STRATEGIES FOR THE SOUTH EUROPEAN ENERGY SECTOR FOR THE NEXT 40 YEARS

(1)

STRATEGIES FOR THE SOUTH EUROPEAN ENERGY SECTOR FOR THE NEXT 40 YEARS

Gerard Salvador López

EXAMINER: MARK HOWELLS; SUPERVISOR: NAWFAL SAADI

DIVISION OF ENERGY SYSTEM ANALYSIS, KTH ROYAL INSTITUTE OF TECHNOLOGY June 2014, Stockholm, Sweden

(2)

To my parents and my two sisters, for always being there.

(3)

1 0. Abstract

This paper discusses the development of an energy systems model for the southern countries of Europe. More precisely, for three main actors of the South of the European Union: Spain, Italy and Portugal. The three of them are currently facing economic difficulties due to the world financial crisis. To satisfy their energy demand at the less cost-effective price and following the EU policies in terms of greenhouse emissions requires a deep analysis of the current situation and an accurate forecast for the upcoming years.

There are several EU (EU 20/20/20, treaty of Lisbon and EU ETS) and UN (Kyoto Protocol) policies that are taken into account in the model to build the most realistic scenarios that can happen in the three countries in the following years. This paper is based on the electricity consumption coming from the residential, industrial and commercial sectors. The model is developed in the open source program OSINDA (OSeMOSYS with INterface and DAtabase). It considers different possible scenarios for the three countries from 2010 to 2050 and asses the paths to follow in terms of infrastructure investments for the upcoming years.

The baseline scenario takes into account the current taxes in CO2 emissions, the current capital, fixed and variable costs and the prices of the imports of fossil fuels. Then, there are plausible futures that analyze different possible scenarios (with the normal uncertainty of the future).

The source code and modelling data is publicly available under the intellectual protection of Creative Commons®.

(4)

2 Table of content

0. Abstract ... 1

0.1 List of figures ... 3

0.2 List of tables ... 5

1. Introduction ... 6

1.1 Aims and objectives ... 6

1.2 Outline ... 6

1.3 Methodology ... 6

1.4 Limitations ... 7

1.5 Why OSINDA? ... 7

2. The South of Europe at a glance ... 9

2.1 Demographic Development. ... 9

2.2 Economic Development. ... 9

3. The Energy Systems ... 10

3.1 Italy ... 10

3.2 Spain ... 12

3.3 Portugal ... 14

3.4 Common traits ... 16

4. The energy model ... 19

4.1 OSINDA (OSeMOSYS with INterface and DAtabase) ... 19

4.2 The electricity demand ... 19

4.3 Distribution of the days and the seasons among the year and electrical demand associated ... 21

4.4 Imports and exports of the countries ... 24

4.5 The model in OSINDA ... 24

5. Analysis methodology ... 29

6. Results ... 31

6.1 Validation of the results ... 31

6.2 Plausible future results ... 32

7. Conclusions ... 49

8. Bibliography ... 52

Annex 1: Abbreviations used in OSINDA ... 55

Annex 2: Technological and economic data ... 56

Annex 3: Documents attached ... 61

(5)

3

0.1 List of figures

Figure 1: Electrical energy share in Italy, 1883-2012 ... 10

Figure 2: Electrical energy production in Italy, 1883-2012 ... 10

Figure 3 Electricity share in Spain, 2013 ... 12

Figure 4: Gross inland consumption of primary energy in Portugal (as % of total Mtoe), 2010 14 Figure 5: Gross electricity generation in Portugal (as % of TWh), 2010 ... 15

Figure 6: Gas natural pipelines that supply Europe, 2009 ... 18

Figure 7: Dependency on energy imports into the EU, 2009 ... 18

Figure 8: OSeMOSYS modular structure, 2010 ... 19

Figure 9: Electrical demand from Italy, Spain and Portugal in PJ ... 21

Figure 10: Spanish electrical demand per month ... 21

Figure 11: Average electrical demand in a winter day ... 22

Figure 12: Average electrical demand in a summer day ... 22

Figure 13: Average electrical demand in an intermediate day ... 23

Figure 14: Imports (in negative) and exports (in positive) of Italy, Spain and Portugal (in PJ) ... 24

Figure 15: The OSINDA model for the three countries ... 24

Figure 16: Efficiencies for the technologies of the model in % ... 25

Figure 17: Annual residual capacity for Italy in GW ... 26

Figure 18: Annual residual capacity for Spain in GW ... 26

Figure 19: Annual residual capacity for Portugal in GW ... 27

Figure 20: Price of the fossil fuels imported in M$ per PJ ... 27

Figure 21: Taxes in the CO2 emissions for the three countries (in $ per ton of CO2) ... 30

Figure 22: Price of the fuel in M$ per PJ for the renewable challenge scenario compared with the other scenarios ... 30

Figure 23: Energy share of Italy in 2010 in reality (left pie) and in the model (right pie) ... 31

Figure 24: Energy share of Spain in 2010 in reality (left pie) and in the model (right pie) ... 31

Figure 25: Energy share of Portugal in 2010 in reality (left pie) and in the model (right pie) .... 32

Figure 26: Annual production of Italy of electricity by technology in the baseline scenario (in PJ) ... 33

Figure 27: Annual capacity to produce electricity in Italy by technology in the baseline scenario (in GW) ... 33

Figure 28: Annual emissions of CO2 of Italy in the baseline scenario (in ktons) ... 33

Figure 29: Annual production of Spain of electricity by technology in the baseline scenario (in PJ) ... 34

Figure 30: Annual capacity to produce electricity in Spain by technology in the baseline scenario (in GW) ... 34

Figure 31: Annual emissions of CO2 of Spain in the baseline scenario (in ktons) ... 35

