Can someone provide recommendations for setting the wall roughness for characterizing turbulent diffusion? Does this parameter scale with the mean of the hydrodynamic roughness? With respect to EFDC is the wall roughness the same as Zo in the logarithmic vertical velocity profile?
Yes, it is similar, but not exactly the same. EFDC does not use a log profile from the wall, but it is a roughness height.
Typically we use quite small numbers for wall roughness, up to a couple of orders of magnitude smaller that bottom roughness. You would need to be simulating a flume for example before wall roughness becomes a significant factor – bathymetry is a bigger shear component.