UNIVERSITY OF GOTHENBURG 
Department of Earth Sciences 
Geovetarcentrum/Earth Science Centre 
 
 
 
 
 
 
 
 
 
 
 
 
 
 Pedestrian movement and 
 
 
 exposure to heat stress 
 
 
 in Gothenburg, Sweden 
 
  
 
 
  
 
 
  
 
 
  
 
 
 Madeleine Cleve rstam Vikström 
  
  
  
 
  
 
 
  
 
 
  
 
  
ISSN 1400-3821                                                                                                                                  B1247 
 Master of Science (120 credits) thesis Göteborg 2023 
 
Mailing address Address Telephone  Geovetarcentrum 
Geovetarcentrum Geovetarcentrum 031-786 19 56  Göteborg University 
S 405 30 Göteborg Guldhedsgatan 5A    S-405 30 Göteborg 
    SWEDEN 
 
 
Pedestrian movement and exposure to heat stress in Gothenburg, Sweden
by Madeleine Cleverstam Vikström
2023
Degree of Master of Science (120 credits) with a major in Geography (30 credits)
at the Department of Earth Sciences, University of Gothenburg.
Supervisor: Fredrik Lindberg (University of Gothenburg) & Meta Berghauser Pont (Chalmers)
Examiner: Heather Reese (University of Gothenburg)
Abstract
An area in Gothenburg, Sweden, representing various planning designs and microclimates, were
investigated, in order to estimate how the study area manifests potential pedestrian movement patterns
and urban climate conditions in terms of heat stress. The study is based on the Geographical
Information System (GIS) where pedestrian movement and heat stress is calculated. Multiple research
suggests that the urban climate has an impact on the use of public space but few studies show people's
movement patterns in the city and its impact on heat stress. The results of this thesis support the
arguments for the use of GIS-analyses in studying pedestrian movement and heat stress in future
urban design and planning projects. By designing the physical component of a place, it can influence
the site-specific urban climate which can have a direct impact on people’s attendance, perceptions and
emotions related to that place.
Keywords: urban climate, heat stress, Space Syntax, pedestrian movement, urban morphology.
Acknowledgements
This thesis is a result of a 30-credit master’s thesis course in Geography conducted in the spring of
2023. Completing this thesis not only marks an ending of a five-year long study period, it also
highlights my interests and knowledge gained from three years of studying urban planning and two
years as a geographer. With these words, I would like to thank a few people without whom this thesis
would not have been feasible.
First, I am glad that I have been given the opportunity to be part of a master class with many
dedicated students who have different interests and skills. Thanks for all the moral support and
interesting discussions.
Secondly, I would like to express my gratitude to Malin Klarqvist (Stadsbyggnadsförvaltningen,
Göteborgs Stad), as well as Alexander Gösta and Cecilia Windh (Liljewall), who provided valuable
tips and guidance at the beginning of my thesis journey, helping me to explore different topics and
approaches to my research.
Lastly, I am also thankful for both my supervisors Fredrik Lindberg (University of Gothenburg) and
Meta Berghauser Pont (Chalmers). Thank you for your enthusiasm, advice and for guiding me
through this scientific maze!
Madeleine Cleverstam Vikström
Gothenburg, May 2023
Table of Contents
1. Introduction........................................................................................................................................ 1
1.1 Background..................................................................................................................................1
1.2 Aim.............................................................................................................................................. 2
2. Literature Review and Key Themes.................................................................................................3
2.1 Pedestrian Movement and Physical Urban Form........................................................................ 3
2.1.1 Theory of Natural Movement............................................................................................. 3
2.1.2 Density of the urban environment...................................................................................... 5
2.1.3 Configuration of the urban environment............................................................................8
2.2 Urban Climate and Physical Urban Form....................................................................................8
2.2.1 Building geometry.............................................................................................................. 8
2.2.2 Material and surface properties.........................................................................................9
2.2.3 Vegetation.........................................................................................................................10
2.3 Human Thermal Comfort - Pedestrian Movement and Behavior in Urban Environments....... 10
3. Study Area........................................................................................................................................ 13
4. Method and Material....................................................................................................................... 16
4.1 Research Design........................................................................................................................ 16
4.2 Quantitative Methods................................................................................................................ 18
4.2.1 Defining urban typologies................................................................................................18
4.2.2 Mapping of Mean Radiation Temperature (Tmrt)............................................................18
4.2.3 Mapping of pedestrian movement.................................................................................... 20
4.2.4 Mapping of pedestrian movement and heat stress........................................................... 21
5. Results............................................................................................................................................... 23
5.1 Spacematrix and Tmrt>55 within Different Urban Typologies.................................................23
5.1.1 Guldheden........................................................................................................................ 24
5.1.2 Krokslätt...........................................................................................................................25
5.1.3 Stigberget......................................................................................................................... 25
5.1.4 Haga.................................................................................................................................26
5.1.5 Änggården........................................................................................................................ 27
5.2 Pedestrian Movement and Tmrt>55 within the Whole Study Area.......................................... 27
5.2.1 High Attraction Reach with high and low Tmrt>55........................................................ 29
5.2.2 High Angular Betweenness with high and low Tmrt>55.................................................30
5.2.3 High Attraction Reach and Angular Betweenness with high and low Tmrt>55..............31
6. Discussion..........................................................................................................................................32
6.1 Urban Typologies and Heat Stress - Insights from Spacematrix Method................................. 32
6.2 Heat Stress and Pedestrian Movement Patterns........................................................................ 33
6.2.1 The urban geometry: Key factors in pedestrian movement & heat stress....................... 33
6.2.2 Planning and designing streets for pedestrian comfort................................................... 34
6.3 Limitations of the Study............................................................................................................ 36
6.4 Implications of the Study, Recommendations and Further Research........................................36
7. Conclusions....................................................................................................................................... 38
References............................................................................................................................................. 39
Appendices............................................................................................................................................43
1. Introduction
1.1 Background
“If cities are to invite people to walk and bicycle as well as to develop lively and
attractive city areas, the climate between buildings is one of the most important target
areas. Careful climate planning should be a requirement for all new buildings.”
(Gehl, 2010:174)
One major problem of the future is heat stress in urban environments (Folkhälsomyndigheten, 2015).
The warm and sunny summer of 2018, in Sweden, prompted numerous inquiries about the country's
forthcoming thermal conditions. As a result of global climate change, Sweden's climate is anticipated
to be impacted with increased summertime temperatures, as well as more frequent occurrences of heat
waves (Thorsson et al., 2017). Research shows that the microclimate in the public space between
buildings affects people's activity and experience of a place, but also people’s health. There is thus a
relationship between climatic factors and perception, emotions and attendance in public spaces. In
Scandinavia, the warmer the air temperature, the less windy and the clearer the sky, the overall mood
elevates as people make comments about the nice weather (Gehl, 2010). Weather parameters affect
both attendance in the place as well as how citizens assessed and felt in these places. An easily
observable trend is therefore that higher temperatures increase the number of people in public spaces
up to a certain temperature, while after that it can cause discomfort and people start to avoid being
outside. Higher temperatures are therefore to be expected due to climate change which will result in
people often going over this threshold (Thorsson et al., 2006) .
Space Syntax, a theory and method on urban form, has contributed to a better understanding of the
relationship between the configuration of public space in urban environments and various societal
aspects such as human behavior, mainly pedestrian movement and social processes (Hillier & Hanson,
1984). However, knowledge of experiencing the city has not been as explored and even less when it
comes to pedestrian movement and their exposure to heat stress. The changing climate entails a need
to create sustainable cities that can absorb and reduce the effects of climate-related risks, such as
extreme temperatures. Eliasson (2000) conducted a study by interviewing urban planners from
different towns in Sweden, one of which was Gothenburg. The study showed that the participants are
interested in climatic aspects but it has a low impact on the planning process. Five explanatory
variables are represented, two of which indicate that there are flaws within technology and knowledge
when it comes to planning a city based on a climate perspective where comfort is central.
1
The urban climate is thus a developing subject that requires a standard. Stewart and Oke (2012)
developed what are called Local Climate Zones (LCZ), a standard typology that classifies urban
environments at the local scale, including building type and the level of impervious surfaces which
has been proven to give a good indication of the local temperature. Urban typologies and
morphologies are therefore essential in order to understand how to design more sustainable and
