Temperature diurnal change of walls and the effect on modeling urban thermal anisotropy

Voogt [J. A. Voogt, “Assessment of an Urban Sensor View Model for thermal anisotropy,” Remote Sensing of Environment, vol. 112, pp. 482-495, 2008.] assessed the surface-sensor-sun relations model (SUM) of thermal anisotropy. The results of that assessment suggested that there is a general tendency to underestimate overall thermal anisotropy. This paper focuses on the effects of wall direction on urban thermal anisotropy. Field experiments were conducted in Beijing. One experiment used several fixed infrared thermometers to record the diurnal temperature change on the south, north, east, and west walls as well as the roof of a building. Another experiment observed directional radiance and building component temperature using an urban scale model with different geometrical viewing conditions. Two component-dividing patterns were then employed to simulate urban thermal anisotropy using SUM model: Pattern I does not consider wall direction, whereas Pattern II divides walls into south-, north-, east-, and west-facing walls. The results show that there are obvious temperature differences among the differently facing walls. Pattern I has a tendency to underestimate the directional temperature. While Pattern II can rectify this underestimation and can simulate urban thermal anisotropy with improved accuracy. The above findings underscore the importance of wall direction in urban thermal anisotropy, especially for areas with sparse buildings, and, to some extent, explain the main reason of underestimation suggested by Voogt.