Urban climatology: History, status and prospects
Gerald Mills
School of Geography, Planning & Environmental Policy, UCD, Dublin, Ireland
a r t i c l e i n f o
Article history:
Received 30 October 2013
Revised 8 June 2014
Accepted 18 June 2014
Keywords:
Descriptive climatology
Physical climatology
a b s t r a c t
This paper is a description of the historical development, current
status and prospects for urban climatology, the field concerned
with the study of the urban effect on the atmosphere and the
application of this knowledge to the better design and planning
of cities. Urban areas have a profound effect on the overlying air
as a result of changes to the nature of surface cover (urban form)
and emissions of heat, water vapour and materials that attend
human activities (urban function). Whilst these changes have been
well known and observed for over 100 years, it is only recently
that urban climatology has developed a coherent structure for
organising this knowledge so that urban observations can be conducted
and urban models developed that are transferrable from
place to place. In this paper, I give a personal perspective on the
history of the field from the vantage point of its current standing.
In addition, I suggest some pathways that the field may take in the
near future.
Ó 2014 Elsevier B.V. All rights reserved.
1. Introduction
Urban climatology is concerned with the study of the climate effect of urban areas and the application
of the knowledge acquired to the better planning and design of cities. It is defined primarily by
its focus on the city and incorporates aspects of many different disciplines, including meteorology,
climatology, air pollution science, architecture, building engineering, urban design, biometeorology,
amongst others. Each of these disciplines has its own focus and has developed distinctive tools and
methods (including vocabulary) appropriate to their interests. As a result, much of the knowledge base
http://dx.doi.org/10.1016/j.uclim.2014.06.004
2212-0955/Ó 2014 Elsevier B.V. All rights reserved.
E-mail address: gerald.mills@ucd.ie
Urban Climate 10 (2014) 479–489
Contents lists available at ScienceDirect
Urban Climate
journal homepage: www.elsevier.com/locate/uclim
that urban climatology draws upon is fragmented and is still in the process of being assimilated into a
comprehensive (and coherent) field of study.
Cities modify the overlying atmosphere significantly in nearly every respect. These modifications
are the result of changes the surface cover, fabric and geometry (urban form) and the attendant
anthropogenic emissions of waste heat, water vapour and materials (urban function). The fact of
the modifications has been well known for nearly two centuries but the nature of the processes
responsible have only been explored in detail in the last four decades. In this article, I present an
overview of the history of, current status of, and prospects for urban climatology (UC) based on my
perspective, which is that of a physical geographer trained in the Anglo-American education system;
a concern for human–environment relationships and the detectable effect of cities on local climate
explains the historic link of Geography with urban climatology. Of course, there are other relevant perspectives
that may emphasise other aspects of the development of UC. Meteorology, in particular, is a
closely related and professional field and has many sub-fields that have urban applications; these
include the study of the Earth’s boundary-layer and of air quality. However, its traditional concern
on forecasting weather and weather-related events has meant that (for the most part) its measurement
programmes and modelling efforts have excluded city scales. Its focus on meeting user needs
(which have changed over time) has also directed its resources toward other topics. Interestingly, over
the last decade changing user needs (e.g., growth in urban population and concerns about the effects
of climate change) and the dramatic improvement in atmospheric modelling has brought urban-scale
issues to the fore (NRC, 2012). This has occurred as meteorological theories and practices have been
incorporated by climatologists into UC so that today the terms ‘urban climatology’ and ‘urban
meteorology’ are largely synonymous. However, the development path that is outlined here has its
origin in the geographic tradition.
The fundamental motivation for the study of urban climates was outlined by Kratzer (1956): Only
when we possess sufficient knowledge of the bright and dark sides of city climate are we in a position to use
this information and to formulate a technique for city construction based on considerations of climate. Yet
something is already accomplished, when we realise that we do not have to accept city climate simply as a
fact but can influence it.
2. Historic development
Although Luke Howard may not have been the first to recognise the influence of urban areas on the
climate elements (e.g., Cerveny, 2009), his study of the climate in and around London represents the
scientific beginnings of UC. He maintained a meteorological station outside the city of London for
26 years and, with the help of his family, recorded air temperature, pressure, precipitation, etc. on a
daily basis. The product of his work was The Climate of London published in three volumes in 1833.
