Please verify that (1) all pages are present, (2) all figures are correct, (3) all

fonts and special characters are correct, and (4) all text and figures fit within the red
margin lines shown on this review document. Complete formatting information is available at http://SPIE.org/manuscripts

Return to the Manage Active Submissions page at http://spie.org/submissions/tasks.aspx and approve or disapprove this submission. Your manuscript will not
be published without this approval. Please contact author_help@spie.org with any questions or concerns.

Can urban design mitigate the UHI effect?

Blanca Arellano
UPC, Technical University of Catalonia, Barcelona, Spain
blanca.arellano@upc.edu

Josep Roca
UPC, Technical University of Catalonia, Barcelona, Spain
josep.roca@upc.edu

ABSTRACT

The Land Surface Temperature (LST) plays a key role in the generation of the Urban Heat Island (UHI),
representing a determining factor of the radiation of the surface and the exchange of energy and control the
distribution of heat between the surface and the atmosphere. The composition of the land covers is one of the
main factors that influence the LST. The literature have highlighted that the different types of land cover
materials, their albedo and roughness, degree of impervious as well as the quality and quantity of vegetation,
can have a significant impact on the generation of the UHI. The overall objective of the paper is to investigate
if urban design, on a detail scale, can mitigate the Urban Heat Island. The question is if the morphology of
public space, especially the form and zoning of green areas and open spaces, influence surface (LST) and air
(LSAT) temperatures and therefore in the UHI. The working hypothesis is that the morphology of public space
plays a key role to control UHI. Thus, an adequate design of open spaces can help to increase the resilience of
cities to Climate Change (CC). To prove the hypothesis, two case studies were chosen. The case studies are two
urban expansion areas of city of Sant Cugat del Vallès (in the metropolitan area of Barcelona), the first one,
called Parc Central and the second one, Coll Favà. Both have similar characteristics, but with a clearly
differentiated public space design; while in Parc Central built-up area is structured around the open space, in
Coll Favà, the vast majority of open spaces are located peripherally, surrounding the built-up areas. The research
methodology consisted in: a) studying the urban and climatic parameters of selected areas; b) analyzing the
spatial distribution of the LST using remote sensing technologies (Landsat 8); c) obtaining LST and LSAT
through field work, during day and night time; and d) constructing a model of surface and air temperatures as a
function of the different types of land cover, combining Remote Sensed data and in situ measurements, for each
of the areas of analysis.

Keywords: Urban Heat Island, Green Areas, Cold Urban Areas, Planning, Urban Design, Climate Change,
Morphology of Public Space

10793 - 11 V. 2 (p.1 of 16) / Color: No / Format: A4 / Date: 9/11/2018 6:43:50 AM

SPIE USE: ____ DB Check, ____ Prod Check, Notes:

 Please verify that (1) all pages are present, (2) all figures are correct, (3) all fonts and special characters are correct, and (4) all text and figures fit within the red
margin lines shown on this review document. Complete formatting information is available at http://SPIE.org/manuscripts

Return to the Manage Active Submissions page at http://spie.org/submissions/tasks.aspx and approve or disapprove this submission. Your manuscript will not
be published without this approval. Please contact author_help@spie.org with any questions or concerns.

1. INTRODUCTION

"Warming of the climate system is unequivocal, and since the 1950s, many of the observed changes are unprecedented
over decades to millennia. The atmosphere and ocean have warmed, the amounts of snow and ice have diminished, and
sea level has risen." (IPCC 20141, page 30). “Each of the last three decades has been successively warmer at the Earth’s
surface than any preceding decade since 1850. The period from 1983 to 2012 was very likely the warmest 30-year period
of the last 800 years in the Northern Hemisphere, where such assessment is possible (high confidence) and likely the
warmest 30-year period of the last 1400 years (medium confidence)” (IPCC 20141, page 40). This temperature increase is
distributed throughout the planet and is more pronounced in the higher northern latitudes. The terrestrial regions (Fig. 1)
have warmed faster than the oceans. The variation of greenhouse gas (GHG) concentrations in the atmosphere and
variations in land cover and solar radiation, alter the energy balance of the climate system. In this sense, the anthropogenic
origin of the observed changes2 seems today an equally incontrovertible fact. The global emissions of GHG due to the
effects of human activities have increased, since the pre-industrial era, by 70% between 1970 and 2004. The result of the
different models of the evolution of the temperatures of the earth's surface, show the prominence of anthropogenic forcing
origin, with respect to those of a natural nature.

Figure n. 1: Evolution of temperatures (natural and anthropogenic forcing. Source: IPCC 20071

There is a consensus on the fact that cities have a special role in the climate change. According to the Center for Human
Settlements (UN-HABITAT), cities are responsible for 75% of global energy consumption, as of 80% of GHG emissions.
The contribution of the urbanization to the climatic change is of double nature: on the one hand, by the urban generation
of GHG, a factor that contributes in a decisive way to the global warming of the planet; on the other, by the radiation
generated by the surface of the urbanized land, which determines a marked flow of sensible and latent heat according to
the type of urban covers, as well as their degree of humidity.

Although the urban climate depends essentially on global and regional factors, local and micro-scale factors, such as the
different characteristics of the urban structure, topography and surface of land covers, vegetation, anthropogenic heat
generated by urban metabolism, among other factors, can modify the local climate on the urban scale.

There are significant differences between the climate of urban and rural areas: the urban heat island (UHI) describes the
influence of urban surfaces on the patterns of temperature in urban areas as opposed to the surrounding areas. The sealed
land and artificial materials (especially asphalt and concrete) used usually in urban areas are one of the main causes.

