West, East or South, which Railway Station in Hudiksvall is Preferable?

Full text


Master Thesis

Spatial planning and development, Umeå university

Spring 2021

West, East or South, which Railway Station in Hudiksvall is Preferable?

A Predictive Study of Future Climate Scenarios from an Accessibility Perspective

Sofia Moberg


Author: Sofia Moberg

Contact: sofiamoobeerg@gmail.com Master thesis (30 ects)

Program: Master in Spatial Planning and Development Department of Geography

Umeå University

Supervisor: Magnus Strömgren


Abstract in English

An expansion of the railway, East Coast Line is essential in order to ensure transportation of passenger and goods back and forth to Northern Sweden. The preliminary studies of the planned expansion to a double track have identified vulnerabilities linked to how our climate changes. Because of these risks and vulnerabilities, the railway station in Hudiksvall needs to be relocated or the Current station needs to be adapted to potential future climate scenarios.

Furthermore, social sustainability and the aspect of accessibility is also a vital perspective to consider during the development of railway infrastructure. This study compares the three different station locations from an accessibility perspective and from different climate

scenarios through Network Analyst in ArcGIS Pro. To visualize future climate scenarios, two RCP-scenarios (Representative Concentration Pathways) are considered, which is RCP 4.5 and RCP8.5. Additionally, the GTFS specification in ArcGIS Pro is used to model public transit to these railway stations in an accessibility perspective. Because one strategy when developing the East Coast Line is to increase the active transportation in comparison to car transportation.

Results from this study indicates that the Current station, which is located in a coastal area will be worst affected of potential future climate scenarios from an accessibility perspective.

Other findings are that vulnerable groups in the society, such as low-income earners and elderly will be most affected, if the railway station remains in the current location. The results from the performed Service area analysis and Location-allocation analysis advocates the Eastern station as a location for the new railway station.

Keywords: accessibility, active transportation, RCP-scenarios, public transit, GTFS.


Abstract in Swedish – Abstrakt på svenska

En expansion av Ostkustbanan är viktigt för att kunna säkerställa transport av passagerare och gods till och från norra Sverige. Förstudierna av den planerade expansionen till ett dubbelspår har identifierat sårbarheter kopplat till hur vårt klimat förändras. Som en följd av dessa risker och sårbarheter behöver järnvägsstationen i Hudiksvall flyttas. Alternativt behöver den nuvarande klimatanpassas. Detta ställer krav på att ta hänsyn till den sociala hållbarheten och tillgängligheten för befolkningen i Hudiksvall. Denna studie ämnar att jämför de tre olika stationslägena ur ett tillgänglighetsperspektiv samt utifrån olika klimatscenarier i ArcGIS Pro.

För att visualisera framtida klimatscenarier beaktas två RCP-scenarier, det vill säga

representativa koncentrationsvägar, vilket är RCP4.5 och RCP8.5. Dessutom används GTFS- specifikationen i ArcGIS Pro för att modellera kollektivtrafik till dessa järnvägsstationer ur ett tillgänglighetsperspektiv, då ett mål med utvecklingen av Ostkustbanan är att utöka andelen av personer som väljer aktiv transport i jämförelse med biltransporter.

Resultatet av denna studie visar att den nuvarande stationen, som är placerad i närheten av Hudiksvalls kust, kommer att vara hårdast drabbad av potentiella framtida klimatscenarier ur ett tillgänglighetsperspektiv. Vidare visar studien på att låginkomsttagare och äldre kommer att drabbas hårdast om järnvägsstationen ligger kvar på den nuvarande platsen. Resultatet från de utförda Service area analyserna och Location-allocation analyserna, visar att det östra alternativet är det alternativ som kommer att vara minst påverkad av ett framtida klimat från ett tillgänglighetsperspektiv.

Nyckelord: tillgänglighet, aktiv transport, RCP-scenarier, kollektivtrafik, GTFS.



First of all, I would like to thank my supervisor at Umeå University, Magnus Strömgren, for helping me to develop my research and introducing me to the GTFS-specification in ArcGIS.

Most of all, thanks for your patience when mine was non-existing.

I would like to express my gratitude to my supervisor at the Swedish Transport

Administration, (STA), Katarina Lind, for enabling me to do this thesis in collaboration with the STA. Thank you for emphasizing the importance of my work and your faith in me when I need it the most. Also, to Isabelle Jonsson, thank you for the time you have spent introducing me to SCALGO Live and helping me to understand the different conditions of the RCP- scenarios. I would also like to take this opportunity to thank everyone in the Planning- department at STA for your warm welcome of me during these special times.

A very big thank you to Hans Gyllow at Hudiksvall’s municipality and to Agneta Davidsson, project leader at STA, for your time going through the different conditions of the East Coast Line project.

To my father-in-law, Roger Söderberg, M.A. in English linguistic, thank you for your time and effort you have spent on this thesis with your carefully reading of my texts.

Additionally, I would like to thank my family and friends who have supported me through this journey. A personal thanks to my father, Hans Moberg, who has stopped asking finally what I am going to be when I grow up.

At last, I would like to express my gratitude to my fiancé Pontus Söderberg, who always

supports and pushes me to challenge myself. Without you I would probably not even have

started my master studies. You are my greatest source of inspiration.


Tabel of Content

1. Introduction ... 10

1.1. Problem Formulation ... 11

1.1.2. Aim ... 13

2. Theoretical framework & Literature review ... 15

2.1. Theoretical Framework ... 15

2.1.1. How can Accessibility & Active transportation be Achieved? ... 15

2.1.2. Concepts of Planning for Climate Changes ... 16

2.1.3. Representative Concentration Pathways (RCP) ... 16

2.2. Literature review ... 18

2.2.1. Accessibility of Public Transit Systems ... 18

2.2.2. Active Transportation & Accessibility ... 20

2.2.3. Challenges with Implementation of Climate Adaptation ... 21

2.2.4. Expected Events in a Future Climate ... 23

3. Method ... 25

3.1. GIS as a Quantitative Method ... 25

3.1.3. Why Network Analyst? ... 27

3.2. Data sample & Methodological settings ... 28

3.2.1. Methodological Choices for the Network Analysis ... 28

3.2.2. Methodological Choices for the Climate Layers ... 30

3.2.3. Restrictions & Barriers in the Network ... 30

3.2.4. Methodological Choices for the Service Area Analysis ... 31

3.2.5. Methodological Choices for Location-allocation Analysis ... 32

3.3. Limitations... 33

4. East Coast Line & Travel Patterns ... 34

4.1. Travel Patterns of Today ... 34

4.2. Routes of the East Coast Line ... 37

5. Result ... 38

5.1. How will the Accessibility be Affected Regarding Different Locations? ... 38

5.1.1. Catchment Area of Various Groups ... 43

5.2. How will the Concerned Locations be Affected Regarding RCP-scenario 4.5 & 8.5? ... 47

5.2.1. Expected Outcomes of Accessibility for Pedestrians ... 51

5.2.2. Expected Outcomes of Accessibility for Bicyclists ... 53

5.2.3. Expected Outcomes of Accessibility for E-bicyclists ... 55

5.2.4. Expected Outcomes for Various Groups ... 56

5.3. What Location is Preferable Regarding both the RCP-scenarios and Accessibility?... 62

5.3.1. Preferable Location with RCP4.5 ... 65


5.3.2. Preferable Location with RCP8.5 ... 67

6. Discussion ... 71

6.1. Preferable Location in a Socio-accessible Perspective ... 71

6.2. Preferable Location in a Socio-environmental Perspective ... 73

6.3. Socio-environmental Preferable Location ... 75

7. Conclusion ... 77

8. References ... 79


List of figures.

