Tree Physiology Advance Access published March 21, 2011
Tree Physiology 00, 1–11
doi:10.1093/treephys/tpr004
Research paper
Provenance-specific growth responses to drought and air
warming in three European oak species (Quercus robur,
Q. petraea and Q. pubescens)
Swiss Federal Research Institute for Forest, Snow and Landscape Research (WSL), Zürcherstrasse 111, Birmensdorf CH-8903, Switzerland;
1Corresponding author ([email protected])
Received October 1, 2010; accepted January 13, 2011; handling Editor Marc Abrams
Provenance-specific growth responses to experimentally applied drought and air warming were studied in saplings of three
European oak species: Quercus robur, Quercus petraea and Quercus pubescens. Four provenances of each species were
grown in large open-top chambers and subjected to four climates: control, periodic drought, air warming or their combination in 3 subsequent years. Overall growth responses were found among species and provenances, with drought reducing
shoot height growth and stem diameter growth and air warming stimulating shoot height growth but reducing stem diameter
growth and root length growth. Differential growth responses in shoots, stems and roots resulted in altered allometric
growth relations. Root length growth to shoot height growth increased in response to drought but decreased in response to
air warming. Stem diameter growth to shoot height growth decreased in response to air warming. The growth responses in
shoots and stems were highly variable among provenances indicating provenance-specific sensitivity to drought and air
warming, but this response variability did not reflect local adaptation to climate conditions of provenance origin. Shoot
height growth was found to be more sensitive to drought in provenances from northern latitudes than in provenances from
southern latitudes, suggesting that genetic factors related to the postglacial immigration history of European oaks might
have interfered with selective pressure at provenance origins.
Keywords: dryness, provenances, temperature.
Introduction
Global warming is considered a serious threat to forest ecosystems in the twenty-first century. The predicted increase in
frequency and intensity of heatwaves and summer droughts
will have strong impacts on tree growth and viability, thereby
changing the stability and productivity of natural and managed forests (Ciais et al. 2005, Leuzinger et al. 2005). The
ability of individual tree species to cope with such environmental stresses needs to be considered in future silvicultural
strategies. For example, a risk assessment for European tree
species revealed potential risks of habitat loss for droughtsensitive species, e.g., Fagus sylvatica, Picea abies and Abies
alba (Ohlemüller et al. 2006, Gessler et al. 2007), while
­ rought-tolerant species, e.g., Quercus spp. (oaks), will face
d
much lower risks of habitat loss. Indeed, oaks might even benefit from climate change as they are expected to become
increasingly competitive with tree species less tolerant to
drought (Zimmermann et al. 2006).
The genus Quercus (oaks) comprises ~400 deciduous and
evergreen tree and shrub species occupying a wide variety of
habitats in temperate, Mediterranean, subtropical and tropical
areas (Kleinschmit 1993, Rushton 1993, Nixon 2006). Accor­
ding to their widespread distribution, oaks evolved different
morphological and physiological traits enabling them to grow
on sites with contrasting environmental conditions. This especially applies to drought- and flood-prone sites where water
© The Author 2011. Published by Oxford University Press. All rights reserved. For Permissions, please email: [email protected]
Downloaded from treephys.oxfordjournals.org at ETH-Bibliothek on March 21, 2011
Matthias Arend1, Thomas Kuster, Madeleine S. Günthardt-Goerg and Matthias Dobbertin
2 Arend et al.
adapted to cope with this environmental heterogeneity.
However, there is a ­considerable lack of information on
whether and how oaks in Switzerland adapted to local differences in temperature and precipitation, especially with regard
to growth traits that have a strong impact on competitive ability and long-term establishment in plant communities (Coomes
and Allen 2007).
The aim of the present study was to (i) evaluate overall
growth responses in oaks to experimentally applied drought
and air warming, (ii) assess species- and provenance-specific
differences in these responses and (iii) test whether provenance-specific differences are related to local climate variables
of provenance origin. For this purpose, four provenances from
Q. robur, Q. petraea and Q. pubescens each were selected
from climatically different sites in Switzerland and Italy and
subjected during three growth periods to drought and air
warming. The oaks were harvested at the end of the growth
experiment and analysed for shoot, stem and root growth
characteristics.
Materials and methods
Provenance sites
Seeds were collected in 2003 from 5–10 mature trees of
Q. robur L., Q. petraea [Matt.] Liebl. and Q. pubescens Willd. in
11 natural oak stands throughout Switzerland and additionally
in 1 natural oak stand in Italy (Tuscany). Species identification
was based on morphological leaf characteristics and published
information on the genetic structure of the selected oak stands
(Finkeldey and Mátyás 2003). The collection sites covered a
wide range of environmental conditions, with annual temperature ranging from 6.1 to 14.0 °C and annual precipitation ranging
from 657 to 1668 mm. Altitude, latitude and longitude of the
collection sites ranged from 310 to 900 m above sea level (a.s.l.),
43°34′ to 47°38′N and 6°41′ to 12°4′E (Table 1). Collected
­germinating seeds were sown in spring 2004 in a nursery field
Table 1. Location and climatic characterization of provenance sites. Mean annual temperature (°C) and mean annual precipitation (mm) were taken
from SWISS METEO stations located near the provenance sites.
