1
The urban heat island (UHI) is the most studied
of the climate effects of settlements. The
UHI refers to the generally warm urban temperatures
compared to those over surrounding,
non-urban, areas. It is important, however,
to distinguish between the ‘types’ of UHI
(for example, one defined by surface or air
temperatures) as the observations and responsible
processes will differ.
Surface heat islands have been detected
using satellite and aerial imagery. At
sufficient heights, the urban surface that is
’seen’ consists of contributions from roofs,
streets, car-parks, etc. At lower heights, observations
from an oblique viewpoint will contain
contributions from walls and the UHI assessment
will depend on how representative these
observations are. However, most UHI studies
examine air temperatures in urban areas. If
these are observed in Urban Boundary Layer
(above the average height of the buildings,
Figure 1), the sampled air has interacted with
the rough ‘surface’ below. More commonly,
UHI studies have focussed on air temperatures
within the Urban Canopy Layer, below
the roof tops in the spaces between buildings
(Figure 1). It is this UHI and the responsible
processes that are discussed here.
The Canopy Layer UHI
The UHI is typically presented as a temperature
difference between the air within the UCL
and that measured in a rural area outside the
settlement (∆Tu-r). Research strategies have
examined the temporal and spatial characteristics
of ∆Tu-r by using observations at fixed
sites (representing urban and rural locations)
and measurements made on mobile platforms
The Urban Canopy Layer Heat Island
IAUC Teaching Resources
(using cars and bicycles) [15]. In both cases,
the selection of sites and routes is critical to
establishing the form and behaviour of the
UHI. Moreover, the choice of the non-urban,
rural, sites is crucial.
Description: Every settlement is capable of
generating an UHI, regardless of its size. Observations
for a host of UHI studies display
common characteristics (Figure 2). ∆Tu-r reveals
itself as a pool of warm air with largest
values closest to the urban centre. At the urban
edges, temperature changes are rapid,
thereafter, ∆Tu-r increases more slowly. However,
in the vicinity of green parks lower temperatures
are observed (Figure 3). The
strength of the UHI is referred to by the maximum
difference recorded. The magnitude of
∆Tu-r is greatest at night, under clear skies and
with little wind. Under such conditions, surface
cooling is associated with radiation exchange.
While exposed rural sites cool rapidly after unset,
urban sites cool more slowly. The difference
between urban and rural sites grows with
time after sunset and reaches a maximum difference
after about 4 hours (Figure 4). The
maximum ∆Tu-r value recorded is usually
found in the centre of the settlement is generally
larger for bigger settlements.
Airflow
UBL
Figure 2: Mean minimum temperature for
November 1981 in Mexico City. Redrawn
from [4].
Figure 1: A vertically exaggerated crosssection
of the urban atmosphere and its
two main layers. The slope of the UBL is
between 1:100 and 1:200 in reality.
3
3
4
6
7 5
8
9
8
3 4
5
7
4
5
3
2
3
4
2
2
1
0
0
1
234
2400
2400
2400
2400
2800
3200
2400
0 8Km
Parks
2
hydrology) []. In warmer climates, the UHI,
particularly when accompanied by other urban
effects (such as poor air quality) may produce
stressful conditions [7].
The UHI has relevance for the study of
regional and global climates that rely on accurate
assessments of climate. The global network
of stations is unevenly distributed; most
are located in the developed world, over land,
and near urban areas. As settlements have
grown in extent, stations located nearby may
have been affected. Thus, the urban temperature
’effect’ needs to be removed so that
global trends can be examined [1].
History: The study of the UHI encapsulates
the history of the field of urban climatology,
which since the late 19th
century can be divided
into periods characterised by different
research approaches. The earlier period was
dominated by a descriptive approach that began
with Luke Howard’s pioneering examination
of London’s climate [2]. This type of research
continues and there is now a large
body of data on UHI characteristics from cities
globally.