Figure 1. Overview of the East Coast Line and the study area. ... 12

Figure 2. Developments curves for emission of greenhouse gases with different RCPs ... 17

Figure 3. Global mean sea level rise depending on RCP- scenarios ... 24

Figure 4. Road network of Hudiksvall’s municipality. ... 25

Figure 5. Road network around the potential railway stations. ... 26

Figure 6. Work commuters from Hudiksvall ... 34

Figure 7. Work commuters to Hudiksvall.. ... 35

Figure 8. High School commuters from Hudiksvall ... 36

Figure 9. Overview of alternative route of the railway depending in chosen location. ... 37

Figure 10. Overview of the catchment areas with different travel modes ... 39

Figure 11. Number & share of population (%) within each of the cutoff times with walking as travel mode. ... 40

Figure 12. Number & share of population (%) within each of the cutoff times with cycle as travel mode. ... 41

Figure 13. Number & share of population (%) within each of the cutoff times with e-bike as travel mode. ... 42

Figure 14. Number & share of the total female population (%) within each of the cutoff times with pedestrian as travel mode ... 46

Figure 15. Share of the elderly population (%) within each of the cutoff times, with pedestrian as travel mode ... 47

Figure 16. Overview of flooded areas with RCP4.5 & RCP8.5 ... 48

Figure 17. Comparison between the usual catchment area for pedestrians and the catchment area of pedestrians with climate barriers RCP4.5 & RCP8.5. ... 49

Figure 18. Comparison between the usual catchment area for bicyclists and the catchment area of bicyclists with climate barriers RCP4.5 & RCP8.5. ... 50

Figure 19. Comparison between the usual catchment area for e-bicyclist and the catchment area of e- bicyclists with climate barriers RCP4.5 & RCP8.5. ... 50

Figure 20. Number & share of population (%) within each of the cutoff times with pedestrian as travel mode & RCP4.5 ... 51

Figure 21. Number & share of the population (%) within each of the cutoff times with pedestrian as travel mode & RCP8.5. ... 52

Figure 22. Number & share of population (%) within each of the cutoff times with bicycle as travel mode & RCP4.5. ... 54

Figure 23. Number & share of population (%) within each of the cutoff times with bicycle as travel mode & RCP8.5. ... 54

Figure 24. Number & share of population (%) within each of the cutoff times with e-bike as travel mode & RCP4.5. ... 55

Figure 25. Number & share of population (%) within each of the cutoff times with e-bike as travel mode & RCP8.5. ... 56

Figure 26. Number & share of the total female population (%) within each of the cutoff times, with pedestrian as travel mode and RCP4.5. ... 59

Figure 27. Number & share of the total female population (%) within each of the cutoff times, with pedestrian as travel mode and RCP8.5. ... 60

Figure 28. Number & share of the elderly population (%) within each of the cutoff times, with pedestrian as travel mode and RCP4.5. ... 61

Figure 29. Number & share of the elderly population (%) within each of the cutoff times, with pedestrian as travel mode and RCP8.5. ... 61

Figure 30. Chosen facility based on location-allocation analysis with pedestrian as travel mode. ... 62

Figure 31. Chosen facility based on location-allocation analysis with bicycle as travel mode. ... 62


Figure 32. Chosen facility based on location-allocation analysis with e-bike as travel mode. ... 64

Figure 33. Chosen facility based on location-allocation analysis with RCP4.5 and pedestrian as travel mode. ... 65

Figure 34. Chosen facility based on location-allocation analysis with RCP4.5 and bicycle as travel mode. ... 66

Figure 35. Chosen facility based on location-allocation analysis with RCP4.5 and e-bike as travel mode. ... 67

Figure 36. Chosen facility based on location-allocation analysis with RCP8.5 and pedestrian as travel mode. ... 68

Figure 37. Chosen facility based on location-allocation analysis with RCP8.5 and bicycle as travel mode. ... 69

Figure 38. Chosen facility based on location-allocation analysis with RCP8.5 and e-bike as travel mode. ... 70

List of tables Table 1. Data gathering and source. ... 28

Table 2. Average speed per travel mode. Source: Bedogni et al., (2016). ... 29

Table 3. Settings in SCALGO Live ... 30

Table 4. Total amount of the population that each of the station captures 2017 & 2019 ... 43

Table 5. Number of Low-income earners in each cutoff time per station (Pedestrian) ... 44

Table 6. Number of Low-income earners in each cutoff time per station (Bicycle) ... 44

Table 7. Number of Middle-high-income earners in each cutoff time per station (Pedestrian). ... 44

Table 8. Number of Middle-high income earners in each cutoff time per station (Bicycle) ... 45

Table 9. Share of Low-income earners in each cutoff time & station for different RCP-scenario & the differential from today (Pedestrian) ... 57

Table 10. Share of Low-income earners in each cutoff time and station with RCP4.5 & RCP8.5 (Bicycle). ... 57

Table 11. Share of Middle-high income earners in each cutoff time and station with RCP4.5 & RCP8.5 (Pedestrian). ... 58

Table 12. Share of Middle-high income earners in each cutoff time and station with RCP4.5 & RCP8.5

(Bicycle) ... 58



1. Introduction

Events of extreme weather all around the world has caused damage on the infrastructure and affect the population. This has shed light on the importance to considering future climate scenarios in spatial planning (Monterio & Ferreira, 2020). But for decades, the significant impact of the population everyday life as a direct effect of environmental changes have been less prioritized in spatial planning (Nyström & Tonell, 2012). Instead, focus has relied on economic growth and urban sprawl was a planning phenomenon that after the Second World War was a way of developing it (Arnstberg, 2005).

Increased urban sprawl and decentralization expanded the car dependence (Banister, 2008). A wealthier Western World contributes to that more people own a car, which increases the usage and is one reason of elevated rate of greenhouse gas emissions (Norström & Losciale, 1995).

In order to measure how the climate changes due to greenhouse gas emissions, different Representative Concentration Pathways (RCP) 1 has been developed (Van Vuuren et al., 2011). These climate scenarios are common in climate adaptive planning (SMHI, 2020).

The national strategy of climate adaptations that the Swedish government has established, as a direct effect of the Paris Agreement, puts work of climate and vulnerability analysis to a central part of the planning procedure (Prop. 2017/18:163). To consider risks for landslides, erosion, cloudburst, droughts, and sea level rise when planning for new development creates the condition to plan for a robust and sustainable transportation system (Nyström & Tonell, 2012), with uncertainty considered (Coaffee & Lee, 2016). Also, this ensures that we can live with developments impacts of nature (Harvey, 2003). An ecological sustainable transport system is the system that ensures good water, ground and air quality and take the biodiversity in account for both humans and nature (Nyström & Tonell, 2012).

However, decentralization and increased usage of car has not only elevated the carbon dioxide emissions (Norström & Losciale, 1995), it has also affected the mobility. Contributing to less attractiveness of active transportation, such as walking, and cycling (Banister, 2008). Another used terminology of active transportation is non-motorized transport (Sallis, Frank, Saelens &

Kraft, 2004) or slow modes travel (Elldér, Larsson, Solá & Vilhelmson, 2018). However,

1 Over the years different climate scenarios has been adapted to describes a possible development of the climate

based on assumptions of the human activities and how much energy that is retained in the atmosphere. RCP-

scenarios one way of describing how the climate changes based on emission of greenhouse gases, air pollution

and land-use. Today, four RCP-scenarios has been estimated, which is 2.6, 4.5, 6 and 8,5. The higher RCP, the

more emission of greenhouse gases. The adapted RCP-scenarios and its meaning will be explained further in the

theoretical part.



Banister, (2008) has also seen in his study that local public transport has also been less attractive because of urban sprawl. Additionally, less active transportation can e.g., lead to disparities in health (Gray et al., 2011). The same study has also stated that a wider sub- regional pattern of housing, economic development, land use and transportation are factors of social exclusion (ibid.). To prevent this, efforts in the planning process are required to achieve a transportation system that is social sustainable. Implying that the system should create opportunities for positive experiences. A good life of culture, public service, good health and include everyone (Nyström & Tonell, 2012).

But planning for a robust transportation network and social sustainability is challenging and complex. Planning for a robust transportation network because future climate scenarios are predictions. Even if we today know much about the climate change, much is yet to be discovered (Lennartsson & Simonsson, 2007). To fulfill social sustainability goal is difficult because they are soft values that are challenging to measure. The challenge with the discourse of social sustainability is that it has become a collective concept across the whole population.

Leading to difficulties to identify which measures are aimed for which individuals (Nyström

& Tonell, 2012).

Questions that need to be asked are whose right’s and whose city? Lefebvre’s theory of the right to the city refers to urban environment and the context to everyday life and social

relations (Lefebvre, 1996; Shields, 2011). While Marcuse (2012, p. 34) defines the right to the city as individual justice for all. The understanding of specific life situations needs to be considered to be able to create a good environment in spatial planning. Life situations is to be understood as the environmental, economic, and social aspects and the differences in

transportation patterns that come with it (Nyström & Tonell, 2012).

1.1. Problem Formulation

Passenger transportation on the East Coast Line is extensive due to the relationship with

metropolitan areas, such as Stockholm and Gothenburg. Capacity utilization is already high

and with increased population growth in the cities around the East Coast Line, there is a

possibility that travelling times will increase. Because of this it is of high priority to with take

measures in order to ensure a robust and sustainable transport system for passenger and goods

to and from Northern Sweden. The plan is to expand the railway line to a double track and

increase the travels with train in comparison to car travels. This would have positive effects

on both the environment and social sustainability (Åström, 2021).



When planning for large-scale infrastructure, it is vital to consider climate change and adaption that may be bedded in a future climate in consideration, as that ensures that we can live with the developments (Harvey, 2003). While it is equally important to consider social sustainability and equity (Garcia, Macário, Menezes & Lourerio, 2018; Åström, 2021). In that way, climate change and social sustainability is interlinked. Because if we manage to fulfill the 13 th Sustainable Development Goal, to fight climate change. We are on a good way to achieve social sustainability (AtKisson, 2017, 17 May).

Considering climate change as a part of the National Climate Adaption Strategy (Prop.

2017/18:163), the planned expansion of the East Coast Line has identified vulnerabilities of the existing transportation system (Åström, 2021). But the scope of this study is not to investigate the whole route Gävle-Västeraspby, because it would have been too extensive in relation to the time.

The aim is rather to investigate the adaptations that is needed and what social impacts it may have on one municipality along the East Coast Line, namely Hudiksvall. See figure 1. By studying the development of the East Coast Line and the

relocation of the railway station in Hudiksvall, this study

contributes to accessibility perspectives in transportation

planning by analyzing how the accessibility for various groups in society can be affected when performing large redevelopments of infrastructure. Also, this study contributes to how

Figure 1. Overview of the East Coast Line and the study area.



accessibility for various groups in society can be affected by climate change. Additionally, a part of this research is to analyze active travel modes, such as, pedestrians, with the possibility to use public transit, and bicyclist’s accessibility to different alternatives of a potential new railway station. As this study aims to apply the GTFS-specification, which makes it possible to apply public transit data to Network analyst in ArcGIS Pro. Therefore, car transportation is excluded in this study.

Identified vulnerabilities around the traffic lane in Hudiksvall are landslides and erosion. Plus, flooding, sea level rise interlinked to different RCP-scenarios. This has led the Swedish Transport administration, (STA) to consider other options for the location of the railway station (Åström, 2021). This study aims to compare the three alternative locations from an accessibility perspective and how accessibility can be affected based on different climate scenarios.

If measures are not taken, these identified risks and vulnerabilities will cause affects from a social sustainability perspective as well (Åström, 2021). After all, the importance of a robust and accessible transportation system is essential to prevent social exclusion (Gray et al., 2011). Because an accessible transportation system makes it possible for marginal groups in the society, without the possibility to own a car, to access employments or other daily activities (Garcia et al., 2018). Therefore, accessibility to public transport creates the conditions for successful integration of vulnerable groups in society (Gray et al., 2011;

Nyström & Tonell. 2012).

1.1.2. Aim

As a consequence of the identified risks and vulnerabilities of the existing system along the East Coast Line, the location of the railway station in Hudiksvall will need to be reviewed.

The STA has developed proposals for three possible locations for a new railway station, which one is the existing one. Depending on where this new station will be located, different conditions are created for access to the transport system from an accessibility and equality perspective. At the same time, different climate scenarios can cause disturbances and affect the accessibility. This study aims to contribute to accessibility planning from a climate perspective, by analyzing various groups of the society’s accessibility towards the possible locations from a climate scenario of today and in the future. Hence, the aim of this study is to conduct a predictive study of the possible locations of the railway station and answer

following questions:



1. How will the accessibility be affected regarding the different locations?

2. How will the concerned locations be affected regarding RCP-scenario 4.5 and 8.5?

3. What location is preferable regarding both the RCP-scenarios and accessibility?



2. Theoretical framework & Literature review

This episode consists of two parts. The first one is theoretical framework, which deals with the theoretical definitions of accessibility, adaptive planning, and definitions of the RCP- scenarios. The second part is the literature review episode, which is there to guide the decisions that were made when setting important parameters in for my Network analysis in ArcGIS.

2.1. Theoretical Framework

The first part of this episode will define definitions of accessibility and how sustainable mobility can be achieved with active transportation in transportation planning. The second part of this episode consists of theories of planning for climate change and the signification of RCP-scenarios.

2.1.1. How can Accessibility & Active transportation be Achieved?

A sustainable mobility paradigm involves overcoming the car dependency and make room for local public transit and active transportation. However, in order to overcome the car

dependency, the public transit system, together with the active transportation, must precede car traffic and be prioritized in planning. But the problem with the sustainable mobility paradigm is that people overall tend to measure their travels in time. At the same time that travels are not something that people wish to undertake for their own sake. Therefore, to live up to the sustainable mobility paradigm, proximity to different facilities, services and housing is needed (Banister, 2008). This is where accessibility comes in. Accessibility has long been associated with speed (Alfonzo, 2005). But in a sustainable mobility paradigm (Bansiter, 2012) it is rather enabled with proximity (Alfonzo, 2005).

