Abstract
Abstract
Background
Species and structural diversity are important for understanding the formation of forest communities, key ecological processes, and improving forest ecological functions and services, but their spatial characteristics have received little attention. Based on the spatial relationships among neighbouring trees, we proposed to divide trees within a structural unit into 15 structural types, and used the univariate distributions of the uniform angle index (W), mingling (M), and dominance (U), along with four common species diversity indices, to analyse the diversity of structural types in natural forests near the Tropic of Cancer.
Results
Only a portion of clumped class maintained aggregation, most exhibited a random pattern. Species mixture increased exponentially across distribution classes, and abundance and richness exhibited an initial increase followed by a slight decrease. The distribution patterns of mixture classes varied from highly clustered to random, and M distributions gradually shifted from an inverted J-shaped curve to a J-shaped curve. Abundance and richness exhibited an exponential distribution, whereas the Shannon–Wiener index increased linearly. The W distribution of differentiation classes approximated a normal distribution, whereas M distributions exhibited a J shape. The U distribution of each structure type was approximately 0.2.
Conclusions
These results reveal the species and structural diversity characteristics of trees at the structural type level and expand our knowledge of forest biodiversity. The new method proposed here should significantly contribute to biodiversity monitoring efforts in terrestrial ecosystems, and suggests that higher standards for the simulation and reconstruction of stand structure, as well as thinning in near-natural forests, is warranted.
Funder
the National Natural Science Foundation of China
Scientific Research Capacity Building Project for Laibin Jinxiu Dayaoshan Forest Ecosystem Observation and Research Station of Guangxi
Publisher
Springer Science and Business Media LLC