Abstract:

Aims: Under the dual pressures of climate change and anthropogenic disturbances, biological invasions by non-native species are intensifying and have become a major driver of global biodiversity loss. Mountain regions are considered biodiversity refugia under global change, yet the mechanisms by which mountain ecosystems resist non-native plant invasions remain poorly understood. Exploring the mechanisms underlying the “barrier effect” of mountain ecosystems against non-native species invasions is crucial for understanding how biodiversity refugia function under global change, informing conservation strategies, and enhancing the resilience of mountain ecosystems to biological invasions. 

Methods: This study investigated the mechanisms underlying invasion resistance in mountain ecosystems using data from 134 plant communities in the Qionglai Mountain Range of China. By integrating phylogenetic metrics, native community attributes, and environmental variables such as elevation and anthropogenic disturbance, we evaluated the relative importance of biotic and abiotic factors in shaping the occurrence probability and abundance of non-native species. Analytical approaches included generalized linear mixed models, Bayesian structural equation modeling, and random forest analysis. 

Results: The results showed that both the occurrence probability and abundance of non-native species significantly decreased with increasing phylogenetic distance between non-native and native species, particularly mean nearest taxon distance (MNTD), supporting the pre-adaptation hypothesis. This indicates that phylogenetic proximity facilitates invasion and that environmental filtering plays a key role in limiting the establishment and expansion of non-native species in mountainous regions. Notably, pre-adaptation effects remained significant even after accounting for the influence of native community diversity. In contrast, native species richness, abundance, and phylogenetic diversity all significantly suppressed non-native invasions, supporting the biotic resistance hypothesis. The analysis of relative factors importance revealed that MNTD was the most important predictor of invasion severity, followed by elevation and anthropogenic disturbance, with native community attributes showing relatively lower importance. 

Conclusion: These findings support the view that the resistance of mountain ecosystems to non-native plant invasions in the Qionglai Mountain Range is primarily driven by environmental filtering, with native community attributes showing comparatively lower importance. However, the effectiveness of environmental filtering can be weakened by human disturbance and elevational gradients. Our study highlights the need to prioritize the management of non-native species closely related to native flora and to implement conservation measures such as maintaining native community diversity and limiting anthropogenic disturbance to safeguard the invasion resistance of mountain ecosystems.

Key words: non-native species, mountain ecosystem, phylogenetic distance, biotic resistance