Figure 32: Annual production of Portugal of electricity by technology in the baseline scenario (in PJ) ... 35

Figure 33: Annual capacity to produce electricity in Portugal by technology in the baseline scenario (in GW) ... 36

Figure 34: Annual emissions of CO2 of Portugal in the baseline scenario (in ktons) ... 36

Figure 35: Annual production of Italy of electricity by technology in the renewable challenge scenario (in PJ) ... 37

Figure 36: Annual capacity to produce electricity in Italy by technology in the renewable challenge scenario (in GW) ... 37

Figure 37: Annual emissions of CO2 of Italy in the renewable challenge scenario (in ktons) .... 37

(6)

4 Figure 38: Annual production of Spain of electricity by technology in the renewable challenge scenario (in PJ) ... 38 Figure 39: Annual capacity to produce electricity in Spain by technology in the renewable challenge scenario (in GW) ... 38 Figure 40: Annual emissions of CO2 of Spain in the renewable challenge scenario (in ktons) .. 39 Figure 41: Annual production of Portugal of electricity by technology in the renewable

challenge scenario (in PJ) ... 39 Figure 42: Annual capacity to produce electricity in Portugal by technology in the renewable challenge scenario (in GW) ... 40 Figure 43: Annual emissions of CO2 of Portugal in the renewable challenge scenario (in ktons) ... 40 Figure 44: Annual production of Italy of electricity by technology in the storage success

scenario (in PJ) ... 41 Figure 45: Annual capacity to produce electricity in Italy by technology in the storage success scenario (in GW) ... 41 Figure 46: Annual CO2 stored in Italy in the storage success scenario (in ktons) ... 41 Figure 47: Annual production of Spain of electricity by technology in the storage success scenario (in PJ) ... 42 Figure 48: Annual capacity to produce electricity in Spain by technology in the storage success scenario (in GW) ... 42 Figure 49: Annual CO2 stored in Spain in the storage success scenario (in ktons) ... 43 Figure 50: Annual production of Portugal of electricity by technology in the storage success scenario (in PJ) ... 43 Figure 51: Annual capacity to produce electricity in Portugal by technology in the storage success scenario (in GW) ... 44 Figure 52: Annual CO2 stored in Portugal in the storage success scenario (in ktons) ... 44 Figure 53: Annual production of Italy of electricity by technology in the nuclear strategy

scenario (in PJ) ... 45 Figure 54: Annual capacity to produce electricity in Italy by technology in the nuclear strategy scenario (in GW) ... 45 Figure 55: Annual emissions of CO2 of Italy in the nuclear scenario (in ktons) ... 45 Figure 56: Annual production of Spain of electricity by technology in the nuclear strategy scenario (in PJ) ... 46 Figure 57: Annual capacity to produce electricity in Spain by technology in the nuclear strategy scenario (in GW) ... 46 Figure 58: Annual emissions of CO2 of Spain in the nuclear scenario (in ktons) ... 47 Figure 59: Annual production of Portugal of electricity by technology in the nuclear strategy scenario (in PJ) ... 47 Figure 60: Annual capacity to produce electricity in Portugal by technology in the nuclear strategy scenario (in GW) ... 48 Figure 61: Annual emissions of CO2 of Portugal in the nuclear scenario (in ktons) ... 48

(7)

5

0.2 List of tables

Table 1: Power reactors operating in Spain, 2014 ... 13

Table 2: Electricity production from renewable sources, 2011 ... 15

Table 3: Coefficients used to define the projections for the electricity demand ... 20

Table 4: Division of the seasons and average electrical consumptions ... 22

Table 5: Time slice distribution depending on the season and the type of day ... 23

Table 6: Time slice distribution of the electrical demand depending on the season and the type of day ... 23

Table 7: Power plant parameters used in the model for every technology ... 28

Table 8: Electricity demand in Italy, Spain and Portugal in PJ ... 32

Table 9: Import price of fossil fuels (in M$/PJ) and efficiencies (in %) between 2010 and 2050 56 Table 10: Italian residual capacity in GW between 2010 and 2050 ... 57

Table 11: Spanish residual capacity in GW between 2010 and 2050 ... 58

Table 12: Portuguese residual capacity in GW between 2010 and 2050 ... 59

Table 13: Taxes in the emissions of CO2 (in M$/kton) for every plausible future and electrical demand (in PJ) for every sector between 2010 and 2050 ... 60

(8)

6

1. Introduction

In this project the analysis turns into the question of understanding the energetic future for the upcoming years of Southern European countries. The model is based on electrical production and consumption. By analyzing the current situation of all those nations and by modeling the different possible scenarios, this thesis draws the future development of the energy panorama. The data used comes from international data sources and is reported at the end of this document.

1.1 Aims and objectives

The goal of this paper is to carefully model Spain’s, Italian’s and Portuguese’s energy system as well as looking into future especially in the growing demand of population and, therefore, in electrical consumption. It starts on the base year 2010 and the forecast is made for the following years up to 2050. In order to make future projections there are four plausible scenarios explained in the following paper and deeply analyzed at the end of it.

1.2 Outline

The outline of the report is as follows. Chapters 2 and 3 focus on information about the countries differences, their history alone and together and some crucial key data. This is very important as the energy system is highly dependent on the infrastructure of the countries and on their economic growth. In the current energy system, the model is analyzed in chapter 4 and its methodology in chapter 5. Chapter 6 presents the results of the four scenarios where current policies are taken into consideration. Finally, the observation and the conclusions of the analysis are set in Chapter 7. These scenarios look into the effects of the introduction of major actors to the energy production. Even more, the paper discusses the solutions for the alarming emissions of greenhouse gases. The program calculates the most cost-effective strategy to follow in order to satisfy the projected supply.

