Journal_Investigating the Climatic Impact of Urban Planning

INTERNATIONAL JOURNAL OF CLIMATOLOGYInt. J. Climatol. 28: 1943–1957 (2008)Published online 19 March 2008 in Wiley InterScience(www.interscience.wiley.com) DOI: 10.1002/joc.1680

Investigating the climatic impact of urban planningstrategies through the use of regional climate modelling:

a case study for Melbourne, Australia

Andrew M. Coutts,* Jason Beringer and Nigel J. TapperSchool of Geography and Environmental Science, Monash University, Melbourne, Vic, 3800, Australia

ABSTRACT: Urban planning is a useful method for improving local climate and human health in cities throughpurposefully modifying urban land surface characteristics. This can reduce the potential risks of elevated city temperaturesdue to the urban heat island (UHI). Unfortunately, simple tools are not readily available for urban planners to assess theclimatological impacts of various urban development scenarios. Urban modelling could be developed into such a tool toachieve this. This study attempts to design and evaluate a suitable tool for application in Melbourne, Australia. The AirPollution Model (TAPM) was chosen to assess the impact of a long-term urban planning strategy on local climate and theabove canopy UHI in Melbourne. Improvements were made to TAPM by increasing the number of urban land-use classesin the model and creating a higher resolution land cover database focused on housing density. This modified version ofTAPM showed a good performance in replicating the surface energy balance compared with an observational flux towersite in suburban Melbourne during summer. TAPM simulated a mean maximum UHI intensity of 3–4 °C at 2 a.m. inJanuary. A future UHI scenario was then examined (year 2030) using an urban land cover database derived from plansin the Melbourne 2030 urban planning strategy. Results highlighted specific areas where planning intervention would beparticularly useful to improve local climates, namely activity centres and growth areas. The appropriateness of the use ofTAPM and climate models as a tool in urban planning is also discussed. Copyright 2008 Royal Meteorological Society

KEY WORDS urban planning; urban climate; climate modelling; Melbourne; surface energy balance

Received 18 August 2006; Revised 1 December 2007; Accepted 10 December 2007

1. Introduction

Unplanned and rapid urbanization in cities can oftenlead to negative environmental impacts, including mod-ifications to the local urban climate. The urban heatisland (UHI) phenomenon is often evident in citieswhereby urban areas are warmer than surrounding ruralareas. UHIs may contribute towards elevated tempera-tures, which can be harmful for vulnerable urban resi-dents, particularly during summer and heat wave episodes(Rankin, 1959). Higher incidences of heat-related ill-nesses including heart disease and even mortality havebeen associated with elevated temperatures within urbanareas. Those particularly at risk include the elderly, low-income earners, and residents in high density, older hous-ing stock with limited surrounding vegetation (Smoyer-Tomic et al., 2003). Fortunately, there is sufficient evi-dence to suggest that urban planning can be a usefulmethod for improving local climate and human health(Jackson, 2003; Stone, 2005). In order to reduce negativeclimatological impacts, those involved in urban devel-opment and design must begin to incorporate climateknowledge into planning strategies.

* Correspondence to: Andrew M. Coutts, School of Geography andEnvironmental Science, Monash University, Wellington Road, Clayton,Victoria, 3800, Australia. E-mail: [email protected]

UHIs form primarily because of high thermal heatcapacity and heat storage of urban surfaces, addedsources of heat from anthropogenic activities, andreduced evapotranspiration (Oke, 1988). Within the urbancanopy (below maximum building height), urban geome-try is also important in controlling radiative exchangesbetween the walls and floor of urban canyons. Smallsky view factors (SVF) and large height to width ratiostrap radiative energy during the day and limit noctur-nal cooling. This leads to the development of peak UHIintensities during the night, as rural areas are allowedto cool uninhibited. Cloud amount and wind speed areimportant meteorological parameters as they affect long-wave cooling and ventilation, which serve as surrogatevariables describing the relative roles of radiative and tur-bulent exchanges in and around the urban region (Morrisand Simmonds, 2000).

While the UHI phenomenon has been well documentedin the climatological literature over the past few decades,few cities have developed comprehensive strategies tomitigate its intensity. Reasons for little consideration ofclimate related understanding in urban planning includea lack of knowledge, economic constraints, and com-munication problems (Eliasson, 2000). Added to thesereasons, planning tools are not often available for plan-ning authorities to assess the implications of projected

Copyright 2008 Royal Meteorological Society

1944 A. M. COUTTS ET AL.

land-use change on local climate (Fehrenbach et al.,2001). According to Eliasson (2000), the development ofsuch tools based on scientific research that can be incor-porated into the urban planning framework should be achallenge and focus for urban climatologists. However,one such tool that can address the issue of climate impactsof urban planning strategies, if adequately developed, isclimate modelling, both local and regional. Climate mod-elling that uses specific treatments at the urban surfacecan significantly help in determining the likely impacts oflarge scale urbanization on local climate and UHI devel-opment, improving weather forecasts, estimating energyconsumption, and aiding in urban planning (Kusaka andKimura, 2004).

Information at all urban scales (global, regional, local,and microscale) can be highly beneficial for planners, butknowledge about climate at the city (regional) and neigh-bourhood (local) scale is specifically relevant as planningauthorities influence/regulate features at this scale, suchas heights of buildings. For climate models to be a usefultool in aiding sustainable urban planning, it is importantthat they are correctly able to simulate the urban climateat this scale. The urban surface is highly complex andmodels require additional inputs, and new and improvedparameterizations, to accurately simulate the urban cli-mate (Zehnder, 2002). In particular, the high heat storageof urban landscapes associated with high thermal admit-tance and radiation trapping, as well as the added sourcesof anthropogenic heat, need to be incorporated. Tools likesatellite imagery (such as MODIS) or databases of urbanland-use and land classification (LULC), now providefiner spatial resolution of the high heterogeneity of urbancharacteristics (albedo, emissivity or heights of buildings)across cities as input databases for models (Dandou et al.,2005; Jin and Shepherd, 2005). While accuracy in mod-elling the urban climate is of prime importance, featuressuch as ease of use and short running time should alsobe important factors, as urban planners require tools thatincorporate such features.

Recent work in regional scale modelling has seen thedevelopment of a number of urban models of varyingdegrees of complexity based on two types of param-eterization schemes. The first type of scheme involvessimple modifications to existing land surface schemesby modifying or ‘fabricating’ the parameters of theland surface to broadly behave like an urban surface,such as increasing roughness lengths and decreasingalbedo (Atkinson, 2003). One simple parameterizationscheme developed by Grimmond and Oke (2002) is calledthe Localscale Urban Meteorological ParameterizationScheme (LUMPS). Using net all-wave radiation, sim-ple information on surface cover and standard weatherobservations, turbulent and storage heat fluxes can becalculated through a series of linked equations. The equa-tions include the Objective Hysteresis Model (OHM),which uses net all-wave radiation and the surface prop-erties of the site to calculate heat storage (Grimmondet al., 1991. Taha (1999) used a bulk parameterizationapproach to better incorporate heat storage and more

explicitly account for urban canopy layer fluxes, whichalso included the OHM. Similarly, Dandou et al. (2005)made modifications to the thermal part of the fifth-generation Penn state/NCAR Mesoscale Model (MM5)that incorporated the OHM. The model also includedanthropogenic heating, while modifications were alsomade to the dynamical part of the model resulting inacceleration to the diffusion processes during unstableconditions.

