The water modeling is something I am familiar with, being a hydraulic engineer. Larry's comments on the second post apply to hydraulic modeling, too. It is hard to scale down the fluid properties such as surface tension, viscosity, etc. to match the smaller model. A full size model is best.
Water is non compressible, unlike air. This must be considered when evaluating the results of hydraulic modeling to determine airflow. Personally, if my life depended on getting accurate results, I would not use water to model airflow.