healthy cities that can adapt to climate change and reduce the effects of heat stress on human health
and well-being.
1.2 Aim
The purpose of the study is to assess pedestrian movement and exposure to urban climate focusing on
heat stress. The main question for this study is how pedestrian movement patterns and urban climate
conditions in terms of heat stress are manifested in Gothenburg. In order to answer the question, a
quantitative approach is applied with GIS-analyses. Which variables that are important for
understanding pedestrian movement and heat stress and how they are apparently linked will therefore
also be examined.
2
2. Literature Review and Key Themes
2.1 Pedestrian Movement and Physical Urban Form
2.1.1 Theory of Natural Movement
One type of modeling cities, focuses on the relationship between humans and the environment
(Marcus, 2018). The theory of reversible occlusion arose from Gibson (1986) who describes that
people navigate in the environment, not through stationary signals but through scanning the
environment while moving. The author illustrates that with a particular illustration (see Figure 1) of
what happens when people move in the urban environment. The author describes that the 'perceptual
spatial unit' changes when the observer moves in the space due to the urban structure of built forms. It
means that when you see something, you then walk towards it (ibid).
Figure 1- Illustration of Gibson (1986:190), visual perception in the urban space, as seen from above.
It is interesting to connect Gibson's (1986) theory with Space Syntax which is an analysis method that
describes the interconnectivity of public spaces and it can be used in the planning of the city's
structure and architecture (Marcus, 2018). The theory was developed by Bill Hillier and Julienne
Hanson at The Bartlett, University College London (Hillier & Hanson, 1984). In 1984, Hillier and
Hanson wrote the book The Social Logic of Space, which deals with the physical configuration of
public spaces. The analysis model is based on well-founded mathematical models that describe the
spatial connection and accessibility found in the street network (i.e. pedestrian and bicycle network)
and how the continuous movement works. The authors point out that the city's spaces constitute a
system that affects the use of the city (ibid). Within Space Syntax there is a central concept, axial
lines, which make up the network segments and can be defined as lines of sight.
“An axial line is the longest sightline that indicates a movement path in a certain space
within the built environment. In urban studies, each axial line represents a public urban
space that connects to other public urban spaces.”
(Yamu et al., 2021:5)
3
Figure 2 - Map A. illustrates ground plan represented in the usual way whereas map B. is an axial map, representation of
spatial form (Hillier & Hanson, 1984:91)
To describe the dynamics of social life, the axial map is, as mentioned, a representation of continuous
urban areas that are structured based on buildings, infrastructure and landscape elements that capture
both accessibility and visibility (see Figure 2) (Hillier & Hanson, 1984). Axial maps are the same as
Gibson (1986) describes but it covers a much bigger area. The lines represent the spatial environment,
not only lines of sight, it is also about accessibility. The axial map, therefore, represents the ability of
people to see things and also to access it by moving (Marcus, 2018). Having that said, it means that all
human activity takes place in the room and this does not only happen with one individual. It rather
takes several people in a room to create patterns and configurations which in addition become
activities such as meeting, avoiding and integrating. When one plans and builds up an area, it is the
relationship between the rooms and the configuration that is at the center, one does not build
configurations based on groups of people but on the basis of the room. The authors also describe that
it is through space that one can explain people's behavior in the city, which in turn creates conditions
(Hillier & Hanson, 1984). Hillier et al. (1993) explain that the theory of Natural Movement Theory
creates conditions for people's natural movement patterns and therefore, often, the buildings have
more activities on ground level. The theory is related to the triangle of attraction, configuration and
movement in the context of Space Syntax (see Figure 3). The authors argue through the theory that
people have a preference in how they move in the city in a natural way and that the planning and
design of the city, including the buildings, should take these natural movement patterns into account.
This means that through a natural and efficient movement pattern it contributes to a more livable and
walkable urban environment.
Within urban morphology, there are therefore multiple terms which go hand in hand. Figure 3
represents how the triadic relation from Natural Movement Theory is connected to the variables of
spatial forms centrality and density that Bobkova et al. (2017) uses to describe urban morphology.
4
Figure 3 - Triadic relation among attraction, configuration and movement from Natural Movement Theory (modified from
Hillier et al., 1993) with its connection to elements of urban space in the gray squares and spatial variables in blue squares
(modified from Beghauser Pont et al., 2017)
2.1.2 Density of the urban environment
Attraction within Natural Movement Theory refers to the social and economic activity that attracts
people to different locations. Attraction depends from person to person, but it can be influenced by
different factors such as shops, public cultural events and public amenities (Hillier et al., 1993).
Attraction goes hand in hand with density within urban morphology. Density is important when it
comes to urban planning and design. Berghauser Pont and Haupt (2010) developed the Spacematrix
method which describes how to measure urban density, a multi variable phenomenon that has a
correlation with the urban typologies. There are four variables that are included - Floor Space Index
(FSI), Ground Space Index (GSI), Open Space Radio (OSR) and number of floors (L). Figure 4
describes the different calculations to obtain the variables. They are based on area (m2), floor area
(m2) and footprint (m2). FSI describes the built intensity of the plan area, GSI correlates with the built
compactness or ground coverage of the plan area, OSR is the spaciousness of the non-built space and
L represents the average number of storeys. The calculations and relationships of these play a decisive
role in the building density, light conditions but also usability (Berghauser Pont & Marcus, 2020).
5
Figure 4 - Morphological variables based in Spacematrix model (Modified from Berghauser Pont et al., 2021:57)
Figure 5- Spacematrix with the various types of urban areas (modified from Berghauser Pont & Haupt, 2007) - Relationship
between GSI, FSI, OSR and L with three different types of urban areas with 75 dwelling per hectare (modified from Nes et
al., 2012)
By combining these variables, a scatterplot can be represented (see Figure 5). In the Spacematrix
diagram, the y-axis represents FSI which indicates the built intensity in an area and the x-axis reflects
the compactness or coverage of the development which is GSI. There are also two variables that fan
out over the diagram, OSR that represents the spaciousness and L that describes the average number
of floors (Nes et al., 2012). Based on the Spacematrix, Berghauser Pont and Haupt (2010) showed that
contrasting typologies can have the same density. In figure 5 there are three different neighborhoods
and buildings (different urban typologies) but they have the same density of 75 dwelling per hectare.
This is due to the fact that the FSI in all cases is the same but the GSI in the first case is lower
6
compared to the other ones. This means that even though they have the same FSI, they still have
different GSI, OSR and L, which means that their position in the Spacematrix is different.
This can be studied further in that depending on where the different neighborhoods are displaced in
the Spacematrix, there are similarities in terms of spatial structures and therefore urban typologies
(see Figure 5). For instance, the low-rise areas are gathered together in one zone, where FSI, GSI and
L are low and OSR is high. Compact blocks are also grouped together in another zone, where FSI,
GSI and OSR are high, but not particularly L (Berghauser Pont & Haupt, 2010). This means that this
method and statistical clustering analysis makes it possible to map urban typologies for an entire city.
Studies have shown that in Gothenburg, spacious low-rise (BC1) building types are dominated and
then comes compact low-rise buildings (BC2). Based on the map in figure 6, it is possible to see a
clear structure in how the city is built up based on its urban typology. In the historical cores of the city,
the densest and most compact building types are built, compact mid-rise buildings (BC5) and dense
mid-rise buildings (BC3). The building type dense low-rise (BC4) is found in the second zone and the
main arteries. Moving on to the third zone, compact mid-rise (BC5) is dominated and is surrounded
by compact low-rise (BC2). Around the city, a blend of Spacious mid-rise (BC6) and Spacious
low-rise (BC1) can be seen. The study also shows that there is a significant difference between low
and high dense building types (Berghauser Pont et al., 2019).
Figure 6 - Spatial distribution of urban typologies in Gothenburg. Spacematrix showing the density profiles of the building
types with FSI on the y-axis and GSI on the x-axis also a circle chart comparing the numerical distribution of urban
typologies (Berghauser Pont et al., 2019:11)
7
2.1.3 Configuration of the urban environment
Spatial configuration of the urban environment is one of the three components of the triangle. Spatial
configuration is a description of how places are interconnected. This can be measured with centrality
measures and is central to Space Syntax (see Figure 3). By studying people's movement patterns, it is
possible to identify various central places that are important to both people and the city (Hillier et al.,
1993). Centrality is thus a concept used to describe the degree of importance and accessibility of a
place in a city (Berghauser Pont & Marcus, 2020).
From the triad of Natural Movement Theory in figure 3, it is possible to see that planning and design
is changing configuration and attraction and influences movement. Movement within the triadic
relation is referred to the way people move and interact with the urban environment (Hillier et al.,
1993). Through many empirical studies, it is proven that there is a significant correlation between
density, centrality and pedestrian movement. It therefore means that built density refers to the intensity
of the pedestrian flow and the centrality of the streets demonstrates the distribution of the intensity
(Berghauser Pont et al., 2019). Density and centrality interact for that reason with each other and
together create effects on the city's experience and use. However, this does not mean that they follow
each other, for example there are places that have high centrality but also low density. On the other
hand, there are studies that show that high closeness drives densification such as gentrification.
Likewise, places with high closeness can have different degrees of diversity. For example, older city
centers often have high closeness and diversity due to fine-grained property division, while central
areas with high closeness are usually converted into larger retail establishments. Having that said, it
means that the combination of these variables constitutes prerequisites for both social and economic
processes, which also contribute to the fact that different human activities fit better or less well in
