A Sustainable Future for Wind Energy in Sweden

Master’s thesis

Environmental Management and Physical Planning,

30 Credits

Department of Physical Geography

A Sustainable Future for Wind

Energy in Sweden

Aurora Øvereng

MA 48

2018

Preface

This Master’s thesis is Aurora Øvereng’s degree project in Environmental Management and

Physical Planning at the Department of Physical Geography, Stockholm University. The

Master’s thesis comprises 30 credits (one term of full-time studies).

Supervisor has been Salim Belyazid at the Department of Physical Geography, Stockholm

University. Examiner has been Håkan Berg at the Department of Physical Geography,

Stockholm University.

The author is responsible for the contents of this thesis.

Stockholm, 30 June 2018

Lars-Ove Westerberg

Vice Director of studies

Abstract

The 2040 governmental goal of 100 % renewable electricity in Sweden means that there will be a shift in electricity production and a phasing out of nuclear power. This nuclear power has to be replaced by some other source. Wind power is a viable alternative, thanks to its reliability and the abundance of wind in Sweden. However, wind power production requires a large amount of land and carries the risk of disrupting the landscape. Wind energy is therefore often difficult to develop, and when developed it is often in rural areas where it disturbs as few people as possible. This study presents an alternative to rural exploitation, it investigates whether it is possible to produce sufficient wind power to satisfy urban demand within 20 000 meters of the 20 largest cities in Sweden.

Firstly, the criteria for areas where wind power can be developed were synthesised. Secondly a numerical model was used to simulate energy demand in TWh considering the future growth in demand and the phasing out of nuclear power. The demand for wind power was then translated

into correlating area in km2_{. Finally, a GIS analysis was conducted to estimate the extent of area}

suitable for wind power development based on the criteria above and within a 20 000m perimeter from the 20 largest cities in Sweden. The analysis showed that only 35 % of the required area for wind power development fulfilled the criteria within the given perimeter. From the GIS analysis

only 940.73 km2 _{was found to be suitable. From the numerical model, the results showed that for}

it to be sufficient, there would have to be at least 2687.1 km2 _{suitable land. The conclusion from}

this study is that in order to phase out the nuclear power, the majority of the wind power has to be located in the rural areas.

Key words: urban wind power, sustainability, renewable electricity production, governmental goal

2040, Sweden.

Acknowledgements

I would first like to thank my supervisor Salim Belyazid. For your selfless time, continuous support, trust and belief in me. This thesis would not have been possible without you. It was a real privilege and honor for me to take part of your exceptional knowledge but also your extraordinary human qualities. I am truly ever grateful.

A special thanks to Robert Salmijärvi, for guidance in my GIS work, your patience and knowledge was indispensable.

Last, to my dear friends and family. Thank you for listening in times of stress and despair, for your constructive comments and uplifting words of encouragement. For forcing me out when I needed a break. Thank you.

2

Table of Contents

1. Introduction ... 4

1.1

Aim ... 5

2 Background ... 5

2.1

History of wind energy in Sweden ... 5

2.2

Potential of wind power in Sweden ... 5

2.3

The sustainability of wind power ... 6

2.3.1

Ecological sustainability and wind power ... 6

2.3.2

Social sustainability and wind power ... 7

2.3.3

Economic sustainability ... 8

2.4

Criteria for geographical exclusion of wind power ... 8

2.4.1

Protected nature in Sweden- ... 8

2.4.2

Swedish Armed Forces ... 10

2.4.3

Swedish transport administration ... 11

3 Methods ... 11

3.1

Literature review ... 11

3.2

Numerical modelling ... 11

3.2.1

Scenarios in Stella Architect

... 12

3.3

Geographic Information System ... 12

3.4

Boundaries & Assumptions ... 12

4 Results ... 14

4.1

Criteria for wind power development ... 14

4.2

Simulated future demand for wind power electricity ... 18

4.3

Suitable land in close proximity ... 19

4.4

Mapping land suitability for wind power ... 21

5 Discussion ... 41

6 Conclusions ... 43

4

1. Introduction

The atmospheric level of CO2 in December 2017 was 406.75 ppm, higher than any levels over the

last 800,000 years (Luthi et al., 2008). In 2018, the level is still rising. This increase is due to anthropogenic emissions, mainly from the combustion of fossil fuels for energy (IPCC, 2014).

Global warming, the result of elevated levels of atmospheric CO2, is a threat to the energy security

for all countries and a clear sign that the anthropogenic CO2 emissions need to decrease (ibid.).

Energy security and access to reliable energy have been an essential cornerstone in the world´s economic growth since the industrial revolution. However, the use of energy in the 21st century does not only have to be reliable, it also needs to be sustainable (Chu & Majumdar, 2012). The seriousness of worldwide climate change, air pollution, and energy security requires a substantial and immediate change in the current energy systems. The transformation needs to aim for 100 %

clean, renewable energy producing zero anthropogenic CO2 emissions (Jacobsen et al., 2017).

The effect of a fossil-based energy production, as well as progressing shortage of resources, have already initiated a shift towards renewable energy production throughout the world (Schiebahn

et al., 2015). In Sweden, the energy production system in has dramatically changed in the almost

half-century since the first nuclear reactor started to operate, with consumption more than

doubling from 65 TWh1 _{in 1971 to 150 TWh in 2014 (Hong et al., 2018). Sweden, unlike many}

other countries, has historically chosen to increase the share of nuclear power to become the principal energy source, instead of expanding the use of fossil fuels (Cherp et al., 2016).

Although nuclear power is viewed as a clean energy source and can be economically competitive, it is also associated with risk in both the mining of uranium and in the operations. Nuclear power also produces radioactive waste (Lehtveer & Hedenus, 2014). Nuclear accidents are not common, but when they happen, the effect of a nuclear accident is not only a national issue but often much greater both geographically and temporally. The consequences are long-term, often several generations. Because of this association of risk and the danger it can be, nuclear power is not seen as a sustainable power source (Hasegawa et al., 2015).

Nuclear power in Sweden is facing an uncertain future due to political discourse aiming to phase out nuclear power over the next decades (Hong et al., 2018). The International Energy Agency (2016) set the framework for Nordic energy production up until 2050. In this framework, the decommissioning of nuclear power plants is included, due to the fact that nuclear power is not viewed as a renewable or sustainable energy source (ibid.).

In 2015, most of the world leaders met in Paris to reach an agreement on how to deal with the issue of global warming. The result of the Paris Agreement was a consensus that the global temperature increase should not exceed 1.5 degrees Celsius above recent historical levels. To

achieve this goal, the sovereign nations of the world will need to reduce their CO2 emissions. In

the same year, the Swedish Government created a parliamentary commission, the Energy Commission, whose task was to propose new policies for the long-term energy supply. The Energy Commission proposed the goal of reaching 100 % renewable electricity production by the year

5

2040 (Government Offices of Sweden, 2015:25). This goal, was in 2016 endorsed and put into the framework by the Cross-Party Committee on Environmental Objectives (Cross-Party Energy Agreement, 2016).

With the goal of 100 % renewable electricity production by 2040, wind power production will continue to increase as it is the most mature renewable technique, with the exception of hydropower, with a strong commercial prospect. Today, wind power is mostly constructed in a large-scale format. The developers seek to optimize area usage and profits. This means that in the future wind power has the potential to become a primary source of electricity (Leung & Yang, 2012).

Leung and Yang (2012) state that wind power is environmentally benign. However, they do not consider the amount of space and road construction and the social impacts for wind power to become the primary major source of electricity. One way to deal with the environmental issues and to achieve a higher level of social justice in electricity production is to place the new wind turbines in areas that already exploited. In areas with lower ecological values and in the landscape of the people actually demanding the electricity.

1.1 Aim

The aim of this study is to investigate whether it is possible to sufficiently expand wind power in close proximity to the largest urban areas in Sweden to satisfy their energy demand. The study therefore evaluates the possibility to reach the governmental goal of renewable energy by 2040 with minimal social-environmental impact.

2 Background

2.1 History of wind energy in Sweden

In 1997, there were approximately 250 wind turbines in Sweden (Swedish energy agency, 2009). At the start of 2017, there were more than 3 300 (Swedenergy, 2017a). The year 2017 was a record year for produced wind power in Sweden with 17.6 TWh, which translates to about 11 % of the total electricity production (Swedenergy, 2017b).

Today, there is a little over 1.5 % land, or 7 886.8 km2_{, reserved as national interest sites for wind}

power in Sweden, including areas in the Baltic Sea (Swedish Energy Agency, 2017a). A national interest site for wind power needs to fulfil certain criteria: Wind conditions needs to be a yearly average of 7 m/s or more at 100 meters above the ground. Wind turbines cannot be closer than

800 meters to houses and churches, and the area needs to be 5 km2 _{or more (ibid.). The national}

interest sites for wind power in Sweden are predominantly located either in the north or in rural areas (Pasqualetti et al., 2002; Swedish Energy Agency, 2013; Swedish Energy Agency, 2015). However, the largest end-use areas are the big cities, where the majority of the population resides.

2.2 Potential of wind power in Sweden

In Sweden, wind power stands for 10.59 % of the total energy production. In comparison, hydropower stands for 40.47 % of electricity production, 39.70 % comes from nuclear power, 9.59 % from heat power and 0.09 % from solar power (SCB, 2016). In Sweden hydropower is up

6

to it limits at 65 TWh (Vattenfall, 2017) and solar power is not optimal if the aim is to have a stable flow of reliable electricity throughout the year.

Jacobsen et al. (2017) estimate that the average wind power per country, both onshore and offshore, could be approximately 35 % of the total energy production. However, the Royal Swedish Academy of Engineering Sciences (IVA) states that wind power could technically supply up to 100 TWh, 62 % of the total electricity production in Sweden (IVA, 2016). In a scenario where wind power was to replace 70 % of the nuclear power, which translates to ∼ 42 TWh, that would be less than half of what is technically possible.

2.3 The sustainability of wind power

2.3.1 Ecological sustainability and wind power

‘the impacts of wind turbines on our environment have not been well-established and remain under

debate’ (Leung & Yang, 2012:1036).

Wind energy is crucial for the renewable-based energy systems and an indispensable part of today´s world of electricity. In many countries, it stands for a substantial part of the total production, e.g. Denmark (40%) and Germany (15%). Although it is indispensable, it can also be a trigger for spatial conflicts. The possibilities for ground-based wind power are constrained in many aspects, such as restrictions linked to environmental issues and areas of nature or culture conservation and social acceptance (Lunney et al., 2016: Eichhorn et al., 2017). Shifting energy production from nuclear to wind that collects energy from a larger spatial area will involve

trade-offs. Wind energy production will result in reduced CO2 emissions, but it will also result in broader

terrestrial and aerial impacts (McDonald et al., 2009).

The main advantages of wind power are that it is virtually infinite, it does not produce any pollution, toxic or radioactive waste in its operation and it needs no fuel. In the production of a wind power turbine, there are some pollution and GHG emissions (Jaber, 2013). Tremeac and Meunier (2009) performed a life cycle assessment (LCA) on a 4.5 MW wind turbine to evaluate its environmental impact. To evaluate this, they used indexes such as the energy payback time

(EPBT) and the CO2 equivalent emissions. The assessment showed that wind energy is an

environmentally good solution if the turbines are highly efficient, the end user sites are where wind resources are good, the transportation is limited in its energy use and if the decommissioning is performed correctly. Vestas (2006) did a LCA of their own V90-3.3 onshore

3.0 MW wind turbine. The results showed emissions of 4.64 CO2e per kWh. Compared to the

European average electricity emission, which is 548 CO2e per kWh, it is evident that wind power

has a much lower environmental impact, at least in CO2 equivalent emissions.

There is an ecologic environmental opposition to the large-scale development of wind power in Sweden, which focuses mainly on the audio-visual impacts on unexploited nature. This opposition is criticised by Anshelm and Simon (2015), according to them the rationale behind building wind power on a large scale is its sustainability. Wind power does not threaten any animal species, it does not contribute to climate change, it is not associated with resources depletion and it does not pollute the environment (Anshelm & Simon, 2015). Sweden´s other main energy sources (nuclear power, fuel combustion, biomass, hydropower) are associated with the environmental effects

7

above. Therefore, the argumentation against these types of energy will also be the argument in defence for developing wind power on a large scale.

The disadvantage of wind power is that it requires land clearing, road construction, grid development, causes noise pollution and it disturbs the current landscape. Another disadvantage which is not caused by wind power but important to notice is the fact that the best wind conditions often are located in remote areas, far away from the cities where the energy is needed and consumed (Jaber, 2013).

2.3.1.1 Birds, bats and wind power

Wind turbines are for the most part a bigger issue for bats than for birds. The bats killed by wind turbines are almost only the species that fly above tree height, and in this context are therefore called high-risk species. Because of this, it is important to accommodate these species when planning for wind power location (Rydell et al., 2017).

The average number of birds killed per wind turbine per year varies but is estimated to be between 5 and 10. While some wind turbines kill almost no birds’, others can kill up to 60 birds per wind turbine per year depending on the placement of the turbine. Wetland areas are where most birds are killed by wind turbines, because this is often where birds breed and migrate to (ibid.).

2.3.2 Social sustainability and wind power

The challenges of energy security and climate change are enormous. The role of spatial planning in adapting to future climate conditions and mitigating their most extreme impacts is substantial. The government is responsible for supplying reliable electricity and, in the effort of reducing the reliance on fossil fuels, renewable electricity (Ellis et al., 2009). Governmental targets have given rise to an increasing interest and development of wind energy. In Sweden, approximately 59 % support an increased development of wind energy (SOM-institute, 2013). However, this is not reflected in the pace or amount of wind energy developed.

Bell et al., (2005) addressed the issue they refer to as the ‘social gap’ between high levels of support for wind and wind power installed. They believe that this gap is due to the opposition towards altering or developing areas that are today seen as valuable in a landscape view. To solve this ‘social gap’ Peel and Lloyd (2007) suggests planning for wind power in an urban setting. They state that planning for wind power in urban areas differs from doing so in rural areas. They are dissimilar in that the urban landscape is not valued as high visually or in terms of tranquillity. Also, in urban areas, there are already ongoing impacts from development and exploitation (ibid.). One of the biggest issues regarding wind power is location and restrictions on land use. Wind power has in many ways been presented as an interference in our landscape. Because it is seen in competition with aesthetic that change our surroundings and instead creates a landscape of power (Pasqualetti et al., 2002). In this regard, there is a conflict of interest. For the most part, because of their perceived disturbance to the landscape, wind turbines have been placed either out at sea, far away from any landscape view or in rural areas where they visually impact as few people as possible (ibid.).

