Urban impacts on climate - The Global Climate System: Patterns, Processes, and Teleconnections - page 237

geometry, and the prevailing wind profile. These conditions create stagnation
and trapping of air pollutants.
It is the cross-canyon airflows that make complex changes to the prevailing
wind structure. Depending on the angle of the cross-wind flow and the wind
speed, vortices within the canyon develop, which are partially controlled but also
independent of the prevailing winds. The prevailing wind will strike the opposite
(windward) wall first, deflect down toward the street, and then move up the
opposite (leeward) wall. Thus the flow at the floor of the canyon becomes
directly opposite that of the prevailing wind. Wind speed on the leeward wall
is about 40% of that on the windward wall. Johnson and Hunter ( 1999 ) and
Arnfield and Mills ( 1994a ) suggest that vortices can occur when the height
to width ratio (H/W ) of the urban canyon exceeds 0.4, in wind speeds as low
as 1-2m s 1 . Below a H/W of 0.4, any interaction with the wind between
adjacent buildings tends to be restricted to isolated roughness or wake inter-
ference influences. When the prevailing wind direction is at an angle (for
example 458) to the canyon direction, spiraling vortices can develop and extend
down the length of the canyon, in a combination of cross-canyon and parallel-
canyon airflow. When canyons are very deep (H/W > 1.0), two vortices can
develop: an upper vortex that is controlled by the ambient airflow; and near-
surface circulation that is opposite in direction (Arnfield and Mills 1994a ).
The wind direction and speed, the turbulence associated with the vortices,
Ys, the types of building and surface materials, and the exchange of air
through the canyon top all influence the components of the energy balance.
There are major differences in effect between individual cities and between
individual canyons. The incoming solar energy, its absorption and reflection
through the canyon, and interactions with atmospheric and artificially released
water vapor influence the exchanges of heat and water vapor between the air
and the canyon surfaces (Arnfield and Mills 1994a ). The wind assists this
process by generating turbulence and mixing. Air temperatures in the canyon
may therefore show little variation, as compared to sunlit versus shaded sur-
face temperatures.
Arnfield and Mills ( 1994b ) review several studies that evaluate the energy
balance of urban canyons, and the effects of wind. In general, daytime Q* mainly
goes into sensible heat flux, which can then be transferred by turbulence and
convection out through the canyon top. There is some storage of Q* in the
surface material, but, depending on surface material, very little is used for latent
heat of evaporation. At night, Q* is mainly negative, and is balanced by heat
release from the canyon surfaces. The advective impact on Q* can be either
positive or negative, depending on the time of day and the wind speed. The
variations in radiative heat in different parts of the canyon, reflected in the SVF
and the angle of the sun, mean that there is variation in the heat balance in
different parts of the canyon at different times of day.