Content uploaded by Carolina Mayoral
Author content
All content in this area was uploaded by Carolina Mayoral on Apr 23, 2015
Content may be subject to copyright.
Modelling the influence of light, water and temperature
on photosynthesis in young trees of mixed Mediterranean
forests
Carolina Mayoral •Rafael Calama •Mariola Sa
´nchez-Gonza
´lez •
Marta Pardos
Received: 9 June 2014 / Accepted: 5 February 2015
ÓSpringer Science+Business Media Dordrecht 2015
Abstract The composition of Mediterranean forests is expected to vary with ongoing
changes in climate and land use. To gain a clearer understanding of the response to global
change of growth and survival during regeneration it is necessary to take a closer look at
the ecophysiological traits underlying seedling performance. Gas exchange, leaf water
potential, chlorophyll fluorescence, soil moisture, temperature and global site factor were
measured over 1 year in naturally regenerated young trees of three coexisting species
(Pinus pinea,Quercus ilex and Juniperus oxycedrus) in two stands of different density. We
modelled the photosynthetic response of plants to micro-climatic conditions via the
parameterization of the non-rectangular hyperbolic model of photosynthesis, which relates
gross photosynthesis to incident light through three biochemical parameters, and the
subsequent expansion of these parameters as a function of environmental variables (light
environment, soil moisture and temperature). We investigated the relationship between
different photosynthetic performance and the species-specific strategies to cope with stress
(stress tolerant or avoiders). The optimal light environment, defined through the global site
factor (GSF), and the regeneration niche to maximize carbon assimilation differed between
the three species. P. pinea showed high sensitivity to water availability in agreement with a
drought avoidance strategy, attaining the maximum photosynthetic capacity of the three
species following the spring rainfall. Q. ilex was the most thermophilic and light-
demanding of the species. Under high light conditions, J. oxycedrus was more drought
tolerant and displayed higher net CO
2
assimilation than P. pinea over the course of a
growing period. Optimal locations for P. pinea regeneration are below-crown environ-
ments, while for J. oxycedrus regeneration the optimal locations are open gaps. Q. ilex
regeneration occupy open gaps where the other two species are unable to establish
themselves because of excessive light, temperature or very low water availability. Com-
petition between species will occur under a canopy gap fraction of 0.5. Higher GSF values
will exclusively favour the regeneration of Q. ilex.
C. Mayoral (&)R. Calama M. Sa
´nchez-Gonza
´lez M. Pardos
Department of Sylviculture and Forest System Management, INIA-CIFOR Ctra, A Corun
˜a km 7.5,
28040 Madrid, Spain
e-mail: mayoral.lopez.c@gmail.com
123
New Forests
DOI 10.1007/s11056-015-9471-y
Keywords Regeneration niches Net assimilation rate Pine-holm oak mixed stands
Juniper Light environment Non-rectangular hyperbolic model
Introduction
Mediterranean mixed forests have been subjected to important changes in recent years not
only in terms of climate but also land-use (Valladares et al. 2004). These changes have an
impact on the complex process of forest regeneration. All physiological, morphological
and architectural traits exhibited by each species to cope with the environmental hetero-
geneity determine its regeneration niche. The geographical dimensions of the regeneration
niches are constantly changing and are closely linked to climate. An example is the
altitudinal upward shift of forest plant species in order to reach suitable conditions (Ruiz-
Labourdette et al. 2012).
The environmental factors that most influence seedling growth and survival in Medi-
terranean ecosystems are water and light availability together with extreme temperature
(e.g., Pardos et al. 2005). In this regard, the more prolonged and intense drought periods in
the summer, combined with high irradiance and high temperature expected in the future
(IPCC 2007), may induce serious restrictions on the establishment of plants (Breda et al.
2006). In addition to summer stress, cold winters combined with high irradiance are
expected to be equally limiting factors, causing photo inhibition (Martinez-Ferri et al.
2004; Fellows and Goulden 2013; Pardos et al. 2014).
In mixed forest, different strategies (tolerance or avoidance) are adopted by each species
to cope with stress conditions. For instance, under the same degree of drought, some
species depict an anisohydric strategy, their xylems being resistant to cavitation, thus
allowing the stomatal regulation of gas exchange to extend much further into the drought.
These species also tend to have lower maximum conductivity under favourable conditions
(West et al. 2008). Other species avoid low negative water pressure in the xylem and
consequently cavitation adopting isohydric stomatal control, thus closing stomata and
suppressing CO
2
assimilation (Baquedano and Castillo 2007). In addition, coexisting
species with similar life form and habitat range, may also have different reproductive
strategies (Grubb 1977). This is the case of seeder species versus resprouting species,
which are capable of recovering from adverse environmental conditions through
resprouting, thus occupying a so-called ‘persistence niche’ (Bond and Midgley 2001).
Photosynthetic models can describe the effect of environment on the physiological
performance of plants during regeneration and thereby allow us to determine the physio-
logical regeneration niches of species under current and changing climate scenarios
(Calama et al. 2013). Photosynthesis can be described by numerous models, varying from
pure empirical models which do not depend on underlying physiological processes and
constrained expandability (e.g. Stegemann et al. 1999) to very complex mechanistic
models where photosynthesis is defined through a network of submodels describing bio-
chemical processes (e.g. Kirschbaum et al. 1998). Semiempirical models provide an
intermediate option, defining photosynthesis according to a mathematical description of a
single biochemical process. The non-rectangular hyperbola (NRH) model of photosynth-
esis (Marshall and Biscoe 1980; Thornley and Johnson 1990) or the more complex and
popular Rubisco kinetics based model (Farquhar et al. 1980) belong to this latter group.
These models exhibit a sound physiological basis, can be easily fitted to gas-exchange
New Forests
123
measurements and show desirable mathematical properties such as inflection and max-
imum–minimum points. While the models based on Farquhar et al. (1980) equations allow
for a better understanding of the mechanisms related to photosynthesis, they require
empirical calculation of species-specific photo- and bio-chemical parameters which are
subjected to environmental control. However, the NRH model, describing photosynthesis
by a single equation can be more useful for comparing photosynthetic light response curves
under growing conditions with numerous variables (Go
´mez-Aparicio et al. 2006; Aspin-
wall et al. 2011; Mengistu et al. 2011; Xu et al. 2014).
Although the response of photosynthesis to environmental conditions has been widely
studied in different species (mainly under controlled conditions), more attention need to be
paid to the combined effect of the main abiotic stress factors that determine the photo-
synthetic rate under natural field conditions (Markos and Kyparissis 2011; Calama et al.
2013), namely, light availability, soil moisture and temperature, which in turn define the
regeneration niche of each species (Maran
˜o
´n et al. 2004;Go
´mez-Aparicio et al. 2006).
In the present work we propose to analyse the photosynthetic performance of three
species coexisting in Mediterranean mixed forests that differ in their strategies for coping
with drought and in their preferred light environment. These species are Pinus pinea L.
(stone pine), Quercus ilex L. (holm oak) and Juniperus oxycedrus L. (prickly juniper).
While Q. ilex and J. oxycedrus are described as drought-tolerant species (Baquedano and
Castillo 2007; Willson et al. 2008), P. pinea is described as a drought-avoidant species
(Pardos et al. 2009). Furthermore, P. pinea and J. oxycedrus have been described as shade
demanding in the first stages of development (Manso et al. 2014), while Q. ilex is able to
perform adequately in sun exposed areas, although drought can decrease seedling survival
in these light environments (Perez-Ramos et al. 2013). Finally, the resprouting character of
Q. ilex which provides with deep-root system for the regeneration, could confer an
advantage with respect to the other two species.
