Dew is assumed to be less in a city compared to its countryside. The validity of this assertion has hardly been tested, yet it is often invoked to explain the observed nocturnal excess of atmospheric moisture content in cities. The amount and distribution of dew is governed by weather and site properties which exhibit great spatial variability in urban environments. This study seeks to measure and model the surface accumulation of moisture in typical suburban residential settings. Field observations of dew in the Vancouver region were undertaken on building lots and in urban parks. Moisture accumulation on natural and built surfaces was measured using blotting, electronic wetness sensors, and micro-lysimetry techniques. Results show the importance of weather and horizon screening in determining both the amount of water deposited and its spatial distribution.
Given the difficulty of measuring dew and the inevitable complications introduced by the variability of real house lots, the notion of using a scale model emerged. The model design was predicated on the assumption that dewfall is closely related to nocturnal surface temperature which in turn is largely controlled by radiative geometry and substrate thermal properties, with turbulence acting as a lesser control. Radiative similtude was preserved through geometric scaling, and thermal similtude was provided by scaling the external geometry but retaining the structure and thermal properties of the walls and roof of full-scale houses. In this manner it was possible to simulate the spatial distribution of both surface temperature and dew over surfaces typically found in a residential lot (roofs, walls, grass, trees, paved area). The model was constructed at 1/8th scale, and run out-of-doors during fine summer weather. First order validation against measured surface moisture accumulation was achieved.