Halden’s (2012) theory about accessibility lands in a new accessibility paradigm.

Accessibility is an overall terminology used to describe proximity to access different services and is defined different for different individuals. The theory strives to achieve accessibility for various groups of people. In order to live up to the accessibility paradigm a more user centric approach needs to be adapted in accessibility planning (ibid.). This is in line with Lefebvre’s theory of the right to the city, referring to different needs for different groups of people. The meaning is that planning needs to consider different life situations due to that it requires different measurements (Lefebvre, 1996; Shield, 2011). But in order to know which

measurements are required for different individuals: Who the new station is planned for needs to be considered. Is it the people in charge or people in society? This is something that

Harvey’s (2003) theory about the right to the city highlights and builds on. Different



perspectives of various groups of people needs to take more space in pre-studies in order to build inclusive societies (Harvey, 2003; Nyström & Tonell, 2012).

2.1.2. Concepts of Planning for Climate Changes

Coaffee & Lee’s (2016) theory about adaptive planning, refers to involving climate change and uncertainty when planning for development. How resilient a system is, depends on adaption and adaptability. Adaptation and adaptability of a system is linked to policymaking by urban governance. In order to classify planning as adaptive planning, (to make sure that the building environment is resilient against perturbation), it has to consider risks and

vulnerabilities for shock events and future climate scenarios (Coaffee & Lee, 2016).

The opposite to adaptive planning is maladaptive planning. The theory about maladaptive planning refers to planning that is no longer fit for purpose or increases vulnerability. Rather than reducing risk and vulnerability to the impacts of climate change. Examples of

maladaptive planning is how the leader of governmental sectors can fail to build resilience and undermine other objectives and increase vulnerability. Building resilience on just one level, e.g., local level, can risk undermining it at others, such as regional and national levels, affecting the local level (Coaffee & Lee, 2016).

2.1.3. Representative Concentration Pathways (RCP)

One climate scenario describes a possible development of the climate based on assumptions of the human activities and how much energy that is retained in the atmosphere (van Vuuren et al., 2011). IPCC 2 has developed four global climate scenarios connected to the average temperature (SGU, 2020) called RCP-scenarios. Those are future climate estimations based on exiting literature and components of radioactive forcing as input for climate modeling.

These components are emission from greenhouse gases, air pollution and land use and aims to give an indicator of how the climate changes bases on emissions (van Vuuren et al., 2011).

The RCP-scenarios can be reached with a combination of economical, technological,

demographic, and political developments (IPCC, 2014). These scenarios extend over a period to year 2100 (van Vuuren et al., 2011). Mapping and studying these different RCP-scenarios, it is called to add a climate factor 3 . This means that for instance the rain intensity increases, as

2 Intergovernmental Panel on Climate Change is UN: s climate panels who have developed an evaluation report (SGU, 2020).

3 Applicate a climate factor is to increase the rain intensity of different rain-events. E.g., 100-year rainfall.

Adopting a climate factor means that the rain intensity corresponds to a future RCP-scenario (MSB, 2017). See

page 21 & 29 for further explanation and adapted climate factor.



an effect of how the climate is changing (MSB, 2017). Figure 2 illustrates the emission of CO 2 and the different scenarios. RCP4.5

In order to live up to RCP4.5, stricter climate policies need to be adapted. We need to have low energy intensity and adapt more plans for afforestation to capture emissions of CO 2 . The needed area for agriculture production is lower as we must make room for larger harvests due to our consumption. The carbon dioxide emissions will increase and stagnate year 2040 (SMHI, 2020). RCP8.5

RCP-scenario 8.5 is the scenario that is predicted if the carbon dioxide

emissions increase to three times more than today. The population of the Earth will increase to 12 billion by the year of 2100, contributing to larger areas of pastureland is needed. The population are dependent on fossil fuels. The energy intensity is high and

there is a no additional climate policy (SMHI, 2020).

Figure 2. Developments curves for emission of greenhouse gases with different RCPs.

The green line represents RCP2.6, the red RCP4.5, the black

RCP 6, and the blue RCP8.5. Source: van Vuuren et al.,




2.2. Literature review

This study focuses on active transportation, also called slow modes travel to access the public transit system. Therefore, this episode consists of three parts; previous studies of

transportation planning and accessibility of public transit system, definitions of active transportation and measures taken to increase the rate of active transportation. The last part consists of previous research about future climate scenarios and expected events caused by these scenarios and its effects on the infrastructure and planning.

2.2.1. Accessibility of Public Transit Systems

Accessibility is a complex part of transport planning because the terminology is defined different for different individuals (Halden, 2012). But lately, different goals have been adapted with the concept of accessibility of the transport system (Bertolini, le Clercq &

Kapoen, 2005). For instance, accessibility has long been associated with travel speed to facilities and services (Alfonzo, 2005), and research has focused on accessibility of the

transportation system to economic clusters (Grengs, 2015). Therefore, an economic and social goal of accessibility has been to develop the ability for workers and customers to access the system for daily use and access employments (Bertolini et al., 2005). However, as a growing interest of active transportation to access public transit system the meaning of accessibility has shift (Chan & Farber, 2020).

Still, conflicts can be seen in the concept of accessibility, because accessibility is defined different among various groups (Halden, 2012). For instance, some have argued that a

transportation system is social sustainable and accessible if people in an urban area can walk, bike, or go by public transport to work (Nyström & Tonell, 2012). This conclusion certainly supports Elldér et al., (2018) study that accessibility to the public transit system is essential.

Since the main part of the Swedish population does not live close enough to the workplace to get there by slow modes of travel (ibid.). Therefore, in order to decrease the car dependency, which would have good impacts on public health and the environment, planning for

accessibility of the public transit system with active transportation is vital (Chan & Farber, 2020). However, this requires a reliability of the transportation network and reasonable travel time for people to choose public transit before car transportation (Noland & Polak, 2002).

Unfortunately, lack of strategic thinking about accessibility and mobility contributes to

inefficiency in the transportation system and makes accessibility a multifaceted concept

(Garcia et al., 2018). A reason for that is that studies of social patterns and accessibility of

various groups of people is have not been studied in an extent that is needed (Nyström &



Tonell, 2012; Garcia et al., 2018). Garcia et al., (2018) suggests that accessibility should be seen as a tool to improve the social equity. With that meaning that every person in a society would have access to a transportation system, independently of their needs (ibid.).

However, how different groups of people use the public transportation system is something that Trivector, (2018a) has studied. They saw a correlation between usage of public transit system and income, education, and gender. Females tend to choose public transportation in a larger extent compared to males, due to environmental benefits (Trivector, 2018a).

Meanwhile other studies have shown that the financial resources and the possibility to own a car is a factor whether to travel by public transit or not. This is in line with Sayfoor, (2015) study that high-income takers prefer the car as a transportation means in front of public transit and that more men are high-income earners (ibid.). This is something that Trivector, (2018c) also has concluded, that more males own a car compared to females (ibid.).

