Originally Posted by FPOBN
While one can scale the model, you cannot scale the air.
The element of 'scale' is represented by the Reynolds number. For some tests it is important, for many it is not. NASA has or had a 'variable density' tunnel that permitted variation of Re independent of tunnel speed, but its draw-back is a relatively high turbulence level.
I have not seen results that match what has been measured in-flight.
What an amazing statement. You can't have looked very far.
I would also point to wake turbulence modelling.
Wind tunnels are not suitable for that. They are however quite suitable and accurate for studying/measuring many other airflow phenomena, in particular for measuring the lift force generated by an airfoil or wing.
P.S.
Did you know that wind tunnel tests are not limited to airplanes but include items like cars, trucks, buildings, bridges, cyclists, skiers, golf balls, jet engines etc. ?