Changes and drivers of bacterioplankton communities within plain river networks during the rainy season (high inflow event): simulation of the water level using the MIKE11 model.

Plain river networks are becoming increasingly sensitive and vulnerable under the combined effects of climate change and landscape alteration. Understanding adaptive changes in bacterioplankton communities in response to the environment is important for improving the health of river networks. Still, little is known about the changes in bacterioplankton communities and their drivers during the rainy season. By applying co-occurrence networks, source-sink analyses, variance partitioning, and the MIKE11 model, we investigated changes in the composition, interaction, and source of the bacterioplankton community at 32 sampling sites along an anthropogenic pressure gradient (urban, suburban, and agricultural area) in the Taihu Lake plain river networks before and after the Plum rain (East Asian rainy season), as well as the drivers (water chemistry, land use, and hydrological factors) of this change. The results showed that the Chao1 diversity index and Shannon diversity index (richness) decreased significantly, and the inverse Simpson's diversity index (evenness) increased significantly after the plum rain, especially in the suburban area, indicating a decrease in the contribution of rare taxa. Related to the reduction of inputs by rare taxa after the plum rain, network analyses revealed that bacterioplankton communities were no longer tightly linked. The analysis demonstrated that water chemistry explained these changes in the cyanobacterial communities better than the hydrological indicators as evidenced by the substantially higher explanatory power of simulated data (pre-plum rain: 86%; post-plum rain: 50%) compared to measured data (pre-plum rain: 36%; post-plum rain: 22%). The use of models to obtain hydrological data was feasible when the integrity of the data analysis was ensured and the analytical process was optimized. IMPORTANCE: Unlike natural rivers, the microbial response mechanisms in human-dominated plain river networks remain poorly understood. These densely populated aquatic ecosystems, under intense anthropogenic influence, face dual pressures from climate change and landscape alteration. Their fragile hydrodynamics increase ecological vulnerability during the rainy season. Focusing on the Plum rain season, this study reveals that after the rainfall, community diversity significantly decreases, the contribution of rare taxa reduces, and bacterioplankton communities were no longer tightly linked. Combining MIKE11 modeling with multi-source environmental data, the study clarified the decisive role of water chemistry parameters in Cyanobacteria community shifts and validated the applicability of model data in ecological process research. These findings advance understanding of microbial adaptation in human-disturbed river networks and support ecosystem health assessment. The model-data fusion analysis method established in this study provides a technical framework for similar water ecosystem research.