Oak species
Provenance site
Latitude
Longitude
Altitude
Temperature
Precipitation
Q. robur
Q. robur
Q. robur
Q. robur
Q. petraea
Q. petraea
Q. petraea
Q. petraea
Q. pubescens
Q. pubescens
Q. pubescens
Q. pubescens
Tägerwilen
Bonfol
Hühnenberg
Magadino
Corcelles
Magden
Wädenswil
Gordevio
Leuk
Roverella (Italy)
LeLanderon
Promontogno
47°38′N
47°28′N
47°11′N
46°09′N
46°51′N
47°32′N
47°14′N
46°12′N
46°18′N
43°34′N
47°04′N
46°20′N
9°08′E
7°09′E
8°25′E
8°53′E
6°41′E
7°48′E
8°38′E
8°44′E
7°38′E
12°4′E
7°03′E
9°33′E
510
450
398
199
550
308
430
450
720
310
700
900
8.7
8.9
9.1
10.5
9
8.9
8.9
11
8.1
14
8
6.1
929
1035
1147
1772
893
974
1353
1668
657
768
932
1459
Tree Physiology Volume 00, 2011
Downloaded from treephys.oxfordjournals.org at ETH-Bibliothek on March 21, 2011
availability is the main environmental constraint affecting individual plant growth and viability. Oaks are generally considered
to be less sensitive to drought because of deep-penetrating
roots, xeromorphic leaf structure and effective stomatal control
of transpirational water loss (Abrams 1990, Kubiske and
Abrams 1993). These adaptational traits enable them to maintain high water potentials during drought conditions, thus
avoiding severe stress- or even drought-induced mortality.
Notwithstanding, drought tolerance may differ considerably
among oak species and provenances, reflecting adaptation to
environments with different water availability (Dickson and
Tomlinson 1996).
Quercus robur, Quercus petraea and Quercus pubescens
­represent the most abundant oak species in Central Europe
growing mainly on sites with less or irregular soil water supply where they are competitively superior to drought-sensitive
tree species, e.g., F. sylvatica and P. abies (Ellenberg 1996,
Aas 1998). Quercus robur is, however, not solely restricted to
drought-prone sites, as it also occupies hydromorphic soils
experiencing periodic flooding. This ecological difference
­suggests a higher requirement for water in Q. robur than in Q.
petraea and Q. pubescens, which is well reflected by lower
water use efficiency (Ponton et al. 2002), higher sensitivity to
soil water depletion (Vivin et al. 1993) or lower stomatal
responsiveness to decreasing air humidity (Gieger and
Thomas 2005). Quercus pubescens, on the other hand, is
commonly considered to be more tolerant to drought and
requiring higher growth temperatures than the two other oak
species because its main distribution is located in the warmer
and dryer climates of the Mediterranean area of southern
Europe.
In Switzerland, these three oak species can be found in
pure or mixed forests on climatically contrasting sites spanning a wide range of annual temperature and precipitation
(Ellenberg and Klötzli 1972, Brändli 1996). Accordingly, it
seems reasonable to expect that these oaks are locally
Growth response of oaks to drought and warming 3
at the Swiss Federal Research Institute WSL, Birmensdorf and
grown for 2 years under ambient site conditions.
Experimental design
Measurement of soil water status and microclimatological
parameters
Volumetric soil water content was measured by time domain
reflection (TDR 100; Campbell Scientific Inc., Logan, UT, USA)
in each lysimeter compartment at 12, 38, 62 and 88 cm depth
at intervals of 1 week throughout the growing seasons and
2–3 weeks out of the growing seasons. Soil–water matrix
potentials were deduced from soil water content and physical
soil properties using the RETC (retention curve) code
(VanGenuchten 1980, VanGenuchten et al. 1991). Air temperature and relative air humidity were measured with shaded
EL-USB-2 data ­loggers (Lascar Electronics Ltd, Salisbury, UK)
Growth measurements
Shoot height growth was determined annually on current-year
leader shoots, stem diameter growth was measured annually
at the stem base (10 cm above ground) and maximal root
length growth was determined on the main root axis after final
root excavation in winter 2009. Annual shoot and stem growth
for the 3 years of the experiment was added to total growth.
Whole-tree leaf area was calculated for each indi­vidual oak tree
from leaf biomass formed in 2009 and leaf specific area. For
determination of leaf biomass, leaves were completely harvested in early September 2009, oven dried for 2 days at
65 °C and weighed. Mean leaf specific area was determined
on five morphologically representative leaves per tree by measuring single leaf area and single leaf dry weight.
Statistical analysis
All statistical calculations were performed with SPSS 15.0
(SPSS Inc., Chicago, IL, USA). Overall treatment effects and
interactions among treatments were analysed as a three-­
factorial design (irrigation, temperature and provenance) by
analysis of ­variance (ANOVA; general linear model procedure).
Significance of control–treatment differences within provenances was tested by the Student’s t-test in combination with
Bonferroni correction. Relationships between provenance-specific growth responses and environmental variables were estimated by Pearson’s correlation coefficients and Spearman’s
rank correlation. Treatments, interactions between treatments,
control–treatment differences and correlations were considered significant when P < 0.05. All statistical calculations are
based on eight replicate trees per treatment and provenance.
Results
Soil water status and microclimate
Soil water status measured at four different soil depths (12,
38, 62 and 88 cm) differed between control and drought treatments, depending on irrigation regime and season. The 62 cm
soil depth was chosen for presentation because it was the
deepest soil layer that could be accessed by all oaks in all
treatments at the end of the 3-year growth period (Figure 1; cf.
Table 3). In control and air warming treatment, soil water contents ranged approximately from 20 to 30% out of the growing
season and from 15 to 25% within the growing season. Minimal
and maximal soil water content in both treatments was equivalent to soil water matrix potentials of approximately − 30 and
−70 hPa, respectively. In drought and combination treatment,
soil water content ranged approximately from 12 to 20% out of
Tree Physiology Online at http://www.treephys.oxfordjournals.org
Downloaded from treephys.oxfordjournals.org at ETH-Bibliothek on March 21, 2011
The growth study was part of the interdisciplinary ‘Querco’
experiment carried out in the model ecosystem facility of the
Swiss Federal Research Institute WSL, Birmensdorf, Switzerland
(47°21′54″N, 8°27′5″E, 450 m a.s.l.). The research facility
comprises 16 large open-top chambers (OTCs) arranged in a
Latin square with four replicates per treatment (control,
drought, air warming and their combination). Each hexagonal
OTC was 3 m in height and 6.7 m2 in area, with movable side
walls of glass. Below ground, each OTC had two concretewalled lysimeter compartments with a depth of 1.5 m and a
surface area of 3 m2. Each lysimeter contained a 50 cm drainage layer of pure quartz gravel, covered by forest soil and an
85 cm layer of acidic Haplic Alison subsoil (loamy sand, pH
4.1) with a 15 cm layer of top soil. Natural precipitation was
excluded from the OTCs by retractable glass roofs closing
automatically at the onset of rainfall.
