IOP Conference Series: Materials Science and Engineering

PAPER • OPEN ACCESS Related content

- Urban Heat Island towards Urban Climate
Key Parameters for Urban Heat Island Widya Ningrum

Assessment in A Mediterranean Context: A - Urban Heat Island Effect on the Energy

Consumption of Institutional Buildings in
Rome
Sensitivity Analysis Using the Urban Weather Claudia Calice, Carola Clemente, Agnese
Salvati et al.
Generator Model - Urban climate effects on extreme
temperatures in Madison, Wisconsin, USA
Jason Schatz and Christopher J Kucharik
To cite this article: Agnese Salvati et al 2017 IOP Conf. Ser.: Mater. Sci. Eng. 245 082055

View the article online for updates and enhancements.

This content was downloaded from IP address 151.97.7.30 on 10/05/2018 at 08:14

WMCAUS IOP Publishing
IOP Conf. Series: Materials Science and Engineering 245 (2017) 082055 doi:10.1088/1757-899X/245/8/082055
1234567890

Key Parameters for Urban Heat Island Assessment in A

Mediterranean Context: A Sensitivity Analysis Using the
Urban Weather Generator Model

Agnese Salvati 1,2, Massimo Palme 3, Luis Inostroza 4, 5

1
Sapienza University of Rome, DICEA department, Roma, Italy
2
Polytechnic University of Catalunya, School of Architecture of Barcelona, Spain
3
School of Architecture, Catholic University of the North, Antofagasta, Chile
4
Institute of Geography, Ruhr-Universität Bochum, Bochum, Germany
5
Universidad Autonoma de Chile, Chile

agnese.salvati@uniroma1.it

Abstract. Although Urban Heat Island (UHI) is a fundamental effect modifying the urban
climate, being widely studied, the relative weight of the parameters involved in its generation is
still not clear. This paper investigates the hierarchy of importance of eight parameters responsible
for UHI intensity in the Mediterranean context. Sensitivity analyses have been carried out using
the Urban Weather Generator model, considering the range of variability of: 1) city radius, 2)
urban morphology, 3) tree coverage, 4) anthropogenic heat from vehicles, 5) building’s cooling
set point, 6) heat released to canyon from HVAC systems, 7) wall construction properties and 8)
albedo of vertical and horizontal surfaces. Results show a clear hierarchy of significance among
the considered parameters; the urban morphology is the most important variable, causing a
relative change up to 120% of the annual average UHI intensity in the Mediterranean context.
The impact of anthropogenic sources of heat such as cooling systems and vehicles is also
significant. These results suggest that urban morphology parameters can be used as descriptors
of the climatic performance of different urban areas, easing the work of urban planners and
designers in understanding a complex physical phenomenon, such as the UHI.

1. Introduction
Urban Heat Island (UHI), is one of the most profound climatic modification linked to urbanization,
being reported and theorized since the seminal studies of Oke during the 80’s [1]. However, regardless
the relevant advances in the understanding of urban climate, the relative weight of urban variables such
as impervious materials, urban form, anthropogenic heat sources and the like contributing to the general
UHI effect remains unclear, needing more empirical research. This shortcoming also constrains the
development of predictive studies with lower levels of uncertainty.

Research carried out by Ryu and Bak [2], Salleh et al. [3], Perini and Magliocco [4], Palme et al. [5]
have highlighted the fundamental contribution of urban form and anthropogenic heat generation on the
UHI effect; the findings also highlight that the presence of impervious materials and the reduction of
vegetation drive the diurnal UHI. UHI intensity has been predicted using four different kinds of climate
models, as proposed by Kolokotsa [6]: 1) Computational Fluid Dynamics (CFD) models (e.g. FLUENT
model), 2) Urban Thermal Environment (UTE) models (e.g. ENVIMET model), 3) Mesoscale models

Content from this work may be used under the terms of the Creative Commons Attribution 3.0 licence. Any further distribution
of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.
Published under licence by IOP Publishing Ltd 1
WMCAUS IOP Publishing
IOP Conf. Series: Materials Science and Engineering 245 (2017) 082055 doi:10.1088/1757-899X/245/8/082055
1234567890

(e.g. Urban Weather Generator, Weather Research Forecasting and Mitigation Impact Screening Tool
models) and, 4) Local Climate Zones (LCZ). These models differ in calculation approaches and accuracy
of results. CFD and UTE are more accurate in calculation, but the spatial extension of the simulation is
limited, since they require high computational power and time. LCZ and Mesoscale models are less
detailed and more uncertain; nevertheless, they have the advantage to be representative of a whole
neighbourhood or even of a whole city.