different places (Berghauser Pont & Marcus, 2020).
2.2 Urban Climate and Physical Urban Form
2.2.1 Building geometry
Building geometry is therefore an important aspect when it comes to the influence of the urban
climate. It includes the direction, distance and height of the buildings. This means, among other
things, that building density in the city plays a large role (Wallenberg et al., 2018). Therefore, the
concept of Sky View Factor (SVF) is widely used in urban climatic studies. It represents the ratio
between 0 and 1, it explains how much of the sky is visible from a point based on a 180-degree
perspective. If SVF = 0, the entire sky is blocked from view by different obstacles such as buildings or
trees and thus vice versa if SVF = 1, the sky is totally unobstructed by obstacles (Lindberg et al.,
2018). Building density can also be studied based on the Height/Width ratio (H/W) which describes
8
the relationship between the width of the street and the height of the buildings. H/W is also one of the
most important parameters when it comes to studying changes in the street canyon climate (Shishegar,
2013). Density is interesting to study from an urban climate perspective, as a denser built-up area have
the ability to absorb and store more heat than rural areas due to the urban structure and human
influence. This phenomenon is called the Urban Heat Island (UHI) effect and is mainly a nocturnal
phenomenon where the largest UHI can occur on clear and calm nights (Oke et al. 2017). At the same
time, it is interesting to reflect on the fact that a densely built-up area generally has fewer incoming
solar radiations that reach the building surfaces and the ground due to more shade during the day. This
therefore creates both lower air temperature (Ta) but also surface temperature (Ts) (Wallenberg et al.,
2018).
Finally, the direction of the streets is also an important factor. This means that depending on the
direction of the buildings and the time of day, the surfaces are exposed with different amounts of
incoming solar radiation. Dense street canyons that are located in an east-west direction have a lot of
incoming solar radiation during the morning and afternoon. Street canyons, on the other hand, which
are located in a north-south direction, generally have few hours of sunlight throughout the day, but are
fully lit in the middle of the day when the sun is in the south. This means that the southern facades in
the east-west street canyons are the warmest as they are exposed by the sun for a large part of the day
(Shishegar, 2013; Wallenberg et al., 2018).
2.2.2 Material and surface properties
How solar radiation is absorbed, emitted and reflected are important aspects when it comes to urban
climate. In other words, this means that the choice of both the material and the surface properties is of
great importance. From an urban climate perspective, the concept of albedo (α) is used when it comes
to studying the reflectivity of the sun's rays. Dark surfaces that have a low albedo and high emissivity
(e.g. asphalt) absorb a lot of incoming solar radiation and vice versa with light surfaces that have a
high albedo (e.g. white painted building surfaces) reflect a lot of solar radiation. This means that the
surfaces with low albedo absorb incoming solar radiation and thereby increase both the Ts and the Ta.
Therefore, it is generally warmer in dense built-up areas as the most common building materials such
as paved materials have low albedo, high density and high thermal admittance, which causes more
heat to be stored within the street canyons. During the daytime, these aspects lead to a significant
accumulation of heat in the materials, which is then released during the night, causing the lower parts
of the atmosphere to warm up (Wallenberg et al., 2018; Oke et al., 2017).
9
2.2.3 Vegetation
In order to reduce the effect of UHI in cities and also heat stress, it is recommended to increase the
amount of vegetation but also create more shade in the public space by increasing the height and
density between buildings (Lindberg et al, 2016). This means that the vegetation is important and also
decisive in reducing UHI. This happens through the evapotranspiration from water surfaces and
transpiration from vegetation which creates a cooling effect from the leaves to the surrounding urban
environment. The water vapor from the leaves of the vegetation is released into the air, which in turn
causes the Ta to cool down. Through the absorption and reflection of incoming solar radiation and the
shade of, for example, the trees, it also contributes to the Ts and the mean radiant temperature (Tmrt)
falling (Moss et al., 2019). The trees can also reduce Ts in the urban environment, such as nearby
buildings and streets, as the trees create shade and block direct sunlight (Yu et al., 2020).
Selection of vegetation, especially trees, is crucial in urban environments as not all species have the
same cooling capacity through their evapotranspiration (Moss et al, 2019). The size of tree canopy,
leaf quantity, water sensitivity and transpiration rate are important factors (Wang et al., 2021).
According to research, deciduous trees are preferable, especially preferable over conifers, as they
create more shade during the summer and let in sunlight during the winter (Lindberg et al, 2016). The
combination of vegetation is also important as it increases the volume of vegetation, which in turn
maximizes the cooling effect (Wang et al., 2021; Thorsson, 2012). The choice of trees can also allow
or limit the wind depending on the needs of the area. Trees with low canopies that are close to the
ground can reduce wind speed at ground level. It can also have the opposite effect, trees whose
canopies start high up on the tree trunk can generate stronger wind speed at ground level (Oke et al,
2017). The placement of trees is also important depending on what effect is to be achieved. To create
as much shade as possible and to contribute with the greatest effect, trees should be placed on warm
surfaces in the urban space, such as in parking lots, squares but also streets (Thorsson, 2012).
2.3 Human Thermal Comfort - Pedestrian Movement and Behavior in Urban
Environments
In the urban environment during daytime, the Ta is quite equally distributed and does not therefore
represent the reality of the urban microclimate. Tmrt is, however, a meteorological parameter that can
measure the spatial variations in thermal comfort conditions. It sums up all the short and long wave
radiation fluxes, i.e both reflected and direct, to which the human body is exposed to (Thorsson et al.,
2007). Studies have shown that depending on whether it is a sunny or shady area, Tmrt can vary by
over 30 °C while Ta remains the same (Ali-Toudert & Mayer, 2007; Chen et al., 2016; Oliveira et al.,
2011). Tmrt is therefore an important factor in determining outdoor human thermal comfort. Thermal
comfort is a condition that occurs when the heat flow between the human body and its surroundings is
10
in balance. When humans are exposed to excessive heat for a prolonged period, the human body
strives to regulate the rising body temperature, which results in physical strain that leads to thermal
discomfort. The consequences of heat stress are the health effects that have been extensively studied
and shows that heat stress may induce dehydration, exhaustion, cardiovascular problems or even death
(Oke et al., 2017). Some population groups are more vulnerable than others because they find it more
difficult to react to heat but also to regulate their body temperature. These risk groups include the
elderly, children, pregnant women, people with disabilities (both physical and mental) etc.
(Wallenberg et al., 2018). According to Lindberg et al. (2016) and Thorsson (2012), Tmrt has the most
significant impact on human thermal comfort during summer days with clear and calm weather
conditions.
“Good weather is one of the most significant criteria for assuring the ease of people’s movement
in cities, or at least weather as good as it gets given the situation, place and season.”
(Gehl, 2010:168)
With this quote, Gehl (2010) means that the quality of the urban environment has a significant impact
on how people move and behave in public spaces. The author emphasizes the importance of creating
urban environments that are comfortable and inviting for pedestrians and that encourage social
interaction and public life. In order to create a city life that is socially sustainable, Jacobs (2005)
believes that it is important to create environments that are inviting and supporting human activities in
the public space. This means that constant movement around streets and squares contributes to a more
vibrant and sustainable urban life. It goes hand in hand with what Gehl (2010) describes, in order to
create an attractive urban environment with life and movement, it is more important that few people
spend more minutes outdoors per day rather than the number of people. The public space needs
therefore to be designed to allow people to spend more time in the public space.
Gehl (2010) has a very local perspective, addressing the design of the street, while Hillier et al. (1993)
addressed the configurational perspective. Hillier creates conditions that Gehl must take care of
through local design. It is difficult the other way around. When the systematic conditions are not there,
the local design will not suddenly create a lot of movement, but making a very bad design can make
people choose another route despite the fact that it has a high centrality. Gehl (2010) notes that a
healthy city is defined when walking or biking is a natural part of the pattern of daily activities. People
are more likely to walk, cycle and spend time outdoors in places that are well-designed, with ample
shade, seating and other amenities that promote comfort and enjoyment. The author highlights also the
negative impacts of urban environments that are uncomfortable for pedestrians, such as heat stress. He
believes that in such places, people may be less likely to walk or spend time outdoors.
11
Previous studies have shown that urban environments affect the use of public spaces. Eliasson et al.
(2007) show that weather parameters affect people's perception, feeling but also the presence of a
public place in Nordic countries. Thorsson et al. (2006), on the other hand, describe that attitudes
towards the sun and time spent outdoors differ between Japan and Sweden. It is attributed to climatic
and cultural differences between the two countries where, in Sweden, the thermal conditions have a
stronger influence on the use of the public spaces than in Japan. More recent studies (Melnikov et al.,
2022) propose a new model of pedestrian behavior called "behavioral thermal regulation" which
explains how people regulate their body heat by making choices about their movement patterns in
response to thermal stimuli. This includes for example slowing down the walking speed or stopping to
rest, seeking after shade and adjusting the clothing. According to the research, pedestrians generally
prefer streets that offer more shade and lower temperatures, i.e paths with more treecover or buildings
that block the sun, but also avoid open spaces or paths with direct sunlight exposure. Melnikov et al.
(2020) proves in their study that higher walking speed results in higher levels of heat stress for the
pedestrians, due to increased metabolic heat production and decreased convective cooling. Both
Melnikov et al (2022) and Melnikov et al. (2020) point out the importance of urban planning and
designing. It is the key to create a safe, comfortable and sustainable city life that helps people stay cool
and safe in the urban environments, in this way it can reduce the impact of heat stress on pedestrians.
12
3. Study Area
On the Swedish west coast, the city of Gothenburg (57.70° N, 11.94° E) (see Figure 7) is the country's
second largest city with 599 405 inhabitants. Gothenburg is a fast-growing city and until 2040 the