Figure 2. Urban Canopy & Urban Boundary Layers. Source: Oke (1987)3

The effects of the UHI manifest in different scales. Two types of UHI can be distinguished: the "canopy layer heat island"
and the "boundary level heat island" (Fig. 2). The first depends on the roughness of the soil generated by the buildings and
the canopy of the trees, with an upper limit located just above the level of the roofs of the buildings. In this layer, air flow
and energy exchanges are governed by microscale processes that depend on the specific characteristics of the surface. The

10793 - 11 V. 2 (p.2 of 16) / Color: No / Format: A4 / Date: 9/11/2018 6:43:50 AM

SPIE USE: ____ DB Check, ____ Prod Check, Notes:

 Please verify that (1) all pages are present, (2) all figures are correct, (3) all fonts and special characters are correct, and (4) all text and figures fit within the red
margin lines shown on this review document. Complete formatting information is available at http://SPIE.org/manuscripts

Return to the Manage Active Submissions page at http://spie.org/submissions/tasks.aspx and approve or disapprove this submission. Your manuscript will not
be published without this approval. Please contact author_help@spie.org with any questions or concerns.

second is situated above the first, with a lower limit subject to the influence of the urban surface. In the urban boundary
layer, which is the part of the atmospheric boundary layer above the level of the buildings whose characteristics are affected
by the presence of the city, the UHI operates in a different way, in the case of a mesoscale local scale phenomenon,
controlled by processes that operate on a larger spatial and temporal scale.

There is a wide consensus that Land Surface Temperature (LST) plays a key role in the generation of the Urban Heat
Island, representing a determining factor in terms of surface radiation and the interchange of the energy4, in addition to
controlling the distribution of heat between the surface and the atmosphere5.

The composition of land covers are one of the main factors that influence the LST. Specialized literature has emphasized
that built up area and the pervious surface (occupied areas provided with vegetation) and impervious surface (paved and
built areas) have a significant impact on the generation of a UHI6. In this sense, the reduction of vegetation affects not only
in an increase of the LST, but also the reduction of precipitation and evapotranspiration7,8. The relationship between the
LST and the Normalized Difference Vegetation Index (NDVI) is especially well-documented.

Remote sensing provides spatially continuous information that allows obtaining the LST at scales from tens of meters to
kilometers. The satellites provide data on characteristics and land uses that make it possible to quantify the UHI using
various indicators9. Remote sensing has made possible the generalized study of the LST and, consequently, of the UHI at
local and regional scales. At present, there are several operational sensors that allow to measure the LST: MODIS,
Advanced Spaceborne Thermal Emission and Reflection (ASTER), Landsat-7 ETM +, Landsat-8 TIRS, Geostationary
Operational Environmental Satellite (GOES), NOAA Advanced Very High Resolution Radiometer. AVHRR), Indian
National Satellite System (INSAT), Geostationary Meteorological Satellite (GMS) as well as the Meteorology Satellite
(Meteosat), among others10.

Although the LST can be easily measured by remote sensing, the UHI has been commonly studied through the air
temperature in the atmospheric surface layer, usually at a height of 2 m above the ground (land air surface temperature,
LSAT). The air temperature, measured in the weather stations, is one of the observations most frequently recorded, with
great precision and temporal resolution11.

The air temperature, despite its lower climatic relevance in relation to the LST, has an essential role at the scale of human
comfort. It is the temperature that people perceive, and in relation to which the exchange of energy works with the
environment. However, there are usually few meteorological stations, which in general terms means that the data obtained
from the stations do not efficiently represent the spatial variation of the air temperature at the intra-municipal or
neighborhood level. For example, at the RMB (Metropolitan Region of Barcelona) there are only 33 weather stations.

Figure n. 3: LST & LSAT, Day & Night time. Source: Environmental Protection Agency (2008)12

The urban heat island (UHI) has an impact on the health and well-being of the citizens, especially worrisome when there
are extreme episodes such as heat waves. In recent years we can highlight the episode of August 2003 in Western Europe,

10793 - 11 V. 2 (p.3 of 16) / Color: No / Format: A4 / Date: 9/11/2018 6:43:50 AM

SPIE USE: ____ DB Check, ____ Prod Check, Notes:

 Please verify that (1) all pages are present, (2) all figures are correct, (3) all fonts and special characters are correct, and (4) all text and figures fit within the red
margin lines shown on this review document. Complete formatting information is available at http://SPIE.org/manuscripts

Return to the Manage Active Submissions page at http://spie.org/submissions/tasks.aspx and approve or disapprove this submission. Your manuscript will not
be published without this approval. Please contact author_help@spie.org with any questions or concerns.

in which between 22,000 and 45,000 deaths were related to the phenomenon. However, the 2003 heat wave was not an
isolated case. In the last 15 years, climatic anomalies due to heat have exploded. A heat wave becomes especially significant
when for three or more nights the air temperature does not drop below 25 degrees Celsius. Then, mortality shoots up.

The temperature at night has, therefore, a special role in the emergence of the UHI. The UHI manifests more intensely at
night.

Figure 4. Heat wave in Europe (2003)

---

Urban planning has a fundamental transcendence to inform, coordinate and implement measures to improve the climate
quality of cities in the face of global climate change13. There have been some initiatives to introduce methodologies, in the
spatial planning, aimed at mitigating climate change. Especially on the "small scale", there is an extensive experience of
urban bioclimatic design. In turn, at the building level there has been a significant effort in recent years to increase energy
efficiency, with the aim of reducing the generation of GHG (Directives 2002/91 / EC and 2010/31 / EU of the European
Parliament and the Council concerning the energy performance of buildings). On the "large scale", however, that of urban
and territorial planning, there does not seem to be a parallel awareness, with few initiatives to adapt it in order to increase
urban resilience to climate change.

