Urban Heat Islands


Types of UHIs and Their Characteristics..................................................265

Causes of the UHI: The Urban Energy Budget.........................................269

Causes of the UHI: Characteristics of the Urban Surface.......................269

James A. Voogt

University of Western Ontario

Weather and Climate Influences................................................................270

Heat Island Impacts......................................................................................270



The urban heat island (UHI) represents the relative warmth of the air, surface, and subsurface in urban areas compared to their nonurbanized surroundings. It arises from the consequences of urban development on the surface and atmospheric characteristics in urban areas and represents an unintentional change to the climate of cities. The magnitude (or intensity) of an UHI is defined by the difference between urban and nonurban temperatures.

Types of UHIs and Their Characteristics

The relative warmth of the urban atmosphere, surface, and substrate materials leads to several distinct heat islands (Figure 31.1, Table 31.1).lwl The urban canopy-layer (UCL) heat island is the best-known heat island.131 It describes the warmth of a near-surface air layer that extends from the ground to the mean height of the buildings and vegetation. It is the most commonly measured layer of the atmosphere in cities for air temperatures because it is easily accessed by ground-based instruments that are either fixed or mobile (e.g., on vehicles). Canopy-layer heat islands are typically the largest at night (Figure 31.2) during weather conditions in which winds are calm and skies are clear. Heat island intensities under such conditions are typically a few degrees Celsius in large parts of most cities and may exceed 10°C for the most densely developed parts of a large city. They are smaller and sometimes may even slightly negative (representing an urban “cool island”) during clear daytime conditions; together, these indicate a strong temporal variability of the heat island over the course of a day under calm, clear weather conditions.!31 When averaged over all weather conditions, the heat island magnitude in the UCL is typically 1°C-3°C.

The spatial structure of the heat island shows a pattern of isotherms—isolines of equal temperature— that follow the border of the city with a relatively tight spacing when winds are calm, and hence the topographic analogy with an island (Figure 31.2). The spatial temperature gradient, the change of temperature over space, is typically large near the edge of the “island,” forming the so-called cliff in response to the large relative change in surface characteristics in the region of rural-to-urban transition. Within the urban area, temperatures in the UCL are strongly controlled by the local characteristics of the urban surface. They may show substantial spatial variability (on the order of several °C) when weather conditions are favorable. The highest nighttime temperatures, the peak of the heat island, are

Schematic diagram showing the different heat island types and their locations within an urban area

FIGURE 31.1 Schematic diagram showing the different heat island types and their locations within an urban area. Shading represents the affected volume, but it does not show the expected spatial variations, (a) Under fair weather conditions, within the first l-2km of the atmosphere, known as the planetary boundary layer (PBL), a boundary-layer heat island exists with a characteristic plume shape extending in the downwind direction, (b) Within the UCL air volume, a canopy-layer heat island exists (warmer air temperatures). The surface heat island consists of all surfaces, but usually only a subset of these surfaces can be seen from any one observation point. Below the surface, a subsurface heat island exists. (Adapted from Oke 1997.)

usually associated with the area of most intense urban development, and some warmer air is often transported horizontally downwind of the city.

Above the UCL, the urban boundary-layer heat island represents an urban-scale warming through the depth of the urban boundary layer (up to 1-2 km during daytime with clear skies and a few 10s to 100s of meters at night) that has a smaller magnitude and is much less spatially and temporally variable than that of the underlying UCL heat island. The heat island magnitude here is positive both day and night. The warmed boundary-layer air above the city is often transported downwind by the mean wind leading to a plume of warmer air above downwind non-rural areas (Figure 31.1). Measurements of the urban boundary-layer heat island are relatively rare as access to this layer is difficult. Thermometers must be mounted on tall towers, balloons, or aircraft; or temperatures can be observed by remote sensing techniques using ground-based instruments (Table 31.1).

The surface UHI represents the relative warmth of the surfaces in cities compared to that in rural areas. Surface temperatures can be observed using remote sensing techniques from satellite or aircraft- mounted instruments. These instruments provide a spatially continuous view of the urban area but tend to be biased toward viewing the highest upward-facing unobstructed surfaces such as roofs, open spaces, and the tops of trees.141 The implications of this method of observing the heat island must be borne in mind when interpreting remotely sensed images of the surface UHI. Under clear skies and light winds, daytime surface temperatures seen from above are much warmer than air temperatures in the canopy layer and show much more spatial variability (Figure 31.2 and Table 31.1) due to the juxtaposition of surfaces with contrasting properties in urban areas (e.g., hot dry rooftops adjacent to an irrigated lawn). However, the variability is not always easily observed—for example, when the sensors have spatial resolutions that are substantially larger than the scale of the surface structure that provides the temperature variability. This can be the case when using satellite observations of surface temperature. The daytime surface heat island is positive (Figure 31.2) and is controlled by the amounts of impervious

TABLE 31.1 Heat Island Types and Their Spatial and Temporal Characteristics

Heat Island Type



How Measured

Spatial Characteristics

Temporal Characteristics

Canopy layer heat island



Fixed or mobile (vehicle-based) measurements using thermometers.

