Overview

Five overarching sectors are developed to represent the broader energy system. Each of these includes a large set of technologies to capture the status of the current energy mix, as well as future technology options for decarbonisation. Specifically, the following sectors are modelled:

  1. Primary energy supply – this relates to fossil fuel supply, nuclear fuel supply, renewable energy potential and hydrogen imports. Primary energy supply and transformation infrastructure (e.g., gas pipelines, LNG regasification terminals, oil refineries, etc.) is not modelled explicitly, with the only exception of electricity interconnectors between countries in the disaggregated version of the CLEWs-EU model and hydrogen production. Supply of biomass into the energy system occurs via an interlinkage with the land module of the model.

  2. Electricity and heat generation – fossil fuel, nuclear and renewable energy technologies are represented, along with a simplified representation of grid networks to capture associated losses. Additional technology options represented include electricity storage, use of Carbon Capture and Storage (CCS) technologies, hydrogen production through electrolysers to produce green hydrogen and steam methane reforming (i.e., using natural gas as feedstock) with or without CCS to produce blue and grey hydrogen respectively.

  3. Transport – the transport sector includes road transport, which is further broken down to passenger (passenger cars and buses) and freight (light commercial vans and heavy trucks), rail transport (passenger and freight), aviation and shipping.

  4. Buildings – this sector basically comprises of the households and service sectors. It is broken down into end-use services (i.e., space heating, space cooling, cooking, sanitary hot water, lighting and appliances).

  5. Industry –up to four distinct industries, based on the most energy-consuming sectors, are represented separately for each country, while the rest of the industries are lumped together in a fifth category . The output of each of the industries is represented in generic energy terms (i.e., PJ of useful energy services).

Figure 2. Simplified representation of CLEWs-EU energy module.

It should be noted that the energy demands in all sectors are inputs to the model . A simplified representation of the structure of the energy module is provided in Figure 2. Further disaggregation of technologies is available in the model than is depicted. For instance, the various vehicle technologies will be further broken down into passenger cars, buses, light commercial vans and heavy trucks. Similarly, the structure of the industrial sector will be constructed for each of the four key industries (and other industries) to be represented in each country.

Overarching assumptions

Among the most crucial assumptions adopted in the model are the international fuel price projections. The model uses as a base the international fuel price projections of the European Commission recommendations (European Commission – DG CLIMA, 2024). The same source provides projections on the existing Emission Trading System (ETS) and the new ETS (Table 1).

The existing ETS initially applied only to electricity and heat generation and energy-intensive industries. The new Emission Trading System (ETS2) will be expanded to include buildings, road transport, and smaller industries (i.e., those not already covered by the existing ETS), becoming fully operational in 2027 (European Commission – Directorate-General for Energy, n.d.-a). Both the ETS and the ETS2 are included in the baseline setup of the CLEWs-EU model.

Table 1. International fuel and ETS price projections.

Fuel / ETS

Unit

2021

2025

2030

2035

2040

2045

2050

Oil

EUR2018/GJ

10.24

10.16

11.39

12.62

12.95

14.10

16.14

Gas (NCV)

EUR2018/GJ

14.83

7.70

7.38

6.72

8.28

8.11

7.87

Coal

EUR2018/GJ

3.69

3.36

3.28

3.11

3.11

3.28

3.28

ETS WEM

EUR2018/tCO₂

53.27

77.85

77.85

81.95

81.95

131.12

155.71

ETS WAM

EUR2018/tCO₂

53.27

77.85

77.85

114.73

237.66

352.39

401.56

ETS2 WEM

EUR2018/tCO₂

0

0

0

0

0

0

0

ETS2 WAM

EUR2018/tCO₂

0

0

45.07

114.73

237.66

352.39

401.56

Electricity and heat generation

Representation of electricity and heat generation is crucial in the CLEWs-EU model as it defines the long-term cost-competitiveness of these commodities as alternatives to fuel combustion at the final consumer level. A large set of technologies is used to represent this sector (Figure 3). The following subsections discuss the structure and key assumptions adopted in this sector.

Figure 3. Simplified representation of electricity and heat generation in the CLEWs-EU model.

Electricity

The development of the electricity supply module has been based on detailed historical data for electricity demand, PV and wind profiles, as well as hydro availability. Data on existing installed capacity has been retrieved primarily from the JRC-IDEES 2021 database (Rózsai et al., 2024) and cross-verified with data from ENTSO-E. Furthermore, demand profiles have been downloaded from the ENTSO-E website for the year 2022 for all member countries (ENTSO‑E, . Moreover, an important constraint is set, related to the maximum total installed capacity of RES power plants: for hydropower, these values are adopted from each country’s NECP projections, whereas for solar and wind energy, from the JRC ENSPRESO database (Ruiz et al., 2019). PV and wind generation profiles have been downloaded from Renewables-Ninja (Pfenninger & Staffell, 2022) and adjusted to get the average annual capacity factor based on Eurostat (Eurostat, 2025b). Wind and PV resources have been broken into high, average and low potential, based on data from the JRC-IDEES database (Rózsai et al., 2024) (Figure 4). For each country, there are 3 PV and wind technologies/resources associated with a different capacity factor (%) and economic potential (in GW).

Figure 4. Capacity factor assumptions for different classes of wind in Spain.

