Comparison Of Physiological And Wood Anatomical Traits Between Upper And Low Layer Trees In Evergreen Broad-leaved Forests In Tiantong, Zhejiang Province

The synusia is a kind of physiognomy for plants to adapt to the heterogeneity environment of the vertical structure. Thus different heights of plants have different light and water use strategies. Our research about the physiological mechanism of plant height differentiation helps to deepen our understanding of the hypothesis tree height restrictions, and to lay the theoretical foundation for the vertical structure of species coexistence mechanisms.In this study, according to the environmental factor heterogeneity of the vertical structure of different community heights, the study sample was divided into upper layer tree and lower layer tree. We chose three successional series of plots in Tiantong Forest Park, including three 10m ×10m plots of early successional shrub, three 20m×20m plots of mid-successional Schima superba forest and three 20m × 20m plots of late-successional Castanopsis fargesii forest. We first measured tree architecture characteristics, hydraulic and light resource utilization functional traits, and anatomical traits. Then we explored the best model of tree height variation through OLS regression.(1) The tree height, crown thickness, crown area, basal diameter of upper layer tree was significantly larger than the lower layer tree (p< 0.001). The 1sh branch number of upper layer tree was significantly less than the lower layer tree (p< 0.01). The upper layer tree was constructed by larger volume, compact crown shape, but the pattern for lower layer tree was the opposite.(2) The light resource utilization traits (including maximum photosynthetic rate, respiration rate, light saturation point and light compensation point) of upper layer tree were significantly higher than the low layer tree (p< 0.05). Similarly, hydraulic use traits (including hydraulic conductivity, specific conductivity, maximum hydraulic conductivity, maximum specific conductivity, sap flux and sap flow density) of upper layer tree was significantly higher than the low layer tree (p< 0.05). It indicated that the upper layer tree and low layer tree had different light and water resource use pattern.(3) The wood vessel density and branch density of both upper layer tree and low layer tree were not significantly different from each other (p> 0.05), while other anatomical traits such as vessel diameter, vessel lumen area, branch lumen area, branch diameter, sapwood area were significantly different (p< 0.01). The results suggested that these anatomical traits of upper layer tree significantly greater than the low layer tree.The correlation between anatomical traits with each of light and water use traits indicated that anatomical traits was significantly associated with plant light and water use traits (p> 0.05). The principal component analysis (PCA) for anatomical traits with light and water use traits showed that the contribution of the vessel diameter to plant light, water use was greater than vessel density. The linear regression between the first axis of PCA for light and water resources utilization traits and the first first axis of PCA for anatomical traits showed that anatomical traits largely determines the vertical structure of the plant light and water use pattern.(4) Significantly positive allometric relationships between 2st branch conductivity-2st branch radius and 2st sapwood area were found in both upper layer tree and low layer tree (p < 0.05). There was no difference in the regression slope between upper layer tree and low layer tree, and the common regression slope was calculated to be greater than 1 (p< 0.05). The intercept was significantly greater in upper layer tree than in low layer tree (p< 0.05), indicating that upper layer tree have higher conductivity at a given branch radius and sapwood area than low layer tree.With respect to biomass and metabolic rate relationship, there was also no difference in regression slope between upper layer tree and low layer tree, and the common regression slope was mostly significantly less than 1 (p< 0.05), except the biomass-sap flux not significantly different from 1 (p> 0.05). The intercept of the upper layer tree in the regression relationship was significantly larger than the low layer tree suggests that the upper layer tree and low layer tree formed two different resource utilization patterns. The upper layer tree was high light and water use species, while opposite pattern was true for low layer tree.(5) The model which integrates of tree architecture, light utilization traits, hydraulic utilization traits and anatomical traits is the best one, for the lowest AIC and highest R2.