The Urban Weather Generator (UWG) model has been developed by Bueno et al. [palme], coupling
an atmospheric model with a Building Energy Model (BEM). The advantage of using UWG is that it
produces a modified weather file compatible with the most common BEMs. The model accuracy has
been evaluated for different urban contexts, such as Singapore [8] Boston [9], Tolosa [7], Rome and
Barcelona [10]. Preliminary sensitivity studies on this tool have been developed by Nakano [11], Palme
et al. [5] and Salvati [12], considering different geographic locations.

This paper presents a detailed sensitivity analysis of UWG variables considering the Mediterranean
context as reference, in order to identify the relative weight of the major urban parameters responsible
for the UHI intensity in this geographic context. The selected parameters are of high interest for urban
designers and planners, as they could use these results to understand direct links between the
morphological features of the built environment and the complex physical phenomenon of the UHI.

2. Methodology
The influence of different urban parameters on the UHI intensity was tested carrying out sensitivity
analyses with UWG model (v1.0). The model calculates hourly values of air temperature in the urban
canyon using two input files: 1) a rural weather file (. epw format) and 2) an XML file which describes
the area where the canyon is located.

The sensitivity analysis was performed by changing systematically the value of one parameter at a
time in the XML file and assessing the impact on the resulting UHI intensity. Eight parameters were
tested: 1) the city radius, 2) the urban morphology, 3) the tree coverage, 4) the anthropogenic heat from
vehicles, 5) the building’s cooling set point, 6) the heat released to canyon from HVAC systems, 7) the
wall construction properties and 8) the albedo of vertical and horizontal surfaces.

The tested values for the eight parameters were identified according to the range of variability
observed in Rome and Barcelona, being representative of significant urban areas in Mediterranean
climate. However, some values are aimed at providing broader assessment, like the city radius. The
input rural weather file used in the simulations refers to Rome-Ciampino (Lat. 41.47, Long.12.34. This
approach makes evident the trend of the monthly average UHI intensity according to the variability of
the single parameters in the Mediterranean climate.

2.1. Range of variation of urban parameters in the Mediterranean context

The eight parameters selected for the study contribute to the UHI intensity through different physical
processes, which modify the amounts of the energy budget in the urban canyon with respect to the rural
environment [1]. The links between the urban parameters and the commonly hypothesized causes of
UHI are reported in figure 2. The eight parameters correspond to four different scales, as reported in
table 1: 1) "Reference site", the entire urban area,2) "Urban area", area of approximately 500 m radius
surrounding the urban canyon, 3) "Building", the typical building of the canyon and, 4) "Elements", the
wall, road and roof of the urban canyon. Table 1 summarises the test values chosen for each parameter.

2.1.1. City radius. The value of the city radius is used in UWG for the calculation of the advection flux
in the “Urban Boundary layer” (UBL). Therefore, five values of city radius were tested in order to

2
WMCAUS IOP Publishing
IOP Conf. Series: Materials Science and Engineering 245 (2017) 082055 doi:10.1088/1757-899X/245/8/082055
1234567890

consider a wide range of city dimensions, from small towns (1,500m radius) up to megalopolis (20,000m
radius). The values for Rome and Barcelona are 6,500m and 4,500m, respectively.

2.1.2. Urban Morphology. Urban morphology contributes to the UHI intensity through multiple effects:
the shortwave radiation absorption is increased because of multiple reflections into the canyon; the
turbulent sensible heat transfers out of the canyon is reduced due to reduction of wind speed and the
long- wave radiation loss from within the canyon is reduced due to the low sky view factor determined
by surrounding buildings [13]. UWG considers three parameters of urban morphology: 1) Average
building height (m), which expresses the average height of buildings in the urban area, normalised by
building footprints; 2) Site coverage ratio (m2/m2), which is calculated as the ratio of the built area to
the urban site area and; 3) Façade to site ratio (m2/m2), which is given by the ratio of the vertical surface
area (walls) to the urban site area. The range of variability of the three parameters has been explored
considering five urban textures representative of the range of urban densities in the Mediterranean
context [14]; the reference textures are Borrel y Soler, Barceloneta and Raval of Barcelona and Don
Bosco and Prati of Rome (figure 1). The three parameters have been calculated over simplified digital
models, obtained through a process of normalisation and replication of the typical urban blocks present
in each texture [15]. The sensitivity analyses have been carried out performing five simulations,
changing concurrently the three parameters, in order to calculate the UHI intensity for the different urban
textures.