It is primarily a description and analysis of climate from the vantage of London rather than an examination
of the climate in cities. However, he identifies one aspect of the urban heat island effect (UHI)
when he compares his air temperature records (the ‘rural’ temperature, TR) against those maintained
by the Royal Society (the official scientific body) in the centre of London (the ‘urban’ temperature, TU),
DTUÀR ¼ TU À TR ð1Þ
The evidence from plotting these data showed the urban area had a distinct warming effect on the
near-surface atmosphere (Fig. 1). Howard concluded that the Mean Temperature of the Climate. . .is
strictly about 48.50° Fahr.: but in the denser parts of the metropolis, the heat is raised, by the effect of
the population and fires, to 50.50°; and it must be proportionately affected in the suburban parts
(Howard, 1833). He speculated on the processes responsible for this UHI and correctly identified most
of the causes we now study: anthropogenic heating; multiple reflection; lack of evaporation and; the
retardation of airflow (Mills, 2008).
The development of the field since this auspicious beginning is summarised in Table 1, which separates
the period since 1900 into two main phases. Prior to the 1970’s the majority of the research was
dominated by descriptive climatology and was based on observations of the weather elements, especially
air temperature and humidity, at urban scales. In the period since, physical climatology, with its
480 G. Mills / Urban Climate 10 (2014) 479–489
emphasis on the principles of conservation of energy, mass and momentum, has come to the fore. This
is not to say that the work that characterised the earlier period is not still continuing. For example,
urban heat island studies probably still represents the majority of urban climate studies currently
undertaken.
2.1. Descriptive climatology
During the early 20th century the number of studies on urban climates expanded. Much of our
knowledge of the urban effect during this period came from research in central Europe that framed
the urban effect within the context of landscape studies and micro-scale climatology. For example,
Geiger’s first edition of the Climate near the Ground appeared in the German language in 1927 and
it was not until 1965 that it was published in English (Geiger, 1965). Similarly, Kratzer completed a
survey of work done on the Climates of Cities (Stadtklima) in 1937 (the English translation, The Climate
Fig. 1. A comparison between the air temperature observations by Luke Howard (solid) against those made by the Royal Society
within London (broken). Source: Howard (1833).
Table 1
Approaches to the study of urban climate effects.
Time
period
Approach
1900 Observation and description of urban effects using conventional meteorological equipment (thermometers,
hygrometers, etc.)
1960 Move toward measurement of ‘process’ variables – radiation, sensible and latent heat exchanges. The use of
statistical methods to summarise and generalise results
1970 Application of conventional (micro-)meteorological theory to urban climates. Use of energy budget as a
framework to explain the urban effect. Observation of process variables: radiation, estimated fluxes. Use of
computer modelling techniques. More rigorous definition of urban ‘surface’, urban scales and observing
urban effects
1980 Adoption of an experimental approach: Select common urban forms (streets become canyons). Use of scaledphysical
models and direct measurement of fluxes
1990 Relationships between real urban forms and climate effect. Urban field projects examined by research teams.
Generalizations based on a range of settlements
2000 Development of realistic urban climate models. Employment of novel techniques for examining urban
climate
G. Mills / Urban Climate 10 (2014) 479–489 481
of Cities was published in 1956). By comparison, there was less work on these topics elsewhere; one of
the best known studies on the local effect (including that of towns) in Britain by Balchin and Pye
(1947) appears a decade later.
Many of the basic methods for examining local climates were established during this period. For
example, to capture the spatial detail of the UHI a dense network of stations was needed. Moreover,
the instruments would normally require visual inspection in situ at the appropriate time to resolve the
temporal character of the phenomenon. For some variables such a wind, observing the urban effect
across the city under these conditions required ingenuity. For example, Okita (1960) used the
opportunity provided by a winter storm that resulted in rime formation (supercooled droplets that
freeze in contact with a surface). In the presence of wind these features were shaped by the airflow
and allowed the study of wind patterns in city streets. For air temperature and humidity that respond
more conservatively over shorter time periods, the transect method was developed. Sundborg’s (1950)
study of the UHI around Lund is based on this method and exemplifies the personal investment of time
and effort required to gather information at this scale. Chandler (1965) work on the Climate of London
represents the epitome for this type of work. This treatise is the culmination of an extraordinary
amount of data acquisition and analysis and describes in a systematic way the effect of the urban area
on each atmospheric property.