Our aims were (1) to parameterize the NRH model for each of the main species coexisting
in typical Mediterranean forests in order to describe and predict the effect of different
environmental factors on the photosynthetic capacity of young trees; and (2) based on the
model fits, to define the optimal regeneration niche in mixed stands in central Spain for the
maximization of carbon assimilation in each species. The hypotheses to be tested were (1) the
differences in the spatiotemporal pattern of photosynthetic response among species were
linked to their specific seasonal responses to environmental stresses (Q. ilex and J. oxycedrus,
drought-tolerant strategies; P. pinea drought-avoidant strategy); (2) observed specific pat-
terns of photosynthetic response were in accordance with other observed physiological traits
and the previously described natural regeneration dynamics; (3) among the three species, Q.
ilex would be the most competitive under the expected climate change scenario because of its
resprouting capacity and its drought tolerance; (4) the photosynthetic performance of
plants during an optimal growing period (i.e. less severe climatic conditions and optimal light
growth environment) will be better for a stress avoidant species such as P. pinea than for -
species better adapted to cope with adverse climatic conditions.
Materials and methods
Study area
Our study was set up in two mixed-forest stands co-dominated by three typical Medi-
terranean species: P. pinea,Q. ilex and J. oxycedrus. These forests have traditionally been
New Forests
123
Table 1 Dasometric features for the tree species in the two plots used
P. pinea Q. ilex J. oxycedrus Total
Plot 1 2 1 2 1 2 1 2
D (N/ha) 118 173 64 24 64 47 9.05 20.3
BA (m
2
/ha) 8.85 19.8 0.08 0.15 0.12 0.32 20.3 47.2
CC (%) 19.1 46.1 1.76 2.15 1.07 2.47 247 244
DBH (cm) 25.3 ±2.4 37.2 ±0.9 2.52 ±0.6 8.41 ±0.9 2.39 ±0.8 7.94 ±1.0 13.5 ±1.6 28.8 ±1.4
DH (m) 7.31 11.7 2.62 4.63 2.18 3.89 5.1 9.47
Dstand density, Nnumber of stems, BA basal area, CC Canopy cover, DBH mean diameter at breast height ±SE, DH dominant height
New Forests
123
managed for the production of pine nuts, hence, the silvicultural intervention has been
oriented towards achieving regular stands, exclusively dominated by over mature, pro-
ductive P. pinea trees. However, in recent decades the forest uses have changed, and the
decrease in silvicultural interventions, increased grazing and recurrent forest fires have led
to a greater presence of Q. ilex and J. oxycedrus (Jovellar et al. 2013) to the point where
they are considered codominant species. Q. ilex is often present as a shrub, occupying large
areas in the form of a chaparral-type understory rather than as a tree. The two 0.45 ha
(90 950 m
2
) plots were set up in the Tie
´tar and Alberche Valleys, central Spain: Hoyo de
Pinares (40°2907.600N, 4°19034.6600W) and San Martin de Valdeiglesias (40°20022.3400N,
4°2203.8300W), 800 m above sea level, on sandy soils and gently sloping terrain (around
12 %). The plots were selected because between the two they covered all the possible
range of microenvironmental conditions for the plants; more intensive silviculture had
been applied in stand 1, while stand 2 is denser and has more intense grazing by goats
(Table 1). The climate in the area is characterized by hot summers and pronounced,
extended drought periods, while winters are relatively cold and wet (Fig. 1). The shrubby
understory was composed of species such as Cistus ladanifer,Rosmarinus officinalis and
Helichrysum stoechas.
Physiological and microclimatic measurements
We measured the net photosynthetic rate (A
n
,lmol CO
2
m
-2
s
-1
), temperature of the leaf
chamber (T, °C), transpiration rate (E, mmol H
2
Om
-2
s
-1
) and stomatal conductance (g
s
,
mmol m
-2
s
-1
) using a portable photosynthesis system LcPro (ADC BioScientific Ltd.,
UK). The LcPro also automatically recorded the photosynthetically active radiation inci-
dent on the leaf at each recording. Gas exchange measurements were carried out under
natural field conditions i.e. conditions matched those outside the chamber. During the
whole period, the average value (and 5–95 percentiles) observed was: incident light
(797 lmol m
-2
s
-1
, 59–2199), atmospheric CO
2
(389 ppm, 371–425), temperature
(26.3 °C, 12.6–37.9, vapour pressure deficit (2.6 kPa, 0.7–5.3) and soil volumetric water
content (9.9 %, 1.8–24.0). Gas exchange measurements were recorded during the
experiment on the same 96 plants. We chose 16 plants per species (P. pinea,Q. ilex and J.
oxycedrus) in each plot, taken from two height classes (8 plants per class and species):
from 20 to 50 cm height (regeneration) and from 50 to 130 cm (advanced regeneration).
Fig. 1 Meteorological data during the study period in plot 1. Tmax monthly maximum temperature, Tmin
monthly minimum temperature and Tmean monthly mean temperature, Precipitation monthly precipitation
New Forests
123
Class limits were set to clearly define a representative sample of the whole range of
heights, and the limit of 130 cm was established on the basis that at this height the species
have reached a higher stage of development (P. pinea plants over 1.30 m have almost
completely lost their juvenile needles and therefore are no longer considered saplings).
Within each height class, plants were selected such that all the light environments found
within the plots would be covered. Measurements were taken on 10 dates between Sep-
tember 2010 and October 2011 (plot 1: Sep 15-10, Oct 19-10, Jan 18-11, Apr 27-11, May
24-11, Jun 22-11, Jul 27-11, Sep 6-11 and Oct 3-11; plot 2: Nov 4-10, Jan 25-11, Mar
18-11, Apr 26-11, Jun 21-11, Aug 1-11, Sep 5-11 and Oct 11-11). On each date, leaf gas-
exchange measurements were taken twice, morning (from 9:30 to 12:30 pm) and midday
(from 13:00 to 16:00), on the same sun-exposed branchlets from the current year. We only
selected branchlets with juvenile needles in the case of P. pinea. Gas exchange parameters
were calculated on a projected leaf area basis. Leaf area was calculated on the measured
branchlets harvested at the end of the measuring period. After scanning the leaves with a
high-resolution scanner (EPSON expression 10000 XL), the projected leaf area of each
branchlet was estimated using the image analysis software Winrhizo (needles) and Win-
folia (leaves) (Regent Instruments Inc. Canada). We monitored the water status of all
plants, measuring predawn and midday leaf water potentials (W, MPa) with a pressure
chamber (Scholander PMS 1000, Instruments Co, Corvallis, USA).
The photochemical efficiency of photosystem II (F
v
/F
m
)(U
PSII
) was measured with a
FMS2 portable fluorometer (Hansatech, UK). F
v
/F
m
at predawn and midday in leaves after
a 30-min dark-adaptation period. U
PSII
was measured in the morning and at midday. We
also measured other environmental factors which affect the photosynthetic capacity of
plants: soil moisture, light environment and temperature. Soil moisture (VWC, % volu-
metric moisture) was measured in the vicinity of every plant on each measuring date using
a portable time domain reflectometer (TDR) equipped with three 18 cm rod probes
(TRIME FM 3, IMKO, Germany). We estimated the light environment of each plant with
hemispherical photographs taken at the location of the selected seedlings. The photographs
were taken once during the experimental period as there were no major canopy dis-
turbances. From these images, the global site factor (GSF), proportion of global solar
radiation at a given location relative to that in the open, was computed using Hemiview 2.1
software (Delta-T Devices, Ltd., Cambridge, UK). The photographs were taken above each
plant, either at predawn or on cloudy days in order to avoid direct solar radiation and to
correct the contrast between the canopy and the sky, using a fish-eye lens (FC-E8, Nikon).