In spring 2006, the 2-year-old oak saplings were transplanted from the nursery field into the OTCs (two individuals
from each provenance per lysimeter in a randomized design).
All trees were grown with sufficient water supply and ambient
air temperature for one season to ensure successful establi­
shment. Water with an ionic composition equivalent to the 30
years’ mean natural precipitation at the experimental site was
supplied at intervals of 2–3 days, from May to October, by
means of six sprinklers in each lysimeter compartment
(10 l m−2). From October to March, the roofs of the OTCs were
left open to allow natural precipitation. The side walls of the
OTCs were fully open to prevent passive air warming. Drought,
air warming and combination treatments started in spring 2007,
with four OTCs assigned for each climate treatment and four
OTCs assigned for controls. In drought and combination treatment, water supply was reduced during the growing season by
temporary interruption of the irrigation. In air warming and
combination treatment, daytime air temperatures were elevated
by reducing the opening angle of the side walls of the OTCs.
in each growth chamber at 120 cm height at intervals of 1 h.
Solar radiation outside the growth chambers was continuously
monitored with an SP-LITE Silicon Pyranometer (Kipp & Zonen,
Delft, The Netherlands).
4 Arend et al.
the growing season and from 5 to 15% within the growing
season. Equivalent soil water matrix potentials in both drought
treatments were less than −2000 hPa when soil water content
decreased below 5%.
Daytime air temperatures in the warming, drought and combination treatment were higher than in the control, especially
during the growing seasons when solar radiation was high
(Figure 2). During these periods, daytime air temperatures
increased by up to 2 °C in the air warming treatment and by up
to 3 °C in the combination treatment in relation to the control.
In the drought treatment, daytime air temperatures were
increased by up to 0.8 °C in the second year and by up to 2 °C
in the third year. Relative air humidity was nearly comparable in
all treatments during the growing seasons with mean values of
Figure 2. Increase in day air temperature (08:00–18:00; monthly
means) in the drought (black dashed line), air warming (grey solid
line) and combination (grey dashed line) treatments in relation to the
control. Grey columns show monthly means of solar radiation.
Overall growth responses to drought and warming
Multi-factorial ANOVA in combination with descriptive statistics
revealed overall effects of drought and air warming on oak
growth irrespective of differences between species or provenances (Table 2; main effects and interactions including provenances). Drought was found to be a strong environmental
constraint inhibiting growth in stems and shoots to different
extents. Shoot height growth and stem diameter growth
decreased by 44.5 and 42.2%, respectively, whereas root
length growth was not significantly affected by drought. These
growth responses in shoots and stems were influenced by significant ‘drought × provenance’ interactions, indicating provenance-specific differences in sensitivity to drought. Growth
inhibition in shoots was additionally influenced by a significant
‘drought × temperature’ interaction, resulting in stronger inhibition of shoot height growth in oaks subjected to a combination
of drought and air warming (48.5%) than in oaks subjected to
drought alone (39.7%). Different effects on oak growth, partially opposite to those in drought treatments, were found in
oaks subjected to air warming alone. Shoot height growth in
these oaks increased by 14.0% while stem diameter growth
and root length growth decreased slightly by 6.4 and 5.1%,
respectively.
Provenance-specific growth responses to drought
and warming
Total growth in shoots, stems and roots of non-treated oaks
was highly variable among provenances, with provenance
means for shoot height growth ranging from 50 to 273 cm,
provenance means for stem diameter growth ranging from 0.4
to 2.0 cm and provenance means for root length growth ranging from 67 to 124 cm (Table 3). For this reason, the collected
growth data were normalized for differences in total growth by
calculating control–treatment relations, thus allowing reliable
comparisons of treatment responses among provenances. Root
length growth was not considered for provenance-specific
comparisons as it was not or less affected by the drought and
air warming treatments (Table 3). Above-ground growth characteristics were often inhomogeneous within provenances,
Table 2. Main effects of drought (well watered vs. interrupted irrigation) and air warming (ambient vs. elevated air temperature) on stem, shoot
and root growth, as calculated with overall ANOVA. Interactions were calculated between climate treatments as well as between climate treatments
and provenance.
Growth trait
Drought (df 1)
Warming (df 1)
Drought × provenance (df 11)
Warming × provenance (df 11)
Drought × warming (df 1)
Stem
Shoot
Root
***
*
***
***
***
***
**
n.s.
*
n.s.
n.s.
n.s.
n.s.
n.s., not significant; df, degree of freedom.
Levels of significance: *P < 0.05; **P < 0.01; ***P < 0.001.
Tree Physiology Volume 00, 2011
n.s.
n.s.
Downloaded from treephys.oxfordjournals.org at ETH-Bibliothek on March 21, 2011
Figure 1. Volumetric soil water content and equivalent soil–water
matrix potential at −62 cm soil depth in the drought (black dashed
line), air warming (grey solid line) and combination (grey dashed line)
treatments compared with the control (black solid line). Grey labelled
areas indicate periods without irrigation in the drought and combination treatments.
73.4, 72.3, 71.4 and 69.7% in control, air warming, drought
and combination treatments, respectively.
Growth response of oaks to drought and warming 5
Table 3. ​Total stem diameter, shoot height and root length growth (cm) of the oak provenances in control, drought, air warming and combination
treatment for the 3 years of the growth study.