The second type of parameterization scheme involvesthe inclusion of a separate urban canopy scheme to theland surface model by incorporating parameters to rep-resent canyon geometry and interactions between thewalls, rooftops, and roads. A number of variations onthis approach have been developed. Some characteristicsof these included using the drag force approach to repre-sent the dynamic and turbulent effects of buildings andvegetation while the thermal modifications of the surfaceinvolve a 3D urban canopy (Dupont et al., 2004; Martilliet al., 2002). This approach calculates the surface tem-perature of each surface type by taking into account theinteractions of shadowing and radiation trapping effects.Single level urban canopy models have also been devel-oped and incorporated into atmospheric models wherethe canopy model simulates turbulent fluxes into theatmosphere at the base of the atmospheric model, param-eterizing both the surface and the roughness sub-layer(Kusaka and Kimura, 2004; Masson, 2000). The TownEnergy Balance model (TEB) is one such scheme andhas been shown to simulate the surface energy balanceand climate well compared with observations (Lemonsuet al., 2004; Masson et al., 2002). A good summary ofurban modelling approaches and developments can befound in Dandou et al. (2005).

As a result of such modifications and developments,the ability of climate models to simulate the urbanclimate has improved, as has their appropriateness asa tool that may aid urban planning. For instance, Taha(1999) modelled effects of increased albedo for all theLULC types in Atlanta (increasing residential albedofrom 0.16 to 0.29 etc.) and showed a decrease in the airtemperature of about 0.5 °C. Atkinson (2003) found thatin London during the day, the albedo, anthropogenic heat,emissivity, SVF, thermal inertia and surface resistance toevaporation (SRE) all aided the formation of an UHI tovarying amounts of between 0.2 and 0.8 °C. SRE wasthe most important factor increasing the UHI intensityduring the day, while the roughness length decreasedintensity. At night, the roughness length, emissivity, SVFand SRE aided UHI formation by 0.3–0.75 °C, but thelargest effect (2 °C) came from anthropogenic heating(Atkinson, 2003). This kind of information is highlyvaluable to urban planners in developing policies forreducing negative climatic impacts to protect vulnerableurban dwellers from the risk of exposure to elevated heatconditions.

Given the growing knowledge and capacity of urbanclimate modelling, this study attempts to investigate therole of climate modelling as a tool for use in urban

Copyright 2008 Royal Meteorological Society Int. J. Climatol. 28: 1943–1957 (2008)DOI: 10.1002/joc

URBAN PLANNING AND CLIMATE IN MELBOURNE 1945

planning and to design and evaluate a suitable tool forMelbourne, Australia. Through the use of a regional scalemodel, The Air Pollution Model (TAPM), the possibleclimatic impacts of a long-term urban planning strategyfor Melbourne, were assessed. Future planning directionsof the strategy aim at encouraging a more compact cityby clustering and increasing the amount of housing inestablished urban areas. Continued urbanization follow-ing existing development patterns is likely to lead to anintensification of the UHI (Coutts et al., 2007b). Usinga modified version of TAPM, we aimed to model theregional climate of Melbourne and its subsequent UHI.Modifications included an improved urban surface param-eterization and an improved land cover input database.Results will highlight to urban planners that the UHI isan issue that needs to be addressed and identify spe-cific areas/regions where planning intervention may berequired. As well as assessing the impact of the urbanplanning strategy, we will comment on the use of regionalscale modelling as a tool in urban planning.

2. Methods

2.1. The urban planning strategy for Melbourne,Australia

The city of Melbourne, Australia is a rapidly growing citywith an anticipated population increase of up to 1 millionpeople by the year 2030, requiring the developmentof approximately 620 000 new households (Departmentof Sustainability and Environment, 2002). In 2002, theVictorian Government introduced a planning strategy toaccommodate this growth titled ‘Melbourne 2030 ’. Thestrategy seeks to achieve a more compact city throughthe development of activity centres (built up centresfor business, shopping, working and leisure with formsof higher density housing) and the establishment ofan urban growth boundary (Figure 1) (Department ofSustainability and Environment, 2002). The anticipateddevelopment of a more compact city, if not plannedin an informed manner, could lead to an exacerbatedUHI intensity. This will be compounded by increased hotweather and hazardous climatic conditions through globalwarming (IPCC, 2007), which will impact especially onvulnerable urban dwellers. Melbourne already shows anUHI signature, with a 20-year mean maximum UHI of2.68 °C found at 6 a.m. (Morris et al., 2001). Duringsummer, anti-cyclonic events often bring warm and dryNorth to Northeast airflow over Melbourne, and cansend temperatures in excess of 35 °C during the day,while mean early morning (6 a.m.) UHI intensity duringthese conditions has been observed at 3.56 °C (Morrisand Simmonds, 2000). These are mean UHI intensities,suggesting that under optimal conditions (clear skiesand low winds) UHI intensity can be much higher. Anautomobile transect across Melbourne in 1992 showeda peak UHI intensity as high as 7.1 °C in the centralbusiness district (CBD) during the evening (9 p.m.)(Torok et al., 2001).

Unplanned and hasty urban development could com-promise the overall goal of the Melbourne 2030 strategy,which aims to achieve a liveable, attractive and prosper-ous city. Cities that are low density and reliant on privatecar transport and strong zoning that separates housing,employment and services are unsustainable. Rather, a sus-tainable city is often described in the urban design liter-ature as compact, high density urban form and supportedby a comprehensive transport network, which empha-sizes connectivity and mixed use developments at criticalnodes (intersecting transport routes) (Mills, 2005). How-ever, this city model can encourage UHI developmentand compromise green-space, potentially threatening theenvironmental quality of the city (Pauleit et al., 2005).Melbourne 2030 aims for a sustainable city and the plan-ning strategy provided a good opportunity to investigatethe use of regional climate modelling in assessing urbanclimate modifications resulting from land-use and plan-ning policies. Our approach consisted of two scenarios:(A) a simulation of the current urban climate and UHIintensity in Melbourne and (B) a year 2030 scenario ofincreased urbanization based on the Melbourne 2030 keydirections to investigate likely future changes to urbanclimate.