Apart from the measurements taken at the same time as the gas-exchange records, con-
tinuous recording of air temperature (20 cm above ground) and soil moisture (20 cm
depth) was performed every 30-min using a HOBO micro-weather station (Micro-HWS,
ONSET, Massachusetts, USA) installed at the center of each plot.
Modelling net photosynthesis rate
To model the relationship between the net photosynthesis (A
n
,lmol CO
2
m
-2
s
-1
) and
incident light (Q,lmol m
-2
s
-1
) we used the non-rectangular hyperbolic (NRH) model of
photosynthesis (Thornley and Johnson 1990). The NRH model relates leaf gross photo-
synthetic rate (A) to the incident photosynthetically active radiation on the leaf surface (Q)
through a light response curve defined by three parameters that have an ecophysiological
basis and make mathematical sense: A
max
, the maximum rate of gross photosynthesis at
saturating irradiance, upper asymptote of the light response curve; a, the quantum yield of
assimilation which is related to the photochemical efficiency, initial slope of the light
New Forests
123
response curve; and h, dimensionless parameter linked to the carbon transport resistance,
denotes the dimensionless convexity of the light response curve. In order to model the
field-measured data of A
n
, we added the rate of leaf dark respiration (R
d
,lmol CO
2
m
-2
s
-1
assimilation rate at irradiance zero) to the NRH model of photosynthesis, pro-
ducing the following expression:
Fig. 2 Three examples of the light response curves of A
n
(lines), after fitting the NRH model to our data
(black dots). EF Modelling efficiency. aJ. oxycedrus, plot 2, 2011, Mar 18th, midday measurement, wet late
winter conditions (average T: 23.5 °C, average VWC: 22.5 %) parameter estimates A
max
=3.80 lmol
CO
2
m
-2
s
-1
and a=0.017. bP. pinea, plot 2, 2011, Sep 5th, morning measurement, dry summer
conditions (average T: 22.4 °C, average VWC: 5.3 %), parameter estimates A
max
=7.30 lmol CO
2
m
-2-
s
-1
and a=0.006. cQ. ilex, plot 2, 2011, Jun 21st, midday measurement, early summer conditions (average
T: 37.3 °C, average VWC: 5.6 %) parameter estimates A
max
=11.7 lmol CO
2
m
-2
s
-1
and a=0.009. It
should be noted that 2011 was a very dry year
New Forests
123
An¼Rdþ1
2haQþAmax ðaQþAmaxÞ24haQAmax
hi
1=2
ð1Þ
Individual fit of the NRH model and estimation of parameters A
max
and a
In a first step, the model (1) was fitted separately to the data from each species, plot, mea-
surement date and moment of measurement (morning/midday), resulting in 120 independent
light response curves (2 plots 93 species 910 dates 92 moments) (Fig. 2). To carry out
this preliminary fit, we assumed a constant value of 0.8 for h(Anten et al. 1995; Hirose et al.
1997), and R
d
as a quadratic function of the temperature (R
d
=-0.00148T
2
) (Calama et al.
2013). Thus, for each fitted curve, a combination of parameters A
max
and awas obtained,
resulting in 107 combinations (convergence was not attained in 13 fittings).
Expansion of the parameters of the NRH model
Results from the preliminary fit were used to explore the relationships between the
parameters in the NRH model and the environmental variables. We explored the rela-
tionship between A
max
and awith the environmental variables VWC and Tby plotting
each parameter estimated in the previous step against the environmental variables
Fig. 3 Estimates of parameters A
max
and aagainst environmental variables VWC and T
ch
, after initial
fitting of NRH model to each separate inventory 9moment 9plot 9species dataset
New Forests
123
observed at that measurement time (Fig. 3). As long as there are no limitations to light or
CO
2
availability and conditions of stress do not cause stomatal closure or photoinhibi-
tion, A
n
increases with temperature, although only up to a critical value which depends
on the species. Similar behaviour is found with regard to photochemical efficiency a
where a tendency to increase up to an optimum temperature is observed (Fig. 3). On the
basis of these assumptions, we consider that the statistical dependence of A
max
and aon
Tcan be described by a quadratic parabolic curve (Battaglia et al. 1996; Markos and
Kyparissis 2011; Calama et al. 2013) where an optimal temperature value which max-
imizes both parameters is shown. We propose the expansion of A
max
and a(Eqs. 2and
3) through four new estimable parameters (p
0
,p
1
,p
2
and p
3
):
AmaxðTÞ¼p0þp1TTref
2ð2Þ
a¼p2þp3TTref
2ð3Þ
where T
ref
is an estimable parameter related to the temperature which maximizes A
max
, and
Tis the chamber temperature at the moment of measurement.
Soil moisture (VWC) is directly related to the water availability for plants (Denmead
and Shaw 1962). In theory, if low temperature is not a limiting factor, low values of VWC
will be related to low values of A
max
, while as a general rule, the highest A
max
values will
be shown at high VWC. We have introduced a modifier function of A
max
, which is related
exponentially to VWC through a new parameter (p
4
):
AmaxðT;VWCÞ¼½p0þp1TTref
2ep4VWC ð4Þ
Light environment is expected to have a greater effect on A
max
at saturating irradiance than
on a(Go
´mez-Aparicio et al. 2006; Calama et al. 2013). We have only expanded the
parameter A
max
, using a new modifier function with a new parameter (p
5
), where the global
site factor (GSF), whose value varies between 0 and 1, is included as a variable related to
the growth irradiance environment reaching each plant (Eq. 5). This function works as a
modulator of A
max
. An optimal value of GSF which maximizes A
max
can be found for each
species.
Amax T;VWC;GSFðÞ¼p0þp1TTref
2
hi
ep4VWC:p5GSFep5GSF ð5Þ
For the R
d
expansion, we assume that its response to temperature is defined by an expo-
nential function that includes a temperature coefficient (Q
10
), which represents the directly
proportional variation in R
d
when temperature changes 10 °C (Tjoelker et al. 2001).
Although Q
10
values are variable throughout the year, typical values of Q
10
are around 2
(Rodrı
´guez-Calcerrada et al. 2012; Lin et al. 2012). We have set a value of 2.4 for our
modelling approach in J. oxycedrus and P. pinea (Calama et al. 2013), whereas in Q. ilex it
has been set to 1.7 according to the estimated Q
10
values obtained by Zaragoza-Castells
et al. (2008) for this species.
Rdch ¼p6Q
T25
10
ðÞ
10 ð6Þ
where Rdch, at leaf chamber temperature T, is a function of p
6
, which represents R
d
at
25 °C and it is also an estimable parameter.
New Forests
123
Final models and simulations
A model capable of predicting A
n
for the regeneration was formulated for each species.
For this purpose, A
max
,aand R
d
(Eq. 1) were replaced by the expanded functions of
these parameters (Eqs. 3,5and 6) and eight new parameters were estimated or
heuristically determined for each species (p
0
–p
6
and T
ref
) by fitting the model to the
data of A
n
,T,Q, GSF and VWC recorded for each plant and moment of measurement.