1.3 Methodology

The working methodology is based on the gathering of information in the modeling program OSINDA (OSeMOSYS with INterface and Database)¹. Regarding the data used, there are some assumptions that have been made and that are explained in the following lines. The data comes from web pages, international reports, regional reports, national reports, books and papers. The approach of those documents is either national or international and the quality and accuracy of the data has been taken into account in the execution of this report. It has been especially helpful the Eurostat database² for macroeconomic data and the ETSAP database³ for the costs projections of the current technologies.

This thesis analyses three different countries in three regions. The policies that the European Commission has planned and the different plausible futures are taken into account to have the most realistic model.

The theory followed by OSINDA is based on optimizing energy models. In other words, this program is driven by an objective function subject to constraints in terms of environment, policies, physical and technical limits. In this case, the objective is to optimize the total cost of

1 (Mark Howells, 2011)

2 (European Comission, 2014)

3 (International Energy Agency, 2010)

(9)

7 the energy model. For example, an EU objective may be considered through emission taxes. In OSINDA there are some assumptions that are needed in order to use this optimizing energy model tool:

- Perfect competition: all the players compete at their marginal production costs

- Perfect information: all the characteristics of each technology that is in the market are known by all the players. The same happens with the projections of those technologies.

- Economically rational consumer behavior: players invest in the cheapest life-cycle technology.

It is important to remain careful with those assumptions because the reality of the market is far from them. In this optimization process the model describes an ideal world where there is no misuse of the actual potential capacity of a country to produce energy. There are several occasions that those presumptions might be inaccurate. For example, a producer might be interested in a less profitable technology that has a tight pay-back over a more cost-effective long-term investment conveying a higher risk.⁴

However, the responsibility of the policies approved by the governments should be to align the real world to the projected ideal scenario to be, in an overall point of view, more cost-effective as a country.

1.4 Limitations

Data availability is one of the largest limitations of this project. Even Europe, who has the majority of its countries with an exhaustive energy program and a complete awareness of the importance of the energy demand and supply, can present lacks of information. It is especially challenging to find data regarding the characteristics of the different types of technology available in the market. All the power plants are unique in a way with their different capital costs, efficiencies, emissions etc. In the model they are gathered by types of technologies and in the different ways of processing the inputs into electricity. This method might sometimes be inaccurate. Furthermore, the data founded for the scenarios is based on projections (regarding prices, electricity demand, etc.). In this types of projects there is a notable uncertainty regarding the supply of energy (especially with the political conflicts that might appear in the oil and gas exporting countries).

1.5 Why OSINDA?

OSINDA (OSeMOSYS with INterface and DAtabase) is an open source code based on Access®

(from Microsoft Office®) that gather input of the principal characteristics of the energy factors and, by optimizing costs, finds the best cost-effective solution for the region. But there are several reasons that make OSINDA the best choice:

- Free and open source, so easy to modify and to code internally. It brings flexibility to the project and gives new possibilities for it.

- User friendly, therefore accessible for non-expert users

- Simple to link and export to MS-Office® programs (Excel®, Word® and Power-Point®).

4 (Welsch, 2013)

(10)

8 - Able to account energy consumption, its prosecution in the different technologies and

to track both energy and non-energy greenhouse emissions.

- Designed to conduct long-range scenario analysis, with the resulting scenarios being self-consistent story lines that illustrate how a particular energy system might evolve over time, in a particular social context and under a particular set of policy measures.

Even more, OSINDA is easy-to-use scenario-based modeling software for energy planning and GHG mitigation assessment. It is based not on a model of a particular energy system but on a tool for modeling different energy systems. Even more, it has no limitation in the length of the model (regarding the amount of years).⁵

(11)

9

2. The South of Europe at a glance

The three countries analyzed in the paper are Italy, Spain and Portugal; 9^th, 13^th and 50^th in the GDP world ranking. Those nations have shared a common history in terms of culture, language and even energy policies with the adhesion to the European Union.

Regarding the energy sector, Europe is a net importer of energy. Therefore, in the geopolitical issue, the countries are profoundly dependent of the outside to cover its increasing demand of energy. This is one of the reasons why several governments are investing and promoting the renewable energies or the nuclear power among the other sources that are, now-a-day, cheaper.

2.1 Demographic Development.

The population of those countries is distributed in a heterogeneous way. It is specifically located in the coast where the commerce with the other European countries is easier. The population of the three countries is 60, 46 and 10 million inhabitants for Italy, Spain and Portugal. The previsions for the upcoming years are that the populations grow till the 65 million for Italy, till the 52 million for Spain and stay stagnant for Portugal around the 10 million⁶. Those changes in population affect the electrical demand prevision for the following years. It is worth to mention that those assumptions are directly connected with the condition that any war or major crisis happens. Even more, the immigration policies (coming from the national and European level) also play a crucial role in this critical aspect of the economy. It is also important to mention that, especially in the last years, there is a problem of child-bearing (this is one of the reasons for accepting immigration from the outside of the continent) added to an ageing population that causes a lot of social challenges for the countries.

2.2 Economic Development.

Europe has been leading the economy of the world (especially in the colonial period) till the World Wars when it has given this leadership to the emergent countries (especially to the United States). Currently the world financial crisis has changed the forecast about the expansion that was going on before 2008. This crisis has especially affected the periphery countries of the European Union such as the countries that are being studied. It is worth mention that those countries are currently facing two main problems in their economies:

- The unemployment, with alarming numbers such as around 23% for Spain, 16% for Portugal and 11% for Italy⁷. Those rates affect notably the purchase power of the governments and, therefore, their ability to prepare and to plan the energetic needs of the country in terms of investments. Their accounts have to support an increase of the expenses (because of the subsidies of the unemployed) and a reduction of the income (because of a cut in the tax revenue).

