Inter-model comparison of simulated Gulf of Mexico hypoxia in response to reduced nutrient loads: Effects of phytoplankton and organic matter parameterization

• Published in: Environmental modelling & software : with environment data news Vol. 151; pp. 1 - 14
• Main Authors: Jarvis, Brandon M., Pauer, James J., Melendez, Wilson, Wan, Yongshan, Lehrter, John C., Lowe, Lisa L., Simmons, Cody W.
• Format: Journal Article
• Language: English
• Published: Oxford Elsevier Ltd 01-05-2022; Elsevier Science Ltd
• Subjects: Dissolved oxygen; Ecosystem models; Gulf of Mexico; Hydrodynamics; Hypoxia; Inter-model comparison; Nutrient load reduction; Nutrient loading; Nutrients; Organic matter; Parameterization; Phytoplankton; Primary production and respiration; Reduction; Shelves; Simulation; Gulf of Mexico; Nutrient load reduction; Hypoxia; Gulf of Mexico; Inter-model comparison; Primary production and respiration
• ISSN: 1364-8152; 1873-6726
• DOI: 10.1016/j.envsoft.2022.105365

Complex simulation models are a valuable tool to inform nutrient management decisions aimed at reducing hypoxia in the northern Gulf of Mexico, yet simulated hypoxia response to reduced nutrients varies greatly between models. We compared two biogeochemical models driven by the same hydrodynamics, the Coastal Generalized Ecosystem Model (CGEM) and Gulf of Mexico Dissolved Oxygen Model (GoMDOM), to investigate how they differ in simulating hypoxia and their response to reduced nutrients. Different phytoplankton nutrient kinetics produced 2–3 times more hypoxic area and volume on the western shelf in CGEM compared to GoMDOM. Reductions in hypoxic area were greatest in the western shelf, comprising 72% (∼4,200 km2) of the total shelfwide hypoxia response. The range of hypoxia responses from multiple models suggests a 60% load reduction may result in a 33% reduction in hypoxic area, leaving an annual hypoxic area of ∼9,000 km2 based on the latest 5-yr average (13,928 km2).
•Complex model formulations significantly altered outcomes of hypoxia simulations.•Hypoxic area and volume varied spatially across the shelf between two models.•Hypoxia responded differently to reduced nutrient load scenarios between models.•Models must be evaluated in context with multiple models to inform decision making.