As in the previous step, we assumed a constant value of 0.8 for h. Non-linear models
for each species were fitted using Ordinary Least Squares technique, by applying the
NLIN procedure of SAS
Ò
/STAT 9.2. In addition, we used the Gauss–Newton opti-
mization algorithm to fit the final models. We used a heuristic procedure to estimate
those parameters which restrained the convergence of the final model or make no
biological sense. This heuristic procedure involved using an identical model structure
for the three species, minimizing the sum of squares error relative to the full set of
climatic variables. For each model, we calculated the modelling efficiency EF (pseudo-
R
2
for nonlinear models), mean error and associated level of significance and root mean
square error to assess the goodness-of-fit.
Once the new parameters of the complete model have been estimated, we can simulate
different environment scenarios in order to predict A
n
. We used the final models to
simulate the interaction effect of the environmental variables (VWC and T) on the A
n
for
each species. We carried out the simulations, setting Qto 2,000 lmol m
-2
s
-1
, along
VWC and Tgradients. From the simulations, we can identify which species is the most
physiologically active in spring, and conversely, determine the most stressed species under
limiting conditions (e.g. summer season).
Additionally, we evaluated the performance of the models at eighteen real positions in
plot 1, covering the whole GSF range found (from 0.22 to 0.93). We used 30-min data
recorded for VWC,Tand Q. Combined light-temperature sensors and HOBO Pendant
Ò
loggers were installed at each position. Continuous soil volumetric water content data were
obtained from the meteorological station located at the centre of the plot. For the
simulation, we considered the same VWC for all the positions. To support this decision, we
made a preliminary analysis evaluating the effect of light environment on water avail-
ability at 18 cm, which revealed no significant differences in soil water content with the
light environment (results not shown). We simulated the photosynthetic activity in dummy
seedlings of each species sited at each position over a growing period (from 1st March to
31st October, 2012). The differences in the GSF between the eighteen GSF environments
involved differences in the daily pattern of Qand Tover the studied vegetative period. The
analysed positions (GSF environments) were compared between species in terms of
monthly net carbon assimilation rates.
Additional physiological attributes
Leaf water potential, fluorescence of chlorophyll and soil moisture were analyzed using
a repeated measure two-way analysis of variance (ANOVA) to test the temporal effect
of the species for these variables. We included the covariate incident light (Q) in the
chlorophyll fluorescence analysis. We used the Tukey–Kramer post hoc test to compare
species means. All the results were considered to be significant if PB0.05. We used
New Forests
123
PROC MIXED (SAS
Ò
/STAT 9.2) for the analysis and tested several covariance
structures.
Results
Model fitting
We fitted the final model (Eq. 7) for each species. Table 2shows the results of the pa-
rameter estimates per species. Convergence was reached after 28, 41, and 17 iterations for
J. oxycedrus,P. pinea and Q. ilex, respectively.
Table 2 Parameter estimates for the photosynthesis model
Initial parameter Parameter Estimate SE LCL95 UPL95
Juniperus oxycedrus
A
max
p
0
10.814 1.0928 8.6667 12.9624
p
1
-0.0268 0.00547 -0.0376 -0.0161
p
4
0.0192 0.00526 0.00889 0.0296
p
5
2.0961 0.2782 1.5492 2.6430
T
ref
24.0000 Heuristically determined
R
d
p
6
-0.2300 0.1102 -0.4466 -0.0135
ap
2
0.0116 0.00171 0.0083 0.0150
p
3
-0.00003 1.40E-06 -6.00E-05 -5.20E-06
Pinus pinea
A
max
p
0
12.2008 1.6834 8.8922 15.5094
p
1
-0.0215 0.00667 -0.0346 -0.00835
p
4
0.0459 0.0070 0.0322 0.0597
p
5
3.5161 0.3179 2.8914 4.1408
T
ref
26.407 2.8068 20.8905 31.9236
R
d
p
6
-0.3579 0.2371 -0.8239 0.1081
ap
2
0.0119 0.00233 0.0073 0.0165
p
3
-0.00002 6.15E-06 -4.00E-05 -1.00E-05
Quercus ilex
A
max
p
0
14.4226 1.5167 14.0872 20.0503
p
1
-0.0135 0.00668 -0.0329 -0.00668
p
4
0.0015 Heuristically determined
p
5
1.3242 0.3207 0.6938 1.9546
T
ref
35.5276 4.3593 26.9579 44.0973
R
d
p
6
-0.0170 Heuristically determined
ap
2
0.0116 0.00168 0.00831 0.0149
p
3
-0.00001 3.735E-06 -0.00002 -4.58E-06
A
max
, maximum CO
2
assimilation rate (lmol CO
2
m
-2
s
-1
); R
d
, leaf dark respiration (lmol CO
2
m
-2
s
-1
);
a, photochemical efficiency (mol CO
2
mol
-1
light), SE standard errors; LCL95 and UPL95, lower and
upper limits of the confidence interval for the parameter estimates at a 95 % significance level
New Forests
123
An
¼1
2h
p2þp3TTref
ðÞ
2
Qþp0þp1TTref
ðÞ
2
ep4H
p5GSFep5GSF
hi
p2þp3TTref
ðÞ
2
Qþp0þp1TTref
ðÞ
2
ep4H
p5GSFep5GSF
hihi
2
4hp2þp3TTref
ðÞ
2
Qp
0þp1TTref
2
ep4H
p5GSFep5GSF
hi
2
6
6
43
7
7
5
1
2
8
>
>
>
>
>
>
>
<
>
>
>
>
>
>
>
:
9
>
>
>
>
>
>
>
=
>
>
>
>
>
>
>
;
p6Q
T25
10
ðÞ
10
ð7Þ
For the heuristic determination of T
ref
in J. oxycedrus, we assigned values every one degree
from 15 to 45 °C. The sum of squares error was minimized at 24 °C. In the case of Q. ilex,
convergence problems for parameters p
4
and p
6
were found. In this case, we heuristically
identified a combination of parameters p
4
(related to soil moisture) and p
6
(related to leaf dark
respiration) which minimized the sum of squares error. However, the estimated values of p
4
and
p
6
involved a very low A
max
dependence on soil moisture in Q. ilex compared to the other two
species as wellas minimum leaf dark respiration for this species. In P. pinea, convergence o f the
complete model was reached without the need for additional model tuning.
Goodness-of-fit statistics (Table 3) indicated highly significant unbiased predictions of
the models for the three species, with percentages of total explained variability ranging
between 34 and 60 %, and RMSE below 2.3 lmol CO
2
m
-2
s
-1
. Our results showed that
instantaneous A
max
is maximized at the specific reference temperature (T
ref
) estimated for
each species: 35.5 °C for Q. ilex, 26.4 °C for P. pinea and 24 °C for J. oxycedrus.
Similarly, values of GSF maximizing A
max
for each species (given by 1/p
5
) are 0.75 for Q.
ilex, 0.477 for J. oxycedrus and 0.284 for P. pinea.
Further analysis of the stomatal conductance (g
s
, mmol m
-2
s
-1
) and transpiration (E,
mmol H
2
Om
-2
s
-1
) have also shown significant differences between species (P\0.0001),
with higher values (mean ±SE) of both variables in Q. ilex during the summer months
(g
s
=0.031 ±0.002; E=0.914 ±0.061) compared to P. pinea (g
s
=0.014 ±0.001;
E=0.380 ±0.039) and J. oxycedrus (g
s
=0.015 ±0.001; E=0.408 ±0.039). These
results suggest greater water availability for Q. ilex at deeper soil layers.