population is estimated to increase to 707 189 inhabitants (Göteborgs Stad, 2023). With Gothenburg's
coastal location, the climate has mild summers with Ta averaging 16.3 °C during the months of June
to August. Between December and February, the winters are also relatively mild with Ta averaging
-o.4 °C. Although Ta will increase in the future, however, Tmrt will probably not increase more in
Gothenburg than in other cities in Europe (Thorsson et al., 2017).
Figure 7 - Map over the city of Gothenburg and the study area - one orthophoto over the study area and another map of the
buildings.
One of the environmental goals in Gothenburg focuses on the citizens, where Gothenburgers shall
have a healthy living environment. It is described that "Gothenburg must be a green and robust city
where ecosystem services are used to meet people's needs, now and in the future." (Göteborgs Stad,
2021:24. Own translation.). The environmental target is followed up by indicators, one of which is the
accessibility of "cool islands". It is described in the city's environmental and climate program that
13
within the city climate, in particular, more development is currently required and until 2030 the goal is
that an annual development is required.
The geographical delimitation for the study is part of Gothenburg's inner city. Five study sites in
Gothenburg were first chosen to be studied in detail based on density measurements; Guldheden,
Haga, Krokslätt, Stigberget and Änggården (see Figure 8).
Figure 8 - Map over the study area and the different urban typologies
Guldheden is predominantly residential and features a mix of low-rise apartment buildings and
single-family homes. However, Guldheden 15:1 and 15:2 are properties that represent grouped
townhouses that are located on a mountain in central Gothenburg. Eight point houses that have an
average of eight floors and where the entrances to the buildings face the street. There are open
boundaries between semi-private and semi-public urban spaces (Göteborgs Stad, 2008 - a) but a lot of
vegetation surrounds the residential area.
14
Haga is a historic neighborhood located in the central part of Gothenburg. In the traditional
neighborhood town of Haga, there is a high density with little access to open space, despite many
interspersed small businesses that attract numerous visitors every day. One could also call the area a
mixed city as there is a mix of services, businesses, trade, culture and housing (Göteborgs Stad, 2008 -
b). Traditional neighborhood city includes buildings on private neighborhood land, public spaces
(streets) as well as parks and squares (Göteborgs Stad, 2008 - a).
Krokslätt is a neighborhood located in the southeastern part of Gothenburg with a predominantly
low-rise building stock. The area features a grid-like street pattern with entrances facing the street or
garden. There are individual gardens with clear boundaries and the street has a clear public space, thus
there are few semi-private and semi-public urban spaces (Göteborgs Stad, 2008 - a).
Stigberget is a more refined residential area built between the 60s and 70s. Where the residential
buildings are V-shaped with south-facing yards to adapt to the mountainous terrain. The urban
structure is more open where there is less better access to playgrounds by the yards (Göteborgs Stad,
2008 - b). The planning idea at the time was to focus on sun, play and parking needs, which reflects
this residential neighborhood. It is the country's largest condominium association with twenty
buildings that contain six-story slab houses and accommodate 1 400 apartments (Lindgren et al.,
2017).
Änggården is a neighborhood located in the northeastern part of Gothenburg. It could be considered as
a suburban area as it is located on the outskirts of the city and therefore has a more residential
character. It is a residential area surrounded by nature, including the Botanical Garden. The
neighborhood is known for its historic architecture with mixed buildings. The area consists of
semi-detached houses, terraced houses and villas as well as a small group of apartment buildings
which are distributed in a partially symmetrical street grid where the streets are slightly angled
(Göteborgs Stad & Stadsbyggnadskontoret, 1999).
15
4. Method and Material
4.1 Research Design
It is possible to summarize the work done in
QGIS shortly, Figure 9 is read from the bottom up
as there are data layers that are processed based
on each other. The first step represents the study
areas, the whole study area but also the five
different urban typologies. The second and third
steps were to study the average Tmrt above 55 °C
(𝑇𝑚𝑟𝑡 > 55) within the study area during a
whole day, both with vegetation and without
vegetation, which was done with SOLWEIG
(SOlar and LongWave Environmental Irradiance
Geometry model) Analyzer (see section 4.2.2
more detailed information). The next step was to
study the potential pedestrian movement pattern
within the study area. In order to get as realistic a
result as possible, two analyses have been made in
Place Syntax Tool (PST). Attraction Reach which
represents the flow of pedestrian movement based
on population density (step 4) and Angular
Betweenness which is a good indicator to show
the flow of pedestrian movement based on
centrality (step 5). In order to only study the heat
stress on where people walk, the layer with
𝑇𝑚𝑟𝑡 > 55 data (with vegetation) was cut out
from the road segment with a buffer of 13 meters
(step 6). The last step (step 7) represents part of
the result where the walkways that have the
highest potential pedestrian movement together
with the highest and lowest 𝑇𝑚𝑟𝑡 > 55 (the
result is represented in Section 5.).
16
Tabel 1 - Used data and their origin
Data Source Description of data
SOLWEIG Digital elevation model Lantmäteriet (n.d) Lidar-data from 2018 processed by
(DEM) (2 x 2 m) Lujic (2023). Ground elevation in
meters above sea level (m.a.s.l.).
Digital surface model Lantmäteriet (n.d) Lidar-data from 2018 processed by
(DSM) (2 x 2 m) Lujic (2023). Ground and building
elevation (m.a.s.l.).
Canopy digital surface Lantmäteriet (n.d) Lidar-data from 2018 processed by
model (CDSM) (2 x 2 m) Lujic (2023). Canopy heights
(m.a.s.l.).
Landcover (2 x 2 m) Lantmäteriet (n.d) Data from 2018. Containing 7
classes: paved, buildings, evergreen
trees, deciduous trees, grass, bare soil
and water.
Meteorological data Shiny weather data (2021) Warm sunny day in Gothenburg,
01-06-2018, between 01:00 - 23:00.
PST Network segments Spatial Morphology Group & Line layer representing street
Chalmers (n.d) network.
Unlinks Spatial Morphology Group & Point layer representing bridges and
Chalmers (n.d) tunnels.
Buildings Baselayer Spatial Morphology Group & Building data contains building
Chalmers (n.d) height in meters, building height in
number of storey, ground floor area,
gross floor area, number of address
points per building, gross floor area
per address point.
Urban typology Wing (2019) Polygon layer representing four urban
typologies and building types in
Gothenburg (Mixed City, Million
Program, Nordic Functionalism and
Traditional Neighborhood city).
Base areas Göteborgs stad (2023) Polygon layer that divides
Gothenburg into several base areas
17
4.2 Quantitative Methods
4.2.1 Defining urban typologies
In order to choose as varied urban typologies as possible to fit into Spacematrix, a data layer has first
been used that represents four urban typologies and building types in Gothenburg - Mixed City,
Million Program, Nordic Functionalism and Traditional Neighborhood city (see Table 1). Based on
Spacematrix, however, there are even more different types of building geometries (see Figure 5). To
find as varied urban typologies as possible, some have therefore been handpicked. In order to place
the different urban typologies based on Spacematrix, various calculations have been made which have
been based on the explanation in figure 4. Area was calculated in QGIS for each urban typology, floor
area and footprints for each building are in the Buildings Baselayer layer (see Table 1) but were
summed based on each urban typology in excel.
4.2.2 Mapping of Mean Radiation Temperature (Tmrt)
Urban Multi-scale Environmental Predictor (UMEP) is a plugin for the software QGIS and is created
by a team of researchers at the University of Gothenburg. It allows analyzing how urban design
affects local weather conditions. The plugin is divided into three parts - pre-processor (prepares data),
processor (models data) and post-processor (analyzes the result) (Lindberg et al., 2018). To know
more about UMEP, read more on the website UMEP (n.d). To be in line with the purpose of the work,
the processor SOLWEIG and post-processor SOLWEIG Analyzer have been used in order to model
and analyze the distribution of Tmrt within the study area during a warm and clear day.
SOLWEIG is a 2.5-dimensional model and takes into account the complex interactions between Tmrt
and the urban environment. SOLWEIG is capable of simulating the spatial distribution of radiation
fluxes and Tmrt in urban environments using three digital surface models (urban morphology and
buildings, vegetation and terrain) and meteorological data (includes a time serie with Air Temperature
(°C), Relative Humidity (%) and incoming shortwave radiation (W m2)) (Lindberg et al., 2018). The
model takes into account how much solar radiation hits people walking at ground level. Tmrt is thus
calculated based on SOLWEIG where solar radiation is studied and calculated by how they are
reflected from the sky. The sky has been divided into 153 patches arranged in eight annuli where each
annulus is divided into patches that have similar size. The model is also based on formulas that
calculate how much radiation hits vertical and horizontal surfaces. To get a better understanding of the
calculation of Tmrt within SOLWEIG, read further in the article written by Wallenberg et al. (2023).
In order to model Tmrt with SOLWEIG, wall heights and aspects for buildings needed to be calculated
for each pixel of the input data, as well as the SVF. The processes have been automated in VSCode,
18
scripted in Python where each input dataset was clipped and calculated based on each square of a grid.
At the end of the script, the post-processor SOLWEIG Analyzer was used to get average in a day how
many hours Tmrt is above 55 °C (𝑇𝑚𝑟𝑡 > 55). This means that each pixel has a value that represents
how many hours in a day that 𝑇𝑚𝑟𝑡 > 55. In this thesis, low Tmrt means that during less than around
2.5 hours of a day, 𝑇𝑚𝑟𝑡 > 55. High Tmrt means that 𝑇𝑚𝑟𝑡 > 55 for more than 5 hours in a day. The
degree is based on an article written by Thorsson et al. (2014) where it is described that from 55.5°C
and above, there is a higher risk of death. In the schematic image below (see Figure 10), the type of
data and processors used within SOLWEIG and SOLWEIG Analyzer is represented and the settings
used in the SOLWEIG model are shown in table 2. Table 1 describes in more detail about all the input
data used. The meteorological data was reserved from Shiny weather data (2021), a warm and sunny
day (1/06/2018 (SMHI, 2018)) (see Appendix 1). It is, however, necessary to point out that the data
sources from Shiny weather data (2021) is the ERA5 reanalysis dataset for the global climate and
weather and is estimated hourly. After calculating 𝑇𝑚𝑟𝑡 > 55 with the data as just described, it was
realized that it is difficult to compare only the variables from the Spacematrix (FSI, GSI, OSR and L)
with the amount of high Tmrt in the urban typologies. It gave a realistic result of what the heat stress
looks like in these areas, but these are not values  that can easily be compared with urban morphology
as Tmrt varies a lot depending on the vegetation in the area. It was therefore chosen to do the same
calculations again in the python script, with the same parameters but without the CDSM layer.
Figure 10 - Flowchart for SOLWEIG Analyzer used in the study. Adapted and modified from Lindberg et al. (2018). Gray