Run-of-River (RoR) hydro generation is modelled similarly as in the case of PV and wind, using a historical profile of generation, retrieved from ENTSO-E (ENTSO‑E, 2025) and adjusted to get the average capacity factor based on historical data by Eurostat (Eurostat, 2025b). The hourly profiles of demand, as well as generation from PV, wind and RoR have been treated and adjusted to the agreed temporal resolution that is used across the CLEWs-EU model (see Section 1.2.3). Hydro plants with dams are modelled as generators running on water (represented as a “fuel”), while the maximum capacity factor is capped based on historical generation. The main difference between RoR and hydro with dams is that hydro with dams can operate flexibly, subject to a seasonal cap on generation that is based on statistical analysis of historical hydro generation data for seven consecutive years. To better reflect the annual fluctuations of hydropower generation, the model considers 7-year cycles where 6 years with average hydrological data are followed by a single ”dry” year (Figure 5). All adopted profiles for hydro, PV, and wind have been compared with historical data based on Eurostat and have been treated to represent an average year. For this purpose, historical annual averages have been obtained from 1980-2019 for PV and Wind and 2015-2022 for hydro.

Figure 5. Capacity factors for hydro dams in Spain for average vs dry year. In RoRs, generation is pre-defined at the per time-slice level, while on hydros with dams, generation is capped on a seasonal level.

The model accounts for a wide range of technologies which comprise a portfolio of projects to be considered in the optimisation. The technological mix of conventional generators can be distinguished based on the thermodynamic cycle (steam (ST) vs gas, open vs closed cycle vs combined cycle (CCGT)), and on the fuel (biomass, nuclear, coal, natural gas, liquid fuels, waste, biogas etc.). The model also includes CCS for specific technologies, coal-ST with CCS, gas CCGT with CCS and biomass-ST with CCS. Utilising data from ENTSO-E (ENTSO‑E, , grid interconnections between the EU member states are represented in the model. Specifically, the existing and potential future capacities of interconnections have been incorporated in the model. The level of electricity trade between the systems is endogenously calculated in the model and it depends on the technoeconomic assumptions adopted for each scenario. The techno-economic parameters of electricity generation technologies (efficiency, capital cost, operation and maintenance cost, etc.), presented in Table 2 and Table 3, are based on the latest available EU Reference Scenario (European Commission - Directorate General for Energy. et al., 2021). The CLEWS-EU model considers two types of electricity storage, namely battery energy storage systems (BESS) and pumped hydro. The techno-economic parameters and trajectories for these storage options have been based on the Annual Technology Baseline (ATB) documentation of NREL (NREL, 2024). The model assumptions include capital cost trajectories for all technologies; this is especially important for technologies where the capital cost is expected to be reduced with time as in the case of PV, wind and BESS. The OSeMOSYS code used to run the CLEWs-EU model contains modified storage equations to ensure proper representation of the operation of these technologies.

Table 2. CAPEX and FOM costs of electricity & heat generation technologies

Technology

CAPEX 2021

CAPEX 2030

CAPEX 2050

FOM 2021

FOM 2030

FOM 2050

Nuclear Generation III

4,770

4,500

4,500

120

115

105

Coal supercritical

1,650

1,650

1,650

33–41

31–36

28–31

Coal Supercritical with CCS

3,432

3,270

3,075

69

66

62

Biomass Steam Turbine

1,980

1,800

1,700

47

40

38

Biomass Steam Turbine with CCS

4,013

3,675

3,205

80

69

61

Steam Turbine CHP

1,980

1,800

1,700

46

40

38

Gas Turbine CC

598

579

570

22

21

19

Gas Turbine OC *

399–497

386–465

380–450

12–29

12–24

12–24

Reciprocating engine

1,897

1,897

1,897

35

35

35

Natural Gas CC with CCS

1,738

1,625

1,500

41

38

34

Geothermal

3,211

2,807

2,614

102

95

99

Waste PP

1,647

1,615

1,600

52

45

39

Hydro reservoir PP

2,100

2,100

2,100

26

26

26

Run of river PP

1,711

1,670

1,650

9

8

8

Batteries **

578–1734

380–1140

300–900

48

31

21

Pumped Hydro Storage

3,062

30,612

3,062

19

19

19

Table 3. CAPEX and FOM costs of solar and wind technologies based on resource potential

Technology

CAPEX 2021

CAPEX 2030

CAPEX 2050

FOM 2021

FOM 2030

FOM 2050

Wind onshore Low

1198

1175

1100

13

13

12

Wind onshore Medium

1045

1000

925

14

14

12

Wind onshore High

995

950

880

21.9

21

20

Wind offshore Low

1778

1650

1503

32.4

27

26

Wind offshore Medium

1756

1622

1468

32.4

27

26

Wind offshore High

1734

1593

1432

32.4

27

26

Solar PV Low

470

400

367

14.8

12.6

8.2

Solar PV Medium

455

387

355

14.8

12.6

8.2

Solar PV High

447

380

348

14.8

12.6

8.2

Solar PV Rooftop Low

470

400

367

18.6

14.9

9

Solar PV Rooftop Medium

635

543

500

18.6

14.9

9

Solar PV Rooftop High

625

536

493

18.6

14.9

9

Heat

The heat module of the power sector of the CLEWs-EU model contains CHP plants with the main purpose of producing electricity, while heat is directed as a by-product into district heating networks. The existing capacity of CHP plants and the level of district heating demand are retrieved from the JRC-IDEES 2021 database (Rózsai et al., 2024). CHP units that are part of industrial facilities are excluded from this module, as end-use energy demand in the industry sector is accounted for separately. To vary the level of electricity and heat production throughout the model horizon and the intra-annual timesteps, we have added two modes of operation for each CHP technology. The first mode represents the maximum rate of electricity production for the CHP unit, while the highest heat-rate production is accounted for in the other mode. The electricity and heat efficiency ranges of CHPs for these two modes are taken from IEA ETSAP (IEA ETSAP, 2010). Depending on the level of electricity versus heat demand in the end-use sectors, the model chooses which of the two modes to prioritise.