2.2. The air pollution model (TAPM) and urbanmodifications

Selecting an appropriate model as a tool for urban climateimpact assessment and use by urban planners is likely todepend on a number of parameters. The accuracy of themodel must be sufficient to robustly simulate the urbanclimate yet not overly complex, computationally expen-sive, and should be user-friendly. Dandou et al. (2005)suggested that despite the simplicity of their bulk urbanparameterization scheme, improvements in results werecomparable with that produced by the complex canopyscheme of Martilli et al. (2002). The ease of use is likelyto be important, and inputs of surface characteristics intothe model should be simply described and readily avail-able, such as through easily obtainable data on typesof surface cover, vegetation cover, albedo, mean build-ing height, anthropogenic heating, and dwelling density.TAPM (Hurley, 2005) was selected for this study as bene-fits included the ability to conduct year-long simulations;the ability to run simulations without surface observa-tional inputs; the ease of a PC-based interface for use inWindows operating systems; user-defined surface coverdatabases; and a range of methods for analysing outputs.Therefore, TAPM has the potential to be adopted as anurban planning tool.

The meteorological component of TAPM is an incom-pressible, non-hydrostatic, primitive equation meteoro-logical model with a terrain following vertical coordi-nate for 3D simulations and a 3D nestable, eulariangrid (Hurley, 2005). As described in Hurley (2005), theprognostic meteorological component solves approxima-tions to the fundamental fluid dynamics equations, andrather than requiring site specific observations, flows suchas sea breezes and terrain flows are predicted against



Figure 1. The Melbourne 2030 compact city with the location of activity centres and the urban growth boundary (Department of Sustainabilityand Environment, 2003) (State of Victoria, Department of Sustainability and Environment, 2003). This figure is available in colour online at

www.interscience.wiley.com/ijoc

a background of larger scale meteorology provided byglobal synoptic analysis. The vertical fluxes are rep-resented by a gradient diffusion approach, including acounter-gradient term for heat flux from turbulence termsdetermined by solving equations for turbulent kineticenergy and eddy dissipation rate. TAPM includes a soil-vegetation-atmosphere transfer (SVAT) scheme, which isused at the model surface, and a radiative flux parame-terization at both the model surface and at upper levelsin the atmosphere.

For urban land surfaces in TAPM, temperature andspecific humidity are calculated depending on the frac-tion of urban surface cover following similar approaches

for non-urban surfaces, except that the surface properties(albedo, thermal conductivity) are given appropriateurban values. The anthropogenic heat flux is also includedin the surface flux equations (Hurley, 2005). A numberof validation studies and evaluations have been con-ducted on TAPM, including one in Melbourne (Hurleyet al., 2003). For the period July 1997 to June 1998,model verification was completed using eight monitor-ing sites across Melbourne. Results showed that the 10-m winds and screen level temperature were predictedvery well with a low average Root Mean Square Error(RMSE) and a high index of agreement (IOA) (Hur-ley et al., 2003). However, TAPM only incorporated a



single homogeneous urban surface class for the entireurban region with single values for land surface parame-ters such as albedo and thermal conductivity derived fromthe literature.

The surface energy balance is simulated in TAPM andconsidered fundamental to an understanding of boundarylayer climates and is basic to an understanding of suchfeatures as thermodynamic behaviour of air and surfacetemperature and humidity, and the dynamics of localairflow (Oke, 1988). Hence, models need to be able toreplicate the partitioning of available energy adequatelyin order to robustly simulate the resultant climate. Theurban surface energy balance is given by (Oke, 1988):

Q∗ + QF = QH + QE + �QS

where Q∗ is net radiation, QF is anthropogenic heating,QH is sensible heat flux, QE is the latent heat flux and�QS is the storage heat flux.

TAPM was modified to improve simulations of urbanenvironments by incorporating four urban land surfacetypes (low, medium and high density, and CBD) replac-ing the existing single urban surface. Surface parameters

in TAPM for the medium density surface type were speci-fied using information from an observational site locatedin suburban Melbourne (Coutts et al., 2007b). The sitewas located in Preston, north of Melbourne (145°0′47′′,37°43′57′′) in a sprawling, moderately developed hous-ing area consisting largely of detached dwellings, typicalof the Melbourne urban landscape (Figure 2). Site char-acterization showed a plan impervious surface cover of62%, which included a plan building area of 45% and hada mean height to width ratio of 0.42. Surface character-istics observed at the site included fraction of urban andvegetation cover (determined from aerial photography);mean building height; anthropogenic heat flux (deter-mined using population, energy, and transport databases);roughness length; and albedo (Table I) and were availableas model input parameters.

Surface energy balance measurements from the Pre-ston site were used to evaluate the performance of themodel for simulated medium density housing in Preston.Observations were taken from instruments mounted ona tall tower at a height of 40 m using the eddy cor-relation technique (Baldocchi et al., 1988) to measurelocal scale fluxes (102 –104 m) of sensible and latent

Figure 2. The medium density observational study site in Preston, located north of Melbourne CBD. This figure is available in colour online atwww.interscience.wiley.com/ijoc

Table I. The original urban surface characteristics from Preston are given along with the values assigned in the model foreach level of urban density: fraction of urban cover σu; albedo αu; anthropogenic heat flux Au (W.m−2); thermal conductivityku (W.m−1.K−1); roughness length z0u (m); building height zH (m); fraction of non-urban area covered by vegetation σf; leaf

area index LAI; minimum stomatal resistance rsi (s−1.m−1).

σu αu Au ku zou zH σf LAI rsi

Observed (Preston) 0.62 0.15 9–12 – 0.4 12 – – –TAPM (3.0.2) default 0.5 0.15 30 4.6 1 10 0.75 2 100Urban (low) 0.5 0.17 10 15 0.4 8 0.75 2 100Urban (medium) 0.65 0.15 15 25 0.6 12 0.75 2 100Urban (high) 0.8 0.13 20 40 0.8 16 0.75 2 100Urban (CBD) 0.95 0.1 40 60 2 100 0.75 2 100



heat and momentum. Radiation sensors measured eachcomponent of the radiation balance, giving net radiation.The storage heat flux was calculated as a residual of theenergy balance equation. Observations of temperature,vapour pressure, wind speed, and friction velocity werealso conducted. Model outputs matching the location ofthe Preston site were compared with the observed sur-face energy balance and meteorological parameters andprovided information to evaluate how well TAPM repli-cated the surface energy balance and simulated the localurban climate at the observational site. This site was oneof three urban flux sites operating at various times andlocations in Melbourne during 2003–2004 (Coutts et al.,2007b).