8

For wind power to be a sustainable energy source it is important that there is a social acceptance for it. Throughout the literature, the different aspects of social acceptance have mainly focused on rural communities instead of focusing on the urban environments (Khorsand et al., 2015). Most wind power development occurs in rural areas; this is also the case in Sweden (Pasqualetti et al., 2002). The conflict happens when urban developers decide to build in rural landscapes despite the wishes of the local rural community to keep the area unexploited. An uneven power structure is then developed where the rural residents lose the right to decide what happens in their close proximity, and the feeling of place identity is challenged. On the other hand, if rural inhabitants want to develop unexploited areas they often face critic and opposition from urban dwellers, whose wishes are to keep these areas as recreational spaces. The landscape is then viewed as pristine, in need of protection and as an area of ‘wilderness’ quality. The rural residents will generally resent the urban dwellers for not allowing economic growth to the local community because they want to keep it ‘wild’ and the urban dwellers will, in turn, resent the rural residents for ‘destroying’ the ‘wilderness’ and their recreational attractions (ibid.).

2.3.3 Economic sustainability

The conflict between wind power and nature conservation is like no other conflict because both sides want to protect the environment. Yet, neither of these side focus on a conflict between the nature conservation and economic growth. This might be because no such conflict exists. With that said, that does not mean that there is no economic gain or interest in wind power development (Ellis et al., 2009).

Wind farms and investment give rise to ‘green’ jobs and provides prospects for economic development (Edjemo & Söderholm, 2015). According to the Energy Commission (2016) if wind power is measured against other newly developed electricity sources it is economically competitive due to the developments in technology and reduction of cost. However, older production facilities will have lower costs and be more profitable. The Swedish Energy Agency (2016) created a production cost report that showed that 50 TWh of land-based wind power can be profitable if the cost per kWh is between 0.40 and 0.50 SEK. However, this is only after 12- 15 years after installation (Vindkraftig, 2018).

2.4 Criteria for geographical exclusion of wind power

2.4.1 Protected nature in Sweden-

About 11.1 % or 55 121 km2 _{of Sweden is protected through the status of either national park,}

nature reserve, nature protection area or biotope protection area (Environmental Protection

Agency, 2018a). About 60 000 km2 _{are protected through the Natura2000 directives and as stated}

earlier, this makes it challenging to find areas for wind power development (Environmental Protection Agency, 2018b).

Biosphere area

There is no legislation which includes the biosphere areas in Sweden. These areas are a part of UNESCO’s program ‘Man and the Biosphere’. They are created with an aim of being modules for how to reach the UN sustainability goals: Agenda 2030 (Environmental Protection Agency, 2018c).

9

The aim of the RAMSAR convention (1971) was to mainly protect valuable wetlands. Today, the focus is on birds and the protection of breeding and migration sites. In Sweden, none of RAMSAR sites are also under any type of legal protection against exploitation. However, these sites are located in areas which are either classed as a Natura2000, national park or nature reserve (Environmental Protection Agency, 2018d).

UNESCO World Heritage Site

There is no legislation in Sweden which includes the UNESCO World Heritage sites. However, these areas are often also covered by either a national park protection or nature reserve protection, and thereby protected by chapter 7 § 2-3 and § 4-8 of the Swedish Environmental Code (SFS 1998:808). There area today 15 of these sites in Sweden. Only three have UNESCO status because of their natural values, ‘Laponia’ and ‘Höga Kusten’ and ‘Södra Ölands odlingslandskap’ (Environmental Protection Agency, 2018e).

Natura2000

An EU based network of protected areas which includes the bird protection directive (Directive 79/409/EEG) and the species and habitat protection directive (Directive 92/43/EEG). In Sweden, these areas are incorporated in the Swedish Environmental Code (SFS 1998:808) and SFS 1998:1252 about area protection in accordance to the Swedish Environmental Code.

Nature protection area

Before the Swedish Environmental Code (SFS 1998:808), there was the Nature Protection Law (SFS 1964:822). In this legislation, there were areas created to protect the landscape and are therefore often located in coastal areas, called nature protection areas. After the Swedish Environmental Code was written the possibility of creating more natural protection areas expired. The areas remaining should now be considered as nature reserves according to SFS 1998:811 chapter § 2 about the insertion of areas in the Swedish Environmental Code (SFS 1998:808).

National Park

There are 29 national parks in Sweden. They are created by the Government and the County Administration Boards. These differ from nature reserves (and other) protection forms, in that the state owns the land. They are created to preserve what is both typical and unique in Sweden. The status of a national park is the strictest protection in Sweden and actions of development are not permitted. National parks are protected by the Swedish Environmental Code chapter 7 §2-3 (SFS 1998:808).

Nature reserve

Nature reserves are created for the purpose of preservation of the biological diversity, valuable natural environments and to complement the need for recreational space. In Sweden, there are approximately 4 500 nature reserves. Nature reserves are protected by the Swedish Environmental Code chapter 7 § 4-8(SFS 1998:808).

Access order area

Areas within national parks where there are restrictions on where people can move. This is either permanent or seasonal, depending on the reason for the access order. These areas are included in the Swedish Environmental Code chapter 7 § 5 (SFS 1998:808).

10

The purpose of these areas is to protect cultural heritage sites or landscapes. The areas are protected by the Swedish Environmental Code chapter 7 § 9 (SFS 1998:808).

Water protection area

Water protection areas are covered by the Swedish Environmental Code chapter 7 § 21-22 (SFS 1998:808). These areas are groundwater and surface water bodies which needs protection for water security reasons, not for nature protection.

Natural monument

Single monuments that for different reasons, e.g. biological diversity, needs protection from exogenous or anthropogenic impacts. Natural monuments are protected by the Swedish Environmental Code chapter 7 § 10 (SFS 1998:808).

Animal and plant protection area

These areas are mainly located by coastal areas or lakes. They are created with the purpose to protect waterfowl and/or seals. This protection can be constant or seasonal depending on the purpose. These areas are protected by the Swedish Environmental Code chapter 7 § 12 (SFS 1998:808).

Biotope protection

This protection form is only used for smaller habitats for nationally threatened species and/or smaller nature types found in agricultural or forest areas. Biotope areas are protected by the Swedish Environmental Code chapter 7 § 11 (SFS 1998:808).

Coastal protection area

The coastal protected areas are protected through the Swedish Environmental Code chapter 7 § 13-18 (SFS 1998:808). The protected area is at a default at 100 meters from the shoreline, in both directions. This can, however, be expanded to 300 meters if the County Board decides it appropriately.

Landscape protection

The landscape protection is an older form of protection from the Nature Conservation Act §19 (1964:822) which is not included in the Swedish Environmental Code. However, the protection is still a judicial one, according to SFS 1998:811 about the insertion into the Swedish Environmental Code.

Key habitat

Forested areas with high natural value. These areas play a vital role in the protection of the forests red-listed and other threatened species. The areas are however not protected in any legal form (Swedish Forest Agency, 2018).

2.4.2 Swedish Armed Forces

In Sweden, the Swedish Armed Forces have a major say in where wind power can be placed. In 2010, they re-established their claim to hinder the development in areas that are close to their airports or other bases. The Swedish Armed Forces require remittance for all taller objects. If a

11

wind turbine is installed in a location outside an area of connected buildings, the remittance is required when that object is over 20 meters. In locations inside areas of connected buildings, the remittance is required when the object over 45 meters (Swedish Armed Forces, 2018). Areas protected by the Swedish Environmental Code chapter 3 §9 (SFS 1998:808) from exploitation to protect the total defence are military shooting areas, military training area, and military protection

areas. In these areas, it is not possible to develop any wind power. Minimum Sector Altitude (MSA) area

This is an area surrounding all airports/bases where taller objects are not permitted. This area has a radius of 55 km centred around the radio and visual navigation aid of the airport/base (Swedish transport administration, 2015).

2.4.3 Swedish transport administration

Roads and railways

The distance between a wind turbine and a public road or railway should be the total height (tower height + half the rotor blade diameter) plus 20 meters, however, it should always be above 50 meters (Swedish transport administration, 2017).

3 Methods

3.1 Literature review

The first step in this study was to research peer-reviewed articles to put the aim in a larger scientific framework. The databases used to perform this research were: Web of Science, Google Scholar, and EBSCO Discovery Services. The search words used were: wind power, sustainability,

wind power + ecological sustainability, wind power + social sustainability, wind power + economic sustainability, life cycle assessment + wind power, wind power+ social justice, sustainable energy, energy production, nuclear power + sustainability. The second search was done on ‘grey’ literature.

This part focused on Swedish legislation and reports from different governmental agencies in regard to where it is possible to build wind power in Sweden.

3.2 Numerical modelling

In the fall of 2017, I was part of a student group who created a numerical model which simulated the electricity production in Sweden from 1990 to 2050. I used this numerical model in this study to calculate the amount of TWh and number of turbines needed in order to phase out nuclear power and install sufficient wind energy. To be able to find the sufficient area required for wind power development in close proximity to the urban centres, it was necessary to first use the model

to find the amount of demanded wind TWh and use that to find the correlating area (km2_).

Some of the assumptions made in the original numerical model are here revised as follows: (1) The average yearly running hours of a wind turbine has been changed from 1840 hours to

2500 hours (Swedish Energy Agency, 2014).

(2) The amount of megawatt per wind turbine has been changed from 2 to 3 MW/wind turbine (IVA, 2016)

12

(3) Number of turbines per km2 _{was changed from 3 to 3.5 (calculated by author, based on}

assumption ‘e’).

3.2.1 Scenarios in Stella Architect

Four scenarios of the extend of nuclear power to be replaced by wind power were tested, to

capture the difference in area (km2_{) needed to produce the corresponding energy (TWh).}

Scenario A

What if 100% of the phased out nuclear power is replaced by wind power?

Scenario B

What if 70 % of the phased out nuclear power is replaced by wind power?

Scenario C

What if 40 % of the phased out nuclear power is replaced by wind power?

Scenario D

What if 10 % of the phased out nuclear power is replaced by wind power?

3.3 Geographic Information System

GIS was used to evaluate the placement of the needed wind turbines and to find out how much area is suitable for wind power in close proximity to the 20 biggest cities in Sweden. The GIS program used in this study was ArcMap 10.5. The data was gathered from ‘Skyddad Natur’, a database from the Environmental Protection Agency, the database ‘GET SLU’ from the Swedish University of Agricultural Sciences, Lantmäteriet database ‘Open data’ and the county administration boards database ‘webbGIS’.

The wind condition data was collected from the Swedish Energy Agency’s homepage where I could directly download the shape-file. The wind data was from 2011 and calculated with the MIUU-model.

3.4

Boundaries & Assumptions

The study focuses on the twenty largest cities4 _{by population in Sweden. The cities considered in}

this study are: Borås, Eskilstuna, Gälve, Göteborg, Halmstad, Helsingborg, Jönköping, Karlstad, Linköping, Lund, Malmö, Norrköping, Stockholm, Södertälje, Umeå, Upplands-Väsby, Uppsala, Västerås, Växjö and Örebro. Together, they account for 39% of the total Swedish population, and we assume that they account for an equal proportion of the Swedish energy demand. Close proximity is defined as within a 20 000 m perimeter from the border of the urban centres mentioned above.

13

a. This study does not consider whether areas that toady are considered suitable will be urbanised in year 2040.

b. The wind turbines used in this study have a capacity of 3 MW and a total height (tower + half the blade diameter) of 200 meters. The total height determines how large the buffer zone from the roads and railway needs to be in order to be within a safety zone (Swedish transport administration, 2017).

c. It is assumed that wind power turbines will not be placed on sites where the yearly wind speed average at 100-meter height is lower than 7 meters per second, even if wind turbines start rotating at about 3 meters per second.

d. It is assumed that each wind turbine needs a radius of 300 meters from wind tower to wind tower (Observation from flight images done by author in Google Earth Pro).

e. The coastal protection area is considered to be 100 meters. f. Only on-shore wind power is considered in this analysis.

g. This study does not consider whether the suitable area is private owned or owned by the state, county or municipality.

h. Roof-mounted wind power is not included

i.

electricity demand. In this study the 20 biggest cities represent 39 % of the population, and therefore, only 39 % of the total need of wind power is expected to be placed within the area considered in this study.

14

4 Results

4.1 Criteria for wind power development

The criteria were created in order to perform the GIS analysis, and are listed in Table 1. Table 1 shows the different names for the geo-types used and a small description of what the geo-types contain. Geo-types describe the characteristics concerning the type of ground properties, protection or land use. The suitability for developing wind power and the databases and sources used are also included in Table 1. The geo-types that are not classified have been used as base maps and were included in the maps for visual reasons only but included from Table 1. In total, 29 geo-types have been identified. Only 7 geo-types are suitable for wind power development, while 22 geo-types were not suitable (Table 1).

The 7 geo-types where it is suitable to build wind power can be grouped in two: the first is ‘protected area’ where no legislation apply. UNESCO sites, Water protection area and Key habitat areas are all geographically defined, however, in their description there is nothing stating that development cannot take place. The second group: Forest, open land, alvar land and bedrock, have been classified on the basis of their physical properties and that it technically is possible to develop wind power on these types of land.

The geo-types that have been classified as unsuitable all have either strict legislation that prevents exploitation e.g. ‘nature reserve’, ground properties where it would not be possible e.g. ‘glacier’. The geo-types also include areas that have been classified unsuitable due to safety reasons e.g. ‘roads’ or areas that are protected by the Swedish Armed Forces e.g. ‘military training area’.