Despite the accumulation of evidence on the urban air temperature effect, much of it was specific
to particular places and used distinct methods that made generalisations difficult. Oke’s (1973) study
on the UHI and city size showed that abstraction from the specific to the general was possible if the
available evidence was screened to remove extraneous effects, such as variations in topography and
weather. The linear relationship between the maximum value of DTUÀR(max) and city size tells us a
great deal about our knowledge of cities and the possible controls on the UHI at this juncture. For
the available North American data the relationship stated
DTUÀRðmaxÞ ¼ 2:96 log P À 6:41 ð2Þ
Population (P) was used as a surrogate variable for actual measures of urban form and function that
were needed to offer a physically based empirical relationship. The fact that different coefficients were
needed to capture the same relationship for European cities was indicative of a failure to acquire
relevant information on cities (e.g., materials, surface cover and geometry). Oke’s (1981) study nearly
a decade later replaced population with a measure of urban geometry (the ratio of building height (H)
to street width (W), or alternatively sky view factor, wsky) that characterised the city centre and
produced a universal relationship based on available data for cities in different climates,
DTUÀRðmaxÞ ¼ 7:45 þ 3:97 ln
H
W
 
¼ 15:27 À 13:88wsky ð3Þ
Yet, although there was a considerable body of empirical data that had been accumulated during
this period that supported, complemented and extended the findings of Howard, UC had not undertaken
the task of measuring and modelling the causative mechanisms.
A final point on the use of observations to assess the urban effect is worth making. There have been
few expositions on the methodology that has underpinned much of the observational evidence for the
urban effect. It has been common practice to compare urban observations with those acquired in a
nearby ‘rural’ setting (which represents the pre-urban environment) and assume that the difference
is the urban effect. This was the framework used by METROMEX, the first grand urban project
(Lowry, 1974). However, Lowry (1977) characterised the problem of identifying this effect amongst
all the other climate effects, including climate change, as fundamentally intractable; to measure the
urban effect one needed a long-term set of in situ measurements that dated from the period prior
to urban settlement. Unfortunately, such information is not usually available as stations are moved
to offset the ‘contamination’ of records caused by urbanisation.
2.2. Physical climatology
UC experienced a significant advance in the 1970’s with an increase in the numbers of investigators
and a changed perspective. For physical geographers in the Anglo-American realm, this resulted in the
482 G. Mills / Urban Climate 10 (2014) 479–489
adoption of a quantitative and systematic approach to research. For example, Sellers’ (1965) Physical
Climatology was a core textbook in the educational training of a new generation of researchers.
The most common expression of the new approach to the study of UC was the surface energy
balance,
QÃ
¼ QH þ QE þ QG ð4Þ
This described the partitioning of net radiation (Q⁄
) into the turbulent exchanges of sensible (QH)
and latent (QE) heat with the atmosphere and the conductive exchange of sensible heat (QG) with
the substrate. Adopting this approach shifted the research focus from describing effects (responses)
to seeking their cause (processes) and linked its study to the broader topic of boundary-layer meteorology.
This field had already conceptualised the nature of surface–air interactions and developed
schemes to estimate surface energy exchanges, albeit for homogeneous surfaces. Terjung (1976), for
example, argued that physical geography should move up the research hierarchy and could investigate
the varying responses to different inputs, throughputs, and outputs of energy, mass, momentum, and
information to portions of the environmental envelope of concern to mankind. I urge that this be the beacon
towards which physical geographers set their sights. . . Climatologists, increasingly, should adopt the
research level of physical process-response systems relevant to the world of man.
Translating this approach into research was not straight-forward however. There were no generally
available measurement systems that could record the turbulent fluxes directly; the typical approach
was to record the vertical gradients in wind-speed, temperature and humidity near the ground and
estimate the energy exchanges from these. On the other hand, researchers could now access computer
resources, which allowed theories to be translated into computer code and simulated. In his wellknown
paper on the UHI, Myrup (1969) stated that in surveying the explanations [for the formation
of the urban heat island] the complete absence of numerical estimates of the order of magnitude of the
suggested mechanisms is striking. To address this issue he converted the individual terms of the energy
balance statement into a series of relationships amongst measurable atmospheric variables (e.g.,
wind-speed, air temperature, etc.) and solved these equations for a single unknown term, the equilibrium
‘surface’ temperature. The UHI was calculated as the difference in surface temperature between a
grass covered surface (rural) and a concrete slab (city). It is difficult at this remove to gauge the effect
of this paper, which was considerable; this simple model showed how complex ideas could be translated
into solvable form to test hypotheses and put numbers to fluxes. Nevertheless, there remained
theoretical obstacles too. One of the most important of these was the definition of the surface itself
where exchanges were calculated; Myrup’s model avoided this issue by simply treating the layer of
air below roof level as isothermal, even though most of the UHI observations were made in this space.
