Abstract:
ObjectiveTo accurately describe and intuitively express forest structure information, and provide basic information for forest structure regulation and reconstruction, N-variate distributions were used to comprehensively and systematically characterize the spatial structure of the Quercus aliena var. acuteserrata natural mixed forest in Xiaolong Mountains, Gansu Province of northwestern China.
MethodEach tree in the 70 m × 70 m permanent sample plot was located and surveyed. The mingling (M), uniform angle index (W), neighborhood comparison (U) and crowding (C) of each tree were calculated by the Winkelmass software. The relative frequency of N-variate distributions was counted by the Excel Pivot Tables, and their corresponding multivariate diagrams were graphed by the R 3.4.3 and Origin 2015.
ResultIn the Quercus aliena var. acuteserrata natural mixed forest, the whole forest and most trees were randomly distributed, well mixed, relatively dense. The whole forest was moderately differentiated and each neighborhood comparison degree occupied near 20% trees. Among the bivariate distribution, trivariate distribution and quadrivariate distribution, the most prominent characteristics of the spatial structure was as follows: no matter how the structural parameters were combined, most trees were randomly distributed or well mixed under different structural combinations.
ConclusionThe N-variate distribution comprehensively and systematically describes the spatial structure characteristics from different levels and angles. Different distributions make their respective advantages complementary to each other and realize the stepwise interpretation process from point to hyperploid. 3D figures with double X horizontal coordinates or double Y vertical coordinates intuitively graph the structure information of the multivariate distribution. The N-variate distributions provide prior information for forest structure regulation and reconstruction.
下载: