease in water availability triggers increased stomatal closure and
subsequently limits photosynthesis. This causes a reduction in carbon uptake, which
finally leads to carbohydrate starvation due to the ongoing metabolic demand for
12
General introduction
carbohydrates and depletion of reserves (McDowell et al. 2008). In contrast, the
hydraulic failure hypothesis implies that a reduction in water availability decreases
plant hydraulic conductivity to the point at which the conducting cells of the xylem are
impaired, and thus its water columns collapse (McDowell et al. 2008).
The relevance of both mechanisms depends, however, on the species-specific hydraulic
properties of the tree as well as the timing and duration of water scarcity. While
anisohydric species maintain a relatively high stomatal conductance during soil water
depletion and/or high evaporative demand, isohydric species prevent desiccation by a
more conservative stomatal control. As a consequence, anisohydric plants are prone to
hydraulic failure (Anderegg et al. 2011; Brodribb and Cochard 2008), while isohydric
plants are more likely to suffer from carbon starvation (Bréda et al. 2006; Galiano et al.
2011) during prolonged droughts.
Tree mortality during chronic water deficits, however, is known to be a complex
phenomenon, and is likely to interact with other factors such as insect (Gaylord et al.
2013) and pathogen attacks (Jactel et al. 2012). Furthermore, carbon and hydraulic
changes during tree mortality are likely to be connected (McDowell 2011; Sevanto et
al. 2014). In addition, recent studies state that only during extreme environmental
conditions, e.g. long lasting droughts, will C supply be limited, due to a higher priority
of carbon storage over growth (Palacio et al. 2014). The identification of these
mechanisms as the cause of tree mortality is difficult under natural conditions.
Drought manipulation experiments
One method of investigating the response of trees to differences in water availability is
the comparison of forest stands either at (1) dry and mesic sites (Addington et al. 2006;
General introduction
13
Lévesque et al. 2014) or (2) on a precipitation gradient (Lévesque et al. 2013; Scherrer
et al. 2011). One major constraint of these approaches is, however, that sites located
over large spatial distances often differ in other specific characteristics, e.g. soil texture
and chemistry (Beier et al. 2012), and responses to water availability are then often
difficult to disentangle from other mechanisms linked to these site-specific differences.
To reduce the effects of specific site characteristics, precipitation manipulation
experiments are a suitable, although more complex approach to investigate the effects
of changing water availability on forests ecosystems. These either reduce water
availability via rainout shelters, or increase it via irrigation. Most of these studies have,
however, been carried out over short time scales (e.g. Cotrufo et al. 2011; Dobbertin et
al. 2010), while to date there are only a few experiments that investigate the effects of
altered water availability on longer time scales (≥ eight years) (Barbeta et al. 2013;
Herzog et al. 2014; Martin-StPaul et al. 2013).
Resilience and non-linear reactions of trees to drought
Within certain boundaries, forest ecosystems show a resilience to environmental
disturbances. That means, forests might withstand a major stress within tolerable
impairments and recover over time to their pre-conditions (Haimes 2009; Holling
1973). Depending on the severity of perturbation, the time of recovery might vary
between one to several years. Indeed, recent drought manipulation experiments could
show that the response of trees to changes in water availability may be non-linear but
changes over time in response to the treatment (Barbeta et al. 2013; Herzog et al.
2014; Leuzinger et al. 2011). These investigated trends are likely to be related to
physiological and morphological adjustments to ambient water availability, such as
changes in leaf morphological traits (Dobbertin et al. 2010) and shoot-to-root ratio
14
General introduction
(Litton et al. 2007). On the other hand, compensatory effects, such as a reduction in
competition due to higher mortality rates, may also contribute to non-linear responses
(Barbeta et al. 2013; Lloret et al. 2012). It is possible that these adjustments are not
detected in short-term experiments, thus long-term investigations focusing on the
mechanisms controlling adjustment processes are needed in order to quantify the
resilience and the resistance of forests.
Dendroecological analysis
Dendroecological studies represent a valuable tool in quantifying the response of trees
to water availability. Tree-ring studies can provide information on growth reactions to
climate over long time periods, as tree-ring width is very sensitive to changes in water
availability (e.g. Affolter et al. 2010; Bigler et al. 2006; Eilmann et al. 2006). At dry sites
in particular, this variable reflects the site-specific variations and extremes of water
availability (Fritts 1976; Speer 2010), and thus contributes to an improved
understanding of climate-tree growth relationships. By using the basal area increment
(BAI) instead of the raw ring-width series, growth of trees with different DBH classes
can be further compared. The interpretation of the radial increment is, however,
limited due to (1) a variety of confounding factors masking climate signals, e.g. intraspecific competition and nutrient availability (Dobbertin 2005; Speer 2010), and (2)
insufficient information of the underlying physiological mechanisms of water
availability on tree-ring width (Kagawa et al. 2006).
The stable carbon isotope composition (δ13C) in tree rings is a straightforward tool to
investigate the past effects of water availability on gas exchange and stomatal
responses to changes in the water regime (Eilmann et al. 2010; Gessler et al. 2014;
Lévesque et al. 2013). The δ13C in tree rings depends as a first approximation on the
General introduction
15
ratio of ambient (c a ) to leaf intercellular (c i ) CO 2 concentration and therefore jointly
reflects the rates of carbon assimilation during photosynthesis and stomatal
conductance (Farquhar et al. 1989; Farquhar et al. 1982). Due to either low soil water
availability and/or dry air and thus increased vapour pressure deficit, reduced stomatal
conductance decreases the supply and fixation of CO 2 under dry conditions. As a
consequence c i decreases leading to a decrease in carbon isotope discrimination (∆13C)
and thus to an increase in δ 13C. Therefore, δ13C in tree rings depends on the carbon
isotope discrimination affected by g s and A as well as on post-photosynthetic
discrimination during wood formation (Werner and Gessler 2011), but also on the δ13C
of the source CO 2 .
Currently the mean δ13C of atmospheric CO 2 is ~ − 8‰. This value, however, is
becoming more negative over time (~0.02 ‰ year–1) caused by progressively increasing
emissions of
13
C-depleted CO 2 during fossil fuel combustion (Affek and Yakir 2014).
Since this trend biases the climatic and physiological signal in the tree-ring record, it is
commonly removed and the discrimination ∆13C is used as indicator of physiological
responses to changes in water fluxes, and is thus useful to quantify the changes in
water availability connected with climate change.
Besides the adjustments in radial growth and tree physiology, the plasticity in leaf area
is crucial for trees to react to changes in water availability. Furthermore, the leaf area is
an important factor controlling light interception, carbon gain and water use efficiency
in forest ecosystems (Niinemets et al. 2001; Raison et al. 1992). There is evidence to
suggest that morphological leaf traits of trees, such as foliage, shoot growth and foliage
density, adjust to changes in water availability (Dobbertin et al. 2010; Grill et al. 2004;
Limousin et al. 2011). With this in mind, variations in leaf and shoot morphological
traits can be of help to understand adjustments of trees to drought.
16
General introduction
Historical water channels
In Valais and Vinschgau, networks of artificial water channels exist that have been
constructed to lead melt water from higher altitudes to agricultural land in the valleys.
These so called Bisse (Valais) or Waale (Vinschgau) carry water from the end of April
until the beginning of October, and can be dated back to the 11th century (Crook and
Jones 1999; Leibundgut and Kohn 2014). As these traditional channels consist of open
ditches, they lose a large amount of water to the surrounding vegetation. Forests along
these channels have profited from this passive irrigation for centuries (Rigling et al.
2003), creating highly diverse habitats of fauna and flora (Crook and Jones 1999). For
the “Grand Bisse de Lens” in Valais, a channel which was also used as one of our study
sites, the water loss was quantified at about 80 l/s (or 27%) along a 3 km transect
(Brühlhart 1999). To reduce water loss, these channels have often been replaced by
impermeable pipes, or have been abandoned in recent decades due to the high
maintenance costs (Leibundgut and Kohn 2014). The subsequent stop of this passive
irrigation had considerable consequences on the surrounding forest vegetation
(Eilmann et al. 2009; Rigling et al. 2003).
Objectives
The main objectives of this study were to investigate the long-term adjustments of
trees to differences in water availability and the impact of changes in water regime on
tree growth, physiology and morphology of dry forest ecosystems in inner-Alpine
valleys in Switzerland (Valais) and in Italy (Vinschgau). Changes in growth rates were
analysed through basal area increment (BAI) measurements. Tree physiological
responses to water availability were investigated by the analyses of stable carbon
General introduction
17
isotopes (δ13C) in tree rings. Morphological adjustments of the canopy were assessed
by measurements of needle and shoot traits.
Trees growing under naturally dry conditions (control) were compared with trees that
had been (1) permanently irrigated since establishment, growing along open water
channels (chapters I-III), (2) undergone a sudden stop in irrigation, where the channels
were drained (chapters I & II), and (3) exposed to a two-year re-irrigation, where a
channel has been re-established (chapter III).
Structure of the dissertation
The present dissertation is divided into three chapters. The first chapter provides a
general overview on the long-term responses of Scots pine, European larch and black
pine to irrigation and drought using dendrochronological methods. The second chapter
presents a detailed analysis of the responses of growth and tree physiology of Scots
pine and European larch to irrigation using dendrochronological and stable isotope
methods. The third chapter gives an insight into the morphological adjustments of the
crown as well as of radial growth of Scots pine to differences in water availability by
analysing needles, shoots and tree rings.
18
General introduction
Chapter I: Growth adjustments of conifers to drought and to century-long
irrigation
Water shortage has been shown to be an important trigger for tree mortality and
forest decline. However, predictions on the long-term responses of conifers to warmer
and drier climate remain uncertain. Therefore, we used open water channels in innerAlpine dry valleys as a long-term irrigation experiment. The basal area increment of
irrigated Scots pine, European larch and black pine was compared to corresponding
control trees growing under naturally dry conditions. Moving-average response
functions were used to assess the temporal patterns of the growth-climate
relationships and drought responses of trees. Furthermore, the adjustment capacity of
Scots pine and larch to a change in the water supply was analysed. The following
research questions were addressed:
(1) Do species differ in their growth responses to long-term differences in water
availability over the past 90 years?
(2) How do the relationships between radial growth and monthly climatic variables
change with differences in water availability?
(3) Have there been temporal changes in these responses during the last 90 years?
(4) Are Scots pine and European larch able to recover after a sudden stop in irrigation
when growing in dry environments?
General introduction
19
Chapter II: Long-term adjustments in growth and physiology of mature Scots
pine and European larch to drought and irrigation
Investigating growth and physiological responses of trees to long-term differences in
water availability is crucial in order to better understand the adjustment capacity of
trees to drought. During periods of low water availabiliy the maintenance of an
efficient water transport is crucial to avoid xylem cavitation and a downregulation of
photosynthetic activity. However, detailed knowledge about the physiological
responses to drought and irrigation on longer time scales is limited. Therefore, Scots
pine and European larch growing under natural dry conditions and irrigated trees were
compared for the period from 1970 to 2009. The basal area increment and stable
carbon isotopes (δ13C) in tree rings were analysed to assess the effects of a change in
water availability on radial growth and physiology. The following specific research aims
were addressed:
(1) What are the long-term adjustments in growth and physiology of irrigated and
control trees?
(2) What are the short-term responses and long-term adjustments of radial growth
and physiology of irrigated trees following an irrigation stop?
20
General introduction
Chapter III: Trait-specific responses of Scots pine to irrigation on a short vs
long time sale.
Leaf area is an important factor controlling light interception, carbon gain and water
use efficiency in forest ecosystems and its placidity is crucial for trees to react to
changes in water availability. Through the combined analysis of basal area increment,
needle length and width, shoot length, needle numbers per shoot, needle longevity
and specific leaf area, it is possible to better understand the adjusting mechanisms of
Scots pine to drought. Differences in radial growth and leaf area between irrigated and
control trees may, however, not be constant over time. To additionally investigate
temporal changes in the response of Scots pine to water availability, trees exposed to a
two-year irrigation and a long-term irrigation were compared. The following questions
were addressed:
(1) What are the effects of irrigation on needle, shoot and radial growth traits of Scots
pine?
(2) Are there differences in these effects between those with long-term irrigation and
those with only two-years of re-irrigation?
(3) What is the reaction of Scots pine to climate?
General introduction
21
Addington, R.N., L.A. Donovan, R.J. Mitchell, J.M. Vose, S.D. Pecot, S.B. Jack, U.G.
Hacke, J.S. Sperry and R. Oren. 2006. Adjustments in hydraulic architecture of
Pinus palustris maintain similar stomatal conductance in xeric and mesic
habitats. Plant, Cell and Environment. 29:535-545.
Affek, H.P. and D. Yakir. 2014. The Stable Isotopic Composition of Atmospheric CO2. In
Treatise on Geochemistry, 2nd Edn. Eds. H.D. Holland and K.K. Turekian.
Elsevier, Oxford, pp 179-212.
Affolter, P., U. Büntgen, J. Esper, A. Rigling, P. Weber, J. Luterbacher and D. Frank.
2010. Inner Alpine conifer response to 20th century drought swings. European
Journal of Forest Research. 129:289-298.
Allen, C.D., A.K. Macalady, H. Chenchouni, D. Bachelet, N. McDowell, M. Vennetier, T.
Kitzberger, A. Rigling, D.D. Breshears and E.H. Hogg. 2010. A global overview of
drought and heat-induced tree mortality reveals emerging climate change risks
for forests. Forest Ecology and Management. 259:660-684.
Anderegg, W.R.L., J.A. Berry, D.D. Smith, J.S. Sperry, L.D.L. Anderegg and C.B. Field.
2011. The roles of hydraulic and carbon stress in a widespread climate-induced
forest die-off. Proceedings of the National Academy of Sciences. 109:233-237.
Anderegg, W.R.L., J.M. Kane and L.D.L. Anderegg. 2012. Consequences of widespread
tree mortality triggered by drought and temperature stress. Nature Climate
Change. 3:30-36.
Barbeta, A., R. Ogaya and J. Peñuelas. 2013. Dampening effects of long-term
experimental drought on growth and mortality rates of a Holm oak forest.
Global Change Biology. 19:3133-3144.
Beier, C., C. Beierkuhnlein, T. Wohlgemuth, J. Penuelas, B. Emmett, C. Körner, H. Boeck,
J.H. Christensen, S. Leuzinger, I.A. Janssens, K. Hansen and J. Arnone. 2012.
Precipitation manipulation experiments - challenges and recommendations for
the future. Ecology Letters. 15:899-911.
Bigler, C., O.U. Bräker, H. Bugmann, M. Dobbertin and A. Rigling. 2006. Drought as an
inciting mortality factor in Scots pine stands of the Valais, Switzerland.
Ecosystems. 9:330-343.
Bréda, N., R. Huc, A. Granier and E. Dreyer. 2006. Temperate forest trees and stands
under severe drought: a review of ecophysiological responses, adaptation
processes and long-term consequences. Annals of Forest Science. 63:625-644.
Breshears, D.D. 2005. Regional vegetation die-off in response to global-change-type
drought. Proceedings of the National Academy of Sciences. 102:15144-15148.
Brodribb, T.J. and H. Cochard. 2008. Hydraulic Failure Defines the Recovery and Point
of Death in Water-Stressed Conifers. Plant Physiology. 149:575-584.
Brühlhart, H. 1999. Einfluss der Bewässerung auf das Jahrringwachstum von Kiefern.
Diplomarbeit. Berner Fachhochschule, Schweizerische Fachhochschule für
Holzwirtschaft, Biel.
Carlisle, A. and A.H.F. Brown. 1968. Pinus Sylvestris L. Journal of Ecology. 56:269-307.
Ceppi, P., S.C. Scherrer, A.M. Fischer and C. Appenzeller. 2012. Revisiting Swiss
temperature trends 1959-2008. International Journal of Climatology. 32:203213.
Cotrufo, M.F., G. Alberti, I. Inglima, H. Marjanović, D. LeCain, A. Zaldei, A. Peressotti
and F. Miglietta. 2011. Decreased summer drought affects plant productivity
and soil carbon dynamics in a Mediterranean woodland. Biogeosciences.
8:2729-2739.
Crook, D.S. and A.M. Jones. 1999. Design principles from traditional mountain irrigation
systems (Bisses) in the Valais, Switzerland. Mountain Research and
Development. 19:79-99.
22
General introduction
Dobbertin, M. 2005. Tree growth as indicator of tree vitality and of tree reaction to
environmental stress: a review. European Journal of Forest Research. 124:319333.
Dobbertin, M., B. Eilmann, P. Bleuler, A. Giuggiola, E. Graf Pannatier, W. Landolt, P.
Schleppi and A. Rigling. 2010. Effect of irrigation on needle morphology, shoot
and stem growth in a drought-exposed Pinus sylvestris forest. Tree Physiology.
30:346-360.
Dobbertin, M., P. Mayer, T. Wohlgemuth, E. Feldermeyer-Christle, U. Graf, N.E.
Zimmermann and A. Rigling. 2005. The decline of Pinus sylvestris L. forests in
the Swiss Rhône Valley - a result of drought stress? Phyton. 45:153-156.
Eilmann, B., N. Buchmann, R. Siegwolf, M. Saurer, P. Cherubini and A. Rigling. 2010.
Fast response of Scots pine to improved water availability reflected in tree-ring
width and δ13C. Plant, Cell & Environment. 33:1351-1360.
Eilmann, B. and A. Rigling. 2012. Tree-growth analyses to estimate tree species'
drought tolerance. Tree Physiology. 32:178–187.
Eilmann, B., P. Weber, A. Rigling and D. Eckstein. 2006. Growth reactions of Pinus
sylvestris L. and Quercus pubescens Willd. to drought years at a xeric site in
Valais, Switzerland. Dendrochronologia. 23:121-132.
Eilmann, B., R. Zweifel, N. Buchmann, P. Fonti and A. Rigling. 2009. Drought-induced
adaptation of the xylem in Scots pine and pubescent oak. Tree Physiology.
29:1011-1020.
Englisch, M., F. Starlinger and H. Lin. 2011. Die Lärche – ein Baum für alle Fälle? BFWPraxisinformation. 25
Farquhar, G.D., J.R. Ehleringer and K.T. Hubick. 1989. Carbon discrimination and
photosynthesis. Annual Review of Plant Physiology and Plant Molecular
Biology. 40:503–537.
Farquhar, G.D., M.H. O'Leary and J.A. Berry. 1982. On the relationship between carbon
isotope discrimination and the intercellular carbon dioxide concentration in
leaves. Australian Journal of Botany. 9:121-37.
Fritts, H.C. 1976. Tree rings and climate. Academic Press, London, UK.
Galiano, L., J. Martínez-Vilalta and F. Lloret. 2011. Carbon reserves and canopy
defoliation determine the recovery of Scots pine 4 yr after a drought episode.
New Phytologist. 190:750-759.
Gaylord, M.L., T.E. Kolb, W.T. Pockman, J.A. Plaut, E.A. Yepez, A.K. Macalady, R.E.
Pangle and N.G. McDowell. 2013. Drought predisposes piñon-juniper
woodlands to insect attacks and mortality. New Phytologist. 198:567-578.
Gessler, A., J.P. Ferrio, R. Hommel, K. Treydte, R.A. Werner and R.K. Monson. 2014.
Stable isotopes in tree rings: towards a mechanistic understanding of isotope
fractionation and mixing processes from the leaves to the wood. Tree
Physiology
Grill, D., M. Tausz, U. Pöllinger, M.J. Jiménez and D. Morales. 2004. Effects of drought
on needle anatomy of of Pinus canariensis. Flora - Morphology, Distribution,
Functional Ecology of Plants. 199:85-89.
Haimes, Y.Y. 2009. On the definition of resilience in systems. Risk Analysis. 29:498-501.
Herzog, C., J. Steffen, E. Graf Pannatier, I. Hajdas and I. Brunner. 2014. Nine years of
irrigation cause vegetation and fine root shifts in a water-limited pine forest.
PLoS ONE. 9:e96321.
Holling, C.S. 1973. Resilience and stability of ecological systems. Annual Review of
Ecology and Systematics. 4:1-23.
IPCC. 2013. Climate change 2013: The physical science basis. Contribution of working
group I to the fifth assessment report of the intergovernmental panel on
General introduction
23
climate change. Eds. T.F. Stocker, D. Qin, G.-K. Plattner, M. Tignor, S.K. Allen, J.
Boschung, A. Nauels, Y. Xia, V. Bex and P.M. Midgle. Cambridge University
Press, Cambridge, United Kingdom and New York, NY, USA, 1535 pp.
Jactel, H., J. Petit, M.-L. Desprez-Loustau, S. Delzon, D. Piou, A. Battisti and J. Koricheva.
2012. Drought effects on damage by forest insects and pathogens: a metaanalysis. Global Change Biology. 18:267-276.
Kagawa, A., A. Sugimoto and T.C. Maximov. 2006. Seasonal course of translocation,
storage and remobilization of 13C pulse-labeled photoassimilate in naturally
growing Larix gmelinii saplings. New Phytologist. 171:793-804.
Leibundgut, C. and I. Kohn. 2014. European traditional irrigation in transition part I:
Irrigation in times past-a historic land use practice across Europe. Irrigation and
Drainage. 63:273-293.
Leuzinger, S., Y. Luo, C. Beier, W. Dieleman, S. Vicca and C. Körner. 2011. Do global
change experiments overestimate impacts on terrestrial ecosystems? Trends in
Ecology & Evolution. 26:236-241.
Lévesque, M., M. Saurer, R. Siegwolf, B. Eilmann, P. Brang, H. Bugmann and A. Rigling.
2013. Drought response of five conifer species under contrasting water
availability suggests high vulnerability of Norway spruce and European larch.
Global Change Biology. 19:3184-3199.
Lévesque, M., R. Siegwolf, M. Saurer, B. Eilmann and A. Rigling. 2014. Increased wateruse efficiency does not lead to enhanced tree growth under xeric and mesic
conditions. New Phytologist:94-109.
Limousin, J.-M., S. Rambal, J.-M. Ourcival, J. Rodríguez-Calcerrada, I.M. Pérez-Ramos, R.
Rodríguez-Cortina, L. Misson and R. Joffre. 2011. Morphological and
phenological shoot plasticity in a Mediterranean evergreen oak facing longterm increased drought. Oecologia. 169:565-577.
Linares, J.C. and P.A. Tiscar. 2010. Climate change impacts and vulnerability of the
southern populations of Pinus nigra subsp. salzmannii. Tree Physiology.
30:795-806.
Litton, C.M., J.W. Raich and M.G. Ryan. 2007. Carbon allocation in forest ecosystems.
Global Change Biology. 13:2089-2109.
Lloret, F., A. Escudero, J.M. Iriondo, J. Martínez-Vilalta and F. Valladares. 2012. Extreme
climatic events and vegetation: the role of stabilizing processes. Global Change
Biology. 18:797-805.
Martín-Benito, D., M. Río and I. Cañellas. 2010. Black pine (Pinus nigra Arn.) growth
divergence along a latitudinal gradient in Western Mediterranean mountains.
Annals of Forest Science. 67:401-401.
Martin-StPaul, N.K., J.-M. Limousin, H. Vogt-Schilb, J. Rodríguez-Calcerrada, S. Rambal,
D. Longepierre and L. Misson. 2013. The temporal response to drought in a
Mediterranean evergreen tree: comparing a regional precipitation gradient
and a throughfall exclusion experiment. Global Change Biology. 19:2413-2426.