However, even if the concept of accessibility among different groups to a large extent is missing, another study by Trivector, (2018b) has identified females and elderly as an exposed group of the public transportation system. Because of the feeling of precariousness, they tend to limit the usage of the transportation system to specific times of the day. This relates to Garcia et al., (2018) suggestion that more studies of social patterns connected to accessibility like Trivector’s study needs to be performed in order to achieve equity in the transportation system. Harvey’s, (2003) theory of the right to the city and the questions of who we plan for also shed lights on equity in planning. Referring to that accessibility problems in spatial environments, individual mobility and travel patterns must take a bigger place in the planning procedure of mobility systems (Shield, 2011; Nyström & Tonell, 2012; Garcia et al., 2018).

Urban planners need to change the mindset and put people in the center of the planning

procedure (Harvey, 2003). This is in line with Lubitow, Rainer and Bassett, (2017) study that

in order to build an equal transit system (Garcia et al., 2018) the transit dependent people need

to be identified. With that meaning, where the people without private transportation, elderly,

youths, and people below a median income level life (Lubitow et al., 2017). Yet, even if the

transit dependent population is identified, this is not an easy task and is something that usually

is failing during developments of public transit systems (Toms & Song, 2016). Planners strive

to develop an equal transportation system for low-income inhabitants and elderly so that they

can access employments and daily activities but usually ends up planning for the majority of

the population (ibid.).



2.2.2. Active Transportation & Accessibility

Another aspect that makes it vital to increase the active transportation is that physical

inactivity in the Western World is high (Bourne et al., 2018). At the same time, increasing the active transportation towards public transit systems has founded to be a good alternative to decreasing the car dependency and achieve more balance of sustainable transportation (Chan

& Farber, 2020). Therefore, to achieve a sustainable mobility paradigm, planning for

accessibility that is enabled with proximity is essential, in order to with take measurements for people to choose active transportation (Banister, 2008). But that accessibility for various groups of people has been downplayed in accessibility research can also be seen when

looking at a factor such as proximity (Halden, 2012; Garcia et al., 2018), which is stated to be a fundamental factor for people to choose active transportation (Bansiter, 2008: Elldér et al., 2018). Improving the accessibility towards facilities and services also improves the social equity for groups like youths, elderly and low-income workers, which has less opportunities for car transportation (Elldér et al., 2018).

Proximity can be defined with terms of physical distance but also travel time. Previous studies use distance and travel time in combination to define proximity, because of that the physical landscape is a vital factor (Elldér et al., 2018). That aspect is also something that Alfonzo, (2005) concludes, that the time it takes to travel is a crucial factor in the decision to walk and bike in contrast to using the car (Alfonzo, 2005). Willingness to walk varies among different people and the purpose of the trip. There is small part of the population that is willing to walk more than one kilometer for everyday errands (Elldér et al., 2018). Relating to that the

accessibility is defined different for different individuals (Halden, 2012).

Simultaneously, physical infrastructure and the landscape in combination with transportation planning can be an obstacle for people’s accessibility and mobility regarding active

transportation. To have accessibility, an overall land use planning that allows a mixture of facilities and services is needed (Bertolini et al., 2005). Therefore, planners have focused on the built environment. Planning for safe active transportation, with shorter distance are factors which contributes to environmentally friendly and slow modes of travel (Carlson et al., 2014;

Elldér et al., 2018). This proves Nyström & Tonell, (2012) statement that a mixture of

facilities is essential in order for people to choose active transportation. At the same time

Banister, (2008) states that the lengths of trips need to be kept below the threshold for walking

and biking modes to attract people to choose active transportation (ibid.). A reason to why this



is challenging is that the threshold differs from individual to individual (Halden, 2012; Garcia et al., 2018).

A general rule for trips lengths, based on European studies, is to have one-way journeys ≤5 km for bicycle, or ≤2.5 for walking (Rabl & de Nazelle, 2012). However, other studies of Carlson et al., (2014) saw a trend among young people that the environmental benefits and sustainability is a contributing driver of active transportation, instead of keeping the travel distance under the accepted threshold for travel distance (ibid.). This is in line with Chan &

Farber, (2020) conclusion that younger population and households with low access to cars tend to combine active transportation with public transit. Although, this conclusion is not surprising since these groups often has less resources, which forces them to combine these two travel modes (ibid.). Meanwhile, other previous studies have come up with the conclusion that habit is the only factor of adult’s choices between active transportation and car

transportation, regardless of the distance (de Bruijn, Kremers, Singh, van den Putte & van Mechelen, 2009). Yet the most positive correlation of active transportation in terms of walking and bicycling is the correlation of high-income and education, meaning that high- educated people with higher income tend to choose active transportation in a larger degree than low educated, low-income earners (ibid.). Travel Speed of Active Transportation

The average speed for the usual biking and walking is studied in previous research by

Bedogni, Felice & Bononi (2016). The study consists of 5,400 sampled GPS-coordinates from smart phone devices of eight different participants. The result ended with an average speed for pedestrians with 5 km/h and 20 km/h for bicyclists (ibid.).

E-bikes has become a major part of active transportation. The technological progress has increased the number of bicycles worldwide and the motor of the bike helps to come up to a speed of 25km/h. This progress has made travel times shorter than with usual bikes and has made it possible to travel longer distances. However, the distances have not yet increased, but the number of trips with a bicycle has (Stenner et al., 2020).

2.2.3. Challenges with Implementation of Climate Adaptation

Planning for new developments involves different priorities, such as economic development,

political will, and protection of nature. In the role of a planner, it is not always clear what to

prioritize as these three priorities usually end up with three conflicts. Economic growth

usually ends in a conflict of fairly growth in society and not degrading the ecosystem in the

process (Campbell, 1996). Additionally, previous studies have shown that population growth



leads to pressure on the environment, resulting in loss of biodiversity, pollution, and natural resource consumption (Monterio & Ferreria, 2020).

Road networks are important part of infrastructure as it physically interconnects people with the world. Previous events of extreme weather around the world shows how much damage it can cause to the infrastructure and the accessibility for the population (Toma-Danila, 2018; Li

& Kaewunruen, 2018). Events related to climate change are the most common cause of disruption of the railway systems (Lindgren, Jonsson & Carlsson-Kanyama, 2009). Expected events such as flooding, sea level rise, erosion and storms will most likely increase in

frequency and hit hard on the already vulnerable high-populated low-elevation coastal areas (Monterio & Ferreira, 2020). Both hot and cold climates can affect railway networks, which may lead to deterioration and rail buckling (Li & Kaewunruen, 2018). The ability of a

network to function and adapt to new and sudden events is an important aspect that need to be achieved in order to prevent social and economic losses (Toma-Danila, 2018). Therefore, by identifying risks and vulnerabilities connected to a changing climate, is a way of planning for developments that is resilient and adaptive. If the risks are considered (Coaffee & Lee, 2016).

Although, this is challenging since long term planning that is associated with infrastructure makes is difficult to plan for climate changes that possibly can occur in the future (Lindgren, Jonsson & Carlsson-Kanyama, 2009). But already today increased frequency of flooding damaging the railway network can be identified in Sweden and other parts of Europe (ibid.).

