Sven Teske *Editor*

# Achieving the Paris Climate Agreement Goals

Part 2: Science-based Target Setting for the Finance industry — Net-Zero Sectoral 1.5˚C Pathways for Real Economy Sectors

Achieving the Paris Climate Agreement Goals

Sven Teske Editor

# Achieving the Paris Climate Agreement Goals

Part 2: Science-based Target Setting for the Finance industry — Net-Zero Sectoral 1.5˚C Pathways for Real Economy Sectors

*Editor* Sven Teske Institute for Sustainable Futures University of Technology Sydney Sydney, NSW, Australia

#### ISBN 978-3-030-99176-0 ISBN 978-3-030-99177-7 (eBook) https://doi.org/10.1007/978-3-030-99177-7

© The Editor(s) (if applicable) and The Author(s) 2022 . This book is an open access publication.

**Open Access** This book is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this book are included in the book's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the book's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specifc statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use.

The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affliations.

This Springer imprint is published by the registered company Springer Nature Switzerland AG The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

*For the next generation. For my son Travis.*

# **Foreword**

Climate change is the defning threat of the twenty-frst century. It is at the centre of our global risk landscape, affecting global societies and economies through extreme weather events, food and water crises, sea-level rise, and large-scale migration.

This decade is decisive—it is far later than hoped, but not too late to avoid the worst consequences for mankind and our planet. Climate scientists, through the Intergovernmental Panel on Climate Change (IPCC), have detailed the strategies necessary to limit global warming to 1.5 °C above the average pre-industrial temperature level by 2100.

The COP26 summit in Glasgow brought parties together to accelerate action towards the goals of the Paris Agreement and the UN Framework Convention on Climate Change. 'Real economy' actors, cities, regions, businesses, investors, and institutions of higher education, responsible for nearly 25% of global CO2 emissions and over 50% of global GDP, made commitments to net-zero greenhouse gas emissions under the UNFCCC's Race to Zero. This is critical mass signalling to governments that non-state actors are already united in meeting the Paris Agreement's goals and in creating a more inclusive and resilient economy.

The pre-COP26 commitments predicted a 2.7 °C warming pathway. It is somewhat encouraging that in Glasgow at COP26, with the revised Nationally Determined Contributions for 2030 and the announcement of new net-zero targets and pledges, limiting global warming to 2 °C or even lower became achievable.

A major announcement at COP26 was the pledge of the Glasgow Financial Alliance for Net Zero (GFANZ)—a global coalition of over 450 fnance frms across 45 countries, jointly managing US\$130 trillion—to structure their fnancial activities to achieve net-zero emissions by 2050. This pledge is indicative of the scale and awareness of the global private sector.

Three UN-convened fnancial alliances, which are part of the GFANZ—the Net-Zero Asset Owner Alliance (NZAOA), the Net-Zero Banking Alliance (NZBA), and the Net-Zero Insurance Alliance (NZIA)—have committed to immediate action in aligning their investment, lending, and underwriting portfolios with *a pathway to limit the global temperature rise to 1.5 °C, with no or only limited overshoot*, consistent with the fndings of the IPCC Special Report.

If current emission levels are maintained, the remaining carbon budget required to limit global warming to 1.5 °C will be exhausted in less than a decade. Unless emissions are urgently and severely limited, the world will overshoot this carbon budget, and therefore exceed 1.5 °C warming. Overshoot scenarios come with a high risk of failing to reach the 1.5 °C target altogether.

To achieve the steep decarbonization of the global economy, all actors require information on how the transition can be achieved. Policies must be adjusted to support a transitioning economy, green technologies must be scaled up, energy effciency must increase, and consumer behaviour must change.

The United Nations Framework Convention on Climate (UNFCCC) Marrakech Partnership for Global Climate Action works to accelerate the implementation of the Paris Agreement by enabling collaborations between governments and cities, regions, businesses, and investors. In November 2020, it launched the *Climate Action Pathways*, which outlines sectorial visions for a 1.5 °C climate-resilient world. These pathways provide an overview of the actions and milestones required for the transformation of systems within sectors. They are supported and enhanced by the growing body of sectorial decarbonization pathways developed by the scientifc community and others, built on industry intelligence. One such effort, a collaboration between the scientifc community and in consultation with investors, is contained within this book.

In this book, Dr. Sven Teske and his research team provide data points for sectorial pathways on a low/no-overshoot basis. These pathways do not rely on carbon removal technologies but instead build on the rapid deployment of renewable energy and the preservation of natural carbon sinks. These detailed roadmaps provide highly ambitious information on the routes for various sectors and businesses. They also inform fnancial institutions of what they must require of their clients or investees to ensure that they participate in the journey to net-zero emissions by 2050.

This book provides a detailed analysis of 12 industry sectors, their interconnections, and their potential decarbonization in the short and longer terms. This assessment may be the frst to translate a global energy system model into 12 fnancial sectors, and to report the *Scope 1*, *2*, and *3* interconnections and therefore the fnal responsibilities for greenhouse gas emissions. This approach allows investors and actors in the real economy to engage with a common map and work together with all stakeholders towards change.

We must make use of all the intelligence at our disposal to move this critical mass of actors towards the fnish line in the race to achieve net-zero greenhouse gas emissions.

UK Nigel Topping

# **Acknowledgement**

The authors thank the experts, asset owners, and other stakeholders who provided peer review and input during the research between May 2020 and November 2021. In particular, the authors thank Elke Pfeiffer (NZAOA UNPRI) and Jes Andrews (UNEPFI) for their input, guidance, support, and collaborative spirit throughout this project. We also acknowledge and thank the researchers involved in the development of the One Earth Climate Model on which this study builds.

This research has been supported and fnanced in parts by the UN-convened Net-Zero Asset Owner Alliance, the Rockefeller Foundation, and the European Climate Foundation (ECF). The ECF stresses that responsibility for the information and views set out in this research lies with the authors. None of the founders can be held responsible for any use which may be made of the information contained or expressed therein. A special thank you to Dr. Anna Irimisch of ECF for suggestions and support.

Furthermore, we would like to thank Deutsche Gesellschaft für Internationale Zusammenarbeit (GIZ) GmbH, the Transformative Urban Mobility Initiative (TUMI), and the Federal Ministry for Economic Cooperation and Development (BMZ) who fnanced the development of the global and regional transport pathways which have been the basis for the 1.5 °C pathways for transport (Chap. 8). Thank you in particular to the GIZ team Daniel Ernesto Moser, Marvin Stolz, and Rohan Shailesh Modi.

The authors would like to thank the One Earth, a philanthropic organization working to accelerate collective action to limit global average temperature rise to 1.5 °C. Especially the One Earth Climate team Karl Burkart, Justin Winters, Edward Bell, and Edith Espejo for ongoing support. Furthermore, we would like to thank the former Leonardo DiCaprio Foundation, which funded the initial research between July 2017 and February 2019.

This project has been supported by numerous people since the book project started with *Achieving the Paris Climate Agreement Goals (Part 1)* in July 2017 and our thanks go to each of them. The ongoing support was key and kept all researchers highly motivated.

Special thanks to Anna Leidreiter, Anna Skowron, and Naemie Dubbles of the World Future Council (https://www.worldfuturecouncil.org/), Dr. Joachim Fuenfgelt of Bread for the World (https://www.brot-fuer-die-welt.de/en/bread-for-the-world), and Stefan Schurig of F20—Foundations 20 (http://www.foundations-20.org/) who provided initial support to make this project possible. Finally, we would like to thank Greenpeace International and Greenpeace Germany for their ongoing support of the Energy [R]evolution energy scenario research series between 2004 and 2015 which resulted in the development of the long-term energy scenario model, the basis for the One Earth Climate Model.

# **Executive Summary**

*Abstract:* To put this research project into context, a short introduction to the status of the climate debate is given. The methodology of the research is presented in brief, followed by the socio-economic assumptions made and key technological parameters used. The storylines of the energy demand projections for the 12 industry and service sectors analysed are described. The supply side of the sectorial pathways for power, heat, and fuels is documented. Finally, the main results are presented in terms of the fnal and primary energy demands, including energy- and non-energyrelated greenhouse gas emissions. Key conclusions are drawn by sector, and policy recommendations are offered.

# **Introduction**

Extreme weather events, such as extreme rainfall and foods, cyclones, and bushfres, have increased in frequency. Australia experienced the worst bushfre season on record between September 2019 and March 2020—known as the *Black Summer* (Cook et al., 2021). In June 2020, the Arctic region of Siberia experienced a heat wave with temperatures of up to 38 °C and wildfres covering almost 1 million hectares. The World Meteorological Organization (WMO) recognized this as a new Arctic temperature record.

Time is running out. In August 2021, the Sixth Assessment Report (AR6) of the United Nations Intergovernmental Panel on Climate Change (IPCC) was published. The First Assessment Report was launched in 1990 and underlined the importance of climate change as a challenge with global consequences that required international co-operation. Thirty years later, the IPCC states unequivocally that the world is already in the middle of climate change. The UN Secretary-General António Guterres said the Working Group's report was nothing less than '*a code red for humanity. The alarm bells are deafening, and the evidence is irrefutable*'.

Our frst book laid out global and regional 100% renewable energy scenarios with non-energy GHG pathways for +1.5 °C or +2 °C warming scenarios and compared them with a reference case. Those scenarios were calculated under the leadership of the Institute for Sustainable Futures (ISF) at the University of Technology Sydney (UTS) in close co-operation with the German Aerospace Center (DLR) and the University of Melbourne, Australia. The energy scenario model used for that project became known as the *One Earth Climate Model* (OECM) in 2020 during the numerous debates that followed the book launch in February 2019.

# **The Second Book Focuses on Sectorial Pathways and Provides Key Performance Indicators (KPIs) for Industry Sectors to Limit the Global Temperature Increase to 1.5 °C**

The book documents all the steps in the scenario development and provides a detailed analysis of the main assumptions and scenario narratives. The results of the OECM 1.5 °C pathways for 12 industry and service sectors include the total remaining carbon budget and the *Scope 1*, *2*, and *3* emissions for each sector.

# **Science-Based Industry Greenhouse Gas (GHG) Targets— Defning the Challenge**

The UN-convened *Net Zero Asset Owners Alliance* is a *Program for Responsible Investment* and a United Nations Environment Programme Finance Initiative (UNEP FI)-supported initiative. The members of the Alliance have committed to transitioning their investment portfolios to net-zero GHG emissions by 2050, consistent with a maximum global temperature rise of 1.5 °C above pre-industrial levels. This requires intermediate targets to be established for 5-yearly intervals, and regular reporting on progress.

# *Outlining the Task—Trend Reversals Until 2025*

The global economy must decarbonize the energy system entirely within the next 30 years—in one generation. In historical terms, this means breaking the connections between population growth, steady economic development fuelled by energy, and the increase in CO2 emissions of the past 120 years, and reversing those trends within the next 5 years. Between 2025 and 2030, global energy-related CO2 must peak and start to decline to zero by 2050.

Figure 1 shows the historic development of the global population, GDP, energy demand, and the resulting annual CO2 emissions between 1950 and 2020 on the left side, and the projected trajectory until 2050 on the right side. Based on the projected

**Fig. 1** Global development of key parameters

population and economic growth until 2050, and under the assumed historical trends of the past two decades, with an annual decline of 1% in both the energy and emissions intensities, the global energy demand will double, whereas CO2 emissions will remain at around current levels.

The OECM does not question the development of the population or the global economy projected by international organizations but focuses on technical measures to increase energy effciencies and decarbonize the energy supply by a transition to renewable energies to achieve the 1.5 °C decarbonization trajectory (marked with the red line). This will require a bottom-up assessment of the energy demand combined with an alternative energy supply concept for power, heating, and transport, which are documented in the following chapters of this book.

# *Science-Based Target Setting*

The latest available scientifc information is IPCC's Sixth Assessment Report *Climate Change 2021: The Physical Science Basis*. According to the IPCC defnition, 67% likelihood is 'good', whereas 50% likelihood is 'fair'. The OECM aims to limit the global mean temperature rise to 1.5 °C with 'good' likelihood. Therefore, the 'science-based target' for the OECM 1.5 °C pathway in terms of the global carbon budget between 2020 and 2050 is set to 400 Gt CO2.

The development of sectorial targets to meet the needs of specifc countries or industries will ensure that the global sum of all energy-related CO2 emissions for all countries or all industry sectors does not exceed the global budget. Therefore, any approach undertaken in isolation, such as for a single industry sector, will involve the risk that one industry sector will demand a high CO2 budget and push the responsibility to reduce CO2 emissions onto other sectors.

# **Methodology**

# *The One Earth Climate Model Architecture*

The One Earth Climate Model has been developed on the basis of established computer models. The energy system analysis tool consisted of three independent modules:


The advanced One Earth Climate Model, OECM 2.0, merges the energy system model EM, the transport model TRAEM, and the power system model [R]E 24/7 into one MATLAB-based energy system. The Global Industry Classifcation System (GICS) was used to defne sub-areas of the economy. The global fnance industry must increasingly undertake mandatory climate change stress tests for GICSclassifed industry sectors in order to develop energy and emissions benchmarks to implement the Paris climate protection agreement. This requires very high technical resolution for the calculation and projection of future energy demands and the supplies of electricity, (process) heat, and fuels that are necessary for the steel and chemical industries. An energy model with high technical resolution must be able to calculate the energy demand based on either projections of the sector-specifc gross domestic product (GDP) or market forecasts of material fows, such as the demand for steel, aluminium, or cement in tonnes per year.

The MATLAB model has an object-oriented structure and two modules—to calculate demand and supply—that can be operated independently of each other. Therefore, an energy demand analysis independent of the specifc supply options or the development of a supply concept based on demand from an external source is possible.

# *The High-Effciency Buildings Model (HEB)*

The HEB model was originally developed in 2012 to calculate the energy demand and CO2 emissions of the residential and tertiary building sectors until 2050 under three different scenarios (Urge-Vorsatz et al., 2012). Since then, the model has been developed and updated several times. With the latest update, the model calculates the energy demand under four scenarios until 2060, based on the most recent data for macroeconomic indicators and technological development. This model is novel in its methodology compared with earlier global energy analyses and refects an emerging paradigm—the performance-oriented approach to the energy analysis of buildings. Unlike component-oriented methods, a systemic perspective is taken: the performance of whole systems (e.g., whole buildings) is studied and these performance values are used as the input in the scenarios. The model calculates the overall energy performance levels of buildings, regardless of the measures applied to achieve them. It also captures the diversity of solutions required in each region by including region-specifc assumptions about advanced and sub-optimal technology mixes. The elaborated model uses a bottom-up approach, because it includes rather detailed technological information for one sector of the economy. However, it also exploits certain macroeconomic (GDP) and socio-demographic data (population, urbanization rate, foor area per capita, etc.). The key output of the HEB model is foor area projections for different types of residential and tertiary buildings in different regions and their member states, the total energy consumption of residential and tertiary buildings, the energy consumption for heating and cooling, the energy consumption for hot water energy, the total CO2 emissions, the CO2 emissions for heating and cooling, and the CO2 emissions for hot water energy.

# *Integration of HEB Results into the OECM*

To capture the complexity of regional and global building demand projection, both in terms of data availability and high technical resolution, the HEB was used to develop four bottom-up demand scenarios. The HEB was developed by the Central European University (CEU) of Budapest under the scientifc leadership of Prof Dr. Diana Uerge-Vorsatz.

# **Classifcation Systems for Setting Net-Zero Targets for Industries**

Investment decisions, such as the decarbonization targets for the fnance industry, are highly complex processes. In November 2020, the European Central Bank published a Guide on Climate-related and Environmental Risks, which maps a detailed process


**Table 1** GICS: 11 main industries

for undertaking 'climate stress tests' for investment portfolios. To achieve the Paris Climate Agreement goals in the global fnance industry, decarbonization targets and benchmarks for individual industry sectors are required. This opens up a whole new research area for energy modelling because although decarbonization pathways have been developed for countries, regions, and communities, few have been developed for industry sectors. The One Earth Climate Model (OECM) is an integrated assessment model for climate and energy pathways that focuses on 1.5 °C scenarios and has been further improved to meet this need. To develop energy scenarios for industry sectors classifed under the Global Industry Classifcation Standard (GICS), the technological resolution of the OECM required signifcant improvement. Furthermore, all demand and supply calculations had to be broken down into industry sectors before the individual pathways could be developed (Table 1).

The GICS has four classifcation levels, and includes 11 sectors, 24 industry groups, 69 industries, and 158 sub-industries. The 11 GICS sectors are: energy, materials, industrials, consumer discretionary, consumer staples, health care, fnancials, information technology, communication services, utilities, and real estate.

# **Sectorial Energy Scenarios for Industry and Services Provide Key Performance Indicators for Investors**

The fnance industry requires sectorial energy scenarios for the industry and service sectors to set sector-specifc decarbonization targets. Increasingly, investment decisions of international and national banks, insurance companies, and investor groups are driven by key performance indicators (KPIs) not only for proftability but also with regard to the embedded GHG emissions of a company. For asset managers, it has become increasingly important to have access to detailed information about GHG emissions, e.g., whether or not a steel manufacturer is on a decarbonization trajectory. The emissions must be further divided according to the responsibility for those emissions. This is done by calculating so-called Scopes 1, 2, and 3.

# **Methodologies for Calculating Scopes 1, 2, and 3**

Reporting corporate GHG emissions is important, and the focus is no longer only on direct energy-related CO2 emissions but includes the other GHGs emitted by industries. These increasingly include the indirect emissions that occur in supply chains. The Greenhouse Gas Protocol, a global corporate GHG accounting and reporting standard, distinguishes between three 'scopes':


The OECM model focuses on the development of 1.5 °C net-zero pathways for industry sectors classifed under the GICS, for countries or regions or at the global level. Emissions-calculating methodologies for entity-level Scope 3 require bottomup entity-level data to arrive at exact fgures. Therefore, data availability and accounting systems for whole industry sectors on a regional or global level present signifcant challenges.

Therefore, the Scope 3 calculation methodology was simplifed for country-, region-, and global-level calculations and to avoid double counting. The OECM reports only emissions directly related to the economic activities classifed by the GICS. Furthermore, the industries are broken down into three categories: Primary Class, Secondary Class, and End-use Activity Class.

Table 2 shows a schematic representation of the OECM *Scope 1*, *2*, and *3* calculation methods according to GICS class, which are used to avoid double counting. The sum of *Scopes 1*, *2*, and *3* for each of the three categories is equal to the actual emissions.

Double counting can be avoided by defning a primary class for the primary energy industry, a secondary class for the supply utilities, and an end-use class for all the economic activities that use the energy from the primary- and secondaryclass companies. The separation of all emissions by the defned industry categories—such as GICS—also streamlines the accounting and reporting systems. The volume of data required is reduced and reporting is considerably simplifed under the OECM methodology.

For a specifc industry sector to achieve the global targets of a 1.5 °C temperature increase and net-zero emissions by 2050 under the Paris Agreement requires that all its business activities are with other sectors that are also committed to a 1.5 °C increase and net-zero emissions.



# **Decarbonization Pathways for Industries**

The global gross domestic product (GDP) in 2019 was US\$87.8 trillion, 3% of which came from agriculture, 26% from industry, 15% from manufacturing, and the remaining 65% from services.

# *Chemical Industry*

The economic development of the global chemical industry is signifcantly more complex than that of the aluminium and steel industries. The product range of the chemical industry is diverse, and the material fow approach used for aluminium and steel is very data intensive, and is therefore beyond the scope of this research. The chemical industry produces materials for almost all parts of the economy from mining to services—and it is therefore intrinsically connected to overall economic development. Consequently, a GDP-based approach has been used to develop the energy demand projections for the chemical industry over the next three decades.

# *Projection of the Chemical Industry Energy Intensity*

The energy demands for the fve sub-sectors—pharmaceuticals, agricultural chemicals, inorganic chemicals and consumer products, manufactured fbres and synthetic rubber, and the petrochemical industry—were calculated with the energy intensities, which are based on the IEA Energy Effciency extended database and our own research. The energy intensities for primary feedstocks were also considered in estimating the effciency trajectories of the different sub-sectors. An increase in the effciency of primary feedstock production of 1% per year over the entire modelling period is required to achieve the assumed effciency gains for all sub-sectors. However, inadequate data are available to calculate the specifc energy intensities of the chemical industry, and no detailed breakdown of the electricity and process heat temperature levels is available in public databases. Therefore, our estimates should be seen as approximate values and more research, in co-operation with the chemical industry, is required. However, the energy requirements of the entire chemical industry are precisely known and were taken from the IEA statistics Advanced Energy Balances. The energy requirements of the sub-sectors were determined on the basis of market shares and GDP and in discussions with representatives of the chemical industry—specifcally members of the Net-Zero Asset Owner Alliance and the Strategic Approach to International Chemicals Management of the United Nations Environmental Program (SAICM UNEP).

# *Projection of the Energy Demand for the Chemical Industry*

The projections of the economic development and energy intensities of an industry yield the overall global energy demand projection for that industry. In another step, the share of electricity required to generate thermal process heat has been estimated. Table 3 shows the calculated electricity demand and the process heat demand by temperature level for the chemical industry sub-sectors.

# *Cement Industry*

Cement is the second most-consumed substance in the world after water and is a central component of the built environment—from civil infrastructure projects and power generation plants to residential houses. Typically made from raw materials such as limestone, sand, clay, shale and chalk, cement acts as a binder between aggregates in the formation of concrete. Cement manufacture is a resource- and emissions-intensive process, and is associated with around 7% of total global CO2 emissions, according to the Intergovernmental Panel on Climate Change.

Beyond the mining of the raw materials, there are fve main steps in the cement production process:


**Table 3** Projected electricity and process heat demand for the chemical industry to 2050


# *Global Cement Production and Energy Intensity Projections*

Table 4 summarizes the assumptions of the 1.5 °C OECM cement industry pathway in terms of the projected volume of global cement production, the development of energy intensities for the relevant processes, and the process emissions per tonne of clinker produced. These assumptions are similar, to a large extent, to those made for the IEA Technology Roadmap—Low-Carbon Transition in the cement industry projections.

# *Projections of the Cement Industry Energy Demand*

Table 5 shows the calculated electricity and process heat demand developments based on the documented assumptions. The breakdown by temperature level is based on the fve cement production steps required and their shares of the overall energy demand. No detailed statistical documentation of the exact breakdown of the process heat demand by temperature level and quantity is available.

# *Aluminium Industry*

Aluminium is among the most important building and construction materials globally. To understand the opportunities and challenges facing the industry, the global fow of aluminium metal must be considered. Since 1880, an estimated 1.5 billion


**Table 4** Assumed global cement market development and production energy intensities

**Table 5** Projected electricity and process heat demands for the cement industry


tonnes of aluminium have been produced worldwide, and about 75% of the aluminium produced is in productive use. In 2019, 36% of aluminium was located in buildings, 25% in electrical cables and machinery, and 30% in transport applications. Aluminium can be recycled, but the availability of scrap is limited by the high proportion of aluminium in use.

# *Bauxite Production*

Primary aluminium production requires bauxite. Bauxite ore occurs in the top-soils of tropical and sub-tropical regions, such as Africa, the Caribbean, South America, and Australia. The largest producers/miners of bauxite include Australia, China, and Guinea. Australia supplies 30% of global bauxite production.

# *Aluminium Production Processes*

An analysis of current and future aluminium production processes is required to understand the decarbonization opportunities within each process.

# *Primary Aluminium Production Involves the Following Processes (Excluding Mining)*


In the *secondary production of aluminium (aluminium recycling process)*, the process of refning the raw material (bauxite) to alumina is not required. Instead, scrap aluminium is re-melted and refned. Therefore, the energy consumption for this process is much lower than for its primary production.

# *Projection of the Aluminium Industry Energy Demand*

Due to the assumed increase in the share of recycled aluminium in global production and the reduced energy intensity per tonne of aluminium produced, a decoupling of the increases in production and energy demand is possible. Between 2019 and 2050, global aluminium production is projected to increase by 75%, whereas the overall energy demand will increase by only 12% (Table 6). Due to the already high electrifcation rates in the aluminium industry—which are projected to increase further—and the decarbonization of the electricity supply based on renewable power generation, the aluminium industry can halve its specifc CO2 emissions by 2035.

# *Global Steel Industry*

Steel is an important material for engineering and the construction sector worldwide, and it is also used for everyday appliances at the domestic and industrial levels. About 52% of steel usage is for buildings and infrastructure: 16% is used for mechanical equipment, such as construction cranes and heavy machinery; 12% is used for automotive vehicles (road transport); 10% is used for metal products, including tools; 5% is used for other means of transport, including cargo ships, aeroplanes, and two-wheeler vehicles; 3% is used for electrical equipment; and 2% is used for domestic appliances, such as white goods.

# *Technological Overview of Steel Production*

On average, 20 GJ of energy is consumed to produce one tonne of crude steel globally. There are two routes by which steel is produced. Primary or crude steel is produced by the coal- or natural-gas-based blast furnace–basic oxygen furnace (BF–BOF) route, in which iron ore is reduced at very high temperatures in a blast furnace. The iron ore is melted to a liquefed form, and then oxidized and rolled. Coal or natural gas is required to generate high temperatures of up to 1650 °C.

In the secondary production route, scrap steel is melted in electric arc furnaces (EAFs). The EAF route has the lowest emissions intensities. In the EAF (gasfuelled) process, scrap is usually blended at a rate of about 10% with direct reduced iron. A more energy-effcient pathway for primary production is to use scrap steel with ore-based inputs in BF–BOF production, usually at a rate of 15–20% scrap (Table 7).

# *Projection of the Steel Industry Energy Demand and CO2 Emissions*

The assumed division between primary and secondary production rates and the assumed production process technologies are key to the energy demand projections. Whereas secondary steel production requires signifcantly more electricity per tonne, its demand for high-temperature process heat is signifcantly lower (Table 8).


**Table 6** Projected electricity and process heat demands for the aluminium industry to 2050

**Table 7** Assumed market and energy intensity developments for the global steel industry according to the production process


(Continued)



#### **Table 7** (Continued)



**Table 8** Projected electricity and process heat demands for the steel industry to 2050

Furthermore, as the share of primary steel is reduced with higher recycling rates, the energy demand for iron-ore mining (volumes) that is required will decrease.

# *Textile and Leather Industry*

The international fashion industry is estimated to be worth US\$2.4 trillion, and the textile and leather industry constitutes a large proportion of it (valued at US\$818.19 billion in 2020). 'Textiles' refers to natural and synthetic materials used in the manufacture of clothing (including fnished garments and ready-to-wear clothing), furniture and furnishings, automotive accessories, and decorative items. Therefore, the textile industry spans activities related to the design, manufacture, distribution, and sale of yarn, cloth, and clothing. The textile and leather industry has close links with the agricultural and chemical industries. Agricultural output provides the raw materials for the textile industry in the form of natural fbres; similarly, the chemical industry outputs are used as synthetic raw materials in the textile industry.


**Table 9** Projected economic development and energy intensities of the textile and leather industry

# *Projections for the Global Textile and Leather Industry: Production and Energy Intensities*

Table 9 shows the assumed economic development and energy intensities for the textile and leather industry used to calculate the 1.5 °C OECM pathway. The energy intensities per product volume (e.g., in tonnes per year) are not available, so the energy demand is calculated as a product of the assumed economic development in \$GDP and the average energy units required per dollar. This simplifcation is necessary because the level of detail in the available energy demand data for the textile and leather industry on the global level does not allow a more exact approach. Textile mills have a signifcantly higher energy intensity than the clothing industry, which manufactures the clothing in downstream processes. The assumed average energy intensities for both the textile and leather sections of the industry are estimated on the basis of the overall energy demand for both industries according to the IEA World Energy Statistics and the GDP shares.


**Table 10** Projected electricity and process heat demands for the textile and leather industry to 2050

# *Projection of the Textile and Leather Industry Energy Demand and CO2 Emissions*

Analogous to the previous industry energy and emissions projections, Table 10 shows the results for the textile and leather industry. All values are calculated on the basis of the documented assumptions. Based on the production processes typical of the industry, it is assumed that the process heat demand does not exceed the temperature level of 100 °C. The 1.5 °C OECM pathway requires that the global textile and leather industry decarbonizes the required energy demand entirely by 2050, whereas a reduction by almost 50% seems achievable by 2030.

# **Decarbonization Pathways for Services**

The *Service* sector contributes 65% of the global GDP (US\$56.9 trillion in 2019). In this analysis, we use the IEA World Energy Balances as the basis for the energy statistics, which defne three main sub-sectors: *Industry*, *Transport*, and *Other Sectors*. Although *Industry* and *Transport* overlap with corresponding GICS classifcation used for the 1.5 °C OECM sectorial pathways, to a large extent, the *Service* sector is scattered across several GICS sectors and the IEA *Other Sectors* and *Industry* groups. In this section, we describe four service sectors that supply essential goods:


The combined share of the global energy demand of these sectors is about 7.5%, which is relatively minor. Although their energy demand is low and their current energy-related CO2 emissions contribute only 6% to global CO2 emissions, their non-energy GHG emissions are signifcant. Agriculture and forestry are among the main emitters of non-energy CO2, methane (CH4), and nitrous oxide (N2O) referred to in climate science as *AFOLU* (agriculture, forestry, and other land uses) emissions.

# *Global Agriculture and Food Sector*

The *Agriculture & Food* sector is an essential economic sector that contributes to food security, livelihoods, and well-being. Valued at US\$3.5 trillion, agriculture, forestry, and fsheries (AFF) accounted for 4% of the global GDP in 2019, with the largest contributions from China and India. The value added in agriculture alone was US\$0.2 trillion. Value is also added in some of the manufacturing sectors supported by AFF. In 2018, the manufacture of food and beverages contributed S\$1.5 trillion and the manufacture of tobacco products contributed U&S\$167 billion.

# *Energy Demand Projection for the Global Agriculture and Food Sector*

Although energy is an important input to agriculture, the sector accounts for only 2.2% of the total fnal energy consumption globally, with oil and oil products meeting most of this demand. Generally, as agriculture is industrialized, this energy consumption increases. In regions where most agricultural systems are industrialized, effciency gains may have plateaued (in the USA, after a peak in 2006) and the sectorial fnal energy consumption may even have decreased (in EU, 10.8% reduction since 1998).

However, the global food system is estimated to account for almost one-third of the world's total fnal energy demand. In high-GDP countries, approximately 25% of the total sectorial energy is consumed behind the farm-gate (in agriculture, including in fsheries): 45% in food processing and distribution, and 30% in retail, preparation, and cooking. In low-GDP countries, a smaller share is spent on the farm and a greater share on cooking.

The estimated global population growth is based on UN population projections and will decrease evenly from about 1% per year in 2020 to 0.5% per year in 2050.


**Table 11** Energy demand projection for agriculture and food processing

The food production volumes for each product will develop accordingly. No dietary or life-style changes are assumed in estimating the future energy demand of the agriculture and food-processing sector. In addition to food for human consumption, agricultural products are also required for animal feed.

The majority of the energy demand is estimated to be for fuel for agricultural machinery, such as tractors and harvesters, whereas 30% of the energy is electricity. Effciency gains are assumed to be higher in the agriculture sector—0.8–1% per year—than in the food-processing industry.

Table 11 shows the calculated energy demand broken down according to the electricity, heat, and fuel requirements for the agriculture and food-processing sector.

# *Global Forestry and Wood Sector*

Forestry contributes to food security, livelihoods, and well-being, supports terrestrial ecosystems and biodiversity, and provides (human)-life-sustaining ecosystem services, and forests act as carbon sinks. Value is also added by some of the manufacturing sectors supported by forestry. In 2018, wood and wood products contributed US\$183 billion, and paper and paper products contributed US\$324 billion to the global economy. Together with agricultural manufacturing, this is about 18% of the value added in total manufacturing globally.

Globally, 30% of all forests are used for production. Of this 30%, about 1.15 billion ha of forest are primarily used for the production of wood and non-wood forest products, and another 749 million ha are designated for multiple uses. In contrast, only 10% is allocated for biodiversity conversation, although more than half of all forests have management plans.

The energy demand for forestry was calculated both as the energy intensity multiplied by the global GDP for this sector, as shown in Table 12, and by subtracting the calculated energy for agriculture from the combined energy demand for agriculture and forestry provided by IEA. This dual calculation of the energy intensity for forestry was confrmed again with data from the literature (Table 13).

# *Global Fisheries Sector*

About 7% of total protein intake globally is from seafood. Over 200 million tonnes of fsh and seafood are produced annually. According to the Organisation for Economic Co-operation and Development (OECD), the fsheries industry employs over 10% of the world's population. Whereas the overall food fsh consumption expanded by 122% between 1990 and 2018, the global capture fsheries—fsh that are caught from natural environments with various fshing methods—only grew by 14%. The main increase in fsh 'production' was in aquaculture, the output of which increased fve-fold. However, the percentage of fsh stocks caught in the open ocean within biologically sustainable levels decreased from 90% in 1909 to only 65.8% in 2018. The economic (frst sale) value of the global fshing industry in 2018 was estimated at US\$401 billion, of which US\$250 billion was from aquaculture production.

Although the fshing industry plays a signifcant role in the food supply and economic income of a large part of the global coastal population, its share of the global energy demand is minor, at <0.1% of the global energy demand. However, in the OECM, we developed a specifc scenario for fsheries because of their importance for small island states. Subsistence fshing is a key economic pillar of island nations in the Pacifc, the Indian Ocean, and the Caribbean. Over the past decades, large fshing vessels have disputed the traditional fshing grounds of local indigenous people.

Among the most unsustainable fshing methods is bottom trawling by large vessels, which accounts for about one-quarter of the global fsh catch. Traditional artisanal fshing boats, which are either entirely unpowered or powered by small outboard engines, cannot compete with industrial fshing vessels. Increasing fuel costs make it increasingly uneconomic for fshermen, because fuel costs often exceed the income from fshing. Moreover, most island states still rely on expensive diesel generators to provide electricity for households and cooling equipment for food preservation.

The economic value of the fshery industry is assumed retain its current global GDP share of 0.2% and to increase, according to the growth projection for global GDP, from US\$272 billion in 2019 to over US\$700 billion in 2050. However, the


**Table 12** Global economic development of the forestry, wood, and wood products industry

**Table 13** Energy demand for the forestry and wood products industry


proportions of marine fshing, aquaculture, and inland fshing will change signifcantly in favour of aquaculture. Table 14 shows all the key assumptions used to calculate the 1.5 °C pathway for fsheries.

The projected development of fsh production, in million tonnes per year, is certainly arguable and no forecasts of the fsh production volumes over the next 30 years are available. Therefore, it is assumed that the volumes of wild fsh catch and fsh from aquaculture will plateau at the 2020 levels, whereas the market value will


**Table 14** Key assumptions for the energy demand projections of the global fsheries industry

steadily increase. The rationale behind this is that marine fshing will be unable to increase the volume of catch, whereas the costs and economic value per tonne of fsh will continue to increase. The catch per unit effort (CPUE)—the amount of energy per tonne—is assumed to remain stable. In this case, the longer distances and sailing times required to catch one tonne of fsh can be compensated by the increased energy effciency of fshing vessels.

Industrial motor power [GW] 87 90 84 77 69 52

[PJ/Mt fsh] 6 6 6 6 6 6

Catch per unit effort (CPUE)—Energy

Units

The 1.5 °C OECM pathway for the fshing industry suggests moving away from large-scale fsh trawlers towards a more decentralized feet of fshing boats.

In terms of the fshing vessel feet, 2.07 million vessels were registered in 2019: 1.16 million were unpowered, 1.63 million were powered artisanal vessels, and 0.43 million were industrial vessels. The overall motor power of the global fshing feet is estimated have a capacity of 144 GW, 87GW of which is from industrial vessels. The 1.5 °C pathways assumes that the power artisanal fshing vessels will steadily increase in number at the expense of industrial vessels, which will lose market shares by volume in a stable fsh market.


**Table 15** Projected energy demand for global fsheries industry

Table 15 shows the resulting energy demands under the documented assumptions. However, the available data on the energy demand of fshing vessels is sparse and the results are estimates. More research is required to develop more-detailed scenarios for and around the fshing industry, their vessels, and electrifcation concepts for artisanal fshing boats.

# *Overview of the Global Water Utilities Sector*

Water is important for basically every process that supports human life on Earth. Potable drinking water of high quality is therefore a basic requirement for the health of humans, the environment, and an intact economy. Therefore, the economic value of water utilities is far beyond the monetary value of this industry. Although the projection of future energy demands for various sectors in this analysis is based on economic values, the energy demand projections for water utilities must be based on production volumes.

According to the OECD, 70% of all water abstracted is used for agriculture. Whereas freshwater dominates the total water extracted, desalination plants are an important parameter because their consumption of energy is high. However, water extraction by desalination plants constitutes only 0.2% of global water extraction. Globally, about one-third of all countries, representing 80% of the global population, are connected to sewerage treatment plants. Table 16 shows the assumed quantities of global water withdrawn—broken down by usage sector—which form the basis for the energy demand projections for water utilities.

# *Projections of the Energy Demand and CO2 Emission for Water Utilities*

The projected global energy demand for water utilities was calculated with the documented assumed global quantities of water required and energy intensities (Table 17). However, the main GHG emissions from water utilities do not originate


**Table 16** Assumed quantities of global water withdrawn, used to predict the energy demands for water utilities

**Table 17** Projected global energy demand for water utilities


from energy-related CO2, but from methane and N2O (or 'laughing gas'), which have signifcant greenhouse potential.

# **Decarbonization Pathways for Buildings**

The *Buildings* sector is responsible for 39% of process-related GHG emissions globally and for almost 32% of the global fnal energy demand, making the *Buildings* sector pivotal in reducing the global energy demand and climate change. With the increasing rates of population growth and urbanization, the building stock is projected to double in developing regions by 2050, so reducing the global energy demand will become challenging. Together with these challenges, new building stocks in developing regions will simultaneously provide opportunities for energyeffcient construction, which could substantially reduce the global energy demand. In developed regions, opportunities to reduce the energy demand will predominantly involve renovating the existing building stock.

To develop detailed energy demand projections for the regional and global *Buildings* sectors, the *High-Effciency Buildings Model* (HEB) was used. The HEB is based on a bottom-up approach and includes rather detailed technological information for the building sector. The model is based on socio-economic data, including population growth rates, urbanization rates, and foor areas per capita. The HEB model uses four different scenarios to understand the dynamics of energy use and to explore the potential of the buildings sector to mitigate climate change by exploiting various opportunities. The four scenarios are:


# *Final Energy Use for Space Heating and Cooling under the HEB Scenarios*

The fnal energy use for space heating and cooling will largely depend upon the calculated foor area. After the foor area is calculated for each region, the thermal energy use is calculated. Like the foor area calculations, thermal energy use is calculated for the four different scenarios.

Key regions, such as China, EU-27, and India, consume most of the global energy, so it is important to know how the building sectors in these regions will perform under different scenarios. Regions such as the USA and EU-27 have much

greater potential to reduce space-heating- and space-cooling-related energy use with the help of best practices.

# *1.5 °C OECM Pathway for Buildings*

Based on the results of the detailed HEB model analysis, the *Deep Effciency* scenario was chosen for commercial buildings and the *Moderate Effciency* scenario for residential buildings. These scenarios were chosen after stakeholder consultation with representatives of the respective industries, members of the Carbon Risk Real Estate Monitor (CRREM), the Net-Zero Asset Owner Alliance, and academia. To integrate the buildings sector into the 1.5 °C pathway as part of the OECM, consistent with all other industry and service sectors and the transport sector, the selection of one specifc pathway for the buildings sector as a whole was necessary.

Table 18 shows the assumed development of foor space for residential and commercial buildings, which was taken from the HEB analysis and the projected economic development of the construction sector. The increase in the construction industry is based on the overall global GDP, developed as documented in Chap. 2, and is therefore not directly related to the HEB foor space projections. The direct link between both parameters was beyond the scope of this analysis and is therefore highlighted as a potential source of error.

Table 19 shows the calculated annual energy demand for residential and commercial buildings and for the construction industry. The energy demand consists of the energy required for space heating and cooling ('heating energy') and the electricity demand, which includes all electrical applications in the buildings but excludes electricity for heating and cooling. This separation is necessary to harmonize the input data from the HEB (which do not include electricity for household applications such as washing machines, etc.) with the OECM.

The electricity demand for residential buildings is based on the bottom-up analysis of households documented in Sect. 3.1.2. The electricity demand for the service


**Table 18** OECM—Global buildings: projected foor space and economic value of construction


**Table 19** OECM—Global buildings: Calculated annual energy demand for residential and commercial buildings and construction

sector is based on a break down of electricity and heating in 2019 across all service sectors. The future values until 2050 are based on the projections for the analysed service and industry sectors.

# **Decarbonization Pathways for Transport**

The transport sector consumed 28% of the fnal global energy demand in 2019 and its decarbonization potential is therefore among the most important of all industries. Given its size and diversity, not only with regard to different transport modes and technologies, but also regional differences, it is also one of the most challenging sectors. In 2019, transport consumed 78% of the total oil demand globally. Therefore, the transition from oil to electric drives and to synthetic fuels and biofuels is key to achieving the goals of the Paris Climate Agreement. The rapid uptake of electric mobility, combined with a renewable power supply, is the single most important measure to be taken to remain within the carbon budget of the 1.5 °C pathway.

As a result of the restricted mobility imposed to stop spread of the COVID-19 virus, the global pandemic led to a signifcant reduction in the oil demand, especially for road transport and aviation, which are responsible for nearly 60% of oil use. The global oil demand is estimated to have dropped by 8% in 2020. At the time of writing, the global pandemic is still on-going, although travel restrictions have been relaxed in many countries, increasing in the transport demand relative to that in 2020. In our transport demand projections, we assume that the demand will continue to increase to pre-pandemic levels by 2025.

The majority of all passenger transport—in terms of overall kilometres—is by road. However, international freight transport is more strongly dominated by rail and shipping, which account for 45% of all tonne–kilometre. The high effciency of rail and shipping means that their share of the global transport energy demand is small relative to the share of global tonnage transported.

# *Shipping and Aviation: Dominated by Combustion Engines for Decades to Come*

Navigation will probably remain predominantly powered by internal combustion engines (ICEs) in the next few decades. Therefore, we did not model the electrifcation of freight vessels. However, pilot projects using diesel hybrids, batteries, and fuel cells are in preparation. We assumed the same increase in the share of bio- and synthetic fuels over time as in the road and rail sectors.

In aviation, energy effciency can be improved by measures such as winglets, advanced composite-based lightweight structures, powertrain hybridization, and enhanced air traffc management systems. We project a 1% annual increase in effciency on a per passenger–kilometre (pkm) basis and a 1% annual increase in effciency on a per tonne–kilometre (tkm) basis (Tables 20 and 21).

A key target for the global transport sector is the introduction of incentives for people to drive smaller cars and use new, more-effcient vehicle concepts. It is also vital to shift transport use to effcient modes, such as rail, light rail, and buses, especially in large expanding metropolitan areas. Furthermore, the 1.5 °C scenario cannot be implemented without behavioural changes. It is not enough to simply exchange vehicle technologies, but the transport demand must be reduced in terms of the kilometres travelled and by an increase in 'non-energy' travel modes, such as cycling and walking (Table 22).

The proportion of battery electric vehicles (BEVs) among all passenger cars and light commercial vehicles in use is projected to be between 8% and 15% by 2030. This will require a massive build-up of battery production capacity in the coming years. New car sales will already be dominated by battery electric passenger vehicles in 2030 under the 1.5 °C scenario. However, with an assumed average lifetime of 15 years for ICE passenger cars, the existing car feet will still predominantly use ICEs. Under the assumption that new ICE passenger cars and buses will not be


**Table 20** Aviation—energy demand and supply

**Table 21** Shipping—energy demand and supply



**Table 22** Road transport—energy demand and supply

produced after 2030, BEVs will dominate the passenger vehicle feet of 2050 under the 1.5 °C scenario. OECD countries and China are assumed to lead the development of BEVs and therefore to have the highest shares, whereas Africa and Latin America are expected to have the lowest BEV shares. Fuel-cell-powered passenger vehicles are projected to play a signifcantly smaller role than BEVs and will only be used for larger vehicles, such as SUVs and buses.

# **Transition of the Energy Industry to (Net)-Zero Emissions**

To reduce emissions to zero in line with a 1.5 °C increase in global temperature, the use of coal, oil, and gas must be phased out by at least 56% by 2030. However, current climate debates have not involved an open discussion of the orderly withdrawal from the coal, oil, and gas industries. Instead, the political debate about coal, oil, and gas has continued to focus on supply and price security, neglecting the fact that mitigating climate change is only possible when fossil fuels are phased out.

The primary energy demand analysis—and therefore the projections for the primary energy industry and possible future operation strategies—is the product of the energy demand projections for all end-use sectors and the energy supply concept. The challenge for the primary energy industry is to supply energy services for sustained economic development and a growing global population while remaining within the global carbon budget to limit the global temperature rise to 1.5 °C.

The trajectories for oil, gas, and coal depend on how quickly an alternative energy supply can be built up and how energy consumption can be reduced technically and/or by behavioural changes. The OECM 1.5 °C pathway represents such a trajectory and is based on a detailed bottom-up sectorial demand and supply analysis. However, for the primary energy industry, it is important to assess whether or not new oil, gas, or coal extraction projects are required to meet the demand, even under an ambitious fossil-fuel phase-out scenario.

A scenario designated the *Existing International Production Trajectory* ('No Expansion') was developed and modelled, specifcally to understand what global fossil fuel production will look like under the following assumptions:

	- Coal: 2% per year
	- Oil: 4% per year onshore and 6% per year offshore
	- Gas: 4% per year on- and offshore

The *No Expansion* scenario was compared with the OECM 1.5 °C pathways for coal, oil, and gas to understand whether security of supply is possible under the immediate implementation of a 'stop exploration' policy.

The decline rates for oil, gas, and coal that would result from the implementation of the 1.5 °C pathway and the assumed annual production decline rates for oil, gas, and coal are compared in Table 23.

Our analysis shows that even with no expansion of fossil fuel production, the current productions levels—especially for coal—will exhaust the carbon budget associated with the 1.5 °C target before 2030. Without the active phase-out of fossilfuel production, production will signifcantly surpass what can be produced under a 1.5 °C scenario by 2025 onwards, for all fossil-fuel types.

# *Power and Gas Utilities*

Throughout the description of the OECM 1.5 °C pathway, the increased electrifcation of the transport and heating sectors is the overarching scenario narrative, and runs across all sectors. Increased electrifcation will lead to 'sector coupling', with


**Table 23** Decline rates required to remain within the 1.5 °C carbon budget versus the production decline rates under 'no expansion'

the interconnection of the heating and transport sectors with the electricity sector. The sectors are still largely separate at the time of writing. However, the interconnection of these sectors offers signifcant advantages in terms of the management of the energy demand and the management of generation with storage technologies. The synergies of sector coupling in terms of the infrastructural changes required to transition to 100% renewable energy systems are well documented in the literature.

The OECM 1.5 °C pathway will lead to an annual increase in electricity generation from about 26,000 TWh in 2019 to 76,000 TWh by 2050, which will require a signifcant increase in renewable generation capacity (Table 24). Although there is clear agreement that the global electricity demand will increase, the predictions of how this electricity will be generated are very different. Despite the signifcant growth in renewable power generation during the last decade, short-term projections from the IEA still expect that fossil-fuel-based power generation will continue to grow.

The changes in gas utilities under the OECM 1.5 °C scenario are more profound than those for power utilities, because the main product—natural gas—will be phased out globally by 2050. However, the OECM acknowledges the signifcant value of the existing gas infrastructure and recommends that the gas distribution network be re-purposed to utilize it for the future decarbonized energy supply. According to the Global Energy Monitor, 900,757 km of natural gas long-distance transmission pipelines were in operation globally at the end of 2020. Research has shown that there are no fundamental technical barriers to the conversion of natural gas pipelines for the transport of pure hydrogen.

Table 25 shows the development of the demand and supply of natural-gas-derived electricity for the global utilities sector under the OECM 1.5 °C pathway. Global renewable electricity generation will increase signifcantly, by a factor of 10. The projected transition of gas utilities to the distribution of hydrogen and synthetic fuels will represent 50% of their sales by 2045. Therefore, the transition is assumed to have a lead time of about 10 years for the implementation of the required technical and regulatory changes.

Based on the OECM 1.5 °C decarbonization pathway, we propose a horizontal integration of all three sub-sectors, to combine the core areas of expertise and avoid


**Table 24** Renewable power, heat capacities, and energy demand for hydrogen and synthetic fuel production under the 1.5 °C scenario

stranded assets by repurposing the existing fossil-fuel infrastructure, such as pipelines.

Figure 2 shows a possible structure for the decarbonized *Energy* and *Utility* sectors. The (primary) energy industry will focus on utility-scale power generation and the production of hydrogen and synthetic fuels for the supply of energy and chemical feedstock. Gas utilities will focus on the transport of hydrogen and fuels and offer decentralized hydrogen production and storage services to the power sector. Power utilities will concentrate on the power grid, the management of the electricity system, and the integration of decentralized renewable power generation and storage systems, including those from 'prosumers'.


**Table 25** Global utilities sector—electricity and gas distribution under the OECM 1.5 °C scenario

# **Climate Sensitivity Analysis—All Greenhouse Gases and Aerosols**

The IPCC Assessment Report 6 (AR6), published in August 2021, contains fve scenarios, each of which represents a different emissions pathway. These scenarios are called the *Shared Socioeconomic Pathway* (*SSP*) scenarios. The most optimistic

**Fig. 2** One Earth Climate Model: Possible structure of a decarbonized Energy and Utilities industries

scenario, in which global CO2 emissions are cut to net zero around 2050, is the SSP1-1.9 scenario. The number at the end (1.9) stands for the approximate end-ofcentury radiative forcing, a measure of how hard human activities are pushing the climate system away from its pre-industrial equilibrium. The most pessimistic is SSP5-8.5.

*Climate Resource*<sup>1</sup> has added CO2 emissions that fall under other fossil fuel and industrial activities, such as fugitive emissions, cement production, and waste disposal and management, from the SSP1-1.9 scenario, with energy-related CO2 emissions data of the OECM 1.5 °C pathway.

Here, we provide the global mean probabilistic temperature projections, including their medians and 5–95% ranges, for the OECM scenarios analysed (Fig. 3). These probabilistic ranges are sourced from the underlying 600 ensemble members, which are calibrated against the IPCC AR6 WG1 fndings.

<sup>1</sup>This section is based on the analysis of *Climate Resource* under contract to the University of Technology Sydney (UTS) as part of the Net-Zero Sectorial Industry Pathways Project (UTS/ISF 2021). The study is an update of the previous One Earth Climate Model (OECM) publication (Teske et al. 2019). However, the Generalized Quantile Walk (GQW) methodology used (Meinshausen & Dooley 2019) has been developed further. The energy and industrial CO2 emissions pathways are based on the OECM 1.5 °C energy scenario described in previous chapters, whereas the non-CO2 GHG emission time series have been described with the advanced GQW methodology.

**Fig. 3** Probabilistic global mean surface air temperature (GSAT) projections relative to 1850–1900

Similar to the SSP1-1.9 scenario in IPCC AR6 WG1, the OECM 1.5 °C pathways slightly overshoot the 1.5 °C pathway in their median values during the middle of the century, before dropping back to below 1.5 °C warming towards the end of the century. The likelihood that the OECM 1.5 °C scenario will stay below 1.5 °C throughout the century, despite strong mitigation actions, does not exceed 67%. Figure 3 shows the probabilistic global mean surface air temperature (GSAT) projections relative to 1850–1900 for the scenarios analysed.

# **OECM 1.5 °C Pathway for the Global Energy Supply**

The supply side of this 1.5 °C energy scenario pathway builds upon modelling undertaken in an interdisciplinary project led by the University of Technology Sydney (UTS). The project modelled sectorial and regional decarbonization pathways to achieve the Paris climate goals—to maintain global warming well below 2 °C and to 'pursue efforts' to limit it to 1.5 °C. That project produced the One Earth Climate Model (OECM), a detailed bottom-up examination of the potential to decarbonize the energy sector. The results of this on-going research were published in 2019 (Teske et al., 2019). For the present analysis, the 1.5 °C supply scenario has been updated to match the detailed bottom-up analysis for the industry and service sectors, as well as the buildings and transport sectors.

# **Global Final Electricity Demand**

Figure 4 shows the development of the fnal electricity demand by sector between 2019 and 2050. The signifcant increase in the demand is due to the electrifcation of heat, for both space and process heating, and to a lesser extent for the manufacture of hydrogen and synthetic fuels. The overall global fnal demand in 2050 will be 2.5 times higher than in the base year, 2019. In 2050, the production of fuels alone will consume the same amount of electricity as the total global electricity demand in 1991. Therefore, the demand shares will change completely, and 47% of all electricity (Fig. 5) will be for heating and fuels that are mainly used in the industry and service sectors. Electricity for space heating—predominantly from heat pumps—will also be required for residential buildings.

Global power plant capacities will quadruple between 2019 and 2050, as shown in Fig. 6. Capacity will increase more than actual power generation because the capacity factors for solar photovoltaic and wind power are lower than those for fuelbased power generation. By 2030, solar photovoltaic and wind will make up 70% of the generation capacity, compared with 15% in 2019, and will clearly dominate by 2050, with 78% of the total global generation capacity.

However, fossil-fuel-based power generation must be decommissioned, and the global total capacity will not increase over current levels but will remain within the greenhouse gas (GHG) emissions limits. By 2025, global capacities of 63 GW from hard coal power plants and 55 GW from brown coal power plants must go offine. All coal power plants in OECD countries must cease electricity generation by 2030, and the last coal plants must fnish operation globally by 2045 to remain within the carbon budget for power generation required to limit the global mean temperature increase to +1.5 °C. Specifc CO2 emission per kilowatt-hour will decrease from 509 g of CO2 in 2019 to 136 g by 2030, and 24 g in 2040, to be entirely CO2 free by 2050.

**Fig. 4** Electricity demand by sector under the OECM 1.5 °C pathway in 2019–2050

**Fig. 5** Electricity demand shares by sector under the OECM 1.5 °C pathway in 2019 and 2050

# *OECM 1.5 °C Pathway for Global Space and Process Heat Supply*

Services and buildings usually do not require temperatures over 100 °C. Therefore, the supply technologies are different from those of the industry sector, which requires temperatures up to 1000 °C and above. The overall fnal heat demand will increase globally under the OECM 1.5 °C pathway, but the demand shares will

**Fig. 7** Electricity demand shares by sector under the OECM 1.5 °C pathway in 2019 and 2050

change signifcantly. With energy effciency measures for buildings, the overall space heating demand will decrease globally, even with increased foor space. However, the industrial process heat demand is projected to increase because energy effciency measures will not compensate for the increasing production arising from the expected increase in global GDP to 2050. In 2019, the industry sector consumed 43% of the global heat demand and the service and buildings sector the remaining 57%. By 2050, these shares will be exchanged, and the industry sector will consume close to 60% of the global heat demand (Fig. 7).

Table 26 shows the assumed trajectory for the generation of industry process heat between 2019 and 2050. In 2019, gas and coal dominated global heat production. Renewables only contributed 9%—mainly biomass, and electricity had a minor share of 1%. District heat—mainly from gas-fred heating plants—contributed the remaining 7% of the process heat supply, whereas hydrogen and synthetic fuels contributed no measurable proportion. The global OECM 1.5 °C pathway phases out coal and oil for process heat generation between 2035 and 2040, and gas is phased out as the last fossil fuel by 2050. The most important process heat supply technologies are electric heat systems, such as heat pumps, direct electric resistance heating, and arc furnace ovens for process heat; the share will increase to 22% by 2030 and to 60% by 2050. Bio-energy will remain an important source of heat, accounting for 25% in 2050—2.5 times more than in 2019. Synthetic fuels and hydrogen are projected to grow to 8% of the total industry heat supply by 2050.

# *Global Primary Energy Demand—OECM 1.5 °C Pathway*

The global primary energy demand under the OECM 1.5 °C pathway is shown in Table 27. Primary energy includes all losses and defnes the total energy content of a specifc energy source. In 2019, coal and oil made the largest contributions to the


**Table 26** Heat supply under the OECM 1.5 °C pathway

**Table 27** Global primary energy demand and supply under the OECM 1.5 °C pathway


global energy supply, followed by natural gas, whereas renewable energies contributed only 15%. The table also provides the projected trajectories for supplies for non-energy uses, e.g., oil for the petrochemical industry. The OECM does not phase-out fossil fuels for non-energy use, because their direct replacement with biomass is not always possible. A detailed analysis of the feedstock supply for nonenergy uses was beyond the scope of this research.

# **Global CO2 Budget**

The remaining carbon budget for each of the following sectors has been defned based on the bottom-up demand analysis of the 12 main industry and service sectors, as documented in Chaps. 5, 6, 7, and 8. Each of those industry and service sectors must complete the transition to fully decarbonized operation within the carbon budget provided. It is very important that the carbon budget shows the cumulative emissions up to 2050, and not the annual emissions. A rapid reduction in annual emissions is therefore vital.

The shares of the cumulative carbon budget required to achieve the 1.5 °C net-zero target are shown in Fig. 8. The total energy-related CO2 for the aluminium industry between 2020 and 2050 is calculated to be 6.1 Gt, 1.5% of the total budget. For the steel industry, the remaining budget is 19.1 Gt of CO2 (4.8%), whereas the chemicals industry has the highest carbon budget of 24.8 GtCO2 or 6.2% of the total carbon budget. All other remaining industries can emit 27.1 GtCO2 (6.8%), and all other energy-related activities, such as for buildings, transport, and residential uses, have a combined remaining emissions allowance of 323 GtCO2, or 80.7% of the budget.

**Fig. 8** Global carbon budget by sub-sector under 1.5 °C OECM pathway in 2020–2050

# **Scopes 1, 2, and 3—Global Summary**

A global assessment of *Scopes 1*, *2*, and *3* for the whole *Industry* sector is a new research area, and changes had to be made to the method of determining those emissions, which was originally developed by the World Resource Institute (WRI), as documented in Chap. 4.

The OECM methodology differs from the original concept primarily insofar as the interactions between industries and/or other services are kept separate. A primary class is defned for the primary energy industry, a secondary class for the supply utilities, and an end-use class for all the economic activities that consume energy from the primary- or secondary-class companies, to avoid double counting. All the emissions by defned industry categories (e.g., those defned by GICS) are also separated, streamlining the accounting and reporting systems. The volume of data required is reduced and reporting is considerably simplifed with the OECM methodology.

Figure 9 shows the global energy-related *Scope 1*, *2*, and *3* CO2 emissions in 2030 as a Sanky fow chart. The primary energy emissions are on the left and the end-use-related emissions are on the right. The carbon budgets remain constant, from production to end-use, apart from losses and statistical differences. A simplifed description is that all *Scope 1* emissions are on the left, with the primary energy industry as the main emitter, and all *Scope 3* emissions are on the right, with the consumers of all forms of energy and for all purposes as the main emitters. In the secondary energy industry, utilities are the link between the demand of end-users and the supply by the primary energy industry. The fgure also shows the complex interconnections between demand and supply.

# **Nature-Based Carbon Sinks: Carbon Conservation and Protection Zones**

# *Ecosystem Restoration Pathways*

The OECM model presents a 1.5 °C-compatible scenario combining ecosystem restoration with deep decarbonization pathways, called the RESTORE scenario. The fve ecosystem restoration pathways involve forests and agricultural lands: forest restoration, reforestation, reduced harvest, agroforestry, and silvopasture.

The median gross cumulative potential of additional CO2 removal under the fve ecosystem restoration pathways is 100 Gt of carbon (C) until 2100, as shown in Table 28. The peak annual sequestration rate for all ecosystem restoration pathways (forest restoration, reforestation, reduced harvest, agroforestry, and silvopasture) is 2.5 Gt of carbon (GtC) per year, although this rate is only maintained for 1–2 decades. The average annual sequestration rate from 2020 to 2100 will be 1.2 GtC


**Table 28** Summary statistics for the cumulative uptake of CO2 in all pathways

per year. This is approximately 9% higher than the carbon uptake that would occur if the same land management pathways were modelled in a dynamic global vegetation model (DGVM). The difference is largely due to the inclusion of the soil carbon response to land-use changes in the DGVM (Littleton et al. 2021). This removal will be offset by on-going net land-use emissions.

# *Creating Carbon Conservation Zones (CCZ)*

*The role of nature and ecosystem services as climate solutions is gaining increasing attention. As well as their climate mitigation and carbon sequestration potential, ecosystem approaches have co-benefts that contribute to sustainable development goals in terms of livelihoods, productivity, biodiversity conservation, health, and ecosystem services. However, it is important to note that* even with ambitious landuse restoration, carbon removal can still only compensate for, at most, 15% of current emissions. The vast majority of emissive activities must cease if we are to achieve an approximately 1.5 °C target, and all the available removal strategies are required to achieve net-negative emissions pathways and reduce the atmospheric concentrations of CO2.

*Feasible approaches to CDR using land-based mitigation options must be predicated on a 'responsible development' framework that includes broader social and environmental objectives. Carbon conservation zones, which implement different ecosystem approaches, must address these broader objectives:*


Forests and forest products are important parts of a number of supply chains for food, consumer goods, transport, etc., and companies and investors can play an important role in protecting and conserving nature through corporate commitments and by infuencing their downstream supply chains.

# **Conclusion—High-Level Summary**

To comply with the Paris Climate Agreement and limit the global mean temperature rise to +1.5 °C, rapid decarbonization of the energy sector with currently available technologies is necessary, and is possible.

However, to achieve the transformation to a fully renewable energy supply, all available effciency potentials must be combined to reduce the total demand. To reach Net Zero by 2050, the complete phase-out of fossil fuels for all combustion processes is essential.

For the Industry sector, the transition from fossil-fuel-based process heat to renewable energy or electrical systems is the single most important measure. The further reduction of non-energy-related process emissions—mainly from cement and steel manufacture—by altering or optimizing manufacturing processes is also essential. The remaining process emissions might be compensated by natural carbon sinks, so the Industry sector must actively support the Service sector in terms of soil regeneration and reforestation measures.

For the Service sector, especially agriculture and forestry, reducing GHG emissions must clearly involve reducing the greenhouse gas (GHG) emissions arising from land-use changes. Increasing yield effciency to avoid the further expansion of agricultural land at the expense of forests and other important ecosystems is key. However, feeding the growing world population without increasing the area committed to agriculture will require more than just an increase in technical effciency. Moreover, there seems to be no alternative to reducing the consumption of meat and dairy products.

The Forestry sector is the single most important sector for the implementation of nature-based carbon sinks. Deforestation must cease immediately. Reforestation with native trees and plants that are typical of specifc regions and climate zones must replace the forest areas that have been lost since 1990.

To reduce the demand of the Transport sector, a shift from resource-intensive air and road transport to more-effcient and electrifed means of transport is required, together with an overall reduction in transport activity, especially in high-income countries. Phasing-out the production of combustion engines for passenger cars by 2030 and introducing synthetic fuels for long-distance freight transport are essential elements for the future transportation sector. Even with this ambitious goal, the full decarbonization of the road transport sector will not be achieved before 2050, because cars are used, on average, for 15–20 years. There is also signifcant potential for effciency gains in shipping and aviation. However, due to the foreseeable further growth in traffc volume and the lack of alternatives, the large-scale use of synthetic fuels from renewable electricity will also be necessary for these modes of transport. Since not all regions will be able to produce this with domestic resources at reasonable costs, a global trade of these new energy sources must be established.

The decarbonization of the Buildings sector will require a signifcant reduction in the energy demand for climatization—heating and cooling—per square metre. A key result of our research is that the global energy demand for buildings can be halved with currently available technologies. The utilization of this effciency potential will require high renovation rates and changed building codes for new constructions. The widespread use of heat pumps and heat grids are important elements on the supply side. In some areas, however, the supply of renewable gases can substitute today's natural gas consumption with a long-term perspective, especially where there is an industrial gas demand. The conversion of today's gas networks and the local/regional availability of resources for the production of green gases play a decisive role here.

Signifcant electrifcation across all sectors before 2030—especially for heating, process heat, and to replace combustion engines in the Transport sector—is the decisive and most urgent step. Increased electrifcation will require sector coupling, demand-side management, and multiple forms of storage (for heat and power), including hydrogen and synthetic fuels. Accelerating the implementation of renewable heat technologies is equally important because half the global energy supply must derive from thermal processes by 2050.

The transition of the global energy sector will only be possible with signifcant policy changes and reforms in the energy market.

The complete restructuring of the Energy and Utilities sectors is required. The primary Energy sector—the oil, gas, and coal industry—must wind-down all fossil -fuel extraction and mining projects and move towards utility-scale renewable energy projects, such as offshore wind and the production of hydrogen and synthetic fuels.

Power utilities will play a key role in providing the rapidly increasing electricity demand, generated from renewable power. The nexus of the global energy transition will be the power grid. Replacing oil and gas with electricity means that power grids must transport most energy, instead of oil and gas pipelines.

Therefore, the expansion of power grid capacities is one of the most important and also most overlooked measures required. In addition, converting existing gas pipelines and using them for the long-range transport of hydrogen and synthetic methane can signifcantly reduce the infrastructural demands on the power system and increase effciency.

According to the scenario, global transmission and distribution grids must transport at least three times more electricity by 2050 than in 2020. The upgrades and expansion of power grids must start immediately because infrastructure projects, such as new power lines, can take 10 years or more to implement. Conversions of existing gas pipelines will be possible frst where industrial users need large quantities of hydrogen for decarbonized processes.

Limiting the global mean temperature rise to +1.5 °C cannot be achieved by the decarbonization of the energy sector alone. As stated earlier, it will also require signifcant changes in land use, including the rapid phase-out of deforestation and signifcant reforestation. These measures are not alternative options to the decarbonization of the energy sector but must be implemented in parallel. If governments fail to act and mitigation is delayed, we face the serious risk of exceeding the carbon budget. In the One Earth Climate Model (OECM) 1.5 °C pathway, the land-use sequestration pathways complement very ambitious energy-mitigation pathways, and should therefore be regarded as necessary to reduce the CO2 concentrations that have arisen from the overly high emissions in the past, and not as compensatory measures that can be extended indefnitely into the future.