From the late 1960’s onwards, research
shifted toward an understanding of the processes
that produce urban effects through the
application of micrometeorological theory. The
explicit recognition of the scales of urban climate
(Figure 1) has become an essential part
of research design, which is now characterised
by the measurement and modelling of energy,
mass and momentum fluxes [9]. The UHI
is treated as a response to changes in surface
geometry and materials and in atmospheric
composition. Concurrently, an ‘experimental’
approach has developed that isolates elements
of urban form whose unique effects can
be explored. For example, the climate of city
streets is examined by considering the properties
of symmetrical ‘canyons’ characterised by
their length, building height (H) and street
width (W) and orientation [10]. This approach
has permitted the discovery of general relationships
linking street geometry and climate
effect.
The formation of the UHI can now be
understood from the viewpoint of the energy
balance of urban area.
The UHI energy balance: The surface energy
balance accounts for all the exchanges of energy.
These include radiative fluxes (both
shortwave (K) and longwave (L)), turbulent
sensible (QH) and latent (QE) fluxes with the
atmosphere and energy stored or withdrawn
from the substrate (∆QS). An additional term
Relevance: The UHI has a significant and unintended
impact on the climate experienced in
cities. However, whether it is desirable or not
will depend on the background climate.
Warmer night-time temperatures will require
less (more) domestic heating (cooling) in cold
(warm) climates. In colder climates, fewer
snowfall and frost events will occur and surface
snow will melt earlier (altering the surface
20
30
40
5
0
50
50
60
60
70
80
70
Figure 3: Isotherms in Chapultec Park
on December 3, 1970 (5:28 to 6:48),
with clear sky and calm air. Redrawn
from [5].
Figure 4: Typical temporal variation of urban
and rural air temperature under clear
skies and weak airflow. The UHI is produced
by the difference between the cooling
rates. Redrawn from [13].
TIME (h)
HEATISLAND
INTENSITY
AIRTEMPERATURE
(a)
(b)
Urban
Rural
2 C
0
12 18 24 06 12
Tu - r
3
that must be considered in the urban setting is
the heat added by human activities (QF). However,
with few exceptions, QF is a small component
of the urban energy balance.
Q* + QF = QH + QE + ∆QS
Q* = K↓ - K↑ + L↓ - L↑ = K* + L*
Net radiation (Q*) is composed of radiation
arriving at (↓) or exiting from (↑) a surface
each component of which can be sub-divided
into radiative sources and sinks. For example,
incoming radiation could be divided into that
derived from the sky and that from the surrounding
terrain,
L↓ = L↓ sky + L↓terrain
Each energy balance term will be altered in
the urban environment and contribute to the
formation of the UHI (Table 1). The lower albedo
of many urban materials results in
greater absorption of solar radiation during the
daytime (Table 2). Multiple reflections within
the canopy layer lowers the albedo of the urban
‘surface’ (when viewed above the rooftops)
still further. Lower average wind speed
within the UCL suppresses turbulent exchanges
while impervious, unvegetated surfaces
hinder evaporative cooling. However, it
is important to recognize that the physical geography
of individual settlements may not fit
this pattern. For example, urban areas in desert
areas may be wetted through irrigation.
Much can be learned by examining ‘ideal’ meteorological
conditions for UHI formation. ∆Tu-r
is greatest at night, under clear and calm
skies. In these circumstances, the energy balance
can be approximated as,
L* = ∆QS
This suggests that the UHI can be examined
as the result of differential surface cooling
governed by rates of radiative exchange and
of heat storage.
Energy Balance term Urban features Urban effect
Increased K* Canyon geometry Increased surface area and multiple reflection
Increased L↓sky Air pollution Greater absorption and re-emission
Decreased L* Canyon geometry Reduced sky view factor
QF Buildings & traffic Direct addition of heat
Increased ∆QS Construction materials Increased thermal admittance
Decreased QE Construction materials Increased water-proofing
Decreased (QH+QE) Canyon geometry Reduced wind speed
Table 1: Suggested causes of canopy layer Urban Heat Island [12]
Although urban materials are known to have a
high thermal admittance (ability to store and
release heat), there is little evidence that urban
areas are distinguished by their thermal
properties [2]. As typical urban materials are
impermeable, their moisture content and thermal
properties do not vary greatly (Table 2).