Leaf water potential, chlorophyll a fluorescence, and soil moisture variations
The factor effect ‘tree species’ showed between-subject differences in all response vari-
ables except ing VWC (Fig. 4). The lowest predawn-F
v
/F
m
was measured in January (the
Table 3 Goodness of fit statistic
Species Mean error pvalue A
¯obs A
¯pred EF (%) RMSE
Juniperus oxycedrus 0.0489 0.5957 2.0280 1.9791 34 1.9372
Pinus pinea 0.0680 0.5257 2.3668 2.2988 60 2.2723
Quercus ilex 0.0768 0.4786 2.6277 2.5508 34 2.1936
Mean error (lmol CO
2
m
-2
s
-1
); pvalue: level of significance for the mean error; A
¯obs, observed average
value for A
n
(lmol CO
2
m
-2
s
-1
); A
¯pred, predicted average value for A
n
(lmol CO
2
m
-2
s
-1
); EF,
modeling efficiency; RMSE, root mean square error (lmol CO
2
m
-2
s
-1
)
New Forests
123
coldest month), but there were no differences among species on this date. Predawn-F
v
/F
m
was significantly lower in J. oxycedrus compared to P. pinea and Q. ilex on three summer
measurement dates: Sep-2010 [P
adj
(P. pinea)=0.0439 and P
adj
(Q. ilex)=0.0282];
midsummer-2011 [P
adj
(P. pinea)=0.0018 and P
adj
(Q. ilex)\0.0001]; and Sep-2011
[P
adj
(P. pinea)=0.0031 and P
adj
(Q. ilex)\0.0001]. An interesting response of J.
oxycedrus and P. pinea was identified, both showing the lowest midday-F
v
/F
m
under
summer conditions, while Q. ilex reached the minimum value in winter (Fig. 4b). The
midday-F
v
/F
m
was significantly lower in J. oxycedrus compared to the other species in the
summer measurements: Sep-2010 [P
adj
(Q. ilex)=0.0282], mid-summer 2011 (all species
differed from each other P
adj
\0.0001) and Sep-2011 [P
adj
(P. pinea)=0.0099 and P
adj
(Q. ilex)\0.0001] (Fig. 4b).
Fig. 4 Seasonal variations (mean ±standard error) in: maximum quantum efficiency of photosystem II
(Fv/Fm) at predawn (a) and midday (b), photosystem II quantum efficiency (UPSII) in the morning (c) and
midday (d), soil moisture content (% v/v) at 18 cm depth (e) and leaf water potential W(MPa 910) at
predawn (f) and midday (g). Note that predawn Win January was not measured
New Forests
123
Morning-U
PSII
, we found overall differences between J. oxycedrus and Q. ilex
P
adj
\0.0188 (Fig. 4c). Midday-U
PSII
in Q. ilex differed significantly from the other
species in Sep-2010 P
adj
(P. pinea)=0.0375 and midsummer-2011 [P
adj
(P.
pinea)=0.0004 and P
adj
(J. oxycedrus)\0.0001] (Fig. 4d). P. pinea and J. oxycedrus
showed differences on several measuring dates in the midday-U
PSII
, highlighting seasonal
Fig. 5 Contour plots representing the interacting effect of environmental variables (VWC and T) on A
n
(lmol CO
2
m
-2
s
-1
) reached by each species (A
n
Qi; Quercus ilex;A
n
Pp: Pinus pinea;A
n
Jo: Juniperus
oxycedrus) under high irradiance (2000 lmol m
-2
s
-1
)
New Forests
123
variations in this physiological variable, while Q. ilex did not show differences. In Sep-
2010, the decrease in U
PSII
from the morning to midday revealed different behaviour
between species: Q. ilex was 28.2 % lower U
PSII
,J. oxycedrus 43.7 % and P. pinea showed
the greatest decrease (66.7 % lower). This trend was maintained by the three species under
summer conditions.
Pinus pinea exhibited the highest predawn and midday Wover the measuring period
(Fig. 4f, g). J. oxycedrus showed the lowest midday Wof the three species on Sep-2010 (-
3.7 ±0.25 MPa), followed by Q. ilex on the same measuring date (-3.1 ±0.16 MPa).
The lowest midday Wof P. pinea was shown at the end of Jun-2011 (-1.9 ±0.05 MPa).
We found significant differences between J. oxycedrus and P. pinea in the predawn Won
six measuring dates (all P
adj
\0.0001), coinciding with the summer and autumn mea-
surements (Fig. 4f). Q. ilex showed similar behaviour to that of J. oxycedrus on both the
late summer measurement dates, differing significantly from P. pinea in both Sep-2010 and
Sep-2011 (P
adj
\0.0001). However, Q. ilex displayed significantly higher Wthan J.
oxycedrus in autumn 2010 (P
adj
=0.0460) and midsummer (P
adj
=0.0010). All the
species differed from each other in predawn Win June (all P
adj
\0.004) (see Fig. 4f).
Only in April did Midday Wnot differ between species. On the rest of the measuring dates,
P. pinea always differed from J. oxycedrus and Q. ilex (P
adj
\0.0001). All species dif-
fered from each other in the midday Win midsummer (P
adj
\0.0001) (Fig. 4g).
It should be mentioned that the physiological variables measured including gas
exchange variables, with certain exceptions, showed no statistical differences between
plots or height classes.
Simulations and physiological regeneration niches
Simulations under high irradiance (2000 lmol m
2
s
-1
) allow us to define the optimal
environmental conditions under which the instantaneous assimilation rate is maximized
and to analyse the interactive effect of soil moisture and temperature (Fig. 5). P. pinea is
able to attain very high A
n
, exhibiting the maximum expected value of 13.5 lmol CO
2
m
-2
s
-1
(confidence interval 8.9–18.1) under typical moist spring conditions (temperature
26 °C, soil moisture close to maximum water holding capacity). J. oxycedrus, displays the
maximum A
n
under similar conditions although in this case the expected values are much
lower 6.3 lmol CO
2
m
-2
s
-1
(confidence interval 2.3–14.2). In the case of Q. ilex, the
maximum expected value of 5.6 lmol CO
2
m
-2
s
-1
(confidence interval 1.1–10.1) is
achieved under midsummer conditions (temperature about 35.5 °C), with very little
dependence on soil moisture at 18 cm. Under extreme temperatures (mainly coinciding
with low soil moisture), both J. oxycedrus and P. pinea show a negative carbon balance,
while, Q. ilex is able to cope with high Twithout exhibiting a negative carbon balance and
maintaining high rates of assimilation. This is consistent with the A
n
values observed in the
field, where Q. ilex barely exhibited negative A
n
and maximum assimilation rate values
were observed under typical summer conditions. It should be noted that while the model
for Q. ilex lead to unbiased estimates (which makes it highly reliable for predicting the
cumulative assimilation rate over a whole period) it also resulted in low accuracy pre-
dictions for higher assimilation values.
To verify and validate our models we plotted the observed field data of A
n
against A
n
-
curves constructed using the models by including average VWC and Tfrom the real dates
of measurement (Fig. 6). Q. ilex shows high variability in A
n
irrespective of increase in
VWC (from 2.2 to 17.5 %) or W. In contrast, the other two species show larger sensitivity
to variations in VWC.