The urban land surface characteristics for Urban(medium) (Table I) were set to be very similar to themedium density observational site described (Preston).The low, high, and CBD urban land surface characteris-tics were then assigned with reference to the values ofUrban (medium) (Table I) using expert knowledge of theMelbourne urban landscape from the observational cam-paign (Coutts et al., 2007b) and other literature values.However, in order to replicate the storage capacity ofa complex 3D urban surface in a one-dimensional sur-face scheme the thermal conductivity was substantiallyincreased in the model. Using the value of a componentmaterial such as concrete in bulk model parameterizationsdoes not capture the full influence of the heterogeneousurban landscape or the effects of the urban canopy. Sug-awara et al. (2001) created a thermal property param-eter (combining the product of specific heat and ther-mal conductivity) that better represented urban surfaces.This parameter was determined to be much larger than ahomogenous surface type such as asphalt and concludedthat the value should be a few times larger than the com-ponent material in bulk urban models that do not dealwith canyon shape explicitly (Sugawara et al., 2001). InTAPM, the thermal conductivity value for the land sur-face was modified in order to match the storage heatflux results from TAPM with the observational results atthe medium density site in Preston during January. Thethermal conductivity needed to be increased well aboverealistic values before the surface began behaving simi-larly to a ‘real’ urban surface, identifying the importanceof canyon geometry in trapping and storing energy.

2.3. Model configuration and database development

Model scenarios for the current and year 2030 scenarioswere performed for January during the Austral summer,as urban residents are exposed to higher ambient tempera-tures at this time. Large scale synoptic analyses were usedto force the model between the periods 1997 and 2004.Synoptic scale forcing meteorology was provided fromthe 6-hourly Limited Area Prediction Systems (LAPS)(Puri et al., 1998) at a 0.75° grid spacing. These 8 yearsof January simulations were conducted so that modelleddifferences were due to land surface changes and notdue to year-to-year climate variability. Moreover, the

same forcing data were used for each experiment. Themodified TAPM version 3.0.2 was configured with threenested grids of 110 × 110 horizontal points with the innergrid encompassing the Melbourne metropolitan area at agrid spacing of 1000 m centred at 145°9′E and 37°59′S.The middle and outer grid spacings were 3000 m and10 000 m, respectively, and 25 vertical grid levels wereselected with the highest level at 8000 m. Other databasesof terrain height (9-s DEM), sea surface temperature andsoil classification data, were also used in the scenarios(Hurley, 2005).

In order to run the scenarios described, relevantland surface databases were developed at a suitableresolution for input into TAPM. For the current dayscenario (Scenario A) a vegetation (land-use) databasewas obtained that provided recent vegetation cover (1988)(Geoscience Australia, 2003). In addition, a surfacedatabase of low, medium, and high density areas, aswell as the CBD, was constructed. Information on censusdistricts for the entire Melbourne metropolitan area werecollected (Australian Bureau of Statistics (ABS), 2001)and the dwelling density calculated for each district(dwellings per km2). This information was converted tomean dwelling density for 0.01 decimal degree grid cells(approximately 1 km). Plans for Melbourne 2030 aim toincrease the average housing density significantly from1000 dwellings per km2 to an average of 1500 dwellingsper km2 (Department of Sustainability and Environment,2002). Therefore, high density areas were deemed to begreater than 15 dwellings per hectare, medium densityareas between 10 and 15 dwellings per hectare and lowdensity areas less than 10 dwellings per hectare (thoughgreater than 1) (Figure 3). This database was overlain onthe vegetation database and used as input into TAPM.

The database for the Melbourne 2030 scenario (Sce-nario B) was based on the document’s key directionsas discussed earlier. Taking the current urban densitydatabase, the urban growth boundary was added and thoseareas not currently developed within the urban growthboundary were assigned to the low density class. Thelocation of the proposed 26 Principal, 82 Major, and 10Specialized activity centres were then added, by assum-ing that the surrounding housing for a 1-km radius wouldbe high density (within walking distance). Housing withinanother 1-km radius was anticipated to increase to at leastmedium density while existing high density housing areasand the CBD areas remained as such (Figure 3).

3. Results and Discussion

3.1. Evaluating TAPM against urban surface energybalance observations

Using the new land surface database of the current Mel-bourne urban landscape, TAPM was run for the monthof January and compared with the observations at themedium density observational site (Preston) (Figure 4).TAPM showed a good performance in replicating thediurnal course and monthly mean surface energy balance



Figure 3. Land surface database of current density scenario A (left) and the Melbourne 2030 scenario B with the urban growth boundary (right).This figure is available in colour online at www.interscience.wiley.com/ijoc

Figure 4. Comparison of the observed (dashed line) and modelled (solid line) diurnal surface energy balance (observed height 40 m and modellevel 50 m) for location (145°0′47′′, 37°43′57′′) and corresponding grid point (043, 084) for the month of January 2004. Regression (fit) equationswere Q∗ (y = 1.032x + 58.626); QH (y = 1.137x + 30.017); QE (y = 0.789x + 18.819); �QS (y = 0.833x + 14.007). This figure is available

in colour online at www.interscience.wiley.com/ijoc

for the month of January 2004. The evaporative flux wasreplicated well by the model, only showing an overesti-mation in the afternoon. The storage heat flux was alsowell replicated, although some discrepancy was evidentin both Q∗ and QH. This is caused by an overestimationof incoming solar radiation due to the inability of themodel to capture cloudy skies and poor air quality (which

can reflect and scatter incoming short wave radiation),so monthly averages of Q∗ were overestimated. Thisextra energy led to additional partitioning into QH. Onthe majority of the January days, the TAPM model per-formed well. The model also captured important featuresof urban energy balance partitioning (Figure 4). Theseincluded the hysteresis pattern in �QS, showing a peak



approximately 1–2 h before the peak in Q∗. The peakwas not as evident in the observations during this monthas what was generally seen during the observational cam-paign (Coutts et al., 2007b). The asymmetry in QH wasalso evident with the peak occurring later in the after-noon. Importantly, both QH and QE remained positiveinto the evening, supported by heat storage release fromthe urban fabric, and remained slightly positive through-out the night.

Despite the discrepancy in Q∗ and QH, the modelaccurately simulated temperature, relative humidity, andwind speed (Figure 5). Slight discrepancies were seenin the diurnal temperature plot, with a reduced lag intemperature approaching its maximum and the nighttimetemperatures were underestimated, due to underestima-tion of nighttime heat storage release. Table II gives theJanuary 2004 monthly comparison of modelled meteoro-logical variables and their associated error for the modelgrid point corresponding with the measurement tower

location, compiled following Willmott (1981). Statisticalcomparisons are also given for the surface energy balancecomponents.

Changes in urban surface characteristics influence hownet radiation is partitioned into each of the surface energybalance components, so the flux ratios and how theyvary between density classes were of particular interest(Figure 6). Also, while the summer month (January) wasof primary interest, there was also observational dataavailable for a full year (Coutts et al., 2007b) and itwas possible to see how well the model reproducedthe partitioning of the urban surface energy balanceseasonally (Figure 6). Therefore, TAPM was also runfrom August 2003 to July 2004 corresponding with theyear-long observational study. Naturally, partitioning inJanuary was good as the model parameters for the urbansurface characteristics were adjusted to match this data,yet over the course of the year, the model did notcapture �QS and QH well. A reasonable replication

Figure 5. Comparison of observed (left) and modelled (right) temperature, relative humidity, and wind speed (observed height 40 m, model level50 m) for location (145°0′47′′, 37°43′57′′) and corresponding grid point (043, 084) for the month of January 2004.

