Much of our current understanding of the physical processes that contribute to urban climate is derived from field studies conducted in real cities using a range of ground-based and airborne measurement techniques. Insight is also gained from wind tunnel modeling because it allows isolation and simplification of climatic processes through control of the impinging flow and the properties of surface structures. Data from both approaches are used to construct, evaluate, and validate numerical models which further aid our understanding of urban climate processes. However, the inherent complexity of surface morphology and energetic exchanges of real-world urban environments still poses challenges to numerical modeling of urban environments. Outdoor physical scale modeling is a potentially powerful compromise between wind tunnel modeling and full-scale observation because it incorporates the experimental control of physical and numerical modeling but preserves some of the real complexities associated with natural environmental forcing (atmospheric turbulence and radiation loading).

Here we describe the project design and preliminary results of an open-air physical scale model experiment to investigate three-dimensional (i.e., complete) facet surface temperatures within an idealized urban array. In addition to providing a detailed analysis of the variation in observed facet temperatures, the resulting dataset can be used in the evaluation of radiative emissions and a sensor placement model. The array is constructed on a rooftop at Arizona State University in Tempe, Arizona and observations were conducted from November 2006 – January 2007, a period characterized by clear skies, modest precipitation and low atmospheric humidity. These conditions provided a robust regime of daytime heating-nighttime cooling of surfaces representative of thermal patterns observed at many full-scale arid urban sites.

The model assembly consists of scaled “buildings” constructed of hollow concrete masonry blocks with solid capping slabs. Experimental configurations of the 13 x 13 m ‘building' array included: three different canyon aspect ratios (1.25, 0.625, and 0.417), three different ‘roof' albedo values and different ‘roof' angles and orientations. Surface temperatures of scaled ‘wall', ‘roof', and ‘road' facets are continuously monitored by infrared radiation thermocouples, to estimate the ‘complete' surface temperature of the array. Supplementary measurements are performed with an infrared camera positioned at different locations outside the array domain and thermal images are batch processed in IDRISI.

The basic performance of the model and its ability to achieve physical and thermal similarity of the diurnal thermal behavior of real urban settings is discussed. Differences in the observed complete surface temperature due to the geometric and roof surface configurations is demonstrated and the magnitude of directional variations in observed surface temperatures arising from the three-dimensional nature of the scaled array is compared to the effective thermal anisotropy observed in full-scale studies.