New Forests
123
We applied the developed models at eighteen real points, covering the whole range of
GSF, by entering real data for VWC,Tand Qin 30-min intervals. Thus, A
n
was transformed
into the A
n
accumulated during each time interval and then summed up for each day of the
growing period (Fig. 7). The total A
n
at each location differed between species, especially
in high GSF environments. Different patterns of total A
n
along the GSF gradient were also
found between species. In this regard, Q. ilex showed a steep increase in total A
n
with GSF
across the whole range, reaching a value of 31.0 ±1.0 mol m
-2
in the highest GSF class
([0.85). P. pinea showed an increasing–decreasing pattern of total A
n
with GSF, showing
the maximum value (16.2 ±0.7 mol m
-2
) in the GSF class from 0.45 to 0.6. J. oxycedrus
Fig. 6 Light response curves of A
n
modeled by entering observed values of soil moisture (VWC) and
temperature (T) from measuring dates, plotted against observed field data (symbols: filled circle,filled
square,asterisk). Each observed point represents the average for the observation taken on the measuring
date in the morning (mr) or at midday (md) in a 200 lmol irradiance class. Average T, VWC and midday
leaf water potential W(MPa) are shown for each condition
New Forests
123
increases total A
n
up to a GSF of 0.5, remaining more or less constant (close to
18 mol m
-2
) up to 0.9, and reaching the optimum GSF between these two values. The
most efficient species at low GSF was P. pinea, while for GSF [0.35 Q. ilex was the most
efficient.
Discussion and conclusions
Ecological significance of the estimated parameters and the physiological variables
A
max
in Q. ilex is maximized at a temperature of 35.5 °C. This temperature would appear to
be relatively high when compared to the range of optimal temperatures given in the
literature for photosynthesis in European oaks (10–35 °C) (Daas et al. 2008; Yamori et al.
2014). However, similar studies concerning the effect of temperature on photosynthesis in
young oaks (Dreyer et al. 2001), carried out using the biochemical model of leaf photo-
synthesis developed by Farquhar et al. (1980), highlight the fact that, in this species, both
the maximum rate of Rubisco activity and the potential electron transport rate are max-
imized at temperatures similarly high. In accordance with a high T
ref
, under summer
conditions, Q. ilex showed the highest values of maximum photochemical efficiency F
v
/F
m
at both predawn and midday, close to the optimal value of 0.83 (Fig. 4a, b), indicating the
absence of photoinhibitory stress, excess irradiance or water stress (Maxwell and Johnson
2000). Furthermore, minimal variations in F
v
/F
m
from predawn to midday during summer
suggest a strong capacity for acclimation to high temperature in Q. ilex. In addition to the
strong thermophilic character of Q. ilex, this species also displays the broadest thermal
range in which the plants are able to assimilate CO
2
. The narrowest temperature range is
exhibited by J. oxycedrus. This fact confers an adaptive advantage to Q. ilex over the two
conifer species.
The drought tolerant strategy of both Q. ilex (Baquedano and Castillo 2007) and J.
oxycedrus (Willson et al. 2008; Jovellar et al. 2013), which is consistent with their more
negative predawn and midday Win the summer (Fig. 4f, g), implies that stomata can
Fig. 7 Predicted values of CO
2
assimilation as a function of the global site factor for a growing period.
Each GSF class is constructed using three different GSF values. Error bars indicate standard deviation
New Forests
123
remain open, thus allowing a certain level of CO
2
assimilation. Furthermore, Q. ilex
showed a minimum disturbance in A
n
throughout the measuring period irrespective of the
low VWC measured at 18 cm in the summer (Fig. 4e), revealing a notable acclimation
capacity to drought. Our results are consistent with previous studies on stress tolerance
under controlled conditions in seedlings of Q. ilex (Gimeno et al. 2009), also displaying the
highest A
n
of the three species under summer stress conditions. In this regard, although
under summer conditions the response of A
max
in Q. ilex should not be affected by water
deficit since non-stomatal limitations to photosynthesis are expected, we relate the general
lack of dependence of A
max
on soil moisture at 18 cm in Q. ilex to the different strategies
adopted by the three species under water stress conditions and the regeneration strategy
inherent to each species. Our results can be linked to the strategy of regeneration by
resprouting from roots in Q. ilex, which provides an established, deep-rooted system for the
regeneration. Q. ilex is able to obtain groundwater from deeper soil horizons in comparison
to non-resprouting species (Ferrio et al. 2003; Baquedano and Castillo 2007), thus main-
taining a better water status. This conclusion is supported in part by the less negative
predawn W. In this regard, predawn Win Q. ilex differed from the other two species,
displaying an intermediate behaviour between J. oxycedrus and P. pinea in the summer
(Fig. 4f), which points to a better water status in Q. ilex compared to J. oxycedrus. The
highest predawn Win P. pinea is explained by its strategy to induce stomatal closure under
drought conditions. Conversely, our results revealed that for both P. pinea and J. oxyce-
drus,A
max
is highly dependent on soil moisture. P. pinea in particular, exhibits rapid
exponential growth of A
max
with soil moisture, making it the most sensitive species to
water availability.
Photoinhibition caused by low temperature can be an important factor limiting photo-
synthetic activity in continental-Mediterranean ecosystems (Larcher 2000). In this regard,
our results revealed different responses of the species. J. oxycedrus differs from the other
two species in that it exhibits almost null efficiency (a=0) at 4 °C while this occurs at
2°CinP. pinea, and at 1.2 °CinQ. ilex. These results are consistent with the maximum
photochemical efficiency of photosystem II (F
v
/F
m
). Low values of F
v
/F
m
provide an
indication of photoinhibition caused by low temperature. In this regard, Q. ilex showed the
highest predawn F
v
/F
m
of the three species in winter (0.74), followed by P. pinea (0.73)
and J. oxycedrus (0.69) (Fig. 4a), although no significant differences between species were
found on this measuring date. We conclude that J. oxycedrus is the species which is most
affected by temperature, reaching the point at which CO
2
assimilation stops before the
other two species, both with rising and falling temperatures.
Definition of the ecological regeneration niches of species
In accordance with the simulations of the interacting effects of VWC and Ton the net
photosynthetic rate (A
n
), we found that the highest A
n
in P. pinea can only be attained when
the soil moisture content is close to field capacity, which occurs during a brief period
following the spring rainfalls (Calama et al. 2013). This is consistent with the drought-
avoidance strategy of P. pinea (Pardos et al. 2009), characterized by the higher Wwhen
soil moisture content is low. P. pinea induces stomatal closure in accordance with an
isohydric behaviour (Schultz 2003), resulting in a sharp decrease in the carbon assimilation
rate to values well below its potential photosynthetic capacity (Calama et al. 2013). As
regards J. oxycedrus, the low predawn and midday Win summer evidences the fact that it
is better adapted to cope with low water availability than P. pinea, exhibiting a drought-
tolerant strategy similar to that of Q. ilex. However, we found that J. oxycedrus is also the
New Forests
123
least tolerant species to high temperature, exhibiting a negative carbon balance at a lower
maximum temperature than the other two species (42 °C). This result is consistent with the
lower F
v
/F
m
showed by J. oxycedrus in the summer, suggesting that the photoinhibition of
the photosynthesis is mainly induced by high temperature. Q. ilex, on the other hand, is a
drought-tolerant species that is able to cope with high as well as low temperatures without
exhibiting a negative carbon balance, highlighting its capacity to modify the thermal
tolerance according to the environmental temperature (Gimeno et al. 2009). This means
that under hot, dry summer conditions (common in our study area) Q. ilex is the only
species which maintains its CO
2
uptake. This behaviour is supported by the high F
v
/F
m
and
U
PSII
observed under summer conditions and the absence of any type of photoinhibition,
which makes Q. ilex significantly different from the other two species.
Although simulations of interacting effects of VWC and Tprovide important data with
regard to the behaviour of species, plant survival and photosynthetic activity of individual
plants is highly dependent on the light growth environment, which is considered the main
abiotic factor defining the regeneration niche of Mediterranean tree species (Gomez-
Aparicio et al. 2008; Perez-Ramos et al. 2013). In this regard, simulations of A
n
at eighteen
points (covering the whole range of GSF) over a vegetative period, provide a realistic
estimation of the physiological regeneration niches of the species (Fig. 7).