The development of the so-called climatopes has become one of the most effective mechanisms for the introduction of
climate in territorial and urban planning to the "large scale" (meso-scale, in climatic terms). The beginnings of this
technique are found in the pioneering works developed by German researchers at the end of the 70's14, which developed
the new technique of the climatopes in the seminal researches of Baumuller et al. (1998)15 and Scherer et al. (1999)16.

Chao et al. (2010) have carried out a systematic review of the specialized research on urban climate analysis, called in the
English-language literature, urban climatic maps (UCM), or more specifically local climate zones (LCZ). Highlights
include work done for Tokyo17, Berlin18, Freiburg, Hong Kong19, Lisbon or Vancouver20.

Especially relevant, for the control of the UHI, are the urban green areas. Due to its abundant vegetation and its pervious
surface, parks present a different behavior from the rest of urban spaces. The possibility of using parks as a strategy to
reduce the UHI of cities has been studied. The specialized literature recognizes that vegetation is one of the main
components of the urban space that causes cooling.

When there is a decrease in land surface and air temperatures due to the abundance of green areas, it is said that there is a
Green Cooling Island21, which consists of a cooling effect inside the green area and in its surroundings. Its influence is
manifested in both the LST and the LSAT, and its effects occur both day and night, although the extent of its extension
depends on the characteristics of the green space and the urban configuration of its surroundings22, 21.

10793 - 11 V. 2 (p.4 of 16) / Color: No / Format: A4 / Date: 9/11/2018 6:43:50 AM

SPIE USE: ____ DB Check, ____ Prod Check, Notes:

 Please verify that (1) all pages are present, (2) all figures are correct, (3) all fonts and special characters are correct, and (4) all text and figures fit within the red
margin lines shown on this review document. Complete formatting information is available at http://SPIE.org/manuscripts

Return to the Manage Active Submissions page at http://spie.org/submissions/tasks.aspx and approve or disapprove this submission. Your manuscript will not
be published without this approval. Please contact author_help@spie.org with any questions or concerns.

As a consequence of the aforementioned, the design of the urban parks as well as the public space represents a fundamental
element in the regulation of the urban climate, and especially of the UHI.

2. OBJECTIVE OF THE RESEARCH

The general objective of the research is to study, using remote sensing techniques as well as "in situ" measurements, how
urban design affects in the generation of the Urban Heat Island, as well as the urban microclimate in general. Specifically,
this paper seeks to clarify whether the design of green areas can mitigate the UHI.

The case of study is the Metropolitan Region of Barcelona (figure 5), which includes 164 municipalities, with a population
of 4.7 million people and an extension of 3,200 square kilometers. At this metropolitan scale, the spatial configuration of
the UHI will be studied.

Figure n. 5. Metropolitan Region of Barcelona

At the local scale, we will study how the morphology of public space, especially of green areas and open spaces, affects
surface temperatures (LST) and air temperatures (LSAT), determining the characteristics of the UHI.

Two case studies will be analyzed and compared in the municipality of Sant Cugat del Vallés (SCV, figure 6): the
urbanizations of Parc Central (PC) and Coll Favà (CF), which have similar urban characteristics, but which have clearly
differentiated design aspects in terms of public space. Buildings in Parc Central are “built around open space”, while in
Coll Favà, the vast majority of open spaces are located peripherally, surrounding the buildings.

Figure n. 6. Sant Cugat del Vallès

10793 - 11 V. 2 (p.5 of 16) / Color: No / Format: A4 / Date: 9/11/2018 6:43:50 AM

SPIE USE: ____ DB Check, ____ Prod Check, Notes:

 Please verify that (1) all pages are present, (2) all figures are correct, (3) all fonts and special characters are correct, and (4) all text and figures fit within the red
margin lines shown on this review document. Complete formatting information is available at http://SPIE.org/manuscripts

Return to the Manage Active Submissions page at http://spie.org/submissions/tasks.aspx and approve or disapprove this submission. Your manuscript will not
be published without this approval. Please contact author_help@spie.org with any questions or concerns.

The hypothesis is that the morphology of public space is a significant element in the generation of UHI. In particular, the
design of the public space, their morphology as well as the quality of their vegetation determine the degree of cooling of
the urban green areas, delimiting the extent and size of the Green Cooling Island.

3. METHODOLOGY

The methodology used in this paper combines the use of remote sensing technologies, of medium resolution (30 meters /
pixel), with the taking of measurements "in situ". The combination of both approaches, remote and data collection "in situ",
allows to validate in a more robust way the hypothesis of research related to the key role played by the design of the public
space, and especially of the green areas, in the regulation of the UHI.

Specifically, the research methodology23 is developed in the following steps:

I. Remote sensing analysis:

a. The daytime LST will be obtained from LANDSAT 8, at resolution scale of 30x30 meters / pixel. The
measurement of the resulting LST will allow an initial comparison between the study cases (Parc Central and
Coll Favà, located in Sant Cugat del Vallès), evaluating which of the two developments (PC and CF) shows
a less pronounced effect of the daytime “land-surface” UHI.

b. The nighttime LST will be obtained from MODIS, at 1 km2 for pixel. Due to the low resolution of MODIS,
the results of the night LST do not allow a detailed analysis at the local level. However, the nighttime LST
obtained through MODIS allows us to know the configuration of the nocturnal UHI at the metropolitan scale.

c. Daytime and nighttime temperatures, obtained by remote sensing in the previous sections, allow to make a
detailed approximation of the UHI of the Metropolitan region of Barcelona.

d. The information obtained through the satellites allows knowing, together with the LST, some indices that
affect the UHI, such as the NDVI or the NDBI. Particularly the NDVI helps to know the effect of the tree
canopy and the quality of the vegetation in the UHI of the MBR, as well as of the cases of study at SCV.

