Revisiting Alwyn H. Gentry’s forest transect data: latitudinal beta diversity patterns are revealed using a statistical sampling-model-based approach

作者：Anne Chao, Chun‐Huo Chiu, Kai‐Hsiang Hu, David Zelený · 发表于：Japanese Journal of Statistics and Data Science · 年份：2023 · DOI：10.1007/s42081-023-00214-1 · 被引用次数：14 · 研究领域：Ecology and Vegetation Dynamics Studies、Species Distribution and Climate Change、Wildlife Ecology and Conservation

Abstract Gentry’s forest transect data have been frequently used to assess global patterns of plant diversity and plant species compositional changes along environmental and geographical gradients. Based on the worldwide woody plant abundance records from Gentry’s 197 localities/plots (each consisting of ten 2 m × 50 m subplots/transects), we apply a recently developed sampling-model-based standardization method (iNEXT.beta3D standardization) to examine how beta diversity among subplots varies with latitude. Beta diversity quantifies the extent of among-subplot differentiation which represents the interacting effect of species abundance distribution and spatial aggregation. Here beta diversity is obtained by a multiplicative decomposition scheme based on the framework of Hill numbers of any order q ≥ 0. Under statistical sampling models, data in nearly all of the 197 localities were incomplete, i.e., there were species present in the assemblage but undetected in the data. For Gentry’s data collected along narrow transects, the dependence among sampled individuals due to spatial aggregation is generally weak. The observed beta diversity depends on the among-subplot differentiation and sampling effort/completeness, which in turn induce dependence of the observed beta diversity on alpha and gamma diversity. To control for sampling effort/completeness, the iNEXT.beta3D method standardizes both alpha and gamma diversity at the same level of sample coverage to formulate coverage-ba...