The new approach was seen in Terjung’s approach to UC which, stated that: In contrast to the many
inductive-observational studies in urban climatology, few have been based on inferences derived from
theory (Terjung and Louie, 1974). The model developed to address this weakness examined the
exchanges along a surface transect that extended from the roof of one building, down its face, across
the street and up the side of the opposite building. At regular intervals along this transect the surface
energy balance was simulated. This technique, of representing the micro-climatic urban environment
using a transect, was adopted by others in later work (e.g., Arnfield, 1982). However, there was still no
definitive statement of what constituted the urban surface in common usage by researchers; Oke
(1976) finally defined the urban canopy layer (UCL) as the lowest part of the urban boundary layer
(UBL), occupying the space between the ground and the mean height of buildings. Observations
(and simulations) made within the UCL were driven by micro-scale processes such as the radiation
exchange between building surfaces. The atmosphere above the UCL responded to exchanges at an
interface at the top of the UCL, which included both the rooftop surfaces and the open canopy. This
simple distinction clarified much of the work to this point and allowed the adoption of simple
configurations (such as the urban canyon to represent the city street) to represent the complex and
heterogeneous UCL.
One of the most important pieces of work from this period is that by Nunez and Oke (1977) on
observations within the UCL. In the introduction they state that although progress has been made, there
is still no comprehensive study available concerning the surface energy balance of an urban area. The
present study was designed to investigate the energy input, partitioning and output of a characteristic
G. Mills / Urban Climate 10 (2014) 479–489 483
urban canopy layer structure – an urban canyon. The results of this work provided valuable data for subsequent
modelling exercises and showed the significant differences in the energy balance that occurs
at surfaces within the canyon, including the contribution of multiple reflections. Remarkably, the
aggregated results at the canyon top interface showed that the variations within the canyon compensated
for each other; the net radiation showed a smooth diurnal curve not much different from those
found over rural environments and the critical urban effect was to partition this energy mainly into QH
and QG, as there was little evaporation. It is worth pointing out that the urban canyon as a unit of study
was already firmly established in the air pollution community as a means of evaluating pedestrian
exposure to vehicular pollutants (e.g., Johnson et al., 1973).
Much of the foundations for our current understanding of the urban climate effect was established
during the 1980’s and 1990’s as the research frontier moved away from a description of the urban climate
in a given city toward quasi-experiments that could be replicated. The urban canyon (with its
geometrical descriptors of height (H) and width (W)) provided a useful context for bringing together
research in a number of different areas and for structuring research projects; for example the importance
of the sky view factor (wsky) as a regulator of night-time surface cooling and a key indicator of
UHI potential was an important outcome of this work. Oke (1984) could draw on this research to provide
guidance for good urban design practice that was based on the evidence, rather than speculation.
Nevertheless, much of the work during the 1980’s originated from remarkably few teams of researchers,
scattered globally with few opportunities for interaction. An indication of the status of the field at
the time can be found in the proceedings of the second international conference dedicated to UC held
in Mexico City, and focussed on the urban climates of the Tropics (Oke, 1986). Many of the papers
were descriptive in nature and there were few that employed the energy balance approach.
During the 1990’s the intensity of UC research increased substantially; the number of researchers
increased and there was greater attention paid to the make-up of the urban landscape and the integration
of UC into the corpus of boundary-layer meteorology. Much of the observational work shifted
from the UCL into the atmosphere above roof level, where the micro-scale effects of the underlying
urban surface might be assumed to have been integrated, effectively producing a distinct neighbourhood,
or local-scale, signal. These projects attempted to close the energy budget of a representative
volume that extended from the substrate, through the UCL to a level above the buildings:
QÃ
þ QF ¼ QH þ QE þ DQS þ DQA ð5Þ
The new terms here are the anthropogenic heat flux (QF) the storage of heat (DQS) and the advection of
energy through the sides of the volume (DQA). Observing these terms in the complex urban setting
was not a simple task; instruments needed to be located within a suitable fetch of a reasonably
homogenous urban area (so that DQA could be assumed negligible) and positioned at a height where
the turbulent fluxes are reasonably constant. Guidelines for dealing with the challenge of making
urban observations have become a core part of UC knowledge (Oke, 2006a).
Numerical modelling of the urban environment has proceeded apace; as computer power has
improved local-scale and micro-scale models of increasing sophistication and detail were developed.
The simple descriptions of the urban surface (e.g., concrete slabs) used in earlier models were replaced
by more realistic descriptions of land cover (e.g., vegetated and impermeable surfaces), material
properties (e.g., asphalt and concrete), geometry (e.g., street dimensions and building plan area)
and functions (e.g., traffic and heating) that define different types of urban neighbourhoods. These
improvements at the urban-scale have been matched by advances in modelling the hierarchy of
climate scales from global to regional to meso-scale which are now capable of incorporating the presence
of cities to varying degrees (e.g., Best, 2006; Masson, 2000; McCarthy et al., 2010). Advances in
computing power has also allowed computational fluid dynamic (CFD) models that have traditionally
been used to simulate flows around individual buildings to be applied to entire neighbourhoods (e.g.,
Hanna et al., 2006). One of the changes that emerged during this period has been the co-operation
amongst research teams that has allowed for more comprehensive research projects that incorporated
different observation and modelling systems (e.g., Rotach et al., 2005).