boxes mark geodata and white boxes indicate other types of data. Dotted boxes are processors within UMEP.
19
Table 2: Settings used in SOLWEIG model
Parameter Value
Environmental Albedo building 0.2
Parameters walls
Albedo ground 0.15
Emissivity building walls 0.9
Emissivity ground 0.95
Radiation transmissivity through 0.03
vegetation
Human Parameters Body longwave 0.95
absorption
Body shortwave absorption 0.7
Body as cylinder Yes
Posture Standing
4.2.3 Mapping of pedestrian movement
PST is an open-source plugin tool in GIS software QGIS. The tool has been developed by the research
group, The Spatial Morphology Group (SMoG), at Chalmers University, the Royal Institute of
Technology (KTH) and consulting firm Spacescape AB. The difference between Space Syntax and
PST is primarily that Space Syntax can describe how people move through a network, but PST
describes how people move to and from different destinations in the network (Ståhle et al., 2005). PST
has eight different functionalities that can be used to perform various types of flow data analysis:
Reach, Network Integration, Angular Integration, Network Betweenness, Angular Choice, Attraction
Distance, Attraction Reach and Attraction Betweenness. To get a deeper insight into all the analyses
that can be used in PST, see SMoG's website (Spatial Morphology Group & Chalmers, n.d).
In Table 1, a more detailed description is provided for the data layers used. To use PST, at least two
data layers are required: one for the network and one for the unlinks. In this study, the network refers
to the road network along the pedestrian streets in the city. The network is represented by segment
maps where intersections are considered edges and lines are considered nodes. This means that the
intersections are points where two or more lines meet (see Figure 11 A.) (Berghauser Pont et al.
2021). The point layer Unlinks represents the level differences of the intersecting lines in the network,
such as bridges and tunnels. To identify the closest intersection, every Unlink is compared with all
intersection points within the line network. The closest intersection is then removed from the graph. It
20
should be noted that for the Segment map, the lines should cross at the points of the Unlinks rather
than meet to ensure effective disconnection (see Figure 11 B.) (ibid).
Figure 11 - Figures on the left (A.) represents graph notation of Segment map whereas on the right (B.) it is representations
of ‘Unlinks’ in Segment map (Berghauser Pont et al. 2021)
In this study, the focus will be on the analyses of Attraction Reach and Angular Betweenness. In order
to get a better understanding of the calculations of the analysis, read further in the article written by
Berghauser Pont et al. (2021). Attraction Reach measures the accessibility in a network. It calculates
the shortest path between a given node and all other nodes in the network and uses a cost function to
weight the importance of different nodes in the network. Within the different analyses in PST, there
are different ways of measuring distance in its analysis. In this study, Attraction Reach is used to study
how much floor space one can reach within a radius from the location. This correlates highly with the
amount of people in the area but instead of using population data a building layer with GFA variables
were used in order to get data of how many GFA one reaches from each street segment within a
walking distance of 500 meters. In QGIS, four input data are used: origin points, segment lines, unlink
points and destinations. In this case, street networks were used as a origin point and as segment lines,
bridge and tunnel data as unlinks and GFA variables in a building layer as a destination (data objects).
Angular Betweenness calculates however the significance of nodes in a network according to their
position and connections. In this study, the analysis is used to calculate the shortest path within a 2 km
walking distance. Shortest in this context means reaching the destination quickly with the least
angular change, reflecting how pedestrians navigate and flow in a city. In QGIS, only network
segment lines as segment lines, no normalization mode, and no weight mode were used.
4.2.4 Mapping of pedestrian movement and heat stress
In order to visualize the relation between the intensity of pedestrian flow and level of heat stress, a
scatterplot is used with Y-axis representing the potential pedestrian movement (Attraction Reach or
Angular Betweenness) and the X-axis indicator of heat stress (𝑇𝑚𝑟𝑡 > 55). This was done in excel
and the results are represented in appendices 2 and 3. In order to map and study further where in the
21
study area there is the highest potential pedestrian movement with high or low 𝑇𝑚𝑟𝑡 > 55, it was
chosen to extract and visualize the pixels within the green and gray boxes from the graphs in
appendices 2 and 3. The pixels in the green boxes represent the walkways that have high potential
pedestrian movement and low 𝑇𝑚𝑟𝑡 > 55, the pixels in the gray boxes show the streets that have
high potential pedestrian movement and the highest 𝑇𝑚𝑟𝑡 > 55 (see Figure 12). It was not
considered interesting to study the areas that have low potential pedestrian movement (pink and blue
box), due to the fact that it is not interesting based on the purpose of the thesis.
Figure 12 - Description of what the content of the graphs in appendix 2 and 3 means, which is also the explanation for the
mapping of potential pedestrian movement and heat stress.
22
5. Results
5.1 Spacematrix and 𝑇𝑚𝑟𝑡 > 55 within Different Urban Typologies
As can be seen, the areas of Guldheden, Haga, Krokslätt, Stigberget and Änggården are differently
positioned in the Spacematrix (see Figure 13). This is due to the fact that they have different values of
FSI, GSI, OSR and L, which indicate different urban morphologies and typologies. These can be
further studied in Table 3, where 𝑇𝑚𝑟𝑡 > 55 also has been calculated for each area, both with and
without vegetation.
Figure 13 - Spatial configuration of different built environments spaced in the Spacematrix (created from Berghauser Pont &
Haupt, 2010)
It's worth mentioning the common denominators when it comes to the maps representing 𝑇𝑚𝑟𝑡 > 55
with and without vegetation. The maps where 𝑇𝑚𝑟𝑡 > 55 with vegetation has been studied (Figure
X, maps B.), the areas that have been assigned a value of low 𝑇𝑚𝑟𝑡 > 55 primarily have a lot of
vegetation or are shaded by buildings. Compared to the maps where 𝑇𝑚𝑟𝑡 > 55 without vegetation
has been studied (Figure X, maps C.), the areas where 𝑇𝑚𝑟𝑡 > 55 is low are areas where shade is
created by only buildings. A common denominator between both maps B and C is, however, that it is
the same areas that have high 𝑇𝑚𝑟𝑡 > 55. Only that 𝑇𝑚𝑟𝑡 > 55 is even higher in the map without
vegetation (B.), which is interesting to study because of the lack of vegetation in the PST theory.
23
Table 3 - Results calculated based on the five study areas. Including the variables from Spacematrix, average Tmrt (𝑇𝑚𝑟𝑡)
over 55 °C with and without vegetation (hours), as well as their urban typologies.
FSI GSI OSR L Urban Typology 𝑇𝑚𝑟𝑡 > 55 (with 𝑇𝑚𝑟𝑡 > 55 (without
vegetation)1 vegetation)2
Guldheden 0.74 0.11 1.20 6.5 Mid-rise closed 1h53 5h51
buildings
Krokslätt 0.24 0.13 3.67 1.8 Low-rise spacious strip 3h18 5h49
development blocks
Stigberget 1.49 0.25 0.5 5.9 Mid-rise compact 2h31 5h02
buildings
Haga 1.77 0.48 0.29 3.7 Mid-rise spacious 2h57 3h51
building blocks
Änggården 0.58 0.25 1.29 2.4 Low-rise compact strip 3h18 5h44
development blocks
5.1.1 Guldheden
In the case of Guldheden, the FSI (0.74) indicates that it is a relatively dense and compact
environment. The GSI (0.11) is low which means that there is little open space in the built
environment. The OSR (1.20) in this area is relatively low compared to the FSI, which indicates that
the built environment is relatively built-up. This leads on to the L value (6.5) which is high in this
case, which suggests that the buildings are fragmented and is categorized as mid-rise closed buildings.
The result in table 3 also shows that for about 2 hours in a day, the 𝑇𝑚𝑟𝑡 > 55, and without
vegetation it is almost 6 hours in a day (see Figure 14).
1 Average number of hours that the average Tmrt is above 55 °C with vegetation within the urban area. For
example in Guldheden, the average Tmrt is above 55 °C for 1.89 hours in a day.
2 Average number of hours that the average Tmrt is above 55 °C without vegetation within the urban area. For
example in Guldheden, the average Tmrt is above 55 °C for 5.85 hours in a day.
24
Figure 14 - Map A. represents study area Guldheden, while map B. shows 𝑇𝑚𝑟𝑡 > 55, depicted in shades of blue to red,
with darker red indicating higher temperatures. Map C. also represents 𝑇𝑚𝑟𝑡 > 55, but without vegetation.
5.1.2 Krokslätt
Krokslätt has a relatively low FSI (0.24), indicating that the built environment is not very dense and
has a low building coverage. The GSI (0.13) suggests that there is more open space than built-up
space in this environment. The OSR (3.67) value indicates that the environment has a lot of open
space relative to its total area. The low L (1.8) value indicates that the built environment has a lot of
low buildings, which is categorized as low-rise spacious strip development blocks. Within Krokslätt,
during a little bit more than 3 hours of the day, the area has an 𝑇𝑚𝑟𝑡 > 55. But if there weren’t any
vegetation, which lowers the temperature, 𝑇𝑚𝑟𝑡 > 55 during almost 6 hours of the day (see Figure
15).
Figure 15 - Map A. represents study area Krokslätt, while map B. shows 𝑇𝑚𝑟𝑡 > 55, depicted in shades of blue to red,
with darker red indicating higher temperatures. Map C. also represents 𝑇𝑚𝑟𝑡 > 55, but without vegetation.
5.1.3 Stigberget
The value of FSI (1.49) in Stigberget indicates that the built environment is very dense and has a lot of
built-up space in a relatively small area. The GSI (0.25) suggests that there is a relatively small
25
amount of open space, with buildings occupying most of the plot area. The OSR (0.5) indicates that
there is very little open space in the built environment. This is further supported by the high FSI and
the low GSI, which indicates that most of the area is used for buildings. The high L value (5.9)
suggests that the buildings are relatively high and based on the different values  it is possible to
categorize Stigberget as mid-rise compact buildings. In Stigberget, on average, 𝑇𝑚𝑟𝑡 > 55 for 2.5
hours, without vegetation, on the other hand, it is on average for 5 hours (see Figure 16).
Figure 16 - Map A. represents study area Stigberget, while map B. shows 𝑇𝑚𝑟𝑡 > 55, depicted in shades of blue to red,
with darker red indicating higher temperatures. Map C. also represents 𝑇𝑚𝑟𝑡 > 55, but without vegetation..
5.1.4 Haga
Haga has an FSI of 1.77, which indicates that the built environment is very dense and has a lot of
built-up space in a relatively small area. The GSI (0.48) suggests that there is some open space, but
still a large proportion of the area is occupied by buildings. The OSR of 0.29 indicates that there is
relatively little open space in the environment, which is supported by the high FSI, it indicates that the
most of the area is used for buildings. The L value of 3.7 suggests that the buildings are relatively low
and are categorized as mid-rise spacious building blocks. Haga has a 𝑇𝑚𝑟𝑡 > 55 during almost 3
hours of a day and without vegetation the 𝑇𝑚𝑟𝑡 > 55 during almost 4 hours of a day (see Figure 17).
Figure 17 - Map A. represents study area Haga, while map B. shows 𝑇𝑚𝑟𝑡 > 55, depicted in shades of blue to red, with
darker red indicating higher temperatures. Map C. also represents 𝑇𝑚𝑟𝑡 > 55, but without vegetation.
26
5.1.5 Änggården
Änggården has an FSI of 0.58, indicating that the built environment is relatively dense and compact,
with a moderate amount of built-up space. The GSI (0.25) suggests that there is a good amount of
