Development of a hybrid algal population prediction (HAPP) model by algae growth potential estimation and time series regression and its application in one reservoir in China.

Algal blooms in eutrophic lake and reservoir water sources deteriorate the water quality and severely compromise water supply safety. Predicting algal population dynamics is essential for developing effective water source control and management strategies. This study proposes a hybrid prediction model for algal population by algae growth potential (AGP) estimation and time series regression. A time lag is incorporated into the regression due to the delayed outbreak of algal blooms. This model was applied to predict the algal population in one reservoir-type water source in northern China and demonstrated commendable simulation performance. The model accurately simulated algal growth trends, with a Nash-Sutcliffe efficiency (NSE) coefficient of approximately 0.58 on the dataset from 2018 to 2023. The temporal heterogeneity of parameters was systematically analyzed, leading to implementation of time-specific parameter optimization that significantly enhanced model performance. This study further analyzed the model application scenarios and the relationships between algal dynamics and environmental variables, offering novel insights for the algal bloom management in the context of climate change.