To investigate the response mechanisms of microbial communities to different hydrological periods in pump station forebays of tributaries in the Qinhuai River Basin, water and sediment samples were collected during the wet season, normal season, and dry season. Based on high-throughput sequencing technology of 16S rRNA and 18S rRNA genes, methods including co-occurrence network, redundancy analysis, neutral community models, β-nearest taxon index, and partial least squares path modeling were used to systematically analyze the diversity and assembly processes of microbial communities in different hydrological periods. The results indicated that prokaryotic microbial community richness was highest during the wet season, while eukaryotic microbial community richness was significantly higher in both the wet season and normal season compared to the dry season. The microbial community network structure was more complex and stable during the wet season, whereas species interactions were closer during the dry season. Prokaryotic microbial community structure was primarily driven by temperature and dissolved oxygen