LOCAL CLIMATE ZONES FOR

URBAN TEMPERATURE STUDIES

I. D. Stewart anD t. r. Oke

Downloaded from http://journals.ametsoc.org/bams/article-pdf/93/12/1879/3739321/bams-d-11-00019_1.pdf by guest on 26 September 2020

by

The new “local climate zone” (LCZ) classification system provides a

research framework for urban heat island studies and standardizes
the worldwide exchange of urban temperature observations.

T
he study of urban heat islands (UHIs) implicates two of
the most serious environmental issues of the twentieth
century: population growth and climate change. This
partly explains why the worldwide stock of heat island
studies has grown so remarkably in recent decades. From
Cairo to Tokyo, London to Dallas, and Delhi to Nairobi, cit-
ies of every cultural and physical description have been the
focus of a formal heat island investigation. The global reach
of this literature reflects both the widespread repercussions
of the heat island effect in all urban areas, and the scientific
curiosity about a phenomenon so seemingly simple.
“Urban heat island,” a term first coined in the 1940s (e.g.,
Balchin and Pye 1947), refers to the atmospheric warmth of a
city compared to its countryside. Heat islands occur in almost
all urban areas, large or small, in warm climates or cold. The
traditionally described heat island is that which is measured
at standard screen height (1–2 m above ground), below the
city’s mean roof height in a thin section of the boundary
layer atmosphere called the urban canopy layer. Air in this
layer is typically warmer than that at screen height in the
countryside. The physical explanation for this is more com-
plex than generally acknowledged in the literature (Table 1).
The main causes of the heat island relate to structural and
land cover differences of urban and rural areas. Cities are
rough with buildings extending above ground level, and are
dry and impervious with construction materials extending
View of LCZ 1 in Seattle, Washington. across natural soils and vegetation. Also important is the 
Photo: I. D. Stewart
 heat and moisture release from people and their urban and rural categories has given researchers a
activities. These urban characteristics alter the simple framework to separate the effects of city and
natural surface energy and radiation balances such country on local climate (e.g., Lowry 1977).
that cities are relatively warm places (Oke 1982; Lowry However, recent research shows that through this
and Lowry 2001). popular use of urban–rural classification, the methods
The extra warmth in cities has several practical and communication in heat island literature have
implications. Whether these are considered to be suffered critically. In a review of many such studies,
positive or negative depends upon the macroclimate Stewart (2011a,b) found that more than three-quarters
of the city. In cities with a relatively cold climate, or of the observational heat island literature fails to
with a cold season, the heat island can convey benefits give quantitative metadata of site exposure or land
such as cheaper house heating costs, improved cover. Most investigators simply rely on the so-called
outdoor comfort, fewer road weather hazards such as urban and rural qualifiers to describe the local land-

Downloaded from http://journals.ametsoc.org/bams/article-pdf/93/12/1879/3739321/bams-d-11-00019_1.pdf by guest on 26 September 2020

surface icing or fog, and more benign conditions for scapes of their measurement sites. Here we develop a
plant growth and animal habitat. On the other hand, climate-based classification of urban and rural sites
heat islands in relatively hot climates or seasons can that applies universally and relatively easily to local
increase discomfort and potentially raise the threat temperature studies using screen-level observations.
of heat stress and mortality, and heighten the cost of Our aim in this classification is twofold: 1) to facilitate
air conditioning and the demand for energy. consistent documentation of site metadata and thereby
Heat islands also have climatological implications. improve the basis of intersite comparisons, and 2)
The fact that temperatures are elevated at urban to provide an objective protocol for measuring the
stations means that their use in databases to assess magnitude of the urban heat island effect in any city.
historical climate series may have “contaminated” the We do not aim to supplant the terms urban and rural
global air temperature record. The concern is whether from heat island discourse, but instead to encourage a
the presence of urban data has created a warm bias in more constrained use of these terms when describing
the time series. Bias could occur if urban stations are the local physical conditions of a field site. The terms
used in the temperature record in greater numbers urban and rural alone cannot sufficiently describe a
than is warranted by their representation as a land field site or its local surroundings.
cover type on Earth.
To fully understand these and other issues, it INADEQUACIES OF SIMPLE URBAN –
has been the preoccupation of researchers for many RURAL DIVISION. Urban is defined in standard
decades to measure the heat island effect through dictionaries as “constituting, forming, or including
simple comparisons of “urban” and “rural” air tem- a city, town…or part of such,” with town being a
peratures. The conventional approach is to gather “densely populated area…opposed to the country or
temperatures at screen height for two or more fixed suburbs,” and characterized physically as a “cluster
sites and/or from mobile temperature surveys. Sites of dwellings or buildings.” Rural, in contrast, is an
are classified as either urban or rural, and their tem- “agricultural or pastoral area . . . characteristic of
perature differences are taken to indicate the heat the country or country life,” with country being “the
island magnitude. Classifying measurement sites into parts of a region distant from cities.” From these
definitions, we interpret rural landscapes to be less
populated than cities, with fewer built structures and
AFFILIATIONS: Stewart anD O ke —Department of Geography,
more abundant natural space for agricultural use,
University of British Columbia, Vancouver, British Columbia,
Canada whereas urban landscapes have significantly more
CORRESPONDING AUTHOR: I. D. Stewart, Department of built structures and larger populations. By extension,
Geography, University of British Columbia, 1984 West Mall, suburban landscapes are those lying immediately
Vancouver, BC V6T 1Z2, Canada outside or adjacent to a town or city, and that have
E-mail: stewarti@interchange.ubc.ca natural and built-up spaces with population densities
The abstract for this article can be found in this issue, following the lower than cities but higher than the country.
table of contents. While such definitions of urban and rural may be
DOI:10.1175/BAMS-D-11-00019.1 evocative of the landscape, they are vague as objects
A supplement to this article is available online (10.1175/BAMS-D-11-00019.2)
of scientific analysis (Stewart and Oke 2006). In the
heat island literature, for example, the term urban
In final form 1 May 2012
©2012 American Meteorological Society evokes an eclectic mix of local settings from which
its observations have originated: the wooden quarters

1880 | december 2012

of old Hiroshima, Japan
Table 1. Causes of the urban heat island effect. Each cause represents an
(Shitara 1957); the parks urban modification to the surface energy and radiation balance.
and playing fields of
Pretoria, South Africa 1. Greater absorption of solar radiation due to multiple reflection and radiation
trapping by building walls and vertical surfaces in the city.
(Louw and Meyer
Greater absorption is not, as often assumed, due solely to lower albedo of urban materials.
1965); the courtyards
and stonework streets 2. Greater retention of infrared radiation in street canyons due to restricted
of London, England view of the radiatively “cold” sky hemisphere.
Sky view becomes increasingly restricted with taller and more compact buildings.
(Chandler 1965); the
skyscraper canyons of 3. Greater uptake and delayed release of heat by buildings and paved surfaces in
Dallas, Texas (Ludwig the city.
1970); the industrial Often incorrectly attributed only to the thermal properties of the materials, this effect is
also due to the solar and infrared radiation “trap” and to reduced convective losses in the

Downloaded from http://journals.ametsoc.org/bams/article-pdf/93/12/1879/3739321/bams-d-11-00019_1.pdf by guest on 26 September 2020