Figure1. Urban textures of Barcelona (Borrel y Soler, Barceloneta, Raval) and Rome (Prati, Don
Bosco) and normalised digital models used for the calculation of the Urban Morphology parameters

2.1.3. Tree coverage. The three coverage is calculated as the ratio of the horizontal area occupied by
trees to the urban site area. This parameter is used by UWG to calculate the latent heat flux in the urban
area and the percentage of shaded surface in the canyon. The test values for this parameter have been
estimated on the same sample of urban textures, where a range of tree coverage between 5% (in Raval)
and 28% (in Borrel y Soler) was measured.

2.1.4. Anthropogenic heat from vehicle. The anthropogenic heat from traffic is a constant input value
for UWG. The test values for this study have been calculated according to the formulation of Sailor and
Lu [16], considering statistics of "daily vehicle distance per capita" [17] and population density of Italian
cities. According to this analysis, the heat released by vehicles ranges between a minimum of 3 W/m2,
for low traffic levels such as during night time, up to 30 W/m2 during peak hours in high-density areas.
The average urban value has been assumed equal to 8 W/m2 based on previous research [18, 19].

3
WMCAUS IOP Publishing
IOP Conf. Series: Materials Science and Engineering 245 (2017) 082055 doi:10.1088/1757-899X/245/8/082055
1234567890

Figure 2. Relationships between the urban parameters and the 7 hypothesized causes of UHI

2.1.5. Building’s cooling set point. The cooling set point is a fundamental parameter used by UWG to
calculate the building’s energy performance and so the heat released into the canyon. It is expressed in
°C and corresponds to the building’s temperature setting during summer time. The sensitivity analyses
have been performed for values from 22°C to 26°C and a value equal to 35°C which indicates the
absence of mechanical cooling. The average urban value has been established to 26°C. This is a quite
high value, aimed at considering that not all the buildings have air conditioning systems in the
Mediterranean context.

2.1.6. Heat released to canyon from HVAC systems. The "heat released to canyon" indicates the amount
of wasted heat from cooling system that is exhausted into the urban canyon and/or in the urban boundary
layer. So, this parameter describes whether the cooling system is composed of outdoor units on the
facade or centralised units located on the rooftop. The parameter ranges between 0 and 1; the value 0
means that there is no waste heat release into the canyon, while the value 1 means that all the waste heat
is released into the canyon. The tested values are 0, 0.5 and 1, assuming 0.5 as the average urban value.

2.1.7. Wall construction properties. As regards the thermal properties of urban materials, in UWG each
layer of walls, roads and roofs, is characterised by thermal conductivity (W/m1K1), volumetric heat
capacity (J/m3K1) and thickness (m). Considering the properties of building materials and the common
construction technologies in the Mediterranean context, the main difference is given by the density and
thickness of the element's materials, which determine its inertia and ability to store heat. Therefore, two
types of walls have been tested: load bearing walls and light weight walls. The former are typical of
masonry constructions, with high thermal inertia and heat capacity, while the latter are light weight
layered elements which ensure insulation from outdoor environment but have low thermal capacity.
Details on the tested materials and thickness of the two types are given in table 1.

2.1.8. Albedo of vertical and horizontal surfaces. The last parameter tested was the albedo of walls,
roads and roofs surfaces. The albedo, also known as reflection coefficient, is the ratio of reflected
radiation from the surface to incident radiation upon it. Its value, which ranges between 0 and 1, is
mainly linked to the colour of surface. Many studies [20] highlighted that an increase of albedo would
have a significant mitigation effect on urban air temperature. However, the range of variation of the
albedo of urban materials is quite low; according to literary review [21], the average values are between
0.2 - 0.5 for walls (from light-coloured to dark-coloured plasters or coatings), 0.04 - 0.2 for roads (from
fresh asphalt to worn asphalt) and 0.1 - 0.35 for roofs (including bituminous membranes, tiles or
floorings). The sensitivity analyses have been conducted for this range of values. The three values of

4
WMCAUS IOP Publishing
IOP Conf. Series: Materials Science and Engineering 245 (2017) 082055 doi:10.1088/1757-899X/245/8/082055
1234567890

albedo of walls, roads and roofs have been changed together to simulate the effect of dark, medium or
light colours, as reported in detail in table 1.

