关键词: 干扰强度, 土壤理化性质, 湿地, 多样性, 植物功能性状

Abstract: Wetlands are ecologically critical ecosystems increasingly impacted by anthropogenic disturbances. Understanding how disturbance shapes plant communities via soil conditions, functional traits and biodiversity is essential for wetland conservation. Here, we aimed to examine how disturbance intensity shapes soil–plant interactions and multi-dimensional biodiversity in a protected wetland. We surveyed four zones of the Crested Ibis National Nature Reserve (Shaanxi, China) along a disturbance gradient defined by zoning categories and field indicators such as vegetation degradation and human activity. Results showed that disturbance significantly altered soil nutrients, salinity and stoichiometry, especially increasing N:P ratios and conductivity in highly disturbed zones. These shifts corresponded with trait changes from acquisitive to conservative strategies (e.g. reduced specific leaf area (SLA), increased leaf dry matter content (LDMC) and C:P, N:P). While species richness peaked under high disturbance, functional diversity (FR_ic, FD_is and Rao’s Q) and Faith’s PD were highest at moderate levels. Mean pairwise phylogenetic distance (MPD) and mean nearest taxon distance (MNTD) increased under moderate to high disturbance, indicating reduced filtering and greater stochasticity. Principal component analysis, redundancy analysis, regression and structural equation modeling (SEM) supported these findings, highlighting soil-mediated effects on trait expression and diversity patterns. Our findings suggest that moderate disturbance enhances functional and phylogenetic diversity, promoting ecosystem resilience through greater niche and evolutionary space. Excessive disturbance, however, leads to homogenization and instability. Integrating trait-based and phylogenetic perspectives supports adaptive, low-intensity management to sustain wetland biodiversity and multi-functionality.

Key words: disturbance intensity, soil physicochemical properties, wetlands, diversity, plant functional traits