Martínez-Vilalta, J. and J. Piñol. 2002. Drought-induced mortality and hydraulic
architecture in pine populations of the NE Iberian Peninsula. Forest Ecology
and Management. 161:247-256.
McDowell, N., W.T. Pockman, C.D. Allen, D.D. Breshears, N. Cobb, T. Kolb, J. Plaut, J.
Sperry, A. West, D.G. Williams and E.A. Yepez. 2008. Mechanisms of plant
survival and mortality during drought: why do some plants survive while others
succumb to drought? New Phytologist. 178:719-739.
McDowell, N.G. 2011. Mechanisms linking drought, hydraulics, carbon metabolism, and
vegetation mortality. Plant Physiology. 155:1051-1059.
24
General introduction
Niinemets, Ü., D.S. Ellsworth, A. Lukjanova and M. Tobias. 2001. Site fertility and the
morphological and photosynthetic acclimation of Pinus sylvestris needles to
light. Tree Physiology. 21:1231-1244.
Oberhuber, W., W. Hofbauer and W. Kofler. 2001. Mortality and growth anomalies of
Scots pine (Pinus sylvestris L.) growing on drought exposed sites at the
Tschirgant Landslide (Tyrol). Berichte des Naturwissenschaftlich-Medizinischen
Vereins in Innsbruck 88: 87-97.
Palacio, S., G. Hoch, A. Sala, C. Körner and P. Millard. 2014. Does carbon storage limit
tree growth? New Phytologist. 201:1096-1100.
Raison, R.J., B.J. Myers and M.L. Benson. 1992. Dynamics of Pinus radiata foliage in
relation to water and nitrogen stress: I. Needle production and properties
Forest Ecology and Management. 52:139-158.
Rebetez, M. and M. Dobbertin. 2004. Climate change may already threaten Scots pine
stands in the Swiss Alps. Theoretical and Applied Climatology. 79:1-9.
Rigling, A., C. Bigler, B. Eilmann, E. Feldmeyer-Christe, U. Gimmi, C. Ginzler, U. Graf, P.
Mayer, G. Vacchiano, P. Weber, T. Wohlgemuth, R. Zweifel and M. Dobbertin.
2013. Driving factors of a vegetation shift from Scots pine to pubescent oak in
dry Alpine forests. Global Change Biology. 19:229-240.
Rigling, A., H. Brühlhart, O.U. Bräker, T. Forster and F.H. Schweingruber. 2003. Effects
of irrigation on diameter growth and vertical resin duct production in Pinus
sylvestris L. on dry sites in the central alps. Forest Ecology and Management.
175:285-296.
Scherrer, D., M.K.-F. Bader and C. Körner. 2011. Drought-sensitivity ranking of
deciduous tree species based on thermal imaging of forest canopies.
Agricultural and Forest Meteorology. 151:1632-1640.
Schuster, R. and W. Oberhuber. 2012. Drought sensitivity of three co-occurring conifers
within a dry inner Alpine environment. Trees
Sevanto, S., N.G. McDowell, L.T. Dickman, R. Pangle and W.T. Pockman. 2014. How do
trees die? A test of the hydraulic failure and carbon starvation hypotheses.
Plant, Cell & Environment. 37:153-161.
Speer, J.H. 2010. Fundamentals of tree-ring research. The University of Arizona Press,
Tucson, Arizona.
Thiel, D., L. Nagy, C. Beierkuhnlein, G. Huber, A. Jentsch, M. Konnert and J. Kreyling.
2012. Uniform drought and warming responses in Pinus nigra provenances
despite specific overall performances. Forest Ecology and Management.
270:200-208.
Vacchiano, G., M. Garbarino, E. Borgogno Mondino and R. Motta. 2011. Evidences of
drought stress as a predisposing factor to Scots pine decline in Valle d’Aosta
(Italy). European Journal of Forest Research. 131:989-1000.
Werner, C. and A. Gessler. 2011. Diel variations in the carbon isotope composition of
respired CO 2 and associated carbon sources: a review of dynamics and
mechanisms. Biogeosciences Discussions. 8:2183-2233.
Zhao, M. and S.W. Running. 2010. Drought-induced reduction in global terrestrial net
primary production from 2000 through 2009. Science. 329:940-943.
Chapter I
25
Chapter I Growth adjustments of conifers to drought and to
century-long irrigation
Published as:
Feichtinger, Linda M.1; Eilmann, Britta2; Buchmann, Nina3; Rigling, Andreas1, 2014.
Growth adjustments of conifers to drought and to century-long irrigation. Forest
Ecology and Management 334, 96-105.
1
Swiss Federal Research Institute WSL, Zürcherstrasse 111, 8903 Birmensdorf, Switzerland
2
Wageningen University, Centre for Ecosystem Studies, Forest Ecology and Forest
Management, 6700 AA Wageningen, The Netherlands
3
ETH Zürich, Institute of Agricultural Sciences, Universitätsstrasse 2, 8092 Zürich, Switzerland
26
Chapter I
Abstract
Our knowledge on tree responses to drought is mainly based on short-term
manipulation experiments which do not capture any possible long-term adjustments in
this response. Therefore, historical water channels in inner-Alpine dry valleys were
used as century-long irrigation experiments to investigate adjustments in tree growth
to contrasting water supply. This involved quantifying the tree-ring growth of irrigated
and non-irrigated (control) Scots pine (Pinus sylvestris L.) in Valais (Switzerland), as well
as European larch (Larix decidua Mill.) and black pine (Pinus nigra Arnold) in Vinschgau
(Italy). Furthermore, the adjustments in radial growth of Scots pine and European larch
to an abrupt stop in irrigation were analysed.
Irrigation promoted the radial growth of all tree species investigated compared to the
control: (1) directly through increased soil water availability, and (2) indirectly through
increased soil nutrients and humus contents in the irrigated plots. Irrigation led to a full
elimination of growth responses to climate for European larch and black pine, but not
for Scots pine, which might become more sensitive to drought with increasing tree size
in Valais. For the control trees, the response of the latewood increment to water
availability in July/August has decreased in recent decades for all species, but increased
in May for Scots pine only. The sudden irrigation stop caused a drop in radial growth to
a lower level for Scots pine or similar level for larch compared to the control for up to
ten years. However, both tree species were then able to adjust to the new conditions
and subsequently grew with similar (Scots pine) or even higher growth rates (larch)
than the control.
Chapter I
27
To estimate the impact of climate change on future forest development, the duration
of manipulation experiments should be on longer time scales in order to capture
adjustment processes and feedback mechanisms of forest ecosystems.
Introduction
Drought has been shown to be a key factor limiting tree growth, and may lead to tree
mortality and forest decline under extreme conditions (Breshears, 2005; Bigler et al.,
2006; Allen et al., 2010). These drought-induced effects on forest ecosystems are likely
to further gain in importance as the frequency, duration and severity of drought
periods in Central Europe are projected to increase in the future (IPCC, 2013). Already
today, high mortality rates have been observed in semi-arid environments after single
drought events and, even more pronounced, after multiple drought years (Allen and
Breshears, 1998; Bigler et al., 2006; Gitlin et al., 2006; McDowell et al., 2008). But the
species-specific differences in the reaction to water limitation, as well as mechanisms
leading to drought-induced growth depression and forest decline, are still poorly
understood.
Current knowledge on the responses of mature forests to drought is either based on
field observations during extreme drought years (Lebourgeois et al., 2010; Schuster
and Oberhuber, 2012) or on drought-manipulation experiments, in which water
availability on drought-prone sites is either reduced (MacKay et al., 2012; Plaut et al.,
2012; Barbeta et al., 2013) or enhanced via irrigation (Dobbertin et al., 2010; Cotrufo et
al., 2011; Eilmann et al., 2011; Ruehr et al., 2012). In most cases, the observation
period of these experiments lasted only from one to six years, but several long-term
manipulation experiments (>ten years) suggest that the response of trees to changes in
environmental conditions might not be constant over time (Leuzinger et al., 2011;
28
Chapter I
Barbeta et al., 2013). These non-linear responses might result from long-term or
delayed feedback mechanisms, such as morphological or physiological growth
adjustments of trees (phenotypic plasticity) to changes in water availability (Leuzinger
et al., 2011; Lloret et al., 2012), which are not detected in short-term experiments.
Consequently, long-term experiments are needed to reveal the mechanisms behind
growth adjustments to drought.
Mountainous regions are highly vulnerable to changes in climate since increasing
temperatures during the last 50 years already now exceeded those of the global
average (Ceppi et al., 2012). An increase in evapotranspiration rates due to higher
temperatures in inner-Alpine dry valleys, which are characterized by low precipitation,
has caused a significant increase in drought stress (Rebetez and Dobbertin, 2004).
Consequently, increasingly higher rates of drought-induced Scots pine (Pinus sylvestris
L.) mortality have been reported for many sites in these valleys (Cech and Perny, 1998;
Oberhuber et al., 2001; Weber et al., 2007; Rigling et al., 2013).
Historical water channels offer an excellent opportunity to study the long-term effects
of contrasting hydrological conditions on tree growth in such inner-Alpine dry valleys.
These channels are typically open ditches and passively irrigate the surrounding trees,
which can then be compared with trees growing nearby under naturally dry conditions.
Furthermore, several channels were drained in recent decades due to the high loss of
water and replaced by impermeable pipes (Leibundgut and Kohn, 2014), exposing the
trees growing along these channels to an abrupt change in water supply.
In the present study, we analyzed the long-term response to water availability of the
radial growth of three conifer species in inner-Alpine dry valleys, Scots pine in Valais
(Switzerland), European larch (Larix decidua Mill.) and black pine (Pinus nigra Arnold) in
Chapter I
29
Vinschgau (Italy). At each site, we compared the irrigated trees with control trees
growing under naturally dry conditions. In addition, the response to an irrigation stop
was studied at two sites. We aimed to answer the following questions: (1) Do species
differ in their growth responses to long-term differences in water availability over the
past 90 years? (2) How do the relationships between radial growth and monthly
climatic variables change with differences in water availability? (3) Have there been
temporal changes in these responses during the last 90 years? (4) Are Scots pine and
European larch able to recover after a sudden stop in irrigation when growing in dry
environments?
Material and Methods
Study sites
The five study sites are situated in two inner-Alpine valleys: we selected three Scots
pine stands in Valais, Switzerland, and one European larch stand as well as one black
pine stand mixed with European larch in Vinschgau, Italy (Figure 1). All sites were
located on south-facing slopes (15-37°) between 800-1000 m a.s.l. in Valais and about
1200 m a.s.l. in Vinschgau on shallow rendzic leptosols (FAO, 2006) (Table 1). In Valais,
the forest stands are dominated by Scots pine that has naturally regenerated over the
last century. In contrast, the forest in Vinschgau, dominated by European larch and
black pine, was planted at the end of the 19th century. The climate in both valleys is
characterized by low annual precipitation, with 664 mm on average in Valais and 657
mm in Vinschgau (Figure 2).
To reveal species-specific responses to contrasting water availability, we compared
trees growing under naturally dry conditions (control) with: (1) trees that had been
30
Chapter I
permanently irrigated since establishment, growing along open water channels; and (2)
trees exposed to a sudden stop in irrigation when the water channel was drained.
Hence, the five study sites consisted of three plots with permanently irrigated trees at
Varen and Planige (Scots pine) in Valais and Schluderns (European larch, black pine) in
Vinschgau and two plots with previously irrigated trees at Lens (Scots pine) in Valais
and Mals (European larch) in Vinschgau, with corresponding control plots (Table 1). No
suitable Scots pine stands were found in Vinschgau, and there were no appropriate
black pine and European larch stands growing along open water channels in Valais.
Figure 1. Location of the two study regions in Central Europe (a): Valais in Switzerland (b)
and Vinschgau in Italy (c). The study sites are marked as dots and the weather stations as
triangles.
Chapter I
31
The traditional irrigation channels had been constructed to conduct melt water from
higher altitudes to agricultural land in the valleys from the end of April until the
beginning of October (Leibundgut and Kohn, 2014). The oldest verified channel used in
our study was constructed in 1150 AD (Varen). As these channels resemble open
ditches, they lose a large amount of water, passively irrigating the surrounding
vegetation. For one of our sites, the Grand Bisse de Lens (Lens), the water loss was
quantified at about 80 l/s or 27% along a 3 km transect (Brühlhart, 1999). To avoid this
water loss, the open channels have often been replaced by impermeable pipes, as at
two of our study sites: Lens in 1984 and Mals in 1980. The channel at Lens was
however, re-established in May 2010 for touristic reasons. All control plots were
located near the channels, but outside of the irrigation zone (upslope or in the
proximity of the irrigated plots). For the irrigated larch trees in Schluderns, the same
control trees as for the irrigated larch trees in Mals were used.
Figure 2. Monthly mean precipitation (bars) and temperature (line) for Valais in Switzerland and for
Vinschgau in Italy for the period from 1961 to 1990. Temperature data were adjusted to the differences in
altitude between the weather stations and the study sites using monthly lapse rates published in Lotter et
al. (2002).
the control trees at Mals were used.
irrigated trees were tested with a two-sample Wilcoxon-rank-sum test (values with different letters: P≤0.05). Note that as a control for the irrigated European larch trees in Schluderns
coordinates, number of sampled trees (n) and mean values ± SE of tree age, diameter at breast height (DBH), and tree height measured in 2010. Differences between control trees and
Table 1. Site description and characteristics of Scots pine in Valais/Switzerland (Varen, Planige, Lens) and European larch and black pine in Vinschgau/Italy (Mals, Schluderns) with
32
Chapter I
Chapter I
33
Vegetation survey
Due to very rocky and shallow soils at our study sites, we were not able to characterize
soil conditions based on soil samples. As an alternative, we used vegetation surveys to
gain site-specific indicator values according to Landolt et al. (2010) to illustrate possible
differences between irrigated and control plots: (1) in soil characteristics (humidity,
humus and nutrient value), and (2) in growing conditions (light value). The indicator
values range from 1 to 5 (for the humus value from 1 to 3) and represent the mean soil
humidity during the vegetation period, ranging from very dry to flooded conditions
(humidity value); the soil nutrient availability, ranging from very poor to very rich soil
(nutrient value); the humus content, ranging from low content to high content (humus
value); and the light intensity at ground level, ranging from very shady to full light (light
value) (Landolt et al. 2010).
The understory vegetation (< 5 m in height) was surveyed in June and July 2011 on a 4
m2 plot around each sampled tree. Subsequently, indicator values of each species were
assigned. An un-weighted mean was calculated for each plot since the total percentage
of coverage per species was often low (<1%). Surveys with (1) less than nine species
per survey plot and (2) species with indifferent indicator values were excluded. In total,
160 of 162 surveys and 213 of 220 species were used for this analysis. At Lens, we
presume the understory vegetation had not yet been affected by the irrigation, which
was restarted one year before we did our survey.
Sampling, tree-ring measurements and detrending
In each plot, 12 to 18 dominant trees with no visible damage were sampled in 2010 by
taking two increment cores perpendicular to each other with an increment borer at
34
Chapter I
breast height (1.3 m). In the irrigated plots, only trees growing less than 1 m upslope or
4 m downslope of the channel were selected, to include only trees benefiting from the
passive irrigation provided by the channel. For the control, only those trees growing
about 50 m upslope or outside the irrigation zone near the irrigated trees were
sampled. During sampling, diameter at breast height (DBH) and tree height were
recorded for each tree. The surface of the increment cores were prepared with a core
microtome (Gärtner and Nievergelt, 2010) to increase the visibility of the tree rings.
Subsequently, early- and latewood widths were measured separately using a LINTAB
digital positioning table and the software TSAP-Win (Rinntech, Heidelberg, Germany). If
the pith was not included, the cambial age was estimated according to Bräker (1981).
Cross-dating was done both visually and subsequently statistically using the software
COFECHA (Holmes, 1994). For the control trees, missing rings had to be added in 2004
and 2007 for European larch in Vinschgau (four trees), and in 1921 and 1949 for Scots
pine in Valais (three trees). Trees that showed unusual and strong competition related
growth suppression and release or that were less than 70 years old were excluded
from further analysis.
In a next step, residual chronologies were computed using the software ARSTAN
(Holmes, 1994). For this purpose, raw tree-ring series were (1) power transformed to
stabilize the variance and to prevent negative residuals (Cook and Peters, 1997), and
(2) standardized by fitting a cubic smoothing spline with a 50% cutoff at 60 years to
remove age and non-climatic related trends (Cook and Peters, 1981). In order to
compare the absolute growth rates between the irrigated and control trees without
the confounding factors of age trend and DBH class, the annual basal area increment
(BAI) was calculated from bark to pith from the raw ring-width series according to
Biondi and Qeadan (2008):
Chapter I
35
BAI t = π ( R2 t - R2 t-1 )
(1)
where R is the stem radius at time t. Subsequently, the individual residual and BAI
chronologies were averaged per plot using the Tukey’s biweight robust mean
(Mosteller and Tukey 1977).
Climate data
Climate data for the sites in Vinschgau were obtained from the weather stations in
Marienberg (1323 m a.s.l.), about 3 km away from Mals, and about 7 km away from
Schluderns. The closest weather station for temperature data in Valais is located in
Sion (492 m a.s.l.), about 9 km away from Lens, 18 km from Planige and 20 km from
Varen. As precipitation is highly spatially variable we have chosen the weather station
in Sierre (539 m a.s.l.), which is closer to our sites but only recorded precipitation data.
This station is about 4 km away from Planige and 8 km from Varen and Lens (Fig. 1).
Temperature data were adjusted to the differences in altitude between the weather
stations and the study sites using monthly lapse rates published in Lotter et al. (2002).
To estimate the monthly water availability for tree growth, a drought index (DRI)
according to Thornthwaite (1948) was determined. The DRI is calculated as the
monthly precipitation sum minus the sum of potential evapotranspiration.
Data analyses
The raw tree-ring series were used to calculate the first-order autocorrelations,
quantifying the impact of the previous year on the current year’s tree-ring growth.
Based on the standardized tree-ring series, the mean sensitivity (year-to-year
variability of the radial growth) and the expressed population signal (EPS, common
36
Chapter I
variability in a chronology) was computed (Schweingruber, 1983; Speer, 2010). EPS is
used as a statistical quality measure for a chronology, with a common threshold of
0.85. Lower values point to a stronger dependence of single tree-ring series on
microsite conditions rather than on common site conditions (Schweingruber, 1983;
Speer, 2010).
Differences in the yearly BAI, indicator values, tree height, DBH and cambial age
between the irrigated and corresponding control trees were tested for significance
(P≤0.05) with a two-sample Wilcoxon-rank-sum test, the resulting P-values were
adjusted for multiple testing with the Holm-Bonferroni method (Holm, 1979).
The growth response to climate was analysed for all residual chronologies (earlywood,
latewood) with bootstrapped response functions of monthly DRI values using the
software DENDROCLIM2002 (Biondi and Waikul, 2004). Since latewood has been
reported to be more sensitive to current year’s climate than earlywood or the entire
ring width (Lebourgeois, 2010), we used the residual latewood chronologies to study
possible temporal changes in growth reactions to drought. Hence, 35-year moving
response coefficients (r) of monthly DRI values from current April to August and
residual latewood chronologies were calculated using the bootRes package (Zang and
Biondi, 2012) of the software R (R core team, 2012). To display possible changes in DRI
over time and to be able to compare these changes with the results of the moving
response analysis, we calculated a 35-year moving average of the DRI. Subsequently,
the trend of this moving average was analyzed using the rank-based non-parametric
Mann-Kendall statistical test.
Chapter I
37
Results
Understory vegetation
The indicator values of the understory vegetation in 2011 revealed significant
differences in soil and site characteristics between the irrigated and control plots
(Figure 3). Permanently irrigated stands of Scots pine (Varen, Planige) and black pine
(Schluderns) were characterized by higher soil humidity, nutrient availability and
humus contents as well as lower light conditions at ground level. For European larch at
Schluderns, however, no differences in the soil characteristics were detected, but the
light conditions in the irrigated plot were lower than in the control plot. More than 30
years after the irrigation was stopped, we found indicator values at Lens and Mals
showing that the soil humidity was significantly higher, soils richer in nutrients and light
conditions lower than in the control plots. In addition, the humus contents at Lens
were significantly higher at the drained channel than at the control.
Tree growth
For the period from 1920 to 2009, the critical mean EPS value of 0.85 was reached for
permanently irrigated and control trees of Scots pine and European larch as well as for
control trees of black pine, but not for irrigated black pine trees, where the EPS value
was 0.82 (Table 2). Comparing the different tree species, the control and permanently
irrigated European larch trees (0.30-0.33) had the highest year-to-year variability in
tree growth, whereas the growth of black pine control trees was characterized by the
highest autocorrelation (0.81). In general, the tree-ring growth of permanently
irrigated trees showed a lower year-to-year variability (mean sensitivity) and a higher
autocorrelation than that of control trees (Table 2).
38
Chapter I
Figure .3. Mean indicator values (± SE, n = 12-16) of the understory vegetation in June/July 2011 at the
control plots and irrigated plots at the three sites in Valais/Switzerland (Varen, Planige, Lens) and the two
sites in Vinschgau/Italy (Mals, Schluderns). Differences between the irrigated and control plots were
tested using two-sample Wilcoxon-rank-sum tests at a 5% significance level (* P≤0.5).
Chapter I
39
Table 2. Summary of tree total ring-width statistics of the control trees and permanently irrigated trees of
Scots pine at two sites in Valais/Switzerland (Varen, Planige) and European larch and black pine at one site
in Vinschgau/Italy (Schluderns) for the period 1920-2009. Mean values ± SD (n = 12-18) of the basal area
2
-1
increment (BAI) in mm year and increment width (IW) in 1/100 mm as well as 1st-order autocorrelation
(AC1), mean sensitivity (MS), and expressed population signal (EPS) are shown, but the values for plots
where irrigation was stopped are not.
Specie
Status
BAI
Varen
Scots pine
Control
22 ± 7
53 ± 17
Irrigated
102 ± 24
Control
Planige
Scluderns
Scots pine
Europ. larch
Black pine
IW
a
Site
AC
a
b
b
MS
EPS
0.55
0.25
0.94
100 ± 27
0.70
0.19
0.88
32 ± 6
50 ± 11
0.62
0.21
0.90
Irrigated
84 ± 13
119 ± 30
0.76
0.15
0.85
Control
60 ± 22
85 ± 47
0.45
0.33
0.97
Irrigated
224 ± 72
159 ± 78
0.71
0.30
0.90
Control
141 ± 48
121 ± 57
0.59
0.28
0.97
Irrigated
284 ± 71
176 ± 76
0.81
0.18
0.82
The mean BAI of Scots pine was lower than that of the other two species. The
permanently irrigated Scots pine trees grew on average 102 mm2 year-1 at Varen
(control: 22 mm2 year-1) and 84 mm2 year-1 at Planige (control: 32 mm2 year-1) during
the period 1920 to 2009. In contrast, black pine at Schluderns had the highest mean
annual growth rates, namely 267 mm2 year-1 for the irrigated trees and 141 mm2 year-1
for the control trees (Table 2). The permanently irrigated European larch trees at
Schluderns grew on average 224 mm2 year-1 and the control 60 mm2 year-1. Annual
differences in the BAI between the permanently irrigated trees and their corresponding
control trees were significant for all tree species (P≤0.05), except for Scots pine at
Varen in the years 1921 and 1944 and for black pine at Schluderns in several (22) years
(Fig. 4, Table 2).