open space, with a relatively small footprint of buildings in relation to the total plot area. The OSR
(1.29) indicates that there is a reasonable amount of open space in the built environment, which is
supported by the relatively low FSI and the moderate GSI. The L value (2.4) suggests that the
buildings are low and based on the various variables, it is possible to categorize the area as low-rise
compact strip development blocks. In Änggården, for a little more than 3 hours in a day the
𝑇𝑚𝑟𝑡 > 55. On the other hand, in the same built environment but without vegetation, the
𝑇𝑚𝑟𝑡 > 55 during almost 6 hours in a day (see Figure 18).
Figure 18 - Map A. represents study area Änggården, while map B. shows 𝑇𝑚𝑟𝑡 > 55, depicted in shades of blue to red,
with darker red indicating higher temperatures. Map C. also represents 𝑇𝑚𝑟𝑡 > 55, but without vegetation..
5.2 Pedestrian Movement and 𝑇𝑚𝑟𝑡 > 55 within the Whole Study Area
The results from the Attraction Reach and Angular Betweenness analysis from PST are represented in
the maps below (see Figure 19). The brighter the street network is, the more people are potentially
walking on those streets, and vice versa. By studying these two maps, it is clear that pedestrians move
locally with different purposes for navigating the city. Attraction Reach is calculated based on how
many GFA one reaches from each street segment within a walking distance of 500 meters. The result
in Figure 19 A. thus shows that most potential pedestrian movement is in the central and dense parts
of the study area, i.e. Haga, Olivedal, Vasastaden and Landala. However, when Angular Betweenness
shows a high value, it means that a line connects areas with many streets but also that there are no
alternative streets that are of approximately the same length, with length it means the smallest angle
change. Based on Figure 19 B, one can see that it is mainly the longer and wider streets that have the
most pedestrians (i.e. Aschebergsgatan, the turn at Sahlgrenska on Per Dubbsgatan and also the turn in
front of Chalmers on Guldhedsgatan).
27
Figure 19 - Map A. representing Attraction Reach, different flow of potential pedestrian movement based on population
density. Map B. representing Angular Betweenness. Different flow of potential pedestrian movement based on centrality.
There aren’t any strong statistically significant correlation between either centrality (using Angular
Betweenness analysis) and 𝑇𝑚𝑟𝑡 > 55 nor density (using Attraction Reach analysis) and
𝑇𝑚𝑟𝑡 > 55. This was then confirmed because before calculations of correlation were made, no
correlation was expected between centrality and 𝑇𝑚𝑟𝑡 > 55. The graph in appendix 2 indicates that
5% of the variance in Angular Betweenness could be explained by the 𝑇𝑚𝑟𝑡 > 55. The graph in
appendix 3 indicates almost the same but only 2% of the variance in Attraction Reach could be
explained by the 𝑇𝑚𝑟𝑡 > 55. It is therefore the patterns of centrality, density and 𝑇𝑚𝑟𝑡 > 55 that
are interesting to study because they show public spaces/streets where many people are exposed to
heat stress and the opposite (represented in gray and green boxes in the graphs in appendices 2 and 3).
In the following maps, it is possible to identify streets with high potential pedestrian movement (both
with Attraction reach and Angular betweenness but also separately) with varying levels of
𝑇𝑚𝑟𝑡 > 55. In appendix 4, the maps are represented separately and not merged as in the following
maps below.
28
5.2.1 High Attraction Reach with high and low 𝑇𝑚𝑟𝑡 > 55
The streets that have high Attraction Reach and high 𝑇𝑚𝑟𝑡 > 55 are several shorter road segments
that are directed north-south. The smaller parts of the streets are, for example, pedestrian crossings at
intersections. The larger areas where there is a high 𝑇𝑚𝑟𝑡 > 55 and high Attraction Reach are, for
example, Götaplatsen, larger streets near Vasaplatsen, streets around Kapellplatsen and Landala Torg,
but also Övre Husargatan. However, the streets that have high Attraction Reach and low 𝑇𝑚𝑟𝑡 > 55
imply that there is a lot of potential pedestrian movement in the form of population density on the cool
streets. These streets are mainly smaller cross streets that are placed in a west-east direction in Haga,
Vasastaden and Linné, but also pedestrian paths in parks such as Vasaparken and in front of
Hagakyrkan (see Figure 20).
Figure 20 - Map of the study area with highlighted examples of streets with high Attraction Reach with high and low
𝑇𝑚𝑟𝑡 > 55.
29
5.2.2 High Angular Betweenness with high and low 𝑇𝑚𝑟𝑡 > 55
The streets that have high Angular Betweenness and high 𝑇𝑚𝑟𝑡 > 55 are primarily the larger and
wider streets such as Övre Husargatan and Gibraltargatan. Likewise more open spaces such as bus
stops - Wavrinskys Plats (part of Guldhedsgatan), Kapellplatsen (Aschebergsgatan) and Per
Dubbsgatan (Sahlgrenska Huvudentrén). The streets that have high Angular Betweenness and low
𝑇𝑚𝑟𝑡 > 55 imply that these are cooler streets that have high potential pedestrian movement in the
form of pedestrian flows through the city. These streets are few and are not very long, e.g a little part
of Södra Viktoriagatan and a small street in Änggården. In addition to a pathway through Slottsskogen
and the cross street Nordenskiöldsgatan (see Figure 21).
Figure 21 - Map of the study area with highlighted examples of streets with high Angular Betweenness with high and low
𝑇𝑚𝑟𝑡 > 55.
30
5.2.3 High Attraction Reach and Angular Betweenness with high and low
𝑇𝑚𝑟𝑡 > 55
After studying high Attraction Reach and Angular Betweenness separately, it is also interesting to
study them together, which shows more clearly and reliably where the highest potential pedestrian
movement is. The streets with the highest potential pedestrian movement and highest 𝑇𝑚𝑟𝑡 > 55 are
mainly open spaces such as the streets passing around the bus stop, Kapellplatsen, and the main
square Götaplatsen. As well as intersections at Aschebergsgatan and the crossing of Övre Husargatan
and Nordenskiöldsgatan. The streets with the lowest 𝑇𝑚𝑟𝑡 > 55 and a lot of pedestrian movement
are mainly Nordenskiöldsgatan and a part of Södra Viktoriagatan, but also the beginning of
Gibraltargatan and a small part at the bus stop of Vasaplatsen (see Figure 22).
Figure 22 - Map of the study area with highlighted examples of streets with high Attraction Reach and high Angular
Betweenness with high and low𝑇𝑚𝑟𝑡 > 55.
31
6. Discussion
6.1 Urban Typologies and Heat Stress - Insights from Spacematrix Method
The Spacematrix method describes how to measure urban density, which has a correlation with urban
typologies (Berghauser Pont & Haupt, 2010). The calculations of FSI, GSI, OSR and L describe the
relationship between building density, light conditions and usability (Berghauser Pont & Marcus,
2020). This goes hand in hand with how people experience the urban climate, which depends
primarily on the microclimate generated by building structures in the urban environment (Wallenberg
et al., 2018). This means that the orientation (Shishegar, 2013; Wallenberg et al., 2018), height and
distance of the buildings (Shishegar, 2013), as well as their surface and material properties, are crucial
for how much incoming solar radiation reaches the ground during the day and how the radiation is
reflected or absorbed (Wallenberg et al., 2018; Oke et al., 2017). With that said, there are clear
connections between different urban typologies and heat stress. By combining the results in Table 3
with previous research (Berghauser Pont et al., 2019; Berghauser Pont & Haupt, 2010), the five
different study areas can be identified, each with their respective urban typologies.
By studying the variables in Table 3 in more detail, several patterns can be observed between the
parameters in the Spacematrix and 𝑇𝑚𝑟𝑡 > 55. Guldheden, Krokslätt and Änggården have almost
the exact same amount of hours of 𝑇𝑚𝑟𝑡 > 55 without vegetation, indicating that approximately 6
hours of a full day. However, when studying 𝑇𝑚𝑟𝑡 > 55 with vegetation, the values are different.
The 𝑇𝑚𝑟𝑡 > 55 with vegetation in Krokslätt and Änggården is the same (little bit more than 3 h), but
in Guldheden, the 𝑇𝑚𝑟𝑡 > 55 is considerably lower (almost 2h). At the same time, L is high in
Guldheden (6.5) compared to Krokslätt (1.8) and Änggården (2.4). This shows that Guldheden has
both taller buildings that create shading and more vegetation compared to the other two urban
typologies, which lowers 𝑇𝑚𝑟𝑡 > 55 (Lindberg et al, 2016).
Haga has the smallest difference in the number of hours of a day where 𝑇𝑚𝑟𝑡 > 55 with and without
vegetation (without vegetation, 𝑇𝑚𝑟𝑡 > 55 for an hour more), which is also evident in Figure 17.
This is explained by the fact that the studied area has limited vegetation that can lower Tmrt, but rather
it is the shade from the buildings that primarily contributes to the reduction in Tmrt. This is reflected in
the high GSI (0.48) and L (3.7), indicating that a larger portion of the ground surface is covered by
relatively high buildings, resulting in a more compact urban development which creates shadows on
the streets.
32
6.2 Heat Stress and Pedestrian Movement Patterns
Previous research shows a strong relation between pedestrian movement and climate (e.g. Gehl, 2010;
Jacobs, 2005), which is not confirmed by the results of this study in Gothenburg. One reason could be
that the pedestrian movement analysis is giving an indication of the potential for movement based on
the streets centrality and built density and is not giving measured flows. It could thus be that streets
with a high potential for pedestrian movement have a lower actual flow if the urban climate is not
comfortable, and vice versa. However, by studying and mapping the streets from the gray and green
boxes in the graphs, a clear spatial pattern can be seen, which is also confirmed by Shishegar (2013)
and Wallenberg et al. (2018). It is mainly that the cold streets are mostly smaller cross streets in a
west-east direction and warmer streets are mostly north-south directed streets (see Figures 20, 21 and
22). In the upcoming subsection, this will be further discussed in relation to pedestrian movement
patterns. To be able to see clearer patterns in the form of quantitative results, box plots were made
where values of 𝑇𝑚𝑟𝑡 > 55 were studied based on different directions that the streets are facing,
such as streets that are facing north-south or east-west (not shown). The values from each studied
walkway from different directions show hardly any difference, which may indicate that density
mainly controls the pattern more than the direction of the street. It is also debatable that the measure
of 𝑇𝑚𝑟𝑡 > 55 used in this study actually shows the time when a pixel is above 55 °C. The actual
value of Tmrt is therefore not used, which may slightly affect the result, but probably not significantly.
6.2.1 The urban geometry: Key factors in pedestrian movement & heat stress
As mentioned, it is possible to see coexistence and patterns between the results from figures 20 to 22.
The results from the figures show a clear pattern where the coldest streets with high potential
pedestrian movement are west-east direction streets. This can be explained by the fact that the results
in the figures describe that 𝑇𝑚𝑟𝑡 > 55 for a maximum of 2.5 hours during a whole day. Previous
research (Shishegar, 2013; Wallenberg et al., 2018) shows that these west-east direction streets receive
the most incoming sunlight during mornings and afternoons, when the sun is not as hot and only for a
very short period. Similarly, with the results presented in the figures, the streets with high potential
pedestrian movement and 𝑇𝑚𝑟𝑡 > 55 for more than 5 hours per day are primarily large and wide
north-south direction streets. Previous research (Shishegar, 2013; Wallenberg et al., 2018) also shows
that north-south direction streets generally have few hours of sunlight throughout the day, but they are
fully lit in the middle of the day when the sun is at its highest. The selection of streets reflects
