Geoscience Reference
In-Depth Information
mainly as a result of geothermal pavement heating
and hot water pipelines.
city heat excess compared with adjacent rural
areas. Differences are most evident during still air
conditions, especially at night under a regional
inversion (
Figure 12.27
). For the heat island effect
to operate effectively there must be wind speeds
of less than 5-6m s
-1
. It is especially apparent on
calm nights during summer and early autumn,
when it has steep cliff-like margins on the upwind
edge of the city and the highest temperatures are
associated with the highest density of urban
dwellings. In the absence of regional winds, a well-
developed heat island may generate its own
inward local wind circulation at the surface. Thus
the thermal contrasts of a city, like many of its
climatic features, depend on its topographic
situation and are greatest for sheltered sites with
light winds. The fact that urban-rural temperature
differences are greatest for London in summer,
when direct heat combustion and atmospheric
pollution are at a minimum, indicates that heat
loss from buildings by radiation is the most
important single factor contributing to the heat
island effect. Seasonal differences are not neces-
sarily the same, however, in other macroclimatic
zones.
The effects on minimum temperatures are
especially marked. For central Moscow, winter
extremes below -28°C occurred only 11 times
during 1950-1989 compared with 23 cases at
Nemchinovka west of the city. Cologne, Germany
has an average of 34 percent fewer days with
minima below 0°C than its surrounding area. In
London, Kew has an average of 72 more days with
frost-free screen temperatures than rural Wisley.
Precipitation characteristics are also affected;
incidences of rural snowfall are often associated
with either sleet or rain in the city center.
Although it is difficult to isolate changes in
temperatures that are due to urban effects from
those due to other climatic factors (see Chapter
13), it has been suggested that city growth is often
accompanied by an increase in mean annual
temperature. At Osaka, Japan, temperatures
have risen by 2.6
Heat islands
The net effect of urban thermal processes is to
make city temperatures in mid-latitudes generally
higher than in the surrounding rural areas. This
is most pronounced after sunset during calm, clear
weather, when cooling rates in the rural areas
greatly exceed those in the urban areas. The energy
balance differences that cause this effect depend
on the radiation geometry and thermal properties
of the surface. It is believed that the canyon
geometry effect dominates in the urban canopy
layer, whereas the sensible heat input from urban
surfaces determines the boundary layer heating.
The urban heat island intensity is mainly a
function of the ratio of canyon height to width
(H/W), though it increases also with an increasing
difference in thermal inertia of the urban and rural
surfaces, and downward infrared radiation from
pollution layers. By day the urban boundary layer
is heated by increased absorption of shortwave
radiation due to the pollution, as well as by
sensible heat transferred from below and
entrained by turbulence from above.
The
heat island
effect may result in minimum
urban temperatures being 5-6°C greater than
those of the surrounding countryside. These
differences may reach 6-8°C in the early hours
of calm, clear nights in large cities, when the
heat stored by urban surfaces during the day
(augmented by combustion heating) is released.
Because this is a
relative
phenomenon, the heat
island effect also depends on the rate of rural
cooling, which is influenced by the magnitude of
the regional environmental lapse rate.
For the period 1931-1960, the center of
London had a mean annual temperature of
11.0
C for the suburbs,
and 9.6°C for the surrounding countryside.
Calculations for London in the 1950s indicated
that domestic fuel consumption gave rise to a
0.6
°
C, compared with 10.3
°
C warming in the city in winter and that
accounted for one-third to one-half of the average
°
C in the past 100 years. Under
calm conditions, the maximum difference in
°