plants and refineries of canopy layer where airflow is retarded.
Ashdod, Israel (Sharon
and Koplowitz 1972); 4. Greater portion of absorbed solar radiation at the surface is converted to
sensible rather than latent heat forms.
t he shaded avenues
This effect is due to the replacement of moist soils and plants with paved and waterproofed
a nd law ns of New surfaces, and a resultant decline in surface evaporation.
Delhi, India (Bahl and
5. Greater release of sensible and latent heat from the combustion of fuels for
P a d m a n a b h a mu r t y urban transport, industrial processing, and domestic space heating/cooling.
1979); the school and Heat and moisture are also released from human metabolism, but this is usually a minor
c o l l e ge g r ou nd s of component of the surface energy balance.
Nairobi, Kenya (Okoola Source: Oke 1982
1980); the factories and
workshops of Cairo, Egypt (Robaa 2003); the brick site in the city center with a rural site 60 km to the
and tin shanties of Sao Paulo, Brazil (Nunes da Silva north. He defined UHI magnitude for Tokyo as the
and Ribeiro 2006); and the high-rise housing estates temperature difference between the urban and rural
of Singapore (Chow and Roth 2006). A significant sites. Despite being located 60 km from the city center,
problem in this literature, and in heat island method- the so-called rural site was still within the mixed
ology, is that the term urban has no single, objective urban–rural surroundings of metropolitan Tokyo,
meaning, and thus no climatological relevance. What in the small city of Kumagaya. This gave a curious
is described as urban in one city or region differs from portrayal of the rural landscape to some urban
that of another city (Fig. 1). The term urban is there- climatologists (Fig. 2), but one that is, nonetheless,
fore impossible to define universally for its physical understandable given the dense settlement patterns
structure, its surface properties, or its thermal climate. of the Kanto Plain. Yamashita’s remark that “the
Equally problematic is that urban and rural are whole area of the Kanto Plain is more or less urban-
becoming outmoded constructs in landscape classi- ized” correctly speaks to the difficulty of classifying
fication, for the developing world and especially Asia urban and rural landscapes in highly dispersed and
(Lin 1994; McGee and Robinson 1995; Lo and Yeung decentralized cities.
1998). In these and other densely populated regions, the
social, political, and economic space that separates cities EXISTING URBAN AND RURAL LAND-
and countrysides is no longer distinguished by a clear SCAPE CLASSIFICATIONS. We recognize that
urban–rural divide. Urban form is becoming increas- all classifications are limited in scope and function,
ingly dispersed and decentralized as traditional and and further that none of the systems we review in this
nontraditional land uses coexist, and as people, capital, section was designed to classify heat island field sites,
commodities, and information flow continuously be- and none makes that claim. Therefore what we iden-
tween city and countryside. Urban theorists now con- tify as advantageous, or restrictive, with these systems
tend that the spatial demarcation between urban and relates only to the aims of the new classification.
rural is artificial, and that the relation between city and Chandler (1965) was perhaps the first heat
country is more accurately described as a continuum, or island investigator to develop a climate-based clas-
a dynamic, rather than as a dichotomy (Gugler 1996). sification of the city. He divided Greater London
The densely populated Kanto Plain surrounding into four local regions, each distinguished by its
Tokyo is a perfect case in point. In a study of the climate, physiography, and built form. Following
Tokyo heat island, Yamashita (1990) paired an urban Chandler’s lead, Auer (1978) proposed an urban–

AMERICAN METEOROLOGICAL SOCIETY december 2012 | 1881

 rural classification for the city of St. Louis, Missouri. configuration, and construction materials of 10
He recognized 12 “meteorologically significant” U.S. cities. His was the first system to represent
land uses in St. Louis, based on the city’s vegeta- city structure and materials, initially for acid rain
tion and building characteristics. Ellefsen (1991) studies. A key feature of Ellefsen’s system is the
derived a system of 17 neighborhood-scale “urban division of building types into “attached” and
terrain zones” (UTZs) from the geometry, street “detached” forms.

Downloaded from http://journals.ametsoc.org/bams/article-pdf/93/12/1879/3739321/bams-d-11-00019_1.pdf by guest on 26 September 2020

Fig. 1. Examples of urban field sites in climate literature. Conventional methodology defines these sites as
universally “urban” despite obvious differences in building structure, land cover, and human activity: (a) modern
core of Vancouver, Canada; (b) old core of Uppsala, Sweden; (c) town center of Toyono, Japan; (d) business
district of Akure, Nigeria; (e) city airport of Phoenix, Arizona; (f) university campus of Szeged, Hungary.

1882 | december 2012

 Combining features of both Auer’s and Ellefsen’s scheme divides city terrain into seven homogenous
schemes, Oke (2004, 2008) designed a simple and regions called “urban climate zones” (UCZs), which
generic classification of city zones to improve siting range from semi-rural to intensely developed sites.
of meteorological instruments in urban areas. His The zones are distinguished by their urban structure
(building/street dimensions), cover (permeability),
fabric (materials), metabolism (human activity), and
potential to modify the natural, or “preurban,” sur-
face climate. Most recently, Loridan and Grimmond
(2011) developed “urban zones for characterizing
energy partitioning,” or UZEs. Their classes are de-
fined by threshold values for the active vegetative and
built surface fractions of cities (“active” here meaning

Downloaded from http://journals.ametsoc.org/bams/article-pdf/93/12/1879/3739321/bams-d-11-00019_1.pdf by guest on 26 September 2020

engaged in energy exchange). The classification helps
atmospheric modelers to distinguish urban areas with
respect to their partitioning of incoming radiation.
National land cover and land use classifications
often include categories for both urban and rural
environments. For example, the U.S. National Land
Cover Dataset (NLCD) divides the coterminous
United States into 16 land cover classes, 4 of which
are deemed “urban oriented” (Homer et al. 2007). In
some European countries, the “climatope” system
has traditionally been used to classify urban terrain
and urban climates, largely for planning purposes
(Wilmers 1991; Scherer et al. 1999). Climatopes derive
from local knowledge of wind, temperature, land use,
building structure, surface relief, and population den-
sity. These data are integrated across an urban area to
reveal special climates of local places, or climatopes.
Wilmers (1991) identified nine such climates for the
city of Hannover, Germany, based on vegetation,
surface structure, and land use criteria. Scherer et al.
(1999) documented many more climatopes in Basel,
Switzerland, based on ventilation and land cover
characteristics.
These previous classifications contain many fea-
tures that align with the aims of heat island observa-
tion. Their limitations, however, must be recognized.
First, not all classifications use a full set of surface
climate properties to define its classes. A complete
set consists of the physical properties of surface
structure, cover, fabric, and metabolism (Oke 2004).
Second, a system that excludes rural landscapes is
not well suited to heat island investigation, nor is one
with class names and definitions that are culture or
region specific. The classifications of Chandler, Auer,
F ig . 2. “Rural” site used by Yamashita (1990) to Ellefsen, and Oke are all predisposed to the form and
measure UHI magnitude in Tokyo (red circles indicate function of modern, developed cities, so their use in
site location). Urban and rural influences on surface
more diverse economic settings is limited. Third,
climate are seen at (top, center) micro and (center,
bottom) local scales. This overlap in landscapes and
although the climatope concept is well adapted to
spatial scales on the Kanto Plain makes the urban– most urban settings, its class names and definitions
rural dichotomy an awkward fit for site classification. vary widely with place, and thus cannot provide
Aerial photographs courtesy of Google Earth. classification systems with a means for comparison.

AMERICAN METEOROLOGICAL SOCIETY december 2012 | 1883

 CONSTRUCTING A NEW CLASSIFICATION from the system while others were added to represent
SYSTEM. In a classic paper on the logic, method, landscapes defined not by their structural or surface
and theory of classification, Grigg (1965) listed cover characteristics, but by building materials or
several criteria that a system should meet. First, it anthropogenic heat emissions. The resulting classes
should invoke a simple and logical nomenclature by were quantified by their surface properties and
which objects/areas can be named and described. A assigned simple, concise names. Throughout this
system’s nomenclature is critical to its validity and process, prospective users of the system in the inter-
acceptance. Second, a classification system should national climate community were asked for feedback
facilitate information transfer by associating objects/ on the general nature of the system, its application
areas in the real world with an organized system of to local settings, and its cultural and regional biases.
generic classes. Users can then make comparative This early exposure of the system to its target com-
statements about the members belonging to each munity resulted in substantial changes to the number,

Downloaded from http://journals.ametsoc.org/bams/article-pdf/93/12/1879/3739321/bams-d-11-00019_1.pdf by guest on 26 September 2020