# **Policy Recommendations**

The OECM is an integrated assessment tool for the development of science-based targets for all major global industries in a granularity. It includes the key performance indicators (KPIs) required to make informed investment decisions that will credibly align with the global net-zero objective in the short, medium, and long terms. The key fnding of our work on the OECM 1.5 °C cross-sectorial pathway is that it is still possible to remain within the 1.5 °C limit if governments, industries, and the fnancial sector act immediately. The technology required to decarbonize the energy supply with renewable energy is available, market ready, and in most cases, already cost competitive. The energy effciency measures needed to reduce the energy demand have also been understood for years and can be introduced without delay. Finally, the fnance industry—for instance, the Net-Zero Asset Owner Alliance—is committed to implementing carbon targets for its investment portfolios. However, policies are required to ensure that all measures are implemented in the rather short time frame required.

# *Implementing Short-Term Targets for 2025 and 2030*

To implement the documented short-term targets for 2025 and 2030, the following actions are required:

### **Government Policies**


# **Actions Needed by Industry and Financial Institutions**

# **Industry**


# **Financial Institutions**

	- climate mitigation strategies
	- short- and mid-term target setting
	- target achievements
	- progress of climate solution investments
	- engagement outcomes

# **Contents**

#### **Part I Introduction**



# **List of Figures**



#### List of Figures














# **Part I Introduction**

# **Chapter 1 Introduction**

**Sven Teske**

**Abstract** This is a brief introduction to the status of the international climate negotiations of the United Nations Framework Convention on Climate Change (UNFCCC) and its latest scientifc publications, the status of global greenhouse gas emissions, and the impact of the pandemic on energy-related CO2 emissions. The research focus of this book is presented, and how the second part of the book relates to our frst book *Achieving the Paris Climate Agreement Goals* is explained.

The background to the creation of the book is given. The parameters upon which the authors focused when documenting the assumptions used for all calculations are explained. The results and their derivation are presented.

**Keywords** United Nations Framework Convention on Climate Change (UNFCCC) · Net-zero targets · Achieving the Paris Climate Agreements

The climate and energy debate continues to be high on the political agenda at intergovernmental summits. However, since the publication of our frst edition *Achieving the Paris Climate Agreement Goals* in February 2019, the situation has changed dramatically. The COVID-19 pandemic dominates almost every conversation, both in the private sphere and in international political discussions. For the frst time since the beginning of the United Nations Framework Convention on Climate Change (UNFCCC) with Climate Conference COP1 in 1995 in Berlin, Germany, a conference was cancelled. COP26 was meant to be in November 2020 but had to be pushed back by 12 months in response to the pandemic.

© The Author(s) 2022 S. Teske (ed.), *Achieving the Paris Climate Agreement Goals*, https://doi.org/10.1007/978-3-030-99177-7\_1

S. Teske (\*)

University of Technology Sydney – Institute for Sustainable Futures (UTS-ISF), Sydney, NSW, Australia e-mail: sven.teske@uts.edu.au

As a consequence of travel restrictions and lockdowns in almost all countries worldwide, the oil demand decreased by nine billion barrels per day compared with 2020 (BP, 2021, p. 23). Industry production dropped because workers could not come to work, restaurants had to close, and public life almost came to a halt in many countries. Global energy-related CO2 emissions declined by 5.8% in 2020, equal to about 2 Gt (IEA, 2021)—an unprecedented event. Even the global fnancial crisis of 2009 did not have such a profound impact on global emissions. At the time of writing this book—December 2021—the pandemic persists. However, global CO2 emissions have bounced back and increased by 4.8% in 2021—to almost the same level as before the pandemic.

Extreme weather events (extreme rainfall and foods, cyclones, and bushfres) have increased in frequency. Australia experienced the worst bushfre season on record between September 2019 and March 2020—known as the *Black Summer* (Cook et al., 2021). In June 2020, the Arctic region of Siberia experienced a heat wave with temperatures up to 38 °C and wildfres covering almost one million hectares. The World Meteorological Organization (WMO) recognised this as a new Arctic temperature record (WMO, 2021).

Time is running out. In August 2021, the Sixth Assessment Report (AR6) of the United Nations Intergovernmental Panel on Climate Change (IPCC) was published. The First Assessment Report was launched in 1990 and underlined the importance of climate change as a challenge with global consequences that required international co-operation (IPCC, 2021). Thirty years later, the IPCC states unequivocally that the world is already in the middle of climate change. UN Secretary-General António Guterres said the Working Group's report was nothing less than 'a code red for humanity. The alarm bells are deafening, and the evidence is irrefutable' (UN, 2021).

On the positive side, the international fnance industry is increasingly engaged in the international and national climate debate. Initiatives such as the UN-convened Net-Zero Asset Owner Alliance (NZAOA, 2021) and the Glasgow Financial Alliance for Net Zero (GFANZ, 2021) represent leading fnancial institutions committed to achieving the goals of the Paris Climate Agreement and transitioning their investment portfolios to achieve net-zero greenhouse gas (GHG) emissions by 2050.

Our frst book laid out global and regional 100% renewable energy scenarios with non-energy GHG pathways for +1.5 °C or +2 °C warming scenarios and compared them with a reference case. Those scenarios were calculated under the leadership of the Institute for Sustainable Futures (ISF) at the University of Technology Sydney (UTS) in close co-operation with the German Aerospace Center (DLR) and the University of Melbourne, Australia. The model used for that project became known as the *OneEarth Climate Model* (OECM) in 2020 during the numerous debates that followed the book launch in February 2019.

*The second book focuses on sectorial pathways and provides key performance indicators (KPIs) for industry sectors to limit the global temperature increase to 1.5 °C.*

The OECM is an integrated energy assessment model to be used for developing science-based net-zero targets for all major industries in a granularity and with the KPIs needed to make short-, mid-, and long-term investment decisions. The 1.5 °C emission pathways developed by UTS are no or low overshoot scenarios (SSP 1), as defned by the IPCC. This means that a carbon budget overshoot is avoided and that the CO2 already released is not assumed to be 'removed' by unproven technologies still under development, such as carbon capture and storage. The OECM does take 'technical' negative emissions into account, but only natural carbon sinks, such as forests, mangroves and seaweed, which will compensate for the process emissions that are currently unavoidable, such as those from cement production.

A number of climate modelling organisations, including the Energy Transitions Commission, the Potsdam Institute for Climate Impact Research, the Science-Based Targets Initiative, the Carbon Risk Real Estate Monitor (CRREM), and the World Wide Fund for Nature (WWF), were invited to peer-review the OECM-derived netzero pathways between mid-2020 and mid-2021.

The book documents all the steps in the scenario development and provides a detailed analysis of the main assumptions and scenario narratives. The results of the OECM 1.5 °C pathways for 12 industry and service sectors include the total remaining carbon budget and *Scope 1*, *2*, and *3* emissions for each sector.

# **References**


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

# **Part II State of Research**

# **Chapter 2 Science-Based Industry Greenhouse Gas (GHG) Targets: Defning the Challenge**

**Sven Teske and Thomas Pregger**

**Abstract** Background information is given on the Paris Climate Agreement and the role of nationally determined contributions and net-zero pledges. An overview of historical energy-related CO2 emissions since 1750 and how they relate to economic development, measured in gross domestic product (GDP), is provided, together with the cumulative energy-related CO2 emissions by region. The future energy demand if historical trends in energy effciency and carbon intensity continue until 2050 is projected. The term 'science-based target setting' is defned, and how it relates to the carbon budget published in the Sixth Assessment Report of the IPCC is discussed. The energy-related CO2 emission pathway required to achieve the 1.5 °C target is outlined.

**Keywords** Science-based GHG targets · GHG development · GDP · Population · Nationally determined contributions (NDCs)

S. Teske (\*)

University of Technology Sydney – Institute for Sustainable Futures (UTS-ISF), Sydney, NSW, Australia e-mail: sven.teske@uts.edu.au

T. Pregger

German Aerospace Center (DLR), Institute for of Networked Energy Systems (VE), Department of Energy Systems Analysis, Stuttgart, Germany e-mail: thomas.pregger@dlr.de

#### **Paris Agreement—Article 2**

	- (a) *Holding the increase in the global average temperature to well below 2 °C above pre-industrial levels and pursuing efforts to limit the temperature increase to 1.5 °C above pre-industrial levels, recognising that this would signifcantly reduce the risks and impacts of climate change;*
	- (b) *Increasing the ability to adapt to the adverse impacts of climate change and foster climate resilience and low greenhouse gas emissions development, in a manner that does not threaten food production; and*
	- (c) *Making fnance fows consistent with a pathway towards low greenhouse gas emissions and climate-resilient development.*

To understand the challenges involved in implementing the Paris Climate Agreement, it is helpful to look at historic trends in the world's population, its economic growth, its increasing energy demand, and—as a result of the energy sources chosen—the trajectory of energy-related CO2 emissions. CO2 concentrations are increasing in the global atmosphere, causing global warming (IPCC, 2021). The Paris Climate Agreement Goal is to limit this temperature increase to 1.5 °C above pre-industrial levels. In 2021, a new scientifc report defned the remaining global carbon budget.

# **2.1 The Sixth Assessment Report of the IPCC: Climate Change Is Here**

The Intergovernmental Panel on Climate Change (IPCC) is the United Nations (UN) body that assesses the science related to climate change. In August 2021, it launched the Working Group I contribution to its Sixth Assessment Report (AR6) *Climate Change 2021: The Physical Science Basis*. The IPCC concluded that the emission of GHGs from human activities is responsible for approximately 1.1 °C of warming that has occurred since 1850–1900. Based on the improved observational datasets that are used to assess historical warming and the progress in scientifc understanding of the climate system's response to anthropogenic GHG emissions, the IPCC expects that the increase in the global temperature will reach or exceed 1.5 °C (IPCC, 2021) (Table 2.1).


**Table 2.1** Assumed population and GDP developments by region in 2020–2050

The IPCC also identifed the global carbon budget required to avoid exceeding 1.5 °C. Between 2020 and 2050, the global cumulative CO2 emissions must not surpass 400 GtCO2 if we are achieving this target with 67% likelihood. This likelihood decreases to 50% if total emissions reach 500 GtCO2 (Table 2.2) between 2020 and 2050 (IPCC, 2021).

The IPCC media statement was unusually clear and unambiguous for a highlevel scientifc organization:

Stabilising the climate will require strong, rapid, and sustained reductions in greenhouse gas emissions, and reaching net zero CO2 emissions. Limiting other greenhouse gases and air pollutants, especially methane, could have benefts both for health and the climate… *IPCC media release, Geneva 9 August 2021*

# **2.2 The OneEarth Climate Model: The Context**

The UN-convened *Net-Zero Asset Owner Alliance* is a *Program for Responsible Investment* and a United Nations Environment Programme Finance Initiative (UNEP FI)-supported initiative. The members of the Alliance have committed to transitioning their investment portfolios to net-zero GHG emissions by 2050, consistent with


**Table 2.2** Estimates of remaining carbon budgets and their uncertainties—IPCC AR6, WG1, Technical Summary

*Source*: IPCC AR6, WG1, Technical Summary, Table TS.3, page 150

a (1) Human-induced global surface temperature increases in 1850–1900 and 2010–2019 are assessed to be 0.8–1.3 °C, with a best estimate of 1.07 °C. Combined with a central estimate of the transient climate response to cumulative carbon emissions (TCRE) of 1.65 °C EgC-1, this uncertainty in isolation results in a potential variation in the remaining carbon budgets of ±550 GtCO2. However, this is not independent of the assessed uncertainty of TCRE and is thus not fully additional

b (2) TCRE: transient climate response to cumulative emissions of CO2, assessed to probably be 1.0–2.3 °C EgC-1, with a normal distribution, from which the percentiles are taken

a maximum global temperature rise of 1.5 °C above pre-industrial levels. This requires intermediate targets to be established for 5-year intervals and regular reporting on progress.

The Alliance commissioned the Institute for Sustainable Futures (ISF) at the University of Technology Sydney (UTS) to utilize its pre-existing OneEarth Climate Model (Teske et al., 2019) to derive 1.5 °C decarbonization pathways for key highemitting sectors, on a global level, to achieve net-zero emissions by 2050, and to inform the development of sector-based targets for decarbonization. This book presents the results of that research, undertaken between late 2019 and December 2021. We hope it will clarify investor expectations for decarbonization strategies for the sectors in which they invest.

# *2.2.1 Development of GHG Emissions: A Look Back*

The global economy must decarbonize the energy system entirely within the next 30 years—in one generation. In historical terms, this means breaking the connection between population growth, steady economic development fuelled by fossil energy, and the increase in CO2 emissions of the past 120 years and reversing those trends within the next 5 years. Between 2025 and 2030, global energy-related CO2 must peak and start to decline to zero by 2050.

**Fig. 2.1** Development of global energy-related CO2 emissions by region in 1750–2020 (Ritchie & Roser, 2020)

Figure 2.1 shows the development of the annual energy-related CO2 emissions between 1750 and 2020 based on data from the *Global Carbon Project* of the *Integrated Carbon Observation System* (Global Carbon Project, 2021). Global annual CO2 emissions rose from 2 Gt in 1900 to 4 Gt 1935, to 6 Gt in 1950, and to 12 GtCO2 in 1966. By 1996, global emissions had reached 24 GtCO2—and only 10 years later, emissions increased by another 10 Gt. Since 2012, the increase in emissions has at least slowed, and in 2020, emissions were around 35 GtCO2.

A closer look into regional emissions shows that Europe was responsible for 43% of all historic CO2 emissions between 1750 and 1990, followed by the USA

**Fig. 2.2** Cumulative global energy-related CO2 emissions by region in 1750–2020 (Ritchie & Roser, 2020)

with 31%, whereas China emitted 5%, Africa 2%, and India only 1%. However, the regional distribution changed dramatically after 1990, with China's double-digit economic growth over the past decades: China and Europe both contributed 21%, followed by the USA with 19%, India 5%, and Africa 4%. Figure 2.2 shows the cumulative CO2 emissions by region between 1750 and 2020, based on data from Integrated Carbon Observation Systems (ICOS 2021). According to these data, Europe emitted 31% of all cumulative CO2, followed by the USA (25%) and China (14%). The remaining 30% was distributed across all other regions and countries outside those three main economic hubs (the USA, Europe, and China).

# *2.2.2 Global Economic Development: A Look Back*

On average, global economic development has steadily increased. Based on the World Bank data (World Bank, 2020), the global median GDP growth between 1970 and 2015 was 3.5%, although with signifcant regional differences. In 1966, the total global output of the world economy increased by over US\$20 trillion and then doubled within 20 years to \$40 trillion by 1986. Thirty years later—in 2006 this value surpassed \$80 trillion (Fig. 2.3). In 2020, the global GDP reached \$132 trillion. For this analysis, we follow the World Bank projection—which was also used for the World Energy Outlook 2017 and was the basis for the frst OECM book published in 2019 (Teske et al., 2019) (Table 2.2).

**Fig. 2.3** Global GDP development in 1700–2015 (Ritchie & Roser, 2020)

# *2.2.3 Socio-economic Assumptions for the OECM 1.5 °C Scenario*

The assumed development of regional populations is based on the projections of the United Nations Department of Economic and Social Affairs (UN DESA, 2019), whereas the regional GDP developments are based on World Bank projections. The global values for population and GDP are identical throughout the entire analysis, across all sectors. Regional values are used for the buildings and transport sectors, whereas for all other sectors, the resulting (summed) global values are used.

# *2.2.4 Outlining the Task: Trend Reversals Until 2025*

The frst step in the development of sectorial 1.5 °C pathways is to decide on the basic drivers of the future energy demand: population growth and economic development. To ensure that the OECM is transparent and comparable with other

**Fig. 2.4** Projection of global energy demand under the assumption that historic effciency trends continue until 2050

scenarios, established projections of the UN and the World Bank were used. The OECM focuses on the development of energy-relevant parameters.

Figure 2.4 shows the global development of GDP per capita since 1950 and the projections from 2020 to 2050. Economic energy intensity is the average amount of energy units required for each dollar of economic value. In 1950, the energy intensity was around 11 GJ per US\$1000 GDP, on a global average. This value includes electricity and fuel demands, e.g. for heating and transport. Energy intensity decreased over time, which indicated the successful implementation of effciency measures. Different economic sectors have very different energy intensities. A highly industrialized country with large manufacturing capacities, e.g. for steel production, has a signifcantly higher energy intensity than a service-based economy that is focused on tourism, for example. Therefore, a low energy intensity is not necessarily a sign of a very effcient economy, but could indicate an economy that is largely based on agriculture. However, on a global average, energy intensity is an important parameter refecting advances in effciency.

Between 1950 and 2020, the global energy intensity decreased by 1.2% annually, leading to an energy intensity of 4.8 GJ per US\$1000 GDP—about half the value in 1950. The projection of the energy demand shown in Fig. 2.4 was calculated under the assumption that the energy intensity will continue to decrease at 1% per year, while GDP continues to grow by 3.5%, on average, between 2020 and 2050.

The third relevant parameter is the average energy demand per capita, which is simply the overall primary energy demand divided by the population. The per capita energy demand doubled from 40.5 GJ per year in 1950 to around 80 GJ per year in

**Fig. 2.5** Global development of key parameters

2020. With the assumed increase in economic energy intensity and the overall economic development, the per capita energy use should increase again by over 50% to 125 GJ per year in 2050. Finally, the emission intensity, or the average amount of energy-related CO2 emissions per capita, results from the energy demand and the energy source selected. If coal is used to supply the entire energy demand, GHG emissions will steadily increase, whereas a supply of renewable energy will lead to a decarbonized economy.

Figure 2.5 shows the historic development of the global population, GDP, energy demand, and the resulting annual CO2 emissions between 1950 and 2020 on the left side and the projected trend development until 2050 on the right side. Based on the projected population and economic growth until 2050 and under an assumed annual decline of 1% in both energy and emission intensities, the global energy demand will double, whereas CO2 emissions will remain at around current levels.

The OECM does not question the development of the population or the global economy projected by international organizations, but focuses on technical measures to increase energy effciencies and decarbonize the energy supply by a transition to renewable energies to achieve the 1.5 °C decarbonization trajectory (marked with the red line). This will require a bottom-up assessment of the energy demand combined with an alternative energy supply concept for power, heating, and transport, which are documented in the following chapters of this book.

# **2.3 Science-Based Target Setting**

Science-based target setting has been discussed widely at the *United Nations Framework Convention on Climate Change* (UNFCCC) Climate Conferences and among stakeholders from industry, non-governmental organizations, and government departments. Although there is no offcial defnition of 'science-based target setting', it basically means that global, regional, and sectorial carbon emission targets are set to achieve the goals of the Paris Climate Agreement based on the latest available scientifc knowledge. Therefore, the overall target is to limit the global mean temperature rise to +1.5 °C with high probability.

The latest available scientifc information is IPCC's Sixth Assessment Report *Climate Change 2021: The Physical Science Basis* (Sect. 2.1). Table 2.2 shows the estimates of the remaining carbon budgets and their uncertainties published in the Technical Summary (IPCC, 2021). According to the IPCC defnition, 67% likelihood is 'good', whereas 50% likelihood is 'fair'.

The OECM aims to limit the global mean temperature rise to 1.5 °C with 'good' likelihood. Therefore, the 'science-based target' for the OECM 1.5 °C pathway in terms of the global carbon budget between 2020 and 2050 is set to 400 Gt CO2.

The development of sectorial targets for the needs of specifc countries or industries will ensure that the global sum of all energy-related CO2 emissions for all countries or all industry sectors does not exceed the global budget. Therefore, any approach undertaken in isolation, such as for only a single industry sector, will involve the risk that one industry sector will claim a higher CO2 budget and push the responsibility to reduce CO2 emissions onto other sectors.

# *2.3.1 Science-Based Targets for the Finance Industry*

Investment decisions for the decarbonization of investment portfolios are underpinned by highly complex considerations. In November 2020, the European Central Bank published a 'Guide on climate-related and environmental risks', which maps out a detailed process for 'climate stress tests' for investment portfolios. For the global fnance industry to implement the Paris Climate Agreement, decarbonization targets and benchmarks for industry sectors are required.

The estimation of carbon budgets for specifc industry sectors requires a holistic approach, and the interconnection of all sectors and regions must be considered. To estimate the carbon budget for a single industry sector in an isolated 'silo approach' based on current emission shares will inevitably lead to inaccurate results because this approach does not consider the possible technical developments in that sector or its interactions with other industry sectors. Therefore, the total of all sub-concepts for certain industries will exceed the actual CO2 emitted, and/or the responsibilities for CO2 reduction will be shifted to other areas.

# *2.3.2 Nationally Determined Contributions (NDCs)*

The Paris Climate Agreement was adopted by 196 countries and regions (e.g. the European Union) in 2015 and came into force on 4 November 2016. It is a legally binding international treaty on climate change. Each signatory country must submit *nationally determined contributions* (NDCs). An NDC is basically a plan for a country that outlines specifc measures that will be implemented to reduce GHG emissions. This usually includes an energy scenario but can also include targets for emissions related to land-use changes, such as in forestry and agriculture.

NDCs play a central role in the Paris Climate Agreement and are defned in Article 4, paragraphs 2, 3, and 4:

#### **Paris Climate Agreement, Article 4 (UNFCCC, 2015)**


Nationally determined contributions must be submitted every 5 years. The frst submission was in 2020, so the subsequent NDCs are required in 2025 and 2030. All NDCs are publicly available and collected at the 'NDC registry'. At the time of writing (December 2021), the modalities and procedures of the NDC registry were still under negotiation, and an interim NDC registry was in place.

All submitted NDCs are regularly analysed, and their targets are summarized in order to maintain an overview of the projected GHG emissions over the next 5-year period and to assess whether global emissions are on track to meet the 1.5 °C target. The estimated emissions are reported in CO2 equivalents, in order to include all GHGs, not only energy-related CO2 emissions.

There is a not-entirely-fxed template for NDCs, and each country structures its NDC differently. However, all NDCs are expected to cover the following fve sectors:


NDCs should include not only CO2 emissions but also CH4, nitrous oxide (N2O), and gases and aerosols that fall under the Montreal Protocol (see Chap. 11).

# *2.3.3 Net-Zero Pledges*

To support the NDC process, the UNFCCC started the 'Race to Zero' campaign, with the aim of obtaining 'net-zero pledges' in the run-up to and during COP26 in November 2021. The target group for these pledges were industry sectors and/or industry companies, fnance sectors, and/or fnance institutions, but also countries, which would submit pledges in addition to NDCs. The campaign received signifcant positive feedback. An analysis by the International Energy Agency (IEA) in November 2021 concluded that all the pledges announced by 3 November 2021 will—under the assumption that they will be implemented by 2050—reduce annual global CO2 emissions from around 35 Gt currently to just over 20 Gt. Although this is already a notable reduction, it will not limit the global temperature increase to 1.5 °C, but instead to around 2.0 °C (Birol, 2021).

According to the UNFCCC Race to Zero website (UNFCCC, 2021), the *net-zero pledge* consists of four steps (see Box 2.1):

#### **Box 2.1: Race to Zero Criteria**


The OneEarth Climate Model aims to support the development of NDCs and Net-Zero Pledges. The following chapters document the detailed bottom-up assessment of the energy demand, the energy supply concept, and the changes in land-use required to achieve the Paris Agreement goals.

# **References**


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

# **Part III Methodology**

# **Chapter 3 Methodology**

**Sven Teske, Jaysson Guerrero Orbe, Jihane Assaf, Souran Chatterjee, Benedek Kiss, and Diana Ürge-Vorsatz**

**Abstract** The OneEarth Climate Model (OECM), its background, and program architecture are described. How the OECM is broken down into two independent modules to calculate demand and supply is explored. The basic program logic of the MATLAB-based bottom-up demand module, with high technical resolution, is described for various sectors, including the input and output parameters. The description includes numerous fgures and tables for both demand and supply modules. The sub-sectors used for the OECM 1.5 °C pathway are listed, including outputs and the areas of use.

The second part of the chapter documents the high-effciency building (HEB) model of the Central European University, which was used for the global and regional bottom-up analyses of the building sector. Its methodology, including the programme architecture, the workfow, and the equations used, is provided.

**Keywords** Methodology · OneEarth Climate Model (OECM) · MATLAB · High-effciency building (HEB) model

The Paris Climate Agreement (UNFCCC, 2015) 'notes that … emission reduction efforts will be required … to hold the increase in the global average temperature to below 2 °C above pre-industrial levels…'*.* The Intergovernmental Panel on Climate Change (IPCC) further quantifed the carbon budget to achieve this target in its Sixth Assessment Report of the Working Group (IPCC, 2021). According to the IPCC, a global carbon budget of 400 GtCO2 is required to limit the temperature rise to 1.5 °C, with 67% likelihood, by 2050.

S. Teske (\*) · J. G. Orbe · J. Assaf

University of Technology Sydney – Institute for Sustainable Futures (UTS-ISF), Sydney, NSW, Australia e-mail: sven.teske@uts.edu.au

S. Chatterjee · B. Kiss · D. Ürge-Vorsatz Central European University, Department of Environmental Sciences and Policy, Budapest, Hungary

To implement these targets, energy and climate mitigation pathways are required. Numerous computer models for the analysis and development of energy and emission pathways have been developed over the last few decades. Many different calculation methods have been established, which mainly differ in the principal task of the model and the level of detail in the GHG emissions and/or energy systems calculated. The various methods of climate-economy modelling use different ways to describe the economy- and climate-relevant parameters as parts of a highly interconnected process (Nikas et al., 2019). In this context, the economy includes all aspects of the energy system and the policy framework, whereas the climate module refects various GHG emissions from energy-related and non-energy-related processes, such as land use.

A comprehensive review of energy models, focusing on the usability of those models for decision-making, found 'that a better understanding of user needs and closer co-operation between modellers and users is imperative to truly improve models and unlock their full potential to support the transition towards climate neutrality …' (Süsser et al., 2022).

# **3.1 The OneEarth Climate Model**

The UN-convened Net-Zero Asset Owner Alliance (NZAOA) is an international group of institutional investors committed to transitioning their investment portfolios to net-zero emissions by 2050 (NZAOA, 2021). Detailed industry sector-based energy scenarios are required to implement those net-zero commitments. On the basis of the OneEarth Climate Model (OECM; Teske et al. 2019a, b), the Institute for Sustainable Futures, University of Technology Sydney (UTS/ISF), in close cooperation with institutional investors, has developed an integrated energy assessment model for industry-specifc 1.5 °C pathways, with high technical resolution, for the fnance sector. In this article, we describe the detailed methodology and the architecture of the energy model in the 2021 edition of the advanced OneEarth Climate Model (OECM 2.0).

# *3.1.1 The OneEarth Climate Model Architecture*

The OneEarth Climate Model has been developed on the basis of established computer models. The energy system analysis tool consisted of three independent modules:


#### 3 Methodology

3. Power system analysis model [R]E 24/7, which simulates the electricity system on an hourly basis and at geographic resolution to assess the requirements for infrastructure, such as grid connections between different regions and electricity storage types, depending on the demand profle and power generation characteristics of the system (Teske, 2015)

The advanced OneEarth Climate Model, OECM 2.0, merges the energy system model (EM), the transport energy model (TRAEM), and the power system model [R]E 24/7 into one MATLAB-based energy system module. The Global Industry Classifcation Standard (GICS) was used to defne sub-areas of the economy. The global fnance industry must increasingly undertake mandatory climate change stress tests for GICS-classifed industry sectors in order to develop energy and emission benchmarks to implement the Paris climate protection agreement. This requires very high technical resolution for the calculation and projection of future energy demands and the supply of electricity, (process) heat, and fuels that are necessary for the steel and chemical industries. An energy model with high technical resolution must be able to calculate the energy demand based on either projections of the sector-specifc gross domestic product (GDP) or market forecasts of material fows, such as the demand for steel, aluminium, or cement in tonnes per year.

To decarbonize the energy supply, fossil fuels must be phased out and replaced by a renewable energy supply. However, the supply of high-temperature process heat for various production processes cannot yet be fully electrical, and a simple fuel switch from oil, gas, or coal to biomass is also impossible, given the limited availability of sustainable bioenergy (Seidenberger et al., 2010; Farjana et al., 2018). To develop a detailed sector-specifc solution, the temperature level required must be considered when developing an energy scenario. An energy model with such high technical resolution can provide detailed results for various industry sectors but requires a highly complex and data-intensive model architecture. Separate modules for the calculation of different sectors of the energy system are not practicable for such high technical resolution because high electrifcation rates lead to increased sector coupling, and the interactions between sectors cannot be captured if the energy model uses separate modules.

Furthermore, the geographic distribution of the energy demand and supply must be accommodated to calculate the import and export of energy, especially for energy-intensive industries. Finally, the simulation of 100% renewable energy systems requires high time resolution to accommodate the high proportions of variable solar and wind energy.

The MATLAB model has an object-oriented structure and two modules—to calculate demand and supply—that can be operated independently of each other. Therefore, an energy demand analysis independent of the specifc supply options or the development of a supply concept based on demand from an external source is possible.

# *3.1.2 The OECM Demand Module*

The demand module uses a bottom-up approach to calculate the energy demand for a process (e.g. steel production) or a consumer (e.g. a household) in a region (e.g. a city, island, or country) over a period of time. One of the most important elements in this approach is the strict separation of the original need (e.g. to get from home to work), how this need can be satisfed (e.g. with a tram), and the kind of energy required to provide this service (in this case, electricity). This basic logic is the foundation for the energy demand calculations across all sectors: *buildings*, *transport*, *services*, and *industry*. Furthermore, the energy services required are defned: electricity, heat (broken down into four heat levels: <100 °C, 100–500 °C, 500–1000 °C, and > 1000 °C), and fuels for processes that cannot (yet) be electrifed. Synthetic fuels, such as hydrogen, are part of both the demand module, because electricity is required to produce it, and the supply module.

The energy requirements are assigned to specifc locations. This modular structure allows regions to be defned and, if necessary, the supply from other areas to be calculated.

Demand and generation modules are independent and can be used individually or sequentially. Energy demands can be calculated either as synthetic load profles, which are then summed to annual energy demands, or as annual consumption only, without hourly resolution. Whether or not hourly resolution is selected depends to a large extent on the availability of data. Load profles, such as those for the chemical industry, are diffcult to obtain and are sometimes even confdential.

#### **3.1.2.1 Input Parameters**

As in basic energy models, the main drivers of the energy demand are the development of the population and of economic activity, measured in GDP. Figure 3.1 shows the basic methodology of the OECM demand module. Tier 1 inputs are population and GDP by region and sector. 'Population' defnes the number of individual energy services, which determines the energy required per capita, and 'economic activity' (in GDP) defnes the number of services and/or products manufactured and

**Fig. 3.1** Tier 1 and tier 2 input parameters for the assessment of energy demand

sold. Tier 1 demand parameters are determined by the effect that a specifc service requires. For population, the demand parameters are defned by the need for food, shelter (buildings), and mobility and—depending on the economic situation and/or lifestyle of the population—the demand for goods and services.

Economic activity (measured in GDP) is a secondary input and is directly and indirectly dependent upon the size of the population. However, a large population does not automatically lead to high economic activity. Both population and projected GDP are inputs from external sources, such as the United Nations or the World Bank. Tier 1 input parameters themselves are strictly non-technical. The need to produce food can be satisfed without electricity or (fossil) fuels, just as a service can be provided with physical strength.

Tier 2 demand parameters are energy-relevant factors and describe technical applications, their energy intensities, and the extent to which the application is used. For example, if lighting is required, the technical application 'light bulb' is chosen to satisfy the demand.

In this example, the energy intensity is the capacity of a light bulb, e.g. 100 W. The use of the application (e.g. for 5 h per day) defnes the daily demand (5 h × 100 W = 500 Wh per day). The quantity of consumption per year is 365 days at 500 Wh per day = 1825 Wh or 182.5 kWh per year. This very basic and simple principle is used for every application in each of the main sectors: *residential + buildings*, *industry*, and *transport*. These sectors are broken down into multiple sub-sectors, such as aviation, navigation, rail, and road for *transport*, and further into applications, such as vehicle types. The modular programming allows the addition of as many sub-sectors and applications as required.

#### **3.1.2.2 Structure of the Demand Module**

Each of the three sectors, *residential and buildings*, *industry*, and *transport*, has standardized sub-structures and applications. The residential sector *R* (frst layer) has a list of household types (second layer), and each household type has a standard set of services (third layer), such as 'lighting', 'cooling', and 'entertainment'. Finally, the applications for each of the services are defned (fourth layer), such as refrigerator or freezer for 'cooling'. The energy intensity of each application can be altered to refect the status quo in a certain region and/or to refect improvements in energy effciency. An illustrative example of the layers of the residential sector is shown in Fig. 3.2.

Figure 3.3 shows an example of the model structure of the *industry* sector. In the second layer are different industries—the OECM uses the GICS classifcation system for industry sub-sectors. The quantity of energy for each of the sub-sectors is driven by either GDP or the projected quantity of product, such as the tonnes of steel produced per year. The market shares of specifc manufacturing processes are defned, and each process has a specifc energy intensity for electricity, (process) heat, and/or fuels.

**Fig. 3.2** Residential sector sub-structures

**Fig. 3.3** Calculation of the *industry* energy demand

**Fig. 3.4** Calculation of *transport* energy demand

Figure 3.4 shows the structure for the t*ransport* sector. Again, the demand is driven by 'non-energy' factors, such as passenger-kilometres and freight-kilometres, and energy-related factors, such as the transport mode and the energy intensity of the different vehicle options.

#### **3.1.2.3 Demand Module Architecture in MATLAB**

The demand module is implemented in MATLAB, a widely used programming language for mathematics and science computing. MATLAB allows the integration of a range of tools and databases and has the fexibility to add and develop new functions. Specifcally, the model has been developed using an object-oriented programming approach, allowing extensibility and modularity.

Figure 3.5 shows the demand module developed in MATLAB. The demand module encompasses eight classes: (1) demand class, (2) household class, (3) household application class, (4) sub-sector, (5) industry class, (6) industry application class, (7) transport modes, and (8) vehicles class (Fig. 3.6).

**Fig. 3.5** A unifed modelling language (UML) diagram of the demand module in MATLAB, showing its classes, attributes, methods, and associations

**Fig. 3.6** An example of a *household* type object, showing the assigned attributes



**Table 3.1** Methods within the demand class


**Fig. 3.7** An example of an *industry* application object, showing the assigned attributes

sub-sectors (e.g. steel, cement, etc.) and incorporate various types of applications under each sub-sector.

• *Transport modes and vehicle classes*: These classes are used to defne the *transport* sector. The vehicle objects are embedded within the transport mode objects. Therefore, multiple types of transport modes can be defned, such as aviation and navigation, as well as various types of vehicles, such as planes and cruise ships.

Figures 3.6 and 3.7 show the high-level class defnitions for *residential* and *industry* sub-sector objects, respectively. The blue-marked text indicates the defned value for each attribute. For example, one household object with fve residents is defned by the name 'Rural–Phase 1' and has a list of 35 appliance objects, defned with a string array. It is assigned a share factor for 2020 of 0.2, which means that 20% of the households in that specifc region and year are defned by this type of household and its attributes. Furthermore, 24 h load profles are defned for each application for every day, with numerical arrays. For example, weekend load profles have a size of 35 rows and 48 columns, representing 35 applications and 24 time slots for each weekend day.

The object-oriented architecture allows all these input attributes to be updated or modifed easily. These attributes can also be read from a predefned Excel spreadsheet. This facilitates a data input process that follows the array structure, such as the load profle.

Figure 3.7 shows an example of an industrial application object that belongs to the sub-sector *iron and steel*. In this case, the energy consumption form is defned as electricity and heat, which means that it considers the electrical and heat demand. The 'share factors' represent the portions of the demand assigned to electricity and heat. The energy-level array also allows the predefned network to which the application is connected to be defned, as well as the temperature levels. In this particular case, the demand is defned based on the total annual primary energy intensity and the material-based production, which are 1184 GJ/tonnes and 1178 Mt, respectively, for the specifed region and year. The input and output units must be predefned when the MATLAB modules are initialized. Other attributes that can be assigned are conversion factors, such as from primary energy to the fnal energy via an effciency factor.

Additional attributes and methods can be defned for each class if required and the data are available. Therefore, the demand module class can be extended by defning new classes, attributes, and methods.

# *3.1.3 The OECM Supply Module*

The supply module consists of three main elements: supply technologies, storage technologies, and the infrastructure for the power supply (capacities of power lines). For the generation of electricity and heat, the programme considers all the technologies of the energy market, from both renewable and non-renewable sources. In addition to the generation of pure electricity and heat, the entire range of combined heat and power systems is included.

Storage technologies include batteries and the use of hydrogen from electrolysers. The calculation of heat storage is possible, but has not yet been used in the OECM scenarios.

A dispatch strategy is defned for electricity and heat generation that refects market and policy factors. Whether electricity from photovoltaics and onshore and offshore wind turbines have priority dispatch ahead of fossil-fuel power plants and how storage systems are used can be determined. Each technology has a specifc conversion effciency.

Heat generation technologies are also defned by the temperature levels they can provide. For example, residential solar collectors can only supply low-temperature heat and will therefore not be considered for high-temperature process heat (Table 3.2).

The regional energy demand—as defned in the previous section—can be met by neighbouring regions, with importation from or, in the case of oversupply, exportation to them. The extent to which electricity can be imported or exported from one region to another is defned by the capacity of regional interconnections, which represent the available power line capacities.

#### **3.1.3.1 The OECM Dispatch Module**

The methodology of the dispatch module of the MATLAB-based OECM is based on the previous version of the model (Teske et al. 2019a). The key inputs are related to the supply technologies, storage types, dispatch strategy, and the


**Table 3.2** Example of generation and storage technologies

**Table 3.3** Input parameters for the dispatch model


interconnections among regions for possible power exchange (Table 3.3). Different supply technologies can be selected, each with its technical characteristics, including its effciency, available installed capacity, fuel type, and regional meteorological data (solar radiation or wind speed). Meteorological data defne the capacity factors of solar and wind energy generators as their levels of availability at 1-h resolution for an entire year (Table 3.4).

The supply technologies can be either dispatchable (e.g. gas power plants) or non-dispatchable (e.g. solar photovoltaic without storage). The model allows the


**Table 3.4** Output parameters for the dispatch model

**Table 3.5** Technology groups for the selection of dispatch order


**Table 3.6** Technology options—variable renewable energy


order in which the supply technologies and storage functions are utilized to be adjusted to satisfy the demand. However, storage and interconnections cannot be selected as the frst elements of supply (Table 3.5).

Tables 3.6, 3.7, and 3.8 provide an overview of the possible supply technologies and examples of different dispatch scenarios. Although concentrated solar power (CSP) plants with storage are dispatchable to some extent—depending on the storage size and the available solar radiation—they are part of the renewable variable group in the MATLAB model. Although the model allows the dispatch order to be changed, the 100% renewable energy analysis always follows the same dispatch logic. The model identifes excess renewable production, which is defned as any potential wind and solar photovoltaic generation greater than the actual hourly demand in MW during a specifc hour. To avoid curtailment, the surplus renewable electricity must be


**Table 3.7** Technology options—dispatch generation

**Table 3.8** Technology options—storage technologies


stored with some form of electric storage technology or exported to a different cluster or region. Within the model, the excess renewable production accumulates through the dispatch order. If storage is present, it will charge the storage within the limits of the input capacity. If no storage is present, this potential excess renewable production is reported as 'potential curtailment' (pre-storage) (Table 3.9).

*Limitations*: It is important to note that calculating the possible interconnection capacities for transmission grids between subregions does not replace technical grid simulations. Grid services, such as the inductive power supply, frequency control, and stability, should be analysed, although this is beyond the scope of the OECM analysis. The results of [R]E 24/7 provide a frst rough estimate of whether increased use of storage or increased interconnection capacities or a mix of both will reduce systems costs.

### **3.1.3.2 Regional Interconnections**

Interconnection capacities are set as a function of the total generation capacity within a cluster. Interconnections between defned regions are the only ones considered, and all intra-regional interconnections or line constraints are excluded. Therefore, a region is considered a 'copper plate'—and a transmission system


**Table 3.9** Dispatch module—inputs, intermediate outputs, and outputs

where electricity can fow unconstrained from any generation site to any demand site is found in most energy modelling tools (Avrin, 2016). This simplifcation is required to achieve a short calculation time while maintaining high technical and time resolution. The algorithm devised for the function of the interconnectors is based on the following information for each region:


The excess generation capacity and unmet load are calculated by running the model without the interconnections to determine the excess or shortfall in generation when the load within the region is met. These excesses and shortfalls are calculated at the point in the dispatch cascade at which the interconnectors provide or consume power, for example, after the variable renewables and dispatchable generators and before the storage technologies.

The interconnection capacity between regions is defned based on a percentage of the maximum regional load. The capacity is defned in a matrix, both to and from each region to every other region. A priority order for each region to every other region is given based on proximity, so that if a region has an unmet load, it will be served sequentially with the excess generation of loads in other regions in their defned order of proximity.

For every hour and every region in each cluster (a cluster is a group of regions), the possible interconnections required for the importation or exportation of energy to balance the load are calculated. Each region is considered in turn, and the algorithm attempts to meet the unmet load with excess generation by other regions, keeping track of the residual excess loads and the interconnector capacities. Each region's internal load is met frst, before its generation resources are considered for other interconnected regions.

For regions sending generation capacity to other regions, the interconnector element behaves as an increase in load, whereas for regions accepting power from neighbouring regions, the interconnector element behaves as an additional generator, from the model's perspective.

Once the total infow and outfow of the interconnectors are calculated, the hourly values for the total supply in each region are updated, together with any residual defcit in supply or any curtailed (= forced to shut down) electricity generator that does not have priority dispatch.

Similar to the supply technologies, different storage technologies (electrical, thermal, or hydrogen) can be defned and selected, together with their technical characteristics, such as their round-trip effciency, new or installed capacities in each year of the modelled period, lifetime, maximum depth of discharge, maximum energy out in a time step, and costs. When the total energy delivered by the supply technologies in a region does not meet the demand, energy is discharged from storage (if the storage technology has energy available), following the constraints of the storage operation (maximum energy out per time step, maximum depth of discharge, maximum depth of charge, state of charge) and the order of operation for the defned storage technologies. In the case of a demand defcit after storage, electricity from other regions will be imported. When there is surplus energy generation, the surplus will charge any storage appliances (if available), also according to the same constraints of energy storage operation and sequential order.

#### **3.1.3.3 Supply Module Architecture in MATLAB**

Analogous to the demand module, inputs can be made directly into the supply module via MATLAB or a standardized Excel sheet. The supply module in MATLAB is also based on an object-oriented structure, in which classes and the objects belonging to those classes are built based on attributes and methods.

Figure 3.8 shows the UML class diagram for the supply module developed in MATLAB. Specifcally, the supply module has three main classes:


**Fig. 3.8** A UML diagram of the supply module in MATLAB, showing its classes, attributes, methods, and associations

the addition of new attributes if required. This class has methods that are used by the main supply class to calculate the primary fuel, emissions, or installed capacity of a specifc technology.

3. *Storage technology class*: This class is used to defne storage technologies. The attributes include name, type, effciency, year, region, and energy storage form and are defned as text inputs. Other numerical attributes include charging and discharging rates, capacity, cost factors, and state of charge.

Figures 3.9 and 3.10 show the high-level class defnitions for supply technologies and storage objects, respectively. The text in blue indicates the defned value for each attribute. For example, the supply technology object in Fig. 3.9 has the name 'coal power plant', its input energy is defned as hard coal, and the object is associated with the electricity energy form. The attributes in Fig. 3.9 consider the year 2020 and a global scenario. For example, the existing capacity is defned as 989.5 GW and the decommissioned capacity is 23 GW. The lifetime of this object is 35 years.

An example of a storage object is shown in Fig. 3.10. The attributes of this object include text inputs, such as its name 'battery lithium' and its type 'electrical'. This object has numerical attributes such as the effciency (equal to 0.95 for this object) and the charging and discharging rates (fxed at 5 kW). Note that the units for each attribute are defned when the module is initialized in MATLAB.


**Table 3.10** Methods within the supply class


**Fig. 3.9** An example of a supply technology object, showing the assigned attributes


**Fig. 3.10** An example of a storage technology object, showing the assigned attributes

The supply module architecture developed is fexible to accommodate different types of supply and storage technologies. Additional attributes or methods can be easily added to the model.

# *3.1.4 Databases and Model Calibration*

The OECM model uses several databases for energy statistics, energy intensities, technology market shares, and other market or socio-economic parameters. The calculation of the energy balance for the base year is based on the International Energy Agency (IEA) Advanced World Energy Balances (IEA, 2020, 2021).

The energy statistics for a calculated country and/or region are uploaded via an interface module. The data for each year from 2005 onwards until the last year for which data are available are used to calibrate the model. This process is based on the energy system model (EM), developed by the German Aerospace Center DLR, and is implemented in the energy simulation platform Mesap/PlaNet (Schlenzig, 1999; Seven2one, 2012). The market shares are calculated based on the IEA statistics and a technical database for energy intensities for various appliances and applications across all sectors. These data are input and the calibration processes performed with a standardized Excel tool. The calibration method is briefy outlined below using the *transport* sector.

To calibrate the model, the transport demand of the past decade is recalculated on the basis of the available energy statistics. The IEA's Advanced World Energy Balances provides the total fnal energy demand by transport mode—aviation, navigation, rail, or road—by country, by region, or globally. However, it provides no further specifcation of the energy use within each of the transport modes. Therefore, further division into passenger and freight transport is calculated using percentage shares. These proportions are determined with a literature search, together with the average energy intensity for each of the transport modes for passenger and freight vehicles.

The annual transport demand in passenger-kilometres per year (pkm/year) or tonne-kilometres per year (tkm/year) is calculated as the annual energy demand

#### 3 Methodology


**Table 3.11** Calibration for calculating the transport demand

divided by the average energy intensity by mode. These results are then compared with the Organisation for Economic Co-operation and Development (OECD) transport statistics, which provide both parameters——pkm/year and tkm/year. Calibrating the model on the basis of historical data ensures that the basis of the scenario projections for the coming years and decades is correctly mapped and ensures that the changes are calculated most realistically (Tables 3.11 and 3.12).


**Table 3.12** Projection of transport demand based on the changing demand in kilometres

For the forward projection of the transport demand, the calculation method is reversed: the transport demand for each transport mode is calculated on the basis of the annual change, as a percentage. The calculated total annual pkm and tkm are the inputs for the energy demand calculation.

This methodology for calibration and projection is used across all sectors.

The developed MATLAB tool can access online data and databases through available *application programming interfaces* (APIs). For example, the API for the World Bank Indicators provides access to nearly 16,000 time series indicators, including population estimates and projections (World Bank, 2021). Likewise, the OECD provides access to datasets through an API. This allows a developer to easily call the API and access data using the code lines in MATLAB.

# *3.1.5 Sectors and Sub-sectors*

The OECM was designed to calculate energy pathways for geographic regions, as documented in Chap. 2. The OECM was developed further to meet the requirements of the fnancial industry and to design energy and emission pathways for clearly defned industry sectors (sectorial pathways). The fnance industry uses different classifcation systems to describe sub-areas of certain branches of industry. The most important system is the Global Industry Classifcation Standard (GICS; MSCI, 2020). However, the GICS sub-sectors do not match the IEA statistical breakdown of the energy demands of certain industries. Table 3.13 shows examples of the fnance sector calculated with the OECM model, the GICS codes, and the statistical

**Table 3.13** Examples of industry sub-sectors based on the Global Industry Classifcation Standard (GICS)


information used. Although the OECM model allows all the GICS code sub-sectors to be calculated, the availability of statistics is the factor limiting the resolution of the sectorial pathways. For example, the statistical data for the textile and leather industry are stored in the IEA database, but the database does not separate the two industries further.

# *3.1.6 Cost Calculation*

The costs linked to the energy supply in each year of the modelled period include the investment costs related to 'new capacities' for technologies and storage (including replacement or decommissioning, based on the assumed technical lifetime = vintaging), operation and maintenance (O&M) costs as a percentage of the total installed capacities, and fuel costs. Other inputs for each technology and storage type include the capital cost per unit (\$/kW), O&M costs as a percentage of the capital cost, and unit fuel cost (\$/GJ).

Therefore, for each technology or storage type:


The total specifc costs (\$/kWh) of a scenario, as practically distributed over the interpolated years, allow the incurred costs for a scenario to be determined. *Limitations*: The economic model does not consider the change in the value of money over time. Each year of the modelled period is regarded as if it were the present year, with the multiple costs incurred. Future additions to the model could include the net present costs and the contemporary value of money.

# *3.1.7 OECM 2.0 Output and Area of Use*

The added value of OECM 2.0 is its high resolution of the sector-specifc parameters for both demand and supply, which are required as key performance indicators (KPIs) by the fnance industry. Table 3.14 provides an overview of the main parameters and the areas of their use, with a focus on the needs of institutional investors.

Commodities and GDP are the main drivers of the energy demand for industries. The projection of, for example, the global steel demand in tonnes per year over the next decades is discussed with the industry and/or client. The OECM 2.0 can calculate either a single specifc sector only or a whole set of sectors. For the development of global scenarios, various industry projections are combined to estimate both the total energy supply required and the potential energy-related emissions. Therefore, a global carbon budget can be broken down into carbon budgets of specifc industries.

Energy intensities are both input data for the base year and a KPI for future projections. The effect of a targeted reduction in the energy intensity in a given year and the resulting energy demand and carbon emissions can be calculated, for example, for the steel industry.

All sector demands are supplied by the same energy supply structure in terms of electricity, process heat (for each level), and total fnal energy. Finally, specifc emissions, such as CO2 per tonne of steel or per cubic metre of wastewater treatment, are calculated and can be used to set industry targets.

All input and output OECM data are available as MATLAB-based tables or graphs or as standard Excel-based reports.

# *3.1.8 Further Research Demand*

Industry-specifc energy intensities and energy demands are not available for a variety of industries. In particular, the energy intensities for sub-sectors of the chemical industry are either totally unavailable or confdential. A database of energy intensities is required to develop more detailed scenarios. Although energy intensities can be estimated based on the available data, the input parameters are usually derived from various sources, which may not follow the same methodology. Energy intensities based on GDP, for example, are calculated with either nominal GDP, real GDP, or purchasing power parity GDP. Furthermore, energy intensities can be provided as fnal energy or primary energy. In some cases, this information is not available at all. A database of industry-specifc energy demands and energy intensities, with a consistent methodology, is required to improve the accuracy of calculations in future research.

To capture the complexity of regional and global building demand projection, both in terms of data availability and high technical resolution, the high-effciency building (HEB) model was used to develop four bottom-up demand scenarios. The HEB was developed by the Central European University (CEU) of Budapest under


**Table 3.14** Energy-related key performance indicators (KPIs) for net-zero target setting, calculated with OECM 2.0

(continued)


**Table 3.14** (continued)

the scientifc leadership of Prof. Dr. Diana Uerge-Vorsatz. The following section documents the methodology of the HEB based on the paper by Chatterjee, S.; Kiss, B.; and Ürge-Vorsatz, D. (2021). The results are documented in Sects. 7.1 and 7.2.

# **3.2 The High-Effciency Building (HEB) Model**

Modelling the energy demand for buildings is a complex task because the building sector-related energy demand depends on several factors, such as spatial resolution, temporal resolution, building physics, and the different technologies of building construction (Prieto et al., 2019; Chatterjee & Ürge-Vorstaz, 2020). The majority of demand models do not incorporate these factors and therefore provide insights into the future energy demand scenarios of the building sector that can be far from realistic (Prieto et al., 2019; Chatterjee & Ürge-Vorstaz, 2020). Therefore, in this study, we use the HEB model to understand the future energy demand potentials for building in key regions across the globe.

The HEB model was originally developed in 2012 to calculate the energy demand and CO2 emissions of the residential and tertiary building sectors until 2050 under three different scenarios (Ürge-Vorsatz & Tirado Herrero, 2012). Since then, the model has been developed and updated several times. With the latest update, the model calculates the energy demand under four scenarios until 2060 based on the most recent data for macroeconomic indicators and technological development. This model is novel in its methodology compared with earlier global energy analyses and refects an emerging paradigm—the performance-oriented approach to the energy analysis of buildings. Unlike component-oriented methods, a systemic perspective is taken: the performance of whole systems (e.g. whole buildings) is studied, and these performance values are used as the input in the scenarios. The model calculates the overall energy performance levels of buildings, regardless of the measures applied to achieve them. It also captures the diversity of solutions required in each region by including region-specifc assumptions about advanced and suboptimal technology mixes. The elaborated model uses a bottom-up approach, because it includes rather detailed technological information for one sector of the economy. However, it also exploits certain macroeconomic (GDP) and socio-demographic data (population, urbanization rate, foor area per capita, etc.). The key output of the HEB model is foor area projections for different types of residential and tertiary buildings in different regions and their member states, the total energy consumption of residential and tertiary buildings, the energy consumption for heating and cooling, the energy consumption for hot water energy, the total CO2 emissions, the CO2 emissions for heating and cooling, and the CO2 emissions for hot water energy.

# *3.2.1 The High-Effciency Building Model Methodology*

The HEB model conducts a scenario analysis for the entire building sector, in which the building sector is distinguished by location (rural, urban, and slum), building type (single-family, multifamily, commercial, and public buildings, with subcategories), and building vintage (existing, new, advanced new, retroftted, and advanced retroftted). This detailed classifcation of buildings is undertaken for 11 regions (Ürge-Vorsatz & Tirado Herrero, 2012), extended with country-specifc results for the EU-27 countries, China, India, and the USA. Furthermore, within each region, different climate zones are considered to capture the differences in building energy uses and renewable energy generation caused by variations in climate. The climate zones are calculated based on four key climatic factors—heating degree days (HDD), cooling degree days (CDD), relative humidity (RH) in the warmest month, and average temperature in the warmest month (T). These parameters are processed using the GIS5 tool—spatial analysis—and performed with the ArcGIS software. The detailed classifcation categories are summarized in Table 3.2.

The purpose of the detailed classifcation of building categories and scenario assessments is to explore the consequences of certain policy directions or decisions that inform policy-making (Table 3.15).

The key input data used in the HEB are region-specifc forecasts of GDP, population, rate of urbanization, and the proportion of the population living in urban slums. The time resolution of the model is yearly, so that socio-economic input data can be easily obtained from various credible sources, such as the databases of the World Bank, United Nations Development Programme (UNDP), EUROSTAT, and the OECD. Besides these socio-economic parameters, many others are included, and in the case of data absence, assumptions are made in the HEB model to calculate the fnal energy demand. Figure 3.11 shows the main workfow of the HEB model.

The HEB model includes several calculation steps, from considering the input data to obtaining the fnal output. Each of these calculation steps is discussed in the sections below.


**Table 3.15** Building classifcation scheme of HEB

**Fig. 3.11** The main workfow of the HEB. Input data and parameters can be modifed by the user (green). Main outputs are the foor areas of different building vintage types and the energy consumption and CO2 emissions of the stock (blue)

# *3.2.2 Disaggregation*

In the frst step of the calculation, after all the socio-economic input data are obtained, the input is disaggregated into the detailed building classifcation scheme, and the total foor area required to satisfy the year-specifc population and GDP needs (the year is denoted with *Y* in subscript) is determined. The core concept for calculating the foor area differs for residential and commercial buildings:


The region-specifc population data—as the input for the calculation—is further disaggregated into urban and rural populations based on the urbanization rate and into the different climate zones based on GIS data:

$$P\_{\mathbf{r},\mathbf{c},\mathbf{u},\mathbf{Y}} = P\_{\mathbf{r},\mathbf{Y}} \times U\_{\mathbf{r},\mathbf{Y}} \times \mathbf{S} \mathbf{c}\_{\mathbf{r},\mathbf{c}} \quad \text{if } \mathbf{u} = \mathbf{u} \mathbf{r} \mathbf{b} \mathbf{an} \tag{3.1}$$

and

$$P\_{\mathbf{r},\mathbf{c},\mathbf{u},\mathbf{Y}} = P\_{\mathbf{r},\mathbf{Y}} \times \left(1 - U\_{\mathbf{r},\mathbf{Y}}\right) \times \mathbf{S} \mathbf{c}\_{\mathbf{r},\mathbf{c}} \quad \text{if } \mathbf{r} = \mathbf{r} \text{ural} \tag{3.2}$$

where

*P*r, c, u, Y [capita] is the total urban/rural population of region r and climate zone c in year Y

*P*r, Y [capita] is the total population of region r in year Y *U*r, Y [−] is the urbanization rate of region r in year Y Scr, c [%] is the share of the population within region r living in climate zone c

The urban population is then further disaggregated into the population living in slums (in regions where a signifcant number of people do not have access to standard living conditions) and the population living in conventional residential buildings. The latter group is split into the populations living in single-family and multifamily houses based on region-specifc fxed values:

$$P\_{\mathbf{r},\mathbf{c},\mathbf{u},\mathbf{b},\mathbf{Y}} = P\_{\mathbf{r},\mathbf{c},\mathbf{u},\mathbf{Y}} \times \mathbf{S} \mathbf{s}\_{\mathbf{r},\mathbf{Y}} \quad \text{where } \mathbf{u} = \mathbf{u} \mathbf{r} \mathbf{b} \mathbf{a} \text{ and } \mathbf{b} = \text{slum} \tag{3.3}$$

and

$$P\_{\mathbf{r},\mathbf{c},\mathbf{u},\mathbf{b},\mathbf{Y}} = P\_{\mathbf{r},\mathbf{c},\mathbf{u},\mathbf{Y}} \times \left(1 - \mathbf{S}\mathbf{s}\_{\mathbf{r},\mathbf{Y}}\right) \quad \text{where } \mathbf{u} = \mathbf{u}\mathbf{r}\mathbf{b}\mathbf{a} \text{ and } \mathbf{b} = \text{residual} \qquad (3.4)$$

then

$$P\_{\mathbf{r},\mathbf{c},\mathbf{u},\mathbf{b},\mathbf{t},\mathbf{Y}} = P\_{\mathbf{r},\mathbf{c},\mathbf{u},\mathbf{b},\mathbf{Y}} \times \mathbf{S} \mathbf{s} \mathbf{f}\_{\mathbf{r}} \quad \text{where } \mathbf{u} = \text{urban}, \mathbf{b} = \text{residual and } \mathbf{t} = \mathbf{S} \mathbf{F} \quad (3.5)$$

and

$$P\_{\mathbf{r},\mathbf{c},\mathbf{u},\mathbf{b},\mathbf{t},\mathbf{Y}} = P\_{\mathbf{r},\mathbf{c},\mathbf{u},\mathbf{Y}} \times \left(1 - \mathbf{S} \mathbf{s} \mathbf{f}\_{\mathbf{r}}\right) \quad \text{where } \mathbf{u} = \mathbf{u} \mathbf{r} \mathbf{b} \mathbf{an}, \mathbf{b} = \text{residual and } \mathbf{t} = \mathbf{M} \mathbf{F} \text{ (3.6)}$$

#### 3 Methodology

where


The population living in rural areas is assumed to live in single-family houses.

The disaggregation of GDP follows the same pattern, except that the share of GDP that can be associated with rural commercial or public buildings is fxed within the modelling period:

$$\mathbf{GDP}\_{\mathbf{r},\mathbf{c},\mathbf{u},\mathbf{Y}} = \mathbf{GDP}\_{\mathbf{r},\mathbf{Y}} \times \left(1 - U\_{\mathbf{r},\mathbf{Y}}\right) \times \mathbf{Sc}\_{\mathbf{r},\mathbf{c}} \quad \text{if } \mathbf{u} = \mathbf{u} \mathbf{r} \mathbf{b} \mathbf{an} \tag{3.7}$$

and

$$\mathbf{GDP}\_{\mathbf{r},\mathbf{c},\mathbf{u},\mathbf{Y}} = \mathbf{GDP}\_{\mathbf{r},\mathbf{Y}} \times U\_{\mathbf{r},\mathbf{Y}} \times \mathbf{Sc}\_{\mathbf{r},\mathbf{c}} \quad \text{if } \mathbf{u} = \mathbf{r} \text{ural} \tag{3.8}$$

where

GDPr, c, u, Y [USD] is the total GDP that can be associated with urban/rural commercial or public buildings in region r and climate zone c in year Y.

GDPr, Y [USD] is the total GDP of region r in year Y. *U*r, Y [−] is the urbanization rate of region r in year Y. Scr, c [%] is the share of climate zone c within region r.

The share of different commercial building types is also determined with fxed ratios based on data from the literature:

$$\text{GDP}\_{\text{r,c,u,t,Y}} = \text{GDP}\_{\text{r,c,u,Y}} \times \text{Scp}\_t \tag{3.9}$$

where

GDPr, c, u, t, Y [USD] is the total GDP that can be associated with urban/rural commercial or public buildings of type t in region r and climate zone c in year Y.

GDPr, c, u, Y [USD] is the total GDP that can be associated with urban/rural commercial or public buildings in region r and climate zone c in year Y.

Scpt [%] is the share of commercial and public buildings of type t in the commercial and public building stock.

# *3.2.3 Determining the Total Floor Area*

Different equations are used for the calculation of the foor area of residential buildings and non-residential buildings. The foor area of residential buildings can be calculated with the following equation, using specifc foor area values (the foor area that is occupied by one person):

$$\text{TFA}\_{\text{r,c,u,b,t,Y}} = \text{P}\_{\text{r,c,u,b,t,Y}} \times \text{SFAc}\_{\text{r,u,b,t,Y}} \quad \text{where } \mathbf{b} = \text{residual/slum} \quad (3.10)$$

where

TFAr, c, u, b, t, Y [m2 ] is the total urban/rural foor area of building category b and building type t in region r and climate zone c in year Y.

Pr, c, u, b, t, Y [capita] is the total urban/rural population of region r, climate zone c, building category b, and building type t in year Y.

SFAcr, u, b, t, Y [m2 /capita] is the specifc foor area of building category b and building type t in region r in year Y.

Similarly, the foor area of commercial and public buildings is calculated using specifc foor area values (the foor area that is required to produce one unit of GDP):

$$\text{TFA}\_{\text{r.c.u.b.t.Y}} = \text{GDP}\_{\text{r.c.u.t.Y}} \times \text{SFAg}\_{\text{r.b.Y}} \quad \text{if } \mathbf{b} = \mathbf{C} \,\&\,\mathbf{P} \tag{3.11}$$

where

TFAr, c, u, b, t, Y [m2 ] is the total urban/rural foor area of commercial or public buildings of building type t in region r and climate zone c in year Y.

GDPr, c, u, t, Y [USD] is the total GDP that can be associated with urban/rural commercial or public buildings of type t in region r and climate zone c in year Y.

SFAgr, b, Y [m2 /USD] is the specifc foor area of commercial or public buildings in region r in year Y.

Specifc foor area values are determined from statistical data for each region. To take socio-economic development into account, the foor area per capita and the foor area per GDP are modelled as values that change yearly, reaching the average for OECD countries by the end of the modelling period in developing regions.

# *3.2.4 Yearly Dynamics of Floor Area Changes*

The yearly dynamics of this foor area model transition the existing building stock into the future state determined by the scenarios. This includes the retroftting or demolition of existing buildings, as well as the introduction of new buildings to the stock. In some cases, the foor area is left abandoned, which might result from a reduction in the population (e.g. in developed regions) or an increased rate of urbanization due to which buildings located in rural areas are abandoned after a certain time. It is important to capture this phenomenon, because abandoned buildings do not contribute to energy consumption or the emissions of the building stock. This yearly dynamic of the vintage types of buildings is presented in Fig. 3.12.

The demolished foor area is calculated with region-specifc demolition rates. After the demolished foor area is subtracted from the existing total, the remaining existing foor area is classifed into different building vintages. Similarly, the retroftted foor area is calculated by applying the yearly changing region-specifc retroftting rate to the total existing building stock. The retroftted foor area is further classifed into two types: advanced retroftted foor area and normal retroftted foor area. For each of the regions, the shares of retroftted and advanced retroftted foor area differ, and the shares of advance retroftted, advance new, and retroftted foor areas also vary under different scenarios. The foor area from new constructions is classifed into two building vintages: new and advanced new. Like the retroftted foor area, the share of advanced new foor area also varies under different scenarios.

**Fig. 3.12** Yearly foor area dynamics in the HEB model

# *3.2.5 Calculating the Energy Consumption of Buildings*

The energy consumption for heating and cooling depends on the foor area. Therefore, in the HEB model, energy consumption is calculated after the yearspecifc foor area is calculated. The key input required to calculate the energy consumption for heating and cooling is the average consumption data for heating and cooling, which are usually obtained from data reported in the literature, for each of the regions, climate zones, and building types, because different building vintages have different consumption requirements. Therefore, different vintage types are modelled by assuming different energy intensities (denoted with subscript v). The values also depend on the scenario (denoted with subscript s). Energy intensity is multiplied by the corresponding foor area to determine the energy consumption for heating and cooling the stock:

$$\text{HCE}\_{\text{r.c.u.b.t.Y.v.s}} = \text{TFA}\_{\text{r.c.u.b.t.Y}} \cdot \text{EUHc}\_{\text{r.c.u.b.t.v.s}} \tag{3.12}$$

where


After the detailed energy consumption is calculated, the data can be summed to arrive at the region-specifc, yearly aggregated results for a given scenario:

$$\text{Total Energy}\_{\text{r,Y,s}} = \sum\_{\text{c}} \sum\_{\text{u}} \sum\_{\text{b}} \sum\_{\text{t}} \text{Total Energy}\_{\text{r,c,u,b,t,Y,v,s}} \tag{3.13}$$

# *3.2.6 Implementation*

The most recent version of the HEB model was developed in the Python programming language, using the PyData ecosystem to handle large datasets. This ecosystem ensures quite large fexibility among the modelling parameters, and the diversity of input data and its granularity can be properly handled. This model is not an openaccess model, but the Central European University has received funding from the European Union's Horizon 2020 research and innovation programme (under grant agreement no. 837089) in the Sentinel1 project, to develop HEB further. In this project, the HEB model will be made an open-source model that users can use without cost.

# **References**


<sup>57</sup>

<sup>1</sup> https://sentinel.energy/


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

# **Chapter 4 Classifcation Systems for Setting Net-Zero Targets for Industries**

**Sven Teske, Kriti Nagrath, and Sarah Niklas**

**Abstract** The structure of the Global Industry Classifcation Standard (GICS) system and how it is used in the OECM are described, as well as how the statistical data of the International Energy Agency (IEA) were merged with the GICS structure. The development of the pathways for the industry and service sectors, based on the GICS and IEA data, is explained, together with the parameters that are important for the fnancial industry. In this context, the defnitions of *Scope 1*, *2*, and *3* emissions newly developed for the OECM are explained, as well as how the systemic error of double counting in the original procedure can now be avoided.

**Keywords** Global Industry Classifcation Standard (GICS) · IEA Statistic · *Scope 1* · *2* · and *3* emissions

Investment decisions, such as the decarbonization targets for the fnance industry (see also Chap. 2), are highly complex processes. In November 2020, the European Central Bank published a *guide on climate-related and environmental risks*, which maps a detailed process for undertaking 'climate stress tests' for investment portfolios. To achieve the Paris Climate Agreement goals in the global fnance industry, decarbonization targets and benchmarks for individual industry sectors are required. This opens up a whole new research area for energy modelling because although decarbonization pathways have been developed for countries, regions, and communities, few have been developed for industry sectors. The OneEarth Climate Model (OECM) is an integrated assessment model for climate and energy pathways that focuses on 1.5 °C scenarios (Teske et al., 2019) and has been further improved to meet this need. To

S. Teske (\*) · K. Nagrath · S. Niklas

University of Technology Sydney – Institute for Sustainable Futures (UTS-ISF), Sydney, NSW, Australia e-mail: sven.teske@uts.edu.au

develop energy scenarios for industry sectors classifed under the Global Industry Classifcation Standard (GICS), the technological resolution of the OECM required signifcant improvement. Furthermore, all demand and supply calculations had to be broken down into industry sectors before the individual pathways could be developed.

# **4.1 Role of the Global Industry Classifcation Standard (GICS) in Achieving Net-Zero Targets**

The GICS was developed by the American investment research frm Morgan Stanley Capital International (MSCI) and Standard & Poor's (S & P), a fnance data and credit rating company, in 1999. According to MSCI, the GICS was designed to defne specifc industry classifcations for reporting, comparison, and investment transaction processes (MSCI, 2020). The GICS has 4 classifcation levels and includes 11 sectors, 24 industry groups, 69 industries, and 158 sub-industries. The 11 GICS sectors are energy, materials, industrials, consumer discretionary, consumer staples, health care, fnancials, information technology, communication services, utilities, and real estate (Table 4.1).

This section provides an overview of the 1.5 °C sectorial pathways and the associated GICS sectors. The individual end-use sectors are subdivided into four major sections:


The focus of each of these sections is documented in dedicated chapters (see above) that focus exclusively on current and future market developments and their energyrelated aspects. The non-energy-related greenhouse gas (GHG) emissions are described in a separate section (Chaps. 11 and 14).


**Table 4.1** GICS: 11 main industries

The primary energy sector, fossil-fuel-producing companies, and the secondary energy industries, energy-distributing utilities, make up their own two GICS groups.

# *4.1.1 OECM 1.5 °C Industry Pathways and the Associated GICS Sectors*

Table 4.2 provides an overview of the OECM 1.5 °C industry pathways. The majority are in the *materials* sector (1510) and the related sub-sectors *chemicals* (1510 10), *construction materials/cement* (1510 20), *aluminium* (15104010), and *steel* (15104050). Textiles and leather, which are classifed as *consumer durables and apparel* (2520) in the subgroup *textiles* (2520 3030), are included because textiles and leather production are part of the *industry* sector in the International Energy Agency (IEA) World Energy Balances. To maintain consistency in the data sources across all the sectors analysed and to integrate the supply side with the OECM, this sub-sector cannot always follow the GICS categorization.


**Table 4.2** OECM 1.5 °C industry pathways and the associated GICS sectors

# *4.1.2 OECM 1.5 °C Service and Energy Pathways and the Associated GICS Sectors*

The four service sectors analysed are distributed across four GICS sectors. Agriculture and food processing is part of *consumer staples* (30), forestry and wood products are part of the *materials* group (1510 50), the fsheries industry is only represented by its actual product (fsh), and the fshing feet is not part of the GICS classifcation (Table 4.3). Finally, water utilities are part of the wider *utilities* group (55). The OECM 1.5 °C pathways for the primary energy supply are all included in the *energy* group (10), whereas the secondary energy supply is part of the *utilities* group (55) (Table 4.4).

# *4.1.3 1.5 °C Pathways for Buildings and Transport and the Associated GICS Sectors*

The OECM pathways for buildings are all included in GICS sector 60—*real estate* (Table 4.5). However, it is unclear to what extent the actual electricity demand, especially of residential buildings, can be considered as part of an economic activity and therefore as the responsibility of the real estate industry itself. Whereas the energy demand for climatization (heating and cooling) is directly related to the building envelope and architecture and is therefore the responsibility of the real estate industry, the electricity demand for appliances is not related and is the responsibility of private households.




**Table 4.4** OECM 1.5 °C energy pathways and the associated GICS sectors

**Table 4.5** OECM 1.5 °C energy pathways and the associated GICS sectors


Finally, the transport sector is part of the *industrials* group (20) and is represented as a subgroup under *transportation* (20) (Table 4.6).


**Table 4.6** OECM 1.5 °C energy pathways and the associated GICS sectors

# **4.2 Adaptation of Energy Statistical Databases to the GICS Industry System**

The OECM uses the IEA World Energy Statistics and Balances (IEA, 2021b) as one of the main input sources for energy demand and supply data for the base year and the historical time series for model calibration, as described in Chap. 3. To develop energy scenarios that are based on the GICS classifcation, the IEA fnal energy demand sectors used for statistical data must be adapted to GICS sectors. This section provides an overview of the two different categorization systems and how they differ.

# *4.2.1 The Industry Sector*

The IEA database documentation (IEA, 2020) provides detailed information about various statistical parameters. Table 4.7 shows the IEA *industry* sector and how it is broken down into four main sub-sectors:


The *manufacturing* sector consists of 11 industries, among the largest of which are iron and steel, chemical and petrochemical industries, and non-metallic minerals, which includes the cement industry. The IEA (IEA DB, 2020) identifes the


**Table 4.7** IEA World Energy Balances—defnition of the *industry* sector (IEA, 2021a; ISIC, 2008)

(continued)


**Table 4.7** (continued)


subgroups of all economic sectors based on the International Standard Industrial Classifcation of All Economic Activities (ISIC) of the United Nations (ISIC, 2008).

The *iron and steel* sector, for example, includes all activities listed under ISIC divisions 241 and 2431. The ISIC lists under Division 241 *manufacture of basic iron and steel* and under 2431 *casting of iron and steel: This class includes the casting of iron and steel,* i.e. *the activities of iron and steel foundries. This class includes:*


However, the IEA statistics do not provide a further breakdown of the energy demand for the specifc economic activities listed under the ISIC divisions but lump them together. In terms of iron and steel, only one value is provided, and no further details are available. To match the IEA sector with the GICS sectors, the *industry* and *service* sectors of the IEA have been grouped according to GICS classes. The iron and steel industry is part of the GICS industry sector 15 *materials* (Table 4.2), subclass 151,040 *metals and mining*, and the sub-industry 15104050 *steel*. This group includes iron ore, as identifed in the documentation. However, the same group (15 *materials*) also lists the *aluminium* industry (15194010)—a separate IEA statistical sector. Although the industry sectors of the IEA and the GICS systems correspond to a large extent, the *service* sector has signifcant differences.

# *4.2.2 The Service Sector*

The IEA statistics do not have a *service* sector category as such. Under *other sectors*, the energy demand is broken down into four subgroups: (1) residential, (2) commercial and public services, (3) agriculture and forestry, and (4) fsheries.

Detailed data for *water utilities*, for example, are not available, and are part of *commercial and public services*, as highlighted in Table 4.8. When sector-specifc data are not available, the energy demand has been estimated from the energy intensities based on GDP ([MJ/\$GDP] or commodities, such as energy demand per cubic meter of water withdrawn [MJ/billion m3 water]). Furthermore, the service sectors *agriculture and food* and *forestry & wood products* (Chap. 6) are partly from IEA's *other sectors* and partly from the *industry* section. Therefore, the current and future energy demand for agriculture and forestry has been derived bottom-up from energy intensities and calibrated with statistical data from the IEA for the years 2005–2019.

# *4.2.3 The Buildings Sector*

The 1.5 °C OECM pathway for buildings (Chap. 7) consists of three sub-sectors: residential and commercial buildings and construction. The IEA statistics for the buildings sector is comprised of 'residential' and 'commercial and public services', excluding water utilities. There are also economic activities, such as Div. 38 ('waste collection, treatment and disposal activities; materials recovery'), that are outside the OECM scenario breakdown. Therefore, the *buildings* sector has been calculated separately with a bottom-up approach, from the foor space and energy intensities per square meter, to project the current and future energy demands. The energy data for *construction*, which is part of the *industry* group, are taken from the IEA statistics.

# *4.2.4 The Transport Sector*

Statistical data for the transport sector in the IEA database best match the GICS classifcation '2030 Transportation', and the development of the OECM 1.5 °C pathways for aviation, shipping, and road transport is based directly on the IEA statistics. Table 4.9 describes the IEA data series for transport.


**Table 4.8** IEA World Energy Balances—defnition of *other sectors*


**Table 4.8** (continued)

**Table 4.9** IEA World Energy Balances—defnition of the transport sector


The reported differences between IEA and GICS categorization systems lead to some inconsistencies, and discrepancies between the available statistical energy data and the actual energy demands for specifc economic activities are unavoidable. The advantage of the high technical resolution of the OECM is also a disadvantage because it requires a signifcant amount of data, which are sometimes unavailable. Therefore, the energy demand projections may vary from those in other sectorial analyses.

# *4.2.5 From Sectorial Energy Scenarios for Industry Sectors to Emissions*

The fnance industry requires sectorial energy scenarios for the industry and service sectors to set sector-specifc decarbonization targets. Increasingly, investment decisions of international and national banks, insurance companies, and investor groups are driven by key performance indicators (KPIs) not only for proftability but also with regard to the embedded GHG emissions of a company. For asset managers, it has become increasingly important to have access to detailed information about GHG emissions, e.g. whether or not a steel manufacturer is on a decarbonization trajectory. The emissions must be further divided according to the responsibility for those emissions. This is done by calculating the so-called *Scopes 1*, *2*, and *3*.

# **4.3 Methodologies for Calculating** *Scopes 1***,** *2***, and** *3*

# *4.3.1 Calculation of* **Scopes 1***,* **2***, and* **3**

Reporting corporate GHG emissions is important, and the focus is no longer only on direct energy-related CO2 emissions but includes other GHGs emitted by industries. These increasingly include the indirect emissions that occur in supply chains (Hertwich & Wood, 2018). The Greenhouse Gas Protocol, a global corporate GHG accounting and reporting standard (WRI & WBCSD, 2021), distinguishes between three 'scopes':


The United States Environmental Protection Agency (US EPA) defnes *Scope 3* emissions as 'the result of activities from assets not owned or controlled by the reporting organization, but that the organization indirectly impacts in its value chain. They include upstream and downstream of the organization's activities' (EPA, 2021). According to the EPA, *Scope 3* emissions include all sources of emissions not within an organization's *Scope 1* and *2* boundaries, and *Scope 3* emissions of one organization are *Scope 1* and *2* emissions of another organization. *Scope 3* emissions, also referred to as 'value chain emissions', often represent the majority of an organization's total GHG emissions (EPA, 2021).

Whereas the methodologies for calculating *Scope 1* and *Scope 2* emissions are undisputed, the method of calculating *Scope 3* emissions is an area of ongoing discussion and development (Baker, 2020; Liebreich, 2021; Lombard Odier, 2021). The main issues discussed are data availability, reporting challenges, and the risk of double counting. MSCI, for example, avoids double counting by using a 'deduplication multiplier of approximately 0.205' (Baker, 2020). This implies that the allocation of emissions based on actual data is not possible. Accounting methodologies for *Scope 3* emissions have been developed for entity-level accounting and reporting (WRI & WBCSD, 2013).

By contrast, the OECM model focuses on the development of 1.5 °C net-zero pathways for industry sectors classifed under the GICS (MSCI, 2020), for countries or regions or at the global level. Emission-calculating methodologies for entitylevel *Scope 3* require bottom-up entity-level data to arrive at exact fgures. Therefore, data availability and accounting systems for whole industry sectors on a regional or global level present signifcant challenges.

Therefore, *Scope 3* calculation methodology must be simplifed for country-, region-, and global-level calculations and to avoid double counting. In the Greenhouse Gas Protocol, *Scope 3* emissions are categorized into 15 categories, shown in Table 4.10.

To include all the upstream and downstream categories shown in Table 4.10 for an entire industry sector is not possible because, frstly, complete data are not available, for example, how many kilometres employees commute—and, secondly, it is impossible to avoid double counting, for example, when calculating *Scope 3* for the car industry.

The OECM methodology is based on the *Technical Guidance for Calculating Scope 3 Emissions* of the World Resource Institute (WRI & WBCSD, 2013) but is


**Table 4.10** Upstream and downstream *Scope 3* emission categories (WRI & WBCSD, 2013; Baker, 2020)

simplifed to refect the higher levels of industry- and country-specifc pathways. The OECM defnes the three emission scopes as follows:

*Scope 1*—All direct emissions from the activities of an organization or under its control, including fuel combustion on site (such as gas boilers), feet vehicles, and air-conditioning leaks.

*Limitations of the OECM Scope 1 analysis*: Only economic activities covered under the sector-specifc GICS classifcation that are counted for the sector are included. All energy demands reported by the IEA *Advanced World Energy Balances* (IEA, 2021a) for the specifc sector are included.

*Scope 2*—Indirect emissions from electricity purchased and used by the organization. Emissions are created during the production of energy that is eventually used by the organization.

*Limitations of the OECM Scope 2 analysis*: Because data availability is poor, the calculation of emissions focuses on the electricity demand and 'own consumption', e.g. that reported by the IEA, 2021b for power generation.

*Scope 3—*GHG emissions caused by the analysed industry that are limited to sectorspecifc activities and/or products classifed by the GICS. *Limitations of the OECM Scope 3 analysis*: Only sector-specifc emissions are included. Traveling, commuting, and all other transport-related emissions are reported under *transport*. The lease of buildings is reported under *buildings*. All other fnancial activities, such as *capital goods*, are excluded because no data are available for the GICS industry sectors and would lead to double counting. The OECM is limited to energy-related CO2 and energy-related methane (CH4) emissions. All other GHG gases are calculated outside the OECM by Meinshausen and Dooley (2019).

The main difference between the OECM and the World Resources Institute (WRI) concept is that the interactions between industries and other services are kept separate. The OECM reports only emissions directly related to the economic activities classifed by GICS. Furthermore, the industries are broken down into three categories: primary class, secondary class, and end-use activity class.

Table 4.11 shows a schematic representation of the OECM *Scope 1*, *2*, and *3* calculation methods according to GICS class, which are used to avoid double counting. The sum of *Scopes 1*, *2*, and *3* for each of the three categories is equal to the actual emissions. Example: The total annual global energy-related CO2 emissions are 35 Gt in a given year.


Double counting can be avoided by defning a primary class for the primary energy industry, a secondary class for the supply utilities, and an end-use class for all the economic activities that use the energy from the primary- and secondaryclass companies. The separation of all emissions by the defned industry categories—such as GICS—also streamlines the accounting and reporting systems. The


**Table 4.11**Schematic representation of OECM *Scopes 1*, *2*, and *3* according to GICS classes, to avoid double counting volume of data required is reduced, and reporting is considerably simplifed under the OECM methodology.

For a specifc industry sector to achieve the global targets of a 1.5 °C temperature increase and net-zero emissions by 2050 under the Paris Agreement requires that all its business activities are with other sectors that are also committed to a 1.5 °C and net-zero emission targets.

The results of the OECM *Scope 3* analysis are documented in Chap. 13.

# **References**


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

# **Part IV Sector-Specifc Pathways**

# **Chapter 5 Decarbonisation Pathways for Industries**

**Sven Teske, Sarah Niklas, and Simran Talwar**

**Abstract** The decarbonisation pathways for the industry sectors are derived. The energy-intensive chemical industry, the steel and aluminium industries, and the cement industry are briefy outlined. The assumptions for future market development used for the scenario calculations are documented, and the assumed development of the energy intensities for product manufacture is presented. An overview of the calculated energy consumption and the resulting CO2 intensities is given, with the assumed generation mix. The textile and leather industry is also included in this chapter because of its strong ties to the chemical industry and meat production (part of the service sector).

**Keywords** Net-zero pathways · Industry · Chemicals · Textile and leather · Steel · Aluminium · Cement · Energy intensities · Bottom-up demand projections

The global gross domestic product (GDP) in 2019 was US\$87.8 trillion, 3% of which came from agriculture, 26% from industry, 15% from manufacturing, and the remaining 65% from services (World Bank, 2021). The aluminium, steel, and cement industries each had a 1% direct share of the global industry GDP value, and the chemical industry's share was 17%, although the indirect effects of those industries on the GDP were signifcantly higher. The materials produced by these four industries are essential for the manufacturing and service industries, which generate over 80% of the global GDP. In the next section, the status quo of the aluminium, steel, cement, and chemical industries, and that of the textile and leather industry, is briefy described. Their current production processes and energy intensities (by product unit or GDP value) and their effciency potentials are documented. The assumptions made for the energy demand of the industry sectors if they are to achieve the OECM 1.5 °C pathways and their energy-related CO2 emissions are also presented.

S. Teske (\*) · S. Niklas · S. Talwar

University of Technology Sydney – Institute for Sustainable Futures (UTS-ISF), Sydney, NSW, Australia e-mail: sven.teske@uts.edu.au

The section discusses the development of the energy demand for the industry sector, as defned in the International Energy Agency (IEA) World Energy Balances (IEA, 2020a). The section focuses on the *materials* group (1510) in terms of the *Global Industry Classifcation Standard* (GICS) classifcation, plus the textile and leather industry, which is included in the IEA industry statistics, but is classifed as *consumer discretionary—textiles* (2520 3030) (see Chap. 4, Sects. 4.1 and 4.2).

# **5.1 Global Chemical Industry: Overview**

The chemical industry is an important intermediate industry, engaged in the conversion of raw materials, such as fossil fuels, minerals, metals, and water, into a variety of chemical products used in other industrial sectors, including pharmaceuticals, fertilisers, pesticides, plastics, dyes, paints, and consumer products. Close overlaps exist between chemical and plastic industries, and many chemical producers are also involved in the manufacture of plastics. Revenue from the global chemical industry increased by 48% to US\$3.9 trillion between 2005 and 2019 (Garside, 2020; ACC, 2021). Pharmaceuticals had the largest share in the segment-wise breakdown of the global chemical shipments in 2019 at 26.4%, followed by bulk petrochemicals and intermediates (16.4%), specialty chemicals (16%), plastic resins (12.2%), agricultural chemicals (8.6%), consumer products (8.3%), inorganic chemicals (7.1%), manufactured fbres (4%), and synthetic rubbers (1%). Together, the world's 100 leading chemical companies generated US\$1.05 trillion in revenue in 2019 (ACC, 2020).

Basic organic and inorganic chemicals account for the highest shares of production and consumption (by volume) in the global chemical industry (UNEP, 2019):


# *5.1.1 Major Chemical Industry Companies and Countries*

BASF (headquarters [HQ] in Germany), Dow (HQ USA), and Sinopec (HQ China) were some of the world's largest chemical-producing companies (based on sales) in 2018. Each of these three leaders exceeded US\$65,000 million in chemical sales. Eighteen countries were represented in the list of the top 50 chemical companies in 2019, and more than 50% of them were headquartered in the USA (10), Japan (8), and Germany (5) (ACS, 2019). German companies BASF, Bayer, and Linde are the foremost international producers. BASF, for example, owns global operations in the chemical industry and is active across the entire value chain, spanning the manufacture of chemicals, plastics, performance products, functional and agricultural solutions, oil, and natural gas. Bayer is a well-known pharmaceutical and chemical manufacturer, and the Linde group owns large industrial gas and engineering facilities, which produce various gas products, including atmospheric oxygen, nitrogen, and argon.

# *5.1.2 Chemical Manufacturing and Energy Intensity*

The chemical industry uses raw materials from natural gas, ethane, oil-refning byproducts (including propylene), and salt to manufacture bulk chemicals, such as sulphuric acid, ammonia, chlorine, industrial gases, and basic polymers, including polyethylene and polypropylene. The manufacturing activity within the chemical industry can be divided into two main categories: basic chemicals and chemical products.

Basic chemicals are those chemicals that feed into the manufacture of other complex chemicals. Petroleum and coal products can be considered basic chemicals because they are used in the manufacture of a variety of polymers, fbres, and other chemicals. The manufacturing processes for basic chemicals, including inorganic chemicals, organic chemicals (such as ethylene and propylene), and agricultural chemicals, are considered energy-intensive industries and require large production facilities.

The second category involves the manufacture of ammonia, polyethylene, and other chemical products. Ammonia production is an energy-intensive process and is considered to be an important contributor to the chemical industry's energy and emission footprints. Ammonium nitrate is used as an agricultural fertiliser and as a blasting explosive in the mining industry. Polyethylene, a by-product of the petrochemical industry, is produced from ethane feedstock and has a variety of uses in the plastic industry. All other chemical products, such as pharmaceuticals, cleaning products and detergents, cosmetics, paints, pesticides and herbicides, fertilisers, and plastic and rubber products, mainly require non-energy-intensive manufacturing processes (USEIA, 2016). Production facilities range from small to large enterprises, with energy supplied by either gas or electricity.

# *5.1.3 Chemical Industry: Sub-sectors Chosen for the OECM Analysis*

To prepare the decarbonisation pathways, we have broken down the chemical industry into the following sub-sectors. These sub-sector classifcations are based on the main applications for chemical feedstocks and follow the categorisation based on the American Chemistry Council (ACC, 2020).


The most important raw materials and chemical products of those fve chemical industry subgroups are described below. The division into these subgroups was based on the available economic data required for market projections. An assessment of market development on the basis of the material fow would be more precise but was beyond the scope of this research because of the large variety of products produced by the chemical industry. The analysis focuses on the development of the chemical industry's energy requirements.

#### **5.1.3.1 Sub-sector 1: Pharmaceuticals**

**Products and materials** There are two key stages in pharmaceutical production: (i) the manufacture of the active pharmaceutical ingredient (API) and (ii) the production of the formulation. An API is the part of the drug that generates its effect. The production of APIs is usually chemically intensive, involving reactors specifc for the manufacture of specifc drug substances. Formulation production is a physical process, in which substances known as 'excipients' are combined with APIs to create consumable products (tablets, liquids, capsules, creams, ointments, and injectables).

**Production and processes** The world's largest pharmaceutical companies are headquartered in the USA and Europe, although production activities are centred in Asia. Some of the biggest pharmaceutical companies are Pfzer (USA), Roche, Novartis (Switzerland), Merck (USA), and GlaxoSmithKline (UK). Until the mid-1990s, the USA, Europe, and Japan supplied 90% of the world's demand for APIs. However, China's low-cost manufacturing sector and weak environmental regulations have meant that a signifcant proportion of API production has now shifted, with almost 40% of all APIs currently supplied by China. Together, China and India supply almost 75% of the API demand of pharmaceutical manufacturers in the USA. China's dominance in API production is balanced by India's leadership in global formulation production and its biotechnology sector. India is also the third largest producer of pharmaceuticals, by volume, supplying most of Africa's demand. India hosts the highest number of United States Food and Drug Administration (US FDA)-sanctioned production facilities outside the USA and supplies 40% of the US generic drug market. Despite India's vast pharmaceutical manufacturing industry, the country still imports 70% of its API demand from China.

**Uses and applications** Pharmaceutical products primarily service the health-care sector, with prescription and over-the-counter drugs, vaccines, and other pharmaceutical applications for human and veterinary use. The biotechnological production of crop seeds, value-added grains, and enzymes is a rapidly growing segment of the industry.

### **5.1.3.2 Sub-sector 2: Agricultural Chemicals**

**Products and materials** Agricultural chemicals are a type of specialty chemical, and the term refers to a broad variety of pesticide chemicals, including insecticides, herbicides, fungicides, and nematicides (used to kill round worms). Agrichemicals can also include synthetic fertilisers, hormones, and other chemical growth agents, as well as concentrated varieties of raw animal manure (Speight, 2017). The main raw materials for nitrogen fertilisers are natural gas, naphtha, fuel oil, and coal, whereas phosphate fertilisers are based on naturally occurring phosphate rocks or synthetic ammonia.

**Production and processes** Some of the large agrichemical chemical producers are Syngenta, Bayer Crop Science, BASF, Dow AgroSciences, Monsanto, and DuPont. The fertiliser industry is structured around a few producers who supply the base chemicals to downstream manufacturers. The production facilities usually specialise in single-nutrient or high-nutrient fertiliser products and are located in close proximity to raw material suppliers (petrochemical producers) or agricultural regions (Roy, 2012).

**Uses and applications** Unsurprisingly, large-scale farming, also referred to as 'industrialised agriculture', is one of the primary users of agrichemicals. In 2010–2011, the global demand for primary plant nutrients was 178 megatonnes (Mt). China (57 Mt), the USA (20 Mt), and India (28 Mt) were the highest consumers.

### **5.1.3.3 Sub-sector 3: Inorganic Chemicals and Consumer Products**

**Products and materials** Inorganic chemicals are materials derived from metallic and non-metallic minerals, such as ores or elements extracted from the earth (e.g. phosphate, sulphur, potash), air (e.g. nitrogen, oxygen), and water (e.g. chlorine). Other examples include aluminium sulphate, lime, soda ash (sodium carbonate), and sodium bicarbonate. The outputs of the chemical industry are used in the manufacture of consumer products, such as soaps, detergents, bleach, toothpaste and other oral hygiene products, and personal care products, such as hair care, skin care, cosmetics, and perfumes.

**Production and processes** Basic chemicals are typically produced in large-scale capital-intensive facilities with high-energy demands. Industrial gases, which are also products of the inorganic chemical industry, are heavily used in the production processes associated with steel, other chemicals, electronics, and health-care products. Many global factors infuence the production of industrial gases. These factors include high capital intensity, increased consolidation of operations and geographic concentration, service orientation, and innovations in key technologies, such as membrane separation. The chemical conversion processes for consumer products are basic, and the key raw materials include fats, oils, surfactants, emulsifers, other additives, and basic chemicals. Consumer products are usually formulated in batchtype operations, which involve equipment for mixing, dispersing, and flling (ACC, 2020).

**Uses and applications** The applications of inorganic chemicals are diverse. For example, chlorine is an important ingredient used to bleach paper pulp and purify drinking water and is used in oil-refning and the steel industry, and caustic soda is used in the production of soaps and detergents. These consumer products are heavily dependent upon vast distribution channels and product segmentation. Therefore, the supply chain and marketing costs are important determinants of the product price, which is also increased by the need for ongoing product development.

### **5.1.3.4 Sub-sector 4: Manufactured Fibres and Synthetic Rubber**

**Products and materials** Manufactured fbres, also referred to as 'synthetic fbres', consist of *cellulosic fbres*, such as acetate and rayon, and petrochemical-derived *polymeric fbres*, such as acrylics, nylon, polyesters, and polyolefns. There are several types of synthetic rubber, including butyl rubber, ethylene-propylene-diene monomer terpolymers, neoprene, nitrile rubber, styrene-butadiene rubber, and specialty elastomers (ACC, 2020).

**Production and processes** Synthetic or artifcial fbres are derived from polymer industries using processes such as wet spinning (rayon), dry spinning (acetate and triacetate), and melt spinning (nylons and polyesters). Synthetic rubbers have highly fexible material characteristics, and the process of 'vulcanisation' is used to crosslink elastomer molecules.

**Uses and applications** Plastics, synthetic rubber, and manufactured fbres account for the second highest share (30%) of the total energy consumed by the chemical industry in the USA, preceded by petrochemicals and other basic chemicals, which have a 49% share (ACC, 2020). Synthetic fbres are heavily used in apparel, home furnishings, and automotive and construction industries. Similarly, synthetic rubber is in high demand in automotive manufacturing, construction, and consumer products. Synthetic fbres are increasingly used in textile manufacture because of their durability and abundance and their ability to be processed into long fbres or to be batched and cut for processing. Natural fbres, such as wool, silk, and leather, are most frequently used for high-quality and long-lasting garments, whereas synthetic fbres are popular in the manufacture of fast fashion garments and accessories (ILO, 2021).

#### **5.1.3.5 Sub-sector 5: Petrochemicals**

**Products and materials** Petrochemicals are chemical products derived from petroleum-refning and from other fossil fuels, such as natural gas and coal. The two main classes of petrochemicals are olefns and aromatics. Ethylene, propylene, and butadiene are examples of olefns—ethylene and propylene are used in the manufacture of industrial chemicals and plastic products, whereas butadiene is used to manufacture synthetic rubber. Olefns also form the base compounds in the manufacture of the polymers and oligomers used in plastics, resins, fbres, elastomers, lubricants, and gels.

Benzene, toluene, and xylene isomers are examples of aromatic compounds and are primarily produced from naphtha derived from petroleum-refning. Benzene is used as a raw material in the manufacture of dyes and synthetic detergents, whereas xylene is used to manufacture plastic products and synthetic fbres.

Apart from olefns and aromatics, other chemical products of the petrochemical industry include synthetic gases used to make ammonia and methanol (in steamreforming plants), methane, ethane, propane, and butanes (in natural gas-processing plants), methanol, and formaldehyde. Ammonia is also used in the manufacture of the fertiliser urea, whereas methanol is used as a solvent and chemical intermediate.

Globally, 190 million tonnes (Mt) of ethylene, 120 Mt of propylene, and approximately 70 Mt of aromatics were produced in 2019.

**Production and processes** The USA and Western Europe are home to the world's largest petrochemical producers. Some of the most notable petrochemicalmanufacturing locations are in the industrial cities of Jubail and Yanbu in Saudi Arabia, Texas and Louisiana in the USA, Teesside in the UK, Rotterdam in the Netherlands, and Jamnagar and Dahej in India. The Middle East and Asia are witnessing increasing investment in new production capacities for petrochemical plants, and a vast majority of the global demand is expected to be met from these regions in the coming decade (Cetinkaya et al., 2018). Some of the fastest-growing petrochemical companies in terms of capacity are PetroChina, Reliance, SABIC, Sinopec, and Wanhua. Both olefns and aromatics can be produced during oilrefning by the fuid catalytic cracking of petroleum fractions or with chemical processes. In chemical plants, the process of steam cracking is used to produce olefns from natural gas liquids, such as ethane and propane. A naphtha catalysis process is used to produce aromatics.

**Uses and applications** The petrochemical sector supplies materials for the vast majority of chemical industry applications, such as the manufacture of petrochemical derivatives, aromatics from bulk petrochemicals, olefns, and methanol. Seven petrochemicals supply more than 90% of all organic chemicals: benzene, toluene, and xylene (aromatics); ethylene, propylene, and butadiene (olefns); and methanol (ACC, 2020). Bulk petrochemicals are also transformed into intermediate products and downstream derivatives, such as plastic resins, synthetic rubbers, manufactured fbres, surfactants, dyes, pigments, and inks. The end-user industries for petrochemical products are the chemical industry, automotive industry, building and construction, consumer products, electronics, furniture, and packaging.

# *5.1.4 GDP Projections for the Global Chemical Industry*

The economic development of the global chemical industry is signifcantly more complex than that of the aluminium and steel industries. The product range of the chemical industry is diverse, and the material fow approach used for aluminium and steel is very data-intensive and is therefore beyond the scope of this research. The chemical industry produces materials for almost all parts of the economy from mining to services—and it is therefore intrinsically connected to overall economic development. Consequently, a GDP-based approach has been used to develop the energy demand projections for the chemical industry over the next three decades.

Table 5.1 provides an overview of the projected economic development of the chemical industry and its fve sub-sectors. It is assumed that the chemical industry will follow the trajectory of the global GDP growth and that the chemical industry's share of the global GDP will remain constant until 2050. The sub-sectors are assumed to grow at the same rate as the overall chemical industry, and the market value share of each sub-sector will also remain stable. For example, the pharmaceutical industry had a 26% share of the global chemical industry GDP, just over US\$1 billion, in 2019. With this approach, we assume that this share of 26% will remain constant until 2050 and that the growth rate of each sub-sector will develop in line with the global GDP projections. This is a simplifcation, and the actual development trajectories may vary across all sectors. However, a more nuanced projection of the development of the chemical industry is beyond the scope of this research.

# *5.1.5 Energy Flows for the Chemical Industry*

Natural gas and petroleum products are important energy sources for the chemical industry. Globally, the chemical industry is responsible for 11% of the primary demand for oil and 8% of the primary demand for natural gas (Levi & Pales, 2018). The chemical industry in the USA consumes almost 9% of all petroleum products as feedstock for fuel and power use, natural gas liquids (or liquefed petroleum gases), and heavy liquids (naphtha and gas oil) (ACC, 2020).


**Table 5.1** Projected economic development of the chemical industry (ACC, 2020; World Bank, 2021)

*Petrochemical feedstocks*, such as olefns and aromatics, are extracted from hydrocarbons produced with cracking processes. These feedstocks are used in plastics, pharmaceuticals, electronics, and fertiliser industries. *Methanol* is directly converted from methane in natural gas and does not undergo the cracking process. In the USA, natural gas liquids are used in the production of 90% of olefns, whereas naphtha is the main source (70%) of petrochemical production in Europe and Asia.

The IEA (2018a) mapped the fows of fuel feedstocks in the chemical and petrochemical industries in 2015. Most of the oil feedstock was converted to high-value chemicals, and a large proportion of raw materials for the chemical industry were directly supplied by oil refneries. Ammonia and methanol, both chemicals in high demand, require natural gas as the raw material. China also uses coal in the production of ammonia and methanol. Petrochemical production occurs in very large-scale facilities, and a number of related products can be produced at a single petrochemical facility. This differs from the set-ups for commodity chemicals, where specialty chemicals and fne chemicals are manufactured in discrete batch processes. Historically, the accelerating demand for chemical products in these enduse industries has had an inevitable impact on the energy demand and resultant CO2 emissions of the upstream and overall chemical industry. Together, base chemicals supply the intermediate raw materials for the majority of aforementioned demand industries (IEA, 2018a; Levi & Pales, 2018).

The energy demand in the pharmaceutical industry is largely driven by the critical environmental requirements for temperature, humidity, room pressurisation, cleanliness, and containment. The manufacturing and R&D phases consume a high proportion of the energy demand (>65%), followed by the formulation, packaging, and flling phases (15%). Overall, heating, ventilation, and air conditioning are the highest energy end uses in the industry (>65%), because of the nature of the products manufactured (Centrica, 2021). Another energy-consuming system is the production of compressed air, which has multiple applications and is one of the least energy-effcient functions in a pharmaceutical production facility. There are opportunities for energy and cost savings in this area (Centrica, 2021). In the production of agrochemicals, the energy demand is spread across manufacturing, packaging, and transportation, and the majority of raw materials are derived from the petrochemical industry. The production of nitrogen fertilisers is energy-intensive because the process that converts the fossil-fuel raw materials used to manufacture the usable fertilisers is energy-intensive. In terms of material throughput, 1 tonne of nitrogen fertiliser output consumes 1.5 tonnes of petrol equivalents (Ziesemer, 2007).

# *5.1.6 Projection of the Chemical Industry Energy Intensity*

This brief overview of the energy usage for the sub-sectors analysed has shown that the chemical industry consists of a highly energy-intensive part, which produces the primary feedstock (basic chemicals) and a secondary product manufacturing part, with a relatively low energy intensity, similar to those of other manufacturing industries with energy intensities of < 10 MJ per \$GDP.

The energy demands for the fve sub-sectors—pharmaceuticals, agricultural chemicals, inorganic chemicals and consumer products, manufactured fbres and synthetic rubber, and petrochemical industry—were calculated with the energy intensities provided in Table 5.2, which are based on the IEA Energy Effciency extended database (IEA, 2021a) and our own research. The energy intensities for primary feedstock were also considered in estimating the effciency trajectories of the different sub-sectors. An increase in the effciency of primary feedstock production of 1% per year over the entire modelling period is required to achieve the assumed effciency gains for all sub-sectors. However, inadequate data are available for the specifc energy intensities of the chemical industry, and no detailed breakdown of the electricity and process heat temperature levels is available in public databases. Therefore, our estimates should be seen as approximate values, and more research, in co-operation with the chemical industry, is required. However, the energy requirements of the entire chemical industry are precisely known and were taken from the IEA statistics *Advanced Energy Balances* (IEA, 2020a)

The energy requirements of the sub-sectors were determined on the basis of market shares and GDP and in discussions with representatives of the chemical industry—specifcally members of the Net-Zero Asset Owner Alliance and the Strategic Approach to International Chemicals Management of the United Nations Environment Programme (SAICM UNEP).

Table 5.2 shows the assumed energy intensities per \$GDP for the analysed subsectors of the chemical industry. The production of primary feedstock is


**Table 5.2** Assumed energy intensities for sub-sectors of the chemical industry

signifcantly higher than other chemicals owing to the process feedstock used in end products. The share of primary feedstock within a certain production process informs the level of energy effciency potential. Because no detailed published data are available, the effciency across all sub-sectors of the industry was assumed to be 1% per year. However, more research and greater access to data are required to allow a more detailed bottom-up energy demand analysis of the chemical industry.

# *5.1.7 Projection of the Energy Demand and CO2 Emissions of the Chemical Industry*

The projections of the economic development and energy intensities of an industry yield the overall global energy demand projection for that industry. In another step, the share of electricity required to generate thermal process heat has been estimated. Table 5.3 shows the calculated electricity demand and Table 5.4 the process heat demand by temperature level for the chemical industry sub-sectors.


**Table 5.3** Projected electricity and process heat demand for the chemical industry to 2050


**Table 5.4** Process and energy-related CO2 emissions—chemical industry

Finally, energy-related CO2 emissions have been calculated on the basis of the 1.5 °C energy supply pathway, which is documented in Chap. 12.

# **5.2 Global Cement Industry**

Cement is the second most consumed substance in the world after water and is a central component of the built environment—from civil infrastructure projects and power generation plants to residential houses. Typically made from raw materials such as limestone, sand, clay, shale, and chalk, cement acts as a binder between aggregates in the formation of concrete. Cement manufacture is a resource- and emission-intensive process and is associated with around 7% of the total global CO2 emissions, according to the Intergovernmental Panel on Climate Change (IPCC; Fischedick et al., 2014, p. 750).

The economic value of the global cement industry was estimated to be US\$450 billion in 2015 (McKinsey, 2015). In 2012, the US cement industry's shipment (to support construction projects) was estimated to be US\$7.5 billion (Portland Cement Association, 2019), equivalent to 1.6% of the global revenue. In the EU, the cement manufacturing industry's turnover was estimated to be €15.2 billion in 2015, with €4.8 billion in value added (European Commission, 2018).

Beyond the mining of the raw materials, there are fve main steps in the cement production process:


**Fig. 5.1** Steps in cement production, from mining to product. (Source: IEA, 2021a)


The literature distinguishes between the energy consumed to produce the intermediary product *clinker* (in the form of small rocklike nodules) and the energy consumed for cement production, which is based on clinker.

# *5.2.1 Major Cement Industry Companies and Countries*

Early estimates from the United States Geological Survey (2020) and IEA (2021a) suggest that global cement production reached 4.1 gigatonnes (Gt) in 2019. Over the past decade, global production has averaged close to 4.0 Gt a year, reaching a high of almost 4.2 Gt in 2014 (United States Geological Survey, 2020).

China has become the largest cement producer worldwide, accounting for around 55% of the total global production in 2019 (IEA, 2020b). The second largest producer was India (8%), followed by the USA, Vietnam, Indonesia, and Egypt, with 2% each, and six countries (Iran, Brazil, Russia, Japan, South Korea, and Turkey) each contributed 1% of the global cement production (IEA, 2021b). The remaining 22% of the global production was distributed across all other countries, with production shares of <1% of the global production.

Swiss company LafargeHolcim is the largest single cement producer in the world (responsible for 9% of the global production). Overall, Chinese-owned companies, including the Taiwan Cement Corporation, together account for 13% of the global cement production.

Cement producers in OECD Europe (Switzerland, Germany, and Italy) and OECD America (Mexico) have headquarters in OCED regions but operate cement plants in 50–60 countries worldwide, so cement production-related CO2 emissions are spread across various countries. It is important to note that the fgures on cement production by company are a combination of annual production and production capacity data. Therefore, it is likely that there are discrepancies in the production values (Mt cement per year), because plants often do not meet the plant capacity. The ten largest cement companies produce 32% of the global production (Table 5.5).


**Table 5.5** Top ten global cement producers, their headquarters, annual production (Mt), and number of operational cement plants

*Source*: Construction Review Online (2021)

# *5.2.2 Impact of COVID-19 on Global Cement Production*

The global cement demand decreased by 3% in 2020, but this decline varied signifcantly by region. The largest impacts on the cement industry occurred in Southeast Asia (−10%), Western Europe (−8%), Australia and the Middle East including North Africa (−7% each), and Latin America (−6%) (International Finance Corporation, 2020). The reduction in cement demand due to COVID-19 resulted in a decline in global emissions from the cement industry, estimated at −7–8% globally relative to those in 2019. However, future emission reduction targets for 2025 and beyond are based on 2019 emissions, and it is assumed that the demand for cement will increase to pre-COVID-19 levels by 2022 and 2023. Therefore, the emission targets are based on planned construction projects estimated before the pandemic (International Finance Corporation, 2020).

# *5.2.3 Energy Effciency Standards and Energy Intensities for the* **Cement** *Sector*

#### **5.2.3.1 Thermal Effciency of Cement Production**

In cement manufacturing, a theoretical minimum energy demand of 1850–2800 MJ/t of clinker is determined by the chemical and mineralogical reactions and drying (European Cement Research Academy and Cement Sustainability Initiative, 2017). This demand includes:


The average global thermal energy intensity of clinker production (grey clinker, excluding the drying of fuels) reduced from 4254 MJ/t clinker in 1990 to 3472 MJ/t clinker in 2017 (GNR, 2021). Table 5.6 (GNR, 2021) shows the average regional


**Table 5.6** Selected regional average thermal energy intensities for grey clinker production excluding drying of fuels (MJ/t clinker)

*Source*: GNR (2021)

thermal intensity of clinker production (MJ/t clinker). The thermal intensity of clinker production is highest in the Commonwealth of Independent States (CIS; regional intergovernmental organisation in Eastern Europe and Asia), followed by OECD North America and Africa. The average global thermal intensities by kiln type are shown in Table 5.7 (GNR, 2021).

All data in Sect. 5.2.3 are drawn from the 'Getting the Numbers Right (GNR)' database, an independent database of energy performance and CO2 information for the cement industry. Managed by the Global Cement and Concrete Association (GCCA), GNR compiles uniform data from 877 cement production facilities, which accounted for 19% of the global cement production in 2017.

#### **5.2.3.2 Thermal Effciency by Kiln Type**

There are considerable variations in the thermal effciency of kiln types, and the best-performing kilns (which dry with preheating and pre-calcining) achieved a weighted average thermal energy intensity of 3350 MJ/t clinker in 2017 and the least-effcient kiln (wet/shaft kiln) a thermal energy intensity of 5900 MJ/t clinker. These data are shown in Table 5.7 (GNR, 2021).

# *5.2.4 Global Cement Industry: Processand Energy-Related Emissions*

The cement industry is a major source of global CO2 emissions. However, the data required to estimate the emissions from global cement production are not well documented (Andrew, 2018). Consequently, there is considerable variation between


**Table 5.7** Average thermal energy intensity by kiln type—excluding drying of fuels (MJ/t clinker)

*Source*: GNR (2021)

different global estimates. Two main aspects of cement production lead to direct CO2 emissions:


Globally, energy-related (fuel) emissions made up 35% of cement emissions (0.8 GtCO2/yr), and process emissions amounted to 65% (1.5 GtCO2/yr) in 2019 (IEA, 2020c). The energy-related emissions from the cement industry amount to 7% of the global energy emissions in that year (IEA, 2020c). The average emissions associated with the total cement manufacturing process are shown in Fig. 5.2 (McKinsey, 2021).

A comprehensive analysis of the global process emissions from cement production revealed a wide variety of existing datasets (Andrew, 2018). The total global process emissions was 1.5 GtCO2 in 2018 (Andrew, 2018). Table 5.8 outlines the global process emissions (GtCO2) from cement production between 2000 and 2018 (Andrew, 2018).

**Fig. 5.2** Current average energy (MJ/t cement) and emissions (CO2/t cement) in cement manufacture (Source: McKinsey, 2021)


**Table 5.8** Global process emissions from cement production in 2000–2018, in GtCO2

#### **5.2.4.1 Reduction of the Clinker/Cement Ratio**

The process CO2 emissions released during the production of clinker can be reduced by integrating alternative cement constituents that reduce the clinker/cement ratio. A global clinker/cement ratio of 0.60 is achieved by 2050 under the IEA's 2DS scenario (IEA, 2018b). This represents a fall from 0.65 in 2014, which translates into a reduction in the process CO2 intensity of cement by 30% over that period (the global average carbon intensity for process emissions is projected to reach 0.24 tCO2/t cement by 2050, which will lead to a saving of 364 million tonnes of CO2 (MtCO2) emissions (IEA, 2018b). The OneEarth Climate Model (OECM) also assumes this estimate of the possible decline in process emissions.

*Carbonation* occurs when CO2 diffuses into the pores of cement-based materials and reacts with hydrated products in the presence of pore water. Carbonation starts at the surface of the concrete or mortar and moves progressively inwards. In contrast to the instantaneous emission of CO2 during the manufacture of cement, carbonation is a slow process that takes place throughout the entire life cycle of cementbased materials (Xi et al., 2016).

Xi et al. (2016) reported that the carbonation of cement materials over their life cycles represents a large and growing net sink of CO2, increasing from 0.10 GtC/yr in 1998 to 0.25 GtC/yr in 2013. In total, they estimated that roughly 43% of the cumulative cement process emissions of CO2 produced between 1930 and 2013 have been reabsorbed by carbonating cement materials. They propose that an average of 44% of the cement process emissions produced each year between 1980 and 2013 has been offset by the annual cement carbonation sink. Moreover, between 1990 and 2013, the annual carbon uptake increased by 5.8% per year on average, slightly faster than the 5.4% per year increase in process cement emissions over the same period (Xi et al., 2016).

### **5.2.4.2 New Technologies to Reduce Process Emissions in the Cement Industry**

The decarbonisation of cement production-related process emissions is being tested and is in various stages of development. These new processes and technologies include clinker displacement by optimising the combination of calcined clay and ground limestone as the cement constituents (European Cement Research Academy and Cement Sustainability Initiative, 2017) and the use of alternative binding materials. Alternative binding materials offer potential opportunities for reducing process CO2 emissions and involve t mixes of raw materials or alternatives from those used in Portland clinker, although the commercial availability and applicability of the alternatives differ widely.

#### **5.2.4.3 Post-combustion Carbon Capture Technologies**

Chemical absorption is the most advanced post-combustion capture technology and allows up to 95% optimum capture yields (European Cement Research Academy and Cement Sustainability Initiative, 2017). A plant began operation in Texas in 2015 to chemically capture and transform 75 ktCO2/yr from a cement plant into sodium bicarbonate, bleach, and hydrochloric acid, which could be sold, so that the sorbents, once saturated, need not be regenerated (IEA, 2018b). The use of membranes as a CO2 separation technique is another proposed technology, which could theoretically produce a yield of more than 80%. However, membranes have only been proven at small or laboratory scales, at which recovery yields of up to 60–70% were achieved (European Cement Research Academy and Cement Sustainability Initiative, 2017).

None of the technologies currently under development are assumed for the OECM 1.5 °C pathway because the time of possible commercialisation is yet to be determined.

# *5.2.5 Global Cement Production and Energy Intensity Projections*

Table 5.9 summarises the assumptions of the 1.5 °C OECM cement industry pathway in terms of the projected volume of global cement production, the development of energy intensities for the relevant processes, and the process emissions per tonne of clinker produced. These assumptions are similar, to a large extent, to those made for the IEA Technology Roadmap—Low-Carbon Transition in the Cement Industry projections (IEA, 2018b).

# *5.2.6 Projections of the Cement Industry Energy Demand and CO2 Emissions*

Table 5.10 shows the calculated electricity and process heat demand developments based on the documented assumptions. The breakdown by temperature level is based on the fve cement production steps required and their shares of the overall


**Table 5.9** Assumed global cement market development and production energy intensities




**Table 5.11** Process- and energy-related CO2 emissions—cement industry

energy demand. No detailed statistical documentation of the exact breakdown of the process heat demand by temperature level and quantity is available. Table 5.11 shows the energy-related CO2 emissions—based on the 1.5 °C energy generation pathway—and the expected process emissions.

# **5.3 Aluminium Industry: Overview**

Aluminium is among the most important building and construction materials globally. To understand the opportunities and challenges facing the industry, the global fow of aluminium metal must be considered. Since 1880, an estimated 1.5 billion tonnes of aluminium have been produced worldwide (IAI, 2018a), and about 75% of the aluminium produced is in productive use (IAI, 2018b). In 2019, 36% of aluminium was located in buildings, 25% in electrical cables and machinery, and 30% in transport applications. Aluminium can be recycled, but the availability of scrap is limited by the high proportion of aluminium in use (IAI, 2018a).

# *5.3.1 Bauxite Production*

Primary aluminium production requires bauxite. Bauxite ore occurs in the top soils of tropical and subtropical regions, such as Africa, the Caribbean, South America, and Australia. The largest producers/miners of bauxite include Australia, China, and Guinea. Australia supplies 30% of global bauxite production (M'Calley, 1894). Table 5.12 shows the global distribution of bauxite mine production, aluminium refneries, and production.


**Table 5.12** Aluminium resources, bauxite mines, alumina refneries, and aluminium production (in thousand tonnes) by country

*Source*: United States Geological Survey (2020)

W = Withheld to avoid disclosing company proprietary data

a Estimated net exporter

b Excludes US production

c Only one of the bauxite producers in Guinea refnes the raw material in the country; the other aluminium refneries are owned by Russian exporters and Chinese operators d Includes Canada

# *5.3.2 Aluminium Production*

Globally, 63.7 million tonnes of primary aluminium were produced in 2019 (IAI, 2021a). About 32 million tonnes of aluminium is recycled every year (IAI, 2021b). Global primary aluminium production accounts for two-thirds of the total production. However, not all bauxite-rich countries are among the main aluminiumproducing nations. China dominates global aluminium production. Overall, nine conglomerates are responsible for global aluminium production (31.5 million tonnes/year), and of those, four have their headquarters in China (Statista, 2021): Chalco, Hongqiao Group, Xinfa, and SPIC Aluminum & Power Investment Co. Ltd. (Statista, 2021). As a result, Chinese aluminium companies produce 17.8 million tonnes per year or 57% of the volume produced by the nine major companies (Statista, 2021). Russian aluminium manufacturer Rusal produces 3.8 million tonnes annually, which is 12% of the amount produced by the nine largest

companies. Like China, Russia also owns an aluminium refnery in Guinea (Human Rights Watch, 2018). The Australian/UK mining giant Rio Tinto produces 3.2 million tonnes per year, equivalent to 10.2% of the aluminium produced by the main producers; the UAE aluminium producer EGA produces 2.6 million tonnes per year (8%), the US-owned company Alcoa produces 2.5 million tonnes per year (6.9%), and Norwegian Norsk Hydro produces two million tonnes per year, which is equivalent to 6% of the aluminium produced by the nine top companies (Statista, 2021). Another 1.9 million tonnes per year is produced by other companies.

The proportion of recycled or 'secondary' aluminium production is a key consideration in determining decarbonisation pathways because secondary aluminium production is up to 95% less energy-intensive than its primary production from bauxite (IAI, 2020). The aluminium sector distinguishes between new aluminium scrap (offcuts generated during the manufacture of aluminium) and old scrap (used, discarded, and collected aluminium products). The proportion of aluminium that is recycled can be measured by quantifying the input rate and the effciency rate:


Once collected, the metal losses from recycling processes are usually <2%, so the net metal yield is >98% (IAI, 2018c; based on a 2005 study). The global recycling input rate has remained constant, at around 32%, since 2000 (IAI, 2020). The most recent data show a global recycling input rate of 32% in 2020, whereas in 2018, the global recycling input rate was 33%, and old scrap accounted for 60% of this.

Globally, up to 30 Mt. of primary aluminium was recycled in 2020, equivalent to a recycling rate of 76% (IAI, 2020):


# *5.3.3 Aluminium Production Processes*

An analysis of current and future aluminium production processes is required to understand the decarbonisation opportunities within each process.

*Primary aluminium production* involves the following processes (excluding mining):

1. *Refning bauxite to produce alumina (Bayer chemical process)*: Bauxite contains ores other than aluminium, including silica, various iron oxides, and titanium dioxide (The Aluminum Association, 2021). Alumina, an aluminium oxide compound, is chemically extracted with the Bayer process (Scarsella et al., 2015), in which bauxite ore is ground and then digested with highly caustic solutions at elevated temperatures. Approximately 70% of the global bauxite production is refned to alumina with the Bayer process (The Aluminum Association, 2021).

2. *Smelting*: It is the process of refning alumina to pure aluminium metal (Hall– Héroult electrolytic process). Alumina is dissolved at 950 °C (1,750 °F) in a molten electrolyte composed of aluminium, sodium, and fuorine, to lower its melting point, allowing easier electrolysis. An electrical reduction line is formed by connecting several electrolysis cells in series (Haraldsson & Johansson, 2018). Electrolysis separates alumina into aluminium metal at the cathode and oxygen gas at the anode (M'Calley, 1894).

In the *secondary production of aluminium (aluminium recycling process)*, the process of refning the raw material (bauxite) to alumina is not required. Instead, scrap aluminium is re-melted and refned. Therefore, the energy consumption for this process is much lower than for its primary production (Haraldsson & Johansson, 2018; IAI, 2020).

# *5.3.4 Aluminium Industry: Energy Demand and Energy Intensities*

The amount of energy used to generate a unit of GDP is referred to as the 'energy intensity of the economy' (IEA, 2020d). The IEA analyses the energy intensity for different sectors of the economy per GDP, based on US currency. The energy intensities of primary and secondary aluminium production are reported under the subsector *basic metals*. In 2018, the production of basic metals was responsible for 27% of the energy consumption in the *manufacturing* sector. The sub-sector *basic metals* includes ferrous metals (22% of energy consumption) and non-ferrous metals, such as aluminium, nickel, lead, tin, brass, silver, and zinc, and accounts for 5% of the *manufacturing* sector's energy consumption (IEA, 2020d). Table 5.13 shows



*Source*: IEA (2020a)

a Calculation derived from the total energy consumption of the *basic metals* sector the energy intensities of the total *basic metal* and *non-ferrous metal* sub-sectors by region.

Compared with aluminium production processes, the energy demand for bauxite mining is relatively small. Bauxite mining requires <1.5 kg of fuel oil (diesel) and < 5 kWh of electricity per tonne of bauxite extracted (IAI, 2018a).

*Refning/smelting* The global average energy use for the electrolysis cell is 13.4 kWh per kg of aluminium produced. If rectifers and other cell auxiliaries, such as pollution control equipment, are included, the global average increases to 14.2 kWh per kg of aluminium produced (Haraldsson & Johansson, 2018; IAI data).

*Process heat* The Bayer process is the most energy-intensive process in primary aluminium production. The energy consumed by the Bayer process varies at 7–21 GJ/ tonne (Scarsella et al., 2015). However, the aluminium industry is moving towards more energy-effcient primary production methods. A study of Columbian aluminium-producing companies showed that this energy intensity can be reduced by changing the core elements of the process, including the size, processes, and temperature of the furnaces (Carabalí et al., 2018). That study suggested that energy consumption could be reduced by 32% by installing an oxy-combustion technology, which preheats the combustion air. The costs related to thermal energy could be reduced by 50.5% per tonne of aluminium. However, the investment cost (purchase) of the technology is high, which hinders its widespread application (Carabalí et al., 2018).

# *5.3.5 Global Aluminium Production and Energy Intensity Projections*

The projections for the overall increase in global aluminium production are driven by technology shifts, including in lightweight vehicles and mounting and framing equipment used for solar photovoltaic (PV) panels and large refectors for concentrated solar power plants (IEA, 2020e). The assumed ratio of primary/secondary aluminium is vital for the calculation of the energy demand, because secondary aluminium production is signifcantly less energy-intensive than primary production.

The projection of the global energy demand for the aluminium industry until 2050 is based on the projected volume of aluminium production, recycling rates, and energy intensities of the different steps of aluminium production, from bauxite mining to the raw product (aluminium). The IEA Sustainable Development Scenario projects an annual growth rate of around 1.2% until 2030 and 15% overall growth in production between 2018 and 2030 (IEA, 2020e). This is a projected overall increase in global aluminium production from the current 85 million tonnes per year to just under 150 million tonnes per year.

Table 5.14 shows the projected global aluminium production for the OECM 1.5 °C pathway. The global recycling rate is projected to increase from 32% in 2019 to 45% in 2050 (IAI, 2021c). The increased recycling rate will lead to a signifcant


**Table 5.14** Assumed development of the global aluminium market

decoupling of global bauxite and alumina production from global aluminium production. The effciency ratio of bauxite to alumina is projected to increase from 40% to 45%, which will lead to a reduction in the energy demand.

*Secondary aluminium production* occurs through recycling schemes, after which the aluminium is re-melted and refned. The energy consumption involved is much lower than for the primary production of aluminium (Haraldsson & Johansson, 2018). The aluminium sector distinguishes between new or pre-consumer scrap and old or post-consumer scrap (discarded aluminium products). Of the 33 million tonnes of aluminium recycled in 2019, 20 million tonnes was from old scrap, and 14 million tonnes was from new scrap, and the share of new scrap is expected to reach 24 million tonnes in 2050 (IAI, 2021d).

The projected energy intensities for bauxite mining and aluminium production are shown in Table 5.15. The fuel demand per tonne of mined bauxite mainly comprises the fuel consumed by mining vehicles. The projections for the electricity and process heat demand for primary and secondary aluminium refect the improvements in the industry's effciency in the past decade and assume incremental improvements based on the effciency assumptions and opportunities noted above, but with no disruptive new production technologies.

The IEA (2020a) has reported improvements in the energy effciency (−3% annually) of alumina refning and aluminium smelting between 2010 and 2018. These were due to the highly energy-effcient production in China. Further reductions in global energy intensity (1.2% annually) are required under the IEA Sustainable


**Table 5.15** Assumed energy intensities for bauxite mining and aluminium production processes

Development Scenario, which can be achieved through a shift towards increasing rates of aluminium recycling (Table 5.15). Secondary production must reach 40% by 2030, with a minimum proportion from old scrap of 70% (IEA, 2020e). The IAI projection to 2050, with maximum recycling rates, is 43% secondary production, but material recycled from old scrap will not exceed 70% (IAI, 2021c).

The production and energy intensity data for the aluminium sector were used to calculate the sectorial decarbonisation pathway presented in the following section (5.3.6).

# *5.3.6 Projection of the Aluminium Industry Energy Demand and CO2 Emissions*

Due to the assumed increase in the share of recycled aluminium in global production and the reduced energy intensity per tonne of aluminium produced, a decoupling of the increases in production and energy demand is possible. Between 2019 and 2050, global aluminium production is projected to increase by 75%, whereas


**Table 5.16** Projected electricity and process heat demands for the aluminium industry to 2050

**Table 5.17** Process- and energy-related CO2 emissions—aluminium industry


the overall energy demand will increase by only 12% (Table 5.16). Due to the already high electrifcation rates in the aluminium industry—which are projected to increase further—and the decarbonisation of the electricity supply based on renewable power generation, the aluminium industry can halve its specifc CO2 emissions by 2035 (Table 5.17).

# **5.4 Global Steel Industry: Overview**

Steel is an important material for engineering and the construction sector worldwide, and it is also used for everyday appliances at the domestic and industrial levels. About 52% of steel usage is for buildings and infrastructure: 16% is used for


**Table 5.18** Global crude steel production data by country (million tonnes per year)

a EU-28 is the abbreviation of European Union (EU) which consists a group of 28 countries (Belgium, Bulgaria, Czech Republic, Denmark, Germany, Estonia, Ireland, Greece, Spain, France, Croatia, Italy, Cyprus, Latvia, Lithuania, Luxembourg, Hungary, Malta, the Netherlands, Austria, Poland, Portugal, Romania, Slovenia, Slovakia, Finland, Sweden, the UK) that operates as an economic and political block

mechanical equipment, such as construction cranes and heavy machinery; 12% is used for automotives (road transport); 10% is used for metal products, including tools; 5% is used for other means of transport, including cargo ships, aeroplanes, and two-wheeler vehicles; 3% is used for electrical equipment; and 2% is used for domestic appliances, such as white goods (World Steel Association, 2020a).

This section provides an overview of global steel production. Table 5.18 shows the data for global crude steel production. The World Steel Association (2020a) production data published in *2020 World Steel in Figures* is not complete for all countries, but is complete for North America (119.2 Mt) and the EU 28 (150.2 Mt) (note: Bulgaria, Croatia, and Slovenia are not included in the report).

# *5.4.1 Primary and Secondary Steel Production*

Steel is produced by various routes. Crude or primary steel is produced from iron ore and secondary steel is produced from recycled steel. These two routes use different technologies and different energy sources. The share of secondary steel production increased by 25% globally in 2013 and by 28% in 2018 (IEA, 2020f).

Secondary steel production is limited by the availability of scrap. Currently, the total global scrap steel collection rate is 85% (IEA, 2020f), i.e. on average, 85% of steel consumed or utilised will be collected and recycled (Gauffn & Pistorius, 2018). However, the scrap collection rate varies for different steel applications: for structural reinforcement, it is as low as 50%, whereas for industrial equipment, it is as high as 97% (IEA, 2020f). Secondary steel production is up to 74% less


**Table 5.19** Share of scrap (%) in crude steel production, by region, 2018

*Source*: Bureau of International Recycling (2019)

energy-intensive than making steel from iron ore (primary production) (ISRI, 2019). Altogether, scrap input accounts for about 35% of the total primary steel production.

By 2030, this share should increase to 40% under the IEA Sustainable Development Scenario (IEA, 2020f). The share of scrap in primary steel production varies among countries and from year to year (Table 5.19):


Global steel production is highly concentrated, and 12 companies are responsible for >50% of the global steel production. Steel companies with headquarters in China dominate the sector (Fig. 5.3). Seven corporations based in China are responsible for 30% of the global steel production. European steel manufacturers produce 9% of the global steel, Japanese companies 7%, South Korean companies 4%, and Indian steel manufacturers 3%.

Regional age profles show that production capacity (manufacturing plants) in the steel sector differs among world regions. The average age profle of steel plants in the Asia Pacifc region, including China, is among the youngest (IEA, 2020g); as a result, energy effciency improvement is signifcant. Considering this region is responsible for one-third of the global production, energy effciency improvement had an effect at the global scale.

*Impact of COVID-19 on global steel production* Global crude steel production decreased by 1.4% in the frst 3 months of 2020 compared with that in the same period of the previous year, and in March, a reduction of 6% was reported (World Steel Association, 2020b). The largest declines in steel production in the frst quarter of the year (Q1) occurred in the EU (−10%), Japan (−9.7%), South Korea (−7.9%), and North America (−4%) (World Steel Association, 2020a). The longterm consequences of COVID-19 for the steel sector are unclear. During the Global Financial Crisis (GFC) in 2009, steel production in Europe alone dropped by 30% compared with that in previous years.

**Fig. 5.3** Largest steel manufacturing companies and shares of global production, 2019

# *5.4.2 Technological Overview of Steel Production*

On average, 20 GJ of energy is consumed to produce 1 tonne of crude steel globally (World Steel Association, 2021). The IEA's Tracking Industry Report (, 2020c) showed a gradual decline in energy intensity between 2009 and 2018. The largest year-to-year fall was in 2017–2018, when energy intensity declined by 3.6%. As mentioned earlier, there are two routes by which steel is produced (Table 5.20). Primary or crude steel is produced by the coal- or natural gas-based blast-furnacebasic oxygen furnace (BF-BOF) route, in which iron ore is reduced at very high temperatures in a blast furnace. The iron ore is melted to a liquefed form (pig iron or direct reduced iron [DRI]) and then oxidised and rolled (Table 5.21). Coal or natural gas is required to generate high temperatures of up to 1650 °C. In the secondary production route, scrap steel is melted in electric arc furnaces (EAFs). The EAF route has the lowest emission intensities. In the EAF (gas-fuelled) process, scrap is usually blended at a rate of about 10% with DRI. A more energy-effcient pathway for primary production is to use scrap steel with ore-based inputs in BF-BOF production, usually at a rate of 15–20% scrap (IEA, 2020f).


**Table 5.20** Steel production—main processes

BF-BOF: production of primary steel from iron ore (oxygen is blown through liquid pig iron, increasing its temperature and releasing carbon)

EAF: production of secondary steel from scrap metal


**Table 5.21** Steel production—main processes and energy requirements

*Emission benchmarks for the steel industry* Table 5.22 shows the emission values allowed for the manufacture of steel under the emission trading scheme of the EU (EU-ETS). The manufacture of secondary steel with EAFs is signifcantly less carbon-intensive—in tonnes of CO2 per tonne of steel (tCO2/tonne)—than the production of primary steel by the iron ore-based route, in which hot metal is produced in blast furnaces (BF-BOF route).


**Table 5.22** EU-ETS benchmark values for iron and steel manufacture, as of February 2020

*Source*: EU (2020)

a tCO2e = tonne carbon dioxide emission equivalent—a term that describes a unique global warming impact, includes all GHG emissions (CO2 and non-CO2 emissions)

# *5.4.3 Projections for the Global Steel Industry: Production and Energy Intensity*

To calculate the future energy demand for the global steel industry requires a range of assumptions—from the actual market volume to the recycling rates and energy intensities, to the actual production process itself. Unlike the aluminium industry, steel manufacturing involves GHG emissions that are not related to energy generation but to the process itself. The emission intensity of the steel sector, specifcally steel plants, depends upon the production route (BF-BOF or EAF) and the energy source (Table 5.23). Both routes can, for example, be fuelled by natural gas (IEA, 2020f). The actual process emissions per tonne product for each of the production process options are assumed to remain at current levels.

Table 5.23 shows the assumed development of global iron ore and steel production in million tonnes per year and the shares of primary and secondary steel production for the 1.5 °C OECM steel pathway. All assumed energy intensities, which are dependent on the production technologies used and process emissions that are used for the energy demand projections, are provided.

The global steel market is estimated to grow by 1–1.5% throughout the entire modelling period. The recycling rates are assumed to increase so that the share of secondary steel will grow from 35% in 2019 to 48% in 2050. The shares of electricity for primary and secondary steel in the overall production process are projected to remain at the current levels. Secondary steel production is, to a large extent, based on electricity, whereas primary steel production is 98% dependent upon process heat for the melting processes. The energy and electricity intensities per tonne of manufactured volume for both secondary and primary steel production are based on IEA projections (IEA, 2020f). Table 5.23 shows all the assumed market and energy intensity developments for the global steel industry according to the production process.


**Table 5.23** Assumed market and energy intensity developments for the global steel industry according to the production process

(continued)


#### **Table 5.23** (continued)

**Table 5.24** Projected electricity and process heat demands for the steel industry to 2050


# *5.4.4 Projection of the Steel Industry Energy Demand and CO2 Emissions*

The assumed division between primary and secondary production rates and the assumed production process technologies are key to the energy demand projections. Whereas secondary steel production requires signifcantly more electricity per tonne, its demand for high-temperature process heat is signifcantly lower (Table 5.24).


**Table 5.25** Process- and energy-related CO2 for the steel industry

Furthermore, as the share of primary steel will be reduced, demand for iron ore mining (volumes) that is required will decrease with higher recycling rates.

The energy-related CO2 emissions and estimated process emissions are shown in Table 5.25. Whereas the energy-related emissions are projected to be phased out by 2050, the process-related emissions are not, although they will be signifcantly reduced due to the predominant use of EAF ovens and the phase-out of highemitting BOF ovens.

# **5.5 Textile and Leather Industry: Overview**

The international fashion industry is estimated to be worth US\$2.4 trillion, and the textile and leather industry constitutes a large proportion of it (valued at US\$818.19 billion in 2020) (SC, 2019; GNW 2021). 'Textiles' refer to natural and synthetic materials used in the manufacture of clothing (including fnished garments and ready-to-wear clothing), furniture and furnishings, automotive accessories, and decorative items. Therefore, the textile industry spans activities related to the design, manufacture, distribution, and sale of yarn, cloth, and clothing. We refer to the textile and leather industry and the fashion industry interchangeably, because some data are available for the fashion industry as a whole, to which textiles and leather contribute almost 35% (SC, 2019; GNW, 2021).

The textile and leather industry has close links with the agricultural and chemical industries. Agricultural output provides the raw materials for the textile industry in the form of natural fbres; similarly, the chemical industry outputs are used as synthetic raw materials in the textile industry. Chemical industry products are also used in the processing of fbres into textiles, especially during dyeing processes. Some of the commonest chemical products used in textile production include spinning oils, lubricants, solvents, adhesives, binders, detergents, bleaches, acids, dyes, pigments, and resins (ChemSec, 2021).

Over 60% of textiles are used in the manufacture of apparel. Natural fbre crops, such as cotton, jute, kenaf, industrial hemp, sun hemp, and fax, are used in the manufacture of yarn for textiles, paper, and rope. Natural fbres can also be extracted from animals (sheep, goats, rabbits, and silkworms) and minerals (asbestos). Synthetic fbres are increasingly used in textile manufacture because of their durability and abundance and as by-products of the chemical and petrochemical industries.

Cotton is the most commonly grown natural fbre. The main processes involved include cultivation and harvesting, spinning (yarn), weaving (fabric), and fnishing (textiles). Most natural fbres are short (only few centimetres) and generally have a rough surface. In contrast, synthetic fbres have the ability to be processed as long fbres or batched and cut to be processed like natural fbres. Synthetic or artifcial fbres are derived from polymer industries using processes such as wet spinning (rayon), dry spinning (acetate and triacetate), and melt spinning (nylons and polyesters). Natural fbres such as wool, silk, and leather most often result in high-quality and long-lasting garments, whereas synthetic fbres are popular in the manufacture of fast fashion garments and accessories (ILO, 2021).

The fashion industry's vast scale has raised international alarm about the environmental effects and social equity of many offshore production facilities. In addition to glaring issues like child labour, unsafe working conditions, and inequitable wages, the industry's increasing dependence on energy, non-renewable synthetic fbres, and water is an issue of global concern. Estimates suggest that textile dyeing and treatment processes are responsible for almost 20% of all water pollution from industrial effuent (Ellen MacArthur Foundation, 2017). The fashion industry's environmental impact is spread across the value chain, although the manufacturing process is the most energy-, water-, and chemical-intensive, with high volumes of toxic chemical effuent and wastewater ending up in marine systems. Some of the estimated environmental impacts of the industry are:


Note: All estimates are calculated annually.

(Kant, 2012; GFA, 2017; Quantis, 2018; UNFCCC, 2018; Niinimäki et al., 2020) The International Labour Organization (ILO, 2021) noted that the stages of yarn and fabric production in textile manufacture consume signifcant quantities of water, chemicals, and energy. These stages are also responsible for a large share of GHG emissions from the textile industry. In the leather value chain, 63–68% of emissions are generated during the manufacture of products such as footwear, whereas the production of raw materials accounts for only 20–29% of emissions (Cheah et al., 2013; Quantis, 2018). The United Nations Framework Convention on Climate Change (UNFCCC) Fashion Industry Charter for Climate Action (2018) aims to achieve a 30% reduction in GHG emissions by 2030.

Although the use of recycled fbres in new textiles is gaining momentum, Dahlbo et al. (2017) have cautioned that more research and empirical evidence is required to determine the impact of recycled fbres on the replacement of virgin fbres in the textile value chain and the rebound effects of the reuse and resale of textiles on the

demand for new production. However, the present analysis focuses on the energy demand and supply of the industry and the resulting GHG emissions.

# *5.5.1 Global Textile and Leather Production: Major Companies and Countries*

The textile, clothing, leather, and footwear (TCLF) industry is characterised by geographically dispersed production and high volatility to factors external to the market, driven by rising fuel and material prices, low agricultural yields of natural fbres, escalating geopolitical tensions around offshore manufacturing, and higher costs of labour and capital in erstwhile havens for textile manufacturing, such as China, Sri Lanka, and Bangladesh. Niinimäki et al. (2020) mapped the environmental impacts of the fashion industry (energy demand, chemical use, water demand, waste output) across various value chain activities and the countries that lead in each stage of the value chain (Fig. 5.4). It is evident that the different stages of yarn and textile manufacture have environmental impacts across all categories (other than GHG emissions). Despite the fashion industry's global footprint, a vast proportion of fbre production and garment manufacture occurs in developing countries (Niinimäki et al., 2020).

In terms of consumer spending, the Asia Pacifc region accounted for 37% of the global sales of apparel and footwear in 2018. China had the largest share of demand at US\$380 billion, followed by the USA at US\$370 billion (Lissaman, 2019). Despite the fashion industry's highly fragmented production and sales operations, it is reported that just 20 multinational companies own 138% of the sector's profts. In 2018, fashion brands such as Nike, Adidas, H&M, Uniqlo, Zara, Levi's, Old Navy,

**Fig. 5.4** Environmental impacts of global fashion industry across the value chain. (Source: Niinimäki et al., 2020)

and Ralph Lauren owned 8% of the global sales. Given the highly competitive industry dynamics and the low-proft margins in most of the upstream value chain activities, the industry is faced with mounting international pressure to incorporate sustainable resource management practices. Compounded by the impact of COVID-19, triggering closures and retail degrowth, the industry is struggling, because of its global labour- and resource-intensive operations.

#### **5.5.1.1 Volume of Global Textile Production**

In terms of the raw material demand of the textile industry, cotton had the highest value in 2019 (US\$378.6 billion). However, in terms of volume, polyester recorded a 28% share of the textile demand, as a result of the diversity of its applications in textiles and apparel. Unsurprisingly, China leads global textile production and exports of both raw textiles and fnished garments. Within Asia, the Indian textile industry constitutes 6.9% of the global textile production, valued at US\$150 billion. India is the second largest textile producer, after China, in terms of production volume, and the textile industry contributed 15% to India's export earnings in 2018–2019. The USA leads global production and exports for raw cotton and is also a strong importer of raw textiles and fnished garments (BV, 2020). It is also the third largest textile producer, with its industry valued at US\$76.8 billion in 2018.

# *5.5.2 Impact of COVID-19 on Global Textile Production*

Global textile and apparel exports were valued at US\$750 billion in 2017 and were projected to grow at a compound annual growth rate of 18.7% to US\$971—38 billion by 2021, before COVID-19. Most of this growth is still expected in Asia, although it will be dependent on the recovery of individual economies from the impacts of COVID-19, especially the adversely affected local manufacturing and retail sales industries. Because many countries, especially in Asia and Europe, are still experiencing lockdowns and a slow return to economic resurgence, the TCLF industry's growth trajectory is expected to take at least a few years to return to pre-COVID levels.

# *5.5.3 Resource Requirements of the Textile and Leather Industry*

Textile production is water-, energy-, and chemical-intensive, and high volumes of liquid effuent are disposed of in natural water systems. Beyond production, the impact of textile and leather products at the end of the value chain is problematic because they generate high volumes of waste and the lifespans of many synthetic materials are short.

Reputable fashion events are increasingly promoting the theme of sustainability, and regenerated materials and accessories are being adopted by leading fashion designers. Whereas such initiatives are mainly targeted at material waste streams, there has also been a conscious effort to stimulate the use of natural and regeneratable materials in fashion. For the fashion industry to reduce its energy and emission intensities, systemic shifts must work in tandem. These shifts range from innovation in product design; the use of regenerative materials; more effcient technologies for processing and manufacture; decentralised production; reduced chemical use and dyeing; water cycling; common effuent treatment, especially in developing countries, which are major producers of fast fashion; and business models that accelerate longer use, reuse, sharing, and recovery.

The water footprint of the textile industry is one of the primary resource challenges for the environmental sustainability of the processing and production phases of this sector. The industry has one of the greatest demands for fresh water in the world, arising from the high water consumption across different stages: farming (especially cotton farming), washing and cleaning, textile processing, printing, dyeing, and fnishing (ILO, 2021). The demand for water in the fashion industry is estimated to be 1.5–2.5 trillion gallons annually. In terms of the most polluting processes, textile dyeing accounts for the greatest shares of water use and pollution (SC, 2019).

# *5.5.4 Textile and Leather Industry: Energy Intensities and Emissions*

Energy consumption in the textile industry is signifcantly high in the wet processing stages of dyeing and fnishing, where it is used to generate steam, heat water, and dry fabrics. Alkaya and Demirer (2014) found that almost 46% of the energy demand was for the conversion of natural gas to steam, most of which is used to heat water for wet processing. The energy demand for the drying process was 30% in the same cotton mill (Alkaya & Demirer, 2014). The ILO (ILO, 2021) reported that energy use was the major contributor to the textile industry's GHG emissions, other than the emissions associated with agriculture and farming, manufacturing in the chemical industry, and livestock breeding for leather production.

The textile industry's carbon intensity relies on the type of energy source and production processes used. For example, hard coal and natural gas are the primary sources of industrial heat in India, thus raising the carbon footprint of apparel manufactured in India. China's textile industry accounts for almost 17% of the industrial sector's overall energy demand; in Bangladesh, the textile industry's energy use accounts for 9% of this demand. The type of input material also affects the energy demand over a product's lifespan. For example, a cotton t-shirt may have a higherenergy demand during the consumption phase than during its production, whereas

the energy demand is highest during the production of a viscose garment (Allwood et al., 2006).

The various stages of textile production have different energy intensities, and these also vary signifcantly across regions. Therefore, the assumptions made about energy intensity must be simplifed for any global analysis. Dyeing and fnishing processes are most energy-intensive and, because they are currently supplied with predominantly fossil fuels, have the highest energy-related emissions (36%), followed by yarn production (28%), fbre production (15%), and fabric manufacture (12%) (Quantis, 2018). Despite the longevity and reuse characteristics of natural fbres, GFA (GFA, 2017) found that leather, silk, and wool processing generates the highest emissions per kilogram of material. In contrast, synthetic materials, such as polypropylene and acrylic fbres, record the lowest emissions, although post-use issues, such as microplastic pollution and the diffculty in recycling composite fbres, make natural fbres more sustainable.

# *5.5.5 Projections for the Global Textile and Leather Industry: Production and Energy Intensities*

Table 5.26 shows the assumed economic development and energy intensities for the textile and leather industry used to calculate the 1.5 °C OECM pathway. The energy intensities per product volume (e.g. in tonnes per year) are not available, so the


**Table 5.26** Projected economic development and energy intensities of the textile and leather industry

energy demand is calculated as a product of the assumed economic development in \$GDP and the average energy units required per dollar. This simplifcation was necessary because the level of detail in the available energy demand data for the textile and leather industry on the global level did not allow a more exact approach. Textile mills have a signifcantly higher energy intensity than the clothing industry, which manufactures the clothing in downstream processes. The assumed average energy intensity for both textile and leather sections of the industry is estimated on the basis of the overall energy demand for both industries according to the IEA World Energy Statistics and the GDP shares.

# *5.5.6 Projection of the Textile and Leather Industry Energy Demand and CO2 Emissions*

Analogous to the previous industry energy and emission projections, Tables 5.27 and 5.28 show the results for the textile and leather industry. All values are calculated on the basis of the documented assumptions. Based on the production processes typical of the industry, it is assumed that the process heat demand does not exceed the temperature level of 100 °C. The 1.5 °C OECM pathway requires that the global textile and leather industry decarbonises the required energy demand entirely by 2050, whereas a reduction by almost 50% seems achievable by 2030.


**Table 5.27** Projected electricity and process heat demands for the textile and leather industry to 2050


**Table 5.28** Process- and energy-related CO2 emissions for the textile and leather industry

# **5.6 Energy Demand Projections for the Five Industry Sectors Analysed**

The industry sectors analysed, aluminium, steel, cement, chemical industries, and the textile and leather sector, consume more than half the electricity and process heat demand of the combined industry sectors (Table 5.29). The remaining large energy consumers are in *machinery*, including the manufacturing industry, food processing, mining, and construction. The aim of this sectorial pathway analysis is to inform the fnance industry, which uses industry and service classifcation systems such as GICS. GICS differs from the IEA in the IEA sectors *industry* and *services*, as described in Chap. 4. The energy demand of *food processing*—a subgroup of the IEA industry sector—in the OECM is part of the demand analysis and projections for the *services* sector, whereas the IEA industry sector *construction* is part of the *buildings* analysis. Furthermore, the *transport equipment* sector has been analysed as part of the OECM 1.5 °C pathway for global transport.

Table 5.30 shows that the high-temperature process heat (>500 °C) accounts for two-thirds of the total process heat demand. Consequently, the generation of process heat for specifc industries, such as in arc furnace ovens for steel, aluminium smelters, and process heat plants for chemical processes, is key to the decarbonisation of the global industry sector.

Therefore, the challenge is less the generation of carbon-free renewable power than the implementation of applications and manufacturing equipment especially designed for the cement, steel, and chemical industries. Timely investments in new manufacturing equipment may lead to the early retirement of existing industrial plants. The 1.5 °C global carbon budget of 400 GtCO2 between 2020 and 2050, identifed by the Intergovernmental Panel on Climate Change (IPCC; see Chap. 2),


**Table 5.29** Total electricity demand of the industries analysed

**Table 5.30** Total process heat demand of the industries analysed


**Table 5.31** Total energy-related CO2 emissions of the industries analysed


has set a clear and hard limit for future emissions, and industries must be supported by government policies to implement the required transition to decarbonisation.

The fve main industry sectors are responsible for about 85% of the energyrelated CO2 emissions of the entire *industry* sector and for almost 20% of all global energy-related CO2 emissions (Table 5.31).

# **5.7 OECM 1.5 °C Pathways for Major Industries: Limitations and Further Research**

The development of energy and emission pathways for industry sectors requires an energy model with high technical resolution. Compared with regional and global energy scenarios, sectorial pathways for industries are based on signifcantly more statistical data and must be developed in close co-operation with industry partners. Furthermore, the estimation of carbon budgets for specifc industry sectors (based on GICS) requires a holistic approach, and all sectors must be considered in order to capture the interactions between the different industries and with the energy sector. To estimate a carbon budget based on current emissions for a single sector, such as the aluminium industry, will inevitably lead to inaccurate results because this approach does not consider the possible technical developments in other individual industries. The current discussions of net-zero targets for specifc industries are often developed for a single industrial sector in isolation. This means that the total of all sub-concepts for certain industries may exceed the actual CO2 emitted, and/or the responsibility for the reduction of CO2 may be shifted to other sectors. In this research, bottom-up projections of the energy demand for the chemical, aluminium, and steel industries formed the basis for the supply scenarios for electricity, process heat, and fuels. The supply of carbon-free electricity is the key to the decarbonisation of all industry sectors. Furthermore, the electricity demand will increase with the electrifcation of process heat to replace fuels. Therefore, power utilities will play a crucial role in those industries reaching their decarbonisation targets. The decarbonisation of process heat will require changes in specifc production processes and is therefore the core responsibility of the industry itself. We found that it is technically possible to decarbonise the energy supply of the analysed industries with available technologies. However, the OECM 1.5 °C pathway is not a prognosis, but a backcasting scenario that shows what must be done to achieve the carbon target. More detailed analyses for specifc industry locations, e.g. China or India, are required because our global analysis simplifes processes and calculates energy demand projections on the basis of average global energy intensities. Moreover, energy demand was calculated with energy intensities (e.g. for steel production) derived with a literature search. Energy statistics, especially for the chemical industry, are sparse, and all the energy demand for sub-sectors are based on GDP projections. More research is required for industries in specifc GICS classes, in terms of both statistical data and the current and future energy intensities of industry-specifc processes. A central database of energy intensities and energy demand for each GICS class would signifcantly enhance the level of detail available for the calculation of net-zero pathways in the future.

# **References**


McKinsey. (2021). *Laying the foundation for zero-carbon cement*.


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

# **Chapter 6 Decarbonisation Pathways for Services**

**Sven Teske, Kriti Nagrath, and Sarah Niklas**

**Abstract** The decarbonisation pathways for the service sector are derived. Brief outlines of the agriculture—food and forestry—wood product sectors, fshing industry, and water utilities are presented. The projected development of product quantities or GDP and the assumed development of energy intensities are given. The industry-specifc energy consumptions and CO2 emission intensities are provided in tables. The non-energy-related CO2 emissions for all sectors analysed in this chapter are discussed and quantifed.

**Keywords** Decarbonisation pathways · Service industry · Agriculture · Food · Forestry · Wood products · Water utilities · Fisheries · Energy intensities · Bottom-up demand projections

The service sector contributes 65% of the global gross domestic product (GDP in 2019, US\$ 56.9 trillion (World Bank, 2021). In this analysis, we use the IEA World Energy Balances as the basis for the energy statistics which defnes three main subsectors: 'industry', 'transport', and 'other sectors'.

While 'industry' and 'transport' overlap with their respective GICS classifcation used for the 1.5 °C OECM sectoral pathways to a large extent, the service sector is scattered across several GICS sectors and the IEA 'other sectors' and 'industry' group (see Chap. 4). In this section, we describe four service sectors that supply essential goods:


The combined share of global energy demand of these sectors at about 7.5% is relatively minor. Even though the energy demand is low and current energy-related CO2 emissions contribute only 6% to global CO2 emissions, the non-energy GHG

S. Teske (\*) · K. Nagrath · S. Niklas

Institute for Sustainable Futures, University of Technology Sydney, Sydney, NSW, Australia e-mail: sven.teske@uts.edu.au; kriti.nagrath@uts.edu.au; Sarah.Niklas@uts.edu.au

<sup>©</sup> The Author(s) 2022

S. Teske (ed.), *Achieving the Paris Climate Agreement Goals*, https://doi.org/10.1007/978-3-030-99177-7\_6

emissions are signifcant. Agriculture and forestry are among the main emitter of non-energy CO2, methane (CH4), and nitrous oxide (N2O)—emissions referred to as *AFOLU* (agriculture, forestry, and other land uses) in climate science.

# **6.1 Overview of the Global Agriculture and Food Sector**

The agriculture and food sector is an essential economic sector contributing to food security, livelihoods, and well-being. Valued at 3.5 trillion USD, agriculture, forestry, and fsheries (AFF)1 accounted for 4% of the global GDP in 2019, with the largest contributions from China and India. The value added2 in agriculture3 alone was 0.2 trillion USD (FAO, 2021b; The World Bank, 2019). Value is also added in some of the manufacturing sectors supported by AFF. In 2018, the manufacture of food and beverages contributed 1.5 trillion USD, and the manufacture of tobacco products contributed 167 billion USD (UNIDO, 2020). The corresponding GICS sectors addressed in this section are listed in Table 6.1 (ISIC, 2008).

The most widely produced commodities in the world are cereals, sugar crops, vegetables, and oil crops. The area under agricultural use has been increasing since the 1960s, until it started to plateau at the beginning of this century, with almost 5 billion hectares under cultivation by 2018. China, the United States, and Australia


**Table 6.1** Relevant Global Industry Classifcation Standard (GICS) sectors

<sup>1</sup>Corresponds to ISIC divisions 1–3 and includes forestry, hunting, and fshing, as well as cultivation of crops and livestock production.

<sup>2</sup>Net output of a sector after all the outputs are summed and the intermediate inputs subtracted.

<sup>3</sup> Includes crop and animal production, hunting, and related service activities (ISIC division A\_01).

have the largest areas of agricultural land (FAO, 2021b). Besides land and energy (discussed in the next section), other major inputs to agriculture are fertilisers and pesticides, which have been increasing progressively over time.

The impacts of agriculture, forestry, and other land uses (AFOLU) can be both positive and negative. The IPCC describes AFOLU emissions as follows: 'Plants take up carbon dioxide (CO2) from the atmosphere and nitrogen (N) from the soil when they grow, re-distributing it among different pools, including above and below-ground living biomass, dead residues, and soil organic matter. The CO2 and other non-CO2 greenhouse gases (GHG), largely methane (CH4) and nitrous oxide (N2O), are in turn released to the atmosphere by plant respiration, by decomposition of dead plant biomass and soil organic matter, and by combustion' (Smith et al., 2014).

# *6.1.1 Energy Demand Projection for the Global Agriculture and Food Sector*

Although energy is an important input to agriculture, the sector accounts for only 2.2% of the total fnal energy consumption globally, with oil and oil products meeting most of this demand (IEA, 2020). Generally, as agriculture is industrialised, this energy consumption increases. In regions where most agricultural systems are industrialised, effciency gains may have plateaued (in the United States, after a peak in 2006 [FAO, 2021a)]), and the sectoral fnal energy consumption may even have decreased (in EU, 10.8% decrease since 1998 [Eurostat, 2020]).

However, the global food system is estimated to account for almost one third of the world's total fnal energy demand. In high-GDP countries, approximately 25% of the total sectoral energy is consumed behind the farm gate (agriculture including in fsheries): 45% in food processing and distribution and 30% in retail, preparation, and cooking (Sims et al., 2015). In low-GDP countries, a smaller share is spent on the farm and a greater share on cooking (FAO, 2011).

In this study, projections of the future energy demand for the agriculture and food processing sector are based on GDP development projections. The assumed global GDP projections until 2050 are based on the World Bank and IEA projections (IEA, 2019). It is anticipated that both agriculture and food and processing industries will grow in proportion to the global economy and that their share of the global GDP will remain between 3.5% and 4%. The production volumes for cereals, pulses, and other agricultural products for 2019, shown in Table 6.2, are taken from the Food and Agriculture Organization (FAO) database (FAO, 2021b).

The estimated global population growth is based on UN population projections (UN DESA, 2019) and will decrease evenly from about 1% per year in 2020 to 0.5% per year in 2050. The food production volumes for each product shown will develop accordingly. No dietary or lifestyle changes are assumed in estimating the future energy demand of the agriculture and food processing sector. In addition to


**Table 6.2** Economic development—agriculture and food processing: 2019 and projections towards 2050

Source: Food and Agriculture Organisation of the United Nations (FAOSTAT: Production)

food for human consumption, agricultural products are also needed for animal feed. However, the impacts of diets on agricultural product demand and emissions are discussed in the next section.

According to the IEA's *Advanced Energy Balances* database structure, the food processing industry is part of the *industry* sector, whereas agriculture is part of *other sectors* group. Furthermore, the statistical data for the relevant energy demand are provided as 'food and tobacco', and separate data for the food processing industry are not available. Similarly, the IEA database provides the energy demand for agriculture and forestry, but no further separation of the two industries is available.

To calculate the energy demand for each sub-sector, the economic values in \$GDP energy for agriculture, forestry, food processing, and tobacco industry are divided by the average energy intensities (in MJ per \$GDP) for each of those sectors. Table 6.3 shows a selection of energy intensities taken from the IEA database for different agricultural products. To calibrate the model and to understand the development in the past, statistical data for the years 2005–2019 are used. To project the future energy demand for each of the sub-sectors, the calculation method then


**Table 6.3** Energy intensities for selected food processing industries

changes, and the projected GDP development (Table 6.2) is multiplied by the average sector-specifc energy intensities, incorporating an assumed effciency factor, giving the projected energy demand. For more details of the OECM methodology, see chap. 3.

The average energy intensity of the food processing industry for 2019 has been calculated to be around 3.5 MJ/GDP, and it is assumed that the annual effciency gain is 0.25% on average (Table 6.3). The main energy demand for food processing is for heating processes in the range of 100–500 °C. Based on the study of Ladha-Sabur et al. (2019), the share of thermal energy is estimated to be 75% of the fnal energy demand on average for food processing and the remaining 25% for electricity. Transport energy is not included in this approach because the transport sector is analysed separately (see the Methodologies for *Scopes 1*, *2*, and *3* section).

Based on the methodology described above, the energy demand for the agriculture and farming sector is calculated with an energy intensity of 1.74 MJ per \$GDP for the base year 2019. The majority of the energy demand is estimated to be for fuel for agricultural machinery, such as tractors and harvesters, whereas 30% of the energy is electricity. Effciency gains for the agriculture sector are assumed to be higher—0.8–1% per year—than for the food processing industry.

Table 6.4 shows the calculated energy demand broken down according to the electricity, heat, and fuel requirements for the agriculture and food processing sector. The energy-related CO2 emissions for the calculated demand are based on the 1.5 °C OECM supply scenario (see Chap. 12) (Table 6.5).


**Table 6.5** Energy-related CO2 emissions for agriculture and food processing

**Table 6.4** Energy demand projection for agriculture and food processing


# *6.1.2 Food Demand and Implications*

#### **Food Equity**

The FAO estimates that suffcient global aggregate food is produced for nearly everyone to be well fed. However, income inequalities and resource constraints in different parts of the world mean that everyone is not well fed. Progress towards eliminating hunger and malnutrition is still lagging, with 821 million people undernourished in 2017 (FAO, 2018). However, while we recognise the need for the redistribution of available food calories and a discussion of nutrition, in this research, we take a global aggregate view of food production, rather than a nuanced view of food security and nutritional equity in the local context.

#### **Demand for Agricultural Products**

The key drivers of food (and consequently feed) demand are population growth and changes in consumption patterns, which are driving a shift to a more meat-based diet. The demand for commodities, such as food grains, is primarily driven by increases in population because the per capita food demand is stagnant or even decreasing in several high-income countries (although the demand for coarse grains for use as feed will increase as meat and dairy consumption increases). Income, individual preferences, and changes in lifestyle and consumption patterns will play a greater role in the demand for vegetable oils, sugar, meat, and dairy products (OECD-FAO, 2020). The use of cereals for feed is projected to grow at 1.2% per year over the coming decade as livestock production expands and intensifes in lowand middle-income countries, compared with the projected growth of 1% per year for food use (OECD-FAO, 2021).

The average dietary energy supply per person per day in low- and middle-income countries is around 2750 kilocalories, whereas in high-income countries, it is around 3350 kilocalories. Both these fgures exceed the minimum requirement of around 1950 kilocalories per person per day (FAO, IFAD, & WFP, 2015). It is expected that overall per capita consumption will increase globally, including in developed countries, even as concerns around obesity increase (Alexandratos & Bruinsma, 2012).

The global demand for food for human consumption is the main component of the overall demand for agricultural products. However, non-food uses of several commodities, mainly animal feed and fuel, are important and have experienced faster growth than food for human consumption over the last decade(s). It is anticipated that in the coming decade, the relative importance of food, feed, and biofuel use will remain constant, because no major structural shifts in the demand for agricultural commodities are expected (OECD-FAO, 2020). The global demand for agricultural commodities (including for non-food uses) is projected to grow at 1.2% per year over the coming decade, which is well below the 2.2% per year growth experienced over the last decade. This projected slowdown is due to a lower global demand for biofuels, especially as many high-income and emerging countries achieve saturation levels (OECD-FAO, 2021).

# *6.1.3 Meeting Global Food Demand While Reducing the Environmental Impact of Food Production*

As noted above, a major source of emissions from the agricultural sector is associated with land use. Key complementary strategies for increasing food production while reducing the impact on land use are discussed below, followed by a discussion of the environmental impacts and emissions specifcally related to animal protein production, including enteric emissions. These impacts are fundamentally driven by the overall demand for agricultural products.

#### **Crop Yield**

The substantial additional amounts of food required in the coming decades will mainly be produced through yield increases, rather than any major expansion of cultivated areas (FAO, 2017). The FAO expects 77% of this increased production to come from increased yields, compared with 9% from the expansion of cultivated land and 14% from increased cropping intensities (Alexandratos & Bruinsma, 2012). A review of the scientifc literature showed that most of the focus on how to feed the world is on increasing food production through technological advances, whereas attention on reducing the food demand through dietary changes to lessintensive patterns has remained constant and low (Tamburino et al., 2020).

In either case, crop yields must increase to meet the needs of the growing population without increasing croplands. Agricultural yields have increased without a signifcant increase in agricultural land use in the past. For example, between 1961 and 2000, the global population more than doubled, and the per capita cereal consumption increased by 20%. However, the area of harvested cereals increased by only 7%, largely because cropping intensities increased (Piesse, 2020). Mueller et al. (2012) found that by maximising crop yields (i.e. closing yield gaps), the global crop production could increase by 45–70% with the same land use.

#### **Food Waste**

Another important consideration to improve the effciency of food systems is the reduction of food waste. The energy embedded in global food losses is 38% of the total fnal energy consumed by the whole food supply chain. This means that more than 10% of the world's total energy consumption is food that is lost and wasted. By one estimate, the food losses and waste that occur every year generate more than 3.3 gigatonnes of CO2 equivalents (FAO, 2013), equal to the combined annual CO2 emissions of Japan and the Russian Federation (FAO, 2017).

Kummu et al. (2012) determined that an additional one billion people could be fed if food waste was halved, from 24% to 12%. The World Resources Institute reported that a 25% reduction in food waste would push food production 12% closer to the level necessary to feed the world in 2050 and would reduce the amount of increased agricultural land needed by 27%, inching closer to fully closing the land gap (Ranganathan et al., 2018).

#### **Dietary Changes**

Most developed countries have largely completed the transition to livestock-based diets, although it is unlikely that all developing countries—including India—will shift to levels of meat consumption typical of western diets in the foreseeable future (Alexandratos & Bruinsma, 2012).

The FAO 2030 Agriculture Outlook suggests that near-saturation levels of meat consumption, as well as health and sustainability concerns, might limit the growth of animal protein consumption in high-income countries, particularly reducing the demand for beef. However, the demand for poultry is expected to increase in highincome countries in the move to a more sustainable and healthy diet and in middleand lower-income countries because it is the most economic animal protein (this will also circumvent religious reasons for the non-consumption of meat, such as the consumption of beef and pork in India and Muslim countries, respectively). However, it is estimated that over the next decade, any gains (emission-wise) made from the reduced demand for animal products in developed countries due to increases in vegetarianism or veganism will be offset by the increased consumption of meat in middle-income countries due to lifestyle changes and increasing per capita caloric consumption.

The projected improvements in production effciency will be insuffcient to meet the future food demand without increasing the total environmental burden posed by food production. By contrast, transitioning to less impactful diets would, in many cases, allow production effciency to keep pace with the growth in human demand while minimising the environmental burden of the food system (Davis et al., 2016). Changing diets to a globally adequate diet of 3000 kcal per capita per day, with 20% animal kcal would allow an additional 2.1–3.1 billion people to be fed in 2050 if yield gaps are closed (Davis et al., 2014). Another study showed that a transition towards more sustainable production and consumption patterns could support 10.2 billion people within the planetary boundaries given if cropland is spatially redistributed, water and nutrient management improved, food waste reduced, and dietary changes imposed (Gerten et al., 2020).

#### **Environmental Impacts**

Increased meat production impacts land use in terms of increased pastureland and increased cropland. To accommodate the increasing ruminant production (especially sheep and goats) in sub-Saharan Africa, pastureland is expected to expand by 1.2 Mha. The projected expansion in livestock production in North America will require additional pastureland (+3.22 Mha), with the conversion of marginal croplands (OECD-FAO, 2021).

The other main contributor to agricultural emissions is methane emissions from the enteric fermentation in livestock. Diets rich in meat, particularly that from ruminants such as cattle, are associated with higher environmental costs and higher emissions of GHGs: methane, from enteric fermentation; CO2, which is released from the clearing of forests for pasture; and nitrous oxide (N2O), which is generated in feed production (FAO, 2017) . Diets with a smaller meat component have signifcantly lower emission intensities. The FAO 2030 Agriculture Outlook projects predict that agricultural GHG emissions will grow by 4% between 2018–2020 and 2030, with livestock accounting for more than 80% of this global increase (OECD-FAO, 2021).

Non-energy-related carbon emissions are calculated with the Generalized Equal Quantile Walk (GQW) method, the land-based sequestration design method, and the carbon cycle and climate model (*Model* for the Assessment of Greenhouse Gas Induced Climate Change, MAGICC) (Meinshausen & Dooley, 2019). The model also accounts for other GHG gas emissions arising from the enteric fermentation of livestock (CH4), crop residues and fertilisers, and manure management (N2O).

An industry sub-sector share has been assigned for each GHG, as explained in the attached supplementary material. Only a small part (20%) of the CO2 emissions attributable to changes in land use are assigned to the agriculture sub-sector, with 80% assigned to forestry. Table 6.6 shows the breakdown of the different emission sources in agriculture. These emissions are multiplied by the global warming potential of other GHG gases to obtain the total CO2 equivalents (CO2e) for the sector.

# **6.2 Overview of Global Forestry and Wood Sector**

Forestry contributes to food security, livelihoods, and well-being; supports terrestrial ecosystems and biodiversity; provides (human) life-sustaining ecosystem services; and acts as a carbon sink. Value is also added by some of the manufacturing sectors supported by forestry. In 2018, wood and wood products contributed 183 billion USD, and paper and paper products contributed 324 billion USD to the global economy. Together with agricultural manufacturing, this is about 18% of the value added in total manufacturing globally (UNIDO, 2020). The corresponding GICS sectors addressed are listed in Table 6.7.

Globally, 30% of all forests are used for production. Of this 30%, about 1.15 billion ha of forest are primarily used for the production of wood and non-wood forest products, and another 749 million ha are designated for multiple uses. In contrast, only 10% is allocated for biodiversity conversation, although more than half of total forests have management plans (FAO, 2020a).


**Table 6.6** Non-energy emissions from the agriculture sector


**Table 6.7** Relevant Global Industry Classifcation Standard (GICS) sectors

**Table 6.8** Global economic development of the forestry, wood, and wood products industry


# *6.2.1 Energy Demand Projection for the Global Forestry, Wood, and Wood Product Sector*

The sectoral fnal energy consumption of forestry has remained stable over the last three decades, and half of this demand is met by oil products.

The energy demand of the forestry and wood sector was calculated with the same methodology as for the agricultural and food processing sector (Table 6.8). The IEA Advanced Energy Balances show the wood and wood products separately but combine the energy demand for forestry with that for agriculture. The energy demand for forestry was calculated both as the energy intensity (Table 6.10) multiplied by the global GDP for this sector, as shown in Table 6.9, and by subtracting the calculated energy for agriculture (see previous section) from the combined energy demand for agriculture and forestry provided by the IEA. With this repeated calculation, the energy intensity for forestry, taken from the literature, was evaluated again. The economic values for forestry were taken from FAO 2015 (Lebedys, 2015).

Selected energy intensities of the wood products and paper industry, as well as the average energy intensities, were used to calculate the energy demand for the


**Table 6.9** Assumed energy intensities for the forestry, wood, and wood product industry



forestry industry and the wood and wood product industry. For forestry, it is assumed that the improvement in energy effciency per year will be relatively small, at only 0.25% per year, because this industry is already highly automated (Ringdahl, 2011).

The wood and wood product industry, as defned in the IEA statistic, includes the manufacture of wood and of products made of wood and cork, except furniture, and the manufacture of articles of straw and plaiting materials, as classifed under the United Nations International Standard Industrial Classifcation of All Economic Activities (ISIC, 2008).

The calculated total fnal energy demand, further broken down to the electricity and heat/fuel demand for the forestry and wood product industry, is shown in Table 6.10. The processing of wood to wood products requires considerably more energy than forestry activities. For this reason, in developing the 1.5 °C energy pathway, the energy effciency in this area is given greater importance than that for timber harvesting.


**Table 6.11** Energy-related CO2 emissions of the analysed sectors under the 1.5 °C energy pathway

Based on the 1.5 °C OECM supply scenario documented in Chap. 12, the energyrelated CO2 emissions for the analysed forestry and wood product sector are provide in Table 6.11. To decarbonise the energy supply of the forestry requires to switch machinery such as chainsaws and other heavy-duty tools from combustion engines to electric motors, and all-terrain vehicles need to be electrifed.

# *6.2.2 Land-Use Demand for Forestry*

There is potential for 'nature-based solutions' to remove CO2 from the atmosphere at the gigatonne scale, with potentially signifcant co-benefts (Meinshausen & Dooley, 2019) (see also Chap. 14). Simulations of nature-based approaches, such as forest restoration, reforestation, reduced harvest, agroforestry, and silvopasture, were combined and found to sequester an additional 93 Gt carbon by 2100. This would require an additional 344 million ha of land for reforestation (Littleton et al., 2021). The key pathway for managing land-use change is reforestation, which is limited to biomes that will naturally support forests, by identifying previously forested land in close proximity to intact or degraded natural forests. This comprises of 274 Mha of land in proximity to intact forests in subtropical and tropical forest biomes and another 70 Mha identifed in temperate biomes.

Decarbonisation pathways are being developed at the global level. At this level, there is little confict between the competing uses of cropland, pastureland, and forests for carbon removal. Adopting nature-based approaches, such as agroforestry or silvopasture, where trees are integrated into cropland or grazing lands, will help to increase the carbon stock while meeting the increasing demand for forestry and agricultural products. It should be noted that a lot of deforestation and the capacity


**Table 6.12** Non-energy GHG emissions in the forestry industry

and demand for increased agricultural and livestock products will occur in tropical and subtropical regions, often in developing countries. At the local level, there must be a more nuanced approach to addressing the balance between environmental, economic, and well-being outcomes.

The OECM model also calculates the non-energy GHG emissions from the forestry sector, as shown in Table 6.12. The OECM 1.5 °C net-zero pathway is based on effcient energy use and renewable energy supply only—leading to full-energy decarbonisation by 2050. No negative emission technologies are used and the OECM leads to zero energy-related carbon emissions. The model assumes no net deforestation from 2030 onwards and the adoption of nature-based approaches to land-use management. Therefore, from 2030 onwards, there will be carbon removal or negative emissions.

# **6.3 Overview of the Global Fisheries Sector**

About 7% of the total protein intake globally comes from seafood (FAO, 2020b). Over 200 million tonnes of fsh and seafood are produced annually (Ritchie & Roser, 2021). According to the OECD, the fsheries industry employs over 10% of the world's population (OECD, 2020b). While the overall food fsh consumption expanded by 122% between 1990 and 2018, the global capture fsheries—fsh that has been caught from natural environments by various fshing methods—only grew by 14%. The main rise of fsh 'production' came from aquaculture, which increased output by factor fve. However, the percentage of fsh stocks caught in the open ocean within biologically sustainable levels decreased from 90% in 1909 to only 65.8% in 2018 (FAO, 2020b).

The economic (frst sale) value of the global fshing industry in 2018 was estimated at USD 401 billion, of which USD 250 billion came from aquaculture production (FAO, 2020b).

#### **The Fishing Industry and Their Relevance Within the Energy Sector**

While the fshing industry plays a signifcant role in food supply and economic income for a large part of the global coastal population, its share on global energy demand is minor with less than 0.1% of the global energy demand (IEA, 2020). The IEA World Energy Statistics lumps the energy demand of agriculture, forestry, and fsheries in one category. And even within this category, the energy demand of fsheries only makes up 3% within that group. The energy demand of the agricultural/

forestry sector is with 8900 PJ per year—compared to around 300 PJ annually for fsheries—about 25 times higher (IEA, 2020).

However, the OECM decided to develop a specifc scenario for fsheries due to its importance for small island states. Subsistence fshing is a key economic pillar for island nations in the Pacifc, the Indian Ocean, and the Caribbean. Over the past decades, large fshing vessels have been in dispute with the traditional fsh grounds of local indigenous people.

Marine and aquatic ecosystems are under stress—from climate change, overfshing and unsustainable fshing, and aquaculture practices in some areas, as well as pollution from various other human activities, which lead to ocean acidifcation and declining biodiversity. Furthermore, illegal, unreported, and unregulated (IUU) fshing continues in many parts of the world, adding excessive pressure on fsh stocks, harming law-abiding fshers through unfair competition, and thereby reducing their proftability, in addition to limiting employment opportunities throughout the value chain (OECD, 2020b).

Among the most unsustainable fshing methods is bottom trawling with large vessels which accounts for about one quarter of fsh catch globally. Traditional artisanal fshing boats which are either entirely unpowered or with small outboard engines cannot compete with industrial fshing vessels. Increasing fuel costs make it increasingly uneconomic for the fsherman as fuel costs can often outweigh income from fsh. Besides, most island states still rely on expensive diesel generators to provide electricity for households and cooling equipment for food preservation.

# *6.3.1 Fisheries: Projection of Economic Development and Energy Intensities*

The economic value of the fshery industry is assumed to maintain its current global GDP share of 0.2% and to increase from US\$ 272 billion in 2019 according to growth projection for global GDP to over US\$700 billion in 2050. However, the shares between marine fshing, aquaculture, and inland fshing change signifcantly in favour of aquaculture. Table 6.13 shows all key assumptions used of the 1.5 °C pathway for fsheries.

The projected development of produced fsh in million tonnes per year is certainly arguable, and forecasts of fsh production volumes over the next 30 years are not available—thus the assumption that the volume of wild fsh catch and fsh from aquaculture plateaus on 2020 level, while the market value steadily increases. The rationale behind this is that marine fshing will not be able to increase fshing volumes, while costs and economic values per tonne of fsh continue to increase. The catch per unit effort (CPUE)—the amount of energy per tonne—is assumed to remain stable. In this case, longer distances and sailing time to catch 1 tonne of fsh can be compensated by increased energy effciency of fshing vessels.


**Table 6.13** Key assumption for the energy demand projection of the global fsheries industry

The 1.5 °C OECM pathway for the fshing industry suggests moving away from large-scale fsh trawlers towards a more decentralised feet of fshing boats.

In regard to the fshing vessel feet, 2.07 million vessels were registered in 2019, 1.16 unpowered, 1.63 million powered artisanal vessels, and 0.43 million industrial vessels (Rousseau et al., 2019). The overall motor power of the global fshing feet is estimated with a capacity of 144 GW, 87 GW of which are from industrial vessels. The 1.5 °C pathways assume that the power artisanal fshing vessels steadily increase in numbers on the expense of industrial vessels which lose market shares in a stable fsh market by volume.

The average motor power of artisanal vessel is estimated with 35 kW that operate with around 500 full load hours per year. The electricity share for fshing vessels increases from 0% in 2020 to 2% in 2025, to 4% in 2030, to 16% in 2040, to 64% in 2050.

Table 6.14 shows the resulting energy demand under those documented assumptions and Table 6.15 the expected energy-related CO2 emissions. However, the available data about energy demand of fshing vessels is scarce and the results are indicative. More research is required in order to develop more detailed scenarios for and around the fshing industry, their vessels, and electrifcation concepts for artisanal fshing boats.

Decarbonising the energy and electricity supply of island nations away from diesel generators for electricity generation and gasoline-fuelled outboard engines to renewable-powered—mainly battery solar systems—mini- and micro-electricity grids will afford the island energy independence from expensive fuel supply via boat and planes. While the electrifcation of road vehicles for passenger and freight transport is already progressing worldwide, the electrifcation of ships and fshing vessels is still in its very frst developments. Electric outboard engines, supplied with batteries charged with renewable electricity, can support subsistence fshing and help moving away from destructive fshing practices. However, electric outboard engines are still signifcantly more expensive than two-stroke or four-stroke outboarder, and the market is small. Economies of scale are required to make electric outboard engines—preferably in the range of 30–50 kW—cost-competitive.


**Table 6.14** Projected energy demand for global fsheries industry

**Table 6.15** Energy-related CO2 emissions of the fsheries industry under the 1.5 °C energy pathway


# **6.4 Overview of the Global Water Utilities Sector**

Water is important for basically every process that supports human life on Earth. Keeping potable drinking water of high quality is therefore a basic requirement for the health of humans, for the environment, and for an intact economy. Thus, the economic value of water utilities is far beyond the monetary values of this industry. While the projection of future energy demand for various sectors in the analysis is based on economic values, the energy demand projection for water utilities must be based on production volumes.

The 1.5 °C OECM pathways are developed according to sectors as defned in the Global Industry Classifcation Standard (GICS). Water utilities (5510 40) are a subsector of the GICS sector 55 *utilities* together with electric utilities (5510 10), gas utilities (5510 20), multi-utilities (5510 30), and independent power and renewable electricity producers (5510 50). According to the GICS defnition, water utilities are companies that purchase and redistribute water to the end consumer, including large-scale water treatment systems.

Only a fraction of water utilities globally has been privatised. The global market value of privatised water utility companies in 2020 was USD 158.79 billion (Statistica, 2021). Globally, the largest privatised water utilities are located in China and in the United States and are worth between USD22 and USD33 billion (Fig. 6.1).

However, the majority of member countries of the European Community decided against a privatisation of the water sector. The European Economic and Social Committee called for a stop of water utility privatisation (EESC, 2018), and the controversial debate kept the sector predominately in public ownership. Therefore, US American, Chinese, and companies from the United Kingdom dominate the overview due to their high share of privatisation.

To ensure that drinking water is of high quality, stricter water regulations have been implemented, and treatment practises have been intensifed. As a consequence, energy consumption of wastewater treatment plants increased (Rothausen & Conway, 2011). The energy intensity for wastewater treatment depends on the

**Fig. 6.1** Market value of leading water utilities companies worldwide in 2020, by country, in billion USD. (Source: Statistica (2021))

process and/or technology and the scale of the treatment plant (Paul et al., 2019). As electricity consumption in the water sector grows, the carbon footprint of the sector becomes larger and more signifcant if fossil fuel-based electricity is used. If this electricity is purchased from power utilities, energy costs might be signifcant. In developed countries, water utilities, on average, spend 15–30% of their budget on energy—this is for large wastewater plants—costs for small wastewater treatment plants are higher and make up 30–40% of their budget (Paul et al., 2019). For drinking water plants, the largest energy use (80%) is used to operate motors for pumping (Copeland & Carter, 2017).

# *6.4.1 Water Utilities: Commodity Demand Projections and Usage*

There are a several international organisations that oversee water governance and also offer comprehensive databases; the most relevant organisations are:


For this analysis, we use the FAO AQUASTAT database for all water-related data, which contains detailed data on water withdrawal, usage, and treatment (FAO, 2021a). The OECD database is most useful for OECD regions (North America and Europe) but not as comprehensive for global data. A comparison of the different data is shown in Table 6.13. IEA data on water extraction aligns best with the FAO data for global and for OECD regional. The OECD and World Bank data is particularly patchy; historical data is therefore displayed as averaged values for different timeframes. Considering the diversity of databases and approaches to compile data, the OECM project decided to use the FAO database for global analysis.

The FAO defnes *total water withdrawal* as the 'annual quantity of water withdrawn for agricultural, industrial and municipal purposes. It can include water from renewable freshwater resources, as well as water from over-abstraction of renewable groundwater or withdrawal from fossil groundwater, direct use of agricultural drainage water, direct use of (treated) wastewater, and desalinated water. It does not include in-stream uses, which are characterised by a very low net consumption rate, such as recreation, navigation, hydropower, inland capture fsheries, etc.'.

The FAO water extraction data is based on the following calculation:

Total water withdrawal Municipal water withdrawal Industrial water withdrawal Agricultural water withdrawal


**Table 6.16** Total water withdrawal in billion cubic meters per year including extraction from desalination

Data source: FAO AQUASTAT (2021a) and OECD (2020c) stats (most recent values)

**Table 6.17** Assumed global water withdrawal quantities for the energy demand projection for water utilities


In addition to total extraction, the database allows to break down the data into water withdrawal by sector and by industry, which links the water sector with energy consumption in agriculture in form of irrigation.

According to the OECD, 70% of all water abstractions is used for agriculture (OECD, 2020a, p. 35). While freshwater extractions dominate total water extractions, desalinisation plants are an important parameter considering their high-energy consumption. However, water extraction through desalination plants only makes up 0.2% of the global water extraction (Table 6.16). Globally, about one third of all countries representing 80% of global population (OECD, 2020a) are connected to sewerage treatment plants. Table 6.17 shows the assumed global water withdrawal quantities—broken down by usage sector—which form the basis for the projection of the energy demand projection for water utilities.

# *6.4.2 Energy Effciency Standards and Energy Intensities of Water Utilities*

The following processes require the use of energy for water utilities:

	- Surface water pumping or
	- Groundwater pumping

In addition, the topography of a region and the climatic conditions—especially seasonal temperature differences and rainfall pattern—affect energy use in the water sector (Copeland & Carter, 2017). In dry regions such as California, 19% of the state's electricity consumption is used for pumping, treating, collecting, and discharging water and wastewater (ibid). The following provides a brief overview of the technical processes and their energy intensities.

*Water Extraction* To lift 1000 litre (1 m3 ) on metre requires 0.0027 kW/h—at 100% effciency (Rothausen & Conway, 2011). But, in practice, the value is higher and dependents on the quality and effciency of water pumps.

*Wastewater Collection* Wastewater is collected from domestic, commercial, or industrial use and processes. In general, the composition of wastewater by weight consists of 99.9% wastewater and 0.1% contaminants, including organic or inorganic matter, or microorganisms that need to be removed (ERC, 2019). Wastewater must be collected and transported; this process requires water pumps.

*Wastewater Treatment Plants (WWTP)* There are four types of wastewater treatment plants: (1) sewage treatment plants (STPs), (2) effuent treatment plants (ETPs), (3) activated sludge plants (ASPs), and (4) common and combined effuent treatment plants (CEPTs).

For water utilities, only sewage treatment plants (STPs) and activated sludge plants (ASPs), which are part of STPs, are important. Effuent treatment plants (ETPs) are typically used to clean industrial wastewater (ERC, 2019)—most of these are integrated into industrial parks for manufacturing and/or the chemical sector.

*Sewage Treatment Plant (STP)* The STP receives wastewater from domestic and commercial use and industrial processes. It also collects rainwater, storm water, and associated debris. The main processes include a basic fltering procedure to remove debris, dirt, grit, and sand:

• *Primary treatment*—settling: In the primary treatment, heavier and lighter organic solids are separated in a clarifcation tank which promotes sinking of heavier and foating of lighter solids, so they can be removed. This primary sludge is then moved into aeration basins, where the secondary treatment takes place.


The secondary treatment is the most energy-intensive process for wastewater treatment plants; aeration—the introduction of air into the biological tank—consumes about 60% of the plant's total energy. There are ways to improve energy effciency, e.g. by removing the aeration process through enhanced primary solid removal, based on advanced micro sieving and fltration processes (Oulebsir et al., 2020).

For the calculation and projection of energy demand for global water utilities, the energy intensities need to be simplifed, and average values are used. Table 6.18 shows the values used for the 1.5 °C OECM pathway for water utilities. It is assumed the water withdrawal quantities (Table 6.17) will have to be pumped and distributed and—after usage—go back into wastewater treatment. The average energy intensity is provided. A small share of the water withdrawal—around 0.2%—will come from desalination plants which have a relatively high energy intensity per cubic metre.

Sewage plants often have onside electricity generation from biological material collected during wastewater treatment. The 1.5 °C pathway assumes that 5% of all sewage plants will utilise this potential in 2020 and that the share increase by 1% annually to 35% in 2050.

Furthermore, water utilities have signifcant non-energy GHG emissions from sewers, biological wastewater treatment, and sludge—mainly CH4 and N2O. Table 6.18 shows the assumed values in CO2 equivalent per cubic metre with the conservative assumption that those specifc values will remain on 2020 level until 2050.

# *6.4.3 Projection of the Energy Demand and CO2 Emission for Water Utilities*

The projected global energy demand for water utilities was calculated with the documented assumed global quantities of required water and energy intensities (Table 6.19). Based on the required energy and the 1.5 °C energy supply scenario,


**Table 6.18** Assumed global energy intensities for process relevant for water utilities

#### **Table 6.19** Projected global energy demand for water utilities



**Table 6.20** Global energy-related CO2 emissions and non-energy GHG for water utilities under the 1.5 °C energy pathway

the global energy-related CO2 emissions have been estimated (Table 6.20). However, the main GHG emissions from water utilities do not originate from energy-related CO2, but from methane and N2O (nitrous oxide or 'laughing gas') which have a signifcant greenhouse potential (see Chap. 11).

# **6.5 Energy Demand Projection for the Four Analysed Service Sectors**

The combined energy demand for the analysed sectors represented 7.5% of the global demand in 2019. The results of the energy demand projection suggest that demand will continue to grow even with energy effciency measures as the volume of their produced commodities—especially food and water—will have to increase to meet the demands of a growing population by 165% by 2050. The two main drivers for the increased energy demand are agriculture and food processing and forestry and wood products. Due to electrifcation of machinery and (process) heat, the overall electricity demand increases signifcantly by 162% in 2050 in comparison to 2019. Especially the electricity demand for fsheries with the projected electrifcation of marine fshing vessel increases by factor 7 between 2019 and 2050 (Table 6.21).

# **6.6 The OECM 1.5 °C Pathways for Major Industries: Limitations and Further Research**

We have shown that the four analysed sectors can phase out their energy-related CO2 emissions (Table 6.22) with a combination of energy effciency and a shift to a renewable energy supply. Key technologies for the decarbonisations are the following:

*Agriculture and Forestry* Heavy-duty machinery for harvesting food products, such as crops, or timber is currently almost entirely based on fossil fuel-driven com-


**Table 6.21** Projected global energy demand for water utilities


**Table 6.22** Global energy-related CO2 emissions and non-energy GHG for water utilities under the 1.5 °C energy pathway

bustion engines. However, biofuels and—after 2030—electric vehicles are assumed to be available to reduce energy-related CO2 emissions to zero by 2050.

The management of forests, croplands, and pastures can lead to both emission and sequestration of CO2 and other GHGs. The need to feed a population of nine billion in 2050 will exert signifcant demands on the global agriculture and food systems. Advances in technology, particularly the increasing role of renewable energy in the agri-food sector, will help to reduce the energy emissions of the sector. However, given the crop intensifcation and agricultural expansion required to meet these food demands, it is expected that the agriculture sector will be unable to achieve zero emissions of non-energy GHGs by 2050. Improving soil management, reducing the yield gap, and initiating substantial shifts in dietary and nutritional patterns will help to reduce emissions. However, an increase of agricultural land at the expense of forests and/or their expansion in order to achieve negative emissions is likely if crop yield effciencies cannot be improved. Further research is required on the individual contributions of each of these pathways to the complete decarbonisation of the sector.

Nature-based approaches, particularly reforestation, also offer offset options. With an increasing focus on saving and regenerating forests, the forestry sector can become not only carbon-neutral but also carbon-negative, as early as 2030. The abolition of carbon emissions or the achievement of negative emissions between 2030 and 2050 will compensate for the unavoidable process emissions in other sectors, such as the cement and steel industries.

The authors found a lack of policy mechanisms to unlock the large potential for nature-based solutions to create carbon sinks, although the scientifc literature confrms the signifcant role of land-use emissions in climate mitigation pathways (IPCC 2021). More research is required into the compensation mechanisms for process emissions and their potential roles in the implementation of nature-based solutions (see also Chap. 11).

*Food Processing* Food processing, in particular, requires process heat, most of which was supplied by fossil fuel-based technologies in 2019. A signifcant increase in the electrifcation of process heat generation is assumed to occur. To achieve the overall CO2 emission targets, the electricity generation under the OECM pathway will increase the average global renewable electricity share from 25% in 2019 to 74% in 2030. Although the transition to renewables under the OECM 1.5 °C pathways that phase out energy-related Scopes 1 and 2 emissions is ambitious, the implementation of the assumed Scope 3 emission pathways is signifcantly more challenging.

*Wood and Paper Products* The wood processing and pulp and paper industry can use organic residuals and biomass as fuel for onside power and heat generation which is already a common practice especially in Scandinavia and Canada. An increase of those applications is assumed in the OECM.

*Water Utilities* Similar to the wood and paper industry, water utilities can use organic residuals and especially methane from sludge to fuel onside power and heat generation to supply their own demand. Those technologies are assumed to become mainstream in the OECM to reduce 'behind the meter' demand and to capture methane emissions which have a high global warming potential (GWP)—see Chap. 11.

*Fisheries* The transition to sustainable fsheries includes to move away from industrial fshing trawlers towards a more decentralised fshing feet. The electrifcation of marine artisanal vessels via electric outboard engines seems a promising way to reduce emissions from ineffcient diesel ship engines. However, the energy intensity for aquaculture farms is diverse, and a global average value in energy units per tonne of fsh is not available. The literate suggests that it is entirely dependent on the region and the fsh species. Thus, the calculated energy demand for the global fshery industry is fraught with very great uncertainties, and more research is needed.

We found that industry-specifc data for energy intensities, although available (especially for the food sector), are often incomparable because they are based on different assumptions and/or methodologies. Therefore, we recommend the standardisation of the calculation and reporting methodologies for industry-specifc energy intensities for the various technical processes. Furthermore, industry-specifc energy statistics, including those for the sub-sectors of industries classifed under the GICS system, would signifcantly enhance the level of detail available for setting net-zero targets in the future.

# **References**


Davis, K. F., D'Odorico, P., & Rulli, M. C. (2014). Moderating diets to feed the future. *Earth's Future, 2*(10), 559–565. https://doi.org/10.1002/2014EF000254


*scenarios with non-energy GHG pathways for +1.5°C and +2°C*. SpringerOpen. https://link. springer.com/chapter/10.1007/978-3-030-05843-2\_4


UNIDO. (2020). *INDSTAT 2 2020*. UNIDO Statistics Data Portal.https://stat.unido.org/ World Bank. (2021). *World development indicators*. http://wdi.worldbank.org/table/4.2#.

**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

# **Chapter 7 Decarbonisation Pathways for Buildings**

**Souran Chatterjee, Benedek Kiss, Diana Ürge-Vorsatz, and Sven Teske**

**Abstract** This section documents the development of four different energy demand pathways on the basis of the high-effciency buildings (HEB) model of the Central European University. The assumptions and the scenario narratives are derived and the results provided in numerous graphs and tables. Of the four derived scenarios, two are selected for the OECM and the selection criteria are justifed. The results in terms of the global energy demand and energy-related CO2 emissions are provided in tables.

**Keywords** Decarbonisation pathways · Buildings · Residential · Commercial · High-effciency buildings (HEB) model · Energy intensities · Floor area · Bottom-up demand projections

The developments of the regional and global energy demand for the building sector are described in this chapter. Sections 7.1 and 7.2 document the development of the bottom-up energy demand projections for buildings with the methodology described in Sect. 3.2 and are authored by Prof. Dr. Diana Ürge-Vorsatz, Dr. Souran Chatterjee, and Benigna Boza-Kiss of the Central European University Budapest, Hungary. The last section describes the implementation of this research in the wider OneEarth Climate Model (OECM) to generate a single 1.5 °C energy pathway for buildings and construction.

S. Chatterjee · B. Kiss · D. Ürge-Vorsatz

Department of Environmental Sciences and Policy, Central European University, Budapest, Hungary e-mail: ChatterjeeS@ceu.edu; KissB@ceu.edu; Vorsatzd@ceu.edu

S. Teske (\*)

Institute for Sustainable Futures, University of Technology Sydney, Sydney, NSW, Australia e-mail: sven.teske@uts.edu.au

# **7.1 Buildings**

The building sector is responsible for 39% of process-related greenhouse gas (GHG) emissions globally and accounts for almost 32% of the global fnal energy demand, making the building sector pivotal in reducing the global energy demand and climate change (Ürge-Vorsatz et al., 2015a, 2020). The building sector is often suggested to have the largest low-cost climate change mitigation potential, achievable by reducing the energy demand (Ürge-Vorsatz & Tirado Herrero, 2012a; Güneralp et al., 2017a). However, with the increasing rates of population growth and urbanisation, the building stock is projected to double in developing regions by 2050, so reducing the global energy demand will become challenging (EIA, 2015a). Along with these challenges, new building stocks in developing regions will simultaneously provide opportunities for energy-effcient construction, which could substantially reduce the global energy demand. In developed regions, opportunities to reduce the energy demand will predominantly involve renovating the existing building stock (Prieto et al., 2019a; Chatterjee & Ürge-Vorsatz, 2020).

The IPCC's ffth assessment report makes clear that the energy demand must be reduced substantially by 2050 to limit the global temperature rise to 1.5 °C (Rogelj et al., 2018a). However, today, most mitigation pathways still rely on supply-side solutions, and little effort has been made to understand the demand-side potential (Creutzig et al., 2018a). More precisely, understanding the global energy demand for the building sector by assessing the future growth in foor area and the corresponding energy demand is crucial in the context of the 1.5 °C target. Therefore, different models of the building energy demand are used to understand the future energy consumption and emission potential of the building sector under different policy scenarios.

# **7.2 The High-Effciency Buildings (HEB) Model: Energy Demand Projections for the Building Sector**

To develop detailed energy demand projections for the regional and global building sectors, the *high-effciency buildings* (HEB) *model* was used. The HEB methodology is documented in Sect. 3.2 and is among the most detailed models for this sector. The key output of the HEB model consists of foor area projections for different types of residential and tertiary buildings in different regions and countries, the total energy consumption of residential and tertiary buildings, the energy consumption for heating and cooling, the energy consumption for hot water energy, the total CO2 emissions, the CO2 emissions for heating and cooling, and the CO2 emissions for hot water energy. The HEB is based on a bottom-up approach, and it includes rather detailed technological information for one sector of the economy. However, it also uses certain macroeconomic and socio-demographic data, including population growth rates, urbanisation rates, and foor areas per capita. The HEB model uses four different scenarios to understand the dynamics of energy use and to explore the potential of the building sector to mitigate climate change by exploiting various opportunities. The four scenarios are:


The aim of the scenario analysis is to determine the importance of different policies for building energy-effciency measures and to show how much the fnal energy consumption of the building sector can be reduced across the world. Table 7.1 summarises the actual parameters of the four scenarios.

# *7.2.1 Regional Breakdown of the High-Effciency Buildings (HEB) Model*

The end-use demand and its corresponding emissions are produced until 2060 at yearly resolution for 11 key regions, which include 28 member states of the European Union and 3 key countries (India, China, and the USA), and cover the world. Those 11 regions shown in Fig. 7.1 differ from the 10 IEA regions used for the regional transport demand analysis. The main differences are as follows: OECD Europe (IEA) is broken down into Western and Eastern Europe (HEB); Africa (IEA) and the Middle East (IEA) are grouped into Middle East and Northern Africa (HEB)


**Table 7.1** Parameters of the four scenarios

**Fig. 7.1** Global coverage of HEB model

and sub-Saharan Africa (HEB); India (IEA) is part of the South Asia (HEB) region, which includes the neighbouring countries Bangladesh, Bhutan, Sri Lanka, Nepal, and Pakistan—countries that are part of the IEA region non-OECD Asia; China (IEA) is part of the group Centrally Planned Asia, which includes Cambodia, Lao, Mongolia, North Korea, and Vietnam, all of which are part of the IEA region non-OECD Asia; Pacifc Asia (HEB) is the remaining part of the non-OECD Asia (IEA) region and all Pacifc Island states.

# *7.2.2 HEB: Data and Assumptions*

Similar to any bottom-up energy demand model, the HEB model is very dataintensive. Therefore, it relies on a broad variety of input sources, including statistical databases and the scientifc peer-reviewed and grey literature, to incorporate the most up-to-date data. The HEB model largely depends on four sources for its basic input data:

*World Bank Databases* Both present and historical data on population and real gross domestic product (GDP) fgures are obtained from the World Bank databases. The GDP forecast data play a particularly crucial role because they determine the growth in foor area of non-residential buildings. The HEB model calculates future GDP values based on historic and present GDP growth rate data obtained from the World Bank database. The future real GDP is predominantly calculated for non-OECD countries for which future forecasts of real GDP are not available. However, for the OECD member states, this model uses the OECD database of real GDP projections. In addition to the forecast GDP and real GDP databases, the HEB model uses the population forecast database of the World Bank to calculate the future population growth for different countries and regions.

*United Nations Development Programme (UNDP), UN-Habitat, and United Nations Conference on Trade and Development (UNCTAD) Population Databases* To calculate the growth in foor area and therefore the fnal energy consumption for heating and cooling, population projection data are required. Together with the World Bank database, the HEB model uses the UNDP population projection database to calculate future populations. Furthermore, because the HEB uses rural and urban classifcations, urbanisation rate data are obtained from the UNCTAD database. However, none of these databases contains data on slums or the informal settlement of different regions. Therefore, urban populations living in slums are calculated based on UN-Habitat projections.

In addition to population and GDP data, other important data points used in HEB, such as building stock data and energy intensity data, have been collected from several project reports and datasets of the European Commission, as well as the Eurostat database in the case of the EU, the US Energy Information Administration (EIA) database, and various literature sources. Further information on the data collection can be found in the previous report of HEB in Urge-Vorsatz et al. (2012a). In some cases, data for some of the parameters are unavailable, and in those cases, the HEB model relies on expert judgement. For instance, the energy intensity (specifc energy consumption) of advanced buildings mainly utilises the 'passive house' principle, meaning that the useful energy demand may not exceed 15 kWh/m2 /year for heating. This concept has been shown to be applicable throughout the world, and various other measures are used to reduce the cooling and dehumidifcation demands. The total useful demand can be supplied by increasingly effcient heat pumps, which results in very low fnal energy demands in such advanced buildings. In the *nearly net-zero* scenario, the energy consumption of advanced buildings is even more reduced by potential local energy production, which is calculated with the Better Integration for Sustainable Energy (BISE) model at the building level. The basic input data used in the HEB model are presented, together with their sources, in Table 7.2.

The key assumptions of the model are presented in Table 7.1, and the sources of the key input data are documented in Table 7.2. Assumptions, such as the retroftting rate, the share of advanced buildings within the new and retroftted stock, and the energy performance of buildings of different vintages, are based on expert judgements and the authors' experience in the feld of modelling building energy. Because data for these parameters are not available, the authors have made several assumptions related to their magnitudes (Table 7.1). Moreover, because the HEB model provides a realistic evaluation of the building energy demands under different policy scenarios, different scenario-specifc assumptions are also used to defne the scenarios.

The fndings are presented in Sects. 7.2.3 and 7.2.4. First, the fndings of the study show the future foor area projections under different scenarios, and then it presents future space-heating and space-cooling demand of the different regions. Space-heating and space-cooling demand largely depends on the foor area growth,


**Table 7.2** Key input data used in the HEB model and their sources

and hence, the results of foor area are presented frst. To calculate foor area and fnal energy demand, HEB model frst calculates region-specifc population and GDP with the help of Eqs. 3.1–3.9 (Sect. 3.2). Based on the region-specifc populations and GDP growth rates, then region-specifc foor area and fnal energy consumption for space heating and cooling are calculated.

# *7.2.3 Floor Area*

The foor area for each of the regions is calculated with Eqs. 3.10 and 3.11 (Sect. 3.2). In accordance with the HEB modelling assumptions, the growth in foor area in the residential sector depends predominantly on the population growth, whereas the growth in non-residential or commercial foor area depends on the GDP growth of the region. Based on these equations and assumptions, the fndings of the HEB model show that the global foor area will increase by 77% from 2022 to 2060 and the global foor area growth will be dominated by the growth in the Asian, Middle Eastern, and African regions. Precisely, substantial growth in foor area will be observed in the Middle East and Africa (180%), followed by Pacifc Asia (174%), Africa (131%), and Latin America (130%) (refer to Fig. 7.2).

Signifcant population and GDP growth is projected for regions such as the Middle East and Africa, Africa, and Pacifc Asia in the future, so the foor area growth in these regions will be substantial. If the global growth in foor area is further analysed according to different building categories and classifcations, it can be seen that the substantial increase in foor area will be dominated by urban foor area (99% growth is projected by 2060 relative to 2022), which will mainly be caused by an increasing rate of urbanisation. As a result of the increasing urbanisation rate, urban slums are projected to increase signifcantly to 176% by 2060. However, the foor area of slums constitutes only a small proportion of the global foor area (2.4%

**Fig. 7.2** Growth of the total foor area and its distribution among the regions of the world

**Fig. 7.3** Total building foor area in the world by building classifcation

**Fig. 7.4** Distribution of the total foor area throughout the world by building vintage across the modelling period

of the global foor area, which is projected to increase to 3.7%), so the growth of slum areas will have little impact on the global foor area growth. Moreover, if foor area growth is analysed per building classifcation, substantial growth can be projected for both residential and commercial buildings. More precisely, the global residential building sector is projected to grow from 186 billion m2 in 2022 to 292 billion m2 by 2060, and the global commercial building sector is projected to grow from 102 billion m2 in 2022 to 217 billion m2 by 2060 (refer to Fig. 7.3).

The fndings of the HEB model are summarised in Figs. 7.4 and 7.5. However, it is important to understand the future proportions of buildings of different vintages, because they have different levels of energy performance and therefore different energy consumption patterns. The foor area growth for buildings of different

**Fig. 7.5** Distribution of the total foor area in China by building vintages across the modelling period

**Fig. 7.6** Distribution of the total foor area in the USA by building vintages across the modelling period

vintages is presented in Figs. 7.4, 7.5, 7.6, 7.7, 7.8 and 7.9, which shows the share of each vintage and its change over the modelling period across the different scenarios in each of the regions. It is important to note that the total foor area remains the same in all scenarios.

The fndings show that the growth in total foor area is mainly dominated by growth in China and India. More precisely, China's share of the global total foor area in 2022 will be around 28%, and by 2060, it will increase by 54%, whereas India's share in 2022 will be 14% and will increase by 96% by 2060. Furthermore, signifcant growth in foor area can be observed by 2060 in key regions, such as the USA (41%), Pacifc OECD (25%), and EU-28 (22%).

The results of the HEB model also show that a very small amount of today's building stock will remain as it is until 2060. Therefore, to reduce the energy demand and the impact of the energy demand of the building sector on climate change, it

**Fig. 7.7** Distribution of the total foor area in India by building vintages across the modelling period

**Fig. 7.8** Distribution of the total foor area in EU-27 countries by building vintages across the modelling period

will be crucial to implement advanced effciency measures for retroftted and new buildings that will be constructed in 2022–2060. If today's best practices of energy effciency are applied to all new and retroftted buildings globally (*deep effciency* scenario), 43% of the building stock will be classifable as 'advanced new' buildings and 41% of the building stock as 'advanced retroftted' buildings in 2060. However, a signifcant amount of stock will remain less energy-effcient based on the assumption that the construction market cannot adjust immediately to the new practices required to build highly effcient buildings. On the contrary, if the current practice is '*frozen*' and no advanced measures are introduced, 99% of the stock will remain less effcient while having the rest unchanged in 2022 values. It is noteworthy that according to the fndings of the HEB model, 66% of the building stock in 2060 does not yet exist in 2022. The *moderate effciency* scenario assumes that only

**Fig. 7.9** Distribution of the total foor area in Pacifc OECD countries by building vintages across the modelling period

present policies will be enforced and there will no further more ambitious goals set throughout the world. Under this scenario, only a minor share (7%) of the foor area will be classifable as 'advanced' (2% new and 5% retroftted). This is because most countries with strong policies for energy-effcient buildings (especially the EU) will only play a minor role in constructing a share of new buildings around the globe.

# *7.2.4 Final Energy Use for Space Heating and Cooling Under the HEB Scenarios*

The fnal energy use for space heating and cooling will largely depend upon the calculated foor areas. After the foor area is calculated for each region, the thermal energy use is calculated. Like the foor area calculations, thermal energy use is also calculated for the four different scenarios.

Among the four scenarios, the fnal energy use for space heating and cooling under two scenarios clearly shows immense potential for reducing the energy demand of the building sector by 2060. At the global level, if best practices in building construction and retroftting become standard, the fnal energy for heating and cooling will decrease from 24 PWh in 2022 to 10 PWh in 2060, which corresponds to a 56% drop, as shown in Table 7.3. However, if existing policies continue in place until 2060, the fnal energy use will increase by 34% by 2060 relative to the 2022 level. In other words, under the *moderate effciency* scenario, the global fnal energy required for space heating and cooling will increase by 34% by 2060 relative to that in 2022. Under the *deep effciency* scenario, the global fnal energy demand in 2060 will be 67% less than under the *moderate effciency* scenario, whereas under the *frozen effciency* scenario, it will be 37% higher, which corresponds to an 83% increase relative to the 2022 level.



There are two key reasons behind the signifcant energy savings in the *deep effciency* and *nearly net-zero* scenario compared with the *frozen effciency* and *moderate effciency* scenarios:


More precisely, in the *deep effciency* and *nearly net-zero* scenarios, the retroftting rate is assumed to be 3% in developed countries and 1.5–1.6% in developing countries after 2027. The same retroftting rates are assumed in the *moderate effciency* scenario. However, in the *frozen effciency* scenario, the retroftting rate is assumed to be no higher than 1.4% across all regions. Similar to the retroftting rate, under the *deep effciency* scenario, it is assumed that all new and retroftted buildings will have a very low energy demand in the EU, NAM, and PAO after 2030 and in the other parts of the world after 2037. Under the *nearly net-zero* scenario, it is assumed that all new and retroftted buildings will have a net-zero energy demand in EU, NAM, and PAO after 2030 and in other parts of the world after 2037 because the local onsite solar electric production is included in the defnition of the *nearly net-zero* scenario. In the *moderate effciency* scenario, advanced buildings are only introduced in Western Europe after 2035 for all new buildings, and after 2045, all retroftted buildings will have a low-energy design. Based on these assumptions, the fndings of HEB highlight the importance of ambitious in-act policies.

Key regions, such as China, EU-27, and India, consume most of the global energy, so it is important to know how the building sectors in these regions will perform under different scenarios. Regions such as the USA and EU-27 have much greater potential to reduce space-heating- and space-cooling-related energy use with the help of best practices. Precisely, 73% and 75% of energy consumption related to thermal comfort can be reduced by 2060 in the USA and EU-27, respectively, if best practices are followed. The *nearly net-zero* scenario goes one step further than the *deep effciency* scenario. The results show that the energy consumption of buildings for heating and cooling can reach almost zero in the EU, the USA, and Pacifc OECD countries by 2055–2057. Although heating- and cooling-related energy consumption in China and India will not reach zero in the modelled period, signifcant reductions in China and India (85% and 27%, respectively) can be achieved relative to 2022 values. Figures 7.10 and 7.11 show the fnal energy demands for space heating and cooling in different parts of the world under the different scenarios (Fig. 7.12).

Globally, commercial and public buildings in urban areas are the largest consumers of space-heating- and space-cooling-related energy. Therefore, best practices should especially focus on commercial and public buildings in urban areas. Commercial and public buildings in urban areas will reduce their consumption by up to 33% by 2060 under the *deep effciency* scenario. Similarly, urban residential buildings will reduce their consumption by up to 57% globally by 2060 under the *deep effciency* scenario. Under the *nearly net-zero* scenario, commercial and public buildings still have a signifcant share of energy consumption in 2060, but the total energy demand is extremely reduced. It is noteworthy that reducing the energy intensity of commercial and public buildings even further will require further

**Fig. 7.10** Final energy consumption for space heating and cooling in the world and key regions (in PWh)

investigation of the usage characteristics of different building types. Therefore, even more effort will be required than merely servicing these building with renewable energy. Similar fndings are obtained from the analysis of the region-specifc fnal energy demands for the USA, the European Community, India, China, and the OECD Pacifc.

**Fig. 7.11** Shares of the total heating and cooling energy consumption attributable to different regions of the world

Commercial buildings are the largest consumer of space-heating- and spacecooling-related fnal energy in low- to middle-income regions, such as India and China. However, in developed regions, such as the Pacifc OECD, EU-28, and the USA, the residential building sector is the largest consumer. The HEB results show that these developed high-income regions can substantially reduce their energy demands in both residential and commercial building sectors if advanced higheffciency energy measures are standardised over the years. In fact, in these regions, if local energy production is included (i.e. *nearly net-zero* scenario), then the building sector can achieve a net-zero status by 2060. In contrast, the low- to middleincome regions will not be able to achieve a net-zero status by 2060, even if the local production of solar electric energy is added into the calculation. However, regardless of the local energy production, these regions can still achieve a substantial reduction in China, and in India, the rate of increase will be slowed by the introduction of advanced effciency energy measures, such as new energy-effcient

**Fig. 7.12** Shares of fnal energy consumption for space heating and cooling for the world by building category (in PWh)

building codes and the rigorous renovation of existing buildings. More precisely, in India, even with advanced energy-effciency measures, the fnal energy demand for heating and cooling will increase by 12% in 2060 relative to that in 2022, which is 65% lower than the fnal energy demand for 2060 if the existing effciency measures are followed until 2060.

# *7.2.5 Key Findings for the HEB Scenarios*

The HEB model analysis demonstrates the potential for reducing the energy demand in the building sector if state-of-the-art high-effciency buildings are implemented worldwide. The fndings of the study show that with a higher share of high-effciency renovations and construction (as assumed in the *deep effciency* and *nearly net-zero* scenarios), it will be possible to reduce the fnal thermal energy used globally in the building sector by more than half by 2060. In some regions, such as the EU and Pacifc OECD, it will even be possible to achieve net-zero status for the thermal energy demand. However, this pathway towards high-effciency or net-zero emissions is ambitious in its assumptions and requires strong policy support. On the contrary, if policy support to implement more high-effciency buildings is not in place (*frozen effciency* scenario) or even if the present policy scenarios are continued (*moderate effciency* scenario), the total thermal energy demand of the building sector could increase by 34–83% by 2060 relative to the 2022 level. Furthermore, if the present rate of energy-effciency measures is continued, 67–80% of the global fnal thermal energy savings will be locked in by 2060 in the world building infrastructure. The lock-in effect of the building sector also indicates that if the present moderate energy performance levels become the standard in new and/or retroftted buildings, it will be almost impossible to further reduce the thermal energy consumption in such buildings for many decades to come.

# **7.3 1.5 °C OECM Pathway for Buildings**

Based on the results of the detailed HEB model analysis, the *deep effciency* scenario was chosen for commercial buildings and the *moderate effciency* scenario for residential buildings. These scenarios were chosen after stakeholder consultation with representatives of the respective industries, members of the Carbon Risk Real Estate Monitor (CRREM), the Net-Zero Asset Owner Alliance, and academia. To integrate the building sector into the 1.5 °C pathway as part of the OECM, consistent with all other industry and service sectors and the transport sector, the selection of one specifc pathway for the building sector as a whole was necessary. The energy demand for the construction sector was also required to calculate the emissions for the Global Industry Classifcation Standard (GICS) (see Chap. 2). This section documents the calculation process and the results for the residential and commercial building sector and construction.

Table 7.4 shows the assumed development of foor space for residential and commercial buildings, which was taken from the HEB analysis and the projected economic development of the construction sector. The increase in the construction industry is based on the overall global GDP, developed as documented in Chap. 2, and is therefore not directly related to the HEB foor space projections. The direct link between both parameters was beyond the scope of this analysis and is therefore highlighted as a potential source of error.

The global energy intensities for residential and commercial buildings (in kilowatt-hours per square metre (kWh/m2 )) are the second main input for the OECM 1.5 °C building pathway and are taken from the documented HEB analysis. The global values were calculated on the basis of the total HEB results for the global energy demand per year divided by the foor space. The global values are the sum of the values for all 11 regions analysed with HEB. Table 7.5 also shows the reductions


**Table 7.4** OECM—global buildings: projected foor space and economic value of construction

**Table 7.5** OECM—global buildings: assumed energy intensities


in the energy intensity for residential and commercial buildings relative to the values in the base year 2019.

The energy intensity of the construction industry was calculated with the total energy demand (in petajoules (PJ)) in 2019, as provided in the IEA World Energy Balances 2019 for Construction and the projected economic values (in \$US) for the same year. The energy demand value for construction in the IEA statistics includes the construction of roads and railways, as well as other civil engineering and utility projects, as defned in IEA (2020). Therefore, the shares of the energy demand for residential and commercial buildings must be estimated. The calculated energy intensity for construction work was compared with published values.

Based on the assumptions and input parameters documented in Tables 7.4 and 7.5, the energy demand for all sub-sectors was calculated. Table 7.6 shows the calculated annual energy demand for residential and commercial buildings and for the construction industry. The energy demand consists of the energy required for space heating and cooling ('heating energy') and the electricity demand, which includes all electrical applications in the buildings but excludes electricity for heating and cooling. This separation is necessary to harmonise the input data from the HEB, which do not include electricity for household applications such as washing machines, etc., with the OECM.

The electricity demand for residential buildings is based on the bottom-up analysis of households documented in Sect. 3.1.2. The electricity demand for the service sector is based on a breakdown of electricity and heating in 2019 across all service


**Table 7.6** OECM—global buildings: calculated annual energy demand for residential and commercial buildings and construction


**Table 7.7** OECM−global buildings: energy supply

sectors, published in the IEA World Energy Balances. The future values until 2050 are based on the projections for the analysed service and industry sectors documented in Chaps. 5 and 6.

The supply side for the building and construction sectors is based on the 1.5 °C pathway for energy utilities, as documented in Chap. 12 . In contrast to the demand side, the supply values for electricity are provided both for room climatisation (heating and cooling) and for appliances (Table 7.7). The total energy-related CO2 emissions were calculated based on the energy supply mix for heating and electricity generation (Table 7.8).


**Table 7.8** OECM—global buildings: energy-related CO2 emissions

(continued)


#### **Table 7.8** (continued)

The specifc energy-related CO2 emissions are also provided for power and heat generation, as well as per square meter of foor area, for residential and commercial buildings. The specifc energy demand and the CO2 emissions per square meter are key performance indicators for the fnance industry for real estate. Moreover, these parameters are used for regulatory frameworks, such as the EU energy performance for building directive (EU, 2010).

**Acknowledgements** The authors of Sect. 7.2 (Souran Chatterjee, Benedek Kiss, Diana Ürge-Vorsatz) are immensely grateful to the DBH Group for the administrative support provided during the HEB research. This research was funded by the project Sustainable Energy Transitions Laboratory (SENTINEL), which received funding from the European Union's Horizon 2020 research and innovation programme under grant agreement no. 837089. This research was also partly funded by the Energy Demand changes Induced by Technological and Social innovations (EDITS) project, which is part of the initiative co-ordinated by the Research Institute of Innovative Technology for the Earth (RITE) and the International Institute for Applied Systems Analysis (IIASA) (funded by the Ministry of Economy, Trade, and Industry [METI], Japan).

# **References**


# *HEB Scenarios*


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

# **Chapter 8 Decarbonisation Pathways for Transport**

**Sven Teske and Sarah Niklas**

**Abstract** An overview of the main drivers of the transport energy demand and the assumed socio-economic development (population and GDP) until 2050 for ten world regions are given. The countries in each world region are tabulated. Detailed documentation of projected shifts in transport modes for all world regions, including technological assumptions and energy intensities, by vehicle type is presented. This section contains the OECM 1.5 °C transport scenarios for aviation, shipping, road, and rail, each broken down into passenger and freight transport. The calculated energy demands and energy-related carbon emissions for all transport modes are provided.

**Keywords** Global and regional transport demand · Mode shift · Transformative Urban Mobility Initiative (TUMI) · GHG development · GDP · Population · Energy intensities

# **8.1 Introduction**

The transport sector consumed 28% of the fnal global energy demand in 2019, and its decarbonisation potential is therefore among the most important of all industries. Given its size and diversity, not only with regard to different transport modes and technologies but also regional differences, it is also one of the most challenging sectors. In 2019, transport consumed 65% of the total oil demand globally. Therefore, the transition from oil to electric drives and to synthetic fuels and biofuels is key to achieving the goals of the Paris Climate Agreement. A rapid uptake of electric mobility, combined with a renewable power supply, is the single most important measure to be taken to remain within the carbon budget of the 1.5 °C pathway.

S. Teske (\*) · S. Niklas

Institute for Sustainable Futures, University of Technology Sydney, Sydney, NSW, Australia e-mail: sven.teske@uts.edu.au; Sarah.Niklas@uts.edu.au

S. Teske (ed.), *Achieving the Paris Climate Agreement Goals*, https://doi.org/10.1007/978-3-030-99177-7\_8

The fnancial sector *Transport* spans civil aviation, shipping, and road transport, including passenger and freight transport, and all related services. For each transport mode, there are two main sub-sectors:


This section is based on multiple closely linked research projects: the One Earth Climate Model (OECM) developed in 2019 (Teske et al., 2019) and 2021 and the TUMI Transport Outlook 1.5 °C (Teske et al., 2021), which was developed within a multi-stakeholder dialogue, including two workshops organised by Deutsche Gesellschaft für Internationale Zusammenarbeit GmbH (GIZ) and the University of Technology Sydney/Institute for Sustainable Futures (UTS/ISF) in June and September 2021. As a result, the OECM methodology described in Chap. 3 has been expanded to achieve higher levels of accuracy and resolution, in both the area of the transport demand projections and the calculation of the regional and global transport energy demands.

The demand projections are based on a bottom-up approach. The actual basis of the passenger transport demand is diverse (e.g. to get groceries, to commute for work, or for leisure and recreation), and the transport demand is expressed in kilometres per person per year. Therefore, the development of this transport demand is dependent upon a number of different factors, among the most important of which are the actual population development and economic situation of a region. Geography and lifestyle also play important roles.

In considering the transport of goods, it is important where the goods are produced, the resources required, and where they are located. Economies with high local production rates have lower transport demands than those with high import/ export dependence. However, calculation of the actual transport demand is based on non-energy-related factors. A transport or travel demand does not necessarily lead to an energy demand if a non-energy transport mode, such as walking or cycling, is used—suffcient to satisfy the demand. However, most transport modes require energy, and the amount of energy per kilometre depends upon the energy intensity of the chosen vehicle.

The demand for transport energy does not inevitably lead to CO2 emissions if the energy is generated from renewable electricity and/or renewable fuels. Therefore, a carbon-neutral global transport sector is possible, while regional and intercontinental travel and global trade are maintained.

The transport demand is dependent upon a huge number of factors—the most important of which are the population size and the economic situation. In general, more people and a higher economic standard entail a higher transport demand. The transport service structure—and therefore the transport mode—also depends on a variety of factors. The actual distance travelled, the travel time required, the availability of certain transport modes, and the costs, among other factors, defne the chosen transport mode. Each transport mode includes a variety of vehicles with

different energy intensities. The transport mode 'road', for example, has by far the largest number of different vehicle options: buses, a huge variety of car types with different drive trains, motorcycles, bicycles, and even walking.

A global scenario requires the simplifcation of the transport demand projections. A detailed analysis of the purpose of each of those transport demands in kilometres per day for the entire population is not possible. Therefore, the methodology focuses on the development of regional person–kilometres (pkm) and tonne–kilometres (tkm) per year. The main factors affecting demand changes are population and economic development.

Whereas the industry and service pathways (Chaps. 5 and 6) were developed with accumulated global gross domestic product (GDP) values and bottom-up product-based projections, such as the annual steel production (in million tonnes per year), the demand projections for the buildings and transport sectors have been developed on the basis of specifc data from ten world regions, to capture the signifcant regional differences. The geographic breakdown is based on IEA's ten world regions used in the World Energy Outlook series (see Table 8.1).

# **8.2 Socio-economic Assumptions**

The assumed development of regional populations is based on the projections of the United Nations Department of Economic and Social Affairs, whereas the regional GDP developments are based on World Bank projections. The global values for population and GDP are identical throughout the entire analysis, across all sectors. The regional values are used for the buildings and transport sectors, whereas for all other sectors, the resulting (summed) global values are used (Table 8.2).

# **8.3 Transport Demand**

# *8.3.1 Global and Regional Transport Demands*

The global pandemic began in early 2020 and led to signifcant travel restrictions across the world. At the time of writing (December 2021), travel restrictions in many countries are still in place.

The global oil demand accounted for 11.5 Gt of energy-related CO2 in 2019 (IEA, 2020a). The transport sector consumes 65% of total oil demand, which included oil for international bunkers (10.4% of the total oil demand). Road transport consumed more than 40% of the total oil demand in 2019. The sector's growth has been responsible for over half the growth in the total oil demand since 2000 (BloombergNEF, 2020). As a result of the restricted mobility imposed to stop spread of the COVID-19 virus, the global pandemic led to a signifcant reduction in the oil


**Table 8.1** World regions used for the 1.5 °C OECM transport scenario

demand, especially for road transport and aviation, which are responsible for nearly 60% of oil use (IEA, 2020a). The global oil demand is estimated to have dropped by 8% in 2020. At the time of writing, the global pandemic is still ongoing, although travel restrictions have been relaxed in many countries, increasing in the transport demand relative to that in 2020. In our transport demand projections, we assume that the demand will continue to increase to pre-pandemic levels by 2025.


**Table 8.2** Assumed population and GDP developments by region in 2020–2050

The pandemic had a dramatic impact on public transport. Fear of being infected with COVID-19 led many people to avoid using public transport and to switch to other transport modes—especially individual transport, such as private cars or (electric) bicycles. *The Future of Public Transport (*C40 *Cities Climate Leadership Group and International Transport Workers' Federation* 2021*)*, published in March 2021, reported that as 'public transport ridership has fallen during the COVID-19 pandemic, so has revenue. Public transport agencies across cities worldwide face a critical funding shortfall that threatens jobs and services'.

The energy demand is likely to increase and there is currently no sign that these increases will slow in the near future. The increasing demand for energy for transport has mainly been met by greenhouse gas (GHG)-emitting fossil fuels. Although (battery) electric mobility has recently surged considerably, it has done so from a very low base, which is why, in terms of total numbers, electricity still plays a relatively minor role as an energy carrier in the transport sector.

Apart from their impact on climate, increasing transport levels—especially by car, truck, and aeroplane—also have unwanted side effects: accidents, traffc jams, noise and other pollutants, visual pollution, and the disruption of landscapes by the large-scale build-up of the transport infrastructure. However, road, rail, sea, and air transport are also integral parts of our globalised and interconnected world and guarantee both prosperity and intercultural exchange. Therefore, if we are to cater to people's desire for mobility while keeping the economy running and meeting the Paris climate goals, fundamental technical, operational, and behavioural measures are immediately required.

In this analysis, we discuss potential pathways of transport activity and technological developments by which we can meet the requirement that warming does not exceed pre-industrial levels by more than 1.5 °C—while at the same time maintaining a reasonable standard of mobility. The scenarios in this analysis are based on global and regional scenarios developed by the German Aerospace Centre (DLR), published in February 2019 (Pagenkopf et al., 2019), which have been updated in more detail as part of the Transformative Urban Mobility Initiative (TUMI) research (Teske et al., 2021).

We structured our scenario designs around the following key energy- and emission-reducing measures:


The fnal global energy demand in the transport sector1 totalled 103 EJ in 2019, according to the IEA Energy Balances (IEA, 2020b). Based on this estimate, the freight and passenger transport demands were estimated from statistical data and energy-effciency fgures.

Figure 8.1 shows that road passenger transport had the largest share of the fnal transport energy (53%) in 2019. Most of this consisted of individual road passenger modes (mostly cars, but also two- and three-wheel vehicles), which accounted for around 40% of all end energy in the transport sector. In total, road transport (passenger and freight) accounted for around 76% of the total fnal energy demand for transport.

The majority of all passenger transport—in terms of overall kilometres—is by road. However, international freight transport is more strongly dominated by rail and shipping, which account for 45% of all tonne–kilometres. The high effciency of rail and shipping means that their share of the global transport energy demand is small relative to the share of global tonnage transported.

Figure 8.2 shows the passenger (pkm) and freight transport (tkm) by transport mode in 2019 (OECD, 2021). Road transport clearly dominates. However, international freight often arrives by ship and is further transported by rail and/or road. OECD America and OECD Europe together make up half the total global energy demand, as shown in Fig. 8.3. China is at nearly the same level as OECD Europe, although it has about twice as many inhabitants as OECD Europe.

<sup>1</sup> International aviation and navigation bunkers are not included in this fgure.

**Fig. 8.1** Global fnal energy use, by transport mode, in 2019 (without international aviation or navigation bunker fuels)

**Fig. 8.2** Transport mode performances for road, rail, and aviation

**Fig. 8.3** Final energy use by global transport in 2019, according to region

# *8.3.2 Global Transport Technologies*

The energy intensities for different vehicle types and for each of the available drive trains play an important role in the fnal energy demand. Each transport mode has various different vehicular options, and each of the available vehicles has different drive train and effciency options. The technical variety of passenger vehicles, for example, is extremely large. The engine sizes for fve-seater cars range from around 20 kW to over 200 kW. Moreover, drive trains can use a range of fuels, from gasoline, diesel, and bio-diesel to hydrogen and electricity. Each vehicle has different energy intensities in MJ/pkm.

Figure 8.4 shows the powertrain shares of all transport modes in 2019 (in pkm or tkm) (IEA, 2020b). With a few exceptions, most modes were still heavily dependent on conventional internal combustion engines (ICEs). A small number of buses had electric powertrains (mainly trolley buses) and battery-powered electric buses also increased, predominantly in China. China also has a particularly large number of electric two- and three-wheel vehicles. Almost all battery-powered electric scooters were in China. Passenger rail was electrifed to a large extent (e.g. metropolitan and high-speed trains), whereas freight trains were predominantly not electrifed.

**Fig. 8.4** Powertrain split for all transport modes in 2019, by transport performance (pkm or tkm)

# **8.4 Aviation**

The 2020 pandemic led to signifcant travel restrictions and signifcantly affected the energy demands of global and domestic aviation (IEA, 2020c). The International Air Transport Association expects fight capacity utilisation to be, on average, 65% below the 2019 level in the second quarter (Q2) of 2020, 40% below in Q3 2020, and 10% below in Q4 2020 (Pearce, 2020). Data show that the global fight numbers were down by 70% at the start of April 2020 relative to those in the previous year. The consumption of kerosene in the whole of 2020 was expected fall by 26% (IEA, 2020c).

# *8.4.1 Energy Intensity and Emission Factors: Aviation*

The energy intensity for aviation freight transport was assumed to be around 30 MJ/ tkm in 2019 (Pagenkopf et al., 2019), decreasing by 1% per year until 2025. By 2050, the energy intensity for freight planes is estimated to be 25 MJ/tkm, 17% below today's value. The energy intensity for aviation passenger transport will decrease from 5.8 to 4.2 MJ/pkm between 2020 and 2050. Technical improvements in the aerodynamics, materials, weight, and turbine effciency for both freight and passenger planes are assumed. The volume of freight (in tkm) and the passenger– kilometres (pkm) are assumed to decrease by 30% globally between 2019 and 2050, an average reduction of around 1% per year.

The emissions factor for kerosene is calculated to be 73.3 g of CO2 per MJ (gCO2/MJ) (Jurich, 2016). The specifc CO2 emissions for aviation freight will decrease from 2.3 to 2.0 kgCO2/tkm in 2025. By 2035, the specifc emissions will more than halve, to 0.8 kgCO2/tkm, and will be completely decarbonised by 2050.

In passenger aviation transport, specifc CO2 emissions will decrease from 425 gCO2/pkm in 2019 to 350 gCO2/pkm in 2025, will halve by 2035, and will be CO2-free by 2050—analogous to freight transport. Both reduction trajectories will be achieved by the gradual replacement of fossil kerosene with organic kerosene, and after 2040, with synthetic kerosene that is generated with renewable electricity. Because aviation is a truly global sub-sector, the assumptions for aviation are the same for all regions.

# **8.5 Shipping**

Of the global energy demand for shipping, 90% is for freight transport, and only around 10% is for passenger transport (mainly cruise ships and ferries). In 2018, the worldwide cruise ship passenger capacity was 537,000 passengers on 314 ships, and 26 million passengers were transported in 2018 (Cruise Market Watch, 2020). In comparison, around 53,000 merchant ships were registered globally in January 2019: approximately 17,000 cargo ships, 11,500 bulk cargo carriers, 7500 oil tankers, 5700 chemical tankers, and 5150 container ships. The remaining ships included roll-on, roll-off passenger and freight transport ships and liquefed natural gas (LNG) tankers (Statista, 2021).

# *8.5.1 Energy Intensity and Emission Factors: Shipping*

The energy intensity for freight transport by ship was assumed to be 0.19 MJ/tkm in 2019 (Pagenkopf et al., 2019) and will decrease only slightly to 0.18 MJ/tkm in 2030 and 0.17 MJ/tkm in 2050. An equivalent trajectory is assumed for shipping passengers, from 0.056 to 0.054 MJ/pkm in 2030 and to 0.052 MJ/pkm in 2050. Shipping is already by far the most effcient transport mode. However, further technical improvements, especially in ship engines, are required. The volume of freight (in tkm) is assumed to increase by around 0.5% per year globally until 2050, whereas passenger transport volumes will remain at today's levels over the entire modelling period.

The emissions factor for heavy fuel oil is calculated to be 81.3 gCO2/MJ (Jurich, 2016). The specifc CO2 emissions for shipping freight will decrease from 15 gCO2/

tkm to 10 kgCO2/tkm by 2030. By 2040, freight shipping will be completely decarbonised. The specifc CO2 emissions for passenger shipping transport will decrease from 5 gCO2/pkm in 2019 to 3 gCO2/pkm in 2030, and analogous to freight shipping, passenger transport by ship will be carbon neutral by 2040. Both reduction trajectories will be achieved by the gradual replacement of fossil fuels with biofuels and, after 2040, with renewables-generated synthetic fuels.

# **8.6 Land Transport**

Although the most-effcient transport mode for long distances over land is railways, vehicular road transport for passenger and freight transport dominates by an order of magnitude.

Road transport is the single largest consumer of oil. In 2018, 64% of the global demand was attributed to road transport vehicles, for both freight and passenger transport. The pandemic in 2020 led to a unique development: as a consequence of global lockdown measures, mobility (57% of the global oil demand) declined at an unprecedented rate. The road transport in regions under lockdown decreased by 50–75%, with the global average road transport activity falling to almost 50% of the 2019 level by the end of March 2020 (IEA, 2020c).

Whereas electric-powered planes or ships are still in the early stages of development, there are no technical barriers to the phasing-out of ICEs or the transition to effcient electric vehicles (EVs) for passenger transport and to hydrogen or synthetic biofuels for heavy-duty vehicles. The vehicle technology required is widely available and market shares are rising sharply. In 2012, only 110,000 battery electric vehicles (BEVs) had been sold worldwide. Since then, sales have almost doubled every year, reaching 1.18 million BEVs in 2016, 3.27 million in 2018, and 4.79 million in 2019 (IEA, 2020d).

# *8.6.1 Energy Intensity and Emission Factors: Land Transport*

#### **Individual Transport**

Passenger transport by road makes up by far the commonest and most important form of travel. There are numerous technical options to 'move people with vehicles'—bicycles, motorcycles, tricycles, city cars, four-wheel drive SUVs—and each vehicle has very different energy intensity per kilometre. Although this research project aims for high-technology resolution, simplifcation is required. First and foremost, the data for all existing vehicles for each of the regions and for the global level are neither available nor practical to use. Figure 8.3 shows the energy intensities for the main vehicle types, which form the basis for the energy scenario calculations (Table 8.3).


**Table 8.3** Energy intensities for individual transport modes—road transport


#### **Public Transport**

There are a wide variety of public transport vehicles, ranging from rickshaws to taxis and from minibuses to long-distance trains. The occupation rates for those vehicles are key to calculating the energy intensity per passenger kilometre. For example, a diesel-powered city bus that transports 75 passengers requires, on average, about 27.5 litres per 100 kilometres. If the bus is operating at full capacity during peak hour, the energy demand per passenger is as low as 400 ml per kilometre—lower than almost all other fossil-fuel-based road transport vehicles. However, if the occupancy drops to 10% (e.g. for a night bus), the energy intensity increases to 3.7 litres, equal to that of a small energy-effcient car. Occupation rates vary signifcantly and depend upon the time of day, day of the week, and season. There are also signifcant regional differences, even within a single country, and even more so across larger regions, such as OECD Europe, which is composed of over 30 countries from Iceland to Turkey.

Again, the parameters shown in Table 8.4 are simplifed averages and are further condensed for the scenario calculations. Although high technical resolution is possible for the scenario model, it would imply an accuracy that does not exist, because the statistical data required for this are not available on either regional or global levels.

#### **Freight Transport**

The energy intensity data for freight transport are not as diverse as those for passenger transport, because the transport vehicle types are more standardised and the fuel demand is well known. However, the utilisation rate of the load capacity varies signifcantly, and consistent data are not available for the regional and global levels calculated. Therefore, the assumed utilisation rate has a huge infuence on the calculated energy intensity per tonne–kilometre. The average energy intensities per tonne–kilometres used in the scenarios are shown in Table 8.5 and are largely consistent with other sources in the scientifc literature. The assumed energy intensities for electric and fuel cell/hydrogen freight vehicles are only estimates, because this technology is still in the demonstration phase. Therefore, none of the scenarios calculated factor in large shares of electric freight transport vehicles before 2035.

# **8.7 Global Transport Demand Projections**

A variety of actions will be required for the transport sector to conform to the limit global warming to 1.5 °C. The set of actions described can be clustered into technical and operational measures (e.g. increases in energy effciency, electrifcation of drive trains), behavioural measures (e.g. shifts to less-carbon-intensive transport carriers and an overall reduction in transport activity), and accompanying policy measures (e.g. taxation, regulations, urban planning, and the promotion of lessharmful transport modes).


**Table 8.4** Energy intensities for public transport—road and rail transport


**Table 8.5**

Energy intensities for freight transport—road and rail transport

**Fig. 8.5** Energy intensities for urban and interurban passenger transport modes in 2019 (world averages). (Source: DLR/ IFFT 2019, Deutsches Zentrum für Luft- und Raumfahrt (DLR), Institut für Fahrzeugkonzepte, Fahrzeugsysteme und Technologiebewertung, Stuttgart, Data from Johannes Pagenkopf et al. 2019)

The key requirements for achieving a reduction of the transport energy demand in the alternative scenarios follow a three-step approach:


# *8.7.1 Projection of the Transport Service Demand*

The frst step in the projection of the global transport demand is calculating the actual service demand in passenger–kilometres travelled and tonnes of goods–kilometres transported. This is essential before the development of the chosen transport mode (road, rail, or ship) is projected.

Under the three scenarios, the global transport demand is the sum of the ten world regions plus bunker fuels. Bunker fuels are all the fuels required for interregional aviation and shipping transport and are therefore not part of any regional demand. The assumed development is based on the population and economic developments in \$GDP provided in Table 8.2. The 1.5 °C scenario assumes a reduction in the global pkm of 30% relative to 2020, whereas the global freight demand will increase by 30% based on the assumption of a growing GDP (Tables 8.6 and 8.7).


**Table 8.6** Global: development of behavioural changes in passenger travel (based on pkm) by transport mode

**Table 8.7** Global: development of changes in freight logistics (based on tkm) by transport mode


**Fig. 8.6** Energy intensities for freight transport modes in 2019 (world averages). (Source: DLR IFFT 2019)

# *8.7.2 Mode-Specifc Technology Effciency and Improvements Over Time*

For passenger transport, trains and buses are much more energy effcient per pkm than passenger cars or airplanes. This situation does not change fundamentally if only electric drive trains are compared (Fig. 8.5). Railways and (especially) ships are clearly more energy effcient than trucks in transporting freight (Fig. 8.6). The effciency data are based on both literature-reported and on transport-operator documents in this study and on Pagenkopf et al. (2019). Effciency levels, in terms of pkm or tkm, depend to a large extent on the underlying utilisation of the capacity of the vehicles, which varies across world regions. The numbers presented are average values and differences are evaluated at the regional level.

# *8.7.3 Powertrain Electrifcation for Road Transport*

Increasing the market penetration of highly effcient (battery and fuel cell) electric vehicles, coupled with the generation of clean electricity, is a powerful lever for decarbonisation and probably the most effective means of moving toward a decarbonised transport system.

All-electric vehicles have the highest effciency levels of all the drivetrain options. Today, only a few countries have signifcant proportions of electric vehicles in their feets. The total number of electric vehicles, particularly for road transport, is insignifcant, but because road transport is by far the largest CO2 emitter of overall transport, it offers a very powerful lever for decarbonisation.

In terms of drivetrain electrifcation, we cluster the world regions into three groups, according to the diffusion theory (Rogers, 2003):


Although this clustering is rough, it suffciently mirrors the basic tendencies implemented in our scenarios. The regions differ in the speed with which novel technologies, especially electric drivetrains, will penetrate the market.

In addition to powertrain electrifcation, there are other potential improvements in energy effciency, and their implementation will steadily improve these energy intensities over time. Regardless of the type of power train and the fuel used, effciency improvements on MJ/pkm or MJ/tkm will result from (for example):


# *8.7.4 Projection of Global and Regional Modal Shifts*

In 2019, road transport predominated over all other transport modes, with almost 95% of all pkm travelled by some form of road vehicle throughout the world. Based on the kilometres travelled, just over 3.5% of journeys were by train and about 2% by plane. Although ship transport is one of the most important means of transport for freight, marine-based passenger transport makes only a very minor contribution at the global level. To implement the 1.5 °C scenario, passenger transport must shift from road to rail. Effcient light rail in cities, commuter trains for short to medium distances, and high-speed trains that offer convenient services are therefore alternatives to individual car journeys.

In the context of urban transport, the use of road transport by cars will be signifcantly reduced and will move towards public transport by other road vehicles, such as buses or trains. The role of electric bikes and walking must also increase under the 1.5 °C scenario. However, road transport will remain dominant, at well over 80% (Fig. 8.7), until 2050. Therefore, the modal shifts within road transport systems, such as from individual cars to public transport, cycling, or mobility services (such as car sharing), are extremely important.

Maritime shipping is the backbone of world trade. It is estimated that some 80% of all goods are carried by sea. In terms of value, the global maritime container trade is estimated to account for around 60% of all seaborne trade, which was valued at around \$US14 trillion in 2019 (Placek, 2021).

In terms of tonnage, aviation plays a comparatively minor role globally. In terms of tonne–kilometres, road transport dominates globally. Every second tonne is transported by road and only 10% by rail (Fig. 8.8). However, the different transport modes cannot be separated because goods delivered by ship are further distributed by road and rail. Therefore, a direct modular shift is often not possible. Ship transport cannot be replaced by trains in most cases, and vice versa. There is competition between road and rail, and modular shifts in favour of rail freight transport will occur. The 1.5 °C pathway assumes that about one-third of the freight transported by trucks will be shifted to rail transport systems.

**Fig. 8.7** World passenger transport by mode under the 1.5 °C scenario—shares based on passenger–kilometres

**Fig. 8.8** World freight transport by mode under the 1.5 °C scenario—shares based on tonne– kilometres

Compared with passenger transport, freight transport is far more diverse, and regional differences are signifcant. In Eurasia, a region very similar to the former USSR, rail transport shoulders about half of all freight transport in terms of tonnage. This refects the signifcance of the Trans-Siberian Railway line connecting the European part of Russia with Mongolia (Ulan Bator) and China (Beijing).

In Non-OECD Asia, water transport is by far the most important transport mode, which refects the situations in the island states Indonesia and the Philippines, as well as the vast coastlines of Southeast Asian countries.

# *8.7.5 Calculation of Transport Energy Demand*

The calculation of the transport demand is based on a two-step approach, with all the parameters described in the previous subsections (Sects. 8.7.1, 8.7.2, 8.7.3, and 8.7.4):


To calibrate the model, the transport demand of the past decade was recalculated on the basis of the available energy statistics. The International Energy Agencies' (IEA) Advanced World Energy Balances provided the total fnal energy demands by transport mode—aviation, navigation, rail, and road—by country, region, or globally. However, there is no further specifcation of the energy usage within each of the transport modes. A further division into passenger and freight transport is therefore calculated using percentage shares. These proportions are determined with a literature research and from the average energy intensity for each of the transport modes for passenger and freight vehicles.

The annual energy demand divided by the average energy intensity by mode generates the annual transport demand in passenger–kilometres per year [pkm/yr]


**Table 8.8** Calibration for transport demand calculations

**Table 8.9** Projection of transport demand based on changing demand in kilometres


and tonne–kilometres per year [tkm/yr]. Those results are then compared with the OECD transport statistics, which provide both parameters, pkm/yr and tkm/yr. Calibrating the model with historical data ensures that the basis of the scenario projection for the coming years and decades has been correctly mapped and that the changes can be calculated more realistically.

For the forward projection of the transport demand, the calculation method is reversed. The transport demand for each transport mode is calculated on the basis of the annual change (as a percentage). The calculated total annual passenger–kilometres and tonne–kilometres are the inputs for the energy demand calculations.

# *8.7.6 Transport Service: Energy Supply Calculation*

Like the transport demand calculation, calculation of the transport energy 'supply' begins with the calibration of the model based on historical data, as part of a twostep approach:


As well as the fnal energy demand for each transport mode, the IEA Advanced World Energy Balances also provide the energy demand by source—soil, gas, biofuels, and electricity. To calculate the exact energy requirement for each transport mode with the corresponding transport requirement (in km), assuming different vehicle technologies, the status quo must be determined. For this purpose, the respective transport energy requirement for each transport mode and fuel type is calculated based on the current vehicle technology market shares and the technologyspecifc energy intensities per kilometre. The results provide a technology-specifc illustration of each sector. Table 8.10 presents an overview of the calculation process for the calibration of the model.

Future energy demands based on the projected pkm and tkm are calculated from market shares and technology-specifc energy intensities. In the frst step, the overall transport energy demand, e.g. in passenger–kilometres, is distributed to each transport mode. A mode shift from road to rail can be assumed, and the sectorspecifc demand is further distributed to specifc vehicle types—again by the assumption of market shares (Table 8.11).

# **8.8 Transport: Energy Demand and Supply**

In the previous sections, the global energy demand was calculated based on the documented assumptions. However, the transport sector is among the most diverse sectors of all the end-use sectors analysed. A whole range of logistical, technical,


**Table 8.10** Calibration for transport demand calculations

All the energy carriers are summed by transport mode to calculate the total energy demand for aviation, navigation, rail, and road

All energy demands are summed by transport mode to calculate the total energy demand for transport

**Table 8.11** Projection of transport supply based on transport mode and vehicle-specifc parameters


and political measures are required to reduce the overall energy demand while maintaining freedom of movement and mobility. The transport sector is closely related to the buildings sector, because urban planning and urban designs go hand in hand with the transport demand—in terms of the distances travelled or goods transported—and the most suitable technical solutions to provide those services. Furthermore, the carbon intensity of the electricity consumed for transport is directly related to the renewable energy share in power generation.

# *8.8.1 Shipping and Aviation: Dominated by Combustion Engines for Decades to Come*

Navigation will probably remain predominantly powered by ICEs in the next few decades. Therefore, we did not model the electrifcation of freight vessels. However, pilot projects using diesel hybrids, batteries, and fuel cells are in preparation (DNV, 2015). We assumed the same increase in the share of bio- and synthetic fuels over time as in the road and rail sectors.

In aviation, energy effciency can be improved by measures such as winglets, advanced composite-based lightweight structures, powertrain hybridisation, and enhanced air traffc management systems (Vyas et al., 2013; Madavan, 2016). We project a 1% annual increase in effciency on a per pkm basis and a 1% annual increase in effciency on a per tkm basis.

Aviation will probably remain predominantly powered by liquid fossil fuels (kerosene and bio- and synthetic fuel derivatives) in the medium to long term because of the limitations in electrical energy storage. We project a moderate increase in domestic pkm fown in electric aircraft starting in 2030, with larger shares in OECD Europe, because the fight distances are shorter than, for example, in the USA or Australia. Norway has announced plans to perform all short-haul fights electrically by 2040 (Agence France-Presse, 2018).

However, no real electrifcation breakthrough in aviation is foreseeable unless the attainable energy densities of batteries increase to 800–1000 Wh/kg, which will require fast-charging post-lithium battery chemistries.

That said, it is estimated that over 200 electric aircraft programs are in progress around the world (Downing, 2019). While small electric planes (up to car size) are in the demonstration phase, long-haul fights with electric planes are currently unviable with contemporary battery technology.

From the perspective of technological innovation, electric aviation is an important feld of engineering, and investment in this sector must occur now to achieve results in the mid-2030s. Domestic aviation—mainly short-distance fights of up to around 700 km—makes up about 45% of all global fights (Downing, 2019). The electrifcation of passenger planes for these distances will most likely start in this market segment.


**Table 8.12** Aviation—energy demand and supply

However, this research has focused on the rapid reduction of CO2 in the global transport sector, and realistically, electric aviation will not play a role in the reduction of large amounts of carbon before 2040. Nevertheless, the development of this technology is important in the long term (Tables 8.12 and 8.13).

A key target for the global transport sector is the introduction of incentives for people to drive smaller cars and use new, more-effcient vehicle concepts. It is also vital to shift transport use to effcient modes, such as rail, light rail, and buses, especially in large expanding metropolitan areas. Furthermore, the 1.5 °C scenario cannot be implemented without behavioural changes. It is not enough to simply exchange vehicle technologies, but the transport demand must be reduced in terms of the kilometres travelled and by an increase in 'non-energy' travel modes, such as cycling and walking.

With population increases, GDP growth, and higher living standards, the energy demand of the transport sector is expected to increase without technical and behavioural changes. Under the 1.5 °C scenario, effciency measures, modal shifts, and the behavioural changes mentioned above will reverse the trend in permanent growth (Table 8.14).

The proportion of BEVs among all passenger cars and light commercial vehicles in use is projected to be between 8% and 15% by 2030. This will require a massive build-up of battery production capacity in the coming years. New car sales will already be dominated by battery electric passenger vehicles in 2030 under the 1.5 °C scenario. However, with an assumed average lifetime of 15 years for ICE passenger cars, the existing car feet will still predominantly use ICEs.


**Table 8.13** Shipping—energy demand and supply

Under the assumption that new ICE passenger cars and buses will not be produced after 2030, BEVs will dominate the passenger vehicle feet of 2050 under the 1.5 °C scenario. OECD countries and China are assumed to lead the development of BEVs and therefore to have the highest shares, whereas Africa and Latin America are expected to have the lowest BEV shares. Fuel cell-powered passenger vehicles are projected to play a signifcantly smaller role than BEVs and will only be used for larger vehicles, such as SUVs and buses (Fig. 8.9).

The shares of electric trains and diesel-powered locomotives vary signifcantly by region (Fig. 8.10). Under the 1.5 °C scenario, all diesel locomotives will be phased out in all regions by 2050. It is assumed that biofuels and synthetic fuels, as well as hydrogen, will play a minor role and that around 90% of all trains—for both passenger and freight transport—will use electric locomotives. The highest utilisation rates of diesel locomotives in 2019 were in the Middle East (98%) and OECD North America (95%), whereas the majority of trains in Europe were electrifed.

Highly effcient drives—with a focus on electric mobility—supplied with renewables will result in large effciency gains. By 2030, electricity will provide 5% of the transport sector's total energy demand under the 1.5 °C scenario, whereas in 2050, the share will be 37%. The majority of electricity consumed in the transport sector


**Table 8.14** Road transport—energy demand and supply

**Fig. 8.9** Proportions of powertrains in (feet) passenger cars and buses by region in 2030 (left) and 2050 (right)

**Fig. 8.10** Proportions of electrifed passenger and freight rail in 2019 (left) and 2030 (right)—1.5 °C scenario


**Table 8.15** Transport sector—fnal energy demand and supply

will be for land transport—road and rail. Hydrogen and other synthetic fuels generated with renewable electricity will be complementary options to further increase the share of renewable energy in the transport sector, especially for aviation and shipping. In 2050, up to 7700 PJ/yr of hydrogen will be required under the 1.5 °C transport pathway (Table 8.15).

The high reliance on renewable electricity, used either directly in BEVs or to produce synthetic fuels, will require close cooperation between the transport sector and the power sector, not only in terms of the decarbonisation of the power sector itself but also in terms of the increasing electricity demand. In our analysis, the electrifcation of the transport sector—especially the replacement of ICEs with BEVs—will roughly double the electricity demand of an industrialised country if no further effciency measures are taken in other sectors, such as the residential and service sectors.

# **8.9 Transport: Energy-Related CO2 Emissions**

The overall energy-related CO2 emissions are directly linked to the power sector, as stated above. Under the assumption that electricity generation is fully decarbonised by 2050 (see Power sector trajectory, Chap. 12), Tables 8.16, 8.17, and 8.18 show the carbon intensities and total CO2 emissions for aviation, shipping, and road transport, respectively, under the 1.5 °C scenario. Both the aviation and shipping values include domestic and international transport. Emissions intensity is an important key performance indicator (KPI) for the fnance industry, for both *Climate Change Stress Tests* (see Chap. 2) and the evaluation of investment portfolios that include transport industry assets. For the automobile industry, carbon intensities (in gCO2/ km) are an important KPI and have already been used for mandatory effciency standards, such as those in the European Community (EU, 2021).


**Table 8.16** Aviation—energy-related CO2 emissions

**Table 8.17** Shipping—energy-related CO2 emissions


**Table 8.18** Road transport—energy-related CO2 emissions


# **8.10 Transport Equipment**

According to the OECD defnition, *'*Transport equipment (assets) consists of equipment for moving people and objects, other than any such equipment acquired by households for fnal consumption'(OECD SP, 2021). According to the 2020 edition of the International Energy Agency's World Energy Balances Database Documentation (IEA, 2020b), the energy demand for 'transport equipment' includes industries under Divisions 29 and 30 of the *International Standard Industrial Classifcation of All Economic* (ISIC) Rev. 4 (ISIC, 2008). Table 8.19 shows the industries that are classifed under 'transport equipment'. Based on this classifcation, the economic values for all sub-sectors were estimated.

Table 8.20 shows the estimated economic breakdown of all sub-sectors of the transport equipment industries. The literature provides various different defnitions and economic values for the global automotive industries and for the aviation and shipping industries. However, some of the much higher values (e.g. for the car industry) include the value added for sales and other related services.

Table 8.21 shows the calculated global values for all sub-sectors of the transport equipment industry. The purpose of this analysis is to estimate the energy demand for the manufacture of vehicles, ships, and planes, because the exact statistics for the energy demands of those industries are not available on the global level. To maintain consistency in our methodology, the energy demand for transport equipment provided by the IEA database was used. However, further research is required to determine the industries' exact energy demands.


**Table 8.19** Industries classifed under 'transport equipment'




**Table 8.21** Global transport equipment—estimated GDP values by sub-sector and projection until 2050

**Table 8.22** Global transport equipment—estimated energy intensities by sub-sector and projection until 2050


In the absence of more-detailed information about the energy intensity of the industries analysed, the same values have been assumed for the manufacture of cars, locomotives, ships, and planes. Consistent with this assumption, the same effciency progress ratio of 0.5% per year has been assumed over the entire scenario period until 2050. More research is required to estimate the energy demand and supply for these industries in the future (Table 8.22).

Based on IEA statistics, the share of electricity in the total energy demand has been calculated as 47%, whereas the remaining 53% is required for heat. The breakdown by temperature level has been estimated as 72% for low-temperature heat (<100 °C) and 10% for medium-temperature heat (100–500 °C), and the remaining demand is for process heat (5% for 500–1000 °C; 13% for >1000 °C). More-detailed assessments of the process heat requirements were not available for this analysis (Table 8.23).

Finally, the calculated energy-related CO2 emissions for transport equipment are shown in Table 8.24. The emissions are based on the 1.5 °C pathways for electricity


**Table 8.23** Global transport equipment—calculated energy demand by sub-sector

**Table 8.24** Global transport equipment—calculated energy-related CO2 emissions


and (process) heat generation (see Chap. 12). The values shown here were used for the Scope 1, 2, and 3 analyses reported in Chap. 12 (Results: industry pathways) and Chap. 13 (Scope 3: industry emissions and future pathways).

# **References**

Agence France-Presse. (2018). Norway aims for all short-haul fights to be 100% electric by 2040 | Air transport | *The Guardian*. https://www.theguardian.com/world/2018/jan/18/norway-aimsfor-all-short-haul-fights-to-be-100-electric-by-2040. Accessed 22 Dec 2021.


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

# **Part V Energy Industry & Engineering**

# **Chapter 9 Renewable Energy for Industry Supply**

**Sven Teske, Thomas Pregger, Sonja Simon, and Carina Harpprecht**

**Abstract** This section focuses on technologies that provide heat, and especially process heat, with renewable energy and electrical systems. All the technologies described, except those that use high-temperature geothermal or concentrated solar heat (CSH) for process heat, are used for the OECM 1.5 °C pathways described in Chaps. 5, 6, 7, and 8. The authors have included geothermal and solar technologies to highlight the further technical options available and to underscore that more research is required in the area of renewable process heat.

**Keywords** Industry process heat by sector · Renewable process heat · Electric process heat · Solar · Bio energy · Geothermal · Heat pumps · Arc furnace · Hydrogen · Synthetic fuels · Power-to-X)

# **9.1 Introduction**

Heat generation relies currently, to a large extent, on combustion processes. In 2019, 77% of global heating for buildings and industrial process heat came from fossil fuels, whereas only 3.2% was provided by electric heating systems, and 23% was supplied by renewable heating almost entirely from biomass. Only 0.9% derived from solar and geothermal heating systems. To decarbonize the global heat supply is more challenging than to decarbonize the electricity sector, because geographic

S. Teske (\*)

T. Pregger · S. Simon

#### C. Harpprecht Institute of Networked Energy Systems, German Aerospace Center (DLR), Stuttgart, Germany

Institute for Sustainable Futures, University of Technology Sydney, Sydney, NSW, Australia e-mail: sven.teske@uts.edu.au

Department of Energy Systems Analysis, Institute for Engineering Thermodynamics (TT), German Aerospace Center (DLR), Stuttgart, Germany e-mail: thomas.pregger@dlr.de; sonja.simon@dlr.de

limitations make it diffcult to provide high-temperature heat with direct solar or geothermal energy due to their dependency on locally available resources. However, the use of renewable electricity for heating is key to a successful 1.5 °C pathway. This section provides a short overview of the suitable technologies available and the temperature levels that these technologies can generate.

Industry involves a large variety of processes that demand heat. These requirements range, for example, from 40 °C to around 300 °C in the food industry to metal production with furnaces well above 800 °C and cement production with dry kilns at around 1500 °C. Figure 9.1 shows that the metal, chemical, and mineral industries require particularly large amounts of high-temperature process heat.

Decarbonizing process heat for energy-intensive industries, such as the steel, aluminium, cement, and chemical industries, is a major prerequisite to remaining within a 1.5 °C increase in the global temperature. Three main groups of technologies can provide renewable process heat at different temperatures:


The energy sources for these heat generation technologies are either biomass, geothermal energy, solar energy, or electricity, used either directly or as fuels produced with electricity, such as hydrogen and other synthetic fuels. Whereas the most effcient transformation to renewable energy is the direct application of renewable heat, many industrial processes require higher temperatures or fuels, which cannot be provided directly by renewables. Therefore, as the next best option in terms of effciency, the direct electrifcation of processes is preferable. However, some processes

**Fig. 9.1** Distribution of process heat demand across all branches of industry in Europe. (Naegler et al., 2015)

will still rely on fuel input in the future. In this case, power-based synthetic fuels will be required, with increasing effciency losses along the chain from hydrogen to synthetic gas to synthetic liquid fuels. To comply with the 1.5 °C carbon budget, all the electricity used for heat or fuel production must be produced from renewable energy.

In Table 9.1, we compare the technology options used in the OECM for generating different levels of heat. Their descriptions are provided in the following subchapters. It can be seen that the use of fuels is most suitable for high-temperature process heat, although direct electrifcation is also suitable to some extent. Geothermal energy is particularly suitable for lower temperatures, whereas direct solar energy can only be used to generate high-temperature heat via central receivers. This illustration does not consider the opportunities for or barriers to renewable heat integration that may arise from technical or local structural constraints.

Table 9.2 shows the average breakdown between heat and fuels and electricity in percentages of the industry sectors analysed, which are included in the OECM. The data demonstrate the large share of high-temperature heat in the primary industries. The average energy demand for the steel industry in 2019, for example, was mainly for the generation of process heat (86%), with 14% for electricity. The value for process heat includes fuels.


**Table 9.1** Technology options to generate renewable heat by temperature level


**Table 9.2** Average electricity and heat shares by industry in 2019 (heat includes electricity for heating and fuels)

However, whereas the share of the actual heat demand will remain stable for each sector in the OECM until the end of the scenario period in 2050, the electricity used to produce heat will increase. A more detailed bottom-up analysis broken down into primary and secondary steel and aluminium and new manufacturing processes has been undertaken. The assumptions for the process heat calculation for each industry sector are presented in Chap. 5.

In the following section, these different technologies are outlined, and their respective areas of application are explained.

# **9.2 Direct Renewable Process Heat**

# *9.2.1 Bioenergy and Biofuels*

'Biomass' is a broad term used to describe materials of recent biological origin that can be used as a source of energy. It includes wood, crops, algae, and other plants, as well as agricultural and forest residues (Teske & Pregger, 2015). Biomass is used to generate electricity, heat, and fuels. The following section focuses on heat generation.

The majority—around 90%—of bioenergy is used in direct combustion processes to generate heat and/or electricity, mostly for domestic and low-temperature applications. However, many studies and scenarios that have considered the potential of biomass have envisaged a shift in its currently limited potential to allow the generation of high-temperature industrial process heat, in the transition towards a renewable energy system (Lenz et al., 2020).

In principle, two biomass conversion routes are available for the production of heat for industry, using several biomass technologies:

	- Direct combustion
	- Gasifcation
	- Pyrolysis
	- Anaerobic digestion
	- Fermentation

#### **9.2.1.1 Thermochemical Processes**

Direct Combustion

The direct combustion technologies relevant to the generation of process heat can be differentiated according to the state in which the biomass is fed into and burned in the process.

In fxed-bed combustion applications, the air is frst passed through a fxed bed for drying, gasifcation, and charcoal combustion. In the second step, the combustible gases produced are burned with air, usually in a zone separated from the fuel bed. Fixed-bed combustion is adaptable to a variety of fuels such as wood, straw chips, and pellets. Therefore, the range of capacities is large, ranging from 10 kW to 60 MW.

The fuidized-bed technology involves the combustion of particulate solid fuel in an inert material bed (usually sand), which is fuidized by the fow of a gas. This type of fow allows effcient gas–solid contact, so it is widely used in covering particles, drying, granulation, blending, combustion, and gasifcation processes (Philippsen et al., 2015). This technology provides almost complete combustion, with very stable temperatures and low emissions. The prerequisites are fuels with particle sizes <100 mm and ash melting temperatures >1000 °C (Kaltschmitt et al., 2009). Entrained-fow combustion is suitable for fuels that are available as small particles, such as sawdust or fne shavings, which are pneumatically injected into the furnace. Fluidized-bed combustion is generally used in larger systems (> 20 MW), because it is expensive (Teske & Pregger, 2015; ARENA, 2019).

#### Gasifcation

Biomass gasifcation is a method for upgrading solid biomass and is especially valuable in processing biomass of low caloric value or moist biomass, e.g. many residues. The partial oxidation of the biomass fuel provides a combustible gas mixture mainly consisting of carbon monoxide (CO). Gasifcation provides a homogeneous fuel and controlled combustion, which can increase the effciency along the whole biomass chain, although at the expense of additional investments in the more sophisticated technology, or the effcient use of low-quality biomass. During the frst step, the volatile components of the fuel are vaporized in a complex set of reactions at temperatures <600 °C. Gasifcation is an intermediate step between pyrolysis and combustion. It is a two-step, endothermic process (IEA BioEnergy Agreement Task 33, 2020).

Biomass gasifcation is increasingly used to generate high temperature levels. The most commonly available gasifers use wood or woody biomass, whereas especially designed gasifers can convert non-woody biomass materials (Norfadhilah et al., 2017). Gasifcation is more effcient than combustion, providing bettercontrolled heating, higher effciencies in power production, and the possibility for co-producing chemicals and fuels (Kirkels & Verbong, 2011). Gasifcation can also reduce emission levels better than power production with direct combustion and a steam cycle. Finally, gasifcation can also be the frst step in the production of synthetic fuels (Malico et al., 2019) (see next section).

#### Pyrolysis

Pyrolysis is a technology that 'upgrades' biomass, providing products of high caloric value for combustion. It has been long used in the production of charcoal (Malico et al., 2019). Technically, thermal decomposition occurs in the absence of oxygen. It is also always the frst step in combustion and gasifcation processes, where it is followed by the total or partial oxidation of the primary products (IEA BioEnergy Agreement Task 34, 2021). Pyrolysis produces a solid (charcoal), liquid (pyrolysis oil or bio-oil), and gas product. The relative amounts of the three products are determined by the operating temperature and the residence time used in the process. Lower temperatures produce more solid and liquid products, and higher temperatures, more biogas. All the products are then available for the production of industrial process heat.

#### **9.2.1.2 Biochemical Conversion Processes**

Anaerobic digestion and fermentation are the two main biochemical processes that provide energy from biomass with high moisture content, such as food waste or agricultural residues, including liquid manure.

#### Anaerobic Digestion

In a biogas plant, organic waste is broken down by bacteria in an oxygen-free (= anaerobic) environment in about two-thirds methane (CH4) and one-third CO2. This gas is used either directly in power, heating, or cogeneration plants, or purifed gas is fed into renewable gas pipelines. For its direct injection into natural gas pipelines, the CH4 content must be increased to approximately 95% (Wall et al., 2018). The quality of the renewable gas produced depends on the energy content of the feedstock. Possible feedstocks include food waste, livestock manure, process effuent, sewage sludge, and domestic biowaste.

# Alcoholic Fermentation

The alcoholic fermentation of sugar and starch is a 'state-of-the-art' technology. Plants with high sugar and starch contents, such as sugar cane, are broken down into ethanol and methanol by microorganisms. Because the use of sugar and starch plants for this purpose is in direct competition with human nutrition, one direction of research focuses on the fermentation of lignocellulose, e.g. from straw or grass. Although lignocellulosic processes are more complex than the fermentation of carbohydrates, the frst production plants have been developed in Germany (DBFZ, 2015). The products can be used as combustible fuels for power, heat, or cogeneration plants and as a vehicle fuel. However, in the future, these products will become more important as low-emission feedstocks in a circular economy, with increased competition for the limited biomass potential (Table 9.3).

