Coupling relationship of structure and tree diversity between upper and lower canopy layer based on structural equation model.

In order to clarify the relationship of structure and tree diversity between upper and lower canopy layers, we built the structural equation model of the site condition-structure of upper canopy layer-tree diversity of upper canopy layer-structure and tree diversity of lower canopy layer based on the secondary multiple species forest beside the Ashi River in southern Heilongjiang Province. The results showed that site condition (ξ1) had a positive impact on structure (ξ2) and tree diversity (ξ3) of the upper canopy layer with the total path coefficients of 0.59 and 0.75, and the direct path coefficients of 0.59 and 0.43, respectively. Site condition (ξ1) had a negative impact on structure and tree diversity of the lower canopy layer (ξ4) with the total path coefficients of -0.58, and the direct path coefficient of -0.12. Structure of the upper canopy layer (ξ2) had a positive impact on tree diversity of the upper canopy layer (ξ3) and a negative impact on structure and tree diversity of the lower canopy layer (ξ4), with the total path coefficients of 0.55 and -0.55, and the direct path coefficient of 0.55 and -0.38, respectively. Trees diversity of upper canopy layer (ξ3) had a directly negative impact on structure and tree diversity of the lower canopy layer (ξ4), with the path coefficient of -0.31. Of all the observation variables of structure and tree diversity for the upper canopy layer, canopy area (x7) was the most important factor affecting density of the lower canopy layer (x11), with the total path coefficient of -0.19 and the direct path coefficient of -0.12. Average height (x6) was the most important factor affecting average height of the lower canopy layer (x12), with the total path coefficient of -0.28 and the direct path coefficient of -0.13. Canopy area (x7) and height diversity (x10) were the most important factors affecting tree species diversity of the lower canopy layer (x14), with the total path coefficients of -0.38 and -0.36, respectively. The impact of structure (ξ2) and tree diversity (ξ3) of the upper canopy layer on the canopy area of the lower canopy layer (x13) was not obvious.