Table 1.
Indicators of water stress degree.
Figure 1.
Gas exchanges (photosynthesis, stomatal conductance and isoprene).
Seasonal course of the net photosynthesis (Pn, graph a), stomatal conductance to water vapor (Gw, graph b) and isoprene emissions (graph c) under control (–⧫–), mild () and severe (···⋄···) water stressed. Note that 70% of isoprene emissions were mostly measured under standard conditions (refer to materials and method for details). Differences are tested using a two-way ANOVA repeated measures (F) followed by Tukey tests. Since interaction between seasonality and treatment is not significant according to the two-way ANOVA, results of water stress impact are shown in the small graph where data of all months are pooled together and differences between treatments are denoted by capital letters (A>B) while seasonality impact is shown in the main graph (in lower case letters: a>b>c). Values are mean ± SE of n = 5.
Figure 2.
Seasonal course of the variable to maximum fluorescence ratio (Fv/Fm, graph a) and maximal electron transport rate (ETRmax, graph b) under control (–⧫–), mild () and severe (···⋄···) water stress. Differences are tested using a two-way ANOVA repeated measures (F) followed by Tukey tests. Since interaction between seasonality and treatment is not significant according to the two-way ANOVA, results of water stress impact are shown in the small embedded graph where data of all months are pooled together and differences between treatments are denoted by capital letters (A>B) (similar capital letters indicate the absence of water stress influence) while seasonality impact is shown in the main graph (in lower case letters: a>b>c >d>e). Values are mean ± SE of n = 5.
Figure 3.
Isoprene emission rate as percentage of carbon re-emitted.
Seasonal course of the percentage of photosynthetically assimilated carbon re-emitted as isoprene under control (–⧫–), mild () and severe (···⋄···) water stress. Differences are tested using a two-way ANOVA (F) repeated measurements followed by Tukey tests. Since interaction between seasonality and treatment is not significant according to the two-way ANOVA, results of water stress impact are shown in the small embedded graph where data of all months are pooled together and differences between treatments are denoted by capital letters (A>B) while seasonality impact is shown in the main graph (in lower case letters: a>b). Values are mean ± SE of n = 5 saplings.
Figure 4.
Experimental and calculated isoprene emission factors.
Comparison between the isoprene emission factor calculated with Guenther et al. (1993) algorithm () and the experimental isoprene emission factor (▪) obtained under standard conditions (30±1°C and 1000±50 µmol m−2 s−1 of PPFD) for control, mild and severely water stressed sapling. t: value of the paired sample comparison tests. N.S.: not significant, *0.01<P<0.05. Values are mean ± SE of n = 6–7 saplings.
Figure 5.
Leaf total phenolic concentration.
Seasonal course of leaf total phenolic concentration under control (–⧫–), mild () and severe (···⋄···) water stress. Differences are tested using a two-way ANOVA (F) followed by Tukey tests. Since interaction between seasonality and treatment was not significant, the impact of water stress is shown in the small embedded graph where data of all months are pooled together and differences between treatments are denoted by capital letters (similar capital letters indicate the absence of water stress influence) while seasonality impact is shown in the main graph (in lower case letters: a>b>c). Values are mean ± SE of n = 5.
Figure 6.
Shoot, foliar and root biomass (gDM ind−1) during leaf senescence (in November) for control, mild and severe water stress. Differences are tested with one-way ANOVA (F). Capital black lower case black and white lower case letters denote statistical differences for shoot, foliar and root biomasses respectively with a>b or A>B given by Tukey tests. Values are mean ± SE. n = 5.
Figure 7.
Mapping control, mild stress and severe stress.
Two-dimensional mapping of the Principal Component Analysis (PCA) performed for all plant traits measured at the end of the experiment (plant growth, emission or concentration of secondary metabolites, gas exchange and water status). This analysis was performed on n = 5 trees per treatment. Traits shown at the end of some arrows correspond to the most explanatory traits.
Figure 8.
Seasonal course of the leaf mass per area ratio (LMA) under control (–⧫–), mild () and severe (···⋄···) water stress. Differences are tested using a two-way ANOVA (F) repeated measurements followed by Tukey tests. Since interaction between seasonality and treatment is not significant according to the two-way ANOVA, results of water stress impact are shown in the small embedded graph where data of all months are pooled together and differences between treatments are denoted by capital letters (A>B) while seasonality impact is shown in the main graph (in lower case letters: a>b). Values are mean ± SE of n = 5 saplings.