previous research (Shishegar, 2013; Wallenberg et al., 2018) as the sun is at its highest in the sky,
making it the strongest during the middle of the day.
33
The maps in figures 20. 21 and 22 also reveal strong patterns based on density. One notable pattern is
observed in the dense residential areas like Haga, Vasastaden and Linné, where streets with the
highest Attraction Reach and lowest 𝑇𝑚𝑟𝑡 > 55 are located. This can be attributed to the dense
placement of buildings, resulting in narrower roads and public spaces covered by shade, which
ultimately reduces Tmrt (Wallenberg et al., 2018). As the results suggest, constructing buildings in a
compact manner can decrease Tmrt due to the shadow cast by buildings. However, it is not ideal from a
theoretical standpoint. This is primarily due to the fact that a high degree of compact urban
development, characterized by buildings closely packed together with dark facades and materials that
have low albedo and high emissivity, absorbs a lot of incoming sunlight, leading to an increase in both
Ts and Ta (Wallenberg et al., 2018). Furthermore, this can contribute to the UHI effect (ibid; Oke et
al., 2017).
In order to reduce UHI, it is not enough to study density and heat, vegetation is also a central factor
(Lindberg et al., 2016; Yu et al., 2020), which PST isn’t created to analyze. Based on the results of low
𝑇𝑚𝑟𝑡 > 55 (see Figures 20, 21 and 22) it is also evident that pathways, such as Vasaparken or
Slottsskogen, are cooler due to the presence of vegetation (e.g grass and trees), which lower
𝑇𝑚𝑟𝑡 > 55 through evapotranspiration and shading (Moss et al., 2019). These areas are dominated
by deciduous trees, which are preferable, especially during summer, as they create denser shade
(Lindberg et al., 2016). Based on previous research by Melnikov et al. (2022), pedestrians generally
prefer shaded paths with lower temperatures and tend to avoid open spaces or paths with direct
sunlight exposure. Pedestrians prefer shading from buildings rather than trees because it provides a
denser and cooler shade. Based on this fact, it also suggests that denser urban development with
buildings may be more effective in creating shade compared to vegetation. However, as mentioned
earlier, this can exacerbate the UHI effect, leading to consequences such as increased heat stress
(Wallenberg et al., 2018; Oke et al., 2017).
6.2.2 Planning and designing streets for pedestrian comfort
The planning and development of public spaces in relation to heat stress varies depending on the
purpose and usage of the streets. According to Gehl (2010), public spaces should be designed in a way
that encourages more people to spend longer periods of time outside on the streets. It is better to have
fewer people outside but staying for a longer time than the other way around, where many people
simply pass through without stopping. Eliasson et al. (2007), Melnikov et al. (2020), Melnikov et al.
(2022) and Thorsson et al. (2007) also emphasize the importance of urban planning and design in
creating safe, comfortable and sustainable urban environments that reduce the impact of heat stress on
pedestrians. This is interesting to discuss in relation to the results in section 5.2. The streets frequently
used by pedestrians who live nearby need to provide a high level of comfort and pleasantness, as
34
many people pass through them (represented from Attraction Reach). This type of comfort differs
from what is required on streets with heavy pedestrian traffic due to city-wide traffic patterns
(represented from Angular Betweenness).
Based on figure 21, some of the streets that require high comfort levels are Övre Husargatan,
Gibraltargatan, parts of Guldhedsgatan and Per Dubbsgatan, as well as larger hubs such as
Kapellplatsen, which has a high 𝑇𝑚𝑟𝑡 > 55. These streets are usually wide and long and often have
public transportation, so these public spaces need to be planned and developed in a way that lowers
the Tmrt and increases comfort. This is especially important at bus stops such as those at Kapellplatsen,
Chalmers in Guldheden, Wavrinskys plats and Sahlgrenska Huvudentré on Per Dubbsgatan. However,
based on the results in figure 20, there is a lot of potential pedestrian movement and high 𝑇𝑚𝑟𝑡 > 55
on the broad and open public spaces in densely populated areas, such as at intersections and
pedestrian crossings. Since the result is based on calculations of Attraction Reach, it means that the
pedestrian movement is due to many people living within 500 meters. Citizens spend short periods of
time, such as a minute, at intersections and crossings. However, the result also shows that there is high
potential pedestrian movement and high 𝑇𝑚𝑟𝑡 > 55 on other larger and more open public spaces,
such as squares. Götaplatsen, Vasaplatsen, Landala Torg and Kapellplatsen are larger public spaces
with high 𝑇𝑚𝑟𝑡 > 55 and high potential pedestrian movement. These places are central hubs in the
urban space and can therefore serve as meeting points for residents, necessitating an increase in
comfort to mitigate heat stress in these public spaces, as people tend to spend longer time at these
locations.
This goes hand in hand with what previous research highlights (Hillier et al., 1993; Berghauser Pont
& Marcus, 2020; Berghauser Pont et al., 2019). In broad strokes, density affects the amount of people
who move and stay in the public space, while centrality has a strong influence on where people move
and stay in the city's space. These together make it interesting to study the pedestrian movement in
public space. Hillier et al. (1993) additionally explains that based on pedestrians' movement patterns,
it is possible to identify central places that are important to humans. This implies that the comfort
level and urban design of walking paths may vary depending on their purpose and intent. However, it
is also important to note that these different factors can be defined differently from person to person,
as Hillier et al. (1993) also describes. This implies that individuals of different age groups may
perceive these factors differently. For example, different age groups may have varying abilities to
navigate in the city, such as elderly finding it more challenging to ascend a hill than a straight road,
which makes them choose an alternative route. Additionally, people may choose to settle in specific
areas based on proximity to various amenities or services. However, these factors have not been
incorporated in the calculations of this study, which has focused on a general movement pattern.
35
6.3 Limitations of the Study
During the course of the study, various deficiencies emerged with the Network segment data. Due to
the fact that pedestrian movement is studied, the data was not so detailed. The used data layer had the
lines in the middle of the road, but in reality, people do not walk in the middle of a road, but rather on
the sidewalks, either on the right or left side. The study area was too large and it would have taken too
long to modify the network segments. In this case, it was not considered to be a major problem as the
study was done in a general overview.
As mentioned, the metrological input data has been retrieved from the ERA5 reanalysis dataset for the
global climate and weather. It doesn't therefore represent spatial variations of air temperature and
relative humidity within the city, which may have affected the result. But after comparing the data
with metrology data collected from the university of Gothenburg (Göteborgs Universitet, 2022), not
much was different. There is also a disadvantage with the Landcover layer which is that it has too
much grass, which in turn may have given too low Tmrt values. It was chosen not to be taken into
account because the grass areas are mainly around the buildings, which are nevertheless cold areas for
studying Tmrt due to the shading of the buildings (Lindberg et al. 2016).
6.4 Implications of the Study, Recommendations and Further Research
There are several points that would have been interesting to study further in relation to this study. Tmrt
and pedestrian movement are in this case studied from a very broad and general perspective.
Therefore, it would have been interesting to further study how different target groups are affected by
heat stress on the walking paths, but also to include several parameters in PST such as slope and
different distances. This is mainly because everyone has different conditions for moving around in the
city. For example, a person sitting in a wheelchair does not have the same movement pattern as
another person who does not have a functional variation. They also do not have the same access to
various points of interest, therefore it would also have been interesting to add different comfort and
well-being indexes in the form of, for example, that there are certain shops, restaurants or attractions
that many like to go to. Right now, PST shows a general pedestrian flow, which does not reflect reality
as there are not the same amount of pedestrians out on the streets during the day as at night.
Furthermore, the development of PST would also have been interesting to see from a safety
perspective, for example which streets have a lot of pedestrian movement during the night and what
do these streets look like. In this way, urban planners can use the tool to point out where, for example,
street lighting could have been implemented.
Then it is also possible to spin further on more development opportunities with this study as a basis.
For example, it would have been interesting to carry out an observation to primarily see pedestrians'
36
choice to move on a hot summer day and the reason why they are outside walking. For example, do
they choose to walk on the shady part of the road and it goes hand in hand with the results in the GIS
analyses. This would also have been interesting to implement in other cities and also during different
times of the year. In addition, other factors in the urban climate such as air pollution and wind can also
be taken into account to get a more realistic and clearer understanding of thermal comfort.
37
7. Conclusions
Summarizing the differences between the five urban areas and the whole study area, it was clear that
the cold streets are mostly smaller cross streets in a west-east direction and warmer streets are mostly
large and wide north-south directed streets. The result also indicates that density (calculated with
Attraction Reach) is a variable that affects both heat stress and pedestrian movement, but centrality
(calculated with Angular Betweenness) does not do so as much. The density of an urban environment
reduces therefore Tmrt, which in turn reduces heat stress. However, by combining PST and SOLWEIG,
it also shows that density is not the only parameter that affects Tmrt, vegetation is also a central factor.
As demonstrated, depending on what kind of pedestrian movement that exists in different places in the
city, it is important to plan and design the right environments for the right purpose in order to increase
people's comfort in the public space. The results of the thesis show that within the study area it is
mainly broad and open public spaces that need to be improved in order not to suffer from heat stress,
such as bus stops and squares.
Finally, it is also important to point out that pedestrian movement and heat stress affect each other. If
planners want to change the design of the public space to reduce heat stress, it will also automatically
affect pedestrian movement. Since the design of the physical component of a place can affect the
place's microclimate and thus influence people's attendance, perceptions and emotions the result
supports the arguments for using GIS and implementing climate aspects in future planning and design
projects.
38
References
Ali-Toudert, F., & Mayer, H. (2007). Thermal comfort in an east-west oriented street canyon in Freiburg
