Spatial Differentiation and Environment-Driven Mechanisms of Locust Community Structure in the Xinjiang Region Along the Sino-Kazakh Border

This study was conducted in the Xinjiang region, China, along the Sino-Kazakh border, an area recognized as high-risk for locust outbreaks and characterized by ongoing shifts in dominant pest species. This study systematically examined the structural characteristics of locust communities across different grassland types and identified the underlying environmental driving mechanisms. Field surveys were conducted to assess the diversity characteristics, density variations, and niche width of the locust communities across the different grassland types. The locust community in the mountain meadows had a significantly lower Shannon diversity index compared with the other grassland types (<i>p</i> < 0.05). Although the Simpson dominance index and Pielou evenness index were also the lowest in the mountain meadows, these differences were not statistically significant (<i>p</i> > 0.05). Permutational multivariate analysis of variance (PermANOVA) revealed highly significant differences in locust density among the grassland types (<i>p</i> = 0.001). Ecological niche analysis revealed stronger interspecific competition in the lowland meadow, and weaker competition in the temperate steppe-enriched deserts and temperate desert grasslands. Structural equation modeling and random forest analysis identified soil organic, plant total potassium, and soil pH as the key factors driving locust community structure across grassland types. This study clarifies the diversity patterns of locust communities in the Xinjiang region along the Sino-Kazakh border and provides empirical data to better understand locust community structure and distribution. It also offers a scientific basis for developing sustainable locust management strategies.