Ten iterative steps for model development and evaluation applied to Computational Fluid Dynamics for Environmental Fluid Mechanics

Computational Fluid Dynamics (CFD) is increasingly used to study a wide variety of complex Environmental Fluid Mechanics (EFM) processes, such as water flow and turbulent mixing of contaminants in rivers and estuaries and wind flow and air pollution dispersion in urban areas. However, the accuracy and reliability of CFD modeling and the correct use of CFD results can easily be compromised. In 2006, Jakeman et al. set out ten iterative steps of good disciplined model practice to develop purposeful, credible models from data and a priori knowledge, in consort with end-users, with every stage open to critical review and revision (Jakeman et al., 2006). This paper discusses the application of the ten-steps approach to CFD for EFM in three parts. In the first part, the existing best practice guidelines for CFD applications in this area are reviewed and positioned in the ten-steps framework. The second and third part present a retrospective analysis of two case studies in the light of the ten-steps approach: (1) contaminant dispersion due to transverse turbulent mixing in a shallow water flow and (2) coupled urban wind flow and indoor natural ventilation of the Amsterdam ArenA football stadium. It is shown that the existing best practice guidelines for CFD mainly focus on the last steps in the ten-steps framework. The reasons for this focus are outlined and the value of the additional - preceding - steps is discussed. The retrospective analysis of the case studies indicates that the ten-steps approach is very well applicable to CFD for EFM and that it provides a comprehensive framework that encompasses and extends the existing best practice guidelines.