From deterministic to stochastic: Reservoir operations alter phytoplankton assembly processes by mediating nutrient...

Reservoir operations create complex environmental heterogeneity by altering hydrodynamic and nutrient regimes. However, their influence on the underlying processes governing phytoplankton community assembly remains inadequately understood. Here, we integrated long-term field monitoring data (2015-2023) with three-dimensional hydrodynamic-transport simulations to investigate how reservoir regulation drives phytoplankton succession and alters assembly processes in Zipingpu Reservoir, Sichuan, China. We observed a fundamental shift in the dominant phytoplankton group from Dinophyta (Peridiniopsis spp.) to green algae (Tetraselmis spp.), with the relative biomass of green algae surging from <10% to 96%. Reservoir operations reduced the water residence time (WRT) by 30% and increased the surface water total phosphorus (TP) concentration by approximately 170%, thereby significantly modulating key environmental filters. Concomitantly, the explanatory power (R²) of the neutral community model increased from 21.2% to 60.8%, and the modified stochasticity ratio increased from 43% to 52%. This quantitatively demonstrates a shift in community assembly from deterministic to stochastic dominance. Structural equation modeling further delineated a key pathway: operations → WRT → TP → phytoplankton. Our findings demonstrate that reservoir operations alter assembly mechanisms by alleviating phosphorus limitation, thereby weakening deterministic selection and enhancing stochastic processes. This study provides a mechanistic foundation for optimizing reservoir operations to manage phytoplankton communities.