Units of electricity and heat generation technologies

OSeMOSYS models do not have a specific unit definition, so the decision on the units to be used is entirely up to the modeller. The parameter “CapacityToActivityUnit” is used to relate the rate of activity of each technology to its capacity. In the energy module of CLEWs-EU, this parameter is either 1 or 31.536. In the former case, this means that the activity is in PJ and the capacity is in PJ/year, while in the latter case, the capacity is in GW.

Table 4. Units for the capacity and cost characteristics of electricity and heat technologies in the CLEWs-EU model

Capacity

Activity

Capital cost

Fixed cost

Variable cost

GW

PJ

Million EUR2018/GW

Million EUR2018/GW

Million EUR2018/PJ

Buildings

The buildings module aggregates the residential and commercial services sectors. It splits demand for energy services into four primary end-uses: heating (i.e., space heating & hot water), space cooling, cooking and electrical appliances. Units for the capacity and cost characteristics of electricity and heat technologies in the CLEWs-EU model. For space heating, the module incorporates several technologies to reflect the heterogeneity of heating systems in buildings. These include solid fuel furnaces for solid fuels, oil boilers for LPG and gas/diesel oil, gas-based systems (including gas heat pumps and heaters), biomass boilers for biomass and waste, district heating systems, heat pumps (indicated as advanced electric heating in JRC-IDEES database (Rózsai et al., 2024)), electric furnaces (representing conventional electric heating and circulation systems), and solar thermal installations. Space cooling is represented by electric space cooling systems due to the predominance of electricity-based cooling technologies in both residential and commercial buildings. Cooking is represented using three primary technologies: LPG stoves, gas stoves, and electric stoves, capturing the most common cooking appliances across the sectors. Lighting and electrical appliances are treated as a single, aggregated category without further technological disaggregation. As in all modules of CLEWs-EU, to facilitate the independent development and testing of the buildings’ module, we introduce additional dummy technologies for electricity and centralised heat provision, as well as dummy technologies that can directly satisfy the defined energy service demands at an exceptionally high cost. These temporary additions allow for standalone operation of the module during development phases. However, it’s important to note that these dummy technologies are also included in the final integrated model version, where electricity and heat are endogenous to the model via the dedicated “Electricity & Heat generation” module, as they allow identification of potential modelling faults.

Figure 6. Simplified representation of the buildings sector in the CLEWs-EU model.

Technoeconomic characteristics of building technology options

The development of the buildings module was based on data from the Joint Research Centre of the European Commission, specifically the Joint Research Centre’s Integrated Database of the European Energy System (JRC-IDEES). The latest version of the publicly-available dataset, JRC-IDEES-2021 (Rózsai et al., 2024) was published in May 2024 and covers the timeframe from 2000 to 2021, making it suitable for the calibration of the model. The release introduces several methodological refinements and incorporates new statistical sources as well as feedback from the user community. As an analytical database, it goes beyond raw statistical data by processing and decomposing energy consumption into specific processes and end-uses, allowing for in-depth analysis of energy system dynamics. JRC-IDEES was created by harmonizing existing statistics with technical assumptions. For the buildings sector (both commercial and residential), JRC-IDEES (Rózsai et al., 2024) offers highly disaggregated data on energy demand, breaking it down by end-use (e.g., space heating, cooling, lighting) and fuel type and is aligned with Eurostat energy balances. As per the overarching model assumptions (Table 5), capital costs are derived from the EU Reference Scenario 2020 technology assumptions (European Commission - Directorate General for Energy. et al., 2021), with the exception of solid fuel boilers/furnaces, which are taken from IEA ETSAP (IEA ETSAP, 2012), which is also consulted to derive operation and maintenance costs.

Table 5. Technoeconomic assumptions of buildings technologies in the CLEWs-EU model.

Technology

Capital cost (EUR2018/kW)

Fixed O&M (EUR2018/kW)

Energy Intensity Index (2021 = 100%)

2021

2030

2050

2021

2030

2050

2021

2030

2050

Oil boiler

170

192

180

5.08

5.08

5.08

100%

100%

100%

Gas boiler

165

187

185

5.08

5.08

5.08

100%

88%

88%

Coal boiler

165

187

185

5.08

5.08

5.08

100%

100%

100%

Biomass boiler

431

488

457

13.22

13.22

13.22

100%

94%

91%

Electric furnace

64

78

71

5.08

5.08

5.08

100%

100%

100%

Heat pump

817

865

697

5.08

5.08

5.08

100%

81%

63%

Solar thermal

1308

1432

1119

5.08

5.08

5.08

100%

95%

92%

District heating

96

111

104

100%

99%

99%

Heat pump (cooling)

817

865

697

5.08

5.08

5.08

100%

81%

66%

LPG stove

198

202

195

100%

95%

91%

Gas stove

198

202

195

100%

95%

91%

Electric stove

189

194

186

100%

96%

91%

General appliances

100%

93%

91%

Renovation (Low)

1134

1134

1134

100%

100%

100%

Renovation (Medium)

705

705

705

100%

100%

100%

Renovation (High)

900

900

900

100%

100%

100%

*The capital cost unit of building renovations is in Million EUR2018/PJ of yearly savings. The values are indicative, as renovation cost and achieved savings vary by country.*