- The financial problem. The international markets are lately getting in trouble. The international financial crisis has provoked that the companies and the governments had difficult access to affordable loans to invest in projects (such as the construction of a power plant). Therefore, the economy as a whole is stagnated and the ventures with a high capital cost are getting more and more difficult to bring into reality.

(12)

10

3. The Energy Systems

The main goal of this chapter is to have a general overview of the energy systems of the three countries that are going to be analyzed in the paper regarding the energy parameters. The three of them have their differences and similitudes that aware of the complexity of the topic but that also establish the first assumptions regarding the future image of those countries. All the peculiarities of the nations have been taken into account in the modelling process.

3.1 Italy

Italy is a country that has consumed around 2000 TWh of primary energy in 2012⁸ and it primary comes from fossil fuels (especially coming from petroleum, natural gas and coal). This country dedicates the 35% of its primary share to the electricity production⁹ coming mostly from natural gas, hydroelectric power and imports. However, there is currently a high increase of the renewable sources of energy (such as solar or wind) partly thanks to the high incentives coming from the European Union and from the national government.

Figure 1: Electrical energy share in Italy, 1883-2012¹⁰

Figure 2: Electrical energy production in Italy, 1883-2012¹¹

8 (BP, 2013)

10 (Terna, 2013)

11 (Terna, 2013)

(13)

11 Regarding the electricity sector, it has been usually accepted that a public monopolistic company would be more efficient and cheap that a complete open market. But this belief has started to disappear since 1980s. The liberalization of the electricity sector has started in the late 1990s after European directives. The new directive 96/92/CE of 1996 followed the new tendency towards liberalization and privatization. Then, the Italian legislative decree 79/1999 ("Decreto Bersani") of 1999 created a path towards the complete liberalization of the market through gradual steps. More than 40% of the electricity was planned to be traded on the free market by 2002. The network was transferred to a new company, Terna®, responsible for the management of the system. Moreover, the limit on Enel® (the government company) property share of Terna® was set at 20%. In order to improve competition and to develop a real free market for production, Enel® was also forced to sell 15,000 MW of capacity to competitors before 2003. Three new production companies were created: Endesa Italia®, Edipower® and Tirreno Power®. Finally, the European directive 2003/54/CE of 2003 and its subsequent Italian decree requested a free electricity trading for all commercial clients from July 2004 and a complete opening of the market for private customers from July 2007¹². Therefore, in the model there is free movement of the actors. Supply and demand reach an equilibrium prince without any government action.

An interesting point about Italy is the absence of nuclear power in its electricity production.

Italy is a powerful country that could afford this technology. In the Figure 2 of the paper there is a shy start in the 1960s but it disappears completely in the 1990s. This was due to the result of the Italian nuclear power Referendum in 1987 (after the Chernobyl accident in 1986). But in 2008 there were some new plans of constructing four new reactors in the country¹³. After the elections of April 2008 and the victory of the party People of Freedom (supporters of the nuclear power), some new plans were set to start the construction of nuclear powered plants in 2013. Even though, because of the strong opposition of the country to this technology and because of the disaster of Fukushima in 2011, the project was slowed down and belated till the second referendum held on 13^th of June 2011. The immense majority of the voters (94%) rejected the construction of new plants and, therefore, the plans of construction were cancelled. As it seems, the nuclear power is not supposed to be in the electricity market share for a long period. But in the framework of 40 years (in this paper the analysis is from 2010 to 2050), this vision might change. Especially if the energy price highly increases due to the scarcity of fossil fuels and to the costs of production or importation. Therefore, in this paper there is one plausible future projection that analyzes this topic.

Regarding the renewables, Italy has performed a high increase in the late years. The primary energy from renewables accounts for about 10% of the total primary energy consumption in Italy¹⁴ and for about 24.5% for the electricity production in 2011¹⁵. In 2011 the total capacity was 41,352 MW and the total energy produced was 84,190 GWh¹⁶. There is an explosion in the investments regarding those sustainable sources of energy, especially in wind and solar technologies. Italy is a country that has largely based its electrical production in hydropower

12 (Wikipedia, 2014)

13 (Herald, 2008)

14 (BP, 2013)

15 (Gestore Servizi Energetici (GSE), 2012)

(14)

12 but the fiscal advantages and incentives from the government and the economic support towards those technologies from the European Commission have made an increase in the demand of those clean technologies. The European Commission required to all the Member States of the European Union to notify a road map regarding the usage of the renewable technologies for the future. This document, called National Renewable Energy Action Plan (NREAP), describes how Italy plans to reach the share of 17% in gross final consumption and 26% in the electricity sector in renewable sources by 2020¹⁷.

At last but not least, Italy has a particular point to take into consideration. A part of his electricity is not produced inside the country; it is therefore imported (especially from France).

The total amount of electricity imported is about 43 PWh in 2010¹⁸, which is the 12% of the total electricity consumption of the country.

3.2 Spain

Spain is the fifth largest energy consumer in Europe because of its population size and industrial development. It is the 14^th biggest country in terms of GDP¹⁹ and one main actor in the European Union politics. This country is completely dependent on the liquid fuels (especially petroleum and natural gas.). The national government regulations limit the percent of total oil and gas imports that can be sold to Spain in order to have a balanced and diversified supply. Even more, the country has been (until the financial crisis of 2008) the fastest growing natural gas markets in Europe. Its imports come primarily from the north of Africa (especially from Algeria).²⁰

Figure 3 Electricity share in Spain, 2013²¹

17 (Italian Ministry for Economic Development, 2010)

19 (International Monetary Fund, 2013)

20 (EIA, 2013)

21 (Red Eléctrica Española, 2013)

(15)

13 Regarding the electricity sector, Spain has an electricity generation of 258 TWh²² and its share is balanced to have the less dependence of the exterior countries and of only one type of energy (Figure 3). This characteristic prepares Spain for fluctuations in prices of one specific technology or fuel. In Spain there is only one transmission agent and operator of the electricity system. This company is called Red Eléctrica Española® and depends directly from the central government.

