Flood disturbances are key drivers in river ecosystems, yet their specific impacts on the assembly mechanisms of planktonic eukaryotic microbial communities remain unclear. Four representative sites were established along the Laoguan River in Nanyang City, southwestern Henan Province, to represent gradients of human disturbance. Surface water samples were collected during pre- and post-flood periods across two consecutive years. Environmental variables, including hydrological and physicochemical parameters such as nutrients, conductivity, and chlorophyll a, were measured to evaluate their influence on community patterns. Microbial DNA was extracted from filtered water samples, and the V4 region of the 18S rRNA gene was sequenced on the Illumina MiSeq platform. After quality control, sequences were clustered into operational taxonomic units (OTUs) at 97% similarity and normalized to ensure comparability. Based on relative abundance thresholds, planktonic eukaryotic microorganisms were classified into dominant and rare ecological groups, enabling investigation of community assembly mechanisms under flood disturbance. Our findings indicate that flood disturbances, through changes in hydrological conditions and physicochemical parameters, induced phase transitions in community assembly mechanisms. After the disturbance, stochastic processes initially dominated community assembly, as indicated by a significant increase in the normalized stochasticity ratio (NST). However, environmental filtering gradually intensified and progressively steered the community toward deterministic processes. These transitions were accompanied by a marked reduction in alpha diversity (α-diversity) immediately after flooding, reflecting the strong disruption of community stability. Subsequent recovery during pre-flood periods highlighted the resilience of planktonic eukaryotic microorganisms. The temporal sequence of these processes underscores the complexity of microbial responses to natural disturbances, suggesting that community assembly is not fixed but highly dynamic. The response strategies of taxa with varying abundance patterns showed notable differences. Dominant taxa managed to maintain their ecological roles by responding consistently to environmental factors, which explained up to 70% of the variation. This was particularly clear in their steady reactions to changes in conductivity and nitrate-nitrogen (NO^-₃-N) levels, indicating strong environmental filtering and stable adaptation strategies. Conditionally rare taxa (CRT) were more sensitive to nutrient variations, making them reliable indicators of environmental variation. Their temporal heterogeneity suggested a potential role in signaling ecological shifts. In contrast, absolutely rare taxa (ART) demonstrated high stochasticity (NST values reaching 100%) and substantial unexplained variation (up to 90%), reflecting the unpredictability in their community dynamics. The weak environmental responses of ART suggest that micro-scale processes such as dormancy, microbial interactions, and viral lysis may underlie their assembly, emphasizing their contribution to maintaining community plasticity under disturbance regimes. Together, these results provide the first comprehensive evidence that flood disturbances trigger stage-dependent shifts in microbial community diversity and assembly processes, offering novel insights into how river ecosystems respond and adapt to hydrological perturbations.