Efficiency factors are projected to evolve and are therefore inserted in the model as time series. For the period 2018-2021, these calculations were based on the JRC-IDEES-2021 database. Since both Final Energy Consumption (FEC) and Useful Energy Consumption (UEC) data for the Households (HH) and Services (SER) sectors are available in the database for all EU Member States and the European Union as a whole, we computed the efficiency factors as:

Efficiency = \(\frac{\text{Total UEC}}{\text{Total FEC}}\)

For 2022-2050 projections, efficiencies are indexed to the estimated efficiencies for 2021 from JRC-IDEES as:

Efficiency \((y_n) = Efficiency_{2021} \times Energy\ Intensity\ Index\)

Where

\(Energy\ Intensity\ Index = \frac{Efficiency\ in\ Reference\ Scenario\ (y_n)}{Efficiency\ in\ Reference\ Scenario_{2021}}\),

derived from the EU Reference Scenario 2020 technology assumptions.

Some technologies, such as oil- and coal-based systems, are assumed to maintain constant efficiency over time. For lighting and electrical appliances, the model incorporates projected efficiency improvements due to equipment replacement and anticipated technological advancements. This is implemented by gradually improving the relevant efficiency parameters over time, reflecting the expected enhancement in energy conversion efficiency for these end-uses.

Furthermore, according to the adopted technoeconomic assumptions, building renovation costs and savings potential are assumed to differ across EU regions, since both climatic and socioeconomic conditions influence renovation performance. EU Member States are grouped as follows:

  • South: Greece, Cyprus, Italy, Spain, Malta, Portugal

  • East: Bulgaria, Estonia, Latvia, Lithuania, Romania

  • Centre/West: Austria, Belgium, Czech Republic, Germany, France, Croatia, Hungary, Ireland, Luxembourg, Netherlands, Poland, Slovenia, Slovakia

  • North: Denmark, Finland, Sweden

A limitation on annual renovation rates is also implemented in the model. Specifically, up to \(1.5\%\) of the building stock may be renovated per year, unless adjusted in a scenario.

Table 6. Regional variation in savings and cost for renovation measures in the buildings sector of the CLEWs-EU model (adapted from the EU Reference Scenario 2020).

Region

Extent of renovation

Energy savings (%)

Capital cost (Million EUR2018/PJ annual savings)

Centre/West

Light renovation

15.8%

1134

Centre/West

Medium renovation

68.7%

705

Centre/West

Deep renovation

78.0%

900

North

Light renovation

22.3%

897

North

Medium renovation

67.4%

749

North

Deep renovation

86.6%

870

South

Light renovation

16.1%

853

South

Medium renovation

56.5%

682

South

Deep renovation

75.3%

802

East

Light renovation

12.9%

780

East

Medium renovation

55.5%

452

East

Deep renovation

68.6%

566

Finally, technology lifetimes are also sourced from available technology briefs (IEA ETSAP, 2012), which provide standardised estimates for the operational lifespan of various heating and cooling technologies in residential and commercial buildings.

Input data transformation

Due to the lower technological complexity adopted in CLEWs-EU, aggregation from the relevant input data is pursued. Specifically, we group similar technologies together to keep the model complexity low and to create meaningful categories for analysis. Energy service demands for the tertiary and residential sectors are aggregated together. JRC-IDEES (Rózsai et al., 2024) provided data for the use of hot water separately, but we group this thermal demand with space heating.

Table 7. Technology aggregation by end-use service and fuel, based on JRC-IDEES 2021 data.

Building demand projections

Demand projections in the baseline scenario of CLEWs-EU are based on final energy demand projections for the residential and commercial sectors, retrieved from the EU Reference Scenario 2020 (European Commission - Directorate General for Energy. et al., 2021). For consistency with the JRC-IDEES-2021 data used for the calibration of the model, we indexed the values of the final energy demand projections to the base year 2021 and calculated coefficients relative to this baseline. Specifically, in the baseline setup of the CLEWs-EU model, the growth rate of useful energy demand for each of the energy service demands in the period 2022-2050 resembles the collective final energy demand growth rate of the residential and tertiary sectors from the EU Reference Scenario 2020 (European Commission - Directorate General for Energy. et al., 2021). This implies that the overall efficiency will remain at current levels until mid-century. However, since the model allows building upgrades for improved energy efficiency, as well as assumes partial improvement of the energy efficiency of the technology stock for the various energy services, the baseline scenario of CLEWs-EU is expected to result to lower final energy demand projections than the EU Reference Scenario 2020. This is an aspect that can be adjusted in the scenario exploration phase of the DIAMOND project (i.e., WP5). More details on the exact demand projections by country and energy service are provided in the Supplementary material within the model’s GitHub repository.

Units of building sector technologies and commodities

Since OSeMOSYS models do not have a specific unit definition, the adopted units must be defined externally prior to the model development. As indicated in the tables below, the output of building technologies is defined in terms of their activity. The existing capacity of technologies is estimated based on the activity of technologies as extracted from the JRC-IDEES 2021 database.

Table 8. Units for the input and output of building technologies in the CLEWs-EU model.

Energy service technologies

Input

Output

Space heating & hot water

PJ fuel

PJ of useful energy for heating

Space cooling

PJ electricity

PJ of useful energy for cooling

Cooking

PJ fuel

PJ of useful thermal energy for cooking

Appliances

PJ electricity

PJ of useful energy for appliances

Energy efficiency upgrades

PJ of useful energy savings (cooling & heating)

Table 9. Units for the capacity and cost characteristics of building technologies in the CLEWs-EU model.