Moreover, it is worth to mention the importance of the nuclear sector in Spain. The first commercial nuclear reactor began operation in 1968. It accounts eight nuclear reactors that supply the 21% of the country’s electricity generation. But, despite its importance, there are critic voices towards this technology. The ownership and operation is mostly directed by Endesa SA® and Iberdrola®. It is worth mention that Endesa is 92% owned by Italy's Enel®.

Spain is notable for power plant uprates. It has a program to add 810 MWe (11%) to its nuclear capacity through upgrading its nine reactors by up to 13%. The licenses for the plants to have activity ends in 2020 and 2021 but the new government has approved that the nuclear stations can ask for an extension of ten more years. Government commitment to the future of nuclear energy in Spain is uncertain, but has firmed up as the cost of subsidizing renewables becomes unaffordable. In the Table 1 there is the list of nuclear stations that are operative in Spain.²³

Reactors Net MWe

Almaraz 1 947

Almaraz 2 956

Asco 1 996

Asco 2 992

Cofrentes 1,063

Trillo 1 1,003

Vandellos 2 1,045

Total 7,002 MWe

Table 1: Power reactors operating in Spain, 2014

Then, the Spanish renewable sector is also important in the electricity production. The wind power plants have the second largest capacity of Europe (just after Germany). It has an installed capacity of 16,740 megawatts at the end of 2008 and rise of 1,609 MW every year.

Even more, in 2007 the Spanish government authorized offshore electricity generating facilities promote the development of offshore wind energy. It is forecasted that this technology has a notable impact on the future electricity share. Moreover, the solar sector has also increased its potential in the country. In 2005 Spain became the first country in Europe to require by law the installation of photovoltaic electricity generation in new buildings. Even more, there are subsidies towards the installation of those types of technologies²⁴. Those investments have permitted the government to slowly phase out the subsidies to coal production. However, and due to the recession of 2008, coal production and consumption increased in 2011 after a new regulation that introduced new domestic coal production subsidies to promote the usage of this cheaper technology and, therefore, decrease the overall price of electricity. The direct consequence being that electricity producers have moved from renewables to coal. The future is uncertain but the path towards the clean production of electricity seems started.

22 (IEA, 2013)

23 (World Nuclear Association, 2014)

24 (Roberts, 2009)

(16)

14

3.3 Portugal

Portugal is a country with around 10.65 million inhabitants and with a GDP of around 200 billion USD. The country has an electricity consumption of 51.19 TWh (2011) and emits around 48 Mt of CO2. The primary energy is originated mainly from fossil fuels. Around 75% of the total gross inland consumption comes from crude oil and petroleum, natural gas and solid fuels. On the other hand, the electricity generation of the country comes mainly from renewable energy sources (53.2%), followed by natural gas and other fossil fuels (Figure 4).

Figure 4: Gross inland consumption of primary energy in Portugal (as % of total Mtoe), 2010²⁵

Regarding the electricity sector, there is a high market concentration. The three biggest electricity producers (EDP®, REN Trading® and Iberdrola®) have a share of approximately 70%

of the whole production being EDP® the main actor (with a 55% of the share). It is important to mention that the Portuguese market has been integrated into the Iberian market in July 2007.

This strategic decision has permitted to the two countries to save money by economies of scale and on 2010 the market was split only the 21% of the time. It is worth to mention that the investment and operating costs of power generation in Portugal is significantly influenced by unsustainable support measures for generation such as the power guarantee mechanism, the Costs with the Maintenance of Contractual Equilibrium (CMECs) and the Power Purchase Agreements (PPAs). All those plans are built by the Portuguese government and have the objective of promoting some specific energy sectors.²⁶

(17)

15

Figure 5: Gross electricity generation in Portugal (as % of TWh), 2010²⁷

Then, there is an important point about the country. There is not a commercial usage of nuclear power but there is one research reactor located in the National Nuclear Research Centre used to science development. But Portugal has had nuclear plans. In 1971, an 8,000 MW nuclear power plant was projected with the deadline for its construction by 2000.

However, the plans were delayed and in 1995 they were definitely dismissed. Even more, in 2004 the government rejected to reconsider its decision. The strong opposition from the environmentalist organizations and companies involved in the renewable energy business forbid a further development of this technology in the near future.²⁸

The next stage, the renewable sector, is an important aspect of the electricity production. As shown in the Figure 5, the 53% of the electricity production comes from renewable sources. Its geographical position gives to Portugal the potential to continue his path towards a full clean production. The most abundant sunlight in Europe, strong winds from the Atlantic Ocean to the west, strong flowing rivers and huge ocean waves have long time been disregarded in this South European countries. In 2001, the government launched the E4 program (Energy Efficiency and Endogenous Energies), consisting of a set of multiple measures aiming to promote energy efficiency and the use of renewable energy (endogenous).²⁹ Portugal’s major actor regarding electricity production is hydropower. However, other technologies are growing fast (Table 2), especially thanks to the subsidies from the government towards this technology.

GWh Percentage

Hydro 12115 89,32

Wind 961 7,09

Solar 277 2,04

Geothermal 210 1,55

TOTAL 13563 100

Table 2: Electricity production from renewable sources, 2011³⁰

28 (World Nuclear Association, 2014)

29 (European Renewable Energy Council, 2004)

30 (IEA, 2011)

(18)

16

3.4 Common traits

As analyzed in the section before, Spain Portugal and Italy have different ways of supplying their electrical needs. Even if they have distant beliefs about what the energy sector should look like and what are the plans for the future, there are several common traits that affect all of them.