# **9.2.1.3 Bioenergy and Reduction of Greenhouse Gas (GHG) Emissions**

Bioenergy is not necessarily carbon neutral. Depending on the feedstock, which can be agricultural or forestry waste, other biogenic residues, or energy crops, bioenergy production has different upstream burdens in terms of the consumption of materials and energy, land-use changes, and emissions that have a signifcant impact on GHG emissions. Given the environmental effects of the production of energy crops, the global use of biomass in the 1.5 °C pathway is limited to 100 EJ per year,


**Table 9.3** Bioenergy for process heat—overview

*Sources*: IEA, ARENA, UTS/ISF, and DLR (own research)

which is the estimated threshold of carbon-neutral sustainable biomass based on residuals and organic waste (Thrän et al., 2011).

# *9.2.2 Geothermal*

Geothermal resources consist of thermal energy from the Earth's interior stored in both rocks and trapped steam or liquid water (IPCC-SRREN CH4, 2011). Although geothermal resources are available in all countries, their utilization is concentrated in regions where geothermal heat is available close to the Earth's surface. Geothermal 'hotspots' with high temperature levels occur in the western part of the USA, west and central Eastern Europe, Turkey, Iceland, and 'the ring of fre' around the Pacifc, from Japan, the Philippines, South-East Asia, and Indonesia to New Zealand.

Geothermal energy resources are classifed by temperature level (Huddlestone-Holmes, 2014). Each temperature level involves different technologies and applications. The global average thermal gradient is around 25–30 °C per km depth (Beardsmore & Cull, 2001), which results in an average crustal temperature of around 150 °C at a depth of 5000 m. Higher temperatures can be achieved by drilling deeper or by focusing on areas with favourable conditions. In such areas, the following temperatures are usually possible:


To date, high-temperature geothermal systems are almost exclusively used for power generation. However, high-temperature geothermal systems, around 200 °C, can also be utilized to provide direct process heat (ARENA, 2019).

Geothermal systems predominantly provide low-temperature process heat, which can be used, for example, in the food-processing industry (see Fig. 9.1).

In high-temperature hydrothermal reservoirs, water occurs naturally underground in its liquid form under pressure. As it is extracted, the pressure drops and the water is converted to steam. The residual salty water is sent back to the reservoir through injection wells, sometimes via another system that uses the remaining heat (Teske & Pregger, 2015). The hot water produced from intermediate-temperature hydrothermal or enhanced geothermal system (EGS) reservoirs can be used in heat exchangers, to generate power in a binary cycle, or directly in heat applications. The recovered fuids are also injected back into the reservoir (Younger, 2015).

The key technologies for EGS are:

#### 9 Renewable Energy for Industry Supply

	- Dry steam plants
	- Flash plants (single, double, or triple)
	- Binary combined-cycle plants or hybrid plants

# *9.2.3 Concentrated Solar*

Concentrating solar technologies generate high-temperature heat that can be used for industry processes (CSH) or to produce electricity via steam turbines (concentrated solar power—CSP).

<sup>1</sup>A Rankine cycle power system is a heat engine that converts thermal energy into work. Similar to the vapor compression heat pump, it comprises four main components: a boiler (sometimes called an 'evaporator'), a turbine, a condenser, and a pump (Fig. 9.2). The working fuid, in a low-pressure slightly subcooled liquid state, is brought to high pressure by the pump. The pump consumes power (ARENA2019).

Like high-temperature geothermal plants, concentrating solar technologies are currently predominantly utilized to generate power. However, the process heat temperature required for direct use in industrial processes (to 400 °C) is technically possible (DLR-ISR, 2021).

Direct normal irradiation—sunlight not dispersed by clouds, fumes, or dust in the atmosphere—is concentrated by mirrors to a single point or line to heat a liquid, solid, or gas to a temperature between 400 °C and >> 1000 °C, depending on the technology used. Concentrating solar plants require direct sunlight, which limits the areas of application to regions with more than 2000 h of direct sunlight per year.

There are several different CSP/CSH system types, but all require four main elements: a concentrator, a receiver, some form of transfer medium or storage, and a power conversion system or a connection that directs process heat to the site of its applications. An overview of the commonest concentrating solar systems is given by Pitz-Paal (2016).

*Parabolic trough plants* use rows of parabolic trough collectors, each of which refects solar radiation into an absorber tube. The troughs track the sun around one axis, which is typically oriented north–south. Synthetic oil circulates through the tubes and is heated to approximately 400 °C. The hot oil from numerous rows of troughs is passed through a heat exchanger. The direct evaporation of water in the parabolic troughs, which has been developed to operational maturity for years, will allow the realization of decentralized plants with relatively small solar felds, because heat exchangers and (possibly) toxic synthetic heat transfer fuids will no longer be required. Increasingly, CSP plants use thermal storage systems, such as molten salt, to store high-temperate heat (up to 400 °C) to allow their operation without sunlight or at night. The land requirements are around 2 km2 for a 100 MW plant, depending on the collector technology and assuming that no storage is available.

*Linear Fresnel systems* use a series of long, narrow, fat Fresnel mirrors instead of a parabolic trough to concentrate solar radiation to a linear absorber positioned above the lenses. All the other parts of the system correspond to those of parabolic trough plants.

*Central receivers or solar towers* focus solar radiation to a single point and achieve higher temperatures than parabolic troughs or Fresnel lenses. This technology uses a circular array of mirrors (heliostats) in which each mirror tracks the sun, refecting the light onto a fxed receiver on top of a tower. Temperatures exceeding 1000 °C can be achieved. A heat-transfer medium absorbs the highly concentrated radiation refected by the heliostats and converts it into thermal energy to be used for the subsequent generation of super-heated steam for turbine operation or as industrial process heat. The heat transfer medium is currently either water/steam, molten salts, liquid sodium, or air and possibly also pressurized gas or air at very high temperatures. The unit sizes range from 20 to 200 MW.

*Parabolic dishes* use a shaped refector to concentrate sunlight onto a receiver located at their focal points. The receiver moves with the dish. The concentrated beam radiation is absorbed into the receiver to heat a fuid or gas to approximately 750 °C. This is then used to generate electricity via Stirling engines or a micro-turbine attached to the receiver. Dishes have been used to power Stirling engines up to 900 °C and also to generate steam. The largest solar dishes have a 485 m2 aperture and are in research facilities or demonstration plants. Individual unit sizes are in the double-digit kilowatt range and can be combined in modular systems to form utility-scale plants. The generation of process heat is possible but is not yet commercially available.

*Concentrated solar heat (CSH) system applications*: In addition to its use in different types of solar refector systems, a CSH system can be used to directly feed into industrial processes, or to desalinate water. The signifcant cost reduction with solar photovoltaic systems has led to a focus on the application of concentrated solar for to heat generation rather than to the generation of electricity.

*Thermal storage*, when integrated into a system, is an additional and increasingly important asset in concentrated solar plants, providing heat outside the hours of sunshine and even during the night. Additional concentrator area can be added to produce heat for storage purposes, increasing the capacity factor. There are three categories of storage medium that can be used in CSP plants (Pitz-Paal, 2020):


The storage capacity currently installed is, on average, around 8 full-load hours. Concentrated solar power plants have been developed to generate electricity, but the technology has signifcant potential to provide high-temperature process heat in sunny regions, such as Australia, Chile, North Africa and the Sahara, parts of Central Asia, India, and China, as well as the Middle East. Research is targeting CSP as a source of high-temperature process heat that can directly feed reactors for endothermic chemical reactions. Currently, solar metal-oxide redox cycles and sulphur cycle processes have been developed that rely on temperatures of 1000–1500 °C (Roeb et al., 2020). The frst applications of this technology, for hydrogen production, have achieved technology readiness levels of 5–6, for example, in the SUN-to-LIQUID project (Koepf et al., 2019). Newer projects go beyond hydrogen and integrate the direct air capture of CO2 for the production of chemical feedstocks, such as methanol (Prats-Salvado et al., 2021).

# **9.3 Electric Process Heat**

# *9.3.1 Heat Pump Technology*

Heat pumps are largely known as electric heating (and cooling) systems that supply buildings with space heat and hot water. However, in general, heat pumps are devices that transfer heat from one medium at a lower temperature to another medium at a higher temperature. Therefore, they allow the effcient recycling of low-temperature heat, such as waste heat.

Heat pumps use a refrigeration cycle to provide heat or cold. They use renewable energy from the ground, water, or air to move heat from a relatively low-temperature reservoir (the 'source') to a temperature level required for a specifc thermal application (the 'output'). Heat pumps commonly use two types of refrigeration cycles:


Compression heat pumps are most commonly used, but thermally driven units are considered a promising future technology. The effciency of a heat pump is described by the coeffcient of performance (COP), the ratio between the annual useful heat output and the annual auxiliary energy consumption of the unit. In the residential market, heat pumps work best for relatively warm heat sources and lowtemperature applications, such as space heating and sanitary hot water. They are less effcient in providing higher-temperature heat and cannot be used for heat over 90 °C. For industrial applications, different refrigerants can be used to effciently provide heat of 80–90 °C, so they are only suitable for part of the energy requirements of industry.

Heat pumps are generally distinguished by the heat source they exploit:


Heat pumps require additional energy apart from the environmental heat extracted from the heat source, so the environmental beneft of heat pumps depends upon both

their effciency and their emissions associated with the production of working energy. When a heat pump has a low COP and a high share of electricity from coal power plants, for example, the CO2 emissions relative to the useful heat produced might be higher than for conventional gas condensing boilers. However, effcient heat pumps powered with renewable electricity are emission-free.

Reversible heat pumps can be operated in both heating and cooling modes. When they operate in cooling mode, heat is extracted from, for example, a building, and 'pumped' into either a reservoir or the open environment, without storage. When a reservoir is used, the heat can be reused. Alternatively, renewable cooling can be provided by circulating a cooling fuid through the relatively cool ground before it is distributed in a building's heating/cooling system ('free cooling'). However, in a GHG-emission-free system, this cooling fuid must not be based on hydrofuorocarbons (HFCs) or chlorofuorocarbons (CFCs) but on ammonia, water, or air (ARENA, 2019).

In principle, high-enthalpy geothermal heat can provide the energy required to drive an absorption chiller. However, only a very limited number of geothermal absorption chillers are in operation throughout the world. Heat pumps have become increasingly important in buildings but can also be used for industrial process heat. Industrial heat pumps offer various opportunities for all types of manufacturing processes and operations and use waste process heat as their heat source. They deliver heat at medium temperatures for use in industrial processes, heating, or preheating or for space heating and cooling in industry. Heat pumps with operating temperatures below 100 °C are state-of-the-art technologies, and high-temperature industrial heat pumps in the range of 160–200 °C are beginning to enter the market. Essential aspects of the future use of heat pumps are effcient system integration and fexibility via heat storage.

# *9.3.2 Electric Heating Systems*

There are four main technological types of electric heating systems, which use different physical methods. Each of them has different temperature levels and applications.

	- (a) Dielectric heating
	- (b) Infrared heating
	- (c) Induction
	- (a) Ultraviolet
	- (b) Pulsed electric feld


**Table 9.4** Electromagnetic process heating technologies

*Source*: ARENA, 2019

(c) Microwave2 .


Electromagnetic heating systems are used to transfer energy to a target material or process without the need for a heat transfer medium. The main advantage of this technology is that heat can be generated and delivered to the point of need, which makes this an energy-effcient technology (ARENA, 2019).

Non-thermal electrical systems generate heat directly on the target object, and no additional medium is required to transfer the heat (Xiong, 2021). Both technology groups use different frequencies to generate heat (Table 9.4).

### **9.3.2.1 Electric Resistance Heating**

Materials conduct electricity to different degrees. The lower the electrical conductivity of a material, the higher the heat developed within that material. This physical law—*ohmic resistan*ce—is used in electric resistance heating devices. There are two types of electrical resistance heating:


<sup>2</sup>Microwaves can generate signifcantly higher temperatures over time. Objects continue to heat while microwaves are emitted.

This technology is among the oldest electrical heating systems and has been used for room heat, industrial ovens, furnaces, and kilns for decades. Different confgurations of indirect resistance heating are:


#### **9.3.2.2 Electric Arc Furnaces**

An electric arc occurs when an electric current jump between electrodes. As the current passes through air (or another gas), it produces a plasma discharge, generating heat and light. Lightning is a natural form of electric arc (ARENA, 2019). Electric arc furnaces are predominately used in steel recycling, to melt scrap steel. However, they are also used in other industries that require temperatures up to 1500 °C, such as the processing of copper and other metals.

# **9.4 Synthetic Fuels and Hydrogen**

When the direct use of renewable heat sources (frst choice) or electrifcation (second choice) is not applicable, industrial processes will still rely on the input of fuels based on renewable electricity. For effciency reasons, hydrogen is the next best choice. However, as a last fuel option, synthetic hydrocarbons can provide the necessary energy.

# *9.4.1 Hydrogen: The Basics*

Hydrogen can be used as a feedstock, a fuel, an energy carrier, and for energy storage and has many possible applications across the industry, transport, energy, and buildings sectors. Molecular hydrogen does not occur in nature but can be produced using any primary source of energy, such as gas, oil, or coal. It can be produced by electrolysis, which requires electricity, or by directly splitting water with a solar high-temperature process. Therefore, hydrogen is not an energy source—it is a secondary energy carrier and an energy storage medium. The combustion of hydrogen gas only generates water and no further GHG emissions are produced.

The chemical formula for this process—the scientifc term is 'oxidation'—is

$$\text{2H}\_2 + \text{O}\_2 \text{ ?}\\2\text{H}\_2\text{O}.$$

Today, most of the world's hydrogen is still produced in CO2-intensive processes: steam–methane reformation (SMR) (gas, approximately 50%), oil product reformation (30%), and coal gasifcation (18%). In SMR, carbon (CO2) is separated from hydrogen by the steam reformation of natural gas. This method involves the conversion of hydrocarbons and steam into hydrogen and CO (known as 'syngas').

According to the London-based *Committee on Climate Change (CCC)*, SMR has an emissions factor of around 285 g of CO2 per kilowatt-hour (kWh) (9.5 kg of CO2 per kg of hydrogen), and coal gasifcation has an emission factor of around 675 g of CO2 per kilowatt of hydrogen, accounting only for energy use and process emissions. Therefore, arguments for the early establishment of an energetic use of hydrogen based on fossil energies are usually combined with arguments for the implementation of carbon capture and storage (CCS) technologies. Counterarguments point to the lock-in effect of investments in high-carbon infrastructure, which comes at the expense of fnancial resources for the expansion of renewable energies.

If hydrogen is to contribute to climate neutrality, it must achieve a far larger scale and its production must become fully decarbonized. According to the International Energy Agency (IEA), the current production of hydrogen—mainly based on natural gas—is responsible for CO2 emissions of around 830 million tonnes per year. By comparison, Germany's total CO2 emissions in 2020 are estimated to have been 722 million tonnes. It is estimated that 6% of global natural gas and 2% of global coal production are used for hydrogen production, whereas only about 0.1% of global dedicated hydrogen production is produced with water electrolysis.

#### **9.4.1.1 Status Quo: Global Demand for Hydrogen in Industry**

There are various applications for hydrogen in industry, as shown in Table 9.5, but only two main areas consume most of the global hydrogen produced today: ammonia production and refning processes. About 90% of ammonia is used for the production of fertilizers and the majority of the remaining 10% for cleaning products. In 2018, 43% of the global hydrogen demand was used for ammonia production and 52% for refning processes. Refneries use hydrogen to lower the sulphur content of diesel fuel. The remaining 6% of global hydrogen production is distributed across the other applications shown in Table 9.5.

The demand for hydrogen has grown continuously over the past decades, and the market shares for ammonia production and refnery processes have remained similar. The use of hydrogen for energy storage does not yet show in the global energy statistics.


**Table 9.5** Current areas of hydrogen use in industry

*Source*: Pregger et al. 2019

#### **9.4.1.2 Possible Applications of Hydrogen in Decarbonization Pathways**

Although there is a huge diversity of market projections and possible future applications for hydrogen, there is a broad consensus among all market analysts that the market for hydrogen will grow signifcantly in the coming decade. As well as its current application as a feedstock in chemical industries, hydrogen is expected to expand in the energy sector. Once electricity has been generated and used to produce hydrogen, this hydrogen can store energy in the form of a gas or (pressurized) liquid and replace fossil and/or biofuels in power plants (including fuel cells), cogenerating or heating plants to generate electricity, as heat, or as a transport fuel for vehicles. Figure 9.2 provides an overview of the possible future applications of hydrogen.

An important new industry sector for hydrogen is primary steel production. Based on current knowledge, the use of hydrogen for steel production is among the most promising processes to decarbonize the steel industry (Recharge 2020).

**Fig. 9.2** Areas of hydrogen application. (BNEF 2020)

### **9.4.1.3 New Processes to Produce Hydrogen**

In the public debate, colours are often used to refer to different processes for hydrogen production (IRENA 2020–1):


However, the only desirable production route is renewable hydrogen ('green') only in this case is it a zero-emissions technology. In all other-coloured methods, hydrogen production still demands fossil fuels, which are the greatest cause of climate change. Consequently, our report focuses on renewable ('green') hydrogen as a key element of climate neutrality.

### **9.4.1.4 Hydrogen and 'Power-to-X'**

When hydrogen is used as a fuel, 'power-to-X' (PtX) is often used as a term for the conversion processes and technologies involved. Power or 'P' is the electricity or input on the production side. 'X' can stand for any resulting fuel, chemical, power, or heat. PtX has received increasing public attention, because these technologies allow the indirect electrifcation of sectors that are (as yet) dependent on fossil fuels. PtX includes:


# *9.4.2 Synthetic Fuels*

Some (chemical) processes require either a liquid fuel or a carbon source, and will also do so in the future. Therefore, synthetic fuels are a prerequisite for carbonneutral industry. On the one hand, these synthetic fuels are based on renewable power. On the other hand, the production of synthetic liquid and gaseous hydrocarbons and methanol requires carbon sources. In a fossil-fuel-free circular carbon economy, only a few carbon sources will be available: carbon from biomass and CO2 emissions—from either waste incineration or fue gases, such as the processrelated emissions from cement production. Therefore, possible CO2 sources are not only industrial plants but also biogas plants or the direct air capture of CO2. Depending upon the carbon source and the output, the different PtX processes for the generation of synthetic fuels are defned as:


Hydrogen, methane, and liquid hydrocarbons are studied in numerous research projects for their possible use in long-term electricity storage, a balancing option for variable wind and photovoltaic power, and fuels for transportation (see, e.g. Pregger et al., 2019). Liebich et al. (2021) give an overview of the main production routes for synfuels from a variety of energy sources, locations, and transport options, as well as their ecological and economic advantages, disadvantages, and future prospects. A variety of technical concepts and test facilities are also available, ranging from PBtL based on biomass and hydropower in Sweden to PtL based on CO2 from cement production in Germany and PV power imports from Saudi Arabia. Because the costs are currently relatively high, the potential generation of synfuels and their use in the short term are not economically feasible, but they are primarily considered from the perspective of political expediency for the extensive decarbonization of the entire energy system.

The sustainable production of biofuels, even BtL, is limited by the availability of biomass feedstocks, e.g. residues and solid biomass. Research into and the development of the most-effcient generation routes for synthetic fuels are therefore very important, both for the decarbonization of the transport sector and for the security of future fossil-free fuel supply. Industry co-benefts can arise when emission reduction targets (e.g. for transport) lead to the accelerated development of synfuel production capacities. Synthetic fuel production processes can also provide the necessary feedstocks, such as methanol, for the chemical industries. The future availability of appropriate carbon sources from biomass or process-related emissions for industry is currently unclear, especially for biomass. The direct use of (solid) biomass in industry or the building sector might signifcantly reduce the remaining potential for biomass, and transport sectors (such as aviation and heavyduty traffc) might also compete for BtL.

Here, the specifc advantages of liquid synthetic hydrocarbons will also play a role. They require no special storage or transport containers, and losses during storage are negligible. The energy density is 100 times higher than today's batteries and 10 times higher than hydrogen at a pressure of 200 bar. Therefore, their handling, transport, and storage are much easier and safer, making the transport sector a major competitor for limited synfuel production. Signifcant improvements in development will be necessary along the complete process chain, which is as yet far from optimized.

# **References**


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

# **Chapter 10 Transition of the Energy Industry to (Net)-Zero Emissions**

### **Sven Teske**

**Abstract** The status quo in the global oil, gas, and coal industries in terms of their economic value, geographic distribution, and company structures is given. The current fossil fuel production volumes and decline rates required under 1.5 °C-compatible pathways for coal, oil, and natural gas are discussed. The assumptions made when calculating *scope 1* and *2* emissions and current and future energy intensities are defned. The role of power and gas utilities under the OECM 1.5 °C scenario is discussed, together with the projected trajectories for renewable power- and heatgenerating plants and those for hydrogen and synthetic fuel. Future structures of the global primary and secondary energy industries are suggested.

**Keywords** Global oil, gas, and coal industries · International production trajectory · Fossil fuel decline rates · 1.5 °C fossil fuel trajectories · Utilities

# **10.1 Introduction**

The Paris Agreement achieved consensus among all member states to maintain global warming well below 2 °C above pre-industrial levels while pursuing efforts to limit the increase to 1.5 ° C, because this will signifcantly reduce the risk and impacts of climate change (UNFCCC, 2015). The role of governments in implementing national climate targets and endeavours to reduce emissions at the country level is crucial for achieving global success. Considering the current concentration of CO2 in the atmosphere (416.2 parts per million) (US DC, 2021), a global effort is required to reduce emissions to as close as possible to zero while removing atmospheric CO2 by restoring ecosystems.

To reduce emissions to zero in line with a 1.5 °C increase, the use of coal, oil, and gas must be phased out by at least 56% by 2030. However, current climate debates have not involved an open discussion of the orderly withdrawal from the coal, oil, and gas industries. Instead, the political debate about coal, oil, and gas has

S. Teske (\*)

Institute for Sustainable Futures, University of Technology Sydney, Sydney, NSW, Australia e-mail: sven.teske@uts.edu.au

<sup>©</sup> The Author(s) 2022

S. Teske (ed.), *Achieving the Paris Climate Agreement Goals*, https://doi.org/10.1007/978-3-030-99177-7\_10

continued to focus on supply and price security, neglecting the fact that mitigating climate change is only possible when fossil fuels are phased out.

The fnance industry set various 'net-zero' targets in the run up to the Climate Conference COP26 in Glasgow in November 2021. One of these target setting organizations is the Net-Zero Asset Owner Alliance (NZAOA) (see Chap. 2). But what does this mean for the primary energy industry?

This section focuses on the fossil fuel trajectory of the OECM 1.5 °C pathways presented in this book and what it means for the primary energy industry and electricity and (natural) gas utilities to supply end users—customers from industry, services, or private households.

# **10.2 The Energy Industry: Overview**

Oil, gas, and coal are all hydrocarbons—combinations of hydrogen and carbon that originate in decomposed organic materials. Different combinations of heat and pressure—depending on geological conditions—create different forms of hydrocarbons: oil, gas, and brown or hard coal (NG, 2021).

Oil and gas often occur together, and with the proximity of both fossil resources, primary energy companies are active in oil *and* gas extraction. Geographically, the largest extraction companies for oil and gas are based in the USA, Saudi Arabia, and Russia, which were responsible for 43% of global production in 2020 (IEA OIL, 2021). By far the largest producer of coal is China, which contributed 53% of global production in 2018 (Statista, 2021c).

The geographic distribution of fossil fuels is also refected in the structure of the industry. In 2020, the top fve oil and gas companies were as follows: (1) China Petroleum & Chemical Corp. (SNP); (2) PetroChina Co. Ltd. (PTR); (3) Saudi Arabian Oil Co. (Saudi Aramco); (4) Royal Dutch Shell PLC (RDS.A); and (5) BP PLC (BP). Only Shell and BP are involved in some coal mining, whereas the top three companies focus on oil, gas, and related products for the chemical industry (IN, 2020).

The largest coal companies are BHP and Rio Tinto, both based in Australia, and China Shenhua Energy and have no or only relatively minor involvement in oil and gas extraction.

The Global Industry Classifcation System (GICS) category *10 Energy* includes all steps in the value chain for the production of primary energy from fossil fuels (oil, gas, and coal), from exploration and extraction to the refnement and processing of fuels as commodity for industry clients, such as the chemical industry and utilities.

Oil, gas, and coal are among the most fundamental commodities of the current global economy. Oil is not only used as a fuel for cars, planes, and ships but also as a commodity to produce, for example, asphalt, plastics, and a variety of other products. In response to the COVID-19 pandemic, the global market size for oil and gas exploration and production in 2020 was at a 10-year low, at US\$1.8 trillion, compared with US\$2.9 trillion in 2019. The market size a decade earlier (2011) was estimated to be US\$5.3 trillion (IBIS, 2021), more than twice as high as 2020. In comparison, the global market value of coal mining companies was US\$0.66 trillion in 2020 and US\$0.79 trillion in 2021 (Statista, 2021a), about half the value of US\$1.27 trillion a decade earlier (2011).

# *10.2.1 1.5 °C Pathway for the Primary Energy Industry*

The primary energy demand analysis—and therefore the projections for the primary energy industry and possible future operation strategies—is the product of the energy demand projections for all end-use sectors, as presented in previous chapters, and the energy supply concept. The challenge for the primary energy industry is to supply energy services for sustained economic development and a growing global population while remaining within the global carbon budget to limit the global temperature rise to 1.5 °C.

The trajectory for oil, gas, and coal depends on how quickly an alternative energy supply can be built up and how energy consumption can be reduced technically and/ or by behavioural changes. The OECM 1.5 °C pathway represents such a trajectory and is based on a detailed bottom-up sectorial demand and supply analysis, as documented in previous chapters.

However, for the primary energy industry, it is important to assess whether or not new oil, gas, or coal extraction projects are required to meet the demand, even under an ambitious fossil-fuel phase-out scenario.

A specifc analysis was undertaken in parallel with the development of the OECM 1.5 °C pathway and with scenario data from a previous version of the OneEarth Climate Model (OECM), published in 2019 (Teske et al., 2019). On the basis of publicly available oil, gas, and coal extraction data, future production volumes were calculated and compared with the 1.5 °C trajectory (Teske & Niklas, 2021).

The calculation was based on the assumptions that no new fossil fuel extraction projects would be developed from 2021 onwards and that all existing projects will see a production decline at standard industry rates. These assumptions are supported by the IEA Net-Zero by 2050 report, which concludes that there can be 'no new oil, gas or coal development if world is to reach net zero by 2050' (IEA NZ, 2021b).

A scenario designated the *Existing International Production Trajectory* ('no expansion') was developed and modelled, specifcally to understand what global fossil fuel production will look like under the following assumptions:


The *no expansion* scenario was compared with the OECM 1.5 °C pathway for coal, oil, and gas to understand whether security of supply is possible under an immediate implementation of a 'stop exploration' policy.

The decline rates for oil, gas, and coal that would result from the implementation of the 1.5 °C pathway and the assumed annual production decline rates for oil, gas, and coal are compared in Table 10.1.


Table 10.2 shows the modelled trajectories for global coal, oil, and gas production under the 1.5 °C scenario and the *no expansion* scenario. Projections beyond 2025 are extrapolated based on the fossil fuel production values for 2018 and 2019, taken from the BP Statistical Review and IEA World Energy Balances (IEA WEB, 2021).

The highest rates of overproduction are for hard coal and brown coal (lignite). On a global average, even existing mines cannot remain in operation until their resources are depleted, when calculations are made under the assumed production


**Table 10.1** Decline rates required to remain within the 1.5 °C carbon budget versus the production decline rates under 'no expansion'


**Table 10.2** Comparison of the 1.5 °C scenario and the *no expansion* scenario (excluding nonenergy use)

decline rates. No new mines need be opened to supply the remaining demand for coal.

The results for natural gas are less clear, and the production decline rates vary signifcantly. Shale gas production wells, in particular, have signifcantly higher production decline rates than conventional onshore or offshore natural gas extraction wells. The demand and supply values under a 1.5 °C scenario are similar, and a large overproduction of gas under the defned scenarios seems unlikely. However, a more detailed and production-side-specifc analysis is required. The demand and supply for oil on a global level are similar—meaning that the assumed average production decline rates for oil wells and the reduction in demand are in the same order of magnitude.

Our analysis shows that even with no expansion of fossil fuel production, the current productions levels—especially for coal—will exhaust the carbon budget associated with the 1.5 °C target before 2030. Without the active phase-out of fossilfuel production, production will signifcantly surpass what can be produced under a 1.5 °C scenario by 2025 onwards, for all fossil-fuel types.

The following section provides an overview of the breakdown of gross production, the losses during fuel processing, refnement, or the production of other fossil fuel products for hard coal, brown coal (lignite), gas, and oil. These parameters are required to calculate the *scope 1* and *2* emissions of the primary energy industry and are therefore documented. All parameters for the base year (2019) in Tables 10.3, 10.5, 10.6, and 10.8 are based on IEA World Energy Balances Statistics and projections under the OECM 1.5 °C pathway. Losses are calculated with statistical data from previous years and remain stable over the entire modelling period until 2050.

#### **10.2.1.1 1.5 °C Trajectory: Hard Coal**

The gross production of hard coal is the second highest for any fossil fuel, after oil. Around 35,000 PJ/yr of all the coal consumed globally is imported from other countries, or in other words, the coal consumed is not a regional energy resource. The main coal producers are China, Indonesia, and Australia. Interestingly, the largest importer of coal in 2019 was China, followed by India and the European Community (IEA Coal, 2020) (Table 10.3).

Table 10.4 shows the assumed losses in the coal industry and 'own energy uses', which are required for secondary projects, such as coking coal and coal liquifcation. The current shares of coal export and import for hard coal are also shown. All


**Table 10.3** Global coal trajectory—OECM 1.5 °C


**Table 10.4** Global coal production—assumptions for transport shares and technical losses in percent

**Table 10.5** Global lignite trajectory—OECM 1.5 °C


parameters are calculated on the basis of 2019 values and remain at the same level for the entire modelling period.

#### **10.2.1.2 1.5 °C Trajectory: Brown Coal**

Brown coal (or lignite) mines are in direct proximity to power plants, so the fuel is on-site and not exported. The use of brown coal is limited to fewer countries than that of hard coal (Table 10.5).

#### **10.2.1.3 1.5 °C Trajectory: Oil**

Crude oil is the largest single energy source globally. Its production is regionally concentrated, and more than 60% of all oil produced crosses borders between its production and consumption. In 2019, about 0.3% of the oil produced was consumed by the extraction process itself—generating part of the *scope 1* emissions of the oil industry—and another 1.7% was losses in refneries and other predictionrelated processes (Table 10.6).

Table 10.7 provides an overview of assumptions for transport shares and technical losses in percent as well as the specifc emissions that are assumed for the calculation of the *scope 1* emissions for oil production.


**Table 10.6** Global oil trajectory—OECM 1.5 °C



The assumed methane emissions are based on the IEA Methane Tracker (IEA MT, 2021). It is assumed that methane emissions will be reduced by 30% according to the *Global Methane Pledge* (EU-US, 2021), as announced at the Climate Conference COP26 in Glasgow, which has been supported by 44 countries (GMI, 2021) at the time of writing (December 2021).

#### **10.2.1.4 1.5 °C Trajectory: Natural Gas**

About one-third of all-natural gas produced crosses a national border between extraction and consumption (Table 10.8). The vast majority is transported via pipelines, which leads to a fractured world market with different prices, roughly broken down into the Americas, Europe, and the Middle East and Russia, as well as the Asia Pacifc Region, which is more focused on liquifed natural gas (LNG) transported by ships.

The share of gas faring in the total production is part of the *scope 1* emissions of the gas industry and is assumed to decrease from 4% currently to 2% in 2025 and to end by 2030, according to *Zero Routine Flaring by 2030* by the World Bank (ZRF, 2030).

Finally, the assumed shares for import and export and various losses, as well as the transport modes, for natural gas are shown in Table 10.9.


**Table 10.8** Global gas trajectory—OECM 1.5 °C


**Table 10.9** Global gas production—assumptions for transport shares and technical losses in percent

### **10.2.1.5 Global Renewables Trajectory (Power, Heat, and Fuels) Under OECM 1.5 °C**

The primary energy industry—oil and gas companies and coal companies—is at the crossroads. The fossil fuel demand and therefore its extraction must decline sharply to remain within the carbon budget. However, both the global population and the global economy are projected to increase over the next three decades. Therefore, the energy demand will remain high. Even under the ambitious energy effciency assumptions of the OECM 1.5 °C pathway, the global fnal energy demand will decrease by less than 10%. Therefore, the (primary) energy industry has an important role to play.


**Table 10.10** Global renewables trajectory—power, thermal and fuels—under OECM 1.5 °C

(\*) Data shows the natural gas reduction trajectory under OECM 1.5 °C in PJ/year

However, the way energy is produced must change and if primary energy companies transition to renewable energy—not just electricity but also heat and fuels the energy industry must move towards new business models that are closer to those of utilities, renewable project developers, and energy technology companies. Largescale renewable energy projects, such as offshore wind farms, are in regard to investment needs within the same order of magnitude as offshore oil and gas projects. The skill sets of the offshore oil and gas workforce can also be accommodated well within the offshore wind industry (see Box 10.2).

Table 10.10 shows the global renewable power, heat, and fuel generation requirements under the OECM 1.5 °C trajectory. The overall renewable energy intensity in petajoules (PJ) per billion \$GDP—is compared with the overall current coal energy intensity. Renewables will take over the role of coal in supplying the global economy with energy by around 2030. The overall renewable energy required to supply the needs of industry, services, transport, and buildings will reach the levels of oil, gas, and coal, at around 150,000 PJ/yr, between 2030 and 2035.

Only 5 years later, renewables will provide energy equal to the current contributions of oil and gas combined. Therefore, the potential new market opportunities for both the 'traditional' primary energy industry and utilities are signifcant, whereas the borders between the primary and secondary energy industries (= utilities) will start to blur.

# **10.3 Global Utilities Sector**

Power and gas utilities are a secondary energy industry. Until now, utilities have purchased (fossil) fuels from the primary energy industry and converted them to electricity in power plants or distributed the fuels—mainly gas—directly to customers to meet their demand for power and heat. Therefore, utilities are positioned between the primary energy industry and the end-use sector. Electricity generation is among the core businesses of utilities. Therefore, the signifcant increase in the electricity demand due to the electrifcation of transport and heat under the OECM 1.5 °C pathway can be seen as a business opportunity.

The global market for the generation, transmission, and distribution of electric power was estimated to be US\$3.2 trillion in 2020 (PRN, 2021). The 20 largest electric utilities had a cumulative market value of US\$686 million (Statista, 2021b). Market analysts expect a signifcant increase in the electricity demand (IEA EMR, 2021a; IRENA & JRC, 2021) (Fig. 10.1).

In a comparison of 14 global and regional energy scenarios, the International Renewable Energy Agency (IRENA) found that all projections agree that the demand for electricity will increase sharply:

Total global electricity generated in 2040 ranges from around 40,000 terawatt hours (TWh) in the IEA Sustainable Development Scenario (SDS) to nearly 70,000 TWh in the Bloomberg New Energy Outlook 2021 (BNEF NCS) where electricity generation grows two-and-a-half times from 2019 to 2040. This is due to electric vehicle uptake, electrifcation in industry and buildings and green hydrogen production. (IRENA & JRC, 2021)

The OECM 1.5 °C pathway will lead to an annual increase in electricity generation from about 26,000 TWh in 2019 to 76,000 TWh. Although there is clear agreement that the global electricity demand will increase, the predictions on how this electricity will be generated are very different. Despite the signifcant growth in renewable power generation during the last decade, short-term projections still expect that fossil-fuel-based power generation will continue to grow.

**Fig. 10.1** World's largest electric utility companies. (Statista, 2021c)

The Electricity Market Report of the International Energy Agency (IEA) expects that fossil-fuel-based electricity will provide 40% of the additional electricity demand in 2022 and that coal-fred power generation will jump back to 2019 levels after a 4.6% decline in 2020 (IEA EMR, 2021a). Therefore, the lead of renewable power generation is fragile.

# *10.3.1 Global Power and Natural Gas Utilities: Infrastructural Changes Under the 1.5 °C Scenario*

The assumed development of new manufacturing technologies, vehicle technologies, and building standards to achieve lower energy intensities for products and services has been presented in Part IV*—Sector-Specifc Pathways* (Chaps. 5, 6, 7, and 8). Power and gas utilities will be signifcantly affected by the suggested changes. Therefore, the business model must be adapted, as well as the operational organization, to supply secure electricity to all customers.

Throughout the description of the OECM 1.5 °C pathway in this book, the increased electrifcation of the transport and heating sectors is the overarching scenario narrative and runs across all sectors. Increased electrifcation will lead to 'sector coupling', i.e. the interconnection of the heating and transport sectors with the electricity sector. The sectors are still largely separate at the time of writing. However, the interconnection of these sectors offers signifcant advantages in terms of the management of the energy demand and the utilization of generation management with storage technologies. The synergies of sector coupling in terms of the infrastructural changes required to transition to 100% renewable energy systems are well-documented in the literature (e.g. Brown et al., 2018; Bogdanov et al., 2021; Bermúdez et al., 2021; Jacobson, 2020).

#### **10.3.1.1 Power Utilities**

Power utilities undertake three main tasks: power generation, the transmission and distribution of electricity, and electricity services. In countries in which the electricity market is liberalized, these tasks are separated and are performed by three independent (unbundeled) companies, for generation, transmission, and distribution. All three areas of responsibility will change signifcantly under the 1.5 °C scenario.

#### Power Generation

Fossil-based and nuclear power generation with average sizes of 500–1000 MW per production site require only a small number of power plants at few locations. Widely distributed solar photovoltaic generation, with an average size of 3–5 kW per system, will often be located at customers' premises or private homes, leading to thousands or even millions of decentralized power plants. Utilities and/or powerplant operators have access to a coal power plant for maintenance, for example, but decentralized power generation is different. Solar photovoltaic generators are usually neither owned by utilities, nor are they serviced by them in terms of technical maintenance, and utilities therefore have little infuence on the quantity of electricity generated or the time of generation. Electricity is also consumed partly locally and may not even reach the public power grid.

Offshore wind farms, in contrast, are centralized power plants with installed capacities within the range of an average conventional coal power plant and are usually not in direct proximity to the electricity demand. In contrast to oil and gas companies, utilities usually have no experience of working offshore, so the skills of the workforce must change.

#### Transmission

Under the 1.5 °C scenario, the power grid will change signifcantly over the next decade in response to three major changes: in the volume, load, and the location of generation.

First, the amount of electricity that must be transported will increase signifcantly. Electricity will replace fuels for heating and mobility, and the additional energy previously transported by other energy infrastructure, such as pipelines, will fow at least partly via power grids. End users—both private households that use heat and charge vehicles with electricity and industry clients—will not just increase the amount of electricity they use in kilowatt hours per outlet but also the loads required in kilowatts or even megawatts.

A home charger for Tesla vehicles, for example, operates at 230 V and 8–32 A, depending on the location and model (Tesla, 2021), resulting in a load of 1.8–7 kW. Therefore, the load of an average household will approximately double. The replacement of a coal-fred process (such as replacing a heating oven for steel production with an electric arc furnace) can increase the load by ≥300 MW.

Higher loads at the customer connection point and increased on-site generation will require a signifcantly stronger power grid. Furthermore, on days of higher wind and/or solar electricity production, electricity can 'reverse the fow'. With centralized power plants, the electricity is fed into the system—the transmission grid at high- or medium-voltage levels and is taken out at medium-voltage levels by industry customers and at low-voltage levels (from the distribution grid) by residential or commercial customers. Solar rooftop systems feed electricity in at a lowvoltage level. During times of high production, solar electricity can fow from a low-voltage level to a medium-voltage level, although this requires special transformer stations.

#### Electricity System Services

Like the sites of electricity in- and output, also the time of generation is not centrally managed by a power-plant operator (who would ramp up and down the power plant) but by a 'swarm' of solar electricity generators and onshore wind turbines, whose operation depends on the availability of sunshine and wind. Weather forecasts, the related power generated, and demand projections will be increasingly important for utilities and grid operators. The operation and management of decentralized storage systems, often operated by private households, must also be considered. Power grid operators are among the most important enablers of the energy transition, because an effcient and safe power grid is the backbone of the decarbonized energy industry.

#### Distribution

Widely distributed generation and storage capacities, in increasing proximity to the electricity demand, will change the relationship between the utility and the customer. The customer is no longer just a consumer but a 'prosumer'—a producer and consumer of electricity. Therefore, the business concept must change signifcantly, and utilities may see themselves competing with the electricity produced their own customers. A utility must increase its services and integrate local electricity generation (Table 10.11).

#### **10.3.1.2 Gas Utilities**

The changes in gas utilities under the OECM 1.5 °C scenario are more profound than those for power utilities, because the main product—natural gas—will be phased out globally by 2050. Tables 10.8 and 10.9 show the projected trajectories. However, the OECM acknowledges the signifcant value of the existing gas infrastructure and recommends that the gas distribution network be repurposed to utilize it for the future decarbonized energy supply. According to the Global Energy Monitor, 900,757 km of natural gas transmission pipelines were in operation globally at the end of 2020. Research has shown that there are no fundamental technical barriers to the conversion of natural gas pipelines for the transport of pure hydrogen.

Box 10.1 summarizes the key results of the comprehensive research project 'Repurposing Existing Gas Infrastructure: Overview of existing studies and refections on the conditions for repurposing' by the European Union Agency for the Cooperation of Energy Regulators (ACER), published in July 2021.


**Table 10.11** Renewable power, heat capacities, and energy demand for hydrogen and synthetic fuel production under the 1.5 °C scenario

Therefore, the OECM assumes the conversion of natural gas pipelines to transport hydrogen, either for direct use as a replacement for natural gas in (process) heating systems, as feedstock for chemical processes, or for energy storage purposes. Therefore, the calculation of *scope 1* and *2* emissions (Chap. 13) factors in a transition to hydrogen and synthetic fuels, with the provided factors repurposed for conversion losses (Table 10.9).

### **Box 10.1 Conversion of Existing Natural Gas Pipelines to Transport Hydrogen**

### **Key results from the research of the European Union Agency for the Cooperation of Energy Regulators (ACER,** 2021**)**

*Pipeline transport capacity: natural gas vs pure hydrogen—technical aspects*

	- (i) Inner coating to chemically protect the steel layer
	- (ii) Intelligent pigging (monitoring)
	- (iii) Operational pressure management (avoiding large pressure changes)
	- (iv) Admixing degradation inhibitors (e.g. 1000 ppm oxygen)

### *Transmission pipeline conversion*

The main advantages of repurposing pipelines are:


# *10.3.2 1.5 °°C Trajectory for Power and Gas Utilities*

Table 10.12 shows the development of the demand and supply of natural gas and electricity for the global utilities sector—including combined heat and power (CHP)—under the OECM 1.5 °C pathway. Figure 10.2 shows the signifcant increase—by a factor of 10—in the global generation of renewable electricity. The projected transition of gas utilities to the distribution of hydrogen and synthetic


**Table 10.12** Global utilities sector—electricity and gas distribution under the OECM 1.5 °C

**Fig. 10.2** Global power utilities sector—electricity under the OECM 1.5 °C scenario

**Fig. 10.3** Global gas utilities sector—gaseous fuel supply under the OECM 1.5 °C scenario

fuels will represent 50% of their sales by 2045. Therefore, the transition is assumed to have a lead time of about 10 years for the implementation of the required technical and regulatory changes (Fig. 10.3).

# **10.4 Energy and Utilities Sectors: A Possible Structure**

Of all the industries analysed, the energy industry—often referred to in this book as the primary energy sector—classifed as GICS *10 Energy*, will experience the most drastic changes. The decarbonization of the global *energy* sector requires the complete phase-out of fossil fuels in combustion processes to generate energy —the very core business of the energy industry.

The OneEarth Climate Model (OECM) 1.5 °C scenario assumes that 100% of the fossil-fuel-based energy supply will be replaced by renewable energy by 2050 complete transition within one generation—which is unprecedented in modern human history.

The purpose of this book is to document the development, calculation, and results of the OECM to provide benchmark key performance indicators for specifc industries. These will support target setting by the fnance industry and those who develop the net-zero targets and/or the National Determined Contributions (NDCs) required under the Paris Climate Agreement.

To develop new business concepts for industry sectors is not the task of this research. Instead, we aim to support our assumptions with technical details and scenario narratives, which have been discussed with the scientifc advisory board of the Net-Zero Asset Owner Alliance (NZAOA) (see Chap. 2).

The *energy* and *utility* sectors must grow together to implement the global energy transition within only three decades. Utility-scale solar power plants and onshore and offshore wind farms are large infrastructural projects that require investments in the range of several hundred millions to billions of dollars. The operation and maintenance of offshore wind farms are very similar to those of offshore oil and gas rigs. The transport and distribution of natural gas from the point of its extraction to the end user is the core business activity of (natural) gas utilities. Power utilities oversee the entire gamut of production, from generation to distribution. Based on the OECM decarbonization pathway described in this book, we propose a horizontal integration of all three sub-sectors, which integrates the core areas of expertise and avoids stranded assets by repurposing the existing fossil-fuel infrastructure, such as pipelines.

Figure 10.4 shows a possible structure for the decarbonized *energy* and *utility* sectors. The (primary) energy industry will focus on utility-scale power generation

**Fig. 10.4** One Earth Climate Model: possible structure of a decarbonized *energy* and *utilities* industries

and the production of hydrogen and synthetic fuels for the supply of energy and chemical feedstock. Gas utilities will focus on the transport of hydrogen and fuels and offer decentralized hydrogen production and storage services to the power sector. Power utilities will concentrate on the power grid, the management of the electricity system, and the integration of decentralized renewable power generation and storage systems, including those from 'prosumers'.

#### **Box 10.2 Occupational Match between Offshore Oil and Gas, and Offshore Wind energy (Briggs et al., 2021)**

Briggs et al. (2021) found that the main occupational pathways into offshore wind with are from other technically related sectors (such as offshore industries and the energy sector), as new entrant apprentices or graduates, and from a workforce with skills that cut across sectors (e.g. business/commercial, IT and data analytics, drone and underwater remotely operated vehicle (ROV) operators, etc.).

Consequently, the development of offshore wind energy could be an important source of alternative employment for the offshore oil and gas workforce.

A major UK study of offshore wind found that there are three main pathways for workers into the industry:


A range of studies have found signifcant movement has occurred from the offshore oil and gas workforce into the offshore wind industry, because these workers often have the foundation skills required to work on offshore installation vessels and offshore platforms and the specialized knowledge of the environmental challenges associated with operating and maintaining offshore infrastructure (IRENA, 2018).

A Scottish study found that only 15% of jobs in these industries have no skills match. For around two-thirds of jobs, there is a 'good' or 'some' skills match, including in many professional jobs, construction and installation, electrical and mechanical trades, technicians, and subsea pipelines. For a range of administrative, quality control, logistic, and project management jobs, there are 'partial' skills overlaps, which suggests that these workers could be transitioned with training (Fig. 10.5).

**Fig. 10.5** Occupational match between offshore oil and gas and offshore wind energy. (Friends of the Earth; Global Witness and Greener Jobs Alliance, 2019)

# **References**


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

# **Part VI Non-energy GHG and Aerosol Emissions**

# **Chapter 11 Climate Sensitivity Analysis: All Greenhouse Gases and Aerosols**

**Sven Teske**

**Abstract** This section provides an overview of all greenhouse gases (GHGs) and aerosols, the sources, their contributions to overall emissions, and their likely cumulative effects on global temperature increases. The non-energy GHG modelling in this chapter is an update of the probabilistic assessment of the global mean temperature published in the frst part of *Achieving the Paris Climate Agreements*, Chap. 12 (Meinshausen 2019). The 1.5 °C energy and non-energy pathways were assessed by Climate Resource—specialists in assessing the warming implications of emissions scenarios. The analysis focuses on the derivation of the trajectories of non-CO2 emissions that match the trajectories of energy and industrial CO2 emissions and evaluates the multi-gas pathways against various temperature thresholds and carbon budgets until 2100. (120).

Section 7.2 is based on the following: 'Documentation of 'UTS scenarios – Probabilistic assessment of global-mean temperatures' by Climate Resource Malte Meinshausen, Zebedee Nicholls, October 2021.

**Keywords** Non-energy GHG modelling · Agriculture, forestry, and other land use (AFOLU) emissions · N2O · CH4 · Global warming potential (GWP), Temperature projections and exceedance probabilities

# **11.1 Introduction**

In previous chapters, we focused on the *energy* sector and the role of land use in certain industry sectors. This section provides an overview of all greenhouse gases (GHGs) and aerosols, the cause of their emission, their contribution to overall

S. Teske (\*)

University of Technology Sydney – Institute for Sustainable Futures (UTS-ISF), Sydney, NSW, Australia e-mail: sven.teske@uts.edu.au

emissions, and the likely cumulative effect of global temperature increases. The non-energy GHG modelling in this chapter is an update of Meinshausen (2019). The major sources of non-energy-related emissions—process emission from cement, steel, and aluminium production—have been quantifed as part of the *industry* demand analysis (Chap. 5).

# **11.2 Overview: Greenhouse Gases and Aerosols (Substances, Origins, and Projected Development)**

# *11.2.1 Energy-Related CO2 Emissions*

Energy-related CO2 emissions all derive from oil, gas, or coal and are defned as '1A' emissions according to the Intergovernmental Panel on Climate Change (IPCC) 2006 guidelines (IPCC 2006) for the National Greenhouse Gas Inventory, shown in Fig. 11.1. These emissions are caused by combustion processes, such as those in power or heating plants, engines of cars, truck, planes, and ships, and any other use of fossil fuels that involves a combustion process. All the pre-2019 values are historical statistical data, whereas all the data points from 2020 onwards are the results of the 1.5 °C energy scenario documented in previous chapters.

#### **11.2.1.1 Fugitive CO2 Emissions**

According to the IPCC (2019), fugitive CO2 emissions can be broken down into energy- and industry-related emissions and are categorized as 'non-1A' emissions. Energy-related fugitive emissions are further subdivided into fugitive coal emissions from underground or surface mines, including the CO2 from methane (CH4) utilization or faring from underground coal mines.

Fugitive emissions from oil and natural gas include the products of unconventional oil and gas exploration, such as tar sand and fracking gas, and emissions from abandoned wells. Fugitive emissions also arise from fuel transformation processes, such as in oil refneries, charcoal and coking coal production, or gasifcation processes. Fugitive emissions constitute only a fraction of the emissions from energyrelated combustion processes. In our analysis of industry-specifc emissions, they are included in the *Scope 1*, *2*, and *3* emissions.

#### **11.2.1.2 Industrial Process Emissions**

The second category according to the IPCC guidelines (IPCC 2019) is emissions from industrial processes and product use. The main emissions in this group are non-energy-related CO2 from steel and cement manufacturing and include chemical

**Fig. 11.1** Main categories of emissions by source and their removal by sinks, as used by the IPCC. (Source: 2006 IPCC Guidelines for National Greenhouse Gas Inventories, p. 6, (IPCC 2006))

substances used in the chemical industry, in aluminium production, or as technical gases for refrigeration. Although the volume of these chemical substances is small, their global warming effect is often signifcant. Details are provided in Sects. 11.2.5 and 11.2.6.

#### **11.2.1.3 Black and Organic Carbon and Carbon Monoxide**

There are three other forms of carbon:


#### **11.2.1.4 Responsible Industry Sectors: CO2**

Energy-related CO2 emissions are obviously caused by all industry sectors that use fossil fuels. However, as reported in Chap. 4, the categorization of emissions into *Scopes 1*, *2*, and *3* helps defne the levels of responsibility for emissions and the extent to which emission can be reduced. The primary energy industry is responsible for the exploitation and mining of fossil fuels and for fuel transformation from, for example, crude oil to kerosene. Therefore, the primary energy industry directly infuences the potential reduction of fugitive emissions.

The conversion of fossil fuels into secondary energy, such as power and heat, and the transport of fuels to industrial, commercial, or private consumers are the responsibility of power, gas and energy utilities. Utilities only have a limited infuence on the overall energy demand but can reduce conversion losses, including in coal or gas power plants. Although the amount of CO2 released from burning a tonne of coal, a litre of oil, or a cubic metre of gas is constant and only varies across different qualities of fuel, the amount of secondary energy units (e.g. electricity) generated depends on the effciency of the power plant. The GHGs emitted for each kilowatthour of electricity can be reduced, although the overall emissions can only be reduced by reducing the use of fossil fuel itself.

Finally, the end-use sector of fossil-fuel-based energy is responsible for the actual total demand for fossil fuels. End users are not responsible for fugitive emissions or conversion losses in power plants but can lower CO2 emissions by using more-effcient end-use applications, such as energy-effcient cars, and by driving less. However, a total phase-out of energy-related CO2 emission is possible with the use of carbon-free renewable energy sources.

# *11.2.2 Agriculture, Forestry, and Other Land Use (AFOLU) Emissions*

In the climate science context, emissions from agriculture, forestry, and other landuses are referred to as *AFOLU* emissions. The AFOLU sector contributes to the emission of multiple GHGs and aerosol species, including CO2, CH4, and nitrous oxide (N2O). More details about AFOLU emissions and industry responsibilities are provided in Sect. 6.1 (Overview of the Global Agriculture and Food Sector) and Chap. 14. In the 1.5 °C pathway, the phase-out of AFOLU emissions by 2030 mainly by the cessation of deforestation and the introduction of negative emissions by the creation of carbon sinks with nature-based solutions, such as are-forestation and soil management—is vital. AFOLU emissions must decline sharply until 2030, in concert with the introduction of negative emissions and the absorption of CO2, between 2035 and 2100 (Fig. 11.9).

# *11.2.3 N2O Emissions*

The long-lived GHG N2O is emitted by human activities, such as fertilizer use, burning fossil and biofuels, and wastewater treatment (IPCC 2007 AR4). However, natural processes in soils and oceans also release N2O. More than one-third of all N2O emissions are anthropogenic and primarily derive from agriculture (IPCC 2007 AR4). In this analysis, we focus on human sources of N2O.

#### **11.2.3.1 Responsible Industry Sectors: N2O**

Of all GHG emissions, 6% are N2O. About 71% of all N2O emissions are caused by the use of synthetic and organic fertilizers in the agricultural sector. Of all N2O emissions, 15% are related to the chemical production of fertilizers, fbres, and synthetic products; around 10% are the by-products of combustion processes; and 4% arise from wastewater treatment plants (IPCC AR4).

# *11.2.4 CH4 Emissions*

Methane is a GHG with an estimated lifetime of 12 years. About 17% of all GHG emissions are CH4. Anthropogenic CH4 is predominantly emitted from manure and as gastroenteric releases from livestock; from rice paddies; as fugitive emissions from the mining of coal, oil, and gas; and in gas transport leakages. There are also natural sources CH4, such as gas hydrates, freshwater bodies, oceans, termites, and wetlands, and other sources such as wildfres. Globally, wetlands are the largest natural source of CH4, with emissions estimated to be 102–200 Mt./year on average in 2008–2017, which constituted approximately one-quarter of global CH4 emissions (UNEP 2021).

### **11.2.4.1 Responsible Industry Sectors: CH4**

Anthropogenic sources contribute to about 60% of total global CH4 emissions, 90% of which come from only three sectors: 40% from the fossil-fuel industry, approximately 35% from the agriculture sector, and approximately 20% from waste and landfll utilities (UNEP 2021).


The remaining CH4 emissions are mainly from wastewater treatment (UNEP 2021).

# *11.2.5 Other GHGs*

Although CO2, CH4, and N2O are the main GHG gases, representing approximately 90% of all GHG emissions, a large number of other GHGs and aerosol precursors are emitted, including substances used as feedstock for the chemical industry, such as ammonia, or chemical substances used for technical processes. The largest group of these chemical substances is controlled by the Montreal Protocol, which phases down the consumption and production of different ozone-depleting substances, including halons, chlorofuorocarbons (CFCs), and hydrofuorocarbons (HFCs) (UNEP MP 2021).

# *11.2.6 Global Warming Potential (GWP)*

Greenhouse gases warm the earth by trapping energy and reducing the rate at which energy escapes the atmosphere. These gases differ in their ability to trap heat and have various radiative effciencies. They also differ in their atmospheric residence


**Table 11.1** Main greenhouse gases and their global warming potential (GWP)

*Source*: IPCC AR4, compilation by the Climate Change Connection, Manitoba/Canada Note: GWP values were changed in 2007. The values published in the 2007 IPCC Fourth Assessment Report (*AR4*) were refned from the IPCC Second Assessment Report (*SAR*) values. However, both values (AR2 and AR4) can be found throughout the literature

times. Each gas has a specifc global warming potential (GWP), which allows comparisons of the amount of energy the emission of 1 tonne of a gas will absorb over a given time period, usually a 100-year average time, compared with the emissions of 1 tonne of CO2 (Vallero 2019). Table 11.1 shows the main GHGs and their GWPs. Although the quantities of substances considered under the Montreal Protocol are small, their GWPs are signifcantly higher than those of the main GHGs.

# **11.3 Assessment of the 1.5 °C Energy and Non-Energy Pathways**

This section is based on the analysis of *Climate Resource* under contract to the University of Technology Sydney (UTS) as part of the Net-Zero Sectorial Industry Pathways Project (UTS/ISF 2021). The study is an update of the previous OneEarth Climate Model (OECM) publication (Teske et al. 2019). However, the Generalized Quantile Walk (GQW) methodology used (Meinshausen & Dooley 2019) has been developed further.

The energy and industrial CO2 emissions pathways are based on the OECM 1.5 °C energy scenario described in previous chapters, whereas the non-CO2 GHG emission time series have been described with the advanced GQW methodology.

The probabilistic global mean temperature, radiative forcing, and concentration implications of the scenarios are also examined with the reduced complexity model MAGICC, in the same set-up used by the IPCC's Sixth Assessment Report (IPCC AR6 2021). The emissions pathways developed are analysed in terms of their 1.5 °C, 2 °C, and 2.5 °C exceedance probabilities over time until 2050 and 2100. The climate projections are performed with a probabilistic modelling set-up that includes additional feedbacks, such as permafrost-related CH4 and CO2 emissions.

# *11.3.1 Accounting for Non-Energy Sectors*

The IPCC Assessment Report 6 (IPCC AR 2021), published in August 2021, contains fve scenarios, each of which represents a different emissions pathway. These scenarios are called the *Shared Socioeconomic Pathway* (*SSP*) scenarios. The most optimistic scenario, in which global CO2 emissions are cut to net zero around 2050, is the SSP1-1.9 scenario. The number at the end (1.9) stands for the approximate end-of-century radiative forcing, a measure of how hard human activities are pushing the climate system away from its pre-industrial equilibrium. The most pessimistic is SSP5-8.5. The SSP1-1.9 scenario, described in detail by Rogelj et al. (2018), assumes that the global community takes strong mitigation action consistent with the sustainable development goals. As a result, this scenario sees strong reductions in GHG emissions.

In this analysis, the energy-related CO2 emissions data are the results of the OECM 1.5 °C pathway documented in previous chapters. These sectorial energy scenarios include key fossil-fuel combustion activities, as defned under category 1A emissions of the IPCC 1996 guideline defnitions (IPCC 2006), shown in Fig. 11.1.

*Climate Resource* has added CO2 emissions that fall under other fossil fuel and industrial activities, such as fugitive emissions, cement production, and waste disposal and management, from the SSP1-1.9 scenario, a scenario in which there is strong mitigation action. The SSP1-1.9 scenario has been chosen, because it has similar reductions of CO2 fossil-fuel emissions as the OECM 1.5 °C scenario. The combination of both time series creates an emission pathway that is likely to include all fossil-fuel and industrial uses.

Within the energy sector category, the non-1A category emissions are those that derive from fugitive emissions and fossil-fuel fres. In adding these emissions, we assume that they will remain constant into the future, and we derive their magnitude based on the detailed sectorial breakdown provided by Hoesly et al. (2018). The data categorization follows the latest scientifc standards (Nicholls et al. 2021; Gidden et al. 2019).

The assumption that the non-1A emissions—industrial and fugitive emissions will remain constant is an oversimplifcation, given the likelihood that changes (such as faring during gas production) will vary into the future. With a complete fossil-fuel phase-out, there will be no further natural gas extraction and therefore no emissions from gas faring. However, these emissions represent <1% of total CO2 emissions, so the effect of this simplifcation will be of the order of hundredths of a degree, even in a baseline scenario. We chose not to assume that these emissions

will continue to represent a fxed fraction of the total energy sector emissions, because the energy sector emissions will become negative in the twenty-frst century under the SSP1-1.9 scenario and negative emissions from fossil-fuel fres seem highly unlikely.

In the industry sector, the non-1A category emissions mainly derive from cement and metal production. We assume that these emissions will represent a fxed fraction of the *industry* sector emissions, with the fxed fraction varying by region. We derive the fxed fraction from the ratio of non-1A category emissions in the *industry* sector to the total emissions in the *industry* sector in 2014 in the data of Hoesly et al. (2018). This assumption is once again a simplifcation. However, in the absence of other data sources, it is a simple and justifable choice. Moreover, given that these emissions represent approximately 6% of the total emissions and that the fxed fraction assumption captures at least some of the underlying scenario dynamics, we expect the effect of this assumption to be limited to the order of a few hundredths of a degree centigrade.

We combined the 1A CO2 emissions of the OECM with the estimate of non-1A emissions to create a complete time series of fossil CO2 emissions (see Fig. 11.2). Whereas the 1A emissions from the OECM will reach zero in 2050, the non-1A sector emissions are generally considered harder to mitigate, so we assume that they will not reach zero in 2050.

**Fig. 11.2** Three non-1A fossil CO2 emissions (Reference, 2.0 °C) and the OECM 1.5 °C pathway

The analysis performed above suggests there will be a small non-zero amount of emissions from these non-1A sectors in 2050, even under an ambitious mitigation scenario. As a result, the total fossil CO2 emissions will generally follow the trajectory provided by the OECM but will be slightly higher, because the non-1A sector emissions are included, and in 2050, the total fossil CO2 emissions will be close to, but not equal to, zero. Creating a scenario in which they reach exactly zero by 2050 would require further analysis of these non-1A sectors. Figure 11.2 shows the inclusion of non-1A fossil CO2 emissions. The data for the additional scenario represent the reference case and the 2.0 °C scenario published by Teske et al. (2019).

# *11.3.2 Harmonization*

In a second step, the projected emissions are harmonized to historical emissions estimates of the Global Carbon Project 2020 (GCP 2021a). To estimate the rebound after the COVID-related reduction in emissions in 2020, we assume that the 2021 emissions will rebound to their 2019 levels, within the same level estimated by the International Energy Agency (IEA PR 2021). This ensures a smooth transition between historical emissions and the three projections—a Reference case, a 2.0 °C scenario (see Teske et al. 2019) and the OECM 1.5 °C—as well as capturing the impact of COVID and the subsequent recovery efforts. The impact of harmonization on each of the OECM scenarios is illustrated in Fig. 11.3.

**Fig. 11.3** Harmonization of fossil CO2 emissions with historical emissions from the Global Carbon Project 2020 (GCP 2021b)

**Fig. 11.4** Extending fossil CO2 emissions from 2050 to 2100

# *11.3.3 Extending Emissions to 2100*

A simple approach is taken to extending emissions to 2100. This process is also called 'inflling'. For the mitigation scenarios OECM 2.0 °C and OECM 1.5 °C, fossil CO2 emissions are simply held constant from 2050 to 2100. For the reference scenarios, fossil CO2 emissions are extended forward by assuming that the emissions follow the evolution of other pathways at a similar level of emissions in 2050. This process has been undertaken with the Silicone Software (Lamboll et al. 2020). The other pathways are taken from the SR1.5 database, i.e. the scenarios that underpinned the IPCC's Special Report on 1.5 °C (Huppmann 2018).

The SR1.5 scenarios are, at the time of writing, the most comprehensive set of strong mitigation scenarios available in the literature. Consequently, they provide the best basis for statistical inferences on how emissions will evolve over time (e.g. as we have done here by inferring the post-2050 emissions based on the emissions in 2050) and how the evolution of one set of emissions (e.g. fossil CO2) is linked to changes in other sets of emissions (e.g. CH4) (Fig. 11.4).

# *11.3.4 Inflling Emissions Other Than Fossil CO2*

#### **11.3.4.1 Emissions in the SR1.5 Database**

The OECM 1.5 °C fossil CO2 time series is inflled with non-fossil-fuel CO2 emissions from the SR1.5 database, whose targets are similar to the OECM 1.5 °C emissions trajectory (Fig. 11.5). This method examines the relationship between fossil

**Fig. 11.5** Inflled emissions time series compared with the SR1.5 scenario database

CO2 based on the OECM 1.5 °C pathways and other emissions of the SR1.5 database. This process requires the re-harmonization of the SR1.5 database to match the historical emissions inputs used by MAGICC v7.5.3 in the probabilistic AR6 set-up in 2015. This ensures that all-time series start from a consistent point, so there are no spurious jumps in the complete emissions time series, which are then passed to the climate model MAGICC (see Sect. 11.3.2 and Gidden et al. 2018).

In Fig. 11.9, the four scenarios analysed (thick lines) are shown in the context of the international integrated assessment model (IAM) scenarios, shown with blue thin lines, which represent 411 scenarios taken from the IPCC Special Report on the 1.5 °C warming scenario database. We show the OECM-modelled fossil and industrial CO2 emissions (top left), the inferred CO2 land-use (AFOLU)-related emissions (panel top right), inferred total CH4 emissions (panel middle left), inferred total N2O emissions (panel middle right), inferred total CF4 emissions (panel bottom left), and inferred total C2F6 emissions (panel bottom right).

#### **11.3.4.2 Emissions Not in the SR1.5 Database**

In addition to the SR1.5 database emissions, as described in the previous section, emissions from the SSP scenarios—see defnition in Sect. 11.3.1—are introduced into the analysis, as shown in Fig. 11.6.