40
Chapter I
Figure 4. Mean basal area increment (BAI, n = 12-18) of control trees (grey line) and irrigated trees (black
line) at three sites in Valais/Switzerland (Varen, Planige, Lens) and two sites in Vinschgau/Italy (Mals,
Schluderns). Grey horizontal lines indicate insignificant relations of BAI between the irrigated and control
trees (P>0.05). Grey shaded areas symbolize periods after the irrigation was stopped at Lens (1983) and
Mals (1980).
The trees at the irrigated plots did not significantly differ (P>0.05) in age from those at
the control plots, except for Scots pine at Varen and European larch at Schluderns
(P≤0.01, Table 1). In 2010, the irrigated trees were significantly taller (P≤0.001) and had
a lager DBH (P≤0.01) than the control trees at all sites.
Scots pine clearly reacted to the irrigation stop in 1984 (Lens) and European larch to
that in 1980 (Mals) by decreasing BAI (Fig. 4). From 1988 to 1990, Scots pine grew at a
significantly lower rate than the corresponding control trees (P≤0.05). However, after
this period, the previously irrigated Scots pine trees had similar growth rates to the
corresponding control trees. For European larch, the BAI dropped to a comparable
level of increment to that of the control trees for the years 1981 to 1989. After 1989,
Chapter I
41
the larch trees growing along the drained channel again showed significantly higher
growth rates than the control trees (P≤0.05). Hence, both species had a lagged
recovery, with a delay in the response of radial growth to a sudden and drastic
reduction in water availability.
Drought index
In Valais and in Vinschgau, we found similar long-term trends in monthly water
availability via climate (DRI) for the period 1920 to 2009 (Figure 5). The Mann-Kendall
test showed a significantly (P≤0.05) increased water availability in April, May and June,
and decreasing trends in July and in August (Figure 5). Despite the similar trends in
both study regions, the water availability quantified by the DRI differed between the
two study regions. Namely, the DRI from April to August in Valais ranged between -62
and 0, whereas in Vinschgau it was between -26 and 8. However, the weather stations
in Valais are situated on the valley floor at about 600 m a.s.l., whereas in Vinschgau the
station is located on a north-facing slope at about 1300 m. Hence, this allowed only a
comparison in the trend of the DRI rather than in its magnitude.
42
Chapter I
Figure 5. Moving averages of monthly drought index (DRI) values for April to August in Valais/Switzerland
(thick line, left axis) and Vinschgau/Italy (thin line, right axis), centred around their mean values (dashed
horizontal line). Note the different scale of both axes. Each time window consists of 35 years starting with
the interval from 1920 to 1954 and subsequently shifting by one year until the year 2009. Lower DRI
values indicate a lower water availability. Rank-based non-parametric Mann–Kendall statistical test was
used to estimate the trend in monthly DRI values, Kendall’s tau (T) statistic is shown. For all T-values, twosided P-values ≤ 0.001, except for T Valais in July P-value ≤ 0.01.
Growth responses to climate
For Scots pine, the earlywood increment of control trees was significantly enhanced
(P≤0.05) by the high climatic water availability in January, March, April (Varen) and May
(Varen, Planige). The latewood increment of control trees was positively related to the
climatic water availability in January (Varen) and from April to July (Varen, Planige) as
Chapter I
43
well as negatively related to the climatic water availability in previous year’s July (Fig.
6). Furthermore, the growth responses of earlywood for the permanently irrigated
Scots pine (Varen, Planige) to climate were identical to those for the control trees, with
the only exception that the irrigated trees also responded positively to climatic water
availability in the previous year’s August. On the other hand, the latewood increment
of the permanently irrigated Scots pine was less sensitive to the climatic conditions
than that of the control trees, and responded only positively to moist climate in May
and July at Varen and August at Planige. In general, the highest growth responses of
Scots pine to climate were found in May (r=0.3 to 0.4) for both the control and the
irrigated trees.
For European larch, the growth of the control trees responded positively to the climatic
water availability in May (earlywood) and June/July (latewood), with the highest
correlations in May and July (P≤0.05, r=0.4 to 0.5). For the permanently irrigated trees
of European larch, the earlywood increment responded only positively to moist climate
in the previous year’s June and otherwise showed no responses to climate.
For black pine, the earlywood increment of the control trees (Schluderns) was
significantly related to the climatic conditions in the previous year’s November and the
current year’s February, April, May and June, whereas the latewood increment
responded positively to moist climate in previous year’s November and the current
year’s June, July and August (P≤0.05, Figure 6). Furthermore, the largest responses to
climate of the control trees were found for the period from May to August (r = 0.3 to
0.4). Similar to European larch, growth of permanently irrigated black pine also
appeared to be hardly related to climate, as the earlywood increment of black pine
responded positively only to the climatic water availability in the previous year’s
44
Chapter I
August and the latewood increment was negatively influenced by climatic water
availability in current year’s August.
Figure 6. Growth responses to climate of control trees and of permanently irrigated trees of Scots pine at
two sites in Valais/Switzerland (Varen, Planige) and of European larch and black pine at one site in
Vinschgau/Italy (Schluderns). Filled squares represent the significant (P≤0.05) coefficients of the relation
between a monthly drought index (DRI) and the entire increment width (IW) of (1) earlywood (EW) and,
(2) latewood (LW) for the period 1920 to 2009. Small letters represent the previous year’s DRI, and capital
letters the current year’s DRI. For positive response coefficients (r), only values larger than 0.2 are
displayed.
Chapter I
45
3.5 Temporal changes in growth responses to climate
For Scots pine in Valais, the response of latewood increment to monthly DRI was not
constant over time for the permanently irrigated and control trees at Varen and
Planige (Figure 7). From around 1920 to 1980, moist climatic conditions in summer
months (July/August) had a significantly positive effect on the latewood increment of
Scots pine (P≤0.05). However, from around 1980, the latewood was mainly enhanced
by high water availability in May and June. In Vinschgau, the impact of climate on the
latewood increment in late summer (July/August) has decreased for the control trees
of European larch at Schluderns since the beginning of the 21st century and for black
pine at Schluderns since the 1980s (Figure 7). On the other hand, the latewood
formation of irrigated European larch and black pine at Schluderns was not significantly
responding to the climatic water availability, but moist conditions in August had a
significant negative impact on the latewood of European larch, as did moist conditions
in May and July on black pine.
46
Chapter I
Figure 7. Response coefficients of the relationship between monthly drought index (DRI) values from April
to August and residual latewood chronologies of (1) control trees (on the left) and (2) permanently
irrigated trees (on the right) of Scots pine at two sites in Valais/Switzerland (Varen, Planige) and European
larch and black pine at one site in Vinschgau/Italy (Schluderns). Average coefficients were calculated for a
35-year time window starting with the interval from 1920 to 1954 and subsequently shifting by one year
until the year 2009. Filled bars indicate positive significant relationships, empty bars indicate significant
negative relationships (P≤0.05). The intensity of grey shading reflects the magnitude of the positive
response coefficients, ranging from 0.2 (light grey) to 0.6 (dark grey). Significant positive coefficients are
not shown (single bars) if adjacent time windows were not significant for the same month.
Chapter I
47
Discussion
Effects of long-term irrigation and drought
Long-term irrigation significantly enhanced radial tree growth for all the investigated
tree species throughout the study period from 1920 to 2009 (Table 2). An increase in
water availability is known to be linked to increased stomatal conductance (Larcher,
2003), resulting in a higher carbon uptake through photosynthesis, and finally to an
enhanced primary production (Wu et al., 2011; Ruehr et al., 2012). Furthermore, the
growth of the irrigated trees seems to be promoted by a prolongation of the growing
period in irrigated trees compared to trees at naturally dry sites (Eilmann et al., 2011).
In our study, the permanently irrigated Scots pine trees grew up to four times more
than corresponding control trees. In addition, the mean BAI of control trees of Scots
pine was extremely low (Table 2), even compared to Scots pine growth in similar
stands in terms of altitude and tree age at the Iberian Peninsula (Hereş et al., 2011).
This suggests that the growth of Scots pine is severely limited by water deficits in
Valais, which is in accordance with the results of Bigler et al. (2006). In Vinschgau, the
BAI of the permanently irrigated European larch trees was more than three times that
of the control trees and of the black pine trees about twice as high. For both species,
the growth rates of control trees were higher than those in similar forest stands
(altitude, tree age) in the Aosta valley in Italy (Lévesque et al., 2013). Comparing the
two species in Vinschgau, the tree growth of black pine at naturally dry sites was more
than twice as high than that of European larch. This is in accordance with the findings
of other studies revealing that the growth of black pine was less sensitive to water
availability than the growth of European larch (Eilmann and Rigling, 2012; Schuster and
Oberhuber, 2012; Lévesque et al., 2013).
48
Chapter I
The increased growth rates of the permanently irrigated trees might not only be a
direct result of more soil water being available. Additionally, the nutrient mobility is
known to be positively related to an increase in water availability, leading to enhanced
primary production (e.g. Larcher, 2003). As a result of both increased water and
nutrient availability, the total leaf area is increasing (Dobbertin et al., 2010) leading to
higher litterfall (Cotrufo et al., 2011), and litter decomposition (Anderson, 1991) and
thus, higher soil C and nutrient inputs further promoting tree growth. In our study,
these indirect effects of irrigation via enhanced soil C and nutrient inputs were
supported by the indicator values of the understory vegetation, reflecting the higher
soil humus contents and nutrient availability in the irrigated plots compared to the
corresponding control plots (Figure 3). Indeed, these differences were still evident 30
years after the irrigation stopped at Lens and Mals, with long-lasting changes in soil
characteristics.
Growth responses to climate
The analysis of growth responses to climate revealed differences between the
permanently irrigated trees and control trees (Figure 6). For the control trees of
European larch, radial growth was mainly related to the climatic water availability in
May for earlywood as well as in June and July for latewood. These very distinct growth
responses to dry climatic conditions (r=0.3 to 0.5) of larch contrasted with the lower
responses of the control trees of black pine (r=0.2 to 0.4) over several months (Figure
6). Namely, growth of the black pine control trees was responding to climate in
February and April to June (earlywood) as well as June to August (latewood). The
growth of larch, as a deciduous species, seems to depend more on the period of leaf
development (May) and on the current year’s climatic conditions than the growth of
Chapter I
49
evergreen species. This strong dependence is also reflected in the high year-to-year
variability (Table 2) of the radial growth of larch compared to black pine. In contrast,
evergreen species are able to rely on the previous year’s foliage and are
photosynthetically active as soon as the temperature is warm enough (Gruber et al.,
2010). The deciduous character of larch might explain why it seems to be much more
drought intolerant than other conifer species studied in inner-Alpine dry valleys
(Eilmann and Rigling, 2012; Schuster and Oberhuber, 2012; Lévesque et al., 2013).
Century-long irrigation nearly fully eliminated the climate sensitivity of the radial
growth of European larch and black pine compared to the control (Figure 6). The radial
growth of the permanently irrigated Scots pine was less limited than that of the control
trees at both sites in Valais, however, it was still depending on the climatic water
availability in the current and previous year. Consequently, the irrigation was not
sufficient to eliminate the growth responses of Scots pine to climate. An additional role
in the response of Scots pine to climate might have been the increase in tree size due
to the century-long irrigation and the resulting increase in annual increment. Namely,
irrigated trees were approximately twice as high than corresponding control trees
(Table 1). This increase in tree size might have led to an increase in the irrigated trees’
vulnerability to dry conditions, as has been reported in other studies (Lévesque et al.,
2013; Schuster and Oberhuber, 2013). Hence, the impact of tree size needs to be taken
into account when evaluating the effects of decreased water availability on tree
growth.
Temporal changes in growth sensitivity
In Valais as well as in Vinschgau, water availability in July, and even more so in August,
has decreased since the 1990s, while it has increased in April, May and June since the
50
Chapter I
1980s (Figure 5), which was reported already by Rebetez and Dobbertin (2004). In our
study, these changes in climatic conditions were reflected in the response of latewood
increment (Figure 7). For Scots pine, we found a shift in the growth response to climate
since about 1980 with (1) decreasing responses to water availability in July and August,
and (2) increasing responses in June and particularly May for both the control and the
irrigated trees. For the other two species, only the growth response of the control trees
to climatic conditions in August (black pine) vanished since the 1980s and those in July
(European larch) at the beginning of the 21st century. Most probably, water availability
in July and August in both valleys has decreased to such an extent (Figure 5) that the
trees of all three species completed most of their latewood growth already before the
start of the dry summer period in July. This drought avoidance strategy is also reported
in other studies (Zweifel et al., 2006; Eilmann et al., 2010; Lévesque et al., 2013). On
the other hand, the increase in growth sensitivity of Scots pine to the climatic water
availability in May might be a consequence of an earlier start of radial growth and
budburst (Menzel and Fabian, 1999; Studer et al., 2005; Gruber et al., 2010; Swidrak et
al., 2011) due to an increase in spring temperatures. As a consequence, a high water
availability in spring favored Scots pine growth at our sites. This increase in early spring
sensitivity for Scots pine growth was also reported by Schuster and Oberhuber (2012)
in the Inn-valley (Austria).
Growth response to an abrupt decrease in soil water availability
The sudden irrigation stop at Lens in Valais and Mals in Vinschgau led to a drop in the
BAI of Scots pine and European larch, followed by a period of reduced tree growth for
up to ten years (Figure 4). This strong decrease in BAI might be caused by a lower
assimilate availability (Galiano et al., 2011), as drought events are known to lead to a
Chapter I
51
reduction in stomatal conductance (Larcher, 2003), to needle shedding in order to
minimize water loss from transpiration (Pouttu and Dobbertin, 2000; Sánchez-Salguero
et al., 2012) and finally to lower primary production (Ciais et al., 2005; Wu et al., 2011).
Consequently, the Scots pine trees exposed to an abrupt reduction in water availability
showed significantly lower BAI values than the control trees for the period 1988 to
1990, while the BAI of European larch dropped to BAI similar to the corresponding
control trees for the period from 1981 to 1989 (Figure 4). This delayed responses may
hint to stored carbohydrates mixing with new assimilates during the first years after
the stop in irrigation. Although recent investigations stated that the variability of
carbohydrates during drought is less critical than originally suggested due to a higher
priority of carbon storage over tree growth (Muller et al., 2011; Palacio et al., 2014),
the stop in irrigation resembled an extreme and long-lasting drought event leading to
increased consumption of carbohydrates in subsequent years (Palacio et al., 2014).
However, after six years for Scots pine and nine years for larch, the surviving trees of
both species were able to adjust to the new conditions and subsequently grew with
similar (Scots pine), or even higher growth rates (European larch) compared to the
control.
These findings demonstrate the high phenotypic plasticity of individual trees, which
may finally lead to growth adjustments to drier conditions (Lloret et al., 2012), e.g., by
enhanced root growth to access water resources in deeper soil layers (Litton et al.,
2007). Thus, the adjustments of trees might diminish the impact of decreased water
availability on tree growth on longer time scales (Barbeta et al., 2013). Consequently,
short-term monitoring of tree growth responses to decreasing water availability might
lead to an over-estimation of the impact if they are projected onto longer time scales
(Leuzinger et al., 2011).
52
Chapter I
In addition, the improved soil conditions in the previously irrigated plots might have
diminished the effects of the irrigation stop, as the indicator values revealed
significantly higher soil humidity at Lens and Mals than in the corresponding control
plots (Figure 3). Accordingly, higher soil water availability, e.g., from greater soil waterholding capacities, have been found to buffer the effects of precipitation deficits on
tree growth (Rigling et al., 2002; Lévesque et al., 2013). The increased growth rates of
previously irrigated trees led, moreover, to larger tree dimensions (Table 1), and thus
to more shaded conditions underneath the tree canopy (Figure 3). Consequently, soil
evaporation might be lower in previously irrigated plots, resulting in higher soil water
availability, which further diminishes the effects of the irrigation stop on tree growth.
Conclusions
This study has revealed marked effects of century-long irrigation on the radial growth
of Scots pine in Valais (Switzerland) and of European larch and black pine in Vinschgau
(Italy). Such effects may be linked not only to the direct influence of water availability,
but also to indirect effects due to long-term improvements in soil characteristics
(increase in nutrient and humus contents), as shown by the indicator values of the
understory vegetation according to Landolt (2010). Our findings demonstrate that the
response of trees to abrupt changes in water availability is decreasing with time due to
growth adjustments, which are often unknown at longer time scale.
The growth responses to long-term changes in water availability highlight the need to
understand the adjustment mechanisms of each species at longer time scales. Further,
it appears necessary to include site-specific soil conditions to capture the indirect
effects of a changing water regime on the C and nutrient cycles. This is basic to
Chapter I
53
accurately predict future forest development and to adapt forest management and
conservation strategies under conditions of climate change.
Acknowledgments
We would like to thank Thomas Wohlgemuth for the help in the realization of the
vegetation surveys, Magdalena Nötzli and Anne Verstege for technical support and
Silvia Dingwall for language corrections. We would also like to thank David Frank,
Mathieu Lévesque, Alexander Bast, Martin Loos, Julia Nabel and Ellen Pflug for valuable
discussions. We are grateful to the forest service of the Swiss Canton Valais and the
local forest services in Leuk and Mals for the support and permissions. This study was
funded by Swiss National Science Foundation (National Research Programme NRP 61).
54
Chapter I
References
Allen CD, Breshears DD (1998) Drought-induced shift of a forest-woodland ecotone:
Rapid landscape response to climate variation. Proceedings of the National
Academy of Sciences. 95:14839-14842.
Allen CD, Macalady AK, Chenchouni H, Bachelet D, McDowell N, Vennetier M,
Kitzberger T, Rigling A, Breshears DD, Hogg EH (2010) A global overview of
drought and heat-induced tree mortality reveals emerging climate change risks
for forests. Forest Ecology and Management. 259:660-684.
Anderson JM (1991) The effects of climate change on decomposition processes in
grassland and coniferous forests. Ecological Applications. 1:326-347.
Barbeta A, Ogaya R, Peñuelas J (2013) Dampening effects of long-term experimental
drought on growth and mortality rates of a Holm oak forest. Global Change
Biology. 19:3133-3144.
Bigler C, Bräker OU, Bugmann H, Dobbertin M, Rigling A (2006) Drought as an inciting
mortality factor in Scots pine stands of the Valais, Switzerland. Ecosystems.
9:330-343.
Biondi F, Qeadan F (2008) A theory-driven approach to tree-ring standardization
defining the biological trend from expected basal area increment. Tree-ring
research. 64:81-96.
Biondi F, Waikul K (2004) DENDROCLIM2002: A C++ program for statistical calibration
of climate signals in tree-ring chronologies. Computers & Geosciences. 30:303311.
Bräker O (1981) Der Alterstrend bei Jahrringdichten und Jahrringbreiten von
Nadelhölzern und sein Ausgleich. Mitteilungen der Forstlichen
Bundesversuchsanstalt Wien. 142:75-102.
Breshears DD (2005) Regional vegetation die-off in response to global-change-type
drought. Proceedings of the National Academy of Sciences. 102:15144-15148.
Brühlhart H. 1999. Einfluss der Bewässerung auf das Jahrringwachstum von Kiefern.
Diplomarbeit. Berner Fachhochschule, Schweizerische Fachhochschule für
Holzwirtschaft, Biel.
Cech T, Perny B (1998) Kiefernsterben in Tirol. Forstschutz Aktuell. 22:12–15.
Ceppi P, Scherrer SC, Fischer AM, Appenzeller C (2012) Revisiting Swiss temperature
trends 1959-2008. International Journal of Climatology. 32:203-213.
Ciais P, Reichstein M, Viovy N, Granier A, Ogée J, Allard V, Aubinet M, Buchmann N,
Bernhofer C, Carrara A, Chevallier F, De Noblet N, Friend AD, Friedlingstein P,
Grünwald T, Heinesch B, Keronen P, Knohl A, Krinner G, Loustau D, Manca G,
Matteucci G, Miglietta F, Ourcival JM, Papale D, Pilegaard K, Rambal S, Seufert
G, Soussana JF, Sanz MJ, Schulze ED, Vesala T, Valentini R (2005) Europe-wide
reduction in primary productivity caused by the heat and drought in 2003.
Nature. 437:529-533.
Cook ER, Peters K (1981) The smoothing spline: a new approach to standardizing forest
interior tree-ring width series for dendroclimatic studies. Tree-ring Bulletin.
41:45-53.
Cook ER, Peters K (1997) Calculating unbiased tree-ring indices for the study of climatic
and environmental change. The Holocene. 7:361-370.
Cotrufo MF, Alberti G, Inglima I, Marjanović H, LeCain D, Zaldei A, Peressotti A,
Miglietta F (2011) Decreased summer drought affects plant productivity and
soil carbon dynamics in a Mediterranean woodland. Biogeosciences. 8:27292739.
Chapter I
55
Dobbertin M, Eilmann B, Bleuler P, Giuggiola A, Graf Pannatier E, Landolt W, Schleppi P,
Rigling A (2010) Effect of irrigation on needle morphology, shoot and stem
growth in a drought-exposed Pinus sylvestris forest. Tree Physiology. 30:346360.
Eilmann B, Buchmann N, Siegwolf R, Saurer M, Cherubini P, Rigling A (2010) Fast
response of Scots pine to improved water availability reflected in tree-ring
width and δ13C. Plant, Cell & Environment:1351-1360.
Eilmann B, Rigling A (2012) Tree-growth analyses to estimate tree species' drought
tolerance. Tree Physiology. 32:178–187.
Eilmann B, Zweifel R, Buchmann N, Graf Pannatier E, Rigling A (2011) Drought alters
timing, quantity, and quality of wood formation in Scots pine. Journal of
Experimental Botany. 62:2763-2771.
FAO. 2006. World reference base for soil resources, vol. 103. In FAO, World soil
resources reports, Rome.
Galiano L, Martínez-Vilalta J, Lloret F (2011) Carbon reserves and canopy defoliation
determine the recovery of Scots pine 4 yr after a drought episode. New
Phytologist. 190:750-759.
Gärtner H, Nievergelt D (2010) The core-microtome: A new tool for surface preparation
on cores and time series analysis of varying cell parameters.
Dendrochronologia. 28:85-92.
Gitlin AR, Sthultz CM, Bowker MA, Stumpf S, Paxton KL, Kennedy K, Muñoz A, Bailey JK,
Whitham TG (2006) Mortality gradients within and among dominant plant
populations as barometers of ecosystem change during extreme drought.
Conservation Biology. 20:1477-1486.
Gruber A, Strobl S, Veit B, Oberhuber W (2010) Impact of drought on the temporal
dynamics of wood formation in Pinus sylvestris. Tree Physiology. 30:490-501.
Hereş A-M, Martínez-Vilalta J, Claramunt López B (2011) Growth patterns in relation to
drought-induced mortality at two Scots pine (Pinus sylvestris L.) sites in NE
Iberian Peninsula. Trees. 26:621-630.
Holm S (1979) A simple sequentially rejective multiple test procedure. Scandinavian
Journal of Statististics. 6:65-70.
Holmes RL (1994) Computer-assisted quality control in tree-ring dating and
measurement. Tree-ring Bulletin. 43:69-78.
IPCC. 2013. Climate change 2013: The physical science basis. Contribution of working
group I to the fifth assessment report of the intergovernmental panel on
climate change. Eds. Stocker TF, Qin D, Plattner G-K, Tignor M, Allen SK,
Boschung J, Nauels A, Xia Y, Bex V, Midgle PM. Cambridge University Press,
Cambridge, United Kingdom and New York, NY, USA, 1535 pp.