Table 1. The tested values of the eight urban parameters used in the sensitivity analysis. The average
values are in bold.
Scale Urban Parameters Tested values
Reference 1) City Radius (m) 1. 2. (Barcelona) 3. (Rome) 4. 5.
site 1,500 4,500 6,500 10,000 20,000

Urban 2) Urban Morphology 1.Borrel y Soler 2.Barceloneta 3.Prati 4.Don Bosco 5.Raval
Area Average Building Height (m) 15 16.5 19.5 25.5 19.5
Site coverage ratio (m2/m2) 0.2 0.52 0.49 0.43 0.8
Facade-to-site ratio (m2/m2) 0.63 2.23 1.43 1.68 1.38
3) Tree coverage (%) 1.Low 2.Average 3.High
5 12 28
4) Anthropogenic heat from 1.low 2.Average 3.high 4.pick hour
Traffic (W/m2) 3 8 15 30

Building 5) Cooling set point 1.No cooling 2.High 3.Medium 4.Low

35° 26° 24° 22°
6) Heat released to canyon 1.HVAC Units 2. Average 3.HVAC units
on rooftop on facade
0 0.5 1
Element 7) Wall construction 1.Heavy Wall 2.Light Wall
Layers material plaster-masonry-plaster plaster-hollow brick-air-insulation-hollow brick-
plaster
Thickness (m) 0.02-0.5-0.01 0.02-0.08-0.05-0.07-0.12-0.01
Conductivity (W/mK) 1.0-0.72-1.0 1.0-0.37-0.026-0.034-0.37-1.0
Volumetric heat capacity 14*105-1612800-14*105 14*105-930000+1206+49000+930000 +14*105
(J/m3 K)
8) Albedo 1.ligh colours 2.Medium 3.dark colours
wall 0.5 0.35 0.2
road 0.2 0.08 0.04
roof 0.35 0.25 0.1

3. Results and discussions

3.1. Impact of the single parameters on the UHI intensity

The variation of the monthly average UHI intensity according to the tested values for each of the eight
parameters is reported in figure 3.

Regarding the city radius, UWG estimates a higher UHI intensity for the smaller cities, above all
during autumn and winter months, while during summer months this parameter is irrelevant. This result
is probably linked to the calculation of circulation velocity [7], which determines if the advection flux
in the "Urban boundary layer" is driven by the geostrophic wind (forced problem) or by the urban breeze
circulation (buoyancy-driven problem). The maximum variation of UHI intensity determined by urban
radius variability occurs in September and October, reaching 0.7 °C. However, considering the radius
of the two tested Mediterranean urban areas, Barcelona, about 4,500m radius and Rome, about 6,500m,
the temperature gap is negligible.

The impact of urban morphology on the UHI intensity is significant during both winter and summer.
The maximum difference is recorded in February, when the UHI intensity for the morphology of Raval
is up to 2.4°C higher with respect to the one of Borrell Y Soler. The annual average temperature
difference between the densest and the rarest urban texture is 1.8°C.

5
WMCAUS IOP Publishing
IOP Conf. Series: Materials Science and Engineering 245 (2017) 082055 doi:10.1088/1757-899X/245/8/082055
1234567890

The tree coverage has, instead, an appreciable effect on the UHI intensity only during summer
months; the variation of the tree coverage from the lowest value (5%) to the highest one (28%)
determines an average decrease of UHI intensity of 0.8°C in July and August. However, it has to be
considered that the average value of tree coverage in the Mediterranean urban context is quite low,
because the built surface is very high, especially in the city centre. In fact, the difference of UHI intensity
between the average case (12% tree coverage) and the worst case (5% tree coverage) is negligible. The
analysis of variability of sensible heat from traffic shows significant impacts on the UHI. During the
summer, the UHI intensity based on the maximum value of the parameter (30 W/m2) is more than 1°C
higher than the UHI corresponding to the average value (8 W/m2). Nevertheless, in UWG model this
parameter is assumed to be constant over the day, while the maximum value considered in this analysis
corresponds to peak traffic hours; during the rest of the day the value would be certainly lower.
Therefore, this result may overestimate the impact of heat from traffic on the UHI. In this regard, the
possibility to implement the daily trend of heat release from vehicles would improve the accuracy of the
UWG predictions.