Technology

Capacity

Capital cost

Fixed cost

Space heating & hot water

GW

Million EUR2018/GW

Million EUR2018/GW

Space cooling

GW

Million EUR2018/GW

Million EUR2018/GW

Cooking

GW

Million EUR2018/GW

Million EUR2018/GW

Appliances

PJ/year

Energy efficiency upgrades

PJ/year

Million EUR2018/PJ yearly savings

Industry

Representation of industry in the CLEWs-EU model is largely based on data retrieved from the JRC-IDEES 2021 database (Rózsai et al., 2024). Specifically, the database provides statistics on the final energy demand and useful energy demand for each sector by process and by technology/fuel for the period 2000-2021, but for the purposes of CLEWs-EU, we only focused on the period 2018-2021.

Figure 7. Simplified representation of industry in the CLEWs-EU model.

As a first step, the key industrial sectors in each country are identified based on the level of final energy demand in 2021 (Table 2); for each country, up to 4 main sectors are represented separately, while the rest of the sectors are aggregated into an “Other industrial sectors” category. In countries where industrial energy demand is low, the number of sectors represented is lower (e.g., Cyprus, Luxembourg, Malta, etc.). As mentioned above, the output of each of the industries is represented in generic energy terms (i.e., PJ of useful energy services). As more data becomes available in the future after completion of the DIAMOND project, a more explicit representation of key industries can be pursued.

Table 10. Industrial sector representation in the disaggregated CLEWs-EU version.

Country

ISI

NFM

CHI

NMM

PPA

FBT

TRE

MAE

TEL

WWP

OIS

AT

X

X

X

X

BE

X

X

X

X

X

BG

X

X

X

X

X

CY

X

X

X

CZ

X

X

X

X

DE

X

X

X

X

DK

X

X

X

X

X

EE

X

X

X

X

X

GR

X

X

X

X

X

ES

X

X

X

X

FI

X

X

X

X

FR

X

X

X

X

X

HR

X

X

X

X

X

HU

X

X

X

X

X

IE

X

X

X

X

X

IT

X

X

X

X

LT

X

X

X

X

X

LU

X

X

X

LV

X

X

X

X

MT

X

NL

X

X

X

X

PL

X

X

X

X

PT

X

X

X

X

X

RO

X

X

X

X

X

SE

X

X

X

X

SI

X

X

X

X

SK

X

X

X

X

*Note: The industrial sectors represented are Iron and Steel (ISI), Non-ferrous metals (NFM), Chemical industry (CHI), Non-metallic minerals (NMM), Pulp, paper and printing (PPA), Food, beverages and tobacco (FBT), Transport equipment (TRE), Machinery equipment (MAE), Textiles and leather (TEL), Wood and wood products (WWP), and Other industrial sectors (OIS).*

Technoeconomic characteristics of industrial technology options

Data on final energy demand and useful energy demand are aggregated for each fuel category in each of the represented sectors, and an efficiency is calculated for the years 2018–2021. Specifically, the technological and fuel options for satisfying useful energy demand are the following:

  • Thermal energy – coal (with and without CCS), oil, natural gas (with and without CCS), biomass/biofuels, hydrogen, solar thermal, centralised steam, electricity.

  • Electrical processes (lighting, motor drives, etc.) – electricity.

Adopting the same approach as for the buildings sector, projections for the evolution of capital costs and efficiencies of industrial boilers until 2050 are based on assumptions from the EU Reference Scenario 2020 (European Commission - Directorate General for Energy et al., 2021). However, efficiency is indexed to the last available year of statistics (i.e. 2021), using data from the JRC-IDEES 2021 database (Rózsai et al., 2024).

An overview of the key technoeconomic assumptions for the industrial sector is provided in Table 11.

Table 11. Technoeconomic assumptions of industrial technologies in the CLEWs-EU model (adapted from EU Reference Scenario 2020).

Technology

Capital cost (EUR2018/kW)

Fixed O&M (EUR2018/kW)

Energy Consumption Index (2021=100%)

2021

2030

2050

2021

2030

2050

2021

2030

2050

Oil

232

252

252

1.32

1.32

1.32

100%

91%

91%

Coal

356

386

386

1.37

1.37

1.37

100%

91%

91%

Coal with CCS*

2203

2085

1888

38.37

35.01

30.46

115%

114%

107%

Gas

119

128

128

1.25

1.25

1.25

100%

92%

92%

Gas with CCS*

1350

1256

1133

22.75

20.14

16.09

132%

126%

121%

Biomass

771

836

836

1.37

1.37

1.37

100%

92%

92%

Electricity (heat)

701

683

673

10.36

10.36

10.36

100%

100%

100%

Electrical processes

100%

94%

87%

Hydrogen

119

128

128

1.25

1.25

1.25

100%

92%

92%

Solar thermal

1308

1432

1119

15.54

15.54

15.54

100%

95%

92%

*Coal with CCS and gas with CCS capital and fixed costs are based on modellers’ own assumptions using cost differences from relevant power generation technologies in the EU Reference Scenario 2020 assumptions. Similarly, their efficiency is estimated in the same manner; in these cases, the Energy Consumption Index refers to a comparison with the respective technology without CCS for the same year. For instance, the efficiency of gas with CCS is 32% higher in 2021 compared to the same technology without CCS.*

Data extraction and preparation from JRC-IDEES statistics to data ready for direct input to the model has been conducted with specially developed Python scripts. These scripts provide detail on the technology/fuel aggregation for each industry and how the selected industries vary across the countries and are made available on the model’s GitHub repository.