T able 1 . Table of criteri a for w ind pow er d

evelopment. The table lists the

geo types names, d

escription of geo types, possibility for

d evelopin g w ind pow er, d atabases and sources. The ge o type s are n ot liste d in the ir origin al shape -file names, but in t heir d

escriptive names since they hav

e been alt er ed in Ar cM ap 10. 5. G eo ty p es * D es cr ip tio n o f g eo ty p es ** Su ita bi lit y fo r d ev elo p in g w in d p owe r D at ab as e a n d S ou rc e W in d co n d iti on s A ll a re as w he re th e w ind co ndit io ns are ov er 7 m/ s at 10 0 ov er th e grou nd Sui ta ble Sw ed ish E ne rg y A ge nc y (2018a )h ttp :/ /w w w .e ne rg im yn dig he te n.s e/ fo rn yb ar t/ vin dk ra ft/ pla ne rin g-o ch -til lst an d/ vin dk ra ftsp la ne rin g1 /n ati on ell -v in dk ar te rin g/ [a cc esse d 2 0.0 2.1 8] Fo re st A ll f ore ste d a re as in S w ed en Sui ta ble La nt m äte rie t ( G SD -T er rä ng ka rta n, ve kt or . 1 :5 0 0 00 ) ftp :/ /d ow nlo ad -o pe nd ata .la nt m ate rie t.se /[a cc esse d 1 4.0 2.1 8] O p en la n d A ll ope n l and in S w ede n Sui ta ble La nt m äte rie t ( G SD -T er rä ng ka rta n, ve kt or . 1 : 5 0 0 00 ) ftp :/ /d ow nlo ad -o pe nd ata .la nt m ate rie t.se /[a cc esse d 1 4.0 2.1 8] Al va r l an d A ll a lv ar la nd in S w ede n Sui ta ble La nt m äte rie t ( G SD -T er rä ng ka rta n, ve kt or . 1 :5 0 0 00 ) ftp :/ /d ow nlo ad -o pe nd ata .la nt m ate rie t.se /[a cc esse d 1 4.0 2.1 8] Be d ro ck A ll b edr oc k a re as in S w ede n Sui ta ble La nt m äte rie t ( G SD -T er rä ng ka rta n, ve kt or . 1 :5 0 0 00 ) ftp :/ /d ow nlo ad -o pe nd ata .la nt m ate rie t.se /[a cc esse d 1 4.0 2.1 8] K ey h ab ita t p ro te ct io n a re as A ll k ey h ab ita ts in S w ede n Sui ta ble En vir on m en ta l P ro te cti on A ge nc y ( 20 18 f), ‘S ky dd ad n atu r’ ht tp :/ /sk yd da dn atu r.n atu rv ar dsv er ke t.se / [ ac ce sse d 0 5.0 2.1 8] W at er p ro te ct io n ar ea A ll ke y ha bit at prot ec tio n ar ea s in Sw ede n Sui ta ble En vir on m en ta l P ro te cti on A ge nc y ( 20 18 f), ‘S ky dd ad n atu r’ ht tp :/ /sk yd da dn atu r.n atu rv ar dsv er ke t.se / [ ac ce sse d 0 5.0 2.1 8] UN ES CO Th e th re e U N EC SO sit es : ‘H ög a ku ste n’, ‘L apo nia ’ a nd ‘S ödr a Ö la nds odl ing sla nds ka p’ Sui ta ble En vir on m en ta l P ro te cti on A ge nc y ( 20 18 f), ‘S ky dd ad n atu r’ ht tp :/ /sk yd da dn atu r.n atu rv ar dsv er ke t.se /[ ac ce sse d 0 5.0 2.1 8] N at io n al p ar k A ll 2 9 na tio na l pa rk s in S w ede n Un sui ta ble En vir on m en ta l P ro te cti on A ge nc y ( 20 18 f), ‘S ky dd ad n atu r’ ht tp :/ /sk yd da dn atu r.n atu rv ar dsv er ke t.se /[ ac ce sse d 0 5.0 2.1 8] N at u re re se rv es A ll na tu re re se rve s and na tu ra l mon ume nt s in S w ede n Un sui ta ble En vir on m en ta l P ro te cti on A ge nc y ( 20 18 f), ‘S ky dd ad n atu r’ ht tp :/ /sk yd da dn atu r.n atu rv ar dsv er ke t.se /[ ac ce sse d 0 5.0 2.1 8] N at u re p ro te ct io n ar ea s A ll re ma inin g na tu re prot ec tio n are as in S w ede n Un sui ta ble En vir on m en ta l P ro te cti on A ge nc y ( 20 18 f), ‘S ky dd ad n atu r’ ht tp :/ /sk yd da dn atu r.n atu rv ar dsv er ke t.se /[ ac ce sse d 0 5.0 2.1 8] Cu ltu ra l Co n se rv at io n si te s A ll cu ltu re co ns erv atio n sit es in Sw ede n Un sui ta ble En vir on m en ta l P ro te cti on A ge nc y ( 20 18 f), ‘S ky dd ad n atu r’ ht tp :/ /sk yd da dn atu r.n atu rv ar dsv er ke t.se /[ ac ce sse d 0 5.0 2.1 8] A n im al an d s p ec ie s p ro te ct io n a re a A ll anima l and spe cie s prot ec tio n are as in S w ede n Un sui ta ble En vir on m en ta l P ro te cti on A ge nc y ( 20 18 f), ‘S ky dd ad n atu r’ ht tp :/ /sk yd da dn atu r.n atu rv ar dsv er ke t.se /[ ac ce sse d 0 5.0 2.1 8] Ac ce ss o rd er a re a A ll ac ce ss a re as in Sw ede n, e xc ept Ö re bro C ou nt y Un sui ta ble En vir on m en ta l P ro te cti on A ge nc y( 20 18 f), ‘Sk yd da d n atu r’ ht tp :/ /sk yd da dn atu r.n atu rv ar dsv er ke t.se /[ ac ce sse d 0 5.0 2.1 8]