Some of the advances in this period could be attributed to demographic change; students trained in
UC during the 1980’s were now themselves teachers/researchers with their own students. The
transformation of UC during this period is described by Arnfield (2003).
484 G. Mills / Urban Climate 10 (2014) 479–489
3. Current status
Much of our understanding of the urban boundary layer is encapsulated in Fig. 2. It distinguishes
between the three scales of effect (micro-, local- and meso-scales) and suggests that the processes that
dominate in each differ. It shows the vertical layering of the urban atmosphere: the surface layer,
which is comprised of roughness (RSL) and inertial (ISL) sub-layers (the urban canopy layer is
immersed in the former); the mixed layer above is where the urban ‘surface’ exchanges are blended
into the wider atmosphere and transported downwind. Critically, the ISL is the level at which the
atmosphere has adjusted to the underlying urban landscape such that observations of energy, mass
and momentum fluxes made at this height are representative of the amalgam of microclimates
(created by gardens, roofs, walls, etc.) that comprise a local-scale ‘neighbourhood’. On the other hand,
instruments positioned within the RSL are exposed to the individual microclimatic effects so that the
sources of measured fluxes are difficult to disentangle. So, if a measurement site is selected over a
reasonably homogeneous urban landscape and within its fetch, then one can link observations to
the character of the underlying surface (for example, the proportion of the surface that is occupied
by buildings, impervious surfaces, vegetation, etc.) This understanding allows the application of traditional
micro-meteorological theory to urban settings, which were hitherto regarded as too complex
owing to their great spatial heterogeneity. A growing number of urban flux measurement sites are
located in city environments based on this understanding are providing greater understanding of
the linkages between background climates and the surface-air exchanges associated with different
urban neighbourhood types (Christen et al., 2009).
Oke (2006b) described the evolution of UC using eight modes of investigation or practice:
Conceptualisation; Theorisation; Field observation; Modelling; Model evaluation; Application in
urban design and planning; Impact assessment (post-implementation) and; Policy development and
modification. He opined that whereas the first four modes had progressed significantly in recent
Fig. 2. The scales of the urban climate effect. Source: Oke (2006a).
G. Mills / Urban Climate 10 (2014) 479–489 485
decades, the evaluation of urban models and the transfer of knowledge into planning practice
remained undeveloped. The process of evaluating models has just begun yet this stage is needed if
they are to be used to simulate the effects of climate-based urban policies, such as albedo modification
or rooftop greening.
Transferring this scientific knowledge into practice on a widespread basis remains a major challenge
(Hebbert and MacKillop, 2013). Part of the problem is due simply to the perceived importance
of such information in making planning decisions; when placed alongside other planning issues,
concerns for the urban heat island (as an example) have been seen as of marginal interest. However,
the failure of UC in general to structure its scientific knowledge in an accessible and applicable manner
has also been part of the problem. This is not universally true however; the example of the urban
climatologist office in Stuttgart is a case in point (Reuter, 2011). Established in the late 1930’s to
manage the air resource over the heavily industrialised city, the office has maintained a presence to
this day that has acted as an exemplar for other places. Moreover, German scholars have maintained
a perspective on the connection between settlements and their landscape context and on the
biometeorological consequences of urbanisation (e.g., Mayer and Höppe, 1987). As a result, the understanding
of links between land cover, topography and urbanisation has resulted in a versatile and
flexible approach to climate mapping at scales of urban planning that has proved adaptable to many
different climates and planning circumstances (Ren et al., 2011).
4. Future prospects
Oke et al. (2006) characterised the recent history of UC in terms of ‘reducing solitudes’, by which he
meant the bringing together of what had been a diverse and scattered community often working in
isolation. Professional associations have played a large part in this: the American Meteorological
Society has a Board of the Urban Environment, which has been involved with symposia on urbanrelated
themes (e.g., boundary layer meteorology and air pollution) since 1982; the Japanese–German
Meeting on Urban Climatology has held at regular intervals since 1994. By the turn of the millennium
it was apparent that there was sufficient interest and expertise to establish an organisation dedicated
to UC; The International Association for Urban Climates (IAUC) was formed in 2000 and has assumed
responsibility for organising a series of international conferences dedicated to the topic of urban
climates. These events have provided opportunities to assess the status of UC in some key areas
and identify lacunae (see for example, Mills et al., 2014 in this journal).