Landolt E, Bäumler B, Erhardt A, Hegg O, Klötzli F, Lämmler W, Nobis M, RudmannMaurer K, Schweingruber F, Theurillat J-P, Urmi E, Vust M, Wohlgemuth T
(2010) Flora indicativa: Ökologische Zeigerwerte und biologische Kennzeichen
zur Flora der Schweiz und der Alpen. Haupt Verlag, Bern, Switzerland.
Larcher W (2003) Physiological plant ecology: Ecophysiology and stress physiology of
functional groups, 4th edn. Springer, Berlin Heidelberg New York.
Lebourgeois F (2010) Climatic signals in earlywood, latewood and total ring width of
Corsican pine from western France. Annals of Forest Science. 57:155-164.
Lebourgeois F, Rathgeber CBK, Ulrich E (2010) Sensitivity of French temperate
coniferous forests to climate variability and extreme events (Abies alba, Picea
abies and Pinus sylvestris). Journal of Vegetation Science. 21:364-376.
56
Chapter I
Leuzinger S, Luo Y, Beier C, Dieleman W, Vicca S, Körner C (2011) Do global change
experiments overestimate impacts on terrestrial ecosystems? Trends in
Ecology & Evolution. 26:236-241.
Lévesque M, Saurer M, Siegwolf R, Eilmann B, Brang P, Bugmann H, Rigling A (2013)
Drought response of five conifer species under contrasting water availability
suggests high vulnerability of Norway spruce and European larch. Global
Change Biology. 19:3184-3199.
Litton CM, Raich JW, Ryan MG (2007) Carbon allocation in forest ecosystems. Global
Change Biology. 13:2089-2109.
Lloret F, Escudero A, Iriondo JM, Martínez-Vilalta J, Valladares F (2012) Extreme
climatic events and vegetation: the role of stabilizing processes. Global Change
Biology. 18:797-805.
Lotter AF, Appleby PG, Bindler R, Dearing JA, Grytnes J-A, Hofmann W, Kamenik C, Lami
A, Livingstone DM, Ohlendorf C, Rose N, Sturm M (2002) The sediment record
of the past 200 years in a Swiss high-alpine lake: Hagelseewli (2339 m a.s.l.).
Journal of Paleolimnology. 28:111–127.
MacKay SL, Arain MA, Khomik M, Brodeur JJ, Schumacher J, Hartmann H, Peichl M
(2012) The impact of induced drought on transpiration and growth in a
temperate pine plantation forest. Hydrological Processes. 26:1779-1791.
McDowell N, Pockman WT, Allen CD, Breshears DD, Cobb N, Kolb T, Plaut J, Sperry J,
West A, Williams DG, Yepez EA (2008) Mechanisms of plant survival and
mortality during drought: why do some plants survive while others succumb to
drought? New Phytologist. 178:719-739.
Menzel A, Fabian P (1999) Growing season extended in Europe Nature. 397:659-659.
Mosteller F, Tukey JW. 1977. Data analysis and regression. Addison-Wesley, Reading.
Oberhuber W, Hofbauer W, Kofler W (2001) Mortality and growth anomalies of Scots
pine (Pinus sylvestris L.) growing on drought exposed sites at the Tschirgant
Landslide (Tyrol). Berichte des Naturwissenschaftlich-Medizinischen Vereins in
Innsbruck 88: 87-97.
Plaut JA, Yepez EA, Hill J, Pangle R, Sperry JS, Pockman WT, McDowell NG (2012)
Hydraulic limits preceding mortality in a piñon-juniper woodland under
experimental drought. Plant, Cell & Environment. 35:1601-1617.
Pouttu A, Dobbertin M (2000) Needle-retention and density patterns in Pinus sylvestris
in the Rhone Valley of Switzerland: comparing results of the needle-trace
method with visual defoliation assessments. Canadian Journal of Forest
Research. 30:1973–1982.
R core team. 2012. R: A language and environment for statistical computing. R
Foundation for Statistical Computing, Vienna, Austria.
Rebetez M, Dobbertin M (2004) Climate change may already threaten Scots pine
stands in the Swiss Alps. Theoretical and Applied Climatology. 79:1-9.
Rigling A, Bigler C, Eilmann B, Feldmeyer-Christe E, Gimmi U, Ginzler C, Graf U, Mayer
P, Vacchiano G, Weber P, Wohlgemuth T, Zweifel R, Dobbertin M (2013)
Driving factors of a vegetation shift from Scots pine to pubescent oak in dry
Alpine forests. Global Change Biology. 19:229-240.
Rigling A, Bräker O, Schneiter G, Schweingruber F (2002) Intra-annual tree-ring
parameters indicating differences in drought stress of Pinus sylvestris forests
within the Erico-Pinion in the Valais (Switzerland). Plant Ecology. 163:105-121.
Ruehr NK, Martin JG, Law BE (2012) Effects of water availability on carbon and water
exchange in a young ponderosa pine forest: Above- and belowground
responses. Agricultural and Forest Meteorology. 164:136-148.
Chapter I
57
Sánchez-Salguero R, Navarro-Cerrillo RM, Camarero JJ, Fernández-Cancio Á (2012)
Selective drought-induced decline of pine species in southeastern Spain.
Climatic Change. 113:767-785.
Schuster R, Oberhuber W (2012) Drought sensitivity of three co-occurring conifers
within a dry inner Alpine environment. Trees
Schuster R, Oberhuber W (2013) Age-dependent climate-growth relationships and
regeneration of Picea abies in a drought-prone mixed-coniferous forest in the
Alps. Canadian Journal of Forest Research. 43:609-618.
Schweingruber FH (1983) Der Jahrring; Methodik, Zeit und Klima in der
Dendrochronologie. Haupt Verlag, Bern
Speer JH (2010) Fundamentals of tree-ring research. The University of Arizona Press,
Tucson, Arizona.
Studer S, Appenzeller C, Defila C (2005) Inter-annual variability and decadal trends in
Alpine spring phenology: A multivariate analysis approach. Climatic Change.
73:395-414.
Swidrak I, Gruber A, Kofler W, Oberhuber W (2011) Effects of environmental conditions
on onset of xylem growth in Pinus sylvestris under drought. Tree Physiology.
31:483-493.
Thornthwaite C (1948) An approach toward a rational classification of climate.
Geographical review. 38:55-94.
Weber P, Bugmann H, Rigling A (2007) Radial growth responses to drought of Pinus
sylvestris and Quercus pubescens in an inner-Alpine dry valley. Journal of
Vegetation Science. 18:777-792.
Wu Z, Dijkstra P, Koch GW, PeÑUelas J, Hungate BA (2011) Responses of terrestrial
ecosystems to temperature and precipitation change: a meta-analysis of
experimental manipulation. Global Change Biology. 17:927-942.
Zang C, Biondi F (2012) Dendroclimatic calibration in R: The bootRes package for
response and correlation function analysis. Dendrochronologia. 31:68-74.
Zang C, Pretzsch H, Rothe A (2011) Size-dependent responses to summer drought in
Scots pine, Norway spruce and common oak. Trees. 26:557-569.
Zweifel R, Zimmermann L, Zeugin F, Newbery DM (2006) Intra-annual radial growth and
water relations of trees: implications towards a growth mechanism. Journal of
Experimental Botany. 57:1445-1459.
58
Chapter I
Chapter II
59
Chapter II Long-term adjustments in growth and
physiology of mature Scots pine and European larch to
drought and irrigation
To be submitted as:
Feichtinger, Linda M.1; Mathieu Lévesque1; Rolf Siegwolf2; Arthur Gessler1; Buchmann,
Nina3; Rigling, Andreas1. Long-term adjustments in growth and physiology of mature
Scots pine and European larch to drought and irrigation.
1
Swiss Federal Research Institute for Forest, Snow and Landscape, WSL, Zürcherstrasse 111,
8903 Birmensdorf, Switzerland
2
Paul Scherrer Institute PSI, Lab for Atmospheric Chemistry, Stable Isotopes and Ecosystem
Fluxes, 5232 Villigen, Switzerland
3
ETH Zürich, Institute of Agricultural Sciences, Universitätsstrasse 2, 8092 Zürich, Switzerland
60
Chapter II
Abstract
The adjustment mechanisms of trees to changes in water availability on long time
scales are poorly understood. Yet they are crucial to improve model-based estimations
and predictions of future forest development under a changing climate. To examine
the long-term adjustments of tree growth and physiology to contrasting water supply,
we compared irrigated Scots pine (Pinus sylvestris L.) and European larch (Larix decidua
Mill.), growing along open water channels since their germination (> 90 years), with unirrigated control trees at three sites in dry inner-Alpine valleys. We combined
measurements of basal area increment (BAI), tree height and carbon isotope
discrimination (Δ13C) to assess the growth and physiology of gas exchange of pine and
larch for the period from 1970 to 2009. At two of our study sites, the water channels
had been drained and we were thus able to investigate the adjustment capacity of
both tree species to an immediate and essential decrease in water availability. The BAI
and height of pine and larch increased with water availability at all three sites.
However, the response in Δ13C to irrigation was highly site specific. At one of our sites
(Scots pine), we found a distinct increase in Δ13C compared to the control, suggesting
increased stomatal conductance with irrigation. In contrast, at the two other sites, we
found no (Scots pine) or only minor increases (larch) in Δ13C compared to the control.
With an increase in BAI and height the negative pressure of the xylem and thus
stomatal conductance might decrease, leading to no or minor differences in Δ13C
between irrigated and control trees. This suggests that the adjustments of Scots pine
and larch to the local water availability occurred mainly through modifications in radial
and height increment, whereas variations in physiology appear to be highly sitespecific. In response to the irrigation stop, larch demonstrated a higher sensitivity in
radial growth and physiology than Scots pine. Surviving trees of Scots pine and larch,
Chapter II
61
however, were able to adjust to the new and drier conditions and showed similar radial
growth rates and Δ13C values than control trees after about ten to 15 years after the
irrigation stopped.
Introduction
The combined effect of low water availability and high temperatures is a key factor for
limitations of primary production (Zhao and Running 2010) and alterations of
ecosystem processes (Anderegg et al. 2012). In mountain regions of Central Europe,
the rise in temperature is exceeding the global average (Ceppi et al. 2012), leading to
higher evapotranspiration and water demand of the vegetation. As a consequence,
forest ecosystems show higher mortality rates, diminished growth, and vegetation
shifts in species composition, especially at dry sites and at the southern distribution
limit of a species (Rigling et al. 2013; Vacchiano et al. 2011). Since climatic conditions in
the European Alps are predicted to become even drier and warmer (Gobiet et al.
2013), a thorough mechanistic understanding of tree responses towards increasing
drought is thus crucial in order to predict the future state and performance of forest
ecosystems.
Estimations and predictions of drought effects on forests are mostly accomplished
using models, which are based on and validated with experimental data (Luo et al.
2008). However, these experiments are mostly restricted to short time scales (Cotrufo
et al. 2011; Eilmann et al. 2010; MacKay et al. 2012). As a result, the accuracy of any
such model prediction strongly depends on how well findings from short-scale
experiments reflect responses to drought on longer time scales. Recent studies found
that the response of trees to changes in water availability is often non-linear over
62
Chapter II
longer time periods, due to the ability of trees and whole forest ecosystems to adjust
to the new conditions (Barbeta et al. 2013; Beier et al. 2012; Leuzinger et al. 2011).
However, investigations focusing on the long-term responses (> 10 years) of trees to
variations in the water regime including adjustment processes are scarce.
Trees can instantly respond to a low water availability via a decrease in stomatal
conductance (g s ) in order to reduce the loss of water to the atmosphere (Larcher
2003). This short-term response leads to a reduced gas exchange, limiting the rates of
both water vapour efflux and CO 2 influx via the stomata and resulting finally in a
decreased carbon aquisition. On seasonal and annual time scales, however, trees
adjust to drought. This can be achieved by adapting the hydraulic achitecture, via
modifications of radial growth (Feichtinger et al. 2014), shoot-to-root ratio (Litton et al.
2007), leaf morphology (Dobbertin et al. 2010; Feichtinger et al. 2015) and/or wood
anatomy (Eilmann et al. 2009).
The basal area increment (BAI) has been found to be a valuable trait to quantify the
effects of long-term changes in water availability on radial tree growth (Barbeta et al.
2013; Feichtinger et al. 2014). By using the BAI instead of the raw ring-width series,
growth of trees with different DBH classes can be compared. However, tree growth is
influenced not only by the amount of water, but also by a variety of biotic and abiotic
factors, e.g. intra-specific competition, nutrient availability or pathogens (Dobbertin
2005; Speer 2010). Thus, variations in the BAI are not always easy to interpret.
The stable carbon isotope composition (δ13C) in tree rings is a straightforward tool to
investigate the past effects of drought on gas exchange and stomatal responses to
changes in the water regime (Eilmann et al. 2010; Gessler et al. 2014; Lévesque et al.
2013). The δ13C in tree rings depends as a first approximation on the ratio of ambient
(c a ) to leaf intercellular (c i ) CO 2 concentration and therefore jointly reflects the rates of
Chapter II
63
carbon assimilation during photosynthesis and stomatal conductance (Farquhar et al.
1989; Farquhar et al. 1982). Due to either low soil water availability and/or dry air and
thus increased vapour pressure deficit, reduced stomatal conductance decreases the
supply and fixation of CO 2 under dry conditions. As a consequence c i decreases leading
to a decrease in carbon isotope discrimination (∆13C) and thus to an increase in δ 13C.
Therefore, δ13C in tree rings depends on the carbon isotope discrimination affected by
g s and A as well as on post-photosynthetic discrimination during wood formation
(Werner and Gessler 2011), but also on the δ13C of the source CO 2 .
Currently the mean δ13C of atmospheric CO 2 is ~ − 8‰. This value, however, is
becoming more negative over time (~0.02 ‰ year–1) caused by progressively increasing
emissions of
13
C-depleted CO 2 during fossil fuel combustion (Affek and Yakir 2014).
Since this trend biases the climatic and physiological signal in the tree-ring record, it is
commonly removed and the discrimination ∆13C is used as indicator of physiological
responses to changes in CO 2 and water fluxes, and is thus useful in global change
studies.
In this study, we aimed to identify the long-term adjustments to variations in water
availability of tree growth and physiology in dry forest ecosystems in inner-Alpine
valleys in Switzerland (Valais) and in Italy (Vinschgau). Changes in the growth rates
were analysed by the assessment of the BAI. Tree physiological responses were
investigated with the help of stable carbon isotopes in tree rings. We analysed Scots
pine (Pinus sylvestris L.) in Valais and European larch (Larix decidua Mill.) in Vinschgau
growing under naturally dry conditions (control) with trees that had been (i)
permanently irrigated since establishment (>90years), i.e. growing along historical
open water channels, and (ii) exposed to a sudden stop in irrigation, where the
channels were drained. To assure that the irrigated trees took up the water from the
64
Chapter II
channels, we further analysed the differences in xylem water δ18Ο in 2011 of the
irrigated and control trees. We addressed the following objectives: (1) to analyse the
long-term adjustments in growth and physiology of Scots pine and larch to irrigation,
and (2) to quantify the short-term responses and long-term adjustments in growth and
physiology towards an abrupt increase in drought exposure due to an irrigation stop.
Material and methods
Study sites
We selected two Scots pine stands in Valais, Switzerland, and one European larch stand
in Vinschgau, Italy (Figure 1). All sites are located on south-facing slopes (15-37°), with
shallow rendzic leptosols (FAO 2006), between 800-1000 m a.s.l. in Valais and at about
1200 m a.s.l. in Vinschgau (Table 1). In Valais, the forest stands have naturally
regenerated over the last century, while the forest in Vinschgau was planted at the end
of the 19th century. The climate in both valleys is characterised by low annual
precipitation, with 664 mm on average in Valais and 657 mm in Vinschgau (Figure 2).
To examine the adjustments in tree growth and physiology to contrasting water
availability, we compared trees growing under naturally dry conditions (control) with
trees that had been permanently irrigated since establishment (>90 years), growing
along open water channels at two Scots pine sites in Valais (Varen, Lens) and one
European larch site in Vinschgau (Mals). At two of these sites (Lens and Mals), the
channels were however drained in 1980 (Mals) and 1984 (Lens) and could thus be used
to quantify the effects of a sudden stop in irrigation.
The irrigation channels had been constructed to lead melt water from higher altitudes
to agricultural lands in the valleys from the end of April until the beginning of October
Chapter II
65
(Leibundgut and Kohn 2014). The channels in Valais can be dated back to the 12th
(Varen) and 15th century (Lens), while the one in Vinschgau (Mals) is of unknown age.
As these channels are mainly open ditches, they passively lose a large amount of water
to the surrounding vegetation. Only for the site Lens, the water loss could be
quantified, namely about 80 l/s or 27% along a 3 km transect (Brühlhart 1999). To
avoid this passive water leakage, the open channels have been replaced by
impermeable pipes at two of our study sites: the irrigation stopped in Lens in 1984 and
in Mals in 1980. To guarantee that the trees categorized as irrigated did benefit from
the passive irrigation, we only chose trees growing less than one meter away from the
water channels. As controls, we sampled trees growing nearby the irrigation channels,
but unaffected from the irrigation and under naturally dry conditions (upslope or in
proximity of the irrigated plots).
Figure 1. Location of the
two
study
regions
in
Central Europe (a), Valais in
Switzerland
(b)
and
Vinschgau in Italy (c). The
study sites are marked as
dots, the weather stations
as triangles.
66
Chapter II
Table 1. Site description and characteristics of Scots pine in Valais/Switzerland (Varen, Lens) and European
larch in Vinschgau/Italy (Mals), mean values ± SE of tree age, diameter at breast height (DBH), and tree
height measured in 2010 for trees used for the isotope measurements (n = 4) and the entire mean
chronologies (n = 12-18) including the isotope trees. Differences between control trees and irrigated trees
of the entire mean chronologies were tested with a two-sample Wilcoxon-rank-sum test (values with
different letters: P≤0.05).
Si te
Sta tus
Speci es
Age
N=4
Va ren
Control
Scots pi ne
Irri ga ted
Lens
Control
Scots pi ne
Irri ga ti on s top*
Ma l s
Control
Europea n l a rch
Irri ga ti on s top**
DBH (cm)
N=12-16
N=4
N=14
Hei ght (m)
N=4
N=15-18
118 ± 5
106 ± 4a
19 ± 1
18 ± 1a
5.4 ± 0.0
5.5 ± 0.2a
120 ± 14
135 ± 14a
38 ± 6
44 ± 3b
11.3 ± 1.0
11.0 ± 0.7b
119 ± 25
118 ± 12a
33 ± 3
31 ± 2a
12.1 ± 1.0
11.6 ± 0.5a
103 ± 4
103 ± 4a
39 ± 2
38 ± 1b
16.9 ± 1.0
16.2 ± 0.5b
103 ± 3
105 ± 1a
31 ± 2
31 ± 1a
21.8 ± 1.0
23.5 ± 0.4a
105 ± 4
114 ± 3a
58 ± 7
63 ± 3b
29.5 ± 1.0
27.1 ± 0.8b
* i n 1984
** i n 1980
Climate data
Climate data for the site in Vinschgau were obtained from the weather stations in
Marienberg (1323 m a.s.l.), about 3 km away from the site Mals. In Valais, the closest
weather station for temperature data is located in Sion (492 m a.s.l.), 9 km away from
the site Lens and 20 km from Varen. As precipitation has a high spatial variability, we
chose a weather station located in Sierre (539 m a.s.l.), which is closer to our study
sites, but which only records precipitation data. This station is about 8 km away from
Varen and Lens (Figure 2). Temperature data were adjusted to the differences in
altitude between the weather stations and the study sites using monthly lapse rates
published in Lotter et al. (2002). To order to estimate the combined effects of low
water availability and high evaporative demand on tree growth and physiology, we
calculated a drought index (DRI) according to Thornthwaite (1948):
𝐷𝐷𝐷𝐷𝐷𝐷 = 𝑃𝑃 − 𝑃𝑃𝑃𝑃𝑃𝑃,
(1)
Chapter II
67
where P is the monthly precipitation and PET the potential evapotranspiration. Low
and high water availability are characterised by negative and positive values of the
drought index, respectively. We have chosen this relatively simple index, as it has been
proven to highly correlate with tree growth in Valais (Dobbertin et al. 2010; Eilmann et
al. 2006; Weber et al. 2007).
Figure 2. Monthly mean precipitation (bars) and temperature (line) for Valais in Switzerland and for
Vinschgau in Italy for the period from 1961 to 1990. Temperature data were adjusted to the differences in
altitude between the weather stations and the study sites using monthly lapse rates published in Lotter et
al. (2002).
Radial growth
We sampled 15 to 18 dominant (>70 years old) trees with no visible damage at each
irrigated and control plot in 2010 and took two increment cores perpendicular to each
other with an increment borer at breast height (1.3 m) from each tree. For the irrigated
trees, we sampled only those growing directly at the water channel in order to
guarantee a high water availability. For each tree sampled, diameter at breast height
(DBH) and tree height were measured. To improve the visibility of the tree rings, the
surfaces of the air-dried increment cores were prepared with a core microtome
68
Chapter II
(Gärtner and Nievergelt 2010). Subsequently, the ring width was measured using a
LINTAB digital positioning table and the software TSAP-Win (Rinntech, Heidelberg,
Germany). The individual tree-ring series were cross-dated visually and subsequently
verified statistically by using the software COFECHA (Holmes 1994). We derived series
of annual basal area increments (BAI) from raw ring width in order to compare the
absolute growth rates between the irrigated and control trees without the confounding
factors of DBH class. The BAI was calculated from bark to pith according to Biondi and
Qeadan (2008) and the BAI chronologies were averaged per treatment and per site.
Stable isotope analysis
For the measurements of the δ13C, we selected four trees per species from all sampled
trees of each irrigated and control plot that were the most synchronous and best
correlated with the mean BAI chronology per plot. For each of these selected trees, an
extra core was taken, and the resin was extracted in 95% ethanol for 24 h using a
Soxhlet apparatus. Subsequently, the cores were washed in boiling distilled water and
air dried. While the latewood of tree rings is mostly influenced by climatic conditions in
the current year, earlywood is partly influenced by climatic conditions in the previous
year since stored carbohydrates are used for its formation (Helle and Schleser 2004).
Thus, we carried out δ13C analyses on latewood only. The latewood was split from each
annual ring with a scalpel for the period from 1970 to 2009, and samples were pooled
per year and by each irrigated and control plot. To estimate the variance of δ13C in the
wood, we analysed the isotopic composition separately for every individual tree every
ten years (1970, 1980, 1990, 2000). For the trees growing along the channel at the site
Lens, the amount of wood in 1990 (after the irrigation stopped) was however too small
to be split, and therefore it was not possible to calculate the standard error in 1990. All
Chapter II
69
latewood samples were milled and homogenized using an ultra-centrifugal mill (ZM,
200, Retsch, Haan, Germany).