II. “In situ” Analysis:

a. The previous information will be completed by obtaining day and night “in situ” LST in the two study cases.
This second approach will allow contrasting the results obtained through remote sensing. The greater level
of detail of the field data will help to ratify the hypothesis of research.

b. In addition, day and night LSAT measurements will be taken in the two study cases. This information will
help us to confirm the hypothesis of the microclimatic effect of public space and green design (both during
daytime and nighttime) in the perspective of human perception. Particularly, at nighttime, if the air
temperature reaches the 25 degrees Celsius that represent the "torrid" nights.

c. Finally, the urban design of both study cases and its green infrastructure will be analyzed.

4. OBTAINING LST

The method used to obtain the LST using information from LANDSAT consists of converting the numerical coding
(Digital Number-DN) of the thermal band (infrared heat) facilitated by the satellite in physical units, i.e. in Celsius degrees.
As such:

10793 - 11 V. 2 (p.6 of 16) / Color: No / Format: A4 / Date: 9/11/2018 6:43:50 AM

SPIE USE: ____ DB Check, ____ Prod Check, Notes:

 Please verify that (1) all pages are present, (2) all figures are correct, (3) all fonts and special characters are correct, and (4) all text and figures fit within the red
margin lines shown on this review document. Complete formatting information is available at http://SPIE.org/manuscripts

Return to the Manage Active Submissions page at http://spie.org/submissions/tasks.aspx and approve or disapprove this submission. Your manuscript will not
be published without this approval. Please contact author_help@spie.org with any questions or concerns.

 The DN is transformed into spectral radiance.

 At-sensor brightness temperature is calculated. This temperature does not take into account the type of material
or land that emits the energy captured, it would therefore be equivalent to the temperature emitted by a black
body.
 The final step for obtaining the LST is to correct the numerical value obtained in the previous steps, by introducing
the emissivity of the soil materials. In this case, the emissivity has been obtained from the NDVI (normalized
difference vegetation index). The use of the NDVI with respect to other alternatives, such as land use classification
has two main advantages: 1. both, temperature and vegetation index come from the same moment in time, and 2.
the immediacy involved in the calculation of this index.

The used image (LC81970312015176LGN00) was in the summer of 2015 (an extremely hot year). Specifically from June
25. LANDSAT 8 image with the following bands: 1 = Coastal; 2 = Blue (30 m); 3 = Green (30 m); 4 = Red (30 m); 5 =
Close-up infra-red (30 m); 6 = SWIR 1 (30 m); 7 = SWIR 2 (30 m); 8 = Pan (15 m); 9 = Cirrus; 10 = TIR 1 (100 m); 11 =
TIR 2 (100 m).

Figure 7. Landsat bands

Figure n. 8: LST process

Figure 9. LST

10793 - 11 V. 2 (p.7 of 16) / Color: No / Format: A4 / Date: 9/11/2018 6:43:50 AM

SPIE USE: ____ DB Check, ____ Prod Check, Notes:

 Please verify that (1) all pages are present, (2) all figures are correct, (3) all fonts and special characters are correct, and (4) all text and figures fit within the red
margin lines shown on this review document. Complete formatting information is available at http://SPIE.org/manuscripts

Return to the Manage Active Submissions page at http://spie.org/submissions/tasks.aspx and approve or disapprove this submission. Your manuscript will not
be published without this approval. Please contact author_help@spie.org with any questions or concerns.

5. METROPOLITAN HEAT ISLAND OF BARCELONA

The LST (Landsat) has allowed to know the UHI (daytime) of Barcelona:

 The “economic activities & logistics” (34.40 ºC), “industrial” (33.29 ºC), “tertiary” and “services” (31.37 ºC)
represent the hottest land uses at metropolitan scale.

 At the same time the “roads network” (31.34 ºC) as well as the “compact residential” (31.05 ºC), LST is clearly
higher than the average of the MRB (28.58 ºC).

 These land uses contribute to the UHI in a clearly more pronounced way than the "traditional urban residential"
(30.43 ºC), “historic center" (30.32 ºC), "residential in grouped houses" (30.30 ºC) or "single family homes"(28.62
ºC).

 The "non-developable land of special protection" (26.17 ºC), the "undeveloped land with protection" (28.16 ºC),
the open spaces and greens areas"(28.23 ºC), as well as the "Hydrographic system" (28.61 ºC) are the best rated
land uses from a climate perspective.

 On the other hand, the "non-developable rustic land", mainly agricultural land, shows an average LST of 29.08
ºC, higher than the “single family homes“.

In summary: the artificialized areas observe an average LST that is 7.2% higher than the non-artificialized ones.

Hottest Uses LST Coolest Uses LST

Logistics Activity 34,4 Rural (Spetial Protection) 26,17
Industry 33,29 Rural (Protection) 28,16
Dev. Land (economic ac 31,89 Green Areas 28,23
Tertiary 31,39 Hydrographic system 28,61
Roads and Streets 31,34 Detached Houses 28,62
Compact City 31,05 Agricultural Land 29,08
Mixed Area 31,04 Residential Blocks 29,6
Mixed Use Development 30,82 Dev. Land (residential) 29,71
Old City 30,32 Semi-Detached Houses 30,3
Table 1. LST by land use

Once the LST was obtained at daytime by means of Landsat (30 m / pixel), the LST at nighttime was obtained through
MODIS (1 km / pixel). Figure n. 10 compares the LST of day (Landsat) with the LST at night. Figure 11 shows the
nighttime UHI of MRB.