Table II. Statistical comparison between variables for the observational location (145°0′47′′, 37°43′57′′) and corresponding modelgrid point (043, 084) for the month of January 2004 of temperature T (°C); wind speed WS (m/s); specific humidity q (g/kg);friction velocity u∗ (m/s); sensible heat flux QH (W.m−2); latent heat flux QE (W.m−2); and storage heat flux �QS (W.m−2).

T WS q u∗ Q∗ QH QE �QS

n 744 744 744 744 744 744 744 744O 16.35 4.74 7.21 0.40 146.26 88.01 40.81 17.43P 16.00 4.27 6.93 0.44 209.58 130.06 51.00 28.52sO 3.76 2.33 1.73 0.21 267.22 116.34 45.58 127.61sP 4.03 1.92 1.38 0.23 307.93 151.91 47.88 131.87CORR 0.89 0.79 0.73 0.79 0.90 0.87 0.75 0.81RMSE 1.84 1.50 1.21 0.15 151.10 87.24 34.61 81.71RMSES 0.39 0.94 0.75 0.07 63.90 44.95 14.03 24.06RMSEU 1.80 1.17 0.95 0.13 136.92 74.76 31.64 78.09MAE 0.35 0.47 0.27 −0.04 −63.33 −42.05 −10.19 −11.09d 0.94 0.86 0.83 0.87 0.93 0.89 0.85 0.89r2 0.80 0.63 0.53 0.63 0.80 0.76 0.56 0.65

n, number of observations; O, observations; P , predicted values; sO sP, observed and predicted standard deviations; CORR, Pearman Correlation;RMSE, Root Mean Square Error; RMSES, Systematic RMSE; RMSEU, Unsystematic RMSE; d, Index of Agreement; r2, Coefficient ofdetermination (Willmott, 1981).



Figure 6. Mean monthly plots of daytime fractions of Q∗ for each energy balance component and the Bowen Ratio. CBD, HIGH, MEDIUM, andLOW correspond to each of the urban classes and OBSERVATIONS correspond to the measured data from Preston. The Bowen Ratio (QH/QE)

is not shown for the CBD as it was significantly higher (≈20).

of observations for the evaporative fraction (QE/Q∗)was seen over the course of the year. Figure 6 alsodemonstrates the differences in partitioning of energybetween each of the urban land surface classes in themodel and shows that the influence of changing the landsurface values alters energy balance partitioning. Themodelled data for these urban classes were not verifiedagainst any observations.

Some differences in energy partitioning over the courseof the year could result from a number of uncertain-ties. Actual deep soil volumetric moisture contents werenot available to initialize the model and we foundthat there was a mismatch in the seasonal course ofQE/Q∗ between the observations and the model out-put (Figure 6). In the model, moisture contents werethe lowest over the Austral summer months (Decem-ber–February). Rainfall in Melbourne during Februaryand March 2004, however, was well below average, so itwas likely that deep soil volumetric moisture contents

were also below average at this time, leading to thereduced energy partitioning into QE. More accurate val-ues of monthly soil moisture content could improve thisresult. As expected, QE/Q∗ decreased with increasingurban density as the vegetated surface cover was replacedwith greater impervious surface cover, restricting evapo-transpiration. Generally, the partitioning of energy intoQE was acceptable over the course of the year andresponded well to the changes in surface cover.

The modelled Urban (medium) heat storage fraction(�QS/Q

∗) during the summer period generally showeda slight underestimation compared with the observations,but were much improved compared with earlier versionsof TAPM. The substantial increases of the values forthermal conductivity in the model were large enough tocapture the significant energy storage by the 3D urbanlandscape. Comparing each of the densities, the amountof heat storage increased with increasing density. How-ever, absorption of energy by the urban surface in the



model appeared to saturate (reach a maximum absorptivecapacity) when increasing thermal conductivity. There-fore, despite the increasing thermal conductivity withurban density, the 1D land surface in the model did notcapture the full influence of the 3D canyon morphologyon heat storage.

During winter, the land surface scheme did not repli-cate the high heat absorption (�QS/Q

∗) by the urbanfabric that was seen in the observations (Figure 6). Inmost Northern Hemisphere energy balance studies, adecrease in heat storage is seen during winter, followingthe reduction in Q∗, as well as added surface moisturefor increased QE/Q∗ (Grimmond, 1992), a pattern thatTAPM did replicate. However, in this case, it may notbe that TAPM inaccurately represented urban heat stor-age, but rather the uncertainty may lie in the observations.Spronken-Smith et al. (2006) found in Christchurch, NewZealand that under settled anti-cyclonic conditions, astrong inversion can occur that can severely restrict tur-bulent mixing and influence the above canopy flux mea-surements. As the observations in Melbourne calculated�QS as a residual of the eddy correlation technique, itcould be that the observational results over emphasizethe importance of heat storage during stable wintertimeconditions and is an area that requires further study.

On account of the slightly underestimated �QS/Q∗,

the sensible heating fraction (QH/Q∗) during the sum-mertime for the Urban (medium) density was also slightlyhigher than observed. The partitioning of energy was verysimilar for each urban density during summer, though allsites were slightly higher than the observations. This isoften why above canopy temperatures are similar acrossan urban area during the day, as higher density sitesabsorb more energy and �QS/Q

∗ increases, restrictingthe availability for atmospheric heating, which sometimesaids Urban Cool Island (UCI) development in combina-tion with shading (Morris and Simmonds, 2000). TheBowen ratio (QH/QE) throughout the year was wellreplicated compared with Urban (medium) and increasedwith higher urban density (Figure 6). The Bowen ratiofrom the model results also preceded the observationsagain as a result of the lack of input data for the soilmoisture content and the influence of this on QE.

The model was not able to accurately replicate thepartitioning of energy outside of the summer months.However, as TAPM was replicating the partitioning ofenergy and meteorological parameters at the surface rea-sonably well in January, it can be used for the scenariosdescribed with a good degree of confidence. While acrude method of parameterizing the model to behavemore like an urban surface was used, and direct validationwas not completed on the energy balance partitioning,the model has vastly improved on the performance ofTAPM version 2.0 before the modifications were made(data not shown). Additionally, the model was only eval-uated for the medium density urban class, so there maybe limitations in the model’s applicability to other urbandensity classes. The lack of an urban canopy scheme

could also limit the model’s capacity to accurately repli-cate urban heat storage across density classes. There isobvious scope for a specific urban parameterization inTAPM.