Figure 3. Monthly average UHI intensity according to the variability of the urban parameters.

Another crucial variable affecting the UHI intensity in summer is the presence of cooling systems in
buildings; the impact is especially high for low cooling set points. In the hottest month, UWG estimates
an increase of the average UHI intensity between 0.8 °C and 1.7°C due to anthropogenic heat released
by cooling systems. The parameter "heat released to canyon", which defines whether the heat from
buildings is exhausted directly into the canyon or not, is less important.

The results also show that the variability of wall's thermal capacity does not affect the UHI intensity
during most part of the year. The only appreciable temperature difference is recorded in July, when
UWG results indicate a 0.3°C increase of UHI intensity for the walls with high thermal capacity (heavy
walls). Therefore, this parameter is not influential on urban climate, in spite of its significance on the
indoor thermal balance in summer.

Similarly, UWG calculations do not show any significant impact of the albedo of surfaces on the
UHI intensity. This can be explained in the light of the limited variability of albedo values in the urban
context. Urban albedo values of approximately 1 would be required for the UHI intensity effect to be
reduced substantially; however, it is not likely that these values can be achieved across a whole urban
area, where the average values are normally below 0.5.

6
WMCAUS IOP Publishing
IOP Conf. Series: Materials Science and Engineering 245 (2017) 082055 doi:10.1088/1757-899X/245/8/082055
1234567890

3.2. Hierarchy of importance of the parameters

A comparative analysis of the results is presented in figure 4 to highlight the relative weight of each
parameter on the UHI intensity in the Mediterranean context. The black bars in the graphs indicate the
maximum difference of the monthly average UHI intensity determined by the maximum variation of
each parameter; the difference varies from less than 0.5 °C to more than 2 °C, depending on the month
and the parameter considered.

Figure 4. Maximum, minimum and average UHI intensity determined by the maximum, minimum
and average value of each parameter

These results indicate a clear hierarchy of importance of the eight analysed urban parameters in terms
of their impact on the UHI effect in the Mediterranean context. This is explained by two reasons: 1) the
range of variability of certain parameters in the Mediterranean urban context is larger than others and,
as a consequence, their impact on the UHI intensity is higher and; 2) some of the "causes" of the UHI
are more important than others, and so the related parameters.

According to UWG results, the most important factor is the urban morphology, which determines
substantial variations of the UHI intensity throughout the year. Considering that the annual average UHI
intensity is about 1.5 °C for the average case, the variability of urban morphology determines changes
up to 1.8°C; this means a relative impact up to the 120% of the average UHI intensity.

Another key parameter is the anthropogenic heat from the air conditioning systems of buildings,
which significantly increases the UHI during summer. The average impact on air temperature is about
1.3°C, over an average UHI intensity in summer of 1.8°C. So, results indicate that the use of air
conditioning systems in the buildings determines a relative change of 72% of the UHI intensity in
summer. The anthropogenic heat released from traffic has an appreciable impact on the UHI intensity
as well, even if lower than the urban morphology and the cooling systems. The variability of this
parameter determines an annual average temperature change of 0.6 °C, which is the 40% of the annual
average UHI intensity. A similar impact is determined by the variability of the city radius. However,
with regards to this parameter, the results predict higher temperatures in smaller cities, probably due to
the methodology adopted by UWG for the calculation of the advection flux in the urban boundary layer.
Since there is no validation of this specific calculation module, these results should be further
investigated.

The remaining four analysed parameters are much less significant on the UHI intensity. Therefore,
according to these results, the most important causes of urban temperature increase in the Mediterranean

7
WMCAUS IOP Publishing
IOP Conf. Series: Materials Science and Engineering 245 (2017) 082055 doi:10.1088/1757-899X/245/8/082055
1234567890

context are the physical processes that occur in the urban canyon due to the urban morphology and the
presence of anthropogenic sources of heat like traffic and cooling systems.