class. This condition led Grigg to his third and most nature, and naming of the individual classes.
important criterion: inductive generalization. A
properly constructed classification system should Data sources. Quantitative data to characterize the
simplify the objects/areas under study, and thereafter classes by their surface properties were selected from
promote theoretical statements about their proper- the urban climate observational and numerical mod-
ties and relations. To Grigg’s criteria, we add that eling literature. Measured and estimated values of
a new classification of urban and rural field sites geometric, thermal, radiative, metabolic, and surface
should be inclusive of all regions, independent of all cover properties were gathered from urban and rural
cultures, and, for heat island assessment, quantifiable field sites worldwide. Quantitative data were also
according to class properties that are relevant to sur- retrieved from the classifications of Anderson et al.
face thermal climate at the local scale (i.e., hundreds (1976), Auer (1978), Häubi and Roth (1980), Ellefsen
of meters to several kilometers). (1990/91), Theurer (1999), and Oke (2004), and from
reviews of empirical urban climate literature (e.g.,
Classification by logical division. Scientific classification Wieringa 1993; Grimmond and Oke 1999).
is essentially a process of definition. It begins with Data to adapt the classes to the real world were
a “universe” class, which is divided into subclasses chosen from the urban design literature, which gives
(Black 1952). The basis for division at each class level qualitative attributes to urban form through expres-
is a differentiating principle, or property, of theoreti- sions of “fabric,” “texture,” and “morphology” (e.g.,
cal interest. The universe for the new classification Brunn and Williams 1983; O’Connor 1983; Vance
is “landscape,” which we define as a local-scale tract 1990; Kostof 1991; Potter and Lloyd-Evans 1998).
of land with physical and/or cultural characteristics These are the same expressions to which urban cli-
that have been shaped by physical and/or cultural matologists give quantitative attributes. This overlap
agents. The landscape universe is divided according was especially useful to assimilate regional urban
to properties that influence screen-height tempera- form into the classification system, and to balance its
ture, namely surface structure (height and spacing of temporal (old vs modern) and spatial (core vs periph-
buildings and trees) and surface cover (pervious or ery) representation. These data also give support to
impervious). Surface structure affects local climate culturally neutral definitions for each class.
through its modification of airflow, atmospheric heat
transport, and shortwave and longwave radiation bal- LOCAL CLIMATE ZONES. Hereafter, all classes
ances, while surface cover modifies the albedo, mois- to emerge from logical division of the landscape uni-
ture availability, and heating/cooling potential of the verse are called “local climate zones” (LCZs; Table 2)
ground. These properties tend to “cluster” spatially, (Stewart 2011a). The name is appropriate because
such that in locations where the building height-to- the classes are local in scale, climatic in nature, and
width ratio is large, so is the fraction of impervious zonal in representation. We formally define local
cover and the density of urban construction materials. climate zones as regions of uniform surface cover,
Dividing the landscape into these properties gen- structure, material, and human activity that span
erates dozens of prototype classes, many having clus- hundreds of meters to several kilometers in hori-
ters that are considered highly improbable or logically zontal scale. Each LCZ has a characteristic screen-
unacceptable in the real world (e.g., closely spaced height temperature regime that is most apparent
buildings on pervious cover or closely spaced trees over dry surfaces, on calm, clear nights, and in areas
on impervious cover). Such clusters were removed of simple relief. These temperature regimes persist

1884 | december 2012

Table 2. Abridged definitions for local climate zones (see electronic supplement for photographs, surface
property values, and full definitions). LCZs 1–9 correspond to Oke’s (2004) urban climate zones.

Built types Definition Land cover types Definition

1. Compact high-rise Dense mix of tall buildings to tens of A. Dense trees Heavily wooded landscape of
stories. Few or no trees. Land cover deciduous and/or evergreen trees.
mostly paved. Concrete, steel, stone, Land cover mostly pervious (low
and glass construction materials. plants). Zone function is natural
forest, tree cultivation, or urban park.

2. Compact midrise Dense mix of midrise buildings (3–9 B. Scattered trees Lightly wooded landscape of
stories). Few or no trees. Land cover deciduous and/or evergreen trees.
mostly paved. Stone, brick, tile, and Land cover mostly pervious (low

Downloaded from http://journals.ametsoc.org/bams/article-pdf/93/12/1879/3739321/bams-d-11-00019_1.pdf by guest on 26 September 2020

concrete construction materials. plants). Zone function is natural
forest, tree cultivation, or urban park.

3. Compact low-rise Dense mix of low-rise buildings (1–3 C. Bush, scrub Open arrangement of bushes, shrubs,
stories). Few or no trees. Land cover and short, woody trees. Land cover
mostly paved. Stone, brick, tile, and mostly pervious (bare soil or sand).
concrete construction materials. Zone function is natural scrubland or
agriculture.

4. Open high-rise Open arrangement of tall buildings to D. Low plants Featureless landscape of grass or
tens of stories. Abundance of pervious herbaceous plants/crops. Few or
land cover (low plants, scattered no trees. Zone function is natural
trees). Concrete, steel, stone, and grassland, agriculture, or urban park.
glass construction materials.

5. Open midrise Open arrangement of midrise buildings E. Bare rock or paved Featureless landscape of rock or
(3–9 stories). Abundance of pervious paved cover. Few or no trees or
land cover (low plants, scattered plants. Zone function is natural desert
trees). Concrete, steel, stone, and (rock) or urban transportation.
glass construction materials.

6. Open low-rise Open arrangement of low-rise buildings F. Bare soil or sand Featureless landscape of soil or sand
(1–3 stories). Abundance of pervious cover. Few or no trees or plants.
land cover (low plants, scattered trees). Zone function is natural desert or
Wood, brick, stone, tile, and concrete agriculture.
construction materials.

7. Lightweight low-rise Dense mix of single-story buildings. G. Water Large, open water bodies such as seas
Few or no trees. Land cover mostly and lakes, or small bodies such as
hard-packed. Lightweight construction rivers, reservoirs, and lagoons.
materials (e.g., wood, thatch,
corrugated metal).

8. Large low-rise Open arrangement of large low-rise VARIABLE LAND COVER PROPERTIES
buildings (1–3 stories). Few or no
trees. Land cover mostly paved. Variable or ephemeral land cover properties that change
Steel, concrete, metal, and stone significantly with synoptic weather patterns, agricultural practices,
construction materials. and/or seasonal cycles.

9. Sparsely built Sparse arrangement of small or b. bare trees Leafless deciduous trees (e.g., winter).
medium-sized buildings in a natural Increased sky view factor. Reduced
setting. Abundance of pervious land albedo.
cover (low plants, scattered trees).
s. snow cover Snow cover >10 cm in depth. Low
admittance. High albedo.
10. Heavy industry Low-rise and midrise industrial struc- d. dry ground Parched soil. Low admittance. Large
tures (towers, tanks, stacks). Few or Bowen ratio. Increased albedo.
no trees. Land cover mostly paved
or hard-packed. Metal, steel, and w. wet ground Waterlogged soil. High admittance.
concrete construction materials. Small Bowen ratio. Reduced albedo.

AMERICAN METEOROLOGICAL SOCIETY december 2012 | 1885

 Table 3. Values of geometric and surface cover properties for local climate zones. All properties are
unitless except height of roughness elements (m).
Building Impervious Pervious Height of Terrain
Local climate zone Sky view Aspect surface surface surface roughness roughness
(LCZ) factora ratiob fractionc fractiond fractione elementsf classg
LCZ 1 0.2–0.4 >2 40–60 40–60 < 10 > 25 8
Compact high-rise
LCZ 2 0.3–0.6 0.75–2 40–70 30–50 < 20 10–25 6–7
Compact midrise
LCZ 3 0.2–0.6 0.75–1.5 40–70 20–50 < 30 3–10 6
Compact low-rise

Downloaded from http://journals.ametsoc.org/bams/article-pdf/93/12/1879/3739321/bams-d-11-00019_1.pdf by guest on 26 September 2020

LCZ 4 0.5–0.7 0.75–1.25 20–40 30–40 30–40 >25 7–8
Open high-rise
LCZ 5 0.5–0.8 0.3–0.75 20–40 30–50 20–40 10–25 5–6
Open midrise
LCZ 6 0.6–0.9 0.3–0.75 20–40 20–50 30–60 3–10 5–6
Open low-rise
LCZ 7 0.2–0.5 1–2 60–90 < 20 <30 2–4 4–5
Lightweight low-rise
LCZ 8 >0.7 0.1–0.3 30–50 40–50 <20 3–10 5
Large low-rise
LCZ 9 > 0.8 0.1–0.25 10–20 < 20 60–80 3–10 5–6
Sparsely built
LCZ 10 0.6–0.9 0.2–0.5 20–30 20–40 40–50 5–15 5–6
Heavy industry
LCZ A <0.4 >1 <10 <10 >90 3–30 8
Dense trees
LCZ B 0.5–0.8 0.25–0.75 <10 <10 >90 3–15 5–6
Scattered trees
LCZ C 0.7–0.9 0.25–1.0 <10 <10 >90 <2 4–5
Bush, scrub
LCZ D >0.9 <0.1 <10 <10 >90 <1 3–4
Low plants
LCZ E >0.9 <0.1 <10 >90 <10 <0.25 1–2
Bare rock or paved
LCZ F >0.9 <0.1 <10 <10 >90 < 0.25 1–2
Bare soil or sand
LCZ G >0.9 <0.1 <10 <10 >90 – 1
Water
a
Ratio of the amount of sky hemisphere visible from ground level to that of an unobstructed hemisphere
b
Mean height-to-width ratio of street canyons (LCZs 1–7), building spacing (LCZs 8–10), and tree spacing (LCZs A–G)
c
Ratio of building plan area to total plan area (%)
d
Ratio of impervious plan area (paved, rock) to total plan area (%)
e
Ratio of pervious plan area (bare soil, vegetation, water) to total plan area (%)
f
Geometric average of building heights (LCZs 1–10) and tree/plant heights (LCZs A–F) (m)
g
Davenport et al.’s (2000) classification of effective terrain roughness (z0) for city and country landscapes. See Table 5 for class descriptions