This proves that the uncertainty about climate change and future climate scenarios will make the resilience process a continuous journey. Especially with identifying the problems and planning for new infrastructure solutions and developments, that either mitigates the already existing risks and vulnerabilities. Alternatively, planning new developments that are adapted to a future climate to get an adaptive planning procedure. Learning from past of extreme events and do risk assessments is also vital in order to prevent the planning for being

maladaptive (Coaffe & Lee, 2016). For that reason, research has started to focus on simulating and estimating disasters and events connected to climate change using GIS (Toma-Danila, 2018).

It is vital to implement climate policy documents and measures for adaptation. Yet, it will be

challenging for decision-makers at every level. These policy documents require navigation of

information and facts generated at different scales into options of adaptation. These options

should also be socially and politically acceptable, due to significant degrees of uncertainty

(McEvoy, 2013). Referring to Campbell, (1996) conclusion that planning often ends up in



conflicts of prioritized areas. This in turn is in line with Bertolini et al., (2005) conclusion that the policy documents should be acceptable on other levels than just the environmental level because of degrees of uncertainty. Because of this governments often want to include other goals in the environmental goal, either economic or social, to access space in plan-making (ibid.). This is in line with Marcuse’s (1976) statement that the role of a planner often comes down to narrow interest of the political will from governance and authorities. A matter of fact is that it has been difficult to adopt adaptation and adaptability in policymaking because it is uncertain and costly. Thus, it is one reason why it has been less prioritized to have adaptive planning (Coaffee & Lee, 2016). All the different aspects need to be weighted and considered (Bertolini et al., 2005) and “the right plan” is the plan that does least harm in a public interest (Campbell, 1996).

2.2.4. Expected Events in a Future Climate

Dahlström (2010) has developed a formula to estimate rainfall intensities with duration from 5 minutes to 24 hours. This formula is commonly used by planners in Sweden to handle rain intensities in urban structures and is applicable if local precipitation statistics are limited or missing in a specific area (SMHI, 2015c). This formula is used to estimate the 100-years rainfall adopted for the climate barriers in the Network analysis (Dahlström, 2010; MSB, 2017). Depending on different RCP-scenarios, the annual precipitation estimates to increase.

Namely 10–30 % for RCP4.5 and 15–40% for RCP8.5 in the whole of Sweden (SMHI, 2015a). Cloudbursts are estimated to increase both in intensity and in frequency in the whole of Sweden due to a warmer climate (SMHI, 2017). The definition of a cloudbursts is that it could rain 50 mm within an hour (SMHI, 2015c). While a 100-years rain events is

precipitation of 44 mm rain with a duration of 30 minutes (Dahlström, 2010; Svenskt vatten, 2016; MSB, 2017). To know how much the rain intensity is increased for events like a 100- year rain and for specific RCP-scenarios, a climate factor could be adapted. Using a climate factor means that the rain volume increases with 20–50 percent based on today’s knowledge (MSB, 2017).

Depending on different RCP-scenarios the global sea level will rise with different magnitude.

The sea level is estimated to rise with one meter by the year of 2100 with RCP-scenario 8.5.

The worst affected areas would be the coastal regions (SMHI, 2015b). Se figure 3.


24 Water Depth and Flooded Areas

These future climate scenarios and the extreme weather will most likely cause floods in built areas, such as areas of housing, which will affect the physical infrastructure (DHI, 2019).

Nevertheless, other factors, such as densification of properties in urban environments causes lack of drainage systems and easily creates floods during extreme weathers (Svenskt vatten, 2016). The consequences will of course depend on the water depth of where the flood occurs.

By water depth, meaning the amount of water that is gathered at the specific location.

Mapping consequences of extreme floods the guideline is to have a value of 10–30 cm of water depth. This guideline is based on that the accessibility is made difficult at those water depths. Water depths around 30–50 cm make the accessibility impossible. While water depth

≥50 cm causes risks of damages on housing and infrastructure (DHI, 2016).

Figure 3. Global mean sea level rise depending on RCP- scenarios. Source: IPCC, (2014).



3. Method

This chapter consists of two major parts. The first part explains the Network Analyst and why this method is suitable for answer the research questions. The second part consists of

information and clarification of the collected data, the attributes, settings, and parameters in the network.

3.1. GIS as a Quantitative Method

To model transportation network, GIS is an appropriate tool. One among many platforms for road network models is ArcGIS, with the extension Network Analyst (Toma- Danila, 2018).

This study is a quantitative study using ArcGIS Desktop and ArcGIS Pro to model the transport network in Hudiksvall. In order to see how the possible re-location of the railway station will affect the accessibility for the population in Hudiksvall and how potential future climate scenarios may affect the accessibility of the different station locations. The result is based on two analyses, Service area analysis and Location-allocation analysis. These analyses will be explained in the second part.

3.1.1. Network Analyst as a Quantitative Tool In the network model, data of links, nodes, edges, and junctions becomes a cohesive set of a network, which can represent car roads, cycle- and pedestrian roads and railways, just to name a few. All the data that characterizes the network, such as the attributes named above, are stored as attribute information in a vector database (Heywood, Cornelius &

Carver, 2011). In this network the node can also represent bus stops and the potential railway stations.

Figure 4 illustrates the overall the network in Hudiksvall. While figure 5 illustrates a closer picture of the network around the railway stations.

Figure 4. Road network of Hudiksvall’s municipality.


26 Figure 5. Road network around the potential railway stations.

When building the network, impedance is a feature, representing the cost of traveling in network links and stops. Impedance values are important in determining routes finding, allocation and spatial interactions (Heywood et al., 2011). In this case the impedance is the time it takes to travel by bike or walk with an assumption of the average speed to the possible locations for the railway stations.

Supply and demand are other factors that stores in the vector database and are equally

important as the impedance. Supply is the quantity of a resource available at a center that can satisfy demand associated with the links of a network (Heywood et al., 2011). Supply

represents the amount of people within an area for each location of the railway. The network centers are the different locations for the railway station. However, other centers that is important to have in mind are center or stops that serves a large amount of people or people that are dependent on public transports. These locations can be hospitals and schools

(Mitchell, 1999; Heywood et al., 2011). Therefore, it is important to have railway stations in

proximity to these other centers. The demand is the utilization of a resource by an entity that

is associated with a network link or node (Mitchell, 1999; Heywood et al., 2011), with other

words it is the catchment area for each location of each railway station.



3.1.3. Why Network Analyst?

A network is a system of interconnected linear features through which materials, goods and people are transported. In GIS, network models are abstract representations of the components and characteristics of their real-world counterparts (Heywood et al., 2011). Network analysis makes it possible analyze movements and model potential paths (ArcGIS Desktop, 2020).

Therefore, Network Analyst is an appropriate tool as this study aims to answer how the accessibility may be affected regarding the different locations and how these locations will be affected in a future climate.