**Fig. 11.6** Inflled emissions compared with the SSP scenarios

The SSP emissions pathways chosen for the analysis are those that are closest to the OECM 1.5 °C pathway. The SSP scenarios were selected using the root mean square (RMS) methodology, which measures closeness based on the difference in emissions for gases that have similar applications and uses.

A scenario for the extremely potent GHG octafuoropropane (C3F8) emissions, for example, was chosen based on its similarity to C2F6 emissions, which is a simplifed way of inferring the appropriate emissions.

Hexafuoroethane (C2F6) is, like C3F8, a substance used in the semiconductor industry. However, this pragmatic technique is appropriate because the climate impact of these species is minor, representing <10% of the total GHG emissions.

In Fig. 11.10, the four scenarios analysed are shown in thick lines in the context of the SSP scenarios, which are marked in thin blue lines, and represent specifc SSP scenarios (O'Neill et al. 2016). As examples, C2F6 emissions are shown (panel top left) as well as C3F8 emissions (panel top right), which follow from the C2F6 emissions, together with CF4 emissions (panel bottom left) and CFC11 emissions (panel bottom right), which follow from the CF4 emissions.

Carbon tetrafuoride (CF4) and trichlorofuoromethane (CFC11) are both substances used in refrigeration.

# *11.3.5 Temperature Projections and Exceedance Probabilities*

Here, we provide the global mean probabilistic temperature projections, including their medians and 5%–95% ranges, for the OECM scenarios analysed (Fig. 11.7). These probabilistic ranges are sourced from the underlying 600 ensemble members, which are calibrated against the IPCC AR6 WG1 fndings.

**Fig. 11.7** Probabilistic global mean surface air temperature (GSAT) projections relative to 1850–1900

Similar to the SSP1-1.9 scenario in IPCC AR6 WG1, both the OECM 2.0 °C and OECM 1.5 °C pathways slightly overshoot the 1.5 °C pathway in their medians during the middle of the century, before dropping back to below 1.5 °C warming towards the end of the century. These probabilistic temperature projections can also be converted into exceedance probabilities (Fig. 11.8), i.e. the likelihood of exceeding a given temperature threshold at each point in time.

Both the OECM 2.0 °C and OECM 1.5 °C pathways are characterized as 1.5 °C low-overshoot pathways, i.e. pathways that end up below 1.5 °C (with a greater than 50% chance) at the end of century but slightly exceed a 50% chance of 1.5 °C over the course of the century. Both pathways are consistent with what is referred to in the SR1.5 report as '1.5 °C-compatible pathways'. However, the likelihood that the OECM 1.5 °C scenario will stay below 1.5 °C throughout the century, despite strong mitigation actions, does not exceed 67%. Figure 11.7 shows the probabilistic global mean surface air temperature (GSAT) projections relative to 1850–1900 for the scenarios analysed.

# **11.4 One Earth Summary Graph**

The OECM 1.5 °C mitigation scenario limits the global average temperature rise to 1.5 °C using a carbon budget of 400 GtCO2 in cumulative emissions, commencing in January 2020, as defned in the IPCC's Sixth Assessment Report, Working Group 1 (AR6 2021).

**Fig. 11.8** Exceedance probabilities for the analysed scenarios relative to 1.5 °C, 2 °C, 2.5 °C, and 3 °C warming until 2100

The OECM calls for net-zero emissions by 2040, achieved by:

	- (a) Approximately 400 GtCO2 of additional carbon to be removed by reforestation and land restoration by 2100.
	- (b) Natural land carbon sinks to absorb CO2 but which will decline in the second half of the century.
	- (c) Natural ocean carbon sinks, which will continue to absorb CO2 throughout the century.

The IPCC AR6 presents the 400 GtCO2 carbon budget as providing a 'good' (67%) chance of limiting warming to 1.5 °C, but it does not incorporate the anthropogenic emissions that occurred between the pre-industrial era (1750–1800) and the early industrial era (1850–1900).

If historical emissions between 1750 and 1900 are included, a 400 GtCO2 carbon budget provides a 'fair' (50%) chance of an increase of 1.5 °C. In this case, a 'good' chance to achieve 1.5 °C warming would require an even steeper decline in emissions—net zero by 2040, instead of 2050—with the possibility of achieving 1.4 °C by 2100.

**Fig. 11.9** Probability of remaining under 1.5 °C. (Source: Creative Commons: Karl Burkart, One Earth)

Figure 11.9 shows the reduction of energy-related CO2 emissions (black), the removal of carbon by reforestation and land restoration (yellow), the natural land carbon sinks (green), and ocean carbon sinks (blue).

# **References**

GCP. (2021a). *The global carbon project*. Online database.


Emissions Data System (CEDS). *Geoscientifc Model Development, 11*, 369–408. https://doi. org/10.5194/gmd-11-369-2018


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

# **Part VII Results**

# **Chapter 12 OECM 1.5 °C Pathway for the Global Energy Supply**

#### **Sven Teske and Thomas Pregger**

**Abstract** This chapter summarizes all the calculated energy demands for the industry, service, transport, and building sectors. The supply side results for the OECM 1.5 °C scenario are documented. Electricity generation and the power generation required globally are provided by technology, together with the corresponding renewable and fossil energy shares. A detailed overview of the heat demand by sector, the heat temperature levels required for industrial process heat, and the OECM 1.5 °C heat supply trajectories by technology are presented, in both total generation and installed capacities. The calculated global fnal and primary energy demands, carbon intensities by source, and energy-related CO2 emissions by sector are given. Finally, the chapter provides the global carbon budgets by sector.

**Keywords** Global electricity generation · Final electricity demand · Power plant capacities · Heat generation capacities · Final energy demands of energy-intensive industries · Global carbon budget

# **12.1 Introduction**

The fnal energy demands for the industries, services, transport, and buildings sectors, including residential buildings, were determined based on the assumed global population and economic development until 2050 (for details see Chap. 2), within the context of increased energy effciencies across all sectors. All supply scenarios were developed on the basis of a global carbon budget of 400 GtCO2 between 2020 and 2050, in order to qualify as an IPCC Shared Socioeconomic Pathway 1 (SSP1) no- or low-overshoot scenario (IPCC 2021).

The supply side of this 1.5 °C energy scenario pathway builds upon modelling undertaken in an interdisciplinary project led by the University of Technology

#### T. Pregger

S. Teske (\*)

University of Technology Sydney – Institute for Sustainable Futures (UTS-ISF), Sydney, NSW, Australia e-mail: sven.teske@uts.edu.au

German Aerospace Center (DLR), Institute for Engineering Thermodynamics (TT), Department of Energy Systems Analysis, Stuttgart, Germany

Sydney (UTS). The project modelled sectorial and regional decarbonization pathways to achieve the Paris climate goals—to maintain global warming well below 2 °C and to 'pursue efforts' to limit it to 1.5 °C. That project produced the OneEarth Climate Model (OECM), a detailed bottom-up examination of the potential to decarbonize the energy sector. The results of this ongoing research were published in 2019 (Teske et al. 2019), 2020 (Teske et al. 2020), and 2021 (Teske et al. 2021). For this analysis, the 1.5 °C supply scenario has been updated to match the detailed bottom-up analysis documented in Chaps. 5, 6, 7, and 8.

# **12.2 OECM 1.5 °C Pathway for the Global Electricity Supply**

The global electricity demand has grown continuously over the past decades. Global electricity generation more than doubled over the past 30 years, from 12,030 TWh (IEA WEO 1994) in 1991 to 26,942 TWh in 2019 (IEA WEO 2020). The COVID-19 pandemic led to a small reduction of about 2%, or 500 TWh (IEA WEO 2020), equal to Germany's annual electricity demand in 2020. The decline in demand was due to lockdowns and the consequent reductions in industrial manufacturing and services. However, the electricity demand increased again to pre-COVID levels in 2021. Increasing market shares of electric vehicles also increased the electricity demand in the transport sector globally. The OECM 1.5 °C pathways will accelerate this trend and the electrifcation of the transport sector and the provision of space and process heat to replace fossil fuels will continue to increase the global electricity demand.

# *12.2.1 Global Final Electricity Demand*

Figure 12.1 shows the development of the fnal electricity demand by sector between 2019 and 2050. The signifcant increase in the demand is due to the electrifcation of heat, for both space and process heating, and to a lesser extent for hydrogen and synthetic fuels. The overall global fnal demand in 2050 will be 2.5 times higher than in the base year, 2019. In 2050, the production of fuels alone will consume the same amount of electricity as the total global electricity demand in 1991. Therefore, the demand shares will change completely, and 47% of all electricity (Fig. 12.2) will be for heating and fuels that are mainly used in the industry and service sectors. Electricity for space heating—predominantly from heat pumps—will also be required for residential buildings.

**Fig. 12.1** Electricity demand by sector under the OECM 1.5 °C pathway in 2019–2050

**Fig. 12.2** Electricity demand shares by sector under the OECM 1.5 °C pathway in 2019 and 2050

# *12.2.2 Global Electricity Supply*

Just as the electricity demand has changed over the past three decades, the global electricity supply has changed signifcantly in the same period. In 1994, 63% of electricity was generated from fossil fuels, 19% from hydropower, and 17% from nuclear power (UN 1996). Since 2010, the share of renewable electricity has increased every year. By the end of 2019, renewables contributed 27.3%, and by 2020, the share was expected to have increased to 29%. For the second consecutive year, electricity production from fossil fuels was estimated to have declined, driven mainly by a 2% reduction in coal-based power generation (REN21 GSR 2020).

The global installed capacity, by power plant technology and as a whole, has also changed rapidly. In 2010, just under 50% of all new annual additions to powergenerating capacities were renewables, and 10 years later, this share had risen to 83%. Since 2012, net additions of renewable power generation capacity have outpaced the net installation of both fossil fuel and nuclear power capacity combined (REN21 GSR 2020). With the cost competitiveness achieved by renewables (mainly solar photovoltaic PV and wind power), this trend is expected to continue. China continues to be the world leader in bringing new renewable power generation on line, and the country contributed nearly half of all renewables-based installations in 2020.

In 2020, 256 GW of new renewable power generation capacity was added globally, leading to a total capacity of 1668 GW, or 2838 GW when hydro power is included (REN21 GSR 2020). By the end of 2020, the combined capacity of all solar photovoltaic installations was 760 GW, and wind power capacity summed to 743 GW. By comparison, the total global power generation capacity was 7484 GW, 2124 GW of which was from coal power plants, 1788 GW from gas power plants, and 415 GW from nuclear power plants (IEA WEO 2020). Thus, the trend in global generation is clearly in favour of cost-competitive new solar PV and wind power.

Table 12.1 shows the development of the projected global electricity generation shares. Under the OECM 1.5 °C pathway, coal- and lignite-based power plants will be phased out frst, followed by gas power plants as the last fossil-fuelled powergeneration technology to be taken out of service after 2040. Renewable power plants, especially solar photovoltaic and onshore and offshore wind, are projected to have the largest growth rates, leading to a combined share of 70% of electricity generation globally by 2050. To fully decarbonize the power sector, the overall renewable electricity share will increase from 25% in 2019 to 74% in 2030 and to 100% by 2050.

Global power plant capacities will quadruple between 2019 and 2050, as shown in Fig. 12.3. Capacity will increase more than actual power generation, because the


**Table 12.1** Global electricity supply shares under the OECM 1.5 °C pathway

**Fig. 12.3** Global installed power plant capacities under the OECM 1.5 °C pathway in 2019–2050

capacity factors for solar photovoltaic and wind power are lower than those for fuelbased power generation. By 2030, solar photovoltaic and wind will make up 70% of the generation capacity, compared with 15% in 2019, and will clearly dominate by 2050, with 78% of the total global generation capacity.

To implement the generation capacity required for the OECM 1.5 °C pathway, the global annual market for solar photovoltaic must increase from 139 GW per year (market in 2020) to an average of 800–1000 GW *additional* capacity per year from 2025 onwards to 2040. Thereafter, the overall additional annual capacity will decrease to under 100 GW to reach the required capacity calculated for 2050. However, solar photovoltaic is likely to remain at an annual market level of around 1000 GW—predominantly to provide the replacement capacity for plants that reach the end of their lifetimes after 25–30 years.

The annual market for onshore wind must increase from 87 GW in 2020 to 134 GW in 2025 and 373 GW in 2035. The total onshore wind capacity will continue to rise by 250 GW per year until 2050—including repowering. The annual onshore wind market is therefore likely to stabilize at around 300 GW per year. The size of the annual offshore wind market must increase from 6 GW in 2020 to 47 GW in 2025 to implement the 1.5 °C pathway and to grow further to around 100 GW per


**Table 12.2** Global power plant capacities—annual changes under the OECM 1.5 °C pathway

year throughout 2050, with increasing market shares for repowering after 2040 (Table 12.2).

However, fossil-fuel-based power generation must be decommissioned and the global total capacity will not increase over current levels but will remain within the greenhouse gas (GHG) emissions limits. By 2025, global capacities of 63 GW from hard coal plants and 55 GW from brown coal power plants must go offine. All coal power plants in the Organization for Economic Cooperation and Development (OECD) must cease electricity generation by 2030, and the last coal plants must fnish operation globally by 2045 to remain within the carbon budget for power generation to limit the global mean temperature increase to +1.5 °C. Specifc CO2 emission per kilowatt-hour will decrease from 509 g of CO2 in 2019 to 136 g by 2030 and 24 g in 2040 to be entirely CO2 free by 2050 (see Table 12.1, last row).

# **12.3 OECM 1.5 °C Pathway for Global Space and Process Heat Supply**

Analogous to electricity, the energy demand for space and process heat has been determined for the industry and service sectors and for residential and commercial buildings. The specifc value for each sub-sector, such as the steel and aluminium industries, has been documented in Chaps. 5, 6, and 7. In this section, we focus on the cumulative heat demand and the supply structure required for the two main sectors, *service and buildings* and *industry*.

Services and buildings usually do not require temperatures over 100 °C. Therefore, the supply technologies are different from those of the industry sector, which requires temperature up to 1000 °C and above. The overall fnal heat demand will increase globally under the OECM 1.5 °C pathway, but the demand shares will change signifcantly. With energy effciency measures for buildings (see Chap. 7), the overall space heating demand will decrease globally, even with increased foor space. However, the industrial process heat demand is projected to increase, because energy effciency measures will not compensate for the increasing production due to the expected increase in global GDP to 2050. In 2019, the industry sector consumed 43% of the global heat demand and the service and buildings sector the remaining 57%. By 2050, these shares will be exchanged and the industry sector will consume close to 60% of the global heat demand (Fig. 12.4).

Table 12.3 shows the supply structure for the *services and buildings* and *industry* sectors. District heat is projected to remain the smallest part of the global heat supply, followed by cogeneration. Direct heating systems installed on-site will continue to supply the majority of the heat demand. The most important technologies required to implement the OEM 1.5 °C pathway for buildings will be heat pumps and solar thermal heating for buildings, while on-site generation for industry will allow the transition from fossil-fuel-based heating plants to electrical systems, such as electric resistance ovens, electric arc furnaces, and, to a lesser extent, bioenergy or synthetic-fuel-based heating plants.

Cogeneration plants for buildings and the service sector will decline as the onsite heating demand decreases with increased effciency. For industry, cogeneration will remain an alternative and slightly increase overall generation. However, cogeneration requires fuel, and after the phase-out of fossil fuels, only biofuels, hydrogen, or synthetic fuels will be an option for CO2-free operation. The limited sustainable potential for bioenergy-based fuels and the relatively high costs of synthetic fuels will allow only minor growth of cogeneration plants or heating plants for the industry sector.

**Fig. 12.4** Electricity demand shares by sector under the OECM 1.5 °C pathway in 2019 and 2050


**Table 12.3** Global heat demand by sectors under the OECM 1.5 °C pathway

To develop the 1.5 °C pathways for process heat based on a renewable energy supply, it is necessary to separate the temperature levels for the required process heat, because not all renewable energy technologies can produce high-temperature heat. Whereas the heat generation for low-temperature heat can be achieved with renewable-electricity-supplied heat pumps or solar collectors, temperatures over 500 °C are assumed to be generated predominantly by combustion processes based on bioenergy up until 2030. After 2030, the share of electric process heat from electric resistance heat and electric arc furnaces is projected to increase to replace fossil fuels. Hydrogen and synthetic fuels will also play increasing roles in supplying high-temperature process heat.

The OECM model differentiates four temperature levels: low (<100 °C), medium low (100–500 °C), medium high (500–1000 °C), and high (>1000 °C).

Figure 12.5 shows the development of the industry process heat demand by temperature level. Whereas the values will increase over time despite energy effciency measures, the shares of the temperature levels will remain constant. This arises from the assumption that all industry products will increase with the assumed development of the global GDP. No replacement of products, e.g. cement produced with alternative materials, is assumed because this was beyond the scope of this research.

Table 12.4 shows the total process heat demand by temperature level for three major industries combined: aluminium, steel, and chemicals. The overall heat demand of these sectors represented 20% of the global heat demand in 2019. This share is projected to increase to 37% due to a signifcant reduction in the heat demand in the building sector (see Chap. 7). The steel and chemical industries had similar process heat demands in 2019, at 13 EJ/year and 12 EJ/year, respectively. In contrast, the process heat demand of the aluminium industry was a quarter of this, at 3 EJ/year. Most of the process heat required by the aluminium industry is

**Fig. 12.5** Global: industry process demand by sector under the OECM 1.5 °C pathway in 2019–2050

**Table 12.4** Total global process demand, by temperature level, in the aluminium, steel, and chemical industries


high-temperature heat (72%), whereas the iron ore and steel industry require only 57% high-temperature heat. The majority of the process heat required by the chemical industry is in the medium–high level (48%), between 500 °C and 1000 °C (Table 12.4).

# *12.3.1 Global Heat Supply*

The process heat supply in 2019 relied heavily on fossil fuels (83%), mainly coal (33%) and gas (36%). Renewables played a minor role and the majority of renewable process heat was from biomass. To increase the renewable energy shares especially for high-temperature heat—is more challenging than for the electricity sector. The fuel switch from coal and gas to biomass requires fewer technical changes than a transition towards geothermal energy, all forms of heat pumps, or direct electricity use (Keith et al. 2019). However, the OECM assumes that the global limit for sustainable biomass is around 100 EJ per year (Seidenberger et al. 2008). The generation of high-temperature heat requires concentrated solar thermal plants. However, solar thermal process heat is limited to low temperatures in most regions, because concentrated solar plants require direct sunlight with no cloud coverage, and can therefore only operate in the global sunbelt range in most regions (Farjana et al. 2018). Therefore, it is assumed that process heat will increasingly derive from electricity-based technologies: heat pumps, for low-temperature levels, and direct resistance electricity and electric arc furnaces, for medium- and hightemperature levels. However, to adapt appliances to generate electricity-based process heat will require signifcant changes in the production process. A signifcant increase in this technology is assumed to be unavailable before 2025 but will increase rapidly between 2026 and 2030. Hydrogen and synthetic fuels produced with renewable electricity will increase after 2030, especially for processes that cannot be electrifed.

A global phase-out of coal for heat production is a priority objective to reduce specifc CO2 emissions. To replace fuel-based heat production, electrifcation, especially for low- (<100 °C) and medium-level (100–500 °C) process heat, is extremely important in achieving decarbonization.

Table 12.5 shows the assumed trajectory for the generation of industry process heat between 2019 and 2050. In 2019, gas and coal dominated global heat production. Renewables only contributed 9%—mainly biomass—and electricity had a minor share of 1%. District heat—mainly from gas-fred heating plants—contributed the remaining 7% of the process heat supply, whereas hydrogen and synthetic fuels contributed no measurable proportion. The global OECM 1.5 °C pathway phases out coal and oil for process heat generation between 2035 and 2040, and gas is phased-out as the last fossil fuel by 2050. The most important process heat supply technologies are electric heat systems, such as heat pumps, direct electric resistance heating, and arc furnace ovens for process heat; the share will increase to 22% by 2030 and 60% by 2050. Bioenergy will remain an important source of heat, accounting for 25% in 2050—2.5 times more than in 2019. Synthetic fuels and hydrogen are projected to grow to 8% of the total industry heat supply by 2050.


**Table 12.5** Heat supply under the OECM 1.5 °C pathway


**Table 12.6** Calculated global capacities for renewable and electric heat generation under the OECM 1.5 °C pathway

Based on the annual heat demand, the generation capacities required by the three renewable heating technologies (solar thermal, geothermal, and bioenergy) have been calculated with average capacity factors.

A *capacity factor* is defned as 'the overall utilization of a power or heatgeneration facility or feet of generators. The capacity factor is the annual generation of a power plant (or feet of generators) divided by the product of the capacity and the number of hours of operation over a given period. In other words, it measures a power plant's actual generation compared to the maximum amount it could generate in a given period without any interruption. As power or heating plants sometimes operate at less than full output, the annual capacity factor is a measure of both how many hours in the year the power plant operated and at what percentage of its entire production' (Pedraza 2019).

The same annual capacity factors are assumed for solar thermal as for photovoltaic, at around 1000 hours per year (h/yr), whereas 3000 h/yr is estimated for geothermal energy, and 4500 h/yr. for bioenergy. For electrical systems, 4500 h/yr. is assumed. The capacities shown in Table 12.6 are indicative; the actual installed capacity required under industrial conditions is dependent on a variety of factors, one of which is the production volume of a specifc manufacturing plant.

# **12.4 OECM 1.5 °C Final and Primary Energy Balances**

The European statistics bureau, EUROSTAT, defnes 'fnal energy consumption' as 'total energy consumed by end users, such as households, industry and agriculture. It is the energy which reaches the fnal consumer's door and excludes that which is used by the energy sector itself' (EUROSTAT 2021). Therefore, fnal energy is the energy actually used from the analysed industry, service sector, or building, or for transport.

Figure 12.6 shows the global fnal energy trajectory for *Industry* as a whole and for fve analysed sectors, as well as for *transport* and *service and buildings*. The overall *Industry* demand will increase signifcantly—as the only sector— from 120 EJ in 2019 to 177 EJ in 2050 (over 40%), whereas the *transport* energy demand will increase by more than 50%, due mainly to electrifcation and the introduction of strict effciency standards for all vehicles. The demand of the *service and buildings*

**Fig. 12.6** Global fnal energy demand by sector under the OECM 1.5 °C pathway in 2019–2050


**Table 12.7** Total global fnal energy demand of the aluminium, steel, and chemical industries

sector will decrease by just over 10%, leading to a global total fnal energy demand in 2050 that is 7% lower than in 2019. A combination of ambitious effciency measures and the replacement of a signifcant amount of fuels for transport and heating with electrifcation will reduce the global energy demand despite a growing population and constant economic growth. The energy demands of energy-insensitive industries—chemicals, cement, steel, and aluminium—will increase continuously throughout the entire modelling period to 2050, but specifc energy demands per production unit will decrease, decoupling economic growth from energy demands.

# *12.4.1 Final Energy Demands of Energy-Intensive Industries: Aluminium, Steel, and Chemicals*

A closer look at the energy-intensive industries shows that the aluminium, steel, and chemical industries combined accounted for 34% of the global industrial fnal energy demand and 11% of the total fnal energy demand in 2019 (Table 12.7). The combined energy share of these three sectors will increase to 41% by 2050 under the

OECM scenario, in response to the assumed higher effciencies in other industry sectors, such as construction and mining. The overall energy demand of the three sectors will increase from 41 EJ/year to 72 EJ/year in this period, driven mainly by the projected increase in the global GDP and therefore their production volumes.

A comparison of the consumption shares of *industry*, *transport*, and *service and buildings* shows that their shares in the OECM 1.5 °C pathway will shift very much in favour of industry. The technical energy effciency potential in the buildings sector (Chap. 7) and the transport sector (Chap. 8) will be signifcant, whereas the energy demand of the service sector (Chap. 6) is projected to increase further mainly with the growing population and therefore the growing demand for products produced by this sector.

Figure 12.7 shows that the energy demand for transport will decrease by more than half (to 16%) and that this share will be taken up by the industry sector. The demand of the *service and buildings* sector will remain at the same level, because the reduced energy demand for buildings—mainly achieved by climatization—will be compensated by the increase in the energy demand of service industries, mainly for food production.

In the next section, we present the generation components for the three main sectors *industry*, *transport*, and *service and buildings*. The latter group is called *other sectors* by the International Energy Agency (IEA). Table 12.8 shows the total fnal energy demand for each of the three sectors and their supply by technology. The *transport* energy demand is almost exclusively supplied by oil, whereas natural gas and electricity provide only minor contributions, and coal is not used at all for transport. *Industry* uses the majority of coal and almost half the global demand for gas.

The data show the transition towards renewable energy and an increased electricity demand between 2025 and 2050. The renewable energy share in the *other* 

**Fig. 12.7** Global fnal energy demand shares by sector under the OECM 1.5 °C pathway in 2019 and 2050


**Table 12.8** Global fnal energy demand and supply under the OECM 1.5 °C pathway

(continued)


#### **Table 12.8** (continued)

*sectors* group will increase fastest, whereas the renewable energy supply for *transport* will grow slowly.

# *12.4.2 Global Primary Energy Demand: OECM 1.5 °C Pathway*

The global primary energy demand under the OECM 1.5 °C pathway is shown in Table 12.9. Primary energy includes all losses and defnes the total energy content of a specifc energy source. In 2019, coal and oil made the largest contributions to


**Table 12.9** Global primary energy demand and supply under the OECM 1.5 °C pathway

the global energy supply, followed by natural gas, whereas renewable energies contributed only 15%. The table also provides the projected trajectories for supplies for non-energy uses, e.g. oil for the petrochemical industry. The OECM does not phaseout fossil fuels for non-energy use, because their direct replacement with biomass is not always possible. A detailed analysis of the feedstock supply for non-energy uses was beyond the scope of this research.

# **12.5 Global CO2 Emissions and Carbon Budget**

In the last step, we calculated the energy-related carbon emissions. The OECM 1.5 °C net-zero pathway is based on effcient energy use and a renewable energy supply only, leading to full energy decarbonization by 2050. No negative emission technologies are used and the OECM results in zero energy-related carbon emissions. However, process emissions are compensated by nature-based solutions, such as increased forest coverage. The details are documented by Meinshausen and Dooley (2019) and in Chaps. 11 and 14.

The global carbon budget identifes the total amount of energy-related CO2 emissions available to limit global warming to a maximum of 1.5 °C with no or only a low overshoot. The Intergovernmental Panel on Climate Change (IPCC) is the United Nations body that assesses the science related to climate change. In August 2021, the IPCC published a report that identifed the global carbon budget required to achieve a global temperature increase of 1.5 °C with 67% likelihood as 400 GtCO2 between 2020 and 2050 (IPCC 2021).

# *12.5.1 Global CO2 Emissions by Supply Source*

The CO2 emissions per petajoule (PJ) of energy depend upon the quality and energy content of the energy source, e.g. coal. The German Environment Agency (UBA) has reported the specifc CO2 emissions for a variety of fossil fuels in order to calculate Germany's annual carbon emissions. In terms of coal, the UBA reports that 'most varieties of hard coal have a carbon content (with respect to the original substance) between 60 and 75%. The average content, which can vary from year to year, ranges between 65 and 66%. Hard coal within the lower range, up to a carbon content of about 56%, and a net calorifc value of no more than 22 MJ/kg, is referred to as low-grade coal. Hard coal within the upper range is of coking-coal quality. The highest carbon content, reaching values over 30%, is found in anthracite coal' (UBA 2016). The OECM uses global average emission factors for hard coal, brown coal, oil, and gas, as shown in Table 12.10.


**Table 12.10** Emission factors: primary energy relative to energy-supply-related CO2 emissions

Based on the development of the primary energy supply from fossil fuels, as defned in Table 12.9, the annual energy-related CO2 emissions are calculated as the average global emission factors. Table 12.11 shows the calculated CO2 emissions for fossil power generation and cogeneration and for specifc sectors. The sectorial breakdown provided follows the IEA sectorial breakdown and therefore varies from the values provided for end-use sectors in Table 12.12. The specifc CO2 intensities for power generation are made available in grams of CO2 per kilowatt-hour across total electricity generation—and therefore include carbon-free electricity


**Table 12.11** Global energy-supply-related CO2 emissions under the OECM 1.5 °C pathway

a Includes CHP auto producers; 2 district heating, refneries, coal transformation, gas transport generation, such as from renewables—and for fossil-fuel-generated power only. The reduction in CO2 intensity for fossil-fuel-based power generation between 2019 and 2040 indicates that the share of natural gas will increase and that power plants will become more effcient and therefore generate more units of electricity per unit of fuel.

Table 12.11 shows the energy-related CO2 from the supply side and therefore defnes the carbon budgets for coal, lignite, oil, and gas. The OECM also determines the carbon budget for end-use sectors.

# *12.5.2 Global CO2 Budget*

The remaining carbon budget for each of the following sectors has been defned based on the bottom-up demand analysis of the 12 main industry and service sectors, as documented in Chaps. 5, 6, 7, and 8. Each of those industry and service sectors must complete the transition to fully decarbonized operation within the carbon budget provided. It is very important that the carbon budget shows the cumulative emissions up to 2050, and not the annual emissions. A rapid reduction in annual emissions is therefore vital.

The shares of the cumulative carbon budget required to achieve the 1.5 °C netzero target are shown in Fig. 12.8. Table 12.12 shows the remaining cumulative CO2 emissions in gigatonnes. The total energy-related CO2 for the aluminium industry between 2020 and 2050 is calculated to be 6.1 Gt, 1.5% of the total budget. For the


**Table 12.12** Global carbon budget by end-use sector under the OECM 1.5 °C pathway (GtCO2)

**Fig. 12.8** Global carbon budget by sub-sector under 1.5 °C OECM pathway in 2020–2050

steel industry, the remaining budget is 19.1 Gt of CO2 (4.8%), whereas the chemical industry has the highest carbon budget of 24.8 GtCO2 or 6.2% of the total carbon budget. All other remaining industries can emit 27.1 GtCO2 (6.8%), and all other energy-related activities, such as for buildings, transport, and residential uses, have a combined remaining emissions allowance of 323 GtCO2, or 80.7% of the budget.

# **References**


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

# **Chapter 13** *Scopes 1***,** *2***, and** *3* **Industry Emissions and Future Pathways**

### **Sven Teske, Kriti Nagrath, Sarah Niklas, Simran Talwar, Alison Atherton, Jaysson Guerrero Orbe, Jihane Assaf, and Damien Giurco**

**Abstract** The *Scope 1*, *2*, and *3* emissions analysed in the OECM are defned and are presented for the 12 sectors analysed: (1) energy, (2) power and gas utilities, (3) transport, (4) steel industry, (5) cement industry, (6) farming, (7) agriculture and forestry, (8) chemical industry, (9) aluminium industry, (10) construction and buildings, (11) water utilities, and (12) textiles and leather industry. The interconnections between all energy-related CO2 emissions are summarized with a Sankey graph.

**Keywords** Scope 1, 2, and 3 emissions · Industry · Service · Transport · Buildings steel · Cement · Aluminium · Chemicals agriculture · Forests water utilities · Textile and leather

# **13.1 Introduction**

The OECM methodology has been presented in previous chapters, based on which energy consumption and supply concepts for sectorial pathways were developed. All 12 sectors analysed have been described, the assumptions presented, and the derivations of the energy pathways explained in detail. The resulting energy-related CO2 levels for the sectors are described in Chap. 12. The present chapter focuses on the results of the calculated *Scope 1*, *2*, and *3* emissions for all the sectors analysed.

The industry-specifc emission budgets are further subdivided into so-called *Scope 1*, *2*, and *3* emissions, which defne the responsibility for those emissions. So far, this system has only been applied to companies, and not yet to entire industry

J. G. Orbe · J. Assaf · D. Giurco

S. Teske (\*) · K. Nagrath · S. Niklas · S. Talwar · A. Atherton

University of Technology Sydney – Institute for Sustainable Futures (UTS-ISF), Sydney, NSW, Australia e-mail: sven.teske@uts.edu.au

sectors or regions. For a better overview, the OECM defnitions of *Scopes 1*, *2*, and *3*, which are explained in detail in Chap. 4, are shown again in Box 13.1.

#### *Box 13.1: OneEarth Climate Model: Defnitions of Scope 1, 2, and 3 Emissions*

*Scope 1* – All direct emissions from the activities of an organization or under their control. Including on-site fuel combustion, such as gas boilers, feet vehicles, and air-conditioning leaks. For this analysis only, the economic activities covered under the sector-specifc GICS classifcation that are counted under the sector are included. All the energy demands reported by the International Energy Agency (IEA) Advanced World Energy Balances (IEA, 2020, 2021) for a specifc sector are included.

*Scope 2* – Indirect emissions from electricity purchased and used by an organization. Emissions are created during the production of this energy, which is eventually used by the organization. For reasons of data availability, the calculation of these emission focuses on the electricity demand and 'own consumption', e.g. reported for power generation.

*Scope 3* – Greenhouse gas (GHG) emissions caused by the analysed industry, limited to sector-specifc activities and/or products, as classifed in the GICS. The OECM only includes sector-specifc emissions. Traveling, commuting, and all other transport-related emissions are reported under *transport*. The lease of buildings is reported under *buildings*. All other fnance activities, such as 'capital goods', are excluded because no data are available for the GICS industry sectors, and their inclusion would lead to double counting. The OECM analysis is limited to energy-related CO2 and energy-related methane (CH4) emissions. All other GHGs are calculated outside the OECM model by Meinshausen and Dooley (2019).

The results and key parameters for the primary and secondary energy sectors are presented frst, followed by those for the *industries* and *services*, *buildings*, and *transport* sectors.

# **13.2 Scope 1, 2, and 3: Energy and Utilities**

The *energy* sector includes the primary production of energy from oil, gas, hard coal, and lignite, and all renewable energies. This includes the exploration for all types of fossil fuels; the operation of oil and gas drilling facilities, mining equipment, and fossil fuel transport to refneries; and further processing facilities, as defned under GICS Sector *10 Energy*. To remain within the defned carbon budget, no new oil, gas, or coal-mining projects can be opened up, an assumption that is


**Table 13.1** GICS Sector 10 *energy*

consistent with the recommendations of the IEA NetZero by 2050 report (IEA-NZ 2050) (Table 13.1).

As documented in section 10, the OECM 1.5 °C trajectory requires a phase-out of brown coal (lignite) and hard coal by 2030 in all Organization for Economic Cooperation and Development (OECD) countries and in all other regions thereafter by 2050, at the very latest. The phase-out of brown coal has priority over that of hard coal, because its specifc CO2 emissions are higher. For the oil and gas sector, it is assumed that existing mines will wind-down, with an average decline in production of minus 2% per year for coal, minus 4% per year for onshore oil felds and 6% per year offshore oil felds, and minus 4% per year on- and offshore gas felds, which represent the average industry standards on the global scale (see Chap. 10). However, the production decline rates will differ signifcantly by region and geological formation. It is assumed that natural gas will be phased out by 2050 and partly replaced by alternative fuels, such as hydrogen and/or synthetic fuels, from 2025 onwards. The *energy* sector is also assumed to transition to utility-scale renewable energy projects and therefore to maintain its core business of energy production. Utilityscale renewables are defned as power plants that produce bulk power that is sold to utilities or end-use customers in the *industry* or *service* sector, such as offshore and onshore wind farms, solar farms, and geothermal and biomass power plants (including combined heat and power) with over 1 megawatt installed capacity.

A signifcant part of the renewable energy production by this sector under the 1.5 °C pathway will be from offshore wind, both to supply utilities with electricity and to produce hydrogen and other synthetic fuels. Figure 13.1 shows the global amounts of annual energy production in petajoules (PJ). The renewable energy production level will reach parity with those of oil and gas by 2030 and will continue to grow throughout the next two decades. The remaining oil and gas production shown for 2050 is for non-energy use.

*Energy—Scope 1* emissions are defned as the direct emissions related to the extraction, mining, and burning of fossils fuels. This analysis covers both the

**Fig. 13.1** Global primary energy sector—energy production under the OECM 1.5 °C pathway

energy-related CO2 emissions and non-energy GHGs, such as methane (CH4) emissions from mining and fossil fuel production.

*Energy—Scope 2* emissions are indirect emissions from the electricity used for the operation of mining equipment, oil and gas rigs, refneries, and other equipment required in the primary energy sector. Their calculation is based on statistical information ('own consumption') from the IEA Advanced Energy Balances. The OECM assumes the global average carbon intensity of electricity generation for each calculated year according to the OECM power scenario, which will reach 100% renewables by 2050 (for details, see Chap. 12).

*Energy—Scope 3* emissions are embedded CO2 emissions, which occur when the fossil fuel produced by the primary energy industry is burnt by end users.

Table 13.2 shows the *scope 1*, *2*, and *3* emissions for coal, oil, and gas and the development of the fuel intensity of the global economy. In 2019, as a global average, 1.25 PJ of coal was used for each billion US\$ of gross domestic product (\$GDP). This coal intensity is assumed to halve by 2025 and to drop by 85% by 2030. The global economy will grow independently of coal use under the OECM 1.5 °C pathway (Table 13.3).

The *utilities* sector covers energy transport and the operation and maintenance of power- and heat-generating equipment and is responsible for the energy transport infrastructure, such as power grids and pipelines to the end user. In this analysis, the *utilities* sector is a secondary energy service provider, whose core function is the generation and distribution of electricity and the distribution of natural gas, as well as hydrogen and synthetic fuels, beyond 2030. It operates and maintains power and cogeneration plants, power grids (all voltage levels), and pipelines and provides energy services, such as balancing, demand-side management, and storage. 'Utilities' are energy service companies linking the primary energy supply with consumers.

Electricity and gaseous fuel supplies are the core commodities of gas and power utilities. With the increased electrifcation of the transport and heating sectors, the


**Table 13.2** Global *energy* sector—*scopes 1*, *2*, and *3* for coal, oil, and gas

**Table 13.3** Global *energy* sector—*scopes 1*, *2*, and *3*


electricity demand—and therefore the potential market value of power utilities will increase signifcantly. Renewable electricity will overtake global coal- and gasfuelled power generation combined by 2025. By 2045, the market volume of hydrogen and synthetic fuels will be as high as that of natural gas for gas utilities, making them important new products.

*Utilities—Scope 1* emissions are defned as the direct emissions from fuels related to the generation and transmission of electricity and the distribution of fossil fuels and/or renewable gas.

*Utilities—Scope 2* emissions are indirect emissions from the electricity used for the production of a sector's core product. This includes the electricity consumption of power plants, losses by power grids, and the operation of pumps for gas pipelines, etc. Their calculations are based on statistical information listed under 'selfconsumption' of the IEA Advanced Energy Balances plus the global average power grids losses, which are assumed to be 7.5%.

*Utilities—Scope 3* emissions are embedded CO2 emissions that occur with the use of electricity or gaseous fuels by end users. Table 13.4 shows all *scope 1*, *2*, and *3* emissions for the *utilities* sector by sub-sector and in total.


**Table 13.4** Global *utilities* sector—*scopes 1*, *2*, and *3*

# **13.3** *Scopes 1***,** *2***, and** *3***:** *Industry*

All results for the *scope* 1, *2*, and *3* emissions for the fve main energy-intensive industry sectors are based on the energy demand assessment documented in Chap. 5.

# *13.3.1* **Scopes 1***,* **2***, and 3: Chemical Industry*

The *chemical industry* is the most complex industry of all the sectors analysed, and the data available on its energy demand are less detailed than for, for example, the steel industry. Furthermore, the production of chemical commodities (see Sect. 5.1) is energy intensive, and they are used not only across the chemical industry but also in other sectors. Therefore, the calculation results shown in Table 13.5 are subject to uncertainties resulting from the paucity of detailed data. The global energy demand data for, for example, the pharmaceuticals industry are not available, and the calculations are based upon sector-specifc energy intensities and the market shares of the pharmaceuticals industry in 2019 (see Sect. 5.1.3).

*Chemicals—Scope 1* emissions are defned as the direct emissions related to the production of raw materials for the chemical industry from natural gas, ethane, oilrefning by-products (such a propylene), and salt, which are used to manufacture bulk chemicals, such as sulfuric acid, ammonia, chlorine, industrial gases, and basic polymers, such as polyethylene and polypropylene.

*Chemicals—Scope 2* emissions are indirect emissions from the electricity used for the production and processing of chemical products and the manufacture of goods that fall under *chemicals*, as classifed under GICS 1510 10.

*Chemicals—Scope 3* emissions are all non-energy-related GHG emissions and aerosols that fall under the Montreal Protocol (UNEP MP, 2021). Montreal Protocol gases are mainly propellants, foams, or liquids and gases used for cooling and refrigeration that are produced by the chemical industry. More details about these gases are given in Chap. 11.

*Scope 1* and *2* emissions will reach zero by 2050, whereas *Scope 3* emissions will only be reduced by 73% compared with 2019 due to the nature of those substances.

# *13.3.2* **Scope 1***,* **2** *and* **3***:* **Cement Industry**

The energy intensity of the cement production processes is well-documented, and data for the energy demands and process emissions are available. This analysis includes all steps in cement production, from quarrying the raw materials to its storage in cement silos. However, the further processing of cement for construction, for example, is not included but is included in the *buildings and construction* sector (Sect. 13.4).


**Table 13.5** Global *scope 1*, *2*, and *3* emissions of the *chemical industry*


**Table 13.6** Global *scope 1*, *2*, and *3* emissions for the *cement industry*

*Cement—Scope 1* emissions are defned as the direct energy-related CO2 emissions related to all steps of cement production, from mining to the fnal raw product that is used in further processes and applications. The fuels for mining vehicles are included, as well as the process heat for clinker production in kilns, etc. Emissions from the calcination process—the decomposition of limestone into quick lime and carbon dioxide (Kumar et al., 2007)—are also included.

*Cement—Scope 2* emissions are the indirect emissions from the electricity used across all steps of the value chain of the cement industry.

*Cement—Scope 3* emissions of the cement industry are *scope 2* emissions of the *buildings* sector, according to the World Business Council for Sustainable Development's Cement Sector Reporting Guidance (WBCSD, 2016).

By 2050, there will be no energy-related CO2 emissions from the cement industry under the OECM 1.5 °C pathway (Sect. 5.2). Process emissions from calcination are assumed to decline from 0.4 tCO2 per tonne of clinker to 0.24 tCO2—an assumption based on the IEA Technology Roadmap (IEA, 2018). Table 13.6 and Fig. 13.2 show the calculated results for the *scope 1*, *2*, and *3* emission of the global cement industry.

# *13.3.3* **Scopes 1***,* **2***, and* **3***:* **Aluminium Industry**

As for the cement industry, all aluminium production processes are well-documented. The processes and their energy demand for each step of aluminium production, from bauxite mining to aluminium sheets or aluminium blocks, which are then delivered to other industries for further processing, are available in the literature. The recycling of aluminium for the production of secondary aluminium is also included. All assumptions for the projected development of the aluminium industry—including bauxite mining—are documented in Sect. 5.3.

*Aluminium—Scope 1* emissions are defned as the direct energy-related CO2 emissions related to the use of fuels for bauxite mining, alumina processing, and all steps of the production of primary and secondary aluminium. The process emissions from anode or paste (IAI, 2006) consumption, which lead to CO2 emissions that are not energy related, are included.

**Fig. 13.2** Global *cement* sector—energy- and process-related CO2 under the OECM 1.5 °C pathway

*Aluminium—Scope 2* emissions are the indirect emissions from the electricity used across all the steps of the value chain of the aluminium industry.

*Aluminium—Scope 3* emissions are solely those emissions caused by tetrafuoromethane, a strong GHG that is produced in certain aluminium production processes. A recent study published in Nature highlights the increased emissions of this gas, which probably derive from aluminium production facilities in Asia (Nature 8/2021). We decided to include tetrafuoromethane emissions in this OECM analysis to highlight the importance of this fnding.

By 2050, all energy-related CO2 emissions of the aluminium industry will be zero and the industry will be fully decarbonized. However, process-related GHG emissions are not expected to be completely phased out (Table 13.7).


**Table 13.7** Global *scope 1*, *2*, and *3* emissions for the *aluminium industry*

**Fig. 13.3** Global *steel* sector—iron-ore mining and steel production under the OECM 1.5 °C pathway

# *13.3.4* **Scope 1***,* **2***, and* **3***:* **Steel Industry**

Global and regional steel industry emissions are among the most discussed of all industry emissions. Various industry- and science-based working groups have developed relevant scenarios over the past decade. However, most of them are consistent with the Iron and Steel Technology Roadmap of the IEA (IEA, 2020). The OECM 1.5 °C pathway for the *steel industry* is based to a large extent on the IEA assumptions for the energy demand side but has added a more ambitious decarbonization scenario for the energy *supply* side. Figure 13.3 shows the development of

iron-ore mining and primary and secondary steel production assumed for the global market between 2019 and 2050. The increase in secondary steel—recycled steel, mainly from scrap—will increase from 35% in 2019 to 48% in 2050, leading to a reduction in the iron and mining demands and the process emissions that are only related to primary steel production. Therefore, a high recycling rate will directly affect process emissions, which are not related to the actual energy supply but to the steel-making process itself. Further information about the assumptions for the *steel industry* is documented in Chap. 5 (Sect. 5.4).

The OECM analysis includes energy-related CO2 emissions that occur from iron-ore mining across all steps of the steel manufacturing processes for primary and secondary steel but exclude manufacturing processes that use steel for product manufacture, such as the automotive industry.

*Steel—Scope 1* emissions are defned as the direct energy-related CO2 emissions related to the use of fuels for iron-ore mining and the production of primary and secondary steel. Process emissions from anode or paste (IAI, 2006) consumption, which lead to CO2 emissions that are not energy related, are included.

*Steel—Scope 2* emissions are the indirect emissions from the electricity used across all steps of the value chain of the steel industry.

*Steel—Scope 3* emissions are only process-related emissions, as defned in the 2006 IPCC Guidelines for National Greenhouse Gas Inventories (IPCC, 2006). It is assumed that process emissions will decline signifcantly from 0.92 tCO2 per tonne currently to 0.08 tCO2 by 2050 as a result of the transition to electric-furnace-based steel production (see Sect. 5.4.3) (Table 13.8).

# *13.3.5* **Scopes 1***,* **2***, and* **3***:* **Textile and Leather Industry**

The *textile and leather industry* is part of the IEA *industry* sector but is not part of the GICS (15) *materials* group (see Chap. 4). The *textile and leather industry* is closely associated with the *chemicals industry*, from which synthetic fbres and plastic for accessories are sourced, and with the *agriculture* sector, for cotton and other natural fbres. The production of leather depends on animal farms, especially those that produce meat. The assumptions made for the calculation of the energyrelated CO2 emissions of this industry are documented in Sect. 5.5.


**Table 13.8** Global *scope 1*, *2*, and *3* emissions for the *steel* industry


**Table 13.9** Global *scope 1*, *2*, and *3* emissions for the *textile and leather industry*

*Textile and Leather—Scope 1* emissions are defned as the direct energy-related CO2 emissions associated with all the steps of textile and leather production that require process heat or fuels. It covers leather production, but not the production of fbres, which is part of the *chemicals* sector. The calculation of these emissions includes the value chain until delivery to retail.

*Textile and Leather—Scope 2* emissions are the indirect emissions from the electricity used for the production of textile and leather products, excluding fbres manufacture and retail.

*Textile and Leather—Scope 3* emissions include 25% of all CH4 emissions from the agricultural sector to refect the non-energy-related GHG emissions associated with the production of natural fbres and animal skins (Table 13.9).

# **13.4** *Scope 1***,** *2***, and** *3***:** *Services*

All the results for the *scope 1*, *2*, and *3* emissions of the four *service* sectors analysed are based on the energy demand assessment documented in Chap. 6. Nonenergy-related GHG emissions form the majority of the *service* sector emissions, whereas energy-related CO2 is a relatively small component compared with that in other sectors, such as *industry* and *transport*. These non-energy-related GHG emissions—referred to as agriculture, forestry, and other land-uses (AFOLU) in climate science—are among the main emitters of non-energy CO2, CH4, and nitrous oxide (N2O). The *service* sectors analysed, *agriculture and food*, *forestry and wood*, *fsheries*, and *water utilities*, are described and the assumptions are documented in Chaps. 6, 11, and 14. Therefore, in this section, we focus solely on the presentation of their calculated *scope 1*, *2*, and *3* emissions.

# *13.4.1* **Scope 1***,* **2***, and* **3***:* **Agriculture and Food** *Sector*

The *agriculture and food* sector includes all economic activities from 'the feld to the supermarket'. With regard to the energy used, this sector is a combination of the service sector *agriculture* and the industry sub-sector *food and tobacco*. Therefore, it includes crop and animal farming and the processing of all commodities into food, beverages, and tobacco products.

*Agriculture and Food—Scope 1* emissions are related to fuel used in agricultural vehicles, such as tractors, machinery for harvesting and other equipment used on farms, as well as heat for food and tobacco processing and packaging.

*Agriculture and Food—Scope 2* emissions include those for electricity purchased from utilities for either farming or any step in food processing or packaging. On-site electricity generation (e.g. on farms via solar photovoltaic, wind power, or bioenergy from residuals) will reduce *scope 2* emissions, but sub-sector-specifc on-site generation is not assumed in this analysis.

*Agriculture and Food—Scope 3* emissions include AFOLU emissions, N2O, and ammonia emissions from fertilizers and CH4 emissions (see Chaps. 11 and 14).

All energy-related CO2 emissions of the *agriculture and food* sector will be reduced by half by 2030 and phased out entirely by 2050. However, it is assumed that AFOLU emissions from agriculture cannot be reduced to zero, because the demand for food for the growing global population will increase (Table 13.10).

# *13.4.2 Sc***opes 1***,* **2***, and* **3***:* **Forestry and Wood** *Sector*

Like the *agriculture and food* sector, the *forestry and wood* sector contains to subsectors: *forestry*, which is part of the IEA's *other sectors*, and the IEA industry subsector *wood and wood products*, which includes the pulp and paper industry. Details of the energy demand of this sector are provided in Sect. 6.2.

*Forestry and Wood—Scope 1* emissions include those from heavy machinery for wood harvesting, all-terrain vehicles, power tools, chainsaws, etc.

*Forestry and Wood—Scope 2* emissions are the indirect emissions from electricity. Like the agricultural sector, the forestry sector has signifcant potential for onsite power and heat generation, e.g. from forestry residuals, which can lower its *scope 2* emissions, but this is not assumed under the OECM 1.5 °C pathway.


**Table 13.10** Global *scope 1*, *2*, and 3 emissions for the *agriculture and food* sector (including tobacco)

*Forestry and Wood—Scope 3* emissions are forestry-related AFOLU emissions. The transition towards sustainably managed forests, the cessation of deforestation, and the commencement of reforestation are integral parts of the OECM 1.5 °C pathway as 'carbon sinks'. Therefore, *scope 3* emissions will become negative by 2030 (see Chaps. 11 and 14) (Table 13.11).

# *13.4.3* **Scopes 1***,* **2***, and* **3***:* **Fisheries** *Sector*

The majority of all energy-related *scope 1* and *2* emissions in this industry are from fshing vessels and other equipment directly related to wild catches and aquaculture fsh farms. Whereas the energy demand for fshing vessels is documented in the literature (see Sect. 6.3), no statistical data on the global energy demand for aquaculture and fsh farming are available. Instead, only accumulated data on the GHG emissions for the global aquaculture sector have been published and have been used to calculate the *scope 3* emissions (MacLeod et al., 2020). Therefore, the energy demand of the fshing industry in 2019 and its projection until 2050 are estimates with uncertainties.

*Fisheries—Scope 1* emissions are defned as the direct energy-related CO2 emissions related to the use of fuels for fshing vessels and directly related to the infrastructure, such as refrigerators and freezers for fsh on board fshing vessels.

*Fisheries—Scope 2* emissions are the indirect emissions from the electricity used for cooling devices as part of the cooling chain for fsh, from 'dock to supermarket'.

*Fisheries—Scope 3* emissions are emissions from aquaculture as defned by MacLeod et al. (2020) as 'emissions arising from fshmeal production, feed blending, transport … and non-feed emissions from the nitrifcation and denitrifcation of nitrogenous compounds in the aquatic system ('aquatic N2O')'. Also included are the estimated energy-use emissions, mainly for pumping water.

Table 13.12 shows the results for all the calculated emissions in this industry. It is assumed that about one-quarter of aquaculture *scope 3* emissions are directly related to energy use and will therefore be reduced to zero with the use of 100% renewable energy.


**Table 13.11** *Global scope 1*, *2*, and *3* emissions for the *forestry and wood* sector


**Table 13.12** Global *scope 1*, *2*, and *3* emissions for the *fsheries* sector

**Table 13.13** Global *scope 1*, *2*, and *3* emissions for *water utilities*


# *13.4.4* **Scopes 1***,* **2***, and* **3***:* **Water Utilities**

Only 13% of the GHG emission from *water utilities* are related to energy use. The bulk of GHG emissions are related to CH4 and N2O emission from sewers or the treatment of biological wastewater and the resulting sludge. Chapter 6 documents all the assumptions and input data used to calculate the *scope 1*, *2*, and *3* emissions for water utilities.

*Water Utilities—Scope 1* emissions are defned as the direct energy-related CO2 emissions associated with the supply of the low- and medium-temperature process heat used in all steps of wastewater treatment.

*Water Utilities—Scope 2* emissions are the indirect emissions from the electricity used across all steps of wastewater treatment processes.

*Water Utilities—Scope 3* emissions are the CH4 and N2O emissions from sewers or biological wastewater treatment. They are calculated with average global emission factors of 0.17 kg CO2 equivalents per cubic metre (kgCO2eq/m3 ) for CH4 emissions and 0.033 kgCO2eq/m3 for N2O emissions.

Water utilities have signifcant potential to use the CH4 from sewage and wastewater treatment for on-site power and heat generation. The identifed *scope 2* emissions for water utilities do not include the implementation of this technology. The *scope 3* emissions shown in Table 13.13 are entirely related to CH4 and N2O emissions and are projected to increase with the growing global population. The use of on-site CH4 emissions with a global warming potential (GWP) of 25 (see Chap. 11) for electricity and heat generation would result in CO2 (GWP = 1), instead of CH4 emissions, and would therefore signifcantly reduce the *scope 3* emissions. Therefore, we strongly recommend the utilization of on-site CH4 emissions for energy generation.

# **13.5** *Scopes 1***,** *2***, and** *3***:** *Buildings*

The *buildings* sector is further broken down into residential and commercial buildings and is based on calculations that include construction. The energy demand for construction is taken from the IEA World Energy Balances, and the demand includes the *construction of buildings* (ISIC Rev. 4, Div. 41), c*ivil engineering* (ISIC Rev. 4, Div. 42), and s*pecialized construction activities* (Div. 43), as documented in Chap. 4. It is assumed that 60% of the energy used for construction is for buildings. The energy demands calculated for residential and commercial buildings are based on a separate research project under the leadership of the Central European University (Chatterjee et al., 2021) and are documented in Chaps. 3 and 7.

*Buildings—Scope 1* emissions are defned as direct energy-related CO2 emissions associated with the construction of those buildings.

*Buildings—Scope 2* emissions are indirect emissions from the residential and commercial use of electricity and energy for space heating. The commercial electricity demand is the remaining electricity that is not allocated elsewhere in the *service*, *industry*, *transport*, or *residential* sectors, to avoid double counting.

*Buildings—Scope 3* emissions are the *scope 1* emissions of the *cement industry* to capture the embedded building emissions from construction materials.

There are no *scope 3* emissions calculated for construction to avoid double counting with the remaining *buildings* sector. Table 13.14 shows the global *scope 1*, *2*, and *3* emissions for all sub-sectors and for the overall *buildings* sector.

# **13.6 Scope** *1***,** *2***, and** *3***:** *Transport*

The *transport* sector includes all travel modes (aviation, shipping, and road transport), and passenger and freight transport have been calculated separately on the basis of current and projected passenger-kilometres (pkm) and tonne-kilometres (tkm), as documented in Chap. 8. The *transport* sector includes the manufacture of vehicles and other transport equipment, as defned in GICS group 2030 (see Chap. 4) and documented in Sect. 8.9.

*Transport—Scope 1* emissions are defned as the direct energy-related CO2 emissions associated with the manufacture of road and rail vehicles, planes, and ships.

*Transport—Scope 2* emissions are the indirect emissions from electricity used for all from the electric drives in vehicles and the electricity required for hydrogen or synthetic fuel production. The emission factors for this electricity—as in all other *scope 2* emission calculations—are based on the OECM 1.5 °C pathway for power


**Table 13.14** Global *scope 1*, *2*, and *3* emissions for *buildings*

generation, with an emission factor of 0.5 kg CO2 per kilowatt-hour in 2019, which will decline to zero by 2050.

*Transport—Scope 3* emissions are all the emissions caused by the utilization of all vehicles, planes, and ships for passenger and freight transport by end users. These emissions are not further allocated to other sectors in which vehicles are used to avoid double counting. Data are unavailable on how freight kilometres are distributed to, for example, the cement or steel industry.

Table 13.15 provides the global *scope 1*, *2*, and *3* emissions for the *transport* sector. Specifc emissions from, for example, airports or single airline offces, as


**Table 13.15** Global *scope* 1, 2, and 3 emissions for the transport sector

defned under GICS 2030 5010, cannot be assessed on a global scale because of lack of data. Furthermore, these emissions are allocated under 'commercial buildings'. *Scope 3* emissions are the 'classic' emissions when consumers drive a car or use a plane. The OECM deliberately includes electricity emissions from, for example, electric cars under *scope 2* emissions, because car manufacturers today include the charging infrastructure in their value chain and are therefore responsible for it.

# **13.7** *Scopes 1***,** *2***, and** *3***: Global Summary**

A global assessment of *scopes 1*, *2*, and *3* for the whole *industry* sector is a new research area, and changes had to be made to the method for determining those emissions, which was originally developed by the World Resource Institute (WRI), as documented in Chap. 4.

The OECM methodology differs from the original concept primarily insofar as the interactions between industries and/or other services are kept separate. A primary class is defned for the primary energy industry, a secondary class for the supply utilities, and an end-use class for all the economic activities that consume energy from the primary- or secondary-class companies, to avoid double counting. All the emissions by defned industry categories (e.g. with GICS) are also separated, streamlining the accounting and reporting systems. The volume of data required is reduced, and reporting is considerably simplifed with the OECM methodology.

Figure 13.4 shows the global energy-related *scope 1*, *2*, and *3* CO2 emissions in 2019 as a Sanky fow chart. The primary energy emissions are on the left and the end-use-related emissions are on the right. The carbon budgets remain constant, from production to end use, apart from losses and statistical differences. A simplifed description is that all *scope 1* emissions are on the left, with the primary energy industry as the main emitter, and all *scope 3* emissions are on the right, with the consumers of all forms of energy and for all purposes as the main emitters. In the secondary energy industry, utilities are the link between the demand of end users

**Fig. 13.4** Global *scope 1*, *2*, and *3* energy-related CO2 emissions in 2019

**Fig. 13.5** Global *scope 1*, *2*, and *3* energy-related CO2 emissions in 2030 under the OECM 1.5 °C pathway

and the supply by the primary energy industry. The fgure also shows the complex interconnections between demand and supply.

Figure 13.5 shows the energy-related CO2 emissions and the interconnections between various sectors and consumers in 2030 under the global 1.5 °C pathway.

# **References**


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

# **Chapter 14 Nature-Based Carbon Sinks: Carbon Conservation and Protection Zones**

### **Kriti Nagrath, Kate Dooley, and Sven Teske**

**Abstract** Basic information on ecosystem-based approaches to climate mitigation is provided, and their inclusion in international climate and nature conservation treaties is discussed. Key concepts around net-zero emissions and carbon removal are examined, as are the roles they play in the One Earth Climate Model, which develops a 1.5 °C-compatible scenario by combining ecosystem restoration with deep decarbonization pathways. The carbon removal potentials of the fve ecosystem restoration pathways—forests and agricultural lands, forest restoration, reforestation, reduced harvest, agroforestry, and silvopasture—are provided. Land-use management options, including the creation of 'carbon conservation zones' (CCZ), are discussed.

**Keywords** Ecosystem-based approaches · Ecosystem restoration pathways · Forest restoration · Reforestation · Reduced harvest · Agroforestry · Silvopasture · 'Carbon conservation zones' (CCZ)

# **14.1 Ecosystem Approaches to Climate Action**

This section looks at the variety of ecosystem approaches available for implementation as climate solutions. It also follows the global developments of ecosystem and nature outcomes from the recent climate summit, the 26th Conference of the Parties (COP26) to the United Nations Framework Convention on Climate Change (UNFCCC).

K. Nagrath · S. Teske (\*) University of Technology Sydney – Institute for Sustainable Futures (UTS-ISF), Sydney, NSW, Australia e-mail: sven.teske@uts.edu.au

K. Dooley University of Melbourne, Melbourne, VIC, Australia

# *14.1.1 Understanding Ecosystem Approaches*

Climate change and climate action can no longer be discussed without reference to their environmental impacts, in particular the crises of biodiversity loss and ecosystem decline. Ecosystem approaches to climate management that restore degraded ecosystems and focus on nature will play important roles in climate solutions, for both mitigation and adaptation. These approaches aim to maintain and increase the resilience of people and the ecosystems upon which they rely and to reduce their vulnerability (Lo, 2016). Healthy, well-managed ecosystems have climate change mitigation potential, through the sequestration and storage of carbon in healthy forests, wetlands, and coastal ecosystems (IPBES, 2019).

Approaches to protecting and restoring nature can take a variety of forms. These include initiatives such as the sustainable management, conservation, and restoration of ecosystems. The ecosystem approach is a strategy for the integrated management of land, water, and living resources, which promotes their conservation and sustainable use in an equitable way, as defned by the Convention on Biological Diversity (CBD). The convention also defnes 'ecosystem-based adaptation' (EbA) as the use of biodiversity and ecosystem services as part of an overall adaptation strategy to help people adapt to the adverse effects of climate change.

Ecosystem services are the benefts people obtain from ecosystems, which have been classifed by the Millennium Ecosystem Assessment as supporting services, such as seed dispersal and soil formation; regulating services, such as carbon sequestration, climate regulation, water regulation and fltration, and pest control; provisioning services, such as the supply of food, fbre, timber and water; and cultural services, such as recreational experiences, education, and spiritual enrichment.

The International Union for Conservation of Nature (IUCN) defnes naturebased solutions (NbS) as 'actions to protect, sustainably manage, and restore natural or modifed ecosystems, that address societal challenges effectively and adaptively, simultaneously providing human well-being and biodiversity benefts'. These solutions include ecosystem restoration strategies, such as ecological restoration, ecological engineering, and forest landscape restoration; issue-specifc ecosystem-related strategies, such as ecosystem-based adaptation or mitigation and disaster risk reduction; infrastructure-related strategies; ecosystem-based management strategies; and area-based ecosystem protection strategies.

'Landscape restoration' refers to the improvement of degraded land on a large scale, to rebuild ecological integrity and enhances people's lives. It involves restoring degraded forests and agricultural lands by reducing the intensity of use or improving productivity with mixed-use approaches, such as agroforestry and climate-smart agriculture (Winterbottom, 2014).

'Ecological engineering' is defned as the design of sustainable ecosystems that integrate human society with its natural environment for the beneft of both. It includes ecosystem rehabilitation (actions that repair the structures and functions of indigenous ecosystem), nature engineering, and habitat reconstruction and reclamation (stabilization, amelioration, increases in utilitarian or economic value). However, indigenous ecosystems are rarely used as models (Mitsch, 2012). Therefore, there is a diversity of approaches that can be adopted to protect and restore the natural world.

# *14.1.2 Ecosystem and Nature Outcomes at COP26*

The 2021 Glasgow Climate Pact recognizes the critical role of protecting, conserving, and restoring nature, while ensuring social and environmental safeguards, through the following text:

Emphasizes the importance of protecting, conserving and restoring nature and ecosystems to achieve the Paris Agreement temperature goal, including through forests and other terrestrial and marine ecosystems acting as sinks and reservoirs of greenhouse gases and by protecting biodiversity, while ensuring social and environmental safeguards*.* (*Decision-/ CP.26 Glasgow Climate Pact*, 2021)*.*

Food, land, and nature were popular topics at COP26 and featured in a series of pledges, speeches, initiatives, and coalitions (Chandrasekhar & Viglione, 2021). These included deforestation pledges, new climate pledges, the methane pledge, and other agricultural innovation and policy announcements. The key pledges included:


Although these declarations are signs that we are moving in the right direction, there are concerns regarding the uncertainty around key defnitions and the transition from promise to action, which must be resolved soon.

A World Wild Fund for Nature (WWF) study found that 92% of countries' new climate action plans now include measures to tackle nature loss. One hundred and fve of the 114 enhanced Nationally Determined Contributions (NDCs) submitted by 12 October included nature in their climate mitigation or adaptation plans. Of the 96 NDCs that cited using nature for climate mitigation, 69 quantifed these as numerical targets, mostly in the forest sector (Bakhtary et al., 2021).

# *14.1.3 Concepts of Consequence*

This section discusses key concepts around net-zero emissions and carbon removal that we must understand to model the pathways in Chap. 11.

'Net zero' refers to the balance between the amount of greenhouse gases (GHGs) produced by humans and the amount removed from the atmosphere. This means that for any remaining emissions produced, an equivalent amount must also be removed through processes such as planting new forests, which reduce the GHGs accumulating in the atmosphere, to reach net-zero emissions.

As discussed in previous chapters, it is imperative for the various industry sectors to reduce their energy emissions (which primarily arise from fossil fuels) to zero. Given the temporal differences between the fossil and terrestrial carbon cycles, any essential residual emissions arising from non-energy sources and processes must be removed via geological storage, to go beyond net-zero emissions and eventually achieve net-negative emissions to reduce atmospheric GHG concentrations.

Carbon dioxide (CO2) removal (CDR) is the process of removing CO2 from the atmosphere and locking it away in a carbon 'sink' for a long period of time. A carbon sink is a natural or human-made reservoir that accumulates and stores carbon and thus lowers the concentration of CO2 in the atmosphere. Forests and oceans are natural carbon sinks and absorb more CO2 from the atmosphere than they emit. CDR requires that we enhance the ability of these natural sinks to remove and store carbon, or store this carbon geologically.