Figure 10. LST (daytime Landsat vs. nighttime Modis)

10793 - 11 V. 2 (p.8 of 16) / Color: No / Format: A4 / Date: 9/11/2018 6:43:50 AM

SPIE USE: ____ DB Check, ____ Prod Check, Notes:

 Please verify that (1) all pages are present, (2) all figures are correct, (3) all fonts and special characters are correct, and (4) all text and figures fit within the red
margin lines shown on this review document. Complete formatting information is available at http://SPIE.org/manuscripts

Return to the Manage Active Submissions page at http://spie.org/submissions/tasks.aspx and approve or disapprove this submission. Your manuscript will not
be published without this approval. Please contact author_help@spie.org with any questions or concerns.

Figure 11. UHI of Metropolitan Region of Barcelona (nighttime)

As can be seen in Figure 10, the morphology of UHI during daytime differs sharply from nighttime UHI. During the day,
the metropolitan center (the city of Barcelona) does not show the highest metropolitan LST. They are the industrial areas
of the first metropolitan ring (that surrounds the city except for the North side, where Collcerola Park closes the ring) that
reveal the highest surface temperatures. The extreme LST are also located in the industrial areas and in the compact centers
located in the Pre-coastal Depression, which runs parallel to the sea, beyond the Coastal Mountains (Garraf, Collserola and
so on). The agricultural areas of the Pre-coastal Depression, especially the vineyards, also observe elevated LST. On the
opposite side, the forested areas show the lowest temperatures at metropolitan scale: the spaces characterized by a high
quantity and quality of the vegetation observe the most moderate LST in the MRB, as evidenced by the high negative
correlation between the NDVI and the daytime LST.

During the night, the spatial distribution of the LST varies in a meaningful way. The compact urban centers, and especially
the city of Barcelona, denote the highest LST. On the other hand, in rural areas, the reduction of night temperatures is
pronounced. The agrarian zones, and especially the vineyards, observe a much more pronounced cooling than the urban
spaces. And the highest LSTs move from the Pre-coastal Depression to the Coastline. The industrial areas, on the other
hand, also show a marked reduction of the LST: they cool much more sharply than the compact residential fabrics. And
the wooded areas, except for those located at very high altitudes (> 1,000 meters), also observe, like the urban centers, a
greater thermal inertia. Forests preserve surface temperature much more than open spaces (table. 2).

CORINE LAND COVERS LST (Daytime) LST (Nighttime) Difference

High and Medium Density 30,35 17,63 12,72
Urban Sprawl 28,8 16,17 12,63
Industrial and Commercial 31,21 16,52 14,69
Other Urban Land Uses 29,72 17,06 12,66
Agricultural Areas 30,07 15,06 15,01
Forest Areas 26,93 15,23 11,7
Other Rural Land Uses 28,35 15,32 13,03
Table 2. LST (day and night) by CLC

6. CASE STUDIES

The cases used in this research are the “planning areas” with a master plan: 1. Parc Central (figure 12) and 2: Coll Favà
(figure 13), located in the municipality of Sant Cugat del Vallés (10 km from Barcelona).

10793 - 11 V. 2 (p.9 of 16) / Color: No / Format: A4 / Date: 9/11/2018 6:43:50 AM

SPIE USE: ____ DB Check, ____ Prod Check, Notes:

 Please verify that (1) all pages are present, (2) all figures are correct, (3) all fonts and special characters are correct, and (4) all text and figures fit within the red
margin lines shown on this review document. Complete formatting information is available at http://SPIE.org/manuscripts

Return to the Manage Active Submissions page at http://spie.org/submissions/tasks.aspx and approve or disapprove this submission. Your manuscript will not
be published without this approval. Please contact author_help@spie.org with any questions or concerns.

The comparison between both planning areas generally provides an improved evaluation of the traditional urban
development indicators at Coll Favà:

 Built-up area index at Coll Favà (0.61) is lower than in Parc Central (0.85).
 Public land (systems) takes up a greater % of the area (65.87 % with respect to 59.27 %).
 The percentage of open space (roads and green areas) is also higher at Coll Favà (59.21 %) than in Parc Central
(53.36 %).
 The percentage of pervious land is 53.99 % in Coll Favà, and 37.26 % in Parc Central.

Figure 12. Parc Central

Figure 13. Coll Favà

10793 - 11 V. 2 (p.10 of 16) / Color: No / Format: A4 / Date: 9/11/2018 6:43:50 AM

SPIE USE: ____ DB Check, ____ Prod Check, Notes:

 Please verify that (1) all pages are present, (2) all figures are correct, (3) all fonts and special characters are correct, and (4) all text and figures fit within the red
margin lines shown on this review document. Complete formatting information is available at http://SPIE.org/manuscripts

Return to the Manage Active Submissions page at http://spie.org/submissions/tasks.aspx and approve or disapprove this submission. Your manuscript will not
be published without this approval. Please contact author_help@spie.org with any questions or concerns.

From an urbanistic point of view, the vast majority of the indicators (table 3) suggests a better climate performance of Coll
Favà: lower built up area, greater proportion of public and open space and a higher level of pervious land. Only one
indicator (the tree canopy) favors Parc Central.

Gross Net Green

Public Private % Open Green
Area % Public Built‐up Building Building Open Land Land
Pervious Impervious % Pervious surface
Canopy
Land Land (planning)
Index Index
Parc Central 324,696 192,449 132,247 59.27% 277,447 0.8545 20.979 173,248 53.36% 64,546 108,702 37.26% 64,146 74,547

Coll Favà 567,861 374,029 193,832 65.87% 348,9 0.6144 1.800 336,213 59.21% 181,512 154,701 53.99% 176,856 64,878

Table 3. Urban indicators

Despite its worst traditional “planning indicators”, Parc Central is potentially likely to experience a better thermal
performance due to the more compact and concentrated nature of their green areas. In the opposite, Coll Favà shows a
more fragmented structure of the green zones. Parc Central will also probably benefit from an improved microclimate,
due to the greater density of its tree canopy, unlike that of Coll Favà, which is much poorer and bare.