3.2. Modelling UHI intensity and the impact ofMelbourne 2030

TAPM was configured as described in Section 2.3 andrun for eight Januaries from 1997 to 2004 to provide anensemble average for current summertime conditions andthen again for the 2030 planning scenario. The currentscenario (A) showed a mean nighttime (0200) UHI ofapproximately 3–4 °C in the CBD, reducing as distancefrom the CBD increased (Figure 7(a)). Variability washigh with anomalous warmer and cooler areas seen acrossthe metropolitan area corresponding with urban densityclass. The modelled UHI intensity was similar in rangeto those previously observed in Melbourne (Morris andSimmonds, 2000; Morris et al., 2001). During the day(1400) the current Scenario (A) screen level UHI was notas intense as at 0200, but still showed an UHI intensityof 1–2 °C, with temperatures being more uniform acrossthe region (Figure 7(b)). The CBD was not warmer thanthe surrounding urban area. Temperatures away from thecoast to the north and east of Port Phillip Bay showedhigher values as a result of mesoscale airflows and aregional sea breeze. During the night, the lower windspeeds reduced the influence of the regional flows and theurban density more strongly controlled the developmentof the UHI. The modelled UHI also varied with synopticconditions that supported maximum UHI developmentunder conditions of anti-cyclonic highs centred just eastof Melbourne, low wind speeds and cloudless skies(Morris and Simmonds, 2000). The Melbourne 2030scenario (B) revealed a slightly modified UHI patternfrom the current scenario (A) (Figure 7(c) and (d)).While the maximum intensity of the UHI did not increase,the areal extent of elevated temperatures expanded. Thenighttime UHI reduced in its spatial variability, becomingmore uniform across the urban area similar to that of thedaytime UHI.

Analysing the difference in screen level tempera-ture between the current and Melbourne 2030 scenariosallows specific areas of significant warming to be iden-tified and is what urban planners are most interested in.The extent of change in the UHI resulting from planningstrategies shows areas that are particularly vulnerable.This information can be used for improved planning deci-sions. The greatest temperature increases during night-time maximum UHI intensity (Figure 8) were seen inareas where development replaced pasture land and innew activity centres. Temperatures in other areas of Mel-bourne also appeared to respond significantly to increasesin housing density especially along the edge of the currenturban-rural boundary. Some of these areas are locatedalong transport links and ‘growth areas’ designated forconcentrated expansion as outlined in Melbourne 2030.While these areas are likely to show the greatest increasein temperatures in 2030, temperatures were only seen to



Figure 7. Spatial variability in mean screen level temperatures for the Melbourne area at 02 : 00 (2 a.m.) and 14 : 00 (2 p.m.) h for each scenario:A – Current development at 02 : 00 (a) and 14 : 00 (b); B – Melbourne 2030 planned development at 02 : 00 (c) and 14 : 00 (d). This figure is

available in colour online at www.interscience.wiley.com/ijoc

increase to levels currently seen within the CBD. Initia-tives that can help reduce temperature increases can bemore easily incorporated into newly developing regions,rather than in existing urban development. Therefore,these growth areas and new or minimally developed exist-ing activity centres could provide excellent opportunitiesfor UHI mitigation strategies to be put in place.

During the day, some portions of Melbourne to thewest and north showed elevated temperatures followingthe planned development (Figure 9). Interestingly, duringthe day a large fraction of the urban area, mainly wheredevelopment increases from low to higher densities, actu-ally showed a very slight decrease (largely insignificant)in temperature due to the increased heat storage limitingthe amount of energy available for atmospheric heatingand reducing temperatures. The areas of greatest tempera-ture increase were the planned growth areas where devel-opment will replace existing natural landscapes. While it

may seem that these mean temperatures are not high,during periods in summer of extreme heat, temperaturescan be much higher. While higher nighttime tempera-tures from restricted nocturnal cooling in urban areas maynot seem like a significant problem, extended periods ofwarmer temperatures can limit nighttime recovery fromdaily heat stress. Inland activity centres that do not feelthe effects of the cooling sea breeze would especiallybenefit from UHI mitigation measures.

The planned increase in urban density through theestablishment of an urban growth boundary and thedevelopment of activity centres in Melbourne will likelylead to a more intense UHI during the night, while duringthe day this is less significant. Coutts et al. (2007b) intheir observational study in Melbourne found that duringthe summer across three urban sites of varying urbandensity, all sites showed a mean daytime Bowen ratioof over 2 and the daily Bowen ratio was sometimes



Figure 8. Change in mean nighttime (02 : 00) screen level temperature change from the current urban development, to that proposed by theMelbourne 2030 planning strategy. Areas within the contours are statistically significant at the 95% confidence level. This figure is available in

colour online at www.interscience.wiley.com/ijoc

Figure 9. Change in mean daytime (14 : 00) screen level temperature from the current urban development, to that proposed by the Melbourne2030 planning strategy. Areas within the contours are statistically significant at the 95% confidence level. This figure is available in colour

online at www.interscience.wiley.com/ijocCopyright 2008 Royal Meteorological Society Int. J. Climatol. 28: 1943–1957 (2008)

DOI: 10.1002/joc


in excess of 5. The increase in Bowen ratios withincreasing urban density modelled in this study werenot found in the observational results as evaporativefluxes were very similar across all housing densitiesdespite varying vegetation cover. This was a result ofpoor moisture availability in response to water restrictionswhen observations were conducted (Coutts et al., 2007b).Therefore during the summer, the entire Melbourneregion experienced warm, dry and hence unfavourableclimatic conditions. Adoption of the Melbourne 2030strategy is not likely to increase Bowen ratios acrossthe city significantly, but there will be an extension ofwarm and dry conditions over longer periods of theday as well as an extension of the seasonal exposure tounfavourable conditions along with an increased spatialextent, especially if water restrictions remain tight.

In this study, the effect of heat trapping and storage inthe urban environment was replicated in the simulationsby significantly increasing the thermal conductivity. The3D nature and complexity of the urban landscape wasnot explicitly included in the model, unlike new urbancanopy schemes. As a result, the model could not deliverwithin-canopy temperatures, which could possibly begreater than the modelled temperatures in this study.Modelled screen level temperatures were also slightlyunderestimated during the evening and night due to anunderestimation of the slow release of heat stored inthe urban fabric due to complex canyon morphology(including walls). Finally, the Melbourne 2030 scenario(B) only accounts for climatic impacts from land coverchange. Mean global temperatures over the last 100 years(1906–2005; 100-year linear trend) have increased by0.74 °C (±0.18 °C) largely as a result of carbon dioxide(CO2) emissions (IPCC, 2007), so projected globaltemperature rises (0.2 °C per decade for the next twodecades (IPCC, 2007)) coupled with heating from furtherurban development will lead to further increases in urbantemperatures. Also, the frequency of extreme warm daysand nights has increased since 1961 (Plummer et al.,1999). Urban areas themselves are a significant source ofCO2 mostly from vehicles emissions with local annualemissions from urban Melbourne as high as 84.9 tCO2 ha−1 y−1 (Coutts et al., 2007a). Urban planningmeasures such as energy-efficient buildings and increasedpublic transport use would help contribute to combatinggreenhouse gas emissions.

