Abstract
Habitat fragmentation dramatically reshapes species richness and community composition. However, most estimates of β-diversity rely on incidence-based metrics, which consider only species presence/absence. Here, we introduce a novel framework that explicitly incorporates species abundance and intraspecific trait variation (ITV) into the quantification of taxonomic, functional, and phylogenetic β-diversity, allowing a more nuanced understanding of community differentiation. To demonstrate the utility of this framework, we quantified the effects of island area and isolation on β-diversity across plant communities in China's Thousand Island Lake. Abundance-weighted taxonomic multiple-site/pairwise β-diversity showed substantially higher nestedness and stronger nestedness-area relationship than incidence-based metrics, indicating that species-poor communities are not only subsets of richer ones but share similar abundance hierarchies, highlighting strong environmental filtering and hierarchical species sorting. We also found that the turnover component was less sensitive to isolation, suggesting limited dispersal effects. Incidence-based functional and phylogenetic distances increased with differences in island area, but these associations weakened or disappeared in abundance-weighted measures, suggesting stronger environmental filtering and functional/phylogenetic clustering among larger islands. Only abundance-weighted standardized effect sizes increased with island area differences. Additionally, ITV further amplified functional nestedness and buffered the influence of isolation on turnover, emphasizing its role in mitigating dispersal limitations. By jointly considering abundance and ITV, two often-overlooked but critical dimensions, this study advances our understanding of how fragmentation shapes β-diversity. These findings highlight the importance of integrating abundance-weighted and trait-based metrics into conservation strategies to better detect functionally important species, prioritize larger habitat patches, and design biodiversity monitoring that captures within-species variation.