Abstract
Multiple wire-sonde observations (Fig. 1) were conducted in the urban area of Hiroshima City (Photo 1) on fine days in July of 1999 and 1991. In the present paper, vertical structure of wind velocity and urban heat island are documented, and the formative process of a daytime urban “heat dome” is discussed on the basis of observation data. Special attention has been paid to the discrimination between the urban canopy layer and urban boundary layer from the point of view of a vertical wind profile, and on the behavior of the heat island in both layers.
Average wind velocity profiles during the daytime along a through-urban section (DJM-HRU-NBR-HKS-GON), which is parallel to the direction of the prevailing sea breeze, were evaluated from wind data up to 50m (Fig. 2). A logarithmic low was applied to these wind profiles to detect the transformation of vertical wind velocity distribution in the urban area from the regression coefficients of estimated equations. Regression of wind profile to the logarithmic equation was carried out by the least square method above and below the height at which inclination of wind velocity profile varies (Fig, 3).
Wind velocity profiles of the upper and lower parts are regarded as that of the lower urban boundary layer and urban canopy layer, respectively (Fig. 4). The height of zero-plain displacement of upper profile (lower urban boundary layer) or the intersection of upper and lower profiles is considered to be a dividing height between urban boundary layer and urban canopy layer. The height of the urban canopy layer indicates that its maximum is in or slightly leeward of the urban core (NBR or HKS) and is approximately 0.7-0.8 times as high as that of typical buildings in the urban area (Fig. 5-a). The momentum of the intruded sea breeze is consumed by the friction against the roughness on the top of the urban canopy layer (Figs. 5-b, c), which could cause a decrease in wind velocity in the lower urban boundary layer and hence a readjustment of the vertical wind profile above the urban area.
Vertical sections of temperature distribution up to 80m along the through-urban section (DJM-HRU-NCR-GON) and the along-river section (DJM-SMY-MTM-GON) are shown in Figs. 6-a, b. During the nighttime, when a weak land breeze blows, a vertically isothermal condition is observed along both sections, and it is slightly warmer in the coastal area than in the inland area. In contrast, during the daytime (sea breeze duration) in the through-urban section, the air temperature at NBR is higher than that at comparable heights at other sites, and the isothermal lines are raised in the central urban area like a dome. The urban heat dome is most intensified between noon and 4 p. m. (Fig. 8), In the along-river section, however, warm air is observed only in the near surface level on the leeward side of the urban area. Decrease of wind velocity in the urban canopy layer (Fig. 7) should contribute to the appearance of a daytime urban heat dome. The temperature difference between NBR and its windward site HRU is larger above 15-20m than near surface level in the early afternoon (Fig. 10). This fact indicates that the daytime thermal effect of the urban area appears more clearly above the urban canopy layer than in the layer. Moreover, diurnal variation of heat island intensity in the lower urban boundary layer differs from that in the urban canopy layer (Fig. 9). Thus, it should be noted that the development of heat islands in the urban canopy layer and the lower urban boundary layer is controlled by varying factors.
During the daytime, the urban air mass is heated by surfaces whose surface temperature is much higher than the air temperature as a result of absorption of solar radiation. Rooftop surfaces are an elevated heat source which is peculiar to urban areas, and become particularly high temperature compared to other urban surfaces (Fig. 11-a).