There remain many gaps in our scientific knowledge; for example, the urban effect on precipitation
has been researched for nearly forty years but progress has been very slow and the results are still not
definitive. This is especially ironic as it was the search for this effect that was the impetus for the
Metromex experiment; however it may be a problem that is ideally suited to numerical modelling
(Shepherd, 2005). Also, there have been very few observations made within the ML of the urban
boundary layer, yet this information is needed to understand the impact of cities beyond their boundaries.
However, the outstanding gap in my view is the paucity of information on the rapidly growing
cities of the less prosperous regions, many of which are located in the tropics. Meeting this challenge
will require a considerable international collaborative effort. Ideally, such a programme would put in
place both an observational and educational infrastructure to conduct some basic research. The cost of
meteorological equipment remains an impediment to acquiring information in the places that it is
most needed. It may be that rapid advances in sensor technologies linked to telecommunication
devices can provide some of the more basic data but acquiring flux information will need a concerted
effort. On the other hand, data on the nature of the urban landscape in such places could be very
valuable; for example, much of the UC knowledge on surface–air exchanges that is encoded in numerical
models could be transferred to new urban settings if we had suitable data on those places. Even a
simple but consistent physically-based description of urban landscapes, using the scheme of Stewart
and Oke (2012) perhaps, could be of great value. This is one area where the potential of remote sensing
for acquiring information on cities globally is great (Miller and Small, 2003).
Overcoming the barriers to knowledge–policy transfer will require both accessible knowledge and
appropriate tools and, I would venture a supportive political context. Recent publications (e.g., Erell
et al., 2011; Grimmond et al., 2010; Mills et al., 2010; NRC, 2012) have discussed the current state
486 G. Mills / Urban Climate 10 (2014) 479–489
of UC knowledge and its value for urban planning and design. Moreover, there are modelling tools
available that can run on inexpensive computers that may be used to explore urban micro-climatic
variations and to test the climatic implications of different design options. As examples: SOLWEIG
(Lindberg, 2007) employs databases on urban morphology to calculate access to sunshine and skyview
factors, two of the most relevant variables in assessing urban micro-climates and; ENVI-MET
(Bruse and Heribert, 1998) is a remarkably versatile local scale model that can simulate three
dimensional flow fields and surface–air exchanges. However, achieving Kratzer’s vision of a
climate-informed urban planning may require the political impetus created by some form of crisis
– at an urban scale, it is likely that poor air quality in many Chinese cities will eventually be simply
unacceptable and prompt action, just as it did for cities elsewhere. At a global scale, concerns about
climate change may provide this impetus. In the past decade, the role of cities as sources of greenhouse
gas emissions and their vulnerability to climate changes has generated a great deal of research
on urban-scale adaptation and mitigation and on urban sustainability (e.g., Rosenzweig, 2011; Stone,
2012). Unfortunately, much of this research makes little or no reference to a large body of existing UC
research. Naturally, it will be important that policies designed to achieve outcomes with respect to
global scale issues are consistent with those designed to improve conditions at the urban scale, where
the decisions on transport, on heating/cooling, etc. that regulate emissions are made. Currently
climate research in cities and that on cities and global climate change are proceeding along parallel
paths and there needs to be far greater interaction.
5. Conclusions
Advances in our understanding of the urban climate effect have come about very quickly and comparatively
recently. Establishing the scientific foundation of UC only really began after the 1970’s;
prior to this most work based on comparing the measured atmospheric properties in urban areas with
those in surrounding rural areas. For some places, such as London, the work produced outstanding
comprehensive climatologies of a city but for most places the information gathered was piecemeal
and did not allow for either an examination of causes for the effect or comparative analysis between
cites. For example, studies of the near-surface urban heat island were prolific but made little
substantial progress on Howard’s work completed over a 100 years earlier. The adoption of the energy
balance/budget approach and all that it implied was critical to the development of the field as it lent
itself to numerical modelling and demanded a different type of observation. Moreover, applying it to
the three dimensional city with its distinct urban canopy layer forced a consideration of the character
of the urban surface to which instruments were exposed (or to which model results referred). The
slow advance in our understanding since its adoption can be attributed partly to the fewness of those
working in the field; most research was done in relative isolation at few places. Major advances in the
last two decades have resulted from: increased interest in the places that most live (from many other
disciplines, including meteorologists); technological advances in instrumentation and computing
power and; the development of institutional frameworks, which have helped develop and foster a
community with shared interests in city climates.