Species/provenance
Q. robur
Tägerwilen
Bonfol
Hühnenberg
Magadino
Magden
Wädenswil
Gordevio
Q. pubescens
Leuk
Roverella
LeLanderon
Promontogno
Drought
Air warming
Combination
Stema
Shoota
Root
Stema
Shoota–c
Rootb
Stema, c
Shoota–c
Root
Stema
Shoota–c
Root
2.0 ± 0.2
273 ± 36
112 ± 17
1.9 ± 0.4
175 ± 33
124 ± 29
1.6 ± 0.3
194 ± 13
95 ± 18
1.8 ± 0.4
189 ± 62
106 ± 15
1.0 ± 0.2
104 ± 28
90 ± 12
1.0 ± 0.3
113 ± 32
119 ± 29
1.0 ± 0.2
132 ± 33
90 ± 12
1.1 ± 0.2
126 ± 38
111 ± 32
1.9 ± 0.3
271 ± 35
89 ± 14
1.7 ± 0.3
249 ± 52
101 ± 18
1.8 ± 0.2
272 ± 51
92 ± 17
1.9 ± 0.4
259 ± 25
102 ± 16
0.9 ± 0.2
125 ± 28
105 ± 13
1.0 ± 0.5
124 ± 19
105 ± 16
0.9 ± 0.2
134 ± 34
98 ± 11
0.9 ± 0.2
120 ± 29
101 ± 8
Stema
Shoot a, b
Root
Stema
Shoota
Roota
Stema
Shoota
Root
Stema
Shoota
Root
1.3 ± 0.3
167 ± 46
82 ± 20
1.3 ± 0.2
192 ± 44
79 ± 11
1.3 ± 0.5
172 ± 49
88 ± 20
1.4 ± 0.4
156 ± 58
67 ± 9
0.7 ± 0.2
104 ± 16
77 ± 12
0.8 ± 0.3
102 ± 27
94 ± 12
0.8 ± 0.3
84 ± 10
84 ± 30
0.9 ± 0.1
92 ± 18
76 ± 9
1.2 ± 0.2
215 ± 34
77 ± 17
1.5 ± 0.4
244 ± 87
73 ± 20
1.1 ± 0.2
182 ± 57
80 ± 31
1.1 ± 0.3
165 ± 71
69 ± 9
0.9 ± 0.3
137 ± 29
93 ± 10
0.9 ± 0.3
123 ± 44
83 ± 8
0.6 ± 0.2
106 ± 48
81 ± 12
0.8 ± 0.3
82 ± 42
73 ± 13
Stema
Shoota
Root
Stem
Shoot
Root
Stema
Shoota
Root
Stema
Shoota
Root
1.6 ± 0.4
201 ± 68
72 ± 10
0.4 ± 0.3
50 ± 30
69 ± 17
1.5 ± 0.5
200 ± 60
86 ± 24
1.3 ± 0.2
170 ± 42
76 ± 17
1.1 ± 0.3
144 ± 23
88 ± 14
0.3 ± 0.2
43 ± 22
73 ± 18
0.9 ± 0.1
110 ± 35
81 ± 14
0.8 ± 0.2
130 ± 28
73 ± 10
1.5 ± 0.3
247 ± 50
76 ± 13
0.4 ± 0.2
72 ± 47
71 ± 10
1.6 ± 0.4
230 ± 73
80 ± 11
1.1 ± 0.3
174 ± 63
66 ± 11
0.8 ± 0.2
122 ± 33
76 ± 8
0.2 ± 0.1
30 ± 21
70 ± 16
0.7 ± 0.2
123 ± 17
76 ± 15
0.7 ± 0.2
103 ± 28
68 ± 11
Lowercase letters indicate significant treatment effects of (a) drought, (b) air warming and (c) their interaction within provenances with P < 0.05
as calculated with provenance-specific ANOVAs according to a two-factorial test design (irrigation and temperature). Values are means ± SD;
n = 8.
making it difficult to find significant ­treatment responses, especially in the air warming treatment where control–treatment
­differences were rather small.
Growth in shoots and stems was found to be strongly inhibited by drought, but with high provenance variation in the
extent of responses, indicating provenance-specific differences
in drought sensitivity (Figure 3). Declines in shoot height growth
and stem diameter growth varied from 15% in Q. ­pubescens
Roverella to 62% in Q. robur Tägerwilen and from 12% in
Q. pubescens Roverella to 50% in Q. robur Tägerwilen, ­respectively.
This high variability was less obvious when treatment responses
were calculated at the species level. Species-specific shoot
height growth decreased by 27.8% in Q. pubescens, 40.6% in
Q. robur and 44.2% in Q. petraea. Species-specific stem diameter growth decreased by 31.2% in Q. pubescens, 40.3% in Q.
petraea and 41.2% in Q. robur. The responses of shoots and
stems to drought were modified in a few provenances by concomitant air warming, albeit general response trends in the
combination treatment were mostly similar to those in the
drought treatment (Figure 3). Drought responses, for instance,
Tree Physiology Online at http://www.treephys.oxfordjournals.org
Downloaded from treephys.oxfordjournals.org at ETH-Bibliothek on March 21, 2011
Q. petraea
Corcelles
Control
6 Arend et al.
were strongly diminished by air warming in Q. petraea Corcelles
but strongly enhanced in Q. pubescens Leuk and Roverella,
indicating provenance-dependent interaction between drought
and air warming. Air warming alone had a comparatively low
effect on oak growth, except in Q. robur Bonfol, Hühnenberg
and Magadino in which shoot height growth increased
­significantly by ~40% in response to this stimulus (Figure 4).
Either shoot height growth in other provenances was unresponsive to air warming or effects were not statistically significant
due to high heterogeneity of above-ground growth. Stem diameter growth did not significantly respond to air warming in any of
the studied provenances (Figure 4).
Growth responses of provenances to drought and air warming were additionally checked for correlations with environmental variables of provenance origin to test for local adaptation.
Annual precipitation, annual temperature, longitude and altitude could not be statistically related to growth responses in
shoots and stems (data not shown). However, there was a distinct trend towards increasing drought sensitivity of shoot and
stem growth with increasing latitude, i.e., shoot height growth
and stem diameter growth were overall more sensitive to
drought in provenances originating from regions north of the
Swiss Alps than in provenances originating from regions at the
southern border of the Swiss Alps or from Italy (Figure 5).