This hypothesis of the best climate performance of Parc Central is confirmed by the following indicators:

 The NDVI analysis reveals a very different structure between Coll Favà and Parc Central. While in the first zone
the areas with dense vegetation is marginal, occupying the perimeter limits at the east and west of the area, Parc
Central shows abundant vegetation in the central core of the area.
 The average NDVI in Coll Favà is 0.3435, which is less than that of Parc Central, which is 0.4529. This means
that PC has an average of vegetation quality far higher than Coll Favà.
 Parc Central has an NDVI close to the municipal average, which includes almost 50% protected natural spaces,
Natural Park of Collserola.
 NDVI analysis at a detailed scale (IGCC) confirms (see figure 14) the highly differentiated quality of the
vegetation in both areas.
 The aboveground vegetation (canopy tree) in Parc Central fulfils a fundamental role, unlike Coll Favà, where the
“green” areas do not have dense foliage in most of the area, even in the central axis of the neighborhood.

Figure 14. NDVI. Source: IGCC

10793 - 11 V. 2 (p.11 of 16) / Color: No / Format: A4 / Date: 9/11/2018 6:43:50 AM

SPIE USE: ____ DB Check, ____ Prod Check, Notes:

 Please verify that (1) all pages are present, (2) all figures are correct, (3) all fonts and special characters are correct, and (4) all text and figures fit within the red
margin lines shown on this review document. Complete formatting information is available at http://SPIE.org/manuscripts

Return to the Manage Active Submissions page at http://spie.org/submissions/tasks.aspx and approve or disapprove this submission. Your manuscript will not
be published without this approval. Please contact author_help@spie.org with any questions or concerns.

The LST analysis from the LANDSAT 8 image confirms the better climatic performance of Parc Central, with an average
of 32.65 degrees Celsius when compared with Coll Favà, with an average LST of 35.54 degrees:

 This difference of almost 3 degrees is highly significant at an intra-urban level, and does not seem to be the result
of differential geographic factors, given the proximity of the two areas.
 The LST histograms for Parc Central and Coll Favà confirm the different climatic results of both areas. LST
distribution in Parc Central (blue) is clearly skewed to the left, i.e. towards lower temperatures. Coll Favà (green)
however shows a clear structure, with a skew to the right.
 The analysis of both robust averages (32.53 PC and 35.44 CF) and medians (32.55 and 35.48 respectively)
confirms the diagnosis.

Figure 15. LST (Landsat 8)

Figure 16. Histograms of land surface temperatures (CF, green; PC, blue)

If we segment the land surface temperatures according to the type of space, both public and private, the results are
noticeably similar to those found for all areas in Parc Central and Coll Favà. Public spaces of Parc Central have an average
temperature of 32.32 Celsius, compared to 35.64 in Coll Favà. If we take a look only on private land, Parc Central obtains
an average LST of 33.15 degrees Celsius, which is also clearly below that of Coll Favà, at 35.25 degrees. The better thermal

10793 - 11 V. 2 (p.12 of 16) / Color: No / Format: A4 / Date: 9/11/2018 6:43:50 AM

SPIE USE: ____ DB Check, ____ Prod Check, Notes:

 Please verify that (1) all pages are present, (2) all figures are correct, (3) all fonts and special characters are correct, and (4) all text and figures fit within the red
margin lines shown on this review document. Complete formatting information is available at http://SPIE.org/manuscripts

Return to the Manage Active Submissions page at http://spie.org/submissions/tasks.aspx and approve or disapprove this submission. Your manuscript will not
be published without this approval. Please contact author_help@spie.org with any questions or concerns.

performance of PC is not limited, therefore, to public space, it also radiates to the rest of the urban fabric, affecting private
areas in a singularly significant manner.

Particularly significant is the fact that public spaces in Coll Favà has an average LST higher than private land, contrary to
what would be expected. This shows the inadequate design of the green and open spaces at Coll Favà. One example of this
is the “Jaume Tubau” sports zone, whose artificial grass surface makes a negative contribution to the UHI in the area. This
is also true for open (pervious) areas without vegetation or bare.

---

In addition to the analysis based on remote sensing, "in situ" analysis of the LST and the LSAT has been carried out in order
to verify the climatic performance of Parc Central and Coll Favà. A set of field measurements was taken with this purpose:

 Parc Central - Night (night from 2 to 3/09/16): 19 thermal images were made, 4 videos and 20 measurements of
climatic air parameters.
 Parc Central - Day (midday, 3/09/16): 63 thermal images and 20 videos and 12 atmospheric readings.
 Coll Favà - Day (midday, from the 3 and 4/9/16): 31 thermal images, 13 vídeos and 12 atmospheric readings.
 Coll Favà - Night (night 4/9/16): 53 thermal images, 16 videos and 12 atmospheric readings.

Figure 17. Sample of LST and LSAT measures

The analysis of day-time temperatures of pavements or ground surfaces confirms the better thermal performance of Parc
Central. All the surfaces obtain an LST that is lower in Parc Central than at Coll Favà. Bare land obtains greater differences
(13.2 degrees), as asphalt (10.3 degrees). The paved areas in parks are in an intermediate level (8.5 degrees), as clay tiles
and sidewalk edge (8.2 degrees). Lawn-type areas present a smaller thermal difference (4.7 degrees cooler at Parc Central
than at Coll Favà).

