Competitive spatial patterns for Moso bamboo on Mount Tianmu
Moso bamboo(Phyllostachys edulis)stands, one of the most economic forest-types in China, possess many advantages over other types, such as rapid growth, wide distribution, and high production. Much research on competition of Moso bamboo, such as intraspecific and interspecific competition, the relationship between biomass and competition, and improvements of competition index, has been conducted, but this study focused on spatial variability and its dynamic analysis of bamboo competition in distance scale dependence, which has been neglected. Based on data collected from 2009-2015 in the Moso bamboo stand of Tianmu Mountain Na-tional Nature Reserve of Zhejiang Province, the Hegyi Competition Index(CI)was calculated based on Geo-graphic Information System(GIS). Spatial semivariogram was used to analyze the auto-correlation range of CI and its anisotropy. Also Moran's I and Empirical Bayesian Kriging (EBK)were used to describe dynamic changes in the competitive spatial pattern.Results showed that spatial distribution of CI for Moso bamboo had a significant general global autocorrelation (P<0.001). Bamboos of the same competitive intensity were dis-tributed evenly. Also, Moso bamboo stands had strong spatial variation due to their structural features. The CI spatial pattern, influenced by micro topography, existing anisotropy, and an extreme climate, reduced spatial heterogeneity and was inclined to be isotropic.The competition range appeared inverse to change in the on-year and off-year growth cycles with the average range being 4.306 m. This range could be used as the distance scale for the basis of determining competition unit radius and for control of the Moso bamboo competitive rela-tionship. Even though Moso bamboo competition increased from off-years to on-years, its competition lessened from on-years to off-years. These research results could provide a reference for adjusting and controlling the competitive relationships in Moso bamboo stand management.