Outdoor materials (e.g. asphalt road and parking
lots) are thick and in contact with a solid
substrate. However, building materials are selected
for their strength, and are formed into a
relatively thin envelope that separates indoor
Table 2: Properties of materials
Thermal properties of selected materials: Density
(ρ) in kg m-3
x103
; heat capacity (C) in J m-
3
K-1
x 106
; thermal conductivity (k) in W m-1
K-
1
and; thermal admittance (µ) in J m-2
s-1/2
K-1
.
Compiled from [13].
Radiative properties of selected materials: Albedo
(α) and emissivity (ε) are nondimensional.
Compiled from [13].
Material ρ C k µ
Dry clay soil 1.60 1.42 0.25 600
Saturated clay soil 2.00 3.10 1.58 2210
Asphalt 2.11 1.94 0.75 1205
Dense Concrete 2.40 2.11 1.51 1785
Surface α ε
Asphalt 0.05-0.20 0.95
Concrete 0.10-0.35 0.71-0.91
Urban areas 0.10-0.27 0.85-0.96
Soils: wet to Dry 0.05-0.40 0.98-0.90
Grass: long to short 0.16-0.26 0.90-0.95
4
and outdoor air. Much of the diurnal heat exchange
is confined to this layer, which has a
large specific heat capacity but a limited volume.
As a result, the replacement of natural
cover by urban materials is not a sufficient explanation
for the formation of the UHI.
Under clear night-time skies, the rate of surface
cooling is driven by net longwave radiation
loss. The magnitude of this loss is proportional
to its exposure to the sky, which may be
measured as the proportion of the viewing
hemisphere that is occupied by sky; this is the
sky view factor (ψsky). Thus, the three dimensional
structure of the urban area (its geometry)
is a good measure of night-time radiative
loss (Figure 5). Oke [11] has shown that under
these ideal conditions the maximum observed
UHI is found in the central, densest part the
settlement and that it is strongly correlated
with ψsky, which can be approximated as the
ratio of building height to street width (H/W).
This formulation is suitable for ideal conditions
and city centres, where the surfaces are dry
and materials can be assumed to be largely
the same.
Measuring the urban effect: Ideally, the urban
effect would be assessed from a continuous
set of observations that begin prior to urban
settlement [5]. Over a stable climatic period,
the unique contribution of the urban area
could be extracted. However, most UHI studies
are based on comparisons between observations
made at existing ‘urban’ and ’rural’
sites. This places a great onus on the selection
of these sites.
Figure 4. A depiction of the sky view factor
in a symmetrical street canyon, described
by its width (W) and its height
(H), ψsky = cos β. [12]
ψsky
ββ
W
H
Lowry [6] identifies three components in a set
of measurements (M):
1. The ‘background’ climate (C),
2. the effects of the local climate (L) and
3. the effects of local urbanization (E).
Values of M are observed, but the contributions
of C, L and E are not. Moreover, these
are likely to vary with time depending on the
frequency and duration of weather types experienced.
The situation is further complicated
Figure 5. Relation between the maximum
heat island intensity (∆Tu-r(max)) and the
height to width ratio (H/W) of the street
canyons in the centres of the settlement
(Based on observations for 31 cities in N.
America (•), Europe (○) and Australasia
(+)). Redrawn from [11]. The relationship
can be expressed as
∆Tu-r(max) = 7.45 + 3.97 ln(H/W)
or
∆Tu-r(max) = 15.27—13.88 ψsky
12
10
8
6
4
2
0
0 0.5 1.0 1.5 2.0 2.5 3.0 3.5
T(C)u-r(max.)
0
H/W
Figure 6. The
urban effect is
present in the
urban area (u)
and in a surrounding
area
(u’), whose
shape and size
depends on synoptic
conditions.