First, they are all members of the United Nations. This international organization gathers the countries of all over the planet and discusses matters that affect the globe beyond their frontiers. The greenhouse emissions are one of the crucial topics that have been negotiated.

The three countries have signed the resolution of the Kyoto Protocol. Many developed countries (including the three that are being studied) have agreed to legally binding reductions in their emissions of greenhouse gases in two commitments periods. The first commitment period applies to emissions between 2008 and 2012, and the second commitment period applies to emissions between 2013 and 2020. The protocol was amended in 2012 in the conference of Doha to accommodate the second commitment period (even if it has not yet entered in the regulations of the countries). Even if there are voices that are very critical with the insufficient scope of the document, it is a legally binding agreement under which industrialized countries have to reduce their collective emissions of greenhouse gases by 5.2%

compared to the year 1990 (but note that, compared to the emissions levels that would be expected by 2010 without the Protocol, this target represents a 29% cut).

Second, they are all members of the European Union. Therefore they are under the supervision and authority of the European Commission in terms of laws and of policy making.

This point is especially important regarding new regulations. Moreover, as part of Europe’s Climate Mitigation Plan, the European Emission Trading Scheme (EU ETS) system has been implemented in 2003 under the Directive 2003/87/EC³¹. The main purpose is to reduce the amount of greenhouse emissions thanks to a market based instrument complying with the Kyoto Protocol. The idea is to set tradable carbon units, which are permissions to emit CO2; one carbon unit is the permission to emit one ton of CO2. The scheme defines how much the countries can emit to attend the objective of the Kyoto protocol and give them those units.

There are countries that are more successful reducing emission, and therefore they might not need all the carbon units. The scheme set that the permission can be bought and sold in international markets. The price of emitting CO2 depends on how many are in circulation, similar to a stock exchange. Therefore, the success of the countries of the world reducing CO2

emissions defines how expensive is to actually emit those tons of CO2.³²

31 (European Parliament and European Council, 2003)

32 (Status, 2010)

(19)

17 Then, the Commission approved in March 2007 a climate and energy package set of binding legislation which aims to ensure that the European Union meets its ambitious climate and energy targets for 2020. These targets, known as the 20-20-20 targets, set three key objectives for 2020:

 A 20% reduction in EU greenhouse gas emissions from 1990 levels

 Raising the share of EU energy consumption produced from renewable resources to 20%

 A 20% improvement in the EU's energy efficiency

The 20-20-20 targets represent an integrated approach to climate and energy policy that aims to combat climate change, increase the EU’s energy security and strengthen its competitiveness. They are also headline targets of the Europe 2020 strategy for smart, sustainable and inclusive growth. This reflects the recognition that tackling the climate and energy challenge contribute to the creation of jobs, the generation of sustainable growth and a strengthening of Europe's competitiveness on the international markets.³³

Even more, after the Treaty of Lisbon of 2009, Europe was given the competences required to carry out a true energy policy. Before it, the action of the EU within the energy sector was limited to producing regulations covering the natural environment and competition. The objectives of the EU’s energy policy were set:

 To establish a single European market for energy; this is planned for 2014.

 To guarantee the security of basic energy supplies within the EU.

 To promote energy efficiency and energy savings and also to develop new and renewable types of energy.

 To foster the interconnection of energy networks.

The advantages of energy integration are to offer a greater efficiency due to a reduction of production costs (economies of scale), to favor the creation of interconnections that favor competition and to diversify the energy sources that ensures greater security of supply.³⁴ With the EU ETS, the 20/20/20 target and the Treaty of Lisbon, the European Commission sets a package of rules that defines and affects the future of the European energy sector. The model has taken into account the repercussion of those policies.

Then, regarding the renewable energies, there is a constraint that has to be mentioned. Even if the renewable energy delivers power till the infinite, there is a limit regarding the maximum potential capacity that can be delivered by it. The hydropower energy has a yearly potential of 121 PJ for Portugal, 467 PJ for Spain and 656 PJ for Italy³⁵. These limits have been modelled using specific parameters within the OSeMOSYS tool.

Another interesting matter to point out from those countries is the amount of refineries they have. In order to save money and to be more independent, the governments invest in this technology to transform crude oil into HFO (High Fuel Oils) and LFO (Light fuels oils). The

34 (Learn Europe, 2010)

35 (Bernhard Lehner, 2010)

(20)

18 current yearly capacity of crude oil processed is 3,200 PJ for Italy, 670 for Portugal and 3,105 for Spain³⁶. And in the process, it is important to mention that those refineries can work with different outputs. Therefore, there are two modes that have been implemented in the model approximating production outputs by linear interpolation. Using this strategy, the program can choose every year according to its need.

At last but not least, it is worth mention that the three countries are completely dependent on the imports of fossil fuels (notably natural gas, oil and coal). Even if there are efforts to reduce it, the reality is that the primary energy and the electricity production are directly connected to the prices of the fossil fuels in the international markets. Any conflict in a far country can have terrible consequences in the economy of those countries. This is one of the reasons to try to be less dependent of the surrounding countries and to try to invest in energy production technologies that do not need fuels from outside. In Spain and Portugal, the main supply comes from the north of Africa and for Italy it also receives from the eastern countries of Europe (Figure 6 and Figure 7).

Figure 6: Gas natural pipelines that supply Europe, 2009

Figure 7: Dependency on energy imports into the EU, 2009

36 (Oil & Gas Journal, 2013)

(21)

19

4. The energy model

The three countries analyzed have their own characteristics. Therefore they need a model that is realistic with the current situation in order to be able to have correct and accurate projections for the next forty years. In this chapter the main structure of the model is described.