Another reason to why Network Analyst is beneficial in this study is that the tool is efficient during strategic decisions due to that it gives an understanding of current market dynamics and potential market dynamics. Network Analyst helps to solving problems like transportation costs and finding best stop, etc. (ArcGIS Desktop, 2020). In this case the focus lies on the accessibility for citizens to the potential railway stations and how the location can be affected by a future climate. General Transit Feed Specifications (GTFS)

With ArcGIS Pro and Network Analyst it is possible to model public transit, such as buses, trains and subways (ArcGIS Pro, 2021a). GTFS is a detached specification that supports the network and the performed analysis (ArcGIS Pro, 2021d). GTFS is data consisted of locations of transit lines and stops and actual time schedules. It makes it possible to answer questions like: how well the public transit system serves its riders? How easily can people access

important destinations using public transit? (ArcGIS Pro, 2021a). This tool is essential for this study of the public transit system, in this case, the bus system is crucial for groups like elderly and socio-economic exposed groups in order for them to access other public transit systems.

With GTFS it is possible to model travel time that involves the time for waiting until the next scheduled transit trip. Plus, the travel time along the line segment from one end to the other, plus the time it takes to walk to the final destination. Before this application was available at ArcGIS Pro, analysis of public transit was made with assumptions about travel time in the network. This new adopted specification contributes to network analyzes that represents the reality. Also, this tool contributes to that pre-studies of relocations of facilities that is

dependent on the accessibility of another transportation system, such as railway stations, has

been developed and corresponds to the real-world (ArcGIS Pro, 2021d).



3.2. Data sample & Methodological settings

Secondary data is data that has been collected previously in case of other studies. The advantage of secondary data is that it saves time. Also, the data is often of good quality.

Disadvantages of secondary data are that other researchers are not familiar with the data and its structure (Bryman, 2008). In this study all collected data are secondary data. The data of the road network are gathered from The Swedish Transport Administration (STA) open database, Lastkajen. While climate related data and other variables are a collected from several other authorities, municipalities, and organizations open databases. See table 1.

Climate barriers, such as flooding is gathered from SCALGO Live which estimates flooded areas and precipitation. See table 1. SCALGO Live is a platform that has gathered data about the Earth’s surface through sensors and satellites. The platform provides a snapshot of flooded areas, with the assumption that the ground is saturated on water. With this platform it is possible to increase and decrease rain intensities to be able to identify areas with risks of flooding due to estimation of where water is gathered (https://scalgo.com).

Table 1. Data gathering and source.

Source Data

Geodata Extraction Tool (GET) Population 250m x 250m (Urban area) (2017, 2019) Geodata Extraction Tool (GET) Population 1000m x 1000m (Periphery area) (2017, 2019) Geodata Extraction Tool (GET) Population by income (2017)

Hudiksvall’s municiapality Student commuters (high school)

Lastkajen National road network, Sweden

SCB.se Work commuters

SMHI Future sea levels (average)

SCALGO Live Precipitation

Trafiklab GTFS Public Transit data: Xtrafik

3.2.1. Methodological Choices for the Network Analysis

Each of the attributes of average speed for pedestrians and bicyclists are set according to the literature studies previously in the text (Bedogni et al., 2016; Stenner et al., 2020). These values are presented in table 2.

Because of the significant impact e-bikes have had on the active transportation (Stenner et al., 2020), the decision to include e-bikes as a separate attribute are essential. Therefore, a

separated cost attribute e-bikes is added. Also, this makes it possible to compare the

catchment areas of the two bicycle types as e-bikes has become very common.


29 Table 2. Average speed per travel mode.

Source: Bedogni et al., (2016).

Traveling mode Average speed (km/h)

E-bike 25km/h

Bicycle 20 km/h

Walk 5 km/h

In order to estimate driving time for walking and bicycling, each of the created attribute of the different travel modes is calculated for each road segment, according to this formula.

([Road length in meters]) ÷ ([Assumption of average speed in km/h] × 1000 ÷ 60)

Regarding the pedestrian travel mode, the different cutoffs times are set to 10, 20 and 30 minutes, considering the general rule to have one-way trip length ≤2,5 km for pedestrians (Rabl & de Nazelle, 2012) and with the assumption that we walk 5 km/h (Bedogni et al., 2016). This means that the break point of choosing active transportation before car

transportation goes with 2,5 km or 30 minutes of walking. Therefore, the maximum cutoff time for pedestrians is set to 30 minutes.

The same assumption is used for the bicyclists, which is also based on previous literature from Rabl & de Nazelle (2012) that a one-way journey with bike should be ≤5 km and Bedogni et al. (2016) research that the average speed of a bicyclist is 20 km/h. The settings for the e-bicyclists are based on the same trip length, considering the studies from Stenner et al., (2020) that has shown that the trip length does not increase with e-bikes, but the number of trips do. Anyhow, this means that is takes 15 minutes for a bicyclist to travel 5 km if the average speed is 20 km/h. While it takes 12 minutes to travel 5 km for e-bicyclists if the average speed is 25km/h. Therefore, the cutoff times for the two bicyclist modes is set to 5, 10 and 15 minutes. However, it is essential to also analyze farther distances, which the distance was increased to 7 km for cyclists so therefore a cutoff time of 20 minutes is also added.

Once this has been applied, the GTFS sources are added in the network dataset. Public transit is added for the pedestrian travel mode. Public transit data are excluded with bicycle and e- bike due to that the city bus service in Hudiksvall does not allow bicycles on the busses (X- trafik, 2021). The time used for travel for all the performed analysis is set to Monday 8 a.m.

This means that the person would start the journey in the network 8 a.m, with the assumption

that the person will leave Hudiksvall by train for work or other activities.



3.2.2. Methodological Choices for the Climate Layers

The Swedish Transport Administration has taken a government decision that estimates of climate change should be made for roads and rail networks where climate change can be affected during the facility’s technical lifespan. These analyses done by SMHI are estimated to year 2100, therefore the same time frame for climate analysis of STA’s facilities is adapted (STA, 2020).

STA has internal guidelines adopting climate factors when mapping rain events and floods.

These internal guidelines have come up with a decision to have a climate factor at 1.3

simulating rain with a duration of ≤60 minutes. A climate factor 1.2 is adapted for rain with a duration ≥60 minutes. Implying that a 100-year rainfall with 44 mm with a duration of 60 minutes is added with a climate factor of 1.2 if it rains ≥60 minutes, representing RCP4.5.

Estimating the same 100-year rainfall with a climate factor of 1.3 represent RCP8.5 (STA, 2020).

According to DHIs guidelines of water depth causing problems with accessibility, already at a 10 cm water depth the traffic could be redirected (DHI, 2016). Therefore 10 cm of water depth has been applied in SCALGO Live estimating 100-years rain events. See table 3 for settings in SCALGO Live.

Table 3. Settings in SCALGO Live

Layer Water depth (cm) Rain (mm) Climate factor

RCP- scenario


100-years rain event

10 44 1.2 4.5 0.4 m

100-years rain event

10 44 1.34 8.5 1 m

3.2.3. Restrictions & Barriers in the Network

Restrictions are set in the Network analysis and can be used for certain travel modes. The alternatives of setting the restriction attributes of traversing streets are prohibited, avoided, or preferred (ArcGIS Desktop, 2019). In this case restrictions for pedestrian and bicycle modes are set to prohibited on highways, due to that vehicle with speed below 40 km/h are not allowed (Trafiksäkerhet, 2021).

Other restrictions can be to prefer certain roads. For instance, to access the closest way to the railway station by walk or bike, other ways than pedestrian- and bicycle roads can be faster.