Statistical analysis revealed significant correlations between
latitude of provenance origin and drought sensitivity of shoot
Tree Physiology Volume 00, 2011
and stem growth (shoot growth: R = 0.699, P < 0.05; stem
growth: R = 0.657, P < 0.05; Spearman’s rank correlation).
Effects on allometric growth relations
Analysis of overall oak growth responses provided evidence
that growth in shoots, stems and roots is differently affected
by drought and air warming (cf. Tables 2 and 3). Ratios of stem
diameter growth to shoot height growth (SD/SH) and root
length growth to shoot height growth (RL/SH) were calculated
for each provenance in control, drought and air warming
­treatments to gain more information on the effects of drought
and air warming on these allometric traits (Figure 6). Only in
some provenances were the effects statistically significant.
Nevertheless, clear response trends were found among provenances, with decreased SD/SH and RL/SH ratios in oaks
­subjected to air warming and increased RL/SH ratios in oaks
subjected to drought. Different behaviour was observed in
Q. pubescens Promontogno in which RL/SH ratios appeared to
be unresponsive to drought and air warming. Quercus pubescens Roverella differed from the other provenances in that it
had exceptionally high RL/SH ratios in all treatments.
Effects of drought and air warming on whole-tree
leaf area
The responses of whole-tree leaf area to drought and air warming were deduced from leaf biomass formed in 2009 and leaf
Downloaded from treephys.oxfordjournals.org at ETH-Bibliothek on March 21, 2011
Figure 3. Inhibition of shoot height growth and stem diameter growth by drought in different oak provenances. Columns are relative differences
between mean values from the control and drought treatments; arrows show enhancement or impairment of drought responses by concomitant air
warming in the combination treatment; open circles represent annual precipitation at the provenance site (filled columns indicate significant
­differences between control and drought treatments with P < 0.05, Student’s t-test in combination with Bonferroni correction).
Growth response of oaks to drought and warming 7
specific mass. Overall ANOVA revealed a significant drought
effect reducing whole-tree leaf area by ~31% across all provenances (P < 0.001). Air warming and interactions between
treatments and with provenance had no significant effect on
this trait. Reduction of whole-tree leaf area by drought was
found in all provenances, but many of the provenance-specific
drought effects were not statistically significant because of the
exceptionally high variability within provenances (Table 4).
Discussion
Responses of oaks to drought and air warming
Increasing drought and air warming are expected to change
tree growth under conditions of climate change. While drought
is known to limit tree growth by suppressing photosynthetic
carbon sequestration and causing severe stress (Jolly et al.
2005, Bréda et al. 2006), warming is generally thought to act
in an opposite way as it facilitates carbon allocation to internal growth processes (Saxe et al. 2001). Such generalization
can however be difficult since the effects of drought and
warming on tree growth depend on the strength of these
environmental factors and may vary between genotypes and
growth processes (Dobbertin 2005). Growth responses may
even change with tree age, resulting in different behaviour of
juvenile and mature trees. In the present long-term study,
responses of stem, shoot and root growth to experimentally
applied drought and air warming were investigated in saplings from 12 oak provenances originating from climatically
different sites in Switzerland and Italy. The following general
growth responses were observed among these provenances:
(i) shoot height growth was reduced by drought and stimulated by air warming, (ii) stem diameter growth was reduced
by drought and to a much lesser extent by air warming and
(iii) root length growth was slightly reduced by air warming
but unresponsive to drought. These overall findings are largely
consistent with previous studies (Kozlowski 1982, Vivin et al.
1993, Collet et al. 1997, Thomas 2000, Saxe et al. 2001,
Ponton et al. 2002), with the exception of the missing root
growth response in the drought treatments, which was contradictory to studies showing impaired root elongation in oaks
grown with severe soil moisture stress (Teskey and Hinckley
1981, Fort et al. 1997).
Root responses to drought are still a matter of debate since
previous studies have reported controversial results (cf. Joslin
et al. 2000). This controversy may be attributable to different
experimental designs in root growth studies, testing trees in
different ontogenetic stages or studying trees with different
root growth characteristics. In our study, the missing drought
response can be mainly explained by the strength of the
applied drought treatments and the deep-reaching root system
of the studied oaks. Indeed, it is well established that root
growth in trees declines as soil water potential decreases
Tree Physiology Online at http://www.treephys.oxfordjournals.org
Downloaded from treephys.oxfordjournals.org at ETH-Bibliothek on March 21, 2011
Figure 4. Shoot height growth and stem diameter growth in oak provenances subjected to air warming. Columns are relative differences between
mean values from the control and warming treatment; open circles represent annual temperature at the provenance site (filled columns indicate
significant differences between control and warming treatment with P < 0.05, Student’s t-test in combination with Bonferroni correction).
8 Arend et al.
Figure 6. Allometric relations of stem diameter to shoot height growth and root length to shoot height growth in oak provenances subjected to air
warming (black columns) or drought (white columns) compared with the control (grey columns). Values are means ± SE. (a, significant difference
between control and warming treatment; b, significant difference between control and drought treatment; P < 0.05, Student’s t-test in combination
with Bonferroni correction.)
Tree Physiology Volume 00, 2011
Downloaded from treephys.oxfordjournals.org at ETH-Bibliothek on March 21, 2011
Figure 5. Inhibition of shoot height growth and stem diameter growth
by drought in relation to latitude of provenance origin (shoot height
growth: R = 0.699, P < 0.05; stem diameter growth: R = 0.657,
P < 0.05; Spearman’s rank correlation).
below a range of −0.3 to −1.0 MPa (Larson 1980, Tesky and
Hinckley 1981, Kuhns et al. 1985, Torreano and Morris 1998).
In our drought treatments, however, soil water potentials in
deeper soil layers were, for most of the growing season, clearly
above this critical range, making it unlikely that substantial
growth decline occurred in the deep-reaching oak roots.
