We consider the development of harmful algal blooms (HABs) in a lake with uncertain nutrients inflow. To quantify the impact of this uncertainty on predictions of the concentrations of various algae groups, we explore two alternative approaches based on the Fokker–Planck equation and PDF methods. Both approaches quantify predictive uncertainty by deriving deterministic equations for joint probability density functions of the algae concentrations. As an example, we study the impact of uncertain initial concentration and inflow–outflow ratio on the evolution of cyanobacteria (the blue-green algae).