4. Conclusions

Simulations of the changes in climate resulting from theproposed land cover changes identified in the directionsof the Melbourne 2030 plan showed that continuedincreases in density would result in an increased intensityof the nighttime Melbourne UHI. Growth areas andparticular activity centres were predicted to have thegreatest temperature increases. During the day, the impactof changes in urban development was not seen toincrease the peak daytime temperature due to increased

storage limiting the amount of sensible heating of theatmosphere. Yet, existing urban climates during summerdays can already be unfavourable with high Bowenratios regularly observed across varying densities of thecity (Coutts et al., 2007b). These results demonstratethe utility of regional scale climate modelling as atool for climate impact assessment and show the abilityto determine likely climate modifications from simpleland-use changes based on planning directions. Theuse of TAPM for the Melbourne urban landscape wasadequate for January, and identified that continued urbandevelopment in Melbourne could lead to higher diurnalexposure to warmer temperatures. Modelling results suchas these are an excellent way to present and conveyinformation and issues to environmental planners.

Planning in urban areas to ameliorate, and limit thedevelopment of degraded local climates has been knownfor decades (Aron, 1984; Oke, 1984; Oke, 2005), yet pol-icy development in this area is still lacking despite callsfor improvements. The concept of sustainable settlementsis recognized within Melbourne 2030 with initiativessuch as those under the direction of ‘A greener city’,including reducing the impacts of storm-water on baysand catchments, and the management of water resources(Department of Sustainability and Environment, 2002).Melbourne 2030 currently notes concern for issues suchas global warming and a livable city but an assessmentof the impact of a more compact city on climate had notbeen undertaken. Our analysis should persuade the devel-opment of new policies for UHI mitigation by planners.This work may be opportune since the Melbourne 2030plan is due for review in 2007. Some initiatives alreadyexist that aid in reducing UHI intensity include energy-efficient buildings and encouraging a shift in travel fromprivate vehicles to public transport, which will reduceanthropogenic heat emissions. While this is good, a com-prehensive UHI mitigation strategy for Melbourne isrequired and it is hoped that this study will prompt theMelbourne 2030 planning group to act and encourage theimplementation of UHI mitigation initiatives. It would bea great opportunity for the Victorian Government, whowish to lead by example in environmental management(Department of Sustainability and Environment, 2002).

Regional scale modelling of urban climate is a pow-erful tool and the use of TAPM as a model for usein urban planning has both benefits and shortfalls. AsTAPM is now set up for Melbourne, further summertimescenarios could be conducted to investigate the poten-tial of mitigation strategies such as alterations in surfacealbedo or the effect of increasing vegetation cover. Also,any type of urban spatial configuration at the neighbour-hood scale could also be easily modelled. This study hasdemonstrated the potential for TAPM to become a rig-orous model for use in urban planning. However, muchimprovement is still required before it could be com-monly used. Operating the model for other Australianor international cities may not be feasible without somemodification of surface parameters (requiring local fieldobservations) or development of new parameterizations.



Also, as the winter performance of the model was poor,further improvements to TAPM would be required beforeit could be used as a year round urban climate modelfor climate impact assessment or forecasting. On accountof the high spatial resolution used in these scenarios,compute time was long and could be improved on high-performance machines.

TAPM does not currently have any urban canopyscheme and so does not explicitly resolve canyon geom-etry effects and hence there is a lot of room for improve-ments through new urban parameterizations. This high-lights the need for further developments in urban climatemodelling within the Australian modelling community,especially as the focus on extreme temperatures in urbanregions grows. The inclusion of an urban canopy schemewould also then permit modelling of diurnal tempera-ture scenarios, rather than just the monthly averages forJanuary presented here. A complex urban canopy modelwith the simplicity and ease of use of TAPM would bean ideal product. The UHI was predicted well comparedwith previous observed ranges of UHI intensity. Futurework could also involve coupling TAPM with a wateruse model such as Aquacycle (Mitchell et al., 2001),which could allow investigation of climate interactionwith human water consumption and irrigation, which isvery important (and ever growing in importance), in theMelbourne urban landscape.

Despite the ease of use of the PC-based TAPM,substantial resources are required to both understandand run the model, and to develop the urban databasesfor their application in the simulations. The potentialoperation of climate models directly by urban plannersmay be unachievable currently and therefore climateimpact studies of urban development scenarios are bestout-sourced to urban climatologists. While continuedmodel improvements and validation are still needed, eventhe best urban climate model would need to be run bythose who know how to use it. An inter-disciplinaryand team-based approach is imperative in order forthis to be effective (Oke, 2005). As a result, plannersand climatologists must work together utilizing theirfull knowledge and allowing the development of moreaccurate urban climate predictions.

Acknowledgements

Thanks to Peter Hurley for assistance and the conductionof modifications to the model. Thanks to Peter Wallace ofP. G. Wallace Communications for permission of the useof the communications tower for the field observationsand to Christopher Barker for assistance in setting upand maintenance of the towers and equipment. The loanof instrumentation by Lindsay Hutley (Charles DarwinUniversity) and Russell Jaycock (James Cook University)is also greatly appreciated.

ReferencesAron J. 1984. Climatic consciousness–a conscious change in the

Australian architectural environment. Energy and Buildings 7:71–75.

Atkinson BW. 2003. Numerical modeling of Urban heat-islandintensity. Boundary-Layer Meteorology 109: 285–310.

Australian Bureau of Statistics (ABS). 2001. 2001 Census, Australia.Baldocchi DD, Hicks BB, Meyers TP. 1988. Measuring biosphere-

atmosphere exchanges of biologically related gases with microm-eteorological methods (in special feature: gas exchange in a newdimension). Ecology 69: 1331–1340.

Coutts AM, Beringer J, Tapper NJ. 2007a. Characteristics influencingthe variability of urban CO2 fluxes in Melbourne, Australia.Atmospheric Environment 41: 51–62.

Coutts AM, Beringer J, Tapper NJ. 2007b. Impact of increasing urbandensity on local climate: spatial and temporal variations in thesurface energy balance in Melbourne, Australia. Journal of AppliedMeteorology 46: 477–493.

Dandou A, Tombrou M, Akylas E, Soulakellis N, Bossioli E. 2005.Development and evaluation of an urban parameterization schemein the Penn State/NCAR Mesoscale Model (MM5). Journal ofGeophysical Research-Atmospheres 110: 10102.

Department of Sustainability and Environment. 2002. Melbourne 2030:Planning for Sustainable Growth, State of Victoria, 192.

