Turbulent flows over rough walls

Fluid flows over surfaces that are rough (like the atmospheric flow over the earth's surface, the water flow over a barnacled ship's hull, or oil flow through industrial pipework) might be expected to differ from those over smooth surfaces. It has for decades been believed that all the roughness does is to effect a change (usually an increase) in the surface drag, without changing significantly the structure of the flow away from the surface. There is increasing evidence that this is not always true and, clearly, for sufficiently large surface protruberances it is intuitively most unlikely. There has never been a concerted effort to explore under what precise circumstances this 'classical' view is inadequate and what are then the distinguishing differences between the rough-surface and smooth-surface flow, for both external flows (boundary layers) and internal flows (pipes & channels). We aim to do just that and thus to explore (i) how processes affected by details of the turbulence (e.g. scalar dispersion or heat transfer) might be influenced and (ii) how the usual modelling of surface effects in prediction techniques becomes inadequate and how it might be improved.