However, in the traffic safety aspect, car roads should be avoided as far as possible. However,

it may be inevitable for pedestrians and bicyclist not to traversing car roads to access the



railway station. Therefore, restrictions for pedestrian and bicyclist are set to prefer pedestrian and bicycle road.

When water is gathered on the streets due to e.g., sea level rise and 100-year rainfall, barriers of water are added to the network analysis. These water barriers are polygons gathered from SCALGO and SMHI. Depending on the analyzed RCP-scenario, different water barriers are added.

3.2.4. Methodological Choices for the Service Area Analysis

To answer the first and the second research question Service area analysis is performed with different settings regarding different travel modes. Service area analysis makes it possible to answer questions like: What areas are within 10 minutes from a railway station? (ArcGIS Pro, 2021c). This is an appropriate analysis to answer these research questions as the catchment area of the three different locations are essential in order to gain understanding of the population’s accessibility of the different locations.

Doing a service area analysis is creating a buffer around a point, with specifications about distance. Although, unlike an ordinary buffer, it represents the maximum distance that can be traveled along the road network. As seen earlier in the method part the cutoff times for pedestrians are set to 10-, 20- and 30 minutes. While for bicyclist it is set to 5-, 10-, 15- and 20 minutes cutoffs. This means that the maximum cutoff time, in this case are therefore 20- or 30 minutes and a buffer is created around the roads that can be reached within that time (ArGIS Pro, 2021c).

Adding the GTFS specification on the Service area analysis, creates the opportunity to combine e.g., the pedestrian travel mode with public transit and model the catchment area of the population. This is significant for this study because it makes it possible to see how well the public transportation system catches the different train stations (ArcGIS Pro, 2021d).

The boundary type for each of the analysis is set to overlap due to that this type creates

individual polygons for each facility and they overlap each other (ArcGIS Pro, 2021c). This is a suitable setting because it makes it possible to visualize the catchment area for each


By setting the parameter “Exclude modes” in the network dataset it is possible to exclude

certain public transit modes (ArcGIS Pro, 2021d). In this case public transit by train is

excluded as buses are the only alternative for local public transport in Hudiksvall.



To answer the second research question the same settings in the Service area analysis is applied with an addition to the climate layers of sea level rise and 100-year rainfalls. To better illustrate that accessibility is deteriorating, the “Exclude lines” parameter is added. This makes is possible to exclude certain bus lines due to that the road e.g., is flooded (ArcGIS Pro, 2021d).

3.2.5. Methodological Choices for Location-allocation Analysis

To answer the third research question a Location-allocation analysis is performed. Location- allocation analysis statistically points out a good location based on the planned expansion and the surroundings (ArcGIS Pro, 2021b). Location-allocation analysis makes it possible to define a set of service facilities within a limited area (Kotavaara, Pohjosenperä & Rusanen, 2018). The goal with location-allocation is to locate the facilities in a way that supplies the demands points most efficiently (ArcGIS Pro, 2021b). Additionally, applying public transit data for pedestrians will result in a time-based Location-allocation to define optimal

catchment areas (Kotavaara et al., 2018). This is an appropriate analysis for the third research question because it can statistically point out a location based on the population. However, a Location-allocation analysis can give multiple solutions based on how many facilities that will be developed, therefore this analysis is a good complement with Service area analysis.

The considered facilities in the Location-allocation analysis are the same locations that STA

has suggested, and the chosen facility is set to one. This means that only one of the three

alternatives can be chosen. Location-allocation analysis has seven different problem types that

answers different kinds of questions. The used problem type for these analyses is Maximize

Coverage. This problem type is used so that so many demand points (the population) as

possible are allocated to solution facilities within the impedance cutoff. This problem type is

useful trying to locate stores or other society services so it can catch most of the people within

a certain drive time (ArcGIS Pro, 2021b). After all, an essential part of locating the railway

station is how many of the population that can reach it within a certain travel time. This

problem type is a good alternative to answer the third research question. The cutoff time is set

to 30 minutes for pedestrians and 20 minutes for bicyclist, due to that these are the maximum

cutoff times used in the Service area analysis.




  1. , https://desktop.arcgis.com/en/arcmap/10.3/guide-books/extensions/network-analyst/understanding-network-attributes.htm#GUID-4BAE3856-
  2. https://desktop.arcgis.com/en/arcmap/latest/extensions/network-analyst/what-is-network-analyst-.htm#GUID-81249557-BEB5-49CC-A393-28FAB997C6EE
  3. data.htm
  4. ro, https://pro.arcgis.com/en/pro-app/latest/help/analysis/networks/location-allocation-analysis-layer.htm
  5. ro, https://pro.arcgis.com/en/pro-app/latest/help/analysis/networks/service-area-analysis-layer.htm
  6. https://pro.arcgis.com/en/pro-app/latest/help/analysis/networks/public-transit-evaluator.htm
  7. https://www.greenbiz.com/article/sdgs-everything-connected
  8. 2541. https://doi.org/10.1002/wcm.2702
  9. https://doi.org/10.1186/s12966-018-0751-8
  10. 312. https://doi.org/10.1080/01944369608975696
  11. http://vav.griffel.net/filer/Rapport_2010-05.pdf
  12. https://infobank.nacka.se/ext/Bo_Bygga/stadsbyggnadsprojekt/Plania%20sydväst/Samråd/Underlag%20-%20Skyfallsanalys.pdf
  13. https://www.lantmateriet.se/contentassets/9161a365093b40b5a583688feacea22c/oversvamningar_vid_skyfall_med_nationell_hojdmodell.pdf
  14. 439. https://doi.org/10.1007/s11067-018-9391-4
  15. https://doi.org/10.1186/1471-2458-11-889.
  16. https://www.regiongavleborg.se/regional-utveckling/om-regional-utveckling/regional-utvecklingsstrategi/
  17. https://doi.org/10.1016/j.rtbm.2012.04.004
  18. . https://doi.org/10.1111/j.0309-1317.2003.00492.x.
  19. , https://www.hudiksvall.se/Sidor/Kommun--politik/Kommunfakta/Hur-manga-bor-det-i-Hudiksvall/Statistik-om-Hudiksvalls-kommun.html
  20. https://www.slu.se/globalassets/ew/org/centrb/cbm/dokument/publikationer-cbm/cbm-skriftserie/bmochklimat.pdf
  21. https://doi.org/10.3390/proceedings2161146
  22. https://doi.org/10.1080/17450101.2016.1253816
  23. https://doi.org/10.2112/SI95-173.1
  24. 54. https://doi.org/10.1080/01441640010022456.
  25. 131. https://doi.org/10.1016/j.tranpol.2011.09.008
  26. https://doi.org/10.1016/j.tra.2003.11.003.
  27. https://www.diva-portal.org/smash/get/diva2:855960/FULLTEXT01.pdf
  28. https://scalgo.com/en-US/scalgo-live-documentation/analysis/flash-flood-map
  29. https://scalgo.com/en-US/scalgo-live-documentation/analysis/sea-level-rise
  30. https://www.scb.se/hitta-statistik/statistik-efter-amne/befolkning/befolkningens-sammansattning/befolkningsstatistik/
Relaterade ämnen :