The fact that the values of tree coverage is less influential on the UHI intensity is consistent with
other urban studies [22–26]. Like in the case of albedo, the influence depends on the limited range of
variability that these variables have in urban areas; a substantial modification of their values in the entire
urban area could produce a more appreciable effect on the UHI intensity. At the same time, it has to be
considered that UWG cannot simulate the effect of large water bodies or big parks on urban climate, so
these variables have not been included in the study so far. They are matter of future studies, since their
effect at urban scale may be significant.

4. Conclusions
This paper analyses the relative impact on the UHI intensity of eight parameters in the Mediterranean
context, through sensitivity analyses using the UWG model.

Urban morphology is the parameter having the higher effect over the UHI intensity, causing a relative
change up to 120% of the annual average UHI intensity. The anthropogenic heat related to traffic and
buildings is also important. The variability of the heat released from vehicles in the urban area
determines a 40% change of the annual UHI average. In summer, the presence of cooling systems in
buildings is even more significant on the UHI intensity; waste heat produced by cooling systems may
determine an increase up to the 72% of the average UHI intensity. All in all, the amount of anthropogenic
heat from both vehicles and cooling systems is depending on the density and morphology of buildings
in the urban area. Therefore, urban morphology is, indeed, the most important parameter with respect to
the UHI intensity. The impact of variability of other parameters is quite negligible on the UHI variation
in the Mediterranean context.

These results show that some common mitigation strategies, such as increase of albedo and tree
coverage, may be ineffective to reduce the UHI intensity in the Mediterranean context, unless the
intervention is able to change the ratio of these variables over a whole urban area; if their values remain,
instead, constrained to small urban extensions, the impact on UHI intensity will be weak. This result is
in line with other similar urban studies [22–27].

These results also suggest that urban density and morphology parameters could be used as predictors
of the UHI intensity in the Mediterranean context, easing the analysis of the phenomenon in the urban
areas for planners and urban designers. The knowledge of the temperature distribution according to
morphology and density of the urban texture could also help to steer the mitigation interventions in the
most affected areas within the city, so as to obtain effective results at urban scale.

Acknowledgments
The authors wish to thank Catholic University of the North, Sapienza University of Rome and
Polytechnic University of Catalunya for the financial support to carry out and present the results of this
study.

References
[1] T.R. Oke, Boundary Layer Climate, 1987.
[2] Y.-H. Ryu, J.-J. Baik, Quantitative Analysis of Factors Contributing to Urban Heat Island
Intensity, J. Appl. Meteorol. Climatol. 51 (2012) 842–854. doi:10.1175/JAMC-D-11-098.1.
[3] S.A. Salleh, Z. Abd.Latif, W.M.N.W. Mohd, A. Chan, Factors Contributing to the Formation of
an Urban Heat Island in Putrajaya, Malaysia, Procedia - Soc. Behav. Sci. 105 (2013) 840–
850. doi:10.1016/j.sbspro.2013.11.086.
[4] K. Perini, A. Magliocco, Effects of vegetation, urban density, building height, and atmospheric

8
WMCAUS IOP Publishing
IOP Conf. Series: Materials Science and Engineering 245 (2017) 082055 doi:10.1088/1757-899X/245/8/082055
1234567890