1886 | december 2012

 year-round and are associated
Table 4. Values of thermal, radiative, and metabolic properties for local
climate zones. All values are representative of the local scale. with the homogeneous environ-
ments or ecosystems of cities
Local climate zone Surface Anthropogenic (e.g., parks, commercial cores),
(LCZ) admittancea Surface albedob heat outputc natural biomes (e.g., forests,
LCZ 1 1,500–1,800 0.10–0.20 50–300 deserts), and agricultural lands
Compact high-rise (e.g., orchards, cropped fields).
LCZ 2 1,500–2,200 0.10–0.20 <75 Each LCZ is individually named
and ordered by one (or more)
Compact midrise
distinguishing surface prop-
LCZ 3 1,200–1,800 0.10–0.20 <75
erty, which in most cases is the
Compact low-rise height/packing of roughness

Downloaded from http://journals.ametsoc.org/bams/article-pdf/93/12/1879/3739321/bams-d-11-00019_1.pdf by guest on 26 September 2020

LCZ 4 1,400–1,800 0.12–0.25 <50 objects or the dominant land
Open high-rise cover. The physical properties
LCZ 5 1,400–2,000 0.12–0.25 <25 of all zones are measurable and
nonspecific as to place or time
Open midrise
(Tables 3 and 4).
LCZ 6 1,200–1,800 0.12–0.25 <25
The landscape universe con-
Open low-rise sists of 17 standard LCZs, of
LCZ 7 800–1,500 0.15–0.35 <35 which 15 are defined by surface
Lightweight low-rise structure and cover and 2 by
LCZ 8 1,200–1,800 0.15–0.25 <50 construction materials and
anthropogenic heat emissions.
Large low-rise
The standard set is divided into
LCZ 9 1,000–1,800 0.12–0.25 <10
“built types” 1–10, and “land
Sparsely built cover types” A–G (Table 2).
LCZ 10 1,000–2,500 0.12–0.20 >300 Built types are composed of
Heavy industry constructed features on a pre-
LCZ A unknown 0.10–0.20 0 dominant land cover, which is
paved for compact zones and
Dense trees
low plants / scattered trees for
LCZ B 1,000–1,800 0.15–0.25 0
open zones. Land cover types
Scattered trees can be classified into seasonal or
LCZ C 700–1,500 0.15–0.30 0 ephemeral properties (i.e., bare
Bush, scrub trees, snow-covered ground,
LCZ D 1,200–1,600 0.15–0.25 0 dry/wet ground).
Low plants
Thermal differentiation of LCZ
LCZ E 1,200–2,500 0.15–0.30 0 classes. The logical structure of
Bare rock or paved the LCZ system is supported by
LCZ F 600–1,400 0.20–0.35 0 observational and numerical
Bare soil or sand modeling data (Stewart and Oke
LCZ G 1,500 0.02–0.10 0 2010; Stewart 2011a). Mobile
temperature observations from
Water
Uppsala, Sweden (Sundborg
a
Ability of surface to accept or release heat (J m –2 s –1/2 K–1). Varies with soil wetness and 1951; Taesler 1980); Nagano,
material density. Few estimates of local-scale admittance exist in the literature; values given Japan (Sakakibara and Matsui
here are therefore subjective and should be used cautiously. Note that the “surface” in LCZ A
is undefined and its admittance unknown. 2005); and Vancouver, Canada
b
Ratio of the amount of solar radiation reflected by a surface to the amount received by it. (T. Oke and A. Christen) were
Varies with surface color, wetness, and roughness. used to measure thermal con-
c
Mean annual heat flux density (W m−2) from fuel combustion and human activity trasts a mong LCZ classes.
(transportation, space cooling/heating, industrial processing, human metabolism). Varies During calm, clear evenings,
significantly with latitude, season, and population density.
thermal contrasts are driven

AMERICAN METEOROLOGICAL SOCIETY december 2012 | 1887

 layer. Urban energy bal-
ance zones (e.g., UZEs;
Loridan and Grimmond
2011) are therefore unlikely
to coincide exactly with
LCZs because similar flux
densities can occur above
canopy layers with dis-
tinctly different microscale
structure, land cover, and
thermal climate.

Downloaded from http://journals.ametsoc.org/bams/article-pdf/93/12/1879/3739321/bams-d-11-00019_1.pdf by guest on 26 September 2020

Communication. Definitions
and physical properties of
all LCZs are given in 17
illustrative datasheets [see
the appendix for sample
sheets 1, 6, and D; see elec-
tronic supplement online
Fig. 3. Thermal differentiation of local climate zones using temperatures from at http://dx.doi.org/10.1175
automobile traverses in Vancouver during calm, clear evenings, Mar 2010. All /BAMS-D-11-00019.2 for
measurements were made with an aspirated and insulated copper-constantan full set]. Using these sheets,
thermocouple (precision ±0.2 K) at 1.5 m above ground. Temperatures are urban climate investigators
adjusted to a standard base time of 2 h after sunset. Raw temperature data can classify landscapes and
provided by Andreas Christen (University of British Columbia). field sites with consistency
and efficiency, and link
largely by building geometry and land cover (Fig. 3). standardized site metadata to all temperature ob-
Contrasts between classes with significant differ- servations. These metadata include photographs of
ences in geometry and cover can often exceed 5 K, LCZs in different regional settings and bar charts
whereas contrasts between classes with lesser physical of zone properties relating to surface structure (sky
difference can be less than 2 K. This pattern is easily view factor, aspect ratio, roughness element height),
disrupted by the dynamical and seasonal effects of surface cover (plan fraction occupied by buildings,
surface wetness, relief, tree cover, snow cover, and vegetation, and impervious ground), surface fabric
anthropogenic heat, all of which can override or offset (thermal admittance, surface albedo), and human
the unvarying effects of building form and surface activity (anthropogenic heat output). For definitions
cover. Numerical surface and atmospheric models of these properties, refer to Tables 3 and 4, and to the
show that the diurnal temperature range in each class datasheet key in the appendix.
decreases with increasing impervious surface fraction
and height/density of buildings (Fig. 4). These simula- G U I D E L I N E S F O R U S I N G T H E LC Z
tions follow the approach of Krayenhoff et al. (2009) CLASSIFICATION SYSTEM. The LCZ system is
in an attempt to aid the assessment of LCZ thermal inherently generic and cannot capture the peculiarities
responsiveness. Computer modeling of canopy layer of every urban and rural site. Its view of the landscape
climates is not sufficiently developed to enable more universe is highly reductionist, and, like all classifica-
accurate predictions of thermal microclimates at tions, its descriptive and explanatory powers are lim-
screen level for individual urban sites. ited. LCZs represent a simple composition of buildings,
The extent to which LCZs reveal characteristic roads, plants, soils, rock, and water, each in varying
regimes for other meteorological variables in the amounts and each arranged uniformly into 17 recog-
urban canopy layer, such as wind and humidity, nizable patterns. The internal homogeneity portrayed
has yet to be explored. Atmospheric properties that by each LCZ is unlikely to be found in the real world,
blend at the top of the roughness sublayer (typically except in planned or intensely managed settings (e.g.,
more than twice the height of the canopy layer) will city parks, new housing estates). The 17 patterns should
not classify discretely into LCZs. This is true of nevertheless be familiar to users in most cities, and
surface energy balance fluxes in the urban boundary should be adaptable to the local character of most sites.

1888 | december 2012

 The LCZ classes are each physically discrete in sur- to nonurban and urban settings, respectively. Their
face structure and land cover, such that the boundaries recommendations [adopted by the World Meteoro-
separating most classes can be delineated on a city map logical Organization (WMO)], together with those
or aerial photograph. However, the thermal climate included here, lead to consistent use of the LCZ
across those boundaries is spatially continuous— system. Furthermore, they require only minimal
the screen-level air temperature of one zone blends site metadata and surface characterization, and they
gradually into that of its neighboring zones. This effect avert potential pitfalls that might compromise the
arises from advection of heat and moisture across classification of a site. [For regional examples of LCZ
areas of different surface structure, land cover, and classification, see Stewart and Oke (2009).]
human activity. As air crosses the border between
neighboring zones, it gradually adjusts to a new set Step 1: Collect site metadata. Users must collect
of internal boundary conditions, forming thermal appropriate site metadata to quantify the surface

Downloaded from http://journals.ametsoc.org/bams/article-pdf/93/12/1879/3739321/bams-d-11-00019_1.pdf by guest on 26 September 2020

transitions—not sudden breaks at zone borders. properties of the source area (as defined in step 2)
The upwind fetch required for air at screen height for a temperature sensor. This is best done by a visit
to become fully adjusted to its underlying surface is to the field sites in person to survey and assess the
typically 200–500 m, depending on surface roughness, local horizon, building geometry, land cover, surface
building geometry, and atmospheric stability condi- wetness, surface relief, traffic flow, and population
tions (Oke 1987; Wieringa 1993). Each LCZ should density [see sample template in Oke (2004)]. If a
therefore have a minimum diameter of 400–1,000 m field visit is not possible, secondary sources of site
(i.e., a radius of 200–500 m) so that the adjusted por- metadata include aerial photographs, land cover/land
tion of its internal boundary layer lies entirely within use maps, satellite images (e.g., Google Earth), and
the zone and does not overlap with surrounding published tables of property values (e.g., Davenport
LCZs of different structure
or cover. If the minimum
diameter for an LCZ is met,
it remains possible for air
aloft—the properties of
which have been modi-
fied by the surfaces of an
upwind LCZ—to be en-
trained down into the sur-
face layer of a leeward zone.