15

La n d sc ap e p ro te ct io n a re a La nds ca pe prot ec tio n are as in D ala rn a co unt y, Gä vle bo rg c ou nt y, Jä mt la nd co unt y, V äs te rno rr la nd co unt y, U pps ala co unt y, H all and co unt y, Sto ck ho lm co unt y, Sk åne co unt y, K ron ob erg co unt y and V äs tra Göt ala nd co unt y. Un sui ta ble Co un ty A dm in ist riv e B oa rd (2 01 8) D ata ba se ‘w eb bG IS’ ht tp :/ /e xtr a.l an sst yr else n.se /g is/ Sv /P ag es/ ka rtt ja nst er .a sp x [a cc esse d 2 0.0 2.1 8] R AMS AR a re a A ll R A M SA R a re as in S w ed en Un sui ta ble En vir on m en ta l P ro te cti on A ge nc y ( 20 18 f), ‘S ky dd ad n atu r’ ht tp :/ /sk yd da dn atu r.n atu rv ar dsv er ke t.se /[ ac ce sse d 0 5.0 2.1 8] H ab ita t d ir ec tiv e A ll are as co ve re d by D ir ec tiv e 92 /43 /E EG in S w ed en Un sui ta ble En vir on m en ta l P ro te cti on A ge nc y ( 20 18 f), ‘S ky dd ad n atu r’ ht tp :/ /sk yd da dn atu r.n atu rv ar dsv er ke t.se /[ ac ce sse d 0 5.0 2.1 8] Bi rd s d ir ec tiv e A ll are as co ve re d by D ir ec tiv e 79 /40 9/ EE G i n S w ed en Un sui ta ble En vir on m en ta l P ro te cti on A ge nc y ( 20 18 f), ‘S ky dd ad n atu r’ ht tp :/ /sk yd da dn atu r.n atu rv ar dsv er ke t.se /[ ac ce sse d 0 5.0 2.1 8] M SA -a re as A ll M SA -a re as in S w ed en Un sui ta ble En vir on m en ta l P ro te cti on A ge nc y ( 20 18 f), ‘S ky dd ad n atu r’ ht tp :/ /sk yd da dn atu r.n atu rv ar dsv er ke t.se /[ ac ce sse d 0 5.0 2.1 8] M ili ta ry tr ai n in g ar ea s A ll m ilit ary t ra inin g a re as in S w ede n Un sui ta ble En vir on m en ta l P ro te cti on A ge nc y ( 20 18 f), ‘S ky dd ad n atu r’ ht tp :/ /sk yd da dn atu r.n atu rv ar dsv er ke t.se /[ ac ce sse d 0 5.0 2.1 8] M ili ta ry sh oo tin g ar ea s A ll mil it ary sh oo ting are as in Sw ede n Un sui ta ble En vir on m en ta l P ro te cti on A ge nc y ( 20 18 f), ‘S ky dd ad n atu r’ ht tp :/ /sk yd da dn atu r.n atu rv ar dsv er ke t.se /[ ac ce sse d 0 5.0 2.1 8] M ili ta ry p ro te ct io n ar ea s A ll mil it ary prot ec tio n are as in Sw ede n Un sui ta ble En vir on m en ta l P ro te cti on A ge nc y ( 20 18 f), ‘S ky dd ad n atu r’ ht tp :/ /sk yd da dn atu r.n atu rv ar dsv er ke t.se /[ ac ce sse d 0 5.0 2.1 8] R oa d s’ b u ffe r z on e Pu blic roa d > 5 m, < 7 m, 5 -7m. , H ig hw ay > 5 m, < 7 m, 5-7m. , M ot orw ay > 5 m, < 7 m, 5 -7m. , A cc es s roa d > 5 m, < 7 m , 5 -7 m, a nd S tre et > 5 m, < 7 m, 5 -7m. A ll w it h a b uff er o f 22 0 me te rs Un sui ta ble La nt m äte rie t ( G SD -V äg ka rta n, ve kt or 1:1 0 0 00 ) f tp :/ /d ow nlo ad -op en da ta .la nt m ate rie t.se /[a cc esse d 1 4.0 2.1 8] R ai lwa y b u ffe r zo n es A ll ra ilw ay tr ac ks in Sw ed en. A ll w it h a b uff er o f 22 0 m ete rs . Un sui ta ble Sw ed ish U niv er si ty o f A gr icu ltu ra l S cie nc es ’G ET S LU ’ ( G SD -V äg ka rta n, ve kt or 1:1 0 0 00 ) h ttp s: // ze us. sl u.se /g et/ ?d ro p= [a cc esse d 1 4.0 2.1 8] G la cie rs A ll g la cie rs in S w ede n Un sui ta ble La nt m äte rie t ( G SD -T er rä ng ka rta n, ve kt or . 1 :5 0 0 00 ) ftp :/ /d ow nlo ad -o pe nd ata .la nt m ate rie t.se /[a cc esse d 1 4.0 2.1 8] Fe n la n d A ll f en l and are as in S w ede n Un sui ta ble La nt m äte rie t ( G SD -T er rä ng ka rta n, ve kt or . 1 :5 0 0 00 ) ftp :/ /d ow nlo ad -o pe nd ata .la nt m ate rie t.se /[a cc esse d 1 4.0 2.1 8] Bui ld in gs A ll h ou se s a nd c hu rc he s in S w ede n, ex ce pt fo r sma lle r err ors in c lo se Un sui ta ble La nt m äte rie t ( G SD -Fa st ig he tsk ar ta ns to po gr afi , v ek to r 1 :1 0 0 00 ) ftp :/ /d ow nlo ad -o pe nd ata .la nt m ate rie t.se /[a cc esse d 1 4.0 2.1 8]

16

prox imit y to H alms ta d a nd H els ing bo rg . A ll h ou se s a nd ch urc he s h av e a b uff er zo ne of 80 0 me te rs Ur b an ce n tr e W it h a m ult ipl e r in g b uff er of 50 00 me te rs , 1 0 00 0 me te rs , 1 5 00 0 me te rs a nd 20 0 00 m ete rs N ot cl as sif ie d La nt m äte rie t(G SD -Fa st ig he tsk ar ta ns to po gr afi , v ek to r 1 :1 0 0 00 ) ftp: //dow nl oa d-op enda ta.l ant m ate rie t.s e/[ ac ce sse d 1 4.0 2.1 8] W at er c ou rs es a n d la k es A ll la ke s and w ate rc ou rs es in Sw ede n N ot cl as sif ie d La nt m äte rie t ( G SD -T er rä ng ka rta n, ve kt or . 1 :5 0 0 00 ) ftp :/ /d ow nlo ad -o pe nd ata .la nt m ate rie t.se /[a cc esse d 1 4.0 2.1 8] Swe d en b as e m ap N atio na l and te rr it oria l w ate r bo rd ers N ot cl as sif ie d La nt m äte rie t ( G SD -T er rä ng ka rta n, ve kt or . 1 :5 0 0 00 ) ftp: //dow nl oa d-op enda ta.l ant m ate rie t.s e/ [a cc esse d 1 4.0 2.1 8] *T ype s of la nd c ov er us ed in de te rm inin g s uit ab ilit y o f w ind po w er de ve lo pm ent ** F or a de ta ile d d es criptio n o f g eo -ty pe s, s ee s ec tio n 2. 4. 1 ab ov e

17

18

4.2 Simulated future demand for wind power electricity

Figure 1 shows the total energy demand for wind power in Sweden between 1990 and 2040 for the four scenarios of substituting nuclear power with wind power. All scenarios show an increase, driven by the expected increase in population. The higher substitution rates produce higher demand for wind energy. The total demand for wind energy is expected to reach between 181 TWh and 75.5 TWh under scenarios A and D, respectively. However, in this study only 39 % of the demanded wind energy is required for 39 % of the population, therefore the reach is 29.45 TWh and 70.59 TWh. Figure 2 shows the correlating area required for the wind energy demand. Figure 2, as Figure 1, shows the total area, but also here the reach that is considered is only 39 %.

Figure 1. Required wind energy (TWh) for decommissioning nuclear power. The graph is driven by the increase in population and demand. The green line is scenario A (100 %), the pink line is scenario B (70 %), the red line is scenario C (40 %) and the blue line is scenario D (10 %).

Figure 2. Required area (km2_{) for correlating wind energy. The graph is driven by the energy}

demand (Figure 1). The green line is scenario A (100 %), the pink line is scenario B (70 %), the red line is scenario C (40 %) and the blue line is scenario D (10 %).

19

In Table 2, both the demanded TWh and the area (km2_{) is listed. In Table 2, 39 % of the total have}

been calculated and it only shows the minimum (Scenario D) and maximum (Scenario A). In Table

2, both the area (km2_{) and the demanded TWh is displayed in 2040 and in 2050. In this study the}

focus is on the year 2040, but because the model already runs until 2050, this is also listed in Table 2.

Table 2. Results from the numerical model. The table lists 39 % of the total from Figures 1 and 2. Both wind power demand

in TWh and correlating area in km2 _{is presented. The TWh and correlating area is listed in minimum (Scenario D) and}

maximum (Scenario A).

