Leaf physiological measurements
Midday leaf pressures (Ψ md, MPa) of each species were measured during two successive sunny days with a pressure chamber (PMS1000, Albany, OR, USA) in July 2018. Eight sun-exposed terminal twigs with each from a different tree were sampled at midday for each species (between 12: 00 and 13: 00) and then samples were sealed in small plastic bags containing moist paper towels and transported to laboratory immediately. All the samples were measured within 1 hr after excision. Hydraulic safety margins (HSM) of each species were calculated as the difference betweenΨ md and P 50.
Leaf osmotic potential at zero turgor (π 0) and at full turgor (π 100) were obtained from the leaf pressure-volume (P-V) relations (Tyree & Hammel, 1972). The branches c.a. 1 m in length of each species were sampled at predawn, and transported to laboratory in dark plastic bags with the ends of branches immerged in water. After a 2 h rehydration, the terminal shoot samples were cut off and used for determining the P-V relations. The mass of saturated terminal shoot was measured using an analytical balance (CPA 225D, Sartorius Inc., Germany). The mass and water potential of the samples were measured periodically during a slow dehydration process. Samples were over-dried at 70 ℃ for 48 hr to measure the dry mass at the end. The P-V curves were fitted using a computer program by Schulte and Hinckley (1985).
In August 2018, the stomatal conductance (g s, mol m-2 s-1) and net CO2assimilation (A ) were measured using a LICOR-6400 gas exchange system (LI-COR, Lincoln, NE, USA) equipped with a 6400-05 conifer chamber (Lincoln, Nebraska, USA) between 0900 and 1030 hours. The measurements were conducted on mature and fully developed sunlit needles of eight south-facing trees of each species at 400 ppm reference CO2 and 1200 μmol m-2s-1 photosynthetic photon flux density. The humidity, temperature and photosynthetic active radiation of leaf chamber were set as ambient conditions. After gas exchange measurements, needles fit in the leaf chamber were excised and measured their area back in the laboratory. Then gas exchange needles were dried and weighed to determine their dry mass. The net photosynthetic rate per leaf area (A a, μmol m-2s-1) and net photosynthetic rate per mass (A m, μmol g-1s-1) were calculated as A divided by leaf area and leaf dry mass, respectively. The intrinsic water use efficiency of photosynthesis (WUEi) was calculated as the ratio ofA to g s for each twig measured.
Statistical analysis
The correlations between functional traits (K s,A m, WD, HSM and π 0) and radial growth resilience indices (RT, RC and RS), growth rate (CBA15), and relationship between mean sensitivity and CBA15 were evaluated by linear regression analyses. One-way analysis of variance (ANOVA) was used to test the differences in functional traits among studied species using the SigmaPlot 12.5 software (Systat Software, Inc., San Jose, CA, USA). Pearson’s correlation coefficients between functional traits, growth resilience indices, mean sensitivity and CBA15 were also calculated.