Deterministic processes shape bacterial community assembly in a karst river across dry and wet seasons.

Karst rivers are particularly vulnerable to bacterial pollution because immigrations are easily diffused from the surrounding environments due to their strong hydraulic connectivity. However, the assembly mechanism in shaping riverine bacterial biogeography is still poorly understood, especially for an ecosystem in the karst area. Here, 16S rRNA genes were used to explore the spatiotemporal and biogeographical patterns of bacterial communities from the Chishui River in the dry and wet seasons, and explore the impact of external immigration on the assembly of water bacterial communities. Our results showed clear spatiotemporal patterns of bacterial communities with a more pronounced seasonal rather than spatial fluctuation, which appeared to be dependent on seasonal-related environmental factors (e.g., temperature and turbidity). The bacterial communities exhibited a significant ( < 0.05) distance-decay pattern in both seasons, and they had a stronger distance-decay relationship in the dry season than in the wet season. However, most of the biomarkers of different external immigrations did not show significant ( > 0.05) distance-decay patterns along the Chishui river, implying that the biomarkers could be used as indicators of external immigration (e.g., OTU_125 and OTU_536). Also, the tributaries were the main external immigration (20.44-83.68%) for the Chishui River, while other terrestrial immigration (e.g., livestock, the soil of the cropland, brewing wastewater treatment plant, and sewages) showed relatively little influence, which could be due to the hydrodynamic conditions (e.g., fragile rock-soil system and hydrological structure) of the karst river. Additionally, the assembly of water bacterial communities in the Chishui river was governed by more determinism (50.7-85.7%) than stochasticity (14.3-49.3%) in both the dry and wet seasons. We demonstrated that the bacterial community's substantial variations are largely shaped by deterministic processes, thereby providing a better understanding of spatiotemporal patterns and mechanisms of the bacterial community in karst river waters.