BENMO|Simulation: A computationally efficient and biologically enhanced model for nutrient dynamics in coastal bays.

Area of Science:

• Environmental modeling
• Coastal ecosystem dynamics
• Water quality management

Background:

• Coastal bay nutrient simulations face challenges with high computational costs and simplified multi-trophic organism processes.
• Existing models often struggle to balance accuracy and efficiency in complex estuarine environments.

Purpose of the Study:

• To develop an efficient and accurate simulation model for nutrient dynamics in coastal bays.
• To integrate multi-trophic organism effects and improve computational performance for water quality management.

Main Methods:

• Decoupled hydrodynamics from biogeochemistry and incorporated a multi-trophic organism module using the Dynamic Energy Budget (DEB) model.
• Developed a graph-neural-network zoning algorithm to reduce grid cells and used a pre-computed matrix for efficient water exchange surrogating.
• Applied the Bay Estuary Nutrient Management Optimization (BENMO) framework to Sansha Bay, China.

Main Results:

• Achieved Nash-Sutcliffe efficiency (NSE) > 0.80 for NO3- and PO43- across most zones.
• Demonstrated comparable or lower RMSE values than Delft3D while reducing simulation time from ~4 hours to 12 minutes.
• The model successfully simulated nutrient dynamics with high accuracy and computational efficiency.

Conclusions:

• The BENMO|Simulation model offers a significant improvement in simulating nutrient dynamics for coastal bays.
• This tool enables rapid exploration of nutrient abatement strategies, aiding policy-making for water quality.
• The integration of efficiency and accuracy provides a powerful new approach for coastal nutrient management.