10793 - 11 V. 2 (p.13 of 16) / Color: No / Format: A4 / Date: 9/11/2018 6:43:50 AM

SPIE USE: ____ DB Check, ____ Prod Check, Notes:

 Please verify that (1) all pages are present, (2) all figures are correct, (3) all fonts and special characters are correct, and (4) all text and figures fit within the red
margin lines shown on this review document. Complete formatting information is available at http://SPIE.org/manuscripts

Return to the Manage Active Submissions page at http://spie.org/submissions/tasks.aspx and approve or disapprove this submission. Your manuscript will not
be published without this approval. Please contact author_help@spie.org with any questions or concerns.

The analysis of nocturnal temperatures on ground surfaces confirms the lower temperatures of Parc Central when
compared to Coll Favà. Bare ground is the surface type that reveals a higher temperature difference, while grass and
asphalt reveal a higher level of similarity in both areas (figure 18).

Figure 18: “In-situ” night-time LST (different surfaces)

In conclusion, the analysis of land surface temperature measurements obtained by using the thermal camera broadly
confirms the better performance of Parc Central:

 The average for the group of measurements obtained for day-time LST levels is 34.08 degrees Celsius, for the
Parc Central area, compared to 44.30 for Coll Favà: a difference of more than 10 degrees.
 At night time, the average Parc Central reading is 24.28 degrees, compared to 24.91 for Coll Favà. The
temperatures tend to equal at night, probably due to the significant cooling of permeable, vegetation-free surfaces
at Coll Favà.
 Average surface temperatures decreases by 9.8 degrees in Parc Central, and by 19.39 degrees in Coll Favà.
 Parc Central is, therefore, much cooler during the day, although at night temperatures tend to be similar.

LST LSAT
Day‐time Night‐time Difference Day‐time Night‐time Difference

Parc Central 34.08 25.78 8.3 33.12 25785 7.33

Coll Favà 44.3 27.05 17.25 38.14 27.05 11.09

Difference 10.22 1.27 5.03 1.27

Table 4. LST and LSAT “in situ” measures

The analysis of the air temperatures (LSAT) in-situ confirms the previous conclusions:

 The average daytime temperature of Parc Central was 33.12 degrees Celsius, 5.03 degrees lower than that of Coll
Favà (38.14).
 The average nighttime temperature is also cooler in Parc Central (25.78 degrees) than in Coll Favà (27.05
degrees), even though the difference drops to 1.27 degrees Celsius.
 This reduction of the differences is due to the relatively higher cooling areas in Coll Favà (especially bare ground)
with respect to Parc Central: 11.09 degrees vs. 7.33.
 However, night-time measurements at Coll Favà exceed the maximum comfort temperatures (24-26 degrees).

10793 - 11 V. 2 (p.14 of 16) / Color: No / Format: A4 / Date: 9/11/2018 6:43:50 AM

SPIE USE: ____ DB Check, ____ Prod Check, Notes:

 Please verify that (1) all pages are present, (2) all figures are correct, (3) all fonts and special characters are correct, and (4) all text and figures fit within the red
margin lines shown on this review document. Complete formatting information is available at http://SPIE.org/manuscripts

Return to the Manage Active Submissions page at http://spie.org/submissions/tasks.aspx and approve or disapprove this submission. Your manuscript will not
be published without this approval. Please contact author_help@spie.org with any questions or concerns.

The best climate performance of Parc Central compared to Coll Favà is evident in all the analyses carried out. Both the
study of the daytime LST resulting from remote sensing, as well as the surface temperatures of day and nighttime resulting
from the field study, show that the design of the public spaces and the green areas of Parc Central offers a better thermal
performance than Coll Favà. Results that are also confirmed by the LSAT analysis.

7. CONCLUSIONS

The main conclusions of the work can be summarized as follows:

1. Despite having better traditional "planning indicators", Coll Favà observes a poorer "climate performance" than
Parc Central.

2. The key factor is the differences in green structures. The tree canopy is better in Central Park than Coll Favà. In
addition, it is likely that Parc Central has a greater thermal reach due to the greater compactness and concentration
of the green infrastructure than the fragmentation of Coll Favà.

3. The NDVI (from Landsat 8) analysis reveals a very different structure between Coll Favà and Parc Central. While
in the first neighborhood the areas with dense vegetation are marginal, occupying the perimeter limits at the east
and west of the area, Parc Central shows abundant vegetation in the central core of the area. Parc Central has a
better microclimate due to the greater density of its tree canopy, unlike that of Coll Favà, much poorer and bare.

4. The LST analysis (from the LANDSAT 8) confirms the better climatic performance of Parc Central, with an
average of 32.65 degrees Celsius when compared with Coll Favà, with an average LST of 35.54 degrees.

5. “In situ” analysis also confirms the better climatic performance of Parc Central. Both day and night, the LST and
the LSAT are cooler in Parc Central than in Coll Favà.

We can conclude that: a) the use of both analysis, remote sensing technologies and “in situ” measures of land and air
surface temperatures, can help to identify UHI; b) urban morphology plays a key role to generate and control UHI; and
c) landscape design and selection of vegetation has a particular relevance to promote resilience to heat waves.

Acknowledgements
The authors wish to thank the help received by the Government of Spain (Ministry of Economy and Competitiveness,
Mineco) as well as by the European Union (European Regional Development Fund).

REFERENCES
[1] Intergovernmental Panel on Climate Change, [Climate Change 2014. Synthesis Report], Contribution of Working
Groups I, II and III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change, IPCC,
Geneva, Switzerland, 151 pp.
[2] Trenberth, K.E., Jones, P.D., Ambenje, P., Bojariu, R., Easterling, D., Klein, T. A., Parker, D., Rahimzadeh, F.,
Renwick, J.A., Rusticucci, M., Soden, B. and Zhai, P., [Observations: Surface and atmospheric climate change.
In Climate Change 2007: The Physics Science Basis], Cambridge University Press: Cambridge, New York,
(2007).