4.1 OSINDA (OSeMOSYS with INterface and DAtabase)

The program used for this model is OSINDA, based on OSeMOSYS. This tool is a systems optimization model for long-term energy planning.

In essence, the model calculates the lowest net present cost (NPC) option of an energy system to meet the given energy demands. The code is written in GNU MathProg programming language and uses the GNU Linear Programing Kit (GLPK) solver. It is open source and has a wide flexibility thanks to its block structure.

Figure 8: OSeMOSYS modular structure, 2010³⁷

With all the demand and the possible technologies, the program optimizes the cost without forgetting the boundaries set in the program and following the modular structure of Figure 8.

4.2 The electricity demand

First, there is the importance of defining the timeframe of the model. The initial year of the project is 2010 and it goes till 2050. The energy system modelled includes only electricity including imports of fossil fuels, transformation, distribution and final consumption by three divisions of the society: industry, households and sectors (that includes transportation, commerce, agriculture and services).

In the model, one of the crucial information that has to be introduced is the electricity demand of the country for the next 40 years. This information directly affects the consistency of the model presented. It is difficult to project the demand but there are two variables that might directly affect it: the GDP and the population. Those two variables are forecasted by the

(22)

20 European Commission³⁸. Regarding the electricity demand, this paper uses an Analysis of Variance between 2000 and 2012 to determinate if there is a correlation between the electrical demand and the other two variables.

Then, if the answer is yes, the following equation is used for the forecast from 2013 till 2050.

On the other hand, if the answer is no, the demand is simply and linearly extrapolated from the data available.

𝐸𝑙𝑒𝑐𝑡𝑟𝑖𝑐𝑎𝑙 𝑑𝑒𝑚𝑎𝑛𝑑 = 𝐼𝑛𝑑𝑒𝑝 + 𝑐𝑜𝑒𝑓𝑓𝐴 ∗ 𝐺𝐷𝑃 + 𝑐𝑜𝑒𝑓𝑓𝐵 ∗ 𝑃𝑂𝑃

The results are based on a 95% of confidence interval. All values in this interval are plausible values for the parameter, whereas the ones outside it are rejected as plausible values.

Therefore, if the variable explains the model, its confidence interval does not contain 0 in the 95% of the cases.

In order to take a decision, the parameters taken into account have been the p-value, the T- Student value and the R-squared adjusted value. The Table 3 points out the different cases encountered. All the tables are able to be examined in the spread sheet attached to this paper.

Spain SEC

Relation YES

Spain HOU

Spain IND

Indep -11961 Indep -9297 Indep 33539

Coeff A 0,077 Coeff A 0,071 Coeff A 0,667

Coeff B 0,0004 Coeff B 0,0003 Coeff B -9E-4

Portugal SEC

Portugal HOU

Relation YES, POP

Portugal IND

Relation NO

Indep -6049 Indep -10865 Indep .

Coeff A 0,078 Coeff A . Coeff A .

Coeff B 0,0006 Coeff B 0,001 Coeff B .

Italy SEC

Italy HOU

Italy IND

Indep -23058 Indep -1649 Indep 57218

Coeff A 0,303 Coeff A 0,078 Coeff A 0,486

Coeff B 0,0004 Coeff B 9,3E-05 Coeff B 1E-04

Table 3: Coefficients used to define the projections for the electricity demand³⁹

There is one important result found in the forecast of the electrical demand of Portugal.

Following this strategy for the projection, this country decreases its demand after 2040 because of a reduction in the population of this nation for those last years of the model. Italy and Spain have a sustained increment of the demand over the years.

Even more, the electrical demand of Spain overcomes the one of Italy in 2030 following this forecast (Figure 9).

39 This table has been elaborated with after analyzing in the ANOVA table the p-value, the T-Student and the R-squared adjusted. SEC stands for sectors (including Agriculture, Services and Transports), HOU for households and IND for industry. The electrical demand is in mTOE, the GDP in EUR and the Population in direct amount.

(23)

21

Figure 9: Electrical demand from Italy, Spain and Portugal in PJ

4.3 Distribution of the days and the seasons among the year and electrical demand associated

The first assumption made was regarding the climate and the demand of the three countries.

They have similar climatological characteristics and therefore their distribution in the demands has been processed following the same pattern. This assumption comes from the fact that the electrical distribution is tightly related with the amount of hours of sun and with the climate in general. The country selected to set the model for the three countries has been Spain. Its distribution is used in the other nations too. The year selected for the measurements has been 2013. The average hourly load values of every month⁴⁰ have defined a division for the seasons:

winter, summer and intermediate. And, regarding the days, there is a division in two: day and night.

40 (ENTSOE, 2013) 0 200 400 600 800 1000 1200 1400 1600 1800

2010 2012 2014 2016 2018 2020 2022 2024 2026 2028 2030 2032 2034 2036 2038 2040 2042 2044 2046 2048 2050

Electrical demand in PJ

Year

Spain Italy Portugal

25000 26000 27000 28000 29000 30000 31000 32000 33000 34000

1 2 3 4 5 6 7 8 9 10 11 12

Average MW load per hour

Month of the year

Real Model

Figure 10: Spanish electrical demand per month

(24)

22

Month Days per month Average MW per hour per season Season

January 31 31380,575 Winter

February 28 31380,575 Winter

March 31 31380,575 Winter

April 30 27930,51667 Intermediate

May 31 27930,51667 Intermediate

June 30 27930,51667 Intermediate

July 31 30501,8125 Summer

August 30 30501,8125 Summer

September 31 27930,51667 Intermediate

October 30 27930,51667 Intermediate

November 31 31380,575 Winter

December 30 31380,575 Winter

Table 4: Division of the seasons and average electrical consumptions

Each season has a different type of day regarding the amount of hours with a high and low demand. The accuracy of the model is also defined by the precision of those calculations.