In ecosystems where the drought periods are not intense, an increase in irradiance leads
to higher CO
2
assimilation rates. However, when low water availability and high irradiance
occur together, as in Mediterranean ecosystems, the plants suffer photoinhibition processes
and their CO
2
assimilation rates will decrease (Valladares et al. 2005). Hence, Q. ilex
exhibits efficient photosynthetic activity with increasing GSF without suffering photo-
inhibition; reaching the optimal light environments at which A
n
is maximized under very
high GSF. Recent studies in Q. ilex stands have also shown that the A
n
of seedlings derived
from acorns increases with GSF when there is no water stress or photoinhibition. However,
findings suggest that the maximum probability of recruitment is attained with intermediate
light availability (Perez-Ramos et al. 2013). In this regard, it should be noted that the best
physiological performance under certain environmental conditions does not necessarily
imply the best regeneration niche for recruitment. However, in our study this fact makes Q.
ilex regeneration the most light-demanding, preferentially establishing in canopy gaps open
to the sky. As for J. oxycedrus, locations ranging from medium-shade conditions to large
gaps are suitable for maximizing A
n
. In the case of P. pinea, we determined that the
optimal conditions for maximizing A
n
are found in medium-shade environments, which
correspond to a GSF of around 0.5. P. pinea also shows the most shade-tolerant behaviour
of the three species. Although Mediterranean pines during their seedling stage are con-
sidered light-demanding (Awada et al. 2003;Go
´mez-Aparicio et al. 2006), several studies
suggest that medium-shade environments provide optimal conditions for P. pinea seedlings
(Pardos et al. 2010; Calama et al. 2013; Manso et al. 2013). Our measurement of U
PSII
,
which is an indicator of both the light saturation behaviour of the species in the field and
their photosynthetic performance, support our simulations. P. pinea was the most sensitive
species to light, the U
PSII
value decreasing from morning to midday by more than 66 % as
well as showing the lowest U
PSII
of the three species in the summer. By contrast, midday
U
PSII
in Q. ilex did not show significant differences between measuring dates, indicating
that the photosynthesis rate is not affected by light conditions over the year, and under-
lining the high light demanding behaviour of Q. ilex. In this regard, our results suggest that
photoinhibition in Q. ilex is caused by low temperature rather than by high irradiance or
high temperature, which is consistent with previous findings (Valladares et al. 2008).
Holmgren et al. (2012), in their meta-analysis of the interactive effects of light and water
New Forests
123
availability, found that very low levels of light affect the physiological and morphological
adaptations for coping with drought more severely in species which are capable of toler-
ating greater light and water stress. Our simulations are consistent with this finding,
suggesting that a reduction in positive canopy effect is more likely to occur in species that
are relatively tolerant to dry, high light conditions such as Q. ilex or to a lesser degree, J.
oxycedrus, whereas in the case of P. pinea the negative effects of drought tend to be
reduced under moderate shade but not under deep shade.
Our findings are supported by the observed survival rate of young plants of these three
species at the study site over the last 3 years. Mortality in P. pinus was 11 and 15 % in
plots 1 and 2 respectively; whereas mortality in J. oxycedrus was less than 2 % in either
plot and in the case of Q. ilex no mortality was observed.
We conclude that the optimal locations for P. pinea regeneration in these forests are
below-crown environments, which prevent photoinhibition of the photosynthetic apparatus
and mitigate the effects of high irradiance and high midday temperature in summer
(Calama et al. 2013; Manso et al. 2013). As for J. oxycedrus regeneration, optimal loca-
tions are found in open gaps as long as high temperature does not endanger survival. We
also suggest that Q. ilex is the best adapted species to cope with the expected changing
climatic conditions, characterized by increasing mean temperature and a drastic reduction
in precipitation. In this regard, optimal locations for Q. ilex regeneration are open gaps
where the other two species are unable to establish themselves because of high midday
temperature, high irradiance or low water availability in summer.
In terms of forest management, our results suggest that maintaining a canopy structure
which relates to an average global site factor (GSF) of around 0.5, will promote compe-
tition between the three species, whereas higher values will greatly favor the regeneration
of Q. ilex. Additionally, the conversion from coppices to high forest stands of Q. ilex may
improve regeneration from seed instead of resprouting, as well as create medium-shade
conditions for future P. pinea and J. oxycedrus seedlings.
Acknowledgments This study was financed by the project REGENFOR (S2009/AMB-1668, Regeneration
of Forest Systems of Madrid Community) and AGL-2010-15521.
References
Anten NPR, Schieving F, Werger MJA (1995) Patterns of light and nitrogen distribution in relation to whole
canopy carbon gain in C3 and C4 mono- and dicotyledoneous species. Oecologia 101:504–513
Aspinwall MJ, King JS, McKeand SE, Domec J-C (2011) Leaf-level gas-exchange uniformity and photo-
synthetic capacity among loblolly pine (Pinus taeda L.) genotypes of contrasting inherent genetic
variation. Tree Physiol 31:78–91
Awada T, Radoglou K, Fotelli MN, Constantinidou HIA (2003) Ecophysiology of seedlings of three
Mediterranean pine species in contrasting light regimes. Tree Physiol 23:33–41
Baquedano FJ, Castillo FJ (2007) Drought tolerance in the Mediterranean species Quercus coccifera,
Quercus ilex,Pinus halepensis, and Juniperus phoenicea. Photosynthetica 45:229–238
Battaglia M, Beadle C, Loughhead S (1996) Photosynthetic temperature responses of Eucalyptus globulus
and Eucalyptus nitens. Tree Physiol 16:81–89
Bond WJ, Midgley J (2001) Ecology of sprouting in woody plants: the persistence niche. Trends Ecol Evol
16:45–51
Breda N, Huc R, Granier A, Dreyer E (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
Calama R, Pue
´rtolas J, Madrigal G, Pardos M (2013) Modeling the environmental response of leaf net
photosynthesis in Pinus pinea L. natural regeneration. Ecol Model 251:9–21
New Forests
123
Daas C, Montpied P, Hanchi B, Dreyer E (2008) Responses of photosynthesis to high temperatures in oak
saplings assessed by chlorophyll—a fluorescence: inter-specific diversity and temperature-induced
plasticity. Ann For Sci 65:305
Denmead OT, Shaw RH (1962) Availability of soil water to plants as affected by soil moisture content and
meteorological conditions. Agron J 54:385–390
Dreyer E, le Roux X, Montpied P, Daudet FA, Masson F (2001) Temperature response of leaf photosyn-
thetic capacity in seedlings from seven temperate tree species. Tree Physiol 21:223–232
Farquhar GD, von Caemmerer S, Berry JA (1980) A biochemical model of photosynthetic CO
2
assimilation
in leaves of C3 species. Planta 149:78–90
Fellows AW, Goulden ML (2013) Controls on gross production by a semiarid forest growing near its warm
and dry ecotonal limit. Agric For Meteorol 169:51–60
Ferrio JP, Florit A, Vega A, Serrano L, Voltas J (2003) D13 C and tree-ring width reflect different drought
responses in Quercus ilex and Pinus halepensis. Oecologia 137:512–518
Gimeno TE, Pias B, Lemos-Filho JP, Valladares F (2009) Plasticity and stress tolerance override local
adaptation in the responses of Mediterranean holm oak seedlings to drought and cold. Tree Physiol
29:87–98
Gomez-Aparicio L et al (2008) Oak seedling survival and growth along resource gradients in Mediterranean
forests: implications for regeneration in current and future environmental scenarios. Oikos
117:1683–1699
Go
´mez-Aparicio L, Valladares F, Zamora R (2006) Differential light responses of Mediterranean tree
saplings: linking ecophysiology with regeneration niche in four co-occurring species. Tree Physiol
26:947–958
Grubb PJ (1977) Maintenance of species-richness in plant communities. Biol Rev 52:107–145
Hirose T, Ackerly D, Traw MB, Ramseier D, Bazzaz EA (1997) CO2 elevation, canopy photosynthesis and
optimal leaf area index. Ecology 78:2339–2350
Holmgren M, Go
´mez-Aparicio L, Quero JL, Valladares F (2012) Non-linear effects of drought under shade:
reconciling physiological and ecological models in plant communities. Oecologia 169:293–305
IPCC (2007) Climate Change 2007: impacts, adaptation and vulnerability. Contribution of Working Group
II to the fourth assessment. Report of the Intergovernmental Panel on Climate Change. Cambridge
University Press, Cambridge
Jovellar LC, Ferna
´ndez L, Mezquita M, Bolan
˜os F, Escudero V (2013) Structural characterization and
analysis of the regeneration of woodlands dominated by Juniperus oxycedrus L. in west-central Spain.
