Figure 1.
Climate diagram according to Walter & Lieth [28] of Xingangshan, the location of the experimental sites.
Elevation: 211 m above sea level. Observation period was March 2009 to October 2012. Mean annual temperature was 17.4°C and total annual precipitation was 1635 mm. Monthly precipitation below 100 mm is scaled 2∶1 with mean monthly temperature (vertically hatched) and above 100 mm 15∶1. Turquoise bars below the x-axis show the months where frosts can occur (when absolute monthly minimums are equal or lower than 0°C). Climate data were recorded by a meteorological station established at the very center of the experimental site (Kühn, unpublished).
Table 1.
Tree species planted in the BEF-China experiment and included in this study.
Figure 2.
Simplified overview of the xylem hydraulics measurement installation.
A) Pressure chamber with the stem segment, B) tank with the perfusion solution, C) flush tank, D) three-way stopcock, E) electronic scale, F) nitrogen pressure cylinder, and G) pressure gauge.
Table 2.
List of the leaf traits measured across the 39 tree species.
Figure 3.
Conductivity rates as a function of decreasing water potential, taking Castanea henryi as an example.
Outline dots show the measured values of three independent vulnerability curves. Dotted lines show the single regression functions obtained from the measured value per vulnerability curve, obtained from a 3-parametric sigmoid function. The bold line shows the regression lines constructed from the means of the parameters from the three single vulnerability curves. Filled dots represent Ψ50 values.
Figure 4.
Plots of modeled specific xylem hydraulic conductivity KS versus water potentials for all 39 species included in the study.
A) and B) show the same data at different scale of KS. Deciduous species are shown in red, evergreen species shown in black. For species abbreviations see Table 1. For details of calculation of regression lines, see Fig. 2 and Methods.
Figure 5.
Boxplots characterizing the species set by leaf habit.
A) xylem pressure at 50% loss of the maximum specific xylem hydraulic conductivity (Ψ50) and B) specific xylem hydraulic conductivity of the xylem (KS) as a function of leaf habit. d = deciduous, e = evergreen. The probability values were derived from an ANOVA.
Figure 6.
Principal component analysis (PCA) biplots showing the relationships between the mean values of all chemical and morphological leaf traits (black arrows), all parameters of stomatal regulation (green arrows) and all parameters of cavitation sensitivity (blue arrows).
A) and C) PCA axes 1 and 2, B) and D) PCA axes 1 and 3. A) and B) loadings of the different traits. C) and D) species scores in the PCA shown separately by leaf habit. Deciduous species are shown in red and evergreen species in black. See Table 1 for abbreviations of species names. Eigenvalues: axis 1 = 7.03, axis 2 = 4.2, axis 3 = 3.48, with cumulative proportion of explained inertia 20.1%, 33.0% and 43.3%, respectively. See Table 2 for abbreviations of trait names.
Figure 7.
Ψ50 as a function of A) leaf area (p = 0.0315, r = 0.34), B) leaf toughness (p<0.0003, r = −0.47), C) leaf nitrogen concentration (p<0.0075, r = 0.43), D) leaf carbon to nitrogen ratio (p<0.0078, r = −0.43), E) leaf magnesium concentration (p<0.042, r = 0.33) and F) hydraulically weighted conduit diameter (p<0.01, r = 0.39).
Filled black dots represent species of evergreen leaf habit; empty dots represent species of deciduous leaf habit.
Figure 8.
KS as a function of A) leaf area (p<0.044, r = 0.1), B) leaf nitrogen concentration (p<0.00001, r = 0.59), C) leaf carbon to nitrogen ratio (p<0.0019, r = 0.48), D) mean area of conducting vessels (p<0.0001, r = 0.85), and E) hydraulically weighted diameter of conducting vessels (p<0.0001, r = 0.72).
Filled black dots represent species of evergreen leaf habit and empty dots represent species of deciduous leaf habit. In B) and C) KS also showed significant interactions with leaf habits. Dotted lines represent species of evergreen and broken lines represent species of deciduous leaf habit. B) leaf habit p = 0.061, interaction leaf nitrogen concentration and leaf habit p = 0.049, C) leaf habit p = 0.026, interaction leaf carbon to nitrogen ratio and leaf habit p = 0.031.
Table 3.
Correlation matrix of hydraulic xylem characteristics with numerical leaf traits and parameters of stomatal control.
Table 4.
Regression equations for all significant correlations of Ψ50 and KS to the functional traits.