CFD simulations of running aerodynamics: Impact of computational parameters

Running is a fundamental discipline in athletics, yet its aerodynamic characteristics have not yet been intensively studied, particularly from a computational perspective. In recent years, Computational Fluid Dynamics (CFD) has become an increasingly valuable tool for advancing research in sports aerodynamics. However, the reliability of CFD predictions depends strongly on the selection of computational parameters which remains insufficiently explored in the context of human running. This paper presents a detailed study on the impact of grid resolution, computational domain size, and turbulence modelling on the computed drag area for a full-scale female runner manikin. The CFD simulations are validated by comparison with wind tunnel measurements performed in a geometrically matched test section. The sensitivity analysis provides practical guidelines for generating grids that balance accuracy and computational economy. The blockage ratio (BR) is found to be a critical parameter: values exceeding 3.5% result in drag overestimations larger than 2.8%. Among the turbulence models tested, transition-sensitive models (γ–SST and T–SST) in pseudo-transient RANS formulation and the hybrid scale-adaptive simulation (SAS) approach showed the best agreement with experimental results. Based on these findings, the study proposes a set of best-practice guidelines for reliable and cost-effective CFD simulations of running aerodynamics.