Industrial demand projections

In the baseline setup of the model, the growth rate of useful energy demand for the period 2022-2050 resembles the final energy demand growth rate of energy-intensive and non-energy intensive industries from the EU Reference Scenario 2020. Energy intensive industries include iron and steel, non-ferrous metals, chemicals, non-metallic minerals and pulp and paper; these are assumed to be subject to the existing Emissions Trading System (ETS). The rest of the industrial sectors are assumed to be subject to the new ETS (i.e., ETS2) from 2027 onwards (European Commission - Directorate-General for Energy, n.d.-a). Adopting the same growth rate for useful demand as for final energy demand from the EU Reference Scenario 2020 entails the assumption that the overall efficiency will remain at current levels until mid-century; this is an aspect that can be amended in the scenario exploration phase of the DIAMOND project (i.e., WP5). Separate demand projections are developed for each of the key industries of each member state. In turn, these are split between demand for useful thermal energy and useful demand for electrical processes; as indicated above, the latter is satisfied by a single technology option. More details on the exact demand projections by country and industrial sector are provided in the Supplementary material within the model’s GitHub repository.

Units of industrial sector technologies and commodities

OSeMOSYS models do not have a specific unit definition; as such, the adopted units must be defined externally before the model development. As indicated in the tables below, the output and capacity of industrial technologies are defined in terms of their activity. The existing capacity of technologies is estimated based on the activity of technologies as extracted from the JRC-IDEES 2021 database.

Table 12. Units for the input and output of transport technologies in the CLEWs-EU model.

Technology

Input

Output

Capacity

Capital cost & Fixed cost

Fuel cost

Thermal & electrical processes

PJ fuel

PJ of useful heat

GW

Million EUR2018/GW

Million EUR2018/PJ

Transport

Similar to the other sectors of the energy system, the representation of the transport sector is constructed using data from the JRC-IDEES 2021 database (Rózsai et al., 2024). Specifically, relevant statistics are retrieved on the biofuel blending ratios in each mode of transport, existing stock of technologies, annual rate of activity and current efficiency levels of each technology. Key assumptions (e.g., efficiencies, annual mileage etc.) vary considerably between member states, hence country-specific information is adopted. Table 13 summarises the technology options represented for each mode of transport. As in the case of the industrial sector, Python scripts have been used for the transport data preparation and are also made available in the model’s GitHub repository.

Table 13. Technology options in the transport module of CLEWs-EU.

Mode of transport

Category

Technology options

Road transport

Passenger cars

Gasoline vehicles, Diesel vehicles, CNG/LNG vehicles, LPG vehicles, Hybrid electric vehicles, Plug-in electric vehicles, Battery electric vehicles, Hydrogen fuel cell vehicles

Buses

Diesel bus, CNG bus, Battery electric bus, Hydrogen fuel cell bus

Light commercial vans

Gasoline light trucks, Diesel light trucks, CNG/LNG light trucks, Hybrid electric light trucks, Plug-in electric light trucks, Battery electric light trucks, Hydrogen fuel cell light trucks

Heavy trucks

Diesel heavy trucks, CNG/LNG heavy trucks, Hybrid electric heavy trucks, Battery electric heavy trucks, Hydrogen fuel cell heavy trucks

Rail transport

Passenger

Diesel train, Electric train

Freight

Diesel train, Electric train

Aviation

Conventional kerosene planes, Electric planes, Hydrogen/e-fuel planes

Shipping

Oil vessels, LNG vessels, Hydrogen/synthetic-fuel vessels

Figure 8. Simplified representation of land transport in the CLEWs-EU model.
Figure 9. Simplified representation of aviation and shipping in the CLEWs-EU model.

Technoeconomic characteristics of mobility technology options

Technoeconomic characteristics of the various technologies and the evolution of the mobility demand for each mode of transport are based on the equivalent assumptions from the EU Reference Scenario 2020 (European Commission - Directorate General for Energy. et al., 2021), but are adjusted to agree with the base year statistics. Specifically, efficiency for each transport technology from JRC-IDEES in 2021 in each country is used as the starting point and its evolution is indexed to the growth rate projected in the EU Reference Scenario for the period up to 2050. In addition, the capital cost of transport technologies is based on the relevant technology assumptions of the EU Reference Scenario 2020, but the adopted number of technologies is lower for the sake of simplification. For instance, in the case of passenger vehicles, the corresponding values for medium cars are chosen for each technology option and these act as representative technologies.

Table 14. Capital cost of transport technologies in the CLEWs-EU model (adapted from EU Reference Scenario 2020).