Plant Ecol 214:61–73
Kirschbaum MUF et al (1998) Modelling forest-growth response to increasing CO
2
concentration in relation
to various factors affecting nutrient supply. Glob Chang Biol 4:23–41
Larcher W (2000) Temperature stress and survival ability of Mediterranean sclaerophyllous plants. Plant
biosystems 134:279–295
Lin YS, Medlyn BE, Ellsworth DS (2012) Temperature responses of leaf net photosynthesis: the role of
component processes. Tree Physiol 32:219–231
Manso R, Pukkala T, Pardos M, Miina J, Calama R (2013) Modelling Pinus pinea forest management to
attain natural regeneration under present and future climatic scenarios. Can J For Res Rev Can Rech
For 44:250–262
Manso R, Pardos M, Calama R (2014) Climatic factors control rodent seed predation in Pinus pinea L.
stands in Central Spain. Ann For Sci 71:873–883
Maran
˜o
´n T, et al. (2004) Heterogeneidad ambiental y nicho de regeneracio
´n. In: Ambiente MdM (ed)
Ecologı
´a del bosque mediterra
´neo en un mundo cambiante
Markos N, Kyparissis A (2011) Ecophysiological modelling of leaf level photosynthetic performance for
three Mediterranean species with different growth forms. Funct Plant Biol 38:314–326
Marshall B, Biscoe PV (1980) A model for C3 leaves describing the dependence of net photosynthesis on
irradiance. J Exp Bot 31:29–39
Martinez-Ferri E, Manrique E, Valladares F, Balaguer L (2004) Winter photoinhibition in the field involves
different processes in four co-occurring Mediterranean tree species. Tree Physiol 24:981–990
Maxwell K, Johnson GN (2000) Chlorophyll fluorescence—a practical guide. J Exp Bot 51:659–668
Mengistu T, Sterck FJ, Fetene M, Tadesse W, Bongers F (2011) Leaf gas exchange in the frankincense tree
(Boswellia papyrifera) of African dry woodlands. Tree Physiol 31:740–750
Pardos M, Jime
´nez MD, Aranda I, Pue
´rtolas J, Pardos JA, Baquedano FJ (2005) Water relations of cork oak
(Quercus suber L.) seedlings in response to shading and moderate drought. Ann For Sci 62:377–384
Pardos M, Calama R, Climent J (2009) Difference in cuticular transpiration and sclerophylly in juvenile and
adult pine needles relates to the species-specific rates of development. Trees 23:501–508
New Forests
123
Pardos M, Pue
´rtolas J, Madrigal G, Garriga E, de Blas S, Calama R (2010) Seasonal changes in the
physiological activity of regeneration under a natural light gradient in a Pinus pinea regular stand. For
Systems 19(3):367–380
Pardos M, Calama R, Mayoral C, Madrigal G, Sa
´nchez-Gonza
´lez M (2014) Addressing post-transplant
summer water stress in Pinus pinea and Quercus ilex seedlings. iForest. doi:10.3832/ifor1256-007
Perez-Ramos IM, Rodriguez-Calcerrada J, Ourcival JM, Rambal S (2013) Quercus ilex recruitment in a
drier world: a multi-stage demographic approach. Perspect Plant Ecol Evol Syst 15:106–117
Rodrı
´guez-Calcerrada J, Limousin JM, Martin-StPaul NK, Jaeger C, Rambal S (2012) Gas exchange and
leaf aging in an evergreen oak: causes and consequences for leaf carbon balance and canopy respi-
ration. Tree Physiol 32:464–477
Ruiz-Labourdette D, Nogue
´s-Bravo D, Saı
´nz H, Schmitz MF, Pineda FD (2012) Forest composition in
Mediterranean mountains is projected to shift along the entire elevational gradient under climate
change. J Biogeogr 39:162–176
Schultz HR (2003) Differences in hydraulic architecture account for near-isohydric and anisohydric be-
haviour of two field-grown Vitis vinifera L. cultivars during drought. Plant, Cell Environ 26:1393–1405
Stegemann J, Timm HC, Ku
¨ppers M (1999) Simulation of photosynthetic plasticity in response to highly
fluctuating light: an empirical model integrating dynamic photosynthetic induction and capacity. Trees
14:145–160
Thornley JHM, Johnson IR (1990) Plant and crop modelling: a mathematical approach to plant and crop
physiology. Clarendon, Oxford
Tjoelker MG, Oleksyn J, Reich PB (2001) Modelling respiration of vegetation: evidence for a general
temperature-dependent Q 10. Glob Change Biol 7:223–230
Valladares F et al (2004) Estre
´shı
´drico: ecofisiologı
´a y escalas de la sequı
´a. In: Ambiente MdM (ed)
Ecologı
´a del bosque mediterra
´neo en un mundo cambiante, pp 163–190
Valladares F, Dobarro I, Sa
´chez-Go
´mez D, Pearcy RW (2005) Photoinhibition and drought in Mediter-
ranean woody saplings: scaling effects and interactions in sun and shade phenotypes. J Exp Bot
56:483–494
Valladares F et al (2008) Is shade beneficial for Mediterranean shrubs experiencing periods of extreme
drought and late-winter frosts? Ann Bot 102:923–933
West AG, Hultine KR, Sperry JS, Bush SE, Ehleringer JR (2008) Transpiration and hydraulic strategies in a
pinon-juniper woodland. Ecol Appl 18:911–927
Willson CJ, Manos PS, Jackson RB (2008) Hydraulic traits are influenced by phylogenetic history in the
drought-resistant, invasive genus Juniperus (Cupressaceae). Am J Bot 95:299–314
Xu JZ, Yu YM, Peng SZ, Yang SH, Liao LX (2014) A modified nonrectangular hyperbola equation for
photosynthetic light-response curves of leaves with different nitrogen status. Photosynthetica
52:117–123
Yamori W, Hikosaka K, Way DA (2014) Temperature response of photosynthesis in C-3, C-4, and CAM
plants: temperature acclimation and temperature adaptation. Photosynth Res 119:101–117
Zaragoza-Castells J et al (2008) Climate-dependent variations in leaf respiration in a dry-land, low pro-
ductivity Mediterranean forest: the importance of acclimation in both high-light and shaded habitats.
Funct Ecol 22:172–184
New Forests
123
