Process-based modeling and prediction of cyanobacterial harmful algal blooms of long-range dependence on environmental conditions

Harmful algal blooms of toxin-producing cyanobacteria (CyanoHABs) increasingly threaten ecosystems, water supplies, and public health. Yet, most CyanoHAB modeling tools ignore the influence of antecedent environmental conditions on CyanoHABs. This paper couples a high-resolution hydrodynamic model of Lake Pontchartrain Estuary with a finite-element-based advection–dispersion–reaction module that embeds time-lagged effects of temperature, salinity, light, and nutrients on CyanoHABs, highlighting the key novelty of this work. Forced by satellite and in-situ observations, the coupled CyanoHAB modeling system reproduced the 2021 basin-wide CyanoHAB bloom including its abrupt collapse following Hurricane Ida, demonstrating the robustness of the coupled CyanoHAB modeling system. It was discovered that favorable antecedent environmental conditions over one to four weeks govern the growth and abundance of CyanoHABs, indicating the long-range dependence of CyanoHABs on antecedent environmental conditions. It was also discovered that extreme mixing events can rapidly suppress the biomass of CyanoHABs. The incorporation of antecedent environmental conditions and the mixing effect into process-based models markedly improves the forecasting skill and offers a robust foundation for early-warning systems as climate change intensifies CyanoHAB episodes.