Mode

Technology

Unit

2021

2025

2030

2040

2050

Passenger road

Gasoline vehicles

EUR2018/vehicle

20,084

20,140

20,794

21,553

23,431

Diesel vehicles

EUR2018/vehicle

23,540

23,685

24,366

25,833

34,858

CNG/LNG vehicles

EUR2018/vehicle

22,239

22,295

22,949

23,709

25,586

LPG vehicles

EUR2018/vehicle

21,652

21,708

22,361

23,121

24,998

Hybrid electric vehicles

EUR2018/vehicle

22,450

21,699

21,854

21,985

22,264

Plug-in electric vehicles

EUR2018/vehicle

26,673

23,419

22,296

21,967

21,973

Battery electric vehicles

EUR2018/vehicle

37,319

34,664

24,709

22,919

22,396

Hydrogen fuel cell vehicles

EUR2018/vehicle

49,782

37,348

31,330

29,985

30,170

Diesel bus

EUR2018/vehicle

287,109

287,844

290,033

291,898

294,695

CNG bus

EUR2018/vehicle

312,165

312,900

315,089

316,954

319,751

Electric bus

EUR2018/vehicle

444,300

420,053

329,131

308,306

299,329

Hydrogen fuel cell bus

EUR2018/vehicle

576,118

402,723

343,597

328,325

328,729

Freight road

Gasoline light trucks

EUR2018/vehicle

18,535

18,608

19,880

22,328

27,952

Diesel light trucks

EUR2018/vehicle

23,336

23,571

25,683

31,935

31,935

CNG/LNG light trucks

EUR2018/vehicle

20,701

20,774

22,046

24,494

30,118

Hybrid electric light trucks

EUR2018/vehicle

25,316

24,340

24,479

24,651

24,872

Plug-in electric light trucks

EUR2018/vehicle

29,346

25,619

24,518

24,224

24,359

Battery electric light trucks

EUR2018/vehicle

37,489

35,018

25,755

22,940

22,117

Hydrogen fuel cell light trucks

EUR2018/vehicle

45,010

35,109

29,449

28,300

28,633

Diesel heavy trucks

EUR2018/vehicle

82,977

83,760

85,202

87,326

90,454

CNG/LNG heavy trucks

EUR2018/vehicle

102,438

103,221

104,663

106,787

109,914

Hybrid electric heavy trucks

EUR2018/vehicle

91,463

88,932

88,031

87,698

88,517

Battery electric heavy trucks

EUR2018/vehicle

182,915

170,477

123,834

107,803

103,601

Hydrogen fuel cell heavy trucks

EUR2018/vehicle

273,991

171,693

132,082

118,724

119,183

Passenger rail

Diesel train

MillionEUR2018/unit

9.2

9.3

9.4

9.7

10.1

Electric train

MillionEUR2018/unit

12.6

12.7

12.8

13.2

14.3

Freight rail

Diesel train

MillionEUR2018/unit

10.1

10.3

10.6

11.1

12.0

Electric train

MillionEUR2018/unit

12.8

13.0

13.3

13.8

14.8

Aviation

Kerosene planes

MillionEUR2018/unit

115.5

117.0

118.8

123.1

129.2

Electric planes

MillionEUR2018/unit

319.8

247.3

228.9

Hydrogen planes

MillionEUR2018/unit

149.8

144.8

146.1

Shipping

Oil vessels

MillionEUR2018/unit

22.3

22.3

22.3

22.3

22.3

Gas vessels

MillionEUR2018/unit

24.9

24.9

24.9

24.9

24.9

Hydrogen/syn fuel vessels

MillionEUR2018/unit

34.2

31.1

30.1

Due to the large variation in the efficiency and mileage of each transport technology across countries, the relevant data extracted from JRC-IDEES 2021 is substantial. For this reason, the Python scripts used to extract and prepare the respective input data are provided in the model’s GitHub repository.

New ETS and automotive fuel price assumptions

The model uses as a starting point the international fuel price projections of the European Commission recommendations (European Commission - DG CLIMA, 2024). Then, statistics for automotive fuel prices from the Weekly Oil Bulletin (European Commission - Directorate-General for Energy, n.d.-b) are used to come up with fuel price projections until 2055. The price projections are initially estimated without taxes and duties, but the relevant information on the tax regime by the end of 2023 has been used to compile fuel prices with taxes and levies. The new Emission Trading System (ETS2) is planned to be expanded to include road transport and will become fully operational in 2027 (European Commission - Directorate-General for Energy, n.d.-a). The ETS2 is included in the baseline setup of the CLEWs-EU model. In the case of electricity cost for battery electric and plug-in hybrid vehicles, the cost of electricity includes all taxes and levies. Specifically, the cost of generated electricity in each country is calculated endogenously by the model, but the relevant technologies representing transmission and distribution also include relevant taxes and levies. Statistics from Eurostat are used to extract country-specific information on this aspect. Due to the higher number of passenger vehicles that correspond for battery electric vehicles, the relevant cost figures for household consumers are used (Eurostat, n.d.).

Mobility demand projections

The activity level of each mode of transport and technology option in 2021 is extracted from the JRC-IDEES 2021 database. The level of activity for each existing road transport technology option is only allowed to decrease within a certain percentage range to prevent unrealistic drops in the usage of specific technologies; this is primarily guided by the assumed operational lifetime of the respective vehicles. Similarly, the registrations of new vehicles are constrained by the actual rate of change of the vehicle fleet stock. In regards to the future evolution of activity for each mode of transport, the relevant projections from the EU Reference scenario are used (European Commission - Directorate General for Energy. et al., 2021). Specifically, the level of activity in 2021 is extracted from the JRC-IDEES 2021 database and used as a starting point. Then, the growth rate foreseen in the EU Reference Scenario is adopted and applied to come up with new projections until 2050. It should be clarified that the respective growth rate varies for each mode of transport and country across the EU.

Charging infrastructure

The future deployment of large numbers of electric vehicles will necessitate equivalent investments in charging infrastructure. A representative charging point technology is used to represent the relevant infrastructure, whose cost characteristics are based on the EU Reference Scenario 2020 technology assumptions.

Table 15. Cost characteristics of electric recharging in the CLEWs-EU model (adapted from EU Reference Scenario 2020).