Scientific progress in UC seems assured but two major challenges remain: acquiring information on
the climates of cities in less prosperous regions, which are growing rapidly and many of which have
Tropical climates in which there have been few studies and; transferring urban climate knowledge
into routine planning/design practice.
References
Arnfield, A.J., 1982. An approach to the estimation of the surface radiative properties and radiation budgets of cities. Phys.
Geogr. 3, 97–122.
Arnfield, A.J., 2003. Two decades of urban climate research: a review of turbulence, exchanges of energy and water, and the
urban heat island. Int. J. Climatol. 23, 1–26.
Balchin, W.G.V., Pye, N., 1947. A micro-climatological investigation of Bath and the surrounding district. Q. J. R. Meteorolog. Soc.
73, 297–323.
Best, M.J., 2006. Progress towards better weather forecasts for city dwellers: from short range to climate change. Theoret. Appl.
Climatol. 84, 47–55.
G. Mills / Urban Climate 10 (2014) 479–489 487
Bruse, M., Heribert, F., 1998. Simulating surface–plant–air interactions inside urban environments with a three dimensional
numerical model. Environ. Modell. Softw. 13, 373–384.
Cerveny, R., 2009. Noah webster: lexicographer, climatologist. Weatherwise 62, 38–43.
Chandler, T.J., 1965. The Climate of London. Hutchinson, London (Reprinted by IAUC and available at <www.urban-
climate.org>).
Christen, A., Grimmond, C.S.B., Roth, M., Pardyjak, E., 2009. The IAUC urban flux network – an international network of
micrometeorological flux towers in urban ecosystems. Eos Trans. AGU 90 (52) (Fall Meet. Suppl., Abstract B31C-06).
Erell, E., Pearlmutter, D., Williamson, T.J., 2011. Urban Microclimate: Designing the Spaces Between Buildings. Routledge.
Geiger, R., 1965. The Climate Near the Ground. Mass, Cambridge, p. 611.
Grimmond, C.S.B., Roth, M., Oke, T.R., Au, Y.C., Best, M., Betts, R., Carmichael, G., Cleugh, H., Dabberdt, W., Emmanuel, R., Freitas,
E., Fortuniak, K., Hanna, S., Klein, P., Kalkstein, L.S., Liu, C.H., Nickson, A., Pearlmutter, D., Sailor, D., Voogt, J., 2010. Climate
and more sustainable cities: climate information for improved planning and management of cities (producers/capabilities
perspective). Procedia Environ. Sci. 1, 247–274.
Hanna, S.R., Brown, M.J., Camelli, F.E., Chan, S.T., Coirier, W.J., Kim, S., Hansen, O.R., Huber, Alan H., Michael Reynolds, R.M., 2006.
Detailed simulations of atmospheric flow and dispersion in downtown Manhattan: an application of five computational
fluid dynamics models. Bull. Am. Meteorol. Soc. 87, 1713–1726.
Hebbert, M., MacKillop, F., 2013. Urban climatology applied to urban planning: a postwar knowledge circulation failure. Int. J.
Urban Reg. Res. 37, 1542–1558.
Howard, L., 1833. The Climate of London, vols. I–III. Harvey and Dorton, London (Republished by IAUC and available at <http://
www.urban-climate.org>).
Johnson, W.B., Ludwig, F.L., Dabberdt, W.F., Allen, R.J., 1973. An urban diffusion simulation model for carbon monoxide. J. Air
Pollut. Control Assoc. 23, 490–498.
Kratzer, A., 1956. The Climate of Cities (Das Stadtklima). Air Force Cambridge Research Laboratories (Reprinted by IAUC and
available at <www.urban-climate.org>).
Lindberg, F., 2007. Modelling the urban climate using a local governmental geo-database. Meteorol. Appl. 14, 263–273.
Lowry, W.P., 1974. Project METROMEX: its history, status, and future. Bull. Am. Meteorol. Soc. 55, 87–88.
Lowry, W.P., 1977. Empirical estimation of urban effects on climate: a problem analysis’’. J. Appl. Meteorol. 16,
129–135.
Masson, V., 2000. A physically-based scheme for the urban energy budget in atmospheric models. Boundary Layer Meteorol. 94,
357–397.
Mayer, H., Höppe, P., 1987. Thermal comfort of man in different urban environments. Theoret. Appl. Climatol. 38,
43–49.
McCarthy, M.P., Best, M.J., Betts, R.A., 2010. Climate change in cities due to global warming and urban effects. Geophys. Res. Lett.
37.