There are both natural and technological strategies to remove carbon from the atmosphere and store it in a sink. Natural strategies include reforestation and the ecosystem restoration approaches discussed above, where carbon is removed from the atmosphere by photosynthesis and stored in vegetation and soil. Although natural solutions, such as restorative agriculture and reforestation, can help remove carbon, they must be thoroughly monitored and balanced against competing demands on land use. Technological strategies, such as direct air capture and enhanced mineralization, that capture carbon underground or under the ocean or in products such as concrete, are also being explored but are yet to be commercialized on a large scale.

The OECM model focuses on natural strategies for carbon removal. The different land management pathways for achieving this are discussed in the next section.

# **14.2 Ecosystem Restoration Pathways**

The OECM model presents a 1.5 °C-compatible scenario combined with ecosystembased approaches. The ecosystem restoration pathways outlined in this section have been published as Littleton et al. (2021) and have been built on previous work by Meinshausen and Dooley (2019) .

# *14.2.1 Pathways*

The fve pathways involve forests and agricultural lands: forest restoration, reforestation, reduced harvest, agroforestry, and silvopasture. The frst three pathways focus on the forestry sector, and the latter two are most relevant for the agriculture and food sector. In all three forest pathways, the intervention is natural regeneration with no active planting of trees (Littleton et al., 2021).

Forest restoration sets aside natural (secondary) forest areas that are partly deforested or degraded for conservation purposes. This pathway is applied to all biomes. Reforestation includes the reforestation of mixed native species maintained for conservation purposes. It is limited to biomes that would naturally support forests, after the identifcation of previously forested land in close proximity within 70–105 km for tropical forests and within 11–18 km of temperate forests. Reforestation in boreal biomes is excluded because the albedo effect accompanies changes from deforested to forested land types, specifcally at high latitudes, which can potentially increase warming. The reforestation pathway is the only land-management intervention in this scenario that requires a change in land use. Reduced harvest describes a reduction in harvest intensity by 25% in commercial forests in boreal and temperate biomes. In tropical and subtropical biomes, commercial timber extraction is halted completely, given the lack of evidence that any form of reducedimpact logging leads to increased carbon stocks. These management interventions only apply to natural managed forests and not to plantations. Areas of shifting cultivation are excluded from consideration for reduced harvesting, to avoid impacting communities dependent on subsistence agriculture (Littleton et al., 2021).

The pathways involving the regeneration of agricultural areas—agroforestry and silvopasture—allow for existing land uses to continue. Temperate, subtropical, and tropical cropland and grazing areas with mean annual precipitation ranging from 400 to 1000 mm per year were targeted for these two pathways. Agroforestry can be implemented in many different ways, but here it is assumed to be the integration of additional trees into agricultural landscapes, which will result in signifcant sequestration across large areas of temperate and tropical croplands. Silvopasture, defned here as a reduction in grazing intensity on managed pastures, results in almost twice the level of carbon sequestration over a similar land area.

# *14.2.2 Methodology*

Spatial distribution for the fve pathways was identifed using WRI's global map of forest condition and the ESA-CCI land cover maps for the forest and agriculture pathways, respectively. The areas identifed for ecosystem restoration were simulated in a community land surface model, the Joint UK Land Environment Simulator (JULES) to get the carbon sequestration potential. JULES also incorporates the dynamic global vegetation model TRIFFID to simulate vegetation and carbon cycle

processes. JULES simulations were run using meteorological forcing output from HadGEM2-ES, covering the period 1880–2014 (historical) and 2015–2100 (SSP1–2.6) on a 3-hour timestep at the N96e grid size.

For temperature projections, MAGICC, a reduced-complexity probabilistic climate emulator, was used, which refects updated climate science knowledge. The scenarios are consistent with limiting warming to 1.5 °C by the end of the century, although at best, with a roughly 50–50 chance of staying below this limit.

A no-removal baseline scenario is modelled under a shared socioeconomic sustainability future (SSP1) and represents CO2 emissions from forestry and land use (including land-use changes) in the absence of the ecosystem restoration measures considered here. To minimize the risk of double-counting sequestration, all carbon sequestration reported in this baseline scenario are set to zero from 2050 onwards.

# *14.2.3 Results*

The results of Littleton et al. (2021) showed the median gross cumulative potential of additional CO2 removal with the fve ecosystem restoration pathways to be 93 Gt of carbon (C) until 2100, as shown in Table 14.1. The peak annual sequestration rate for all ecosystem restoration pathways (forest restoration, reforestation, reduced harvest, agroforestry, and silvopasture) is 3.1 GtC per year in 2041, as forests reach maturity. Then on, the fux declines, with an average annual sequestration rate of 1.1 GtC per year from 2050 to 2100. This removal will be offset by ongoing net landuse emissions but still has a signifcant contribution to temperature reduction. Combined with a 100% renewable energy scenario by 2050 in the OECM, this additional carbon uptake reduced 2100 temperature by a further 0.12 °C when compared to a no-removal scenario (Littleton et al., 2021).

The most successful restoration pathway identifed in terms of carbon sequestration is *reduced harvest* in the tropics, with carbon gains of 80–100 tC per hectare by 2100 in Southeast Asia and the Amazon basin. Reduced harvest means that less timber is harvested. The pathway assumes that harvest intensity in temperate and


**Table 14.1** Summary statistics for the cumulative uptake of CO2 in all pathways

boreal biomes is decreased, and commercial harvesting is completely stopped in tropical forests. Reduction in harvest can be achieved through either reduced harvest intensity in natural forests or doubling the length of rotation of managed forests. This will have a signifcant impact on timber supply and on the economics of forestry industries. Strategies to continue to meet the timber demand include shifting away from wood products, increasing effciency, and recycling of wood-based products, to avoid the expansion of plantation forests.

The next highest gains are seen in reforestation in China, Latin America, and Southeast Asia in the decade leading up to 2050. Reforestation target areas are adjacent to existing intact forests and are consistent with targets in national policies and international commitments like the Bonn Challenge. The analysis acknowledges that natural succession to native vegetation is more cost-effective and has a greater success rate than planting new forests. As the carbon sequestration potential of full regrowth of deforested land to forested land is higher than in recovering carbon stocks in degraded forests, this is the only pathway that requires a land-use change of 344 Mha converted from deforested areas to reforested land. All other pathways maintain the existing land uses.

As seen in Table 14.2, the largest concentrations of carbon storage occur where humid tropical and warm temperate forests are allowed to regenerate. Higher rates of sequestration will be seen in Asia, Latin America, and Africa, where tropical biomes have higher net primary productivity than elsewhere, but also because greater land areas are forested in the tropics.

The pathways were designed to ensure that they do not negatively impact agriculture production; it does not completely eliminate the competition for land. Agroforestry should enhance agricultural productivity and has wide geographic applicability. Silvopasture could enhance it or could require reduced stocking rates. Silvopasture results in lower uptake, due to higher initial soil carbon content in temperate pasture lands compared to croplands. Both pathways result in rapid but temporary increase in carbon stocks (Littleton et al., 2021).


**Table 14.2** Gross regional carbon sequestration rates in ten world regions as categorized for the RCP database (Littleton et al., 2021)

Importantly, all ecosystem restoration pathways presented here reach the full extent in terms of area by 2040 and then held constant over the rest of the study period. This is coupled with the assumption that net deforestation will be halted by 2030. Without frm action to stop deforestation, gains made through the proposed ecosystem pathways will be offset by forest loss.

It is also important to realize that these ecosystem pathways do not and should not be used to offset fossil fuel emissions. Carbon uptake from land-based mitigation is slow and offers long-term temperature reduction. However, this approach needs to be implemented in conjunction with net-zero targets for other sectors not as a substitute. While removing more carbon from the atmosphere than is emitted into it would begin to reverse some aspects of climate change, some changes would still continue in their current direction for decades to millennia. The reversal of global surface warming lags the decrease in the atmospheric CO2 concentration by a few years (IPCC, 2021).

# **14.3 Managing Land Use**

This section discusses the impact of the ecosystem restoration pathways on existing land use and the land-use changes required for agriculture to meet the future food demand.

# *14.3.1 Mapping Land Use for Agriculture*

One of the biggest challenges in managing land use is the agricultural expansion required to feed 9 billion people in 2050. Based on the 2012 Food and Agriculture Organization (FAO) projections, the overall demand for agricultural products is expected to grow at 1.1% per year from 2005/2007 to 2050, which will result in a 60% increase globally by 2050 to meet the increased demand. Meeting this demand will require additional land for agriculture, but there is no consensus in the literature on how much more land will be required. The FAO projections indicate that about 70 million ha of additional land will be required for agricultural use in 2050 (Alexandratos & Bruinsma 2012). Several studies have discussed doubling production to meet the 2050 demand, particularly given the shift towards protein-rich diets and the consequent need for land to grow animal feed (Ray et al. 2013). Scenarios that do not link production with health and nutrition involve the expansion of agricultural lands into forests (Maggio et al. 2018). However, Hunter et al. (2017) disagree with the call to double agriculture production, largely because of recent production gains and because it is claimed that an increase of approximately 25%–70% above the current production levels should be suffcient to meet the 2050 demand. Conijn et al. (2018) noted that the planetary boundary for agricultural land

was already exceeded in 2010, and a 2050 scenario without effciency gains to meet the increased demand for food would require an increase of >3.5 Gha in agricultural land (grassland and cropland areas would increase by 78% and 67%, respectively). The FAO's latest alternative pathways to 2050 estimate that arable land must increase by 86 million ha from 2012 in the sustainability scenario and by 165 million hectares in the business-as-usual scenario.

Therefore, projections of the increased land required for agriculture range from 70 million ha to 3.5 billion ha. The FAO (2018) has identifed a global reserve of at least 400 million ha of suitable and unprotected land that could be brought under rain-fed cultivation. However, when losses to urbanization and degradation are considered, less than half of this reserve will be available. Data from the FAO– International Institute for Applied Systems Analysis (IIASA) Global Agro-ecological Zones (GAEZ v4) suggest that around 360 million ha of additional and unprotected areas and areas that are highly suitable for rain-fed crop production will be available by 2050. The majority of this land is situated in low- and medium-income countries.

All these scenarios involve increasing agricultural land at the expense of forests, and the resulting deforestation will have drastic consequences for the emission intensity of the sector. However, if a small expansion is coupled with the other strategies discussed in Chap. 6, there may be enough land to feed the 9 billion people estimated to exist in 2050 (FAO Forecast).

# *14.3.2 Mapping Land Use for Forestry*

Unlike agricultural land, forested land has been declining over time, and in 2020, 4 billion ha were recorded as under forest. An estimated 420 million ha of forest was lost through deforestation between 1990 and 2020, although the rate slowed over that period and the net reduction in the global forest area was about 178 million ha (FAO 2020a). Agriculture has driven an estimated 80% of the deforestation worldwide (FAO 2017). The global expansion of agricultural land has stabilized over the last 20 years at around 4.9 billion ha (FAO 2017).

The rate of net forest loss has been decreasing substantially as deforestation declines in some countries, whereas an increase in forest area has been seen in other countries, with both afforestation and the natural expansion of forests. However, there has been a reduction in the rate of forest expansion in the last decade (FAO 2020a).

Regional inequalities are not refected in this global overview. In tropical and subtropical regions, annual forest losses still amounted to 7 million ha in 2000–2010, whereas the agricultural area expanded by 6 million ha per year in the same period (FAO 2018). The largest reductions were observed in Brazil (down 53.2 million ha) and Indonesia (down 27.5 million ha). However, small increases were seen in Europe and the United States. The largest increase was in China, where the forest area was 51.2 million ha larger in 2015 than in 1990 (EUROSTAT 2020).

# *14.3.3 Implications for Decarbonization*

As seen in previous chapters, the *services* and *industry* sectors can decarbonize their energy emissions (i.e. *Scope 1* and *Scope 2* emissions) by incorporating energy effciency and transitioning to a 100% renewable energy source. The electrifcation of industry process heat, although harder to achieve, is another key step in the decarbonization pathway, and there is increasing innovation and technological development to support this. The largest challenge in reaching net-zero emissions remains the management of non-energy process emissions. The OECM model estimates 2.2 GtC will be released in unavoidable emissions annually in 2050 from the nine industrial sectors modelled in this study.

Ecosystem approaches can potentially remove CO2 from the atmosphere at the gigatonne scale, with potentially signifcant co-benefts, as discussed above (Meinshausen & Dooley, 2019). To achieve 93 GtC sequestration until 2100, land use must shift towards forest on over 350 million ha of land (Littleton et al., 2021).

The annual peak uptake calculated by Littleton et al. (2021) for all fve ecosystem pathways is 3.1 Gt/year in 2041 and 1.1 Gt C per year from 2050 to 2100. While in the short term this appears to provide an opportunity to offset non-energy-related industrial process emissions (e.g. from cement and steel production) that are diffcult to avoid with currently available technologies by using ecosystem approaches, in the long term these emissions must be reduced to zero or removed and stored geologically to prevent further warming.

Decarbonization pathways are being developed at the global level. At this level, there is little confict between the competing uses of cropland, pastureland, and forests for carbon removal. Adopting ecosystem approaches, such as agroforestry or silvopasture, where trees are integrated into cropland or grazing lands, will help to increase the carbon stock while meeting the increasing demand for forestry and agricultural products. It should be noted that a lot of deforestation and the capacity and demand for increased agricultural and livestock products will occur in tropical and subtropical regions, often in developing countries. At the local level, there must be a nuanced approach to addressing the balance between environmental, economic, and well-being outcomes.

# **14.4 Creating Carbon Conservation Zones (CCZ)**

The role of nature and ecosystem services as climate solutions is gaining increasing attention. As well as their climate mitigation and carbon sequestration potential, ecosystem approaches have co-benefts that contribute to sustainable development goals in terms of livelihoods, productivity, biodiversity conservation, health, and ecosystem services. However, it is important to note that even with ambitious landuse restoration, carbon removal can still only compensate for a small part of current emissions. The vast majority of emissive activities must cease if we are to achieve an approximately 1.5 °C target, and all the available removal strategies are required

to achieve net-negative emission pathways and reduce the atmospheric concentrations of CO2.

Feasible approaches to CDR using land-based mitigation options cannot be implemented in a vacuum but must address broader social and environmental objectives. Carbon conservation zones, which implement different ecosystem approaches, must address these broader objectives:

• Respecting indigenous rights and knowledge of land

Indigenous peoples and their connection to land play an important role in protecting and conserving nature and advancing climate solutions. This connection and their stewardship in protecting nature is particularly important in forested areas around the world. Several studies have found that the best forest protection is provided by people with collective legal titles to their land, i.e. by indigenous people (Fa et al., 2020; FAO and FILAC, 2021), and have recognized the contributions of indigenous knowledge to ecosystem-based climate solutions. For the frst time, COP26 formally acknowledged the roles and contributions of indigenous people's culture and knowledge in climate action and nominated indigenous peoples to engage directly with governments as knowledge holders and experts (2021a; UN Climate Change News, 2021).

Assisted natural regeneration strategies based on indigenous knowledge are promising ways to restore degraded lands (Schmidt et al., 2021). Formal recognition of indigenous people's rights over their forested lands can slow deforestation (Ricketts et al. 2010; Ceddia et al. 2015). These efforts must be supported by policies and actions that recognize collective territorial rights, provide compensation for environmental services, and allow community forest management, the revitalization of ancestral knowledge, and the strengthening of grassroots organizations and mechanisms for territorial governance (FAO and FILAC, 2021).

• Understanding fnancial implications

A study investigating the benefts of investing in ecosystem restoration found that tropical forests offered one of the highest value for restoration investment (after coastal and inland wetlands) (De Groot et al., 2013). Case studies across the world have also established that natural regeneration is signifcantly cheaper than tree planting, while simultaneously providing much higher carbon sequestration, but need to be incentivized by long-term funding mechanisms (Di Sacco et al., 2021). Much of the restoration opportunity identifed in this study lies in tropical forested developing countries, and fnancing incentives and support will be critical to ongoing success.

**R**educing **E**missions from **D**eforestation and Forest **D**egradation (REDD) is an effort to provide incentives through payment for results, allowing developing countries to reduce emissions from forested lands. REDD+ goes beyond addressing deforestation and forest degradation and fosters conservation, the sustainable management of forests, and the enhancement of forest carbon stocks. Initiatives like the Reforestation Accelerator are working with impact investment funds and innovative incubation ideas to provide seed funding to unlock ecosystem-based solutions (The Nature Conservancy, 2022). Such mechanisms can address the lack of fnancial support that is a major barrier to implementing ecosystem approaches.

• Protecting and conserving biodiversity

Reversing land degradation and limiting climate change depend upon retaining forests with high ecological integrity. A wide diversity of values and services tends to be found at higher levels in the more-intact forests of a given type. Biomass carbon stocks are a good example (Keith et al. 2009; Mackey et al. 2020), and forests and other ecosystems with no history of signifcant disturbance collectively absorb around 30% of anthropogenic carbon emissions annually (Friedlingstein et al. 2020).

Ambitious policies that prioritize the retention of forest integrity, especially in the most-intact areas, are now urgently required, in parallel with the current efforts to halt deforestation and restore the integrity of forests globally (Grantham et al., 2020). Higher levels of biodiversity generally support greater levels of ecosystem service production (e.g. carbon sequestration) than lower biodiversity levels, and ecosystem properties, such as resilience, are important considerations when managing human-modifed ecosystems (Ferreira et al., 2012). It is necessary to build on the synergies between climate action and activities directed towards conserving biodiversity.

• Infuencing supply chains and investment portfolios

Over the last decade, there has been a swell of industry-led commitments to zerodeforestation supply chains, but they are not yet implemented and many companies are yet to act (NYDF Assessment Partners 2020). The Carbon Disclosure Project's (CDP) Investor Report fagged that industry targets for net-zero deforestation are unlikely to be met unless commodity producers in the supply chain manage of their deforestation risk. This highlights the issue that certifcation is not enough and that companies require initiatives, such as education and fnancing, to promote sustainable agriculture and demonstrate strong policy commitments to end deforestation (Sin et al., 2020).

Forests and forest products are important parts of a number of supply chains for food, consumer goods, transport, etc., and companies and investors can play an important role in protecting and conserving nature through corporate commitments and by infuencing their downstream supply chains.

# **References**


*Global Environmental Change, 35*, 316–322, ISSN 0959-3780. https://doi.org/10.1016/j. gloenvcha.2015.09.010


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

Winterbottom, R. (2014). *Restoration: It's about more than just the trees*.

# **Part VIII Synthesis**

# **Chapter 15 Discussion, Conclusions, and Policy Recommendations**

# **Sven Teske, Thomas Pregger, Sarah Niklas, Kriti Nagrath, Simran Talwar, Souran Chatterjee, Benedek Kiss, and Diana Ürge-Vorsatz**

**Abstract** This section summarizes the main fndings of all parts of the research, with priority given to the most important fndings to avoid the repetition of previous chapters. The key fndings for the industry, services, buildings, and transport sectors, including the 12 sub-sectors analyzed, are provided and discussed. Policy recommendations for each sector and recommendations for the actions for governments, industries, the real economy, and fnancial institutions are offered.

**Keywords** Conclusion · Policy recommendations · Industry · Services · Financial institutions · Government policies

# **15.1 Background: Discussion of the Results with Academia, Industry, Government Agencies, and Financial Institutions**

In this section, we focus on the outcomes and conclusions of qualitative research rather than on the quantitative results documented in previous chapters. The most important technical measures are highlighted for each sector, followed by policy recommendations. This section refects extensive discussions and workshops with

University of Technology Sydney – Institute for Sustainable Futures (UTS-ISF), Sydney, NSW, Australia

e-mail: sven.teske@uts.edu.au

T. Pregger

S. Chatterjee · B. Kiss · D. Ürge-Vorsatz

Central European University, Department of Environmental Sciences and Policy, Budapest, Hungary

S. Teske (\*) · S. Niklas · K. Nagrath · S. Talwar

German Aerospace Center (DLR), Institute for of Networked Energy Systems (VE), Department of Energy Systems Analysis, Stuttgart, Germany

stakeholders from various industries and includes the recommendations of Teske et al. (2019). This chapter documents the key outcomes of two key research projects conducted between 2020 and late 2021:


# **15.2 Conclusion: High-Level Summary**

To comply with the Paris Climate Agreement and limit the global mean temperature rise to +1.5 °C, rapid decarbonization of the energy sector with currently available technologies is necessary and is possible.

However, to achieve the transformation to a fully renewable energy supply, all available effciency potentials must be combined to reduce the total demand. To reach *Net Zero* by 2050, the complete phase out of fossil fuels for all combustion processes is essential.

For the *industry sector*, the transition from fossil-fuel-based process heat to renewable energy or electrical systems is the single most important measure. The further reduction of non-energy-related process emissions—mainly from cement and steel manufacture—by altering or optimizing manufacturing processes is also essential. The remaining process emissions might be compensated by natural carbon sinks, so the *industry* sector must actively support the *service* sector in terms of soil regeneration and reforestation measures.

For the *service sector*, especially agriculture and forestry, reducing GHG emissions must clearly involve reducing the greenhouse gas (GHG) emissions arising from land-use changes. Increasing yield effciency to avoid the further expansion of *agricultural land* at the expense of forests and other important ecosystems is key. However, feeding the growing world population without increasing the area committed to agriculture will require more than just an increase in technical effciency. Moreover, there seems to be no alternative to reducing the consumption of meat and dairy products.

The *forestry sector* is the single most important sector for the implementation of nature-based carbon sinks. Deforestation must cease immediately. Reforestation with native trees and plants that are typical of specifc regions and climate zones must replace the forest areas that have been lost since 1990.

To reduce the demand of the *transport sector*, a shift from resource-intensive air and road transport to more-effcient and electrifed means of transport is required, together with an overall reduction in transport activity, especially in high-income countries. Phasing out the production of combustion engines for passenger cars by

2030 and introducing synthetic fuels for long-distance freight transport are essential elements for the future transportation sector. Even with this ambitious goal, the full decarbonization of the road transport sector will not be achieved before 2050, because cars are used, on average, for 15–20 years. There is also signifcant potential for effciency gains in shipping and aviation. However, due to the foreseeable further growth in traffc volume and the lack of alternatives, the large-scale use of synthetic fuels from renewable electricity will also be necessary for these modes of transport. Since not all regions will be able to produce this with domestic resources at reasonable costs, a global trade of these new energy sources must be established.

The decarbonization of the *buildings sector* will require a signifcant reduction in the energy demand for climatization—heating and cooling—per square metre. The key result of our research is that the global energy demand for buildings can be halved with currently available technologies. The utilization of this effciency potential will require high renovation rates and changed building codes for new constructions. The widespread use of heat pumps and heat grids is an important element on the supply side. In some areas, however, the supply of renewable gases can substitute today's natural gas consumption with a long-term perspective, especially where there is an industrial gas demand. The conversion of today's gas networks and the local/regional availability of resources for the production of green gases play a decisive role here.

Signifcant electrifcation across all sectors before 2030—especially for heating, to process heat, and to replace combustion engines in the *transport* sector—is the decisive and most urgent step. Increased electrifcation will require sector coupling, demand-side management, and multiple forms of storage (for heat and power), including hydrogen and synthetic fuels. Accelerating the implementation of renewable heat technologies is equally important because half the global energy supply must derive from thermal processes by 2050.

The transition of the global energy sector will only be possible with signifcant policy changes and reforms in the energy market.

The complete restructuring of the *energy and utilities sectors* is required. The primary *energy* sector—the oil, gas, and coal industry—must wind down all fossilfuel extraction and mining projects and move towards utility-scale renewable energy projects, such as offshore wind and the production of hydrogen and synthetic fuels.

Power utilities will play a key role in providing the rapidly increasing electricity demand, generated from renewable power. The nexus of the global energy transition will be the power grid. Replacing oil and gas with electricity means that power grids must transport most energy, instead of oil and gas pipelines.

*Therefore, the expansion of power grid capacities is one of the most important and also most overlooked measures required. In addition, converting existing gas pipelines and using them for the long-range transport of hydrogen and synthetic methane can signifcantly reduce the infrastructural demands on the power system and increase effciency.*

According to the scenario, global transmission and distribution grids must transport at least three times more electricity by 2050 than in 2020. The upgrades and expansion of power grids must start immediately because infrastructure projects, such as new power lines, can take 10 years or more to implement. Conversions of existing gas pipelines will be possible frst where industrial users need large quantities of hydrogen for decarbonized processes.

Limiting the global mean temperature rise to +1.5 °C cannot be achieved by the decarbonization of the energy sector alone. As stated earlier, it will also require signifcant changes in land use, including the rapid phase out of deforestation and signifcant reforestation. These measures are not alternative options to the decarbonization of the energy sector but must be implemented in parallel. If governments fail to act and mitigation is delayed, we face the serious risk of exceeding the carbon budget. In the *One Earth Climate Model* (OECM) 1.5 °C pathway, the land-use sequestration pathways complement very ambitious energy-mitigation pathways and should therefore be regarded as necessary to reduce the CO2 concentrations that have arisen from the overly high emissions in the past and not as compensatory measures that can be extended indefnitely into the future.

# **15.3 Industry Sector**

Policies to achieve the implementation of new highly effcient technologies and to replace fossil-fuel use in industry must be defned region-wide or even on the global level and will require stringent and regulated implementation. Economic incentives, national initiatives, and voluntary agreements with branches of industry will probably not, by themselves, achieve rapid technological change. Concrete standards and requirements must be defned in great detail, covering as far as possible all technologies and their areas of application. The systematic implementation of already-identifed best-available technologies should begin immediately.

Mandatory energy management systems must be introduced to identify effciency potentials and to monitor effciency gains. The sustainability features of process chains and material fows must also be considered when designing political measures. Particular attention must be paid to the material effciency of both production processes and their products, because this can open up major energy effciency potentials and reduce other environmental effects. Public procurement policies and guidelines will help to establish new markets and to introduce new, more-effcient products and opportunities. The effectiveness of policy interventions must be assessed by independent experts, and the further development of effciency programs and measures will require ongoing co-ordination by independent executive agencies. The public provision of low-interest loans, investment risk management, and tax exemptions for energy-effcient technologies and processes will signifcantly support technological changes and incentivize the huge investments required. Knowledge transfer between sectors and countries can be achieved through networks initiated and co-ordinated by governments. Public funding for research and development activities with regard to technological innovation, low-carbon solutions, and their process integration will be vital to push the technological limits further. Innovative approaches to the realization of material cycles and recycling options, the recovery of industrial waste heat, and low-carbon raw materials, and process routes in industry must also be identifed and implemented.

# *15.3.1 Steel Industry*

There are two key policy recommendations for the steel industry:

	- EAF processes
	- Hydrogen-based steel production

Although policies to support the transition towards a renewable energy supply are identical to those described for the *energy* and *utilities* sectors, support for mainstreaming steel production processes to reduce process emissions must be developed specifcally for the regional steel industry.

Research and development grants are required, as well as product certifcation schemes, to fnancially encourage changes towards new production lines. Steelprocessing industries, such as the automotive and construction sectors, require binding purchase quotas for CO2-neutral steel. CO2-intensive steel should gradually be made more expensive with a special 'steel tax', to further promote the production of 'green steel'.

# *15.3.2 Cement Industry*

Just as in the steel industry, the decarbonization of energy production for the cement industry has the highest priority in achieving short-term emission reductions. Reducing process emissions requires increased effciency along all steps of the production line. However, to date, no processes are available for the production of emissions-free cement. Therefore, nature-based carbon sinks must be established to compensate for the residual process emissions.

The Global Cement and Concrete Association (GCCA 2020) published a 2050 road map that set a 'long-term vision for the industry' that covers the following topics:


# *15.3.3 Chemical Industry*

The production of the main feedstocks for the chemical industry, such as ammonia, methanol, ethylene, and propylene, is almost entirely based on oil and gas but also on some biomass and coal. The refnery and production processes are very energy intensive. The production facilities are signifcantly different in each country and depend upon the company's product range. Therefore, universal policy recommendations are not possible.

However, the decarbonization of the chemical industry must focus on the following key areas:


The electrifcation of process heat will signifcantly increase the electric load for the production side. Therefore, in the transition from fossil- to electricity-based process heat generation, upgrading the power grids must also be considered, and planning must involve the local power-grid operator.

# **15.4 Land-Use and Non-energy GHGs in the Service Sector**

The key recommendations for the *service* sector focus on non-energy GHG emissions and especially the emissions associated with changes in land use (agriculture, forestry, and other land use, AFOLU). Although the transition to a renewable energy supply is a prerequisite for the decarbonization of the *service* sector, deforestation and other forms of land conversion must decline much more rapidly. Moreover, reductions in methane and nitrogen must also be achieved in the agriculture sector. Without nature-based solutions, the 1.5 °C limit will not be possible, even with a rapid decline in fossil-fuel emissions.

Four main natural sequestration pathways are utilized in the OECM, divided into temperate and tropical zones—reforestation, natural forest restoration, sustainable forest management, and cropland afforestation (trees in croplands):

1. *Wild lands* cover approximately 50% of the Earth's terrestrial area and are vital to the world's carbon cycle, sequestering up to one-quarter of anthropogenic carbon emissions and storing approximately 450 gigatonnes of solid carbon (Erb et al. 2018). Preserving these land and forest intact is key to maintaining our global carbon sinks, making the 1.5 °C limit possible.


#### **Planting Trees on Croplands**

Tree cropping—a strategy in which trees are planted within croplands—can signifcantly increase carbon storage on agricultural lands. The OECM estimates that planting trees on 400 Mha of cropland will achieve approximately 30 Gt of carbon removal by 2100.

The four sequestration pathways occur in all countries and regions, although we have excluded reforestation in the boreal forest zone because of the albedo effect.

All four sequestration pathways commence in 2020 but have different phase-in and phase-out rates, which also differ between the boreal/temperate and tropical/ subtropical biomes.


# **15.5 Transport Sector**

There are actionable measures in three key areas to decarbonize *transport* in line with the 1.5 °C target: *avoiding* or reducing the need to travel, *shifting* to moreeffcient transport modes, and *improving* effciency through vehicular technology. The implementation of these measure must take place until 2030 in order to reduce emissions suffciently rapidly.


# **15.6 Buildings Sector**

The in-depth HEB analysis (Chap. 7) demonstrates the potential to reduce the energy demand in the *buildings* sector with state-of-the-art high-effciency buildings, implemented worldwide. The fndings of the HEB analysis show that with a greater proportion of high-effciency renovations and construction, it will be possible to reduce the fnal thermal energy use globally in the building sector by more than half by 2060. For some regions, such as the EU and the Pacifc OECD, it will even be possible to achieve net-zero status for the thermal energy demand. However, this pathway towards high-effciency or net-zero emissions in the *buildings* sector is ambitious in its assumptions and requires strong policy support. On the contrary, if policy support to implement more high-effciency buildings is not in place, then the total thermal energy demand of the building sector will increase signifcantly over coming decades. Furthermore, if the use of energy effciency measures continues at the present rate, 67–80% of the fnal global thermal energy savings will be locked in by 2060 in the world building infrastructure. This lock-in effect in the *buildings* sector also means that if the present moderate energy performance levels become standard in new and/or retroftted buildings, it will be almost impossible to further reduce thermal energy consumption in these buildings for many decades to come.

Therefore, to realize the immense potential of the *buildings* sector, strong and ambitious policies are required. The fndings of our study are translated into the following policy recommendations:


buildings. Therefore, more education about low-carbon lifestyles must be provided.

4. Even with ambitious policy assumptions, the building sector will still consume substantial thermal energy globally, which may hinder the transition towards climate neutrality. Therefore, reducing the building energy demand must be accompanied by the promotion of building-integrated solar electric production. The fndings of the nearly net-zero scenario show that in developed regions, it will be possible to achieve net-zero status by 2060. Therefore, positive incentives should be given for on-site building-integrated solar energy production.

# **15.7 Energy and Utilities Sector**

The *energy* and *utilities* sectors may constitute separate categories for the fnancial sector, but for the energy sector, they are two sides of the same coin. The 1.5 °C pathway will lead to a 100% renewable electricity supply, with a signifcant share of variable power generation. The framework of the traditional electricity market has been developed for central suppliers operating dispatchable and limited dispatchable (base-load) thermal power plants. However, the electricity markets of the future will be dominated by variable generation, with no marginal or fuel costs. The power system will also require the build-up and economic operation of a combination of dispatch generation, storage, and other system services, the operation of which will be conditioned by renewable electricity feed-ins. For both reasons, a signifcantly different market framework is urgently required, in which the technologies can be operated economically and refnanced. Renewable electricity must be guaranteed priority access to the grid. Access to the exchange capacity available at any given moment should be fully transparent, and the transmission of renewable electricity must always have preference. Furthermore, the design of distribution and transmission networks, particularly for interconnections and transformer stations, should be guided by the objective of facilitating the integration of renewables and achieving a 100% renewable electricity system.

To establish fair and equal market conditions, the ownership of electrical grids should be completely disengaged from the ownership of power-generation and supply companies. To encourage new businesses, relevant grid data must be made available by the operators of transmission and distribution systems. This will require establishing communication standards and data protection guidelines for smart grids. Legislation to support and expand demand-side management is required to create new markets for the integration services for renewable electricity. Public funding for research and development is required to further develop and implement technologies that allow variable power integration, such as the smart-grid technology, virtual power stations, low-cost storage solutions, and responsive demand-side management. Finally, a policy framework that supports the electrifcation and sector coupling of the *heating* and *transport* sectors is urgently required to ensure a successful and cost-effcient transition process.

# **15.8 Policy Recommendations**

The OECM is an integrated energy assessment tool for the development of sciencebased targets for all major global industries in a granularity. It includes the key performance indicators (KPIs) required to make informed investment decisions that will credibly align with the global net-zero objective in the short, medium, and long term. The key fnding of our work on the OECM 1.5 °C cross-sectorial pathway is that it is still possible to remain with the 1.5 °C limit if governments, industries, and the fnancial sector act immediately. The technology required to decarbonize the energy supply with renewable energy is available, market ready, and in most cases, already cost competitive. The energy effciency measures needed to reduce the energy demand have also been understood for years and can be introduced without delay. Finally, the fnance industry—for instance, the Net-Zero Asset Owner Alliance—is committed to implementing carbon targets for its investment portfolios. However, policies are required to ensure that all measures are implemented in the rather short time frame required.

### **Implementing Short-Term Targets for 2025 and 2030**

To implement the documented short-term targets for 2025 and 2030, the following actions are required:

### **Government Policies:**


# **Actions Needed by Industry and Financial Institutions**

# *Industry:*


# *Financial Institutions:*

	- Climate mitigation strategies
	- Short- and mid-term target setting
	- Target achievements
	- Progress of climate solution investments
	- Engagement outcomes

# **References**


**Open Access** This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.

# **Annex**


#### **Fig. 1** Global electricity generation—OECM 1.5 °C

© The Editor(s) (if applicable) and The Author(s) 2022

S. Teske (ed.), *Achieving the Paris Climate Agreement Goals*, https://doi.org/10.1007/978-3-030-99177-7


#### Annex


3) incl. process heat, cooking

**Fig. 3** Global fnal heat demand—OECM 1.5 °C


**Fig. 4** Global transport fnal energy demand—OECM 1.5 °C




**Fig. 6** Global primary energy demand—OECM 1.5 °C


**Fig. 7** Global energy-related CO2 emissions—OECM 1.5 °C

# **Index**

#### **A**

Achieving the Paris Climate Agreements, 3 Active pharmaceutical ingredient (API), 84 Agrichemicals, 85 Agricultural chemicals production and processes, 85 products and materials, 85 uses and applications, 85 Agricultural product demand, 134, 137 Agricultural sector crop yield, 138 dietary changes, 139 environmental impacts, 139, 140 food waste, 138 land use, 138 Agriculture, 278 mapping land use, 344, 345 Agriculture and food sector, 327, 328 Agriculture, forestry, and fsheries (AFF), 132 Agriculture, forestry, and other land use (AFOLU) emissions, 133, 277, 327 Agroforestry, 341, 343 Alcoholic fermentation, 231 All-electric vehicles, 205 Aluminium, 102 Aluminium industry alumina refneries, 103 aluminium production, 103–108 aluminium resources, 103 bauxite mines, 103 bauxite production, 102 CO2 emissions, 108 electricity and process heat demands, 109 energy demand, 105, 106, 108 energy intensities, 105–108

global aluminium market development, 107 process- and energy-related CO2 emissions, 109 scope 1, 2, and 3 emissions, 323–325 Aluminium production, 103–108 Ammonia, 87, 89 Ammonium nitrate, 83 Anaerobic digestion, 230 Artifcial fbres, 86, 118 Automobile industry, 363 Aviation, 206 electric, 211 emissions factors, 196 energy demand and supply, 212 energy intensity, 195 energy-related CO2 emissions, 216 liquid fossil fuels, 211

#### **B**

Basic chemicals, 82, 83, 86 Battery electric vehicles (BEVs), 197 Bauxite production, 102 Bayer process, 106 Benzene, 87 Better Integration for Sustainable Energy (BISE) model, 166 Binary-cycle plants, 233 Bioenergy and biofuels biochemical conversion processes alcoholic fermentation, 231 anaerobic digestion, 230 biomass technologies, 228 direct combustion processes, 228 GHG emissions reduction, 231

© The Editor(s) (if applicable) and The Author(s) 2022 S. Teske (ed.), *Achieving the Paris Climate Agreement Goals*, https://doi.org/10.1007/978-3-030-99177-7

Bioenergy and biofuels (*cont.*) process heat, overview, 231 thermochemical processes direct combustion, 229 gasifcation, 229 pyrolysis, 230 Biomass, 228 Biomass carbon stocks, 348 Biomass gasifcation, 229, 230 Biomass technologies, 228 Black and organic carbon, 276 Black Summer, 4 Bottom trawling, 145 Bottom-up energy demand projections, 161, 165 Brown coal, 253 Buildings commercial, 168, 175, 177 residential, 168, 175, 177 Building sector, 69, 331, 332, 361 challenges, 162 HEB model (*see* High-effciency buildings (HEB) model) IPCC's ffth assessment report, 162 low-cost climate change mitigation, 162 process-related GHG emissions, 162

#### **C**

Calculation demand methods, 32 Carbon budgets, 12, 126 Carbon capture and storage (CCS) technologies, 240 Carbon conservation zones (CCZ), 346–348 Carbon dioxide (CO2) removal (CDR), 340 Carbon monoxide (CO), 276 Carbon Risk Real Estate Monitor (CRREM), 177 Carbonation, 99 Catch per unit effort (CPUE), 145 Cellulosic fbres, 86 Cement industry cement production, 100, 101 CO2 emissions, 100 companies and countries, 94, 95 COVID-19 impact, global cement production, 96 economic value, 93 electricity and process heat demand, 101 energy effciency standards, 96, 97 energy intensities, 96, 97, 100, 101 manufacture, 93 process-and energy-related CO2 emissions, 97–100, 102

scope 1, 2, and 3 emissions, 321, 323 steps in cement production, 93, 94 thermal energy intensity grey clinker production, 96 kiln type, 97 CH4 emissions, 277, 278 Chemical absorption, 100 Chemical industry, 358 basic chemicals, 82 chemical manufacturing, 83 CO2 emissions, 92, 93 companies and countries, 83 economic development, 89 electricity and process heat demand, 2050, 92 energy demand, 92, 93 energy fows, 88, 90 energy intensity, 83, 90, 91 GDP projections, 88 inorganic chemicals, 82 pharmaceuticals, 82 process and energy-related CO2 emissions, 93 raw materials, 82 scope 1, 2, and 3 emissions, 321, 322 sub-sectors, OECM analysis agricultural chemicals, 85 inorganic chemicals and consumer products, 85 manufactured fbres and synthetic rubber, 86, 87 petrochemicals, 87, 88 pharmaceuticals, 84, 85 Climate change, 247, 338 Climate modelling organisations, 5 Climate resource, 280 CO2 emissions, 92, 93, 99, 100, 108, 116, 123 CO2 separation technique, 100 Coeffcient of performance (COP), 236, 237 Commercial buildings, 175 Commodity chemicals, 82 Component-oriented methods, 49 Compression heat pumps, 236 Concentrated solar heat (CSH) system applications, 235 Concentrated solar power (CSP) advanced sensible heat-storage systems, 235 central receivers/solar towers, 234 CSH applications, 235 direct normal irradiation, 234 electricity, 235 elements, 234 generate power, 234

#### Index

hydrogen production, 235 latent heat-storage systems, 235 parabolic dishes, 234 parabolic trough plants, 234 thermal storage, 235 thermochemical energy storage, 235 Cost-effcient transition process, 362 Cotton, 118 COVID-19, 3, 96, 111, 120, 191, 248, 294 Crude oil, 253 Crude/primary steel, 110

#### **D**

Decarbonisation pathways, 143 transport (*see* Transport) Decarbonisation pathways for industries aluminium (*see* Aluminium industry) energy demand projections, 124, 125 global cement industry (*see* Cement industry) global chemical industry (*see* Chemical industry) OECM 1.5 °C pathways further research, 126 limitations, 126 textile and leather (*see* Textile and leather industry) total energy-related CO2 emissions, 125 total process heat demand, 125 Decarbonization pathways, 346 Decarbonizing process heat energy-intensive industries, 226 Deep effciency scenario, 163, 173 Deforestation, 358 Demand module architecture in MATLAB classes, 31 demand class, 32 energy-level array, 33 household and appliance classes, 32 household type object, 31, 33 industry application object, 33 object-oriented architecture, 33 object-oriented programming approach, 31 sub-sector and industry application classes, 32 transport modes and vehicle classes, 33 UML, 31 Dietary/lifestyle changes, 133 Diets, 139 Direct combustion technologies, 229 Direct energy-related CO2 emissions, 326 Direct heating systems, 299 Direct renewable process heat

bioenergy, 228–232 biofuels, 228–232 CSP, 233–235 geothermal, 232, 233 Dispatch strategy, 34 Dry steam plants, 233

#### **E**

Ecological engineering, 338 Economic energy intensity, 16 Ecosystem approaches to climate action climate action, 338 climate change, 338 climate management, 338 consequence concepts, 340 EbA, 338 ecological engineering, 338 landscape restoration, 338 millennium ecosystem assessment, 338 NbS, 338 variety of forms, 338 Ecosystem restoration pathways agricultural areas regeneration, 341 agroforestry, 341 cumulative uptake of CO2, 342 forest restoration, 341 forests and agricultural lands, 341 gross regional carbon sequestration rates, world regions, 343 methodology, 341, 342 results, 342–344 Electric arc furnaces, 239 Electric aviation, 211 Electric heating systems, 225 electric arc furnaces, 239 electric resistance heating, 238, 239 electromagnetic, 238 heat pumps, 236 non-thermal, 238 types, 237 Electric outboard engines, 147 Electric process heat electric heating systems, 237–239 heat pumps, 236, 237 Electric resistance heating, 238, 239 Electric vehicles (EVs), 197 Electromagnetic heating systems, 238 Emissions intensity, 216 Energy, 19, 98, 316–318 Energy and climate mitigation pathways, 26 Energy and industrial CO2 emissions pathways, 279 Energy and utilities sectors, 265, 267

Energy-consuming system, 90 Energy consumption, 133 Energy demand, 90, 92, 93, 100, 105, 106, 108, 116, 123, 191 Energy industry, 265 coal, 248 COVID-19 pandemic, 248 fossil fuel decline rates, 250 GICS, 248 global coal production, 253 global coal trajectory—OECM 1.5 °C, 252 global gas production, 256 global gas trajectory—OECM 1.5 °C, 255 global lignite trajectory—OECM 1.5 °C, 253 global oil production, 254 global oil trajectory—OECM 1.5 °C, 254 global renewables trajectory under OECM 1.5 °C, 256, 257 hydrocarbons, 248 oil and gas, 248 Energy intensities, 29, 47, 56, 83, 90, 91, 100, 101, 105–108, 114, 121–123, 135, 195–197, 200 Energy intensity of economy, 105 Energy Performance of Building Directive (EPBD), 163 Energy-related CO2 emissions, 10, 18, 117, 135, 136, 143, 216, 288 black and organic carbon and carbon monoxide, 276 combustion processes, 274 fugitive CO2 emissions, 274 industrial process emissions, 274, 275 industry sectors, 276 Energy-related emissions, 98 Energy services, 28 Energy statistical databases adaptation to GICS industry system buildings sector, 69 industry sector, 66–68 sectorial energy scenarios, industry sectors to emissions, 72 service sector, 69 transport sector, 69, 71, 72

Energy system model (EM), 26, 42 Enhanced geothermal system (EGS)

exploration and drilling, 233 geothermal power and heat plants, 233 reservoir engineering, 233 Entrained-fow combustion, 229 Environmental burden, 139 EU energy performance, 182

European Economic and Social Committee, 148 Excipients, 84 Extreme weather events, 4

#### **F**

FAO 2030 Agriculture Outlook, 139 Fashion industry, 118, 121 Final electricity demand demand shares by sector under the OECM 1.5 °C pathway 2019 and 2050, 295 electrifcation of heat, 294 sector under the OECM 1.5 °C pathway in 2019–2050, 295 Final energy consumption, 303 Final energy demands of energy-intensive industries, 304–307 Final global energy demand, 192 Finance industry, 18, 72, 248 Financial institutions, 364 Find demand method, 32 Fisheries, 157 Fisheries sector, 329, 330 Fixed-bed combustion applications, 229 Flash plants, 233 Fluidized-bed combustion, 229 Fluidized-bed technology, 229 Food and Agriculture Organization (FAO), 133, 344 Food equity, 137 Food processing, 156 Forest restoration, 341 Forestry, 345 Forestry and wood sector, 328, 329 Forestry-related AFOLU emissions, 329 Formulation production, 84 Fossil-fuel-based energy, 276 Fossil-fuel-based power generation, 298 Fossil-fuel-free circular carbon economy, 243 Freight transport, 200, 202 Frozen effciency scenario, 163, 173 Fuel cell-powered passenger vehicles, 213 Fuel emissions, 98 Fugitive CO2 emissions, 274

#### **G**

Gas pipelines, 356 Gas utilities, 267 existing natural gas pipelines to transport hydrogen, 263

#### Index

gaseous fuel supply under the OECM 1.5 °C scenario, 265 OECM 1.5 °C scenario, 261 GDP development projections, 133 Generalized Equal Quantile Walk (GQW), 140 Geothermal, 232, 233 Geothermal energy, 227 Geothermal energy resources, 232 GHG emissions, 12–14, 26, 358 scopes 1, 2, and 3, 72 Global agriculture and food sector commodities, 132 economic sector, 132 energy demand projection, 133–136 food demand and implications, 137 meeting global food demand, 138–140 Global and regional transport demands, 189, 192 Global carbon budget, 11 bottom-up demand analysis, 311 end-use sector under the OECM 1.5 °C pathway, 311 sub-sector under 1.5 °C OECM pathway, 2020–2050, 312 Global CO2 emissions, 4 emission factors, 309, 310 OECM 1.5 °C pathway, 310 supply source, 309 Global economic development, 14 Global electricity generation, 294, 296, 367 Global electricity supply annual onshore wind market, 297 China, 296 COVID-19 pandemic, 294 fnal electricity demand, 294 fossil-fuel-based power generation, 298 global installed capacity, 295 global installed power plant capacities OECM 1.5 °C pathway, 2019–2050, 297 global power plant capacities, 296 OECM 1.5 °C pathways, 294 shares under the OECM 1.5 °C pathway, 296 Global energy demand, 131, 209 Global energy-related CO2 emissions, 4, 13, 372 Global fnal heat demand, 369 Global fsheries sector aquaculture, 144 economic (frst sale) value, 144 energy and electricity supply, 147 energy demand projection, 145–147 marine and aquatic ecosystems, 145

OECM, 145 total protein intake, 144 unsustainable fshing methods, 145 Global food system, 133 Global forestry and wood sector assumed energy intensities, 142 energy demand projection, 141–143 food security, 140 GICS, 140, 141 global economic development, 141 land-use demand, 143, 144 products, 140 Global gross domestic product (GDP), 81, 88 Global heat supply, 225, 302, 303 calculated capacities for heat generation, 303 capacity factor, 303 electric heat systems, 302 global phase-out of coal, heat production, 302 high-temperature heat generation, 302 OECM 1.5 °C pathway, 302 process heat supply in 2019, 301 solar thermal process, 302 Global Industry Classifcation Standard (GICS), 27, 44, 45, 131, 132 industries, 62 net-zero targets individual end-use sectors, 62 OECM 1.5 °C industry pathways, 63 OECM 1.5 °C service and energy pathways, 64–66 scope 1, 2, and 3 emissions, 72–76 Global installed power generation capacities, 368 Global mean surface air temperature (GSAT) projections, 286 Global oil demand, 190 Global power plant capacities, 296, 298 Global primary energy demand, 307, 308, 372 Global space and process heat supply cogeneration plants, buildings and services, 299 electricity demand shares by sector OECM 1.5 °C pathway in 2019 and 2050, 299 global heat demand by sectors under the OECM 1.5 °C pathway, 300 industry process heat demand by temperature level aluminium, steel, and chemical industries, 301 OECM 1.5 °C pathway in 2019–2050, 300, 301

Global space and process heat supply (*cont.*) industry sectors, 299 services and buildings, 299 Global steel production, 111 Global transmission and distribution grids, 355 Global transport demand projections actions, 200 behavioural changes development in passenger travel, transport mode, 204 changes in freight logistics, transport mode, 204 changing demand in kilometres, 208 energy intensities freight transport modes, 2019, 204 urban and interurban passenger transport modes, 2019, 203 global and regional modal shifts, 206 improvements over time, 204 mode-specifc technology effciency, 204 powertrain electrifcation, road transport, 205 service demand, 203 transport energy demand, 203, 207–209 transport service, 209 world freight transport by mode under the 1.5 °C scenario, 207 world passenger transport by mode under the 1.5 °C scenario, 206 Global transport fnal energy demand, 370 Global transport technologies, 194 Global warming potential (GWP), 157, 278, 279 Global water utilities sector assumed global energy intensities, 153 commodity demand projections, 149, 150 energy consumption, 148 energy demand projection, 152–155 energy effciency standards, 151, 152 energy-related CO2 emissions, 154, 156 fraction, 148 GICS, 148 OECM pathways, 148 potable drinking water, 148 privatisation, 148 Global water withdrawal quantities, 150 Greenhouse gas (GHG), 162, 231, 316 Greenhouse gases and aerosols AFOLU, 277 CH4 emissions, 277, 278 emissions categories by source/removal by sinks, 275 energy-related CO2 emissions, 274, 276 GWP, 278, 279

N2O emissions, 277 Grid services, 37 Gross domestic product (GDP), 165

### **H**

Hard coal, 252 Harmonization, 282 Heat generation, 34, 225 Heat pump technology, 236, 237 Heavy-duty machinery, 154 Hexafuoroethane (C2F6), 285 High-effciency building (HEB) model bottom-up approach, 162 building classifcation scheme, 51 building sector-related energy demand, 49 component-oriented methods, 49 data and assumptions, 165–167 disaggregation, 51–53 energy consumption, 56 energy demand, 169 fnal energy use, 171, 173–176 foor area projections, 162 foor area, 167–171 implementation, 56 key fndings, 176 key output, 50 methodology, 49, 50 regional breakdown, 163 scenarios, 163, 164 total foor area, 54 workfow, 51 yearly dynamics of foor area changes, 54, 55 High-enthalpy geothermal heat, 237 High-temperature geothermal plants, 234 High-temperature geothermal systems, 232 Hydrocarbons, 248 Hydrogen, 262 CO2-intensive processes, 240 demand, 240 feedstock, 239 future applications, 241 in industry, 240, 241 natural gas, 240 oxidation, 239 processes for hydrogen production, 242 refneries, 240 SMR, 240

### **I**

IEA Advanced Energy Balances, 141 IEA statistics, 68, 69

IEA World Energy Balances, 131 industry sector defnition, 67–68 other sectors defnition, 70–71 transport sector defnition, 71 Illegal, unreported, and unregulated (IUU), 145 Indigenous ecosystems, 338 Individual transport modes, 198–199 Industrial process emissions, 274, 275 Industrial Processes and Product Use (IPPU), 19 Industry energy demand, 30 Industry in Europe process heat demand distribution, 226 Industry sector, 66–68 Industry-specifc data, 157 Industry-specifc energy intensities, 47 Industry-specifc energy statistics, 157 Inorganic chemicals, 82 Inorganic chemicals and consumer products production and processes, 86 products and materials, 85 uses and applications, 86 Input parameters, 28, 29 Intergovernmental Panel on Climate Change (IPCC), 25, 309 International fashion industry, 117 International Production Trajectory, 249 International Renewable Energy Agency (IRENA), 258 International Union for Conservation of Nature (IUCN), 338 Investment decisions, 61 Iron and steel sector, 68

#### **J**

Joint UK Land Environment Simulator (JULES), 341

#### **K**

Kerosene, 196 Key performance indicators (KPIs), 47–49, 72, 363

#### **L**

Land transport electric-powered planes/ships, 197 emission factors, 197, 200 energy intensity, 197, 200 road, 197 Land use management

agriculture, 344, 345 forestry, 345 implications, decarbonization, 346 Lightning, 239 Linear Fresnel systems, 234 Liquid synthetic hydrocarbons, 244

#### **M**

Mandatory energy management systems, 356 Manufactured fbres and synthetic rubber production and processes, 86 products and materials, 86 uses and applications, 86 Manufacturing sector, 66 Mapping land use agriculture, 344, 345 forestry, 345 Maritime shipping, 206 MATLAB, 27, 31, 39, 40, 42 MATLAB-based OECM, 34 Methanol, 89 Moderate effciency scenario, 163, 170, 173 Molecular hydrogen, 239 Morgan Stanley Capital International (MSCI), 62

#### **N**

Nationally Determined Contributions (NDCs), 19, 20 Natural fbre crops, 117 Natural fbres, 87, 117, 118 Natural gas, 88, 240, 251, 255 Nature-based approaches, 143 Nature-based solutions (NbS), 338 Nearly net-zero scenario, 163 Net-Zero Asset Owner Alliance (NZAOA), 266 Net-zero pledges, 20, 21 Net-zero targets, 4 Nitrogen fertilisers, 90 Non-1A category emissions, 280–282 Non-energy emissions, 140 Non-energy GHG emissions, 144 Non-energy GHG modelling emissions in the SR1.5 database, 283, 284 emissions not in the SR1.5 database, 284 energy and industrial CO2 emissions pathways, 279 exceedance probabilities, 285–287 extending emissions to 2100, 283 GSAT projections relative to 1850–1900, 286

Non-energy GHG modelling (*cont.*) harmonization, 282 inflled emissions compared with the SSP scenarios, 285 non-1A category emissions, 280–282 SSP scenarios, 280 SSP1-1.9 scenario, 280 temperature projections, 285, 286 Non-energy GHG pathways, 4 Non-energy-related GHG emissions, 327 Non-residential buildings, 165 Non-thermal electrical systems, 238 Nuclear power generation, 259

#### **O**

OECM 1.5°C global electricity generation, 367 global energy-related CO2 emissions, 372 global fnal energy demand, 371 global fnal heat demand, 369 global installed power generation capacities, 368 global primary energy demand, 372 global transport fnal energy demand, 370 OECM 1.5 °C fnal and primary energy balances fnal energy demands of energy-insensitive industries, 304–307 global primary energy demand, 307 industry, 303, 304 service and buildings sector, 304 OECM 1.5 °C industry pathways, 63 OECM 1.5 °C net-zero pathway, 309 OECM 1.5 °C service and energy pathways, 64, 65 OECM 2.0, 47–49 OECM demand module bottom-up approach, 28 energy requirements, 28 energy services, 28 input parameters, 28, 29 load profles, 28 MATLAB, 31 sectors, 28 structure industry sector, 29 residential sector, 29, 30 transport sector, 29, 30 synthetic load profles, 28 OECM supply module dispatch model input parameter, 35

inputs, intermediate outputs, and outputs, 38 limitations, 37 output parameters, 36 technology groups, 36 technology options—dispatch generation, 37 technology options—storage technologies, 37 technology options—variable renewable energy, 36 dispatch module, 34 dispatch strategy, 34 elements, 34 generation technologies, 35 heat generation technologies, 34 regional energy demand, 34 regional interconnections, 37, 39 storage technologies, 34, 35 Offshore wind energy, 267, 268 Offshore wind farms, 260 Ohmic resistance, 238 Oil and gas, 248 The One Earth Climate Model (OECM), 4, 161, 188, 266 architecture, 26, 27 assumed population and GDP developments, region in 2020–2050, 11 climate and energy pathways, 61 cost calculation, 46 databases, 42, 44 demand module (*see* OECM demand module) energy demand per capita, 16 further research demand, 47, 49 future energy demand, 15 GHG emissions development, 12–14 global development of GDP per capita, 16 global development of key parameters, 17 global economic development, 14 global energy demand, 16 global energy intensity, 16, 17 global energy-related CO2 emissions, region in 1750–2020, 13, 14 global GDP development in 1700–2015, 15 industry sector-based energy scenarios, 26 industry sub-sectors based on the GICS, 45 model calibration, 42–44 NZAOA, 26 OECM 2.0 output and area of use, 47 sectors, 44, 46 socio-economic assumptions, OECM 1.5 °C Scenario, 15

sub-sectors, 44, 46 supply module, 34–42 UN-convened *Net-Zero Asset Owner Alliance*, 11 One Earth summary graph, 286, 287 Operation and maintenance (O&M) costs, 46

#### **P**

Pandemic, 191 Parabolic trough plants, 234 Paris Agreement, 247 Paris Climate Agreement, 10, 18, 19, 25, 61, 354 Passenger transport, 207 Petrochemicals feedstocks, 89 production, 89 production and processes, 87 products and materials, 87 uses and applications, 88 Petroleum products, 88 Pharmaceutical products, 85 Pharmaceuticals, 82 production and processes, 84 products and materials, 84 uses and applications, 85 Policy framework, 362 Policy recommendations, 353, 361 Population, 28 Post-combustion carbon capture technologies, 100 Power-based synthetic fuels, 227 Power plant technology, 295 Power system analysis model, 27 Power utilities, 266, 267 distribution, 261 electricity system services, 261 electricity under the OECM 1.5 °C scenario, 265 power generation, 259 secondary energy industry, 258 transmission, 260 Power-to-X (PtX), 242 Primary energy demand analysis, 249 Primary energy industry, 249, 256 challenge, 249 OECM 1.5 °C pathway, 249 Primary energy sector, 278 Printing result methods, 32 Process-related emissions, 98 Public transport, 200, 201, 360 PyData ecosystem, 56

Pyrolysis, 230 Python programming language, 56

#### **Q**

Qualitative research, 353

#### **R**

Rail transport, 201 Reducing Emissions from Deforestation and Forest Degradation (REDD), 347 Refning/smelting, 106 Reforestation, 341, 354 Regional energy demand, 34 Regional interconnections, 37, 39 Renewable electricity for heating, 226 Renewable power, 262 Renewable process heat direct (*see* Direct renewable process heat) electric (*see* Electric process heat) electricity and heat share*s,* industry in 2019, 227, 228 hydrogen, 239–242 synthetic fuels, 243, 244 technologies, 226 temperature level, 227 Reservoir engineering, 233 Residential building sector, 175 Residential sector, 167 Reversible heat pumps, 237 Road transport, 189, 197–199, 201, 206 energy demand and supply, 214 energy-related CO2 emissions, 216 powertrain electrifcation, 205

#### **S**

Science-based GHG targets AR6 of the IPCC, 10, 11 assumed population and GDP developments, region in 2020–2050, 11 energy demand per capita, 16 global GDP development in 1700–2015, 15 OECM future energy demand, 15 GHG emissions development, 12–14 global development of GDP per capita, 16 global development of key parameters, 17 global economic development, 14

Science-based GHG targets (*cont.*) global energy demand, 16 global energy intensity, 16, 17 global energy-related CO2 emissions, region in 1750–2020, 13, 14 socio-economic assumptions, OECM 1.5 °C Scenario, 15 UN-convened Net-Zero Asset Owner Alliance, 11 science-based target setting, 18–21 Science-based target setting fnance industry, 18 NDCs, 19, 20 net-zero pledges, 20, 21 OECM, 18 Paris Climate Agreement, 18 sectorial targets development, 18 UNFCCC, 18 Scope 1, 2, and 3 emissions, 72–74, 76 buildings sector, 331, 332 defnitions, 316 energy, 316–319 energy-related CO2 emission*s* in 2019, 334 in 2030 under the OECM 1.5 °C pathway, 335 industry aluminium, 323–325 cement, 321, 323 chemical, 321, 322 steel, 325, 326 textile and leather, 326, 327 industry-specifc emission budgets, 315 service sectors agriculture and food, 327, 328 fsheries sector, 329, 330 forestry and wood, 328, 329 water utilities, 330 transport sector, 331–333 utilities sector, 318, 320 Secondary aluminium production, 107 Secondary steel production, 110 Sequestration pathways, 359 Service sector, 69, 131 Sewage treatment plant (STP), 151 Shared Socioeconomic Pathway (SSP) scenarios, 280 Shipping emission factors, 196, 197 energy demand and supply, 213 energy intensity, 196, 197 energy-related CO2 emissions, 216 global energy demand, 196 Silvopasture, 341, 343

Sixth Assessment Report (AR6), 10 Solar photovoltaic generators, 260 Solar radiation, 234 Solar thermal process, 302 SSP emissions pathways, 285 State-of-the-art technology, 231 Steam–methane reformation (SMR), 240 Steel, 109, 110 Steel industry, 357 assumed market and energy intensity developments, 115–116 CO2 emissions, 116, 117 crude steel production, 111 electricity and process heat demands, 116 energy demand, 116, 117 energy intensity, 114 EU-ETS benchmark values, iron and steel manufacture, 114 global crude steel production data by country, 110 global production, 2019, 112 primary and secondary steel production, 110, 111 process-and energy-related CO2, 117 production, 114 scope 1, 2, and 3 emissions, 325, 326 steel production energy requirements, 113 main processes, 113 technological overview, steel production, 112, 113 Storage technologies, 34, 35 Subsistence fshing, 145 Supply module architecture in MATLAB object-oriented structure, 39 storage technology class, 40 storage technology object, 40, 42 supply class, 39, 41 supply technology class, 39 supply technology object, 40, 41 UML diagram, 39, 40 Synthetic fbres, 86, 117 Synthetic fuels, 243, 244

### **T**

Tetrafuoromethane, 324 Textile and leather industry agricultural output, 117 artifcial fbres, 118 chemical industry products, 117 CO2 emissions, 123 companies and countries, 119, 120 cotton, 118

#### Index

COVID-19 impact, global textile production, 120 economic development, 122 electricity and process heat demands, 123 emissions, 121, 122 energy demand, 123 energy intensities, 121–123 fashion industry's environmental impact, 118, 119 international fashion industry, 117 natural fbre crops, 117 natural fbres, 117, 118 process- and energy-related CO2 emissions, 124 production, 122, 123 recycled fbres, 118 resource requirements, 120, 121 scope 1, 2, and 3 emissions, 326, 327 synthetic fbres, 117 yarn and fabric production, 118 Textile production, 120 Textiles, 117 Thermal energy, 361 Thermal storage, 235 Total water withdrawal, 149 Traditional electricity market, 362 Transformative Urban Mobility Initiative (TUMI), 192 Transport aviation, 195, 211, 212 demand (*see* Transport demand) economies, 188 electrifed passenger and freight rail in 2019, 215 energy-related CO2 emissions, 216 equipment, 217 calculated energy-related CO2 emissions, 219, 220 defnition, 217 energy demand by sub-sector, 220 energy intensities, 219 GDP values, 217, 219 global GDP shares, 217, 218 industries classifcation, 217 fnal energy demand and supply, 215 fnal global energy demand, 2019, 187 land, 197–200 mode, 188 OECM in 2021, 188 powertrains in passenger cars/buses by region, 2030 and 2050, 214 region in 2020–2050 GDP developments, 191 population, 191 renewable electricity, 215

shipping, 196, 197, 213 socio-economic assumptions, 189 sub-sectors, 188 world regions, 1.5°C OECM, 190 Transport demand factors, 188 fnal energy use by global transport, 2019, 194 global and regional, 189, 192 global fnal energy use by transport mode, 2019, 193 global transport technologies, 194 non-energy-related factors, 188 transport mode performances aviation, 192 rail, 193 road, 193 Transport demand calculations calibration, 210 Transport energy, 135 Transport energy demand, 30, 207–209 Transport energy model (TRAEM), 27 Transport sector, 69, 71, 72, 331–333 Transport service, 203, 209 Transport supply transport mode, 210 vehicle-specifc parameters, 210 Tree cropping, 359

#### **U**

UN-convened Net-Zero Asset Owner Alliance (NZAOA), 11, 26 Unifed modelling language (UML), 31 United Nations Development Programme (UNDP), 165 United Nations Framework Convention on Climate Change (UNFCCC), 3, 18 United Nations Intergovernmental Panel on Climate Change (IPCC), 4 United States Environmental Protection Agency (US EPA), 72 Urban residential buildings, 173 Urban transport, 206 US Energy Information Administration (EIA) database, 165 Utilities sector electricity and gaseous fuel supplies, 318 fossil-fuel-based electricity, 259 gas existing natural gas pipelines to transport hydrogen, 263 OECM 1.5 °C scenario, 261 secondary energy industry, 258 global energy transition, 266

Utilities sector (*cont.*) OECM 1.5 °C pathway, 258 power distribution, 261 electricity and gas distribution under the OECM 1.5 °C, 264 electricity system services, 261 power generation, 259 secondary energy industry, 258 transmission, 260 primary energy industry and end-use sector, 258 scopes 1, 2, and 3, 320 secondary energy service, 318 total global electricity, 258 world's largest electric utility companies (Statista 2021), 258

### **V**

Vulcanisation, 86

# **W**

Wastewater collection, 151 Wastewater treatment plants (WWTP), 151 Water extraction, 151 Water transport, 207 Water utilities, 69, 330 Wood processing, 157 World Bank Databases, 165 World Meteorological Organization (WMO), 4