Three-step process. Users

should observe the follow-
ing steps when classify-
ing field sites into LCZs,
whether for a multiyear cli-
mate study or a short-term
field survey. The temporal
scope of a study dictates
when, or if, the seasonal
or ephemeral properties
of the sites need to be clas-
sified (e.g., soil moisture,
snow cover, tree leaf cover).
Classification of sites into
LCZs draws substantively
Fig . 4. Simulated values for diurnal temperature range in select local cli-
from Aguilar et al. (2003)
mate zones. Temperature values are representative of the surface-layer air.
a nd Oke (20 04, 20 08), All zones except low plant cover and bare soil or sand were modeled with an
whose recommendations updated version of the Town Energy Balance scheme (Masson 2000). Low
for siting instruments and plant cover and bare soil or sand were modeled with the Coupled Atmosphere–
collecting metadata apply Plant–Soil scheme (Mahrt and Pan 1984).

AMERICAN METEOROLOGICAL SOCIETY december 2012 | 1889

 terrain roughness lengths; see Table 5). These data the upwind direction of the sensor (Schmid 2002).
are usually available at little or no cost to the user. Because of temporal changes in stability, and espe-
It is important that LCZ users quantify the local- cially wind direction, the ellipses are centered upon,
scale character of their sites: anomalous microscale but oscillate around, the measurement site so that,
features of land cover, building geometry, and human over time, they form a misshapen circle. The dimen-
activity must not distract an assessment of the broader sions of the source area should be computed, ideally
surroundings. However, if the area of influence for a with a footprint model (e.g., Kljun et al. 2002; Schmid
field site is too heterogeneous to derive locally rep- 2002). The approximate size and orientation of the
resentative values from its surface properties, users area can otherwise be estimated from (a) the vectors
are advised to designate the site “unclassified” and of a wind rose for that location and time period, and
to find a more homogeneous setting for observation. (b) an empirically based “rule of thumb” that states
the source area for a screen-height temperature sensor

Downloaded from http://journals.ametsoc.org/bams/article-pdf/93/12/1879/3739321/bams-d-11-00019_1.pdf by guest on 26 September 2020

Step 2: Define the thermal source area. The thermal in a neutrally stable atmosphere extends, on average,
source area for a temperature measurement is the no more than a few hundred meters away (Tanner
total surface area “seen” by the sensor (sometimes 1963; Mizuno et al. 1991; Runnalls and Oke 2006).
called its “footprint” or “circle of influence”). In other Estimating the source area for a screen-height
words, it is the surface area from which the tempera- thermometer in an urban area is a challenging task
ture signal is derived and subsequently is carried, via that has hardly been scientifically investigated. It
turbulent transport, to the sensor. The source area can be anticipated that screen-height temperature
of standard contact thermometers extends upwind measurements in a canopy layer with compact build-
for meters to kilometers depending on instrument ings have smaller and more poorly defined source
height, surface geometry, and boundary layer wind areas than those of open urban and rural zones. The
and stability conditions (Kljun et al. 2002). sources will include upwind buildings, the walls and
The size, shape, and orientation of turbulent floor of an upwind street, and perhaps a branching
source areas evolve with time (Fig. 5). Over periods network of more distant street canyons that channel
of, say, an hour, the area is elliptical and oriented in surface air toward the measurement site.

Table 5. Davenport classification of effective terrain roughness. Correspondence with LCZs is given for
each Davenport class.
Davenport Roughness LCZ
class length, z 0 (m) Landscape description correspondence
Open water, snow-covered flat plain, featureless desert, tarmac, and
1. Sea 0.0002 E, F, G
concrete, with a free fetch of several kilometers.
Featureless landscape with no obstacles and little if any vegetation
2. Smooth 0.0005 E, F
(e.g., marsh, snow-covered or fallow open country).
Level country with low vegetation and isolated obstacles separated by
3. Open 0.03 D
50 obstacle heights (e.g., grass, tundra, airport runway).
Low crops or plant covers; moderately open country with occasional
4. Roughly open 0.10 obstacles (e.g., isolated trees, low buildings) separated by 20 obstacle 7, C, D
heights.
High crops, or crops of varying height; scattered obstacles separated
5. Rough 0.25 by 8 to 15 obstacle heights, depending on porosity (e.g., buildings, 5–10, B, C
tree belts).
Intensely cultivated landscape with large farms and forest clumps
6. Very rough 0.5 separated by 8 obstacle heights; bushland, orchards. Urban areas with 2, 3, 5, 6, 9, 10, B
low buildings interspaced by 3 to 7 building heights; no high trees.
Landscape covered with large, similar-height obstacles, separated by
7. Skimming 1.0 1 obstacle height (e.g., mature forests). Dense urban areas without 2, 4
significant building-height variation.
Landscape with irregularly distributed large obstacles (e.g., dense
8. Chaotic ≥2 urban areas with mix of low and high-rise buildings, large forest with 1, 4, A
many clearings).
Source: Davenport et al. (2000)

1890 | december 2012

Fig. 5. Hypothetical source areas and surface wind directions for a screen-height temperature sensor in prevailing

Downloaded from http://journals.ametsoc.org/bams/article-pdf/93/12/1879/3739321/bams-d-11-00019_1.pdf by guest on 26 September 2020

southeast winds. (left) Short-term (<1 h) source area. (right) Mean daily source area.

Quantifying the surface properties for field sites its nearest LCZ class should be identified by these
and source areas located on or near the border of two differences and described with the appropriate LCZ
(or more) zones is problematic. If the location of the name.
sensor can be moved, it should be placed where it Alternatively, users can create new subclasses for
samples from a single LCZ. Land cover and exposure sites that deviate from the standard set of classes
at that location should then be representative of the shown in Table 2. New subclasses represent combina-
designated LCZ: for compact built zones (e.g., LCZs tions of built types, land cover types, and land cover
1–3), “representative” implies a sheltered street canyon properties (Fig. 6). The notation for new subclasses
with paved ground; for open built zones (e.g., LCZs is LCZ Xai, where X is the higher parent class in the
4–6), it implies an exposed setting with vegetated standard set of LCZs, a is the lower parent class (if
ground, scattered trees, and nearby buildings. If the applicable) from the standard set, and i is a variable
location of the sensor cannot be moved, temperature or ephemeral land cover property (if applicable).
data retrieved from that site should be stratified Land cover properties are selected from the subset
first according to wind direction, then to LCZ. In of bare trees (b), snow cover (s), dry ground (d), and
transitional urban–rural areas, or along breaks in wet ground (w) (Table 2). Identifiers a and i may
land use, split classifications will occur. A site with each be assigned one or more classes or properties.
split classification is less ideal for heat island studies For example, a site whose building geometry is open
because changes in airflow and stability conditions midrise (LCZ 5) but whose surface cover is pre-
confuse the relation between surface form/cover and dominantly bush, scrub (LCZ C) should be notated
air temperature. It is recommended that transitional as LCZ 5C to represent the main features of its two
areas be avoided when siting meteorological instru- parent classes, the first in a higher position of the
ments (Oke 2004, 2008). LCZ order (baseline character 5) and the second in a
lower position (subscript C). By this convention, other
Step 3: Select the local climate zone. Metadata collected subclasses include open midrise with paved ground
in step 1 should lead users to the best, not necessarily (LCZ 5E ), open midrise with open low-rise (LCZ 56 ),
exact, match of their field sites with LCZ classes. open low-rise with dense, bare trees (LCZ 6Ab ), low
Metadata are unlikely to match perfectly with the sur- plants with wet ground (LCZ Dw ), and bare soil with
face property values of one LCZ class. If the measured dry ground (LCZ Fd ).
or estimated values align poorly with those in the LCZ All standard LCZ classes with extensive pervi-
datasheets, the process of selecting a best-fit class ous cover (A–D, F, 4–6, and 9) are prescribed an
becomes one of interpolation rather than straight intermediate soil moisture status, meaning that soil
matching. Users should first look to the surface cover water content is below field capacity but above wilting
fractions of the site to guide this process. If a suitable point. In this state, soil moisture is available to plants
match still cannot be found, users should acknowl- and the soil feels pliable and slightly sticky. If soils
edge this fact and highlight the main difference(s) are saturated (i.e., water content is above field capac-
between their site and its nearest equivalent LCZ. For ity, and water is easily squeezed from the soil), the
example, a site having considerably more traffic flow site should be subclassified as “wet” (with subscript
(and therefore anthropogenic heat flux) or noticeably w). Soils submerged in shallow water due to natural
taller, shorter, or more variable building heights than drainage or flood irrigation (e.g., wetlands, paddy