Therefore, the analysis has been focused on the case by case (Figure 10, Figure 11, Figure 12).

- Normal winter day

- Normal summer day

20000 25000 30000 35000 40000

01:00:00 02:00:00 03:00:00 04:00:00 05:00:00 06:00:00 07:00:00 08:00:00 09:00:00 10:00:00 11:00:00 12:00:00 13:00:00 14:00:00 15:00:00 16:00:00 17:00:00 18:00:00 19:00:00 20:00:00 21:00:00 22:00:00 23:00:00 24:00:00

Hours during the day

Real Model

Figure 11: Average electrical demand in a winter day

20000 25000 30000 35000 40000

01:00:00 02:00:00 03:00:00 04:00:00 05:00:00 06:00:00 07:00:00 08:00:00 09:00:00 10:00:00 11:00:00 12:00:00 13:00:00 14:00:00 15:00:00 16:00:00 17:00:00 18:00:00 19:00:00 20:00:00 21:00:00 22:00:00 23:00:00 24:00:00

Real Model

Figure 12: Average electrical demand in a summer day

(25)

23 - Normal intermediate day

After analyzing the three type of day in each type of season, there are two main results that have been used for the model: the time slice distribution and the electrical distribution in those time slices. The Table 5 and Table 6 sum up the result of the analysis.

Winter Summer Intermediate

Day Night Day Night Day Night

0,708 0,292 0,625 0,375 0,708 0,292

Winter 0,415 0,294 0,121

Summer 0,168 0,105 0,063

Intermediate 0,418 0,296 0,122

Table 5: Time slice distribution depending on the season and the type of day

Winter Summer Intermediate

Day Night Day Night Day Night

0,774 0,226 0,686 0,314 0,599 0,401

Winter 0,349 0,271 0,079

Summer 0,340 0,233 0,107

Intermediate 0,311 0,186 0,125

Table 6: Time slice distribution of the electrical demand depending on the season and the type of day 20000

25000 30000 35000

01:00:00 02:00:00 03:00:00 04:00:00 05:00:00 06:00:00 07:00:00 08:00:00 09:00:00 10:00:00 11:00:00 12:00:00 13:00:00 14:00:00 15:00:00 16:00:00 17:00:00 18:00:00 19:00:00 20:00:00 21:00:00 22:00:00 23:00:00 24:00:00

Real Model

Figure 13: Average electrical demand in an intermediate day

(26)

24

4.4 Imports and exports of the countries

The countries that have been analyzed import (Italy and Portugal) or export (Spain) electricity.

The data for the forecast from the European Comission⁴¹ is sum up in the Figure 14.

Figure 14: Imports (in negative) and exports (in positive) of Italy, Spain and Portugal (in PJ)

4.5 The model in OSINDA

The model used in the program including all the technologies and all the interconnections between them are schematized in the Figure 15.

Figure 15: The OSINDA model for the three countries

41 (Directorate general for energy, 2009) -180

-160 -140 -120 -100 -80 -60 -40 -20 0 20 40

2010 2050

Spain exports Italy imports Portugal imports

(27)

25 In terms of efficiency, the values are very different depending on the technologies that are considered. The efficiencies and its forecast used in the program has been interpolated and extrapolated from the International Energy Agency⁴² and is sum up in the following Figure 16.

It is also important to remember that the efficiencies of the technologies based on renewable sources have been set at 100% because the input is supposed infinite and free in the model.

Figure 16: Efficiencies for the technologies of the model in %

Then, the projection needs to have values about the residual capacities of the three countries.

Even more, it is needed to know how they are going to disappear during the forecast. The following Figure 17, Figure 18 and Figure 19 define how they are going to behave in the model.

It is important to remember that those figures do not take into account the new investments.

They only project the amount of capacity that is available each year for the already built capacity before 2010.

42 (International Energy Agency, 2010) 20

25 30 35 40 45 50 55 60 65 70

2010 2012 2014 2016 2018 2020 2022 2024 2026 2028 2030 2032 2034 2036 2038 2040 2042 2044 2046 2048 2050

Efficiencies (in %)

Year

NG Open Cycle NG Combined Cycle NG Stream turbine NG Steam turbine CHP CL CHP

CL Steam turbine CL Combined Cycle HF Open cycle HF Combined cycle HF Combined cycle CHP LF Engine

UR LWR BM CHP

(28)

26

0 5 10 15 20 25 30

Annual residual capacity in GW

Year

NG Open Cycle NG Combined Cycle NG Stream turbine NG Steam turbine CHP HY Small

HY Large CL CHP

UR LWR WI Inshore WI Offshore SO Steam turbine SO PV

BM CHP 0

5 10 15 20 25 30 35

Year

HY Large CL CHP

UR LWR WI Inshore WI Offshore SO Steam turbine SO PV

BM CHP

Figure 18: Annual residual capacity for Spain in GW Figure 17: Annual residual capacity for Italy in GW

(29)

27 Regarding the price of the imports of the fossil fuel used in the model is the same for the three countries. The current information⁴³ and the forecast⁴⁴ are based on two papers from the International Energy Agency and are sum up in the following Figure 20.

Figure 20: Price of the fossil fuels imported in M$ per PJ

43 (International Energy Agency, 2013)

44 (International Energy Agency, 2012) 0

5 10 15 20 25 30

2010 2012 2014 2016 2018 2020 2022 2024 2026 2028 2030 2032 2034 2036 2038 2040 2042 2044 2046 2048 2050

Price of the fuel in M$ per PJ

Year

Natural gas Coal Crude Oil Uranium 0

0,5 1 1,5 2 2,5 3 3,5 4 4,5 5

Year

HY Large CL CHP

UR LWR WI Inshore WI Offshore Figure 19: Annual residual capacity for Portugal in GW