Department of Sustainability and Environment. 2003. Page 1 ofAddendum: Figure of Activity Centers and Principle PublicTransport Network 2003 Only, State of Victoria, 2.

Dupont S, Otte TL, Ching JKS. 2004. Simulation of meteorologicalfields within and above urban and rural canopies with a mesoscalemodel (MM5). Boundary-Layer Meteorology 113: 111–158.

Eliasson I. 2000. The use of climate knowledge in urban planning.Landscape and Urban Planning 48: 31–44.

Fehrenbach U, Scherer D, Parlow E. 2001. Automated classification ofplanning objectives for the consideration of climate and air qualityin urban and regional planning for the example of the region ofBasel/Switzerland. Atmospheric Environment 35: 5605–5615.

Geoscience Australia. 2003. Vegetation–Post-European Settlement,1988. National Mapping Division.

Grimmond CSB. 1992. The suburban energy balance: methodologicalconsiderations and results for a mid-latitude west coast city underwinter and spring conditions. International Journal of Climatology12: 481–497.

Grimmond CSB, Oke TR. 2002. Turbulent heat fluxes in urban areas:observations and a local-scale urban meteorological parameterizationscheme (LUMPS). Journal of Applied Meteorology 41: 792–810.

Grimmond CSB, Cleugh HA, Oke TR. 1991. An objective urban heat-storage model and its comparison with other schemes. AtmosphericEnvironment Part B-Urban Atmosphere 25: 311–326.

Hurley PJ. 2005. CSIRO Atmospheric Research Technical Paper No.71, 54.

Hurley P, Manins P, Lee S, Boyle R, Ng YL, Dewundege P. 2003.Year-long, high-resolution, urban airshed modeling: verification ofTAPM predictions of smog and particles in Melbourne, Australia.Atmospheric Environment 37: 1899–1910.

IPCC. 2007. Summary for Policymakers. In Climate Change 2007:The Physical Science Basis. Contribution of Working Group I to theFourth Assessment Report of the Intergovernmental Panel on ClimateChange, Solomon S, Qin D, Manning M, Chen Z, Marquis M,Averyt KB, Tignor M, Miller HL (eds). Cambridge University Press:Cambridge and New York.

Jackson LE. 2003. The relationship of urban design to human healthand condition. Landscape and Urban Planning 64: 191–200.

Jin M, Shepherd JM. 2005. Inclusion of urban landscape in a climatemodel–How can satellite data help? Bulletin of the AmericanMeteorological Society 86: 681–689.

Kusaka H, Kimura F. 2004. Coupling a single-layer urban canopymodel with a simple atmospheric model: impact on urban heatisland simulation for an idealized case. Journal of the MeteorologicalSociety of Japan 82: 67–80.

Lemonsu A, Grimmond CSB, Masson V. 2004. Modeling the surfaceenergy balance of the core of an old Mediterranean city: Marseille.Journal of Applied Meteorology 43: 312–327.

Martilli A, Clappier A, Rotach MW. 2002. An urban surfaceexchange parameterisation for mesoscale models. Boundary-LayerMeteorology 104: 261–304.

Masson V. 2000. A physically-based scheme for the urban energybudget in atmospheric models. Boundary-Layer Meteorology 94:357–397.



Masson V, Grimmond CSB, Oke TR. 2002. Evaluation of the TownEnergy Balance (TEB) scheme with direct measurements fromdry districts in two cities. Journal of Applied Meteorology 41:1011–1026.

Mills G. 2005. Progress toward sustainable settlements: a role for urbanclimatology. Theoretical and Applied Climatology 84: 69–76.

Mitchell VG, Mein RG, McMahon TA. 2001. Journal of environmentalmodeling and software. Modeling the Urban Water Cycle 16:615–629.

Morris CJG, Simmonds I. 2000. Associations between varyingmagnitudes of the urban heat island and the synoptic climatologyin Melbourne, Australia. International Journal of Climatology 20:1931–1954.

Morris CJG, Simmonds I, Plummer N. 2001. Quantification of theinfluences of wind and cloud on the nocturnal urban heat islandof a large city. Journal of Applied Meteorology 40: 169–182.

Oke TR. 1984. Towards a prescription for the greater use of climaticprinciples in settlement planning. Energy and Buildings 7: 1–10.

Oke TR. 1988. The Urban energy balance. Progress in PhysicalGeography 12: 471–508.

Oke TR. 2005. Towards better scientific communication in urbanclimate. Theoretical and Applied Climatology 84: 179–190.

Pauleit S, Ennos AR, Golding Y. 2005. Modeling the environmentalimpacts of urban land use and land cover change–a study inMerseyside, UK. Landscape and Urban Planning 71: 295–310.

Plummer N, Salinger MJ, Nicholls N, Suppiah R, Hennessy KJ,Leighton RM, Trewin B, Page CM, Lough JM. 1999. Changes inclimate extremes over the Australian region and New Zealand duringthe twentieth century. Climatic Change 42: 183–202.

Puri K, Dietachmayer GS, Mills GA, Davidson NE, Bowen RA,Logan LW. 1998. The new BMRC limited area prediction system,LAPS. Australian Meteorological Magazine 47: 203–223.

Rankin DW. 1959. Mortality associated with heat wave conditionsin the Melbourne metropolitan area, January and February, 1959.Australian Meteorological Magazine 26: 96–98.

Smoyer-Tomic KE, Kuhn R, Hudson A. 2003. Heat Wave Hazards: anoverview of heat Wave impacts in Canada. Natural Hazards 28:463–485.

Spronken-Smith RA, Kossmann M, Zawar-Reza P. 2006. Where doesall the energy go? Surface energy partitioning in suburbanChristchurch under stable wintertime conditions. Theoretical andApplied Climatology 84: 137–150.

Stone B. 2005. Urban heat and air pollution–an emerging role forplanners in the climate change debate. Journal of the AmericanPlanning Association 71: 13–25.

Sugawara H, Narita K, Mikami T. 2001. Estimation of effectivethermal property parameter on a heterogeneous Urban surface.Journal of the Meteorological Society of Japan 79: 1169–1181.

Taha H. 1999. Modifying a mesoscale meteorological model to betterincorporate urban heat storage: A bulk-parameterization approach.Journal of Applied Meteorology 38: 466.

Torok SJ, Morris CJG, Skinner C, Plummer N. 2001. Urban heat islandfeatures of southeast Australian towns. Australian MeteorologicalMagazine 50: 1–13.

Willmott CJ. 1981. On the validation of models. Physical Geography2: 184–194.

Zehnder JA. 2002. Simple modifications to improve fifth-generationPennsylvania State University-National Center for AtmosphericResearch Mesoscale Model performance for the Phoenix,Arizona, metropolitan area. Journal of Applied Meteorology 41:971–979.


Documents

Journal_Investigating the Climatic Impact of Urban Planning