The missing response of root length growth to drought contrasted sharply with severe growth reduction in shoots and
stems. Such differential response of below- and above-ground
growth to drought is common in plants, and the resulting
increase in root length to shoot height ratio has been widely
considered as an adaptive mechanism by which plants readjust
the balance between soil water absorption in roots and canopy
water loss (Kozlowski et al. 1991). The control of root to shoot
growth relations is a complex issue that is not fully understood
yet. To date, several hypotheses have been raised to explain
the underlying mechanisms and how environmental variables
affect this adaptive trait (cf. Farrar and Jones 2000). Knowledge
about organ-specific growth responses to environmental constraints is crucial to all these hypotheses. In oak saplings we
show that drought-induced changes in root length to shoot
height ratio are solely triggered by severe growth depressions
in shoots, thus supporting the view that oaks adapt to drought
by restricting canopy water loss rather than by facilitating soil
water absorption from deeper soil layers. Some circumstantial
evidence for this hypothesis comes from the reduction of transpiring leaf area in drought-subjected oaks that was associated
with reduced shoot development. This effect was not, however,
found to be significant in all provenances because of the high
Growth response of oaks to drought and warming 9
Table 4. Whole-tree leaf area (m2) of the oak provenances in control,
drought, air warming and combination treatments in 2009.
Species/
provenance
Drought
Air warming Combination
1.1 ± 0.6
1.1 ± 0.5
0.9 ± 0.4
0.8 ± 0.2
0.4 ± 0.2
0.9 ± 0.3
0.8 ± 0.3
0.6 ± 0.2
1.1 ± 0.4
1.2 ± 0.5
1.2 ± 0.4
1.1 ± 0.5
0.6 ± 0.2
0.9 ± 0.4
0.8 ± 0.2
0.6 ± 0.3
0.6 ± 0.3
0.7 ± 0.2
0.6 ± 0.3
0.5 ± 0.3
0.3 ± 0.1
0.6 ± 0.2
0.4 ± 0.1
0.4 ± 0.2
0.6 ± 0.1
0.9 ± 0.5
0.5 ± 0.2
0.4 ± 0.2
0.5 ± 0.2
0.8 ± 0.4
0.3 ± 0.2
0.4 ± 0.2
0.6 ± 0.3
0.2 ± 0.1
0.8 ± 0.4
0.5 ± 0.2
0.5 ± 0.2
0.1 ± 0.1
0.4 ± 0.2
0.4 ± 0.2
0.5 ± 0.2
0.2 ± 0.1
0.8 ± 0.3
0.6 ± 0.2
0.4 ± 0.1
0.1 ± 0.1
0.4 ± 0.1
0.3 ± 0.1
Lowercase letter (a) indicates significant treatment effects of drought
within provenances with P < 0.05 as calculated with provenance-specific ANOVAs according to a two-factorial test design (irrigation and
temperature). Effects of air warming were not statistically significant.
Values are means ± SD; n = 8.
variability of whole-tree leaf area, which was a result of strong
lateral shoot development in a few individuals. This is a ­common
phenomenon in young oaks, which possess weak apical control (Collet et al. 1997).
Air warming, in contrast to drought, stimulated shoot height
growth while stem diameter growth and root length growth
were slightly reduced. The resulting decrease in root length
to shoot height ratio and stem diameter to shoot height ratio
is a well-known response to moderate warming, but its eco-­
physiological significance remains obscure (Hawkins and
McDonald 1994, Ericsson et al. 1996, Saxe et al. 2001, Way
and Oren 2010). Nevertheless, it might be speculated that
such altered growth allometry has a negative impact on the
mechanical stability of juvenile trees, making them more vulnerable to storms and heavy snow loads (Moore et al. 2008),
which are important environmental constraints in colder environments (Peltola and Kellomäki 1993, Nykänen et al. 1997).
However, it remains unclear whether this is generalizable to
adult trees, which exhibit different biomass partitioning.
Provenance-specific responses to drought
and air warming
The sensitivity of shoot height growth and stem diameter
growth to drought and air warming varied considerably among
the studied provenances but much less when these growth
traits were compared at the species level. The latter finding
was unexpected as the studied oak species are commonly
thought to differ from each other by their temperature and soil
water demands, with Q. pubescens being much more tolerant
Conclusions
In conclusion, our study on oak saplings shows differential
growth responses to drought and air warming in shoots, stems
and roots, resulting in altered allometric growth relations.
Increased root length to shoot height ratios may contribute to
higher drought resistance in water-limited oaks by re-balancing
soil water uptake and canopy water loss. Decreased root length
Tree Physiology Online at http://www.treephys.oxfordjournals.org
Downloaded from treephys.oxfordjournals.org at ETH-Bibliothek on March 21, 2011
Q. robur
Tägerwilena
Bonfol
Hühnenberga
Magadinoa
Q. petraea
Corcellesa
Magden
Wädenswil
Gordevio
Q. pubescens
Leuk
Roverella
LeLanderona
Promontognoa
Control
to drought and requiring higher growth temperatures than
Q. robur and Q. petraea (Ellenberg 1996, Timbal and Aussenac
1996). However, this general belief is based on empirical
knowledge mainly derived from ecological field studies rather
than on experimental evidence, and differences among provenances have not been considered so far. In oak saplings, we
show that sensitivity of oak growth to drought and air warming
is not strictly determined at the species level, and furthermore
we demonstrate that provenances from Q. pubescens are not
necessarily more tolerant to drought than those from Q. robur
and Q. petraea. For example, Q. pubescens LeLanderon was
found to be among the most sensitive provenances to drought
when all studied provenances of the three oak species were
compared with each other.