Miller, R.B., Small, C., 2003. Cities from space: potential applications of remote sensing in urban environmental research and
policy. Environ. Sci. Policy 6, 129–137.
Mills, G., 2008. Luke Howard and the climate of London. Weather 63, 153–157.
Mills, G., Cleugh, H., Emmanuel, R., Endlicher, W., Erell, E., McGranahan, G., Ng, E., Nickson, A., Rosenthal, J., Steemer, K., 2010.
Climate information for improved planning and management of mega cities (needs perspective). Procedia Environ. Sci. 1,
228–246.
Mills, G., Di Sabatino, S., Erell, E., Martilli, B., 2014. Introduction to ICUC8 special issue. J. Urban Climate 10 (P2), 201–203.
Myrup, L.O., 1969. A numerical model of the urban heat Island. J. Appl. Meteorol. 8, 908–918.
National Research Council (NRC), 2012. Urban Meteorology: Forecasting, Monitoring, and Meeting Users’ Needs. The National
Academies Press, Washington, DC.
Nunez, M., Oke, T.R., 1977. Energy balance of an urban canyon. J. Appl. Meteorol. 16, 11–19.
Oke, T.R., 1973. City size and the urban heat island. Atmos. Environ. 7 (8), 769–779.
Oke, T.R., 1976. The distinction between canopy and boundary-layer urban heat islands. Atmosphere 14, 268–277.
Oke, T.R., 1981. Canyon geometry and the nocturnal urban heat island: comparison of scale model and field observation. Int. J.
Climatol. 1, 237–254.
Oke, T.R., 1984. Towards a prescription for the greater use of climatic principles in settlement planning. Energy and Buildings 7,
1–10.
Oke, T.R. (Ed.), 1986. In: Proceedings WMO Technical Conference on Urban Climatology and its Applications with Special Regard
to Tropical Areas. WMO No. 652, World Meteorological Organisation, Geneva, p. 534.
Oke, T.R., 2006a. Initial guidance to obtain representative meteorological observations at urban sites, World Meteorological
Organization. Instrument and Observing Methods Report No. 81.
Oke, T.R., 2006b. Towards better scientific communication in urban climate. Theoret. Appl. Climatol. 84, 179–190.
Oke, T., Kłysik, K., Bernhofer, C., 2006. Editorial: progress in urban climate. Theoret. Appl. Climatol. 84, 1–2.
Okita, T., 1960. Estimation of direction of air flow from observation of rime ice. J. Meteorol. Soc. Jpn 38, 207–209.
Ren, C.R., Ng, E., Katzschner, L., 2011. Urban climatic map studies: a review. Int. J. Climatol. 31, 2213–2233.
Reuter, U., 2011. Implementation of Urban Climatology in City Planning in the City of Stuttgart. Powerpoint presentation at City
Weathers June 2011 at Manchester University. Available from: <http://www.sed.manchester.ac.uk/architecture/research/
csud/workshop/programme/>.
Rosenzweig, C. (Ed.), 2011. Climate Change and Cities: First Assessment Report of the Urban Climate Change Research Network.
Cambridge University Press.
Rotach, M.W., Vogt, R., Bernhofer, C., Batchvarova, E., Christen, A., Clappier, A., Feddersen, B., Gryning, S.-E., Martucci, G., Mayer,
H., Mitev, V., Oke, T.R., Parlow, E., Richner, H., Roth, M., Roulet, Y.-A., Ruffieux, D., Salmond, J.A., Schatzmann, M., Voogt, J.A.,
2005. BUBBLE – an urban boundary layer meteorology project. Theoret. Appl. Climatol. 81, 231–261.
Sellers, W.D., 1965. Physical Climatology. University of Chicago Press, Chicago.
Shepherd, J.M., 2005. A review of current investigations of urban-induced rainfall and recommendations for the future. Earth
Interact. 9.
488 G. Mills / Urban Climate 10 (2014) 479–489
Stewart, I.D., Oke, T.R., 2012. Local climate zones for urban temperature studies. Bull. Am. Meteorol. Soc. 93, 1879–1900.
Stone Jr., B., 2012. The city and the coming climate: climate change in the places we live. Cambridge University Press.
Sundborg, A., 1950. Local climatological studies of the temperature conditions in an urban area. Tellus 2, 221–231.
Terjung, W., 1976. Climatology for geographers. Ann. Assoc. Am. Geogr. 66, 199–222.
Terjung, W.H., Louie, S.S.F., 1974. A climatic model of urban energy budgets. Geog. Anal. 6, 341–367.
G. Mills / Urban Climate 10 (2014) 479–489 489