Shows significant spatial variability associated with important elements of surface structure: building height-to- width ratio, amount of vegetation, topographic features.

Largest at night. Magnitude grows rapidly in the late afternoon and early evening. May be negative (a cool island) during the daytime. Highly sensitive to wind and cloud.

Boundary -layer heat island



Thermometers mounted on very tall towers, balloons, kites, or aircraft. Remote-sensing from ground-based instruments.

Exhibits a “domed" structure in near-calm conditions and a distinct downwind “plume” as winds increase. Boundary-layer depth is 1-1.5 km by day, but only 50-300 m at night. Heat island magnitude decreases with height in the boundary layer by night and is approximately constant by day.

Shows relatively small diurnal variation. Sensitive to wind and cloud.

Surface heat island



Remote sensing from towers, aircraft, or satellites.

Significant spatial variability associated with variations in surface characteristics, including shading, surface orientation, moisture status, thermal properties, surface reflectivity, and vegetation coverage. Variability of rural surface temperature is also large and affects heat island magnitude.

Largest during daytime in the summer season. Nighttime value is also positive and largest in summer. Highly sensitive to weather conditions. Varies with season especially if there are significant changes to moisture or vegetation characteristics or where substantial winter heating is used.

Subsurface heat island



Ground or borehole temperature measurements.

Spatial variability occurs due to variations in surface characteristics and subsurface heat sources from urban infrastructure. Urban areas show a greater depth of temperature decrease from the surface before temperatures reverse to show the geothermal gradient.

Temporal response is increasingly lagged with depth so that subsurface patterns reflect past conditions at the surface. Temporal influences below the first meter are typically only seasonal or longer.

and vegetated surfaces as well as the surface characteristics of the rural reference. At night, the overall spatial structure and magnitude of the surface heat island are similar to those of the canopy-layer heat island, reflecting the importance of surface controls (Table 31.2) on the near-surface air temperature. The spatially averaged surface heat island magnitude under favorable weather conditions is larger by day than at night,151 making it the reverse of that in the canopy-layer air. The use of simple correlations to relate surface and air temperatures and their heat islands is problematic without the use of fully coupled surface-atmosphere energy budget models that can represent the physical processes that govern the exchanges of energy between the surface and the air.

The subsurface UHI is influenced by the relative warmth of the surface and the atmosphere above it as well as by contributions of heat from the basements of buildings and subsurface infrastructure. It

Top: A plan view of the nighttime canopy-layer heat island. Bottom

FIGURE 31.2 Top: A plan view of the nighttime canopy-layer heat island. Bottom: Conceptual cross sections of the canopy layer and surface UHIs by day and night. (Oke et al. 2017, with permission.)

TABLE 31.2 Factors That Lead to the Formation of and/or Influence the Magnitude of UHIs

Heat Island Influence

Effect on UHI Magnitude

Surface geometry

UHI magnitude increases as the ratio of building height to street width increases and view of the nighttime sky is obstructed.

Surface thermal properties

UHI magnitude increases with materials that make the city a better storer of heat; those with higher heat capacity and/or thermal conductivity relative to rural materials. Variations in rural moisture can influence UHI magnitude.

Anthropogenic heat input

UHI magnitude increases as anthropogenic heat increases. This input can have large seasonal variations in some climates as well as intraurban spatial variability related to the density of development and magnitude of energy use.

City size

UHI magnitude tends to increase with city size up to a limiting amount.

Wind speed

UHI magnitude decreases rapidly as wind speed increases.

Cloud cover

UHI magnitude decreases as cloud cover increases.


UHI magnitude is typically the largest in the warm season in mid-latitudes. In high latitudes, the UHI is the largest in the winter due to anthropogenic heat input. In tropical cities with distinct wet and dry seasons, the UHI is typically largest in the dry season.

Time of day

The canopy-layer UHI is the largest at night (air temperatures).

The surface UHI magnitude is larger during the day (clear, sunny conditions).

Source: Adapted from Arnfield.161

These factors are most directly related to the canopy-layer UHI but also affect other UHI types.

too shows spatial variability, although much less than that of the surface, and it decreases with depth. It can be measured from temperature measurements deep in the soil or from boreholes or wells. Where substrate materials are water-saturated, subsurface flows may also yield a warm plume associated with the subsurface heat island.

The exact spatial configuration and behavior of each heat island type in a given city depend on the layout of the city and its topographic setting; cities in coastal or mountainous areas will be influenced by sea-valley or mountain-valley breeze systems that can significantly impact the spatial pattern of the heat islands and their temporal development.111

< Prev   CONTENTS   Source   Next >