AMERICAN METEOROLOGICAL SOCIETY december 2012 | 1891

 fields, detention storage ponds) are also considered undermines this principal function. While subclasses
wet. If soils are parched (i.e., water content is below may enhance the physical description of a site, they
wilting point), soil moisture is unavailable to plants may not improve communication or comparison of
and the soil feels loose, powdery, or hard-crusted. that site with other studies. Subclasses are justified
The site in this case should be subclassified as “dry” when they describe sites whose secondary features
(subscript d). Finally, if a site is covered with >10 cm are thought to affect the local climate, or whose fea-
of snow, it should be subclassified as “snow cover” tures may otherwise relate to the particular aims of
(subscript s). a climate investigation.
Although subclasses add flexibility to the LCZ Despite preference for using the standard set
system, we give several caveats for their use. First, of 17 LCZs, we recognize that subclassification is
the system as proposed does not provide surface unavoidable if the ground at a site is snow-covered or
property values for subclasses, and thus their thermal extremely dry/wet, or if the deciduous trees are bare.

Downloaded from http://journals.ametsoc.org/bams/article-pdf/93/12/1879/3739321/bams-d-11-00019_1.pdf by guest on 26 September 2020

climates may not be known a priori. At best, one can When classifying a site for a short-term study (e.g.,
estimate the surface property values—and surface days, weeks, or months), the LCZ notation should
air temperatures—by weighting the known values of include variable land cover properties (i.e., subscripts
two (or more) standard classes. Second, the thermal b, s, d, and w), but only if the site differs from the
climate of an LCZ subclass is not expected to differ prescribed definitions of the standard LCZ classes
significantly from either of its two (or more) parent (Table 2). Most sites with extensive tree or pervious
classes, because maximum thermal contrasts between cover will experience significant change in land
successive zones in the standard set are typically no cover properties due to synoptic weather patterns
more than 1–2 K (Stewart and Oke 2010). Third, the (e.g., drought, monsoon), agricultural practices (e.g.,
purpose of LCZs is to ease the process of site classifi- planting, harvesting), or seasonal cycles (e.g., leaf
cation and metadata reporting for heat island investi- growth/leaf drop). If, however, the sites are classified
gators. Creating too many, or too complex, subclasses for long-term studies (e.g., years), subclassification

Fig. 6. LCZ subclasses to represent combinations of “built” and “land cover” types.

1892 | december 2012

is unnecessary because seasonal and/or ephemeral elevation, and water bodies on UHI magnitude. If such
change is either implicit in the name (as in classes “control” is not possible—and thus the dynamical
with extensive tree or pervious cover) or irrelevant effects of local relief, elevation, and/or water bodies
to the local climate (classes with no tree or pervious are at risk of being confused with the thermal effects of
cover). Long-term temperature data from sites buildings and land cover—the sites should be described
experiencing seasonal and/or ephemeral change for more than just their LCZ designations. In this way,
should be stratified into time periods that corre- the LCZ system prompts investigators to recognize
spond with different land cover properties, so that and separate urban from nonurban influences on
underlying trends in the temperatures can be found. temperature. [See Lowry (1977), Lowry and Lowry
A site covered by low plants year-round, the soils of (2001), Oke (2004), and Stewart (2011b) for discussions
which are saturated during a rainy season, will be of experimental design and heat island observation.]
notated as LCZ Dw if used in a short-term study during

Downloaded from http://journals.ametsoc.org/bams/article-pdf/93/12/1879/3739321/bams-d-11-00019_1.pdf by guest on 26 September 2020

the wet period and as LCZ D if used in a long-term Public scrutiny of heat island literature. The assessment
study of several seasons or years. of Stewart (2011b) that “poor scientific practice”
has compromised nearly one-half of the published
Updating zone designations. Correspondence between heat island literature should give pause. Several of
field sites and LCZ classes will vary with physical and the weaknesses contributing to that assessment can
cultural changes to the landscape. This is a virtual be improved upon with the new LCZ classification
certainty in any active urban area as land clearance system. While using LCZs, heat island investigators
and development occur. Updating LCZ designations must aim to move their studies to a higher level of
is crucial for all sites, particularly those used in inquiry by connecting observations and explanations
long-term temperature studies. A changing site will with underlying energetics, or with other heat-related
“progress” or “regress” through the natural, built, phenomena of the urban climate. Towards that goal,
compact, and open forms of the LCZ classification. it is their responsibility to follow the relatively simple
This allowance for landscape change ensures that the guidelines for instrument siting, exposure, metadata,
LCZ system stays relevant through time. Sites located and classification as stated in the present paper and in
on the edges of cities where urban growth and envi- the WMO (2008) Guide to Meteorological Instruments
ronmental change are rapid, or in the cores of cities and Methods of Observation. Reviewers and editors
where land redevelopment and large-scale greening have a critical role to play in this movement: they must
projects are taking place, should be surveyed and advocate and uphold a more responsible, ethical code
classified annually. For sites used in mobile or short- for gathering and reporting heat island observations.
term stationary surveys, the frequency of updates is
dictated largely by day-to-day variations in weather Climate modeling, weather forecasting, and historical
and soil moisture (e.g., rainfall, snow melt, irrigation). temperature analysis. Parameterization of the urban
canopy in numerical climate models and weather
APPLICATION OF LOCAL CLIMATE ZONES. prediction schemes can be improved with LCZs and
Defining UHI magnitude. The urban–rural temperature their surface morphological and land cover data.
difference, or UHI magnitude, is the most widely cited LCZs can provide input data for numerical climate
measure of city climate modification in the environ- models that incorporate urban canopy parameters
mental sciences. It is also the most poorly represented into their formulations to forecast temperature, wind,
(Stewart 2011a; Stewart and Oke 2006). We therefore precipitation, etc. Likewise, statistical regression
propose a new framework to extract UHI magnitudes models can use LCZ input data to predict intra-urban
from local temperature observations. With this new temperature patterns and UHI magnitudes.
framework, UHI magnitude is an LCZ temperature In global climate change studies, the classifica-
difference (e.g., ΔTLCZ 1 – LCZ D ), not an “urban–rural” tion of weather stations and observatories into LCZ
difference (ΔTu–r ). LCZ differences are more conducive classes, rather than “urban” and “rural” classes, could
to analysis, and less prone to confusion, because they lead to more accurate assessments of urban bias in the
highlight the common surface and exposure charac- climate record. Climate change researchers often use
teristics of the compared field sites, and they invite observations from many hundreds, even thousands,
physically based explanations of UHI magnitude. The of stations that they cannot or do not visit in person,
former is especially true if the temperature measure- and for which site metadata are often lacking.
ments of a heat island investigation are strategically Information on the surrounding exposure and
located to reduce the variable effects of surface relief, surface conditions of such stations is generally vague

AMERICAN METEOROLOGICAL SOCIETY december 2012 | 1893

 or incomplete. Davey and Pielke (2005) concluded matology in individual cities. A more appropriate use
this for the U.S. Historical Climate Network, and for LCZs is to build spatial databases of urban form
we contend that the situation is no better in other and cover—and the associated effects on thermal
parts of the world. Many climatologists instead use climate—for cities worldwide.
surrogate measures of surface cover and exposure in
their attempts to classify actual ground conditions CONCLUSIONS. We present local climate zones
at poorly documented sites, and especially to differ- as a comprehensive climate-based classification of
entiate between “urban” and “rural” stations. These urban and rural sites for temperature studies. The
measures—which include population data (e.g., Karl cultural and geographic appeal of the classifica-
et al. 1988), satellite-derived vegetation indices (Gallo tion has been demonstrated, as has the potential to
et al. 1993), and night-light frequencies (Hansen improve consistency and accuracy in urban climate
et al. 2001; Peterson 2003)—are not reliably linked to reporting. The system functions easily and inexpen-