For the analysis of δ13C, 0.6 to 0.8 mg of wood powder from each sample was weighed
into a tin capsule and burned to CO 2 at 1020°C in an elemental analyser (EA-1110,
Carlo Erba Thermoquest, Milan, Italy) connected to a Delta S mass spectrometer by a
CONFLO II (Finnigan MAT, Bremen, Germany). The obtained isotopic values are given in
delta notation in ‰ relative to the international standard:
𝑅𝑅𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠
𝛿𝛿𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠 = �𝑅𝑅
𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠
− 1�,
(2)
where 𝑅𝑅𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠 is the molar fraction of the 13C/12C ratio of the sample and 𝑅𝑅𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠
that of the international standard. For carbon, we related our values to the VPDB
(Vienna Pee Dee Belemnite) standards. The precision of the analysis was estimated as
the standard deviation of laboratory cellulose standards, which was ≤ 0.10‰ for δ13C.
Correcting for δ13C atmospheric CO 2
The δ13C series were corrected for the depletion in δ13C in atmospheric CO 2 of recent
decades by calculating the carbon isotopic discrimination (∆13C) using the simplified
equation 3, adapted from Farquhar et al. (1982):
∆13 𝐶𝐶 (‰) =
𝛿𝛿 13 𝐶𝐶𝑎𝑎𝑎𝑎𝑎𝑎 −𝛿𝛿 13 𝐶𝐶𝑤𝑤𝑤𝑤𝑤𝑤𝑤𝑤
,
1+𝛿𝛿 13 𝐶𝐶𝑤𝑤𝑤𝑤𝑤𝑤𝑤𝑤
(3)
where δ 13C atm is the isotopic value of atmospheric CO 2 and δ13C wood is the isotopic
value of the wood. For this purpose, we used δ13C atm values published in McCarroll and
Loader (2004) for the period from 1970 to 2003, and values for the period 2004 to
2009 available online (http://www.esrl.noaa.gov/gmd/).
70
Chapter II
Xylem water δ18O
To proof that irrigated and control trees took up different source water δ18O, we
measured the oxygen isotopic signature (δ18O ) in xylem water of each sampled tree
(four trees per irrigated and per non-irrigated control plots) for the period from June to
August 2011. The sampling was performed at two sites: at Varen, where the irrigation
continued until the present day, and at Lens (Figure 3b), where the irrigation was
stopped in 1984 and re-started in 2010, i.e., one year before the xylem water samples
were taken. We hypothesized that xylem δ18O values of trees growing close to the
water channels differ from those of control trees growing further away, supporting our
experimental set-up. Water was extracted from two twigs per tree using the cryogenic
vacuum extraction method (West et al. 2006) at about 100°C and 0.01 mbar
Data analysis
The adjustments in BAI and Δ13C of irrigated and corresponding control trees were
compared using the Wilcoxon-Rank sum test. Furthermore, we investigated the
13
C
discrimination in latewood (Δ13C), related to changes in stomatal conductance and/or
photosynthesis for the irrigated and corresponding control trees, by correlating Δ13C
with the mean DRI during spring (March to May), early summer (June, July), and late
summer (July, August). Subsequently, we repeated these correlations between the BAI
and the DRI. All statistical analysis were performed using the software R (R
Development Core Team 2011).
Chapter II
71
Results
Long-term differences in water availability
The irrigated and the control trees took up different water sources in 2011 (P≤0.001,
Figure 3), as shown by the lower δ18O in the xylem water for the irrigated trees for the
period from June to September 2011. At all three sites, the irrigation led to a
significantly enhanced radial tree growth (P<0.001, Figures 4, 5, 6), as well as to taller
and larger (DBH) trees than the corresponding control trees in 2010 (P<0.05, Table 1).
18
Figure 3. δ O in the xylem water of the trees growing along the water channels and the control trees (n =
4) at Varen (a) and Lens (b) in Valais (Switzerland). Samples were collected during the growing season of
18
2011. Monthly mean δ O values were connected with dotted (control) and black lines (irrigated). At both
18
sites, the yearly δ O values of irrigated and control trees (n = 12) were significantly different (Wilcoxonrang-sum test, P≤0.001).
72
Chapter II
Irrigation led to an increase in Δ13C compared to the controls but with site-specific
variations. At the site Varen, irrigation led only to a slight but non-significant increase
in the mean Δ13C values (Figure 4c, d). In contrast, for pine trees at the site Lens, the
higher water availability led to significantly higher values of Δ13C (P≤0.001, Figure 5c,
d). The irrigated larch trees at the site Mals also showed a significant (P≤0.05) but
relatively moderate increase in mean Δ13C values (Figure 6c, d). This increase in Δ13C of
larch was most pronounced in the years 1971, 1975-76 and 1979 (Figure 6c).
13
Figure 4. Mean annual basal area increments (BAI) ± SE (n = 4), and Δ C ± SE every tenth year (n = 4)
measured in latewood of the irrigated and control trees of Scots pine at the site Varen in Valais (a, c) as
well as the mean drought index (DRI) for the period from March to May and from June to August (e).
18
13
Mean values of BAI, Δ O and Δ C for the period from 1970 to 2009 are displayed as box plots (b, d)
(control: empty boxes, irrigated: filled boxes), different letters indicate significant differences between
mean values of irrigated and control trees (Wilcoxon-rang-sum test, P≤0.05).
Chapter II
73
Abrupt decrease in water availability
The stop in irrigation led to a significant decrease in BAI and Δ13C for both pine (Lens)
and larch (Mals, Figures 5, 6). For pine, the mean BAI of the trees growing along the
channel responded with a lag of two years to the sudden reduction in water
availability, and gradually decreased to a similar level as the one of the control trees
(Figure 5a). In contrast, the BAI of larch decreased immediately to a similar level of BAI
as that of the control trees (Figure 6a). At the end of the study period, the BAI of both
species showed similar or even higher BAI than corresponding control trees.
The Δ13C of pine and larch responded immediately to the stop in irrigation (Figures 5,
6). Specifically, Δ13C of Scots pine decreased in response to the lower water availability
but still remained on a higher level than the control trees until 2009. In contrast, Δ13C
of larch trees first decreased to a lower level than that of the control trees for about
two years after the stop in irrigation.
Similarly, Δ13C of pine and larch responded immediately to the stop in irrigation
(Figures 5, 6). Specifically, Δ13C of Scots pine decreased in response to the lower water
availability but still remained on a higher level than the control trees for about five
years. In contrast, Δ13C of larch trees decreased to a lower level than the control trees
for about two years after the stop in irrigation. Subsequently, tress of both pine and
larch showed a similar signal of 13C than control trees.
74
Chapter II
13
Figure 5. Mean annual basal area increments (BAI) ± SE (n = 4), and Δ C ± SE every tenth year (n = 4)
measured in latewood of the irrigated and control trees of Scots pine at the site Lens in Valais (a, c) as well
as the mean drought index (DRI) for the period from March to May and from June to August (e). Irrigation
13
was stopped in 1984 (dashed vertical line). Mean values of BAI and Δ C for the period before (1970-1983)
and after (1984-2009) the irrigation stop are displayed as box plots (b, d) (control: empty boxes, irrigated:
filled boxes), different letters indicate significant differences between mean values of irrigated and control
trees (Wilcoxon-rang-sum test, P≤0.05).
Chapter II
75
13
Figure 6. Mean annual basal area increments (BAI) ± SE (n = 4), and Δ C ± SE every tenth year (n = 4)
measured in latewood of the irrigated and control trees of European larch at the site Mals in Vinschgau (a,
c) as well as the mean drought index (DRI) for the period from March to May and from June to August (e).
18
13
Irrigation was stopped in 1980 (dashed vertical line). Mean values of BAI, Δ O and Δ C for the period
before (1970-1979) and after (1980-2009) the irrigation stop are displayed as box plots (b, d) (control:
empty boxes, irrigated: filled boxes), different letters indicate significant differences between mean values
of irrigated and control trees (Wilcoxon-rang-sum test, P≤0.05).
76
Chapter II
Discussion
Long-term responses of growth and physiology to water availability
At Varen and Lens, the mean δ18O of the xylem water for irrigated trees measured in
2011 was significantly lower than that of the control trees (Figure 3), clearly indicating
different source water taken up by these trees and proofing the assumptions for our
experimental set-up. Water of higher elevations and melt water is typically highly
depleted in
18
O compared to precipitation water (Siegenthaler and Oeschger 1980;
Tang and Feng 2001), leading to lower δ18O in the water taken up by irrigated trees. In
contrast, the xylem water δ
18
O of control trees resembled mainly the signal in
precipitation water (data not shown).
Under low water availability both species need to reduce stomatal conductance in
order to minimize water loss at the cost of decreasing carbon uptake (Larcher 2003),
resulting in low values of Δ13C for pine and larch in the control plots (Farquhar et al.
1989, Figs. 4, 5, 6). For the irrigated trees, we would thus expect a less conservative
stomatal control (McDowell et al. 2008), and an increase in Δ13C compared to the
control. This was indeed found for Scots pine at the site Lens (Figure 5c, d), with
irrigated trees showing on average a 2.2‰ higher carbon discrimination than control
trees. Similar results were found by Eilmann et al. (2010) conducting a short-term
irrigation experiment for three years in Valais.
On the other hand, the irrigated trees of pine at Varen showed no and larch at Mals
only slightly higher Δ13C values than those of their controls (Figures 4, 6). Similar
findings were found by Addington et al. (2006), reporting equal leaf and canopy
conductance for Pinus palustris when comparing dry versus mesic sites. They further
related these findings to differences in hydraulic architecture between sites and
Chapter II
77
primarily to an increased tree height at the mesic site. At all our study sites, irrigated
trees were indeed up to twice as high as control trees (Table 1) and had larger BAIs
(Figures 4, 5, 6). Moreover, the leaf area per branch of Scots pine was significantly
higher for irrigated than for control trees as has been found in a previous study at our
sites in Valais (Feichtinger et al. 2015). An increase in tree height and in leaf area are
associated with (1) an increase in water demand (Sohn et al. 2013), and (2) a decrease
in the negative pressure of the xylem (Martínez-Vilalta et al. 2006; Woodruff and
Meinzer 2011). Both mechanisms may lead to a decrease in stomatal conductance and
thus to a lower Δ13C (Martínez-Vilalta et al. 2006; McDowell et al. 2011) for the
irrigated trees at Varen and Mals.
Our findings may suggest, that long-term adjustments to differences in water
availability are regulated mainly via the adaptation of the radial and height growth and
the increase in leaf area, while adjustments in the physiology of Scots pine is highly
depending on site characteristics. Thus, Scots pine and larch trees tend to grow close to
their capacity of hydraulic adjustment (Martínez-Vilalta and Piñol 2002). The findings of
Eilmann et al. might support this assumptions, as the comparison of the lumen
diameter of irrigated Scots pine and the control in Valais revealed a minor but
significant decrease in earlywood cells and no differences in latewood cells with
irrigation. The question arises, why we found an high increase vs. no increase in Δ13C
between the irrigated Scots pine and the control depending on the site. In contrast to
the site Lens, the irrigation at Varen and Mals might have not been sufficient, still
entailing low values of Δ13C in tree rings of irrigated trees. Further research needs to be
conducted in order to investigate these site-specific effects on Δ13C on a long term
scale.
78
Chapter II
Adjustments in growth and physiology to decreasing water availability
The stop in irrigation resembled an extreme and long-lasting drought for trees used to
grow at mesic soil conditions. The BAI of Scots pine responded to the stop in irrigation
with a two-year delay and then decreased gradually to similar growth rate than that of
the control trees (Figure 5a). Scots pine at our site Lens might have had large amounts
of stored carbohydrates as the trees already established under irrigated well watered
conditions and as has been discussed by Eilmann et al. (2010).
The Δ13C values of the former irrigated Scots pine decreased instantly to a lower level
(Figure 5c), suggesting a reduction in stomatal conductance in response to the new
drought conditions. However, only five years after the stop in irrigation, the level of
Δ13C of trees exposed to the sudden reduction in water availability was reaching the
same level as the control trees. For the carbon source of the tree ring, this indeed
implies that a certain amount of old carbohydrates has been mixing with new
assimilates (Keel et al. 2007). This did thus not only result in delayed responses in BAI
after the irrigation was stopped, but also might led to a gradual decrease in ∆13C.
For European larch (Mals; Figure 6), we found no lagged but an immediate response in
BAI and Δ13C of trees exposed to a stop in irrigation, decreasing to a similar level in BAI
and even lower level in Δ13C compared to control trees. This is in agreement with other
studies, reporting a higher sensitivity to dry conditions for European larch than for
Scots pine (Eilmann and Rigling 2012; Lévesque et al. 2013; Schuster and Oberhuber
2012). The higher growth sensitivity and immediate responses to the stop in irrigation
of larch might be linked to its deciduous character. This is in contrast to evergreen
Scots pine, with needle longevity up to five years in Valais (Dobbertin et al. 2010) that
Chapter II
79
might have buffered the initial effect of drought due to foliage developed during times
of sufficient water supply.
Overall, Scots pine and European larch trees were able to adjust to the drier conditions
after the stop in irrigation as shown by the similar level of BAI and Δ13C compared to
the control. Similar growth adjustments were found for Holm oak in Spain following a
13-years period of rain exclusion (Barbeta et al. 2013). However, while the BAI of pine
and larch growing along the drained channels exceeded the BAI of the control trees at
the end of the study period, Δ13C of both irrigated and control trees were about on a
similar level. This phenomena might on the one hand be explained by the differences
evolved from an improvement in soil conditions (nutrient availability and humus
contents) at the previously irrigated sites shown by Feichtinger et al. (2014). On the
other hand, the increased tree height of previously irrigated trees in comparison to the
control might be another indication that Scots pine in Valais and larch in Vinschgau is
growing close to their maximum capacity of hydraulic adjustments as mentioned
above.
Conclusion
We have provided evidence that long-term differences in water availability led to
changes in BAI by a combination of increased canopy-level photosynthesis and
physiological adjustments of the metabolic processes. This was reflected in high
differences in BAI between the irrigated and control trees, but varying differences in
Δ13C resulting from hydraulic changes with tree height in response to long-term
differences in water availability. Models that extrapolate the effects of changes in
water availability on photosynthesis and stomatal conductance at longer time scales
80
Chapter II
therefore need to consider adjustments in the hydraulic architecture, e.g. tree size to
ambient water availability and site-specific effects.
Changes in water availability caused non-linear and lagged responses over time in both
Scots pine and larch due to adjusting processes, which lasted from one to several
years. Both tree species were able to adjust to the new and drier conditions. This
suggests that, when climate gets warmer and drier, Scots pine and European larch
growing at mesic sites and at lower elevations in the Alps may face lower mortality
rates than previously anticipated.
Acknowledgments
We would like to thank Britta Eilmann for support during project planning, Arnaud
Giggiola and Ellen Pflug for valuable input, Catharina Lötscher, Lola Schmid for support
and isotope measurements, Markéta Jetel and Chantal Freymond for lab assistance,
and Magdalena Nötzli and Loïc Schneider for technical support. This study was funded
by the Swiss National Science Foundation (National Research Programme NRP 61).
Chapter II
81
References
Addington, R.N., L.A. Donovan, R.J. Mitchell, J.M. Vose, S.D. Pecot, S.B. Jack, U.G.
Hacke, J.S. Sperry and R. Oren. 2006. Adjustments in hydraulic architecture of
Pinus palustris maintain similar stomatal conductance in xeric and mesic
habitats. Plant, Cell and Environment. 29:535-545.
Affek, H.P. and D. Yakir. 2014. The Stable Isotopic Composition of Atmospheric CO2. In
Treatise on Geochemistry, 2nd Edn. Eds. H.D. Holland and K.K. Turekian.
Elsevier, Oxford, pp 179-212.
Anderegg, W.R.L., J.M. Kane and L.D.L. Anderegg. 2012. Consequences of widespread
tree mortality triggered by drought and temperature stress. Nature Climate
Change. 3:30-36.
Barbeta, A., R. Ogaya and J. Peñuelas. 2013. Dampening effects of long-term
experimental drought on growth and mortality rates of a Holm oak forest.
Global Change Biology. 19:3133-3144.
Beier, C., C. Beierkuhnlein, T. Wohlgemuth, J. Penuelas, B. Emmett, C. Körner, H. Boeck,
J.H. Christensen, S. Leuzinger, I.A. Janssens, K. Hansen and J. Arnone. 2012.
Precipitation manipulation experiments - challenges and recommendations for
the future. Ecology Letters. 15:899-911.
Biondi, F. and F. Qeadan. 2008. A theory-driven approach to tree-ring standardization
defining the biological trend from expected basal area increment. Tree-ring
research. 64:81-96.
Brühlhart, H. 1999. Einfluss der Bewässerung auf das Jahrringwachstum von Kiefern.
Diplomarbeit. Berner Fachhochschule, Schweizerische Fachhochschule für
Holzwirtschaft, Biel.
Ceppi, P., S.C. Scherrer, A.M. Fischer and C. Appenzeller. 2012. Revisiting Swiss
temperature trends 1959-2008. International Journal of Climatology. 32:203213.
Cotrufo, M.F., G. Alberti, I. Inglima, H. Marjanović, D. LeCain, A. Zaldei, A. Peressotti
and F. Miglietta. 2011. Decreased summer drought affects plant productivity
and soil carbon dynamics in a Mediterranean woodland. Biogeosciences.
8:2729-2739.
Dobbertin, M. 2005. Tree growth as indicator of tree vitality and of tree reaction to
environmental stress: a review. European Journal of Forest Research. 124:319333.
Dobbertin, M., B. Eilmann, P. Bleuler, A. Giuggiola, E. Graf Pannatier, W. Landolt, P.
Schleppi and A. Rigling. 2010. Effect of irrigation on needle morphology, shoot
and stem growth in a drought-exposed Pinus sylvestris forest. Tree Physiology.
30:346-360.
Eilmann, B., N. Buchmann, R. Siegwolf, M. Saurer, P. Cherubini and A. Rigling. 2010.
Fast response of Scots pine to improved water availability reflected in tree-ring
width and δ13C. Plant, Cell & Environment. 33:1351-1360.
Eilmann, B. and A. Rigling. 2012. Tree-growth analyses to estimate tree species'
drought tolerance. Tree Physiology. 32:178–187.
Eilmann, B., P. Weber, A. Rigling and D. Eckstein. 2006. Growth reactions of Pinus
sylvestris L. and Quercus pubescens Willd. to drought years at a xeric site in
Valais, Switzerland. Dendrochronologia. 23:121-132.
82
Chapter II
Eilmann, B., R. Zweifel, N. Buchmann, P. Fonti and A. Rigling. 2009. Drought-induced
adaptation of the xylem in Scots pine and pubescent oak. Tree Physiology.
29:1011-1020.
FAO. 2006. World reference base for soil resources, vol. 103. In FAO, World soil
resources reports, Rome.
Farquhar, G.D., J.R. Ehleringer and K.T. Hubick. 1989. Carbon discrimination and
photosynthesis. Annual Review of Plant Physiology and Plant Molecular
Biology. 40:503–537.
Farquhar, G.D., M.H. O'Leary and J.A. Berry. 1982. On the relationship between carbon
isotope discrimination and the intercellular carbon dioxide concentration in
leaves. Australian Journal of Botany. 9:121-37.
Feichtinger, L.M., B. Eilmann, N. Buchmann and A. Rigling. 2014. Growth adjustments
of conifers to drought and to century-long irrigation. Forest Ecology and
Management. 334:96-105.
Feichtinger, L.M., B. Eilmann, N. Buchmann and A. Rigling. 2015. Trait-specific
responses of Scots pine to irrigation on a short vs long time scale. Tree
Physiology. 35:160-171.
Gärtner, H. and D. Nievergelt. 2010. The core-microtome: A new tool for surface
preparation on cores and time series analysis of varying cell parameters.
Dendrochronologia. 28:85-92.
Gessler, A., J.P. Ferrio, R. Hommel, K. Treydte, R.A. Werner and R.K. Monson. 2014.
Stable isotopes in tree rings: towards a mechanistic understanding of isotope
fractionation and mixing processes from the leaves to the wood. Tree
Physiology
Gobiet, A., S. Kotlarski, M. Beniston, G. Heinrich, J. Rajczak and M. Stoffel. 2013. 21st
century climate change in the European Alps - A review. Science of The Total
Environment
Helle, G. and G.H. Schleser. 2004. Beyond CO 2 -fixation by Rubisco – an interpretation
of 13C/12C variations in tree rings from novel intra-seasonal studies on broadleaf trees. Plant, Cell & Environment. 27:367–380.
Holmes, R.L. 1994. Computer-assisted quality control in tree-ring dating and
measurement. Tree-ring Bulletin. 43:69-78.
Keel, S.G., R.T.W. Siegwolf, M. Jäggi and C. Körner. 2007. Rapid mixing between old and
new C pools in the canopy of mature forest trees. Plant, Cell & Environment.
30:963-972.
Larcher, W. 2003. Physiological plant ecology: Ecophysiology and stress physiology of
functional groups, 4th Edn. Springer, Berlin Heidelberg New York.
Leibundgut, C. and I. Kohn. 2014. European traditional irrigation in transition part I:
Irrigation in times past-a historic land use practice across Europe. Irrigation and
Drainage. 63:273-293.
Leuzinger, S., Y. Luo, C. Beier, W. Dieleman, S. Vicca and C. Körner. 2011. Do global
change experiments overestimate impacts on terrestrial ecosystems? Trends in
Ecology & Evolution. 26:236-241.
Lévesque, M., M. Saurer, R. Siegwolf, B. Eilmann, P. Brang, H. Bugmann and A. Rigling.
2013. Drought response of five conifer species under contrasting water
availability suggests high vulnerability of Norway spruce and European larch.
Global Change Biology. 19:3184-3199.
Litton, C.M., J.W. Raich and M.G. Ryan. 2007. Carbon allocation in forest ecosystems.
Global Change Biology. 13:2089-2109.
Chapter II
83
Lotter, A.F., P.G. Appleby, R. Bindler, J.A. Dearing, J.-A. Grytnes, W. Hofmann, C.
Kamenik, A. Lami, D.M. Livingstone, C. Ohlendorf, N. Rose and M. Sturm. 2002.
The sediment record of the past 200 years in a Swiss high-alpine lake:
Hagelseewli (2339 m a.s.l.). Journal of Paleolimnology. 28:111–127.
Luo, Y., D. Gerten, G. Le Maire, W.J. Parton, E. Weng, X. Zhou, C. Keough, C. Beier, P.
Ciais, W. Cramer, J.S. Dukes, B. Emmett, P.J. Hanson, A. Knapp, S. Linder, D.A.N.
Nepstad and L. Rustad. 2008. Modeled interactive effects of precipitation,
temperature, and [CO 2 ] on ecosystem carbon and water dynamics in different
climatic zones. Global Change Biology. 14:1986-1999.
MacKay, S.L., M.A. Arain, M. Khomik, J.J. Brodeur, J. Schumacher, H. Hartmann and M.
Peichl. 2012. The impact of induced drought on transpiration and growth in a
temperate pine plantation forest. Hydrological Processes. 26:1779-1791.
Martínez-Vilalta, J. and J. Piñol. 2002. Drought-induced mortality and hydraulic
architecture in pine populations of the NE Iberian Peninsula. Forest Ecology
and Management. 161:247-256.
Martínez-Vilalta, J., D. Vanderklein and M. Mencuccini. 2006. Tree height and agerelated decline in growth in Scots pine (Pinus sylvestris L.). Oecologia. 150:529544.
McCarroll, D. and N.J. Loader. 2004. Stable isotopes in tree rings. Quaternary Science
Reviews. 23:771-801.
