Important Parameters for a Predictive Model of ks for Zero Pressure Gradient Flows

To predict drag on a rough surface under turbulent flow conditions, practitioners rely on roughness correlations that map topographical features of the surface to the equivalent sand-grain roughness k(s). However, details of the data that underpin these empirical correlations are not always immediately evident for comparison and discussion. Therefore, here we compile a table of roughness correlations with unified notation, in chronological order, listing the parameter ranges and the roughness types used in their development, noting idiosyncrasies. Overall, the table shows that tested roughness types have increased in generality from regular roughness features to random surface elevations, and that the independent parameters of primary importance measure size (e.g., height), frontal area (e.g., slope), and coverage (e.g., skewness). In addition to the need for more data to populate the parameter space, outstanding questions facing practitioners are filtering and sampling of multiscale surfaces, and the treatment of heterogeneous surfaces, with answers appearing on the horizon.