Downloaded from http://journals.ametsoc.org/bams/article-pdf/93/12/1879/3739321/bams-d-11-00019_1.pdf by guest on 26 September 2020

local site conditions. This provoked a leading global sively in any city or region. We therefore anticipate
change scientist (Peterson 2003, p. 2957) to say, “As a that it can meet a basic requirement in urban climate
community, we need to update our understanding of studies through standardized description of surface
urban heat islands, to realize that this phenomenon structure and cover; meaningful definition and
is more complex than widely believed by those not intercity comparison of UHI magnitude (ΔTLCZ X – Y);
immersed in the field.” guided exploration of heat island causes and controls;
Reconstructing a climate station’s LCZ history clear communication of site metadata; and inter-
through archival techniques could reveal surface disciplinary transfer of urban climate knowledge.
inf luences on long-term temperature trends that Our primary motive behind the system, however, is
might otherwise have remained unknown (e.g., to enhance the description of surface conditions in
Runnalls and Oke 2006). At sites where historical urban and rural areas, and thereby ease the process
temperature records are unavailable, one should of site selection and metadata reporting.
instead consult, or construct, an LCZ history of the
area to infer temperature trends back into the past. ACKNOWLEDGMENTS. A wide network of col-
leagues has given us valuable input on the LCZ system.
Architecture, city planning, and landscape ecology. To We were hosted in many cities while visiting field sites
date, the integration of urban climate knowledge and gathering site metadata. To the hosts of these visits—
with city planning has not been especially useful or Anthony Brazel, Phoenix (AZ); Roger Taesler, Uppsala
successful, in part because urban climatology has (Sweden); Janos Unger, Szeged (Hungary); Reinhard Böhm,
advanced slowly around issues of scale and com- Vienna (Austria); Mariusz Szymanowski, Wroclaw (Poland);
munication (Oke 1984, 2006; Mills 2006). The Krzysztof Fortuniak, Lodz (Poland); Shuji Yamashita, Tokyo;
LCZ system could advance these issues because it Yasushi Sakakibara, Nagano (Japan); Kiyotaka Sakaida,
offers a basic package of urban climate principles Sendai (Japan); Yeon-Hee Kim, Seoul; and Janet Nichol,
for architects, planners, ecologists, and engineers. Hong Kong—we offer sincere thanks for their important
The system conveys these principles through spa- contributions to this research. We gratefully acknowledge
tial scales (micro, local) and design elements (e.g., Gerald Mills (University College Dublin), Andreas Christen
building height, “green” cover ratio) that are relevant (University of British Columbia), Matthias Roth (National
to the many cognate disciplines of urban climatology. University of Singapore), and Ahmed Balogun (Federal
The LCZ system could also support well-established University of Technology, Akure, Nigeria) for engaging the
planning projects such as “climatope maps” (Scherer urban climate community with the LCZ system, through
et al. 1999) and “urban climatic maps” (Ren et al. class projects, graduate theses, and international confer-
2011). To help quantify the thermal and morpho- ences. Michael Church (University of British Columbia) is
logical layers of an urban climatic map (or UCMap), acknowledged for sharing his philosophical views on the
standardized metadata for urban structure, cover, LCZ system, and for offering helpful comments on early
fabric, and metabolism can be extracted from the versions of this manuscript. We owe special recognition
LCZ datasheets and adapted to a specific area. LCZs to Scott Krayenhoff (University of British Columbia) for
should not, however, be used for climatic mapping developing and running the numerical models to simulate
alone because, unlike climatopes, they are developed LCZ temperatures. Finally, we acknowledge the Natural
from generalized knowledge of built forms and land Sciences and Engineering Research Council of Canada for
cover types that are universally recognized, not from their financial support to IS through a Doctoral Fellowship
specialized knowledge of local topography and cli- and to TO through a Discovery Grant.

1894 | december 2012

 er look at United States and global surface tempera-
REFERENCES ture change. J. Geophys. Res., 106, 23 947–23 963.
Aguilar, E., I. Auer, M. Brunet, T. C. Peterson, and Häubi, F., and U. Roth, 1980: Wechselwirkung zwischen
J. Wieringa, 2003: Guidance on metadata and ho- der Siedlungsstruktur und Wärmeversorgungssyste-
mogenization. WMO Tech. Doc. 1186, 53 pp. men (Interaction between settlement structure and
Anderson, J. R., E.E. Hardy, J. T. Roach, and R. E. Witmer, heating supply systems). Forschungsprojekt BMBau
1976: A land use and land cover classification system RS II 4-704102-77.10, 270 pp.
for use with remote sensor data. Geological Survey Homer, C., and Coauthors, 2007: Completion of the 2001
Professional Paper 964, 28 pp. national land cover database for the coterminous
Auer, A. H., 1978: Correlation of land use and cover United States. Photogramm. Eng. Remote Sens., 73,
with meteorological anomalies. J. Appl. Meteor., 17, 337–341.
636–643. Karl, T. R., H. F. Diaz, and G. Kukla, 1988: Urbanization:

Downloaded from http://journals.ametsoc.org/bams/article-pdf/93/12/1879/3739321/bams-d-11-00019_1.pdf by guest on 26 September 2020

Bahl, H. D., and B. Padmanabhamurty, 1979: Heat island Its detection and effect in the United States climate
studies at Delhi. Mausam, 30, 119–122. record. J. Climate, 1, 1099–1123.
Balchin, W. G. V., and N. Pye, 1947: A micro-climato- Kljun, N., M. Rotach, and H. P. Schmid, 2002: A three-
logical investigation of Bath and the surrounding dimensional backward Lagrangian footprint model
district. Quart. J. Roy. Meteor. Soc., 73, 297–323. for a wide range of boundary-layer stratifications.
Black, M., 1952: Critical Thinking: An Introduction to Bound.-Layer Meteor., 103, 205–226.
Logic and Scientific Method. 2nd ed. Prentice Hall, Kostof, S., 1991: The City Shaped: Urban Patterns and
459 pp. Meanings through History. Thames and Hudson,
Brunn, S. D., and J. F. Williams, Eds., 1983: Cities of the 352 pp.
World: World Regional Urban Development. Harper Krayenhoff, E. S., I. D. Stewart, and T. R. Oke, 2009:
and Row, 506 pp. Estimating thermal responsiveness of local-scale
Chandler, T. J., 1965: The Climate of London. Hutchinson, climate ‘zones’ with a numerical modeling approach.
292 pp. Extended Abstracts, Timothy R. Oke Symp./Eighth
Chow, W., and M. Roth, 2006: Temporal dynamics of Symp. on Urban Environment. Phoenix, AZ, Amer.
the urban heat island of Singapore. Int. J. Climatol., Meteor. Soc., J12.5. [Available online at http://
26, 2243–2260. ams.confex.com/ams/89annua l/techprogram
Davenport, A. G., S. B. Grimmond, T. R. Oke, and /paper_149911.htm.]
J. Wieringa, 2000: Estimating the roughness of Lin, C. G., 1994: Changing theoretical perspectives on
cities and sheltered country. Preprints, 12th Conf. on urbanization in Asian developing countries. Third
Applied Climatology, Asheville, NC, Amer. Meteor. World Plann. Rev., 16, 1–23.
Soc., 96–99. Lo, F., and Y. Yeung, Eds., 1998: Globalization and the
Davey, C. A., and R. A. Pielke Sr., 2005: Microclimate World of Large Cities. United Nations University
exposures of surface-based weather stations: Impli- Press, 530 pp.
cations for the assessment of long-term temperature Loridan, T., and C. S. B. Grimmond, 2011: Charac-
trends. Bull. Amer. Meteor. Soc., 86, 497–504. terization of energy f lux partitioning in urban
Ellefsen, R., 1990/91: Mapping and measuring buildings environments: Links with surface seasonal proper-
in the urban canopy boundary layer in ten US cities. ties. J. Appl. Meteor. Climatol., 51, 219–241.
Energ. Buildings, 15–16, 1025–1049. Louw, W. J., and J. A. Meyer, 1965: Near-surface noc-
Gallo, K. P., A. L. McNab, T. R. Karl, J. F. Brown, J. J. turnal winter temperatures in Pretoria. Notos, 14,
Hood, and J. D. Tarpley, 1993: The use of NOAA 49–65.
AVHRR data for assessment of the urban heat island Lowry, W. P., 1977: Empirical estimation of the urban
effect. J. Appl. Meteor., 32, 899–908. effects on climate: A problem analysis. J. Appl.
Grigg, D., 1965: The logic of regional systems. Ann. Meteor., 16, 129–135.
Assoc. Amer. Geogr., 55, 465–491. —, and P. P. Lowry II, 2001: The Biological Environ-
Grimmond, C. S. B., and T. R. Oke, 1999: Aerodynamic ment. Vol. 2, Fundamentals of Biometeorology: Inter-
properties of urban areas derived from analysis of actions of Organisms and the Atmosphere. Peavine,
surface form. J. Appl. Meteor., 38, 1262–1292. 680 pp.
Gugler, J., Ed., 1996: The Urban Transformation of the Ludwig, F. L., 1970: Urban temperature fields. Urban
Developing World. Oxford University Press, 327 pp. Climates, WMO Tech. Note 108, 80–107.
Hansen, J., R. Ruedy, M. Sato, M. Imhoff, W. Lawrence, Mahrt, L., and H.-L. Pan, 1984: A two-layer model of soil
D. Easterling, T. Peterson, and T. Karl, 2001: A clos- hydrology. Bound-Layer Meteor., 29, 1–20.