McDowell, N., W.T. Pockman, C.D. Allen, D.D. Breshears, N. Cobb, T. Kolb, J. Plaut, J.
Sperry, A. West, D.G. Williams and E.A. Yepez. 2008. Mechanisms of plant
survival and mortality during drought: why do some plants survive while others
succumb to drought? New Phytologist. 178:719-739.
McDowell, N.G., B.J. Bond, L.T. Dickman, M.G. Ryan and D. Whitehead. 2011.
Relationships between tree height and carbon isotope discrimination. In Sizeand Age-Related Changes in Tree Structure and Function Eds. F.C. Meinzer, B.
Lachenbruch and T.E. Dawson. Springer Netherlands, pp 255-286.
R Development Core Team. 2011. R: A language and environment for statistical
computing. In R Foundation for Statistical Computing, Vienna, Austria.
Rigling, A., C. Bigler, B. Eilmann, E. Feldmeyer-Christe, U. Gimmi, C. Ginzler, U. Graf, P.
Mayer, G. Vacchiano, P. Weber, T. Wohlgemuth, R. Zweifel and M. Dobbertin.
2013. Driving factors of a vegetation shift from Scots pine to pubescent oak in
dry Alpine forests. Global Change Biology. 19:229-240.
Schuster, R. and W. Oberhuber. 2012. Drought sensitivity of three co-occurring conifers
within a dry inner Alpine environment. Trees
Siegenthaler, U. and H. Oeschger. 1980. Correlation of 18O in precipitation with
temperature and altitude. Nature. 285:314-317.
Sohn, J.A., T. Gebhardt, C. Ammer, J. Bauhus, K.-H. Häberle, R. Matyssek and T.E.E.
Grams. 2013. Mitigation of drought by thinning: Short-term and long-term
effects on growth and physiological performance of Norway spruce (Picea
abies). Forest Ecology and Management. 308:188-197.
Speer, J.H. 2010. Fundamentals of tree-ring research. The University of Arizona Press,
Tucson, Arizona.
Tang, K.L. and X.H. Feng. 2001. The effect of soil hydrology on the oxygen and
hydrogen isotopic compositions of plants' source water. Earth and Planetary
Science Letters. 185:355-367.
Thornthwaite, C. 1948. An approach toward a rational classification of climate.
Geographical review. 38:55-94.
84
Chapter II
Vacchiano, G., M. Garbarino, E. Borgogno Mondino and R. Motta. 2011. Evidences of
drought stress as a predisposing factor to Scots pine decline in Valle d’Aosta
(Italy). European Journal of Forest Research. 131:989-1000.
Weber, P., H. Bugmann and A. Rigling. 2007. Radial growth responses to drought of
Pinus sylvestris and Quercus pubescens in an inner-Alpine dry valley. Journal of
Vegetation Science. 18:777-792.
Werner, C. and A. Gessler. 2011. Diel variations in the carbon isotope composition of
respired CO 2 and associated carbon sources: a review of dynamics and
mechanisms. Biogeosciences Discussions. 8:2183-2233.
West, A.G., S.J. Patrickson and J.R. Ehleringer. 2006. Water extraction times for plant
and soil materials used in stable isotope analysis. Rapid Communications in
Mass Spectrometry. 20:1317-1321.
Woodruff, D. and F. Meinzer. 2011. Size-dependent changes in biophysical control of
tree growth: the role of turgor. In Size- and Age-Related Changes in Tree
Structure and Function Eds. F.C. Meinzer, B. Lachenbruch and T.E. Dawson.
Springer Netherlands, pp 363-384.
Zhao, M. and S.W. Running. 2010. Drought-induced reduction in global terrestrial net
primary production from 2000 through 2009. Science. 329:940-943.
Chapter III
85
Chapter III Trait-specific responses of Scots pine to
irrigation on a short vs long time sale.
Published as:
Feichtinger, Linda M.1; Eilmann, Britta2; Buchmann, Nina3; Rigling, Andreas1, 2015.
Trait-specific responses of Scots pine to irrigation on a short vs. long time scale. Tree
Physiology 35, 160–171.
1
Swiss Federal Research Institute WSL, Zürcherstrasse 111, 8903 Birmensdorf, Switzerland
2
Wageningen University, Centre for Ecosystem Studies, Forest Ecology and Forest
Management, 6700 AA Wageningen, The Netherlands
3
ETH Zürich, Institute of Agricultural Sciences, Universitätsstrasse 2, 8092 Zürich, Switzerland
86
Chapter III
Abstract
In xeric environments, an increase in drought is related to reduced forest productivity
and to enhanced mortality. However, predictions of future forest development remain
difficult as the mechanisms underlying the responses of mature trees to long-term
variations in water availability are not well understood. Here, we aimed to compare the
adjustments in radial growth and morphological needle and shoot traits of mature
Scots pine (Pinus sylvestris L.) growing along open water channels with those of control
trees growing under naturally dry conditions at three sites in Valais, an inner-Alpine dry
valley of Switzerland. The trees growing along two channels had been irrigated since
germination (>70 years), whereas those along another previously drained channel had
been irrigated only from 2010 to 2012, when the channel was re-established, and could
thus be used to quantify the short-term effects of re-irrigation. Linear mixed models
revealed that needle and shoot lengths as well as early- and late-wood basal area
increments (BAIs) were most responsive to short-term and long-term irrigation.
However, the magnitude of the response to the short-term irrigation exceeded that of
the long-term irrigation. An extreme drought during the first half of 2011 led to an
immediate decrease in the needle length, needle width, and early- and late-wood BAIs
of the control trees, whereas the shoot length and needle numbers of control trees
reacted with a 1-year delay to the extreme drought, as the shoots were responding to
water availability of previous year's summer. Such negative responses to dry climatic
conditions were even found in irrigated trees at one of our sites, which might be linked
to tree growth becoming more sensitive to drought with increasing tree height and leaf
area. In order to improve predictions of future forest development, long-term studies
are necessary that consider lagged responses and adjustment processes of trees to
changes in water availability.
Chapter III
87
Introduction
Drought is one of the most important environmental constraints on forest productivity
and the key factor in climate-related forest dieback (Allen et al. 2010). As climate
projections predict an increase in the frequency and severity of future drought events
(IPCC 2013), drought-induced effects on forest ecosystems are likely to gain in
importance. Species growing at their distribution limit are particularly sensitive to such
effects (Hampe and Petit 2005; Parmesan 2006). These phenomena have recently been
investigated for Scots pine (Pinus sylvestris L.), which has experienced high rates of
drought-induced mortality at its southern distribution limit in Spain (Martínez-Vilalta
and Piñol 2002; Sánchez-Salguero et al. 2012), and in the Central Alps (Dobbertin et al.
2005; Oberhuber et al. 2001; Rebetez and Dobbertin 2004). In general, Scots pine has a
wide spatial distribution, limited by high temperature and summer drought at its
southernmost range (Carlisle and Brown 1968). Thus, an increase in drought due to
climate change may severely affect current populations (Matías and Jump 2012).
Topographic characteristics such as slope, aspect, altitude and soil depth influence soil
water availability (Western et al. 2002) and consequently the responses of trees to
climate. Accordingly, the highest mortality rates of Scots pine in the southernmost
populations were found in inner-Alpine dry valleys on steep, south-facing slopes at
lower altitudes and on shallow soils (Oberhuber et al. 1998; Rigling et al. 2013; Weber
et al. 2007). Inner-Alpine valleys, such as Valais in Switzerland, are characterized by low
annual precipitation and are regularly exposed to drought due to very high rainshading
mountains.
In
addition,
higher
temperatures
have
increased
evapotranspiration in Valais during recent decades, leading to reduced water
availability and thus to higher drought stress for Scots pine (Rebetez and Dobbertin
2004; Zweifel et al. 2009). The effects of drought on the future development of Scots
88
Chapter III
pine in inner-Alpine dry valleys are still uncertain and the mechanisms leading to
drought-induced mortality are not yet fully understood.
Scots pine has several adaptive traits to avoid drought stress to a certain extent. These
reduce water loss by minimizing transpiration and increasing water uptake through the
tree’s deep and extensive root system. Consequently, Scots pine trees are able to delay
the onset of water stress by combining physiological and morphological whole-tree
adjustments. In addition, the adjustments of leaf, shoot and radial growth in response
to soil water availability are important mechanisms that influence a tree’s light
interception, carbon gain, and water use efficiency and thus its survival (Niinemets et
al. 2001; Raison et al. 1992) and are crucial aspects of functional plant ecology
(Ordoñez et al. 2009; Walters and Gerlach 2013; Wright et al. 2004). However, little is
known about the intraspecific ability of mature trees to adjust needle and shoot traits
to drought.
Drought manipulation experiments enable estimations of the future impacts of climate
change on forest development. The responses of trees to differences in water
availability are assessed by either enhancing or reducing drought (via irrigation) in
experiments, which are typically short term, lasting less than ten years (e.g. Albaugh et
al. 2004; Cotrufo et al. 2011; Dobbertin et al. 2010; MacKay et al. 2012). Initial studies
suggest that adjustment processes and feedback mechanisms may dampen the
responses of trees to changes in water availability on longer time scales (Barbeta et al.
2013; Beier et al. 2012; Feichtinger et al. 2014; Leuzinger et al. 2011). Therefore, it is
crucial to gain a better understanding of the drought impact over long time periods in
order to be able to predict future forest development under a changing climate.
Chapter III
89
In the present study, we investigated the adjustment capability of Scots pine to longterm irrigation and drought by studying different traits in trees growing along historical
water channels at two sites in Valais, Switzerland. These channels are mainly open
ditches that have irrigated the surrounding trees passively since their germination. The
comparison of irrigated trees with those growing nearby under naturally dry conditions
allows conclusions to be drawn about Scots pine’s drought-adjustment strategies. An
additional channel, which had been dry for several decades and re-established in 2010,
allowed us to quantify the effects of short-term re-irrigation on tree growth.
The following questions were addressed: (1) What are the effects of irrigation on
needle, shoot and radial growth traits of Scots pine? (2) Are there differences in these
effects between those with long-term irrigation and those with only two-years of reirrigation? (3) What is the reaction of Scots pine to climate?
Material and Methods
Study sites
We studied three Scots pine stands (Varen, Planige and Lens) located in Valais, an
inner-Alpine dry valley of Switzerland (7°E/46°N, Figure 1). All stands are situated on
south-facing slopes (15-37°) at about 800 to 1000 m a.s.l. (see Table 1) on shallow
rendzic leptosoils (FAO 2006). These forests, which are dominated by Scots pine,
belong to the vegetation type Erico-Pinetum sylvestris. The climate in Valais is
characterized by a low total annual precipitation of 664 mm and by an average annual
temperature of 9.2 °C (see Supplementary Figure S1).
90
Chapter III
Figure 1. Location of the study region
in Central Europe (a), Valais in
Switzerland (b). The study sites are
marked as dots and the weather
stations as triangles.
We studied the effects of water availability on different traits by comparing trees
growing under naturally dry conditions (control) with irrigated trees growing along
open water channels. These open channels, so called Bisse, conduct melt water from
high elevations to agricultural land from the end of April until the beginning of October
(Leibundgut and Kohn 2014). At the site Varen, the channel can be dated back to the
12th, at Planige and Lens, to the 15th century. These channels are mainly open ditches,
which lose a large amount of water passively to the surrounding vegetation. For one of
these water channels (Lens), the water loss was quantified at 27% or about 80 l/s along
a 3 km transect (Brühlhart 1999). Due to this leakage, the open channel at Lens was
replaced by an impermeable pipe in 1984. However, it was re-established for touristic
reasons in May 2010.
Chapter III
91
All trees classified as irrigated were growing less than one meter away from the
irrigation channel to ensure that they did benefit from the passive irrigation. As
controls, we chose trees growing nearby the irrigation channels, but outside the
irrigation zone (upslope or in the proximity of the irrigated plots) under naturally dry
conditions.
Table 1. Description of sampled trees per control and irrigated plot at the three Scots pine sites with
coordinates, mean values ± SE (n = 8) of tree age, DBH, height and needle longevity. Differences between
(i) the control and the irrigated trees and (ii) the trees measured in the campaigns 2010 (2010) and 2012
(2012) were tested with two-sample Wilcoxon-rank-sum tests (values with different letters: P ≤ 0.05).
Needle longevity was only measured in the campaign 2012. Trees at the site Lens were re-irrigated since
2010.
Climate data
To estimate the combined effects of low precipitation and high evaporative demand on
tree growth and physiology, a drought index (DRI) was calculated according to
Thornthwaite (1948):
𝐷𝐷𝐷𝐷𝐷𝐷 = 𝑃𝑃 − 𝑃𝑃𝑃𝑃𝑃𝑃,
where P is the monthly precipitation and PET the potential evapotranspiration.
(1)
92
Chapter III
Temperature data were obtained from the weather stations located in Sion at 492 m
a.s.l about 9 km away from Lens, 17 km from Planige and 20 km from Varen. As
precipitation has a high spatial variability, we chose the weather station in Sierre (539
m a.s.l.), which is closer to our sites, but which only records precipitation data. This
station is about 4 km away from Planige and 8 km from Varen and Lens (Figure 1).
Sampling and measurements
In a first campaign (campaign 2010), we cut three branches from eight mature Scot
pine trees (> 50 years) per irrigated and control plots in September 2010, using clippers
extendable up to 8 m. To exclude confounding factors such as self shading, we
exclusively sampled branches from the sunlit, south-exposed crown parts, and avoided
branches with a damaged main shoot leader, mistletoe infection, or visible insect
infestations. Immediately after cutting, the branches were put into plastic bags, which
were sealed and stored in cooling boxes for transportation to the lab. All subsequent
measurements were conducted within 36 hours. In a second campaign in 2012
(campaign 2012), needle sampling was repeated following the same protocol as in the
campaign in 2010. However, a wet snow event in the winter 2011/2012 had damaged
several trees at the study sites and consequently, only 32 of the 48 trees sampled in
2010 were re-sampled in 2012. The damage was most severe at Lens, and the control
plot was therefore heavily thinned in 2012. Furthermore, two increment cores were
taken perpendicular to each other at breast height (1.3 m) with an increment borer,
and the diameter at breast height (DBH) as well as the tree height were recorded for
each tree in 2012.
In the lab, the shoot length of the main shoot leader per branch was measured with an
electronic gauge (± 0.1 mm), going back to the year 2000. Here, it was only possible to
Chapter III
93
measure 13 years because morphological markers fade with time, making it difficult to
visualize different shoots. Subsequently, the needles of the last three years were
counted per shoot, and 20 needles per year were then selected (or all needles per year
if fewer than 20 needles were present) and scanned with a resolution of 300 dpi using
a scanner with back illumination. The curved needle length and width as well as the
projected needle area were quantified from the digital images using the software
winSEEDLE (winSEEDLE 2006 Régent Instruments, Quebec, QC). The needles were then
oven-dried for at least 48 h at 80°C (Pérez-Harguindeguy et al. 2013) and weighed to an
accuracy of 0.01 g to assess the needle dry weight. Finally, the specific leaf area (SLA)
was calculated as the ratio of needle area to needle dry weight, the total leaf area as a
product of needle area and the number of needles per shoot and the needle density as
the number of needles per mm2 shoot.
At the site where the irrigation had been re-started (Lens), the needles were
additionally ground and 25 mg weighed into tin capsules. Total C and N were
subsequently measured with a CN analyzer NC 2500 (CE Instruments, Milan, Italy). The
surface of the tree cores of all sites were prepared with a core microtome (Gärtner and
Nievergelt 2010) to improve the visibility of the tree rings. Subsequently, the early- and
latewood widths were measured using a LINTAB digital positioning table and the
software TSAP-Win (Rinntech, Heidelberg, Germany), and cross-dated both visually and
statistically using the software COFECHA (Holmes 1994). To compare the absolute treering growth between the irrigated and control trees, the annual basal area increment
(BAI) was calculated for the period 2000 to 2012 according to Biondi and Qeadan
(2008):
94
2 ),
𝐵𝐵𝐵𝐵𝐵𝐵 = 𝜋𝜋 (𝑅𝑅𝑡𝑡2 − 𝑅𝑅𝑡𝑡−1
Chapter III
(2)
where R is the stem radius at time t.
Vegetation survey
We were not able to take soil samples in order to characterize soil conditions due to
very rocky and shallow soils at our study sites. We therefore used vegetation surveys of
the understory vegetation (< 5 m in height) to gain site-specific indicator values
according to Landolt (2010) and to assess possible differences in soil characteristics
(humidity, humus and nutrient values) between irrigated and control plots. The
understory vegetation of a 4 m2 subplot around each tree was surveyed during June
and July 2011. The humidity value indicated the mean soil humidity during the
vegetation period, and ranged from 1 (very dry) to 5 (flooded). The nutrient value
indicated the soil nutrient availability, and ranged from 1 (very poor) to 5 (very rich),
and the humus value indicated the soil humus content, and ranged from 1 (low
content) to 3 (high content) (Landolt et al. 2010).
Un-weighted means of the indicator values were calculated per plot because the total
percentage of coverage per species was often low (<1%). Species for which no specific
indicator values could be assigned were excluded from the analysis. In total, 131 of 137
species from 69 surveys were used in this analysis. The average number of species per
subplot was 20, with a minimum of ten species and a maximum of 34 species. At Lens,
we assumed the understory vegetation had been unaffected by the one year reirrigation at the time of the survey.
Chapter III
95
Statistical analysis
All statistical analyses were performed using the software R (R Development Core
Team 2011). The differences between the needle and the shoot traits of the irrigated
and control trees were analysed with a generalized linear mixed model, using the
function “lmer” of the package “lme4” (Bates et al. 2014), fitted by restricted maximum
likelihood (REML). This meant that, each site (Varen, Planige, Lens), each needle and
shoot trait as well as the early- and latewood BAI, were modelled separately as a
function of treatment (irrigation/control) and the current year with the individual tree
as the random variable:
𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑖𝑖𝑖𝑖 = 𝛼𝛼 + 𝛽𝛽 1 ∙ 𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑖𝑖 + 𝛽𝛽2 ∙ 𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡 𝑖𝑖𝑖𝑖 + 𝛽𝛽 3 ∙ 𝑦𝑦𝑦𝑦𝑦𝑦𝑦𝑦𝑖𝑖𝑖𝑖 + 𝛽𝛽4 (𝑡𝑡𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟 ∙
𝑦𝑦𝑦𝑦𝑦𝑦𝑦𝑦) + 𝜀𝜀𝑖𝑖𝑖𝑖 ,
(3)
with α as the overall intercept, β 1 as the random effect on the intercept associated
with the individual tree and β 2 , β 3 and β 4 as the parameters of the fixed effects, i as
the index for the tree, j as the index for the year of growth and ε ij as the error term.
We transformed the data to achieve normality and homoscedasticity of the residuals
when necessary. Furthermore, orthogonal contrasts were calculated to quantify the
effects of the treatment for each year separately as well as the differences of each trait
between the years. The needle traits, needle length and width, SLA and total SLA were
analysed for a five-year period (2008-2012), and the tree-ring width and shoot length
for a 13-year period (2000-2012). Both periods were too short to detect any temporal
autocorrelation. Differences in tree age, DBH and height, as well as in the indicator
values between irrigated and control plots, were quantified with a Wilcoxon-rank-sum
test and the resulting P-values were adjusted for multiple testing using the HolmBonferroni method (Holm 1979).
96
Chapter III
To test which seasonality in climate could best explain the variations in the BAI (earlyand latewood) and the shoot length of Scots pine, these series were correlated with
the monthly mean DRI for (1) previous summer (June to August), (2) current spring
(March to May) and (3) current summer (June to August) and Spearman’s rank
correlation coefficients were calculated.
Results
Climate
During the study period from 2000 to 2012, the mean annual precipitation sum in
Valais was 636 mm, with a mean annual temperature of 10.7°C. For the same period,
the climatic water availability expressed as the DRI and estimated as a function of
precipitation and potential evapotranspiration revealed a mean DRI of 0.03 (see
Supplementary Figure S2). The years with the lowest climatic water availability were
2003 (DRI= -0.60) and 2011 (DRI= -0.52), while 2002 (DRI= 0.55) and 2007(DRI= 0.48)
were the years with the highest water availability .
Effects of long-term differences in water availability
The understory vegetation in June/July 2011 indicated significantly higher soil
humidity, nutrient availability and humus contents in the long-term irrigated plots at
the sites Varen and Planige than in the corresponding control plots (P≤0.05, Figure 2).
Furthermore, long-term irrigated trees were significantly taller and wider (DBH) than
corresponding control trees in 2012 (P≤0.01, Table 1) and revealed significantly higher
BAI and longer shoots for the period from 2000 to 2012 (P≤0.05, Figure 3). Irrigated
trees had on average about three times higher early- and latewood BAIs than control
Chapter III
97
trees, and twice as long shoots, respectively (Figure 3). Furthermore, the needles of
irrigated trees were by over a third longer and approx. 10% wider for all years between
2008 and 2012 (P≤0.05, Figure 4). In addition, significantly more needles per shoot
(approx. 60%) developed in the irrigated trees between 2010 and 2012, while the
needle density was approx. 20% lower (P≤0.05, Figure 5). Despite the lower needle
density, the total needle area per shoot was significantly higher for irrigated trees than
for control trees (approx. 130%). No significant differences in the mean SLA between
the irrigated and control trees were found at the site Varen, whereas the control trees
at Planige had significantly higher SLA values in the years 2009, 2010 and 2011 than the
irrigated trees (P≤0.05, Figure 4). The SLA showed a clear needle age effect, decreasing
with the age of the needles.
Figure 2. Mean indicator values ± SE (n = 9–14) of the understory vegetation in June/July 2011 in the
control plots and irrigated plots at the three Scots pine sites (Varen, Planige and Lens). Differences in
indicator values between the control plots and long-term irrigated (Varen and Planige) or re-irrigated plots
(Lens) were tested using two-sample Wilcoxon-rank-sum tests. Differences between all indicator values
were significant at all three sites (P ≤ 0.05).
98
Chapter III
Effects of short-term re-irrigation
Before the irrigation re-started at the site Lens in May 2010, no significant differences
were found in the shoot length (Figure 3), needle length, SLA (Figure 4) and needle N
concentration (Figure 6) between the trees growing along the drained channel and the
corresponding control trees (P<0.05). However, prior to the re-irrigation, trees tended
to have a higher annual BAI than the control trees, which was significant for the years
2000 to 2004 and 2008 for earlywood , as well as for the years 2000, 2003 and 2008 for
latewood (P≤0.05, Figure 3). Furthermore, soil conditions in the re-irrigated and control
plots were significantly different, as the understory vegetation indicated the soil
humidity, nutrient availability and humus content in the re-irrigated plots was higher in
2011 (Figure 2). Measurements in 2012 revealed no differences in DBH, but trees were
higher in the re-irrigated plots than in the control plots (Table 1).
In 2010, the year of the re-establishment of the channel, the latewood BAI (Figure 3),
needle length (Figure 4) and needle N concentration (Figure 6) of the re-irrigated trees
reacted significantly (P≤0.05) to the increase in water availability. In the following year,
these differences became even more pronounced. The latewood BAI of the re-irrigated
trees increased by over six times compared to the control (Figure 3), the needles were
more than twice as long (Figure 4), and the needle N concentration increased by one
third (Figure 6). In contrast, the earlywood BAI (Figure 3), the shoot length (Figure 3)
and the needle width (Figure 4) reacted with a one year delay to the re-irrigation. In
2011, the earlywood BAI was more than five times higher for the re-irrigated trees than
for control trees, the shoots were nearly twice as long and the needles approx. 20%
wider. In contrast, there was no increase in the number of needles per shoot after two
years of irrigation, but a significant (P≤0.05) decrease in needle density by 30% (Figure
3.5) and in SLA by 10% (Figure 4) in the irrigated than in the control trees in 2011.