conditions on local temperatures and thermal comfort, Urban For. Urban Green. 13 (2014)
495–506. doi:10.1016/j.ufug.2014.03.003.
[5] M. Palme, C. Carrasco, A. Lobato, Quantitative Analysis of Factors Contributing to Urban
Heat Island Effect in South American Cities, Procedia Eng. 169 (2016) 199–206.
[6] M. Santamouris, D. Kolokotsa, Urban climate mitigation techniques, 2016.
[7] B. Bueno, L. Norford, J. Hidalgo, G. Pigeon, The urban weather generator, J. Build. Perform.
Simul. 6 (2013) 269–281. doi:10.1080/19401493.2012.718797.
[8] B. Bueno, M. Roth, L. Norford, R. Li, Computationally efficient prediction of canopy level
urban air temperature at the neighbourhood scale, Urban Clim. 9 (2014) 35–53.
doi:10.1016/j.uclim.2014.05.005.
[9] M. Street, C. Reinhart, L. Norford, J. Ochsendorf, Urban heat island in Boston – An evaluation
of urban air- temperature models for predicting building energy use, in: Proc. BS201313th
Conf. Int. Build. Perform. Simul. Assoc., Chambéry, France, 2013: pp. 1022–1029.
[10] A. Salvati, H. Coch, C. Cecere, Urban Heat Island Prediction in the Mediterranean Context: an
evaluation of the urban weather generator model, ACE Archit. City Environ. = Arquit.
Ciudad Y Entorno. 11 (2016) 135–156. doi:10.5821/ace.11.32.4836..
[11] A. Nakano, Urban weather generator user interface development: Towards a usable tool for
integrating urban heat island effect wothin design process, Igarss 2014. (2015) 1–141.
doi:10.1007/s13398-014-0173-7.2.
[12] A. Salvati, The compact city in Mediterranean climate: Heat Island, Urban Morphology and
Sustainability, PhD thesis, 2016.
[13] T.R. Oke, Street design and urban canopy layer climate, Energy Build. 11 (1988) 103–113.
doi:10.1016/0378-7788(88)90026-6.
[14] C. Cecere, H. Coch, M. Morganti, G. Clementella, Dalla Riqualificazione Energetica al
Recupero Sostenibile, In_BO Ric. E Progett. per Territ. La Città E L’architettura. 3 (2012).
[15] J. Zhang, C.K. Heng, L.C. Malone-Lee, D.J.C. Hii, P. Janssen, K.S. Leung, et al., Evaluating
environmental implications of density: A comparative case study on the relationship between
density, urban block typology and sky exposure, Autom. Constr. 22 (2012) 90–101.
[16] D.J. Sailor, L. Lu, A top – down methodology for developing diurnal and seasonal
anthropogenic heating profiles for urban areas, 38 (2004) 2737–2748.
doi:10.1016/j.atmosenv.2004.01.034.
[17] European Commission, Analysis of National Travel Statistics in Europe, 2013.
[18] G. Pigeon, D. Legain, P. Durand, V. Masson, Anthropogenic heat release in an old European
agglomeration (Toulouse, France), Int. J. Climatol. 27 (2007) 1969–1981.
doi:10.1002/joc.1530.
[19] B. Bueno, L. Norford, G. Pigeon, R. Britter, Combining a Detailed Building Energy Model
with a Physically-Based Urban Canopy Model, Boundary-Layer Meteorol. 140 (2011) 471–
489. doi:10.1007/s10546-011-9620-6.
[20] M. Santamouris, Cooling the cities - A review of reflective and green roof mitigation
technologies to fight heat island and improve comfort in urban environments, Sol. Energy.
103 (2014) 682–703. doi:10.1016/j.solener.2012.07.003.
[21] L. Gartland, Heat Islands: Understanding and Mitigating Heat in Urban Areas, London, 2008.
[22] L. Inostroza, Open space and urban ecosystem services, J. L. Use, Mobil. Environ. (2014) 523–
534.
[23] L. Inostroza, Informal urban development in Latin American urban peripheries. Spatial
assessment in Bogotá, Lima and Santiago de Chile, Landsc. Urban Plan. In Press (2016).
doi:10.1016/j.landurbplan.2016.03.021.
[24] L. Inostroza, M. Palme, F. De La Barrera, A heat vulnerability index: Spatial patterns of
exposure, sensitivity and adaptive capacity for Santiago de Chile, PLoS One. 11 (2016) 1–
26. doi:10.1371/journal.pone.0162464.
[25] M. Palme, G. Villacreses, L. Inostroza, A. Lobato, C. Carrasco, From urban climate to energy

9
WMCAUS IOP Publishing
IOP Conf. Series: Materials Science and Engineering 245 (2017) 082055 doi:10.1088/1757-899X/245/8/082055
1234567890

consumption. Enhancing building performance simulation by including the urban heat island
effect, Energy Build. 145 (2017) 107-120. doi:10.1016/j.enbuild.2017.03.069.
[26] A. Salvati, H. Coch, C. Cecere, Assessing the urban heat island and its energy impact on
residential buildings in Mediterranean climate: Barcelona case study, Energy Build. 146
(2017) 38–54. doi:10.1016/j.enbuild.2017.04.025.
[27] M. Palme, L. Inostroza, C. Carrasco, G. Villacreses, A. Lobato, Urban climate in South-
American coastal cities in Chile, Peru and Ecuador, in: Prep., SPRINGER.

10