AMERICAN METEOROLOGICAL SOCIETY december 2012 | 1895

 Masson, V., 2000: A physically-based scheme for the canopy characteristics in settlements of Nagano
urban energy budget in atmospheric models. Bound.- basin, Japan. Geogr. Rev. Japan, 78, 812–824.
Layer Meteor., 94, 357–397. Scherer, D., U. Fehrenbach, H.-D. Beha, and E. Parlow,
McGee, T., and I. M. Robinson, Eds., 1995: The Mega- 1999: Improved concepts and methods in analysis
Urban Regions of Southeast Asia. UBC Press, 384 pp. and evaluation of the urban climate for optimizing
Mills, G., 2006: Progress toward sustainable settlements: urban climate processes. Atmos. Environ., 33,
A role for urban climatology. Theor. Appl. Climatol., 4185–4193.
84, 69–76. Schmid, H. P., 2002: Footprint modeling for vegetation
Mizuno, M., Y. Na kamura, H. Mura kami, and atmosphere exchange studies: A review and perspec-
S. Yamamoto, 1990/91: Effects of land use on urban tive. Agric. For. Meteor., 113, 159–183.
horizontal atmospheric temperature distributions. Sharon, D., and R. Koplowitz, 1972: Observations of the heat
Energ. Buildings, 15, 165–176. island of a small town. Meteor. Rundsch., 25, 143–146.

Downloaded from http://journals.ametsoc.org/bams/article-pdf/93/12/1879/3739321/bams-d-11-00019_1.pdf by guest on 26 September 2020

Nunes da Silva, E., and H. Ribeiro, 2006: Temperature Shitara, H., 1957: Effects of buildings upon the winter
modifications in shantytown environments and temperature in Hiroshima city (in Japanese). Geogr.
thermal discomfort. Rev. Saude Publ., 40, 663–670. Rev. Japan, 30, 468–482.
O’Connor, A., 1983: The African City. Hutchinson, 359 pp. Stewart, I. D., 2011a: Redefining the urban heat island.
Oke, T. R., 1982: The energetic basis of the urban heat Ph.D. dissertation, Department of Geography,
island. Quart. J. Roy. Meteor. Soc., 108, 1–24. University of British Columbia, 352 pp. [Available
—, 1984: Towards a prescription for the greater use online at https://circle.ubc.ca/handle/2429/38069.]
of climatic principles in settlement planning. Energy —, 2011b: A systematic review and scientific critique
Build., 7, 1–10. of methodology in modern urban heat island litera-
—, 1987: Boundary Layer Climates. 2nd ed. Routledge, ture. Int. J. Climatol., 31, 200–217.
435 pp. —, and T. R. Oke, 2006: Methodological concerns
—, 2004: Initial guidance to obtain representative surrounding the classification of urban and rural cli-
meteorological observations at urban sites. IOM Rep. mate stations to define urban heat island magnitude.
81, WMO/TD-No. 1250, 47 pp. [Available online at Proc. Sixth Int. Conf. on Urban Climate. Goteborg,
www.wmo.int/pages/prog/www/IMOP/publications Sweden, International Association for Urban
/IOM-81/IOM-81-UrbanMetObs.pdf.] Climate, 431–444. [Available online at http://urban-
—, 2006: Towards better communication in urban climate.com/wp3/wp-content/uploads/2011/04
climate. Theor. Appl. Climatol., 84, 179–190. /ICUC6_Preprints.pdf.]
—, 2008: Urban observations. Guide to meteorologi- —, and —, cited 2009: Classifying urban climate
cal instruments and methods of observation, Part field sites by local climate zones: The case of Nagano,
II of Observing Systems, 7th ed., WMO-No. 8, II- Japan. Preprints, Seventh Int. Conf. on Urban
11-1–II-11-25. Climate, Yokohama, Japan, International Associa-
Okoola, R. E., 1980: The Nairobi heat island. Kenya J. tion for Urban Climate. [Available online at www.ide
Sci. Technol., 1A, 53–65. .titech.ac.jp/~icuc7/extended_abstracts/pdf/385055-
Peterson, T. C., 2003: Assessment of urban versus 1-090515165722-002.pdf.]
rural in situ surface temperatures in the contiguous —, and —, 2010: Thermal differentiation of local
United States: No difference found. J. Climate, 16, climate zones using temperature observations from
2941–2959. urban and rural field sites. Extended Abstracts, Ninth
Potter, R. B., and S. Lloyd-Evans, 1998: The City in the Symp. on Urban Environment, Keystone, CO, Amer.
Developing World. Longman, 244 pp. Meteor. Soc., 1.1. [Available online at http://ams
Ren, C., E. Y. Ng, and L. Katzschner, 2011: Urban cli- .confex.com/ams/19Ag19BLT9Urban/webprogram
matic map studies: A review. Int. J. Climatol., 31, /Paper173127.html.]
2213–2233. Sundborg, A., 1951: Climatological studies in Uppsala with
Robaa, S. M., 2003: Urban-suburban/rural differences special regard to the temperature conditions in the
over Greater Cairo, Egypt. Atmosfera, 16, 157–171. urban area. Geographical Institute of Uppsala, 111 pp.
Runnalls, K. E., and T. R. Oke, 2006: A technique to Taesler, R., 1980: Studies of the development and
detect microclimate inhomogeneities in historical thermal structure of the urban boundary layer in
records of screen-level air temperature. J. Climate, Uppsala. Uppsala University Meteorological Institute
19, 959–978. Rep. 61, 221 pp.
Sakakibara, Y., and E. Matsui, 2005: Relation between Tanner, C. B., 1963: Basic Instrumentation and Measure-
heat island intensity and city size indices/urban ments for Plant Environment and Micrometeorology.

1896 | december 2012

 Vol. 6, Soils Bulletin, University of Wisconsin Soil Wieringa, J., 1993: Representative roughness parameters for
Science Dept., 200 pp. homogeneous terrain. Bound.-Layer Meteor, 63, 323–363.
Theurer, W., 1999: Typical building arrangement for Wilmers, F., 1991: Effects of vegetation on urban climate
urban air pollution modelling. Atmos. Environ., 33, and buildings. Energ. Build., 15–16, 507–514.
4057–4066. WMO, 2008: Guide to Meteorological Instruments and
Vance, J. E., Jr., 1990: The Continuous City: Urban Methods of Observation. 7th ed., WMO-No. 8.
Morphology in Western Civilization. John Hopkins Yamashita, S., 1990: The urban climate of Tokyo. Geogr.
University Press, 552 pp. Rev. Japan, 63B, 98–107.

APPENDIX: SAMPLE DATASHEETS FOR LOCAL CLIMATE ZONES.

Downloaded from http://journals.ametsoc.org/bams/article-pdf/93/12/1879/3739321/bams-d-11-00019_1.pdf by guest on 26 September 2020

AMERICAN METEOROLOGICAL SOCIETY december 2012 | 1897

Downloaded from http://journals.ametsoc.org/bams/article-pdf/93/12/1879/3739321/bams-d-11-00019_1.pdf by guest on 26 September 2020

december 2012
1898 |
Downloaded from http://journals.ametsoc.org/bams/article-pdf/93/12/1879/3739321/bams-d-11-00019_1.pdf by guest on 26 September 2020

| 1899
december 2012
AMERICAN METEOROLOGICAL SOCIETY
Downloaded from http://journals.ametsoc.org/bams/article-pdf/93/12/1879/3739321/bams-d-11-00019_1.pdf by guest on 26 September 2020

december 2012
1900 |