This paper focuses on urban environmental system identification. Wind-flow modeling is a challenging task because of simplifications of complex processes as well as geometry simplifications. Even a more detailed model may not be accurate because of uncertainties in parameter values. Measurement data can be employed in order to improve our understanding of the system. This paper underlines the challenges related to identification of input-parameter values of physics-based models and proposes a system-identification approach that is appropriate for wind-flow modeling. The approach has been tested with the case study of an experimental facility in the Future Cities Laboratory in Singapore, called the 'BubbleZERO'. © 2013 IEEE.