Unit

Capital cost (EUR2018/kW)

Fixed O&M cost (EUR2018/kW)

2021

2030

2050

2021

2030

2050

EUR2018/kW

603

440

367

7.24

5.28

4.40

Biofuel and alternative fuel supply

The current energy mix of the transport sector in the EU is dominated by oil products, but this is gradually changing as alternative fuels are introduced in the stock of technologies. The model has the option to invest in technologies that are powered by natural gas, LPG, hydrogen and biofuels. The extent to which these fuels are used in the modelling horizon depends on the cost-competitiveness of the respective technology option is combination with the fuel cost and the respective emissions penalty, if applicable (i.e. through ETS2). In the case of biofuels, we allow conventional internal combustion engine (ICE) vehicle technologies to use blended fuel (e.g. diesel with biodiesel or gasoline with bio gasoline). The blending ratio at the start of the model horizon is based on the 2021 fuel statistics as retrieved for each country from the JRC-IDEES 2021 database. In the Baseline scenario setup of the CLEWs-EU model, this ratio is kept constant, but in future scenarios, it can be increased according to the relevant scenario narrative. Similarly, if one would like to explore the possibility of vehicles run by 100% biofuels, this adjustment can easily be accommodated in the model either through the aforementioned blending ratio or by altering the input fuel of ICE vehicles to the respective biofuel supply. A similar approach is followed for both road transport, rail transport and shipping.

Figure 10. Biofuel use in transport technologies in the CLEWs-EU model.

In the case of aviation, due to the very specific requirement of the ReFuelEU Aviation regulation, the provision for Sustainable Aviation Fuels (SAF) is included in the model but is kept deactivated in the Baseline Scenario setup of CLEWs-EU. SAF can refer to (a) synthetic aviation fuels, (b) aviation biofuels, or (c) recycled carbon aviation fuels; only the first two are represented in the CLEWs-EU model. Similar to the blending of biofuels in land transport, a minimum use of SAF is enforced based on the relevant legislation (Regulation (EU) 2023/2405 of the European Parliament and of the Council of 18 October 2023 on Ensuring a Level Playing Field for Sustainable Air Transport (ReFuelEU Aviation), 2023). It should be noted that the respective minimum requirements for synthetic aviation fuels are not forced in the Baseline scenario of the CLEWs-EU model. The adopted minimum shares of SAF are as follows:

  1. From 1 January 2025, each year a minimum share of 2% SAF.

  2. From 1 January 2030, each year a minimum share of 6% SAF.

  3. From 1 January 2035, each year a minimum share of 20% SAF.

  4. From 1 January 2040, each year a minimum share of 34% SAF.

  5. From 1 January 2045, each year a minimum share of 42% SAF.

  6. From 1 January 2050, each year a minimum share of 70% SAF.

Units of transport sector technologies and commodities

As mentioned previously in the CLEWs-EU model documentation, OSeMOSYS models do not have a specific unit definition, and the adopted units must be defined externally prior to the model development. This is comparatively more complicated in the transport sector than for the rest of the energy module. As indicated in the tables below, the output and capacity of transport technologies is defined in terms of their activity. The stock of technologies is converted externally into the required units by multiplying the stock with the annual usage of each technology (i.e. mileage), which is extracted from the JRC-IDEES 2021 database.

Table 16. Units for the input and output of transport technologies in the CLEWs-EU model.

Mode of transport

Input

Output

Road transport

PJ

Billion vehicle kilometres (G-veh km)

Passenger rail

PJ

Billion passenger kilometres (Gpkm)

Freight rail

PJ

Billion tonne kilometres (Gtkm)

Aviation

PJ

Million vehicle kilometres (M-veh km)

Shipping

PJ

Billion tonne kilometres (Gtkm)

Table 17. Units for the capacity and cost characteristics of transport technologies in the CLEWs-EU model.

Mode of transport

Capacity

Capital cost

Fuel cost

Road transport

G-veh km/year

Million EUR2018/G-veh km/year

Million EUR2018/PJ

Passenger rail

Gpkm/year

Million EUR2018/Gpkm/year

Million EUR2018/PJ

Freight rail

Gtkm/year

Million EUR2018/Gtkm/year

Million EUR2018/PJ

Aviation

M-veh km/year

Million EUR2018/M-veh km/year

Million EUR2018/PJ

Shipping

Gtkm/year

Million EUR2018/Gtkm/year

Million EUR2018/PJ

Hydrogen

Hydrogen is a fuel whose role may be critical to the decarbonisation of certain sectors. As such, the supply and production of this fuel have to be represented in the CLEWs-EU model. This fuel enters the energy module of the CLEWs-EU model either through imports or domestic production. Production of hydrogen can occur through steam methane reforming (SMR with or without Carbon Capture and Storage) using natural gas as feedstock, or through electrolysis. Hydrogen can then be used in the transport sector or in industry. For use in the aviation sector, hydrogen is first converted into jet fuel.

Figure 11. Hydrogen supply in the CLEWs-EU model.

The technoeconomic assumptions for the relevant hydrogen production technologies are drawn from EU Reference Scenario 2020 assumptions (European Commission - Directorate General for Energy. et al., 2021). Specifically, the adopted input is shown in Table 16.

Table 18. Technoeconomic characteristics of hydrogen production technologies (adapted from EU Reference Scenario 2020).

Technology

Capital cost (EUR2018/kW)

Fixed O&M cost (EUR2018/kW)

Efficiency (energy output vs input)

2021

2030

2050

2021

2030

2050

2021

2030

2050

Electrolysis

1241

621

186

27.5

14.5

9.3

72%

79%

85%

SMR

564

518

466

22.6

20.7

18.6

70%

72%

73%

SMR with CCS

927

880

829

37.1

35.2

33.1

41%

42%

42%