The wide range in sensitivity to drought and warming that
was observed among the provenances is in good agreement
with exceptionally high degrees of intra-specific diversity in
oaks (Jensen 1993, Kleinschmit 1993, Kriebel 1993, Liepe
1993, Finkeldey 2001), but surprisingly the observed sensitivity pattern did not reflect local adaptation. In fact, growth sensitivity of the provenances could not be related to any local
climate variable of the provenance sites from which seeds
were collected. This might well be explained by the postglacial
history of European oaks, which immigrated into Switzerland
from distinct glacial refugia located in southern Europe. Regions
north of the Swiss Alps were predominantly re-colonized
from refugia in the Balkans while regions at the southern border of the Swiss Alps were re-colonized from refugia in Italy
(Mátyás and Sperisen 2001, Finkeldey and Mátyás 2003).
Given that oaks from these isolated glacial refugia diverged
genetically, it seems reasonable to assume that genetic factors
related to the immigration history of the studied provenances
have interfered with selective pressure at the provenance sites
and this might account for our failure to correlate the observed
provenance sensitivity pattern with the local climate conditions.
Some support for this assumption comes from the north–
south-directed variations in drought sensitivity of shoot and
stem growth, which were found when these growth traits were
related to the latitude of the provenance sites, i.e., shoot height
growth and stem diameter growth were more sensitive to
drought in provenances from northern latitudes than in provenances from southern latitudes, irrespective of any climate
factor.
10 Arend et al.
to shoot height and stem diameter to shoot height ratios reflect
mainly shoot-specific growth stimulation in response to air
warming, which might have implications for resistance to
mechanical stresses. Differences between provenances were
not related to local climatic factors, suggesting that selection
pressure at the provenance sites was either low or dominated
by genetic factors related to the postglacial immigration history
of the provenances. Taken together, the present study shows
provenance-specific growth responses of oak saplings to
drought and air warming that might be important for the establishment and regeneration of oaks in a dryer and warmer
­climate. However, it remains to be elucidated in future studies
whether adult trees show similar growth responses.
We thank Peter Bleuler for technical assistance and Kim Krause,
Stefan Eichenberger, Tobias Bregy and Michael Bühlmann for
their help with growth measurements and root harvests.
Funding
This research was supported by grants of the VELUX Stiftung
to M.A. and T.K.
References
Aas, G. 1998. Morphologische und ökologische Variation mitteleuropäischer Quercus-Arten: Ein Beitrag zum Verständnis der
Biodiversität. Libri Botanici: Band 19. IHW Verlag, Eching, 221 S.
Abrams, M.D. 1990. Adaptations and responses to drought in Quercus
species of North America. Tree Physiol. 7:227–238.
Brändli, U.B. 1996. Die häufigsten Waldbäume der Schweiz. Ergebnisse
aus dem Landesforstinventar 1983–85: Verbreitung, Standort und
Häufigkeit von 30 Baumarten. Ber. Eidgenöss. Forsch. WSL 342:​
107–120.
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.
Ann. For. Sci. 63:625–644.
Ciais, P., M. Reichstein, N. Viovy et al. 2005. Europe-wide reduction in
primary productivity caused by the heat and drought in 2003.
Nature 437:529–533.
Collet, A., F. Colin and F. Bernier. 1997. Height growth, shoot elongation and branch development of young Quercus petraea grown under
different levels of resource availability. Ann. Sci. For. 54:65–81.
Coomes, D.A. and R.B. Allen. 2007. Effects of size, competition and
altitude on tree growth. J. Ecol. 95:1084–1097.
Dickson, R.E. and P.T. Tomlinson. 1996. Oak growth, development and
carbon metabolism in response to water stress. Ann. Sci. For. 53:​
181–196.
Dobbertin, M. 2005. Tree growth as indicator of tree vitality and of
tree reaction to environmental stress: a review. Eur. J. For. Sci.
124:319–333.
Ellenberg, H. 1996. Vegetation Mitteleuropas mit den Alpen. 5th edn.
Ulmer Verlag, Stuttgart.
Ellenberg, H. and F. Klötzli. 1972. Waldgesellschaften und Waldstandorte
der Schweiz. Mitt. Eidgenöss. Forsch. WSL 48:589–930.
Tree Physiology Volume 00, 2011
Downloaded from treephys.oxfordjournals.org at ETH-Bibliothek on March 21, 2011
Acknowledgments
Ericsson, T., L. Rytter and E. Vapaavuori. 1996. Physiology of carbon
allocation in trees. Biomass Bioenergy 11:115–127.
Farrar, J.F. and D.L. Jones. 2000. The control of carbon acquisition by
roots. New Phytol. 147:43–53.
Finkeldey, R. 2001. Genetic variation of oaks (Quercus spp.) in
Switzerland. 2. Genetic structures in ‘pure’ and ‘mixed’ forests of
pedunculate oak (Q. robur L.) and sessile oak (Q. petraea (Matt.)
Liebl.). Silv. Genet. 50:22–30.
Finkeldey, R. and G. Mátyás. 2003. Genetic variation of oaks (Quercus
spp.) in Switzerland. 3. Lack of impact of postglacial recolonization
history on nuclear gene loci. Theor. Appl. Genet. 106:346–352.
Fort, C., M.L. Fauveau, F. Muller, P. Lanbel, A. Granier and E. Dreyer.
1997. Stomatal conductance, growth and root signaling in young
oak seedlings subjected to partial soil drying. Tree Physiol.
17:281–289.
Gessler, A., C. Keitel, J. Kreuzwieser, R. Matyssek, W. Seiler and
H. Rennenberg. 2007. Potential risks for European beech (Fagus
sylvatica L.) in a changing climate. Trees 21:1–11.
Gieger, T. and F.M. Thomas. 2005. Differential response of two CentralEuropean oak species to single and combined stress factors. Trees
19:607–618.
Hawkins, B.J. and S. McDonald. 1994. The influence of temperature
and soil water on growth, photosynthesis, and nitrogen fixation of
red alder (Alnus rubra) seedlings. Can. J. For. Res. 24:1029–1032.
Jensen, J.S. 1993. Variation of growth in Danish provenance trials with
oak (Quercus robur L and Quercus petraea Mattuschka Liebl). Ann.
Sci. For. 50:203–207.
Jolly, W.M., M. Dobbertin, N.E. Zimmermann and M. Reichstein 2005.