10793 - 11 V. 2 (p.15 of 16) / Color: No / Format: A4 / Date: 9/11/2018 6:43:50 AM

SPIE USE: ____ DB Check, ____ Prod Check, Notes:

 Please verify that (1) all pages are present, (2) all figures are correct, (3) all fonts and special characters are correct, and (4) all text and figures fit within the red
margin lines shown on this review document. Complete formatting information is available at http://SPIE.org/manuscripts

Return to the Manage Active Submissions page at http://spie.org/submissions/tasks.aspx and approve or disapprove this submission. Your manuscript will not
be published without this approval. Please contact author_help@spie.org with any questions or concerns.

[3] Oke, T.R., [Boundary Layer Climates. Second], Taylor & Francis, 416 pp. (1987).
[4] Weng, Q., “Thermal Infrared remote sensing for urban climate and environmental studies: Methods, applications,
and trends”, ISPRS Journal of Photogrammetry and Remote Sensing, Volume 64, Issue 4, July 2009, Pages 335-
344 (2009).
[5] Oke, T.R., “Towards better scientific communication in urban climate”, Theoretical and Applied Climatology,
84: 179–19 (2006).
[6] Zhou, W., Huang, G., Cadenasso, M. L., “Does spatial configuration matter? Understanding the effects of land
cover pattern on land surface temperature in urban landscapes”, Landscape and Urban Planning, Volume 102,
Issue 1, 30 July 2011, Pages 54-63 (2011).
[7] Guillevic, P., Koster, R., Suarez, M., Bounoua, L., Collatz, G., Los, S., et al., “Influence of the interannual
variability of vegetation on the surface energy balance - A global sensitivity study”. Journal of Hydrometeorology,
617–629 (2002).
[8] Meng, C. L., Li, Z. L., Zhan, X., Shi, J. C., and Liu, C. Y., “Land surface temperature data assimilation and its
impact on evapotranspiration estimates from the Common Land Model”, Water Resources Research, 45,1-14
(2009).
[9] Schwarz, N., Lautenbach, S., and Seppelt, R., “Exploring indicators for quantifying surface heat islands of
European cities with MODIS land surface temperatures”, Remote Sensing of Environment, 115, 3175–3186
(2011).
[10] Tomlinson, C. J., Chapman, L., Thornes, J. E., and Baker, C. J., “Derivation of Birmingham's summer surface
urban heat island from MODIS satellite images”, Journal of Climatology, 32: 214-224 (2012).
[11] Stisen, S., Sandholt, I., Norgaard, A., Fensholt, R. and Eklundh, L., “Estimation of diurnal air temperature using
MSG SEVIRI data in West Africa”, Remote Sensing of Environment, 10(2): 262-274 (2007).
[12] US Environmental Protection Agency, “Reducing urban heat islands: Compendium of strategies. Draft”,
https://www.epa.gov/heat-islands/heat-island-compendium, (2008).
[13] Alcoforado, M. J. and Matzarakis, A., “Planning with Urban Climate in Different Climatic Zones”,
Geographicalia, 57: 5-39 (2010).
[14] Matzarakis, A., “Urban climate research in Germany”, IAUC Newsletter, 11: 4-6 (2005).
[15] Baumüller, J., Flassak, T., Schädler, G., Keim, M. and Lohmeyer, A., [‘Urban climate 21’—Climatological basics
and design features for ‘Stuttgart 21’], Report of Research Center for Urban Safety and Security, Kobe University,
1: 42-52 (1998).
[16] Scherer, D., Fehrenbach, U., Beha, H-D. and Parlow, E., “Improved concepts and methods in analysis and
evaluation of the urban climate for optimizing urban planning process”, Atmospheric Environment, 33: 4185-
4193 (1999).
[17] Architectural Institute of Japan, [Urban Environmental Climatic Atlas: urban development utilizing climate
information], Tokyo, Japan (2000).
[18] Senate Department for Urban Development Berlin (SDUDB), “Environmental Atlas”,
https://www.stadtentwicklung.berlin.de/umwelt/umweltatlas/edua_index.shtml, (2003).
[19] Ren, C., Ng, E. and Katzschner, L, “An investigation into developing an urban climatic map for high density
living – initial study in Hong Kong”, 2nd PALENC, Greece (2007).
[20] Stewart, I. D., Oke, T. R. and Krayenhoff, E.S.: Evaluation of the ‘local climate zone’ scheme using temperature
observations and model simulations. International Journal of Climatology, DOI: 10.1002/joc.3746. (2014).
[21] Du, H., Cai, W., Xu, Y., Wang, Z., Wang, Y., and Cai, Y., “Quantifying the cool island effects of urban green
spaces using remote sensing Data”, Urban Forestry & Urban Greening(27), 24-31 (2017).
[22] Oliveira, S., and Andrade, H. V., “The cooling effect of green spaces as a contribution to the mitigation of urban
heat: A case study in Lisbon”, Building and Environment, 2186-2194 (2011).
[23] Arellano, B. and Roca, J., “Identifying Urban Heat Island at Medium Scale: The Barcelona Case”, 11th Congress
Virtual City and Territory, 11 VCT Proceedings book, 798-812 (2016).

10793 - 11 V. 2 (p.16 of 16) / Color: No / Format: A4 / Date: 9/11/2018 6:43:50 AM

SPIE USE: ____ DB Check, ____ Prod Check, Notes: