How do materials properties are considered in k-epsilon turbulent model?

Hi everyone,

I have a doubt about how material properties are considered in k-epsilon turbulent models. I'm simulating air flow around buildings and I specified materials for the ground and the buildings facades (concrete). I've realized that the wind speed on the top of a building in the middle of a very dense region of interest is exactly the same as the velocity measured in the same height from the ground in the meteorological station. I have three questions:

1 - If surface roughness may reduce air velocity, how does it doesn't change in the middle of the region of interest? Even if it the measure point is on the top of buildings, shouldn't the top surfaces of buildings influence on velocity air flow?

2 - I didn't find in the turbulent model equation any variable related to materials properties. So how does COMSOL consider their influence on air flow?

3 - I read a discussion in this forum about the parameter B of turbulent model (www.comsol.com/community/forums/general/thread/33544) and I was wondering if I should change this to represent the roughness height of this urban context.

Thank you very much for your help.

Marilia