Chapter III
99
Figure 3. Mean early- and latewood BAI and shoot length ± SE (n = 8) of irrigated and corresponding
control trees at the three Scots pine sites for the period 2000–12 measured in September 2012: Varen
(left), Planige (middle) and Lens (right). The grey shaded areas indicate the non-irrigated period before the
re-establishment of the water channel at the site Lens in 2010. The data for the site Lens in 2012 are not
shown, because the thinning in winter 2011/2012 was a confounding factor. Contrasts of linear mixed
models were used to test for differences between the control trees and irrigated trees. For Varen and
Planige, differences were significant for all years and traits (P ≤ 0.05) and were thus not marked. For the
site at Lens, only single years were significantly different and are marked (*P ≤ 0.05).
Effects of climate
For trees growing under naturally dry conditions at the sites Varen and Planige, the
earlywood BAI did not appear to be related to the climatic water availability (DRI,
P<0.05, Table 2). However, the latewood BAI of the control trees at Varen (Table 2) was
significantly enhanced by high water availability during the current summer (June to
August, r=0.63, P≤0.05). On the other hand, the shoot growth of the control trees at
both sites was significantly influenced by wet climate during the previous year’s
100
Chapter III
summer (r=0.70 to 0.77, P≤0.01). The early- and latewood BAIs of the long-term
irrigated trees at the sites Varen and Planige, did not seem to be influenced by the
climatic water availability, whereas the shoot growth of the trees growing along the
channels at Varen (but not at Planige), was still positively influenced by wet climatic
conditions during the summer of the previous year (r= 0.70, P≤0.01).
Figure 4. Mean needle traits ± SE (n = 8) for three needle generations of the irrigated Scots pine and
corresponding control trees at the three sites (Varen, Planige and Lens). The following needle traits are
compared: needle length, needle width and specific leaf area (SLA) measured in 2010 and 2012. The
needle traits for the site Lens in 2012 are not shown, because the thinning in winter 2011/2012 was a
confounding factor. The grey shaded areas indicate the unirrigated period before the re-establishment of
the water channel in 2010 at the site Lens. Contrasts of linear mixed models were used to test for
differences between the irrigated trees and control trees (*P ≤ 0.05, lower row of stars for earlier and
upper row for later years).
Chapter III
101
Figure 5. Mean needle traits ± SE (n = 8) for three needle generations of the irrigated Scots pine and
corresponding control trees at the three sites (Varen, Planige and Lens). The needle traits compared are:
number of needles, needle density and total needle area per year of needle emergence measured in the
campaign 2012. The needle traits for the site Lens in 2012 are not shown, because the thinning in winter
2011/2012 was a confounding factor. Contrasts of linear mixed models were used to test for differences
between the irrigated and control trees (*P ≤ 0.05).
The dry period in the first half of 2011 led to a significant reduction in the earlywood
BAI of the control trees at the sites Varen and Planige, but the reduction of the
latewood was only significant at the site Varen (P≤0.05, Table 3). The shortest shoots in
the control trees were found in 2012 at both sites (Figure 3), even though the shoot
growth in 2012 was not significantly different from that in 2011. For the period 2008 to
2012, the needles of the control trees were shortest and thinnest in 2011, with
significantly shorter needles in 2011 than in 2010 (P≤0.05). The numbers of needles per
102
Chapter III
shoot in the control trees at the sites Varen and Planige were lowest in 2012, but not
significantly fewer than in the previous years (Figure 5). The early- and the latewood
BAIs of the irrigated trees at the site Varen were significantly reduced (P≤0.05) in 2011,
as were the shoot and the needle lengths and widths. The irrigated trees at the site
Planige showed, however, only a significant reduction in needle length (Table 3).
Figure 6. Mean Needle nitrogen
concentration ± SD (n=8) of the living
needles of the irrigated Scots pine and
corresponding control trees measured
in the campaigns 2010 and 2012 at the
site Lens. Data for 2012 are not shown,
because
the
thinning
in
winter
2011/2012 was a confounding factor.
The grey shaded area indicates the unirrigated time before the re-irrigation
in May 2010. Contrasts of the linear
mixed models were used to test
between the irrigated and control
trees (* P≤0.05, lower row of stars for
earlier and upper for later years).
Chapter III
103
Table 2. Spearman's correlation coefficients calculated for the mean DRI during the previous summer
(June to August), current spring (March to May) and current summer (June to August), and the annual
mean values of early- and late-wood basal area increments and shoot lengths of control Scots pine trees
and irrigated trees for the period 2000–12. (*P ≤ 0.05; **P ≤ 0.01).
Table 3. P-values from the calculated contrasts of the linear mixed models between the irrigated and
control trees of the extreme dry year 2011 and the years before 2010 and after 2012. Significant
differences are formatted in bold. All traits were sampled in 2012.
104
Chapter III
Discussion
In this study, we investigated the effects of irrigation on mature Scots pine trees on
two different time scales. The results showed that: (1) the direction of the responses to
differences in water availability were similar for all affected traits, although (2) the
magnitude of these responses to the short-term irrigation exceeded that of the longterm irrigation. The needle and shoot lengths, as well as the early- and latewood BAIs,
were most responsive to both short- and long-term differences in water availability,
while other traits were less responsive (needle width) or did not seem to be affected by
water availability (SLA and needle longevity).
Effects of long-term vs. short-term irrigation
The long-term effects of irrigation on the investigated pine forests and their soil
conditions were demonstrated by the increased indicator values at the sites Varen and
Planige, where the long-term irrigated soils had higher humidity, nutrient availability
and humus contents than the control soils in 2011 (Figure 2). These effects were still
visible 30 years after the irrigation was stopped at the site Lens, which indicates a longlasting improvement of the soils. As a consequence, the tree growth of both long-term
and short-term re-irrigated Scots pine increased and the trees’ total leaf area per shoot
was significantly larger than that of the trees growing under naturally dry conditions
(Figure 5), mainly due to the irrigated trees having longer shoots and needles (Figure
4). An increase in leaf area depends on water availability via the trade-off between
additional carbon gains and water loss through evapotranspiration (Girard et al. 2011),
as has also been found for pine species in other drought manipulation experiments
(Cinnirella et al. 2002; Grill et al. 2004; Lebourgeois et al. 1998). Moreover, the higher
nutrient availability for irrigated trees at the sites Varen and Planige (Figure 2) may
Chapter III
105
have led to an increase in the photosynthetic capacity, as indicated by the higher
needle N content of the re-irrigated trees in Lens (Figure 6). In addition, a higher water
availability has been found to increase the integrated stomatal conductance (Eilmann
et al. 2010). Thus, the trees growing at the water channels benefited from the
irrigation with a higher early- and latewood BAIs compared to control trees (Figure 3).
We found no differences in needle longevity for trees at the sites Varen and Planige
(Table 1), but a significantly lower SLA for irrigated trees than for control trees in 2009,
2010 and 2011 at the site Planige (Figure 4). At the site Lens, the SLA was significantly
reduced after two years of re-irrigation (in 2011). This is in contrast to findings from
interspecific comparisons, with leaves showing a higher longevity and a lower SLA for
species adapted to dry sites, and a shorter longevity and a higher SLA for those
adapted to mesic environments (Wright et al. 2001). Moreover, studies comparing
different pine species either found no relationships between water availability and
both SLA (P. radiata: Raison et al. 1992) and needle longevity (P. resinosa: Walters and
Gerlach 2013), or a negative relationship during very dry years (P. sylvestris Dobbertin
et al. 2010). Furthermore, Niinemets et al. (2001) found a decrease in SLA of Scots pine
with an increase in quantum flux density. However, at nutrient-limited sites, the trait
plasticity was lower than at nutrient-rich sites, leading to comparably higher SLA values
under nutrient-limited conditions (Niinemets et al. 2001). Thus, the high SLA values in
control plots in our study might be linked to a lower availability of soil nutrients (Figure
2) and the inability of SLA to adjust to high quantum flux rates rather than a direct
cause of lower water availability.
The traits of Scots pine that reacted most to the long-term irrigation at Varen and
Planige were also most responsive to the short-term effects of the re-irrigation at the
site Lens. However, the magnitude of the responses to the short-term irrigation greatly
106
Chapter III
exceeded that of trees irrigated since their establishment at the sites Varen and
Planige. In 2011, after two years of re-irrigation, early- and latewood BAIs increased by
about five to six times, while the long-term irrigated trees grew only about four times
more than corresponding control trees. In a previous investigation of Feichtinger et al.
(2014), Scots pine trees growing at the site Lens were investigated before and after the
irrigation stop in 1984: before the stop (1920 to 1983), the BAI of irrigated trees was
less than twice as high than that of the control trees. Thus, we can exclude site-specific
differences or differences in the quantity of water flowing through the channel as the
main reason for these high initial responses in radial growth at the site Lens. In our
study, the total leaf area per shoot of the trees re-irrigated for a period of two years
was more than three times larger compared to the corresponding control trees and the
needles were twice as long, while the total leaf area per shoot of long-term irrigated
trees was only twice as large and needles 50 % longer. In contrast, the magnitude of
the responses of the shoots to differences in water availability was similar for both the
short and long-term irrigated trees. Hence, the magnitude of response seems to
depend highly on the duration of the change in water availability.
A recent meta-study by Leuzinger et al. (2011) investigated the effect size of several
climate change drivers on tree growth and found in almost all cases a reduction in
response with time (>10 years). This effect was also studied by Barbeta et al. (2013),
who reported a dampening in the response of radial growth and stem survival to a
reduction in water availability with time (≥13 years) in Quercus ilex and Phillyrea
latifolia. Thus, the initially high increase in BAI (Figure 3) and needle growth (Figures 4,
5) in response to the short-term irrigation might decrease with time. Namely, to a
lower level of response with ongoing irrigation similar to the one we found in the longterm irrigated trees. One reason for this high response of recently re-irrigated trees is
Chapter III
107
that water and nutrient availability are tightly linked (e.g. Larcher 2003), as indicated
by the increase in the needle nitrogen concentration (Figure 6). However, the higher
nutrient availability related to the re-irrigation may decrease with time due to an
increased nutrient accumulating in plant biomass. As a consequence of this feedback,
the responses of trees to ongoing irrigation may be dampened with time.
Effects of climate
The extreme drought during the first half of 2011 (DRI January
to May =
-0.81, see
Supplementary Figure S2) led to trait-specific responses in Scots pine, with the control
trees having shorter and narrower needles than in 2010 (Table 3). This finding is in
agreement with that of other studies, showing that water availability during the
current year’s spring has a high impact on the needle area of pine species (Dobbertin et
al. 2010; Grill et al. 2004). In contrast to these direct responses, we found no significant
reductions in the number of needles and the shoot length in control trees in 2011
(Table 3). This is not surprising because buds would have already formed in the
previous year, and the shoot growth at our study sites is highly correlated with the
climatic water availability of the previous summer (Table 2). As a consequence, the
total needle area per shoot (Figure 5) of Scots pine trees in Valais in 2011 was less
reduced than the mean needle area (Figure 4). Hence, these delayed reactions to water
availability might buffer the negative impacts of extreme drought events on tree
performance.
As direct responses to the drought in 2011, the earlywood (Varen and Planige) and
latewood (Varen only) BAIs of control trees were significantly reduced. Surprisingly, the
BAI was not (earlywood) or only at one of our sites (latewood) correlated with the
climatic water availability during the period from 2000 to 2012, although radial growth
108
Chapter III
has been shown to be a good proxy for the effects of drought stress on trees in our
study region (see Affolter et al. 2010; Bigler et al. 2006; Eilmann et al. 2006). Our BAI
chronologies may have been too short (13 years) to detect the influence of climate on
tree-ring growth. Hence, a more suitable proxy for drought stress than radial growth
during shorter time periods might be the trait shoot length.
For irrigated trees at the site Varen, not only were the needle length significantly
reduced in 2011 but also the earlywood BAI. Moreover, the shoot growth of these
irrigated trees depended on the same extent on the climatic water availability as the
control trees (Table 2). This growth sensitivity to climate despite irrigation also affected
the radial growth of irrigated Scots pine trees growing along a water channel near our
site at Lens, investigated over a period of 100 years (Eilmann et al. 2009). Therefore,
the drought stress in 2011 must have been exceptionally high and irrigation insufficient
to fully eliminate any growth responses of needles, shoots and radial growth to
climate. In addition, the size of the trees may have influenced their responses, as the
irrigated trees at the site Varen were nearly twice as high (Table 1) and had a higher
total leaf area per shoot (Figure 5) than the corresponding control trees. As a
consequence, the irrigated trees at Varen were probably equally vulnerable to the
exceptionally dry conditions in 2011 than the control trees as has been found for
several other coniferous species (Lévesque et al. 2013; McDowell et al. 2006; Schuster
and Oberhuber 2013).
Conclusion
The morphological mechanisms underlying the responses of mature forests to longterm variations in water availability are still poorly understood, however, this
knowledge is crucial to be able to predict future forest developments in the presence
Chapter III
109
of climate change. We provided evidence that the responses of trees to long-term
changes in water availability are likely to be different in magnitude from short-term
responses. Consequently, studies are necessary to uncover the long-term processes
including changes in soil characteristics and the adjustment processes of trees over
time.
Furthermore, our study could show that drought mainly reduces the total leaf area per
shoot of Scots pine by limiting growth of needles and shoots as well as the numbers of
needles. As the needle longevity at our sites was about four to five years (Table 1),
needle and shoot reductions during drought years significantly lowered the potential
photosynthetic capacity of the trees for several subsequent years (Girard et al. 2011).
This smaller photosynthetic carbon uptake may contribute to the delayed recovery of
tree growth after extreme climatic events (Bigler et al. 2007; Girard et al. 2010).
We also found that the traits like needle, shoot and radial growth are affected
differently by climate, depending on the timing of water scarcity. Thus, understanding
the effects of drought on tree performance requires integrated approaches that
consider multiple traits. We suggest using the approved tree-ring analysis to track
growth responses over longer periods, combined with measurements of highly
responsive needle and shoot traits, which are more appropriate for short-term
tracking. Including additional morphological traits such as root growth (e.g. Brunner et
al. 2009; Herzog et al. 2014) and cell lumen and cell wall thickness (e.g. Eilmann et al.
2009; Fonti et al. 2013), as well as tree physiological traits such as stable carbon
isotopes in tree rings (Eilmann et al. 2010) would help to produce a more complete
picture of how trees respond to drought in both the short and long term.
110
Chapter III
Acknowledgments
We would like to thank Thomas Wohlgemuth for supporting us in organizing and
conducting the vegetation surveys, Anna Drewek for statistical advice, Arnaud
Giuggiola, Michael Mahlberg and Martin Loos for field assistance, Silvia Dingwall for
English correction, and Magdalena Nötzli and Loïc Schneider for technical help.
Funding
This study was funded by the NFP 61 - Swiss National Science Foundation (SNF)
Chapter III
111
Affolter, P., U. Büntgen, J. Esper, A. Rigling, P. Weber, J. Luterbacher and D. Frank.
2010. Inner Alpine conifer response to 20th century drought swings. European
Journal of Forest Research. 129:289-298.
Albaugh, T.J., H. Lee Allen, P.M. Dougherty and K.H. Johnsen. 2004. Long term growth
responses of loblolly pine to optimal nutrient and water resource availability.
Forest Ecology and Management. 192:3-19.
Allen, C.D., A.K. Macalady, H. Chenchouni, D. Bachelet, N. McDowell, M. Vennetier, T.
Kitzberger, A. Rigling, D.D. Breshears and E.H. Hogg. 2010. A global overview of
drought and heat-induced tree mortality reveals emerging climate change risks
for forests. Forest Ecology and Management. 259:660-684.
Barbeta, A., R. Ogaya and J. Peñuelas. 2013. Dampening effects of long-term
experimental drought on growth and mortality rates of a Holm oak forest.
Global Change Biology. 19:3133-3144.
Bates, D., M. Maechler, B. Bolker and S. Walker. 2014. lme4: Linear mixed-effects
models using Eigen and S4. R package version 1.1-7, <URL: http://CRAN.Rproject.org/package=lme4>. (13 January 2015, date last accessed).
Beier, C., C. Beierkuhnlein, T. Wohlgemuth, J. Penuelas, B. Emmett, C. Körner, H. Boeck,
J.H. Christensen, S. Leuzinger, I.A. Janssens, K. Hansen and J. Arnone. 2012.
Precipitation manipulation experiments - challenges and recommendations for
the future. Ecology Letters. 15:899-911.
Bigler, C., O.U. Bräker, H. Bugmann, M. Dobbertin and A. Rigling. 2006. Drought as an
inciting mortality factor in Scots pine stands of the Valais, Switzerland.
Ecosystems. 9:330-343.
Bigler, C., D.G. Gavin, C. Gunning and T.T. Veblen. 2007. Drought induces lagged tree
mortality in a subalpine forest in the Rocky Mountains. Oikos. 116:1983-1994.
Biondi, F. and F. Qeadan. 2008. A theory-driven approach to tree-ring standardization
defining the biological trend from expected basal area increment. Tree-ring
research. 64:81-96.
Brühlhart, H. 1999. Einfluss der Bewässerung auf das Jahrringwachstum von Kiefern.
Diplomarbeit. Berner Fachhochschule, Schweizerische Fachhochschule für
Holzwirtschaft, Biel.
Brunner, I., E.G. Pannatier, B. Frey, A. Rigling, W. Landolt, S. Zimmermann and M.
Dobbertin. 2009. Morphological and physiological responses of Scots pine fine
roots to water supply in a dry climatic region in Switzerland. Tree Physiology.
29:541-550.
Carlisle, A. and A.H.F. Brown. 1968. Pinus Sylvestris L. Journal of Ecology. 56:269-307.
Cinnirella, S., F. Magnani, A. Saracino and M. Boeghetti. 2002. Response of a mature
Pinus laricio plantation to a three-year restriction of water supply: structural
and functional acclimation to drought. Tree Physiology. 22:21-30.
Cotrufo, M.F., G. Alberti, I. Inglima, H. Marjanović, D. LeCain, A. Zaldei, A. Peressotti
and F. Miglietta. 2011. Decreased summer drought affects plant productivity
and soil carbon dynamics in a Mediterranean woodland. Biogeosciences.
8:2729-2739.
Dobbertin, M., B. Eilmann, P. Bleuler, A. Giuggiola, E. Graf Pannatier, W. Landolt, P.
Schleppi and A. Rigling. 2010. Effect of irrigation on needle morphology, shoot
and stem growth in a drought-exposed Pinus sylvestris forest. Tree Physiology.
30:346-360.
Dobbertin, M., P. Mayer, T. Wohlgemuth, E. Feldermeyer-Christle, U. Graf, N.E.
Zimmermann and A. Rigling. 2005. The decline of Pinus sylvestris L. forests in
the Swiss Rhône Valley - a result of drought stress? Phyton. 45:153-156.
112
Chapter III
Eilmann, B., N. Buchmann, R. Siegwolf, M. Saurer, P. Cherubini and A. Rigling. 2010.
Fast response of Scots pine to improved water availability reflected in tree-ring
width and δ13C. Plant, Cell & Environment. 33:1351-1360.
Eilmann, B., P. Weber, A. Rigling and D. Eckstein. 2006. Growth reactions of Pinus
sylvestris L. and Quercus pubescens Willd. to drought years at a xeric site in
Valais, Switzerland. Dendrochronologia. 23:121-132.
Eilmann, B., R. Zweifel, N. Buchmann, P. Fonti and A. Rigling. 2009. Drought-induced
adaptation of the xylem in Scots pine and pubescent oak. Tree Physiology.
29:1011-1020.
FAO. 2006. World reference base for soil resources, vol. 103. In FAO, World soil
resources reports, Rome.
Feichtinger, L.M., B. Eilmann, N. Buchmann and A. Rigling. 2014. Growth adjustments
of conifers to drought and to century-long irrigation. Forest Ecology and
Management. 334:96-105.
Fonti, P., O. Heller, P. Cherubini, A. Rigling and M. Arend. 2013. Wood anatomical
responses of oak saplings exposed to air warming and soil drought. Plant
Biology. 15:210-219.
Gärtner, H. and D. Nievergelt. 2010. The core-microtome: A new tool for surface
preparation on cores and time series analysis of varying cell parameters.
Dendrochronologia. 28:85-92.
Girard, F., M. Vennetier, F. Guibal, C. Corona, S. Ouarmim and A. Herrero. 2011. Pinus
halepensis Mill. crown development and fruiting declined with repeated
drought in Mediterranean France. European Journal of Forest Research.
131:919-931.
Girard, F., M. Vennetier, S. Ouarmim, Y. Caraglio and L. Misson. 2010. Polycyclism, a
fundamental tree growth process, decline with recent climate change: the
example of Pinus halepensis Mill. in Mediterranean France. Trees. 25:311-322.
Grill, D., M. Tausz, U. Pöllinger, M.J. Jiménez and D. Morales. 2004. Effects of drought
on needle anatomy of of Pinus canariensis. Flora - Morphology, Distribution,
Functional Ecology of Plants. 199:85-89.
Hampe, A. and R.J. Petit. 2005. Conserving biodiversity under climate change: the rear
edge matters. Ecology Letters. 8:461-467.
Herzog, C., J. Steffen, E. Graf Pannatier, I. Hajdas and I. Brunner. 2014. Nine years of
irrigation cause vegetation and fine root shifts in a water-limited pine forest.
PLoS ONE. 9:e96321.
Holm, S. 1979. A simple sequentially rejective multiple test procedure. Scandinavian
Journal of Statististics. 6:65-70.
Holmes, R.L. 1994. Computer-assisted quality control in tree-ring dating and
measurement. Tree-ring Bulletin. 43:69-78.
IPCC. 2013. Climate change 2013: The physical science basis. Contribution of working
group I to the fifth assessment report of the intergovernmental panel on
climate change. Eds. T.F. Stocker, D. Qin, G.-K. Plattner, M. Tignor, S.K. Allen, J.
Boschung, A. Nauels, Y. Xia, V. Bex and P.M. Midgle. Cambridge University
Press, Cambridge, United Kingdom and New York, NY, USA, 1535 pp.
Landolt, E., B. Bäumler, A. Erhardt , O. Hegg, F. Klötzli, W. Lämmler, M. Nobis, K.
Rudmann-Maurer, F. Schweingruber, J.-P. Theurillat, E. Urmi, M. Vust and T.
Wohlgemuth. 2010. Flora indicativa: Ökologische Zeigerwerte und biologische
Kennzeichen zur Flora der Schweiz und der Alpen. Haupt Verlag, Bern,
Switzerland.
Chapter III
113
Larcher, W. 2003. Physiological plant ecology: Ecophysiology and stress physiology of
functional groups, 4th Edn. Springer, Berlin Heidelberg New York.
Lebourgeois, F., G. Lévy, G. Aussenay, B. Clerc and F. Willm. 1998. Influence of soil
drying on leaf water potential, photosynthesis, stomatal conductance and
growth in two black pine varieties. Annals of Forest Science. 55:287-299.