Divergent growth responses of Alpine forests to 2003 heat wave.
Geo­phys. Res. Lett. 32:Art. No. L18409, DOI: 10.1029/2005
GL023252.
Joslin, J.D., M.H. Wolfe and P.J. Hanson. 2000. Effects of altered water
regimes on forest root systems. New Phytol. 147:117–129.
Kleinschmit, J. 1993. Intraspecific variation of growth and adaptive
traits in European oak species. Ann. Sci. For. 50:166–185.
Kozlowski, T.T. 1982. Water supply and tree growth. For. Abstr.
43:57–95.
Kozlowski, T.T., P.J. Kramer and S.G. Pallardy. 1991. The physiological
ecology of woody plants. Academic Press Ltd, San Diego, CA, USA.
Kriebel, H.B. 1993. Intraspecific variation of growth and adaptive traits
in North American oak species. Ann. Sci. For. 50:153–165.
Kubiske, M.E. and M.D. Abrams. 1993. Stomatal and nonstomatal
­limitations of photosynthesis in 19 temperate tree species on
­contrasting sites during wet and dry years. Plant Cell Environ.
16:1123–1129.
Kuhns, M.R., H.E. Garrett, R.O. Tesky and T.M. Hinckley. 1985. Root
growth of black walnut trees related to soil temperature, soil water
potential, and leaf water potential. For. Sci. 31:617–629.
Larson, M.M. 1980. Effects of atmospheric humidity and zonal soil
water stress on initial growth of planted northern red oak seedlings.
Can. J. For. Res. 10:549–554.
Leuzinger, S., G. Zotz, R. Asshoff and C. Körner. 2005. Responses of
deciduous forest trees to severe drought in Central Europe. Tree
Physiol. 25:641–650.
Liepe, K. 1993. Growth-chamber trial on frost hardiness and field trial
on flushing of sessile oak (Quercus petraea Liebl). Ann. Sci. For.
50:208–214.
Mátyás, G. and C. Sperisen. 2001. Chloroplast DNA polymorphism
provide evidence for postglacial re-colonisation of oaks (Quercus
spp.) across Swiss Alps. Theor. Appl. Genet. 102:12–20.
Moore, J.R., J.D. Tombleson, J.A. Turner and M. Van der Colff. 2008.
Wind effects on juvenile trees: a review with special reference
to toppling of radiata pine growing in New Zealand. Forestry
81:377–387.
Growth response of oaks to drought and warming 11
tree-line sites on shallow calcareous soils: physiological reactions of
seedlings to severe drought. Flora 195:104–115.
Timbal, J. and G. Aussenac. 1996. An overview of ecology and silviculture of indigenous oaks in France. Ann. Sci. For. 53:649–661.
Torreano, S.J. and L.A. Morris. 1998. Loblolly pine root growth and
distribution under water stress. Soil Sci. Soc. Am. J. 62:818–827.
VanGenuchten, M.T. 1980. A closed-form equation for predicting
the hydraulic conductivity of unsaturated soils. Soil Sci. Soc. Am. J.
44:892–898.
VanGenuchten, M.T., F.C. Leij and S.R. Yates. 1991. The RETC code for
quantifying the hydraulic functions of unsaturated soils. R. S. Kerr
Environmental Research Laboratory, US Environmental Protection
Agency, Ada, OK.
Vivin, P., G. Aussenac and G. Levy. 1993. Differences in drought resistance among 3 deciduous oak species grown in large boxes. Ann.
Sci. For. 50:221–233.
Way, D.A. and R. Oren. 2010. Differential responses to changes in growth
temperature between trees from different functional groups and
biomes: a review and synthesis of data. Tree Physiol. 30:​669–688.
Zimmermann, N.E., J. Bolliger, J. Gehrig-Fasel, A. Guisan, F. Kienast, H.
Lischke, S. Rickebusch and T. Wohlgemuth. 2006. Wo wachsen die
Bäume in 100 Jahren? In Wald und Klimawandel. Forum für Wissen
2006. Ed. T. Wohlgemuth. WSL, Birmensdorf, Switzerland, pp 63–71.
ISSN 1021–2256.
Tree Physiology Online at http://www.treephys.oxfordjournals.org
Downloaded from treephys.oxfordjournals.org at ETH-Bibliothek on March 21, 2011
Nixon, K.C. 2006. Global and neotropical distribution and diversity of
oak (genus Quercus) and oak forests. Ecol. Stud. 185:1–13.
Nykänen, M.L., H. Peltola, C.P. Quine, S. Kellomäki and M. Broadgate.
1997. Factors affecting snow damage of trees with particular reference to European conditions. Silva Fenn. 31:193–213.
Ohlemüller, R., E.S. Gritti, M.T. Sykes and C.D. Thomas. 2006.
Quantifying components of risk for European woody species under
climate change. Global Change Biol. 12:1788–1799.
Peltola, H. and S. Kellomäki. 1993. A mechanistic model for calculating
wind throw and stem breakage of Scots pines at stand age. Silva
Fenn. 27:99–111.
Ponton, S., J.L. Dupouey, N. Bréda and E. Dreyer. 2002. Comparison of
water-use efficiency of two sympatric oak species: genotype × ­environment interactions. Tree Physiol. 22:413–422.
Rushton, B.S. 1993. Natural hybridization within the genus Quercus L.
Ann. Sci. For. 50:73–90.
Saxe, H., M.G.R. Cannell, O. Johnsen, M.G. Ryan and G. Vourlitis. 2001.
Tree and forest functioning in response to global warming. New
Phytol. 149:369–400.
Tesky, R.O. and T.M. Hinckley. 1981. Influence of temperature and water
potential on root growth of white oak. Physiol. Plant. 52:363–369.
Thomas, F.M. 2000. Growth and water relations of four deciduous
tree species (Fagus sylvatica L., Quercus petraea [Matt.] Liebl.,
Q. ­pubescens Willd., Sorbus aria [L.] Cr.) occurring at Central-European