Enhanced assessment of the eReefs biogeochemical model for the Great Barrier Reef using the Concept/State/Process/System model evaluation framework

• Published in: Environmental modelling & software : with environment data news Vol. 129; p. 104707
• Main Authors: Robson, B.J., Skerratt, J., Baird, M.E., Davies, C., Herzfeld, M., Jones, E.M., Mongin, M., Richardson, A.J., Rizwi, F., Wild-Allen, K., Steven, A.D.L.
• Format: Journal Article
• Language: English
• Published: Oxford Elsevier Ltd 01.07.2020; Elsevier Science Ltd
• Subjects: Assessment; Biogeochemistry; Computer simulation; Data interpretation; Ecosystem model; Emergent properties; Evaluation; Great barrier reef; Hydrodynamic model; Mathematical models; Model evaluation; Numerical models; State variable; World Heritage Areas; Great Barrier Reef; Emergent properties; Model evaluation; Hydrodynamic model; Ecosystem model; Great barrier reef; Assessment
• ISSN: 1364-8152; 1873-6726
• DOI: 10.1016/j.envsoft.2020.104707

The Great Barrier Reef is a UNESCO World Heritage Area that has been assessed as having a very poor outlook and needing urgent intervention. The eReefs hydrodynamic-biogeochemical models are used to complement monitoring, facilitate data interpretation and support policy decisions. Management and policy for the Great Barrier Reef are politically contentious, so a high standard of model evaluation is required. We report the application of a recently-developed four-level “CSPS” (Concept/State/Process/System) model evaluation framework to the eReefs biogeochemical model. The framework considers: Level 0, conceptual evaluation; Level 1, simulated state variables; Level 2, process rates; and Level 3, system-level patterns and emergent properties. This paper is the first complete application of this model assessment framework. It highlights both strengths and weaknesses of the eReefs biogeochemical model that might not have been apparent from a traditional model evaluation. The framework can be applied to evaluation of any complex, process-based numerical model. •In politically-contentious contexts, models require a high standard of assessment.•We demonstrate the four-level (Concept/State/Process/System) assessment framework.•The performance of the eReefs marine models is assessed.•The approach highlights strengths and weaknesses of the models.•Emergent properties are particularly useful to enhance model assessment.