(Germany) under hot summer conditions. Theoretical and Applied Climatology, 87(1–4), 223–237.
https://doi.org/10.1007/s00704-005-0194-4
Berghauser Pont, M. & Haupt, P. (2007). The relation between urban form and density. Urban Morphology. 11,
62-65. https://doi.org/10.51347/jum.v11i1.4495
Berghauser Pont, M. & Haupt, P. (2010). Spacematrix: Space, density and urban form. NAi.
Berghauser Pont, M. & Marcus, L. (2020). Teorier om stadsform för att mäta städer. Fusion Point Gothenburgs.
https://alvstranden.com/app/uploads/2020/12/FP02_Teorier_om_stadsform_for_att_mata_stader.pdf
[pdf] (Retrevied: 05/10/2023)
Berghauser Pont, M., Stavroulaki G., Fitger, M., Koch, D., Legeby, A., Marcus L. (2021) PST Documentation 3
(2.3), Chalmers University of Technology, KTH. https://doi.org/10.13140/RG.2.2.25718.55364
Berghauser Pont, M., Stavroulaki, G., Gil, J., Marcus, L., Olsson, J., Sun, K., . . . Legeby, A. (2019). The spatial
distribution and frequency of street, plot and building types across five European cities. Environment
and Planning. B, Urban Analytics and City Science, 46(7), 1226-1242.
https://doi.org/10.1177/2399808319857450
Bobkova, E., Marcus, L. & Berghauser Pont, M. (2017). Multivariable measures of plot systems: describing the
potential link between urban diversity and spatial form based on the spatial capacity concept.
Chen, L., Yu, B., Yang, F., & Mayer, H. (2016). Intra-urban differences of mean radiant temperature in different
urban settings in Shanghai and implications for heat stress under heat waves: A GIS-based approach.
Energy and Buildings, 130, 829–842. https://doi.org/10.1016/j.enbuild.2016.09.014
Eliasson, I. (2000). The use of climate knowledge in urban planning. Landscape and Urban Planning, 48(1– 2),
31–44. https://doi.org/10.1016/S0169-2046(00)00034-7
Eliasson, I., Knez, I., Westerberg, U., Thorsson, S., & Lindberg, F. (2007). Climate and Behaviour in a Nordic
city. Landscape and Urban Planning, 82(1–2), 72–84.
https://doi.org/10.1016/j.landurbplan.2007.01.020
Folkhälsomyndigheten (2015). Hälsoeffekter av höga temperaturer – En kunskapssammanställning.
https://www.folkhalsomyndigheten.se/contentassets/e39b425555f44a3ba05aa0dbaa956c43/halsoeffekte
r-hoga-temperaturer-15048-webb.pdf [pdf] (Retrieved 2023-03-28)
Gehl, J. (2010). Cities for people. Washington: Island Press.
Gibson, J.J., (1986). The ecological approach to visual perception, Psychology Press, New York, NY, USA.
Göteborgs Stad (2008 - a) Om stadens utformning. Stadsbyggnadskvaliteter Göteborg.
https://goteborg.se/wps/wcm/connect/f6c03c8f-10c6-41cd-85d7-bb72e2f8e50f/OPA_stadsbyggnkvalite
ter.pdf?MOD=AJPERES [PDF] (Received: 3/15/2023)
39
Göteborgs Stad (2008 - b) Linnéstaden - Beskrivning av stadsdelen.
https://goteborg.se/wps/wcm/connect/31156a4a-c7a1-41cb-895d-2c4dc2d6dbd1/OPALinnestaden.pdf?
MOD=AJPERES [PDF] (Received: 3/15/2023)
Göteborgs Stad (2021). Göteborgs Stads miljö- och klimatprogram 2021-2030.
https://goteborg.se/wps/wcm/connect/4578bcdd-0a21-4d90-98c5-8ec4e68b366b/G%C3%B6teborgs+St
ads+milj%C3%B6-+och+klimatprogram+2021-2030.pdf?MOD=AJPERES [PDF] (Received:
3/15/2023)
Göteborgs Stad (2023). Kartor som GIS-skikt (stadens områdesindelning).
https://goteborg.se/wps/portal/enhetssida/statistik-och-analys/geografi/gisskikt-for-stadens-omradesind
elning (Received 03/02/2023)
Göteborgs Stad (2023). Statistikdatabas Göteborgs Stad. http://statistikdatabas.goteborg.se/pxweb/sv/ (Received
03/02/2023)
Göteborgs Stad (Sverige : 1983-). Stadsbyggnadskontoret (1999). Kulturhistoriskt värdefull bebyggelse i
Göteborg: ett program för bevarande. D. 1. Göteborg: Stadsbyggnadskontoret.
Göteborgs Universitet (2022). Väderstationer vid institutionen för geovetenskaper.
https://www.gu.se/geovetenskaper/vaderstationer-vid-institutionen-for-geovetenskaper (Received
04/19/2023)
Hillier, B. & Hanson, J. (red.) (1984). The Social Logic of Space [Elektronisk resurs]. Cambridge: Cambridge
University Press.
Hillier, B., Penn, A., Hanson, J., Grajewski, T., & Xu, J. (1993). Natural movement: or, configuration and
attraction in urban pedestrian networks. Environment and Planning B: Planning and Design, 20(1),
29-66. https://doi.org/10.1068/b200029
Jacobs, J. (2005). Den amerikanska storstadens liv och förfall. Göteborg: Daidalos. Translated by Charlotte
Hjukström.
Lantmäteriet (n.d). E-tjänster för privatpersoner.
https://www.lantmateriet.se/sv/om-lantmateriet/Sjalvservice/e-tjanster-for-privatpersoner/#geolex
(Received: 4/26/2023)
Lindberg, F., Grimmond, C.S.B., Gabey, A., Huang, B., Kent, C.W., Sun, T., Theeuwes, N., Järvi, L., Ward, H.,
Capel- Timms, I., Chang, Y.Y., Jonsson, P., Krave, N., Liu, D., Meyer, D., Olofson, F., Tan, J.G.,
Wästberg, D., Xue, L., Zhang, Z. (2018) Urban Multi-scale Environmental Predictor (UMEP) - An
integrated tool for city-based climate services. Environmental Modelling and Software 99, 70-87
https://doi.org/10.1016/j.envsoft.2017.09.020
Lindgren, A., Peter, S., Reuter Metelius, A., & Göteborgs stad . Stadsbyggnadskontoret. (2017). Kulturhistoriskt
värdefull bebyggelse Del III Moderna Göteborg : Ett kulturmiljöprogram för Göteborgs stad : En
översikt och ett kunskapsunderlag över utbyggnadsperioden 1955-1975. Göteborg: Göteborgs stad,
Stadsbyggnadskontoret.
Lindberg, F., Thorsson, S., Rayner, D., & Lau, K. (2016). The impact of urban planning strategies on heat stress
in a climate-change perspective. Sustainable Cities and Society 25, 1-12.
https://doi.org/10.1016/j.scs.2016.04.004
40
Lujic, S. (2023). Behaviours of older adults in urban outdoor environments during warm days. [Masteruppsats,
Göteborgs universitet].
Marcus, L. (2018). Overcoming the Subject-Object Dichotomy in Urban Modeling: Axial Maps as Geometric
Representations of Affordances in the Built Environment. Frontiers in Psychology, 9. 449-449.
https://doi.org/10.3389/fpsyg.2018.00449
Melnikov, V.R., Christopoulos, G.I., Krzhizhanovskaya, V.V., Lees, M.H. & Sloot. P.M.A. (2022). Behavioural
thermal regulation explains pedestrian path choices in hot urban environments. Sci Rep 12, 2441.
https://doi.org/10.1038/s41598-022-06383-5
Melnikov, V.R, Krzhizhanovskaya, V.V., Lees, M.H. & Sloot. P.M.A. (2020). The impact of pace of life on
pedestrian heat stress: A computational modelling approach. Environmental Research, 186.
https://doi.org/10.1016/j.envres.2020.109397
Moss, J.L., Doick, K.J., Smith, S., & Shahrestani, M. (2019). Influence of evaporative cooling by urban forests
on cooling demand in cities, Urban Forestry & Urban Greening 37. 65-73.
https://doi.org/10.1016/j.ufug.2018.07.023
Nes, A., Berghauser Pont, M. & Mashhoodi, B. (2012). Combination of Space syntax with spacematrix and the
mixed use index: The Rotterdam South test case. Anesthesiology.
Oke, T., Mills, G., Christen, A., & Voogt, J. (2017). Urban Climates. Cambridge: Cambridge University Press.
https://doi.org/10.1017/9781139016476
Oliveira, S., Andrade, H., & Vaz, T. (2011). The cooling effect of green spaces as a contribution to the
mitigation of urban heat: A case study in Lisbon. Building and Environment, 46(11), 2186–2194.
https://doi.org/10.1016/j.buildenv.2011.04.034
Shiny weather data (2021). https://www.shinyweatherdata.com/ (Received 03/02/2023)
Shishegar, N. (2013). Street Design and Urban Microclimate : Analyzing the Effects of Street Geometry and
Orientation on Airflow and Solar Access in Urban Canyons. Journal of Clean Energy Technologies,
1(1). https://doi.org/10.7763/JOCET.2013.V1.13
Spatial Morphology Group & Chalmers (n.d). Place Syntax tool. https://www.smog.chalmers.se/pst (Received
03/02/2023)
Stewart, I.D. & Oke, T.. (2012). Local Climate Zones for Urban Temperature Studies. Bulletin of the American
Meteorological Society. 93. 1879-1900. https://doi.org/10.1175/BAMS-D-11-00019.1
Ståhle, A., Marcus, L. & Karlström, A. (2005). Place Syntax - Geographic Accessibility with Axial Lines in
GIS. International Symposium on Space Syntax, 5. 131-144.
Svenska Meteorologiska Hydrologiska Institutet [SMHI]. (2018). Sommaren 2018 - Extremt varm och solig.
https://www.smhi.se/klimat/klimatet-da-och-nu/arets-vader/sommaren-2018-extremt-varm-och-solig-1.
138134 (Recieved: 2023-03-28)
Thorsson, S. (2012). Stadsklimatet: åtgärder för att sänka temperaturen i bebyggda områden. Stockholm:
Avdelningen för försvarsanalys, Totalförsvarets forskningsinstitut (FOI).
41
Thorsson, S., Honjo, T., Lindberg, F., Eliasson, I. & Lim, E. (2006). Thermal Comfort and Outdoor Activity in
Japanese Urban Public Places. Environment and Behavior, 39(5).
https://doi.org/10.1177/0013916506294937
Thorsson, S., Lindberg, F., Eliasson, I., & Holmer, B. (2007). Different methods for estimating the mean radiant
temperature in an outdoor urban setting. International Journal of Climatology, 27(14), 1983-1993.
Thorsson, S., Rayner, D., Lindberg, F., Monteiro, A., Katzschner, L., Lau, K. K.-L., Holmer, B. (2017). Present
and projected future mean radiant temperature for three European cities. International Journal of
Biometeorology, 61(9), 1531–1543. https://doi.org/10.1007/s00484-017-1332-2
Thorsson, S., Rocklöv, J., Konarska, J., Lindberg, F., Holmer, B., Dousset, B. & Rayner, D. (2014). Mean
radiant temperature – A predictor of heat related mortality. Urban Climate, 10(2).
http://dx.doi.org/10.1016/j.uclim.2014.01.004
UMEP (n.d). UMEP documentation. https://umep-docs.readthedocs.io/en/latest/index.html# (Received:
3/28/2023)
Wallenberg, N., Holmer, B., Lindberg, F. & Rayner, D. (2023). An anisotropic parameterization scheme for
longwave irradiance and its impact on radiant load in urban outdoor settings. Int J Biometeorol 67,
633–647. https://doi.org/10.1007/s00484-023-02441-3
Wallenberg, N., Thorsson, S., Lindberg, F. & Holmer, B. (2018). Värmestress I Urbana Utomhusmiljöer –
Förekomst Och Möjliga åtgärder I Befintlig Bebyggelse. [PDF]
https://www.folkhalsomyndigheten.se/contentassets/e5286456e91c442a923c6884d84f79be/varmestress
-urbana-utomhusmiljoer-18061-webb-181112.pdf
Wang, X., Dallimer, M., Scott, C.E., Shi, W., & Gao, J. (2021). Tree species richness and diversity predicts the
magnitude of urban heat island mitigation effects of greenspaces, Science of The Total Environment,
770, 145211. https://doi.org/10.1016/j.scitotenv.2021.145211
Wing, C. (2019). EVALUATING GREENERY IN URBAN TYPOLOGIES - A study with a mixed method
approach in Gothenburg, Sweden. [Masteruppsats, Göteborgs universitet]. Gothenburg University
Publications Electronic Archive. https://gupea.ub.gu.se/handle/2077/68595
Yamu, C.; van Nes, A.; Garau, C. (2021). Bill Hillier’s Legacy: Space Syntax—A Synopsis of Basic Concepts,
Measures, and Empirical Application. Sustainability, 13, 3394. https://doi.org/10.3390/su13063394
Yu, Q,. Ji, W., Pu, R., Landry, S., Acheampong, M., O’ Neil-Dunne, J., Ren, Z,. & Tanim,S.H. (2020). A
preliminary exploration of the cooling effect of tree sade in urban landscapes, International Journal of
Applied Earth Observation and Geoinformation, 92, 102161.
https://doi.org/10.1016/j.jag.2020.102161h
42
Appendix I: Meteorological input data
43
Appendix II: Correlation of Attraction Reach and 𝑇𝑚𝑟𝑡 > 55
44
Appendix III: Correlation of Angular Betweenness and 𝑇𝑚𝑟𝑡 > 55
45
Appendix IV: High Pedestrian movement with high or low heat stress
Appendix 4, figure 1 - Map of the study area with highlighted examples of streets with high Attraction Reach with high
𝑇𝑚𝑟𝑡 > 55.
46
Appendix 4, figure 2 - Map of the study area with highlighted examples of streets with high Attraction Reach with low
𝑇𝑚𝑟𝑡 > 55.
47
Appendix 4, figure 3 - Map of the study area with highlighted examples of streets with high Angular Betweenness with high
𝑇𝑚𝑟𝑡 > 55.
48
Appendix 4, figure 4 - Map of the study area with highlighted examples of streets with high Angular Betweenness with low
𝑇𝑚𝑟𝑡 > 55.
49
Appendix 4, figure 5 - Map of the study area with highlighted examples of streets with high Attraction Reach and Angular
Betweenness with high𝑇𝑚𝑟𝑡 > 55.
50
Appendix 4, figure 6 - Map of the study area with highlighted examples of streets with high Attraction Reach and Angular
Betweenness with low𝑇𝑚𝑟𝑡 > 55.
51