关键词: 微塑料, 多物种入侵, 植物物种丰富度, 土壤非生物因子, 结构方程模型

Abstract: Aims Microplastic pollution and multispecies invasions are escalating global threats to native plant communities. However, how plant communities respond to microplastics in the context of multispecies invasions, and the mechanisms driving these responses, are still largely unknown. To address this gap, we conducted a controlled experiment. Methods We established synthetic plant communities composed of three grass species, three legume species, and three invasive species along a 12-level abundance gradient of polypropylene microplastics. To elucidate the underlying ecological drivers, we measured soil physicochemical properties, community-level fluorescence traits, species richness, community biomass, and species relative abundance. Furthermore, structural equation modeling was used to identify key pathways. Important findings (1) Overall, polypropylene (PP) microplastics significantly decreased plant species richness, especially at intermediate levels, without a consistent monotonic trend. (2) Unlike plant species richness, PP microplastics did not decrease overall community biomass, and community biomass exhibited a U-shaped response, decreasing at low to medium microplastic levels and increasing significantly at high levels. (3) PP microplastics had no significant effect on the relative abundance of invasive plants, but significantly affected community-level fluorescence traits, and soil phosphorus and potassium. Structural equation modeling revealed distinct pathways by which microplastics influenced plant species richness, invader relative abundance, and community productivity. These findings suggest that microplastics could regulate plant diversity and community productivity primarily by altering soil nutrient availability and plant functional traits. Furthermore, soil nutrients and pH could exert contrasting influences on invasion processes under microplastic stress.

Key words: microplastics, multiple species invasion, plant species richness, soil abiotics, structural equation modeling