J. A. Voogt, T. R. Oke
Theoretical and Applied Climatology, Volume 60, Numbers 1-4, September 1998, Pages: 199 – 217
Abstract
Building walls form an important component of the total surface area in cities, but are not generally considered in most attempts to measure urban surface temperatures. A procedure which allows sampling of the spatial and temporal variation of wall surface temperatures in urban areas is presented. Radiometric surface temperatures are monitored using an array of infrared thermometers mounted on a moving vehicle. It is necessary to remove observations which view sky, or mixed sky and building scenes. Distribution truncation and distribution modelling are investigated as methods to remove the unwanted observations. Use of the traverse method in several urban areas reveals strong temporal variations in wall temperatures due to solar loading. Significant spatial variations in temperatures suggest caution be used in the extrapolation of data from single canyon studies to larger scales. Times of strong surface temperature contrast between different walls are linked to canyon and solar geometry. In the light industrial and residential observation areas, north-south streets show mid-morning and late afternoon peaks of approximately equal magnitude in the difference between wall surface temperatures. East-west streets show a single peak in the hour following solar noon. The downtown study area, with streets oriented northeast-southwest and northwest-southeast, shows a double wave pattern with one peak dominant. These patterns are of interest because they can lead to large biases in remotely-observed surface temperature with view direction.