Leibundgut, C. and I. Kohn. 2014. European traditional irrigation in transition part I:
Irrigation in times past-a historic land use practice across Europe. Irrigation and
Drainage. 63:273-293.
Leuzinger, S., Y. Luo, C. Beier, W. Dieleman, S. Vicca and C. Körner. 2011. Do global
change experiments overestimate impacts on terrestrial ecosystems? Trends in
Ecology & Evolution. 26:236-241.
Lévesque, M., M. Saurer, R. Siegwolf, B. Eilmann, P. Brang, H. Bugmann and A. Rigling.
2013. Drought response of five conifer species under contrasting water
availability suggests high vulnerability of Norway spruce and European larch.
Global Change Biology. 19:3184-3199.
MacKay, S.L., M.A. Arain, M. Khomik, J.J. Brodeur, J. Schumacher, H. Hartmann and M.
Peichl. 2012. The impact of induced drought on transpiration and growth in a
temperate pine plantation forest. Hydrological Processes. 26:1779-1791.
Martínez-Vilalta, J. and J. Piñol. 2002. Drought-induced mortality and hydraulic
architecture in pine populations of the NE Iberian Peninsula. Forest Ecology
and Management. 161:247-256.
Matías, L. and A.S. Jump. 2012. Interactions between growth, demography and biotic
interactions in determining species range limits in a warming world: The case
of Pinus sylvestris. Forest Ecology and Management. 282:10-22.
McDowell, N.G., H.D. Adams, J.D. Bailey, M. Hess and T.E. Kolb. 2006. Homeostatic
maintenance of ponderosa pine gas exchange in response to stand density
changes. Ecological Applications. 16:1164-1182.
Niinemets, Ü., D.S. Ellsworth, A. Lukjanova and M. Tobias. 2001. Site fertility and the
morphological and photosynthetic acclimation of Pinus sylvestris needles to
light. Tree Physiology. 21:1231-1244.
Oberhuber, W., W. Hofbauer and W. Kofler. 2001. Mortality and growth anomalies of
Scots pine (Pinus sylvestris L.) growing on drought exposed sites at the
Tschirgant Landslide (Tyrol). Berichte des Naturwissenschaftlich-Medizinischen
Vereins in Innsbruck 88: 87-97.
Oberhuber, W., M. Stumböck and W. Kofler. 1998. Climate-tree-growth relationships of
Scots pine stands (Pinus sylvestris L.) exposed to soil dryness. Trees-Structure
and Function. 13:19-27.
Ordoñez, J.C., P.M. van Bodegom, J.-P.M. Witte, I.J. Wright, P.B. Reich and R. Aerts.
2009. A global study of relationships between leaf traits, climate and soil
measures of nutrient fertility. Global Ecology and Biogeography. 18:137-149.
Parmesan, C. 2006. Ecological and evolutionary responses to recent climate change.
Annual Review of Ecology, Evolution, and Systematics. 37:637-669.
Pérez-Harguindeguy, N., S. Díaz, E. Garnier, S. Lavorel, H. Poorter, P. Jaureguiberry,
M.S. Bret-Harte, W.K. Cornwell, J.M. Craine, D.E. Gurvich, C. Urcelay, E.J.
Veneklaas, P.B. Reich, L. Poorter, I.J. Wright, P. Ray, L. Enrico, J.G. Pausas, A.C.
de Vos, N. Buchmann, G. Funes, F. Quétier, J.G. Hodgson, K. Thompson, H.D.
Morgan, H. ter Steege, L. Sack, B. Blonder, P. Poschlod, M.V. Vaieretti, G. Conti,
A.C. Staver, S. Aquino and J.H.C. Cornelissen. 2013. New handbook for
standardised measurement of plant functional traits worldwide. Australian
Journal of Botany. 61:167.
114
Chapter III
R Development Core Team. 2011. R: A language and environment for statistical
computing. In R Foundation for Statistical Computing, Vienna, Austria.
Raison, R.J., B.J. Myers and M.L. Benson. 1992. Dynamics of Pinus radiata foliage in
relation to water and nitrogen stress: I. Needle production and properties
Forest Ecology and Management. 52:139-158.
Rebetez, M. and M. Dobbertin. 2004. Climate change may already threaten Scots pine
stands in the Swiss Alps. Theoretical and Applied Climatology. 79:1-9.
Rigling, A., C. Bigler, B. Eilmann, E. Feldmeyer-Christe, U. Gimmi, C. Ginzler, U. Graf, P.
Mayer, G. Vacchiano, P. Weber, T. Wohlgemuth, R. Zweifel and M. Dobbertin.
2013. Driving factors of a vegetation shift from Scots pine to pubescent oak in
dry Alpine forests. Global Change Biology. 19:229-240.
Sánchez-Salguero, R., R.M. Navarro-Cerrillo, J.J. Camarero and Á. Fernández-Cancio.
2012. Selective drought-induced decline of pine species in southeastern Spain.
Climatic Change. 113:767-785.
Schuster, R. and W. Oberhuber. 2013. Age-dependent climate-growth relationships and
regeneration of Picea abies in a drought-prone mixed-coniferous forest in the
Alps. Canadian Journal of Forest Research. 43:609-618.
Thornthwaite, C. 1948. An approach toward a rational classification of climate.
Geographical review. 38:55-94.
Walters, M.B. and J.P. Gerlach. 2013. Intraspecific growth and functional leaf trait
responses to natural soil resource gradients for conifer species with contrasting
leaf habit. Tree Physiology. 33:297-310.
Weber, P., H. Bugmann and A. Rigling. 2007. Radial growth responses to drought of
Pinus sylvestris and Quercus pubescens in an inner-Alpine dry valley. Journal of
Vegetation Science. 18:777-792.
Western, A.W., R.B. Grayson and G. Blöschl. 2002. Scaling of soil moisture: a hydrologic
perspective. Annual Review of Earth and Planetary Sciences. 30:149-180.
Wright, I.J., P.B. Reich and M. Westoby. 2001. Strategy shifts in leaf physiology,
structure and nutrient content between species of high- and low-rainfall and
high- and low-nutrient habitats. Functional Ecology. 15:423–434.
Wright, I.J., P.B. Reich, M. Westoby, D.D. Ackerly, Z. Baruch, F. Bongers, J. CavenderBares, T. Chapin, J.H.C. Cornelissen, M. Diemer, J. Flexas, E. Garnier, P.K.
Groom, J. Gulias, K. Hikosaka, B.B. Lamont, T. Lee, W. Lee, C. Lusk, J.J. Midgley,
M.-L. Navas, Ü. Niinemets, J. Oleksyn, N. Osada, H. Poorter, P. Poot, L. Prior,
V.I. Pyankov, C. Roumet, S.C. Thomas, M.G. Tjoelker, E.J. Veneklaas and R.
Villar. 2004. The worldwide leaf economics spectrum. Nature. 428:821-827
Zweifel, R., A. Rigling and M. Dobbertin. 2009. Species-specific stomatal response of
trees to drought – a link to vegetation dynamics? Journal of Vegetation
Science. 20:442-454.
Chapter III
115
Figure S1. Mean monthly values of
precipitation
(bars)
from
the
weather station in Sierre (539 m
a.s.l.) and temperature (line) from
the weather station in Sion (492 m
a.s.l.) in Valais for the reference
period
from
1961
to
1990.
MAT=mean average temperature,
MAP=mean average precipitation.
Figure S2. Monthly values of the mean drought index (DRI) in Valais for the period from 2000 to 2012. The
black line represents a cubic smoothing spline.
116
Chapter III
Syntheses
117
Synthesis
The main objectives of this dissertation were to investigate the long-term adjustments
of radial growth, tree physiology and morphology to differences in water availability
and the impact of changes in water regime for dry forest ecosystems in inner-Alpine
valleys in Switzerland (Valais) and in Italy (Vinschgau). For this purpose, changes in
radial growth rates, stable carbon isotope discrimination and needle and shoot
morphological traits were assed.
The following chapter summarizes the methodological aspects and key findings of the
presented dissertation and concludes with recommendations for future research.
Methodological aspects
Multiproxy analysis
This thesis combines an analysis of tree-rings, tree physiology and morphological
needle and shoot traits in a multi-trait approach. For the investigations that used treering data, moving average correlations with monthly climate data gave a good
overview of the impact of climatic conditions on tree growth over time. In addition, an
analysis of stable carbon isotopes in tree rings was included to promote the
understanding of the physiological responses underlying radial growth patterns. This
provided novel insights on the adjustments in photosynthesis and stomatal
conductance in response to long-term differences in water availability. Such
physiological adjustments over time are often linked to changes in leaf area, and thus
needle and shoot morphological traits were also analysed to promote insights into
whole-tree adjustments to water availability.
118
Syntheses
This study is unique for covering a long temporal scale by comparing mature trees
irrigated since their establishment with trees growing under naturally dry conditions. In
fact, few long-term precipitation manipulation experiments have been conducted to
date. The presented results of this long-term investigation of species response can
therefore contribute to validating and extrapolate findings of manipulation
experiments at shorter temporal scales. Furthermore, long-term studies are crucial in
providing information for the creation of realistic model predictions of future forest
development at temporal scales relevant for ecosystem responses to climate change.
Moreover, the response of trees to (1) a sudden stop in irrigation, and (2) a two-year
re-irrigation were analysed, providing new information about the effects of irrigation
and drought at differing temporal scales. It was thus possible to disentangle traitspecific responses, including the magnitude of each specific response and the
identification of potential lags in these responses.
Study sites
Using dendrochronological methods, the response and sensitivity of tree growth to
differences in water availability of different conifers was investigated. In this
dissertation, the results from three Scots pine stands in Valais (Switzerland), as well as
two European larch and one black pine stand in Vinschgau (Italy) are outlined (chapter
I). Herein, a replicated sampling design was employed in order to gain representative
conclusions of the responses and the adjustment capacities of the investigated tree
species to drought. However, finding comparable forests stands with regard to
exposure, elevation and species composition, as well as sites with adequate numbers
of individual trees in the same age class growing along open water channels, provided
a challenge. For this reason it was not possible to sample larch and black pine stands
Syntheses
119
growing in Valais, whilst no suitable Scots pine stands were found in Vinschgau.
Nevertheless, by analysing three Scots pine as well as two European larch stands close
to each other, we were able to account for spatial heterogeneities and thereby any
site-specific effects. Overall, these forest stands have provided a unique set-up to
analyse tree species responses to irrigation and drought on the proposed long time
scale.
Further investigations covered the detailed analysis of growth and physiological
responses to water availability (chapter II). This latter part of the study focused on the
sudden stop in irrigation, which was only available for the Scots pine and European
larch stands. Even more restricted, only Scots pine could be considered for an in-depth
analysis of needle and shoot morphological traits (chapter III), as branches were not
accessible for the trees in Vinschgau. Namely, sampling relied on the use of extendable
clippers (up to 8 m), whilst both the larch and black pine trees were between 22 and 30
m in height, with a mean crown height of 10 m (data not shown). In contrast, Scots
pine trees were only between 6 and 17 m in height and thus, the lower enlightened
crown parts were always reachable.
120
Syntheses
Key findings of tree adjustments to irrigation and drought
This dissertation gave new insights into both the responses of trees to long-term
differences in water availability and their capacity to adjust to abrupt changes in the
water regime. Analysis of radial growth, tree physiology and morphology led to
corresponding results and helped to determine the mechanisms underlying tree
adjustment to drought. The direction and the magnitude in the response to irrigation
of all measured traits of Scots pine is summarised in Table 1.
Table 1. Summary of measured traits of Scots pine in Valais/Switzerland and the direction of their
response to (1) long-term irrigation, (2) short-term irrigation (2 years) and (3) a stop in irrigation compared
to corresponding control trees. Number of arrows indicate the magnitude in response.
Analysis
Measured trait
Response
Long-term
irrigation
Basal area increment
Tree physiology
Total ring width
↑
Earlywood
↑
↑↑
Latewood
↑↑
↑↑↑
Stable carbon isotopes
↓
Needle length
↑↑
↑↑↑
Needle width
↑
↑↑
Specific leaf area
↘
↓
↑
Needle numbers per shoot
↑
→
Needle density
↓
↓
Total leaf area
↑
↑
Needle longevity
→
→
↑↑
↑↑
Shoot length
↑ : increase
↓ : decrease
↗ : increase (but not at all sites/in all years)
↘ : decrease (but not at all sites/in all years)
→ : no response
Irrigation
stop
↓
↗
Needle nitrogen concentration
Tree morphology
Short-term
irrigation
Syntheses
121
Scots pine, European larch and black pine trees exposed to natural dry conditions in
Valais and Vinschgau were characterised by low rates of annual radial growth rates,
small tree size and leaf area per branch (Scots pine) when compared to irrigated trees
(chapters I & III). Adjustments in tree size and leaf area in response to low water
availability have previously been reported as mechanisms to minimize water loss
through transpiration (Larcher 2003; Martínez-Vilalta et al. 2009). Further adjustments
to drought include a more conservative stomatal control (Farquhar et al. 1989), as is in
line with the low values of Δ13C found for Scots pine and European larch growing under
natural dry conditions (chapter II). Adjustments in leaf area to water availability of
Scots pine was mainly achieved by developing shorter needles and shoots, as well as
decreasing the abundance of needles per shoot. Reductions in leaf area per dry mass
(SLA) and a prolongation of leaf longevity were also considered as mechanisms of
adjustments to xeric conditions (McDowell et al. 2011), but could not be confirmed in
our long-term study.
The long-term increase in water availability through the irrigation of Scots pine,
European larch and black pine trees led to an enhanced radial growth and taller trees
(chapters I, II & III), as well as to an increased leaf area per branch (Scots pine, chapter
III). In contrast, we found site-specific adjustments in tree physiology (Δ13C) between
irrigated and control trees of Scots pine and European larch (chapter II), which may be
a result of a decrease in hydraulic conductivity with increasing tree height (McDowell
et al. 2011). Furthermore, the investigated increase in leaf area is known to be
connected to a higher tree demand for water (Sohn et al. 2013). Consequently, the
positive effects of higher water availability on stomatal conductance may be
compensated by changes in the tree growth and height of Scots pine and European
larch, leading to similar Δ13C values in irrigated and control trees at two of our sites.
122
Syntheses
The higher growth rates of irrigated trees compared to corresponding control trees
may be achieved by higher C gains connected to the increase in leaf area rather than to
an increase in leaf-level photosynthesis.
Moreover, the response of Scots pine and European larch to changes in the water
regime was found to be non-linear over time. This non-linearity applied to (1) the
impact of (1) the abrupt stop in irrigation on radial tree growth and physiology
(chapters I & II) and (2) the short-term re-irrigation on tree growth and morphological
traits (chapter III). In both cases the magnitude in the response of trees to short-term
changes in water availability (< 10 years) exceeded those of the long-term changes.
This latter diminished response was particularly evident for radial growth and traits
affecting the leaf area, whereas, changes in the magnitude over time were less
pronounced for other parameter (e.g. Δ13C). This supports findings of a meta-study of
long-term investigations (>10 years) by Leuzinger et al. (2011), indicating that certain
tree responses to alterations in the drivers of climate change decrease with time.
Hence, the accuracy of model predictions strongly depend on how well findings from
manipulation experiments at a given temporal scale reflect those on longer time scales.
The observed non-linearity in the response of Scots pine and European larch to water
availability demonstrates that both species have a high phenotypic plasticity and are
able to adjust to substantial decrease in water availability. Although we were not able
to reconstruct possible tree mortality after the stop in irrigation, we know from the
local foresters that several trees died. In turn, the resulting decrease in competition
might have enhanced survival and growth of the remaining individuals (Lloret et al.
2012), leading to higher radial increments of Scots pine and larch growing at the
drained channels as compared to the control stands. This suggests that, when climate
gets warmer and drier, Scots pine and European larch growing at mesic sites and at
Syntheses
123
lower elevations in the Alps may face lower mortality rates than previously anticipated
(Lévesque et al. 2013).
This study further highlights trait-specific differences in the timing of the responses to
water availability. A lag in response to changes in the water regime of Scots pine
ranged between one (shoot length, needle numbers per shoot) and up to several years
(radial increment). On the one hand, such delayed responses may be attributed to a
growth sensitivity to previous, rather than current year’s climate conditions. On the
other hand, long-lasting increases in the humus content, nutrient availability and water
holding capacity induced by the century-long irrigation may have buffered the
response of pine to the decrease in water availability (chapters I & III). Thus, including
changes in the interactions between soils and trees are crucial to consider.
Recommendation for future research
The re-establishment of a channel at the site Lens was used to study the response of
trees to a two-year re-irrigation. Based on the effects of this re-irrigation on radial
growth and leaf area, it can be concluded that the initial responses of these traits
exceed the responses at a longer time scale. It may hence be desirable to monitor the
ongoing development of tree performance at this re-established channel for several
more years. Especially, soil properties and needle nitrogen concentrations, as well as
needle and shoot lengths, should be covered in following campaigns.
In order to further distinguish between the two possible physiological mechanisms
explaining drought-mediated forest mortality (McDowell et al. 2008), future
measurements may explore the changes in hydraulic architecture in response to longterm differences in water availability, e.g. wood anatomical traits, namely cell wall
124
Syntheses
thickness and cell size. Recent studies have already investigated changes in these traits
in response to differences in water availability for Scots pine (Eilmann et al. 2009;
Martínez-Vilalta et al. 2009; Sterck et al. 2008). These results were, however, often
contradicting, showing either smaller, larger or no differences in lumen sizes for trees
at dry sites compared to mesic sites. It would therefore be of interest to investigate all
three Scots pine stands in Valais, as well as the two European larch stands in
Vinschgau, in order to account for site-specific differences in the response of wood
anatomical traits to drought and irrigation.
Consequently, variations in the carbohydrate content of trees in response to changes in
water availability may provide further insights into the mechanisms of tree responses
to drought. A recent letter of Palacio (2014) concluded that the variability of
carbohydrates during drought is less critical than originally suggested due to a higher
priority of carbon storage over tree growth, and that only under extreme conditions,
such as long-lasting droughts, may carbon be depleted. To further investigate these
assumptions, the analysis of carbohydrates in combination with tree-ring widths of
Scots pine, European larch and black pine at our study sites may be of value.
Syntheses
125
References
Eilmann, B., R. Zweifel, N. Buchmann, P. Fonti and A. Rigling. 2009. Drought-induced
adaptation of the xylem in Scots pine and pubescent oak. Tree Physiology.
29:1011-1020.
Farquhar, G.D., J.R. Ehleringer and K.T. Hubick. 1989. Carbon discrimination and
photosynthesis. Annual Review of Plant Physiology and Plant Molecular
Biology. 40:503–537.
Larcher, W. 2003. Physiological plant ecology: Ecophysiology and stress physiology of
functional groups, 4th Edn. Springer, Berlin Heidelberg New York.
Leuzinger, S., Y. Luo, C. Beier, W. Dieleman, S. Vicca and C. Körner. 2011. Do global
change experiments overestimate impacts on terrestrial ecosystems? Trends in
Ecology & Evolution. 26:236-241.
Lévesque, M., M. Saurer, R. Siegwolf, B. Eilmann, P. Brang, H. Bugmann and A. Rigling.
2013. Drought response of five conifer species under contrasting water
availability suggests high vulnerability of Norway spruce and European larch.
Global Change Biology. 19:3184-3199.
Lloret, F., A. Escudero, J.M. Iriondo, J. Martínez-Vilalta and F. Valladares. 2012. Extreme
climatic events and vegetation: the role of stabilizing processes. Global Change
Biology. 18:797-805.
Martínez-Vilalta, J., H. Cochard, M. Mencuccini, F. Sterck, A. Herrero, J.F.J. Korhonen, P.
Llorens, E. Nikinmaa, A. Nolè, R. Poyatos, F. Ripullone, U. Sass-Klaassen and R.
Zweifel. 2009. Hydraulic adjustment of Scots pine across Europe. New
Phytologist. 184:353-364.
McDowell, N., W.T. Pockman, C.D. Allen, D.D. Breshears, N. Cobb, T. Kolb, J. Plaut, J.
Sperry, A. West, D.G. Williams and E.A. Yepez. 2008. Mechanisms of plant
survival and mortality during drought: why do some plants survive while others
succumb to drought? New Phytologist. 178:719-739.
McDowell, N.G., B.J. Bond, L.T. Dickman, M.G. Ryan and D. Whitehead. 2011.
Relationships between tree height and carbon isotope discrimination. In Sizeand Age-Related Changes in Tree Structure and Function Eds. F.C. Meinzer, B.
Lachenbruch and T.E. Dawson. Springer Netherlands, pp 255-286.
Palacio, S., G. Hoch, A. Sala, C. Körner and P. Millard. 2014. Does carbon storage limit
tree growth? New Phytologist. 201:1096-1100.
Sohn, J.A., T. Gebhardt, C. Ammer, J. Bauhus, K.-H. Häberle, R. Matyssek and T.E.E.
Grams. 2013. Mitigation of drought by thinning: Short-term and long-term
effects on growth and physiological performance of Norway spruce (Picea
abies). Forest Ecology and Management. 308:188-197.
Sterck, F.J., R. Zweifel, U. Sass-Klaassen and Q. Chowdhury. 2008. Persisting soil
drought reduces leaf specific conductivity in Scots pine (Pinus sylvestris) and
pubescent oak (Quercus pubescens). Tree Physiology. 28:529-536.
126
Syntheses
Curriculum vitae
1
Acknowledgments
Many people supported me throughout my dissertation project and with their support,
the final version of this dissertation was possible. Working on this thesis was a valuable
experience for me and I am glad that I could share this time with all those involved. In
particular I want to thank:
-
Andreas Rigling, for his constant support and motivation in numerous discussions
and for giving me the opportunity to study the forests of Valais and Vinschgau.
-
Britta Eilmann, for her supervision from WSL and Wageningen, her practical advice,
encouragements on scientific and personal matters and valuable comments on the
manuscripts,
-
Nina Buchmann, for the careful revisions of my manuscripts and support in
planning the “story”,
-
Kerstin Treydte for accepting to act as external supervisor,
-
Rolf Siegwolf, for the support concerning isotopes and the possibility to work at PSI
-
Mathieu Lévesque, for being a co-author, for valuable discussions and input and for
being a super office mate,
-
my other and not less super office mates Martina Hobi, Caroline Heiri and Lucía
Galiano, for having valuable discussions and companionship,
-
Julia Nabel and Ellen Pflug, for corrections of manuscripts and their caring and
support,
-
Alexander Bast, for all the statistical discussions, support, coffee and cake,
-
Arnaud Giggiola, for help in the field, discussions about science and (mostly) life,
-
Diego Galván, for his support and encouragements during the final stage of my
PhD,
-
Lisa Hülsmann, Lena Hellmann, Kathrin Kramer-Priewasser, Christoph Bachofen
and Oliver Jakoby for all the support, walks and talks,
-
Silvia Dingwall and Gregory Tomlinson for language corrections,
-
Thomas Wohlgemuth for the help with indicator values and the taming of herb
layers,
Curriculum vitae
-
2
Anne Verstege, Magdalena Nötzli, Loïc Schneider, Michael Mahlberg, Gustav
Schneiter, Dieter Trummer, for technical support and help in the field ,
-
David Frank and Arthur Gessler, for valuable discussions,
-
Anna Drewek, for statistical advice.
Finally, I want to thank my family for their caring, support and understanding during
my PhD. And of course I want to thank Martin Loos, for his love and companionship, his
encouragement and support with field work and manuscripts.