39 % of the minimum and maximum in Figures 1 and 2 is shown in Table 2. In 2040, as a minimum, there will be a wind power demand of 29.45 TWh and this will require 1123.2 km2. Scenario A, which is 100 % replacement of nuclear power with wind power. The demand in 2040

is 85.8 TWh and this requires 3276 km2_{. In 2050, the minimum wind power demand is 32.64 TWh}

and the maximum is 85.8 TWh. The correlating area is 1244.1 km2 _{and 3276 km}2_{, respectively.}

In Figure 1, the total required area from 2040 is shown. The movements in the lines are determined by the decommissioning of nuclear reactors, set by a time interval in the model. The blue line represents wind power replacing 10 % of the nuclear power, the red line is 40 % of the nuclear power being replaced by wind power, the pink line represents 70 % and the green line represents wind power replacing 100 % of the nuclear power. The same colour –scheme is used in Figure 2.

4.3 Suitable land in close proximity

In Table 3, the results from each city and the possible geo types are shown. The results from the GIS analysis show that the total area suitable for wind power from all cities amounts to 940.72

km2_{. The three largest cities; Stockholm, Göteborg and Malmö, have a total of suitable land in}

their close proximity 65.90 km2_{, 15.73km}2 _{and 6.28 km}2 _{respectively. The cities with most}

suitable land are: Helsingborg with 140.15 km2_{, Halmstad with 118.24 km}2_{, Gävle with 114.66}

km2 _{and Umeå with 95.89 km}2_.

The geo type which has the most suitable land is forested area, followed by open land, water

protection area, key biotopes and biosphere area. The total sum of suitable land is 940.72 km2

(Table 3). Year

Wind power demand (TWh) Correlating area (km2₎

Min (Scenario D) Max (Scenario A) Min (Scenario D) Max (Scenario A)

2040 29.45 70.59 1123.2 2687.1

Table 3 .Suitable land in close proximity . Table 3 show s all the cities consid ere d in this stu d y an d how much area ( km 2

) each city has

that i s suitab le. It also s how s ho w much w ould be required from each c ity in regard to scenario A an d scenario D . The tabl e s ho w s ho w

the suitable area d

ivid ed per geo -type. G eo -ty p e Ci ty n am e Su itabl e ar ea fo r win d po we r , al l ge o ty pe s in clu de d (km 2 ) R eq uir ed ar ea (km 2 ) Sc en ar io A (10 0 % ) 2040 R eq uir ed ar ea (km 2 ) Sc en ar io D (1 0 % ) 2040 Su ita ble fo r w in d in F ore st are as (km )₂ O pe n l an d i n (km 2 ) W ate r pro te cti on are a ( km 2 ) K ey b io to pe s i n (km 2 ) Bi os phe re a re a i n (km 2 ) Um eå 95.89 58.71 24.54 93 .6 9 0.5 4 0.6 7 0.9 9 - G äv le 114.66 51.72 21.61 10 5.8 2 0.1 3 5.6 9 0.7 9 2.2 2 St oc kho lm 65.90 1062.67 444.19 62 .7 5 1.9 8 1.1 7 - - M alm ö 6.28 212.25 88.72 3.1 2 3.1 4 - 0.0 3 - G öt eb or g 15.73 401.75 167.93 8.6 1 6.4 5 0.6 7 0.0 2 - H els in gb or g 140.15 73.69 30.80 14 .5 6 12 5.6 0 0.5 6 0.1 9 - V äx jö 23.55 45.84 19.16 21 .4 4 1.8 0 0.1 2 0.1 9 - B or ås 38.18 49.69 20.77 24 .2 9 1.2 5 12 .6 4 0.0 5 - H alm sta d 118.24 46.31 19.36 52 .7 7 61 .3 3 4.1 4 0.5 5 - N or rkö pin g 42.65 65.75 27.48 37 .1 1 2.0 9 3.1 1 0.3 4 - Lun d 6.28 61.96 25.90 3.6 0 1.9 6 0.7 1 0.0 1 - Jö nkö pin g 45.32 65.37 27.32 44 .5 8 0.4 4 0.1 3 0.1 8 - Sö de rtä lje 23.78 49.45 20.67 22 .3 2 0.7 1 0.7 6 - - K ar lst ad 5.68 42.71 17.85 4.6 9 0.9 3 0.0 7 - - Up pla nd s V äs by 46.62 97.67 40.82 39 .9 5 6.2 7 0.4 1 - - Es kils tun a 46.86 46.94 19.62 42 .5 8 4.1 9 0.0 9 - - Ö re br o 53.53 81.01 33.86 32 .3 2 21 .1 0 - - - V äs te rå s 40.45 82.86 34.63 32 .5 8 7.8 1 0.0 6 - - Up ps ala 10.97 106.70 44.60 10 .0 9 0.8 8 - - - Lin kö pin g - - - - - - - - T ot al su m 940.72 - - 65 6.8 7 24 0.6 0 31 .0 0 3.3 4 2.2 2

20

21

The results from Table 3, shows that four cities: Umeå, Gävle, Helsingborg and Halmstad, have the area required to decommission nuclear power and exchange it for wind power to support the % of the population residing in these cities. Uppsala, Karlstad, Stockholm, Malmö and Göteborg does not have the required area to decommission any nuclear power. The majority of the cities: Växjö, Borås, Norrköping, Jönköping, Södertälje, Örebro and Västerås, have the area required to decommission 10 % or more of the nuclear power and replace it wind power.

If the cities are combined to one sum, 940.73 km2_{, then there is not sufficient area for either}

scenario A, B, C or D, for as Table 2 show the minimum requirement for area in 2040 is 1123.2 km2 and 1244.1 km2 in 2050.

With the total sum of 940.73 km2, the total energy production would be 2.01 TWh with average running hours being 2500, number of wind turbines per km2 being 3.5 and the MW per turbine being 3. Of the 2.01 TWh, 1 would be produced in four cities alone (Umeå, Gävle, Helsingborg and Halmstad) and only 25 wind turbines and 0.18 TWh would be produced by the three largest cities; Stockholm, Göteborg and Malmö, combined.

The city of Linköping has no suitable area in any of the geo-types (Table 3). Linköping is located within a MSA-area, which means that no wind power can be built here. The city of Linköping is therefore not mapped in the next chapter.

4.4 Mapping land suitability for wind power

Figures 3 through to 21 show maps of land cover suitability for wind power in proximity to the

19 largest cities considered in the

study.

area for wind power development are shown in orange, water bodies are blue and the land is grey. The ring buffers for each urban centre are 5000 meters, 10 000 meters, 15 000 meters and 20 000 meters. They are all displayed as black circles. In the top right corner of each map there is an overview map that shows where in Sweden the city is located. The scale of the maps varies from 1:250 000 to 1:350 000, depending of the size of the city centre. All information can be found in the bottom right corner of the maps.

The three largest cities, in population, are shown in Figures 3 to 5. The cities which have the most suitable land are displayed in Figures 6- 9. The data missing in both Helsingborg and Halmstad are evident in Figure 6 and Figure 7, respectively. The missing data is sourced back to the ‘Building’ geo-type. In Helsingborg and Halmstad, the data data was only suitable for approximately 50 % of the area considered. This means that the results from the GIS analysis in regard to these two cities are not entirely correct. The other cities included in this study is mapped in Figures 10 through 21.