In a., the shape
and size of u’ is
regulated by a
particular weather event. In b., u’ is defined
by the sum of all weather events, that is,
the climate of the area. Redrawn from [6].
Airflow
u u’r
r
u
u’
b.
a.
5
in that we cannot, at the outset of a UHI study,
identify the spatial limits of the urban effect
(Figure 6). In the absence of pre-urban observations,
the urban effect may be only be estimated.
Observational studies of the UHI must
give careful consideration to the selection of
measurement sites and to the ‘background’
climate that is likely to enhance or diminish
∆Tu-r and the spatial extent of the urban effect.
Applied Climatology: The UHI is an unintended
outcome of urbanization. Climateconscious
urban design seeks to modify climate
processes to achieve a particular end.
For the UHI, a starting point would be a consideration
of the energy balance to decide
which term(s) could be readily altered (Table
1). Figure 5 indicates that the maximum potential
value for ∆Tu-r can be managed by
modifying the geometry of the UCL (that is,
the building heights and street widths). However,
opportunities for changing the physical
form of existing settlements are rare.
Another approach is to select surface materials
with desirable properties. Increasing the
surface albedo will reduce the amount of heat
stored during the daytime and limit the surface-air
sensible heat flux [14]. Increasing the
vegetated area of the city will have a similar
effect, as much of the available energy at the
surface will be expended as latent (rather than
sensible) heat energy. In addition, the leafy
canopy will shade surfaces during the daytime.
However, at night this canopy will limit
radiative heat loss from the surface and could
contribute to the night-time UHI.
Selected Readings
1. Changnon S.A. 1992: Inadvertent weather
modification in urban areas: Lessons for
global climate change. Bulletin of the American
Meteorological Society 73, 619-627.
2. Goward, S.N. 1981. Thermal behavior of
urban landscapes and the urban heat island.
Physical Geography 2, 19-33.
3. Howard L. 1818: The climate of London /
deduced from meteorological observations.
London: W. Phillips.
4. Jauregui E. 1986: The urban climate of
Mexico city in Urban Climatology and its Applications
with special regard to Tropical Areas.
WMO 652.
5. Jauregui E. 1990/91: Influence of a large
urban park on temperature and convective
precipitation in a tropical city. Energy and
Buildings 15-16, 457-463.
6. Landsberg, H.E. and Maisel, T.N. 1972:
Micrometeorological observations in an area
of urban growth. Boundary-Layer Meteorology
2, 365-370.
7. Lowry W.P. 1977: Empirical estimation of
urban effects on climate: A problem analysis.
Journal of Applied Meteorology 16, 129-135.
8. Matzarakis A. and Mayer H. 1991: The extreme
heat wave in Athens in July 1987 from
the point of view of human biometeorology.
Atmospheric Environment 25B, 203-211.
9. Myrup, L.O. 1969: A numerical model of the
urban heat island, Journal of Applied Meteorology
8, 908-918.
10. Nunez M. and Oke T.R. 1977: The energy
balance of an urban canyon. Journal of Applied
Meteorology 16, 11-19.
11. Oke, T.R. 1981: Canyon geometry and the
nocturnal urban heat island: Comparison of
scale model and field observations. International
Journal of Climatology 1, 237-254.
12. Oke, T.R. 1982: The energetic basis of the
urban heat island. Quarterly Journal of the
Royal Meteorological Society 108, 1-24.
13. Oke, T.R. 1987: Boundary Layer Climates.
2nd edition. Routledge.
14. Rosenfeld A.H., Akbari H., Bretz S.,
Fishman B.L., Kurn D.M., Sailor D. and Taha
H. 1995: Mitigation of urban heat islands: materials,
utility programs, updates. Energy and
Buildings 22, 255-265.
15. Sundborg Å. 1950: Local climatological
studies of the temperature conditions in an urban
area. Tellus 2, 221-231.
Figure 7. The relationship between the
surface-air temperature difference (∆T)
and the albedo of selected paints and
roofing materials facing the sun. Redrawn
from [14].