Adaptive Algal Cultivation Enabled by a Monthly Biomass Forecasting System

Microalgae offer a promising pathway for sustainable biofuel and bioproduct development, but outdoor cultivation is highly sensitive to environmental variability. To address this, the authors present an experimental monthly biomass forecasting system designed to guide operational decisions such as strain selection and pond water depth. Using the Biomass Assessment Tool (BAT) coupled with two forecasting approaches, a climatology-based method using data from Phase 2 of the North American Land Data Assimilation System (NLDAS-2) and three models from the North American Multi-Model Ensemble (NMME), the authors evaluated biomass production strategies for two high-performing algal strains (Picochlorum celeri and Tetraselmis striata) across four pond depths (15-30 cm) from 2020 to 2024 in Arizona. One of the NMME models achieved the highest selection accuracy, correctly identifying the optimal strain and pond depth in 84% of the months, with model accuracies across the NMME suite ranging from 74% to 84%. In comparison, the NLDAS-2 climatology-based approach achieved a 78% accuracy. Strain selection was consistently more accurate than pond depth selection across all methods, with one NMME model and the NMME multi-model ensemble achieving up to 92% accuracy in strain prediction. Simulation results show that forecast-informed approaches increased average biomass yields by 15% over the current State-of-Technology strategy, with gains exceeding 40% in certain months. These results highlight the potential of forecast-guided strategies to enhance biomass production and enable more adaptive, weather-resilient microalgae cultivation. The system is scalable to additional strains and geographic regions, offering a flexible tool for advancing sustainable algal production under increasingly variable environmental conditions.