Generally, Figures 3-21 shows that there is more suitable area in the outer buffer-rings and that the further away you are from the urban centre the more area is suitable. There is no trend that shows that there is a linear relationship that smaller cities have more suitable area or that larger cities have less suitable area.

22

Figure 3. The city of Göteborg. The map shows that there are almost no areas within 5000m buffer zone and in the outer buffer of 20 000 meters. In Göteborg most of the suitable area is within 10 000m -15 000m from the urban centre.

23

Figure 4. Map over the city of Stockholm. Within the ring buffers of 10 000-meters, 15 000-meters and 20 000-meters there are some area where wind area can be developed in the south east and in the north, but almost none in the mid-part of the map.

24

Figure 5. Map over the city of Malmö. Within the ring buffers of 5000 meters, 10000meters, 15000 meters and 20000 meters there are almost no area suitable for wind power development. The only visually detectable area is located in the east at 10 000 meters and 20 000 meters.

25

Figure 6. Helsingborg urban area. The majority of suitable area is located in the north, this is due to the fault in the

‘Building’ layer, which did not cover this area and therefore the criteria of distance to houses and churches is not included in this area in Helsingborg.

26

Figure 7. Halmstad urban centre. The majority of suitable area is located in the south-east, this is due to the fault in the ‘Building’ layer, which did not cover this area and therefore the criteria of distance to houses and churches is not included in this area of Halmstad.

27

Figure 8. Map of Gävle. In Gävle there is a substantial amount of area suitable for wind power development. The area is spread out within the ring buffers. The smallest amount of suitable land is in the inner ring buffer (5000 meters). In the south-eastern part of the map, the issue of layer fault can be seen and there is therefore no suitable land in this area.

28

Figure 9. Map of Umeå. The suitable land in Umeå is substantial in relation to the other cities. The suitable area is almost evenly spread out within the buffer zones.

29

Figure 10. Map of Borås city. In Borås there are visually no suitable area in the inner ring buffer (5000 meters), most of the area is located in the 10 000 meters’ buffer and some evenly divided between the 15 000-meter buffer and the 20 000-meter buffer.

30

Figure 11. Map of Eskilstuna. In Eskilstuna most of the suitable area is in the south and in the 15 000-meter buffer and the 20 000-meter buffer.

31

Figure 12. The city of Jönköping. The map shows that in the inner 5000 meter-buffer, there are visually no suitable area for wind power development. The suitable are is located in the 10 000 meter-buffer, the 15 000 meter-buffer and the 20 000 meter-buffer. Most of the area is also located to the south-west.

32

Figure 13. In Karlstad, the area that can be detected in the map is located in the far east in the 20 000 meter-buffer and some in the 15 000 meter-buffer.

33

Figure 14. Map of Lund. Within the area of Lund, most of the suitable land is located in the south in the 20 000 meter-buffer. There also are some smaller patches of suitable area in the east and in the north-east.

34

Figure 15. Map of Norrköping city.In Norrköping most of the suitable area is located in the north distributed over the 10

000 meter-buffer, the 15 000 meter-buffer and the 20 000 meter-buffer. In the Norrköping map there are white jagged line in the upper right area. This is a result from how the GIS analysis was conducted.

35

Figure 16. Map over the city of Örebro with ring buffers. The area which is suitable for developing wind power is mostly located in the north-west 15 000 meter-buffer and 20 000 meter-buffer. The area is also located in the south, within the 5000 meter-buffer, 10 000 meter-buffer and 15 000 meter-buffer.

36

Figure 17. Map over the urban centre Södertälje. The suitable area is mostly distributed in the south and the majority of the area is in the 15 000 meter-buffer and in the 20 000 meter-buffer.

37

Figure 18.Map over Upplands Väsby. In Upplands Väsby the suitable areas shown are located from the middle to the north.

There is almost no area suitable in the inner circle (5000 meter-buffer). The areas suitable are spread out, and in the outer 15 000 meter-buffer and 20 000 meter-buffer.

38

Figure 19. Map over Uppsala. In the area surrounding Uppsala urban centre there is almost no suitable area for developing wind power. The only area visible in the map is located in the south east and spread out in the 10 000 meter-buffer, 15 000 meter-buffer and the 20 000 meter-buffer.

39

Figure 20. Map over Västerås. In the map over Västerås the suitable area is mostly located in the northern parts and in the outer buffer rings (10 000 meters – 20 000 meters).

40

Figure 21.Map over Växjö. In the area covered by this map the only larger suitable areas are located in the 20 000

41

5 Discussion

The study shows that there is not enough combined area in close proximity to the largest urban areas in Sweden to build sufficient wind power in order to phase out nuclear power. However, when individually analysed, four of the studied cities have the area required to become 100 % fossil and nuclear free. Eight cities can decommission 10 % of their nuclear power and replace it with wind power. The three largest cities in Sweden: Stockholm, Göteborg and Malmö do not have the required area to decommission any nuclear power.

There is generally more suitable area further away from the urban centre (Figures 3-21). The majority of the suitable area is in the outer bands of 10 meters to 15 meters and 15 000-meters to 20 000-000-meters. This is because housing is not as densely built in the outer areas of the cities.

The buffer zone between buildings and wind turbines had the most significant impact on the suitability of the studied landscapes for wind power. In urban areas this buffer makes it difficult to find any space for wind power development. The impact of this buffer is evident in the maps over Helsingborg and Halmstad (Figures 6 and 7) where the data is missing from parts of the city boundaries. Up to 2004, this criterion was only half of what it is today, 400 meters (Swedish

Energy Agency, 2017).Future changes in the distance criterion is possible since legislation makes

no mention of its extent. Because the distance has such a great impact on urban wind power, one could argue for more flexibility in urban contexts. If the criterion was reduced from 800 meters to 400 meters, there could be as much as double the amount of suitable area. In this scenario 40 % of the nuclear power needed in the urban areas could be replaced with wind power (39 % of scenario ‘C’).

The key findings in this study support and highlight the fact that the most difficult aspect of wind power development is location, as brought forward by Pasqualetti (2002) Lunney et al. (2016) and Eichhorn et al. (2017). The number of identified geo-types where wind power cannot be developed and the area they cover are much larger than the suitable geo-types (Table 1 and Table 3). This is also evident in Figures 3-21. As wind power in Sweden is continuously expanding, planners and developers have to navigate the issue of location.

The Swedish Energy Agency is working on the ongoing expansion of wind power. However, in the long-term, the placement is often uncertain. As indicated by the location of the national interest sites for wind power production, put forward by the Swedish Energy Agency (2017), the areas they consider are rural. The results of this study show that to phase out nuclear power, the majority of wind power has to be placed in Sweden’s rural areas. With regard to social sustainability, this causes conflicts. It is therefore important for the developers and planners of wind power production to carefully navigate this conflict. To even partially alleviate this problem, planners could aim at using as much of the suitable area as possible within the urban areas as investigated in this study.

Urbanisation in Sweden is continuing, and increasingly more people are moving into the larger cities. This means that in the future there will be even less suitable area in close proximity to the cities if no actions are taken today, as these areas will be used for housing and other infrastructures. Higher urbanisation would also mean that the public would have a greater need